Smithsonian contributions to the marine sciences

mithsonian Institution
Scholarly Press

Proceedings of the
Smithsonian
Marine Science Symposium

Edited by
Michael A. Lang,
Ian G. Macintyre, and Klaus Riitzler

SERIES PUBLICATIONS OF THE SMITHSONIAN INSTITUTION

Emphasis upon publication as a means of “diffusing knowledge” was
expressed by the first Secretary of the Smithsonian. In his formal plan
for the Institution, Joseph Henry outlined a program that included

the following statement: “It is proposed to publish a series of reports,
giving an account of the new discoveries in science, and of the changes
made from year to year in all branches of knowledge.” This theme of
basic research has been adhered to through the yea:s by thousands of
titles issued in series publications under the Smithsonian imprint, com-
mencing with Smithsonian Contributions to Knowledge in 1848 and
continuing with the following active series:

Smithsonian Contributions to Anthropology
Smithsonian Contributions to Botany

Smithsonian Contributions in History and Technology
Smithsonian Contributions to the Marine Sciences
Smithsonian Contributions to Museum Conservation
Smithsonian Contributions to Paleobiology
Smithsonian Contributions to Zoology

In these series, the Institution publishes small papers and full-scale
monographs that report on the research and collections of its various
museums and bureaus. The Smithsonian Contributions Series are
distributed via mailing lists to libraries, universities, and similar institu-
tions throughout the world.

Manuscripts submitted for series publication are received by the
Smithsonian Institution Scholarly Press from authors with direct affilia-
tion with the various Smithsonian museums or bureaus and are subject
to peer review and review for compliance with manuscript preparation
guidelines. General requirements for manuscript preparation are on the
inside back cover of printed volumes. For detailed submissions require-
ments and to review the “Manuscript Preparation and Style Guide for
Authors,” visit the Submissions page at www.scholarlypress.si.edu.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES @© NUMBER 38

re,
Proceedings of the
Smithsonian
Marine Science
Symposium

Edited by
Michael A. Lang,
Ian G. Macintyre, and Klaus Riitzler

oe Issued

Smithsonian Institution NOV 2
Scholarly Press :
WASHINGTON D.C.
2009

Smithsonian Institution

ABSTRACT

Lang, Michael A., Ian G. Macintyre, and Klaus Rutzler, editors. Proceedings of the Smithsonian
Marine Science Symposium. Smithsonian Contributions to the Marine Sciences, number 38, 529
pages, 217 figures, 47 tables, 2009.—The Smithsonian Marine Science Symposium was held on
15-16 November 2007 in Washington, D.C. It represented the first major dissemination of ma-
rine research results since the establishment of the Smithsonian Marine Science Network (MSN).
The 39 papers in this volume represent a wide range of marine research studies that demonstrate
the breadth and diversity of science initiatives supported by the MSN. The first section contains
an overview of the MSN along with papers describing the multidisciplinary investigations span-
ning more than 37 years for the four Smithsonian marine facilities that constitute the Network:
the Smithsonian Environmental Research Center at the Chesapeake Bay, Maryland; the National
Museum of Natural History’s Smithsonian Marine Station at Fort Pierce, Florida; the Carib-
bean Coral Reef Ecosystems Program, with its Carrie Bow Marine Field Station in Belize; and
the Smithsonian Tropical Research Institute in Panama. Subsequent papers represent findings
by Smithsonian scholars and their collaborators on overarching topics of marine biodiversity,
evolution, and speciation; biogeography, invasive species, and marine conservation; and forces of
ecological change in marine systems.

Cover images: (left) Aurelia aurita sea jelly with juvenile carangid jacks in its bell, Carrie Bow
Cay, Belize; (middle) Dendronephthya soft corals and Anthias school, The Brothers Islands, Red
Sea, Egypt; (right) grey reef shark Carcharhinus amblyrhynchos, Kingman Reef, Northern Line
Islands (all photos by Michael A. Lang).

Published by Smithsonian Institution Scholarly Press
P.O. Box 37012

MRC 957

Washington, D.C._20013-7012
www.scholarlypress.si.edu

Library of Congress Cataloging-in-Publication Data
Smithsonian Marine Science Symposium (2007 : Washington D.C.)

Proceedings of the Smithsonian Marine Science Symposium / edited by Michael A. Lang, Ian
G. Macintyre, and Klaus Riitzler.

p. cm. — (Smithsonian contributions to the marine sciences, ISSN 0196-0768 ; no. 38)

Includes bibliographical references and index.

1. Marine sciences—Congresses. I. Lang, Michael A. II. Macintyre, Ian G. III. Rutzler, Klaus.
IV. Title

GC2.S57 2007

578.77—dc22

2009028023

ISSN (print): 0196-0768
ISSN (online): 1943-667X

6°) The paper used in this publication meets the minimum requirements of the American National
Standard for Permanence of Paper for Printed Library Materials Z39.48-1992.

Contents

FOREWORD by Ira Rubinoff
EXECUTIVE SUMMARY by Michael A. Lang
OVERVIEW OF SMITHSONIAN MARINE SCIENCE

Introduction to the Smithsonian Marine Science Network
Michael A. Lang, Smithsonian Office of the Under Secretary
for Science

Land-Sea Interactions and Human Impacts in the Coastal Zone
Anson H. Hines, Smithsonian Environmental Research Center

Smithsonian Marine Station at Fort Pierce: Thirty-Eight Years
of Research on the Marine Biodiversity of Florida
Valerie J. Paul, Smithsonian Marine Station at Fort Pierce
Julianne Piraino, Smithsonian Marine Station at Fort Pierce
Laura Diederick, Smithsonian Marine Station at Fort Pierce

Caribbean Coral Reef Ecosystems: Thirty-Five Years of Smithsonian
Marine Science in Belize
Klaus Riitzler, National Museum of Natural History,
Smithsonian Institution

The Smithsonian Tropical Research Institute: Marine Research,
Education, and Conservation in Panama
D. Ross Robertson, Smithsonian Tropical Research Institute
John H. Christy, Smithsonian Tropical Research Institute
Rachel Collin, Smithsonian Tropical Research Institute
Richard G. Cooke, Smithsonian Tropical Research Institute
Luis D’Croz, Smithsonian Tropical Research Institute
Karl W. Kaufmann, Smithsonian Tropical Research Institute
Stanley Heckadon Moreno, Smithsonian Tropical Research Institute
Juan L. Maté, Smithsonian Tropical Research Institute
Aaron O’Dea, Smithsonian Tropical Research Institute
Mark E. Torchin, Smithsonian Tropical Research Institute

xi

11

25

43

73

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

MARINE BIODIVERSITY, EVOLUTION, AND SPECIATION

Protandric Simultaneous Hermaphroditism Is a Conserved Trait
in Lysmata (Caridea: Lysmatidae): Implications for the Evolution
of Hermaphroditism in the Genus 95
J. Antonio Baeza, Smithsonian Marine Station at Fort Pierce

Reconciling Genetic Lineages with Species in Western Atlantic
Coryphopterus (Teleostei: Gobiidae) 111
Carole C. Baldwin, National Museum of Natural History,
Smithsonian Institution
Lee A. Weigt, National Museum of Natural History,
Smithsonian Institution
David G. Smith, National Museum of Natural History,
Smithsonian Institution
Julie H. Mounts, National Museum of Natural History,
Smithsonian Institution

Recent Insights into Cnidarian Phylogeny 139
Allen G. Collins, National Marine Fisheries Service Systematics
Laboratory, National Museum of Natural History,
Smithsonian Institution

Biodiversity and Abundance of Sponges in Caribbean Mangrove:
Indicators of Environmental Quality 151
Maria Cristina Diaz, Museo Marino de Margarita, Venezuela
Klaus Riitzler, National Museum of Natural History,
Smithsonian Institution

Internal Transcribed Spacer 2 (ITS2) Variation in the Gorgonian Coral
Pseudopterogorgia bipinnata in Belize and Panama 173
Daniel Dorado, Universidad de los Andes, Colombia
Juan A. Sanchez, Universidad de los Andes, Colombia

Obvious Invaders and Overlooked Infauna: Unexpected Constituents
of the Decapod Crustacean Fauna at Twin Cays, Belize 181
Darryl L. Felder, University of Louisiana at Lafayette
Peter C. Dworschak, Naturhistorisches Museum in Wien
Rafael Robles, University of Louisiana at Lafayette
Heather D. Bracken, University of Louisiana at Lafayette
Amanda M. Windsor, University of Louisiana at Lafayette
Jennifer M. Felder, University of Louisiana at Lafayette
Rafael Lemaitre, National Museum of Natural History,
Smithsonian Institution

Imposex in One of the World’s Busiest Shipping Zones 189
Carter Li, McGill University
Rachel Collin, Smithsonian Tropical Research Institute

Shorefishes of the Tropical Eastern Pacific Online Information System 197
D. Ross Robertson, Smithsonian Tropical Research Institute

Nephasoma pellucidum: A Model Species for Sipunculan Development? 209
Anja Schulze, Texas A@’M University at Galveston
Mary E. Rice, Smithsonian Marine Station at Fort Pierce

NUMBER 38 ¢ v

Mitochondrial Phylogeography of the Intertidal Isopod Excirolana
braziliensis on the Two Sides of the Isthmus of Panama 219
Renate Sponer, Smithsonian Tropical Research Institute
Harilaos A. Lessios, Smithsonian Tropical Research Institute

Stability and Change in the Indian River Area Bryozoan Fauna
over a Twenty-Four Year Period 229
Judith E. Winston, Virginia Museum of Natural History

BIOGEOGRAPHY, INVASIVE SPECIES, AND MARINE CONSERVATION

The Turtles’ Tale: Flagships and Instruments for Marine Research,
Education, and Conservation 241
John G. Frazier, National Zoological Park, Smithsonian Institution

Latitudinal Gradients in Recruitment and Community Dynamics
in Marine Epifaunal Communities: Implications
for Invasion Success 247
Amy L. Freestone, Smithsonian Environmental Research Center
Richard W. Osman, Smithsonian Environmental Research Center
Robert B. Whitlatch, University of Connecticut

Ex Situ Culture of Caribbean and Pacific Coral Larvae Comparing
Various Flow-Through Chambers 259
Mary Hagedorn, National Zoological Park,

Smithsonian Institution
Virginia L. Carter, National Zoological Park,
Smithsonian Institution
Lea Hollingsworth, University of Hawaii
JoAnne C. Leong, University of Hawaii
Roland Kanno, University of Hawaii
Eric H. Borneman, University of Houston
Dirk Petersen, Rotterdam Zoo
Michael Laterveer, Rotterdam Zoo
Michael Brittsan, Columbus Zoo and Aquarium
Mark Schick, John G. Shedd Aquarium

Worldwide Diving Discoveries of Living Fossil Animals
from the Depths of Anchialine and Marine Caves 269
Thomas M. Iliffe, Texas A&M University at Galveston
Louis S. Kornicker, National Museum of Natural History,
Smithsonian Institution

Decimating Mangrove Forests for Commercial Development

in the Pelican Cays, Belize: Long-Term Ecological Loss

for Short-Term Gain? 281

Ian G. Macintyre, National Museum of Natural History,
Smithsonian Institution

Marguerite A. Toscano, National Museum of Natural History,
Smithsonian Institution

Ilka C. Feller, Smithsonian Environmental Research Center

Maria A. Faust, National Museum of Natural History,
Smithsonian Institution

vi ¢ SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Using the Panama Canal to Test Predictions about Tropical
Marine Invasions 291
Gregory M. Ruiz, Smithsonian Environmental Research Center
Mark E. Torchin, Smithsonian Tropical Research Institute
Katharine Grant, Smithsonian Environmental Research Center

Ciguatera Fish Poisoning in the Caribbean 301
Patricia A. Tester, National Ocean Service, NOAA
Rebecca L. Feldman, RLF Environmental
Amy W. Nau, National Ocean Service, NOAA
Maria A. Faust, National Museum of Natural History,
Smithsonian Institution
R. Wayne Litaker, National Ocean Service, NOAA

FORCES OF ECOLOGICAL CHANGE IN MARINE SYSTEMS

History of Reef Coral Assemblages on the Rhomboid Shoals of Belize 313
Richard B. Aronson, Florida Institute of Technology
Tan G. Macintyre, National Museum of Natural History,
Smithsonian Institution
Anke M. Moesinger, Dauphin Island Sea Lab
William F. Precht, Florida Keys National Marine
Sanctuary, NOAA
Michael R. Dardeau, Dauphin Island Sea Lab

Climate and Hydrological Factors Affecting Variation in Chlorophyll
Concentration and Water Clarity in the Bahia Almirante, Panama 323
Rachel Collin, Smithsonian Tropical Research Institute
Luis D’Croz, Smithsonian Tropical Research Institute
Plinio Gondola, Smithsonian Tropical Research Institute
Juan B. Del Rosario, Smithsonian Tropical Research Institute

Nutrient and Chlorophyll Dynamics in Pacific Central
America (Panama) 335
Luis D’Croz, Smithsonian Tropical Research Institute
Aaron O’Dea, Smithsonian Tropical Research Institute

Growth and Nutrient Conservation in Rhizophora mangle in Response
to Fertilization along Latitudinal and Tidal Gradients 345
Ilka C. Feller, Smithsonian Environmental Research Center
Catherine E. Lovelock, University of Queensland
Cyril Piou, Institut Nacional de la Recherche Agronomique, France

Underwater Spectral Energy Distribution and Seagrass Depth Limits
along an Optical Water Quality Gradient 359
Charles L. Gallegos, Smithsonian Environmental
Research Center
W. Judson Kenworthy, NOAA Center for Coastal Fisheries
and Habitat Research
Patrick D. Biber, University of Southern Mississippi
Bret S. Wolfe, University of Virginia

NUMBER 38

Interannual Variation in Gelatinous Zooplankton and Their Prey
in the Rhode River, Maryland
Eileen S. Graham, Smithsonian Environmental Research Center
Danielle M. Tuzzolino, Smithsonian Environmental
Research Center
Rebecca B. Burrell, Smithsonian Environmental Research Center
Denise L. Breitburg, Smithsonian Environmental Research Center

Patterns of Water Quality and Movement in the Vicinity
of Carrie Bow Cay, Belize
Karen H. Koltes, U.S. Department of the Interior
Thomas B. Opishinski, Interactive Oceanographics

Global Change and Marsh Elevation Dynamics: Experimenting
Where Land Meets Sea and Biology Meets Geology
J. Adam Langley, Smithsonian Environmental Research Center
Marc V. Sigrist, Smithsonian Environmental Research Center
James Duls, Smithsonian Environmental Research Center
Donald R. Cahoon, U.S. Geological Survey
James C. Lynch, U.S. Geological Survey
J. Patrick Megonigal, Smithsonian Environmental Research Center

Herbivory, Nutrients, Stochastic Events, and Relative Dominances

of Benthic Indicator Groups on Coral Reefs: A Review

and Recommendations

Mark M. Littler, National Museum of Natural History,
Smithsonian Institution

Diane S. Littler, National Museum of Natural History,
Smithsonian Institution

Barrett L. Brooks, National Museum of Natural History,
Smithsonian Institution

Impacts of Human Disturbance on Soil Erosion Potential and
Habitat Stability of Mangrove-Dominated Islands in the
Pelican Cays and Twin Cays Ranges, Belize
Karen L. McKee, U.S. Geological Survey
William C. Vervaeke, U.S. Geological Survey

An Overview of Symbiont-Bleaching in the Epiphytic Foraminiferan
Sorites dominicensis
Susan L. Richardson, Smithsonian Marine Station at Fort Pierce

New Perspectives on Ecological Mechanisms Affecting Coral
Recruitment on Reefs
Raphael Ritson-Williams, Smithsonian Marine Station
at Fort Pierce
Suzanne N. Arnold, University of Maine
Nicole D. Fogarty, Florida State University
Robert S. Steneck, University of Maine
Mark J. A. Vermeij, CARMABI
Valerie J. Paul, Smithsonian Marine Station at Fort Pierce

vii

369

379

391

401

415

429

437

viii

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Do Indian River Lagoon Wetland Impoundments (Eastern Florida)
Negatively Impact Fiddler Crab (Genus Uca) Populations?
Bjorn G. Tunberg, Smithsonian Marine Station at Fort Pierce

Dynamic Hydrology of a Mangrove Island: Twin Cays, Belize
Daniel W. Urish, University of Rhode Island
Raymond M. Wright, University of Rhode Island
Ilka C. Feller, Smithsonian Environmental Research Center
Wilfrid Rodriguez, University of Rhode Island

Ecological Characteristics of Batis maritima in Florida and Belize
Dennis E Whigham, Smithsonian Environmental Research Center
Michael C. Whigham, University of New Hampshire
Ilka C. Feller, Smithsonian Environmental Research Center
Wilfrid Rodriguez, University of Rhode Island
Ryan S. King, Baylor University

Sponge Community Dynamics on Caribbean Mangrove Roots:
Significance of Species Idiosyncrasies
Janie L. Wulff, Florida State University

INDEX

459

473

491

501

515

Foreword

early two-thirds of Earth’s surface is covered by the ocean, a global

system essential to all life. Impacts on one part of the ocean can have

worldwide effects. The ocean moderates our climate, provides valu-

able resources, and produces at least half the oxygen we breathe: it
makes our planet livable. We know little, however, about the physical, chemical,
geological, and biological aspects of this crucial life support system.

The Smithsonian Institution, in efforts to increase knowledge about the
ocean, has established a network of marine laboratories that monitors coastal
habitats along a latitudinal gradient from the Chesapeake Bay through the
Indian River Lagoon to the Mesoamerican barrier reef and on both sides of
the Isthmus of Panama. The maintenance of long-term research projects and
environmental monitoring is crucial to understanding changes that exceed in
time the professional career of any given scientist. The information gained from
such studies at stable sites enables scientists to differentiate between long-term
changes and local or short-term environmental variations. Results contribute
to our knowledge of systematics and ecology, physiology, behavioral sciences,
geology, and paleoecology.

Our marine science universe comprises Smithsonian staff scientists and ex-
ternal collaborators and encourages the next generation of scientists, graduate
students, and fellows. This symposium presents a Smithsonian-wide sample of
marine science results.

Ira Rubinoff
Smithsonian Institution
Acting Under Secretary for Science, 2007-2008

a “ 7 1 n os oy
e
i i
rs + | : — 4 : ie ; — = ae as 7. & -_ - es :
| | en inno? ee
i : 7s Re oo #, - a ne
= ; i : : ; . 26% Lif sae Ooms nile

+> P ab

+
¢
- >?
— + 5
;
= " ca 3
f
P .
}
; ®
$
= i a > 4 —
a
fj — fi
ae ~
7
+ = 7 _
a
r 4 or :
i =

Executive Summary

he results of the Smithsonian Marine Science Symposium, convened by
the Marine Science Network on 15-16 November 2007 in Washington,
D.C., are reported in 39 papers in this volume. These proceedings cover
a wide range of marine research studies that demonstrate the breadth
and diversity of science initiatives supported by the Smithsonian Marine Science
Network. The first section treats an overview of the Smithsonian Marine Science
Network established in 1998, and a brief background and history of multidisci-
plinary investigations spanning more than 37 years for each of the four marine
facilities that constitute the Network: the Smithsonian Environmental Research
Center at the Chesapeake Bay, Maryland; the National Museum of Natural His-
tory’s Smithsonian Marine Station at Fort Pierce, Florida; the Caribbean Coral
Reef Ecosystems Program, with its Carrie Bow Marine Field Station in Belize; and
the Smithsonian Tropical Research Institute in Panama. Subsequent papers in this
volume represent findings by Smithsonian scholars and their collaborators on over-
arching topics of marine biodiversity, evolution, and speciation; biogeography, in-
vasive species, and marine conservation; and forces of ecological change in marine
systems. The volume includes contributions on historical and geological aspects of
coral reef and mangrove development; on biodiversity, developmental biology, and
evolution (including molecular genetics) of sponges, cnidarians, sipunculan worms,
crustaceans, and fishes; on ecology and population dynamics of algae, sponges,
bryozoans, zooplankton, and miscellaneous invasive species; on environmental pa-
rameters, including pollutants, oceanographic factors, and hydrological regimes,
and their effects on primary and secondary productivity, bleaching of symbiotic fo-
raminiferans, benthic community structure, herbivory, development of toxic algal
blooms, and land-sea connectivity in coastal habitats, that is, temperate bays and
lagoons and tropical reefs and mangroves; and on conservation and education ini-
tiatives encompassing a range of organisms from sponges and corals to sea turtles,
as well as communities such as tidal marshes, mangrove swamps, and coral reefs.
As we prepare to face the challenges of rapidly accelerating biodiversity loss and
global environmental stresses, particularly in highly vulnerable tropical shallow-
water ecosystems such as reefs and mangroves, the focus of our scientific expertise
on the member laboratories of the Marine Science Network has, during decades
of documentation, established ecological standards that will help us monitor and
evaluate future changes or trends and contribute to forthcoming education and
conservation initiatives.
Michael A. Lang
Smithsonian Institution
Office of the Under Secretary for Science
April 2009

: %
J - fy c ;
1
ry y i. P,
9
é %
= f a ~S
F
: ‘ a > a > Pe > +
: : a ore = os ey b= ° r
| 7 BBR Amo ; ‘Svitnioay.
= ih : * J d
A a 3: L > -
i
&, ‘
a “= ae Pal
by ;
3 44
a bats
; i os
> | + , tT
af =
\ a =
i r * <7
3 es S
% F 2
- 7
=~ ~~ i .
Ay ce My e
fi .
a 5}
j
- 3;
- ony
Ef i
‘ = 7 cy
Fa = \ i
2
yi, oa t
}
: j #7 a h ‘ i f i
es, J
Widy Its nje ‘ ‘ £/ ae |
~ a d
r = , - t
eat 4 Bt
x, f >
af j
2 “gyre A. cin

ie wcuwditent, eA SINS.
ol Wy, eneaney aR aan pearly AW 2
i ee) . I ane

iP

Introduction to the Smithsonian
Marine Science Network

Michael A. Lang

Michael A. Lang, Smithsonian Institution, Of-
fice of the Under Secretary for Science, P.O. Box
37012, MRC-009, Washington, D.C. 20013-
7012, USA (langm@si.edu). Manuscript received
10 April 2009; accepted 20 April 2009.

ABSTRACT. The “Smithsonian Marine Science Symposium” contained more than 70
oral and poster presentations by Smithsonian scholars and collaborators and represented
the first major dissemination of marine research results since the establishment of the
Marine Science Network (MSN) in 1998. The MSN operates a unique array of labo-
ratories and research vessels that spans the latitudinal gradient of the western Atlantic
(Chesapeake Bay, Indian River Lagoon, Mesoamerican Barrier Reef, and Panamanian
Coast) and crosses the isthmus of Panama. The Network is dedicated to understanding
the rich biodiversity and complex ecosystem dynamics that sustain coastal processes and
productivity. We study evolutionary, ecological, and environmental change in the ocean’s
coastal zones, increasing scientific knowledge of these environments and improving soci-
ety’s appreciation of the ocean’s effect on our lives. Coastal environments are of immense
economic and environmental importance and comprise 95% of the ocean’s fisheries. Our
coasts are the most densely populated and fastest growing communities in the USA. The
MSN ensures integrated support of “Discovering and Understanding Life’s Diversity,” a
core Smithsonian scientific mission. The MSN’s goals are to ensure that the whole of the
integrated Network is larger than the sum of its parts, leading to enhanced productivity
through collaborative and comparative research, marine infrastructure development and
support, professional training and outreach, and effective allocation of resources.

INTRODUCTION

The “Smithsonian Marine Science Symposium” was held 15-16 Novem-
ber 2007 to celebrate individual and long-term pan-institutional marine re-
search, with a particular focus on highlights of the first ten years since the
establishment of the Marine Science Network (MSN) in 1998. The symposium
was convened by the Office of the Under Secretary for Science and represented
the first gathering, of this magnitude, of Smithsonian marine scientists. The
symposium presented marine research findings by Smithsonian scholars and
their collaborators with emphasis on marine biodiversity, evolution, and spe-
ciation; biogeography, invasive species, and marine conservation, including
life histories and microbial and behavioral ecology; and forces of ecological
change in marine systems. The symposium carried on a tradition of Smithson-
ian marine science that began nearly 150 years ago and resulted in some of the
world’s foremost collections of marine specimens. More than 70 presentations

2 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

and posters discussed results of marine research from the
Chesapeake Bay, Indian River Lagoon and Florida Keys,
the Mesoamerican Barrier Reef in Belize, the Atlantic
and Pacific Coasts of the Isthmus of Panama, and other
international research sites. Thirty-nine papers from this
symposium are presented in this 38th volume of Smith-
sonian Contributions to the Marine Sciences, and ad-
ditional marine education posters reside on http://www
.si.edu/marinescience. Smithsonian speakers included ma-
rine research leaders, collaborators, and fellows from the
Smithsonian Environmental Research Center, National
Zoological Park, National Museum of Natural History,
Smithsonian Marine Station at Fort Pierce, Caribbean
Coral Reef Ecosystems Program, Smithsonian Tropical
Research Institute, and the Office of the Under Secretary
for Science.

The Smithsonian Institution operates a unique net-
work of coastal laboratories and long-term research sites
on the east coast of North and Central America that ex-
tends along the western Atlantic Ocean and bridges the
Panamanian isthmus from the Caribbean Sea to the Pacific
Ocean (Figure 1). Scientific diving supports a significant
amount of Smithsonian marine research throughout the
Network and internationally (Lang and Baldwin, 1996;
Lang, 2007).

The Marine Science Network concept was developed
in 1998 from the bottom up and has achieved the follow-
ing important milestones:

1998: Formalization of a pan-institutional Smithsonian
Marine Science Network initiated at two-day in-
augural workshop at Smithsonian Environmental
Research Center, with more than 50 Smithsonian
Institution participants.

1999: Dedication of new Carrie Bow Cay Marine Field
Station.

1999: Dedication of new Smithsonian Marine Station at
Fort Pierce.

2000: MSN concept and infrastructure allocations ap-
proved by the Under Secretary for Science.

2001: Launch of the MSN website
www.si.edu/marinescience.

2001: Annual MSN Calls for Proposals for infrastructure,
marine research awards, and postdoctoral fellow-
ships.

2003: Dedication of Bocas del Toro Marine Laboratory.

2006: Science Executive Committee review of Smithson-
ian marine science, including MSN.

2007: Formulation of Big Questions in Marine Science:

1. What are the major spatial and temporal
patterns in distribution of biodiversity?

2. How does biodiversity, and the loss of biodi-
versity, affect the functioning of ecosystems?
3. How are humans changing the magnitude
and distribution of biodiversity and what
are the patterns and consequences?
2007: Smithsonian Marine Science Symposium.

The MSN is administered as a pan-institutional pro-
gram through the Office of the Under Secretary for Sci-
ence. It is governed by a seven-member Steering Committee
composed of Michael Lang (Office of the Under Secreta-
ray for Science), Anson Hines (Smithsonian Environmen-
tal Research Center), Eldredge Bermingham (Smithsonian
Tropical Research Institute), Klaus Ruetzler (National
Museum of Natural History), Robert Fleischer (National
Zoological Park), Valerie Paul (Smithsonian Marine Sta-
tion at Fort Pierce), and Phillip Taylor (National Science
Foundation). Additional Smithsonian scientists partici-
pate by invitation in MSN research proposals and post-
doctoral fellowship review panels, MSN symposia and
workshop committees, and special projects. Support for
MSN infrastructure, research, and postdoctoral fellow-
ships is provided by the Office of the Under Secretary
for Science’s Johnson and Hunterdon Oceanographic Re-
search Endowments.

There are four main unifying disciplinary themes to
Smithsonian marine research: systematics, evolutionary
biology, ecology, and geology. Biogeography is a key re-
search element, linking systematics, ecology, and evolu-
tionary biology. Mechanisms of biogeographic isolation
are central elements in evolutionary theory, population
dynamics, conservation biology, and patterns of biodiver-
sity. Biogeographic patterns are crucial data in the deter-
mination of introduced and native species. Site-specific,
long-term measurements of environmental variables allow
for analysis of change over multiple time scales, which
is necessary to detect patterns in typically complex eco-
logical data. The Smithsonian Marine Science Network
is uniquely positioned to monitor long-term change at its
component sites. It has an extensive array of programs
that address many of the most pressing environmental is-
sues in marine ecosystems, including biological invasions,
eutrophication, harmful species and parasites, plankton
blooms and red tides, linkages among coastal ecosystems,
global warming including sea-level rise, El Nifio/La Nifia
effects, UV radiation impacts, habitat destruction, fish-
eries impacts, ecology of key habitats (estuaries, coral
reefs, mangroves, seagrasses, wetlands), and biodiversity
inventories.

The Smithsonian’s marine education programs con-
sist of public outreach and professional training. A series

NUMBER 38

‘

J National
Museum of
Natural History

e Mich,

2 Lak

x

i

Smithsonian
Environmental

des 2 Nox Hatteras

g 20 Sound
“ZJ, Cape Lookout

SeANiimington

nytt =
® Galveston Bay x0\ @ 4
2 Galveston £y , ;
ry ce a
arene , , am

me

Ei Smithsonian
= preoettswn a 4, Marine Station
=" Ze at Fort Pierce

eSbleOeS
at One |.
Tort eee : '
Pm aru us?

%

Morant Poj
ingstant Poi

Carrie Bow Cay

“Corn Ie ‘Albuquerque Ca

ee Smithsonian
oe Tropical
nag 7, Research

# Institute

Buenavexgy

Buenavert™ o Bet

*Malpelo I.

FIGURE 1. Map showing locations of the Smithsonian Marine Science Network members.

Research Center

3

4 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

of these activities are aimed at promoting awareness and
conservation of marine environments and at communi-
cating the Smithsonian’s research findings to the general
public. By integrating research with education, the Smith-
sonian produces tomorrow’s discoverers while pursuing
today’s discoveries. The public is engaged with interac-
tive exhibits and scientists (e.g., the National Museum of
Natural History’s Sant Ocean Hall), symposia, popular
books, lectures, and films about the marine environment.
The Smithsonian Marine Science Network contributes
to the public interest by disseminating novel environ-
mental information around the globe. Its research helps
build a solid foundation for informed decisions about
environmental policy, natural resource management, and
conservation.

Other recent coordinated Smithsonian marine sci-
ence efforts of note are marine-terrestrial flora and fauna
of Cayos Cochinos Archipelago, Honduras (Guzman,
1998), natural history of the Pelican Cays (Macintyre and
Ritzler, 2000), the Twin Cays mangrove ecosystem, Be-
lize (Macintyre et al., 2004), and investigations of the ma-
rine fauna and environments of Bocas del Toro, Panama
(Collin, 2005). Smithsonian taxonomic field guides and
keys of algae, invertebrates, and fishes (Littler and Littler,
2000; Collin et al., 2005; Robertson, 2009) are valuable
tools for biologists, divers, and fishermen alike.

MARINE SCIENCE Network MISSION AND VISION

The MSN mission is dedicated to understanding the
rich biodiversity and complex ecosystem dynamics that
sustain coastal processes and productivity, and its vision
is “to increase scientific knowledge of marine coastal en-
vironments and to improve society’s appreciation of the
ocean’s effect on our lives.”

MARINE SCIENCE Network GOALS

The MSN provides integrated support of “Discover-
ing and Understanding Life’s Diversity,” a core scientific
mission of the 2005-2010 Smithsonian Science Strate-
gic Plan. The MSN ensures that the whole of the inte-
grated Network is greater than the sum of its parts, lead-
ing to enhanced productivity through collaborative and
comparative research facilitated by increased inter-unit
coordination, marine infrastructure development and
support, professional training and outreach, effective al-
location of research funding, and transparent manage-
ment, participation, and support for Smithsonian marine

scientists through availability of shared resources and
facility access.

SMITHSONIAN ENVIRONMENTAL
RESEARCH CENTER (CHESAPEAKE BAY)

The Smithsonian Environmental Research Center
(SERC) advances stewardship of the biosphere through
interdisciplinary research and education. With a resident
staff of more than 100 scientists, technicians, fellows, and
students, SERC has experienced significant growth in the
last few years. The SERC laboratories, educational facili-
ties, and primary field sites are located 25 miles east of
Washington, D.C., on the western shore of Chesapeake
Bay. Its campus includes a growing complex of offices,
laboratories, maintenance shops, a library, housing, and
facilities for public programs. A dock, fleet of research
vessels, dive locker, wet laboratory, aquarium room, and
large fish-weir are used in support of estuarine research.

The greatest resource at SERC is its main research site
on the Rhode River subestuary, which includes more than
1,072 ha of land and 26 km of undeveloped shoreline of
the Chesapeake Bay. Since 1965, SERC’s long-term stud-
ies have focused on the interactions among ecosystems in
complex landscapes, tidal marshes, and estuaries. With
the Rhode River site as its hub, SERC research radiates to
sites around the world to address effects of global change,
landscape ecology, coastal ecosystems, and population
and community ecology. Much of SERC’s comparative
research across latitudes extends to the other sites of the
MSN and includes studies of mangrove biocomplexity, in-
vasive and native species biodiversity, estuarine food webs,
land use impacts linked to water quality, carbon process-
ing and global change, nutrient loading and low dissolved
oxygen, ecosystem management of fisheries and crucial
habitats, and life history patterns and evolution. Research
at SERC focuses on five grand environmental challenges
(Hines, 2009): (1) impacts of atmospheric change on cli-
mate, sea level, ultraviolet radiation, pollutant deposition,
and carbon balance; (2) impacts of watershed nutrient dis-
charges causing harmful algal blooms, depletion of oxy-
gen, and destruction of submerged vegetation; (3) food
web disruption by pollution and overfishing; (4) invasive
species; and (5) landscape disturbance by agriculture and
development.

Goals of SERC in marine education include profes-
sional training (interns, graduate students, postdoctoral
fellows, and visiting scientists), teacher training, site visits
and public programs, and distance learning.

NATIONAL ZOOLOGICAL PARK
(WASHINGTON, D.C.)

The Smithsonian’s National Zoological Park (NZP)
was founded in 1889. Its mission is to provide leadership
in animal care, science, education, and sustainability. Ap-
proximately 2,000 individuals of 400 different species
constitute its animal collection. The NZP consists of a 163
acre urban park located in Rock Creek Park in northwest
Washington, D.C., and the 3,200 acre Conservation and
Research Center in Front Royal, Virginia, emphasizing re-
productive physiology, analysis of habitat and species rela-
tionships, and the training of conservation scientists.

The National Zoological Park conducts international
marine research on sea turtles and sea birds, ecology of
bottlenose dolphins, Weddell seal lactation, life history
and reproductive strategies of gray and harbor seals, nu-
tritional ecology of sea otters, and cryopreservation of en-
dangered coral species. Marine exhibits include the Seal
and Sea Lion Pool and the Invertebrate Exhibit, which
opened in 1987, where marine invertebrates comprise
75% of its live collections on display. The NZP’s tools
to inspire, train, and empower successive generations to
care for the world’s biological diversity are its exhibits,
science, outreach, and education programs. Ultimately,
efforts must be oriented toward protecting wildlife and
other forms of biological diversity so that we, and future
societies, continue to enjoy the incalculable benefits of our
natural world.

NATIONAL MUSEUM OF NATURAL HISTORY
(WASHINGTON, D.C.)

The National Museum of Natural History (NMNH)
has a distinguished history of more than 150 years of sam-
pling and collections-based research. Major collections
represent algae and dinoflagellates, foraminifera, sponges,
cnidarians, ctenophores, worms, crustaceans, mollusks,
bryozoans, echinoderms, tunicates, fishes, marine rep-
tiles, birds, and mammals), now numbering more than
33,000,000 specimens of plants and animals. Of approxi-
mately 2,415 families of marine invertebrates, nearly 67%
are represented in the NMNH invertebrate collection,
which is not limited solely to the diversity-rich tropics.
The NMNH provides professional collection management
services to the National Science Foundation United States
Antarctic Program (USAP) and the international scientific
community. A primary focus of this project is improving
access to the collections through its cataloging (inventory)

NUMBER 38 e¢ 5

program (more than 900,000 USAP specimens) and loan
program. More than 170,000 USAP specimens in 138
separate transactions were either lent or returned from
loan between 1995 and the end of 2004, supporting the
research efforts of scientists in 22 countries. Several hun-
dred lots of archive samples from the Palmer Long-Term
Ecological Research Program were also accessioned (Le-
maitre et al., 2009).

The focus of marine science at NMNH addresses the
diversity of marine life, where species occur, how they are
related to each other, how marine diversity developed and
how it is maintained, what are the human impacts on ma-
rine life, and how marine life-forms are used by people.

The Museum administers the Laboratories of Analyti-
cal Biology for state-of-the art molecular work and two
marine field stations (Carrie Bow Cay, Belize, and Smith-
sonian Marine Station at Fort Pierce, Florida), member
facilities of the MSN. Since 1966 the Museum has funded
the Atoll Research Bulletin, which publishes research re-
ports on the geology and ecology of islands and their adja-
cent coral reef and mangrove communities in tropical sites
around the world.

The NMNH Ocean Initiative comprises the Sant
Ocean Hall, the Ocean Web Portal, the Sant Chair in
Marine Science, and interdisciplinary marine research at
NMNH. Virtual access to the Museum’s key marine col-
lections is being created. The Initiative aims to train future
generations of marine scientists and educate the public
about, and raise awareness of, the importance of the ocean
as a global system.

SMITHSONIAN MARINE STATION
AT FORT PIERCE (FLORIDA)

The Smithsonian Institution has had a presence in Fort
Pierce, Florida, since 1969 and was known then as the
Fort Pierce Bureau. From 1969 to 1981, the Fort Pierce
Bureau carried out studies including underwater oceanog-
raphy with research submersibles, a survey of the Indian
River Lagoon, coral reef research, and research on life
histories of marine invertebrates, partly in collaboration
with the newly formed Harbor Branch Foundation (now
the Harbor Branch Oceanographic Institution at Florida
Atlantic University). In 1981, the Fort Pierce Bureau was
dissolved, and in its place the Smithsonian Marine Station
at Link Port was formally recognized as an organizational
unit under the auspices of the National Museum of Natu-
ral History. The Station took over the barge, acquired origi-
nally by the Smithsonian in 1973 from federal surplus, that

6 ¢ SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

was docked at the Harbor Branch campus. In 1996, the
Smithsonian purchased, from the MacArthur Foundation,
8 acres of property 7 miles south near the Fort Pierce Inlet
with easement access to the Indian River Lagoon. St. Lucie
County enacted a 25 year lease of a county dock and adja-
cent land at a site on the inlet across from the Station, whose
main building was completed and dedicated in 1999.

The Smithsonian Marine Station at Fort Pierce
(SMSFP) is a marine science research center located on the
Indian River Lagoon along 156 miles of Florida’s central
Atlantic coast. The Indian River Lagoon is a long, nar-
row, and shallow estuary adjacent to the Atlantic Ocean,
separated by a strip of barrier islands. Biologists at SMSFP
have the advantage of working just 20 miles from the Flor-
ida current, a stream of warm water from the Caribbean
that moves northward past Florida’s coastline as part of
the larger, complex system of currents known as the Gulf
Stream. The current carries with it many tropical marine
organisms, allowing researchers to work at the interface of
the Northern Hemisphere’s tropical and temperate regions.
Situated in a biogeographic transitional zone between the
temperate and subtropical provinces, the SMSFP facility
provides access to an extraordinary diversity of marine
and estuarine species and to a variety of habitats, which
include mangroves, salt marshes and sandy beaches, rocky
intertidal substrates, seagrass beds, mud and sand flats,
coral reefs, worm reefs, Coquina hard bottoms, deep coral
rubble zones, shallow- to deep-water sandy plains, and the
blue waters of the Gulf Stream.

The Marine Station supports and conducts scholarly
research in the marine sciences, emphasizing studies of
biodiversity, life histories, and ecology of marine organ-
isms (Paul et al., 2009). The results of this research enable
policy makers to make informed environmental decisions
in guiding conservation and sustainable management of
marine resources, as well as providing the basis for inno-
vative applications in medicine, aquaculture, and the ef-
fective balance between development and conservation.
For Smithsonian scientists, the SMSFP provides an impor-
tant link with other MSN facilities in the tropics at the
Smithsonian Tropical Research Institute (STRI) in Panama
and Carrie Bow Cay in Belize and in the temperate region,
the Smithsonian Environmental Research Center on the
Chesapeake Bay.

The facilities at the Smithsonian Marine Station
at Fort Pierce include an 8,000 square foot facility that
houses a histology laboratory, an electron microscopy
lab, a confocal microscope, a combination electrophore-
sis/DNA/chemistry laboratory, a photographic darkroom,
flow-through seawater tables and aquaria, an industrial

shop, and offices and laboratories for visiting scientists
and fellowship recipients. The 39-foot R/V Sunburst and
two smaller vessels are used for scientific diving, dredg-
ing, and trawling in the Indian River Lagoon, Continental
Shelf, and Gulf Stream.

The Marine Station’s educational efforts include post-
doctoral fellows and interns, public events and lectures,
school programs and public tours, a web site, the Indian
River Lagoon Species Inventory, and the Marine Ecosystems
Exhibit, which was established in 2001 with the following
ecosystems on display: coral reef, seagrass, mangrove, hard-
bottom and nearshore habitats, and Oculina reef).

CARIBBEAN CORAL REEF ECOSYSTEMS
PROGRAM (CARRIE BOW CAY, BELIZE)

Coral reefs are unique biogeological structures that
thrive in clear, nutrient-poor (oligotrophic) tropical oceans
and support a rich and diverse biological community. Reef
systems are driven by the symbiosis between scleractinian
corals and microscopic dinoflagellate algae (zooxanthellae)
as their chief energy source. The largest, best developed, least
polluted, and least commercially exploited coral reef in the
Atlantic region is the Mesoamerican Barrier Reef in Belize. It
is a complex of reefs, atolls, islands, oceanic mangroves, and
seagrass meadows that extends over 160 km. For its unique
characteristics and unperturbed condition, the Belize barrier
reef has been declared a World Heritage Site.

In the early 1970s, Riitzler et al. (2009) discovered
the formidable qualities of the Belize (then British Hon-
duras) barrier reef. After careful comparison with other
locations in the western Caribbean, it was chosen as the
site of an interdisciplinary long-term study of systematics,
ecology, behavior, and evolution of reef organisms and the
dynamics and historical development of reef communi-
ties (Rutzler and Macintyre, 1982). Carrie Bow Cay, only
three hours by plane and boat from Miami, was found to
be the ideal logistical base because of its location on top of
the barrier reef, only meters away from a variety of habi-
tat types (reef flat, spur and groove, deep fore-reef slope,
patch reefs, seagrass meadows, and mangroves), and its
undisputed ownership by a Belizean family able to cater to
all Smithsonian needs for lodging, food, local transporta-
tion, and contacts with government.

In 1985, as part of the U.S. Congress Caribbean Ba-
sin Initiative, the National Museum of Natural History
received an increase to its budget base to continue and
intensify study of Caribbean coral reef ecosystems. These
funds allowed for the expansion of research facilities on

Carrie Bow Cay and the update of CCRE equipment. In
the years since, CCRE has accomplished the following:
amassed thousands of specimens of marine plants, inver-
tebrates, and fishes, which are organized in an enormous
database; assisted the government of Belize in shaping and
justifying its coastal conservation policy; participated con-
tinuously in the Caribbean-wide reef monitoring network
(CARICOMP); established the first meteorological ocean-
ographic monitoring station in coastal Belize; and, above
all, published well over 850 scientific papers in reviewed
journals, as well as several books, doctoral dissertations,
popular articles, and photo and video documentaries. Be-
tween 60 and 80 scientists use Carrie Bow Cay each year
as a part of ongoing CCRE research.

The Carrie Bow Cay Laboratory serves primarily in
support of SI marine scientists’ research projects and their
external collaborators. Seasonal hurricanes during the
past 35 years could not destroy Carrie Bow Cay facilities
to the extent that a devastating fire did in December 1997.
Improved facilities now include dry and wet labs, housing,
generator, compressor, small boats and scuba cylinders,
and essential facilities such as solar power, a running-
seawater system, and weather station.

CCRE’s educational and outreach programs include
its Belize teachers’ mangrove workshops, publications,
symposia, advisory consults with Belizean Ministries, and
fellows and interns.

SMITHSONIAN TROPICAL RESEARCH
INSTITUTE (REPUBLIC OF PANAMA)

The Smithsonian Tropical Research Institute’s (STRI)
marine research program in the Republic of Panama dates
to 1964 when small laboratories were established on the
Pacific and Caribbean coasts within the former Canal
Zone. Today, STRI operates marine stations at Bocas del
Toro and Galeta Point in the Caribbean and the Naos ma-
rine laboratory complex in the Pacific. Until 2008, the R/V
Urraca, a 96 foot nearshore coastal oceanographic ves-
sel, was outfitted with remotely operated vehicle, scientific
diving, and dredging capabilities, and was operated under
University National Oceanographic Laboratory System
(UNOLS) research fleet standards.

At the Panama Canal, the Isthmus of Panama narrows
to less than 100 km, separating oceans that are very dif-
ferent tropical marine ecosystems. The Caribbean is a rela-
tively stable ocean, with small fluctuations in temperature
and relatively low tidal variation. Its transparent, nutrient-
poor waters are ideal for the growth of reefs, and it ranks

NUMBER 38 e@ 7

just behind the Indian Ocean and the Indo-West Pacific in
terms of numbers of marine species. The tropical eastern
Pacific, in contrast, exhibits much greater fluctuations in
tides and temperature, with seasonal upwelling locally and
longer-term variation resulting from the El Nifio south-
ern oscillation cycle. Its more nutrient-rich waters support
commercial fisheries of major importance. The creation of
these two distinct marine realms by the rise of the Isthmus
of Panama during the past 10 million years also contrib-
uted to the formation of the modern biological and geo-
logical world. During this interval, the Gulf Stream was
established, the mammals of North America conquered a
newly connected South America, the Ice Ages began, and
modern man arose. The Isthmus played a major role in
this history, and set in motion a fascinating natural experi-
ment, as the animals and plants of the two oceans went
their separate evolutionary ways.

There are also major differences within each ocean.
In the Pacific, seasonal upwelling of nutrient-rich wa-
ters is strong in the Gulf of Panama, where trade winds
blow freely across the Isthmus, but absent in the Gulf of
Chiriqui, where the high terrain blocks these winds. The
more stable conditions in the Gulf of Chiriqui support the
best developed coral reefs in the tropical eastern Pacific.
On the Caribbean side, the San Blas Archipelago is bathed
in clear oceanic waters, whereas the reefs and mangroves
of the enormous Chiriqui Lagoon of Bocas del Toro are
enriched by runoff from the land. Thus, Panama can be
considered a nation of four ocean types, providing unique
opportunities for understanding how and why marine eco-
systems function as they do.

Understanding the history and ecology of Panama’s
diverse marine environments has been a major theme of
STRI’s research over the past four decades (Robertson et
al., 2009). Major programs include between-ocean com-
parisons of physical and biological oceanography, geologi-
cal reconstruction of events leading up to and following
the rise of the Isthmus, studies of marine biodiversity, and
analyses of the vulnerability of marine habitats to natural
and anthropogenic change. In celebration of STRI’s role in
coral reef research, the Smithsonian’s 150th anniversary,
and the International Year of the Reef, the Smithsonian
hosted the Eighth International Coral Reef Symposium
in Panama in 1996. This meeting brought 1,500 reef sci-
entists and managers to Panama from around the world
and resulted in the publication of a two-volume proceed-
ings (Lessios and Macintyre, 1997) and an international
traveling exhibit that has already brought STRI’s marine
discoveries to Miami, the District of Columbia, Honduras,
and Jamaica.

8 © SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

The Marine Environmental Sciences Program (MESP)
at STRI collects and analyzes fundamental oceanographic
information that provides critical information for studies
such as E] Nifio and coral bleaching. The Panama Paleon-
tology Project in Bocas del Toro seeks to record the history
of the divergence between the two oceans over the past
10 million years and the evolutionary response of marine
organisms to these changes. Results from this project are
the geological reconstruction of the closure of the Isth-
mus of Panama 3 million years ago and the discovery of a
major extinction event in the Caribbean about 2 million
years ago. Through a combination of molecular and pale-
ontological information, STRI’s molecular evolution pro-
gram has developed a model system for determining the
rate at which organisms diverge genetically through time
(Panama molecular clock). This achievement allows for
the phylogenetic reconstruction of marine life elsewhere
in the world.

Educational programs within STRI include a marine
fellowship program, school group visits (Culebra Nature
Center, Galeta Point Marine Lab, Bocas del Toro Lab),
public seminars, advisory consults with the Panamanian
government, and graduate courses.

CONCLUSION

The Smithsonian Institution is unique among federal
agencies, research organizations, and universities with its
investment in comprehensive, long-term marine studies
of crucial ecosystems at a latitudinal gradient of stable
sites. Thousands of marine science publications provide
results and data for synthesis and a new baseline for mod-
eling and forecasting. Continued opportunities remain for
many important marine organisms to be identified by con-
ventional and molecular techniques and described. Deep
reefs are becoming increasingly important as a focal area
to understand how the reef system in toto functions and to
quantify their physical, chemical, and biological contribu-
tions to the shallow reefs that we have studied for more
than three decades (Lang and Smith, 2006). Life histories
require further analysis to aid ecological understanding
and fisheries management. Thirty-five-year multidisci-
plinary databases allow for early detection and evaluation
of community changes, invertebrate diseases, invasive spe-
cies, and recruitment caused by environmental degrada-
tion and catastrophic events.

The MSN continues to provide support for individual
and pan-institutional collaborative research, postdoctoral
marine science fellows, marine science staff and infra-

structure support, marine outreach and education, and
workshops and symposia: for example, Bocas del Toro
taxonomy; coral reef management; mangrove ecology of
Twin Cays, Belize; marine genetics; sea turtle conserva-
tion and population management; seagrass and mangrove
ecosystems; neogastropod evolution; marine invasives of
the Gulf of Mexico; and marine invasive species across
latitudinal gradients. Outcomes of the integration of the
Smithsonian marine facilities and programs since 1998 are
the facilitated freedom of movement of scientists between
units and the increased collaborations and co-authored
publications. The MSN was highlighted as a model for
pan-institutional Smithsonian programs by the Smithson-
ian Science Commission in 2003. The most likely keys to
its success were the bottom-up development of the Net-
work concept, starting with the Institution’s staff scien-
tists, and the availability of research funding through the
Office of the Under Secretary for Science to enable marine
research and postdoctoral fellowships.

The Smithsonian Marine Science Network and the
Smithsonian Scientific Diving Program provide the facili-
ties and support for the efficient conduct of marine re-
search. The primary objective of the marine research effort
is the advancement of science. The deliverable is mainly in
the form of peer-reviewed publications for dissemination
throughout the scientific community and to the public.

The importance of the MSN is its contribution to the
knowledge of complex ecosystems including seagrasses,
mangrove islands, bays, estuaries, and coral reefs, and the
preservation of these precious resources by learning about
their rich biodiversity, function, and interconnectedness.
Only a long-term commitment will allow us to understand
the dynamics of coastal processes and organisms, obtain the
cooperation of the public, and educate a new generation.

ACKNOWLEDGMENTS

Support for the Smithsonian Marine Science Network
is provided by the Hunterdon and Johnson Oceanographic
Research Endowments in the Office of the Under Secretary
for Science. I wish to extend appreciation to the Marine
Science Symposium Committee: Biff Bermingham, Tuck
Hines, Ian Macintyre, Olav Oftedal, Valerie Paul, Klaus
Riitzler, and Phil Taylor; and to Session Co-Chairs: Carole
Baldwin, Allen Collins, Candy Feller, Ian Macintyre, Jon
Norenburg, Valerie Paul, and Mark Torchin. I thank Ira
Rubinoff for his long-term support of the Scientific Div-
ing Program and MSN. Editorial collaboration with Ian
Macintyre and Klaus Ritzler on this volume was much
appreciated, as was the assistance of Ginger Strader and

her staff at the Smithsonian Institution Scholarly Press. I
also thank the many anonymous peer reviewers. The au-
thors’ contributions to this symposium proceedings vol-
ume highlighting the breadth and scope of Smithsonian
marine research were invaluable.

LITERATURE CITED

Collin, R., ed. 2005. Marine Fauna and Environments of Bocas del Toro,
Panama. Caribbean Journal of Science, 41(3):367-707.

Collin, R., M. C. Diaz, J. L. Norenburg, R. M. Rocha, J. A. Sanchez,
A. Schulze, M. L. Schwartz, and A. Valdes. 2005. Photographic
Identification Guide to Some Common Marine Invertebrates of
Bocas Del Toro, Panama. Caribbean Journal of Science, 41(3):
638-707.

Guzman, H. M., ed. 1998. Marine—Terrestrial Flora and Fauna of Cayos
Cochinos Archipelago, Honduras. Revista Biologia Tropical, 46(4)
(Suppl.):1-200.

Hines, A. H. 2009. “Land-Sea Interactions and Human Impacts in
the Coastal Zone.” In Proceedings of the Smithsonian Marine
Science Symposium, ed. M. A. Lang, I. G. Macintyre, and K.
Riutzler, pp. 11-23. Smithsonian Contributions to the Marine
Sciences, No. 38. Washington, D.C.: Smithsonian Institution
Scholarly Press.

Lang, M. A. 2007. Scientific Diving in the United States: The Value of
Scuba as Research Methodology. Journal of the Society for Under-
water Technology, 27(3):95-107.

Lang, M. A., and C. C. Baldwin, eds. 1996. Methods and Techniques
of Underwater Research. Proceedings of the American Academy
of Underwater Sciences Scientific Diving Symposium. Washington,
D.C.: Smithsonian Institution. ;

Lang, M. A., and N. E. Smith, eds. 2006. Proceedings of the “Advanced
Scientific Diving Workshop,” February 23-24, 2006. Washington,
D.C.: Smithsonian Institution.

Lemaitre, R., M. G. Harasewych, and J. Hammock, eds. 2009. ANTIZ v
1.0: A Database of Antarctic and Subantarctic Marine Invertebrates.
National Museum of Natural History, Smithsonian Institution.
World Wide Web electronic publication. URL http://invertebrates
.si.edu/ANTIZ.

NUMBER 38 e¢ 9

Lessios, H. A., and I. G. Macintyre, eds. 1997. Proceedings of the 8th
International Coral Reef Symposium, Volumes 1 and 2. Smithson-
ian Tropical Research Institute, Republic of Panama.

Littler, D. S., and M. M. Littler. 2000. Caribbean Reef Plants: An Identifi-
cation Guide to the Reef Plants of the Caribbean, Bahamas, Florida
and Gulf of Mexico. Offshore Graphics, Inc., Washington, D.C.

Macintyre, I. G., and K. Riutzler, eds. 2000. Natural History of the Peli-
can Cays, Belize. Atoll Research Bulletin, Nos. 466-480:1-333.
Washington, D.C.: Smithsonian Institution.

Macintyre, I. G.,-K. Riitzler, and I. C. Feller, eds. 2004. The Twin Cays
Mangrove Ecosystem, Belize: Biodiversity, Geological History, and
Two Decades of Change. Atoll Research Bulletin, Nos. 509-530.
Washington, D.C.: Smithsonian Institution.

Paul, V. J., J. Piraino, and L. Diederick. 2009. “Smithsonian Marine Sta-
tion at Fort Pierce: Thirty-Seven Years of Research on the Marine
Biodiversity of Florida.” In Proceedings of the Smithsonian Ma-
rine Science Symposium, ed. M. A. Lang, I. G. Macintyre, and
K. Riutzler, pp. 25-41. Smithsonian Contributions to the Marine
Sciences, No. 38. Washington, D.C.: Smithsonian Institution Schol-
arly Press.

Robertson, D. R. 2009. “Shorefishes of the Tropical Eastern Pacific
Online Information System.” In Proceedings of the Smithsonian
Marine Science Symposium, ed. M. A. Lang, I. G. Macintyre, and
K. Ritzler, pp. 197-208. Smithsonian Contributions to the Marine
Sciences, No. 38. Washington, D.C.: Smithsonian Institution Schol-
arly Press.

Robertson, D. R., J. H. Christy, R. Collin, R. G. Cooke, L. D’Croz,
K. W. Kaufmann, S$. Heckadon Moreno, J. L. Mate, A. O’Dea, and
M. E. Torchin. 2009. “The Smithsonian Tropical Research Insti-
tute: Marine Research, Education, and Conservation in Panama.”
In Proceedings of the Smithsonian Marine Science Symposium, ed.
M. A. Lang, I. G. Macintyre, and K. Ritzler, pp. 73-93. Smithso-
nian Contributions to the Marine Sciences, No. 38. Washington,
D.C.: Smithsonian Institution Scholarly Press.

Ritzler, K. 2009. “Caribbean Coral Reef Ecosystems: Thirty-five Years
of Smithsonian Marine Science in Belize.” In Proceedings of the
Smithsonian Marine Science Symposium, ed. M. A. Lang, I. G.
Macintyre, and K. Ritzler, pp. 43-71. Smithsonian Contributions
to the Marine Sciences, No. 38. Washington, D.C.: Smithsonian
Institution Scholarly Press.

Ritzler, K., and I. G. Macintyre, eds. 1982. The Atlantic Barrier Reef
Ecosystem at Carrie Bow Cay, Belize. 1: Structure and Communi-
ties. Smithsonian Contributions to the Marine Sciences, 12:1-539.

Land-Sea Interactions and Human Impacts
in the Coastal Zone

Anson H. Hines

Anson H. Hines, Smithsonian Environmental Re-
search Center, P.O. Box 28, 647 Contees Wharf
Road, Edgewater, Maryland 21037-0028, USA
(hinesa@si.edu). Manuscript received 29 August
2008; accepted 20 April 2009.

ABSTRACT. The Smithsonian Environmental Research Center (SERC) conducts re-
search on land-sea interactions to understand natural processes and human impacts in
linked ecosystems of the coastal zone. Coastal ecosystems support great biological pro-
ductivity and are of immense ecological and economic importance. In addition, more
than two-thirds of the human population resides in the coastal zone, where human
activities cause chronic and acute disturbance of every habitat and marked degrada-
tion of ecological balance and productivity. The Chesapeake Bay and its Rhode River
subestuary are used by SERC as model study systems to conduct long-term, intensive
monitoring and experiments. Research at SERC focuses on five grand environmental
challenges: (I) impacts of atmospheric change on climate, sea level, ultraviolet radiation,
pollutant deposition, and carbon balance; (II) impacts of watershed nutrient discharges
causing harmful algal blooms, depletion of oxygen, and destruction of submerged veg-
etation; (III) food web disruption by pollution and overfishing; (IV) invasive species; and
(V) landscape disturbance by agriculture and development. Research by SERC on these
grand challenges serves to advise policy and management from improved stewardship of
coastal resources.

INTRODUCTION

The coastal zone is of immense economic and environmental importance.
More than 50% of the Earth’s human population (3 billion people) resides in
the coastal zone and relies on the goods and services of coastal ecosystems, and
this number is expected to double by 2045 (Creel, 2003). Coastal communities
are the most densely populated and fastest growing areas in the United States:
14 of the nation’s largest 20 cities are in coastal locations; more than 50% of the
U.S. population lives in 17% of the country’s land, comprising coastal counties;
this population concentration is expected increase to 70% within 25 years; and
23 of the 25 most densely populated counties encompass coastal cities and their
surrounding sprawl (Crossett et al., 2004). The coastal environment includes the
Earth’s most biologically productive ecosystems, and this diverse environment
includes unmeasured reserves of strategic minerals, oil and gas, and other non-
living resources. The coastal zone encompasses major hubs of global transporta-
tion and commerce and unparalleled opportunities for recreation and tourism,
as well as the majority of fisheries and aquaculture industries. At the same time,

12 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

these activities cause chronic and acute disturbance of ev-
ery coastal habitat: overfishing has removed most large
species at the top of the food web, and coastal waters re-
ceive most of the waste of urban centers and agricultural
runoff of the coastal plain.

Research at the Smithsonian Environmental Research
Center (SERC) focuses on land-sea interactions. Scientists
at SERC study linked coastal ecosystems to understand
natural processes and human impacts in the coastal zone.
Ocean productivity is concentrated in the coastal fringe
where nutrients run off the land and well up from the
deep. The coastal environment includes the Earth’s most
biologically diverse ecosystems: estuaries, wetlands, man-
groves, seagrasses, coral and oyster reefs, kelp forests, and
pelagic upwelling areas. Bottom communities and water
column processes of the photic zone are most tightly cou-
pled in the nearshore shallows. Coastal waters comprise
95% of the oceans’ fisheries. Thus, SERC research focuses
on improved stewardship of these marine resources.

CHESAPEAKE BAY AND THE RHODE RIVER
SUBESTUARY AS A MODEL SYSTEM

The Smithsonian Environmental Research Center
utilizes the nation’s largest estuary, Chesapeake Bay and
its 177,000 km? watershed including six states and the
District of Columbia (Figure 1), as its primary research
landscape and main study site. In addition to SERC, this
study area includes the Smithsonian’s museum complex,
zoological exhibits, and administrative offices. An area
with a long American history of exploitation of coastal re-
sources, the Chesapeake watershed is home to 17 million
people, who are mostly concentrated in the urban centers
and suburban sprawl of Baltimore, Washington, D.C., and
Norfolk. Agriculture, particularly row crops, is the major
land use of the Chesapeake watershed, and farming has
been the major source of disturbance to the eastern de-
ciduous forest for 400 years.

Established in 1965, SERC owns a unique 1,072 ha
land holding for long-term descriptive and experimental
studies of linked ecosystems in a model subestuary and
subwatershed of Chesapeake Bay—the Rhode River,
which is located 40 km east of Washington, D.C., and
10 km south of Annapolis, Maryland (Figure 2). The
property at SERC includes cropland, forests in various
successional stages, wetlands, and 26 km of undeveloped
shoreline; this is the largest contiguous block of land
dedicated to environmental research, science education,
public access, and stewardship on the western shoreline

of Chesapeake Bay. The 585 ha Rhode River subestuary
is a shallow (maximum depth = 4 m), soft-bottom em-
bayment in the lower mesohaline zone of the Bay. The
facilities at SERC provide strategic support for research
at the site and ready access to the rest of the Chesapeake
watershed and estuary.

GRAND CHALLENGES OF COASTAL
ENVIRONMENTAL RESEARCH

The purpose of this paper is to present examples that
highlight SERC’s coastal research on five grand environ-
mental challenges. With data sets extending back to the
1970s and 1980s, SERC research monitors decadal-length
changes to distinguish seasonal and annual fluctuations
from long-term trends in the environment. Importantly,
SERC research seeks to determine mechanistic under-
standing of the causes of change at multiple spatial scales
ranging from global change to landscape, watershed, eco-
system, and community levels of organization. The land
and long-term studies at SERC’s Rhode River site afford
multidisciplinary experimental analyses of mechanisms
controlling ecological interactions. The research there
addresses the grand challenges and advises environmen-
tal policy and management for improved stewardship of
coastal resources.

GRAND CHALLENGE |: IMPACTS OF ATMOSPHERIC CHANGE

Human alterations of the atmosphere are causing
rapid changes in climate, sea level, ultraviolet radiation,
pollutant deposition, and ecosystem carbon balance. Re-
search by SERC on the salt marshes of the Rhode River
subestuary provides a good example of the ecological com-
plexities of this challenge. B. G. Drake and colleagues have
been conducting the world’s longest running experimental
manipulation of CO) on natural plant communities (1985
to present), which has been testing the effects of rising at-
mospheric CO, concentration in these salt marshes. The
experiment measures response of the two dominant plant
species at the site: Spartina patens and Scirpus olneyi. The
experiment applied nine treatment combinations of three
CO, levels in open-top chambers (ambient air at 340 ppm;
elevated CO, at a twofold increase in concentration of 680
ppm; and a control treatment without chambers) crossed
with types of patches (nearly monospecific S. patens; nearly
monospecific S. olneyi; and patches with mixes of the two
species) (Drake et al., 1989). Chambers were replaced
exactly on replicate marked plots of the nine treatment

NUMBER 38 e¢ 13

Chesapeake Bay

50 km

FIGURE 1. Map of Chesapeake Bay and its watershed with six physiographic provinces. Arrow
indicates the location of the Smithsonian Environmental Research Center on the Rhode River
subestuary and watershed. Darkened areas indicate 17 clusters of 500 subwatersheds that dif-
fered in land use and were monitored for stream discharges of nutrients.

14 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

[-] Rhode River Watershed
HMM SERC property
Private with SERC covenant
Private with easement
(P Private unprotected
Private

Smithsonian Environmental
Research Center

a
By * 4
Os

FIGURE 2. Map of land holdings (shaded green) of the Smithsonian Environmental Research Center (SERC) surrounding the Rhode River

subestuary. Red outline shows the boundary of the watershed.

combinations for the duration of the growing season for
the past 23 years (1995-2008). Photosynthesis and respi-
ration were measured in each chamber during the growing
season, and plant production was measured at the end of
each season. As predicted, Spartina patens is a C, plant
that responds weakly to rising CO, whereas growth and
production were greatly stimulated in Scirpus olneyi as a
C; plant (Drake and Rasse, 2003). However, the amount
of stimulation of S. olneyi is significantly inversely depen-
dent on salinity (i.e., water stress), with lower production
in years of high salinities (i.e., low rainfall) (Rasse et al.,
2005; and Figure 3).

Salt marsh research at SERC’s Rhode River site also
explores other ecosystem complexities. New research is
tracking the fate of the carbon added by growth stimula-

tion of the plants, which appears to be sequestered in the
peat-forming roots of the salt marsh (Carney et al., 2007).
Research conducted by J. P. Megonigal and colleagues at
the same marsh study site compares effects of increased
CO, interacting with nutrient additions to the marsh to
determine whether peat accumulation is sufficient to keep
up with rising sea level. Their initial results indicate that
the peat accumulation is equivalent to the current rate
of sea-level rise of approximately 3 mm year’, allow-
ing the marsh to persist instead of becoming submerged.
Additionally, a nonnative species, Phragmites australis, is
rapidly invading the marsh site, similar to most others
in the region (King et al., 2007); and its responses to the
interaction of rising CO, and nutrients are unknown. The
Chesapeake region has high levels of mercury deposition

NUMBER 38

(A)

Stimulation (%)

200 300 400 500 600 700 800

Precipitation (mm)

(B)

Stimulation (%)

Salinity (ppt)

FIGURE 3. Effect of (A) precipitation and (B) salinity (ppt = parts per thousand) on the stimulation of
photosynthesis by twofold increase in CO, concentration on the sedge Scirpus olneyi in open-top chambers
placed on a salt marsh of the Rhode River subestuary during a 17-year period (1989-2003). (After Rasse
et al., 2005.)

15

16 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

that is derived from coal-burning power plants. New work
at the SERC salt marsh site shows that microbes rapidly
activate the mercury (mercury-methylation) (Mitchell et
al., 2008) deposited into marshes, thus feeding it into bio-
logical processes on the coastal food web (C. Mitchell and
C. Gilmour, Smithsonian Environmental Research Center,
2008, personal communication).

GRAND CHALLENGE II:
IMPACTS OF NUTRIENT LOADING

Over-enrichment of coastal waters with nutrients
causes harmful algal blooms, depletion of oxygen, and
destruction of submerged vegetation. Eutrophication in
Chesapeake Bay and many other coastal systems is caus-
ing “dead zones” of anoxic and hypoxic waters along
deeper bottom areas. A major focus of the restoration
efforts of the Environmental Protection Agency’s Chesa-
peake Bay Program has been to reduce nutrient loading
by phosphorus and nitrogen runoff into the Bay. Long-
term watershed and estuarine water quality monitoring
by SERC at the Rhode River site and throughout Chesa-
peake Bay shows the dynamic interactions of stream dis-
charge, nutrient inputs, and plankton responses affecting
oxygen levels.

Watershed nutrient discharge occurs primarily in
storm events and is related to both geologic position
(e.g., Piedmont or Coastal Plain provinces of the Chesa-
peake watershed) and land use, especially development
and agriculture (Figure 4). Plankton productivity is much
higher in years with high runoff, which leads to plank-
ton blooms (Figure 5). Long-term monitoring from 1986
to 2004 shows that water clarity (Secchi disc depth) and
near-bottom oxygen levels have declined significantly in
the Rhode River subestuary (Figure 6). Although oxygen
levels at SERC’s long-term monitoring station in the shal-
low edge of the Bay generally do not fall below alarm-
ing levels of approximately 6 ppm, oxygen levels in the
deeper mainstem of the Bay drop to very low levels (Hagy
et al., 2004) and occasionally spill into the mouth of the
Rhode River, killing benthic organisms (A. Hines, personal
observations).

With the decline in water clarity, light levels are not
sufficient to support growth of seagrasses and other sub-
merged aquatic vegetation, which had largely disappeared
from the Rhode River subestuary and much of Chesapeake
Bay by the early 1970s. These structured ecosystems are
important nursery habitats for fish and crabs in coastal
systems such as Chesapeake Bay. Recent SERC research

Oo

a

o @410
308 @409

Nitrate-N (mg/L)

ND

—

0 10 20 30 40 50 60 70
Percentage of Cropland

FIGURE 4. Effects of cropland on stream discharge of nitrogen
for watersheds in the Piedmont and Coastal Plain physiographic
provinces of Chesapeake Bay (see Figure 1). Nitrogen is shown as
nitrate concentration on the y-axis; cropland is shown as a per-
centage of land use of the subwatershed area on the x-axis. (After
Jordan et al., 1997.)

emphasizes the linkage of submerged aquatic vegetation to
watershed characteristics (Li et al., 2007).

GRAND CHALLENGE III: FooD Wes DISRUPTION
BY POLLUTION AND OVERFISHING

Pollution and overfishing result in severe disruptions
of coastal food webs (Jackson et al., 2001). The combined
effects of low dissolved oxygen and loss of submerged
aquatic vegetation comprise much of the major impact
of pollution in coastal systems such as Chesapeake Bay.
However, inputs of mercury and other toxic chemicals also
markedly affect the food web as they become concentrated
at its upper levels, often causing serious effects on seafood
that affect human health (Krabbenhoft et al., 2007). Im-
pacts of overfishing and habitat loss have resulted in the
loss of sustainable stocks for nearly every fishery species in
Chesapeake Bay and in nearly every coastal system world-
wide. After a century of intense exploitation, disease, and
ecosystem impacts, oysters, as the Bay’s most productive

Non-bloom
year (1999)
—— Bloom year

Daily Production (g C m* d’')

=. =o ae - > 1 L S l
0 60 120 180 240 300 360
Day of Year

FIGURE 5. Comparison of carbon production in the Rhode River
subestuary during two years, one with and one without a spring
plankton bloom, which is mainly regulated by variation in spring
precipitation and watershed discharge. (After Gallegos and Jor-
dan, 1997.)

Rhode River, Maryland
2.0 a

February Secchi Depth (m)

1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006

NUMBER 38 ¢ 17

fishery historically, are now at only 1% of their biomass
in 1900 (Rothschild et al., 1994). Eutrophication and
overfishing act as multiple stressors on coastal food webs,
and management’s too narrow focus on single factors may
have adverse consequences for restoring ecosystem health
and fishery production (Breitburg et al., 2009).

Blue crabs are the remaining major lucrative fishery in
the upper Bay, but the blue crab stock has also declined by
60% since 1991 (CBSAC, 2008). Research by SERC at the
Rhode River subestuary provides the most detailed analy-
sis of blue crab ecology available (Hines, 2007). Nearly 30
years of SERC experiments show that blue crabs are the
dominant predator on benthic communities in the estuary,
and their foraging limits abundance and species compo-
sition of infaunal invertebrates as well as causing major
bioturbation of the upper 10 cm of sediments (Hines et
al., 1990). Long-term monitoring of fish and blue crabs
throughout the Rhode River subestuary shows the marked
seasonal and annual variations in population abundance
(Figure 7), as blue crabs migrate from the nursery habitat
and become inactive below 9°C in winter. Annual varia-
tion in recruitment into the Rhode River causes more than
a 10-fold fluctuation in abundance, with obvious variation
in effects of predation on infaunal invertebrates. Many up-
per Chesapeake Bay nursery habitats now appear to be be-
low carrying capacity for juvenile blue crabs (Hines et al.,
2008). Recent SERC blue crab research has focused on de-

Bottom layer D.O. (g m”)

December-March average

1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006
Year

FIGURE 6. Long-term trends in water clarity as determined by Secchi (disk) depth (left) and in oxygen concentration (D.O. = dissolved oxygen;

right) in the Rhode River subestuary. (Figure courtesy of C. Gallegos.)

18 e© SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

250

200

150

100

50

Mean monthly number of blue crabs per trawl

81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08

Year

FIGURE 7. Seasonal and annual variation in abundance of blue crabs caught in 3 m otter trawls in the Rhode
River subestuary. Abundance is the monthly mean of three trawls at each of three permanent stations within

the estuary.

veloping innovative approaches to restoring the blue crab
population in the Bay, especially by testing the feasibility
of releasing hatchery-reared juvenile blue crabs into nurs-
ery areas such as the Rhode River (Hines et al., 2008).

GRAND CHALLENGE IV: INVASIVE SPECIES

Invasions of nonindigenous species are drastically alter-
ing biodiversity, structure, and function of coastal ecosys-
tems (Ruiz et al., 2000). The largest, most comprehensive
research program on marine invasive species in the USA
is conducted by SERC. Rates of invasion into coastal
ecosystems are increasing markedly as a result of a wide
range of human-mediated vectors, but most importantly
as a result of shipping, both ballast water discharge and
hull fouling (Ruiz et al., 2000). The SERC database for
invasive species (NEMESIS) documents more than 500
invasive species of invertebrates, algae, and fish in North
American coastal waters. For Chesapeake Bay approxi-
mately 176 species are documented as established inva-

sions (Figure 8). Invasions are dynamic and ongoing in
Chesapeake Bay, as indicated by recent records of Chi-
nese mitten crabs (Ruiz et al., 2006). Many species are
having large but poorly understood impacts in Chesa-
peake ecosystems, such as the salt marsh reed Phragmites
australis (King et al., 2007).

GRAND CHALLENGE V:
LANDSCAPE DISTURBANCE BY AGRICULTURE
AND DEVELOPMENT

Agriculture and urbanization are causing widespread
modifications of landscape structure. Researchers at SERC
recently analyzed various indicators of estuarine habitat
quality for 31 Chesapeake subwatersheds that differed in
five categories of land use composition: forest, agriculture,
developed, mixed agriculture, and mixed developed (Fig-
ure 9). These land uses have profound effects on estuarine
habitat quality because they increase stormwater runoff and
loading of nutrients. Nitrogen discharge into subestuaries of

NUMBER 38 e¢ 19

Taxonomic groups of NIS introduced and established in the Chesapeake

Other
Vertebrates

Fishes

Invertebrates

Vascular Plants

Algae

Bay region (n= 176)

@ Regular Residents
@ Boundary Residents

40 60 80

Number of species

FIGURE 8. Numbers of invasive species documented for algae, vascular plants, invertebrates, fishes, and other

vertebrates (total number = 176 species) in Chesapeake Bay. Regular residents are species living in habitats

below tidal influence; boundary residents are species primarily living either above the intertidal zone or in non-

tidal freshwater and that occasionally move into tidal portions of the Bay. (NIS = noninvasive species.)

agricultural and developed watersheds was high in both wet
and dry years, but in dry years it was high only in developed
watersheds, which continue to have high human water use
regardless of rainfall (Figure 10) (Brooks et al., 2006). Land
use also has marked effects on levels of toxic chemicals in
the food webs of the subestuaries. Level of polychlorinated
biphenyls (PCBs) was highly correlated with percentage of
developed lands on the subwatershed (Figure 11).

In addition to effects on the watershed, development
of the shoreline has large impacts on coastal ecosystems.
Research by SERC in the Rhode River shows that the
shallowest fringe of the subestuary serves as a critical ref-
uge habitat for juvenile fishes and crabs to avoid larger
predators, which are restricted to deeper water (Ruiz et
al., 1993; Hines and Ruiz, 1995). Coarse woody debris
from forested shores also plays a valuable role as struc-
tural habitat and refuge from predators (Everett and
Ruiz, 1993). As development results in cutting down the
riparian forest and hardening the shoreline with bulk-
heads and riprap to prevent erosion, water depth at the
shoreline increases and the source of woody debris is
lost. With the loss of functional refuge in the nearshore
shallows, juvenile fish and crabs become increasingly ac-
cessible to predators.

CONCLUSION

The decadal data sets generated by SERC for the
linked ecosystems of the Rhode River and Chesapeake Bay
clearly show the importance of sustaining long-term, in-
tensive studies to distinguish natural variation and trends
of human impacts. The rate of change associated with hu-
man impacts is increasing markedly as the effects of global
change become manifest and as the human population of
the watershed continues to grow rapidly, with another
50% increase predicted in the next 25 to 50 years. The
interactive effects of these multiple stressors require much
more research to define improved management solutions
to restore and sustain these resources. Scientists at SERC
also extend studies of the large-scale systems of the Rhode
River and Chesapeake Bay through comparative studies
with other coastal areas, especially latitudinal compari-
sons of systems in the Smithsonian Marine Science Net-
work along the western Atlantic. Although each site has
its idiosyncratic traits, the common impacts of the grand
challenges of atmospheric change, nutrient loading, food
web disruption by pollution and overfishing, invasive spe-
cies, and land development are all manifested pervasively
in the linked ecosystems throughout the coastal zone.

20 * SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Jones Falls

Gwynns Falls

Patapsco eu angford

Curt . BZ \ Southeast

Magoth Corsica
Seven : ss

out
Rhe ian

d Avon
Battle ae

Wicomico

t: Leonard

“7 St. Clements 4
A" si il

Bis oni “4 Tisees > Ne" Manokin

Nomini

Totuskey

LANDUSE
— Agricultural
ard Developed
eal Forested
Mixed-Ag
ened Mixed-Dev

Elizabeth

a
ees Kilometers

G12'5'25 50 75 100

FIGURE 9. Map of 31 subwatersheds of Chesapeake Bay that were sampled for effects of land
use on estuarine habitats. Watersheds were categorized in the five predominant categories shown:
forest, agriculture, developed, mixed-agriculture, and mixed-developed.

NUMBER 38

2500

N
(2)
Oo
Oo

1500

Surface-water Total N (pg/L)

Forested Mixed-Dev Developed Forested Mixed-Dev Developed
Mixed-Ag Agricultural Mixed-Ag Agricultural

Watershed Land-use Class

FIGURE 10. Effect of land use on nitrogen discharge from watersheds in the five land use categories shown in
Figure 9. Stream surface discharges are compared among land use categories between a dry year with record
low rainfall (2002, left) and a wet year (2003, right) with high rainfall. (After Brooks et al., 2006.)

y = 19.781x - 8.9308
R’= 0.99

Lyi

Total PCBs in White Perch (ppb)

0) 10 20 30 40
IDW (d-‘) Percent Commercial Land in Watershed

FIGURE 11. Concentration of toxic polychlorinated biphenyls (PCBs) in white
perch (Morone americana) sampled from Chesapeake subestuaries with watersheds
of varying percentages of commercially developed land use (IDW = inverse distance
weighted). Watersheds sampled are shown in Figure 9. (After King et al., 2004.)

21

PAPA O

ACKNOWLEDGMENTS

I thank my colleagues at SERC for use of their data
and publications to illustrate this paper, especially De-
nise Breitburg, Bert Drake, Chuck Gallegos, Cindy
Gilmour, Tom Jordan, Pat Megonigal, Rick Osman,
and Greg Ruiz. I thank the Smithsonian Marine Science
Network and the Smithsonian Environmental Sciences
Program for long-term funding to SERC’s Rhode River
research program. Thanks go to Michael Lang for his
steady assistance in managing the Marine Science Net-
work and for organizing the symposium and editing
these proceedings.

LITERATURE CITED

Breitburg, D. L., J. K. Craig, R. S. Fulford, K. A. Rose, W. R. Boyn-
ton, D. Brady, B. J. Ciotti, R. J. Diaz, K. D. Friedland, J. D. Hagy
Ill, D.R. Hart, A. H. Hines, E. D. Houde, S. E. Kolesar, S. W.
Nixon, J. A. Rice, D. H. Secor, and T. E. Targett. 2009. Nutrient
Enrichment and Fisheries Exploitation: Interactive Effects on Es-
tuarine Living Resources and Their Management. Hydrobiologia,
http://www.springerlink.com/content/02q0214638972r18 (accessed
1 April 2009).

Brooks, R. P., D. H. Wardrop, K. W. Thornton, D. F. Whigham, C. Her-
shner, M. M. Brinson, and J. S. Shortle, eds. 2006. Integration of
Ecological and Socioeconomic Indicators for Estuaries and Water-
sheds of the Atlantic Slope. Final Report to U.S. Environmental
Protection Agency STAR Program, Agreement R-82868401, Wash-
ington, D.C. Prepared by the Atlantic Slope Consortium, University
Park, Penn.

Carney, K. A., B. A. Hungate, B. G. Drake, and J. P. Megonigal. 2007.
Altered Soil Microbial Community at Elevated CO, Leads to Loss
of Soil Carbon. Proceedings of the National Academy of Sciences of
the United States of America, 104:4990-4995.

Chesapeake Bay Stock Assessment Committee (CBSAC). 2008. 2008
Chesapeake Bay Blue Crab Advisory Report. National Oceanic and
Atmospheric Administration, National Marine Fisheries, Chesa-
peake Bay Office, Annapolis, Md.

Creel, L. 2003. Ripple Effects: Population and Coastal Regions. Population
Reference Bureau, 1875 Connecticut Ave., NW, Suite 520, Washing-
ton, D.C. 20009 USA. http://www.prb.org/Publications/PolicyBriefs/
RippleEffectsPopulationandCoastalRegions.aspx (accessed 1 April
2009).

Crossett, K. M., T. J. Culliton, P. C. Wiley, and T. R. Goodspeed. 2004.
Population Trends along the Coastal United States: 1980-2008.
Coastal Trends Report Series, National Ocean Service, National
Oceanic and Atmospheric Administration, Silver Spring, Md.

Drake, B. G., P. W. Leadly, W. J. Arp, D. Nassiry, and P. S. Curtis. 1989.
An Open Top Chamber for Field Studies of Elevated Atmospheric
CO, Concentration on Saltmarsh Vegetation. Functional Ecology,
3:363-371.

Drake, B. G., and D. P. Rasse. 2003. The Effects of Elevated CO, on
Plants: Photosynthesis, Transpiration, Primary Productivity and
Biodiversity. Advances in Applied Biodiversity Science, 4:53-59.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Everett, R. A., and G. M. Ruiz. 1993. Coarse Woody Debris as a Refuge
from Predation in Aquatic Communities. An Experimental Test.
Oecologia (Berlin), 93:475-486.

Gallegos, C. L., and T. E. Jordan. 1997. Seasonal Progression of Fac-
tors Limiting Phytoplankton Pigment Biomass in the Rhode River
Estuary, Maryland (USA). I. Controls on Phytoplankton Growth.
Marine Ecology Progress Series, 161:185-198.

Hagy, J. D., W. R. Boynton, C. W. Keefe, and K. V. Wood. 2004. Hypoxia
in Chesapeake Bay, 1950-2001: Long-Term Change in Relation to
Nutrient Loading and River Flow. Estuaries, 27:634-658.

Hines, A. H. 2007. Ecology of Juvenile and Adult Blue Crabs. In Biology
of the Blue Crab, V. S. Kennedy and L. E. Cronin, eds., pp. 565-654.
College Park: Maryland Sea Grant College Program.

Hines, A. H., A. M. Haddon, and L. A. Wiechert. 1990. Guild Struc-
ture and Foraging Impact of Blue Crabs and Epibenthic Fish in a
Subestuary of Chesapeake Bay. Marine Ecology Progress Series,
67:105-126.

Hines, A. H., E. G. Johnson, A. C. Young, R. Aguilar, M. A. Kramer,
M. Goodison, O. Zmora, and Y. Zohar. 2008. Release Strategies
for Estuarine Species with Complex Migratory Life Cycles: Stock
Enhancement of Chesapeake Blue Crabs, Callinectes sapidus. Re-
views in Fisheries Science, 16:175-185.

Hines, A. H., and G. M. Ruiz. 1995. Temporal Variation in Juvenile Blue
Crab Mortality: Nearshore Shallows and Cannibalism in Chesa-
peake Bay. Bulletin of Marine Science, 57:885-902.

Jackson, J. B. C., M. X. Kirby, W. H. Berger, K. A. Bjorndal, L. W. Bots-
ford, B. J. Bourque, R. H. Bradbury, R. Cooke, J. Erlandson, J. A.
Estes, T. P. Hughes, S. Kidwell, C. B. Lange, H. S. Lenihan, J. M.
Pandolfi, C. H. Peterson, R. S. Steneck, M. J. Tegner, and R. R.
Warner. 2001. Historical Overfishing and the Recent Collapse of
Coastal Ecosystems. Science, 293:629-638.

Jordan, T. E., D. L. Correll, and D. E. Weller. 1997. Relating Nutrient
Discharges from Watersheds to Land Use and Stream Flow Vari-
ability. Watershed Resources Research, 33:2579-2590.

King, R., M. Baker, A. H. Hines, D. Weller, and D. F. Whigham. 2004.
Watershed Land Use Is Strongly Linked to PCBs in White Perch in
Chesapeake Bay Subestuaries. Environmental Science and Technol-
ogy, 38:6546-6552.

King, R. S., W. V. Deluca, D. EF Whigham, and P. P. Marra. 2007. Thresh-
old Effects of Coastal Urbanization on Phragmites australis (Com-
mon Reed) Abundance and Foliar Nitrogen in Chesapeake Bay.
Estuaries and Coasts, 30:469-481.

Krabbenhoft, D., D. Engstrom, C. Gilmour, R. Harris, J. Hurley, and
R. Mason. 2007. “Monitoring and Evaluating Trends in Sediment
and Water.” In Ecosystem Responses to Mercury Contamination:
Indicators of Change, ed. R. Harris, pp. 47-86. Boca Raton, Fla.:
CRC Press.

Li, X., D. E. Weller, C. L. Gallegos, T. E. Jordan, and H.-C. Kim. 2007.
Effects of Watershed and Estuarine Characteristics on the Abun-
dance of Submerged Aquatic Vegetation in Chesapeake Bay Sub-
estuaries. Estuaries and Coasts, 30(5):840-854.

Mitchell, C. P. J., B. A. Branfireun, and R. K. Kolka. 2008. Spatial Char-
acteristics of Net Methylmercury Production Hot Spots in Peat-
lands. Environmental Science and Technology, 42(4):1010-1016;
doi:10.1021/es0704986.

Rasse, D. P., G. Peresta, and B.G. Drake. 2005. Seventeen Years of El-
evated CO, Exposure in a Chesapeake Bay Wetland: Sustained but
Contrasting Responses of Plant Growth and CO, Uptake. Global
Change Biology, 11:369-377.

Rothschild, B. J., J. S. Ault, P. Goulletquier, and M. Heral. 1994. De-
cline of the Chesapeake Bay Oyster Population: A Century of Habi-
tat Destruction and Overfishing. Marine Ecology Progress Series,
111:29-39.

Ruiz, G., L. Fegley, P. Fofonoff, Y. Cheng, and R. Lemaitre. 2006. First
Records of Eriocheir sinensis H. Milne Edwards, 1853 (Crustacea:
Brachyura: Varunidae) for Chesapeake Bay and the Mid-Atlantic
Coast of North America. Aquatic Invasions, 1:137-142.

Ruiz, G. M., P. W. Fofonoff, J. T. Carlton, M. Wonham and A. H.
Hines. 2000. Invasion of Coastal Marine Communities of North

NUMBER 38 e¢ 23

America: Patterns and Processes. Annual Reviews in Ecology and
Systematics, 31:481-531.

Ruiz, G. M., A. H. Hines, and M. H. Posey. 1993. Shallow Water as a
Refuge Habitat for Fish and Crustaceans in Non-vegetated Estuar-
ies: An Example from Chesapeake Bay. Marine Ecology Progress
Series, 99:1-16.

H yy ’ Mee i ba
= ;
= = 5
x
é
ry *:
art) & ie io
oo 4, aia a
: ee ye
z
1 ¢ y
<i
]
é «
= =) ae

Smithsonian Marine Station at Fort Pierce:
Thirty-Eight Years of Research on the Marine
Biodiversity of Florida

Valerie J. Paul, Julianne Piraino, and Laura

Diederick

Valerie J. Paul, Julianne Piraino, and Laura Died-
erick, Smithsonian Marine Station at Fort Pierce,
701 Seaway Drive, Fort Pierce, Florida 34949,
USA. Corresponding author: V. Paul (paul@si.edu).
Manuscript received 29 August 2008; accepted 20
April 2009.

ABSTRACT. The Smithsonian Marine Station at Fort Pierce, located on South Hutchin-
son Island in Fort Pierce, Florida, has had an ongoing program in the marine sciences
since the early 1970s. Funded by a private trust from J. Seward Johnson, Sr., to the
Smithsonian, the marine program has supported the research of Smithsonian scientists
and their associates, postdoctoral fellows, resident scientists, and the operations of the
station, including a small support staff. The station is administered by the National Mu-
seum of Natural History as a facility for research dedicated to the marine sciences. The
Smithsonian Marine Station at Fort Pierce has developed a strong, broadly based re-
search program focusing on ecology, evolution, systematics, and life histories of marine
organisms. Ongoing studies address important issues in biodiversity, including global
climate change, invasive species, harmful algal blooms, larval ecology, and evolutionary
developmental biology.

INTRODUCTION

The Smithsonian Marine Station at Fort Pierce (SMS) is dedicated to study-
ing the rich diversity of marine life of the Indian River Lagoon and Florida coast.
In sharing its findings with the scientific community, resource managers, and the
general public, the Marine Station promotes the conservation and stewardship
of Florida’s vast marine resources. Research activities focus on the Smithsonian
Institution’s core scientific emphasis of discovering and understanding life’s di-
versity. Although most research projects focus on biodiversity, life histories, and
ecology of marine and estuarine organisms, complementary studies of physi-
cal and chemical processes related to the marine environment are also part of
the Station’s investigations. The insights gained by the research conducted at
SMS are widely disseminated through scientific publications (more than 780 to
date; see complete listing on the Station’s website www.sms.si.edu), scientific
and public presentations, popular articles, and the media, thus contributing to
the broader mission of the Smithsonian Institution for the “increase and diffu-
sion of knowledge.”

The Smithsonian’s presence in Fort Pierce, Florida, began in 1969 through
an association with Edwin Link, an inventor and engineer who was involved
at that time in the design of research submersibles, and J. Seward Johnson, Sr.,

26 ¢ SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

founder of the Harbor Branch Foundation, now known
as the Harbor Branch Oceanographic Institute (HBOI) at
Florida Atlantic University (Figure 1). In late 1969, the
Smithsonian was given two trust funds through the gener-
osity of J. Seward Johnson, Sr., for the development and
maintenance of a submersible (then under construction)
and for research in underwater oceanography. At its com-
pletion in 1971, the submersible, the Johnson Sea-Link I,
was donated to the Smithsonian. In 1973, after a tragic
accident in the Johnson Sea-Link in which two men died,
the Smithsonian transferred ownership of the submersible
to Harbor Branch. Following the transfer of the submers-
ible, the Smithsonian’s marine research program in Fort
Pierce continued to be supported by income from both
trust funds, then later, after certain legal resolutions, by
one of the two funds, designated as the Hunterdon Fund.
The Smithsonian carried out its activities in Fort Pierce on
the grounds of the Harbor Branch Foundation (Link Port)
under the auspices of the Fort Pierce Bureau, a unit admin-
istered directly by the Office of the Secretary and then later
by the Assistant Secretary for Science. In March 1981 this
Bureau was dissolved as an organizational entity, and the
administrative responsibility for the Smithsonian research

programs at Link Port was transferred to the Director
of the National Museum of Natural History (NMNRH).
The organization was then retitled by the Secretary of the
Smithsonian as the Smithsonian Marine Station at Link
Port. At the time of the transfer of administrative respon-
sibility, the directive from the Office of the Assistant Sec-
retary for Science was that a strong research program in
marine science should be established and that the program
should be open to all marine scientists in the Smithson-
ian Institution. In response, Richard Fiske, then Director
of NMNH,, established an inter-unit advisory committee,
appointing Catherine Kerby, his administrative assistant,
as chair of the committee, and Mary Rice, Department
of Invertebrate Zoology (on assignment to the Fort Pierce
Bureau), as director of the facility and research programs
at Link Port. Rice held this position until her retirement
in 2002, at which time Valerie Paul was selected as her
successor.

The Smithsonian Marine Station at Link Port was ini-
tially set up with a small on-site staff and well-equipped
laboratories and field facilities to provide opportunities for
Smithsonian scientists and their colleagues to conduct field
research in a highly diverse subtropical marine environ-

FIGURE 1. J. Seward Johnson, Sr. (left), and inventor Edwin Link were instrumental in pro-

viding funding and submersibles for the Smithsonian’s marine research program in Fort Pierce,

Florida.

ment. This plan gave Museum scientists the opportunity
to extend and broaden their research from museum col-
lections to field studies in such areas as behavior, ecology,
physiology, and life histories. Moreover, it provided the
opportunity for all Smithsonian marine scientists to carry
out comparative studies of the diverse ecosystems and bi-
ota within the Fort Pierce vicinity and peninsular Florida
and, most importantly, to establish long-term databases
and to conduct long-term studies. An important compo-
nent of the plan was to include postdoctoral fellows, both
to complement the research of Smithsonian scientists and
to contribute to training of future generations of marine
scientists. In addition, by serving many Smithsonian scien-
tists (as opposed to a few resident scientists), the program
was conceived to yield maximum productivity of high-
quality modern science and to be the most equitable and
effective use of available funds.

For the first 18 years the Smithsonian Marine Station
at Link Port used a vintage WW II barge as a floating labo-
ratory docked at Harbor Branch (Figure 2) as the base
of operations for its highly successful research program,
which was carried out primarily by visiting scientists from
the Smithsonian, their colleagues, and postdoctoral fel-

NUMBER 38 e¢ 27

lows. Restrictions imposed by the space and structural
limitations of the barge for many research activities as well
as its high maintenance requirements led the Smithsonian
to pursue plans for a land-based laboratory.

In May 1999 these plans were realized when, with the
approval of J. Seward Johnson, Jr., and a signed Memo of
Understanding, the Smithsonian Marine Station relocated
to an 8 acre site acquired from the MacArthur Foundation
near the Fort Pierce Inlet, 7 miles south of Harbor Branch.
At this time the official name of the station was changed to
the Smithsonian Marine Station at Fort Pierce. The move
was made into a newly constructed 8,000 square foot
building with offices and laboratories for visiting scien-
tists, resident staff and postdoctoral researchers, general-
use laboratories for chemistry, microscopy, and molecu-
lar research (Figures 3, 4), and a wet laboratory supplied
by a small seawater system. In March 2003, a 2,400
square foot storage building was completed. The build-
ing includes a workshop and storage for scientific supplies,
scuba equipment, and other marine research equipment.
In April 2004 a research dock was completed on the In-
dian River Lagoon, which is accessible by an easement on
adjacent property. A flow-through seawater building was

FIGURE 2. A retired World War II Army barge was remodeled to include two levels of offices and laboratories for use by the Smithsonian
Marine Station scientists from the early 1970s to 1999.

28 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

ATLANTIC OCEAN

FIGURE 3. This aerial view shows the location of the Smithsonian Marine Station on the Fort Pierce Inlet of the Indian

River Lagoon in Florida.

added to the campus in August 2005. The relocation to
the new research building and campus provided the op-
portunity for the Smithsonian Marine Station to increase
and strengthen the breadth and diversity of its research
as well as to establish new collaborative interactions. The
move also made it possible to expand the Station’s educa-
tional mission, initiating new cooperative ventures in edu-
cation and public outreach.

In the struggle to understand life, how its diversity has
come about, and the current rapid loss of biodiversity on a
global scale, the Smithsonian Marine Station is positioned
as are few laboratories in the world to study this excep-
tional diversity from an array of environments. The Smith-
sonian Marine Station is located on the Fort Pierce Inlet of
the Indian River Lagoon (IRL) (see Figure 3), an estuary
extending along one-third the length of the east coast of
Florida. The IRL is widely recognized as one of the most
diverse estuaries in North America, and it has been desig-
nated an estuary of national significance by the Environ-
mental Protection Agency. The Marine Station’s unique
location on the Fort Pierce Inlet puts it in a prime position
to access oceanic waters and to sample organisms from the
Florida Current and other offshore habitats. This region of
Florida’s coast, characterized as a transitional zone where
temperate and tropical waters overlap, offers access to a
great variety of habitats and an extraordinary diversity of
species. To the south of Fort Pierce, within a few hours of
travel, are Florida Bay and the Florida Keys, the only living
tropical coral reefs in the continental United States.

Specialized equipment and instrumentation at the
Smithsonian Marine Station include temperature-con-
trolled aquaria and incubators, equipment for preparing
tissues for light and electron microscopy, an ultracold
freezer, equipment for electrophoresis, a thermocycler
for DNA amplification, high performance liquid chro-
matographs, a gas chromatograph/mass spectrometer,
and a UV-visual spectrophotometer. For microscopic
studies, equipment is available for light, epi-fluorescent,
and Nomarski microscopy, time-lapse and normal-speed
cinematography, photomicrography, video recording and
editing, inverted microscopy, scanning and transmission
electron microscopy (Figure 5), and confocal laser scan-
ning microscopy.

Confocal laser scanning microscopy (CLSM) has be-
come an increasingly important tool in modern environ-
mental microbiology, larval ecology, developmental biol-
ogy, and biochemistry. CLSM involves the use of a light
microscope, laser light sources, a computer, and special
software to image a series of in-focus optical sections
through thick specimens. The specimens, which can be
live or fixed, are stained with fluorescent dyes that high-
light specific structures when excited by the lasers. Once
the stacks of two-dimensional (2-D) images are collected,
the computer software constructs spectacular, information-
rich, three-dimensional (3-D) images that yield a wealth of
information. In June 2008, the Smithsonian Marine Station
acquired a Zeiss LSM510 confocal system that is already
providing data in the cutting-edge studies of Postdoctoral

NUMBER 38 e¢ 29

FIGURE 4. Smithsonian Marine Science Network Postdoctoral Fellow Koty Sharp uses molecular methods to deter-
mine the diversity of bacteria associated with corals.

30 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 5. Julie Piraino, Laboratory Manager, examines the larva of a sipunculan worm on the scanning electron microscope.

Fellow Koty Sharp on the presence and transmission of
bacteria in corals, and in research conducted by Postdoc-
toral Fellow Kate Rawlinson on the fate of individual cells
in the development of embryos of polyclad flatworms.
This new microscope will greatly increase the capabilities
of Smithsonian marine scientists to conduct probe-based
subcellular studies in biochemistry, microbiology, and de-
velopmental biology (Figure 6).

The Marine Station owns four boats for use in field
studies: a 17 foot Boston Whaler and a 21 foot Carolina
Skiff for work in the shallow waters of the IRL, a 21 foot
center-console boat to access nearshore waters, and a 39
foot vessel, the R/V Sunburst, for offshore research activi-
ties. These vessels provide access to the diverse marine
and estuarine environments in the vicinity of SMS. The
excellent location, facilities, instrumentation, and skilled
staff of the Smithsonian Marine Station facilitate research

on many diverse topics in marine biology and marine
biodiversity.

RESEARCH ACTIVITIES

The Smithsonian Marine Station at Fort Pierce is an
important contributor to the marine research and col-
lections at the National Museum of Natural History. It
provides a vital link between tropical and temperate eco-
systems in a coastal network of marine research stations
known as the Smithsonian Marine Science Network. The
Marine Science Network (MSN) is an array of laborato-
ries spanning the western Atlantic coastal zone and across
the Isthmus of Panama, facilitating long-term interdisci-
plinary, comparative research among MSN sites, including
the Smithsonian Environmental Research Center (SERC)

FIGURE 6. Left: Confocal microscopy captures
an oxidative burst reaction by the green alga Dic-
tyosphaeria cavernosa following exposure to the
fungus Lindra thalassiae. An oxidative burst is
an explosive production of reactive oxygen spe-
cies (hydrogen peroxide is an example) intended
to act as first defense against invading pathogens.
Middle: Confocal image shows development of
musculature and nervous system in a larva of a
sipunculan worm. Right: Confocal laser scan-
ning micrograph of the musculature of a Miller’s
larva of the flatworm Cycloporus variegatus.
Phalloidin staining shows circular and longitudi-
nal muscles (cm, Im, respectively) and the ciliary
band (cb). Scale = 30 xm.

in Maryland, the Carrie Bow Cay Marine Field Station
in Belize, and the Smithsonian Tropical Research Institute
(STRI) in Panama.

Research at SMS continues to be carried out by Smith-
sonian scientists from various units within the Institution

NUMBER 38 e 31

along with their colleagues from other national and inter-
national institutions, as well as by resident SMS scientists,
postdoctoral fellows, and graduate students (Figure 7).
Ongoing research programs by resident scientists at the
Smithsonian Marine Station involve coral reef research,

32 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 7. Visiting Scientist Anastasia Mayorova (kneeling) collects sipunculan worms with the assistance of Mary
Rice, Director Emeritus of SMS (left), and Research Technician Woody Lee.

monitoring that is supporting restoration of the Florida Ey-
erglades, harmful algal blooms, marine natural products,
and invertebrate larval life histories, evolution, and devel-
opment. The Smithsonian Marine Station promotes the
education of emerging scientists by offering pre- and post-
doctoral research fellowships and supporting the work of
student interns. Examples of ongoing research activities
are discussed below.

MARINE BIODIVERSITY

The Smithsonian Marine Station has long had a cen-
tral focus on documenting biodiversity of marine life in the
most diverse coastal waters of the continental United States.
NMNuH invertebrate zoologist David Pawson has discovered
and documented echinoderms (sea urchins, sand dollars, sea
cucumbers) in shallow and deep waters of Florida for more
than 25 years (Hendler et al., 1995). He has found sand dol-
lars that are probably hybrids between two species in the

offshore waters of Fort Pierce. Other groups of organisms
that have been well studied by NMNH researchers for many
decades include the marine algae (Mark and Diane Littler),
foraminifera (Marty Buzas), crustaceans (Rafael Lemaitre
and colleagues), deep- and shallow-water mollusks (Jerry
Harasewych and Ellen Strong), and meiofaunal organisms
(animals less than 1 mm in size that live in sand and sedi-
ments) (Jon Norenburg and coworkers). Additionally, many
SMS scientists, including former director Mary Rice, have
focused on understanding the diversity and distribution of
larval forms of different groups of marine invertebrates.
These larval stages are morphologically and ecologically
very different from adult life stages and are extremely im-
portant for the transport and propagation of marine species,
sometimes over long distances (Figure 8). Tuck Hines and
Richard Osman (SERC) have also studied recruitment pat-
terns and larval ecology for a variety of invertebrate larval
forms in the IRL. A few examples of the many biodiversity
studies conducted at SMS are highlighted below.

FIGURE 8. Examples of marine invertebrates, the larval develop-
ment of which has been studied by Mary Rice and colleagues at SMS
for more than 30 years. Organisms shown here, clockwise from up-

per left, are a starfish brachiolarian larva, a sipunculan pelagosphera
larva, a brachiopod larva, and a flatworm Muller’s larva.

With much of their research focused on ecology, phys-
iology, and pollution-oriented work, Mark and Diane Lit-
tler observed a growing need for an easier means for field
scientists and resource managers to identify the diverse
and abundant marine algal species in the field. They have
published user-friendly field guides, including the award-
winning book Caribbean Reef Plants (Littler and Littler,
2000), with much of their laboratory and field research
based at SMS. More recently, with co-author M. Dennis
Hanisak (Harbor Branch Oceanographic Institute), they
published Submersed Plants of the Indian River Lagoon:
A Floristic Inventory and Field Guide (Littler et al., 2008),
a book rich with photography and illustrations depicting
the taxonomy and distributional patterns of more than
250 species of submersed plants in the Indian River La-
goon. The book was based on six years of field and labora-
tory work along the central east coast of Florida.

D. Wayne Coats, a protistan ecologist at the Smithso-
nian Environmental Research Center, has worked on the
biology and ecology of free-living and symbiotic protists
for 20 years. His work has enabled comparisons between
the Chesapeake Bay and Indian River Lagoon estuaries
and provided enhanced understanding of how eukaryotic
microbes influence the structure of marine food webs (Sno-
eyenbos-West et al., 2004). Much of Coats’ work at SMS

NUMBER 38 e¢ 33

has considered the biodiversity and trophic biology of pro-
tists living in coastal waters of Florida or associated with
local marine fauna. He has shown parasitism of plank-
tonic ciliates in the Indian River Lagoon to be a major
pathway for recycling material within the microbial loop.
Coats and his graduate students have also shown that many
free-living photosynthetic dinoflagellates have the ability
to feed on ciliate protozoa. Although ingestion rates are
typically low, the high densities attained by red-tide dino-
flagellates in the Indian River Lagoon and Chesapeake Bay
make their ability to ingest ciliates an important microbial
food web interaction. Feeding on ciliates and other pro-
tists may help sustain blooms when nutrient resources for
photosynthesis are limited. Coats and his colleagues have
also revealed a rich and poorly known ciliate fauna asso-
ciated with the respiratory tract of bottle-nosed dolphins
and other cetaceans (Ma et al., 2006). Previously reported
to be parasitic, these ciliates appear not to directly impact
the health of animals held in captivity. Through his work
at SMS, Coats has helped define the significance of pro-
tists within the marine ecosystem. In some instances, these
protists compete directly with zooplankton for food re-
sources, thus limiting the upward movement of energy and
matter in the food web. In other instances, they can recycle
biomass not readily grazed by zooplankton, thus repack-
aging it in a form that can move more readily through the
food web.

Mary Rice, former director of SMS, established a pro-
gram of life history studies more than 30 years ago involv-
ing numerous postdoctoral fellows and visiting scientists
who have worked on a variety of marine invertebrates.
Her research has focused on an enigmatic group of ma-
rine worms known as sipunculans. Presumably a primitive
group related to annelids and mollusks, sipunculans are
unique in their complete lack of segmentation and single
unpaired ventral nerve cord. One of several objectives of
her studies has been the use of developmental studies to
understand phylogenetic affinities both within the group
and with other spiralian phyla (Schulze and Rice, 2009).
Other objectives have been comparative studies of repro-
ductive biology and ecology of shallow-water and deep-sea
species, an investigation of the biology of oceanic larvae,
including their metamorphosis and their role in species
distribution, and a systematic survey of the Sipuncula of
Florida and the Caribbean.

In studies of reproductive biology, comparative in-
formation was gathered for numerous species on gameto-
genesis, spawning, egg sizes, egg maturation, fertilization,
and reproductive seasonality (Rice, 1989). Year-long ob-
servations of reproductive activity in Phascolion cryptum,

34 * SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

a small species inhabiting discarded gastropod shells in
the subtidal waters of the Indian River Lagoon, revealed
that—in contrast to temperate species—animals were
reproductive throughout the year. A collaborative ultra-
structural study of spermiogenesis was also conducted.
The most abundant sipunculan of the Indian River La-
goon, this species was recorded in densities up to 2,000
to 3,000 per square meter. No longer found in these den-
sities, the population has declined for reasons unknown
(Rice et al., 1983).

Studies of larval biology by Rice and collaborators
have concentrated on the oceanic pelagosphera larvae of
sipunculans that occur in abundance in the Florida Cur-
rent, a component of the Gulf Stream System that flows
along the edge of the Continental Shelf offshore from Fort
Pierce. Reported in warm water currents throughout the
world’s oceans, these larvae are known to be long lived,
existing in the larval stage for 6 to 7 months, and hence to
have the potential for widespread species dispersal (Rice,
1981). Continuing for more than three decades, the stud-
ies have included descriptions of the various larval morpho-
types through light and scanning microscopy, as well as an
investigation of factors inducing metamorphosis and the
identification of species by rearing larvae to adulthood.
Several of the larvae were identified by rearing, surviving
in the laboratory for periods of 3 to 26 years as adults. In
more recent collaborative studies (with Postdoctoral Fel-
low Anja Schulze and staff of the NMNH Laboratory of
Analytical Biology), genomic analysis, comparing larval
and known adult sequences, was utilized to identify addi-
tional larval types. These analyses suggested the presence
of two cryptic species, characterized by morphologically
similar adults but different larval types.

Jon Norenburg (NMNH), together with students,
postdoctoral fellows, and collaborators, has been and is
focused on discovering nemertean diversity in Florida and
using that diversity to address broader questions. In short
visits over the course of 20 years they have collected as
many as 70 putative species, primarily from the shoreline
and shallow coastal waters, from a region with 24 previ-
ously known species. Many of the additional species are
potential range extensions that await confirmation with
specimens from type locales, especially those in southern
Brazil, which is the nearest subtropical nemertean fauna
that also is well documented. There also are tantalizing
preliminary data for close genetic links with European
species (Maslakova and Norenburg, 2008). New spe-
cies have been named, and another 10 to 15 potential
new species await additional specimens or genetic work.
Most nemerteans have few to no external diagnostics to

characterize and discriminate species unambiguously. Al-
most all the species collected in Florida by Norenburg
and coworkers in the past 15 years were processed with
genetic work in mind, which will resolve some questions
of identity and yield realistic estimates of true diversity
and contribute important samples for studying diversifi-
cation of the phylum (Thollesson and Norenburg, 2003).
That effort in Florida is an important component of two
global-scale nemertean projects headed by Norenburg:
(1) diversity and coevolution of the specialized, ectosym-
biotic carcinonemertid worms with their decapod crus-
tacean hosts (mostly crabs), and (2) phylogeny and bio-
geography of Ototyphlonemertes, which are specialized
and miniaturized worms occupying the aqueous pore
space in coarse sediments, such as coarse sand beaches
and in high-current subtidal habitats. Norenburg’s study
of nemerteans in Florida has contributed important orig-
inal observations about developmental biology of nemer-
teans, and one species in particular has revolutionized
our understanding of nemertean evolution (Maslakova
et al., 2004).

Carole Baldwin and Lee Weigt from the National
Museum of Natural History are studying fish diversity,
including larval fishes, through DNA barcoding methods.
Fish taxonomists have traditionally classified fishes based
on morphological features that can be seen and described.
However, many families of fishes, such as parrotfishes and
gobies, have members that look so similar they are virtu-
ally indistinguishable without examining the genetic mate-
rial. Baldwin and her research team have now cataloged
more than 200 species (from more than 1,000 specimens)
from the Indian River Lagoon. Processing involved identi-
fying and measuring each fish, photographing its live col-
oration, taking a tissue sample for DNA analysis, and pre-
serving the rest of the specimen as a voucher for NUNH
archival collections. Tissue samples from each specimen
were used to create a DNA barcode, which is unique to
the individual fish species and can be used for identifica-
tion purposes. Not only will this work be important for
establishing a database of genetic information for fishes
of the Indian River Lagoon, it will greatly increase our
understanding of shorefish diversity. The overall goal of
the work is to provide a new, more realistic estimate of
species diversity in the Caribbean, Florida, and adjacent
areas. Having amassed DNA extractions from fishes from
a variety of taxa and from multiple localities in the tropi-
cal Atlantic, the investigators can now examine interspe-
cific phylogenetic relationships to investigate patterns of
speciation and potential patterns of morphological diver-
gence accompanying speciation.

Important reasons often cited for understanding
biological diversity are the possible benefits these spe-
cies might yield as foods, medicines, or for other human
uses. Valerie Paul, Director of SMS, and members of her
research group investigate the chemical diversity of ma-
rine organisms by studying marine natural products, small
molecules produced as chemical signals or as toxins or
chemical defenses. Members of Paul’s research team iso-
late and characterize natural products from Florida’s ma-
rine life (seaweeds and invertebrates) and have discovered
compounds that have never previously been found in na-
ture. A current area of interest for her research group is
the biodiversity and chemical diversity of benthic marine
Cyanobacteria. Through collaborations with medicinal
chemists, they are actively investigating the beneficial uses
of these compounds for treatment of human diseases such
as cancer and bacterial infections.

MARINE ECOLOGY

Valerie Paul studies marine plant—animal interactions
in coral reef habitats. Coral reefs in Florida and through-
out the world are declining, in part the consequence of
shifts from coral- to algal-dominated communities. Paul
and members of her research team study grazing by reef
fishes and sea urchins and the effects of herbivory on
coral reef community structure (Paul et al., 2007). They
have found that chemical defenses of marine algae al-
low some well-defended marine plants to dominate on
coral reefs despite grazing pressure. Key to the recovery
of coral reefs is the successful recruitment of coral larvae
to become juvenile and eventually adult corals (Ritson-
Williams et al., 2009). Paul’s research group and their
collaborators examine positive and negative interactions
between coral larvae and the marine algae that dominate
coral reef habitats. Some of the same species of algae
that are chemically protected from grazers can inhibit the
settlement of coral larvae, thus preventing the successful
recovery of coral reefs.

Algae are an essential part of marine ecosystems
and when maintained in balance can provide food, shel-
ter, oxygen, and more to millions of organisms, includ-
ing people. But some algae can produce harmful toxins
and, under certain conditions, can grow out of control.
These so-called harmful algal blooms have been increas-
ing in frequency and severity along the world’s coastlines.
NMNH scientist Maria Faust has been investigating the
types of planktonic harmful algae, often called red tides,
which occur along Florida’s east coast (Faust and Tester,
2004). Valerie Paul, NMNH Statistician Lee-Ann Hayek,

NUMBER 38 e¢ 35

and Postdoctoral Fellows Karen Arthur and Kate Semon
have been studying formation of blooms of marine cya-
nobacteria in Florida’s estuaries and coral reefs and try-
ing to elucidate environmental factors that contribute to
bloom formation (Paul et al., 2005). Increased nutrients
from land-based sources, such as runoff from fertilizers
and sewage treatment plants, may help to fuel some of
these algal blooms. The biological and biochemical diver-
sity of harmful algae is the subject of ongoing research,
which has led to the discovery of novel toxins produced
by these cyanobacteria.

Estuaries and coasts around the world are approaching
critical levels of degradation, and the southern Indian River
Lagoon is no exception. Large-scale, collaborative efforts
are underway to restore biodiversity and the vital ecologi-
cal functions these ecosystems provide. Bjorn Tunberg and
members of his benthic ecology research team at SMS are
involved in one of the most ambitious of these projects, the
Comprehensive Everglades Restoration Plan (CERP). Ex-
tensive modifications to the southern IRL watershed over
the past 100 years have decreased the system’s ability to
store water and have increased nutrient-rich stormwater
runoff. The CERP plan, under the direction of the South
Florida Water Management District and the U.S. Army
Corps of Engineers, aims to restore wetlands and build
water storage basins to improve estuarine health. Tunberg
and his team established a benthic monitoring program
five years ago to provide a baseline data set of species dis-
tribution and abundance in the sediments of the Indian
River Lagoon. This team is acquiring quarterly data that
will allow them to detect and predict long-term changes
in the benthic communities throughout the central and
southern Indian River Lagoon.

The location of the Smithsonian Marine Station on
the Indian River Lagoon for 37 years has allowed Smith-
sonian researchers to establish long-term and intensive
research projects that are valuable in understanding and
assessing marine biodiversity. Long-term biological moni-
toring is most effectively carried out on organisms with
high densities, many species, short generation times, quick
responses to changes in environmental variables, and a
long history of extensive study on a worldwide basis.
The benthic foraminifera fit these requirements, and their
populations have been monitored in the Indian River La-
goon for more than 30 years by Marty Buzas (NMNH).

At one station near the Harbor Branch Oceanographic
Institute, monthly replicate sampling of foraminifera liv-
ing in the sediment has been carried out since 1977. These
data indicate significant differences between seasons, as
well as among years, but no overall increase or decrease

36 ¢© SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

over a longer time span. The spatial distribution of the
foraminifera forms an environmental mosaic of patches
whose densities change with time. This newly discovered
phenomenon was termed pulsating patches (Buzas et al.,
2002; Buzas and Hayek, 2005). At the St. Lucie Inlet, ob-
servations were made in 1975-1976 and again 30 years
later in 2005. Species richness had greatly declined over
30 years, and the community structure of the foraminifera
in this area was completely destroyed (Hayek and Buzas,
2006). Monitoring at this Inlet during 2007-2008 has
shown that species richness has increased; however, ex-
cept for the abundant species, the fauna does not contain
the same species as it did 30 years ago. Monitoring efforts
are continuing, and Buzas and Hayek have also begun a
coring program to determine the effects of both natural
and anthropogenic effects on community changes during
the past 150 years.

Candy Feller, Dennis Whigham, coworkers from the
Smithsonian Environmental Research Center (SERC), and
national and international collaborators have conducted
long-term studies of the mangrove ecosystems of the In-
dian River Lagoon. The overall goal of this project is to
collect hydrological, nutrient, microbial, and vegetation
data in support of their long-term ecological studies of
factors that control the structure and function of man-
grove ecosystems (Figure 9, top). Feller has continued a
study of how nutrient enrichment affects the mangrove
communities along the Atlantic coast of Florida for the
past 10 years. Fertilization experiments designed to en-
rich nitrogen (N) and phosphorus (P) in sediments have
shown that black mangrove forests in Florida are nitrogen
limited. When nitrogen was added in the IRL, the black
mangroves grew out of their dwarf form (Feller et al.,
2003). Addition of N also affected internal dynamics of
N and P, caused increases in rates of photosynthesis, and
altered patterns of herbivory (Lovelock and Feller, 2003).
These findings contrast with results for mangrove forests
in Belize and Panama where the seaward fringe was N-
limited but the dwarf zone was P-limited. Their studies
have demonstrated that patterns of nutrient limitation in
mangrove ecosystems are complex, that not all processes
respond similarly to the same nutrient, and that similar
habitats are not limited by the same nutrient when differ-
ent mangrove forests are compared (Lovelock et al., 2006;
Feller et al., 2007).

Feller and her colleagues have also studied the ef-
fects of the 2004 hurricanes Frances and Jeanne on the
mangrove communities (Figure 9, bottom). Over the past
4 years they have continued to monitor and quantify the
recovery of the mangroves, documenting tree height, leaf

area index, mangrove type, mangrove defoliation and
recovery, hydrology, and salinity. Damage to the man-
groves was higher in the fringe and transition zones than
in the dwarf zone. The N-fertilized trees sustained sig-
nificantly higher damage than controls in all zones and
have been slower to recover. After 2.5 years, the leaf area
index (LAI) of P-fertilized and control trees was equal to
pre-storm levels, whereas +N trees were less than 90%
recovered. LAI again decreased dramatically in Janu-
ary 2007, presumably as the result of an intense 2 year
drought in Florida.

Dennis Whigham and colleagues from the University
of Utrecht, The Netherlands Institute for Ecology—Centre
for Limnology, and the University of South Florida are
determining the relationships between the structure and
productivity of different mangrove habitat types and
hydrological processes and nitrogen cycling, including
characteristics of the microbial community associated
with nitrogen cycling. Their hydrological studies have
shown that there is no evidence of freshwater input from
groundwater into their study site and that the ground-
water chemistry is primarily influenced by evapotrans-
piration. Subsequently, salt pans and dwarf mangrove
communities develop in areas that are characterized by
hypersaline conditions associated with evapotranspira-
ton. Growth rates are lower in the salt pan and dwarf
mangrove habitats, and preliminary results indicate that
the microbial community in those habitats differs from
other habitats.

The studies described above document the morpho-
logical, genetic, and biochemical diversity of Florida’s
marine life. Collectively, these biodiversity and ecological
studies and investigations of long-term changes in Flor-
ida’s coastal waters, including the Indian River Lagoon,
mangrove ecosystems in Florida, and coral reef habitats
of southeast Florida and the Florida Keys, give us the
background essential to document changes in biodiversity
resulting from human and climatic impacts on Florida’s
coastal environments.

EDUCATION AND OUTREACH

As a resource for educators, students, researchers, and
the public, the Marine Station maintains a species inven-
tory of plants and animals in the Indian River Lagoon.
The Indian River Lagoon Species Inventory website (www
.sms.si.edu/IRLspec) is continually expanding and now in-
cludes more than 3,000 species with many photographs
and scientific references. In addition to individual species

NUMBER 38 ¢ 37

FIGURE 9. Top: From front to back, Smithsonian Marine Station graduate fellow Juliane Vogt (University
of Dresden), postdoctoral fellow Cyril Piou, and volunteer Rainer Feller explore the mangrove at Hutchinson
Island, Florida, looking for light gaps. Bottom: Sharon Ewe (on left) and Anne Chamberlain examine damage
to the mangroves at Hutchinson Island, Florida, immediately after Hurricane Jeanne.

38 © SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

reports that give habitat, distribution, life history, popula-
tion biology, physical tolerance, and community ecology
information, the database includes information on non-
native and endangered, threatened, and special-status spe-
cies. An electronic companion publication to the Species
Inventory is the Field Guide to the Indian River Lagoon
(www.sms.si.edu/IRLfieldguide). Both projects have been
supported by the Indian River Lagoon National Estuary
Program administered by the St. Johns River Water Man-
agement District. Features such as an interactive glossary,
enhanced indexing, and links to other relevant websites
add to the educational value of these websites.

The Smithsonian Marine Ecosystems Exhibit in the
St. Lucie County Marine Center celebrated its seventh an-
niversary in August 2008. Administered by the Smithson-
ian Marine Station, the exhibit showcases the Caribbean
coral reef ecosystem that was a popular exhibit at the Na-
tional Museum of Natural History for more than 20 years
and the first living model of an Atlantic coral reef ecosys-
tem available for public viewing. Through an outpouring
of local interest and support, the exhibit was transferred
to Fort Pierce to a building constructed and maintained by
St. Lucie County for the sole purpose of housing this edu-
cational attraction. At the Ecosystems Exhibit, visitors are
invited to explore six Florida marine habitats and learn
about the complexity and importance of these ecosystems
(Figure 10). The largest aquarium houses a Caribbean
coral reef display. Additional aquaria depict a seagrass bed,
red mangrove coastline, estuarine and nearshore habitats,
and a deep-water Oculina coral reef. Smaller aquarium
displays highlight single species of interest, and a touch
tank offers visitors personal interaction with various lo-
cal invertebrates, such as horseshoe crabs, sea urchins, sea
cucumbers, and peppermint shrimp.

This public aquarium is unlike any other, providing an
accurate representation of the underwater worlds of the In-
dian River Lagoon and Atlantic Ocean. Although these wa-
ters are a common sight to many coastal Florida residents,
few have experienced the unsurpassed diversity of life just
below their surface. By highlighting this diversity and dis-
playing local ecosystems as complex communities of organ-
isms interacting in their environments, the Exhibit aims to
provide the public with a better understanding of the fragile
coastal ecosystems of the Indian River Lagoon and the sur-
rounding area, including the impacts people have on them.

The Smithsonian Marine Ecosystems Exhibit is a
field trip destination for thousands of school-aged chil-
dren each year (Figure 11). Although some choose a
self-guided visit, most participate in one of several struc-
tured programs facilitated by Education staff members.

Program options are age- and grade-appropriate and are
structured in compliance with Florida’s Sunshine State
Standards. Activities include scavenger hunts, water qual-
ity experiments, food web and energy transfer studies,
simulated benthic sampling, and field experiences in the
Indian River Lagoon.

In 2005, education staff at the Exhibit began offer-
ing community and visitor programs. Ranging from in-
formative breakfast programs to sleepovers and summer
camps, the new programs target traditional visitor groups
in new ways, providing more focused and in-depth learn-
ing experiences for those interested in taking advantage
of the many resources the Smithsonian has to offer. The
enthusiastic response from the community has resulted in
continual additions to the events calendar.

In addition to being a physical destination for the
local community, the Ecosystems Exhibit has also estab-
lished itself as a valuable resource for local schools and
community organizations that do not have the means to
travel. Education staff provides classroom outreach pro-
grams, bringing the wonders of the underwater world to
hundreds of students each school year. Education staff
members have also developed Resource Loan Kits for area
teachers to borrow for in-classroom use for a two-week
time period. The Exhibit website also hosts three webcams
that provide live feeds to three of the Exhibit’s displays.
Online visitors have alternate, unparalleled views into
the seagrass and coral reef model ecosystems, as well as
through the lens of a laboratory microscope. Future plans
include the development of online curricula and activities
based on observations made via the webcams.

LOOKING TO THE FUTURE

During the past 37 years the Smithsonian Marine Sta-
tion at Fort Pierce has developed a strong, broadly based
program in marine biodiversity research focusing on sys-
tematics, ecology, and life histories of marine organisms.
With nearly four decades of research along the IRL, the
Smithsonian Marine Station has been able to establish
long-term and intensive research projects that are valuable
in understanding and assessing marine biodiversity as well
as the changes in biodiversity occurring on a global scale.
As a result of its excellent location, modern facilities, and
experienced staff, the Marine Station is well positioned
to continue to address important research topics includ-
ing global climate change, invasive species, harmful algal
blooms, systematics, larval ecology, and evolutionary de-
velopmental biology.

NUMBER 38 ¢ 39

FIGURE 10. The coral reef at the Smithsonian Marine Ecosystems Exhibit.

The Smithsonian Marine Ecosystems Exhibit makes
the work of Smithsonian marine researchers accessible to a
broad, non-scientific audience. The living displays capture
the dynamic quality of natural ecosystems, and the educa-
tional offerings are a reflection of the same. Programs, dis-
plays, and live exhibits are constantly changing, evolving,
and taking on new life, providing a foundation to ensure
the Exhibit’s future in an ever-changing community.

ACKNOWLEDGMENTS

The authors thank the Hunterdon Oceanographic Re-
search Fund and the Florida Fish and Wildlife Conserva-
tion Commission (agreement numbers 05011 and 08017)
for financial support. We are grateful to the many Smith-

sonian scientists who provided input about their past and
ongoing research activities at SMS. Mary Rice provided
many helpful comments that greatly improved this manu-
script. This is Smithsonian Marine Station at Fort Pierce
Contribution No. 764.

LITERATURE CITED

Buzas, M. A., and L. C. Hayek. 2005. On Richness and Evenness within
and between Communities. Paleobiology, 31:199-220.

Buzas, M. A., L. C. Hayek, S. A. Reed, and J. A. Jett. 2002. Foraminif-
eral Densities over Five Years in the Indian River Lagoon, Florida:
A Model of Pulsating Patches. Journal of Foraminiferal Research,
32:68-93.

Faust, M. A., and P. A. Tester. 2004. “Harmful Dinoflagellates in the
Gulf Stream and Atlantic Barrier Coral Reef, Belize.” In Harmful

40

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 11. Top: A young girl takes a closer look at the inhabitants of the seagrass ecosystem.
Bottom: Excited children view the nearshore reef ecosystem.

Algae 2002, ed. K. A. Steidinger, J. H. Landsberg, C. R. Tomas,
and G. A. Vargo, pp. 326-328. Florida Fish and Wildlife Con-
servation Commission, Florida Institute of Oceanography, and
Intergovernmental Oceanographic Commission of UNESCO, St.
Petersburg, Fla.

Feller, I. C., C. E. Lovelock, and K. L. McKee. 2007. Nutrient Addition
Differentially Affects Ecological Processes of Avicennia germinans
in Nitrogen Versus Phosphorus Limited Mangrove Ecosystems.
Ecosystems, 10:347-359.

Feller, I. C., D. F Whigham, K. L. McKee, and C. E. Lovelock. 2003.
Nitrogen Limitation of Growth and Nutrient Dynamics in a Dis-
turbed Mangrove Forest, Indian River Lagoon, Florida. Oecologia
(Berlin), 134:405-414.

Hayek, L. C., and M. A. Buzas. 2006. The Martyrdom of St. Lucie: Deci-
mation of a Meiofauna. Bulletin of Marine Science, 79:341-352.

Hendler, G., J. E. Miller, D. L. Pawson, and P. M. Kier. 1995. Starfish,
Sea Urchins and Allies: Echinoderms of Florida and the Caribbean.
Washington, D.C.: Smithsonian Institution Press.

Littler, D. S., and M. M. Littler. 2000. Caribbean Reef Plants. Washing-
ton, D.C.: Offshore Graphics, Inc.

Littler, D. S., M. M. Littler, and M. D. Hanisak. 2008. Submersed Plants
of the Indian River Lagoon: A Floristic Inventory and Field Guide.
Washington, D.C.: Offshore Graphics, Inc.

Lovelock, C. E., and I. C. Feller. 2003. Photosynthetic Performance and
Resource Utilization of Two Mangrove Species Coexisting in a Hy-
persaline Scrub Forest. Oecologia (Berlin), 134:455—-462.

Lovelock, C. E., I. C. Feller, M. C. Ball, B. M. J. Engelbrecht, and
M. L. Ewe. 2006. Differences in Plant Function in Phosphorus-
and Nitrogen-Limited Mangrove Ecosystems. New Phytologist,
172:514-522.

Ma, H., R. M. Overstreet, J. H. Sniezek, M. Solangi, and D. W. Coats.
2006. Two New Species of Symbiotic Ciliates from the Respi-
ratory Tract of Cetaceans with Establishment of a New Genus
Planilamina n. gen. (Dysteriida, Kyaroikeidae). Journal of Eu-
karyotic Microbiology, 53:407-419.

Maslakova, S. A., M. Q. Martindale, and J. L. Norenburg. 2004. The
Vestigial Prototroch in a Basal Nemertean Carinoma tremapho-
ros (Nemertea; Palaeonemertea). Evolution and Development, 6:
219-226.

NUMBER 38 e¢ 41

Maslakova, S. A., and J. L. Norenburg. 2008. Revision of the Smiling
Worm, Genus Prosorhochmus Keferstein, 1862 and Descriptions of
a New Species, Prosorhochmus belizeanus sp. nov. (Prosorhochmi-
dae, Hoplonemertea, Nemertea) from Florida and Belize. Journal of
Natural History, 42:1219-1260.

Paul, V. J., K. E. Arthur, R. Ritson-Williams, C. Ross, and K. Sharp.
2007. Chemical Defenses: From Compounds to Communities. Bio-
logical Bulletin, 213:226-251.

Paul, V. J., R. W. Thacker, K. Banks, and S. Golubic. 2005. Benthic Cya-
nobacterial Bloom Impacts the Reefs of South Florida (Broward
County, USA). Coral Reefs, 24:693-697.

Rice, M. E. 1981. Larvae Adrift: Patterns and Problems in Life Histo-
ries of Sipunculans. Past-Presidential Address. American Zoologist,
21:605-619.

. 1989. “Comparative Observations of Gametes, Fertilization,
and Maturation in Sipunculans.” In Reproduction, Genetics and
Distributions of Marine Organisms, 23rd European Marine Biol-
ogy Symposium, ed. J. S. Ryland and P. A. Tyler, pp. 167-182.
Fredensborg, Denmark: Olsen & Olsen.

Rice, M.E., J. Piraino, and H. FE. Reichardt. 1983. Observations on the
Ecology and Reproduction of the Sipunculan Phascolion cryptus in
the Indian River Lagoon. Florida Scientist, 46:382-396.

Ritson-Williams, R., S. N. Arnold, N. D. Fogarty, R. S. Steneck, M. Ver-
meij, and V. J. Paul. 2009. “New Perspectives on Ecological Mecha-
nisms Affecting Coral Recruitment on Reefs.” In Proceedings of
the Smithsonian Marine Science Symposium, ed. M. A. Lang, I. G.
Macintyre, and K. Ritzler, pp. 437-457. Smithsonian Contribu-
tions to the Marine Sciences, No. 38. Washington, D.C.: Smithson-
ian Institution Scholarly Press.

Snoeyenbos-West, O. L. O., J. Cole, A. Campbell, D. W. Coats, and L. A.
Katz. 2004. Molecular Phylogeny of Phyllopharyngean Ciliates
and Their Group I Introns. Journal of Eukaryotic Microbiology,
51:441-450.

Schulze, A., and M. E. Rice. 2009. Musculature in Sipunculan Worms:
Ontogeny and Ancestral States. Evolution and Development,
11(1):97-108.

Thollesson, M., and J. L. Norenburg. 2003. Ribbon Worm Relation-
ships-A Phylogeny of the Phylum Nemertea. Proceedings of the
Royal Society of London, B Biological Sciences 270:407-415.

aa

I 5, “N i
i
{
S sn Aj Ss
r a
oe .
E ey , 1
\ ¥ ’ I u
v { he
7 i ‘
: — : ~ a
i g =
= b
1
lf i
1
Ey r ' ‘
v
u t
1 7
i ‘ 1 a
f
| i
1
i
Way
(
i
t 2 n
o j
i
s i
. 7
i ’ ;
f | - y
om °
. Z is
R _
i ~ rs 7 “

} es 3 ee oo

Caribbean Coral Reef Ecosystems:
Thirty-Five Years of Smithsonian
Marine Science in Belize

Klaus Riitzler

Klaus Ruetzler, Department of Invertebrate Zool-
ogy, National Museum of Natural History, Smith-
sonian Institution, Washington, D.C. 20560-
0163, USA (ruetzler@si.edu). Manuscript received
9 June 2008; accepted 20 April 2009.

ABSTRACT. With foresight and tenacity, Smithsonian Institution marine scientists have
devoted more than three decades to understanding and preserving one of the planet’s vital
natural resources: the coral-reef ecosystem. In the late 1960s marine scientists from the
Smithsonian National Museum of Natural History, Washington, founded a long-term
Caribbean coral-reef field program, now known as Caribbean Coral Reef Ecosystems
(CCRE), to investigate the biodiversity, community structure and dynamics, and envi-
ronmental processes that control this ecosystem. Its core group of botanists, zoologists,
paleobiologists, and geologists found an ideal study site—with high biological diversity,
significant geological features, and minimal anthropogenic disturbance—on the barrier
reef off Southern Belize, and in 1972 established a field station on one of its tiny islands,
Carrie Bow Cay. Within a radius of less than 2 km lie a great variety of richly populated
habitats, from mangrove to fore-reef. The Belize mainland and three offshore atolls are
within easy reach by small boat. Each year, up to 120 Smithsonian staff and associated sci-
entists, with assisting students and technicians, study the area’s reefs, nearby mangroves,
and seagrass meadows. Their “whole-organism” expertise encompasses many fields of
biology—systematics, evolution, paleobiology, ecology, and ecophysiology—supported
by molecular techniques to expand upon traditional morphological taxonomic analyses.
An oceanographic-meteorological monitoring station on Carrie Bow Cay records envi-
ronmental data, now available on the World Wide Web, and monitors the productivity
of selected reef, mangrove, and seagrass communities. Field research is complemented by
the large resources of the Smithsonian home base. Today, the CCRE program is a mem-
ber of the Smithsonian’s Marine Science Network, which includes coastal laboratories in
Panama, Florida, and Maryland. In these and other respects—CCRE now has more than
800 papers in print—the program’s accomplishments are indeed impressive.

INTRODUCTION

How does one summarize in a few pages 35 years of research on a com-
plex ecosystem by more than 200 investigators? Clearly, it cannot be done in a
complete fashion. With apologies for any omissions, I present this review as a
tribute to every single participant in the Caribbean Coral Reef Ecosystems pro-
gram (CCRE) dating back to the late 1960s, when it was titled Investigations
of Marine Shallow-Water Ecosystems (IMSWE). The founders’ unifying objec-
tive was to apply a multidisciplinary, long-term team approach to studies of
marine shallow-water animals and plants, and to examine their interactions in

44 e¢ SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

their environment—today as well as in the past—for infor-
mation on the determinants of community structure and
evolutionary change. A coral-reef ecosystem, we agreed,
is the most extensive and biologically productive shallow-
water community on Earth and thus would fully meet our
purposes. After conducting literature reviews and several
joint surveys throughout the Caribbean, we chose Belize
(then British Honduras) as the program’s locale because
of its pristine environment and high diversity of organisms
and reef types.

PROGRAM FOUNDERS AND OBJECTIVES

Coral reefs are among the true wonders of the world:
they cover 190 million km? of the world’s ocean floors,
are tremendously productive, protect tropical continental
coasts and islands from the eroding forces of the oceans,
supply humans with large quantities of high-quality pro-
tein, and are a unique recreational resource. For all their
aesthetic and economic value, coral reefs remain invisible
to most people unless they live close to tropical coasts
or engage in skin or scuba diving. Without such contact,
many are insensitive to the catastrophic effects of pollu-
tion and uncontrolled land development, which can rap-
idly decimate entire communities and thus their benefits,
or are unaware of the effects of natural phenomena such
as global warming and acid rain.

Fortunately, the unique composition of the Smithson-
ian Institution, with specialists in many disciplines of the
life and earth sciences, provided a substantial number of
researchers interested in reefs and willing to team up for
the common good of an integrated study. Some experts
from other institutions were expected to join for specific
tasks. Our original team, all staff of the Natural Museum
of Natural History (NMNH), consisted of Walter H. Adey,
Department of Paleobiology, a specialist in fossil and mod-
ern coralline algae; Ian G. Macintyre, Paleobiology, a car-
bonate sedimentologist studying calcification, reef-building
organisms, and reef evolution; Arthur L. Dahl, Botany, an
algal ecologist; Mary E. Rice, Invertebrate Zoology, an ex-
pert in sipunculan worm systematics and developmental
biology; Tom Waller, Paleobiology, a malacologist focus-
ing on the systematics and distribution of scallops in time
and space; Arnfried Antonius, a postdoctoral fellow in In-
vertebrate Zoology working on stony corals; and myself,
Invertebrate Zoology, a sponge biologist with an interest
in reef ecology and bioerosion. We were joined in our early
search for the optimal research site by David R. Stoddart,
a geographer at the University of Cambridge, England;
Porter M. Kier, an actuopaleontologist (later a director of

the Natural History Museum) looking for modern clues
to interpreting fossil echinoderm assemblages; Richard S.
(Father Joe) Houbrick, a former priest turned malacolo-
gist and working at the Smithsonian Marine Sorting Cen-
ter; Ernst Kirsteuer, an invertebrate zoologist specializing
in nemertine worms at the American Museum of Natural
History, New York; and Fred Hotchkiss, a postdoctoral
fellow in Invertebrate Zoology specializing in ophiuroid
echinoderms. David Stoddart was a particular asset be-
cause he had a wealth of research experience with the dis-
tribution, geomorphology, terrestrial botany, and dynam-
ics of Belizean islands (cays), having been a member of the
1959 Cambridge Expedition to British Honduras (Carr
and Thorpe, 1961) and participant in numerous post-
Hurricane Hattie (1961) surveys (Stoddart et al., 1982).

Our main objective was to study the historical and pres-
ent conditions in a well-developed coral reef far removed
from the stressful impacts of an industrial society with a
view to compiling baseline data on how an established reef
community adjusts to natural environmental parameters.
These data would include information on diagenetic altera-
tion of the reef structure, as revealed in drill cores. With
the resulting information, we hoped to develop a predic-
tive model of the impact of anthropogenic stress. As we
quickly discovered, most previous reef studies consisted of
short-term surveys during large-scale expeditions, with su-
perficial sampling during a single season; moreover, many
of the reports on reef fauna and flora had been prepared
by specialists who had never observed the organisms and
processes in the field. A complex ecosystem such as a coral
reef obviously required a more rigorous, long-term, and
multidisciplinary approach if we had any hope of deter-
mining the relative importance of diversity, biomass, en-
ergy flow, and environment to community function.

Carrie Bow Cay, BASE OF A New MARINE FIELD STATION

The team chose a Caribbean reef site for several rea-
sons: most of us had already worked in that area, and it
would be “close to home,” would permit comparison with
the already-stressed reefs of Florida, and would minimize
travel time and cost. Equally important, to be sure, was
the fact that all the characteristic reef types and zones were
within workable distance, reef growth was vigorous with
a good geological record of past development, and the lo-
cale was remote from terrestrial and human influences.

Moving ahead with small grant awards from Smith-
sonian Institution endowments, we purchased an inflatable
boat with an outboard motor, dive tanks, a small compres-
sor, and tents for reconnaissance trips across the Carib-

bean. In another step forward, the U.S. National Science
Foundation provided support for a planning meeting on
Glovers Reef atoll, Belize, attended by representatives of
some 40 academic institutions. We envisioned starting up
the program there and eventually conducting comparative
studies on an Indo-Pacific atoll. As fate would have it, the
proposal emanating from this meeting was not funded.

Returning to Glovers Reef in February 1972 to retrieve
our IMSWE equipment from storage, Arnfried Antonius and
I discovered a small unoccupied islet with three shuttered
buildings on the southern Belize barrier reef. Its name, we
learned, was Carrie Bow Cay (16°48'N, 88°05’W; origi-
nally spelled Caye), and it was owned by the family oper-
ating the Pelican Beach Motel in Dangriga (Stann Creek
District), a small town on the mainland (Figures 1, 2). To
our happy surprise, this reef tract met all our scientific re-
quirements, studies there would garner generous coopera-
tion from Belize’s Fisheries Department, and excellent lo-
cal logistical support would be available. The motel, now
called Pelican Beach Resort, was owned and operated by
Henry Bowman, Jr., and his wife Alice. After negotiating
storage for our equipment, we initiated a contract to lease
part of Carrie Bow Cay, including the two smaller cot-
tages, for a three-month research period that spring and
summer.

Carrie Bow Cay was owned by Henry Junior’s father,
Henry T. A. Bowman, a third-generation descendant of
Scottish settlers. The enterprising Henry senior was a cit-
rus grower, businessman, and one-time legislator who had
bought the island from his father as a vacation retreat,
changed its name (from Ellen or Bird Caye) to Carrie for
his wife, and put up an old farmhouse that he had bought
on the mainland and carried out to the cay in sections.
With his love of fishing, “Sir Henry” (as I referred to him
when we became friends) and some of his relatives (daugh-
ter Norma and daughter-in-law Alice, in particular) de-
veloped a keen interest in the sea and the reef’s myriad
animals and plants. This interest persuaded him to allow
us unrestricted access to most of his island and provided
many opportunities to share our observations over drinks
during the sunset hour.

A great concern for both of us then, and for all of
CCRE today, was the rate of coastal erosion, mainly the
consequence of frequent hurricanes, which had reduced
the size of the island from 2 acres (0.8 ha) in the 1940s to a
little more than half that in the 1970s. In his delightful au-
tobiography (Bowman, 1979), Henry took the blame him-
self, admitting that he had carelessly removed mangrove
trees “that build and bind these cayes.” At the same time,
he did make a significant contribution to the island’s mor-

NUMBER 38 e¢ 45

phology: in 1942 he built a 27 m long concrete boat dock
on the leeward (lagoon) side. It has remained unchanged
to this day and has served as a reference in our mapping of
the geomorphology and communities nearby.

Since then, both Henrys have passed away, but their
naturalist spirit lives on. Therese Rath, who is Junior and
Alice’s daughter (Sir Henry’s granddaughter), runs Pelican
Beach with her mother and continues to offer us logistical
support on Carrie Bow. Therese’s husband, Tony Rath—
one of our early volunteer station managers who moved to
Dangriga from Minnesota two decades ago—is a success-
ful nature photographer and runs the premier web design
business in Belize; he still helps us out as a naturalist ad-
viser and provides documentary photography.

FACILITIES AT START-UP

Between 1972 and 1975, our team operated on a
shoestring. The relatively small grants available to us
(the Smithsonian has no direct access to National Science
Foundation funding) kept the field station open for no
more than four months a year and supported up to 25
scientists and assistants per season. Our facilities consisted
of a small three-room building with a tin roof to the south
of the main house (it contained our lab, living quarters for
two, and a kitchen); a 4 m2 shed that could house two; and
a tent, when needed, that could accommodate up to six.
The dive compressor and a small generator were installed
in improvised shelters.

Our shower consisted of a spray-head on a pipe
screwed into the bottom of a huge wooden vat that col-
lected rainwater running off the roof of the main build-
ing. To preserve decency, there was an enclosure (its sign
read: “Save Water, Shower with a Friend”). The toilets for
all island occupants were two outhouses accessed from a
wooden pier extending over the reef flat to the island’s
east. The cabin’s seats were rough-cut planks with holes.
However, its window opening allowed a spectacular view
of the reef flat, barrier reef, and unobstructed horizon,
with pelicans and 1 m long parrotfish jumping and feed-
ing in the foreground.

After getting used to us, Sir Henry turned his chil-
dren’s “museum” in the main house into a station man-
ager’s quarters by adding some wood siding for walls and
a door. It was a roofed-over corner of the house’s wide,
upper-level porch, where Norma and Alice had kept and
displayed an assortment of shells, corals, quirky drift-
wood, and stranded and mummified algae, invertebrates,
and fishes. Working for a museum ourselves, we found
that a quaint step forward.

Mexico

“* Columbus Cay,

ot @ Mosquito Cay p 7-00N
+: @Sandfly Cay a r
Hutson Cay Grose Cay 3
Belize City ; 4 (f ee aa

-- SGarbuttCay . .”
o

i ; i Colurtibus Reef
Dangriga iz )

Tobacco Range

Guatemala

‘® Tobacco Cay

Coco Plum Cay) ;

Ragged Cay‘e:

Blue Ground “

40; South Water Cay
~~ Sittee Point

“s; Carrie Bow Cay
<>. Curlew Bank

Stewart Cay-g Be : Patch Reefs
i” Wee Wee Cay © Sand Bores

Q Fe

ettttte
'*

rath at!

Oo Gf

8g See were”

Spruce Cayzy ...,

aier*
.

Douglas Cayig ee

ar * South Cut

Elbow Cays* :
Pelican Cays Bo}
Vo

es

—fY sees Point

Quamino Cays ¢

. .
ae
eee

Lagéan Cays:%"

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

@ False Cay

False Point Crawl cr,

Bakers Rendezvous ,

°
Gladden Cays ”

FIGURE 1. Map of research area in coastal Belize, Central America. The barrier reef and other reef tracts appear in pink.

NUMBER 38 e

FIGURE 2. The original Investigations of Marine Shallow-Water Ecosystems (IMSWE) survey team, the Belize barrier reef, and Carrie Bow
facilities in the early 1970s. Upper left: The team included (left to right) Walter Adey, Arthur Dahl, Tom Waller, Klaus Ruetzler, and Arnfried
Antonius (missing from the picture are Porter Kier, Ian Macintyre, and Mary Rice). Upper right: Belize barrier reef looking south, with South
Water Cay in foreground and Carrie Bow Cay near center. Center left: Carrie Bow Cay looking southwest, with ocean-side reef flat in fore-
ground. Center right: Carrie Bow facilities looking south, the Bowmans’ “Big House” to the right and our lab building in the center. Lower left:
Photographer Kjell Sandved working in the aquarium area. Lower right: Scientists in the lab are Anne Cohen (left) and Jim Thomas.

48 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

For the “lab,” we removed some of the cottage parti-
tions that originally defined the bedroom space for the
parents and three children, built a long bench along the
oceanfront wall with a supply cabinet and photo table
opposite it, and weatherproofed windows by insert-
ing acrylic panes in the lower half to allow the wooden
shutters to remain open under most weather conditions
and thus let in more light. The sun gave us light for mi-
croscopy and photography, with a couple of small gaso-
line-driven generators doing the job whenever needed.
We brought microscopes, cameras, some portable in-
struments, labware, and boating and dive gear from
home but improvised on most additional laboratory or
field needs. Our original IMSWE inflatable boat and 25
horsepower outboard engine were still in working order,
supplemented by a similar inflatable recently added. A
shortwave radio provided contact with Pelican Beach in
Dangriga for ordering supplies, brought out once a week.
A local cook prepared our meals and lived in a room
under the big house; next to her was the simple residence
of a native fisherman who served as caretaker and watch-
man, particularly when the island was deserted during
the off-season. Our station manager was usually one of
us, or one of our enthusiastic young museum technicians,
or some other volunteer with technical know-how. When
the lab was closed, all valuables were stored in high
places (in case of storm floods), the windows shuttered,
and the door padlocked and nailed to its frame. (“This,”
the locals said, “does not keep the crooks out but keeps
the honest people honest.”) During hurricane season, all
major equipment was taken to Dangriga and stored in
the Maya Hut behind Pelican Beach.

ANALYZING A COMPLEX ECOSYSTEM

THE EARLY YEARS

The program’s first targets were to map the reefs and
other habitats near the field station, including Carrie Bow
Cay itself, and to identify the key organisms in the commu-
nities (Figures 3, 4). Because the north—south-oriented bar-
rier reef is the dominant feature separating the lagoon from
open ocean, we established a transect perpendicular to its
trend, originating well inside the lagoon in a seagrass bed
2 m deep; it then crossed the barrier-reef crest some 150 m
north of Carrie Bow and extended due east across the reef
and down the fore-reef slope to a depth of 30 m. This tran-
sect would become the baseline reference for all our topo-
graphic studies and future observations and experiments.

We also tried to develop some standard methods of
sampling, extracting interstitial organisms, and deter-
mining biomass (Dahl, 1973; Macintyre, 1975; Ritzler,
1978a). Because of the complexity of the reef framework
(with its three-dimensional structure) and the diversity
and size range of its inhabitants (which varied by at least
three orders of magnitude), we had to modify many of the
commonly used ecological methods to ensure compatible
results.

Unable to employ self-contained recording instru-
ments to monitor important environmental parameters,
we established a manual routine for taking tide and tem-
perature readings, and for observing solar radiation, wind
speed and direction, precipitation, humidity, cloud cover,
wave action, and turbidity with simple handheld devices.
For specific projects, we measured salinity, oxygen con-
centration, pH values, water current speed, and submarine
daylight with off-the-shelf instruments for which we built
waterproof housings when necessary. These data, along
with the first reef maps and results from transect surveys,
were summarized in our 1975 progress report and distrib-
uted to program participants and supporters.

Many colleagues helped identify key organisms and
determine biomass and spatial and temporal distribution
(Adey and Macintyre, 1973; Kier, 1975; Pawson, 1976).
Early on, we discovered unexpectedly high numbers of
new species in almost all taxa, which was surprising be-
cause the Caribbean Sea is generally considered among the
best-studied oceanic regions of the world. Using in situ
methods, we identified and quantified environmental pa-
rameters such as light, water flow, and sediments making
up the “microclimate” of particular organisms (Graus and
Macintyre, 1976). We also investigated important associa-
tions and interactions between organisms, such as symbio-
ses and space competition, predation, diets, and behavioral
patterns. In addition, we measured primary production of
benthic macroalgae and symbiotic microalgae, and growth
and reproduction rates of reef-forming organisms, the first
steps toward determining key metabolic processes (Mac-
intyre et al., 1974).

Some geologists and biologists collaborated in the
study of geological processes such as the construction and
destruction of the coral-reef framework and the calcifica-
tion rates of corals, coralline algae, and other bioherms.
Others studied physical and biological erosion and sedi-
ment production rates (Rutzler, 1975), sediment sorting
and colonization by meiofauna, and processes of recemen-
tation. Ian Macintyre initiated a drilling project with col-
leagues from the U.S. Geological Survey’s Energy Resource
Division to learn about the historical development of the

NUMBER 38

FIGURE 3. Some early program participants. Clockwise, from top left: Ian Macintyre about to enter the submarine Colombus Cay Cave;

Arnfried Antonius setting up time-lapse camera for study of black band coral disease; Klaus Ruetzler catching the evening sun for a microscope

examination; view of entrance of station; Mike Carpenter fixing an underwater viewer; Joan Ferraris measuring oxygen consumption of benthic
community on the Carrie Bow reef flat; and Ilka Feller surveying red-mangrove insects at Twin Cays.

50 e¢ SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 4. Principal marine habitats near Carrie Bow Cay. Top row: left, patch reef near the south tip of the island; right, barrier reef
crest composed of elkhorn and fire coral (Acropora, Millepora). Middle row: left, outer fore-reef with corals, sponges, and gorgonians;

center, small cave in the fore-reef framework; right, seagrass stand in the barrier-reef lagoon. Bottom row: left, red mangrove at Twin
Cays, with Carrie Bow Cay in the background; center, gorgonian and barrel-sponge community on the outer fore-reef; right, diver on
the fore-reef slope.

barrier reef. Cores from a series of holes yielded informa-
tion on past community patterns and successions as well as
the distribution of contemporaneous submarine cements
within the reef structure (Macintyre et al., 1981).

Detailed maps and inventories of terrestrial plants on
several Belizean cays, including Carrie Bow, from 1960,
1962, and 1972 (see Stoddart et al., 1982) aided in our
observation of morphological and floral changes, particu-
larly in relationship to the frequent hurricanes in the re-
gion. In 1974, only two years into our presence on Carrie
Bow, Hurricane Fifi, which destroyed large coastal areas of
Honduras, hit the barrier reef just south of our island. Al-
though our reef habitats experienced only minor changes,
primarily local breakdown of the framework and accumu-
lation of rubble, most terrestrial life on Carrie Bow, par-
ticularly vascular plants, was killed by flooding—except
for coconut trees, which suffered about a 20% loss (16
trees)—and there was severe coastal erosion. Upon remap-
ping the island in the wake of this event, we noted some
redepositing of beach sand and recolonization by plants
through drift and windblown seeds.

GAINING MOMENTUM

Our venture took a significant step forward with the
award, in 1975, of an annual grant by the Exxon Cor-
poration from the company’s public relations budget for
Central America. Although relatively small, the funds
nearly doubled our support and had no strings attached,
except they were to be dedicated to Caribbean coral reef
research. Added to this welcome development was a new
and beneficial relationship with Captain Graham Thomas
of the Royal Signals Detachment in Belize, a helicopter
pilot detailed to support the training of British forces in
jungle environments. Graham was able to equip his heli-
copter with an aerial camera and take vertical pictures of
the Carrie Bow reefs that provided excellent photo cover-
age for a detailed mapping of the area’s reef structures at
a scale of 1:800 and to a depth of 10 m. This informa-
tion constituted enormous progress over available nauti-
cal charts (with a scale of 1:125,000) that dated back to
British surveys in the 1830s and were only partly updated
in the 1940s. Even greater resolution in aerial mapping
(but at the expense of areal coverage) was achieved by in-
troducing a helium weather balloon equipped with a re-
motely operated camera. This technique (Riitzler, 1978b),
like several others devised by our team, was documented
in a volume on coral reef research methodology sponsored
by the United Nations Educational, Scientific and Cultural
Organization (UNESCO) (Stoddart and Johannes, 1978).

NUMBER 38 e 51

In 1977, in association with the Third International
Symposium on Coral Reefs in Miami, several of our team
organized a well-received field trip to Belize, highlighted
by a detailed field guide based on maps, transect data, and
aerial and underwater habitat photographs emanating
from the program (Miller and Macintyre, 1977). In short
order, the program launched several new projects (Fig-
ure 5) to investigate the fate of siliceous skeletons in the
calcium carbonate environment of the reef (Riitzler and
Macintyre, 1978), the feeding behavior of scyphomedu-
sae (Larson, 1979), and the systematics of the unexpect-
edly diverse ostracod crustaceans (Kornicker and Cohen,
1978). Other innovative and pioneering work focused on
the fine structure of bivalve anatomy as revealed through
scanning electron microscopy (Waller, 1980) and, with
collaborators from the Scripps Institution of Oceanog-
raphy, on the chemistry of marine plant and invertebrate
secondary metabolites that showed promise as antibiotics
or other therapeutical substances (Kokke et al., 1979). By
the end of 1978, more than 50 papers had been published
or were in press to document the biology, ecology, and
geology of the Belize barrier reef in the vicinity of Carrie
Bow Cay.

The program’s success appeared to be short lived,
however: in late September 1978, Hurricane Greta passed
across the Belize barrier reef just 6 km north of Carrie
Bow. Four lives were lost in Dangriga, the citrus harvest in
the valley to the west was destroyed, and there was heavy
beach erosion at Pelican Beach Motel. Although no mem-
bers of our group were on location because we had closed
down for the season, part of our equipment was damaged
when ocean storm surge and rain flooded the storage area.
Storm waves from the east and strong backlashing winds
from the northwest caused severe erosion of beach sand
on Carrie Bow and wiped out some 30 coconut trees, the
small house, and the outhouses. The ocean-side wall of the
laboratory building also caved in, exposing equipment and
supplies to saltwater spray. Following the storm, visibility
in the usually very clear ocean water remained at less than
3 m for two weeks, most elkhorn (Acropora palmata) and
fire coral (Millepora complanata) near the reef crest was
reduced to rubble, and the salinity in the lagoon dropped
from the usual 35%o to 25%o.

Despite this setback, we decided to press our team
and collaborators to complete work in progress and pre-
pare a state-of-the-art summary of our accomplishments.
The resulting volume (Riitzler and Macintyre, 1982), later
known as the Blue Book (for the color of the hard cover),
became a platform for the next phase of investigations, as
well as for raising funds. The first section presented a

52 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 5. A few examples of research activities of the 1970s to 1980s. Clockwise, from top left: boat operator Frank (Pelican Beach Resort)
helps Ron Larson to lower plankton net from the stern of the boat; Ian Macintyre’s group drill-coring down a sand groove on the fore-reef;
Joan Ferraris (left) tending incubation chambers to measure temperature-salinity tolerance of invertebrates and Sara Lewis preparing aquaria
for fish herbivory experiments; Sara measuring algal abundance in a quadrat frame on the reef flat; Klaus Ruetzler retrieving trace paper from
tide recorder on Carrie Bow dock; Mark and Diane Littler’s team assessing effects of nutrients on algal growth.

detailed overview of the physical and biological environ-
ment of our study site—its habitats and community struc-
ture, geological history, tides, water currents, climate, and
terrestrial conditions—compiled by Jan Macintyre and
myself and various outside collaborators, including Bjorn
Kjerve (University of South Carolina, Colombia), Joan Fer-
raris (Mount Desert Island Biological Laboratory, Maine),
and Eugene Shinn (U.S. Geological Survey, Miami). The
next section focused on the benthic and planktonic com-
munities—the carbonate microborers, micro- and macro-
benthos, zooplankton, and the populations of a large
submarine cave at nearby Columbus Cay—and their pro-
ductivity. The principal collaborators were Joan Ferraris,
Paul Hargraves (University of Rhode Island, Kingston),
Jeffrey May (Rice University, Houston), and David Young
(Department of the Navy, Mississippi). A section on biodi-
versity included many important groups of reef organisms,
notably the algae and seagrasses (James Norris, NMNH),
hydroids (Barry Spracklin, University of New Hampshire,
Durham), medusae (Ronald Larson, a former NMNH
technician who moved on to the University of Victoria,
Colombia), stony corals (Stephen Cairns, NMNH), oc-
tocorals (Katie Muzik, postdoctoral fellow, NMNH), si-
punculan worms (Mary Rice, NMNH), crustaceans and
pycnogonids (Brian Kensley, Allan Child, NMNH), and
echinoderms ( Frederick Hotchkiss, postdoctoral fellow,
NMNH;; Bradford Macurda, Jr., University of Michigan,
Ann Arbor). The most unusual discovery was that chiron-
omid insect larvae, caught in emergence traps, are part of
the offshore benthic community and live in fore-reef sand
bottoms to depths of 30 m (Gernot Bretschko, Biological
Station Lunz, Austria). A section on species interactions
and responses to the environment addressed chemical de-
fense in algae (James Norris), the life history and ecology
of cnidarians (Ronald Larsen), growth patterns of reef
corals (Richard Graus, NMNH), sponge-zoanthid asso-
ciations (Sara Lewis, Duke University, Durham), bivalve
larval settlement (Thomas Waller, NMNH), and resource
partitioning in chaenopsid, coral-associated fishes (David
Greenfield, Field Museum of Natural History, Chicago).
The concluding chapter puts Carrie Bow Cay and its reefs
in the larger context of the Belize barrier reef complex
(contributed by Randolf Burke, North Dakota Geological
Survey, Grand Forks, and David Stoddart).

Having overcome many of the start-up problems,
including setbacks caused by the hurricanes, we forged
ahead in the new decade with increasing productivity and
innovation. CCRE members authored a number of impor-
tant monographs and other reports on reef biodiversity.
We started a series of papers on the fungi (Kohlmeyer,

NUMBER 38 °¢ 53

1984) and algae (Littler and Littler, 1985); prepared anal-
yses of several large crustacean groups, including parasitic
copepods on fishes (Cressey, 1981), decapods (Kensley
and Gore, 1981), isopods (Kensley, 1984), and amphipods
(Thomas and Barnard, 1983); and published the first sur-
vey of local moss animals, bryozoans, by a colleague then
at the American Museum of Natural History in New York
(Winston, 1984).

We also conducted a series of day and night plankton
tows over the fore-reef and over lagoon seagrass bottoms,
which were surprisingly devoid of larval stages of some of
the area’s common animals, such as an assortment of cni-
darians and sponges. We speculated that the larvae might
be swimming close to or within the reef framework, where
our boat-towed nets could not be operated, and decided to
tow or push the plankton nets by hand, while swimming
close to the bottom. Although more successful, this tech-
nique took time and effort to obtain sufficient samples.
Eventually, we hit on the idea of building a stationary
net supported by a frame that could be placed close to or
among the coral heads or branches. For locations without
strong directional currents, we added a waterproof elec-
tric motor with propeller and a flow meter to measure the
volume of water that passed through the net. This setup
ultimately produced excellent samples of great diversity
considerably beyond the composition of plankton tows by
boat (Rutzler et al., 1980).

Having a small budget and intent on disturbing our
study environment as little as possible, we sought creative
field and laboratory techniques that would not require
sophisticated instrumentation or climate control. In keep-
ing with these goals, our colleague Sara Lewis, for one,
completed the experimental fieldwork for her entire dis-
sertation on fish herbivory on the Carrie Bow reef flat,
just a few meters east of the lab building (Lewis, 1986).
Some of our Museum’s phycologists experimented with
the influence of algal growth forms on herbivores at the
same location (Littler et al., 1983). Ecophysiological work
on temperature and salinity tolerance of polychaetes and
other reef invertebrates was accomplished in situ and with
simple, specially designed acrylic incubation chambers
(Ferraris, 1981; Ferraris et al., 1989). Submarine cementa-
tion processes were determined experimentally in the karst
cave habitat of Columbus Cay (Macintyre, 1984). Benefits
of algal symbionts to sponge hosts were explored by in situ
trials on a nearby patch reef (Rutzler, 1981). And, with an
innovative underwater time-lapse camera with strobe light
borrowed from its inventor, Harald Edgerton at the Mas-
sachusetts Institute of Technology, we recorded several
unattended day-night activities on the reef, including the

54 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

nocturnal feeding behavior of basket stars (Astrophyton)
(Hendler, 1982).

During reef surveys in the early phase of our program,
we were already seeing a number of dead or damaged cor-
als with no clear sign of the common physical impacts
related to storms or boat anchors. Our postdoctoral fel-
low Arnfried Antonius pioneered these observations at
Carrie Bow and elsewhere in the Caribbean, as well as
in the Indo-Pacific (Antonius, 1982). One notable feature
of many of these flagging corals was a black line between
live coral tissue and recently dead (white) skeleton. During
collaborative studies, we determined that the black band
consisted of a mat of entangled filamentous cyanobacteria,
with a number of associated microbes, and that the photo-
synthetic bacteria had an appetite for coral tissue, thereby
causing what has been called “black band disease” (Ruitz-
ler et al., 1983).

OCEANIC MANGROVE SWAMPS

Anyone looking through our 1982 “Blue Book” will
notice that mangroves are barely mentioned, except for a
few remarks about Twin Cays, a mangrove island in the
lagoon just over 3 km northwest of Carrie Bow (Figure 6).
This lack should not be taken as a sign of little interest.
CCRE workers have in fact been highly impressed by the
relatively clear (for a swamp) water in the tidal channels
and the rich flora and fauna, particularly the sponges, coy-
ering the stilt roots of red mangrove (Rhizophora mangle).

On an earlier visit to a very similar mangrove is-
land, East Bimini in the Bahamas, I had been so struck
by its subtidal diversity that it seemed an ideal commu-
nity for multidisciplinary study. The “discovery” of Twin
Cays during the early 1980s rekindled this interest, and
coincidentally our Exxon supporters indicated that they
wanted to diversify their generosity in Central America
beyond coral reef research. We therefore submitted a new
proposal to their open competition for the comprehen-
sive study of a Caribbean mangrove ecosystem at Twin
Cays. A factor in our favor was that oil pollution caused
by tanker ballast-water discharge or wrecks was affect-
ing Caribbean beaches and reefs at that time. Indeed, a
colleague and I had studied the effect of such an oil spill
at Galeta Island, Caribbean Panama, a decade earlier and
found that the subtidal reef corals were barely affected by
the spill but that the oil slick had caused severe damage
to the nearby intertidal mangrove community (Ritzler
and Sterrer, 1971). Our proposal won another five years
of research grants, and we named our initiative SWAMP
(Smithsonian Western Atlantic Mangrove Program). This

support was supplemented by internal grants for specific
purposes, notably Fluid Research Funds travel awards, a
Scholarly Studies grant for mangrove research, a Smithso-
nian Associates Women’s Committee award for scientific
illustration, W. R. Bacon Scholarships (for external col-
laborators), Seidell Funds for library enhancements, and
National Science Foundation grants to outside collabora-
tors (who were to some extent also supported by their
home institutions).

Because mangroves are tidal communities with terres-
trial, intertidal, and subtidal components, we could expand
our fields of interest, adapt our methods to the new envi-
ronment, and add a number of disciplines to our study that
are not usually applicable in the subtidal reef environment
(Rutzler and Feller, 1988; Figure 7). With a wider biodiver-
sity horizon, we could now conduct surveys of microbes,
fungi, algae, sponges and their endofauna, polychaetes,
crustaceans, echinoderms, and bryozoans. We were fortu-
nate to have a rare expert on the quantitatively important
ascidian tunicates join us at this time, Ivan Goodbody of
the University of the West Indies, Jamaica. Our team also
explored the geological history of the mangrove by coring
through massive peat accumulations and dating the dif-
ferent horizons and also initiated terrestrial studies of the
mangrove’s lichens, insects, spiders, reptiles, and birds.

An important first step was to explore and map Twin
Cays and name the many bays, ponds, creeks, mud flats,
and lakes and give them coordinates (before Global Posi-
tioning System [GPS] devices were available) that would
allow us to relocate research sites. We also wanted to gar-
ner more interest in the mangrove ecosystem, but because
swamps tend to be viewed as undesirable environments,
it took considerable effort to win over our sponsors, local
hosts, and even many colleagues. Good photography was
a decided help, but even the best pictures convey but a tiny
segment of a process in nature, although they are absolutely
necessary for documenting shapes, expressions, or colors,
of course. To depict the entirety of, say, an animal—plant
association, we needed to capture the obvious and the hid-
den, the large and the small (in proper detail and perspec-
tive), and the dynamics of day versus night—in a word, we
needed to combine art with science. We did just that in a
new project called Art in a SWAMP (Figure 8). The lead
artist was Ilka (“Candy”) Feller, a contract illustrator at
the time with vast experience in the fields of botany and
entomology, having worked with numerous colleagues in
those departments in our Museum over many years. Candy
not only employed her artistic talent in the illustration of
mangrove communities, but she was so captivated by the
entire ecosystem that she resumed academic studies (after

NUMBER 38

FIGURE 6. Twin Cays mangrove habitats. Clockwise, from top left: the island viewed southwest toward Carrie Bow Cay and the barrier reef;
mangrove fringe lining the Main Channel; sponge clusters in one of the tidal channels supported by red-mangrove stilt roots; diverse community
of sponges and ascidians on a root substrate; a newly discovered and described sponge (genus Haliclona) anchored on and in the mangrove-peat

bank that lines many channels; juvenile barracuda hiding among mangrove roots; a snorkeler exploring Hidden Creek, which connects a shal-
low mangrove lake with the open Main Channel.

NUMBER 38 ¢ 57

FIGURE 7. (facing page) Examples of projects initiated at Twin Cays. Clockwise, from top left: mapping and exploring team landing on the west-
ern shore; our weather station erected in a large tidal mud flat; scientific illustrator Mary Parrish sketching mangrove communities; Molly Ryan

photographing community samples returned to the lab at Carrie Bow for one of her scientific illustrations; student volunteer helping to collect
specimens; ichthyologists Will Davis and C. Lavett Smith comparing catches; Ilka Feller measuring salinity in Hidden Lake, the location of one
of her mangrove fertilization and growth experiments; Ian Macintyre, with Ilka, dissecting termite nest during an exploration of the mangrove’s

interior.

having raised two daughters), completed a dissertation on
the Twin Cays mangrove, and became one of the foremost
mangrove ecologists working on communities between
Florida and Brazil and as far away as Australia and New
Zealand. Other artists on the project were Mary Parrish,
first a staff member in my department and now illustrator
in the Department of Paleobiology; Molly Kelly Ryan, In-
vertebrate Zoology staff illustrator; and Jennifer Biggers,
then my contract research assistant and illustrator. This
team, along with Paleobiology’s Ian Macintyre, Natural
History Museum photographer Chip Clark, Invertebrate
Zoology technician Mike Carpenter, and several more as-
sociates and volunteers engaged in detailed surveys and
mapping of Twin Cays geomorphology and in analysis and
graphic reconstruction of habitats and communities, rang-
ing from epiphytic sponges to intertidal algae—invertebrate
associations to red mangrove-insect interactions, and an
entire mudflat population (Rutzler and Feller, 1996).

Our growing familiarity with mangrove organisms
raised a number of significant ecological and behavioral
questions concerning the composition and ecology of
floating cyanobacterial mats (addressed by Maria Faust
in the Department of Botany; Faust and Gulledge, 1996)
and herbivory in macroalgal communities (investigated
by Mark and Diane Littler and colleagues; e.g., Littler
et al., 1983). A rare immune disease was discovered in
a sponge (Rutzler, 1988) in which the usually beneficial
microbial symbionts turn against their host. The dynamics
and behavioral patterns of swarming copepod crustaceans
among mangrove roots were investigated by Frank Fer-
rari, Invertebrate Zoology, and colleagues (Buskey, 1998;
Ferrari et al., 2003). Also, new work was done on inverte-
brates living in complex burrows in the sediment substrata
of mangrove channels (Dworschak and Ott, 1993), and on
the importance of mangroves for the recruitment and pro-
tection of commercial species such as spiny lobster (Acosta
and Butler, 1997). In research on another puzzling ques-
tion—the role of sponge cyanobacterial symbionts in both
mangrove and reef nutrient cycling—it was shown that
nitrogen fixing by bacteriosponges is indeed an important
input to the community (Diaz and Ward, 1997). With the

new emphasis on the biology of the Twin Cays mangrove,
Ian Macintyre applied geological techniques over several
years to reveal its biological history: using a portable vi-
bracore, he mapped and carbon-dated the subsurface lay-
ering of peat, sand, and rubble, all the way to Pleistocene
level (Macintyre et al., 2004b).

As our financial situation improved, we were able to
address the bothersome problem of coping without auto-
mated instrumentation for continuous monitoring of ba-
sic meteorological and oceanographic parameters. Up to
then we had documented the tidal regime at Carrie Bow
Cay with a pressure sensor (Kjerfve et al., 1982) and re-
lied on various borrowed or leased instruments to meet
the needs of specific projects for monitoring water cur-
rents, temperature, solar radiation precipitation, or wind
(Greer and Kjerfve, 1982; Rutzler and Ferraris, 1982).
None of these improvised methods provided long-term,
reliable records even if we were on site. With the help of
contract engineer George Hagerman, we adapted a leased
Anderaa (Bergen, Norway) automatic weather station for
our purposes and installed it on a massive, elevated wood
platform in an extended tidal mud flat to the north of
Twin Cays. This setup provided us with continuous data
for several years until it became outdated and fell victim
to vandalism.

CARIBBEAN CORAL REEF ECOSYSTEMS:
A BREAKTHROUGH

In the early 1980s, Richard Fiske, then Director of
the National Museum of Natural History, asked for pro-
posals that would interest our sponsors in the U.S. Con-
gress, who at that point appeared to favor the expansion
of promising research already in progress. To our surprise,
the Museum received an increase to its budget base for
the study of Caribbean Coral Reef Ecosystems beginning
in 1985. Modeled on the IMSWE-SWAMP initiatives, the
CCRE program encompassed reef, mangrove, seagrass,
and plankton communities, with a primary focus on the
Carrie Bow Cay region.

58 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

(As
x

ors ie NEN Se ley a
E ai 7A 3)

wen =

FIGURE 8. Examples of illustrations of mangrove swamp communities. Clockwise, from top left: characteristic intertidal red algal cover (Bos-
trychietum) on red-mangrove prop root, with mangrove oyster, mangrove crab, and periwinkle; tidal mudflat showing developed red-mangrove
seedlings, black-mangrove pneumatophores, driftwood, and a marbled godwit; cut-away view of mangrove channel bottom showing charac-
teristic benthic organisms including decapod burrowers; close-up of peat-bottom community with algae, fallen mangrove leaves, sea anemone,
and sabellid tube worms.

Under a new administrative structure approved by
Director Fiske, CCRE would be governed by a steering
committee chaired by me and composed of representa-
tives of different departments and disciplines, including
outside advisers. Marsha Sitnik, who worked with all the
Museum’s interdepartmental biodiversity initiatives, be-
came program administrator. Now we could afford the
important position of operations manager, filled by Mike
Carpenter, who serves as field station logistics director and
has trained and supervised our volunteer station manag-
ers, each of whom typically spends three to six weeks at
Carrie Bow Cay.

The new funding allowed us to lease all of Carrie
Bow Cay year round and remodel the big house for much-
needed dry-lab space for instruments, a library, computer,
and additional living accommodations. We built a sepa-
rate, sound-insulated compressor-generator shed, added
propane gas refrigerators to kitchen and labs, and im-
proved radio-communication and other safety features for
boating and diving. New equipment included microscopes,
balances, centrifuge, and other analytical equipment. We
also upgraded our weather station with real-time data ac-
cess and connected it with the Belize Meteorological Of-
fice, which had no offshore monitoring facility. Even more
important, we had a modest budget for travel stipends to
attract outside collaborators to work on organisms or dis-
ciplines not covered by Smithsonian staff scientists.

At the end of the first CCRE program year, our pub-
lication list exceeded 200 entries. Several of the projects
mentioned earlier were continued or completed and new
ones begun with like-minded colleagues whose expertise
filled the gaps in our experience. To name a few of these
projects, some focused on the control of reef zonation by
light and wave energy (Graus and Macintyre, 1989), the
taxonomy and ecology of hydroids (Calder, 1991), oligo-
chaete worms (Erséus, 1988), mysid crustaceans (Modlin,
1987), and ascidians (Goodbody, 1995); the predation
and feeding ecology of sponges (Wilkinson, 1987), echi-
noderms (Aronson, 1987), and fishes (Wainwright, 1988);
the ecophysiology of invertebrate—bacterial symbiosis
supporting life in hydrogen sulfide environments (Ott and
Novak, 1989; Ott et al., 1991) and mangrove-tree me-
tabolism (McKee et al., 1988); and island groundwater
hydrology (Urish, 1988).

In 1988 we held a workshop at the Calvert Marine
Museum in Solomons, Maryland, to review the accom-
plishments and gaps in our research on the Twin Cays
mangrove ecosystem. Close to 40 program participants
summarized the progress of their work on internal struc-
ture, development over time, sedimentology, meteorology,

NUMBER 38 e¢ 59

hydrology, vegetation, productivity, nutrient cycling,
temperature-salinity tolerance, and biodiversity of fauna
and flora from microbes to amphibious fishes. The most
obvious deficiencies were in oceanography, a number of
important organism groups such as mollusks and fishes,
and marine benthic and terrestrial ecology. One of the
highlights was a report on a complementary team study of
the Holocene geological history, peat composition, and ter-
restrial and marine vegetation of Tobacco Range, another
large mangrove island about 3.5 km north of Twin Cays
(Littler et al., 1995). This atoll-like range drew CCRE’s at-
tention when a large area of fractured and slumped fossil
peat was discovered off its west shore.

At a subsequent planning workshop in Jamaica, CCRE
established a protocol for studies at Twin Cays and Carrie
Bow Cay. This initiative, called Caribbean Coastal Marine
Productivity (CARICOMP), calls for simple but univer-
sally applicable methodologies in the monitoring of major
oceanographic parameters and health of the Caribbean’s
principal communities. To this end, we established repre-
sentative plots and transects in mangrove, seagrass meadow,
and fore-reef, which are being evaluated yearly for changes
in structure and productivity, while climatic factors are de-
termined on a weekly basis (Koltes et al., 1998).

As our scientific drawing and photography of swamp
communities gained scientific importance and aesthetic
value, we were invited display some of this work to the
public at the Smithsonian’s S. Dillon Ripley Center in an
exhibition titled “Science as Art.” It included a video doc-
umentary on mangrove swamp biology, produced in col-
laboration with colleagues from the University of Vienna
(Joerg Ott and Alexander Bochdansky). The video also
served as a teaching aid in an annual educational work-
shop for Belize high school teachers, conducted by Candy
Feller and Marsha Sitnik in collaboration with the Belize
Fisheries and Forestry departments and titled “Mangrove
Conservation through Education.” This was a timely
workshop, indeed: our research area at Twin Cays was
showing the first signs of anthropogenic stress in response
to tourist visitation, garbage dumping, vandalism (of our
weather station and boat dock), and the clear-cutting of
mangrove trees to gain land for development. These de-
velopments had a particularly adverse impact on Candy
and her colleagues, whose work on mangrove plant ecol-
ogy required extended undisturbed natural conditions to
single out parameters (nutrients, in particular) that en-
hance or impede growth (Feller, 1995; Feller et al., 1999).
Fortunately, at our urging, Belize’s Forestry and Natural
Resources Departments helped slow the disturbances and
started work on a conservation plan for the South Water

60 e© SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Cay Marine Protected Area (MPA) that would include
Carrie Bow and Twin Cays.

Side-tracked by the discovery of the Pelican Cays
biodiversity hotspot (see the next section), the impact
of a hurricane, and coral bleaching events, we were un-
able to convene another Twin Cays symposium until
2003. Meeting at the Smithsonian Marine Station in Fort
Pierce, Florida, we found our mangrove program had ac-
cumulated enough scientific results not only to fill a vol-
ume of multidisciplinary papers but also to demonstrate
changes in the structure of the ecosystem over a span of
two decades (Macintyre et al., 2004a). Articles on geo-
logical history and sedimentary conditions were spear-
headed by Ian Macintyre and those on aquatic ecology
by Ritzler and colleagues. Other contributions covered a
wide range of topics: changes in the mangrove landscape,
documented through aerial and satellite imagery by Wil-
frid Rodriguez and Ilka Feller at the Smithsonian Environ-
mental Research Center; marine botany, investigated by
Maria Faust and the Littler team; Foraminifera, by Susan
Richardson of the Smithsonian Marine Science Network
(MSN); symbiotic ciliates, by Joerg Ott; sponge ecology,
by Cristina Diaz (then an MSN Fellow with me) and Janie
Wulff (now at the Florida State University, Tallahassee);
and two very different worm groups, the interstitial gna-
thostomulids by Wolfgang Sterrer (Natural History Mu-
seum, Bermuda), and the burrowing sipunculans, by Anja
Schulze, postdoctoral fellow, and Mary Rice, emeritus
scientist, both at SMS, Ft. Pierce, Florida (Mary is also
the founding director of that laboratory). William Browne
(University of Hawaii, Honolulu) summarized years of
genetic and developmental research on mangrove crusta-
ceans, Judith Winston her work on bryozoans, and Ivan
Goodbody his observations on ascidian diversity. Years of
genetic research at Twin Cays on a highly unusual am-
phibious fish, the mangrove rivulus, was summarized by
Scott Taylor and his collaborators, and terrestrial biology
received a welcome boost from observations by Seabird
McKeon (then at SERC) and Stephen Mitten (University

of Missouri, St. Louis, now based in Belize). Energy flow
was also examined in a paper on nutrients derived from
microbial mats by Samantha Joye (University of Georgia,
Athens), and another on the planktonic food web by Ed-
ward Buskey (University of Texas at Austin). To round
out the reports, Mary Parrish explained the important role
of scientific field illustration—a collaboration between sci-
entist and artist—in analyzing and explaining mangrove
communities.

CCRE’s recent accomplishments also include two
far-reaching initiatives. The first, begun by collabora-
tor Emmett Duffy from the Virginia Institute of Marine
Science (Gloucester Point), is a study of the systematics
and ecology of snapping shrimp (Alpheidae) that live in
reef sponges with a large interior cavity system, such as
the genus Agelas. As the work progressed with various
specialists and graduate students coming on board, al-
pheids were found to have much more genetic diversity
and ecological complexity than previously thought. An-
other discovery, a first among marine life, was that these
crustaceans have the same advanced social structure (eu-
sociality) as some well-studied terrestrial animals, such
as termites (Duffy, 1996). Second, a logistical break-
through, made possible through our collaboration with
colleagues at the University of Rhode Island, was the
development of a new integrated environmental sens-
ing system with a radio-telemetry link to the Internet
(Opishinski et al., 2001).

PELICAN Cays, BIODIVERSITY HOTSPOT

In the early 1990s, our neighbors on Wee Wee Cay,
Paul and Mary Shave, alerted us to another amazing
ecosystem: the Pelican Cays (16°59.8'’N, 88°11.5'W), a
biologically rich mangrove island group less than 20 m
south-southwest of our Carrie Bow station (Figure 9). We
now had an 8 m boat, more substantial than any before,
that could take us there in about an hour. Ivan Good-
body, the first to visit the Pelicans on the rumor of an

FIGURE 9. (facing page) The Pelican Cays “mangreef.” Clockwise, from top left: aerial photograph of Cat Cay showing mangrove, reef ridges,
and deep lagoons (Manatee Cay, left, and Fisherman’s Cay are in the background); diver swimming over coral (Agaricia) below the canopy of a
red-mangrove tree; Rich Aronson and Ian Macintyre operating a hand corer to retrieve subbottom coral and other deposits; close-up of ascidian
(Clavellina)-sponge (Monanchora, Spirastrella) community, enveloped by brittle star arms, on Manatee Cay mangrove root; Klaus Ruetzler
sampling sponges (Aplysina) in a marine pond on Fisherman’s Cay; Coral Ridge (Agaricia, Palythoa) with sponges (Chondrilla) at Cat Cay

lagoon entrance (sponge-covered mangrove roots in background).

NUMBER 38

62 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

ascidian paradise there, found the area teeming with as-
cidians—and much more. Many of us followed in short
order, eager to investigate the Pelicans’ atoll-like reef, an
elongate north-south-oriented structure measuring almost
10 X 3 km and studded with about a dozen mangrove
cays on its northern rim. Most of the islands enclose deep
circular ponds that support and protect a diverse commu-
nity of marine plants and sessile filter feeders—particu-
larly sponges and ascidians—flourishing on red-mangrove
stilt roots and peat banks. A wealth of other invertebrates
and fishes live just below the tide line, a mix of reef and
mangrove organisms, some species previously seen only
in much deeper water on the barrier fore-reef, although
the Pelican Cays are situated deep inside the barrier-reef
lagoon, halfway between the reef and the mainland.

We were so impressed by the unusual diversity and
ecological complexity of the Pelican Cays ecosystem that
we asked the Belize Coastal Zone Management and Fish-
eries units to include the region in the South Water Cay
MPA. Tony Rath (NaturaLight, Dangriga) and Jimmie
Smith (Islands from the Sky, Houston, Texas) helped with
aerial photography, Molly Ryan with mapping, and our
research team along with outside collaborators addressed
the new scientific perspectives (Macintyre and Ritzler,
2000). Macintyre and his team, and Karen McKee (U.S.
Geological Survey, Lafayette, Louisiana), spearheaded the
study of geological underpinnings and vegetation history,
Dan Urish (University of Rhode Island) and Tracy Villar-
eal (University of Texas at Austin) the hydrography of the
ponds, Thomas Shyka (National Oceanographic and At-
mospheric Agency, Silver Spring, Maryland) the nutrient
cycle and water flow patterns in the ponds, Steve Mor-
ton (Bigelow Laboratory, West Boothbay Harbor, Maine)
and Maria Faust the phytoplankton, Mark and Diane
Littler the marine algae and seagrasses, Susan Richard-
son (then at Yale University, New Haven, Connecticut)
the epiphytic foraminiferans, Rutzler and colleagues the
sponges, Janie Wulff (then at Middlebury College, Ver-
mont) sponge predation, Wolfgang Sterrer the gnathosto-
mulids, Gordon Hendler (Natural History Museum, Los
Angeles, California) the echinoderms, and Ivan Good-
body the tunicates. At the height of these investigations,
we were able to introduce the spectacular coral commu-
nities of the cays to participants of the 8th International
Coral Reef Symposium, along with other points of inter-
ests, such as community changes in the reef zones of the
Carrie Bow reference transect over the past two decades
(Macintyre and Aronson, 1997). The fishes of the Peli-
cans were investigated (just after the edited volume was
published) by a team of ichthyologists led by our Muse-

um’s James Tyler and included former American Museum
of Natural History curator (now retired) C. Lavett Smith
(Smith et al., 2003). More recently, important suspension
feeders that had not been covered by the earlier surveys,
the bryozoans, were studied by our long-time collabora-
tor Judith Winston of the Virginia Museum of Natural
History (Winston, 2007).

Although still uncertain of the causes of this archipel-
ago’s unusually high biodiversity, CCRE researchers saw
ample evidence of its fragility and warned of the irrepa-
rable harm that could come to its delicate communities
from careless visitors or water warming during long pe-
riods of calm (as observed in course of some hurricanes).
Little did we know that our concerns would soon prove
to be well founded. In the course of a number of routine
survey flights over the reef, we noted disturbing signs of
land “development” on several of the Pelican islands, sub-
sequently confirmed by ground-truthing: we found large
areas of mangrove clear-cut and bottom sediments near
the cays dredged to obtain fill material on which homes
and resorts could be built. We reported our observations
to the authorities because by that time the cays were al-
ready part of the South Water Cay MPA and mangrove
cutting was illegal without a special permit. At the time of
this writing, the activities have stopped and are under in-
vestigation by the government of Belize. Unfortunately, a
great natural treasure has been severely damaged, without
any clear sense of whether and how soon a recovery will
be possible.

A MEMORABLE YEAR, 1997

In CCRE’s 35-year history, 1997 stands out for its
remarkable highs—and lows. Scientifically, many signifi-
cant field projects were launched or carried to completion:
investigations of coral bleaching, a new and unsettling
phenomenon on the barrier reef (Aronson et al., 2000);
ecophysiological analysis of periodic crustacean swarming
among red-mangrove stilt roots (Buskey, 1998); a pioneer-
ing initiative to match poorly known fish larvae to the
adults of the species, first by morphological means after
rearing in the laboratory, later by DNA analyses (Baldwin
and Smith, 2003); and a workshop on Caribbean sponge
systematics with experts from five nations that led to a
better understanding of the barrier-reef and mangrove
poriferan fauna (Riitzler et al., 2000). This was also the
International Year of the Reef, and to celebrate the occa-
sion we made every effort to share our enthusiasm for this
unique environment with students and the general public
through numerous lectures, poster sessions, and demon-

strations, on site in Belize and back home at the National
Museum of Natural History. To add to the festivities, our
Carrie Bow field station, the logistical base and catalyst
of our program, had reached the respectable age of 25
(1972-1997).

But the Gods of the Sea must have had other plans
for this venerable facility. On 6 December 1997 an acci-
dental fire broke out, aided by old, termite-riddled lum-
ber and fanned by a strong northerly wind. Most of the
station was reduced to ashes—laboratory, kitchen, living
quarters, even wooden vats filled with water, all except
a small cottage and the generator hut, which were iso-
lated on the south end of the island. The blaze destroyed
much valuable equipment, including microscopes, bal-
ances, solar system, weather station, and the contents of
the library. As a result, little fieldwork could be done in
the following year, although we did investigate the fire’s
damage to the island (20 or more coconut trees were
lost) and the impact of recent complete flooding (which
caused substantial coastal erosion). Two other points of
interest were the impact on the reef after being subjected
to stormy seas with waves up to 6 m and to an extensive
calm period with shallow-water warming that appeared
to precipitate the bleaching and death of large numbers
of corals.

At this juncture, we gave serious consideration to ter-
minating the CCRE program at this location—but not for
long. Buoyed by the positive spirit of the Bowman family
(and some insurance payback) and the talent of a young
Cuban-trained architect, Hedel Gongora, we designed
a new field laboratory to take the place of the old main
house. It was built by local carpenters with lumber from
pine forest in the west of the country, complete with wet
lab, dry lab, library, running seawater system, workshop,
and kitchen (Figures 10, 11). The facility was rededicated
as the Carrie Bow Marine Field Station in August 1999,
with more than 100 visitors in attendance to celebrate the
occasion, including local fishermen, cooks, the Minister
of Environment, the U.S. ambassador to Belize, scientists,
and representatives of all major conservation societies.
Over the next two years, we added one cottage for liv-
ing quarters and rebuilt the one spared by the fire. With
generous donations from a number of U.S. companies and
individuals, we replaced and improved most laboratory
equipment and instrumentation, and Tom Opishinski in-
stalled a new meteorological-oceanographic monitoring
station enhanced by COASTMAP software (donated by
the University of Rhode Island). By the beginning of 2000,
CCRE was back on its feet, functioning as a year-round
scientific program.

NUMBER 38 °¢ 63

RESURGENCE AND BIOCOMPLEXITY

With a renovated field station, CCRE’s scientific mo-
mentum took off once again, with new scientific oppor-
tunities as well as challenges. Nearly 80 scientific staff
resumed field research disrupted by the fire or initiated
new projects. A number centered on the sad effects of en-
vironmental stress or degradation on delicate but essen-
tial reef-building corals. To aid in the understanding of
possible coral revival, Ken Sebens (then at the University
of Maryland) and colleagues evaluated the benefit of wa-
ter currents for the growth of the reef-building shallow-
water coral Agaricia, which has been adversely affected
by extended calm periods during hurricanes (Sebens et
al., 2003). A parallel ecophysiological study found dif-
ferent tolerances to elevated temperature among species
of Agaricia and speculated that their abundance may
therefore vary with environmental disturbance (Robbart
et al., 2004). However, corals that have survived such
events may have their recovery impeded by the grazing
of parrotfish, which are otherwise considered beneficial
to the health of reefs (Rotjan et al., 2006). According to
a series of investigations on predators and competitors of
reef and mangrove sponges, these aggressors help defense
mechanisms in sponges evolve (Wulff, 2005). In another
sponge study, we concluded that encrusting excavating
sponges have a competitive edge over reef corals weak-
ened by elevated temperatures: the sponges can under-
mine the weakened opponent as well as displace its liv-
ing tissue (Ritzler, 2002). Assessments by Rich Aronson
(Dauphin Island Sea Lab, Alabama), with collaborators,
showed alarming recent changes in the composition of
reef-building coral species as a result of stress and dis-
ease (Aronson et al., 2002, 2005), while John Pandolfi
(then, Paleobiology) identified trends responsible for the
decline of coral-reef ecosystems worldwide (Pandolfi et
al., 2003). A discovery with harmful implications for
human consumption of seafood (ciguatera poisoning)
was the increase in toxic dinoflagellate algal blooms in
our area (Faust and Tester, 2004), a phenomenon attrib-
uted to increased nutrient levels in lagoon waters, earlier
considered a potential threat (Morton and Faust, 1997).
On a more positive note, studies by Ana Signorovitch
(then a graduate student at Yale University) applying
innovative molecular methods to the enigmatic Tricho-
plax in the one-species phylum Placozoa found consid-
erable genetic diversity there as well as signs of sexual
reproduction (Signorovitch et al., 2005). Using a new
approach to sponge systematics from the cytochrome
oxidase subunit 1 gene tree, Sandra Duran (postdoctoral

FIGURE 10. New Carrie Bow Marine Field Station (2000) and Biocomplexity Program. Top row: left, aerial view of Carrie Bow Cay look-
ing north (see details on map, Figure 11); right upper, the island with laboratory and living facilities as seen from the barrier reef (open-ocean
side) and, image immediately below, view from the dock (lagoon side). Center row: left, flow-through seawater system photographed from the
storage tank above; center, dock-mounted oceanographic sensors; right, view of upper-level dry lab. Bottom row: Starting the Twin Cays Bio-
complexity Program on mangrove nutrient cycle: left, Ilka Feller with experimental enclosure in a tidal mudflat surrounded by black mangrove;
right, subtidal bacterial mat with decaying mangrove leaves, an early stage in the cycle.

NUMBER 38

Oceanographic
Sensors

Data Acquistition

&
Communications
Meteorological Sensors System
& ——>* 9
Communications
Tower ; “ Aeew
Main Building
° <— Rain Guage
Generator
&
Dive shed
Scientists’
y Quarters
y "
Scientists’
Quarters
10m

FIGURE 11. Map of Carrie Bow Cay (in 2000) identifying principal structures (number of coconut palms re-
duced for clarity).

65

66 © SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

fellow with Valerie Paul at SMS, Ft. Pierce) uncovered
important information on rDNA phylogenies and an in-
teresting case of ecological speciation wherein popula-
tions from reefs and mangroves in close vicinity were
genetically more distant than those from similar habitats
separated by thousands of kilometers (Duran and Riutz-
ler, 2006; Erpenbeck et al., 2007).

One breakthrough during this period was the five-
year multidisciplinary “Mangrove Biocomplexity Study
at Twin Cays: Microbial and Nutrient Controls,” funded
by the National Science Foundation and headed by Ilka
Feller, with nine collaborators from outside institutions
(see Figure 10). Several important contributions to an
understanding of nutrient production, cycling, limita-
tion, effects on organisms, and related ecophysiologi-
cal phenomena have already been published (Feller et
al., 2002, 2007; Lovelock et al., 2006; Cheeseman and
Lovelock, 2004), but many more are expected in the near
future now that the fieldwork phase of the Biocomplex-
ity Program has come to a conclusion. One surprising
discovery, in a complementary project by CCRE post-
doctoral fellow Amy Erickson, was that the intertidal
tree crab Aratus, long thought to be a mangrove leaf
eater, actually prefers an animal diet if given a choice
(Erickson et al., 2008). Another important finding, by
a group from the University of Vienna, related to the
microenvironment of sessile ciliates growing on man-
grove peat banks and associated with chemoautotrophic
bacteria coating their surface. By analyzing motion be-
havior of the ciliate and measuring microelectrodes in
situ, scientists could show that the hydrogen sulfide re-
quired by the symbionts seeped into the boundary layer
between the peat surface and oxygenated water column
(Vopel et al., 2002).

CONCLUSION AND OUTLOOK

In 1972 a group of enthusiastic, like-minded young
scientists embarked on a comprehensive, long-term field
investigation of unprecedented dimensions for the Mu-
seum of Natural History. The team was unified in its belief
that organisms had to be studied in their natural settings
for a clear understanding of their features and role in their
community (Figures 12, 13). Only then could a preserved
museum collection aid in documenting the building blocks
of an ecosystem. This approach was particularly true for
coral reefs, which were all but inaccessible to scientists
until scuba diving allowed in situ observations and experi-
mental study. Similarly, it was essential to probe and sam-

ple substrata to better understand the structure of com-
munities, present and past. The team directed its attention
to the Caribbean because it is the Americas’ tropical sea,
to which our own nation is connected by weather, ocean
currents, and marine resources, as well as by cultural and
economic exchange.

In the beginning, we were convinced that together,
and with the cooperation of selected specialized collabo-
rators, we could pierce most of the secrets of a functioning
coral reef in little more than a decade and generate models
for predicting future trends. It did not take long for the re-
strictions of space, limits of available talents, and chronic
shortage of funds to show us how naive we were. Besides,
as all scientists know, every resolved problem opens up
new questions. Nevertheless, we can look back on over a
third of a century of substantial progress, with more than
800 research papers in print and many more under way,
all focused on a particular reef ecosystem and covering a
significant time span. Our investigators addressed a vast
array of subjects: biodiversity, from microbe to mammal;
the geological and sedimentological setting and its devel-
opmental history; the physical and chemical factor regime;
developmental biology, genetics, food chains, nutrient
production, and cycling; behavior, competition, predation,
and disease; and fisheries and conservation. We produced
an impressive database that a new generation of motivated
researchers can build upon with the benefit of technical ad-
vancements such as molecular analysis, which should shed
further light on eutrophication, climate change, and other
stress factors responsible for the increasing occurrence of
algal blooms and devastating invertebrate diseases. These
topics and more will need our full attention to help guide
resource management and conservation efforts—and,
above all, to preserve the aesthetic and economic value of
the world’s reefs.

Over the past few years our program has once again
come up against a number of hurdles. Financial shortages
in the Natural History Museum’s budget have eroded our
funding, while staff has been reduced and not replaced,
leaving our scientific and management capabilities some-
what shaky. However, some of the slack was picked up by
endowed funds, and our field station became part of the
Smithsonian-wide Marine Science Network, joining the
ranks of our “big brothers,” the Environmental Research
Center at Chesapeake Bay, the Marine Station in Florida,
and the Tropical Research Institute in Panama.

It is gratifying to find that half the papers in this vol-
ume of the Smithsonian Contributions to Marine Science
series emanated from our CCRE program and the Car-
rie Bow Marine Field Station. The scientific advances

FIGURE 12. Examples of recent projects conducted at Carrie Bow. Top row: left, diver sampling fish larvae in situ, a project on larval rearing
and molecular identification headed by Carole Baldwin and Lee Weigt; right, Juan Sanchez setting up in situ experiment for study of gorgo-
nian ecology and growth. Middle row: left, Randi Rotjan recording fish bites on coral (Porites) on the reef shallows (inset below: larval fish
[Rypticus] reared by the Baldwin team in the Carrie Bow seawater system to develop characteristics used in identification of adults); center,
colonial ciliates (Zoothammnium), barely 15 mm tall, with bacterial ectobiont, dwell on mangrove peat and are studied by Joerg Ott’s group;
right, collaborator Kay Vopel measuring the microclimate surrounding Zoothamnium, primarily the oxygen versus hydrogen sulfide balance.
Bottom row: left, Klaus Ruetzler recording progress of an excavating encrusting sponge (Cliona) that competes with temperature-stressed coral
(Diploria) (center); right, new records of sponge disease are exemplified by this specimen (Callyspongia).

om A

FIGURE 13. Research in progress and unanticipated new opportunities. Top row: left, Klaus Ruetzler initiated (with Carla Piantoni) a study
of cryptic and cave-dwelling reef communities in shallow-water (center, upper photo) and deep-water (center, lower photo) habitats with little
or no light exposure; right, Laurie Penland assists recording cave fauna using a digital HD video camera. Middle row: left, water warming
during hurricanes killed many shallow-water corals in the Pelican Cays, which became overgrown by sponges (Chondrilla) that benefit starfish
(Oreaster) as a source of food, thus starting a new ecological cycle; right, clear-cutting of mangrove and filling in resulting tidal flats with coral
sand for “land development” started here at Twin Cays in the 1990s and continued at the Pelican Cays. Bottom row: left, Manatee Cay shown
in 2008; right, this environmental disaster, recently stopped by the government of Belize, offers opportunities to study parameters of recovery
of stressed and depleted marine and terrestrial communities.

achieved through CCRE research indicate that our deci-
sions and actions over the years have blazed a fertile trail
for the future of our science.

ACKNOWLEDGMENTS

A program of this length would not have been pos-
sible without the help of countless supporters—too many
to name individually. But all associated with us will know
that they are being thanked because the spirit and cama-
raderie generated by our joint endeavor is unprecedented.
No more than a handful of naysayers have attempted to
slow our progress or dampen our enthusiasm—obviously
without success. Above all, we are grateful to the Bowman
family of Dangriga for leasing their Carrie Bow Cay to our
program and for taking a serious interest in our work from
the first day. The government of Belize, the Department
of Fisheries in particular, hosted our scientific efforts and
granted us necessary permits. Many of the people of Be-
lize, from Dangriga Town in particular, helped us with lo-
gistics by sharing their knowledge of the country’s coastal
resources, their skilled and dedicated cooks, boatmen, car-
penters, hotel services, and provisions. Back home, we re-
ceived invaluable cooperation and support from scientific
colleagues, private and federal funding sources, and Smith-
sonian staff from all levels of administration and science
units, including the Natural History Museum’s Director’s
Office, the Smithsonian Scientific Diving Program and its
staff, and many of the Museum’s research assistants, scien-
tific illustrators, photographers, and collection managers.
For my part, I could not have directed and inspired our
program without the assistance of many dedicated collabo-
rators: Mike Carpenter, CCRE operations manager; Mar-
sha Sitnik, program administrator; and Kathleen Smith,
Michelle Nestlerode, Martha Robbart, Robyn Spittle, and
Carla Piantoni, research assistants.

Photographs and art in this contribution were con-
tributed by Mike Carpenter, Chip Clark, Ilka Feller, Ron
Larson, Kathy Larson, Sara Lewis, Diane Littler, Ian Mac-
intyre, Vicky Macintyre, Julie Mount, Aaron O’Dea,
Tom Opishinski, Mary Parrish, Laurie Penland, Carla
Piantoni, Tony Rath, Mary Rice, Randi Rotjan, Klaus
Ruetzler, Molly Kelly Ryan, Juan Sanchez, Kjell Sandved,
Jimmie Smith, Kathleen Smith, Kay Vopel, all of whom
are (or were at the time) affiliated with the Smithsonian
Institution, and two undetermined photographers.

This is contribution number 850 of the Caribbean
Coral Reef Ecosystems Program (CCRE), Smithsonian
Institution, supported in part by the Hunterdon Oceano-
graphic Research Fund.

NUMBER 38 ¢ 69

LITERATURE CITED

Acosta, C. A., and M. J. Butler. 1997. Role of Mangrove Habitat as a
Nursery for Juvenile Spiny Lobster, Panulirus argus, in Belize. Ma-
rine and Freshwater Research, 48:721-727.

Adey, W. H., and I. G. Macintyre. 1973. Crustose Coralline Algae: A
Re-Evaluation in the Geological Sciences. Geological Society of
America Bulletin, 84:883-904.

Antonius, A. 1982. “Coral Reef Pathology: A Review.” In The Reef and
Man: Proceedings of the Fourth International Coral Reef Sympo-
sium, Volume 2, pp. 3-6. Quezon City, Philippines : Marine Sci-
ences Center, University of the Philippines.

Aronson, R. B. 1987. Predation on Fossil and Recent Ophiuroids. Paleo-
biology, 13(2):187-192.

Aronson, R. B., I. G. Macintyre, S. A. Lewis, and N. L. Hilbun. 2005.
Emergent Zonation and Geographic Convergence of Coral Reefs.
Ecology, 86:2586-2600.

Aronson, R. B., I. G. Macintyre, W. F. Precht, T. J. T. Murdoch, and C. M.
Wapnick. 2002. The Expanding Scale of Species Turnover Events on
Coral Reefs in Belize. Ecological Monographs, 72:233-249.

Aronson, R. B., W. F. Precht, I. G. Macintyre, and T. J. T. Murdoch.
2000. Coral Bleach-Out in Belize. Nature (London), 405:36.

Baldwin, C., and D. Smith. 2003. Larval Gobiidae (Teleostei: Perci-
formes) of Carrie Bow Cay, Belize, Central America. Bulletin of
Marine Science, 72:639-674.

Bowman, H. T. A. 1979. Emerald Valley and Twinkling Town: My Auto-
biography. Stone Haven, Stann Creek Valley, Belize: Author.

Buskey, E. J. 1998. Energetic Costs of Swarming Behavior for the Cope-
pod Dioithona oculata. Marine Biology, 130:425-431.

Calder, D. R. 1991. Abundance and Distribution of Hydroids in a Man-
grove Ecosystem at Twin Cays, Belize, Central America. Hydrobio-
logia, 216/217:221-228.

Carr, D., and J. Thorpe. 1961. From the Cam to the Cays; the Story of
the Cambridge Expedition to British Honduras 1959-60. London:
Putnam.

Cheeseman, J., and C. Lovelock. 2004. Photosynthetic Characteristics of
Dwarf and Fringe Rhizophora mangle L. in a Belizean Mangrove.
Plant, Cell and Environment, 27:769-780.

Cressey, R. 1981. Parasitic Copepods from the Gulf of Mexico and Ca-
ribbean Sea, I: Holomolochus and Novabomolochus. Smithsonian
Contributions to Zoology, No. 339. Washington, D.C.: Smithson-
ian Institution Press.

Dahl, A. L. 1973. Surface Area in Ecological Analysis: Quantifications of
Benthic Coral-Reef Algae. Marine Biology, 23:239-249.

Diaz, M. C., and B. B. Ward. 1997. Sponge-Mediated Nitrification in
Tropical Benthic Communities. Marine Ecological Progress Series,
156:97-107.

Duffy, J. E. 1996. Eusociality in a Coral-Reef Shrimp. Nature (London),
381:512-514.

Duran, S., and K. Riitzler. 2006. Ecological Speciation in Caribbean Marine
Sponges. Molecular Phylogenetics and Evolution, 40(1):292-297.

Dworschak, P. C., and J. Ott . 1993. Decapod Burrows in Mangrove-
Channel and Back-Reef Environments at the Atlantic Barrier Reef,
Belize. Ichnos, 2:277-290.

Erickson, A. A., I. C. Feller, V. P. Paul, L. M. Kwiatkowski, and W. Lee.
2008. Selection of an Omnivorous Diet by the Mangrove Tree Crab
Aratus pisonii in Laboratory Experiments. Journal of Sea Research,
59:59-69.

Erpenbeck, D., S. Duran, K. Riitzler, V. Paul, J. N. A. Hooper, and G. Worhe-
ide. 2007. Towards a DNA Taxonomy of Caribbean Demosponges:
A Gene Tree Reconstructed from Partial Mitochondrial COI Gene
Sequences Supports Previous rDNA Phylogenies and Provides a New
Perspective on the Systematics of Demospongiae. Journal of the Ma-
rine Biological Association of the United Kingdom, 87:1563-1570.

70 e¢ SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Erséus, C. 1988. Taxonomic Revision of the Phallodrilus rectisetosus
Complex (Oligochaeta: Tubificidae). Proceedings of the Biological
Society of Washington, 101(4):784-793.

Faust, M. A., and R. A. Gulledge. 1996. Associations of Microalgae
and Meiofauna in Floating Detritus at a Mangrove Island, Twin
Cays, Belize. Journal of Experimental Marine Biology and Ecology,
197:159-175.

Faust, M. A., and P. A. Tester. 2004. “Harmful Dinoflagellates in the Gulf
Stream and Atlantic Barrier Coral Reef, Belize.” In Harmful Algae
2002. Proceedings of the Xth International Conference on Harmful
Algae, ed. K. A. Steidinger, J. H. Landsberg, C. R. Tomas, and G. A.
Vargo, pp. 326-328. St. Petersburg, Fla.: Florida Fish and Wildlife
Conservation Commission, Florida Institute of Oceanography, and
Intergovernmental Oceanographic Commission of UNESCO.

Feller, I. C. 1995. Effects of Nutrient Enrichment on Leaf Anatomy of Dwarf
Rhizophora mangle (Red Mangrove). Biotropica, 28(1):13-22.

Feller, I. C., C. E. Lovelock, and K. L. McKee. 2007. Nutrient Addition
Differentially Affects Ecological Processes of Avicennia germinans
in Nitrogen vs. Phosphorus Limited Mangrove Ecosystems. Ecosys-
tems, 10:347-359.

Feller, I. C., K. L. McKee, D. Fk Whigham, and J. P. O’Neill. 2002. Ni-
trogen vs. Phosphorus Limitation across an Ecotonal Gradient in a
Mangrove Forest. Biogeochemistry, 62:145-175.

Feller, I. C., D. F Whigham, J. P. ONeill, and K. L. McKee. 1999. Effects
of Nutrient Enrichment on Within-Stand Cycling in a Mangrove
Forest. Ecology, 80(7):2193-2205S.

Ferrari, E, J. Fornshell, L. Ong, and J. W. Ambler. 2003. Diel Distri-
bution of Copepods across a Channel of an Overwash Mangrove
Island. Hydrobiologia, 477:1-13.

Ferraris, J. D. 1981. Oxygen Uptake with Acute Variation in Tempera-
ture and Salinity in Two Coral Reef Polychaetes. Marine Ecology,
2:159-168.

Ferraris, J. D., K. Fauchald, and B. Kensley. 1989. “Degrees of Fitness in
Water and Ion Regulation in Alpheid, Polychaete, and Sipunculid
Inhabitants of a Belizean Mangrove.” In Proceedings of the Inter-
national Union of Physiological Sciences XVII, p. 427. Helsinki:
International Congress of Physiological Sciences.

Goodbody, I. G. 1995. Ascidian Communities in Southern Belize—A
Problem in Diversity and Conservation. Aquatic Conservation:
Marine and Freshwater Ecosystems, 5:355-358.

Graus, R. R., and I. G. Macintyre. 1976. Light Control of Growth
Forms in Colonial Reef Corals: Computer Simulation. Science,
193:895-897.

. 1989. The Zonation Patterns of Caribbean Coral Reefs as Con-
trolled by Wave and Light Energy Input, Bathymetric Setting and
Reef Morphology: Computer Simulation Experiments. Coral Reefs,
8:9-18.

Greer, J. E., and B. Kjerfve. 1982. “Water Currents Adjacent to Car-
rie Bow Cay, Belize.” In The Atlantic Barrier Ecosystems at Carrie
Bow Cay, Belize, I: Structure and Communities, ed. K. Ritzler, and
I. G. Macintyre, pp. 53-58. Smithsonian Contributions to Marine
Science, No. 12. Washington, D.C.: Smithsonian Institution Press.

Hendler, G. 1982. Slow Flicks Show Star Tricks: Elapsed-Time Analysis
of Basketstar (Astrophyton muricatum) Feeding Behavior. Bulletin
of Marine Science, 32(4):909-918.

Kensley, B. 1984. The Atlantic Barrier Reef Ecosystem at Carrie Bow
Cay, Belize, III: New Marine Isopoda. Smithsonian Contributions
to Zoology, No. 24. Washington, D.C.: Smithsonian Institution
Press.

Kensley, B., and R. H. Gore. 1981. Coralaxius abelei, New Genus and
New Species (Crustacea: Decapoda: Thalassinidea: Axiidae): A
Coral-Inhabiting Shrimp from the Florida Keys and the Western
Caribbean Sea. Proceedings of the Biological Society of Washing-
ton, 93:1277-1294.

Kier, P. M. 1975. The Echinoids of Carrie Bow Cay, Belize. Smithsonian
Contributions to Zoology, No. 206. Washington, D.C.: Smithson-
ian Institution Press.

Kjerfve, B., K. Riitzler, and G. H. Kierspe. 1982. “Tides at Carrie Bow
Cay, Belize.” In The Atlantic Barrier Ecosystems at Carrie Bow Cay,
Belize, I: Structure and Communities, ed. K. Ritzler, and I. G. Mac-
intyre, pp. 47-52. Smithsonian Contributions to Marine Science,
No. 12. Washington, D.C.: Smithsonian Institution Press.

Kohlmeyer, J. 1984. Tropical Marine Fungi. Marine Ecology, 5(4):
329-378.

Kokke, W. C. M. C., W. Fenical, C. Pak, and C. Djerassi. 1979. Minor
and Trace Sterols in the Caribbean Sponge Verongia cauliformis.
Helvetica Chimica Acta, 62:1310-1380.

Koltes, K. H., J. J. Tschirky, and I. C. Feller. 1998. “Carrie Bow Cay,
Belize.” In Caribbean Coastal Marine Productivity (CARICOMP):
Coral Reef, Seagrass, and Mangrove Site Characteristics, ed. B.
Kjerfve, pp. 79-94. Paris: UNESCO.

Kornicker, L. S., and A. C. Cohen. 1978. Dantyinae, a New Subfamily of
Ostracoda (Myodocopina: Sarsiellidae). Proceedings of the Biologi-
cal Society of Washington, 91(2):490-508.

Larson, R. J. 1979. Feeding in Coronate Medusae (Class Scyphozoa,
Order Coronatae). Marine Behaviour and Physiology, 6:123-129.

Lewis, S. M. 1986. The Role of Herbivorous Fishes in the Organization
of Tropical Reef Communities. Ecological Monographs, 56(3):
183-200.

Littler, D. S., and M. M. Littler. 1985. An Illustrated Marine Flora of the
Pelican Cays, Belize. Bulletin of the Biological Society of Washing-
ton, 9:1-149.

Littler, M. M., D. S. Littler, and P. R. Taylor. 1983. Evolutionary Strate-
gies in a Tropical Barrier Reef System: Functional-Form Groups of
Marine Macroalgae. Journal of Phycology, 19:229-237.

Littler, M. M., I. G. Macintyre, and D. S. Littler, eds. 1995. Biology and
Geology of Tobacco Range, Belize, Central America: A Mangrove
System Containing Unique Fractured Peat Deposits. Atoll Research
Bulletin, Nos. 426-431.

Lovelock, C. E., M. C. Ball, B. Choat, B. M. J. Engelbrecht, N. M. Hol-
brook, and I. C. Feller. 2006. Linking Physiological Processes with
Mangrove Forest Structure: Phosphorus Deficiency Limits Canopy
Development, Hydraulic Conductivity and Photosynthetic Carbon
Gain in Dwarf Rhizophora mangle. Plant, Cell and Environment,
29:793-802.

Macintyre, I. G. 1975. A Diver-Operated Hydraulic Drill for Coring Sub-
merged Substrates. Atoll Research Bulletin, 185:21-26.

. 1984. Extensive Submarine Lithification in a Cave in the Be-
lize Barrier Reef Platform. Journal of Sedimentary Petrology, 54:
221-235.

Macintyre, I. G., and R. B. Aronson. 1997. Field Guidebook to the Reefs
of Belize. Proceedings of the 8th International Coral Reef Sympo-
sium, 1:203-222.

Macintyre, I. G., R. B. Burke, and R. Stuckenrath. 1981. Core Holes in
the Puter Fore Reef off Carrie Bow Cay, Belize: A Key to the Ho-
locene History of the Belizean Barrier Reef Complex. Proceedings,
Fourth International Coral Reef Symposium, 1:567-574.

Macintyre, I. G., and K. Ritzler, eds. 2000. Natural History of the Peli-
can Cays, Belize. Atoll Research Bulletin, Nos. 466-480.

Macintyre, I. G., K. Riitzler, and I. C. Feller, eds. 2004a. The Twin Cays
Mangrove Ecosystem, Belize: Biodiversity, Geological History, and
Two Decades of Change. Atoll Research Bulletin, Nos. 509-530.

Macintyre, I. G., S. V. Smith, and J. C. Zieman. 1974. Carbon Flux
through a Coral-Reef Ecosystem: A Conceptual Model. Journal of
Geology, 8:161-171.

Macintyre, I. G., M. A. Toscano, R. G. Lightly, and G. Bond. 2004b.
Holocene History of the Mangrove Island of Twin Cays, Belize,
Central America. Atoll Research Bulletin, 510:1-16.

McKee, K. L., I. Mendelssohn, and M. Hester. 1988. A Reexamination
of Pore Water Sulfide Concentrations and Redox Potentials near
the Aerial Roots of Rhizophora mangle and Avicennia germinans.
American Journal of Botany, 75:1352-1359.

Miller, J. A., and I. G. Macintyre. 1977. Field Guidebook to the Reefs
of Belize. Miami Beach, Fla.: Atlantic Reef Committee, University
of Miami.

Modlin, R. E 1987. Mysidacea from Shallow Water in the Vicinity of
Carrie Bow Cay, Belize, with Description of Two New Species.
Journal of Crustacean Biology, 7(1):106-121.

Morton, S. L., and M. A. Faust. 1997. Survey of Toxic Epiphytic Di-
noflagellates from the Belizean Barrier Reef Ecosystem. Bulletin of
Marine Science, 61(3):899-906.

Opishinski, T., M. L. Spaulding, K. Riitzler, and M. Carpenter. 2001.
“A Real Time Environmental Data Monitoring, Management and
Analysis System for the Coral Reefs off the Coast of Belize.” In Pro-
ceedings, Oceans 2001 Conference, Nov. 5-8. Oceanic Engineering
Society and Marine Technology Society, Volume 2, pp. 1188-1197.
Honolulu, Hawaii: IEEE.

Ott, J. A., and R. Novak. 1989. “Living at an Interface: Meiofauna at the
Oxygen/Sulfide Boundary of Marine Sediments.” In Reproduction,
Genetics and Distribution of Marine Organisms, ed. J. S. Ryland,
and P. A. Taylor, pp. 415-422. Denmark: Olsen and Olsen.

Ott, J. A., R. Novak, E. Schiemer, U. Hentschel, M. Nebelsick, and M.
Polz. 1991. Tackling the Sulfide Gradient: A Novel Strategy Involv-
ing Marine Nematodes and Chemoautotrophic Ectosymbionts.
Marine Ecology, 12(3):261-279.

Pandolfi, J. M., R. H. Bradbury, E. Sala, T. P. Hughes, K. A. Bjorndal,
R. G. Cooke, D. McArdle, L. McClenachan, M. J. H. Newman,
G. Paredes, R. R. Warner, and J. B. C. Jackson. 2003. Global Tra-
jectories of the Long-Term Decline of Coral Reef Ecosystems. Sci-
ence, 301:955-958.

Pawson, D. L. 1976. Shallow-Water Sea Cucumbers (Echinodermata:
Holothuroidea) from Carrie Bow Cay, Belize. Proceedings of the
Biological Society of Washington, 89:369-382.

Robbart, M. L., P. Peckol, S. P. Scordilis, H. A. Curran, and J. Brown-
Saracino. 2004. Population Recovery and Differential Heat Shock
Protein Expression for the Corals Agaricia agarites and A. tenuifo-
lia in Belize. Marine Ecology Progress Series, 283:151-160.

Rotjan, R. D., J. L. Dimond, D. J. Thornhill, B. Helmuth, J. J. Leich-
ter, D. W. Kemp, and S. Lewis. 2006. Chronic Parrotfish Graz-
ing Impedes Coral Recovery after Bleaching. Coral Reefs, 25:
361-368.

Ritzler, K. 1975. The Role of Burrowing Sponges in Bio-Erosion. Oeco-
logia (Berlin), 19:203-216.

. 1978a. “Sponges in Coral Reefs.” In Coral Reefs: Research

Methods. Monographs on Oceanographic Methodology, No.

5, ed. D. R. Stoddart and R. E. Johannes, pp. 299-313. Paris:

UNESCO.

. 1978b. “Photogrammetry of Reef Environments by Helium Bal-

loon.” In Coral Reefs: Research Methods. Monographs on Oceano-

graphic Methodology, No. 5,ed. D. R. Stoddart and R. E. Johannes,
pp. 45-52. Paris: UNESCO.

. 1981. An Unusual Blue-Green Alga Symbiotic with Two New

Species of Ulosa (Porifera: Hymeniacidonidae) from Carrie Bow

Cay, Belize. Marine Ecology, 2:35-50.

. 1988. Mangrove Sponge Disease Induced by Cyanobacterial

Symbionts: Failure of a Primitive Immune System? Diseases of

Aquatic Organisms, 5:143-149.

. 2002. Impact of Crustose Clionid Sponges on Caribbean Reef
Corals. Acta Geologica Hispanica, 37(1):61-72.

Ritzler, K., M. C. Diaz, R. W. M. van Soest, S. Zea, K. P. Smith, B. Al-
varez, and J. Wulff. 2000. Diversity of Sponge Fauna in Mangrove
Ponds, Pelican Cays, Belize. Atoll Research Bulletin, 476:230-250.

NUMBER 38 °¢ 71

Ritzler, K., and I. C. Feller. 1988. Mangrove Swamp Communities: An
Approach in Belize. Oceanus, 30(4):16-24.

. 1996. Caribbean Mangrove Swamps. Scientific American,
274(3):70-75.

Rutzler, K., and J. D. Ferraris. 1982. “Terrestrial Environment and Cli-
mate, Carrie Bow Cay, Belize.” In The Atlantic Barrier Ecosystems
at Carrie Bow Cay, Belize, I: Structure and Communities, ed. K.
Ritzler, and I. G. Macintyre, pp. 77-92. Smithsonian Contribu-
tions to Marine Science, No. 12. Washington, D.C.: Smithsonian
Institution Press.

Rutzler, K., J. D. Ferraris, and R. J. Larson. 1980. A New Plankton Sam-
pler for Coral Reefs. Marine Ecology, 1:65-71.

Ritzler, K., and I. G. Macintyre. 1978. Siliceous Sponge Spicules in Coral
Reef Sediments. Marine Biology, 49:147-159.

, eds. 1982. The Atlantic Barrier Ecosystems at Carrie Bow Cay,
Belize, I: Structure and Communities. Smithsonian Contributions
to Marine Science, No. 12. Washington, D.C.: Smithsonian Institu-
tion Press.

Riutzler, K., D. L. Santavy, and A. Antonius. 1983. The Black Band Dis-
ease of Reef Corals, 3: Distribution, Ecology and Course of Disease.
Marine Ecology, 4(4):329-358.

Rutzler, K., and W. Sterrer. 1971. Oil Pollution: Damage Observed in
Tropical Communities along the Atlantic Seaboard of Panama. Bio-
science, 20:222-224.

Sebens, K. P., B. Helmuth, E. Carrington, and B. Aguis. 2003. Effects of
Water Flow on Growth and Energetics of the Scleractinian Coral
Agaricia tenuifolia in Belize. Coral Reefs, 22:35-47.

Signorovitch, A. Y., S. L. Dellaporta, and L. W. Buss. 2005. Molecu-
lar Signatures for Sex in the Placozoa. Proceedings of the Na-
tional Academy of Science of the United States of America,
102(43):15518-15522.

Smith, C. L., J. C. Tyler, W. P. Davis, R. S. Jones, D. G. Smith, and C. C.
Baldwin. 2003. Fishes of the Pelican Cays, Belize. Atoll Research
Bulletin, 497:1-88.

Stoddart, D. R., F R. Fosberg, and D. L. Spellman 1982. Cays of the Be-
lize Barrier Reef and Lagoon. Atoll Research Bulletin, 256:1-74.

Stoddart, D. R., and R. E. Johannes, eds. 1978. Coral Reefs: Research
Methods. Monographs on Oceanographic Methodology, No. 5.
Paris: UNESCO.

Thomas, J. D., and J. L. Barnard. 1983. The Platyischnopidae of America
(Crustacea, Amphipoda). Smithsonian Contributions to Zoology,
No. 375. Washington, D.C.: Smithsonian Institution Press.

Urish, D. W. 1988. Freshwater on a Coral Island—Carrie Bow Cay. Mar-
itime, University of Rhode Island, 32(3):4-6.

Vopel, K. C., G. Arlt, M. Pohn, and J. Ott. 2002. Flow Microenviron-
ment of Two Marine Peritrich Ciliates with Ectobiotic Chemoauto-
trophic Bacteria. Aquatic Microbial Ecology, 29:19-28.

Wainwright, P. 1988. Morphology and Ecology: The Functional Ba-
sis of Feeding Constraints in Caribbean Labrid Fishes. Ecology,
69:635-645.

Waller, T. R. 1980. Scanning Electron Microscopy of Shell and Mantle
in the Order Arcoidea (Mollusca: Bivalvia). Smithsonian Contribu-
tions to Zoology, No. 313. Washington, D.C.: Smithsonian Institu-
tion Press.

Wilkinson, C. 1987. Interocean Differences in Size and Nutrition of
Coral Reef Sponge Populations. Science, 236:1654-1657.

Winston, J. E. 1984. Shallow Water Bryozoans of Carrie Bow Cay,
Belize. American Museum Novitates, 2799:1-38.

. 2007. Diversity and Distribution of Bryozoans in the Peli-
can Cays, Belize, Central America. Atoll Research Bulletin, 546:
1-24.

Wulff, J. L. 2005. Trade-Offs in Resistance to Competitors and Preda-
tors, and Their Effects on the Diversity of Tropical Marine Sponges.
Journal of Animal Ecology, 74:313-321.

= aw SE ay yet

nie

‘ 4
er

pyle are elt eared
ee a ,

The Smithsonian Tropical Research Institute:
Marine Research, Education,
and Conservation in Panama

D. Ross Robertson, John H. Christy, Rachel
Collin, Richard G. Cooke, Luis D’Croz, Karl

W. Kaufmann, Stanley Heckadon Moreno, Juan
L. Maté, Aaron O’Dea, and Mark E. Torchin

D. Ross Robertson, John H. Christy, Rachel Col-
lin, Richard G. Cooke, Luis D’Croz, Karl W.
Kaufmann, Stanley Heckadon Moreno, Juan L.
Maté, Aaron O’Dea, and Mark E. Torchin, Smith-
sonian Tropical Research Institute, Box 0843-
03092, Balboa, Panama. Corresponding author:
R. Robertson (drr@stri.org). Manuscript received
15 August 2008; accepted 20 April 2009.

ABSTRACT. A grand geological experiment with a global reach to its biological impact,
the formation of the isthmus of Panama between 15 and 3 million years ago split the
tropical Interamerican Seaway into two and substantially changed the physical oceanog-
raphy of each part. That event subjected the now-divided halves of the neotropical ma-
rine biota to new environmental conditions that forced each along a different evolution-
ary trajectory. For the past 45 years the Smithsonian Tropical Research Institute (STRI)
marine sciences program has taken full advantage of this event by sponsoring research on
a great diversity of topics relating to the evolutionary effects of the formation of the isth-
mus. That research, which has been supported by multiple laboratories on each coast and
a series of research vessels, has produced more than 1,800 publications. Here we provide
an overview of the environmental setting for marine research in Panama and an historical
perspective to research by STRI’s scientific staff at the different marine facilities.

INTRODUCTION

The unique geological history of Panama encourages a wider variety of re-
search on tropical marine organisms than can be accomplished anywhere else in
the world. The Central American Isthmus narrows in Panama to approximately
70 km, separating oceans that have very different oceanographic regimes and
marine ecosystems. The formation of the central American isthmus, starting
approximately 15 million years ago (Ma) and finishing in Panama about 3 Ma,
had wide ramifications not only for the nature of the modern biological and geo-
logical world of the Americas but also for the entire global oceanic circulation.
With the completion of this process the Gulf Stream strengthened, changing the
Atlantic circulation. That change was soon followed by Northern Hemisphere
glaciation, which in turn brought about a period of climate change in Africa that
may have stimulated the origins of modern man. From a more local perspective,
the completion of the isthmus set in motion a vast natural experiment: single
populations of marine animals and plants were split and isolated in different
and changing environments that forced their evolutionary divergence and fun-
damental changes in their biology.

74 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

The Smithsonian Tropical Research Institute (STRI)
marine research program in the Republic of Panama has
taken full advantage of this globally significant geological
event. In 1964 STRI established its first laboratories on the
Pacific and Caribbean coasts within what then constituted
the Panama Canal Zone. Since that time, marine research at
STRI has expanded greatly and made major contributions
to understanding of tropical marine biodiversity: the geo-
logical history of the isthmus and the development of envi-
ronmental differences in the Caribbean and eastern Pacific,
patterns of biodiversity in neotropical marine habitats, coral
reef development, coral symbioses and diseases, the modes
and tempo of species formation and diversification, evolu-
tionary change within many groups of organisms relating to
environmental differences on the two sides of the isthmus of
Panama, and invasions by marine organisms facilitated by
the Panama Canal and its shipping activity. To date marine
research at STRI has resulted in more than 1,800 scientific
publications; half of these have been produced by staff sci-
entists and more than 200 published in high-profile journals
such as Science, Nature, Proceedings of the National Acad-
emy of Sciences of the United States of America, Proceed-
ings of the Royal Society, American Naturalist, Evolution,
Ecology, and Annual Review of Systematics and Ecology.

In celebration of its role in coral reef research, the
Smithsonian’s 150th anniversary, and the International Year
of the Reef, STRI hosted the Eighth International Coral
Reef Symposium in Panama in 1996. This meeting brought
1,500 reef scientists and managers to Panama from around
the world, resulting in the publication of a two-volume
proceedings (Lessios and Macintyre, 1997), and an interna-
tional traveling exhibit of coral reefs that is now resident at
the Bocas del Toro Research Station.

Here we present an overview of the marine setting of
Panama that clearly indicates its potential for research,
and a historical summary of the diversity of marine stud-
ies conducted at the different STRI marine facilities. We
then briefly outline STRI’s marine education and outreach
activities. Although this review focuses on the research ac-
tivities of STRI’s marine staff scientists, a strong fellow-
ship program and a larger suite of visiting students and
scientific collaborators have acted as substantial multipli-
ers of STRI scientists’ activities.

THE COASTAL OCEANOGRAPHIC SETTING
OF THE ISTHMUS OF PANAMA

The emergence of the Isthmus of Panama likely was
the most crucial event for tropical marine ecosystems
in the last 15 million years of earth’s history. In Cen-

tral America the marine environment experienced drastic
changes in the two seas formed by the isthmus. As the
isthmus approached closure, the Caribbean gradually be-
came cut off from the eastward flow of Pacific water and
became warmer, saltier (westward winds carried away
evaporated moisture), and more oligotrophic. The Ca-
ribbean now is a relatively stable sea, with small fluctua-
tions in temperature, relatively low tidal variation, and
a relatively high salinity. Its relatively clear and nutrient-
poor waters (D’Croz and Robertson, 1997; D’Croz et
al., 2005; Collin et al., 2009) are ideal for the growth
of coral reefs, and the wider Caribbean area ranks third
behind the Indian Ocean and the Indo-West Pacific in
terms of numbers of marine species. Annual rainfall is
high on the Caribbean coast of Panama and follows
the same basic seasonal pattern as on the lower-rainfall
Pacific side of the isthmus (Kaufmann and Thompson,
2005). Relative to the Caribbean, the Tropical Eastern
Pacific (TEP) exhibits much greater fluctuations in tides
and temperature and has substantially lower salinity as
a consquence of an area of very high rainfall along the
Intertropical Convergence Zone. The TEP also has more
and much larger areas of seasonal upwelling than the Ca-
ribbean. In addition, and in contrast to the Caribbean,
the TEP also experiences strong longer-term variation
in temperature and productivity from the influence of El
Nino-Southern Oscillation Cycle (ENSO) events (D’Croz
and O’Dea, 2009). Sea warming related to ENSO (which
occurs at intervals of four to nine years) has drastic af-
fects on coral reef development in the TEP. The direct
marine effects of ENSO events in the tropical and warm
temperate parts of the eastern Pacific are stronger than
anywhere else in the world. Although coastal biological
productivity is strongly related to benthic communities in
the Caribbean, pelagic productivity and high availability
of ocean-derived dissolved nutrients dominate the TEP,
with high variability in those nutrient levels producing
matching variability in the abundance of pelagic organ-
isms (Miglietta et al., 2008). In Panama the nutrient-rich
waters of its Pacific coast support commercial fisheries of
major importance, fisheries that have no counterpart on
the Caribbean coast.

The coastal marine communities of Panama are af-
fected not only by inter-ocean differences in oceanog-
raphy but also by marked local variation in shoreline
environmental conditions along each coast. The Pacific
shelf of Panama is wide and is divided, by the southward-
projecting Azuero Peninsula, into two large areas, the
(eastern) Gulf of Panama and the (western) Gulf of
Chiriqui. The Gulf of Panama is subject to strong sea-
sonal wind-driven upwelling, but the Gulf of Chiriqui is

not (Figure 1; and see D’Croz and Robertson, 1997). In
the latter Gulf, high mountains block the wind and pre-
vent wind-induced upwelling (D’Croz and O’Dea, 2007).
In contrast, the Caribbean coast of Panama is relatively
straight and has a narrow continental shelf, except in the
(western) Bocas del Toro Archipelago. Hydrological con-
ditions vary substantially along the Caribbean coast, rang-
ing from the nutrient- and plankton-poor waters in the

Eastern Pacific
Warm Pool

21

NUMBER 38 e¢ 75

(eastern) San Blas Archipelago, where river discharge is
low and the influence of open ocean water is high (D’Croz
et al., 1999), to the more turbid environments of the Bocas
del Toro Archipelago, where rainfall! and river discharge
are higher as a result of the blockage of westward mois-
ture flow by the highest mountains in Panama (D’Croz et
al., 2005; Collin et al., 2009). Thus, Panama lays claim to
having “four oceans,” providing unique opportunities for

Caribbean Sea

Costa Rica
Sea Surface Temperature (°C)

(Cano Is.)

Pacific Ocean

FIGURE 1. Temperature regimes on the Atlantic and Pacific coasts during the seasonal upwelling in the Gulf of Panama.

76 ¢* SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

understanding how and why marine ecosystems vary and
function as they do.

A HISTORICAL RESUME OF RESEARCH
AT STRI MARINE FACILITIES

Marine research began at STRI in 1961 with the doc-
toral work of STRI Director (Emeritus) Ira Rubinoff on
trans-isthmian sister species of fishes (Rubinoff and Ru-
binoff, 1971), which led to STRI’s first marine publica-
tion, based on work done near Punta Galeta (Rubinoff
and Rubinoff, 1962). Since then the marine program has
undergone exponential growth in its productivity. STRI
currently operates two land facilities on the Caribbean
coast of Panama, Punta Galeta Laboratory and Bocas del
Toro Research Station, and two on the Pacific coast, Naos
Island Laboratory complex and Rancheria Island field
station. Between 1977 and 1997 STRI also operated a
small, highly productive field station in the San Blas islands
(Figure 2). In addition, STRI has maintained a series of
small coastal research vessels that greatly expanded the
geographic reach of its activities well beyond STRI’s land
facilities and, in fact, far beyond Panama.

MARINE ENVIRONMENTAL SCIENCES PROGRAM (MESP)
Monitoring the Physical Environment

In 1974, the Smithsonian Institution Tropical En-
vironmental Sciences Program began monitoring physi-
cal environmental variables at Galeta, recording rain-
fall, wind speed and direction, solar radiation, reef flat
water level, and air and water temperature hourly with
automated equipment. Today, such physical data are re-
corded more frequently, automatically sent to a central
processing facility via radio and internet, and added to a
database that is available online at http://striweb.si.edu/
esp/physical_monitoring/index_phy_mon.htm. Organiza-
tion of physical data collection from Galeta has now been
combined with that from Barro Colorado Island, Bocas
del Toro, and several other sites, under the management
of Karl Kaufmann. Recording of sea-surface temperature
started at Galeta in 1988, and this monitoring now covers
33 shallow-water stations throughout the coastal waters
on both coasts of Panama. Published summaries of the
marine physical data include Meyer et al. (1974), Cubit et
al. (1988), and Kaufmann and Thompson (2005). Physical
environmental monitoring was conducted at the San Blas
station from 1991 until its closure in 1998 and has been in
progress at Bocas del Toro since 1999.

Monitoring the Biological Environment

The first phase of biological monitoring consisted of
work done at Galeta that was stimulated by the two oil
spills and formed part of their resultant studies. At San
Blas, biological (plankton) monitoring co-occurred with
the physical monitoring. At Bocas del Toro, biological
monitoring that started in 1999 includes minor activity
directed at seagrasses and mangroves in connection with
CARICOMP. The major activity however, has been an ex-
panding set of monitoring surveys of coral reefs by Hec-
tor Guzman, which now cover reefs scattered along both
coasts of Panama (see http://striweb.si.edu/esp/mesp/reef
_monitoring_intro.htm). This program developed from a
survey of coral reefs in the general vicinity of Galeta made
in 1985 (Guzman et al., 1991; and see also Guzman et al.,

2008b).
GALETA POINT MARINE LABORATORY

The Galeta Point installation became a STRI labora-
tory in 1964 when a military building constructed in 1942
on a fringing reef flat was turned over to STRI, thanks
to the efforts of Ira Rubinoff. From his research on in-
shore fishes in that area (Rubinoff and Rubinoff, 1962)
Rubinoff recognized its value as an easily accessible Carib-
bean study site. By 1971 Galeta Point was STRI’s primary
marine research site, providing access to a fringing reef
flat, seagrass beds, and mangroves within a few meters of
a laboratory building, with housing in nearby Coco Solo.
Early work emphasized the comparison of reefs on both
sides of the isthmus (Glynn, 1972) and the geological
structure and history of the reefs (Macintyre and Glynn,
1976). Fundamental insights into differences between the
Caribbean and eastern Pacific at Panama also were devel-
oped by Chuck Birkeland (Birkeland, 1977) when, during
his long-term residence at Galeta, he analyzed patterns of
biomass accumulation on settling plates deployed on both
sides of the isthmus.

Permanent monitoring of the biota at Galeta Point
was started in 1970 by Chuck Birkeland, David Meyer,
and Gordon Hendler to provide baseline data on a tropi-
cal Caribbean reef flat; this was done to determine the ef-
fect of the Witwater oil spill, which occurred in December
1968 about 5 km east of Galeta. Because no baseline data
were available to determine effects of that spill on reef
communities, the US Environmental Protection Agency
provided funds to set up the study and to perform experi-
ments testing susceptibility of corals to oil. Transects were
established at both Galeta Point and Punta Paitilla on the

77

ER 38

NUMB

“SonToey OUTTVU (NYLS) aNINsuy Yorvasay feoidosy, urTuosy US Jo UOANGIAsSIpP ay} SuMoys eureUe,Y Jo dey *7 AANO

uoneas ple}4 seig ues

40 \ue> ounsen “sj eaqoyn>

SONALUO}}

| Woneas Piety ©1410

| UoNers yr1eesey ser0g

SaizIpIrDey aulsaeyw
®INIIFSUT YXARAaSaY PerxIdoA’] UeLUOSYyyIWS

78 e¢ SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Pacific side to compare community structure, recruitment
patterns, and the effect of oil on both communities (Birke-
land et alt, 11973).

In April 1986 a storage tank ruptured at an oil refin-
ery about 4 km east of the laboratory, spilling 60,000-
100,000 barrels of oil into the sea. The reef flat, grass
beds, and mangroves around the Galeta laboratory were
heavily oiled. This time a substantial amount of baseline
data was available, thanks to the original transects set up
by Birkeland, Meyer, and Hendler, whose monitoring had
ended in 1982, and to the wider reef surveys in that area
by Ernesto Weil. The Minerals Management Service of the
U.S. Department of the Interior bestowed a 5 year grant
to STRI to study the effects of the second oil spill in tropi-
cal areas (Keller and Jackson, 1993). That effort involved
a considerable expansion of the types of data gathered,
organisms studied, and habitats monitored over those in
the first oil spill study. Subsequent to the second oil spill
study the center of STRI research on coral reefs shifted
first to San Blas then to Bocas del Toro. A long-term study
(since 1988) of mangrove forest dynamics by Wayne
Sousa (Sousa, 2007), occasional short-term projects, and
physical environmental monitoring by MESP (see below)
have continued at Galeta. The site also supports public
education and outreach programs organized by Stanley
Heckadon (see below). To date 315 publications include
data obtained at Galeta laboratory, and the lab itself has
produced 288 marine publications.

SAN BLAS FIELD STATION

The sparsely populated San Blas archipelago, in the au-
tonomous Kuna Yala comarca, consists of several hundred
sand cays scattered along the relatively sparsely populated
eastern third of the Caribbean coast of Panama. The archi-
pelago has the richest and most extensive development of
coral reefs and associated fauna (including reef fishes) and
flora in Caribbean Panama. Marine research sponsored
by STRI began in San Blas in 1970, and research activity
increased greatly in the late 1970s following the gradual
construction by STRI from 1977 onward of a small field
station that provided basic living accommodations and so
allowed year-round research.

The San Blas field station, with its year-round access
to a 15 km? area of rich reefs in calm, clear water, was the
Caribbean base for many of STRI’s comparative studies of
the biology of closely related organisms on the Atlantic and
Pacific sides of the Isthmus of Panama. Early research by
STRI staff in San Blas included studies by STRI’s founding
director, Martin Moynihan, on the behavior of cephalo-

pods (Moynihan, 1975; Moynihan and Rodaniche, 1982)
and by Peter Glynn in the 1970s on coral reef development
(Glynn, 1973). These investigations were followed by oth-
ers on a broad range of organisms: Ross Robertson on
the sexual patterns of labroid fishes, with Robert Warner
(Warner et al., 1975; Robertson and Warner, 1978; War-
ner and Robertson 1978), and the recruitment dynamics
and demography of reef-fishes (Robertson et al., 1999,
2005); Haris Lessios on the evolution and biology of
echinoderms on the two coasts of the isthmus of Panama
(Lessios, 1979, 1981, 2005); deputy director Eric Fischer
on the sexual biology of simultaneously hermaphroditic
groupers (Fischer, 1980, 1981; Fischer and Petersen,
1987); Nancy Knowlton on the biology and evolution of
snapping shrimps and the reproductive biology, coral—
algal symbioses, and evolution of corals (Rowan et al.,
1997; Knowlton et al., 1977, 1992; Knowlton and Weigt,
1998); Jeremy Jackson on the comparative population and
reproductive biology and evolutionary history of bryozo-
ans on both sides of the Isthmus of Panama (O’Dea and
Jackson, 2002; Dick et al., 2003; O’Dea et al., 2004); Luis
D’Croz on comparative oceanographic conditions on the
Caribbean and Pacific coasts of Panama (D’Croz and Rob-
ertson, 1997); and Hector Guzman on coral reef distribu-
tion and conservation (Andrefouet and Guzman, 2005).
During this period STRI also sponsored several anthro-
pological projects on traditional Kuna society, acted as a
conduit for international funding of Kuna marine manage-
ment and conservation activities, and provide fellowships
to Kuna University students.

The San Blas station provided essential support for
projects on long-term ecological change on surrounding
coral reefs. The combination of ease of access to shallow
reefs, access as good as anywhere in the world, and the
ability to do much work while on snorkel rather than
scuba meant that it was possible to accumulate enormous
data sets involving daily or shorter time period observa-
tions over months or years. These kinds of data are all too
rarely available for tropical marine systems.

In early 1983 a Caribbean-wide mass die-off of an
ecologically key organism on Caribbean reefs, the black
sea urchin Diadema antillarum, began near San Blas and
spread within the year throughout the entire Greater Ca-
ribbean. The year-round presence of biologists conducting
long term studies of reef organisms at STRI’s field station
enabled the documentation of the start and spread of that
event, which produced large, long-term effects on algal and
coral growth on Caribbean coral reefs. Haris Lessios has
followed the population and evolutionary consequences
of that event for the urchin since it started (Lessios et al.,

1984; Lessios, 2005). Year-round monitoring of reef-fish
populations on reefs around that station over a 20 year
period contributed key information to a meta-population
study that documents a gradual Caribbean-wide decline
in the overall abundance of reef fishes since the Diadema
dieoff (Paddack et al., 2009). Long-term monitoring of cli-
matic and oceanographic conditions by MESP enabled de-
tailed examination of linkages between environmental dy-
namics and the dynamics of recruitment of pelagic larvae
of reef-fishes (Robertson et al., 1999). In addition regular
station visitors accumulated the world’s only long-term
data sets on gorgonians and sponges. The former work in-
cludes data on a combination of population dynamics and
genetics of clone structure obtained by Howie Lasker and
Mary-Alice Coffroth (Coffroth et al., 1992; Lasker, 1991;
Lasker et al., 1996). The latter work includes data on the
dynamics of sponge communities collected by Janie Wulff
(Wulff, 1991, 1997).

Over the 20 years of its existence, research supported
by the San Blas field station produced 363 publications
on the biology of plants and animals living on the coral
reefs around the station, at a peak annual operating cost
of about US$100,000. The cheapness of this operation
provides a startling example of how effective a small sta-
tion can be for very little expense, so long as the necessary
tools for field research are supplied: grass huts for living,
rainwater for drinking and washing, communal kitchens,
small boats, a scuba compressor, and, above all, field sites
in calm clear water at the station’s doorstep.

Local political events in this autonomously governed
indigenous reserve led to the closure of the San Blas station
in 1998. Although this closure terminated the activities of
STRI staff scientists in that area for some time, several ex-
ternal researchers were able to make private arrangements
to continue their work there. After the closure of the San
Blas station the center of STRI’s Caribbean research ef-
forts moved to Bocas del Toro Province, at the opposite
end of the Caribbean coast of Panama.

BOCAS DEL TORO RESEARCH STATION (BRS)

The Smithsonian Institution (SI) has a long history
of terrestrial and geological research in Bocas del Toro
Province. In the 1970s and 1980s Charles Handley of the
Natural History Museum mounted a number of expedi-
tions to the province to survey the mammal and bird fauna
(Handley, 1993; Anderson and Handley, 2002). This phase
was followed with ground-breaking geological work by
STRI’s deputy director, Anthony Coates. He used the rock
outcrops in the province, which contain the most complete

NUMBER 38 ¢ 79

record of marine environments of the last 10 million years
in the southern Caribbean, to help clarify events associ-
ated with the rise of the Isthmus of Panama (Collins et al.,
1995; Collins and Coates, 1999; Coates et al., 2005).

In 1998 STRI purchased 6 hectares (ha) just outside
the town of Bocas del Toro on Isla Colon. A dormitory
was built on the site in 2001 and a modern, well-equipped
laboratory in 2003. The BRS now houses 28 resident sci-
entists and will soon add accommodation for 16 more.
BRS can now host approximately 325 scientific visitors
from more than 30 countries each year: 40% undergradu-
ates, 25% graduate students, 10% postdoctoral fellows,
and 25% researchers. About half the postdoctoral fellows
and researchers are SI scientists. Research at the station
has resulted in 201 scientific publications in the five years
since its inauguration in 2003, with Rachel Collin as its
director.

The BRS is now among the preeminent research sta-
tions in the Caribbean. It is better equipped and provides
access to a larger diversity of habitats than almost any
other research facility in that region. The wealth of natu-
ral diversity available near BRS combined with technical
support facilities is reflected in the wide range of research
projects that are conducted at the station. Significant re-
search has focused on the coral bleaching response to stress
and disease. These studies have shown that sugars are one
of the most damaging components in pollution from rain
runoff (Kline et al., 2006) and that coral disease is related
to temperature stress. An SI fellow identified candidate
genes that participate in coral’s bleaching response to el-
evated temperature (DeSalvo et al., 2008). Research at the
laboratory also has shown that some coral disease and
death in the Caribbean results from a protozoan infection.
Another strong line of research at the BRS is the investi-
gation of the factors that lead to speciation in the marine
environment. Groundbreaking work on hamlet fishes has
shown that mate choice based on color pattern may drive
divergence and that color patterns may evolve via aggres-
sive mimicry, a previously undemonstrated mechanism of
diversification (Puebla et al., 2007, 2008).

The BRS is also a local focus of taxonomic work and
studies aimed at documenting marine biodiversity that
were published in a special issue of the Caribbean Journal
of Science dedicated to the marine environment and fauna
of Bocas del Toro (Collin, 2005a, 2005b). Extensive work
has been done there on the taxonomy of marine shrimps
(Anker et al., 2008a, 2008b, 2008c). Bocas del Toro is a
global hotspot of shrimp diversity and ranks within the
top 10 sites in the world. More than 20 new shrimp species
from Bocas del Toro have been described in the past five

80 e¢ SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

years. New species of other marine organisms, including
snails, tunicates, sponges, flatworms, and meiofauna, have
also been described on the basis of work at the BRS. As
a result of these taxonomic and faunal studies, Bocas del
Toro has the highest recorded tunicate diversity anywhere
in the Caribbean and the third highest sponge diversity.
Other long-term projects based at BRS include stud-
ies focused on breeding success at major Caribbean turtle
nesting beaches, the effects of noise polluction and tour
boat operations on dolphin vocalization and behavior, ef-
fects of anthropogenic substrates, such as docks, on inva-
sive tunicate abundance, effect of nutrient limitation on
mangrove forest structure and diversity, emerging sponge
diseases, and Caribbean-wide speciation in Montastraea
corals caused by temporal shifts in spawning cycles.

NAos ISLAND LABORATORIES

Naos is one of a cluster of four islands at the end of a
2 km long causeway at the Pacific entrance to the Panama
Canal. STRI’s first marine laboratory was established there
in 1964 in an old military bunker and has since expanded
to four buildings, three of them ex-US Navy buildings re-
furbished by STRI. This laboratory provides ready access
to the upper bay of Panama with its extensive mangroves
and a scattering of inshore islands, plus the coral reefs of
the Perlas Archipelago, 50 km away. The laboratory com-
plex, with a flow-through aquarium system, diving locker,
small boat support, research vessel, and molecular labora-
tories, supports a wide range of research by all the marine
scientific staff. Organismal studies based primarily at Naos
cover or have covered the following topics: the Panama
Canal as a hub for marine invasions, rocky intertidal com-
munity ecology, physiological ecology, behavioral ecology
of intertidal organisms, coral reef development in the TEP,
molecular evolution of marine organisms, life history evo-
lution and evolution of mode of development, and marine
zooarchaeology.

Panama as a Hub for Marine Invasions

Biological invasions are a potent force for change
across the globe. Once established, introduced species can
become numerically or functionally dominant, threaten-
ing native biodiversity and altering ecosystem processes.
The flip side to the damage they cause is that introduced
species can provide opportunities for insight as large-scale
experiments to understand ecological and evolutionary
processes. In marine and coastal environments, shipping
is a major pathway for biological invasions and appears

largely responsible for the recent dramatic increase in
invasions.

Beginning with the studies of Hildebrand (1939) in the
1930s, followed by several investigations surrounding the
potential problems associated with the construction of a
sea-level Canal in Central America (Rubinoff, 1965, 1968;
Rubinoff and Rubinoff, 1969), STRI has been central in
evaluating the role of the Canal as a passageway for shore-
fishes. Interestingly, despite the Canal’s 100 years of exis-
tence and the occurrence of approximately 1,500 species
of marine and brackish-water fishes on the two coasts of
Panama, only a handful of such species have successfully
passed through the Panama canal and established popula-
tions in the “other” ocean. For example, only 8 species
of such successful immigrants are known in the tropical
eastern Pacific and only 3 have spread beyond the immedi-
ate confines of the Pacific entrance to the canal. Why are
there so few successful invasions through a short, suitable
corridor? Why do some invasions fail and others succeed?
Panama and its canal have much to offer studies aimed at
determining success or failure of invasions.

STRI is ideally situated to study marine and coastal
invasions. Panama is one of the world’s largest hubs for
shipping. The Canal serves as an aquatic corridor con-
necting the Atlantic and Pacific Ocean basins, and ports
on either side serve as hubs for international trade. Since
its opening in 1914, approximately 800,000 ocean-going
commercial vessels have passed through the Canal. To-
day, approximately 12,000 to 14,000 commercial ships
transit the Canal annually (Ruiz et al., 2009). Moreover,
Panama is initiating a major effort to expand shipping in
the Canal by constructing additional locks on the Pacific
and Atlantic coasts. Although the freshwaters of Lake Ga-
tun, a large lake that constitutes the bulk of the canal,
have strongly limited the inter-oceanic invasions of purely
marine species, the new locks being added to the canal
have the potential to increase the salinity of Gatun Lake
and increase such interchange. With the Naos and Galeta
marine laboratories strategically located at the Pacific and
Atlantic entrances to the Canal, STRI is well positioned
to continue to conduct a broad range of basic research on
marine invasions.

In contrast to the limited exchange of fishes across
the Isthmus, various introduced invertebrate species have
been documented recently in the Canal, underscoring the
fact that invasions are occurring. Some examples include
a North American mud crab that has established a popu-
lation in the Panama Canal expansion area (Roche and
Torchin, 2007) and an invasive Japanese clam that reaches
densities greater than 100 m~ in the Canal, as well as an

invasive snail that is known to host medically important
trematode parasites. Although there are likely other such
species, with few exceptions (Abele and Kim, 1989) inver-
tebrate diversity of the Canal remains largely unexplored.
Recently, STRI and SERC scientists teamed up to evaluate
the role of the Canal in biological invasions and determine
how patterns and processes driving invasions in tropical
and temperate regions may differ.

Although the potential for invasions in Panama is likely
to be high, with the exception of studies on fishes that have
passed through the Canal in the past 40 to 50 years, we
know surprisingly little about other coastal invasions that
have resulted, and many established invasions probably
have been overlooked (Miglietta and Lessios, 2009). With
the current expansion of the Panama Canal, evaluating the
importance of the Canal in regional and global invasions
is arguably an imperative goal for the conservation of our
coastal resources and an ideal opportunity to illuminate
our understanding of biological invasions.

Rocky Intertidal Community Ecology

The rocky intertidal zone of the TEP appears to be
largely bare rock, with very little macroalgal cover and few
sessile invertebrates, which are not distributed in clear zones
according to tidal height or wave exposure. The striking
contrast between this system and those of temperate North
America and Europe, which are well vegetated and have
an abundance of invertebrates in regularly arranged zones,
drew researchers such as Jane Lubchenco (currently director
of the NOAA) and Bruce Menge to STRI in the late 1970s to
seek an explanation. Their field exclusion experiments indi-
cated that year-round predation and herbivory by a diverse
community of highly mobile fishes, crabs, and mollusks
forces their prey into refugia in cracks and under boulders
and regulates directly, or indirectly, species interactions such
that species capable of dominating space are kept in check
(Menge and Lubchenco, 1981; Menge et al., 1986).

Physiological Ecology

The marine environment of the eastern Pacific is much
more variable than that of the Caribbean, especially so
during upwelling and in shallow-water and intertidal hab-
itats. Temperatures in tidal pools at Naos range between
approximately 18°C and more than 50°C. Jeffery Graham
made contributions to basic understanding of how fishes
and sea snakes contend with this and other physiologi-
cal challenges in the TEP (Graham, 1970, 1971) and later
investigated heat regulation in tunas (Graham, 1975). Ira

NUMBER 38 e¢ 81

Rubinoff, together with Graham and Panamanian cardi-
ologist Jorge Motta, performed pioneering work on the
temperature physiology and diving behavior and respira-
tory physiology of the neotropics’ only sea snake species
(Graham et al., 1971; Rubinoff et al., 1986).

Behavioral Ecology of Intertidal Animals

Marine behavioral and estuarine (soft-bottom) ecol-
ogy has been the focus of long-term research programs by
John Christy and his students on the reproductive ecology
(larval release cycles in relation to predation risks; Mor-
gan and Christy, 1995) and behavior (burrow ornaments
as sexual signals; Christy et al., 2002) of intertidal crabs,
particularly fiddler crabs. The latter reach their highest
species diversity in the world on the Pacific coast of Cen-
tral America (Sturmbauer et al., 1996). Christy recently
completed five years of daily observations of the reproduc-
tive timing of a fiddler crab on Culebra beach, the results
of which demonstrate that these crabs have a remarkable
ability to track, on several time scales, complex variation
in environmental conditions suitable for larval release. Re-
search by Christy’s lab on mechanisms of mate choice in
fiddler crabs has shown that male courtship signals elicit
responses in females that have been selected by predation,
not because the signals lead to choice of the best male as
a mate. This research has provided the best empirical sup-
port to date for the “sensory trap” mechanism of sexual
signal evolution (Christy, 1995; Backwell et al., 2000, Kim
et al., 2009). Together with work by terrestrial biologists
at STRI, this research has made STRI a global center for
the study of the evolution of sexual signals.

Coral Reef Development
in the Tropical Eastern Pacitic (TEP)

Following the closure of the isthmus, different com-
ponents of the tropical biota of the TEP reacted in differ-
ent ways to resultant dramatic changes in the local ma-
rine environment. Most of the coral fauna was wiped out
(~85% of the current, depauperate fauna is derived from
Indo-Central Pacific immigrants), probably largely by ex-
treme environmental fluctuations during ENSO events.
Documentation of effects of environmental changes on
coral reef development in that area has been the focus of
35 years of studies by Peter Glynn and his collaborators,
not only in Panama but also further afield in the TEP in
places such as the Galapagos (Glynn et al., 1979; Glynn
and Wellington, 1983). STRI research on Panama’s Pacific
coral reefs began in the earlier 1970s, when, contrary to

82 ¢ SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

previous ideas, fully developed coral reefs were found in
the Gulfs of Panama and Chiriqui (Glynn, 1972; Glynn
et al., 1972; Glynn and Stewart, 1973). It also became
evident that differences in reef growth in those gulfs were
related to their different temperature regimes (Glynn and
Stewart, 1973). Coral reefs in the Gulf of Panama are
mainly confined to the warmer sides of the Pearl islands
and grow at lower rates than reefs in the year-round
warmth of the Gulf of Chiriqui (Glynn and Macintyre,
1977). The latter reefs grow at rates comparable to those
on the Caribbean coast of Panama (Macintyre and Glynn,
1976), and corals in each gulf differ in their responses to
temperature (D’Croz et al., 2001; Schloeder and D’Croz,
2004). A major thrust of work on TEP reefs has been to
understand the effects of ENSO warming events on the
survival and dynamics of reef ecosystems. Observations
linked coral bleaching in Panama to high-temperature
anomalies of the severe 1982-1983 ENSO (Glynn et
al., 1988; Glynn and D’Croz, 1990; Glynn et al., 2001).
Such bleaching led to region-wide mass coral mortality
during the intense 1982-1983 and 1997-1998 ENSO
events (Glynn, 1984; Glynn et al., 2001). Microcosm ex-
periments at STRI confirmed that temperature stress pro-
duced bleaching and mass mortality of corals (Glynn and
D’Croz, 1990) and that slow-growing massive species are
more resistant than branching types to temperature-induced
beaching (Huerkamp et al., 2001). There has been continu-
ous monitoring of reef recovery since the mass coral mor-
tality produced by the 1982-1983 ENSO, providing one of
the longest term databases of this type in the world (Glynn,
1984, 1990; Glynn and Colgan, 1992; Glynn et al., 2001).
Major efforts have also been made to investigate the re-
productive ecology of corals, relating fecundity, spawning
activity, and recruitment of surviving species to community
recovery and reef resilience in Pacific Panama (Glynn et al.,
1991, 1994, 1996, 2008; Colley et al., 2006; Manzello et
al., 2008). Bleaching patterns have been related not only to
the diversity of zooxanthellae symbionts of corals (Glynn
et al., 2001) but also to coral genotypes (D’Croz and Maté,
2004), with both factors likely playing an important role
in adaptive responses by corals to climate change. Research
on corals in Pacific Panama additionally involves the tax-
onomy and biogeography of gorgonian soft corals (Vargas
et al., 2008; Guzman and Breedy, 2008).

Molecular Evolution of Marine Organisms
STRI has played a leading role in development of mo-

lecular techniques for studies of marine organisms, not
only in relationship to trans-isthmian biology of neotropi-

cal taxa (reviewed by Lessios, 2008) but also in studying
the global biogeography of pantropical groups. A 30 year
history of such work, the longest in SI, began with studies
of sea urchins by Haris Lessios (Lessios, 1979). That work,
although centered at the molecular laboraties at Naos
Laboratory, has relied on all other STRI marine facilities
for collections and maintenance of live organisms. Since
that start, molecular evolution studies at STRI have under-
gone explosive growth. Such studies include assessments
of effects of the rise of the isthmus on the ecology and biol-
ogy of neotropical organisms (Collin, 2003a) and patterns
and processes involved in the evolutionary divergence of
such taxa (Knowlton and Weigt, 1998; Hurt et al., 2009).
Molecular studies also have led to the delineation of spe-
cies boundaries in marine organisms (Knowlton, 2000)
and understanding of global historical biogeography of
pantropical groups (Lessios et al., 1999, 2001; Collin,
2003a, 2003b, 2005a; Quenoiville et al., 2004), invasions
of the tropical Atlantic by Indo-Pacific taxa around south-
ern Africa (Bowen et al., 2001; Rocha et al., 2005a), pat-
terns of dispersal among different tropical biogeographic
regions within the Atlantic (Lessios et al., 1999; Rocha
et al., 2002, 2005b), physiological mechanisms involved
in species formation (McCartney and Lessios, 2002; Zig-
ler and Lessios, 2004), non-allopatric speciation within
biogeographic regions (Rocha et al., 2005a; Puebla et al.,
2007, 2008), patterns and processes involved in speciation
of corals (Fukami et al., 2004), and the history of two-way
transfers of species across the 4,000-7,000 km wide East-
ern Pacific Barrier, the world’s widest stretch of deep open
ocean (Lessios and Robertson, 2006). Molecular evolu-
tion studies at STRI have produced 163 marine-themed
publications to date.

Marine Archaeology: Historical Human
Reliance on Marine Resources in Panama

Zooarchaeology has played an important role in
STRI’s anthropology program for the past 40 years (Lin-
ares and Ranere, 1980) through studies originated by Rich-
ard Cooke of pre-Columbian usage of marine resources
in Panama, primarily in Panama Bay (Cooke, 1981). The
expanding reference collection of 1,540 skeletons of 340
species of fishes and other organisms used in this research
has also enhanced knowledge of the zoogeography of these
organisms (Cooke and Jiménez, 2008b). This work has
charted the time course of geographic changes in patterns
of marine resource usage in Panama Bay. By 7,000-4,500
bp, humans on the shores of that bay exploited a wide
variety of inshore marine resources, including more than

80 species of marine fishes (bony fishes, sharks, sawfish,
sting rays) taken in a variety of different habitats (beaches,
mangroves, estuaries, reefs, open water) using various
methods (hook-and-line, nets, stationary wood-and-stone
traps) (Cooke, 1992; Cooke and Jiménez, 2004, 2008a;
Cooke et al., 2008). Other marine resources used include
sea turtles, dolphins, manatees, and seabirds. The ritual
importance of marine animals in pre-Columbian Panama
is underlined by frequent images of sea turtles, fish, and
marine invertebrates on pottery and goldwork (Linares,
1977; Cooke, 2004a, 2004b). Although currently there
is no convincing zooarchaeological evidence for overfish-
ing in pre-Columbian times in Panama, ongoing research
in the Pearl Islands seems likely to identify pressures that
produced changes to populations of mollusks and reef fish
around individual islands. Intensive collection of color-
ful marine shells and marine birds for making ornaments
likely led to local impacts on populations of these taxa.

RANCHERIA ISLAND FIELD STATION

Rancheria Island is situated in the center of the largest
and best managed marine reserve in Panama: the Coiba
National Park (and World Heritage Site) in the Gulf of
Chiriqui. The park area has a long history of environ-
mental protection (Coiba acted as a “free-range” prison
island for almost 85 years) and hosts the largest area of
coral reefs and richest [number of species] accumulation
of corals on the entire continental shore of the TEP. A tiny,
relatively undeveloped field station at Rancheria has sup-
ported research on coral reefs in the surrounding area by
Peter Glynn and his collaborators (see above).

THE RESEARCH VESSELS

Four vessels were operated by STRI between 1970 and
2008: the 65-foot Tethys (1970-1972), the 45-foot RV
Stenella (1972-1978), the 63-foot RV Benjamin (1978-
1994), and the 96-foot RV Urraca from 1994 to 2008.
None of those vessels was purpose built. The equipping of
the Urraca, after its purchase, with an A-frame and ocean-
ographic winch allowed intensive trawling and dredging
activities (to depths of 250 m) and thus greatly extended
the range of studies that could be supported beyond the
previous emphasis on scuba-based research. These re-
search vessels, and particularly the Urraca, enabled field-
work in remote locations that lacked land bases for marine
research and thus vastly extended the geographic reach of
STRI biologists. The Urraca acted as such a base not only
throughout Panama’s territorial waters but also in locali-

NUMBER 38 e¢ 83

ties as far afield as Clipperton Island (1,000 km west of
Acapulco) in the Pacific (Robertson and Allen, 2008) and
Honduras in the Caribbean (Guzman, 1998).

To date, 14 years service by the Urraca has produced
350 scientific publications. Research supported by the Ur-
raca proved vital to the declaration of two large Marine
Protected Areas (MPAs) on the Pacific coast of Panama,
principally through the research activities of H. Guzman
on coral diversity and conservation (see below). Urraca
support of collecting along the entire Pacific coast of Pan-
ama, as well as Costa Rica, Clipperton and Cocos Islands
(remote oceanic islands in the eastern Pacific), and El Sal-
vador was essential for the development of the world’s first
online information system on a regional shorefish fauna
(www.stri.org/sftep). In addition the Urraca provided ex-
tensive and extended support to the Panama Paleontology
Program (see below) and for collecting fishes (Berming-
ham, Robertson), echinoderms (Lessios), soft corals (Guz-
man), and mollusks (Collin) for taxonomic and evolution-
ary studies, and hydrologic surveys (D’Croz).

HISTORICAL MARINE ECOLOGY:
THE PANAMA PALEONTOLOGY PROJECT

STRI is unique in having an institutional marine pro-
gram that includes both biology and geology, as well as
a series of strong programs in various aspects of tropical
terrestrial biology. Intellectual cross-fertilizations between
scientists steeped in terrestrial and marine systems have
maintained STRI as a place known for creative research.

The striking differences in environmental conditions
and ecology from opposite sides of the Isthmus of Panama
today, and their changes over time during Isthmus closure,
provides marine paleontologists with a “natural experi-
ment” with which to address, on an evolutionary and eco-
logically large scale, the impact of environmental change
and genetic isolation on marine invertebrate faunas. In
1986 the Panama Paleontology Project (PPP) was initiated
by Jeremy Jackson and Anthony Coates. Their aim was to
survey coastal sediments of the isthmian area to establish
if the fossil record were sufficiently complete to explore
the evolutionary responses of marine communities to the
gradual emergence of the Isthmus of Panama.

Stratigraphically complete Neogene deposits were
soon discovered in the Panama Canal basin and Bocas
del Toro, and excavations were subsequently extended
to several other richly fossiliferous regions of Panama
and Costa Rica, Venezuela, Ecuador, Jamaica, and the
Dominican Republic. In addition, large-scale benthic

84 e¢ SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

surveys of modern shallow-water communities across the
Caribbean and Tropical Eastern Pacific serve as a baseline
for understanding biotic changes through geological time.

The PPP has so far involved more than SO scientists
from 20 institutions in seven countries and undertaken
almost 40 expeditions to eight countries. The resultant
collections comprise thousands of replicated samples and
many millions of individual, quantitatively collected fossil
specimens. The rigorous paleontological framework of the
PPP presents evolutionary biologists with a unique view
of 15 million years of life and environments in a tropical
region. Using these samples and framework, the PPP has
documented the environmental, lithologic, and biological
changes in Isthmian nearshore marine habitats from 15
Ma to the present day, producing almost 200 publications
to date (see http://www.fiu.edu/~collinsl/pppcon.html).

Placing igneous and sedimentary rock formations in
sequence established a high-resolution stratigraphic sys- ONG EAN
tem that was critical to effectively reconstruct patterns
of biological change (Coates et al., 1992, 2005; Collins
et al., 1996b; Collins and Coates, 1999). Aligned with
taxonomic and paleoenvironmental analyses, these geo-
logical studies also permit reconstructions of land and
water masses as the isthmus shoaled, providing dates of
final closure that are essential for estimates of the timing
of divergence of modern marine organisms (Collins et al.,
1995; Coates and Obando, 1996) (Figure 3).

Data from PPP studies have revealed the following.
(i) Faunal composition of Caribbean and Pacific fossil as-
semblages and the timing of paleoenvironmental change
demonstrate that major cross-isthmian marine connec-
tions ceased approximately 3 Ma (Collins et al., 1995,
1996a; Coates et al., 2003, 2005; O’Dea et al., 2007a),
consistent with dates from previous (non-PPP) oceano-
graphic studies. (ii) Seasonal upwelling was strong in
what is now the southwestern Caribbean (SWC) before
isthmian closure, and constriction of the forming isthmus
led to a rapid decline in upwelling intensity, resulting in
a collapse in primary productivity from around 5 to 3
Ma (Collins, 1996). The increasing oligotrophy allowed
reefal habitats to expand in the SWC while reducing the
amount of filter-feeding molluscan habitat, and the cessa-
tion of upwelling also stabilized environments to modern-

CAR LBB RE Amy

PEA GMim cilia G:

Today

FIGURE 3. Formation of the Isthmus of Panama during the last 20
million years (Ma = million years ago). Arrows indicate direction of
principal water flow through the Central American Seaway. (From
O'Dea et al., 2007b.)

day conditions (O’Dea et al., 2007a; Jackson et al., 1999).
Meanwhile, upwelling continued in what is now the TEP
to the present day. (iii) A wide assortment of marine taxa
experienced a major turnover in the now-SWC during the
last 10 million years (Jackson et al., 1993; Jackson and
Johnson, 2000; O’Dea et al., 2007a; Smith and Jackson,
2009). Origination of new species in all major groups of
macroinvertebrates peaked at about 5-3 Ma, coincident
with the formation of new habitat along the SWC coast
of the Isthmus. (iv) From approximately 5-3 Ma the SWC
remained connected to the TEP but coastal conditions be-
came instable. This transition period saw most SWC fau-
nas reach their peaks in diversity (Jackson and Johnson,
2000; Todd et al., 2002; Smith and Jackson, 2009). As old
and new species coexisted in time, richness of most groups
was around 30% to 60% higher than in the modern SWC.
(v) Following isthmus closure and the birth of the modern-
day Caribbean, a widespread extinction reduced numbers
of gastropod, bivalve, coral, and bryozoan taxa by 30%
to 95%. (vi) This massive extinction was strongly selec-
tive against nutriphilic taxa, indicating that the collapse
in primary productivity was the causal mechanism. How-
ever, fine-scale environmental and community composi-
tion data reveal that extinction in most groups lagged well
behind the shift to more oligotrophic conditions as the

NUMBER 38 °¢ 85

Isthmus closed (O’Dea et al., 2007a) (Figure 4). Time lags
of this scale challenge the conventional wisdom that cause
and effect have to be contemporaneous in macroevolu-
tion. (vii) Other ecological characteristics of organisms
also shifted dramatically. Average coral colony and snail
egg-size increased, larval durations of scallops decreased,
and rates of clonality in free-living bryozoans declined
dramatically. Ongoing field and laboratory work aims to
analyze the fates and trajectories of clades that preserve
modes of life, life histories, and feeding strategies in fos-
sils within the rigorous framework provided by the PPP.
This approach will help tease apart the drivers of macro-
evolutionary change in the neotropical seas (Jackson and
Erwin, 2006).

MARINE EDUCATION AND OUTREACH

At the level of both the institution and the individual
scientist, STRI, along with other SI bureaus, has become
deeply involved in two global efforts connected with
marine biodiversity: the Census of Marine Life (COML)
and the Consortium for the Barcode of Life (CBOL).
The COML aims to provide rapid and full documenta-
tion of marine biodiversity, while CBOL provides easy

Re Recent Caribbean ©

0 a Recent Pacific @
=O
{ Fossil<3.5Ma A
=x 2» = Fossil 3.5-5 Ma O
s 3 Be Fossil>5 Ma <>
= :
4 2
S z
ce 5
oO
uO
®
=> 7 @
®
8 8
=
9
10
2 4 6 8 10 0 20 40 60 £80 -1 (0) 1 0.5 0 0.5
Mean Mean Ecological structure Extinction
MART (°C) % Carbonate (PCA axis 1) rate

FIGURE 4. The sequence of environmental and ecological changes in the southwest Caribbean in response to the closure of the isthmus of
Panama (Ma = million years ago). A. Upwelling intensity, as estimated by the mean annual range of temperature (MART), shifted rapidly from
high values similar to the modern-day tropical eastern Pacific values to modern Caribbean values. B. Carbonate levels in sediments foliowed
suit, with an increase in the Caribbean. C. Biotic assemblages shifted from mollusk-dominated to a mix of coral-, algae-, and mollusk-dominated
communities (PCA = principal components analysis). D. Extinction rates of corals and mollusks peaked 1-2 million years after the environmen-
tal and ecological changes. (From O’Dea et al., 2007a.)

86 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

oo

molecular means to confirm the identities of a broad ar-
ray of species in both marine and terrestrial ecosystems.
Substantial contributions of information on neotropi-
cal marine organisms have been made by STRI to both
those efforts. Recently, the Encyclopedia of Life (EOL)
began to make use of information generated by STRI sci-
entists, and STRI also recently became part of the United
States Geological Survey’s Caribbean tsunami monitor-
ing network.

Educational and outreach programs at STRI include a
marine fellowship program for graduate students (world-
wide, plus targeted to Latin America), hosting of K-12
school groups and teacher training (at Galeta Point Ma-
rine Laboratory and Bocas del Toro Research Station),
conducting public seminars, responding to requests for
information from Panamanian government entities, and
supporting graduate student courses. The public marine
education program at STRI consists of a series of activi-
ties aimed at promoting awareness and conservation of
marine environments and communicating its research to
the general public. Since 1992 the program has consisted
of docent-led educational visits, seminars for teachers, and
the development of educational materials (posters, news-
papers and supplements, exhibits), and curricular materi-
als for the classroom.

CULEBRA ISLAND EDUCATION CENTER

The Punta Culebra Nature Center (PCNC) of STRI
lies at the Pacific entrance to the Panama Canal imme-
diately adjacent to the Naos Laboratory. For nearly a
century, access to Culebra was restricted to U.S. military
personnel, a practice that protected Culebra’s shoreline
organisms, which now exist in abundances not seen else-
where in Panama Bay. The general health of the intertidal
and shallow-water marine communities at Culebra makes
the area especially attractive for research. Culebra has
been a major research site for John Christy (since 1983),
Mark Torchin (since 2004), and their students.

The PCNC relies on the support of the Smithsonian
Foundation of Panama and international entities. The aca-
demic and public programs at Culebra encourage direct
experiences with organisms in the local habitats and in
touch pools. Exhibits promote environmental awareness,
understanding, and conservation, emphasizing marine sys-
tems. Since it opened in 1996, 750,000 people have visited
PCNG, with about 25,000 schoolchildren annually taking
part in its educational program. The PCNC also fosters re-

earch on site, which allows visitors to see STRI scientists
id students “in action.”

GALETA POINT MARINE LABORATORY

The education and outreach program at Galeta Labo-
ratory was initiated by Stanley Hecakdon in 2000 to build
bridges between research at Galeta on coral reefs, seagrass
beds, and mangrove forests and the schools of Colon and
wider Panama. The program seeks to motivate public
interest on the importance of the sciences and the value
of coastal tropical habitats, currently under severe threat
because of a destructive style of economic development.
Private donors have been vital to the success of this pro-
gram, funding the construction of enhancements to Galeta
buildings, a 300 m long mangrove boardwalk, and science
equipment used by the program.

Attendance in the student education program climbed
from 200 from an orphanage in nearby Colon in 2000 to a
current 10,000 per year from all over Panama. These pro-
grams are hosted by 12 nature guides and 19 volunteers.
Recently, the first live Internet broadcast was made from
Galeta to elementary schools in New Jersey. The next step
will be an online program to schools in Colon and, even-
tually, the rest of Panama. Galeta’s public outreach pro-
gram began in 2003 with the support of students from
McGill University’s “Panama Field Semester Studies.” The
first project was a socioeconomic study of a local fishing
community, with fishermen then being trained in nature
tourism to provide an alternative source of income. In
2006 Galeta began the Smithsonian Talk of the Month,
at which STRI researchers share their work with the peo-
ple of Colon. Teacher training aimed at raising the qual-
ity of science education in Colon started in 2007. To date
120 local elementary and high school teachers have been
trained. Galeta laboratory also participates in a variety of
community events: the yearly community beach cleanup;
scientific and environmental fairs; and events such as Bio
Diversity Day, World Mangrove Day, and Earth Day.

BOCAS DEL TORO RESEARCH STATION

The BRS has had active public programs, almost en-
tirely funded by income from station fees, since the com-
pletion of the main laboratory building in 2003. Activi-
ties organized by the BRS for the general public include
bimonthly public seminars given by researchers working
at the station as well as weekly open houses and an annual
Earth Day beach clean-up. In addition the Station opens
its doors to the public during the annual Feria Ambien-
tal weekend, at which environmental non-governmental
organizations (NGOs) and governmental organizations
from the region present information to the public, debate

local conservation issues in a round-table format, and give
public lectures on their projects. This Feria has proven to
be highly successful, with representatives form organiza-
tions such as IUCN (International Union for Conservation
of Nature) and The Nature Conservancy attending from
Costa Rica and Panama City.

The BRS also has an active program working with lo-
cal schools. School groups visit the station three days a
week during the school year, and a biodiversity summer
program is available for interested children on Isla Colon
and Bastimentos. More than 1,000 children per year par-
ticipate in these programs or, in more remote areas, receive
visits from presenters of the public programs. Finally, the
Station presents an annual teacher training workshop,
which offers teachers development credit for learning
about environmental issues and conservation.

The BRS is also active in undergraduate and graduate
teaching. The station hosts undergraduate courses from 12
institutions from the USA, Colombia, Canada, and Ger-
many and trains graduate students in the advanced Train-
ing in Tropical Taxonomy Program. This program aims to

bring taxonomic experts and experts in training together -

in the field to provide hands-on training in taxonomy. This
program focuses on groups for which taxonomic expertise
is in immediate danger of disappearing. This program, the
only one of its kind in the Neotropics, has so far trained
100 students from 30 countries and receives some fund-
ing from the National Science Foundation Pan-American
Advanced Studies Institutes (NSF PASI) program as well
as individual Assembling-the-Tree-of-Life grants.

The Online BRS Bilingual Biodiversity Database

The public face of the Bocas del Toro Research Station
extends into cyberspace. The Online BRS Bilingual Bio-
diversity Database, available at http://biogeodb.stri.si.edu
/bocas_database/?&lang=eng, has resulted from work at
the BRS and now includes 6,000 species and 8,000 photos
of organisms from Bocas del Toro province. This website
is supplemented by printed identification guides to local
organisms (Collin et al., 2005).

MARINE ZOOARCHAEOLOGY

The zooarchaeology reference collection at STRI is
frequently used by students and researchers to identify
archaeofaunal materials. Specimens are often loaned or
donated to outside institutions. Panamanians have strong
interests in their cultural heritage, and STRI zooarche-
ologists frequently give public lectures in Panama on the

NUMBER 38 e¢ 87

history of human-animal interactions in Panama and the
relevance of zooarchaeology to tropical zoogeography
and biodiversity. STRI’s Bioinformatics office recently
started work on an online database that will provide pho-
tographic, geographic, and biometric information on all
identified zooarchaeological materials and specimens from
Panamanian sites.

ONLINE INFORMATION SYSTEM ON TROPICAL EASTERN
PACIFIC SHOREFISHES

This Shorefishes of the Tropical Eastern Pacific Online
Information System (www-stri.org/sftep) exemplifies the
Smithsonian’s commitment to carrying information that
its research generates to the widest possible audience. It
provides free, public access to comprehensive information
on the biology of almost 1,300 shorefish species. Systems
such as these are useful for managers, biologists, students,
and fishers wanting to identify fishes and obtain informa-
tion about their biology. The information that systems such
as this bring together allows comprehensive assessments
of our level of knowledge about biodiversity (Zapata and
Robertson, 2006) and regional geographic distribution of
that diversity (Mora and Robertson, 2005; Robertson and
Cramer, 2009).

MARINE CONSERVATION ACTIVITIES

The work that STRI biologists, notably Hector
Guzman, have done on organisms as diverse as corals,
sea cucumbers (Guzman et al., 2003), conchs (Tewfik and
Guzman, 2003), lobsters, and crabs (Guzman and Tew-
fik, 2004) has been instrumental in the establishment not
only of management regulations for specific organisms
but also of a large marine reserve on the Pacific coast of
Panama: the Pearl Islands Special Managment Area in the
Gulf of Panama (Guzman et al., 2008a). In addition, ef-
forts by Todd Capson and research on corals by Hector
Guzman (see Guzman et al., 2004) were instrumental in
the declaration of Coiba National Park (where Rancheria
Island is situated) as a World Heritage Site in 2005. In
2009 Panama’s government established the Matumbal Re-
serve, a STRI-managed marine reserve that protects 34 ha
of reefs, seagrass beds, and mangroves immediately adja-
cent to BRS. This reserve will ensure maintenance of the
research potential of the station in an area of explosive
tourism and developmental growth. During 2008-2009
STRI (primarily through the efforts of Juan Maté) has
been involved with the recently completed development

of a comprehensive zoning and management plan for
Coiba Park and workshops aimed at informing govern-
ment resource managers about the utility, methods, and
needs of STRI’s marine research activities.

The online information system on TEP shorefishes
(see above) provided the primary database used in the
first comprehensive IUCN Redlist Assessment of an en-
tire regional shorefish fauna through workshops held in
Costa Rica (2008) and Panama (2007). An equivalent in-
formation system encompassing more than 1,500 species
of Greater Caribbean shorefishes, currently in production,
will facilitate an equivalent Redlist assessment planned for
the Greater Caribbean regional shorefish fauna.

Marine conservation activities by STRI staff also have
a global and historical reach through the work of J. B. C.
Jackson and colleagues on historical declines of coral reef
growth and organisms induced by human activities, and
the depletion of their marine resources, in the Caribbean
area and throughout the rest of the tropics (Jackson, 1997,
2001; Jackson et al., 2001; Pandolfi et al., 2003, 2005;
Pandolfi and Jackson, 2006).

BRS has been a member of CARICOMP (the Carib-
bean Coastal Marine Productivity Program) since 1997,
contributing data to Caribbean-wide monitoring of sea-
grasses, corals, and mangroves (Collin, 2005a; Collin et
al., 2009; Guzman et al., 2005). BRS also recently became
part of a global IUCN program to assess the resilience of
coral reefs worldwide. As part of this program, rapid as-
sessments of the state of coral reefs at each site are linked
to long-term monitoring of physical environmental data to
predict the local response to future bleaching stress from
elevated temperatures. Since 2000 STRI has also been in-
volved with Conservation International, the United Na-
tions Environmental Program, the International Union
for the Conservation of Nature, and the governments of
Panama, Costa Rica, Colombia, and Ecuador in an effort
to develop the Eastern Tropical Pacific Seascape. This 2.1
million km* marine conservation area, in the equatorial
part of the TEP, is based on a cluster of Marine Protected
Areas, among them the Coiba National Park (see also
Guzman et al., 2008a).

2008 —A TIME OF TRANSITION

After 48 years and 1,800 publications the marine pro-
gram, which remains an integral part of research at STRI,
is undergoing rapid change. The year 2008 marked the
end of an era, with the retirement of Ira Rubinoff and the
succession of Eldredge Bermingham as STRI director. It

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

also marked the start of a hiatus in the research vessel pro-
gram, with the retirement of the RV Urraca, as its absence
leaves a significant gap in research capability that STRI
seeks to rapidly fill. The continuing development of the
laboratory at Bocas del Toro will open up new opportuni-
ties for research. The development of a facility at Ranche-
ria Island, and, perhaps, the Pearl Islands would greatly
enhance accessibility of coral reefs and other marine habi-
tats in the two largest nearshore archipelagos in the equa-
torial part of the eastern Pacific, archipelagos that to date
have experienced relatively low impacts from economic
development. STRI geologist Carlos Jaramillo is currently
taking advantage of a unique event—major excavations
to widen the Panama Canal—to clarify the history of the
formation of the isthmus and thus help shed light on the
history of changes in the neotropical marine ecosystems
and the evolution of their organisms. In future STRI also
will emphasize the development of tools that exploit the
World Wide Web to enhance the diffustion of knowledge
derived from its marine research, both through its own
Bioinformatics office and through participation in global
enterprises such as the Census of Marine Life, the Consor-
tium for the Barcode of Life, and the Encyclopedia of Life.
STRI’s marine program will play an increasingly impor-
tant role in efforts to understand the role of the oceans in
global climate variability, interactions between terrestrial
and marine ecosystems, and the response of marine eco-
systems to climate change and more direct human-induced
stresses.

ACKNOWLEDGMENTS

STRI’s marine program thanks Panama for its co-
operation in hosting STRI, and for the long-term sup-
port of the Panamanian government entities that man-
age marine resources and marine reserves and cooperate
with STRI’s research activities (the Autoridad Maritima
de Panama, the Autoridad Nacional del Ambiente, and
the Autoridad de los Recursos Acuaticos de Panama).
Many local and international donors have contributed
generously to the development of STRI facilities and to
its education and research programs, notably D. Cofrin,
FE. Hoch, P. Peck, K. and E. Himmelfarb, the Upton Trust,
and the Fundacion Smithsonian de Panama. Marine sci-
ence at STRI also has benefitted immensely over the years
from grants made by the National Science Foundation,
the Secretaria Nacional de Ciencia Tecnologia de Pan-
ama (SENACYT), Conservation International, the Dar-
win Initiative, the Nature Conservancy, the National
Geographic Society, and various Smithsonian entities: the

Scholarly Studies Program, the Women’s Committee, the
Hunterdon and Johnson Oceanographic Research Funds,
and the Marine Science Network.

LITERATURE CITED

Abele, L. G., and W. Kim. 1989. The Decapod Crustaceans of the Pan-
ama Canal. Smithsonian Contributions to Zoology, 482:1-50.
Anderson, R. P., and C. O Handley Jr. 2002. Dwarfism in Insular
Sloths: Biogeography, Selection, and Evolutionary Rate. Evolution,

56:1045-1058.

Andrefouet, S., and H. M. Guzman. 2005. Coral Reef Distribution, Status
and Geomorphology: Biodiversity Rrelationship in Kuna Yala (San
Blas) Archipelago, Caribbean Panama. Coral Reefs, 24:31-42.

Anker, A., C. Hurt, and N. Knowlton. 2008a. Revision of the Alpheus
cristulifrons Species Complex (Crustacea: Decapoda: Alpheidae)
with Description of a New Species from the Tropical Eastern At-
lantic. Journal of the Marine Biological Association of the United
Kingdom, 88:543-562.

. 2008b. Revision of the Alpheus formosus Gibbes, 1850 Com-

plex, with Redescription of A. formosus and Description of a New

Species from the Tropical Western Atlantic (Crustacea: Decapoda:

Alpheidae). Zootaxa, 1707:1-22.

. 2008c. Revision of the Alpheus websteri Kingsley, 1880 Species
Complex (Crustacea: Decapoda: Alpheidae), with Revalidation of
A. arenensis (Chace, 1937). Zootaxa, 1694:51-68.

Backwell, P. R. Y., J. H. Christy, S. R. Telford, M. D. Jennions, and
N. I. Passmore. 2000. Dishonest Signalling in a Fiddler Crab. Pro-
ceedings of the Royal Society of London, B Biological Sciences,
267:719-724.

Birkeland, C. 1977. The Importance of Rate of Biomass Accumulation
in Early Successional Stages of Benthic Communities to the Survival
of Coral Recruits. Proceedings, Third International Symposium on
Coral Reefs, 1:16-21.

Birkeland, C., A. Alvarez Reimer, and J. R. Young. 1973. Effects of Oil
on Tropical Shore Natural Communities in Panama. Washington
DC: Smithsonian Institution, Federal Water Quality Administra-
tion, U.S. Environmental Protection Agency.

Bowen, B. W., A. L. Bass, L. A. Rocha, W. S. Grant, and D. R. Robertson.
2001. Phylogeography of the Trumpetfishes (Aulostomus): Ring
Species Complex on a Global Scale. Evolution, 55:1029-1039.

Christy, J. H. 1995. Mimicry, Mate Choice, and the Sensory Trap Hy-
pothesis. American Naturalist, 146:171-181.

Christy, J. H., P. R. Y. Backwell, S. G. Goshima, and T. J. Kreuter. 2002.
Sexual Selection for Structure Building by Courting Male Fiddler
Crabs: An Experimental Study of Behavioral Mechanisms. Behav-
ioral Ecology, 13:366-374.

Coates, A. G., M.-P. Aubry, W. A. Berggren, and L. S. Collins. 2003.
Early Neogene History of the Central American Arc from Bocas
del Toro, Western Panama. Geological Society of America Bulletin,
115:271-287.

Coates, A. G., J. B. C. Jackson, and L. S. Collins. 1992. Closure of the
Isthmus of Panama: The Near-Shore Marine Record of Costa Rica
and Western Panama. Geological Society of America Bulletin,
104:814-828.

Coates, A. G., D. F McNeill, M. P. Aubry, W. A. Berggren, and L. S. Col-
lins. 2005. An Introduction to the Geology of the Bocas del Toro
Archipelago, Panama. Caribbean Journal of Science, 41:374-391.

Coates, A. G., and J. A. Obando. 1996. “The Geologic Evolution of
the Central American Isthmus.” In Evolution and Environment
in Tropical America, ed. J. B. C. Jackson, A. F. Budd, and A. G.
Coates, pp. 21-56. Chicago: The University of Chicago Press.

NUMBER 38 °¢ 89

Coffroth, M. A., R. L. Lasker, M. E. Diamond, J. A. Bruenn, and E. Berm-
ingham. 1992. DNA Fingerprints of a Gorgonian Coral: A Method
for Detecting Clonal Structure in a Vegetative Species. Marine Biol-
ogy, 114(2):317-325.

Colley, S. B., P. W. Glynn, A. S. May, and J. L. Maté. 2006. Species-
Dependent Reproductive Responses of Eastern Pacific Corals to the
1997-98 ENSO event. Proceedings of the 10th International Coral
Reef Symposium, 61-70.

Collin, R. 2003a. World-wide Patterns of Development in Calyptraeid
Gastropods. Marine Ecology Progress Series, 247:103-122.

. 2003b. Phylogenetic Relationships among Calyptraeid Gastro-
pods and Their Implications for the Biogeography of Speciation.
Systematic Biology, 52:618-640.

Collin, R., ed. 2005a. Marine Fauna and Environments of Bocas del
Toro, Panama. Caribbean Journal of Science, 41:367-707.

Collin, R. 2005b. Ecological Monitoring and Biodiversity Surveys at the
Smithsonian Tropical Research Institute’s Bocas del Toro Research
Station. Caribbean Journal of Science, 41:367-373.

Collin, R., L. D’Croz, P. Gondola, and J. B. Del Rosario. 2009. “Climate
and Hydrological Factors Affecting Variation in Chlorophyll Con-
centration and Water Clarity in the Bahia Almirante, Panama.” In
Proceedings of the Smithsonian Marine Science Symposium, ed.
M. A. Lang, I. G. Macintyre, and K. Ritzler, pp. 323-334. Smith-
sonian Contributions to the Marine Sciences, No. 38. Washington,
D.C.: Smithsonian Institution Scholarly Press.

Collin, R., M. C. Diaz, J. L. Norenburg, R. M. Rocha, J. A. Sanchez, A.
Schulz, M. L. Schwartz, and A. Valdes. 2005. Photographic Identi-
fication Guide to Some Common Marine Invertebrates of Bocas Del
Toro, Panama. Caribbean Journal of Science, 41:638-707.

Collins, L. S. 1996. “Environmental Changes in Caribbean Shallow Waters
Relative to the Closing Tropical American Seaway.” In Evolution and
Environment in Tropical America, ed. J. B. C. Jackson, A. F. Budd, and
A. C. Coates, pp. 130-167. Chicago: University of Chicago Press.

Collins, L. $., A. E Budd, and A. G. Coates. 1996a. Earliest Evolution
Associated with Closure of the Tropical American Seaway. Proceed-
ings of the National Academy of Sciences of the United States of
America, 93:6069-6072.

Collins, L. $., and A. G. Coates, eds. 1999. A Paleobiotic Survey of Ca-
ribbean Faunas from the Neogene of the Isthmus of Panama. Law-
rence, Kans.: Allen Press.

Collins, L. S., A. G. Coates, W. A. Berggren, M.-P. Aubry, and J. Zhang.
1996b. The Late Miocene Panama Isthmian Strait. Geology,
24:687-690.

Collins, L. S., A. G. Coates, J. B. C. Jackson, and J. A. Obando. 1995.
Timing and Rates of Emergence of the Limon and Bocas del Toro
Basins: Caribbean Effects of Cocos Ridge Subduction? Geological
Society of America Special Paper, 295:263-289.

Cooke, R. G. 1981. Los Habitos Alimentarios de los Indigenas Preco-
lombinos de Panama. Revista Medica de Panama, 6:65-89.

. 1992. Prehistoric Nearshore and Littoral Fishing in the Eastern

Ttropical Pacific: An Ichthyological Evaluation. Journal of World

Archaeology, 6:1-49.

. 2004a. “Observations on the Religious Content of the Animal

Imagery of the ‘Gran Coclé’ Semiotic Tradition of Pre-Columbian

Panama.” In Behaviour behind Bones. The Zooarchaeology of Rit-

ual, Religion, Status and Identity, ed. S. O’Day, W. van Neer, and A.

Ervynck, pp. 114-127. Liverpool: Oxbow.

. 2004b. “Rich, Poor, Shaman, Child: Animals, Rank, and Status
in the ‘Gran Coclé’? Culture Area of Pre-Columbian Panama.” In
Behaviour behind Bones. The Zooarchaeology of Ritual, Religion,
Status and Identity, ed. S. O’Day, W. van Neer, and A. Ervynck, pp.
271-284. Liverpool: Oxbow.

Cooke, R. G., and M. Jiménez. 2004. Teasing Out the Species in Diverse
Archaeofaunas: Is It Worth the Effort? An Example from the Tropi-
cal Eastern Pacific. Archaeofauna, 13:19-35.

90 ¢« SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

. 2008a. “Marine Catfish (Ariidae) of the Tropical Eastern Pacific:

An Update Emphasizing Taxonomy, Zoogeography, and Interpreta-

tion of Pre-Columbian Fishing Practices.” In Archéologie du Poisson:

30 Ans d’Archéo -ichtyologie au CNRS, ed. P. Béarez, S. Grouard,

and B. Clavel, pp. 161-180. Antibes: Editions APDCA.

. 2008b. Pre-Columbian Use of Freshwater Fish in the Santa Ma-
ria Biogeographical Province, Panama. Quaternary International,
185:46-58.

Cooke, R. G., M. Jiménez, and A. J. Ranere. 2008. “Archaeozoology,
Art, Documents, and the Life Assemblage.” In Case Studies in En-
vironmental Archaeology, 2nd Edition, ed. E. J. Reitz, C. M. Scarry,
and S. J. Scudder, pp. 95-121. New York: Springer.

Cubit, J. D., R. C. Thompson, H. M. Caffey, and D. M. Windsor. 1988.
Hydrographic and Meterological Studies of a Caribbean Fringing
Reef at Punta Galeta, Panama: Hourly and Daily Variations for
1977-1985. Washington, D.C.: Smithsonian Institution Press.

D’Croz, L., J. B. Del Rosario, and P. Gondola. 2005. The Effect of Fresh-
water Runoff on the Distribution of Dissolved Inorganic Nutrients
and Plankton in the Bocas del Toro Archipelago, Caribbean Pan-
ama. Caribbean Journal of Science, 41:414-429.

D’Croz, L., and J. L. Maté. 2004. Experimental Responses to Elevated
Water Temperature in Genotypes of the Reef Coral Pocillopora
damicornis from Upwelling and Non-upwelling Environments in
Panama. Coral Reefs, 23:473-483.

D’Croz, L, J. L. Maté, and J. E. Oke. 2001. Responses to Elevated Sea
Water Temperature and UV Radiation in the Coral Porites lobata
from Upwelling and Non-upwelling Environments in the Pacific
Coast of Panama. Bulletin of Marine Science, 69:203-214.

D’Croz, L., and A. O’Dea. 2007. Variability in Upwelling along the
Pacific Shelf of Panama and Implications for the Distribution of
Nutrients and Chlorophyll. Estuarine, Coastal, and Shelf Science,
73:325-340.

. 2009. “Nutrient and Chlorophyll Dynamics in Pacific Central
America (Panama).” In Proceedings of the Smithsonian Marine Sci-
ence Symposium, ed. M. A. Lang, I. G. Macintyre, and K. Ritzler,
pp. 335-344. Smithsonian Contributions to the Marine Sciences,
No. 38. Washington, D.C.: Smithsonian Institution Scholarly Press.

D’Croz, L., and D. R. Robertson. 1997. Coastal Oceanographic Conditions
Affecting Coral Reefs on Both Sides of the Isthmus of Panama. Proceed-
ings of the 8th International Coral Reef Symposium, 2:2053-2058.

D’Croz, L., D. R. Robertson, and J. A. Martinez. 1999. Cross-Shelf Dis-
tribution of Nutrients, Plankton, and Fish Larvae in the San Blas
Archipelago, Caribbean Panama. Revista de Biologia Tropical,
47:203-215.

DeSalvo, M. K., C. R. Voolstra, S$. Sunagawa, J. A. Schwarz, J. H. Still-
man, M. A. Coffroth, A. M. Szmant, and M. Medina. 2008. Dif-
ferential Gene Expression during Thermal Stress and Bleaching in
the Caribbean Coral Monastraea faveolata. Molecular Ecology,
17:3952-3971.

Dick, M. H., A. Herrera-Cubilla, and J. B. C. Jackson. 2003. Molecular
Phylogeny and Phylogeography of Free-Living Bryozoa (Cupuladri-
idae) from Both Sides of the Isthmus of Panama. Molecular Phylo-
genetics and Evolution, 27:355-371.

Fischer, E. A. 1980. Speciation in the Hamlets (Hypoplectrus: Serrani-
dae): A Continuing Enigma. Copeia, 1980:649-659.

. 1981. Sexual Allocation in a Simultaneously Hermaphroditic
Coral Reef Fish. American Naturalist, 117:64-82.

Fischer, E. A., and C. W. Petersen. 1987. The Evolution of Sexual Pat-
terns in the Seabasses. Bioscience, 37:482-489.

Fukami, H., A. F. Budd, D. R. Levitan, J. A. Jara, R. Kersanach, and
N. Knowlton. 2004. Geographic Differences in Species Boundaries
among Members of the Montastraea annularis Complex Based on
Molecular and Morphological markers. Evolution, 58:324-337.

Glynn, P. W. 1972. “Observations on the Ecology of the Caribbean and
Pacific Coasts of Panama.” In The Panamic Biota: Some Obser-

vations Prior to Sea Level Canal, ed. M. L. Jones. Bulletin of the

Biological Society of Washington, 2:13-30.

. 1973. “Aspects of the Ecology of Coral Reefs in the Western

Atlantic Region.” In Biology and Geology of Coral Reefs, ed.

O. A. Jones and R. Endean, pp. 271-324. New York: Academic

Press.

. 1984. Widespread Coral Mortality and the 1982-83 El Nino

Warming Event. Environmental Conservation, 11:133-146.

. 1990. “Coral Mortality and Disturbances to Coral Reefs in the
Tropical Eastern Pacific.” In Global Ecological Consequences of
the 1982-1983 El Nino-Southern Oscillation, ed. P. W. Glynn, pp.
55-126. Amsterdam: Elsevier.

Glynn, P. W., and M. W. Colgan. 1992. Sporadic Disturbances in Fluctu-
ating Coral Reef Environments: El Nifio and Coral Reef Develop-
ment in the Eastern Pacific. American Zoologist, 32:707-718.

Glynn, P. W., S. B. Colley, C. M. Eakin, D. B. Smith, J. N. Cortes,
N. J. Gassman, H. Guzman, J. B. Del Rosario, and J. S. Feingold.
1994. Reef Coral Reproduction in the Eastern Pacific: Costa Rica,
Panama, and Galapagos Islands (Ecuador). II. Poritidae. Marine Bi-
ology, 118:191-208.

Glynn, P. W., S. B. Colley, N. J. Gassman, K. Black, K Cortez, J. Nunez,
and J. L. Maté. 1996. Reef Coral Reproduction in the Eastern Pa-
cific: Costa Rica, Panama, and Galapagos Islands (Ecuador). III.
Agariciidae (Pavona gigantea and Gardineroseris planulata). Ma-
rine Biology, 125:579-601.

Glynn, P. W., S. B. Colley, J. L. Maté, J. Cortés, H. M. Guzman, R. L.
Bailey, J. S. Feingold, and I. C. Enochs. 2008. Reproductive Ecol-
ogy of the Azooxanthellate Coral Tubastraea coccinea in the Equa-
torial Eastern Pacific: Part V. Dendrophylliidae. Marine Biology,
153:529-544.

Glynn, P. W., J. Cortés, H. M. Guzman, and R. H. Richmond. 1988.
El Nino (1982-83)-Associated Coral Mortality and Relationship
to Sea Surface Temperature Deviations in the Tropical Eastern
Pacific. Proceedings of 6th International Coral Reef Symposium,
3:237-243.

Glynn, P. W., and L. D’Croz. 1990. Experimental Evidence for High
Temperature Stress as the Cause of El Nifio-Coincident Mortality.
Coral Reefs, 8:181-191.

Glynn, P. W., N. J. Gassman, C. M. Eakin, J. Cortés, D. B. Smith, and
H.M. Guzman. 1991. Reef Coral Reproduction in the Eastern
Pacific: Costa Rica, Panama, and Galapagos Islands (Ecuador).
I. Pocilloporidae. Marine Biology, 109:355-368.

Glynn, P. W., and I. G. Macintyre. 1977. Growth Rate and Age of Coral
Reefs of the Pacific Coast of Panama. Proceedings of 3rd Interna-
tional Coral Reef Symposium, 2:251-259.

Glynn, P. W., J. L. Maté, A. C. Baker, and M. O. Calderon. 2001. Coral
Bleaching and Mortality in Panama and Ecuador During the
1997-1998 El Nifio-Southern Oscillation Event: Spatial/Temporal
Patterns and Comparisons with the 1982-1983 Event. Bulletin of
Marine Science, 69:79-110.

Glynn, P. W., and R. H. Stewart. 1973. Distribution of Coral Reefs in the
Pearl Islands (Gulf of Panama) in Relation to Thermal Conditions.
Limnology and Oceanography, 18:367-379.

Glynn, P. W., R. H. Stewart, and J. E. McCosker. 1972. Pacific Coral
Reefs of Panama: Structure, Distribution and Predators. Geolo-
gischen Rundschau, 61:483-519.

Glynn, P. W., and G. M. Wellington. 1983. Corals and Coral Reefs of the
Galapagos Islands. Berkeley: University of California Press.

Glynn, P. W., G. M. Wellington, and C. Birkeland, C. 1979. Coral Reef
Growth in the Galapagos: Limitation by Sea Urchins. Science,
203:47-48.

Graham, J. B. 1970. Temperature Sensitivity of Two Species of Inter-tidal
Fishes. Copeia, 1970:49-56.

. 1971. Temperature Tolerances of Some Closely Related Tropical

Atlantic and Pacific Fish Species. Science, 172:861-863.

. 1975. Heat Exchange in Yellow Fin (Thunnus albacares) and
Skipjack (Katsuwonus pelamis) Tunas and the Adaptive Signifi-
cance of Elevated Body Temperatures in Scombrid Fishes. Fishery
Bulletin, 73:219-229.

Graham, J. B., I. Rubinoff, and M. K. Hecht. 1971. Temperature
Physiology of the Sea Snake Pelamis platurus: An Index of Its
Colonization Potential in the Atlantic Ocean. Proceedings of the
National Academy of Sciences of the United States of Amerca,
68:1360-1363.

Guzman, H. M., ed. 1998. Marine-Terrestrial Flora and Fauna of Cayos
Cochinos Archipelago, Honduras. Revista Biologia Tropical Sup-
plement), 46:1-200.

Guzman, H. M., P. A. G. Barnes, C. E. Lovelock, and I. C. Feller. 2005. A
Site Description of the CARICOMP Mangrove, Seagrass and Coral
Reef Sites in Bocas del Toro, Panama. Caribbean Journal of Science,
41:430-440.

Guzman, H. M.., S. L. Benfield, O. Breedy, and J. M. Mair. 2008a. Broad-
ening Reef Protection across the Marine Conservation Corridor of
the Eastern Tropical Pacific: Distribution and Diversity of Reefs
in Las Perlas Archipelago, Panama. Environmental Conservation,
35:46-54.

Guzman, H. M., and O. Breedy. 2008. Leptogorgia christiae (Octo-
corallia: Gorgoniidae): A New Shallow Water Gorgonian from
Pacific Panama. Journal of the Marine Biological Association,
88:719-722.

Guzman, H. M., R. Cipriani, and J. B. C. Jackson. 2008b. Historical
Decline in Coral Reef Growth after the Panama Canal. Ambio,
37:342-346.

Guzman, H. M., C. A. Guevara, and O. Breedy. 2004. Distribution, Di-
versity, and Conservation of Coral Reefs and Coral Communities
in the Largest Marine Protected Area of Pacific Panama (Coiba Is-
land). Environmental Conservation, 3:111-121.

Guzman, H. M., C. A. Guevara, and I. C. Hernandez. 2003. Reproduc-
tive Cycle of Two Commercial Species of Sea Cucumber (Echino-
dermata: Holothuroidea) from Caribbean Panama. Marine Biol-
ogy, 142:271-279.

Guzman, H. M., J. B. C. Jackson, and E. Weil. 1991. Short-Term Eco-
logical Consequences of a Major Oil Spill on Panamanian Subtidal
Reef Corals. Coral Reefs, 10:1-12.

Guzman, H. M., and A. Tewfik. 2004. Population Characteristics and
Co-occurrence of Three Exploited Decapods (Panulirus argus,
P. guttatus, Mithrax spinosissimus) in Bocas del Toro, Panama.
Journal of Shellfish Research, 23:575-580.

Handley, C. 1993. “Conservacion de la Fauna y Flora en las Islas de
Bocas del Toro.” In Agenda Ecoldgica y Social para Bocas del Toro,
ed. S. Heckadon-Moreno, pp. 43-48. Panama: Paseo Pantera and
Smithsonian Tropical Research Institute.

Hildebrand, S. F 1939. The Panama Canal as a Passageway for Fishes,
with Lists and Remarks on the Fishes and Invertebrates Observed.
Zoologica, 24:15-4S.

Hueerkamp, C., P. W. Glynn, L. D’Croz, J. L. Maté, and S. B. Colley.
2001. Bleaching and Recovery of Five Eastern Pacific Corals in an
El Nifio-Related Temperature Experiment. Bulletin of Marine Sci-
ence, 69:215-236.

Hurt, C., A. Anker, and N. Knowlton. 2009. A Multilocus Test of Simul-
taneous Divergence across the Isthmus of Panama Using Snapping
Shrimp in the Genus Alpheus. Evolution, 63:514-530.

Jackson, J. B. C. 1997. Reefs since Columbus. Coral Reefs, 16(Suppl.):
$23-S32.

. 2001. What Was Natural in the Coastal Oceans? Proceedings of
the National Academy of Sciences of the United States of America,
98:5411-5418.

Jackson, J. B. C., and D. H. Erwin. 2006. What Can We Learn about
Ecology and Evolution from the Fossil Record? Trends in Ecology
and Evolution, 21:322-328.

NUMBER 38 e¢ 91

Jackson, J. B. C., and K.G. Johnson. 2000. Life in the Last Few Million
Years. Paleobiology, 26:221-235.

Jackson, J. B., M. X. Kirby, W. H. Berger, K. A. Bjorndal, L. W. Botsford,
B. J. Bourque, R. H. Bradbury, R. G. Cooke, J. Erlandson, J. A.
Estes, T. P. Hughes, S. M. Kidwell, C. B. Lange, H. S. Lenihan,
J. M. Pandolfi, C. H. Peterson, R. S. Steneck, M. J. Tegner, and
R. R. Warner. 2001. Historical Overfishing and the Recent Collapse
of Coastal Ecosystems. Science, 293:629-637.

Jackson, J. B. C., J. A. Todd, H. Fortunato, and P. Jung. 1999. “Diversity
and Assemblages of Neogene Caribbean Mollusca of Lower Cen-
tral America.” In A Paleobiotic Survey of Caribbean Faunas from
the Neogene of the Isthmus of Panama, ed. L. S. Collins and A. G.
Coates, pp. 193-230. Lawrence, Kans.: Allen Press.

Kaufmann, K., and R. C. Thompson. 2005. Water temperature variation
and the meteorological and hydrographic environment of Bocas del
Toro, Panama. Caribbean Journal of Science, 41:392-413.

Keller, B. D., and J. B. C. Jackson, eds. 1993. Long-Term Assessment of
the Oil Spill at Bahia Las Minas, Panama: Synthesis Report. OCS
Study MMS 93-0047. New Orleans: U.S. Department of the Inte-
rior, Minerals Management Service, Gulf of Mexico OCS Regional
Office.

Kim, T. W., J. H. Christy, S. Dennenmoser, and J. C. Choe. 2009.
The Strength of a Female Mate Preference Increases with Preda-
tion Risk. Proceedings of the Royal Society B Biological Sciences,
276:775-780.

Kline, D. I., N. M. Kuntz, M. Breitbart, N. Knowlton, and F. Rohwer.
2006. Role of Elevated Organic Carbon and Microbial Activity in
Coral Mortality. Marine Ecology Progress Series, 314:119-125.

Knowlton, N. 2000. Molecular Genetic Analyses of Species Boundaries
in the Sea. Hydrobiologia, 420:73-90.

Knowlton, N., J. L. Mate, H. M. Guzman, R. Rowan, and J. Jara. 1997.
Direct Evidence for Reproductive Isolation among the Three Species
of the Montastraea annularis Complex in Central America (Panama
and Honduras). Marine Biology, 127:705-711.

Knowlton, N., and L. A. Weigt. 1998. New Dates and New Rates for
Divergence across the Isthmus of Panama. Proceedings of the Royal
Society (London) Series B Biological Sciences, 265:2257-2263.

Knowlton, N., E. Weil, L. A. Weigt, and H. M. Guzman. 1992. Sibling
Species in Montastraea annularis, Coral Bleaching and the Coral
Climate Record. Science, 255:330-333.

Lasker, H. R. 1991. Population Growth of a Gorgonian Coral: Equilib-
rium and Non-equilibrium Sensitivity to Changes in Life History
Variables. Oecologia (Berlin), 86:503-509.

Lasker, H. R., D. A. Brazeau, J. Calderon, M. A. Coffroth, R. Coma,
and K. Kim. 1996. In Situ Rates of Fertilization among Broadcast
Spawning Gorgonian Corals. Biological Bulletin, 190:45-55.

Lessios, H. A. 1979. Use of Panamanian Sea Urchins to Test the Molecu-
lar Clock. Nature (London), 280:599-601.

. 1981. Reproductive Periodicity of the Echinoids Diadema and

Echinometra of the Two Coasts of Panama. Journal of Experimen-

tal Marine Biology and Ecology, 50:47-61.

. 2005. Diadema antillarum Populations in Panama Twenty Years

Following Mass Mortality. Coral Reefs, 24:125-127.

. 2008. The Great American Schism: Divergence of Marine Or-
ganisms after the Rise of the Central American Isthmus. Annual
Review of Ecology and Systematics, 39:63-91.

Lessios, H. A., B. D. Kessing, and J. S. Pearse. 2001. Population Structure
and Speciation in Tropical Seas: Global Phylogeography of the Sea
Urchin Diadema. Evolution, 55:955-975.

Lessios, H. A., B. D. Kessing, D. R. Robertson, and G. Paulay. 1999.
Phylogeography of the Pantropical Sea Urchin Eucidaris in Relation
to Land Barriers and Ocean Currents. Evolution, 53:806-817.

Lessios, H. A., and I. G. Macintyre, eds. 1997. Proceedings of the 8th
International Coral Reef Symposium, Volumes 1 and 2. Smithson-
ian Tropical Research Institute, Republic of Panama.

One

Lessios, H. A., and D. R. Robertson. 2006. Crossing the Impassable:
Genetic Connections in 20 Reef Fishes across the Eastern Pacific
Barrier. Proceedings of the Royal Society (London) B Biological Sci-
ences, 273:2201-2208.

Lessios, H. A., D. R. Robertson, and J. D. Cubit. 1984. Spread of
Diadema Mass Mortality Throughout the Caribbean. Science,
226:335-337.

Linares, O. FE. 1977. “Ecology and the Arts in Ancient Panama: On the
Development of Rank and Symbolism in the Central Provinces.” In
Studies in Precolumbian Art and Archaeology, Volume 17. Wash-
ington, D.C.: Dumbarton Oaks,

Linares, O. EF, and A. J. Ranere, eds. 1980. Adaptive Radiations in Pre-
historic Panama. Peabody Museum Monographs No 5. Cambridge:
Harvard University Press.

Macintyre, I. G., and P. W. Glynn. 1976. Evolution of Modern Carib-
bean Fringing Reef, Galeta Point, Panama. American Association
of Petroleum Geologists Bulletin, 60:1054-1072.

Manzello, D. P., J. A. Kleypas, D. A. Budd, C. M. Eakin, P. W. Glynn,
and C. Langdon. 2008. Poorly Cemented Coral Reefs of the Eastern
Tropical Pacific: Possible Insights into Reef Development in a high-
CO, World. Proceedings of the National Academy of Sciences of
the United States of America, 105:10450-10455.

McCartney, M. A., and H. A. Lessios. 2002. Quantitative Analysis of
Gametic Incompatibility between Closely Related Species of Neo-
tropical Sea Urchins. Biological Bulletin, 202:166-181.

Menge, B. A., and J. Lubchenco. 1981. Community Organization in
Temperate and Tropical Rocky Intertidal Habitats: Prey Refuges in
Relation to Consumer Pressure Gradients. Ecological Monographs,
51:429-450.

Menge, B. A., J. Lubchenco, L. R. Ashkenas, and F. Ramsey. 1986. Ex-
perimental Separation of Effects of Consumers on Sessile Prey in
the Low Zone of a Rocky Shore in the Bay of Panama: Direct and
Indirect Consequences of Food Web Complexity. Journal of Experi-
mental Marine Biology And Ecology, 100:225-269.

Meyer, D. L., C. M. Birkeland, and G. Hendler. 1974. “Environmental
Sciences Program Marine Studies, Galeta Point.” In Environmental
Monitoring and Baseline Data, ed. D. M. Windsor, pp. 273-409.
Washington, D.C.: Smithsonian Institution Environmental Sciences
Program.

Miglietta, M. P., and H. A. Lessios. 2009. A Silent Invasion. Biological
Invasions, 11:825-834.

Miglietta, M. P., M. Rossi, and R. Collin. 2008. Hydromedusa Blooms
and Upwelling Events in the Bay of Panama, Tropical East Pacific.
Journal of Plankton Research, 30:783-793.

Mora, C., and D. R. Robertson. 2005. Causes of Latitudinal Gradients in
Species Richness: A Test with the Endemic Shorefishes of the Tropi-
cal Eastern Pacific. Ecology, 86:1771-1782.

Morgan, S. G., and J. H. Christy. 1995. Adaptive Significance of the Tim-
ing of Larval Release by Crabs. American Naturalist, 145:457-479.

Moynihan, M. H. 1975. Conservatism of Displays and Comparable
Stereotyped Patterns among Cephalopods. In Function and Evolu-
tion in Behavior, ed. G. Baerends, C. Beer, and A. Manning, pp.
276-291. Oxford: Clarendon Press.

Moynihan, M. H., and A. Rodaniche. 1982. The Behavior and Natural
History of the Caribbean Reef Squid, Sepioteuthis sepioidea. Berlin:
Verlag Paul Parey.

O’Dea, A., A. Herrera-Cubilla; H. M. Fortunato, and J. B.C. Jackson.
2004. Life History Variation in Cupuladriid Bryozoans from Either
Side of the Isthmus of Panama. Marine Ecology Progress Series,
280:145-161.

O’Dea, A., and J. B. C. Jackson. 2002. Bryozoan Growth Mirrors Con-
trasting Seasonal Regimes across the Isthmus of Panama. Palaeo-
geography, Palaeoclimatology, Palaeoecology, 185:77-94.

O'Dea, A. A., J. B. C. Jackson, H. Fortunato, J. Travis Smith, L. D’Croz,
K. G. Johnson, and J. A. Todd. 2007a. Environmental Change Pre-

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

ceded Caribbean Extinction by 2 Million Years. Proceedings of the
National Academy of Sciences of the United States of America,
104:5501-5506.

O’Dea, A., FE. Rodriguez, C. DeGracia, and A.G. Coates. 2007b. La
Paleontologia Marina en el Istmo de Panama. Canto Rodado,
2:149-179.

Paddack, M. J., C. Aguilar, R. S. Appeldoorn, J. Beets, E. W. Burkett,
P. M. Chittaro, K. Clarke, R. Esteves, A. C. Fonseca, G. E. For-
rester, A. M. Friedlander, J. Garcia-Sais, G. Gonzdalez-Sans6n,
L. K. B. Jordan, D. McClellan, M. W. Miller, P. Molloy, P. J.
Mumby, I. Nagelkerken, M. Nemeth, R. Navas-Camacho, J. Pitt,
N. V. C. Polunin, M. C. Reyes-Nivia, D. R. Robertson, A. R.
Ramirez, E. Salas, S. R. Smith, R. E. Spieler, M. A. Steele, I. D.
Williams, C. Wormald, A. R. Watkinson, J. Reynolds, and I. M.
Coté. 2009. Recent Caribbean-wide Declines in Reef Fish Abun-
dance. Current Biology, 19:1-6.

Pandolfi, J. M., R. H. Bradbury, E. Sala, T. P. Hugues, K. A. Bjorndal,
R. G. Cooke, D. McArdle, L. McClenachan, M. J. H Newman,
G. Paredes, R. R. Warner, and J. B.C. Jackson. 2003. Global Tra-
jectories of the Long-Term Decline of Coral Reef Ecosystems. Sci-
ence, 301:955-958.

Pandolfi, J. M., and J. B.C. Jackson. 2006. Ecological Persistence Inter-
rupted in Caribbean Coral Reefs. Ecology Letters, 9:818-826.
Pandolfi, J. M., J. B.C. Jackson, N. Baron, R. H. Bradbury, H. M. Guz-
man, T. P. Hughes, C. V. Kappel, E. Micheli, J. C. Ogden, H. P. Pos-
singham, and E. Sala. 2005. Are U.S. Coral Reefs on the Slippery

Slope to Slime? Science, 307:1725-1726.

Puebla, O., E. Bermingham, and F. Guichard. 2008. Population Genetic
Analyses of Hypoplectrus Coral Reef Fishes Provide Evidence That
Local Processes Are Operating during the Early Stages of Marine
Adaptive Radiations. Molecular Ecology, 17:1405-1415.

Puebla, O., E. Bermingham, F. Guichard, and E. Whiteman. 2007. Colour
Pattern as a Single Trait Driving Speciation in Hypoplectrus Coral
Reef Fishes? Proceedings of the Royal Society (London) B Biological
Sciences, 274:1265-1271.

Quenouille, B., E. Bermingham, and S. Planes. 2004. Molecular System-
atics of the Damselfishes (Teleostei: Pomacentridae): Bayesian Phy-
logenetic Analyses of Mitochondrial and Nuclear DNA Sequences.
Molecular Phylogenetics and Evolution, 31:66-88.

Robertson, D. R., J. D. Ackerman, J. H. Choat, J. M. Posada, and J. Pitt.
2005. Ocean Surgeonfish Acanthurus bahianus. 1. The Geography
of Demography. Marine Ecology Progress Series, 295:229-244.

Robertson, D. R., and G. R. Allen. 2008. Shorefishes of the Tropical
Eastern Pacific Online Information System. www:stri.org/sftep.

Robertson, D. R., and K. Cramer. 2009. Marine Shore-fishes and Biogeo-
graphic Subdivisions of the Tropical Eastern Pacific. Marine Ecol-
ogy Progress Series, 380(1):1-17.

Robertson, D. R., S. Swearer, K. Kaufmann, and E. B. Brothers. 1999.
Settlement vs. Environmental Dynamics in a Pelagic Spawn-
ing Reef Fish in Caribbean Panama. Ecological Monographs,
69:195-218.

Robertson, D. R., and R. R. Warner. 1978. Sexual Patterns in the Labroid
Fishes of the Western Caribbean, II: The Parrotfishes (Scaridae).
Smithsonian Contributions to Zoology, 255:1-25.

Rocha, L. A., L. Anna, D. Bass, D. R. Robertson, and B. W. Bowen.
2002. Adult Habitat Preferences, Larval Dispersal, and the Com-
parative Phylogeography of Three Atlantic Surgeonfishes (Teleostei:
Acanthuridae). Molecular Ecology, 11:243-251.

Rocha, L. A., D. R. Robertson, C. R. Rocha, J. L. Van Tassell, M. T.
Craig, and B. W. Bowen. 2005a. Recent Invasion of the Tropi-
cal Atlantic by an Indo-Pacific coral reef fish. Molecular Ecology,
14:3921-3928.

Rocha, L. A., D. R. Robertson, J. Roman, and B. W. Bowen. 2005b. Eco-
logical Speciation in Tropical Reef Fishes. Proceedings of the Royal
Society (London) B Biological Sciences, 272:573-579.

Roche, D. G., and M. E. Torchin. 2007. Established Population of the
North American Harris Mud Crab, Rhithropanopeus harristi
(Gould 1841) (Crustacea: Brachyura: Xanthidae) in the Panama
Canal. Aquatic Invasions, 2:155-161.

Rowan, R., N. Knowlton, A. Baker, and J. Jara. 1997. Landscape Ecol-
ogy of Algal Symbionts Creates Variation in Episodes of Coral
Bleaching. Nature (London), 388:265-269.

Rubinoff, I. 1965. Mixing Oceans and Species. Natural History, 74:69-72.

. 1968. Central American Sea Level Canal: Possible Biological Ef-
fects. Science, 161:857-861.

Rubinoff, I., J. B. Graham, and J. Motta. 1986. Diving of the Sea Snake
Pelamis platurus in the Gulf of Panama. I. Dive Depth and Dura-
tion. Marine Biology, 91:181-191.

Rubinoff, I., and R. W. Rubinoff. 1962. New Records on Inshore Fishes
from the Atlantic Coast of Panama. Breviora, 1962:1-7.

Rubinoff, R. W., and I. Rubinoff. 1969. Observations on Migration of a
Marine Goby Through the Panama Canal. Copeia, 1969:395-397.

. 1971. Geographic and Reproductive Isolation in Atlantic and Pa-
cific Populations of Panamanian Bathygobius. Evolution 25:88-97.

Ruiz, G. M., M. E. Torchin, and K. Grant. 2009. “Using the Panama
Canal to Test Predictions about Tropical Marine Invasions.” In
Proceedings of the Smithsonian Marine Science Symposium, ed.
M. A. Lang, I. G. Macintyre, and K. Ritzler, pp. 291-299. Smith-
sonian Contributions to the Marine Sciences, No. 38. Washington,
D.C.: Smithsonian Institution Scholarly Press.

Schloeder, C., and L. D’Croz. 2004. Responses of Massive and Branching
Coral Species to the Combined Effects of Water Temperature and
Nitrate Enrichment. Journal of Experimental Marine Biology and
Ecology, 313:255-268.

Smith, J. T., and J. B. C. Jackson. 2009. Ecology of Extreme Faunal Turn-
over of Tropical American scallops. Paleobiology, 35:77-93.

Sousa, W. P. 2007. Mangrove Forest Structure and Dynamics, Punta Galeta,
Panama. Bulletin of the Ecological Society of America, 88:46-49.

NUMBER 38 e¢ 93

Sturmbauer, C., J. Levington, and J. H. Christy. 1996. Molecular Phylog-
eny of Fiddler Crabs: Test of the Hypothesis of Increasing Behav-
ioral Complexity in Evolution. Proceedings of the National Acad-
emy of Sciences of the United States of America, 93:10855-10857.

Tewfik, A., and H. M. Guzman. 2003. Shallow-Water Distribution and
Population Characteristics of Strombus gigas and S. costatus (Gas-
tropoda Strombidae) in Bocas del Toro, Panama. Journal of Shell-
fish Research, 22:789-794.

Todd, J. A., J. B. C. Jackson, K. G. Johnson, H. M. Fortunato, A. Heitz,
M. Alvarez, and P. Jung. 2002. The Ecology of Extinction: Mol-
luscan Feeding and Faunal Turnover in the Caribbean Neogene.
Proceedings of the Royal Society of London, B Biological Sciences,
269:571-577.

Vargas, S., H. M. Guzman, and O. Breedy. 2008. Distribution Patterns of
the Genus Pacifigorgia (Octocorallia: Gorgoniidae): Track Compat-
ibility Analysis and Parsimony Analysis of Endemicity. Journal of
Biogeography, 35:241-247.

Warner, R. R., and D. R. Robertson. 1978. Sexual Patterns in the Labroid
Fishes of the Western Caribbean. I. The Wrasses (Labridae). Smith-
sonian Contributions to Zoology, 254:1-27.

Warner, R. R., D. R. Robertson, and E. G. Leigh Jr. 1975. Sex Change
and Sexual Selection. Science, 190:633-638.

Wulff, J. L. 1991. Asexual Fragmentation, Genotype Success, and Popu-
lation Dynamics of Erect Branching Sponges. Journal of Experi-
mental Marine Biology and Ecology, 149:227-247.

. 1997. Mutualism among Species of Coral Reef Sponges. Ecol-
ogy, 78:146-159.

Zapata, E, and D. R. Robertson. 2006. How Many Shore-Fish Species
Are There in the Tropical Eastern Pacific? Journal of Biogeography,
34:38-51.

Zigler, K. S., and H. A. Lessios. 2004. Speciation on the Coasts of the
New World: Phylogeography and the Evolution of Bindin in the Sea
Urchin Genus Lytechinus. Evolution, 58:1225-1241.

i
- °
- (4)
7 = i
= * U
7 A ‘ol
n= G -
= 4 f
{inca | ?
ae é ae Ae oe oe — “See eet A ee
x, - ‘ ss

Ss A Pet ce ae ; 2 a. ; in wil

: a a Ae © fen
a z Rees sen ae adhe 2 resi its WOM
a : cael nm PSR bere OF onde ree

Minar bah oly AAR

* oa = toe 3 i 24 ae. oct

ciel: qrecmbia: Lighnscsnbusali
=I Cle | Pe = “7s
a (f
i
#4 j
ro
= '
x rs ou
«
a)
~ zy = = a /
- ; :
iy 7 are
<: = = ih
q i :
> a = 2

Protandric Simultaneous Hermaphroditism
Is a Conserved Trait in Lysmata (Caridea:
Lysmatidae): Implications for the Evolution
of Hermaphroditism in the Genus

J. Antonio Baeza

J. Antonio Baeza, Smithsonian Tropical Research

Institute, Panama, and Smithsonian Marine Sta-
tion at Fort Pierce, 701 Seaway Drive, Fort Pierce,
Florida 34949, USA (baezaa@si.edu). Received 9
June 2008; accepted 20 April 2009.

ABSTRACT. Shrimps from the genus Lysmata are unusual because of their peculiar
sexual system. Individuals in a population first reproduce as males, to change later in life
to functional simultaneous hermaphrodites. The evolutionary origin of this sexual sys-
tem, called protandric simultaneous hermaphroditism (PSH), is a longstanding question
overdue for consideration. A previously proposed “historical contingency” hypothesis
suggested that PSH evolved in the tropics from an ancestral protandric species of Lys-
mata that became socially monogamous and symbiotic with sea anemones. The restricted
probability of encountering mating partners by shrimps because of their association with
their hosts would have favored PSH. Here, I first provide evidence that PSH is a fixed
trait within the genus. Second, I examine whether the historical contingency hypothesis
appropriately explains the origin of PSH in the genus. Using anatomical observations
and laboratory experiments combined, I demonstrate that two shrimps from the genus
Lysmata, L. galapagensis and L. boggessi, feature PSH. Study of museum specimens sug-
gests that nine other species of Lysmata are protandric simultaneous hermaphrodites.
The foregoing information indicates that PSH represents a fixed trait in the genus Lys-
mata. Ancestral character state reconstruction using Bayesian inference allowed testing
whether the ancestral Lysmata featured a symbiotic lifestyle and a socially monogamous
mating system, as proposed by the historical contingency hypothesis. In agreement with
this hypothesis, analysis indicated that the most common recent ancestor of Lysmata was
most likely socially monogamous. However, the ancestral lifestyle was equally likely to
be free-living or symbiotic. Thus, the present study provides partial support for the his-
torical contingency hypothesis. Studies on the sexual system and lifestyle of more species
and development of a more robust phylogeny are needed to reveal the evolutionary origin
of PSH in the genus Lysmata.

INTRODUCTION

In decapod crustaceans, the greatest diversity of sexual systems is found in
the infraorder Caridea. Most caridean shrimps are gonochoric, with individuals
in a population producing only male or female gametes during their entire life.
Well-studied examples include Rhynchocinetes typus (Correa et al., 2000), Hip-
polyte obliquimanus (Terossi et al., 2008), Pontonia margarita (Baeza, 2008a),
and Hippolyte williamsi (Espinoza-Fuenzalida et al., 2008). The second most
common sexual system is protandry. In at least 31 species of shrimps, individu-
als in a population reproduce first as males and change to females later in life

96 e* SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

(Bauer, 2000). Although several variants of protandry
have been reported (e.g., protandry with primary females
in Crangon crangon; Schatte and Saborowski, 2006), no
study has reported protogyny (changing sex from female to
male) among shrimps. Most recently, a particular variant
of simultaneous hermaphroditism, that is, adolescent pro-
tandry sensu Ghiselin (1974), protandric cosexuality sensu
Policansky (1982), or protandric simultaneous hermaph-
roditism (PSH) sensu Bauer (2000), has been described for
shrimps from the genera Lysmata (Baeza et al., 2008) and
Exhippolysmata (Kagwade, 1982; Braga et al., 2009). It
must be noticed that a recently developed molecular phy-
logeny for Lysmata and other related genera demonstrated
that the genus Exhippolysmata represents a derived group
of shrimps within the genus Lysmata (Baeza et al., 2009).
Thus, species of Exhippolysmata are treated here as mem-
bers of the genus Lysmata.

In protandric simultaneous hermaphroditic shrimps,
juveniles consistently mature as functional male individu-
als (also called male-phase [MP] shrimps; Bauer and Holt,
1998) bearing typical caridean male characters (1.e., cou-
pling hoods and appendix masculina on the first and sec-
ond pleopods, respectively) (Bauer and Holt, 1998; Baeza,
2008b; Baeza and Anker, 2008; Baeza et al., 2008). These
functional males later attain female sexual function and
develop into functional simultaneous hermaphrodites
(hereafter, hermaphrodites; but also called female-phase
[FP] shrimps; Bauer and Holt, 1998). Resembling females
of caridean gonochoric species, hermaphrodites mate as
females shortly after molting, spawn oocytes to an abdom-
inal chamber where fertilization takes place, and brood
their embryos for relatively long periods of time (e.g.,
10-15 days in Lysmata wurdemanni; Baeza, 2006). These
hermaphrodites retain testicular tissue, male ducts, and
gonopores and thus have the ability to reproduce as both
male and female (Bauer and Holt, 1998). After becoming
hermaphrodites, individuals do not revert to males (Baeza,
2007a), and no self-fertilization has been demonstrated
(Bauer and Holt, 1998; Baeza, 2008b; Baeza and Anker,
2008; Baeza et al., 2008).

So far, the various studies on the sexual biology of
shrimps from the genus Lysmata suggest that all species
exhibit PSH. Protandric simultaneous hermaphroditism
is suspected to be a fixed trait in the genus. Nonetheless,
additional information from more species is needed to
confirm this notion. In turn, other life history traits differ
within these two genera. Shrimps have been reported to
inhabit the shallow subtidal and intertidal of subtropical
and tropical rocky and coral reefs around the world. Some
species of Lysmata live in groups, others are solitary, while

some species are socially monogamous (pair-living; e.g.,
L. grabhami (Gordon, 1935)) (Wirtz, 1997). Several spe-
cies with an inconspicuous coloration dwell freely among
rocks in temperate zones, while other more colorful spe-
cies inhabit tropical sponges (L. pederseni Rhyne and Lin,
2006) (Rhyne and Lin, 2006). Other strikingly brilliant
species clean fishes (L. amboinensis (De Man, 1888))
(Limbaugh et al., 1961). Species from this genus represent
ideal candidates to explore the role of ecological condi-
tions in explaining evolutionary innovations in the marine
environment (see Baeza and Thiel, 2007).

Recent studies have examined various aspects of the
biology of various Lysmata and Exhippolysmata shrimps
(Baeza, 2008b; Baeza and Anker, 2008; Baeza et al., 2008;
Lopez-Greco et al., 2009). Furthermore, shrimps from the
genera Lysmata and Exhippolysmata are currently being
used as models in evolutionary biology and behavioral
ecology because of their peculiar sexual system (Baeza and
Bauer, 2004; Baeza, 2006, 2007a, 2007b, 2007c). In spite
of the increasing knowledge regarding the behavior and
ecology of several species of Lysmata, the evolutionary
origins of PSH in the genus remain uncertain. Although
recent studies have shown that the variety of lifestyles
of Lysmata is greater than originally recognized (Baeza,
2008b; Baeza and Anker, 2008; Baeza et al., 2008), an
emerging dichotomy in social organization and ecology
was noted in initial studies. One group of species (named
“Crowd” species by Bauer, 2000) was described as inhab-
iting warm subtropical environments, occurring as dense
aggregations in their refuges, and exhibiting no specialized
fish-cleaning behavior (i.e., L. californica: Bauer and New-
man, 2004; L. wurdemanni: Baeza, 2006). A second group
(named “Pair” species by Bauer, 2000) was described as
mostly tropical, occurring at low densities in the subtidal,
and dwelling as socially monogamous pairs on sea anemo-
nes used as spots for fish-cleaning activities (i.e., L. grab-
ham: Wirtz, 1997; L. amboinensis: Fiedler, 1998). Based
on this initial dichotomy, Bauer (2000) proposed that PSH
evolved in the tropics from an ancestral symbiotic pro-
tandric species of Lysmata that became a specialized fish
cleaner. Restricted mobility of individuals resulting from
their association with the host and, hence, reduced prob-
ability of encountering mating partners would have fa-
vored PSH (also see Bauer, 2006). Under such a scenario,
the “Crowd” warm temperate species that do not exhibit
specialized cleaning behaviors would have evolved from
tropical species with specialized cleaning behaviors and
more complex mating systems (Bauer, 2006). A recent
phylogeny developed for the genus found no support for
Bauer’s hypothesis because socially monogamous species

presented a more derived position than gregarious species
(Baeza et al., 2009). However, no formal testing of Bauer’s
ideas was conducted. Current advances in ancestral char-
acter state reconstruction using Bayesian inference (Pagel
et al., 2004) make it possible to test whether the ancestral
Lysmata featured a symbiotic lifestyle and a socially mo-
nogamous mating system, as proposed by Bauer (2000).
Here, I provide evidence that PSH is a fixed trait
within the genus Lysmata (including Exhippolysmata),
as suspected by previous studies (see Bauer, 2000; Baeza,
2008b; Baeza and Anker, 2008; Baeza et al., 2008).
For this purpose, I examined the sexual system of two
shrimps from the genus, L. galapagensis Schmitt, 1924
and L. boggessi Rhyne and Lin, 2006, using anatomi-
cal observations and laboratory experiments. I also ex-
amined specimens from another nine species deposited
at the National Museum of Natural History (NMNH),
Washington, D.C. The information altogether strongly
suggests that PSH is a conserved trait within the genus
Lysmata. My second goal was to examine Bauer’s (2000)
hypothesis regarding the evolution of PSH in Lysmata.
I tested whether the ancestral Lysmata was socially mo-
nogamous (1) and strictly symbiotic with, for example,
sea anemones (2), as proposed by this author. To accom-
plish this second goal, a review of the literature on the
socioecology of Lysmata was conducted. Next, the life-
style of shrimps was mapped onto the phylogeny of the
genus, and the likelihood of specific traits to occur at
particular ancestral nodes in the phylogeny was tested.

METHODS
COLLECTION AND MAINTENANCE OF SHRIMPS

Individuals from the two studied species were collected
between February and August, 2006, at different localities
in Panama and Florida, USA. Individuals from L. boggessi
were collected at night during low tides from seagrass beds
at Madelaine Key (27°38'51.87”N, 82°42'56.50” W), Fort
De Soto National Park, Florida. Specimens from L. gala-
pagensis were collected from Islas Secas (7°58'37.54’N,
82°02'18.02”W), Gulf of Chiriqui, Panama. Immedi-
ately after collection, specimens were transported to the
R/V Urraca and then to the Naos Marine Laboratories,
Panama (L. galapaguensis) or directly to the Smithsonian
Marine Research Station at Fort Pierce, Florida (L. bog-
gessi). Individuals were maintained in 15—70 L aquaria at
a water temperature of 22°-33°C and 34-36 ppt salinity
and were fed every other day with shrimp pellets before
being selected for dissections or experiments.

NUMBER 38 ¢ 97

DISSECTIONS

Observations on reproductive anatomy were con-
ducted as in Baeza (2008b) in a total of six specimens of
each species, three presumptive males (3.6-3.8 and 4.0-4.6
mm carapace length [CL] in Lysmata galapagensis and L.
boggessi, respectively) and three presumptive hermaphro-
dites that were brooding embryos (4.4—5.1 and 6.5—5.6 mm
CL in Lysmata galapagensis and L. boggessi, respectively).
First, the presence or absence of male gonopores on the
coxae of the fifth pereiopods was recorded for each indi-
vidual. Individuals with male gonopores (all) had sperm
collected from the ejaculatory ducts using short electric
shocks that results in the ejection of a spermatophore (as
noted in Baeza, 2006, 2007c). Each individual was then
dissected to extract the gonad for examination under the
stereomicroscope. Finally, the first and second pleopods
were dissected and the presence or absence of appendi-
ces internae and masculinae, respectively, were recorded.
Specimens were defined as males or hermaphrodites by the
presence (males) or absence (hermaphrodites) of coupling
hooks (cincinnuli) and appendices masculinae on the en-
dopods of the first and second pleopods, respectively (see
Baeza, 2007c, 2008b).

EXPERIMENTS

Three experiments, as described in Baeza et al.
(2008) and Baeza (2008b), were conducted to determine
the sexual system of the three species under study. In
summary, the different experiments determined whether
(1) brooding shrimps (reproducing as females) were capa-
ble of mating as males, (2) brooding shrimps were capable of
self-fertilization, and (3) males were capable of becoming
hermaphrodites with time (see Results). In the first experi-
ment (7 = 5), pairs of brooding shrimps were maintained
in 21 L aquaria. In the second experiment, five brooding
shrimp were each maintained alone. In the third experi-
ment (n = 5), pairs of males (small nonbrooding shrimp
with no externally visible female gonads and visible cin-
cinnuli and appendices masculinae) were maintained
separately in 21 L aquaria for at least 50 days. Individu-
als were examined daily for hatching of the embryos, the
presence of exuvia from molting, development of mature
oocytes in the gonad (visible through the carapace), and
spawning of a new batch of eggs. The development of any
newly spawned embryos was examined in detail after four
days of spawning.

Following the rationale developed by Baeza et al.
(2008), if in the first experiment ovigerous shrimps that

98 © SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

paired together produced normally developing broods,
then it was inferred that either the other ovigerous shrimp
in the aquarium acted as a male to inseminate its partner,
or that the shrimp was capable of self-fertilization. If in the
second experiment shrimps in isolation failed to success-
fully produce and brood developing eggs, then the possibil-
ity of self-fertilization was eliminated. If in the third experi-
ment individuals identified as males at the beginning of the
experiment developed the ovarian portion of the ovotestis
and produced eggs, then I inferred that male shrimps ma-
ture as hermaphrodites (see Baeza et al., 2008).

POPULATION STRUCTURE, SEX RATIO, AND ABUNDANCE

Information on the abundance, population structure,
and sex ratio (males to hermaphrodites) of each species
was collected from the field. The carapace length (CL) and
number of shrimps of each sexual phase and each species
captured during the different samplings were recorded.
The sampling effort (total number of hours spent collect-
ing shrimps) was calculated for each sampling event. Rela-
tive abundance of shrimps was estimated by dividing the
sample abundance (number of shrimps captured) by the
sampling effort.

MUSEUM SPECIMENS

Specimens from nine different species of Lysmata de-
posited at the Collection of Crustaceans, National Mu-
seum of Natural History (NMNH; Smithsonian Institu-
tion, Washington, D.C.) were examined. Dissection of
specimens pertaining to the collection was not possible be-
cause only a few individuals were available from several of
the examined species and many of the specimens were part
of the type series used to describe the species. Therefore,
the identification of males and hermaphrodites was mostly
based on external morphological characters (see forego-
ing). When identifying sexual phases, particular attention
was given to the presence of male gonopores at the base
of the coxae of the fifth pair of pereiopods in brooding
shrimps as a likely indicator of simultaneous hermaphro-
ditism (see Results).

TESTING THE HISTORICAL CONTINGENCY HYPOTHESIS

To examine whether the historical contingency hy-
pothesis proposed by Bauer (2000) appropriately explains
the origins of PSH in shrimps from the genus Lysmata, the
lifestyle (in terms of the propensity to develop symbiotic
partnerships and natural group size) was reconstructed

using BayesTraits (Pagel and Meade, 2006; available at
www.evolution.rdg.ac.uk).

A pruned set of sequences (from the 16S mitochondrial
gene) recently published by Baeza et al. (2009) was used to
generate a phylogenetic hypothesis for the group on which
to reconstruct the evolution of lifestyles in shrimps. The
sequences pertained to 20 species of Lysmata and Exhip-
polysmata plus 3 other species (Merguia rhizophorae, Hip-
polyte williamsi, and H. inermis) used to root the trees
during the initial phylogenetic analysis. The set of aligned
sequences was first imported to BayesPhylogenies (Pagel
et al., 2004) to obtain a Bayesian posterior distribution
of phylogenetic trees. Metropolis coupled—Markov chain—
Monte Carlo analyses were conducted using a GIR + I
(invariant) + G (gamma) model of nucleotide substitu-
tion. The analysis was run on two different simultaneous
chains. A total of 6,000,000 iterations were conducted,
and sampling was performed every 100th tree. The last
1,000 posterior probability trees generated by BayesPhy-
logenies were then imported to BayesTraits. The submod-
ule MultiState in BayesTraits uses Markov chain Monte
Carlo (MCMC) methods to infer values of traits (that
adopt a finite number of discrete states) at ancestral nodes
of phylogenies. Additionally, this method permits testing
for particular ancestral characters at specific nodes taking
phylogenetic uncertainty into account (Pagel et al., 2004).

The two traits here analyzed have three states each.
For group size, the states were (1) aggregations (includ-
ing swarms), (2) small groups, and (3) pair-living (social
monogamy). The three character states used for describ-
ing the symbiotic propensity of different shrimp species
were (1) free-living, (2) facultative associate (with differ-
ent moray eel species, such as L. californica and L. seti-
caudata; with sea anemones, such as L. ankeri), and (3)
strictly symbiotic with either sponges (e.g., L. pederseni)
or sea anemones (L. amboinensis, L. grabhami). Informa-
tion on the lifestyle of each species was obtained by direct
observation of shrimps in nature (personal observations),
from the literature (see literature review), or from both
sources.

During the analysis, a reversible-jump MCMC search
was used with two independent chains that were run for
6,000,000 iterations with a burn-in of 50,000. I choose
the prior distribution of the parameters in the model with
the option Hyperprior (see Pagel et al., 2004), seeding an
exponential distribution from uniform on the interval 0.0
to 30 and a rate deviation of 18. These values were se-
lected considering preliminary runs and were used to keep
the acceptance rate at approximately 0.3, as recommended
by Pagel et al. (2004). Character states at internal nodes

were reconstructed using the most recent common ances-
tor method. I tested hypotheses about particular character
states at specific nodes when comparing the MCMC run-
in which the node was “fossilized” (constrained) to one
state versus an alternative. The command Fossil allows
testing whether a particular state is “significantly” more
likely at a specific node than an alternative state. For each
tested character, the same set of conditions (prior distri-
bution, burn-in) as used in the ancestral character state
reconstructions already described were used. However, the
MCMC was run 5 times for each trait state tested, and
a total of 100,000,000 iterations were conducted. Bayes
factors were calculated as the difference between the high-
est harmonic mean of the marginal likelihood from the
five MCMC runs for each state (Pagel et al., 2004). The
strength of support for one model over another was mea-
sured using the scale from Kass and Raftery (1995).

RESULTS
DISSECTIONS

Dissections demonstrated that all shrimps (brooding
or nonbrooding) from the two species had male gonopores
at the coxae of the fifth pair of pereiopods (Figure 1A).
Female gonopores at the coxae of the third pair of pe-
reiopods were more difficult to reliably observe. From all
shrimps (brooding or nonbrooding), sperm cells shaped in
the form of an inverted umbrella were retrieved from the
male gonopores by electroshocks (Figure 1A,B). Dissec-
tions of the gonads from small shrimps not brooding em-
bryos (presumptive males) demonstrated the presence of
an ovotestes (Figure 1C) with an undeveloped anterior fe-
male portion full of immature oocytes (lacking coloration)
(Figure 1D) and a posterior male gonad containing sperm
cells with the same morphology as the sperm retrieved
from the gonopores (see Figure 1B). Gonads dissected from
brooding (presumptive hermaphrodites) shrimps also had
ovotestes, but with a large ovarian portion full of mature
oocytes and a relatively small posterior testicular portion
with sperm (Figure 1E). In both brooding and nonbrood-
ing shrimps, vas deferentia and oviducts extended later-
ally from the testicular and ovarian portions, respectively
(Figure 1C,E).

Shrimps brooding embryos invariably lacked cincin-
nuli and appendices masculinae in the endopod of the first
and second pereiopods, respectively. In contrast, appendi-
ces masculinae bearing relatively long spines and numerous
cincinnuli were observed in the second and first pleopods,
respectively, of nonbrooding shrimps (Figure 1F—H). Some

NUMBER 38 °° 99

minor differences between the two species were noticed
regarding the relative length and number of spines borne
by the appendix masculinae; in L. boggessi, the spines
were more numerous and longer than those of L. gala-
pagensis (Figure 1G,H). Overall, all the anatomical dif-
ferences observed between brooding and nonbrooding
shrimps indicate that the populations of all the Lysmata
species studied herein are indeed composed of males and
hermaphrodites.

EXPERIMENTS

When two brooding individuals (presumed her-
maphrodites) were paired, all individuals in the two
species examined successfully hatched their embryos as
larvae, molted, and spawned a new batch of oocytes
below the abdomen. The oocytes remained attached to
the pleopods and showed embryonic development as em-
bryos (i.e., early blastulae formation) after three days.
This embryological development suggests the ability of
the other hermaphrodites in the same aquarium to re-
produce as males or, alternatively, the possibility of self-
ing by the hermaphrodites acting as females. However,
none of the 10 hermaphrodites (5 of each species) main-
tained in isolation from conspecifics successfully reared
their embryos to larvae. These solitary shrimps molted
and spawned oocytes to beneath the abdomen. However,
the oocytes invariably disappeared from the pleopods
within a few days after spawning. Overall, the observa-
tions from these first two experiments strongly suggest
that brooding hermaphrodites do not have the capability
of self-fertilization. Therefore, brooding shrimps (her-
maphrodites) maintained in pairs indeed acted as males
and fertilized eggs when their partners molted and repro-
duced as females.

In the experiment conducted to determine whether
males mature as hermaphrodites later in life, all six males
of L. galapagensis turned into simultaneous hermaphro-
dites within four months. Males showed signs of ovarian
maturation during intermolt periods. When the gonad was
full of large green (vitellogenic) oocytes, the male shrimps
molted into hermaphrodites. Most probably, these shrimps
mated as females shortly after molting for the first time in
their lifetime because the spawned embryos beneath the
abdomen were observed developing normally several days
after spawning.

In contrast to L. galapagensis, all six male shrimps
from L. boggessi died of unknown reasons within the
first month of the experiment. However, observations on
three males of L. boggessi in the maintenance aquaria

indicated that they turn into hermaphrodites before four
months. This change of sexual phase was accomplished
after a single month, as observed in L galapagensis. Thus,
it may be concluded that L. galapagensis and L. boggessi
are protandric simultaneous hermaphrodites, incapable of
self-fertilization.

POPULATION STRUCTURE, SEX RATIO, AND ABUNDANCE

Abundances of L. galapagensis and L. boggessi at
the different sampling locations were high and low, with
a mean of 2.79 and 0.317 individuals collected per min-
ute per sampling period, respectively. In the two species,
population was biased toward males. The ratio of males
to total shrimps collected during the sampling period was
0.024 and 0.16 for L. galapagensis and L. boggessi, re-
spectively. The range of body size registered for males
varied from 1.9 to 3.8 and from 3.13 to 5.75 mm CL in
L. galapagensis and L. boggessi, respectively. Hermaph-
rodites ranged in size between 4.1 and 5.1 and 5.63 and
6.5 mm CL in L. galapagensis and L. boggessi, respec-
tively (Figure 2).

MUSEUM SPECIMENS

A variable number of specimens from L. anchisteus, L.
argentopunctata, L. chica, L. kuekenthali, L. moorei, L.
philippinensis, L. rathbunae, L. trisetacea, and L. vittata
were available at the NMNH. Small shrimps in each spe-
cies appear to be males as they have cincinnuli and ap-
pendices masculinae in the second and first pleopod, re-
spectively. In turn, shrimps brooding embryos (the great

FIGURE 1. (facing page) Lysmata galapagensis and Lysmata bog-
gessi: anatomical and morphological differences between males and
hermaphrodites. A, spermatophore (arrow) retrieved from gono-
pores of hermaphrodite (L. boggessi); B, sperm from male (L. gala-
pagensis); C, ovotestes from male (anterior female and male portions
on top and bottom, respectively; arrow points at left vas deferentia)
(L. galapagensis); D, close-up of female gonad portion in male (arrow
points at immature oocyte) (L. boggessi); E, ovotestes from dissected
hermaphrodite (anterior female and male portions on the top and
bottom, respectively; top and bottom arrows point at right oviduct
and left vas deferentia, respectively) (L. galapagensis); F, endopod of
first pleopod in male (arrow points at cincinulli) (L. galapagensis);
G, endopod of second pleopod in male (arrow points at appendix
masculina) (L. galapagensis); H, endopod of second pleopod in male
(arrow points at appendix masculina) (L. boggessi).

NUMBER 38 e¢ 101

majority of them above average size) invariably lacked
cincinnuli and appendices masculinae in the endopod of
the first and second pereiopods, respectively. This last ob-
servation suggests they were hermaphrodites. It was not
possible to detect transitional individuals in these species
because no dissections were possible and gonad condi-
tion was not easily observed. The carapace of formalde-
hyde- and alcohol-fixed specimens is not translucent as it
is in living or recently preserved specimens. Also, shrimps
less than 3.0 mm CL were not sexed because of the risk
of inflicting damage. For all species examined except L.
anchisteus, L. argentopunctata, and L. philippinensis,
a relatively large sample of specimens was available.
The size-frequency distribution of the different species
strongly resembled that of the two species studied above,
with small shrimps resembling males and large shrimps
resembling hermaphrodites (Figure 3). Observations of
the coxae of the fifth pair of pereiopods of the largest
brooding shrimps in each species demonstrated the pres-
ence of male gonopores. Overall, the distribution of the
sexes across size classes and the limited observations on
the external male and female anatomy suggest that all
these other Lysmata shrimps are protandric simultaneous
hermaphrodites.

LITERATURE REVIEW

The literature review of the 41 species of Lysmata (in-
cluding Exhippolysmata) described to date revealed that
the geographic and bathymetric distribution, coloration,
and habitat of these species are relatively well known.
Shrimps from the genus Lysmata occur in tropical, sub-
tropical, and temperate waters around the world and can
be found among rocks or fossilized coral, live coral, sea-
grass blades, on muddy and shell bottoms, or associated
with sponges or sea anemones in the intertidal or subtidal
to 360 m depth. Most species have an inconspicuous color-
ation (red striped, translucent reddish with reddish flagella
on both pairs of antenna). Only 4 species are reported as
featuring a striking color pattern (contrasting body colors,
bright white flagella on both antenna). This dichotomy in
coloration was previously noticed by Bauer (2000). Lys-
mata splendida, one of the 4 species with a brilliant color-
ation, most probably is a cleaner shrimp. However, noth-
ing is known about its reaction to fish and its propensity
to clean them. Similarly, information regarding the degree
of specialization of the cleaning behavior is unknown for
most of the species (Table 1).

Information on the socioecology and sexual system
is, in general, poorly known. Information on lifestyle

102 e

oO Male

i Transitional |_| Hermaphrodite

90
L. galapagensis

60

Number of shrimps

1 (2 3 4 5 6 u 8

Size (CL, mm)

FIGURE 2. Lysmata galapagensis and L. boggessi population struc-
ture (7 = 178 and 22 shrimps, respectively, from L. galapagensis and
L. boggessi; CL = carapace length).

(socioecological attributes) is available only for 18 of the
41 described species. Of these, 7 species live in crowds
(aggregations), 7 species live in small groups, 3 species
live in pairs (i.e., they are socially monogamous), and 1
species is reported as living in extremely large aggrega-
tions (in swarms; Exhippolysmata oplophoroides). Dem-
onstration of PSH using a combination of experimental,
morphological, and anatomical findings and population
structure is available for 12 species. A strong indication
of PSH exists for another 10 species. Although the in-
formation is incomplete (PSH has been reported for a
total of 22 species, or 54% of the described species), this
review clearly demonstrates that the lifestyle and socio-
ecology of shrimps from this genus are more complex
than originally thought and further confirms the idea
that PSH is a fixed trait in the genus Lysmata (including
Exhippolysmata).

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

TESTING THE HISTORICAL CONTINGENCY HYPOTHESIS

The 50% majority-rule consensus tree obtained dur-
ing the initial phylogenetic analysis confirms the existence
of the three natural clades (tropical-American, cosmopoli-
tan, and cleaner) noticed previously by Baeza et al. (2009).
However, one important difference between the present
consensus tree and that previously published is that L.
olavoi is not supported as the most basal species within
the genus. This difference between trees might (1) be an ef-
fect of the different set of species used for the phylogenetic
analysis or (2) perhaps have occurred because the differ-
ent software programs used for phylogenetic inference
function with different algorithms. On the other hand, the
monophyly of Lysmata is well supported in this new tree,
with a 100% posterior probability (Figure 4; Baeza et al.,
2009: fig. 1).

The lifestyle of shrimps mapped onto the consensus
tree indicated that the most recent common ancestor of
the species pertaining to the neotropical and cosmopoli-
tan clades was gregarious. In contrast, the ancestor of
the species comprising the cleaner clade most probably
was socially monogamous (see Figure 4). On average,
the node of the most common recent ancestor of all Lys-
mata species is reconstructed to be in state 2 (social mo-
nogamy) with 80% of certainty. The degree of certainty
varied from tree to tree but was generally high, as indi-
cated by the low standard deviation of this value (SD
= 0.03, calculated from 2,000,000 iterations using 1 of
1,000 randomly sampled posterior probability distribu-
tion trees at each iteration). The largest harmonic log-
likelihood obtained from five independent runs when the
node was fossilized to state 0 and 2 was -—22.309507
and —20.865237, respectively. The almost three log-unit
improvement in likelihood (Bayes factor = 2.89) of the
model when the node was fossilized to state 2 represents
evidence that the ancestral lifestyle of Lysmata was so-
cial monogamy.

With regard to the propensity for developing symbi-
Otic interrelationships, the reconstructions suggest that
the ancestor of the neotropical and cosmopolitan clades
most probably had a free-living lifestyle and did not de-
velop any symbiotic partnership with other macroinver-
tebrates. It should be noticed that the degree of certainty
of these two inferences is relatively low, as indicated by
the large standard deviations of the distribution of the
character (see Figure 4). Also, the reconstructions indi-
cate that, with a probability of 0.46 + 0.20 or 0.41 +
0.18, either facultative partnerships or strict symbiosis,

[1 Male

L. trisetacea

L. vittata

0 4 8

Number of shrimps

NUMBER 38 e¢ 103

MB Hermaphrodite

L. moorei
12

L. rathbunae

L. kuekenthali

0

IZ 4 8 12

size (CL, mm)

FIGURE 3. Lysmata spp. Population structure of selected species from the National Museum of Natural His-
tory (7 = 71, 56, 22, 31, 70, and 57 shrimps from L. trisetacea, L. mooret, L. chica, L. rathbunae, L. vittata,

and L. kuekenthali, respectively).

respectively, was the ancestral state of the genus Lys-
mata. The improvement in the likelihood of the model
(Bayes factor = 1.51) when the node was fossilized to
state 2 (strict symbiosis) was low compared to when the
node was fossilized to state 0 (free-living). Therefore,
there is no evidence indicating that symbiosis is signifi-

cantly more likely than a free-living lifestyle in the an-
cestral Lysmata.

Overall, the present ancestral character state recon-
struction provides partial support for Bauer’s (2000) hy-
pothesis about the evolution of PSH in shrimps from the
genus Lysmata.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

104

panuiquoo

ETET dW ¢HSd é ihsXat é jeordoay, é 6TI-€L uvaqqiied avUungyIDs “T
MOC Wad HSd n 1d sdnoi Jeordoay, u $I uvaqqie) vfD4 “T
ajeroduiay oytoedg “7
I é é é é é ‘qestdonqns Y TI-0 ‘zapueuso, uen{ 1404404 “J
T é é é é é Jestdoary, é L9T sourddryryd sisuauiddiyiqd 7
KOGG dWad HSd iat ma sdno1n jeotdosy, S S uvaqqied tuassapad *T
“s] SEALS
87 9'T W ¢HSd é Imad é yestdory, é O9E-SET pur salozy 100v]0 “T
suey “A
ITT Wa HSd é ld é Jestdongqns é é “UBUPIIOUpI[ DIYIU “TT
077 dwWad HSd nN md PMOID jestdosy, YU SI syed “y sisuaplAvaDU “T
T é é é é é jestdory, é é vouyy “anoqilq DSSIISIYINUL “T
L é é é é é ajeroduiay, U S 9.9) I purleo7 MIN ipuvjosOu. “AL
DUeNY WN
LT6T W ¢HSd n 1a sdnoin jestdory, Eis I “uvaqqiie) 1a400UL “T
BOY yinos
‘oyloeg-Opu]
6C LTO T é HSd é "dq é yestdory, é II-0 “Bag Poy yogjuayany “7
if é é é é é jeordory, é Le ewing 1duay 7]
OUR
OVE T dW ad HSd an nat sdnoiny Jeotdoxy, OUV 77-0 “uvaqqiieD vipawsaqut “T
61 dW Ad HSd ie} IM" sdnoin jestdory, a I uvaqqiied 1q204 “T
IGIDeT
LEG W ¢HSd ie} "da é jestdoary, U 8€I-0 jeoidory “gq SUAISOANIIVAS “T
OUP NY
9T BL ET Wd HSd S queNyag sie jeordory, S SS> [epagqns ‘uvaqqie) MUDYQDAS “7
IUIDeg
6T TT dWaAd HSd ie} nat PMOID yestdory, YU 7-0 Jesidory, “J sisuasvdyps -T
IT TTT é HSd S quenyl3g sue jestdoary, su 87-01 oyioe g-Opu] sniyjagap “T
G é é é naa é festdory, ou SI sosedeyey 2149 “TT
LUTT dWad HSd nN Id PMOID yeotdory, su SI oytoed “yf vIIUAOfI VI “TT
67 dW d HSd ie} "dq PMOID jeotdory, Us 7-0 uvaqqiied 1s8a880q “T
oT dWad HSd ie} 2d sdnoiy —yesidonqng ad I SURLY “A piqng “T
© é é é 1q é yestdory, on! ¢$¢€-0 oyiseg “q-vivjaundoquasv ~T
LT dW Ad HSd n naar sdnoin jestdory, O'S 7-0 uevoqqiied MayUuv “T
eT é é é é é yeoidoary, We €-0 uvaqqiied SNAaISIqIUD “TT
oytoe g-Opu]
9T SIT aq HSd S quel sued jestdory, S S “Pag poy SISUDUIOQULD *T
I é é é é é jestdongns é 8r-CT BOLIFY YING anjasn] “J
aaa dW HSd é é suemsg —-feotdonqns u Gol SURLY “AV sapio1oqdojdo “y
I é é é é é jeotdoay, é 8r-T1 POLITY “AN saploqvysvy “J
CLOUT Wa ¢HSd é é é jeotdoay, é SRI Igoe g-Opu] SLqSOASUa “FJ
pSUsTojoY —-g SSJOF pasn pp (SS) waisks ,JOIABYDq UONeIO[OD = aAIS9FTT apnineyT = g }eqey py (Ut) uOTNqIIysIp sa19adg
dDUDPIAT yenxes suluvay) wOTNIIysIp s1yde1s005)
snoewAWeg

“O[Qe]IeAK JOU eIep So}edIpuT (¢) YIeUT UONsoNb y ‘YyDIUsKT pue vpus{joddigxy es1auss sy) Woy sdumtszys ul ASOJOIqOIDOS pue safAIsajI] JO AWSIIATG *T ATAV.L

105

NUMBER 38

“Apnas styi “67 SuoeotuNUIWOD |eUOSiad ‘uasURIy ‘gz ‘suUOTIBAIASqO paysI{qndun ‘77

‘0007 Fone ‘97 *paysyqndun “ye 19 rayUY “SZ ‘OS6T ‘sINY OH pue usYyod “pz ‘9007 “UIT pu audyyY ‘EZ £2007 JOyUY pue oudyy ‘ZZ $7007 ‘ZO9V.P WAPPA.P ‘TZ £8007 “Te 19 ezarg “OZ “8007 JOYUY

pur vzorg “61 L661 “ZI “8ST POOT “URWIMaN pur Joneg “ZT ‘GROOT “PZ9Pq “OT ‘866T AIPA “ST ‘6007 “Te 19 Seg “pT '7ZET ‘=pemBey ‘ET [0007 ‘AAsaoynang ‘ZT Seger ‘onIg “TT {Sper ‘sMYyIOH ‘OT
“9061 “UNguIeY “6 *7S6T “UNL pur sinyAOH “g *EZ6] “Isao pue sJorUsoID */ *[ 66] ‘uosuesy “9 ‘plET ‘dway ‘¢ ‘per ‘sweIIM “p STZET PPD ‘E {O00T “UOISIOUM “Z {ZE6T “2"VYD ‘T SEER ELENA
‘ammjonsys uoneindod = g ‘A8o0joydiow = yy ‘sjusutiedxe = q ‘suonsassip = C 5
‘wisiIporydeumsay snosurinurs spurioid = YSg P
‘paziyeisadsun = 4) ‘pezieisads = ¢ 5

*S[JPM JOOI JaaI UO ‘SUIDARD = A\ ‘pnut = YY ‘sarjjol = [
$s29B119} [2109 pazifIssoy = Jf ‘s[e109 Suoure = FD ‘9]qqn4 10O/pue suU10}10q AXD01 = Y ‘pues asivOd = eg {(pIVpHYINAS “T pur voIULOf109 “J JO aSed ay) UI sjea APIOU YUM) JUOIqUIAS = ¢ faRSTe YIM pajelsosse = VY q
‘sIajowl ul yidap 0} Jajor sioquinu ‘yepaqns = ¢ ‘[epiasoquI = | p
ee ee eee ee

I é é é é € é é é 9Y1eq-Opu] NLS 7
OIXSIN IONS
‘onUETTY “A

LTITET JW Ad HSd fel 1d pmol eordomqns {a 0€-0 ‘uvaqqiie) mMupimapénn “T
OPEN fal
“elpersny

Cr é é é é é jesidory, Vv vS—0 ‘oytoeg-Opuy PID WT,

ST é é é 1q é yesrdomqns Y S-p POLY AVN S1udOI19UN “'T
puryes7

Cl é é é é é jestdory, on: | OST-0 MAN “PIS poy IN MONEE ST
PISQUOPU]

i é é é "qd é jestdory, ES) C9> ‘anoqulq SETI) 7

CL é é é quel iag é jeordory, Om Sey SYP Opuy ppipuads “T

HG SCN ad Wad HSd nN "dq [peexon) jeoidory, Wwe I ET GEMS) EYEE ©

LS, é é é ¢ é jeotdouy, é SLT-OST aps odep sisdajouays “T
BOS PEA

‘onuepYy “aq
PTITT dWad HSd ie} 1d PMoIy —yeoidoanqns su Sit “uvsUPIIONpPay DIDPNvINIAS *]

106 e

DISCUSSION

The present study suggests that the sexual system in
shrimps from the genus Lysmata (including Exhippolys-
mata) represents a fixed trait. Anatomical observations,
behavioral experiments, and field samples demonstrated
that the 2 species studied here are protandric simultane-
ous hermaphrodites, as reported for all other Lysmata spe-
cies (Table 1). Size-frequency distributions and additional
but limited anatomical observations of museum specimens
further suggest that at least 9 other species are protandric
simultaneous hermaphrodites. Including the information
generated in the present study, PSH has been reported for
a total of 22 species, or 54% of the 41 species described
worldwide.

The well-conserved sexual system in Lysmata con-
trasts with that reported for other genera from the closely
related family Hippolytidae. For instance, two different
genera of Hippolytidae shrimps, Thor and Hippolyte, are
known to contain both gonochoric and strict sequentially
hermaphroditic species (Espinosa-Fuenzalida et al., 2008,
and references therein). The reasons for PSH to be fixed
in Lysmata are not clear, especially when considering
the diversity of environments inhabited by these species
(see Table 1). Different habitats with varying degrees of
structural complexity, seasonality, and predation regimes
should favor different sexual systems. For instance, the
rather heterogeneous environment (i.e., seagrass beds, sea-
weed meadows) in which the gregarious L. boggessi and
L. wurdemanni occur is expected to favor sequential her-
maphroditism over PSH. In these complex environments,
male mating success most likely decreases with increasing
body size because small body size is expected to increase
searching ability and, ultimately, male mating success
when encounter rate among conspecifics is high (Baeza
and Thiel, 2007). This small-male advantage together
with the well-reported exponential relationship between
fecundity and body size in female shrimps is expected to
favor strict protandry over simultaneous hermaphroditism
in these species (Charnov, 1982).

On the other hand, hermaphroditic shrimps are known
to experience brooding constraints (e.g., L. wurdemanni;
Baeza, 2007c), a condition that theoretically favors simul-
taneous hermaphroditism (see Charnov, 1982, and refer-
ences therein). Similarly, in socially monogamous Lysmata
(e.g., L. grabhami; Wirtz, 1997), infrequent encounter
rates among conspecifics should be favoring strict simulta-
neous hermaphroditism over PSH. It should pay (in term
of fitness) for each individual in a pair to reproduce both
as male and female as soon as possible during their life-

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

time because this strategy increases reproductive success
through both sperm donation to the partner and female
reproduction. Thus, an early male phase in these socially
monogamous species should not be adaptive. On the other
hand, differing costs between the sex functions might ex-
plain the existence of an early male phase before the si-
multaneously hermaphroditic phase in these monogamous
species. The relatively large energetic and temporal costs
of producing ova might delay maturation of the female
function, resulting in a functional adolescent male phase
previous to the simultaneously hermaphroditic phase (see
Baeza, 2006). Additional studies in gregarious and socially
monogamous cleaner shrimp species should improve our
understanding about the conditions favoring PSH under a
social monogamous mating system in Lysmata.

The literature review conducted herein indicates that
the diversity of lifestyles in the genus is greater than pre-
viously recognized. Initial studies reported a distribution
for the genus restricted to tropical-subtropical waters.
The present review suggests that shrimps also inhabit cold
temperate environments. Lysmata porteri is reported from
southern Chile, and L. morelandi inhabits New Zealand
(see Table 1). Because Exhippolysmata spp. represents a
derived group of Lysmata, the deep water environment
represents another environment colonized by the species
in this group (see Baeza et al., 2009). Also, the dichotomy
in social organization (“Crowd” versus “Pair” species)
noted in initial studies (Bauer, 2000) is not supported. In
addition to tropical pair-living and temperate gregarious
species, the present review indicates other species forming
swarms (extremely large aggregations) in temperate deep
water soft-bottom environments (i.e., E. oplophoroides) or
living in small groups in the tropical or subtropical intertidal
that might or not associate with sea anemones (L. ankeri;
Table 1). The possibility of an adaptive radiation in this
group of shrimps is currently being explored. The rather
unusual sex allocation pattern of this shrimps might repre-
sent the key innovation allowing species in these two gen-
era to colonize and persist in environments where species
with conventional sexual systems might fail.

The ancestral character state reconstruction analysis
conducted in this study provides partial support for Bau-
er’s (2000) hypothesis about the evolution of PSH in Lys-
mata. The analysis suggested that the ancestral Lysmata
shrimp lived as socially monogamous pairs either faculta-
tively associated to other macroinvertebrates or featuring
a strictly symbiotic lifestyle (with sea anemones, for exam-
ple). The free-living condition of several species pertaining
to the cosmopolitan and neotropical clades is likely to be
derived according to the present analysis. PSH might have

107

NUMBER 38

“UMOYS ST 991] YDEI UT SIPPID JUIIOFFIP O} JOJSIOUB UOUWIWWODS JUIDII JSOUT IYI JO 97e)S [e]

-sadue oy) SUIN.QSUOSAI UIYM pouTe}go (930.9 IYSII) sisoriquIAs pure (991) Ya) 9z1s dnors 104 AyIqeqord Jo11a3sod ayj ‘apou Yea OJ *sapou JosayJIp ye (9zIs dnoss pur sisorquids)

sa}e]S JeSsUR Pa}oNYsuUOdAI dy} JO saisusp Jolajsod ay SuIMOYs spoyjour InaussoyAyd aduaIoyUT ULIsaAeg SUISN PsUTe}qO 394} sNsUssUOD 9[NI-AWIOLeUT %~OS VW “b TUNOW

QL Ofer 0-2
vl OFOP 0: +

02 0F9b'0 :Z
BL OFLPO: Lb

LL OF6Z'0 : Z
SL OF6S'0: Lb

LL OFSE0:b
8zOFee 0-0

(Z) snoiquiks AyoINS A
(1) eaneynoey fj
(0) Guinj-ee14 [J

sisoiquwAs

8z0FZEO: 4
Ze0FSr0:0

OL OF Pr'0 :

£0'0FZE0: 4b
20'0FZE'0: 0

SL OF820:

QL 0F87'0 -Z
6L OF8h 0: b

91 OFE8'0-0

€L OFZ80 :

60-0 0:
QL OF8E'0 :

9} OFZ8'0 =z

ZL OFLG0:

02 0FPP'0 :
6L OFPS0 :

LV OFEEO: 4b
8L 0FZp'0:0

vLOFL80:

SZ'OF9EO:b
Ze OF8E'0: 0

90°0F96'0 :

0

-

QOQOOSMOOBOBBBOOOOOOO0O8BO00

1 (aisea) pwuewepinm °

Oo

ISWIENIIIM ajAjOddl}4
siuuaul ajAjodadl-}
aeloydoziys einBayy
eyepneaijas °
BIPEULIa}U! °
ISINYJOY ©

eyyjiu

JaJOOWw *
sisuebedeeBb °
saplosoydoydo *°
1y90y

snijaqgap *
imeygeib °
sisuauloquie *
JOABjO *
BOILUOYIED *
sisuajueeu *
sujsosoes6 *
luassapad *
ueyue *
isseb6og *

eye:

(XL) JuueWwAapINM °

QOOBSEEBOOB OBOBOA

OOBGoOa Ge

pO'OFPE'O: L
LO'0F8E0 0

£0 OF FEO: b
ZO 0FFE'0 : 0

60 0F€v'0 - 4
60'0F9b'0 0
60 0F67'0: b
pL OFSS'0:0
60'0F PEO: b
ZL OFES'0:0
80°0F62'0- +
LL OFSS'°0:0

OF9E 0 -
€L OF0E'0-0
£0'0F08'0 : Z

vl OFS6'0 :Z
60°0F26'0 2

8 0F6S'0:0

LO0FZEO- +b
QL OFLL'O: 4b LL OFPb'0: 0
LL OFSPO:b
OL OFZv'0: 0 LO0FCE0-4b
LL OFZS°0-0

(2) Bun tied fi
(1) sdnag fj

(0) suoneBaibby [J

ZL OF8S'0 :0
80 03 O-b
Z}0FZS'0:0

80'0FS6'0:0
azIS dnoi5

LEOF6L'0 :2

108 e

evolved in the tropics from an ancestral protandric species
of Lysmata that became a specialized fish cleaner, as sug-
gested by Bauer (2000). Restricted mobility of individuals
resulting from their association with the host, and, hence,
the reduced probability of encountering mating partners,
would have favored PSH (Bauer, 2000) (see foregoing for
further details about this hypothesis). Nonetheless, the in-
ferences about ancestral character states from the present
analysis need to be considered with caution. Indeed, the
present analysis did not support an ancestral symbiotic
condition as significantly more likely than a free-living
condition. Also, several internal nodes in the phylogenetic
tree were not well supported by the Bayesian analysis of
phylogenetic inference (see Baeza et al., 2009). This low
support for internal nodes, together with the breadth of
the posterior distributions of the character inferred for
these nodes, means that other alternative routes to the
evolution and maintenance of this peculiar sexual system
in Lysmata cannot be ruled out.

Among alternative historical scenarios (to that pro-
posed by Bauer, 2000), PSH might well have evolved
from a strict simultaneous hermaphrodite or even from
a strict gonochoric free-living ancestor inhabiting tropi-
cal environments. The evolution of PSH from an ances-
tral strictly simultaneous hermaphroditic condition has
been reported previously for the worm Ophryotrocha di-
adema, one of the few other marine invertebrates in which
PSH has been demonstrated (Dahlgren et al., 2001). Act-
ing together with the conditions favoring simultaneous
hermaphroditism (i.e., low abundance), sex-dependent
energetic costs might have favored an early maturation
of the male reproductive function compared to that of
the female function in the ancestral free-living shrimp (re-
gardless of its sexual system), ultimately resulting in the
evolution of PSH as we observe it today in Lysmata (and
Exhippolysmata). Similarly, brooding constraints experi-
enced by hermaphroditic shrimps might have favored the
retention of the male function later in life. If the space
for brooding embryos in the abdomen becomes saturated,
allocation of energy to sperm production is expected to
maximize fitness. This argument is similar to that of Ghis-
elin (1987) to explain apparent protogynous simultane-
ous hermaphroditism in chitons. In some species of poly-
placophorans, individuals brood eggs along the side of
the body. Early in life, they reproduce strictly as females
until they reach a size at which the space in which they
brood is saturated. At that point, the same individuals
start producing sperm while they are brooding. Brooding
constraints have been previously reported for at least one

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

species of Lysmata (L. wurdemanni; Baeza, 2006). New
studies are needed to confirm whether brood constraints
are common in the genus.

In the scenarios depicted here, we should expect
that, in a phylogeny of the group, “tropical-low abun-
dance” species would have a more basal position than
the “Pair” and “Crowd” species (“Pair” and “Crowd”
sensu Bauer, 2000). The rather complex mating system
(social monogamy) and specialized fish-cleaning behav-
ior of the “Pair” species most probably evolved from
“tropical-low abundance” species without complex
cleaning behavior and with rather simple mating systems
(i.e., without long-lasting associations between mating
partners), as appears to be the case for most shrimps
from the closely related family Hippolytidae. The unre-
solved position of the different natural clades with re-
spect to each other in the current phylogeny (see also
Baeza et al., 2009) constrain testing this last hypothesis
against Bauer’s (2000) ideas. Future studies attempt-
ing to resolve the natural relationships among species
of Lysmata, Exhippolysmata, and other related taxa to-
gether with the detailed examination of their sexual sys-
tem should allow explaining the origin of simultaneous
hermaphroditism in shrimps from the genus Lysmata.

Last, it is worth mentioning one of the main assump-
tions of the present analysis. PSH was treated as a sin-
gular innovation only originating in the genus Lysmata
(which contains Exhippolysmata), as initially suggested
by Bauer (2000). To the best of my knowledge, shrimps
from the genus Merguia, apparently the sister group
to Lysmata, seem to have a gonochoric sexual system.
However, this observation needs experimental confirma-
tion. Most importantly, future studies need to test for
the existence of protandric simultaneous hermaphrodit-
ism in members from other closely related genera (t.e.,
Mimocaris, Parahippolyte, Merguia, Merhippolyte).
These studies might reveal that PSH is not a singularity.
Indeed, PSH has independently evolved in the past at
least four other times outside the Caridea. In addition to
Lysmata shrimps, PSH has been confirmed in the poly-
chaete worm Ophryotrocha diadema (Premoli and Sella,
1995), the land snail Achatina fulica (Tomiyama, 1996),
the tunicate Pyura chilensis (Manriquez and Castilla,
2005), and the symbiotic barnacle Chelonibia patula
(Crisp, 1983). If simultaneous hermaphroditism turns
out not to be a singularity in shrimps from the families
Hippolytidae and Lysmatidae, then it should be possible
to explore the environmental conditions that favor this
unique sexual system in shrimps.

ACKNOWLEDGMENTS

I appreciate the support from the Smithsonian Tropi-
cal Research Institute (STRI, Panama City and Bocas del
Toro, Panama) Marine Fellowship and the Smithsonian
Marine Station at Fort Pierce (Fort Pierce, Florida, USA)
Fellowship. I thank Dr. Rachel Collin for inviting me to
join the 2007 Research Cruise to La Coiba Island and Las
Perlas Archipelago, off the Pacific coast of Panama, aboard
the R/V Urraca (STRI), during which a portion of this
study was conducted. I thank Dr. Klaus Ruetzler (National
Museum of Natural History, Smithsonian Institution), Dr.
Anson (Tuck) Hines (SERC) and Dr. Valerie Paul (SMSFP)
for funding various research visits to Carrie Bow, Belize,
during which a major portion of this manuscript was com-
pleted. I also thank Paula Rodgers and Rafael Lemaitre for
their support during my visit to the National Museum of
Natural History. The comments by two anonymous refer-
ees substantially improved this manuscript. This work is
contribution number 776 of the Smithsonian Marine Sta-
tion at Fort Pierce and contribution number 840 of the Ca-
ribbean Coral Reef Ecosystems Program (CCRE), Smithso-
nian Institution, and supported in part by the Hunterdon
Oceanographic Research Fund. This study was partially
funded by a National Geographic Research Grant from the
National Geographic Society, USA.

LITERATURE CITED

Baeza, J. A. 2006. Testing Three Models on the Adaptive Significance
of Protandric Simultaneous Hermaphroditism in a Marine Shrimp.
Evolution, 60:1840-1850.

. 2007a. Male Mating Opportunities Affect Sex Allocation in a

Protandric Simultaneous Hermaphroditic Marine Shrimp. Behav-

ioral Ecology and Sociobiology, 61:365-370.

. 2007b. No Effect of Group Size on Sex Allocation in a Protan-

dric-Simultaneous Hermaphroditic Shrimp. Journal of the Marine

Biological Association of the United Kingdom, 87:1169-1174.

. 2007c. Sex Allocation in a Simultaneously Hermaphroditic Ma-

rine Shrimp. Evolution, 61:2360-2373.

. 2008a. Social Monogamy in the Shrimp Pontonia margarita, a

Symbiont of Pinctada mazatlantica, in the Tropical Eastern Pacific

Coast. Marine Biology, 153:387-395.

. 2008b. Protandric Simultaneous Hermaphroditism in the
Shrimps Lysmata bahia and L. intermedia. Invertebrate Biology,
127:181-188.

Baeza, J. A., and A. Anker. 2008. Lysmata hochi n. sp., a New Species of
Hermaphroditic Shrimp from the Southern Caribbean. Journal of
Crustacean Biology, 28:148-155.

Baeza, J. A., and R. T. Bauer. 2004. Experimental Test of Social Media-
tion of Sex Change in a Protandric Sequential Hermaphrodite; The
Marine Shrimp Lysmata wurdemanni (Crustacea: Caridea). Behav-
ioral Ecology and Sociobiology, 55:544-550.

NUMBER 38 e¢ 109

Baeza, J. A., J. Reitz, and R. Collin. 2008. Protandric Simultaneous Her-
maphroditism and Sex Ratio in the Shrimp Lysmata nayaritensis.
Journal of Natural History, 41:2843-2850.

Baeza, J. A., C. D. Schubart, P. Zillner, S. Fuentes, and R. T. Bauer. 2009.
Molecular Phylogeny of Shrimps from the Genus Lysmata (Cari-
dea: Lysmatidae): The Evolutionary Origins of Protandric Simulta-
neous Hermaphroditism and Pair-Living. Biological Journal of the
Linnaean Society, 96:415—424.

Baeza, J. A., and M. Thiel. 2007. “The Mating System of Symbiotic
Crustaceans: A Conceptual Model Based on Optimality and
Ecological Constraints.” In Evolutionary Ecology of Social and
Sexual Systems: Crustaceans as Model Organisms, ed. J. E.
Duffy and M. Thiel, pp. 250-267. Oxford, UK: Oxford Univer-
sity Press.

Bauer, R. T. 2000. Simultaneous Hermaphroditism in Caridean Shrimps:
A Unique and Puzzling Sexual System in the Decapoda. Journal of
Crustacean Biology, 20(Spec. No. 2):116-128.

. 2006. Same Sexual System but Variable Sociobiology: Evolution
of Protandric Simultaneous Hermaphroditism in Lysmata Shrimps.
Integrative and Comparative Biology, 46:430-438.

Bauer, R. T., and G. J. Holt. 1998. Simultaneous Hermaphroditism in
the Marine Shrimp Lysmata wurdemanni (Caridea: Hippolytidae):
An Undescribed Sexual System in the Decapod Crustacea. Marine
Biology, 132:223-235.

Bauer, R. T., and W. A. Newman. 2004. Protandric Simultaneous Her-
maphroditism in the Marine Shrimp Lysmata californica (Caridea:
Hippolytidae). Journal of Crustacean Biology, 24:131-139.

Braga, A. A., L. S. Lopez-Greco, D. C. Santos, and A. Fransozo. 2009.
Morphological Evidence for Protandric Simultaneous Hermaphro-
ditism in the Caridean Exhippolysmata oplophoroides. Journal of
Crustacean Biology, 29:34-41.

Bruce, A. J. 1983. Lysmata debelius, New Species, a New Hippolytid
Shrimp from the Philippines. Revue Francaise d’Aquariologie et
Herpetologie, 4:115-120.

Burukovsky, R. N. 2000. Lysmata splendida sp. nov., a New Species of
Shrimp from the Maldives (Crustacea: Decapoda: Hippolytidae).
Senckenbergiana Maritime, 30:223-227.

Chace, F. A. Jr. 1972. The Shrimps of the Smithsonian—Bredin Caribbean
Expeditions with a Summary of the West Indian Shallow-Water
Species. Smithsonian Contributions to Zoology, No. 98. Washing-
ton, D.C.: Smithsonian Institution Press.

. 1997. The Caridean Shrimps (Crustacea: Decapoda) of the Alba-
tross Philippine Expedition, 1907-1910, Part 7: Families Atyidae,
Eugonatonotidae, Rhynchocinetidae, Bathypalaemonellidae, Pro-
cessidae, and Hippolytidae. Smithsonian Contributions to Zoology,
No. 587. Washington, D.C.: Smithsonian Institution Press.

Charnoy, E. L. 1982. The Theory of Sex Allocation. Princeton: Princeton
University Press.

Correa, C., J. A. Baeza, E. Dupré, I. A. Hinojosa, and M. Thiel. 2000.
Mating Behavior and Fertilization Success of Three Ontogenetic
Stages of Male Rock Shrimp Rhynchocinetes typus (Decapoda: Ca-
ridea). Journal of Crustacean Biology, 20:628-640.

Crisp, D. J. 1983. Chelonobia patula (Ranzani), a Pointer to the Evo-
lution of the Complemental Male. Marine Biology Letters, 4:
281-294.

Crosnier, A., and J. Forest. 1973. “Les Crevettes Profondes de I’ Atlantique
Oriental Tropical.” In Faune Tropicale 19. Paris: O.R.S.T.O.M.

Dahlgren, T. G., B. Akesson, C. Schander, K. M. Halanych, and P. Sund-
berg. 2001. Molecular Phylogeny of the Model Annelid Ophryotro-
cha. Biological Bulletin, 201:193-203.

Dohrn, P. FE. R., and L. B. Holthuis. 1950. Lysmata nilita, a New Species
of Prawn (Crustacea Decapoda) from the Western Mediterranean.
Pubblicazioni della Stazione Zoologica di Napoli, 22:339-347.

110 e

d’Udekem d’Acoz, C. 2002. Lysmata seticaudata (Risso, 1816) and
L. nilita Dohrn and Holthuis, 1950 Are Protandrous Simultaneous
Hermaphrodites (Decapoda, Caridea, Hippolytidae). Crustaceana,
75:1149-1152.

Espinoza-Fuenzalida, N. L., M. Thiel, E. Dupre, and J.A. Baeza. 2008.
Is Hippolyte williamsi Gonochoric or Hermaphroditic? A Multi-
Approach Study and a Review of Sexual Systems in Hippolyte
Shrimps. Marine Biology, 155:623-635.

Fiedler, G. C. 1998. Functional, Simultaneous Hermaphroditism in Fe-
male-Phase Lysmata amboinensis (Decapoda: Caridea: Hippolyti-
dae). Pacific Science, 52:161-169.

Fransen, C. H. J. M. 1991. Lysmata olavoi, a New Shrimp of the Fam-
ily Hippolytidae (Decapoda, Caridea) from the Eastern Atlantic
Ocean. Arquipelago-Life Marine Sciences, 9:63-73.

Ghiselin, M. T. 1974. The Economy of Nature and the Evolution of Sex.
Berkeley: University of California Press.

. 1987. “Evolutionary Aspects of Marine Invertebrate Reproduc-
tion.” In Reproduction of Marine Invertebrates: General Aspects:
Seeking Unity in Diversity. Volume 9, ed. A. C. Giese, J. S. Pearse, and
V. B. Pearse, pp. 609-665. Pacific Grove, Calif.: The Boxwood Press.

Holthuis, L. B. 1948. Note on Some Crustacea Decapoda Natantia from
Suriname. Proceedings of the Koninklijke Nederlandse Akademie
van Wetenschappen, Series C, Biological and Medical Sciences,
51:1104-1113.

Holthuis, L. B., and C. Maurin. 1952. Note sur Lysmata uncicornis
nov. spec. et sur deux autres espéces intéressantes de crustacés dé-
capodes macroures de la céte Atlantique du Maroc. Proceedings
of the Koninklijke Nederlandse Akademie van Wetenschappen,
55:197-202.

Kagwade, P. V. 1982. The Hermaphrodite Prawn Hippolysmata ensiros-
tris Kemp. Indian Journal of Fisheries, 28:189-194.

Kass, R. E., and A. E. Raftery. 1995. Bayes Factors. Journal of the Ameri-
can Statistical Association, 90:773-795.

Kemp, S. 1914. Notes on Crustacea Decapoda in the Indian Museum,
V. Hippolytidae. Records of the Indian Museum, 10:81-129.
Limbaugh, C., H. Pederson, and E. A. Chace, Jr. 1961. Shrimps That

Clean Fishes. Bulletin of Marine Science, 11:237-57.

Manriquez, P. H., and J. C. Castilla. 2005. Self-Fertilization as an Alter-
native Mode of Reproduction in the Solitary Tunicate Pyura chilen-
sis. Marine Ecology Progress Series, 305:113-125.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Pagel, M., and A. Meade. 2006. Bayesian Analysis of Correlated Evo-
lution of Discrete Characters by Reversible-Jump Markov Chain
Monte Carlo. The American Naturalist, 167:808-825.

Pagel, M., A. Meade, and D. Barker. 2004. Bayesian Estimation of Ancestral
Character States on Phylogenies. Systematic Biology, 53:673-684.

Policansky, D. 1982. Sex Change in Plants and Animals. Annual Review
of Ecology and Systematics, 13:417-495.

Premoli, M. C., and G. Sella. 1995. Sex Economy in Benthic Polychaetes.
Ethology, Ecology, and Evolution, 7:27-48.

Rathbun, M. J. 1906. South American Crustacea. Revista Chilena de
Historia Natural, 11:45—5S0.

Rhyne, A. L., and A. Anker. 2007. Lysmata rafa, a New Species of
Peppermint Shrimp (Crustacea, Caridea, Hippolytidae) from
the Subtropical Western Atlantic. Helgoland Marine Research,
61:291-296, doi:10.1007/s10152-007-0077-4.

Rhyne, A. L., and J. Lin. 2006. A Western Atlantic Peppermint Shrimp
Complex: Re-Description of Lysmata wurdemanni, Descrip-
tion of Four New Species, and Remarks on Lysmata rathbunae
(Crustacea: Decapoda: Hippolytidae). Bulletin of Marine Science,
79:165-204.

Schatte, J., and R. Saborowski. 2006. Change of External Sexual
Characteristics During Consecutive Moults in Crangon crangon
L. Helgoland Marine Research, 60:70-73, doi:10.1007/s10152-
005-0013-4.

Terossi, M., L. S. Lopez-Greco, and FE. L. Mantelatto. 2008. Hippolyte
obliquimanus (Decapoda: Caridea: Hippolytidae): A Gono-
choric or Hermaphroditic Shrimp Species? Marine Biology, 154:
127-135.

Tomiyama, K. 1996. Mate-Choice Criteria in a Protandrous Simulta-
neously Hermaphroditic Land Snail Achatina fulica (Férussac)
(Stylommatophora: Achatinidae). Journal of Molluscan Studies,
62:101-111.

Wicksten, M. K. 2000. The Species of Lysmata (Caridea: Hippolytidae)
from the Eastern Pacific Ocean. Amphipacifica, 2:3-22.

Williams, A. B. 1984. Shrimps, Lobsters and Crabs of the Atlantic Coast
of the Eastern United States, Maine to Florida. Washington, D.C::
Smithsonian Institution Press.

Wirtz, P. 1997. Crustacean Symbionts of the Sea Anemone Telmatactis
cricoides at Madeira and the Canary Islands. Journal of Zoology,
242:799-811.

Reconciling Genetic Lineages with Species
in Western Atlantic Coryphopterus
(Teleostei: Gobtidae)

Carole C. Baldwin, Lee A. Weigt, David G.
Smith, and Julie H. Mounts

Carole C. Baldwin, Lee A. Weigt, David G. Smith,
and Julie H. Mounts, National Museum of Natural
History, Smithsonian Institution, P.O. Box 37012,
Washington, D.C. 20013-7012, USA. Correspond-
ing author: C. Baldwin (baldwinc@si.edu). Received
13 May 2008; accepted 20 April 2009.

ABSTRACT. Species identification of western Atlantic Coryphopterus can be problem-
atic because some of the species are morphologically similar, there is confusing morpho-
logical variation within some species, no taxonomic key includes all currently recognized
species, and the validity of some species is questionable. The most recently published keys
do not include Coryphopterus tortugae or C. venezuelae, the validity of which as dis-
tinct from C. glaucofraenum has been questioned. Neighbor-joining trees derived from
mitochondrial cytochrome c oxidase I (COI) sequences (DNA barcoding) were used to
determine the number of genetically distinct lineages of Coryphopterus from collections
made off Belize, Curacao, and Florida. Additional specimens for genetic and morpho-
logical analysis were obtained from Panama, Venezuela, and the Bahamas. Subsequent
comparative analysis of preserved voucher specimens from which DNA was extracted
and digital color photographs of those specimens taken before preservation yielded, in
most cases, sufficient morphological information to separate the genetic lineages. Species
identification of the lineages was then determined based on review of original and sub-
sequent descriptions of Coryphopterus species and examination of museum specimens,
including some type material. Many museum specimens are misidentified. Twelve species
of Coryphopterus are herein recognized in the western Atlantic and Caribbean: C. alloi-
des, C. dicrus, C. eidolon, C. glaucofraenum, C. hyalinus, C. kuna, C. lipernes, C. per-
sonatus, C. punctipectophorus, C. thrix, C. tortugae, and C. venezuelae. Coryphopterus
bol Victor, 2008 is a synonym of C. venezuelae (Cervigon, 1966). Although genetically
distinct, C. glaucofraenum and some specimens of C. venezuelae are extremely similar
and cannot be separated on the basis of morphology 100% of the time. Comments on
the identification of each Coryphopterus species and a revised key to western Atlantic
species are provided.

INTRODUCTION

To provide specific identifications of larvae of Caribbean reef fishes at Car-
rie Bow Cay, Belize, a small coral-fringed island on the Belizean Barrier Reef
(16°48.5'N, 88°05’W), we have been matching larvae to adults through DNA
barcoding (mitochondrial cytochrome c oxidase I [COI] sequences). In addition
to greatly increasing our success rate of identifying larvae, DNA barcoding is
also providing a method of checking existing species-level classifications by re-
vealing the numbers of distinct genetic lineages within genera.

73 0

Attempts to identify Belizean Coryphopterus spe-
cies using the most recently published keys (Bohlke and
Robins, 1960, 1962; Bohlke and Chaplin, 1968; Murdy,
2002) proved problematic for certain species. None of
those keys includes C. tortugae (Jordan) or C. venezue-
lae Cervigon, presumably because the validity of both
species as distinct from C. glaucofraenum Gill has been
questioned (e.g., Bohlke and Robins, 1960; Cervigon,
1966; Thacker and Cole, 2002). Longley and Hilde-
brand (1941) and Bohlke and Robins (1960) consid-
ered C. tortugae (Jordan; type locality, Dry Tortugas,
Florida) a synonym of C. glaucofraenum Gill. Garzon-
Ferreira and Acero (1990) redescribed C. tortugae as
distinct based on new collections from the Colombian
Caribbean. Thacker and Cole (2002) acknowledged the
latter work but did not recognize C. tortugae in their
phylogenetic analysis of Coryphopterus species. Victor
(2008) recognized C. tortugae as distinct from C. glau-
cofraenum and identified what he considered a cryptic
new species within Garzon-Ferreira and Acero’s (1990)
C. tortugae, which he named Coryphopterus bol. Cer-
vigon (1994) elevated C. venezuelae from a subspecies
of C. glaucofraenum to a distinct species, but it was not
included in Murdy’s (2002) key or Thacker and Cole’s
(2002) and Victor’s (2008) molecular phylogenies of Cory-
phopterus species.

Another problem with identification of western Ca-
ribbean Coryphopterus is that stated distributions of
many species are conflicting, and some do not include
the western Caribbean. Greenfield and Johnson (1999)
identified nine species of Coryphopterus from Belize (all
of the 12 recognized herein except for C. venezuelae, C.
punctipectophorus, and the recently described C. kuna
(Victor, 2007)). Murdy (2002) listed only C. alloides,
C. dicrus, C. glaucofraenum, C. hyalinus, C. lipernes,
and C. personatus as having ranges that include Central
America, western Caribbean, or Caribbean. A search
for reef-associated species in Belize in FishBase (www
.fishbase.org) returned only C. alloides, C. eidolon, C.
glaucofraenum, and C. personatus.

The purposes of this paper are to assess the number
of valid Coryphopterus species known from the western
Atlantic and to provide comments on the identification
of, and a revised key to, those species based on results
of DNA barcoding, subsequent examination of voucher
specimens and color photographs of them, examination
of museum specimens, and reference to original and other
descriptions of the species. A neotype for C. glaucofrae-
num is designated because the location of Gill’s (1863)
holotype is unknown.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

METHODS

Depending on the locality, fish specimens were col-
lected using the fish anesthetic quinaldine sulfate or ro-
tenone. Specimens were measured to the nearest 0.5 mm,
photographed with a Fujifilm FinePix 3 digital camera
to record color patterns, sampled for genetic analysis,
and then preserved as vouchers. Tissue sampling for mo-
lecular work involved removing a muscle biopsy, eye, or
caudal body portion (depending on size) and storage in
saturated salt buffer (Seutin et al., 1990). Genomic DNA
was extracted from up to approximately 20 mg minced
preserved tissue via an automated phenol:chloroform
extraction on the Autogenprep965 (Autogen, Holliston,
MA) using the mouse tail tissue protocol with a final
elution volume of 50 wL. For polymerase chain reaction
(PCR), 1 wL of this genomic DNA is used in a 10 pL reac-
tion with 0.5 U Bioline (BioLine USA, Boston, MA) Taq
polymerase, 0.4 wL 50 mM MgCh, 1 pL 10 buffer, 0.5
wL 10 mM deoxyribonucleotide triphosphate (dNTP),
and 0.3 wL 10 uM each primer FISH-BCL (5’-TCAA-
CYAATCAYAAAGATATYGGCAC) and FISH-BCH (5’-
TAAACTTCAGGGTGACCAAAAAATCA). The thermal
cycler program for PCR was 1 cycle of 5 min at 95°C; 35
cycles of 30 s at 95°C, 30 s at 52°C, and 45 s at 72°C;
1 cycle of 5 min at 72°C; and a hold at 10°C. The PCR
products were purified with Exosap-IT (USB, Cleveland,
OH) using 2 pL 0.2X enzyme and incubated for 30 min
at 37°C. The reaction was then inactivated for 20 min at
80°C. Sequencing reactions were performed using 1 pL of
this purified PCR product in a 10 pL reaction containing
0.5 wL primer, 1.75 wL BigDye buffer, and 0.5 pL BigDye
(ABI, Foster City, CA) and run in the thermal cycler for
30 cycles of 30 s at 95°C, 30 s at 50°C, 4 min at 60°C,
and then held at 10°C. These sequencing reactions were
purified using Millipore Sephadex plates (MAHVN-4550;
Millipore, Billerica, MA) per manufacturer’s instructions
and stored dry until analyzed. Sequencing reactions were
analyzed on an ABI 3730XL automated DNA sequencer,
and sequence trace files were exported into Sequencher 4.7
(GeneCodes, Ann Arbor, MI). Using the Sequencher pro-
gram, ends were trimmed from the raw sequences until
the first and last 10 bases contained fewer than 5 base
calls with a confidence score (phred score) lower than 30.
After trimming, forward and reverse sequences for each
specimen were assembled, each assembled contig was
examined and edited by hand, and each sequence was
checked for stop codons. Finally the consensus sequence
from each contig was aligned and exported in a nexus for-
mat. Neighbor-joining trees (Saitou and Nei, 1987) and

distance matrices were generated using Paup*4.1 (Swof-
ford, 2002) on an analysis of Kimura 2-parameter (K2P)
distances (Kimura, 1980).

MATERIAL

The Coryphopterus material examined is listed in the
Appendix (Table A.1). This table includes the voucher
specimens represented in the neighbor-joining tree (Fig-
ure 1), as well as non-voucher specimens collected as
part of this or other projects. Most specimens exam-
ined genetically for this chapter are juveniles or adults,
except those of C. kuna; that species is represented in
our samples only by larvae. For most specimens ana-
lyzed genetically, a digital color photograph of the speci-
men taken before dissection and preservation is housed
at the Smithsonian Institution. Cytochrome c oxidase I
(COI) sequences for specimens analyzed genetically are
deposited in GenBank (accession numbers GQ367306-
GQ367475). Genetic information for several specimens
collected in the Bahamas was not available in time for
inclusion in the neighbor-joining tree, but identifications
of those specimens based on that information are dis-
cussed in the text.

RESULTS

Twelve distinct genetic lineages of Coryphopterus
are present in our material (see Figure 1). One of those
lineages, a single specimen identified as C. alloides from
Curacao is under additional investigation and is not dis-
cussed further here. Tissue samples of C. punctipectoph-
orus were not available for genetic analysis. The other
lineages, from top to bottom in Figure 1, are C. lipernes,
C. hyalinus, C. personatus, C. tortugae, C. glaucofrae-
num, C. venezuelae, C. dicrus, C. thrix, C. eidolon, C. al-
loides, and C. kuna. Comments on the identification of
each lineage, as well as C. punctipectophorus, are pro-
vided below. The COI sequence of Coryphopterus bol
Victor, 2008 (PR SIO0869, fig. 1 [SIO = Scripps Institu-
tion of Oceanography]) is part of the C. venezuelae clade,
and the synonymy of that species is discussed below.
Intra- and interspecific differences in percent sequence
divergence for COI for all species are provided in Table
1. We have not plotted distribution maps of Coryphop-
terus species because our samples are from a limited
number of locations, and historical confusion about the
identification of some species precluded our relying on

NUMBER 38 e¢ 113

geographic information based on museum catalogues.
Based on extensive recent collecting throughout the
Caribbean, Ross Robertson (Smithsonian Tropical Re-
search Institute, personal communication, 8 June 2009)
and James Van Tassell are providing distribution maps of
Coryphopterus species in their Shorefishes of the Greater
Caribbean CD, expected to be released in 2009.

Coryphopterus lipernes Bohlke and Robins,
1962

FIGURE 2

Our specimens of C. lipernes from Belize and Cura-
cao form a close genetic clade. Identification of C. liper-
nes presents no problems: It is distinguished from all Cory-
phopterus species except C. hyalinus and C. personatus
by the presence of black pigment surrounding the anus;
from C. hyalinus by the presence of a single (vs. two) an-
terior interorbital pore; and from C. personatus by color
pattern (see Figure 2). We did not make fin-ray counts for
C. lipernes, but according to Bohlke and Robins (1962),
C. lipernes also differs from C. personatus in having 10
(vs. 11) second dorsal- and anal-fin elements. Murdy
(2002) distinguished C. lipernes and C. personatus from
C. hyalinus by the presence of two pores between the eyes
(vs. three), but as noted by Bohlke and Robins (1962),
there is one anterior interorbital pore in C. lipernes and
C. personatus and two in C. hyalinus.

Coryphopterus hyalinus Bohike and Robins,
1962

FIGURE 2

The validity of C. hyalinus as distinct from C. per-
sonatus has been questioned (e.g., Smith et al., 2003), but
the two are genetically distinct (see Figure 1, Table 1).
Of the Coryphopterus gobies with a black ring of pig-
ment around the anus (C. hyalinus, C. personatus, and
C. lipernes), C. hyalinus is the only one with two anterior
interorbital pores (BOhlke and Robins, 1962; Bohlke and
Chaplin, 1968). Because head pores can be difficult to see
in fresh material (considerably easier to see in preserved
specimens), separation of C. hyalinus and C. personatus
in the field can be difficult. We have observed no con-
sistent differences in pigmentation in fresh or preserved
specimens of the two species, but we often collect C. hya-
linus in deeper water than C. personatus.

BZE 4067

BZE 4082
: BZE 7999
& lipernes CUR 8326
CUR 8051
CUR 8327 Agi
BZE 4512
BZE 6221
CUR 8264
BZE 6222
BZE 7760
C. hvali COR 8268
5 ya Inus CUR 8266
- CUR 8046
- CUR 8044
BZE 4014.
CUR 8045
ZE 4307
E 4309
N 7712 1
PAN 7712 5
C. personatus BEE 08
BZE 4079
BZE 5067
BZE 4016
7725 6
736 4
4530
7708
BZE 5238
7106
BZE 7692
BZE 7734
ZE 7107
BZE 7693
709
690
5s
7736 1
C.tortugae “es

C. venezuelae

C. glaucofraenum

BZE 5099

C. thrix

C. eidolon

C. alloides

- BZE 4089

CUR 8325

BZE 4586
BZE 6049
PAN GB EF550211
- BZE ey

— 0.005 substitutions/site

Coryphopterus personatus (Jordan
and Thompson, 1905)

FIGURE 2

Identification of C. personatus also presents no prob-
lems using published keys. It can be distinguished from
C. hyalinus by the presence of a single interorbital pore
and from C. lipernes by pigment pattern (see Figure 2).
According to Bohlke and Robins (1962), C. personatus
also can be separated from C. lipernes by having 11 (vs.
10) total elements in the second dorsal and anal fins.

Coryphopterus tortugae (Jordan, 1904)
FIGURE 3

Longley and Hildebrand (1941) and Bohlke and
Robins (1960) considered C. tortugae (Jordan: type lo-
cality, Dry Tortugas, Florida) to be a synonym of C. glau-
cofraenum Gill. Garzon-Ferreira and Acero (1990) rede-
scribed C. tortugae as distinct based on new collections
from the Colombian Caribbean. Victor (2008) concurred
with Garz6n-Ferreira and Acero’s (1990) recognition of
C. tortugae but noted that their Santa Marta specimens
constitute a distinct species, which he described as C. bol.
As noted below (see “Synonymy of Coryphopterus bol”),
C. bol appears to be a synonym of C. venezuelae.

We had initially identified all specimens of the C. tor-
tugae, C. glaucofraenum, and C. venezuelae clades as C.
glaucofraenum using published keys (Bohlke and Robins,
1960; Bohlke and Chaplin, 1968; Murdy, 2002). How-
ever, those specimens separate into three well-defined lin-
eages based on COI sequences. Specimens in one of those
lineages are usually paler than those of the other two and
almost always have a central bar of basicaudal pigment
(vs. usually two spots or a dumbbell- or C-shaped mark-
ing), characters described by Garz6n-Ferreira and Acero
(1990) as diagnostic for C. tortugae. Bohlke and Robins
(1960), who considered C. tortugae to be a pallid form of
C. glaucofraenum, noted that the pigment markings along
the side of the body are round (upper row) or vertically
elongate (lower row) versus X-shaped as in C. glauco-
fraenum, usually a consistent feature in our specimens of

FIGURE 1. (facing page) Neighbor-joining tree derived from cyto-
chrome c oxidase I sequences showing genetically distinct lineages
of western Atlantic Coryphopterus.

NUMBER 38 e¢ 115

C. tortugae. The pigment spots in the lower row of mark-
ings along the side of the body in C. tortugae are usually
vertically elongate (crescents or some part of an X), but
they are rarely distinct X-shaped markings. If some of the
anterior markings do resemble X’s (Figure 3D), the height
of each X is considerably smaller than the height of the X’s
in C. glaucofraenum and, when present, in C. venezuelae
(half or less of eye diameter in C. tortugae, approximately
three-quarters of or equal to eye diameter in the other two
species). The pigment spots in the lower row also are not
rounded, as they are in pale specimens of C. venezuelae.

We have not found the basicaudal pigment to be a
reliable character for separating C. tortugae from C. glau-
cofraenum and C. venezuelae, as all three species may
have a central bar of pigment; however, C. tortugae does
not have two distinct spots in any of our material, so if
that feature is present in a specimen, it is not C. tortugae.
Coryphopterus tortugae shares with C. glaucofraenum
and C. venezuelae the presence of a distinct dark blotch
or triangle behind the eye above the opercle and with C.
glaucofraenum the absence of a pigment spot on the lower
portion of the pectoral-fin base. Garzon-Ferreira and Ace-
ro’s (1990) redescription of C. tortugae did not mention
the absence of this spot, presumably because the Santa
Marta specimens included in their description do have the
spot and appear to be C. venezuelae (see “Synonymy of
Coryphopterus bol,” below). Our investigations indicate
that the absence of this pigment spot on the pectoral-fin
base, along with the presence of vertically elongate versus
round pigment spots in the lower row of markings on the
body, is significant in separating C. tortugae from pale
specimens of C. venezuelae. Examination of photographs
of the holotype of Ctenogobius tortugae (SU 8363) con-
firms that there is no pigment on the lower portion of the
pectoral-fin base.

Coryphopterus tortugae is most easily separated from
all other Coryphopterus by the following combination of
characters: a dark blotch or triangle of pigment above the
opercle is present; large X-shape markings on the side of
the body and a spot on the lower pectoral-fin base are
absent; at least some of the pigment markings in the lower
row along the side of the body are vertically elongate or
crescent shaped; and the overall coloring is pale.

Coryphopterus glaucofraenum Gill, 1864
FIGURE 4
The location of the single type specimen upon which

Gill described C. glaucofraenum is unknown (Eschmeyer,
2008). Bohlke and Robins (1960:108-109) described

nnn

(pTI-ST'0) (O€'¥7-6S°ET) (S9'ET-LL7TZ) (OT 9T-L9' FT) (LT 9T-96' HT) (LZS°S7-9EET) (¥9'9T-BH'HTZ) (€S°8~T-6L°9TZ) (LT 9T-8S'b7Z) (LE°9T-98'bZ) (S9°E7Z-Z9'TT) (00°L7—6S°SZ)

%LS5°0 %L6ET WOE'ET WS ST WEI ST %lL6'bT %8L ST %L8°LT AUS ST WOL'ST WET ET Iv 97 puny
eu (88°6-8b'6) (06°61-7Z8°8L) (8E°6I-OL6L) (O8°ZI-OZ'9T) (€7'O7T-8S'8T) (T8'LZ-PO'LZ) (SI'ZZ-8EEZ) (F66I-40°6L) (96°LI-Z9'LL) (89°LZ-ZO'EZ)

= eu %89'6 Arve 61 WIV EL IV LI %89' 61 %OE VT HEL VT %LT 61 ASLLV ST IT 7 Sapiojjp
(T€0-91'0) (80°6I-8T LT) (SL TZ-L661) (€0°6I-@S°ZT) (OL 61-Z9'Z1) (8h T7-86'07) (FI'OT-6Z'81) (€6°81-69°ZT) (¥9°8I-60°ZT) (€5°7Z-ES‘TZ)

- = %1T0 %90°8T WEE OT WST'8L TIBI %69 IT Abr 6l WIV BI %06°LT EV CT [ sapioyy
(66°0-00'0) (Lb'07-96°81) (Z8°07-69°81) (9E'07-SL'9T) (¥O'ST-86'ITZ) (ZH IT-PE'SL) (Z8'8I-ZE'LI) (LEOC-61'81) (61°97-Sh' +7)

- = = rT 0 Avs 61 WEE 61 ATL BI MLV ET rl 61 %T6 LI %6l 61 WIV ST uojopia
(8¢'0-00'0) (F8°LT-hZL'07) (ZL OT-19'LZT) (FS TZ-O€ 07) (E9'6I-EF'BT) (Sh'OT-6L'6L) (bb'ZZ-IET'0Z) (69°7Z-§ 8°07)

= = = = %IL‘0 WOE TT %WOE 61 %EO IT % 00°61 %OL 61 YOU TT WIS TT x1q)
(78°7-00'0) (0061-991) (TETT-Z8'6L) + (LO'8I-6S°9T) (F66I-OE'8I) (ZEO7T-19'8T) (98°7Z-OL'0Z)

= = = = = %19'0 W8T BI %S90T HES LL *WEO6T %89I 61 ACL IT sndzip
(v7 1-00'0) (IT'O1-65°8) (ZZ°OI-O1'8) (8T°'L7-85°8T) (69° 17-6S°6L) (€L'TZ-LT'6T)

= - = = = = %ESO %1S6 V8 6 ATL OT %I8' OT ALE TT apjanzauan
(76'0-00°0) (66°TI-ZO'LL) (O€ 7Z-8¥'07) (he'~7-19'07) (OT IT-6F'61)

- - - = = = = %6l'0 LOT %OS TT %8I1T %IL OT unuav.joonvs
(19°0-00'0) (ZO'Lt-9'61) (78° 17-9°07) (€7'0T-S9'8T)

- - = = = = = = %0T'0 %80'0T MvV IT %09'61 avsny.0}
(9¢°0-00'0) (S9°Z-6Z°9) (€0°9T-OT'ST)

= - - = = = = = = %brl'0 WyVL %IIST snypuosiad
(T€0-00'0) (Or ST-I7' rT)

= - - - = = = = = = %90°0 88 vl snuyonky
($€°0-00'0)

= = = = = = = = = = - YEO sausady

(¢ = 4) (L=4)7 (€ =4) T (61 = 4) (Z = 4%) (TZ = 4) (€€ = 4) (67 = 4) (IZ = 4) (OL = 4) (IL = 4) (Z = 4) “ds

puny saployjv sap1ojjv uojopia XIG4 snArip avjanzaueaa wnuapsjoonyjs apsny10}4 snypuossad snuypiy sausagy snsaydoqdaioy

‘ayqeyieae jou ejep = e/U pjog ul UMOYs are sasesaAK yIDadseIIU] *] VINS1J ul 9939 Surutol-roqysrou

yi ul pajuasasdas sjenpratpul Jo saouanbas ([QD) J aseplxo 2 aut01y901A9 UO pasegq saideds smsazdoqdd1oy soy AeuruNs aouLIsIp Jaj}auUIeIed-om) vINUIDY (asUeI puL) ddeIOAY “] FIAV.L

SSS SSS a a TS I ES a SS RE SR RT TSE SS RE I PSE SE ES TT 9 SET ERR SE]

NUMBER 38 e¢ 117

FIGURE 2. Coryphopterus lipernes: A, Curacao, 20 mm SL, DNA 8326, USNM 394896; B, Curacao, 21 mm SL, DNA
8051, USNM 394895. Coryphopterus hyalinus: C, Curacao, 20 mm SL, DNA 8044, USNM 394890; D, Curacao, 17
mm SL, DNA 8265, USNM 294889. Coryphopterus personatus: E, Curacao, 21 mm SL, DNA 8045, USNM 294897;
F, Belize, 15 mm SL, DNA 7163, USNM 394742.

118 e© SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

; ee air Pa il es ee = = :
a ee eo
C
a : a
% ; ’ :
? } f } ' 2
D
Ce ea
‘ a dio ue eS
is " ied i Fs k oy + Fr ;* @2

Bee aes

FIGURE 3. Coryphopterus tortugae: A, Belize, 25 mm SL, DNA 7333, USNM 394744; B, Belize, 34 mm SL, DNA 5237, USNM
394743; C, Belize, 36 mm SL, DNA 7107, USNM 394733; D, Belize, 40 mm SL, DNA 4530, USNM 394730; E, Belize, 40 mm SL,
DNA 4530, USNM 394730, preserved; F, Venezuela, 37 mm SL, DNA 7736 4, AMNH 247340, alcohol preserved.

two forms of C. glaucofraenum: “(D]ark inshore form
(typical glaucofraenum)” and” [P]allid white-sand form.”
Specimens in our genetic clade identified as C. glauco-
fraenum match the Bohlke and Robins (1960) “typical
glaucofraenum,” an identification supported by the fact
that the pallid form is now recognized as C. tortugae.
Below (see “Designation of Neotype for Coryphopterus
glaucofraenum”) we select a neotype for C. glaucofrae-
num Gill.

In our material, adult C. glaucofraenum can always
be separated from C. tortugae by having at least some
large, well-formed X-shaped markings along the side of
the body. It can almost always be separated from C. ven-
ezuelae by lacking a prominent dark marking on the lower
portion of the pectoral-fin base and sometimes by having
10 total anal-fin elements. Rarely, C. glaucofraenum has a
dark pectoral-fin base that includes pigment on the lower
portion (Figure 4G), and C. venezuelae may have 9-11
anal-fin elements, 10 being the typical count in our mate-
rial (Table 2). Coryphopterus glaucofraenum usually can
be separated from both C. tortugae and C. venezuelae by
the shape of the pigment marking above the opercle: a
two-peaked blotch in C. glaucofraenum, and a triangular
or rounded blotch in C. tortugae and C. venezuelae.

If a specimen has a two-peaked blotch of pigment
above the opercle, has at least some large (height ap-
proximately three-quarters of or equal to diameter of
eye) X-shaped markings along the side of the body, has
10 anal-fin elements, and lacks pigment on the lower
portion of the pectoral-fin base, it is unquestionably
C. glaucofraenum.

Coryphopterus venezuelae (Cervigon, 1966)
FIGURE 5

The most recent keys to western Atlantic Cory-
phopterus (Bohlke and Robins, 1960, 1962; Bohlke and
Chaplin, 1968; Murdy 2002) do not include C. venezue-
lae, originally described as a subspecies of C. glaucofrae-
num by Cervigon (1966), but recognized as a separate
species by Cervigon (1994) and known at the time only
from Venezuela. In the Coryphopterus material from the
northeast coast of Venezuela that we examined are speci-
mens that are clearly C. venezuelae based on Cervigon’s
(1966, 1994) descriptions: most notably the presence of
11 second dorsal- and anal-fin elements, a dark blotch of
pigment on the lower portion of the pectoral-fin base, and
two dark spots on the base of the caudal fin (e.g., Figure
5D herein). However, those Venezuelan specimens are part

NUMBER 38 e¢ 119

of a clade based on COI analysis (see Figure 1) that in-
cludes specimens from Venezuela, Curacao, Panama, Belize,
Puerto Rico, and the Bahamas (the last not shown on the
tree) that usually have 10 second dorsal- and anal-fin ele-
ments and various patterns of pigment on the base of the
caudal fin, including a central bar, two spots joined by a
bar, and a C-shaped blotch (Figure 5A—C,E). The Venezue-
lan specimens on the tree (Figure 1), including two that
have 10 second dorsal- and anal-fin elements (VEN 7733
1 and VEN JV12), cluster within the C. venezuelae clade,
but the genetic distance between the Venezuelan specimens
and other members of the clade is only 0.41% to 0.85%.
This distance is extremely small relative to the genetic
distance between the C. venezuelae clade and other spe-
cies on the tree (9.51%-20.86%; see Table 1), suggest-
ing that the individuals in this clade represent a single
species. Corroborating the identification of the clade as
Cervigon’s C. venezuelae is the presence in all individu-
als in the clade of a dark spot on the lower portion of the
pectoral-fin base. Among western Atlantic Coryphopterus,
only C. punctipectophorus and C. dicrus have a promi-
nent pigment spot on the lower portion of the pectoral-fin
base: C. punctipectophorus is not known from the Carib-
bean, and it differs morphologically from C. venezuelae
in, among other features, lacking a dark blotch of pigment
behind the eye above the opercle; in C. dicrus, there is also
a prominent spot of equal size on the dorsal portion of the
pectoral base that is lacking in C. venezuelae (which may
have a slash of pigment but never a well-defined dorsal
spot equal in size and intensity to the lower spot); C. di-
crus also lacks the dark pigment behind the eye above
the opercle and lacks a pelvic frenum (both present in
C. venezuelae).

Our data thus suggest that C. venezuelae is a much
more widespread species than previously recognized, and
fin-ray counts alone are not sufficient in diagnosing the
species. Cervig6n (1994) believed that the presence of 10
second dorsal- and anal-fin elements in C. glaucofraenum
distinguished it from C. venezuelae. In his material of the
latter, all specimens had 11 second dorsal-fin elements and
most had 11 anal-fin elements (two had 10). Most of our
specimens of C. glaucofraenum have 10 second dorsal-
and anal-fin elements, but two specimens have 11 second
dorsal-fin elements, and two have 9 anal-fin elements (see
Table 2). Both 10 and 11 second dorsal- and anal-fin ele-
ments are Common in specimens in our C. venezuelae clade
(Table 3), although we found 11 in both fins only in some
of our material from Venezuela. It is significant that one
of the C. venezuelae specimens from Venezuela that has
10 elements in both fins was caught in the same sample as

120

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 4. Coryphopterus glaucofraenum: A, Belize, 44 mm SL, DNA 6367; B, Belize, 25 mm SL, DNA 7352,
USNM 394354; C, Belize, 35 mm SL, DNA 7351, USNM 394353, preserved; D, Venezuela, 31 mm SL, DNA 7744
2, AMNH 2473339, alcohol preserved; E, Venezuela, 27 mm SL, DNA 7744 3, AMNH 2473339, alcohol preserved;
F, Panama, 34 mm SL, DNA 7712 2, AMNH 247335, alcohol preserved; G, Panama, 37 mm SL, DNA 7701 1,
AMNH 247334, alcohol preserved.

several with 11 in both fins. There is thus more variability
in numbers of second dorsal- and anal-fin elements than
Cervigon indicated, and those fin-ray counts are of value
in separating C. glaucofraenum and C. venezuelae only
when 11 elements are present in both fins—a condition we
have not observed in C. glaucofraenum, which may have
11 second dorsal-fin elements but no more than 10 anal-
fin elements (see Table 2).

If a specimen has a dark blotch or triangle of pigment
above the opercle, 11 second dorsal-fin and 11 anal-fin
elements, and a prominent pigment spot on the lower por-
tion of the pectoral-fin base, it is C. venezuelae.

If a specimen has those features and has 10 second
dorsal- and anal-fin elements, it is usually C. venezuelae
but could be C. glaucofraenum: as noted under “Cor-
yphopterus glaucofraenum,” rarely specimens of that
species may have pigment on the ventral portion of the
pectoral-fin base. The shape of the pigment marking
above the opercle (with two peaks in C. glaucofraenum,
a single triangular or rounded blotch in C. venezuelae;
see “Coryphopterus glaucofraenum”) will frequently re-
solve the species identification.

There are two distinct forms of C. venezuelae in
terms of body pigment: one has at least some large X-
shaped markings in the ventral row of markings similar
to those of C. glaucofraenum (Figure 5B,D,E); the other
is a much paler form, and the ventral pigment mark-
ings along the side of the body are usually fairly small,
somewhat circular blotches (Figure 5A,C). Both forms,
including the palest specimens, have a pigment spot on the
lower pectoral-fin base, but this spot may be composed
primarily of yellow chromatophores versus melanophores
in pale specimens. The less-pigmented form is most easily
confused with C. tortugae, but some of the pigment spots
in the ventral row of C. venezuelae are usually more cir-
cular than the vertically elongate ones of C. tortugae. Ad-
ditionally, none of our specimens of C. tortugae has a
spot of pigment (yellow or black) on the ventral portion
of the pectoral-fin base. Although unusually divergent
intraspecifically in patterns of pigmentation (see Figure
5) relative to, for example, the very similar patterns be-
tween species such as C. personatus and C. hyalinus, the
two forms of C. venezuelae form a tight genetic clade
(intraspecific variation, 0.53%; see Figure 1, Table 1).
The different pigment patterns do not correspond to dif-
ferent fin-ray counts, as we have observed 10 and 11
second dorsal- and anal-fin elements in both forms. For
example, note the similar patterns of pigmentation in a
specimen of C. venezuelae from Venezuela (Figure 5D)
that has 11 second dorsal- and anal-fin elements and
a specimen of C. venezuelae from Panama (Figure 5E)

NUMBER 38 e¢ 121

TABLE 2. Frequency distributions of numbers of second dorsal-
fin and anal-fin elements in two species of Coryphopterus.

No. of
anal-fin elements

No. of second
dorsal-fin elements

Species 10 11 9 10 11
C. glaucofraenum 22 2 2 20 —
C. venezuelae 20 ils} 1 22 11

that has 10 second dorsal- and anal-fin elements. Fur-
thermore, the differences are not attributable to sexual
dimorphism or geography, but they could reflect differ-
ences in local habitat. Some specimens of C. venezuelae
collected in mangrove areas tend to be dark, and those
collected in reef areas pale, although we note that a dark
form was collected on a reef off Panama (Figure SE).

There is some correlation with size: the pale form
of C. venezuelae is more common among small speci-
mens (<30 mm standard length [SL]), and the form with
prominent X-shaped markings is more common among
larger specimens (>40 mm SL). Adults of the pale form
of C. venezuelae (e.g., Figure 5A) look remarkably simi-
lar to juveniles (e.g., see Figure 7C). There is also a trend
toward increasing depth of the head and anterior body in
larger specimens. Similar differences in body shape and
pigment with increasing size are evident in C. glaucofrae-
num (compare the juvenile in Figure 7B with adults in
Figure 4). Possibly in C. venezuelae, growth is not always
accompanied by a transformation in pigment and body
depth, and adults retain more of the juvenile features.
More investigation is needed to determine the relation-
ships in C. venezuelae among pigment pattern, body
shape, size, maturity, and local habitat. Cervig6n (1966,
1994) did not illustrate any of his type specimens of C.
venezuelae, but we obtained digital photographs of two
of his paratypes (MOBR-P-0867 [Museo Oceanoldégico
Hermano Benigno Roman, Venezuela]; one is shown in
Figure 6). The holotype is not in good condition (J. C.
Capelo, MOBR, personal communication, 4 July 2008).
The pigment of the paratypes most closely resembles that
in Figure 5D herein: a triangular to rounded mark above
the opercle, a roughly circular dark spot on the ventral
pectoral-fin base, and two basicaudal spots joined by
a light dusky bar. There is some evidence of X-shaped
markings on the side of the body, but the body pigment is
mostly faded. Cervigon (1966, 1994) did not mention X-
shaped markings in his descriptions; rather, he noted that
there are three longitudinal rows of dark spots.

122

e =SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 5. Coryphopterus venezuelae: A, Curacao, 29 mm SL, DNA 8260, USNM 394740; B, Venezuela, 54.4 mm SL,
no DNA, Photo No. 1907 VT-05-530, photo by J. V. Tassell and D. R. Robertson; C, Belize, 35 mm SL, DNA 7248,
USNM 394736; D, Venezuela, 50 mm SL, DNA JV15, AMNH 247345, alcohol preserved; E, Panama, 42.5 mm SL,
DNA 7725-1, AMNH 247341, alcohol preserved.

Synonymy of Coryphopfterus bol

Victor (2008) described Coryphopterus bol as a spe-
cies that heretofore had been masquerading under C. tor-
tugae (e.g., Garzon-Ferreira and Acero 1990:107, fig. 1A,
Santa Marta specimens). We believe that Victor (2008)
was correct in recognizing that the Santa Marta specimens
are not C. tortugae, but our investigation indicates that
they are C. venezuelae. The COI sequence that Victor
(2008) provided for the new species (from the holotype
from Puerto Rico) places it solidly with our C. venezuelae
clade (PR SIO 0869, fig. 1). The average genetic distance
between C. bol and individuals of C. venezuelae is 0.38%
(range, 0.00%-0.85%) and, for comparison, the aver-
age genetic distance between the holotype of C. bol and
the next most closely related clade (C. tortugae) is more
than 20-fold greater, or 8.47% (range, 8.10%—9.21%).
Diagnostic features of Victor’s (2008:4) C. bol include
10 second dorsal- and anal-fin elements; 19 pectoral-fin
rays; pelvic fins fully joined and with a distinct frenum; a
prominent, dark, upward-pointed triangular marking on
the stripe behind the eye; a discrete blotch of small mela-
nophores on the lower third of the pectoral fin base; and a
basicaudal marking that resembles a thick “C.” The com-
bination of the triangular marking on the stripe behind
the eye above the opercle, the pigment blotch on the lower
portion of the pectoral-fin base, and 10 second dorsal- and
anal-fin elements matches most of our C. venezuelae speci-
mens. Victor (2008) distinguished his new species from C.
venezuelae based on the presence of 11 second dorsal- and
anal-fin elements in C. venezuelae, but, as noted above
(also see Table 2), specimens matching Cervig6n’s C. ven-
ezuelae based on the pre-pectoral pigment may have 10 or
11 second dorsal- and anal-fin elements.

Coryphopterus bol also matches C. venezuelae in
number of pectoral-fin rays (19 in C. bol, 61% of speci-
mens with 19 in Cervigén’s [1994] C. venezuelae ma-
terial), pelvic-fin morphology, and other pigment. For
example, the basicaudal mark in C. venezuelae may be
C-shaped, but it ranges in our material from two sepa-
rate spots to a central bar of pigment (some examples
are shown in Figure 5). The basicaudal pigment is also
somewhat variable in the type material of C. bol (Victor,
2008:fig. 1). Two of the type specimens of C. bol most
closely resemble the pale form of C. venezuelae; that is,
the form with round spots on a relatively slender body
(holotype and a 32.1-mm SL paratype). Two paratypes
(24.5 and 29 mm SL) are darker and have at least some
X-shaped markings. None of Victor’s type material is
larger than 32 mm SL, and, as noted under C. venezue-

NUMBER 38 ¢ 123

TABLE 3. Frequency distributions of the combinations of second
dorsal-fin and anal-fin elements in Coryphopterus venezuelae by
country.

No. of second dorsal-fin elements / anal-fin elements

Country 10/9 10/10 10/11 11/10 11/11
Belize — 2 — — —
Curacao 1 11 1 1 —
Panama — 6 1 — —
Venezuela — D} — 1 9
Puerto Rico — 14 = — =

4 Holotype of Coryphopterus bol.

lae, above, most of our dark, deeper-bodied specimens of
C. venezuelae are >40 mm SL.

In summary, one cannot distinguish C. bol and C. ven-
ezuelae on the basis of numbers of second dorsal- and
anal-fin elements because there is more variation in those
counts than previously reported. One might argue that
specimens from Venezuela that have 11 elements in both
the second dorsal and anal fins and heavy pigment with
X-shape markings are C. venezuelae and that everything
else in our C. venezuelae clade is C. bol. However, some
specimens with those features, except with 10 elements in
the second dorsal and anal fins, were taken in the same
station off Venezuela as those with 11 elements (AMNH
247345 [American Museum of Natural History]), so
would one identify the former as C. venezuelae or C. bol?
Species identification of specimens with 11/10 or 10/11
second dorsal-/anal-fin elements also would be nebulous,
as would species identification of dark forms with 10/10
but otherwise virtually identical to those with 11/11 (e.g.,
Figure 5D,E). Variation in COI among all specimens in the
C. venezuelae clade is well within typical intraspecific lev-
els for the genus. However, even if COI is masking recent
divergence within the clade, there is a diagnostic morpho-
logical feature for the clade: a conspicuous spot or blotch
of pigment on the lower pectoral-fin base; in combination
with a triangular or circular blotch of pigment behind the
eye above the opercle, this character is unique to C. ven-
ezuelae. The more common presence of 11 second dor-
sal- and anal-fin elements in some Venezuelan specimens
may best be interpreted as regional variation. Known
currently from Belize, Panama, Curacao, Venezuela, the
Bahamas, the U.S. Virgin Islands, Puerto Rico, Saba, and
Brazil, C. venezuelae appears to be a widespread species.
It is misidentified in the USNM (U.S. National Museum;

124 e¢ SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 6. Paratype of Coryphopterus venezuelae, MOBR-P-0867, 42 mm SL (length estimated from
ruler included with original photograph; this is likely Cervigon’s 41.2 mm SL paratype).

i.e., National Museum of Natural History, Smithsonian
Institution)—and likely other museum collections—as
C. glaucofraenum or C. tortugae.

Key Notes for C. tortugae, C. glaucofraenum,
and C. venezuelae

Juveniles (Figure 7), and occasionally adults, of C. tor-
tugae, C. glaucofraenum, and C. venezuelae may lack the
black marking or triangle above the opercle, or it is not as
dark as other pigment in the stripe posterior to the eye. As
we have used this feature in the “Revised Key to Western
Atlantic Coryphopterus” (see below) to separate C. tor-
tugae, C. glaucofraenum, and C. venezuelae from other
species, absence of this feature in specimens of any of those
species could present identification problems. If there are
well-defined X’s of pigment along the sides of the body (C.
glaucofraenum and some C. venezuelae) or the basicaudal
pigment comprises two spots or a dumbbell-shaped mark-
ing (most C. glaucofraenum and some C. venezuelae), users
of the key should follow the option in the couplet that indi-
cates the dark marking is present above the opercle (4b). If
a specimen lacks the dark pigment spot above the opercle,
has 11 second dorsal- and anal-fin rays, and has a promi-
nent dark blotch on the lower portion of the pectoral-fin
base, it can only be C. venezuelae. Coryphopterus puncti-
pectophorus is similar in lacking the pigment spot above
the opercle and having 11 second dorsal-fin elements, but
it has 10 anal-fin elements (Springer, 1960). Furthermore,
geography will currently separate those two species: C.
venezuelae occurs in the Caribbean, and C. punctipectoph-
orus is known only from the Gulf of Mexico and off the
southeastern USA.

Coryphopterus dicrus Bohlke and Robins, 1960
FIGURE 8

Numerous features, in combination, separate C. dicrus
from other western Atlantic Coryphopterus, including the
following: no black ring of pigment around anus; no dis-
tinct dark spot behind eye above opercle; anal-fin elements
10; pelvic frenum absent; pectoral-fin base with two promi-
nent dark spots of equal intensity, one above the other; and
sides of body freckled with scattered large and smaller pig-
ment blotches. The last two characters are the quickest way
to make the identification. The only other species that usu-
ally have pigment dorsally and ventrally on the pectoral-fin
base are C. venezuelae and C. thrix, but the dorsal pigment
on the pectoral-fin base in C. venezuelae, when present, is a

NUMBER 38 e¢ 125

slash versus a spot, and the dorsal pigment on the pectoral-
fin base in C. thrix is usually much more pronounced than
the ventral marking. Additionally, both species have a pelvic
frenum, which is lacking in C. dicrus.

Coryphopterus thrix Bohlke and Robins, 1960
FIGURE 8

Coryphopterus thrix is the only western Atlantic spe-
cies of Coryphopterus that lacks black pigment around the
anus and has the second dorsal-fin spine elongated into a
filament. If the spine is broken, however, the species is still
identifiable by the combination of features presented in
the key, most notably the absence of a distinctive pigment
blotch above the opercle, presence of a conspicuous dark
blotch on the dorsal portion of the pectoral-fin base, and
presence of a pelvic frenum.

Coryphopterus alloides Bohlke and Robins,
1960

FIGURE 9

Distinguishing features of C. alloides include having a
low anal-fin count (8-9 total elements), a dark blotch of pig-
ment on the lower portion of the membrane between the
second and third dorsal spines, and the pelvic fins almost
completely separate. Only C. kuna, among the Coryphop-
terus species lacking a black ring of pigment around the
anus, has as few as 9 anal-fin elements, but that species has 9
second dorsal-fin elements and 15 pectoral rays (vs. usually
10 and 16-17, respectively, in C. alloides). Coryphopterus
kuna may have a stripe and distal flag of pigment on the first
dorsal fin, but it never has the pigment blotch on the lower
portion of the first dorsal fin characteristic of C. alloides.
The living color pattern of C. alloides is also distinctive: the
head and anterior portion of the body bear a considerable
amount of orange pigment, whereas the posterior portion
of the body is yellow. An apparently cryptic species related
to but genetically distinct from C. alloides and known only
from Curacao is currently under investigation.

Key Note

In some preserved specimens of C. alloides, there are
melanophores above the opercle that may lead the user
of the key to select “4b. Distinct black blotch or triangle
behind eye above opercle . . .” However, this pigment is
never as consolidated and prominent in C. alloides as in

126 © SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

A . :

FIGURE 7. Coryphopterus juveniles: A, C. tortugae, Belize, 20 mm SL, DNA 7693, USNM 394800; B, C. glaucofraenum,
Belize, 17 mm SL, DNA 7769, USNM 394793; C, Coryphopterus venezuelae, Belize, 17 mm SL, DNA 7728, USNM 394881,
D, Coryphopterus thrix, Curacao, 16 mm SL, DNA 8261, USNM 394760; E, Coryphopterus dicrus, Belize, 13 mm SL, DNA
6110, USNM 394779. F, Coryphopterus eidolon, Belize, 18 mm SL, DNA 6223, USNM 394788.

NUMBER 38 ¢ 127

FIGURE 8. Coryphopterus dicrus: A, Florida, 38 mm SL, DNA 7348, USNM 394345; B, Curacao, 30 mm SL, DNA 8135,
USNM 394747; C, Belize, 13 mm SL, DNA 6110, USNM 394779. Coryphopterus thrix: D, Belize, 23.5 mm SL, DNA 7816,
USNM 394914; E, Curacao, 23 mm SL, DNA 8426, USNM 394761; F, Venezuela, AMNH 244983, 26 mm SL, alcohol pre-
served, no DNA.

128 © SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 9. Coryphopterus alloides: A, Belize, 24 mm SL, DNA 7233, USNM 394754; B, Belize, 19 mm SL, DNA 7264, USNM
394755; C, Belize, 24 mm SL, preserved, DNA 7233, USNM 394754. C. eidolon: D, Curacao, 38 mm SL, DNA 8050, USNM
394885; E, Belize, 34 mm SL, DNA 7109, USNM 394752; F, Belize, 33 mm SL, preserved, DNA 5070, USNM 394750.

C. tortugae, C. glaucofraenum, and C. venezuelae; fur-
thermore, C. alloides lacks a pelvic frenum, a conspicuous
feature in the other three species.

Coryphopterus eidolon Béhlke and Robins,
1960

FIGURE 9

Pigment, except for basicaudal and scattered small
body melanophores, is yellow, which disappears during
preservation, typically rendering this a very pale goby. In
life there is a yellow stripe behind the eye bordered by
small melanophores that remain in preserved specimens af-
ter the color fades. There are no dark markings above the
opercle, on the pectoral-fin base, or on the first dorsal fin.
The absence of distinctive markings (other than the basi-
caudal mark) is the easiest way to recognize C. eidolon, a
very abundant species in many of our samples, particularly
from Belize and Curacao.

Coryphopterus kuna Victor, 2007
FIGURE 10

Baldwin and Smith (2003) described Coryphopterus B
larvae from Belize as likely belonging to an unidentified
species based on the low second dorsal- and anal-fin counts
(9 in both fins) and low pectoral-fin count (15). Victor
(2007) described C. kuna, which has the low fin-ray counts
of the Coryphopterus B larvae, as a new species from off
Panama. Incorporation of the COI sequence published in
the original description of C. kuna into our analysis re-
vealed that Coryphopterus B larvae are C. kuna. This spe-
cies is distinctive in typically having 9 second dorsal- and
anal-fin elements, as well as a low pectoral-ray count of 15
(found elsewhere only in C. personatus and C. hyalinus).
Apparently a small fish—the adult male holotype is 17.1
mm SL—C. kuna has little dark pigment: numerous small

NUMBER 38 ¢ 129

spots on the pelvic fin of the holotype, a few scattered small
spots on the sides of the body, no markings on the pectoral-
fin base, and no basicaudal spot. It lacks a pelvic frenum.

Coryphopterus punctipectophorus Springer,
1960

FIGURE 10

Coryphopterus punctipectophorus is similar to C. tor-
tugae, C. glaucofraenum, and C. venezuelae in having
three rows of pigment spots along the side of the body, but
it differs from those species in lacking a dark blotch or tri-
angle behind the head above the opercle. It is most similar
to C. venezuelae in having a prominent dark spot on the
lower portion of the pectoral-fin base, and juvenile (and
occasionally adult) specimens of C. venezuelae that lack
the pigment blotch above the opercle will typically key to
C. punctipectophorus based on the ventral pigment spot
on the pectoral-fin base. Like C. venezuelae, C. puncti-
pectophorus was originally described as having 11 second
dorsal-fin elements, but as noted above (see C. venezue-
lae), the former has 10 or 11 second dorsal elements. The
“dusky light buff” pigment spots along the dorsal contour
and “coral pink” spots along the sides of the body in fresh
material of C. punctipectophorus (Springer, 1960:240; see
Figure 10B,E herein) apparently fade in preserved material
(see Figure 10D). The known distribution of C. puncti-
pectophorus includes both coasts of Florida, the Gulf of
Mexico (including the southern Gulf where it meets the Ca-
ribbean), and South Carolina. It apparently inhabits deeper
water than some Coryphopterus species: the type material
was collected at 62 and 120 feet. It has not been reported
from the Caribbean. We have not collected C. punctipec-
tophorus, and fresh material of the species was not avail-
able for inclusion in our genetic analysis. Thacker and
Cole’s (2002) C. punctipectophorus from Belize (GenBank
Accession No. AF391396) is C. dicrus, based on incorpora-
tion of their COI sequence into our data set.

REVISED KEY TO THE WESTERN ATLANTIC SPECIES OF CORYPHOPTERUS

la. Black ring of pigment surrounding anus..............
ibaa blackeninpzaroundianusabsentem rer terrier ccna
DA © ileniniterorpitaluporeyanteLiOclyaware lari ile:

As, INGO ANGLO All OIES AINA 6 ous ogo oe Cod dou eb oO UAC UM OOO Dae Meee MaeDeeo OD OOo Coryphopterus hyalinus

3a. Second dorsal and anal fins each typically with 11 total elements; head with some orange pigment in life; body trans-

lucent, with several squares or rectangles of pale orange pigment internally; preserved specimens lacking conspicuous
postorbital stripes of melanophores but with “mask” of pigment around eye ............ Coryphopterus personatus

(continued on p. 130)

130

3b.

Aa.

4b.

Sa.
Sb.
6a.

6b.

7a.

7b.

8a.

8b.

9a.

9b.

10a.

10b.

11a.

11b.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

REVISED KEY TO THE WESTERN ATLANTIC SPECIES OF CORYPHOPTERUS (continued)

Second dorsal and anal fins typically with 10 total elements; head and body predominantly yellow in life; a dusky inter-
nal stripe along posterior section of vertebral column; preserved specimens with postorbital stripes of melanophores and
scatteredispotsiovenentire) bodyareeeeee een oer one-one nee Gee ae eae Coryphopterus lipernes
No distinct black blotch behind eye above opercle in adults; pigment above opercle, if present, no larger or darker than
other markings behind eye; pelvic frenum present or absent (see “Key Note” for C. alloides in text)............... 5
Distinct black blotch or triangle behind eye above opercle in adults, blotch usually larger and darker than other pig-
ment in stripe behind eye; pelvic frenum present (see “Key Notes for C. tortugae, C. glaucofraenum, and C. venezue-

NC Xie hil (<>. <6) ere nee oer ae er tea er ar een RO RMS nord ney eA tad Store oO Ob bo pulp do o00000000 10
Anal-fin elements 8-9 (usually 9), pectoral-fin rays 15-17, pelvic frenum absent..............0.0 cece cece ceees 6
Anal-fin elements 10-11, pectoral-fin rays 17-20, pelvic frenum present or absent ...............-2-2eeeeeeeee 7

Second dorsal and anal fins each with 9 elements; pectoral-fin rays 15; pelvic fins fully joined; first dorsal fin with stripe
of black pigment; in life, head and body with orange spots and blotches and sometimes with flag of dark pigment on
Ist=3rdidorsal spines:s:..cerss ce RRS ue Doo oOEE OD nana eee Coryphopterus kuna
Second dorsal fin with 10 elements, anal fin with 9 (rarely 8); pectoral-fin rays 16-17; pelvic fins almost completely sepa-
rate; black blotch or bar between 2nd and 3rd dorsal spines; head and anterior body mottled orange in freshly caught
Specimens». postenon bodysmottlediyellowaemenen cme ate eee eee te ein eet Coryphopterus alloides
Pectoral-fin base with two prominent dark spots of equal intensity, one dorsally and one ventrally; upper spot usu-
ally with swath of melanophores extending posteriorly onto pectoral-fin rays; sides of body freckled with scattered
large and smaller blotches of melanophores (blotches associated with coral, tan, yellow pigment in life); pelvic
frénum: absent: .!eesti¢%.a75- Ae | ld tke oe ee ee eee Coryphopterus dicrus
Pectoral-fin base not with two prominent dark spots (or, if two spots present, upper spot more intense); sides of body
with few dark markings (with few to many yellow spots in life) or with three rows of light markings (coral pink/orange
in life); pelvic frenum present’. ce) Aut Aare PAO 0.5 es ae Gas bee occ Oe ene Oreo eee eee 8
Pectoral-fin base without prominent dark markings but sometimes with a few to many scattered melanophores; sides of
body with few if any dark markings (with yellow spots in life) except for basicaudal spot...... Coryphopterus eidolon
Pectoral-fin base with prominent markings; sides of body with or without numerous dark markings .............. 2)
Pectoral-fin base with distinct pigment spot dorsally, spot usually dark above, diffuse below, often with dots trailing
ventrally; ventral dots coalescing into a separate spot in some specimens (ventral spot, if present, less intense than dorsal
spot); second dorsal-fin elements 9-10; second dorsal spine filamentous ....................-- Coryphopterus thrix
Pectoral-fin base with prominent dark spot or blotch only on ventral portion; second dorsal-fin elements 11; second
dorsalispineinot filamentous4% 2 jocce eaaeeee eeeeeeee Coryphopterus punctipectophorus
Body usually pale, pigment primarily comprising three rows of markings on side of body; lower row comprising small,
mostly vertically elongate markings, some of which may be crescent shaped or some part of an X-shape but rarely well-
defined X’s; if X-shaped markings present, their height is considerably shorter than eye diameter; pigment marking above
opercle usually a triangle, and basicaudal pigment usually a central bar .................. Coryphopterus tortugae
Body heavily pigmented or pale but without vertically elongate or crescent-shaped markings in ventral row of pigment
on side of body; height of X-shaped markings, if present, three-quarters of or equal to diameter of eye; pigment marking
above opercle triangular, rounded, or with two peaks; basicaudal pigment comprising two separate spots, two spots con-
nected by a line of pigment and resembling a dumbbell, a central bar, or a C-shaped marking................... 11
Pigment on pectoral-fin base variable but always with dark spot or rectangular-shaped blotch ventrally (may be
associated with bright yellow pigment in life); one or two additional bars, blotches, or concentrations of pigment
sometimes present dorsally; three rows of dark markings on side of body, some in lower row large, X-shaped mark-
ings in heavily pigmented specimens, small, circular blotches in paler specimens; pigment marking above the opercle
trian gularorcoumd ee ea irs 5, asec ole toee Se atotenits Gaertn a Rene cn ee Coryphopterus venezuelae
Pectoral-fin base rarely with prominent dark marking ventrally, although melanophores may form one to three light to
moderate concentrations on base; body with three rows of dark markings, most of those in the lower row large, distinc-
tive X-shaped markings; pigment marking above opercle usually with two well-defined peaks ................-.-.
eS CS rw hae i si A Ieee ce co re REN een nd. eee ee Coryphopterus glaucofraenum

NUMBER 38 e¢ 131

FIGURE 10. A, Coryphopterus kuna, San Andres, Colombian Caribbean (photo by Keri Wilk, ReefNet Inc.); B, Co-
ryphopterus punctipectophorus, Holbox Island, Mexico (photo by Hilario Itriago); C, Coryphopterus kuna, Panama,
17.1 mm SL, holotype, SIO-07-5, preserved, DNA GB EF55021 (reproduced from B. Victor, 2007, fig. 1A, Zootaxa
1526:53); D, Coryphopterus punctipectophorus, South Carolina, 28 mm SL, USNM 315530, preserved, no DNA; E,
Coryphopterus punctipectophorus, Florida, Gulf of Mexico, 42 mm SL, holotype, ANSP 90103, preserved, no DNA.

132 e

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 11. Coryphopterus glaucofraenum, neotype, USNM 393907, Belize, 44 mm SL, DNA 6367: A,
fresh; B, preserved.

Designation of Neotype for
Coryphopterus glaucofraenum

FIGURE 11

Eschmeyer (2008) noted the need for designating a
neotype for Coryphopterus glaucofraenum Gill, because
the whereabouts of the holotype are unknown. He also
noted that four MCZ specimens assumed to be syntypes do
not constitute type material because Gill’s (1863) descrip-
tion was clearly based on a single specimen. Because of the
historical confusion regarding the validity of C. tortugae
and C. venezuelae as distinct from C. glaucofraenum, and
because the three species can be difficult to separate, we
have elected to designate a neotype for C. glaucofraenum
from which we have successfully obtained a COI sequence
that places the specimen in the C. glaucofraenum clade.
We hereby make the following type designation:

Neotype

Coryphopterus glaucofraenum Gill, USNM 393907,
44 mm SL, DNA 6367, Twin Cays, Belize, mangrove
edge on interior channel, 0-6 ft. (GenBank accession no.
GQ367355.)

SUMMARY AND FUTURE WORK

Cytochrome c oxidase I sequences (DNA barcoding)
were useful in determining the number of distinct genetic
lineages within Caribbean Coryphopterus. We used the
neighbor-joining tree (see Figure 1) derived from those se-
quences to assemble voucher specimens (and color photo-
graphs of them taken before preservation) into clades and
then compared the morphology of specimens among those
clades. Assigning clades to species was relatively easy based

on review of original literature and examination of some
type specimens (or photographs of them). Resolving the
identities of many Caribbean Coryphopterus in the absence
of the DNA data would have been extremely difficult.

We are continuing to expand our geographic coverage
of Coryphopterus sampling and will continue sequencing
COI, and ultimately other genes, from specimens from a
diversity of locations. The precise geographic distributions
of most western Atlantic Coryphopterus are not known,
and our genetic analyses have revealed the presence of one
or more additional cryptic species. Additionally, the exis-
tence of two morphological forms within the genetic clade
identified as C. venezuelae warrants further investigation.
Ultimately, our multi-locus data set will enable us to re-
analyze phylogenetic relationships among Coryphopterus
species, from which we can investigate patterns of specia-
tion and morphological divergence. Finally, testing of the
species identifications of Coryphopterus larvae proposed
by Baldwin and Smith (2003) based on morphology is cur-
rently in progress based on COI sequences of larvae col-
lected as part of this study.

ACKNOWLEDGMENTS

Cody Payne contributed to the organization of our Beliz-
ean Coryphopterus material, made radiographs and counts
of numerous specimens, helped distinguish C. tortugae from
C. glaucofraenum, and provided data helpful in develop-
ing the revised species key. James Van Tassell contributed
numerous specimens, tissue samples, and photographs of
Venezuelan and Panamanian Coryphopterus and engaged
in many helpful discussions about Coryphopterus with the
first author. Amy Driskell and Andrea Ormos provided lab-
oratory and logistical assistance. Jon Fong provided images
of the holotype of Ctenogobius tortugae. Victor Springer,
Lisa Palmer, and Hilario Itriago provided images of Cory-
phopterus punctipectophorus. Benjamin Victor provided
the photograph of the holotype of C. kuna, and Zhi-Qiang
Zhang, Chief Editor of Zootaxa, allowed us to reproduce
this image. Keri Wilk provided the in situ image of C. kuna.
Annemarie Kramer allowed us to include her sequences
of C. tortugae from Curacao in our analysis. Oscar M.
Lasso-Alcala, Juan C. Capelo, and Ramon Varela provided
digital images of two paratypes of C. venezuelae. Michael
Carpenter, Zachary Foltz, Amy Driskell, and Justin Bagley
provided field assistance in Belize and Florida. Research in
Florida was conducted pursuant to Special Activity License
07SR-1024. Amos Gazit, Kate Wilson, and Maureen Kunen
made it possible for us to collect fish samples through the

NUMBER 38 e¢ 133

CARMABI laboratory in Curacao. Fieldwork in the Baha-
mas was conducted under the auspices of the Perry Institute
of Marine Science, with logistical assistance from Brenda
Gadd. Fourteen members of the first author’s family contrib-
uted to the fieldwork in the Bahamas, and a portion of that
work was funded by a generous donation from Christine B.
Lang in memory of David E. Baldwin and Richard A. Lang.
The Smithsonian Marine Science Network provided most of
the funding for fieldwork, and the Smithsonian DNA Bar-
coding Initiative provided funding for molecular analyses.
This is contribution number 837 of the Caribbean Coral
Reef Ecosystems Program (CCRE), Smithsonian Institution,
supported in part by the Hunterdon Oceanographic Re-
search Fund, and Smithsonian Marine Station at Fort Pierce
(SMSFP) Contribution No. 756.

LITERATURE CITED

Baldwin, C. C., and D. G. Smith. 2003. Larval Gobiidae (Teleostei: Per-
ciformes) of Carrie Bow Cay, Belize, Central America. Bulletin of
Marine Science, 72:639-674.

Bohlke, J. E., and C. G. Chaplin. 1968. Fishes of the Bahamas and Ad-
jacent Tropical Waters. Wynnewood, Pa.: Livingston Publishing
Company.

Bohlke, J. E., and C. R. Robins. 1960. A Revision of the Gobioid Fish
Genus Coryphopterus. Proceedings of the Academy of Natural Sci-
ences of Philadelphia, 112(5):103-128.

. 1962. The Taxonomic Position of the West Atlantic Goby,
Eviota personata, with Descriptions of Two New Related Species.
Proceedings of the Academy of Natural Sciences of Philadelphia,
114(5):175-189.

Cervigon, F. 1966. Los Peces Marinos de Venezuela. Volume II. Caracas:
Estacion de Investigaciones Marinas de Margarita Fundacion La
Salle de Ciencias Naturales.

. 1994. Los Peces Marinos de Venezuela. Volume III. Caracas:
Fundacién Cientifica Los Roques.

Eschmeyer, W.N., ed. 2008. Catalog of Fishes: Electronic Version (updated
23 April 2008), http://www.calacademy.org/research/ichthyology/
catalog/fishcatsearch.html (accessed January 2008).

Garz6n-Ferreira, J., and A. P. Acero. 1990. Redescription of Coryphop-
terus tortugae (Jordan) (Osteichthyes: Gobiidae), a Valid Spe-
cies of Goby from the Western Atlantic. Northeast Gulf Science,
11(2):105-112.

Gill, T. N. 1863. Descriptions of the Gobioid Genera of the Western
Coast of Temperate North America. Proceedings of the Academy of
Natural Sciences of Philadelphia, 15:262-267.

Greenfield, D. W., and R. K. Johnson. 1999. Assemblage Structure and
Habitat Associations of Western Caribbean Gobies (Teleostei: Go-
biidae). Copeia, 1999(2):251-266.

Jordan, D. S. 1904. Notes on the Fishes Collected in the Tortugas
Archipelago. Bulletin of the U.S. Fish Commission, 22(1902):
539-544.

Jordan, D. S., and J. C. Thompson. 1905. The Fish Fauna of the Tortugas
Archipelago. Bulletin of the Bureau of Fisheries, 24:229-256.
Kimura, M. 1980. A Simple Method for Estimating Evolutionary Rates
of Base Substitutions Through Comparative Studies of Nucleotide

Sequences. Journal of Molecular Evolution, 16:111-120.

134 e

Longley, W. H., and F. Hildebrand. 1941. Systematic Catalogue of the
Fishes of Tortugas, Florida: With Observations on Color, Habits,
and Local Distribution. Washington, D.C.: Carnegie Institution of
Washington.

Murdy, E. O. 2002. “Gobiidae.” In The Living Marine Resources of the
Western Central Atlantic. Volume 3: Bony Fishes, Part 2 (Opisto-
gnathidae to Molidae), Sea Turtles and Marine Mammals, ed. K. E.
Carpenter, pp. 1375-2127. Rome: The Food and Agriculture Orga-
nization of the United Nations.

Saitou, N., and M. Nei. 1987. The Neighbor-Joining Method: A New
Method for Reconstructing Phylogenetic Trees. Molecular Biology
and Evolution, 4:406-425.

Seutin, G., B. N. White, and P. T. Boag. 1990. Preservation of Avian
Blood and Tissue Samples for DNA Analysis. Canadian Journal of
Zoology, 69:82-90.

Smith, C. L., J. C. Tyler, W. P. Davis, R. S. Jones, D. G. Smith, and C. C.
Baldwin. 2003. Fishes of the Pelican Cays, Belize. Atoll Research

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Bulletin, No. 497. Washington, D.C.: National Museum of Natural
History, Smithsonian Institution.

Springer, V. G. 1960. A New Gobiid Fish from the Eastern Gulf of Mexico.
Bulletin of Marine Science of the Gulf and Caribbean, 10(2):237-240.

Swofford, D. L. 2002. PAUP*: Phylogenetic Analysis Using Parsimony
(and Other Methods). Version 4, Beta 10 (CD-Rom). Sunderland,
Mass.: Sinauer Associates.

Thacker, C. E., and K. S. Cole. 2002. Phylogeny and Evolution of the Gobiid
Genus Coryphopterus. Bulletin of Marine Science, 70(3):837—850.

Victor, B. C. 2007. Coryphopterus kuna, a New Goby (Perciformes: Go-
biidae: Gobiinae) from the Western Caribbean, with the Identifica-
tion of the Late Larval Stage and an Estimate of the Pelagic Larval
Duration. Zootaxa, 1526:51-61.

. 2008. Redescription of Coryphopterus tortugae (Jordan) and a

New Allied Species Coryphopterus bol (Perciformes: Gobiidae: Go-

biinae) from the Tropical Western Atlantic Ocean. Journal of the

Ocean Science Foundation, 1:1-19.

APPENDIX

TABLE A.1. Coryphopterus material. A number in the DNA column indicates that the specimen was analyzed for cytochrome c oxidase
1. An asterisk beside this number indicates the entry appears in the neighbor-joining tree in Figure 1; because of space constraints, not
all specimens for which DNA was successfully sequenced are included in Figure 1. Extracting DNA was not attempted on formalin-fixed
specimens. If the specimen was not sampled for DNA, “no DNA” is recorded in this column; BZE, Belize; FLA, Florida; CUR, Curacao;

BAH, Bahamas; PAN, Panama; VEN, Venezuela.

Species DNA Standard Specimen Photo voucher
length (mm) voucher at NNUNH
C. lipernes BZE 4067* — No voucher No
BZE 4082* 23 No voucher No
BZE 4083* 21 No voucher No
BZE 7729* 18 USNM 394796 Yes
CUR 8051* Dil USNM 394895 Yes
CUR 8326* 20 USNM 394896 Yes
CUR 8327* 17 USNM 394894 Yes
C. hyalinus BZE 4511* 15 No voucher No
BZE 4512* 15 No voucher No
BZE 5066* 13 No voucher Yes
BZE 6221* 13.5 USNM 394795 Yes
BZE 6222* 14.5 USNM 394794 Yes
BZE 7760* 7 No voucher Yes
CUR 8044* 20 USNM 394890 Yes
CUR 8046* 19.5 USNM 394891 Yes
CUR 8264* 19 USNM 394893 Yes
CUR 8265* 17 USNM 394889 Yes
CUR 8266* 16.5 USNM 394892 Yes
C. personatus BZE 4014* — No voucher No
BZE 4079* 19) No voucher Yes
BZE 4307* 24 USNM 394756 Yes
BZE 4308* 21 USNM 394757 Yes
BZE 4309* 18 USNM 394758 Yes
BZE 5067* 19 USNM 394913 Yes
BZE 7163* 15 USNM 394742 Yes
CUR 8045* 19.5 USNM 394897 Yes

NUMBER 38

e 135

Species DNA Standard Specimen Photo voucher
length (mm) voucher # at NUVNH
BAH 8263 23 USNM 394904 Yes
BAH 8264* 22 USNM 394905 Yes
PAN 7712-1* 22 AMNH 247346 No
PAN 7712-5* DD) AMNH 247346 No

C. tortugae BZE 4016* 28 No voucher Yes
BZE 4530* 40 USNM 394730 Yes
BZE 5237* 34 USNM 394743 Yes
BZE 5238* 30 USNM 394731 Yes
BZE 7106* 20 USNM 394732 Yes
BZE 7107* 36 USNM 394733 Yes
BZE 7333* DS) USNM 394744 Yes
BZE 7677* 31 USNM 394801 Yes
BZE 7690 37/ USNM 394878 Yes
BZE 7691* 29 USNM 394802 Yes
BZE 7692* 36 USNM 394879 Yes
BZE 7693* 20 USNM 394800 Yes
BZE 7708* 33 USNM 394877 Yes
BZE 7709* 29 USNM 394798 Yes
BZE 7734* 26 USNM 394799 Yes
BZE (no DNA) 40 USNM 329834 No
BZE (no DNA) 33 USNM 334838 No
CUR CG25* — No voucher No
CUR CG26* — No voucher No
PAN 7725-6* 36 AMNH 247347 No
VEN (no DNA) 45 USNM 194103 No
VEN 7736-1* 33 AMNH 247340 No
VEN 7736-4* 37 AMNH 247340 No
VEN 7736-6* 46 AMNH 247340 No
Bermuda (no DNA) 9 (15-31) USNM 330023 No
FLA (no DNA, photo = SU 08363 No

of holotype)

C. glaucofraenum BZE 6037* 35 USNM 394347 Yes
BZE 6367* 44 USNM 393907 Yes
BZE 7343* 6 No voucher Yes
BZE 7351* 35 USNM 394353 Yes
BZE 7352* QS: USNM 394354 Yes
BZE 7353* 17.5 USNM 394355 Yes
BZE 7733* 25 USNM 394748 Yes
BZE 7768* 22 USNM 394792 Yes
BZE 7769* 17 USNM 394793 Yes
BZE 7796* 8.5 No voucher Yes
BZE 7798* 8.5 No voucher Yes
FLA 7341 49 USNM 394348 Yes
FLA 7342 42 USNM 394349 Yes
FLA 7343* 35 USNM 394350 Yes
FLA 7344 36 USNM 394351 Yes
FLA 7345 30 USNM 394352 Yes
FLA 7674 49 USNM 394356 Yes
FLA 7675 44 USNM 394357 Yes
FLA 7676 38 USNM 394358 Yes
FLA 7677 32 USNM 394729 Yes
PAN 7701-1* 39 AMNH 247334 No
PAN 7701-2* 40.5 AMNH 247334 No
PAN 7701-3* 32 AMNH 247334 No
PAN 7701-4* 26.5 AMNH 247334 No
PAN 7701-5* 333) AMNH 247334 No
PAN 7712-2* 35 AMNH 247335 No

continued

136 e¢ SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

TABLE A.1. continued

Species DNA Standard Specimen Photo voucher
length (mm) voucher # at NUNH
VEN 7729-1* 31 AMNH 247336 No
VEN 7729-2* 30 AMNH 247336 No
VEN 7729-3* 31 AMNH 247336 No
VEN 7736-2* 37.5 AMNH 247337 No
VEN 7738-1* 38 AMNH 247338 No
VEN 7738-2* 36 AMNH 247338 No
VEN 7738-3* 39 AMNH 247338 No
VEN 7744-2* 32 AMNH 247339 No
VEN 7744-3* 27 AMNH 247339 No
VEN 7744-4* 28.5 AMNH 247339 No
Bahamas (no DNA) 31 USNM 386863 No
Bahamas (no DNA) 2 (30-32) USNM 386955 No
Bermuda (no DNA) 4 (27-35) USNM 178908 No
Bermuda (no DNA) 2 (45-46) USNM 178555 No
C. venezuelae BZE 5099* 16 USNM 394735 Yes

BZE 5319* 8.5 No voucher Yes
BZE 7248* 35 USNM 394736 Yes
BZE 7362* ToS) No voucher Yes
BZE 7704* 20 USNM 394880 Yes
BZE 7728* 17 USNM 394881 Yes
BZE 7797* 8.5 No voucher Yes
CUR 8052* 30.5 USNM 394737 Yes
CUR 8053* 30 USNM 394764 Yes
CUR 8054* 26.5 USNM 39475 Yes
CUR 8055 28 USNM 394766 Yes
CUR 8208* 31.5 USNM 394738 Yes
CUR 8259* 29 USNM 394739 Yes
CUR 8260* 29 USNM 394740 Yes
CUR 8427* 35 USNM 394741 Yes
BAH 8048* 43 USNM 394908 Yes
BAH 8049* 42 USNM 394906 Yes
BAH 8262* 39 USNM 394909 Yes
PAN 7725-1* 42.5 AMNH 247341 No
PAN 7725-2* 38 AMNH 247341 No
PAN 7725-3* 33 AMNH 247341 No
PAN 7725-4* 39 AMNH 247341 No
PAN 7725-5* 42.5 AMNH 247341 No
VEN 6670-3* 41 AMNH 247342 No
VEN 6670-4* 45 AMNH 247342 No
VEN 7733-1* 29 AMNH 247343 No
VEN JV07* 20 AMNH 247344 No
VEN JV08* 29.5 AMNH 247344 No
VEN JV09* 36 AMNH 247345 No
VEN JV10* 29 AMNH 247345 No
VEN JV11* 29 AMNH 247345 No
VEN JV12* 52 AMNH 247345 No
VEN JV13* 50 AMNH 247345 No
VEN JV14* 50 AMNH 247345 No
VEN JV15* 50 AMNH 247345 No
VEN JV16* 29 AMNH 247345 No
VEN (no DNA; photo ~42 MOBR-P-0867 No

of paratype)
Puerto Rico; holotype 26.8 SIO 0869 No

of C. bol* (DNA

from Victor, 2008)
Saba (no DNA) 15 USNM 387726 No

NUMBER 38

e 137

Species DNA Standard Specimen Photo voucher
length (mm) voucher # at NMVNH
Brazil 4 (2-39) USNM 357709 No
C. dicrus BZE 4213* 22 USNM 394337 Yes
BZE 5239* Daf USNM 394763 Yes
BZE 6274* 25 USNM 394774 Yes
BZE 6110* 13 USNM 394779 Yes
BZE 7238 29 USNM 294338 Yes
BZE 7266 24 USNM 294339 Yes
BZE 7354* 22 USNM 394745 Yes
BZE 7410 Dy, USNM 394746 Yes
BZE 7700* 19 USNM 394778 Yes
BZE 7701* 7/ USNM 394776 Yes
BZE 7707* 21 USNM 394777 Yes
BZE 7745* 23 USNM 394780 Yes
BZE 7818* 22 USNM 394775 Yes
FLA 7346* 43 USNM 394343 Yes
FLA 7347* 41 USNM 394344 Yes
FLA 7348* 38 USNM 394345 Yes
FLA 7680 39 USNM 394340 Yes
FLA 7681 42 USNM 394341 Yes
FLA 7682 44 USNM 394342 Yes
CUR 8135* 30 USNM 394747 Yes
BAH 8134* 43 USNM 394900 Yes
BAH 8135* 38 USNM 394898 Yes
BAH 8232 36 USNM 394899 Yes
VEN 7736-3* 35 AMNH 247332 No
VEN JV01* 33 AMNH 247333 No
VEN JV02* 35 AMNH 247333 No
VEN JV03* 36 AMNH 247333 No
VEN JV04* 20.5 AMNH 247333 No
VEN JV05* Ail AMNH 247333 No
VEN JV06* 20 AMNH 247333 No
Saba (no DNA) 4 (25-28) USNM 388525 No
Tobago (no DNA) 35 USNM 318808 No
Tobago (no DNA) 3 (23-25) USNM 318818 No
Dominica (no DNA) 11 (13-27) USNM 325165 No
C. thrix BZE 6111* 15 USNM 394797 Yes
BZE 7265* 10 USNM 394734 Yes
BZE 7267* 30 USNM 394759 Yes
BZE 7816* 23 USNM 394914 Yes
BZE 7817* 22 USNM 394915 Yes
BZE (no DNA) 3 (20-28.5) USNM 328240 No
CUR 8261* 16 USNM 394760 Yes
CUR 8426* 23 USNM 394761 Yes
Venezuela (no DNA) 26 AMNH 244983 No
Navassa (no DNA) 31 USNM 359403 No
Tobago (no DNA) 32 USNM 318811 No
Tobago (no DNA) 2 (23-24) USNM 317133 No
C. eidolon BZE 4017* 31 USNM 394749 Yes
BZE 4080* 20 USNM Yes
BZE 4081* 29 No voucher No
BZE 4089* - No voucher No
BZE 5070* 33 USNM 394750 Yes
BZE 5099 16 No voucher Yes
BZE 6223* 18 USNM 394788 Yes
BZE 6224* 24 USNM 394789 Yes
BZE 6246* 25 USNM 394787 Yes
BZE 6268* DBS) USNM 394790 Yes
BZE 6302* 33 USNM 394751 Yes

continued

138 © SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

TABLE A.1. continued

Species DNA Standard Specimen Photo voucher
length (mm) voucher # at NUNH
BZE 7108 21 USNM 394785 Yes
BZE 7109* 34 USNM 394752 Yes
BZE 7152 19 USNM 394346 Yes
BZE 7232* 31 USNM 394762 Yes
BZE 7350* 36 USNM 394753 Yes
BZE 7671* 28 USNM 394786 Yes
BZE 7672 24 USNM 394784 Yes
BZE 7673* 22 USNM 394781 Yes
BZE 7702 31 USNM 394783 Yes
BZE 7703* 26 USNM 394782 Yes
BZE 7726 24 USNM 394912 Yes
BZE 7727 7 USNM 394911 Yes
BZE 7735 23 USNM 394791 Yes
CUR 8047 37 USNM 394886 Yes
CUR 8048* 39 USNM 394884 Yes
CUR 8049 33 USNM 394883 Yes
CUR 8050* 38 USNM 394885 Yes
CUR 8262* 24 USNM 394887 Yes
CUR 8263 33 USNM 394888 Yes
BAH 8046* 41 USNM 394903 Yes
BAH 8047* 37 USNM 394902 Yes
Navassa (no DNA) 3 (32-33) USNM 360458 No
C. alloides BZE 7233* 24 USNM 394754 Yes
BZE 7264* 19 USNM 394755 Yes
BZE 7761* 12 USNM 394910 Yes
BZE (no DNA) 21 USNM 267843 No
CUR 8325* 18 USNM 394882 Yes
C. kuna BZE 4586* 6 No voucher No
BZE 5134* IS No voucher Yes
BZE 6049* 7 No voucher Yes
BZE 6387* VES) No voucher Yes
PAN; holotype* DNA 7A SIO-07-5 No
from GenBank
C. punctipectophorus FLA; paratype 28 USNM 179307 No
(no DNA)
South Carolina 28 USNM 315530 No
(no DNA)

4USNM = U.S. National Museum (National Museum of Natural History), Smithsonian Institution; AMNH = American Museum of Natural History; MOBR = Museo
Oceanol6gico Hermano Benigno Roman, Venezuela; SIO = Scripps Institution of Oceanography.

Recent Insights
Allen G. Collins

Allen G. Collins, National Marine Fisheries Ser-
vice Systematics Laboratory, National Museum
of Natural History, Smithsonian Institution, P.O.
Box 37012, MRC-153, Washington, D.C. 20013-
7012, USA (collinsa@si.edu). Manuscript received
29 August 2008; accepted 20 April 2009.

into Cnidarian Phylogeny

ABSTRACT. With representatives of more than 10,000 species from diverse clades scat-
tered throughout the world’s oceans, Cnidaria is a moderately diverse phylum of Meta-
zoa. As such, various taxa within Cnidaria have been the subjects of recent phylogenetic
analyses. Because of its diversity, it has not yet been possible to conduct any extensive
phylum-level phylogenetic analyses. In addition, new information suggests that the large
group of parasites known as Myxozoa is part of Cnidaria. The present contribution sum-
marizes recent findings to create a picture of a current working hypothesis of cnidarian
phylogeny. This summary, which treats the relationships among taxa down to the ap-
proximate level of order, likely provides a suboptimal estimation of cnidarian phylogeny
as compared to a detailed phylogenetic analysis of data sampled densely from all the
Cnidaria component clades. Nevertheless, it should provide points of comparison for
upcoming efforts to more comprehensively assess cnidarian phylogeny. Even at the basic
level of order, many taxa are thought to be polyphyletic. Understandably, current clas-
sifications are not fully reflective of recent phylogenetic advances.

INTRODUCTION

Early in the history of Metazoa, the nematocyst evolved. This capsular or-
ganelle encloses venom and a tightly coiled, hollow, dart-like thread that is dis-
charged at incredibly rapid accelerations of up to 5 million g (Nuchter et al.,
2006). This explosive discharge can be achieved because of extreme osmotic
pressures (Holstein and Tardent, 1984; Weber, 1989) within the highly stable
nematocyst wall, the molecular structure and function of which are becom-
ing ever clearer (e.g., Meier et al., 2007). Cnida is the more general term for
this organelle, the nematocyst being just one type. However, it is reasonably
clear, based on the distribution across cnidarian taxa, that the ancestral form
of the cnida was as a nematocyst (Marques and Collins, 2004). The lineage in
which the nematocyst originated gave rise to the moderately diverse phylum
Cnidaria, most likely during the Ediacaran period (Peterson and Butterfield,
2005; Cartwright and Collins, 2007). Since this time, cnidarians have evolved
an enormous variety of forms and a great diversity of life history strategies.
Representative cnidarians build reefs, fish the depths, and parasitize other spe-
cies. Extant valid species number a bit more than 11,000 (Daly et al., 2007),

140 e

or more than 13,000 when roughly 2,200 myxozoan spe-
cies (Lom and Dykova, 2006) are included (see follow-
ing), and can be found living in all marine environments.
Many myxozoans infect freshwater taxa, but just a small
number of other cnidarian species live in freshwater
(Jankowski et al., 2008).

Daly et al. (2007), in honor of the 300th anniversary
of the birth of Linnaeus, recently provided a summary
of phylogenetic knowledge about currently recognized
cnidarian taxa (exclusive of Myxozoa) typically ranked
at ordinal and family levels in current classifications. I
aim to provide a summary of recent insights into cnidar-
ian phylogeny focusing on relationships among the dif-
ferent cnidarian orders. No phylum-level analyses of the
evolutionary relationships among these taxa have been
carried out, although attempts have been made to assess
phylogenetic hypotheses for large subclades of Cnidaria;
i.e., Anthozoa (Berntson et al., 1999; Won et al., 2001),
Hexacorallia (Daly et al., 2003; Brugler and France,
2007), Octocorallia (Bertnson et al., 2001; McFadden et
al., 2006); Medusozoa (Collins, 2002; Marques and Col-
lins, 2004; Collins et al., 2006a; Van Iten et al., 2006),
and Myxozoa (Kent et al., 2001; Fiala, 2006). In addi-
tion, several recent studies have assessed the phylogenetic
affinities of taxa that have been problematic (Collins et
al., 2006b; Van Iten et al., 2006; Dykova et al., 2007;
Jiménez-Guri et al., 2007). The approach taken here is
to cobble together results from these various analyses to
provide a reasonable picture of our present understand-
ing of cnidarian relationships (Figure 1). Representative
cnidarians are illustrated in Figures 2 and 3.

As it concentrates on recent insights, the present pa-
per does not provide a thorough review of the history
of ideas about relationships among cnidarian orders, nor
does it attempt to summarize what recent phylogenetic
results tell us about cnidarian character evolution. For
that type of information, one should consult the studies
referenced herein. The working hypothesis of cnidarian
relationships (see Figure 1), as well as the summary pro-
vided by Daly et al. (2007), should provide points of com-
parison for phylum-wide analyses of cnidarian phylogeny,
which will soon be attempted by researchers engaged in
the cnidarian tree of life project (http://CnidToL.com).
Because it is a representation of a hypothetical history
of Cnidaria, every node in Figure 1 is uncertain and is
subject to change in light of new information. In a couple
of instances, question marks are inserted on the working
hypothesis to indicate relationships that are particularly
tentative at present.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

A WORKING HYPOTHESIS
OF CNIDARIAN PHYLOGENY

Cnidaria is one of the earliest diverging clades within
Metazoa, and surprisingly its precise position within the
early diverging animal lineages—Porifera, Placozoa, Bi-
lateria, Ctenophora, and Cnidaria—has remained elusive
(Collins et al., 2005b; Dunn et al., 2008). That said, it
has become ever clearer that Cnidaria is more closely re-
lated to Bilateria than is Ctenophora, a finding based on
a synthetic consideration of morphology (Salvini-Plawen,
1978), later supported by 18S rDNA data (Wainright et al.,
1993; Collins, 1998), and most recently confirmed by a
large analysis of many sequences of data from expressed
gene transcripts (derived from large-scale sequencing of
messenger RNA; known as expressed sequence tags, or
ESTs) (Dunn et al., 2008; although note that the analy-
ses published therein suggest that Ctenophora is the ear-
liest diverging extant metazoan lineage, which is either
a radical new finding or an indication of bias in the re-
sults). Ribosomal data, both 18S (e.g., Collins, 1998)
and combined 18S and 28S (Medina et al., 2001; Cart-
wright and Collins, 2007), strongly suggest that Cnidaria
forms a clade with Bilateria and the little-known phylum
Placozoa, and that Cnidaria may be the sister group of
either taxon or both together. More recently, phylogenetic
analyses using entire genomes (unfortunately without any
representatives of Ctenophora) found Placozoa to be the
sister group of a clade composed of Cnidaria plus Bila-
teria (Srivastava et al., 2008).

Myxozoa is an interesting group of parasites that very
well may be part of Cnidaria. Although some early work-
ers suggested that they are cnidarians, based on the simi-
larity between nematocysts and myxozoan polar capsules
(Weill, 1938), they were mainly considered as protists
throughout the twentieth century. In 1995, an analysis of
18S and morphological data suggesting that myxozoans
were derived from within Cnidaria was published (Siddall
et al., 1995). However, this conclusion was doubted by
many because the 18S gene of myxozoans appears to have
evolved very quickly relative to that of most other metazo-
ans, and different analyses involving different sets of taxa
came to conflicting conclusions about the precise position
of Myxozoa within Metazoa (Smothers et al., 1994; Sid-
dall et al., 1995; Hanelt et al., 1996; Siddall and Whiting,
1999; Kim et al., 1999; Zrzavy and Hypsa, 2003). This
uncertainty was claimed to have been resolved when it
was discovered that an unusual worm-shaped animal
known as Buddenbrockia was a myxozoan (Okamura

141

NUMBER 38

“IopsO JO YURI oY} IeIU JO Ye payisse|o Ayjensn exe] UeTIeprud JueIxXa suowre sdiysuoneyot AIVUOTINJOAS JO sisoyjoddy SuUPDPIOA\ *T ARADIA

(viavaino)

VOZOSNGAWN
VOZOHLNV

VOZOXAW
VGAdSVYOV é | VOZOMNVLS
VWOZONGAH : VITIVYOOVXSH VITIWYOOOLDO

VOZOHdAOS VOZOIGOdA10d

VNIAHOVEL VWauOdSOXAW
AVSNGAWOOSIC VOZOEND VNITOGIONGAH

5A 5A
BS 6999 29 © Ss © S.5* 3458 © ©
OS SOW ~~ Ke RSS yer Se PEOseAe’ . Sy oe ONS Seek > O97? oO
3 YO Bear od SY \ NY DX \ ») NS 2) oS er OY 7 SHON NY SO DS Oo oY vx
a ee Poi? & we & & ier Ky SR gy oY > & Sor Le oe » oo see & oo
Le OEP SUS BP OP) Or OF SY SPS oe Oe Eee
© Ree x we oe & RS 2 J
ee & oF a os
iv) oe x \e 9 B)
& AN NZ
Bo

FIGURE 2. Representative cnidarians: Anthozoa, Staurozoa, and Polypodiozoa. A, Octocorallia, Holaxonia, Plaxaura from St. John, U.S.
Virgin Islands; B, Octocorallia, Alcyonacea (part of the unnamed clade including Holaxonia), Dendronephthya from Shirahama, Japan; C,
Octocorallia, Alcyonacea, Briareum from St. John, U.S. Virgin Islands; D, Hexacorallia, Scleractinia, Acropora from St. John, U.S. Virgin
Islands; E, Hexacorallia, Actiniaria, Thalassianthus from Shirahama, Japan; F, Hexacorallia, Zoanthidea, Palythoa from St. John, U.S. Vir-
gin Islands; G, Staurozoa, Eleutherocarpida, Haliclystus from Hokkaido, Japan; H, Staurozoa, Cleistocarpida, Manania from Hokkaido,
Japan. I, Polypodiozoa, Polypodium (photographed by N. Evans).

et al., 2002; Okamura and Canning, 2003). Okamura and was a close relative of nematodes and firmly derived from
colleagues showed that the morphology and DNA of Bud- within Bilateria. Most recently, this hypothesis was falsi-
denbrockia firmly placed it within Myxozoa. However, fied by analyses of EST data taken from Buddenbrockia
they also argued that the presence of four muscles located and other metazoans indicating that Myxozoa is part of
between the endoderm and ectoderm of Buddenbrockia Cnidaria (Jiménez-Guri et al., 2007), as suggested by ear-

indicated that it, and by extension Myxozoa as a whole, lier workers (Weill, 1938; Siddall et al., 1995).

NUMBER 38 ¢ 143

FIGURE 3. Representative cnidarians: Hydrozoa, Cubozoa, Scyphozoa, and Myxozoa. A, Hydroidolina, Leptothecata, Lytocarpia from Shi-
rahama, Japan; B, Hydroidolina, other Capitata, Cladonema from Hokkaido, Japan; C, Trachylina, Actinulida, Halammohydra from Bocas
del Toro, Panama (still taken from video by J. Norenburg); D, Trachylina, Limnomedusae, Gonionemus from Hokkaido, Japan; E, Cubozoa,
Carybdeida, Carybdea from Bocas del Toro, Panama; F, Cubozoa, Chirodropida, Chironex from Southern Japan (photographed in Enoshima
Aquarium); G, Scyphozoa, Semaeostomeae, Chrysaora from Bocas del Toro, Panama; H, Scyphozoa, Rhizostomeae, Cassiopea from St. John,

USS. Virgin Islands; I, Myxozoa, Myxosporea, Bivalvulida, actinospore stage of Myxobolus (photographed by A. Nawrocki and N. Evans).

Not surprisingly, given the type of data analyzed, taxon
sampling for the EST analysis was rather limited, with just
two anthozoans, two hydrozoans, and one scyphozoan
included. The myxozoan was shown to fall as the earliest
diverging lineage in a clade including the two hydrozoans
and the scyphozoan (Jiménez-Guri et al., 2007). In contrast,
even more recent analyses of rDNA data with excellent
taxon sampling resulted in best trees in which Myxozoa was
the sister group of Bilateria, a result thought to be biased by
long-branch attraction (Evans et al., 2008). Figure 1 shows

Myxozoa as the sister lineage of Medusozoa, as indicated by
the EST results, but the branch also includes a question mark
because of the small number of taxa included in the analysis
by Jiménez-Guri et al. (2007). Knowing whether myxozoans
possess linear or the typical circular mitochondrial genomes
could help place Myxozoa within Cnidaria, as this is a ma-
jor distinction between Anthozoa and Medusozoa (Bridge
et al., 1992, 1995; see following).

The inclusion of Myxozoa’s 2,200 species (Lom and
Dykova, 2006) within Cnidaria increases the richness of

144 e

the phylum significantly. As parasites with complex life
cycles involving multiple hosts, the diversity of Cnidaria
is substantially enriched as well. In recent years, relation-
ships within Myxozoa have mainly been addressed using
18S rDNA data. These data give a strong signal that spe-
cies parasitic of freshwater bryozoans in the class Malaco-
sporea (=order Malacovalvulida) form a small clade (just
three species are known) that is sister to the remaining
myxozoans classified in the class Myxosporea (Canning
et al., 2000; Kent et al., 2001). Recent classifications of
Myxosporea break the class into two orders, Bivalvulida
and Multivalvulida. While neither taxon as traditionally
recognized is monophyletic, including one aberrant bi-
valvulidid member in Multivalvulida renders Bivalvulida
paraphyletic and Multivalvulida monophyletic (as shown
in Figure 1; Kent et al., 2001; Whipps et al., 2004; Fiala,
2006; Lom and Dykova, 2006). The great majority of
myxosporeans appear to fall into two large clades, one
dominated by species inhabiting freshwater hosts, and
the other including Multivalvulida and other species that
primarily infect marine hosts (Kent et al., 2001; Holzer et
al., 2004). Examples of reversals in freshwater and marine
habits continue to accumulate, and a third smaller clade
has been identified (Fiala, 2006). Considerable diversity of
Myxozoa remains to be sampled, but existing studies indi-
cate that many myxozoan taxa, even genera, are polyphy-
letic. Continued efforts to identify morphological features
reflecting shared ancestry, which could be used to improve
the existing classification, are necessary (Fiala, 2006).

In analyses of Cnidaria (exclusive of Myxozoa), it has
generally become accepted that Anthozoa is the sister clade
of Medusozoa, a hypothesis that is buttressed by morphol-
ogy (Salvini-Plawen, 1978; Bridge et al., 1995), mitochon-
drial genome structure (Bridge et al., 1992), and rDNA
sequences (e.g., Berntson et al., 1999; Kim et al., 1999;
Medina et al., 2001; Won et al., 2001; Collins, 2002). One
recent exception is a study by Kayal and Lavrov (2008)
based on complete mitochondrial genome sequences, which
found Medusozoa (just two representatives) derived from
within Anthozoa as the sister group of three sampled rep-
resentatives of Octocorallia. Although certainly worthy of
consideration and future testing, limitations in taxon sam-
pling in the Kayal and Lavrov (2008) analysis cast some
doubt on the veracity of this finding. Similar arrangements
were also presented in early rDNA analyses that similarly
suffered from poor taxon sampling, as shown by pioneer-
ing work of Bridge et al. (1995). As indicated in Figure 1,
Anthozoa is hypothesized to consist of two well-supported
sister clades with diverse representatives, Octocorallia and
Hexacorallia. Anthozoa is usually considered to be a class

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

within the phylum (e.g., Daly et al., 2007), but making it
a subphylum, with Hexacorallia and Octocorallia as its
classes, would go some way toward balancing the classifi-
cation of Anthozoa with that of Medusozoa.

The phylogeny of Octocorallia has posed some of the
most troublesome questions in recent cnidarian systemat-
ics because of a relatively dramatic incongruence between
traditional taxonomy and molecular-based hypotheses of
relationships (Berntson et al., 2001). Nevertheless, consis-
tent progress has been made; many of the alliances first
suggested by the rDNA analyses of Berntson et al. (2001)
have been confirmed, and some morphological synapo-
morphies of recently recognized clades have been identified
(McFadden et al., 2006). It has been premature, given the
great diversity of Octocorallia remaining to be sampled, to
erect a new classification for the group, but some patterns
are emerging. There appear to be two major clades and a
minor clade or grade that branched early in the history of
Octocorallia (McFadden et al., 2006). One of the three
octocoral orders, Alcyonacea (soft corals and sea fans),
which is by far the most diverse and least distinctive, is
clearly paraphyletic. The other two orders, Pennatulacea
(sea pens) and Helioporacea (blue corals), are monophy-
letic, and each appears to be independently descended
from a paraphyletic Calcoxonia (one group of sea fans), a
suborder of Alcyonacea (McFadden et al., 2006). Another
group of sea fans known as Holaxonia all appear in one of
the major clades, along with other alcyonaceans, but there
is no strong evidence for holaxonian monophyly.

In contrast to Octocorallia, the overall picture of the
phylogeny of Hexacorallia has been relatively clear. Of the
six hexacorallian orders, several lines of evidence indicate
that Ceriantharia (tube anemones) is the earliest diverging
lineage (Berntson et al., 2001; Won et al., 2001; Daly et
al., 2002, 2003; Brugler and France, 2007). Similarly, this
same set of studies all concur in finding a close relationship
between Scleractinia (stony corals) and Corallimorpharia.
However, there has been some confusion about whether
Corallimorpharia might be derived from within Sclerac-
tinia, that is, one version of the “naked coral hypothe-
sis,” which posits that one or more hexacorallian groups
without skeletons are derived from stony coral ancestors.
Abundant evidence refutes the idea that Actiniaria (true
anemones) or Zonanthidea are derived from scleractinian
ancestors (Berntson et al., 2001; Won et al., 2001; Daly
et al., 2002, 2003; Brugler and France, 2007), but several
analyses of mitochondrial genes, including whole mito-
chondrial genomes, have found corallimorpharians to be
derived from within Scleractinia (France et al., 1996; Ro-
mano and Cairns, 2000; Medina et al., 2006). In contrast,

however, better taxon sampling of mitochondrial genomes
(Brugler and France, 2007) and analyses of other genes
with better taxon sampling (Fukami et al., 2008) favor
the hypothesis that Corallimorpharia and Scleractinia are
monophyletic sister groups. No clear picture of the re-
lationships between this clade, Actiniaria, Antipatharia
(black corals), and Zoanthidea has emerged from recent
work, as different data sets or analytical approaches
have yielded conflicting results (Berntson et al., 2001;
Won et al., 2001; Daly et al., 2002, 2003; Brugler and
France, 2007).

In present classifications, Medusozoa is usually pre-
sented as including four classes: Cubozoa, Hydrozoa,
Scyphozoa, and Staurozoa. As mentioned earlier, a recent
analysis of EST data suggested that Myxozoa is derived
from within Cnidaria, as the sister group of the three me-
dusozoans included in the analysis (Jiménez-Guri et al.,
2007). Another taxon, Polypodiozoa, which is sometimes
considered a class because of the unusual nature of the par-
asitic species within its single genus, Polypodium (Raikova,
1988), has also recently been hypothesized to fall within
Medusozoa, most likely as a close relative of Hydrozoa
(Evans et al., 2008). Thus, Medusozoa may have as many
as six classes representing rather distinct, evolutionarily
independent lineages. Evidence for the monophyly of Me-
dusozoa, albeit with the exclusion of Myxozoa and Poly-
podoizoa, comes from rDNA data (Collins, 1998, 2002;
Medina et al., 2001; Collins et al., 2006a) and observa-
tions of morphology (Werner, 1973; Salvini-Plawen, 1978;
Schuchert, 1993; Bridge et al., 1995).

Attempts to incorporate data from Myxozoa and
Polypodiozoa in analyses of cnidarian phylogeny have
been complicated by the relatively rapid rate of molecular
evolution in these two taxa. In many analyses of rDNA,
representatives of these two groups appear to be artifi-
cially attracted to bilaterian examplars and end up form-
ing sister group relationships with Bilateria (Siddall et al.,
1995; Kim et al., 1999; Zrzavy and Hypsa, 2003). In a
recent study of 18S and 28S data, dense taxon sampling
of medusozoans appears to have overcome some of this
long-branch attraction problem, at least so far as Polypo-
diozoa is concerned (Evans et al., 2008). Although the op-
timal trees of Evans et al. (2008) had Myxozoa branching
as the sister group of Bilateria, perhaps because the 28S
marker was only partially sampled for the myxozoans,
Polypodiozoa consistently fell within Medusozoa, as one
would expect based on its morphology (Raikova, 1980,
1994). Unfortunately, however, the exact position of Poly-
podiozoa within Medusozoa was shown to be dependent
upon method of analysis and the inclusion or exclusion of

NUMBER 38 ¢° 145

myxozoan representatives (Evans et al., 2008), prompting
the question mark shown in Figure 1.

Among the taxa more traditionally considered as part
of Medusozoa, Staurozoa (or Stauromedusae) may pos-
sibly be the earliest diverging lineage, a result obtained
through the analysis of both molecular and morphologi-
cal data (Collins, 2002; Dawson, 2004; Collins and Daly,
2005; Collins et al., 2006a; Van Iten et al., 2006). As ben-
thic, so-called stalked medusae, the finding that Staurozoa
might branch early in the history of Medusozoa was of
some interest because it very clearly implied that the pe-
lagic medusa stage was a feature derived within this clade.
However, Collins et al. (2006a) noted that some method-
ological choices in their phylogenetic analyses impacted
the position of Staurozoa. Further, although not specifi-
cally addressed, the position of Staurozoa was not stable
in the analyses of Evans et al. (2008). Thus far, no strong
evidence has been published suggesting that Staurozoa
is not an early diverging lineage of Medusozoa. Within
Staurozoa, there are two main taxa, Cleistocarpida and
Eleutherocarpida, neither of which appears to be mono-
phyletic despite the fact that taxon sampling was relatively
limited (Collins and Daly, 2005).

Another small class of Medusozoa is Cubozoa (box
jellyfishes). Although 18S data provide no clear signal
about the precise position of Cubozoa within Cnidaria
(Collins, 2002; Evans et al., 2008), 28S data strongly sug-
gest that Cubozoa is the sister group of Scyphozoa (true
jellyfishes), together forming the clade Acraspeda (Col-
lins et al., 2006a; Evans et al., 2008). Both 18S and 28S
strongly support cubozoan monophyly, as well as that
of its two main subtaxa, Carydeida and Chirodropida
(Collins, 2002; Collins et al., 2006a). Similarly, there is
relatively strong and stable evidence concerning the evo-
lutionary relationships among the scyphozoan orders
Coronatae, Rhizostomeae, and Semaeostomeae, although
it should be noted that taxon sampling has been sparse.
The earliest diverging lineage is Coronatae, and Rhizosto-
meae is a well-supported clade that is derived from within
Semaeostomeae (Collins, 2002; Collins et al., 2006a).

The largest and most diverse class within Medusozoa
is Hydrozoa. As indicated in Figure 1, an ancient diver-
gence within Hydrozoa divides the group into two clades,
Trachylina and Hydroidolina (Collins, 2002; Marques and
Collins, 2004; Collins et al., 2006a). Each clade has been
the subject of recent papers aimed at bringing increased
taxon and genetic marker sampling to bear on the evo-
lutionary relationships among their respective component
groups (Cartwright et al., 2008; Collins et al., 2008). As
Figure 1 shows, relationships among the major lineages of

146 e

Hydroidolina are uncertain (Collins et al., 2006a; Cart-
wright et al., 2008). In terms of taxonomy, the clade in-
cludes Leptothecata (thecate hydroids and leptomedusae)
and Siphonophora (colonial siphonophores including the
Portuguese man o’ war), two groups with ample evidence
for monophyly (Collins, 2002; Collins et al., 2006a; Cart-
wright et al., 2008). Hydroidolina also includes the large and
diverse taxon Anthoathecata (athecate hydroids and antho-
medusae), which is typically subdivided into Capitata and
Filifera. There is no evidence supporting the monophyly of
Capitata, Filfera, or Anthoathecata (Collins, 2002; Collins et
al., 2006a; Cartwright et al., 2008). Despite the difficulty in
working out the relationships among hydroidolinan clades,
some advances have been made in identifying large clades
that had not been previously recognized. For instance, Capi-
tata appears to be composed of two well-supported clades,
one dubbed Aplanulata (includes the well-known model
organisms of Hydra) in reference to the group’s lack of a
ciliated planula stage (Collins et al., 2005a, 2006a) and the
other consisting of all the other capitate groups (Cartwright
et al., 2008). The name Capitata has recently been applied
to this more restrictive clade (Cartwright et al., 2008). Simi-
larly, within Filifera, a previously unrecognized alliance of
species that bear gonophores, but not on their hydranth
bodies, has been given the name Gonoproxima. There is no
support for the monophyly of the remaining filiferans.

Trachylina is composed of four orders: Actinulida,
Limnomedusae, Narcomedusae, and Trachymedusae. The
monophyly of Narcomedusae seems to be relatively cer-
tain (Collins, 2002; Collins et al., 2006a, 2006b, 2008),
whereas the monophyly of Actinulida has yet to be tested
because just a single representative has been included in
any phylogenetic analysis (Collins et al., 2008). Trachy-
medusae, a group of pelagic species that lack polyp stages,
appears to be polyphyletic. One family (Geryonidae) has a
close relationship with a subgroup of Limnomedusae (Col-
lins et al., 2006a, 2008), whereas another (Rhopalonema-
tidae) may have given rise to the interstitial Actinulida
(Collins et al., 2008). Limnomedusae appears to represent
a grade at the base of Trachylina (Collins, 2002; Collins et
al., 2006a, 2006b, 2008). As with many cnidarian groups,
the classification of Trachylina requires refinement to bet-
ter reflect our phylogenetic knowledge.

CONCLUSION AND CLASSIFICATION

The working hypothesis of cnidarian phylogeny
presented here (see Figure 1), as do all others, requires
continued testing and refinement. Many of the studies

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

behind it have limitations, especially in taxon sampling,
and the original papers should be consulted for more
detailed assessments of strengths and weaknesses of the
analyses that they report. As the working hypothesis
results from no single analysis and was instead put to-
gether from numerous sources, some of my biases, in
the form of judgments, have had an impact on the final
form of the working hypothesis. This effect is certainly a
weakness in such an exercise and demonstrates why it is
less preferable than an analysis that relies on data sam-
pled from diverse representatives across Cnidaria. When
such an analysis is conducted, the working hypothesis
presented here may provide a helpful reference point for
comparison.

Figure 1 makes it clear that the current classifica-
tion of Cnidaria, even at the basic level of order, has
not kept up with phylogenetic advances. A new classi-
fication using taxa hypothesized to be monophyletic is
not feasible until more thorough and robust phyloge-
netic analyses are conducted. Conflicting results from
different phylogenetic studies create one hindrance to
advances in classification, but this is not really new, as
different taxonomists have always offered different clas-
sifications to reflect their changing perceptions of taxa.
More detrimental to progress in classification is the lack
of completeness in existing phylogenetic analyses. With
molecular data, individuals are sampled, and assess-
ments of the phyletic status (monophyletic, paraphyletic,
or polyphyletic) of larger taxa are not very strong until
large numbers of component species are included in an
analysis. Moreover, the relevant morphological features
that distinguish any particular clade (especially if not
corresponding to a traditional taxon) are not easily dis-
cerned without thorough sampling and examination of
its members.

Nevertheless, classifications are made to enhance
communication. Therefore, it may be prudent to attempt
classifications that better reflect ongoing advances in phy-
logenetics. Below I present one such attempt for Cnidaria.
It is not meant to be adopted, as this author has little ex-
pertise in non-medusozoan cnidarians. Instead, it is pre-
sented to illustrate one possible system for classifying tra-
ditional taxa in light of ongoing phylogenetic advances. It
relies on annotation indicating whether a given taxon is
likely to be monophyletic, paraphyletic, or polyphyletic.
Taxa for which reasonable evidence suggests monophyly
are followed by a superscript M. Taxa thought to be para-
phyletic are followed by superscript P and a list of taxa [in
brackets] hypothesized to be derived from it. Taxa that are
likely polyphyletic are placed in quotation marks. Finally,

taxa whose phyletic status is essentially unknown are left
with no annotation.

Phylum CnidariaM
Subphylum Anthozoa™
Class HexacoralliaM
Order Actiniaria
Order AntipthariaM
Order Ceriantharia™
Order CorallimorphariaM
Order ScleractiniaM
Order ZoanthidaM
Class OctocoralliaM
Oder “Alcyonacea”? [Calcoxonia, Helio-
poracea, Holaxonia, Pennatulacea]
Order Calcoxonia’ [Helioporacea, Pen-
atulacea|
Order HelioporaceaM
Order Holaxonia
Order PennatulaceaM
Subphylum Medusozoa™
Class CubozoaM
Order CarybdeidaM
Order ChirodropidaM
Class HydrozoaM
Subclass HydroidolinaM
Order AplanulataM
Order Capitata™ (excluding Aplanulata)
Order Filifera (excluding Gonoproxima)
Order GonoproximaM
Order LeptothecataM
Order Siphonophora™
Subclass TrachylinaM
Order Actinulida
Order Limnomedusae (including Ger-
yonidae)
Order Narcomedusae™
Order Trachymedusae? [Actinulida, Nar-
comedusae]
Class PolypodiozoaM
Genus Polypodium
Class ScyphozoaM
Order Coronatae™
Subclass DiscomedusaeM
Order Rhizostomeae™
Order Semaeostomeae? [Rhizostomeae]
Class StaurozoaM
Order “Cleistocarpida”
Order “Eleutherocarpida”
Subphylum Myxozoa™

M

NUMBER 38 e¢ 147

Class MalacosporeaM
Order MalacovalvulidaM

Class MyxosporeaM
Order Bivalvulida’, [Multivalvulida]
Order MultivalvulidaM

ACKNOWLEDGMENTS

I thank Michael Lang for organizing the Smithsonian
Marine Science Symposium and for inviting me to par-
ticipate. I also thank the Smithsonian Institution Marine
Science Network, which has supported me and other col-
leagues in collection activities from Smithsonian marine
laboratories. Many of these specimens yielded critical data
that were used in several of the papers referenced within
this manuscript. Finally, thanks are owed to three review-
ers, including Antonio Carlos Marques, whose comments
improved an earlier version of the manuscript. This work
was supported by “Assembling the Tree of Life” Grant no.
0531779 from the National Science Foundation.

LITERATURE CITED

Berntson, E. A., EF M. Bayer, A. G. McArthur, and S. C. France. 2001.
Phylogenetic Relationships Within the Octocorallia (Cnidaria: An-
thozoa) Based on Nuclear 18S rRNA Sequences. Marine Biology,
138:235-246.

Berntson, E. A., S. C. France, and L. S. Mullineaux. 1999. Phylogenetic
Relationships within the Class Anthozoa (Phylum Cnidaria) Based
on Nuclear 18S rDNA Sequences. Molecular Phylogenetics and
Evolution, 13:417-433.

Bridge, D., C. W. Cunningham, R. DeSalle, and L. W. Buss. 1995.
Class-Level Relationships in the Phylum Cnidaria: Molecular
and Morphological Evidence. Molecular Biology and Evolution,
12:679-689.

Bridge, D., C. W. Cunningham, B. Schierwater, R. DeSalle, and L. W.
Buss. 1992. Class-Level Relationships in the Phylum Cnidaria: Evi-
dence from Mitochondrial Genome Structure. Proceedings of the
National Academy of Sciences of the United States of America,
89:8750-8753.

Brugler, M. R., and S. C. France. 2007. The Complete Mitochondrial
Genome of the Black Coral Chrysopathes formosa (Cnidaria: An-
thozoa: Antipatharia) Supports Classification of Antipatharians
Within the Subclass Hexacorallia. Molecular Phylogenetics and
Evolution, 42:776-788.

Canning, E. U., A. Curry, S. W. Feist, M. Longshaw, and B. Okamura.
2000. A New Class and Order of Myxozoans to Accommodate
Parasites of Bryozoans with Ultrastructural Observations on Tet-
racapsula bryosalmonae (PKX Organism). Journal of Eukaryotic
Microbiology, 47:456-468.

Cartwright, P., and A. G. Collins. 2007. Fossils and Phylogenies: Inte-
grating Multiple Lines of Evidence to Investigate the Origin of Early
Major Metazoan Lineages. Integrative and Comparative Biology,
47:744-751.

Cartwright, P., N. M. Evans, C. W. Dunn, A. C. Marques, M. P. Mig-
lietta, P. Schuchert, and A. G. Collins. 2008. Phylogenetics of

148 e

Hydroidolina (Hydrozoa: Cnidaria). Journal of the Marine Bio-
logical Association of the United Kingdom, 88(8), doi:10.1017/
$0025315408002257.

Collins, A. G. 1998. Evaluating Multiple Alternative Hypotheses for the
Origin of Bilateria: An Analysis of 18S rRNA Molecular Evidence.
Proceedings of the National Academy of Sciences of the United
States of America, 95:15458-15463.

. 2002. Phylogeny of Medusozoa and the Evolution of Cnidarian
Life Cycles. Journal of Evolutionary Biology, 15:418-432.

Collins, A. G., B. Bentlage, A. Lindner, D. Lindsay, S$. H. D. Haddock,
G. Jarms, J. L. Norenburg, T. Jankowski, and P. Cartwright. 2008.
Phylogenetics of Trachylina (Cnidaria: Hydrozoa) with New In-
sights on the Evolution of Some Problematical Taxa. Journal of
the Marine Biological Association of the United Kingdom, 88(8),
doi:10.1017/S0025315408001732.

Collins, A. G., B. Bentlage, G. I. Matsumoto, S. H. D. Haddock, K.
Osborn, and B. Schierwater. 2006b. Solution to the Phylogenetic
Enigma of Tetraplatia, a Worm-Shaped Cnidarian. Biology Letters,
2:120-124.

Collins, A. G., P. Cartwright, C. S. McFadden, and B. Schierwater. 2005a.
Phylogenetic Context and Basal Metazoan Model Systems. Integra-
tive and Comparative Biology, 45:585-594.

Collins, A. G., and M. Daly. 2005. A New Deepwater Species of Stauro-
medusae, Lucernaria janetae (Cnidaria, Staurozoa, Lucernariidae),
and a Preliminary Investigation of Stauromedusan Phylogeny Based
on Nuclear and Mitochondrial rDNA Data. Biological Bulletin,
208:221-230.

Collins, A. G., P. Schuchert, A. C. Marques, T. Jankowski, M. Medina,
and B. Schierwater. 2006a. Medusozoan Phylogeny and Character
Evolution Clarified by New Large and Small Subunit rDNA Data
and an Assessment of the Utility of Phylogenetic Mixture Models.
Systematic Biology, 55:97-115.

Collins, A. G., S. Winkelmann, H. Hadrys, and B. Schierwater. 2005b.
Phylogeny of Capitata and Corynidae (Cnidaria, Hydrozoa)
in Light of Mitochondrial 16s rDNA Data. Zoologica Scripta,
34:91-99.

Daly, M., M. R. Brugler, P. Cartwright, A. G. Collins, M. N. Dawson,
D. G. Fautin, S. C. France, C. $. McFadden, D. M. Opresko, E. Rod-
riguez, S. L. Romano, and J. L. Stake. 2007. The Phylum Cnidaria:
A Review of Phylogenetic Patterns and Diversity 300 Years After
Linnaeus. Zootaxa, 1668:127-182.

Daly, M., D. G. Fautin, and V. A. Cappola. 2003. Systematics of the
Hexacorallia (Cnidaria: Anthozoa). Zoological Journal of the Lin-
nean Society, 139:419-437.

Daly M., D. L. Lipscomb, and M. W. Allard. 2002. A Simple Test: Evalu-
ating Explanations For The Relative Simplicity of the Edwardsiidae
(Cnidaria: Anthozoa). Evolution, 56:502-510.

Dawson, M. N. 2004. Some Implications of Molecular Phylogenetics for
Understanding Biodiversity in Jellyfishes, with Emphasis on Scy-
phozoa. Hydrobiologia, 530/531:249-260.

Dunn, C. W., A. Hejnol, D. Q. Matus, K. Pang, W. E. Browne, S. A.
Smith, E. Seaver, G. W. Rouse, M. Obst, G. D. Edgecombe, M. V.
Serensen, S. H. D. Haddock, A. Schmidt-Rhaesa, A. Okusu, R. M.
Kristensen, W. C. Wheeler, M. Q. Martindale, and G. Giribet. 2008.
Broad Phylogenomic Sampling Improves Resolution of the Animal
Tree of Life. Nature (London), 452:745-749.

Dykova, I., T. Tyml, I. Fiala, and J. Lom. 2007. New Data on Soricimy-
xum fegati (Myxozoa) Including Analysis of Its Phylogenetic Posi-
tion Inferred from the SSU rRNA Gene Sequence. Folia Parasito-
logica, 54:272-276.

Evans, N. M., A. Lindner, E. V. Raikova, A. G. Collins, and P. Cart-
wright. 2008. Phylogenetic Placement of the Enigmatic Parasite,
Polypodium hydriforme, Within the Phylum Cnidaria. BMC Evo-
lutionary Biology, 8:139.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Fiala, I., 2006. The Phylogeny of Myxosporea (Myxozoa) Based on Small
Subunit Ribosomal RNA Gene Analysis. International Journal for
Parasitology, 36:1521-1534.

France, S. C., P. E. Rosel, J. E. Agenbroad, L. S$. Mullineaux, and T. D.
Kocher. 1996. DNA Sequence Variation of Mitochondrial Large-
Subunit rRNA Provides Support for a Two-Subclass Organization
of the Anthozoa (Cnidaria). Molecular Marine Biology and Bio-
technology, 5:15-28.

Fukami, H., C. A. Chen, A. F. Budd, A. G. Collins, C. Wallace, Y.-Y.
Chuang, C. Chen, C.-F. Dai, K. Iwao, C. Sheppard, and N. Knowl-
ton. 2008. Mitochondrial and Nuclear Genes Suggest That Stony
Corals Are Monophyletic but Most Families of Stony Corals Are
Not (Order Scleractinia, Class Anthozoa, Phylum Cnidaria). PLOS
One, 3(9):1-9.

Hanelt, B., D. Van Schyndel, C. M. Adema, L. Lewis, and E. S. Loker.
1996. The Phylogenetic Position of Rhopalura ophiocomae (Ortho-
nectida) Based on 18S Ribosomal DNA Sequence Analysis. Molecu-
lar Biology and Evolution, 13:1187-1191.

Holstein, T., and P. Tardent. 1984. An Ultrahigh-Speed Analysis of Exo-
cytosis: Nematocyst Discharge. Science, 223:830-833.

Holzer, A. S., C. Sommerville, and R. Wootten. 2004. Molecular Rela-
tionships and Phylogeny in a Community of Myxosporeans and
Actinosporeans Based on Their 18S rDNA Sequences. International
Journal for Parasitology, 34:1099-1111.

Jankowski T., A. G. Collins, and R. Campbell. 2008. Global Diversity of
Inland Water Cnidarians. Hydrobiologia, 595:35—-40.

Jiménez-Guri, E., H. Philippe, B. Okamura, and P. W. H. Holland. 2007.
Buddenbrockia Is a Cnidarian Worm. Science 317:116-118.

Kayal, E., and D.V. Lavrov. 2008. The Mitochondrial Genome of Hy-
dra oligactis (Cnidaria, Hydrozoa) Sheds New Light on Animal
mtDNA Evolution and Cnidarian Phylogeny. Gene (Amsterdam),
410:177-186.

Kent, M. L., K. B. Andree, J. L. Bartholomew, M. El-Matbouli, S. S.
Desser, R. H. Devlin, S. W. Feist, R. P. Hedrick, R. W. Hoffmann,
J. Khattra, S. L. Hallett, R. J. G. Lester, M. Longshaw, O. Palen-
zeula, M. E. Siddall, and C. Xiao. 2001. Recent Advances in Our
Knowledge of the Myxozoa. Journal of Eukaryotic Microbiology,
48:395-413.

Kim, J. H., W. Kim, and C. W. Cunningham. 1999. A New Perspective
on Lower Metazoan Relationships from 18S rDNA Sequences. Mo-
lecular Biology and Evolution, 16:423-427.

Lom, J., and I. Dykova. 2006. Myxozoan Genera: Definition and Notes
on Taxonomy, Life-Cycle Terminology and Pathogenic Species. Fo-
lia Parasitologica, 53:1-36.

Marques, A. C., and A. G. Collins. 2004. Cladistic Analysis of Meduso-
zoa and Cnidarian Evolution. Invertebrate Biology, 123:23—42.

McFadden, C. S., S. C. France, J. A. Sanchez, and P. Alderslade. 2006.
A Molecular Phylogenetic Analysis of the Octocorallia (Cnidaria:
Anthozoa) Based on Mitochondrial Protein-Coding Sequences.
Molecular Phylogenetics and Evolution, 41:513-527.

Medina, M., A. G. Collins, J. D. Silberman, and M. L. Sogin. 2001.
Evaluating Hypotheses of Basal Animal Phylogeny Using Complete
Sequences of Large and Small Subunit rRNA. Proceedings of the
National Academy of Sciences of the United States of America,
98:9707-9712.

Medina, M., A. G. Collins, T. L. Takaoka, J. V. Kuehl, and J. L. Boore.
2006. Naked Corals: Skeleton Loss in Scleractinia. Proceedings of
the National Academy of Sciences of the United States of America,
103:9096-9100.

Meier, S., P. R. Jensen, P. Adamezyk, H. P. Bachinger, T. W. Holstein,
J. Engel, S. Ozbek, and S. Grzesiek. 2007. Sequence-Structure and
Structure-Function Analysis in Cysteine-Rich Domains Forming
the Ultrastable Nematocyst Wall. Journal of Molecular Biology,
368:718-728.

Niichter, T., M. Benoit, U. Engel, S$. Ozbek, and T. W. Holstein. 2006.
Nanosecond-Scale Kinetics of Nematocyst Discharge. Current Biol-
ogy, 16:R316-R318.

Okamura, B., and E. U. Canning. 2003. Orphan Worms and Homeless
Parasites Enhance Bilaterian Diversity. Trends in Ecology and Evo-
lution, 18:633-639.

Okamura, B., A. Curry, T. S. Wood, and E. U. Canning. 2002. Ultrastruc-
ture of Buddenbrockia Identifies It as a Myxozoan and Verifies the
Bilaterian Origin of the Myxozoa. Parasitology, 124:215-223.

Peterson, K. J., and N. J. Butterfield. 2005. Origin of the Eumetazoa:
Testing Ecological Predictions of Molecular Clocks Against the Pro-
terozoic Fossil Record. Proceedings of the National Academy of
Science of the United States of America, 102:9547-9552.

Raikova, E. V. 1980. Morphology, Ultrastructure and Development of
the Parasitic Larva and Its Surrounding Trophamnion of Poly-
podium hydriforme Ussov (Coelenterata). Cell Tissue Research,
206:487-500.

. 1988. “On the Systematic Position of Polypodium hydriforme

Ussov (Coelenterata).” In Sponges and Cnidaria—Contemporary

State and Perspectives of Investigations, ed. V. M. Koltun and S. D.

Stepanjants, pp. 116-122. Leningrad: Zoological Institute of the

Academy of Sciences of the USSR.

. 1994. Life Cycle, Cytology, and Morphology of Polypodium
hydriforme, a Coelenterate Parasite of the Eggs of Acipenseriform
Fishes. Journal of Parasitology, 80:1-22.

Romano, S. L., and S. D. Cairns. 2000. Molecular Phylogenetic Hypoth-
eses for the Evolution of Scleractinian Corals. Bulletin of Marine
Science, 67:1043-1068.

Salvini-Plawen, L. 1978. On the Origin and Evolution of the Lower
Metazoa. Zeitschrift fuer Zoologische Systematik und Evolutions-
forschung, 16:40-88.

Schuchert, P. 1993. Phylogenetic Analysis of the Cnidaria. Zeitschrift fiir
Zoologische Systematik und Evolutionsforschung, 31:161-173.

Siddall, M. E., D. S. Martin, D. Bridge, S. S. Desser, and D. K. Cone. 1995.
The Demise of a Phylum of Protists: Phylogeny of the Myxozoa and
Other Parasitic Cnidaria. Journal of Parasitology, 81:961-967.

Siddall, M. E., and M. F. Whiting. 1999. Long-Branch Abstractions. Cla-
distics, 15:9-24.

NUMBER 38 ¢ 149

Smothers, J. F, C. D. von Dohlen, L. H. Smith Jr., and R. D. Spall. 1994.
Molecular Evidence That the Myxozoan Protists Are Metazoans.
Science, 265:1719-1721.

Srivastava, M., E. Begovic, J. Chapman, N. H. Putnam, U. Hellsten, T.
Kawashima, A. Kuo, T. Mitros, A. Salamov, M. L. Carpenter, A. Y.
Signorovitch, M. A. Moreno, K. Kamm, J. Grimwood, J. Schmutz,
H. Shapiro, I. V. Grigoriev, L. W. Buss, B. Schierwater, S. L. Del-
laporta, and D. S. Rokhsar. 2008. The Trichoplax Genome and the
Nature of Placozoans. Nature (London), 454:955—960.

Van Iten, H., J. De Moraes Leme, M. G. SimGdes, A. C. Marques, and
A. G. Collins. 2006. Reassessment of the Phylogenetic Position
of Conulariids (?Ediacaran-Triassic) Within the Subphylum Me-
dusozoa (Phylum Cnidaria). Journal of Systematic Paleontology,
4:109-118.

Wainright, P. O., G. Hinkle, M. L. Sogin and S. K. Stickel. 1993. The
Monophyletic Origins of the Metazoa; An Unexpected Evolution-
ary Link with Fungi. Science, 260:240-243.

Weber, J. 1989. Nematocysts (Stinging Capsules of Cnidaria) as Don-
nan-Potential Dominated Osmotic Systems. Journal of Biological
Chemistry, 265:9664-9669.

Weill, R. 1938. LInterpretation des Cnidosporidies et Ia Valeur Tax-
onomique de Ieur Cnidome. Leur Cycle Comparé ii Ia Phase Iarvaire
des Narcomeduses Cuninides. Travaux de la Station Zoologique de
Wimereaux, 13:727-744.

Werner, B. 1973. New Investigations on Systematics and Evolution of the
Class Scyphozoa and the Phylum Cnidaria. Publications of the Seto
Marine Biological Laboratory, 20:35-61.

Whipps, C. M., G. Grossel, R. D. Adlard, H. Yokoyama, M. S. Bryant,
B. L. Munday, and M. L. Kent. 2004. Phylogeny of the Multivalvu-
lidae (Myxozoa: Myxosporea) Based on Comparative Ribosomal
DNA Sequence Analysis. Journal of Parasitology, 90:618-622.

Won, J., B. Rho, and J. Song. 2001. A Phylogenetic Study of the An-
thozoa (Phylum Cnidaria) Based on Morphological and Molecular
Characters. Coral Reefs, 20:39-50.

Zrzavy, J., and V. Hypsa. 2003. Myxozoa, Polypodium, and the Ori-
gin of the Bilateria: The Phylogenetic Position of “Endocnido-
zoa” in Light of the Rediscovery of Buddenbrockia. Cladistics,
19:164-169.

Ar : : = F. ‘a FAL : ; ? wine 9

BS Mes» sik i aie Poe
og oleae |

a id uaa ta Pay O Sane

us pine aur iy phe we J
i pe A) Fad. rh haiti ,
Bile plo) Me oe
: ” Al eT TE ave is 7
- 7 i Peet a ~
a
v } ¥
- ‘ y sg : 7 Big 28 eS
D Ly Oey A ne? Fel) : bik Ai aot f
: ; *
; vail ie
: “ i
ore
: (LD Seat ale Presi i
7 Oper ene, a ae id eae |
-
“y ui Wag ly hs
£ , vee _ we sn one IWR «
ay - e | qe 4 ;
ay ‘ aes Ve Ea oe * ¢ ii
Spf ro 5) + % i coimianel <5 Weide i
sy
ret wh
a 7% 404
f a
I ?
&
‘ i n es ; f
ye ? © i; c a
i ’ Fae eS
tr
7h,
t a He ,
¢ a i

Biodiversity and Abundance of Sponges
in Caribbean Mangrove: Indicators
of Environmental Quality

Maria Cristina Diaz and Klaus Ruitzler

Maria Cristina Diaz, Museo Marino de Margarita,
Blvd. El Paseo, Boca del Rio, Margarita, Edo. Nueva
Esparta, Venezuela. Klaus Ruetzler, Department of
Invertebrate Zoology, National Museum of Natural
History, Smithsonian Institution, Washington, D.C.
20560-0163, USA. Corresponding author: M. C.
Diaz (taxochica@gmail.com). Manuscript received
9 June 2008; accepted 20 April 2009.

ABSTRACT. We have long been fascinated by the lush biological diversity seen on sub-
tidal substrates in Caribbean saltwater mangroves. Several groups of plants and sessile
invertebrates flourish on the submerged prop roots of red mangrove (Rhizophora man-
gle), competing for space and tolerating a stressful range of ecological variables (tempera-
ture, salinity, nutrients, sedimentation) that is quite different from the more stable climate
on nearby coral reefs. To test the limits of tolerance, we monitored populations of these
organisms, the abundant sponges in particular, at environmentally and geographically
dissimilar locations in Panama and Belize. We used relative abundance estimates and
frequency counts of major ecologically functional groups and common sponge species to
establish baselines, and we repeated our surveys over long time spans (months to years)
to find correlations between community and environmental changes. Both study loca-
tions demonstrated environmental quality decline during the time of observation, mainly
through mangrove clear-cutting, followed by increase of suspended fine sediments from
dredging reef sands and filling in intertidal land, and elevation of nutrient levels from
terrestrial inputs. Although our methods are still in a stage of refinement, our data are
leading the way to responsible monitoring of our most precious coastal resources in the
tropics. We find that photosynthetic organisms (cyanobacteria, algae) and filter-feeding
invertebrates (sponges, ascidians, bivalves, bryozoans) count among the “canaries in the
coal mine” as effective indicators of environmental change.

INTRODUCTION

Red mangrove trees, Rhizophora mangle, grow along thousands of kilo-
meters of Caribbean shorelines, protecting them from storm erosion and of-
fering habitat to many organisms (Ritzler and Feller, 1988, 1996; de Lacerda
et al., 2002). Caribbean mangroves harbor from a handful to more than 100
sponge species at any one particular site (Table 1). Available data indicate that
sponges may make up 10% to 70% of epiphytic species diversity on submerged
mangrove roots. The best studied mangrove sponge faunas are described from
islands off southern Belize, with species richness reported between 50 and 147
species (Rutzler et al., 2000; Wulff, 2000; Diaz et al., 2004), followed by faunas
from a few islands in the Bocas del Toro Archipelago, Panama, with 65 spe-
cies (Diaz, 2005), from various mainland and island sites off Venezuela with
62 species (Sutherland, 1980; Diaz et al. 1985; Orihuela et al., 1991; Pauls,

152 e

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

TABLE 1. Number of species of Porifera and other epifaunal taxa reported from Caribbean mangroves (n.a. = no data available).

Country Locality Porifera Other taxa Author
Antilles Guadalupe, Trinidad, Puerto Rico 4-10 32-70 Toffart (1983)
Bahamas Bimini 13 n.a. Ritzler (1969)
Belize Four cays 24 59 Farnsworth and Ellison (1996)
Twin Cays 54 n.a. Ritzler et al. (2000)
Pelican Cays 147 217 See Macintyre and Riitzler (2000)
Cuba n.a. 48 n.a Alcolado (unpublished data)
Panama Bocas del Toro 60 n.a. Diaz (2005)
USA Indian River, Florida 3 25 Bingham and Young (1995)
Venezuela Buche Bay 16 32 Sutherland (1980)
Morrocoy National Park 23 n.a Diaz et al. (1985)
Turiamo Bay 10 n.a Pauls (2003)
Cienaga Bay 26 n.a. Pauls (1998)
La Restinga National Park 18 35 Orihuela et al. (1991)
La Restinga National Park 40 n.a Diaz et al. (2003)

1998, 2003; Ramirez, 2002; Diaz et al., 2003; Pérez,
2007), and various mangrove sites in Cuba, with 41 spe-
cies (Alcolado, unpublished data). Other reports are from
Colombia, with 26 species (Zea, 1987; S. Zea, National
University of Colombia, personal communication, 2006);
Jamaica, with 18 species (Hechtel, 1965; Lehnert and van
Soest, 1998); and Trinidad and Guadalupe, with 6 species
(Toffart, 1983) (clearly representing only a portion of the
mangrove sponge diversity there).

Most of the mangrove systems in the Caribbean re-
main unexplored, leaving a large void in biodiversity in-
formation. Most studies just cited show that the more
closely these communities are investigated, the more new
species are being discovered. An example is the research
by the Caribbean Coral Reef Ecosystems Program in Be-
lize during the past 25 years (Rutzler et al., 2000, 2004).
In particular, specialists on certain sponge taxa discov-
ered and described numerous species in the families
Suberitidae (order Hadromerida) (Ritzler and Smith,
1993), Chalinidae (order Haplosclerida) (de Weerdt et
al., 1991) and Mycalidae (order Poecilosclerida) (Hadju
and Ritzler, 1998). A recent revision of Caribbean Lis-
sodendoryx allowed the reinterpretation of L. isodyc-
tyalis (Carter, 1882) and seven other species, four of
them new to science (Ritzler et al., 2007). Similarly, two
unique haplosclerids were found in Belizean and Pana-
manian mangroves: a thin, erect, and fragile undescribed
species of Haliclona from Twin Cays, and Xestospongia
bocatorensis, a thin crust occurring in Bocas del Toro
mangroves and reefs. Both are in an endosymbiotic rela-
tionship with filamentous Cyanobacteria, a very unusual

occurrence in this order of sponges (Diaz et al., 2007,
Thacker at al., 2007).

Besides the importance of sponges species richness,
they may be one of the most abundant animal groups
in mangrove habitats. In Belize, for instance, on the lee-
ward sides of islands, sponges cover 10% to 50% of the
root surfaces, followed in importance by sea anemones,
ascidians, and algae (Farnsworth and Ellison, 1996; Diaz
et al., 2004). In the Caribbean, epibiont mangrove com-
munities have been interpreted as highly heterogeneous
(Riitzler, 1969; Sutherland, 1980; Alcolado, 1985; Al-
varez, 1989; Calder, 1991a; Bingham, 1992; Diaz et al.,
2004) as a result of low recruitment rates (Zea, 1993;
Maldonado and Young, 1996), low and fragmented
available space (Jackson and Buss, 1975; Sutherland,
1980), and stochastic processes in the long term (Bing-
ham and Young, 1995; Ellison et al., 1996). Abundance
and distribution for sponges and algae in these commu-
nities have been related to environmental factors, such
as light intensity, tides, wave impact, air exposure, and
sedimentation (Ritzler, 1995), and to biological factors,
such as larval supply (Farnsworth and Ellison, 1996),
root abundance, competition, and predation (Calder,
1991b; Littler et al., 1985; Taylor et al., 1986; Ellison
and Farnsworth, 1992; Riitzler, 1995; Riutzler et al.,
2000; Wulff, 2000). Algae abound on the shallow, well-
lit parts of stilt roots, and their abundance and species
composition are highly susceptible to the presence of
grazers. Sponges are most abundant on the lower sub-
tidal portions of the stilt roots and dominate peat bank
walls and undercuts.

The major physical and biological processes are mod-
ulated by competitive abilities, such as growth rates and
chemical defenses against predation (Wulff, 2000, 2004,
2005; Engel and Pawlik, 2005). Short-term epibiont abun-
dances are likely to be determined by interspecific com-
petitive interactions and predation, while long-term abun-
dances are limited by seasonal environmental changes,
such as freshwater inputs during periods of rain, strong
tidal currents, waves, and stochastic processes that make
these communities unstable (Bingham and Young, 1995;
Ellison et al., 1996). Despite important generalizations
about mangrove benthos ecology, we lack understanding
of the temporal or spatial variation within most epibiont
groups and knowledge about species occurrence, abun-
dance, dominance, and interactions. For example, we do
not know which species are generally abundant in these
communities, how the hierarchy changes with the year’s
seasons, and if there are predictable succession patterns.
Our current lack of knowledge prevents us from discerning
between natural variations, for instance, seasonal or yearly
dynamics, and artificial disturbances caused by humans.

The present work pursues the overall goal of a better
understanding of diversity, biogeography, and ecological dy-
namics and their causes among the sponges in Caribbean
mangroves. It encompasses two major aspects: evaluation
of our current knowledge of epiphytic sponge taxa and the
contribution of new data on causes for species richness, dis-
tribution, abundance, and dynamics, particularly from the
examples of mangrove in Panama and Belize. The survey
carried out in Bocas del Toro (Panama) intends to follow
short-term changes (over one year) in the epiphytic fauna of
mangrove roots, whereas the study in Belize will clarify shifts
in distribution of taxa over a longer period (four years).

METHODS
SPONGE SPECIES DISTRIBUTION IN CARIBBEAN MANGROVES

The distribution of species in Caribbean mangroves was
determined from currently published data or unpublished
data provided to the authors. Faunas from different regions
were compared by using a cluster analysis with the Bray—
Curtis dissimilarity coefficient, which is part of the Multi-
variate Statistical Package (MVSP 3.1) (van Soest, 1993).

SPONGE IDENTIFICATION

Specimens were identified in situ or, when necessary,
briefly characterized and photographed, with a sample

NUMBER 38 °¢ 153

preserved in ethanol. In the laboratory, routine microscope
preparations were made by cleaning spicules in household
bleach and hand-cutting perpendicular and tangential sec-
tions, which were dehydrated and mounted in Permount
and examined under the light microscope.

PHYSICOCHEMICAL VARIABLES

Temperature and salinity were measured at 0 and 50
cm depth at the Belize sites (January and August 2004),
and in Bocas del Toro (February, June, and September
2004). Sedimentation rates were estimated from accu-
mulations in buried sediment traps (plastic pipes, 10
cm diameter, 50 cm length) left in place for 210 days in
Belize and 150 days in Panama. The trapped sediment
was oven-dried (50°C), and its composition was deter-
mined as percentage of mud (including the very fine clay
fraction) (<0.002-0.05 mm) and sand (0.05-2 mm).
Approximate values of calcium carbonate content were
determined from weight loss after exposure to changes
of dilute hydrochloric acid, and sediment deposition
rates were calculated (g/m7/day). Seawater samples (500
mL each) were taken in Belize (September 2003) and at
Bocas del Toro (September 2004) at low and high tide,
filtered (0.2 tm, GF/F filter) and frozen for nutrient
analysis (Astor, 1996). Nutrient values were determined
by spectrophotometric technique using the procedure
described earlier (Diaz and Ward, 1997). Qualitative
observations about habitat types surrounding the man-
grove fringe were recorded, as well as an estimate of the
level of human disturbance.

SURVEY SITES, BOCAS DEL TORO, PANAMA

Four sites within a perimeter of 10 km were selected
and surveyed during 2004 (Figures 1, 2): (1) STRI Point:
location of the Smithsonian Tropical Research Institute’s
laboratory, several mangrove stands close to reef patches
in the southwest of Colon Island, and near a well-
developed area with a housing complex that is part of
the station; (2) Solarte In: a protected lagoon in the east
of Solarte Island, site of a modest housing development;
(3) Solarte Out: a pristine mangrove island close to reef
patches to the west side of Solarte Island; and (4) Big
Bight: a pristine, mangrove-lined lagoon surrounded by a
well-developed terrestrial forest on Colon Island, less than
5 km northwest of STRI Point. General physicochemical
characterization and geographic location of the sites are
presented in Table 2 and the nutrient regime in Table 3.

154 e

82°17'34" W

_

j ’ Colon Island

Si He

2
q
oA
Big Bight bye

STRI Point

Almirante Bay

g

= Pond
L

CO)
ta)
i)

q
Gan Cristobal

Island

Solarte Out 6

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Caribbean Sea

Pacific Ocean

ASS

& "a

Bj

FIGURE 1. Map of research area at Bocas del Toro, Panama (STRI = Smithsonian Tropical Research

Institute).

Mangrove prop roots (25 per site) were haphazardly se-
lected within a 30 m length of the mangrove fringe. The
front side (facing the channel) of each selected root was
photographed along its entire length and set to scale by
tying a transparent measuring tape to the high-tide water
mark. The tape was prevented from floating by attaching
a metal weight to its lower end. Three to seven photo-
graphs were taken, depending on the root length. Roots
were rephotographed four times during the year (Feb-
ruary 2004, June 2004, September 2004, and February
2005). From these images, abundance values of epifauna
were estimated by measuring the projected area of each
species using the CPCe program. The area (cm*) covered
by each taxon was divided by the root length (m), so that
the relative abundance values are related to a measure
of available substrate. Cover of each taxon is reported
as the sum of its abundances (cm’/m root) on all roots
at a particular site. Eight categories of epiphytes were
distinguished: cyanobacterial mats (monospecific stands

of cyanobacteria), green algae, red algae (including both
crustose calcareous and fleshy species), turf (mixture of
densely packed red, green, and cyanobacterial filaments),
sponges, hydroids, bivalves, and ascidians; the ninth cat-
egory was “empty” (spaces not occupied by macrofauna
or macroalgae). When small organisms were found over-
growing a large one (such as Spongia tubulifera, Hyrtios
proteus) the projected area of both species was included.
The number of roots finally analyzed per studied site was
reduced to 14-22 because there was some accidental loss
of photographic data.

Survey SITES, TWIN CAYS AND PELICAN CAYS, BELIZE

Three sites at Twin Cays and one in the Pelican Cays
were surveyed in August 2003 and four years later in Au-
gust 2007 (Figures 3, 4; see Tables 2, 3). Two of the Twin
Cays sites, the Lair Channel and Hidden Creek, are deep
creeks that branch off the Main Channel; Sponge Haven

NUMBER 38 e¢ 155

FIGURE 2. Views of research locations at Bocas del Toro, Panama. Top row from left: STRI Point looking south, where the transect was
located in the right foreground; Solarte Island, with transect location Solarte In near the center. Bottom row from left: mangrove fringe at
Solarte In; underwater view of mangrove prop roots showing a specimen of sponge, Chalinula molitba; mangrove roots covered by the en-
crusting sponge Halisarca sp. (undescribed; note scale in centimeters [cm] to the left), along with bivalves, algae, bryozoans, ascidians, and
other fouling invertebrates.

is a bay in the southwest of the Main Channel. The Peli-
can Cays site was in the northern part of the lagoon of
Manatee Cay. Transects (30 m long) were placed along
the red mangrove fringe at each site, with number of roots
ranging between 52 and 143. In all, the presence of the
six most conspicuous epiphyte categories was recorded
on each root within each transect: cyanobacterial mats,
macroalgae, sponges, sea anemones (Aiptasia pallida), bi-
valves, and ascidians.

RESULTS

CARIBBEAN MANGROVE SPONGE SPECIES
RICHNESS AND DISTRIBUTION

The distribution of 177 sponge species currently re-
ported from Caribbean mangroves is presented in Table 4.
A cluster analysis (Figure 5) of the best studied sites (Belize,
Cuba, Panama, Venezuela) shows the highest similarity
between Venezuela (62 species) and Panama (65 species).

156 °

SMITHSONIAN

CONTRIBUTIONS TO THE MARINE SCIENCES

TABLE 2. Characterization of study sites in Panama and Belize.

Sedimentation
Country and Habitat? Human Temperature Salinity CaCO3 Rate
locality (depth, m) impact range(°C) range(ppm) Type (% dry wt) (g/m?/day) ‘Turbidity’ Coordinates
Panama
STRI Point PR + 26-29 29-34 Mud 14-23 34-41 /+ 09°21'29.1’N,
(1.5-2) 82°16'28.9°W
Solarte In SG =e 26-29 27-32 Sand 4-24 28-58 = 09°17'05.0’N,
(2-2.5) 82°10'03.3'W
Solarte Out PR = 27-29 29-33 Sand 80-98 88-248 = 09°17'35.6"N,
(1) 82°12'08.3’W
Big Bight SG =z 27-29 27-34 Mud/ 25 40 = [22 09°22'31.1’N,
(1.5-2) sand 82°17'38.3'W
Belize
Sponge Haven SG zt 26-32 33-35 Mud 48.75 25 Zs 16°49'40.5’N,
(1-1.8) 88°06'16.5"W
Hidden Creek TC Se//AP 25.5-33 32-36 n.a n.a n.a. ae 16°49'40.5’N,
(2-2.5) 88°06'16.5"W
Lair Channel TC = 25.3-33 32-36 Mud MS) 44 = 16°49'33.7'N,
(1.5-1.8) 88°06'11.6"W
Manatee Lagoon PR/SG ~c4d 25.5-32 35-36 Mud 38.9 45 = 16°40'03.3’N,
(1-2) 88°11'32.4"W

4 Habitat abbreviations: PR = mangrove prop roots; SG = seagrass (Thalassia); TC = tidal creek with peat walls and undercuts.
Human impact and turbidity designations: +, high; +, medium; —, low.

© Survey of 2003.
d Survey of 2007.

TABLE 3. Ranges of nutrient concentrations (low tide to high
tide) at the study sites: Panama samples taken in September 2004

and Belize samples taken in September 2003.

Country and Phosphate Ammonium Nitrate
locality (umol/L) (umol/L) (umol/L)
Panama
STRI Point 0-0.048 1.32-0.988 0.26-0.253
Solarte In 0.02-0.85 0.845-0.88 0.264-0.23
Solarte Out 0.024-0.048 0.096-0.071 0.345-0.276
Big Bight 0.048-0.048 1.55-1.100 0.345-0.230
Belize
Sponge Heaven 0.528-0.624 4.79-1.88 0.5-0.41
Hidden Creek 0.336-0.786 3.19-2.35 1.1-0.39
Lair Channel 0.384-0.672 2.72-1.59 1.06-1.24
Manatee Lagoon # 0.576 1.41 0.32

4 Only one sample taken, at intermediate tide.
J Pp >

These faunas were paired with Twin Cays (54 species) and
Cuba (48 species). The most dissimilar fauna in the analy-
ses resulted from comparison with the Pelican Cays man-
groves (147 species).

MANGROVE SuRVEYS AT BOCAS DEL TORO, PANAMA
Changes in Abundance of Major Epifaunal Taxa

The relative abundance of major taxa at each of the
four localities studied between February 2004 and Feb-
ruary 2005 is shown in Figure 6. In terms of the hierar-
chy of major taxa, sponges were first or second in abun-
dance on mangrove roots at all sites, followed by algal
turfs. An exception to this pattern was found in Solarte
In (see Figure 1), where large mats of green algae, mostly
Caulerpa verticillata and Halimeda spp., dominated over
the sponges in February 2004 and 2005. Bivalves were the
third most abundant group, followed closely by unoccu-
pied (empty) spaces.

The abundance of the two most dominant groups,
sponges and algae/cyanobacteria, showed a considerable
decrease at STRI Point and Solarte In by the end of the

MEXICO

Belize City» ©

cy

BELIZE

7° N
Dangriga ®

GUATEMALA HONDURAS

88° W|

/’ Caribbean

NUMBER 38 e¢ 157

TWIN CAYS

Sea

J c
( NS ) |
\ = “Lair Channel \

ai

¢ at

)
y L__4 100 m

) % PELICAN CAYS

FIGURE 3. Map of research areas at Twin Cays and Pelican Cays, Belize.

study (February 2005), whereas abundance of both groups
increased or stayed at similar levels at the other two sites.

Sponge Species Abundances per Site

Because of the high level of heterogeneity in sponge
composition and dominance among sites, the relative
abundance of the most conspicuous epiphytic sponge spe-
cies is presented separately, by sites.

STRI Point

Sixteen of a total of 23 species found at this site com-
prise 99% of total sponge abundance. Most of them (13
species) belong to the order Haplosclerida, specifically the
family Chalinidae, and to the order Poecilosclerida. Tedania

ignis was the most abundant, followed by Clathria schoe-
nus, Spongia tubulifera, Mycale microsigmatosa, Chalinula
molitba, Haliclona manglaris, and H. tubifera. Figure 7a
shows the relative abundance of the six most common spe-
cies at this site, which added up to 87% of the total sponge
abundance. It is interesting to note that the presence of both
T. ignis and Clathria schoenus had decreased considerably
by February 2005, whereas S. tubulifera remained with
similar abundance throughout the year. Chalinula molitba
shows a considerable increase (>200%) for June 2004 and
a decrease to its initial values by February 2005.

Solarte In

Eight of 14 species found at this site constituted 99%
of the total sponge abundance. Figure 7b demonstrates

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

NUMBER 38 e¢ 159

FIGURE 4. (facing page) Views of research locations at Twin Cays and Pelican Cays, Belize. Top left, aerial view of Twin Cays looking south,
where transect locations were in the Lair channel (branching from the Main Channel center toward the left), in Sponge Haven (the small
bay at the top right), and Hidden Creek (a narrow, deep tidal channel hidden by mangrove canopy, connecting the Main Channel in the far
right background with Hidden Lake in the center background); top right, aerial view of Manatee Cay where a transect was placed in the
large lagoon to the left (Cat Cay is in the background); middle left, mangrove fringe at Sponge Haven; middle right, red mangrove prop roots
hanging free near the Pelican Cays site and covered mainly by the ropy sponge Iotrochota birotulata; bottom left, close-up of Tedania ignis
and Tedania sp. (probably T. klausi Wulff, a species described after this survey was made), both red, attached to exposed roots in the main
channel of Twin Cays; bottom right, close-up of purple ascidian (Clavelina puertosecensis) with sponges (turquoise Haliclona curacaoensis,
primarily) on root at Manatee Cay lagoon.

TABLE 4. Distribution of sponge species reported from Caribbean mangrove localities by various researchers (X = presence). Locali-
ties are abbreviated as follows: BEL, Belize; TC, Twin Cays; PC, Pelican Cays; PAN, Panama; COL, Colombia; VEN, Venezuela; TRI,
Trinidad; GUA, Guadalupe; JAM, Jamaica; and CUB, Cuba. Data sources are given in table footnotes.

BEL>

se}
O

Species 2 TC PANS COL4 VEN ¢ TRIf GUA JAMs CUB 4

Plakina jamaicensis —
Plakinastrella onkodes —
Plakortis halichondriodes Xx
Plakortis angulospiculatus —
Oscarella sp. 1 (purple) x

x

|
|
|
|
|
|
|

Oscarella sp. 2 (drab)

Cinachyrella apion

Ecionemia dominicana

Myriastra kallitetilla

Erylus formosus —

Geodia gibberosa —

Geodia papyracea x

Dercitus sp. —
x

| > x >< |
|

|
|
|
|
|

Chondrilla caribensis

Chondrosia collectrix

Cervicornia cuspidifera —
Cliona caribbaea —
Cliona raphida =
Cliona varians xX
Cliona sp. —
Placospongia intermedia —
Diplastrella megastellata —
Spirastrella coccinea —
Spirastrella hartmani —
Spirastrella mollis x
Aaptos duchassaingi —
Aaptos lithophaga —
Terpios fugax —
Terpios manglaris Xx
Prosuberites laughlini —
Suberites aurantiaca —
Tethya actinia

Tethya aff. seychellensis _—
Discodermia dissoluta —
Paratimea ? sp. x
Timea unistellata _—
Agela conifera —

| ect] est |
|

|
|

| Se Set] BSS |] BAR I BS Bd YT BBS | BB |] BB

|

| |

| Stes | |

| |

| |

| x | > | |
|

| SSB | Bet |
|

* |
|
|
|
|
|
|
|

continued

160 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

a a a a Ss
TABLE 4. continued

Species 2 AKG PC PAN COL4 VEN ¢ TRI GUAf JAM®& CUB 4

Phorbas amaranthus —

Coelosphaera raphidifera —

Lissodendoryx colombiensis —

Lissodendoryx isodictyalis x

Lissodendoryx sigmata Xx

Monanchora arbuscula —

Desmapsamma anchorata —

Biemna caribea x

Desmacella janiae —

Desmacella meliorata —

Neofibularia nolitangere —

Hymedesmia sp. —

Acarnus sp. —

Artemisina melana —

Clathria affinis —

Clathria cf. ferrea

Clathria microchela

Clathria schoenus

Clathria aff. schoenus

Clathria spinosa

Clathria venosa

Clathria virgultosa

Mycale cf. americana

Mycale angulosa

Mycale arenaria

Mycale arndti

Mycale carmigropila

Mycale citrina

Mycale escarlatei

Mycale laevis

Mycale laxissima

Mycale magniraphidifera

Mycale aff. magniraphidifera

Mycale microsigmatosa

Mycale aff. microsigmatosa

Mycale paresperella

lotrochota birotulata —

Strongylacidon sp. —

Ectyoplasia ferox _—

Eurypon laughlini —

Tedania ignis x

Tedania aff. ignis —

Dragmacidon reticulata —

Pseudaxinella ? sp. —

Ptilocaulis walpersi —

Dictyonella sp. xX

Scopalina hispida —
x

|) Bt Bat Wo Rt ||
} StS |] |] Bet [Bet Bet Bt Bt Bett KT Set |] Bet Bet Bet Bx
|
|

|
|
|
| <>< | >< |
|
|
|
|

|
|

|
|

| o% |
|
| << |
|
|
|
|

|
|
|

|
|

| mx MK mM | Od OM |

Scopalina ruetzleri

Scopalina ? sp.

Ulosa funicularis —
Amorphinopsis sp. 1 —
Amorphinopsis sp. 2 =
Ciocalypta ? sp. —
Halichondria corrugata —
Halichondria magniconulosa ? x
Halichondria melanadocia x
Halichondria poa ? xX

|

|
| > |

|

|

|

|

| Ss |
x
|
| >< >< >< |

Sd >< dd | Dd dd Dd Dd dd Dd Dd dd Dd Dd _dd DK Dd -Dd DK Dd Dd DK Dd DH Dd DK DE |
|
a |
| ES || |
|
|
|
|

NUMBER 38

e 161

Species #

PAN ¢

COL 4

VEN *

TRIf

GUA

JAMS

CUB 4

Hymeniacidon caerulea
Myrmekioderma rea
Topsentia ophiraphidites
Callyspongia arcesiosa
Callyspongia fallax
Callyspongia pallida
Callyspongia vaginalis
Haliclona caerulea
Haliclona curacaoensis
Haliclona aff. curacaoensis
Haliclona implexiformis
Haliclona aff. implexiformis
Haliclona magnifica
Haliclona manglaris
Haliclona mucifibrosa
Haliclona picadaerensis
Haliclona tubifera
Haliclona aff. tubifera
Haliclona twincayensis
Haliclona vermeuleni
Chalinula molitba
Chalinula zeae
Amphimedon compressa
Amphimedon erina
Amphimedon aff. erina
Amphimedon viridis
Niphates caicedoi
Niphates digitalis
Niphates erecta
Niphates sp.

Petrosia pellasarca
Petrosia weinbergi
Strongylophora davilai
Xestospongia carbonaria
Xestospongia muta
Xestospongia proxima
Xestospongia subtriangularis
Aka coralliphaga

Aka siphona

Aka sp.

Calyx podatypa
Oceanapia nodosa
Oceanapia oleracea
Cacospongia sp.
Fasciospongia? sp.
Hyrtios proteus

Hyrtios sp.
Smenospongia aurea
Ircinia campana

Ircinia felix

Ircinia strobilina
Spongia pertusa

Spongia tubulifera
Dysidea etheria

Dysidea fragilis

Dysidea janiae

Aplysilla glacialis
Chelonaplysilla aff. erecta
Darwinella rosacea

| xx mK | em | mK] OX | Cd

| oes | | es

| < |

| P< PS DM DM PK DKS DS ODS PS OK ODM DS OK ODS | OD OS | OK OS OK OM DS OK OM OO OK OM OS OM OM OM OK OM OO | OOO

Sal gt Sh Bt] Bl Bt Sd BS Bd Bg BB Bt |

|

| mm | Om KK | Om | OK | Om |

| Bets || 4 Be |] Sa Be Be |] |] Ba

Ba os || eat |] BSP ee est Bel est

continued

162 e

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

TABLE 4. continued

Species # TC PC PAN ©

COL4

VEN ¢ TRI! GUA! JAM ®& CUB 4

Halisarca caerulea —
Halisarca sp. (white) XxX
Aiolochroia crassa —
Aplysina archeri —
Aplysina fistularis —
Aplysina insularis —
Aplysina fulva
Aplysina lacunosa —
Verongula rigida —_
Clathrina primordialis xX
Sycon sp. Xx

| Ba Be Bel Bea Se Be Be Bx
|

Leucandra aspera —

4 Species are listed in taxonomic order according to class, order, and family.
b Riitzler et al., 2000.

© Diaz, 2005; Lehnert and van Soest, 1998.

d Zea, 1987; unpublished data.

© Sutherland, 1980; Diaz et al., 1985; Orihuela et al., 1991; Pauls, 1998, 2003; Ramirez, 2002; Diaz et al., 2003; Perez, 2007.

' Toffart, 1983.
& Hechtel, 1965.
h Alcolado, unpublished data.

Pelican Cays, Belize

Cuba

Twin Cays, Belize

Panama

Venezuela
0.6 0.5 04 0.3 0.2 0.1 0
Bray Curtis

FIGURE 5. Similarities of mangrove sponge faunas from Belize,
Panama, Venezuela, and Cuba. The dendrogram is built from a bi-
nary matrix (presence or absence) of species distribution using an
unweighted pair-group method with arithmetic mean (UPGMA)
clustal analysis program, with the Bray—Curtis distance index.

the relative abundance of the six most common species,
which comprised 96% of all sponges. Tedania ignis and
Mycale microsigmatosa were among the top species; Hal-
isarca sp. (a species so far undescribed) and Mycale car-
migropila appeared to be among the major components.
Similar to STRI Point, most of the dominant species de-
creased in abundance or disappeared by the end of the
study, while Halisarca remained steady in abundance
throughout the study period. Three of the common spe-
cies at this site (Dysidea etheria, Haliclona curacaoensis,
and Mycale carmigropila) show an increase of sponge
growth in the warmer periods (either June or September
2004), followed by a decrease in size during cooler peri-
ods (February 2005).

Solarte Out

Six of nine species found at this site made up 99%
of total sponge abundance (Figure 7c). Tedania ignis con-
tinues to dominate, followed by three species not seen in
the previously discussed sites: Spirastrella mollis, Hali-
clona vermeuleni, and H. caerulea. It is notable that the
(projected) area coverage of the dominant species is much
lower here than that at the other sites (most values are less
than 500 cm?/m).

| Feb. 2004
Jun. 2004 —
_ Sep. 2004 _
Feb. 2005

CYA GRN RED TUR POR HYD BIV ASC EMP

| Feb. 2004
WH) vun. 2004
J Feb. 2005 _

CYA GRN RED TUR

NUMBER 38 ¢ 163

RED TUR POR HYD BIV ASC EMP

BIV ASC EMP

FIGURE 6. Relative abundance of major functional groups growing on mangrove roots (expressed as projected area [cm] per length [m]
of root) at four Bocas del Toro sites, between February 2004 and 2005: a, STRI Point; b, Solarte In; c, Solarte Out; d, Big Bight. (ASC =
ascidians; BIV = bivalves; CYA = Cyanobacteria; EMP = empty space; GRN = green algae; HYD = hydroids; POR = sponges [Porifera];

RED = red algae; TUR = algal-cyanobacterial turf.)

Big Bight

Twelve of 17 species found at Big Bight comprised
99% of the total sponge abundance; 6 of these amounted
to 90% (Figure 7d). The most abundant species—Teda-
nia ignis, Mycale microsigmatosa, and Haliclona man-
glaris—increased in size throughout the year, whereas
Lissodendoryx colombiensis and Dysidea etheria gained
in size up to September 2004 but disappeared altogether
in February 2005. It is worth pointing out the large val-
ues for area coverage, as compared to the other loca-
tions. The September 2004 data from this site were ac-
cidentally lost.

Sponge Species Ranks

The most common sponges at each site amount to 21
species, from a total of 40 distinguished in the studied ar-
eas. Abundance ranks from each site are listed in Table
5. Only one species, Tedania ignis, maintained the same
rank at all sites, as the most abundant species. The second
and third most abundant species were Clathria schoenus
and Spongia tubulifera at STRI Point, Mycale microsig-
matosa and Halisarca sp. at Solarte In, Spirastrella mollis
and Haliclona manglaris at Solarte Out, and M. microsig-
matosa and H. manglaris in Big Bight. Seven of these 21
common sponges were only found at one site.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

3000

[__] Feb. 2004

2500 eerie Te 2500--
Jun. 2004

GR Sep. 2004
Wi Feb. 2005

2000 2000-}-
1500-}-

40003 ———$—————

| 0 »
Csch Tign Hman Htub Cmol Stub Mcar Mmic Tign Heur Deth Hasp
a b
3000 3000
Feb. 2004
2500 SS 2500-
Jun. 2004
Feb. 2005
2000 = = sss = : Taeee eerie 2000—
1500 1500
1000 1000-—
500 500-
0 ce —i iH | B yy bo 2 (0) 69 SES pee fi sed
Smol Mcar Tign Hman Hver Hcae Lcol Mmic Tign Hman Nere Deth
c d

FIGURE 7. Relative abundance of most common sponge species growing on mangrove roots (expressed as projected area [cm] per length
[m] of root) at four Bocas del Toro sites between February 2004 and 2005: a, STRI Point; b, Solarte In; c, Solarte Out; d, Big Bight. (Cmol
= Chalinula molitba; Csch = Clathria schoenus; Deth = Dysidea etheria; Hasp = Halicarca sp.; Hcae = Haliclona caerulea; Hcur = H.
curacaoensis; Hman = H. manglaris; Hpro = Hyrtios proteus; Htub = Haliclona tubifera; Hver = H. vermeulei; Lcol = Lissodendoryx co-
lombiensis; Mcar = Mycale carmigropila; Mmic = M. microsigmatosa; Nere = Niphates erecta; Smol = Spirastrella mollis; Stub = Spongia
tubulifera; Tign = Tedania ignis).

sites, followed by colonial ascidians, macroalgae, and cya-
nobacteria. Only at Manatee Cay had sponge occurrence
on roots decreased since 2003, whereas at the other three
sites it either increased or stayed nearly the same. Ascid-

MANGROVE SurRVEYS IN BELIZE

Changes in Frequency of Occurrence
of Major Functional Groups

We determined the frequency of occurrence of impor-
tant functional groups growing on mangrove roots to be
able to assess changes over time (Figure 8). The six com-
pound groups recorded in our surveys were cyanobacteria,
algae, sponges, sea anemones, bivalves, and ascidians. In
terms of the hierarchy, sponges were first or second at all

ian occurrence decreased considerably (10%-26%) at all
sites between 2003 and 2007. These changes in sponge
and ascidian populations were accompanied by cyanobac-
terial blooms at three sites (Lair Channel, Hidden Creek,
and Manatee Cay), where increases of 10% to 57% of
these organisms were recorded. One of the less abundant

NUMBER 38 e¢ 165

TABLE 5. Ranking of the most common sponge species according to their abundance at each studied site in the Bocas del Toro region,

2004-2005.
Rank in abundance

Species STRI Point Solarte In Solarte Out Big Bight
Spirastrella mollis 0 0 2 0
Lissodendoryx colombiensis 0 0 0 5
Lissodendoryx isodicyialis 11 0 0 11
Clathria schoenus 2 0 0 12
Mycale carmigrophila 16 4 6 16
Mycale microsigmatosa 5 2 0 D
Totrochota birotulata 7 0 0 0
Tedania ignis 1 1 1 1
Haliclona caerulea 0 0 5 0
Haliclona curacaoensis 13 5 0 0
Haliclona implexiformis 10 0 0 0
Haliclona manglaris 6 7 3 3
Haliclona tubifera 8 0 0 0
Haliclona vermeuleni 0 0 4 0
Chalinula molitba 4 0 0 10
Amphimedon sp. 0 0 0 8
Niphates erecta 0 0 0 4
Hyrtios proteus 15 0 0 6
Spongia tubulifera 3 8 0 13
Dysidea etheria 0 6 0 7
Halisarca sp. 0 3 0 18

groups, the sea anemone Aiptasia pallida (Cnidaria), is
worth mentioning for its striking change in occurrence at
the Twin Cays sites. Although the population remained
steady at Hidden Creek (8%), it doubled in Lair Channel
(10%-24% of roots occupied), but it apparently disap-
peared from Sponge Haven where it had been present on
20% of the roots in 2003. The number of roots available
for settlement per site increased considerably at Sponge
Haven and Manatee Cay lagoon, although it decreased in
Hidden Creek and Lair Channel.

Sponge Species Frequencies per Site

The distinctive species composition and richness at
each site warrant separate presentations.

Sponge Haven
Most mangrove-specific species, such as Halichondria

magniconulosa, Haliclona curacaoensis, H. manglaris, H.
implexiformis, Hyrtios proteus, Lissodendoryx isodyctia-

lis, and Spongia tubulifera, remained the most common
among sponges, and some even increased in frequency be-
tween 2003 and 2007 (Figure 9a).

Lair Channel

In this mangrove channel most species remained in
place between survey periods; some increased in root oc-
currence (Tedania ignis, Haliclona manglaris, H. tubifera,
Dysidea etheria) and a few decreased (Haliclona cura-
caoensis, H. implexiformis, Hyrtios proteus) (Figure 9b).
Overall, this change was accompanied by a slight decrease
in root numbers (from 105 to 91) and an increase in all
non-sponge groups except ascidians.

Hidden Creek

This tidal channel site is opposed to Sponge Haven in
its changes between 2003 and 2007 (Figure 9c). Ten of the
12 sponge species found on the transect decreased con-
siderably in occurrence on roots; only the opportunistic

166 e

Aug. 2003
Wh Aug. 2007

CYA ALG POR ANE BIV ASC

CYA ALG POR ANE BIV ASC

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

CYA ALG POR ANE BIV ASC

CYA ALG POR ANE BIV ASC

FIGURE 8. Frequency of occurrence (% of roots occupied) of major functional groups growing on mangrove roots at four sites in Belize: a,
Sponge Heaven, Twin Cays; b, Lair Channel, Twin Cays; c, Hidden Creek, Twin Cays; d, Manatee Lagoon, Pelican Cays. (ALG = algae; ANE
= sea anemones [Aiptasia pallida|; ASC = ascidians; BIV = bivalves; CYA = cyanobacteria; POR = sponges [Porifera]).

generalist Tedania ignis and Lissodendoryx isodictyalis
increased slightly.

Manatee Cay

At this lagoon site, abundance of most typical man-
grove sponge species decreased considerably during the
survey period while two common opportunistic species
(Tedania ignis, Clathria schoenus) experienced a consid-
erable boost in their populations (Figure 9d). This trend
coincided with a major increase in root numbers (from 89
to 123), similar to that which took place at Sponge Haven
during the same time span.

DISCUSSION
BIOGEOGRAPHY OF CARIBBEAN MANGROVE SPONGES

Available reports describing sponge species distribution
in Caribbean mangroves suggest the importance of geo-
graphic vicinity, with high similarities between the faunas of
Panama and Venezuela. On the other hand, this geographic
concept is upset by the incongruence of faunas encountered
at two nearby sites in Belize (Twin Cays and the Pelican
Cays). This dissimilarity is caused mostly by the presence
of several unique or usually coral reef-associated species in
the mangroves of Manatee Lagoon, an environment of par-
ticular geomorphological structure and prevailing ecologi-

Aug. 2003

ga Aug. 2007

3° ot as po or ac? ei oor a eo? go” ew? ow

NUMBER 38 °¢ 167

b VF goF a oH yeh 0a Ao a? yu? oa rok? co

CA cE WH 0 Ol WOH 2? ae OO po oP col
d

FIGURE 9. Frequency of occurrence (% of roots occupied) of sponge species growing on mangrove roots at four sites in Belize: a, Sponge
Heaven, Twin Cays; b, Lair Channel, Twin Cays; c, Hidden Creek, Twin Cays, d, Manatee Lagoon, Pelican Cays. (Aatl = Amorphinopsis
atlantica; Aeri = Amphimedon erina; Bcar = Biemna caribbea; Ccar = Chondrilla caribensis; Cmic = Clathria microchela; Cpri = Clathrina
primigenia; Csch = Clathria schoenus; Cven = Clathria venosa; Deth = Dysidea etheria; Hasp = Halisarca sp.; Hcur = Haliclona curacaoensis;
Himp = H. implexiformis; Hmag = Halichondria magniconulosa; Hman = Haliclona manglaris; Hpro = Hyrtios proteus; Htub = Haliclona
tubifera; lfel = Ircinia felix; Liso = Lissodendoryx isodictyalis; Mlae = Mycale laevis; Mmag = M. magniraphidifera; Mmic = M. microsigma-

tosa; Stub = Spongia tubulifera; Tign = Tedania ignis.)

cal conditions in the Pelican Archipelago (Macintyre and
Riutzler, 2000; Ritzler et al., 2000; Wulff, 2000).

SHORT-TERM DYNAMICS OF MANGROVE
EPIFAUNA IN PANAMA
Major Functional Groups

As previously reported, mangrove-root epiphytic
communities in Bocas del Toro are dominated either by

sponges or by algae/cyanobacteria (Farnsworth and Elli-
son, 1996; Diaz et al., 2004; Peréz, 2007). Elsewhere in
the Caribbean, other groups, such as bivalves, anemones,
or ascidians, may rival these taxa in abundance (Suther-
land, 1980; Toffart, 1983; Bingham, 1992). The dominance
of macroalgae at the protected lagoon of “Solarte In” might
be a consequence of the eastward orientation of this site
(as opposed to the westward orientation of the other three
sites), which would expose the mangrove fringe to sunlight

168 e

for longer periods, thus promoting the growth of typical
shallow-water algal species. However, further studies are
required to sustain this hypothesis. There were no observa-
tions of seasonal changes in the composition of epiphytic
taxa from one sampling period to the other at any site. The
decrease in abundance (20%-35%) found for the most
dominant groups at two sites (STRI Point and Solarte In)
coincides with housing developments that occurred since
the study started in 2004. Increases in suspended sediments
and incidences of sponges covered by silt, which were ob-
served at STRI Point during September and February 2005,
may have impacted the community. In contrast, at the more
pristine sites (Solarte Out and Big Bight), these same organ-
isms demonstrated considerable quantitative increases.

Sponge Species

The six most common species at each site consti-
tute from 87% to 99% of the total root area covered by
sponges. These dominant species differed between sites,
bringing the number of the most abundant sponges to 21,
of a total diversity of 40 species. Only Tedania ignis was
the most common species at all sites. At Solarte In the sec-
ond most common species was a thin crust of the genus
Halisarca, whereas at STRI Point it was Clathria schoe-
nus, a species with a highly variable growth form (thick
crusts to branching), supporting the common observation
that mangrove fauna can be highly heterogeneous within
one biogeographic region. It is interesting to note that 5
species that ranked near the top at the four sites were en-
crusting sponges (Mycale microsigmatosa, Dysidea ethe-
ria, Haliclona manglaris, Halisarca sp., Clathria schoenus,
and Spirastrella mollis). This result suggests that, at least
in Bocas del Toro, encrusting species are highly success-
ful competitors. The dominance of Tedania ignis was also
reported from other Caribbean locations (Toffart, 1983;
Sutherland, 1980; Wulff, 2004; Diaz et al., 2004) and is
probably related to its high and nearly year-around pro-
duction of larvae (Ruetzler, unpublished data) and rapid
growth rate (Wulff, 2005).

Dominant species were not always consistent in
abundance at all sites. For example, during the observa-
tion time Tedania ignis decreased considerably at STRI
Point and Solarte In but increased at Solarte Out and Big
Bight. Furthermore, increase or decrease in abundance
was not necessarily restricted to certain species or locali-
ties. Certain locality trends, however, were observed. At
STRI Point, where T: ignis and Clathria schoenus de-
creased or disappeared entirely from the roots, the few
large specimens of Spongia tubulifera remained with

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

only slight size changes throughout the year. At least at
one location, Solarte In, deterioration of sponges ap-
peared to be coinciding with the aforementioned housing
development, which caused an increase in suspended and
deposited sediment.

An interesting trend is the predominance of large
sponges at Big Bight versus the much smaller sizes at
Solarte Out. Solarte Out is a shallow habitat in an ex-
posed position and subjected to strong wave action and
scouring by predominantly sandy sediments (see Table 2).
These parameters must impede the growth of large in-
dividuals, with the result that small and better adapted
forms, such as Haliclona vermeuleni, H. caerulea, and
Spirastrella mollis, become very abundant. Even an oppor-
tunistic species such as Tedania ignis, common and large-
growing elsewhere, tends to be considerably restricted
in size there. On the other hand, Big Bight sponges were
found to have rapid growth rates that can be attributed to
high nutrient concentrations measured at this site, possibly
related to runoff from the dense forest that surrounds this
lagoon (see Table 3). The high variability of sponge species
composition between contiguous sites corroborates previ-
ous reports that the mangrove sponge fauna is rather het-
erogeneous in species distribution and dominance within
relatively small geographic areas (Farnsworth and Ellison,
1996; Riitzler et al., 2000; Diaz et al., 2004). This char-
acteristic is probably the result of low recruitment rate in
most species studied and, in some cases, selective physi-
cochemical variables, such as those described for Solarte
Out. A third aspect that became evident in this study is
the intrinsic growth dynamics of species over time, high
in species such as Tedania ignis and Chalinula molitba,
and low or barely noticeable in Hyrtios proteus and Spon-
gia spp. It must be recognized that species have different
lifespans, growth rates, growth periods, and frequency of
reproduction. Understanding these processes is essential to
the interpretation of community dynamics.

LONG-TERM DYNAMICS OF MANGROVE EPIFAUNA IN BELIZE
Major Functional Groups

The distribution of the four primary components of
mangrove-root ephiphytic communities in Belize—cyano-
bacteria, macroalgae, sponges, and ascidians—varied dif-
ferently at each of the four studied sites between August
2003 and 2007. Sponges were the most frequent occupants
at all four locations in 2003; by 2007, the population had
either increased (Sponge Heaven, The Lair), decreased
(Manatee Lagoon), or remained steady. The decrease at
Manatee Cay seemed to be related to macroalgal blooms

that coincided with the recent clear-cutting of the man-
grove adjacent to this lagoon and to dredging for land-fill
that released large quantities of fine sediments. Ascidian
occurrence followed a similar pattern, indicating that all
filter feeders are impacted by environmental events such
as increase of sedimentation and blockage of vents by cya-
nobacterial blooms. The effect of changing root numbers
seems to be obscured by the environmental factors, be-
cause there was no obvious relationship between changes
in root number and frequency of any of the major taxa in
the community.

Sponge Species

Comparing species composition and frequency at
the four study sites in Belize, we found that they varied
considerably during the four years between observation
periods. The most obvious parameters affecting sponge
populations were space competitors (cyanobacteria, mac-
roalgae), number of roots available for settlement, and
anthropogenic destructive events. The considerable de-
crease in cyanobacteria and macroalgae and increase in
root numbers (from 99 to 143) in Sponge Haven may be
related to the strong increase of mangrove-specific sponges
because important competitors were no longer present
and new substrata became available. In contrast, at Hid-
den Creek, the increase of filamentous cyanobacteria (to
57% of substrate area) and decrease in root numbers
(from 59 to 52) must have caused the dramatic reduction
of most mangrove-specific sponge species. In Manatee Cay
Lagoon, mangrove-specific species lost in frequency while
opportunistic species (Tedania ignis, Clathria schoenus)
gained. Overall, however, there was a reduction of sponge
populations despite an addition in root numbers. This
trend can be explained by increased algal competition and
an artificial incursion, the clear-cutting of mangrove trees
and dredging of fill material for a housing development
sometime before the 2007 survey. The dredge operation
in particular can be blamed in the short term as it causes
suspension of fine sediment, affecting the delicate filtra-
tion system of the sponges. A shift of species toward more
robust opportunists rather than typical mangrove forms is
therefore not surprising.

COMMENTS ON METHODS FOR EVALUATING MANGROVE
PRoP-ROOT COMMUNITIES

Two criteria were used in the present study to evaluate
epiphytic communities on mangrove roots. To determine
short-term dynamics (within one year; Bocas del Toro),

NUMBER 38 e¢ 169

it was expected that specimen size rather than numbers
would change. Therefore, a photographic record was made
of a specific number of roots (25) along their entire lengths
(the side facing the open water), and planimetry was used
to measure projected area cover of the fouling organisms.
From these values and the record of root length, an index
of species abundance could be calculated. Area cover has
been extensively used to compare the abundances of plant
and sessile animal communities, and it has been proven a
most practical and reliable method for reef surveys (Wein-
berg, 1981). Considering that in mangroves substrate
availability is quite low, measuring area cover gives a good
indication of how important an organism is in this com-
munity. The limitation of this method applies mostly to
stoloniferous organisms for which cover may underesti-
mate their importance. The photo-transect method proved
to be most useful in areas where visibility was very good,
but it was problematic in locations with high freshwater
or sediment input. Such conditions caused whole sets of
photographs to be impossible to interpret. This method is
also time consuming, both the work underwater and that
during photo analysis in the lab. For this reason there was
a limit to the root numbers that could be included in each
survey. Usually, to complete a survey of 25 roots in one site
it was necessary to visit twice, and evaluation of all (3-8)
photos for one root took from 30 to 60 min. In the end,
after excluding useless images, the data set was reduced to
only 14 to 22 recorded roots, depending on the site.

Alternatively, in Belize we used data on the presence or
absence of taxa on each root and thus were able to survey a
much larger number of roots (50-150), from which we de-
termined the frequency of occurrence of major taxonomic
groups and species of sponges. These data allowed moni-
toring the presence of each group or species and change
in distributions over time. This type of survey follows the
fate of the community rather than fluctuations in biomass.
The method also aids detection of a species or community
reaction to particular environmental disturbances. In terms
of time investment, it takes only 2 to 5 h to obtain frequency
data from a 30 m transect along the mangrove fringe. The
data were in hard copy once the fieldwork was completed
and were independent of visibility conditions and other
variables that may ruin photographic data.

CONCLUSIONS

Many more Caribbean mangroves must be studied be-
fore we can expect a full understanding of the biodiversity
and the biogeographic relationships of their unique and

170 e

fascinating prop-root fouling communities, particularly
the sponges. The rather disjunct pattern of sponge species
distribution found in the Panama and Belize study sites
suggests that biodiversity is better evaluated by surveying
extended stretches of mangrove fringe at numerous sites in
any region rather than short lengths of transects. Interpre-
tation of species composition and interactions can be based
on smaller-scale levels of inquiry. The most abundant or-
ganisms in the studied sites were sponges, macroalgae, cya-
nobacteria, ascidians, and bivalves. The hierarchical rank-
ing of these groups showed great variability on spatial and
temporal scales, making generalization and prediction of
structure and dynamics of communities very difficult.

The one-year study of four sites in the Bocas del Toro
region, Panama, showed various important aspects of
abundance changes in these fouling communities. First, a
few sponge species contribute most of the abundance; sec-
ond, the identity of major community components varies
within a small geographic scale; third, species have ad-
opted distinct life strategies (in growth potential, recruit-
ment rates, and asexual reproduction capabilities) that
allow for adaptations to resist stressful environmental
variables; and fourth, the combination of the factors of
large sediment grain size and energy from wave or current
action limits species habitat access, survival, and growth,
as demonstrated by the increase in turbidity from land-
filling and development in some areas.

The four-year observations in Belize made it evi-
dent that the frequency of occurrence of sponge species
and other taxa, such as cyanobacterial and macroalgal
blooms, is a relatively simple and fast measure to detect
major environmental changes. Even if sponge frequency
on the roots is not much affected by algal blooms, the
presence of mangrove-specific species certainly shows a
decline; only a couple of generalist species seem to profit
from such stressful events. The degree of disappearance
of ascidians at all four sites in Belize suggests that these
organisms may be even more sensitive to algal and cyano-
bacterial competition, as well as suspended fine sediments,
than sponges. We find, both in Belize and in Panama, that
two sponge growth forms are highly successful among
sponge root occupiers: encrusting and irregularly massive.
This observation is in contrast to open reef environments
where tubular and ramose forms predominate.

Close monitoring of the abundance and frequencies of
key mangrove benthos at specific sites and their correlation
with short-term or long-lasting environmental impacts and
stress will be a useful tool for assessing mangrove health
throughout the Caribbean region in the future.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

ACKNOWLEDGMENTS

We thank Mike Carpenter, Kathleen Smith, Estrella
Villamizar, and Martha Nicholas for support of fieldwork
based at Carrie Bow Cay, Belize; likewise, we thank Ga-
briel Jacome and Plinio Gondola for assistance with logis-
tics at the Smithsonian Bocas del Toro field station, Pan-
ama. Molly Kelly Ryan designed the maps and rendered
the graphs in final form; Carla Piantoni prepared the color
figures and helped with editorial tasks. We acknowledge
the National Coral Reef Institute for lending us the CPCe
(Coral Point Count with Excel extensions) program for
area calculations. Photographs for this paper were taken
by the following Smithsonian staff or associates: Cristina
Diaz, Ilka Feller, Diane Littler, Elisabeth McLean, Tony
Rath, and Klaus Ruetzler. This is contribution number
855 of the Caribbean Coral Reef Ecosystems Program
(CCRE), Smithsonian Institution, supported in part by the
Hunterdon Oceanographic Research Fund.

LITERATURE CITED

Alcolado, P. M. 1985. Distribucién de la abundancia y composition de
las communidades de esponjas en la macrolaguna del Golfo de Bata-
bané. [Distribution and Abundance of Sponge Communities in the
Macrolagoon of the Gulf of Bataban6.] In Proceedings of the Sym-
posium on Marine Sciences and 7th Scientific Meeting of the Institute
of Oceanography (20th Anniversary), pp. 5-10. Havana: Academy
of Sciences of Cuba.

Alvarez, A. I. 1989. Establecimiento, desarrollo y mantenimiento de una
comunidad epibenténica tropical. [Settlement, Development and
Maintenance of a Tropical Epibenthic Community.] Licenciate’s
Thesis, Universidad Central de Venezuela, Caracas, Venezuela.

Astor, Y. 1996. “Manual de andlisis de aguas para la acuicultura y las
ciencias del mar.” [“Manual for Water Analysis in Aquaculture and
Ocean Studies.”] In Fundacion La Salle de Ciencias Naturales Con-
tribuci6n Numero 192. Colecci6n Cuadernos FLASA. Serie Ciencia y
Tecnologia, 8:1-89.

Bingham, B. L. 1992. Life Histories in an Epifaunal Community: Cou-
pling of Adult and Larval Processes. Ecology, 73:2244-2259
Bingham, B. L., and C. M. Young. 1995. Stochastic Events and Dynam-
ics of Mangrove Root Epifaunal Community. Marine Ecology,

16:145-163.

Calder, D. R. 1991a. Abundance and Distribution of Hydroids in a Man-
grove Ecosystem at Twin Cays, Belize, Central America. Hydrobiol-
ogy, 216/217:221-228.

. 1991b. Associations Between Hydroid Species Assemblages and
Substrates Types in the Mangal Twin Cays, Belize. Canadian Jour-
nal of Zoology, 69:2067-2074.

de Lacerda, L. D., J. E. Conde, B. Kjerfve, R. Alvarez-Leén, C. Alarcon,
and J. Polania. 2002. “American Mangroves.” In Mangrove Eco-
systems: Function and Management, ed. L. D. de Lacerda, pp. 1-62.
Berlin: Springer-Verlag.

de Weerdt, W. H., K. Riitzler, and K. Smith. 1991. The Chalinidae (Po-
rifera) of Twin Cays, Belize, and Adjacent Waters. Proceedings of
the Biological Society of Washington, 104:189-205.

Diaz, H., M. Bevilaqua, and D. Bone. 1985. Esponjas en Manglares
del Parque Nacional Morrocoy. [Mangrove Sponges of Morrocoy
National Park.| Caracas: Fondo Editorial Acta Cientifica Vene-
zolana.

Diaz, M. C. 2005. Common Sponges from Shallow Marine Habitats
from Bocas del Toro Region, Panama. Caribbean Journal of Sci-
ence, 41(3):466-475.

Diaz, M. C., S. Pauls, E. Villamizar, A. Alvizu, M. E. Amaro, M. Cellamare,
S. Grune, I. Hernandez, S. Narciso, A. Perez, J. Perez, I. Ramirez,
R. Ramos, M. P. Romero, and P. Young. 2003. Porifera Biodiversity
in Nueva Esparta, Venezuela: Common Species from La Restinga
and Cabugua Island. Abstract. The Twin Cays Mangrove Ecosys-
tem, Belize: Biodiversity, Geological History, and Two Decades of
Change. Report from a Workshop, Smithsonian Marine Station,
Fort Pierce, Florida, December 2003. Washington, D.C.: Smithson-
ian Institution.

Diaz, M. C., K. A. Smith, and K. Riitzler. 2004. Sponge Species Richness
and Abundance as Indicators of Mangrove Epibenthic Community
Health. Atoll Research Bulletin, 518:1-17.

Diaz, M. C., R. W. Thacker, K. Riitzler, and C. Piantoni. 2007. “Two
New Haplosclerid Sponges from Caribbean Panama with Sym-
biotic Filamentous Cyanobacteria, and an Overview of Sponge-
Cyanobacteria Associations.” In Proceedings of the 7th Inter-
national Sponge Symposium: Porifera Research: Biodiversity,
Innovation and Sustainability, pp. 31-39. Rio de Janeiro, Brazil:
Museu Nacional.

Diaz, M. C., and B. B. Ward. 1997. Sponge-mediated Nitrification in
Tropical Benthic Communities. Marine Ecology Progress Series,
156:97-107.

Ellison, A. M., and E. J. Farnsworth. 1992. The Ecology of Belizean
Mangrove-Root Fouling Communities: Patterns of Epibiont Distri-
bution and Abundance, and Effects on Root Growth. Hydrobiolo-
gia, 247:87-98.

Ellison, A. M., E. J. Farnsworth, and R. R. Twilley. 1996. Facultative
Mutualism Between Red Mangroves and Root-Fouling Sponges in
Belizean Mangal. Ecology, 77(8):2431-2444.

Engel, S., and J. R. Pawlik. 2005. Interactions among Florida Sponges.
II. Mangrove Sponges. Marine Ecology Progress Series, 303:
145-152.

Farnsworth, E. J., and A. M. Ellison. 1996. Scale Dependent Spatial and
Temporal Variability in Biogeography of Mangrove-Root Epibiont
Communities. Ecological Monographs, 66:45-66.

Hadju, E., and K. Ritzler 1998. Sponges, Genus Mycale (Poecilosclerida:
Demospongiae: Porifera), from a Caribbean Mangrove and Com-
ments on Subgeneric Classification. Proceedings of the Biological
Society of Washington, 11(4):737-773.

Hechtel, G. J. 1965. A Systematic Study of the Demospongiae of Port
Royal, Jamaica. Bulletin of the Peabody Museum of Natural His-
tory, 20:1-103.

Jackson, J. B. C., and L. Buss. 1975. Allelopathy and Spatial Competi-
tion Among Coral Reef Invertebrates (Species Interactions/Commu-
nity Structure). Proceedings of the National Academy of Sciences,
72(12):5160-5163.

Lehnert, H., and R. W. M. van Soest. 1998. Shallow Water Sponges of
Jamaica. Beaufortia, 48(5):71-103.

Littler, M. M., P. R. Taylor, D. S. Littler, R. H. Sims, and J. N. Norris.
1985. The Distribution, Abundance, and Primary Productivity of
Submerged Macrophytes in a Belize Barrier-Reef Mangroves Sys-
tem. Atoll Research Bulletin, 289:1-20.

Macintyre, I. G., and K. Riitzler, eds. 2000. Natural History of the Peli-
can Cays, Belize. Atoll Research Bulletin, 466—480:1-333.

Maldonado, M., and C. M. Young. 1996. Bathymetric Patterns of
Sponge Distribution on the Bahamian Slope. Deep-Sea Research I,
43(6):897915.

NUMBER 38 e¢ 171

Orihuela, B., H. Diaz, and J. E. Conde. 1991. Mass Mortality in a Man-
grove Roots Fouling Community in a Hypersaline Tropical Lagoon.
Biotropica, 23:592-601.

Pauls, S. 1998. Estudio Sistemdtico y Biodiversidad de Porifera y Cni-
daria en la Bahia de la P. N. Henri Pittier. [Systematic and Diversity
Study of Porifera and Cnidaria of the Ciénaga de Ocumare de la
Costa Bay, Henri Pittier National Park.] Caracas: Escuela de Bi-
ologia, Universidad Central de Venezuela.

. 2003. “Esponjas.” In Biodiversidad en Venezuela. |“Sponges.”
In Biodiversity of Venezuela.|, ed. M. Aguilera, A. Azocar, and E.
Gonzales, pp. 210-219. Caracas, Venezuela: Polar Foundation and
Ministry of Science and Technology.

Pérez, A. 2007. Biodiversidad y Estructura Comunitaria de Poriferos
Asociados a Raices del Manglar, Rhizophora mangle, en el P. N.
Laguna de La Restinga, Nueva Esparta, Venezuela. [Biodiversity
and Community Structure of Porifers Associated to Mangrove
Roots, Rizophora mangle, in Laguna de La Restinga National Park,
Nueva Esparta, Venezuela.] Licenciate’s Thesis, Universidad Cen-
tral de Venezuela, Caracas, Venezuela.

Ramirez, I. 2002. Taxonomia de Esponjas (Porifera: Demospongiae) de
la Laguna de Bocaripo, Estado Sucre, e Islote Caribe, Dependencia
Federal, Venezuela. [Taxonomy of the Sponges (Porifera: Demo-
spongiae) of Bocaripo Lagoon, Estado Sucre, and Caribe Island,
Dependencia Federal, Venezuela.] Licenciate’s Thesis, Universidad
de Oriente, Cumana.

Rutzler, K., 1969. “The Mangrove Community: Aspects of Its Structure,
Faunistics, and Ecology.” In Memorias, Simposio Internacional
Lagunas Costeras, 1967, pp. 515-536. Mexico City: Universidad
Nacional Autonoma de México.

. 1995. Low-Tide Exposure of Sponges in a Caribbean Mangrove
Community. P.S.Z.N. I: Marine Ecology, 16:165-179.

Riutzler, K., M. C. Diaz, R. W. M. van Soest, S. Zea, K. P. Smith, B.
Alvarez, and J. Wulff. 2000. Diversity of Sponge Fauna in Man-
groves Ponds, Pelican Cays, Belize. Atoll Research Bulletin,
477:231-248.

Rutzler, K., and I. C. Feller. 1988. Mangrove Swamp Communities.
Oceanus, 30(4):16-24.

1996. Caribbean Mangrove Swamps. Scientific American,
274(3):94-99.

Riutzler, K., I. Goodbody, M. C. Diaz, I. C. Feller, and I. G. Macintyre.
2004. The Aquatic Environment of Twin Cays, Belize. Atoll Re-
search Bulletin, 512:1-49.

Rutzler, K., C. Piantoni, and M. C. Diaz. 2007. Lissodendoryx: Re-
discovered Type and New Tropical Western Atlantic Species (Po-
rifera: Demospongiae: Poecilosclerida: Coelospharidae). Journal
of the Marine Biological Association of the United Kingdom,
87:1491-1510.

Riitzler, K., and K. P. Smith. 1993. The Genus Terpios (Suberitidae)
and New Species in the “Lobiceps” Complex. Scientia Marina,
57(4):381-393.

Sutherland, J. P. 1980. Dynamics of the Epibenthic Community on Roots
of the Mangrove Rhizophora mangle, at Bahia de Buche, Venezu-
ela. Marine Biology, 58:75-84.

Taylor, P. R., M. M. Littler, and D. S. Littler. 1986. Escapes from Hervi-
bory in Relation to the Structure of Mangrove Island Macroalgal
Communities. Oecologia (Berlin), 69:481-490.

Thacker, R. W., M. C. Diaz, K. Ritzler, P. M. Erwin, S. J. A. Kimble,
M. J. Pierce, and S. L. Dillard. 2007. Phylogenetic Relationships
Among the Filamentous Cyanobacterial Symbionts of Caribbean
Sponges and a Comparison of Photosynthetic Production Between
Sponges Hosting Filamentous and Unicellular Cyanobacteria. Pro-
ceedings of the 7th International Sponge Symposium: Porifera Re-
search: Biodiversity, Innovation and Sustainability, pp. 621-626.
Rio de Janeiro, Brazil: Museu Nacional.

172 e

Toffart, J. L. 1983. Les Peuplements des racines de palétuviers en Gua-
deloupe (Antilles Frangaises). I. Analyse floristique et faunistique;
méthodologie et premier résultats. [Populations of Mangrove Roots
in Guadeloupe (French Antilles). I. Floristic and Faunistic Analysis;
Methodology and First Results.] Bulletin d’Ecologie, 14(4):227-239.

Van Soest, R.W. M. 1993. Affinities of the Marine Demosponge Fauna
of the Cape Verde Islands and Tropical West Africa. Courier, For-
schungsinstitut Senckenberg, 159:205-219.

Weinberg, S. 1981. A Comparison of Reef Survey Methods. Bijdragen
tot de Dierkunde, 5:199-218.

Wulff, J. L. 2000. Sponge Predators May Determine Differences in
Sponge Fauna Between Two Sets of Mangrove Cays, Belize Barrier
Reef. Atoll Research Bulletin, 477:250-263.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

. 2004. Sponges on Mangrove Roots, Twin Cays, Belize: Early
Stages of Mangrove Community Assembly. Atoll Research Bulletin,
519:1-10.

. 2005. Trade-Offs in Resistance to Competitors and Predators,
and Their Effects on the Diversity of Tropical Marine Sponges.
Journal of Animal Ecology, 74:313-321.

Zea, S. 1987. Esponjas del Caribe Colombiano. [Sponges of the Colom-
bian Caribbean.] Bogota, Columbia: Instituto de Investigaciones
Marinas y Costeras “José Benito Vives de Andréis.”

. 1993. Recruitment of Demosponges (Porifera, Demospongiae)

in Rocky and Coral Reef Habitats of Santa Marta, Colombian Ca-

ribbean. Marine Ecology, 14:1-21.

Internal Transcribed Spacer 2 (ITS2)

Variation in the Gorgonian Coral
Pseudopterogorgia bipinnata in Belize

and Panama

Daniel Dorado and Juan A. Sanchez

Daniel Dorado and Juan A. Sanchez, Departa-
mento de Ciencias Biologicas—Facultad de Cien-
cias, Laboratorio de Biologia Molecular Marina-
BIOMMAR, Universidad de los Andes, P.O. Box
4976, Bogota, Colombia. Corresponding author:
J. Sanchez (juansanc@uniandes.edu.co). Manuscript
received 9 June 2008; accepted 20 April 2009.

ABSTRACT. One of the most intriguing aspects of molecular evolution is the concerted
evolution of ribosomal genes, yet the presence of intragenomic rDNA variants is still not
well understood. We studied the intragenomic variation of the internal transcribed spacer
2 (ITS2, rDNA) in the gorgonian coral Pseudopterogorgia bipinnata (Gorgoniidae: Octo-
corallia) using a combined approach of denaturing gradient gel electrophoresis (DGGE),
DNA sequencing, and RNA secondary structure prediction. We examined intragenomic
variants of colonies from Carrie Bow Cay (Belize) and Bocas del Toro (Panama). Despite
frequent intragenomic ITS2 variation in P. bipinnata, predicted RNA secondary struc-
tures exhibited no signs of including pseudogenes and comprised functional copies. Given
the low divergence among the ITS2 sequences recovered from DGGE gels, intragenomic
variation was restricted to a few mutations that did not compromise the functionality of
the ITS2 secondary structure. The presence of common ITS2 intragenomic variants at
two distant populations raises new questions such as whether sharing similar copies can
be the product of gene flow. Regardless of the limited number of individuals analyzed in
this study, the method used here, excising bands from DGGE gels for further amplifica-
tion and sequencing, examined the reliability of the technique to separate intragenomic
variants with up to one nucleotide difference. Studying the intragenomic variation of
ITS2 has potential to provide us with information on recent population events such as
introgressive hybridization.

INTRODUCTION

Ribosomal DNA (rDNA) intragenomic variation has puzzled molecular sys-
tematists and ecologists during the past few years. The rDNA is a multigene
family arranged in tandem repeats, frequently achieving several hundreds of
repetitions per chromosome. Each repetition is composed of three ribosomal
subunits (18s, 5.8s, and 28s), separated by two internal transcribed spacers
(ITS1 and ITS2, or ITSs), an external transcribed spacer (ETS), and the non-
transcribed intergenic spacers, IGS. The ITS1 and ITS2 spacers form secondary
structures that are crucial for ribosomal maturation as well as important for
the maturation of the rRNA (Coté and Peculis, 2001). The ITSs are known to
have conserved core structures throughout the metazoans (see reviews in Cole-
man, 2003; Schultz et al., 2005). Changes in the ITS secondary structure are
known to produce inhibition of the maturation of rRNA as a consequence of

174 e

coevolution between RNA secondary structures and the
processing molecular machinery responsible for its re-
moval (Van Nues et al., 1995). As multicopy genes, the
rDNA is assumed to evolve via concerted evolution, re-
sulting in the homogenization of the sequences throughout
the genome (Harris and Crandall, 2000; Hillis and Davis,
1988), that is, homogenization of copies through unequal
crossing-over and gene conversion processes (Liao, 2000).
However, variations within individuals have been reported
primarily as a result of slow concerted evolution (Harris
and Crandall, 2000; Coté and Peculis, 2001), hybrid-
ization, or the presence of pseudogenes (Marquez et al.,
2003; Harpke and Peterson, 2006). The latter can appear
because of the presence of highly divergent rDNA types
in different chromosomes (Arnheim et al., 1980), which
retain ancestral rDNA polymorphisms for long periods of
time (Marquez et al., 2003). Hybridization phenomena
between species per se could increase the rDNA diversity
in an individual, but as an additional consequence could
result in silencing some rDNA loci by chromatin modifica-
tions in a nucleolar dominance process (Chen et al., 1998;
Frieman et al., 1999; Muir et al., 2001), which can drive
some rDNA loci by neutral selection toward pseudogenes
(Muir et al., 2001). However, the presence of ITS2 intrage-
nomic variants is a phenomenon that we do not clearly
understand.

Pseudopterogorgia bipinnata Pallas is one of the most
abundant shallow-water gorgonian corals in the Carib-
bean Sea (Bayer, 1961; Sanchez et al., 1997). This species
has two particular characteristics: it exhibits large pheno-
typic plasticity along the depth-wave exposure gradient,
and it presents clear intragenomic variation in the ITS2
sequence (Sanchez et al., 2007). Consequently, P. bipin-
nata constitutes an appropriate model species to study the
nature and genetics of ribosomal intragenomic variation.
In this study we had two main objectives: (1) to isolate
sequences of intragenomic ITS2 variants in P. bipinnata
from populations at Belize and Panama and (2) to exam-
ine if intragenomic ITS2 variants were functional copies
using predicted RNA secondary structures.

MATERIALS AND METHODS

Samples from Pseudopterogorgia bipinnata colonies
were obtained by scuba diving at Carrie Bow Cay (n = 27),
Belize, and Cristobal Island (n = 11), Bocas del Toro,
Panama. A few P. bipinnata from the Bahamas (San Sal-
vador) and Colombia (Bancos de Salmedina, Cartagena),

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

as well as a sequence of Gorgonia mariae, were chosen
as outgroups. However, there was no a priori informa-
tion on the genetic distance between western and eastern
Caribbean populations. Total DNA was extracted using a
cetyltrimethylammonium bromide (CTAB), proteinase K,
phenol-chloroform-isoamyl alcohol extraction method
(Coffroth et al., 1992); DNA was resuspended and con-
served in TE buffer at —70°C; DNA quality was checked
in agarose (1%) electrophoresis at 80 V for 30 min. Us-
ing the best DNA extraction quality, primers 5.8s 5’-
AGCATGTCTGTCTGAGTGTTGG-3’ and 28s 5’-GGG-
TAATCTTGCCTGATCTGAG-3’, designed by Aguilar and
Sanchez (2007), were used for the ITS2 amplification.
Conditions for polymerase chain reaction (PCR) were as
follows: an initial denaturizing step of 2 min at 94°C; fol-
lowed by 35 cycles of 30 s at 94°C, 30 s at 56.8°C, and 1
min at 72°C; and a final extension step of 2 min at 72°C;
using 1 unit Taq polymerase (Invitrogen), 3.5 mM MgCh,
0.2 mM deoxynucleoside triphosphates (dNTPs; Biorad
Mix), 0.15 wM primers (each), and 4 wL DNA (dilution
1/50) in 20 wL as the final volume. The amplification was
standardized with an efficiency of 95%. PCR reactions were
screened in denaturing gradient gel electrophoresis (DGGE)
containing 8% polyacrylamide, 1 < TAE buffer, and a lin-
ear urea-formamide denaturing gradient from 45% to 80%.
The gels were pre-run at 60°C and 90 V for 30 min, fol-
lowed by electrophoresis at 60°C and 90 V for 13 h. Gels
were stained with ethidium bromide during 15 min and vi-
sualized using a BIORAD Chemidoc system. DGGE sepa-
rates DNA fragments not only by the fragment size but also
by the DNA sequence, where GC-richer sequences migrate
further independently of small differences in size (Figure 1).
All reactions were conducted without a CG-clamp in the
primers, which is a 40 bp GC-rich sequence added before
the 5’-primer that adds an additional denaturing domain
allowing further migration of the DNA before denaturing
(LaJeunesse and Pinzon, 2007). In the case of gorgonian
corals, there was no need for the GC-clamp owing to the
great migration in the DGGE of gorgonian ITS2 sequences,
which avoided the problems involved with PCR reaction
tailed primers. Bands visualized in the DGGE gel were
excised using sterilized micropipette tips in the Bio-Rad
Chemidoc system and placed in 0.5 mL tubes with 100 pL
sterilized double distilled water. The tubes were incubated
in a shaker at room temperature for 24 h at 150 rpm. Each
band extract was collected in a 0.5 mL tube and the DNA
was precipitated with 300 wL cold absolute ethanol; tubes
were placed at —20°C for 24 h and then centrifuged at
13,000 rpm for 30 min; the supernatant was discarded,

eeeoeonknuqneuenun eK

FIGURE 1. Runs (2) of internal transcribed spacer 2 (ITS2) dena-
turing gradient gel electrophoresis (DGGE) banding patterns from
Pseudopterogorgia bipinnata colonies from Panama (Bocas del Toro;

stars) and Belize (Carrie Bow Cay; circles). Numbers correspond to
the sequence size when available. The gels have a common artifact
in the form of a “smile” (more accentuated in the upper gel), where
lateral wells tend to migrate slightly further because of the pressure
acting on the gel edges.

and the pellet was dried and resuspended in 15 pL steril-
ized double distilled water.

Reamplification of bands was conducted using PCR
as just described, using the same set of primers, except
that DNA was used without dilution and the annealing
temperature was raised a few degrees to increase specific-
ity. Purification of PCR products for sequencing was per-
formed by the Exo-SAP (Exonuclease 1 and shrimp alka-
line phosphatase) method using 1 unit Exonuclease, 0.2
units shrimp alkaline phosphatase, and 2 wL SAP buffer
10X per 20 pL in a 0.2 mL tube. Reactions were held at
37°C for 1 hand at 80°C for 15 min. Sequencing reactions
were performed with the BigDye 3.1 system according
to the manufacturer’s instructions (Applied Biosystems)
and sequenced in a capillary electrophoresis automated
sequencer (ABI310). Each sample was sequenced with
forward and reverse primers. The consensus sequences
were obtained by assembling the two complementary
electrophenograms in Sequencer 4.7 software.

NUMBER 38 e¢ 175

Secondary structures of all sequences were obtained
by reconstructing by comparison via Pairwise Alignment
(Bioedit) with previously reported structures in octo-
corals (Aguilar and Sanchez, 2007). The sequences were
then submitted with a few constraints and restrictions in
MFOLD at a default temperature of 37°C (Zuker, 2003).
Constraints force bases to be double stranded whereas
restrictions cause them to be single stranded, which are
chosen depending on the sequence homology between the
sequences with known structure against each problem
sequence without known structure. A good example for
a constraint are the two complementary sequences that
make a stem; an example of a restriction is a string of
free nucleotides between helices or any kind of loop. The
structure chosen was the one with the greater negative free
energy but conserving the ring model known for ITS2. The
obtained secondary structures were used to construct a
matrix for cladistic analysis as described by Aguilar and
Sanchez (2007). Phylogenetic analyses included maximum
parsimony and maximum likelihood as well a Bayesian in-
ference for a combined sequence-molecular morphometric
analysis (see details in Grajales et al., 2007).

RESULTS

Denaturing gradient gel electrophoresis (DGGE) anal-
ysis revealed that most individuals from Belize and Panama
contained intragenomic variants of ITS2 (see Figure 1).
There were as many as three different bands per individual
that were similar or nearly equal in length because of their
closeness in the DGGE gel (Figure 2). Some banding pat-
terns were identical for individuals from both Belize and
Panama, which indicated exact ITS2 copies, although some
patterns unique to each population were also observed (see
Figure 1). Great effort was made to obtain sequences from
most bands, but not all of them were successfully recov-
ered. The sequences had an average GC content of 55.6%,
which afforded the great migration of intragenomic ITS2
variants in the DGGE. The sequences from P. bipinnata
had more than 85.6% of sequence similarity, contrasting
with just 48% with respect to G. mariae.

Predicted secondary structures from all excised bands
exhibited functional structures with the conserved six heli-
coidal ring model previously reported for octocorals (Agui-
lar and Sanchez, 2007), but great variability was observed
in the length and complexity of each stem and spacer (Fig-
ures 2, 3). Intragenomic differences were frequently dis-
crete changes that did not affect the predicted secondary

176 °

_, ec

FIGURE 2. Two different intragenomic ITS2 variants from an in-
dividual colony of Pseudopterogorgia bipinnata from Belize. The
variants were excised from the two bands indicated by arrows in
the DGGE gel above, reamplified, and sequenced. The arrows be-
low show the differences between the two predicted RNA secondary
structures corresponding to one INDEL (insertion or deletion) only.

structures (see Figure 2). The ITS2 in P. bipinnata from
Panama and Belize varied from 212 to 224 nucleotides
(Figure 3). In general, the stems 2, 3, and 6 were shorter
than the stems 4 and 5, with stem 5 being the longest.
Multiple internal loops were frequent in stems 3, 4, and
5, with more nucleotides (nt) on stem 5, where up to six
internal loops were observed (Figure 3). The spacers were
often short, ranging from 1 to 4 nt. Spacer ‘11’ showed a
conserved sequence, UG, with little variation across indi-
viduals, while spacer ‘41’ was the longest, with 4 to 12 nt
and a conserved core sequence (AGUNCAGC) observed in
most of individuals (Figure 3). Phylogenetic results from
sequence and alignments or combined data sets, including
11 excised bands from individuals from Belize and 3 from
Panama, showed little divergence between Panama and Be-
lize despite the long distance with respect to a few individu-
als from Bahamas and Colombia (Figure 4). Very modest
support was found within individuals from Panama or Be-
lize, and no particular grouping could be discerned (data
not shown). In addition, no particular features of the ITS2
secondary structure as seen with helix 5, which showed the
largest number of characters, were supporting any particu-
lar clade or group of individuals (Figure 4).

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

DISCUSSION

The intragenomic ITS2 variation in Pseudopterogor-
gia bipinnata individuals involved functional copies, as
corroborated by reconstructing their predicted RNA sec-
ondary structures. Given the low divergence among the
ITS2 sequences recovered from DGGE gels, intragenomic
variation was restricted to a few mutations that did not
compromise the functionality of the ITS2 secondary struc-
ture. Despite frequent intragenomic ITS2 variation in P. bi-
pinnata, predicted RNA secondary structures exhibited no
signs of including pseudogenes or structural degeneration.
Having in mind that the ITS2 secondary structure has a
major role in the maturation of the ribosomal RNA (Coté
and Peculis, 2001), little tolerance of changes is expected
as a result of the restrictions imposed by the ITS2 splicing
machinery (Van Nues et al., 1995; Coleman, 2003); this
means purifying selection is acting on secondary structural
constraints (Coté and Peculis, 2001) or concerted evolu-
tion mechanisms are acting similarly (Liao, 2000; but
see Nei and Rooney, 2005; Harpke and Peterson, 2006).
Similarly, compensatory base changes (CBC), occurring at
the stem regions, are very unlikely to occur at the intra-
specific level (Miller et al., 2007). Thus, it is expected that
only variants or alleles carrying only minor changes occur,
which was evident with the functionality of co-occurring
secondary structures found at the intraspecific level.

ITS intragenomic variation has been also observed
in scleractinian corals. Van Oppen et al. (2001) exam-
ined diverse nuclear and mitochondrial DNA sequences,
concluding that paraphyly from intragenomic ITS copies
could be explained by extensive introgressive hybridization
and reticulate evolution. Similarly, Marquez et al. (2003)
found the presence of ribosomal pseudogenes as a pos-
sible consequence of multiple hybridization events. How-
ever, Vollmer and Palumbi (2004) examined the multiple
copies of the Caribbean Acropora species and concluded
that there is no proper way to evaluate if the intragenomic
shared variation of genes such as ITS1 and ITS2 was the
result of incomplete lineage sorting or recent hybridiza-
tion processes. Nonetheless, all the studies mentioned
studied the intragenomic variation of ITS using the DGGE
technique, and it is clear that traditional cloning methods
overestimate the intragenomic diversity (LaJeunesse and
Pinzon, 2007).

Regardless of the limited number of individuals ana-
lyzed in this study, the method used here, excising bands
from DGGE gels for further amplification and sequenc-
ing, probed its reliability to separate intragenomic variants
up to one nucleotide difference (see Figure 2). DGGE is a

P. bipinnate Pb7 B1

.e,

P. bipinnata Phip6

Belize, shat.
dG -66.19

Belize,sha §

dG -66.16 8 Belize, deep

dG -62.26

cr Wy
Bahamas [
dG-5705 H

Colombia
dG -70.9

NUMBER 38 ¢ 177

P. bipinnata Pb25 B

14,
e, e J a
$
Y e
i

g

dG -62.88

FIGURE 3. Predicted ITS2 RNA secondary structures in Pseudopterogorgia (P.) bipinnata. The upper
right structure shows the characters for the molecular morphometrics analysis used in the combined Bayes-
ian inference analysis. The number within the ring structure refers to the total number of nucleotides at

each structure.

method useful to detect the most prevalent intragenomic
variants of ribosomal genes, whereas traditional methods
to screen intragenomic variation such as cloning miscalcu-
late the codominance of the different copies (LaJeunesse
and Pinzon, 2007; Thornhill et al., 2007). There are two
main approaches for depicting the nature of intragenomic
ITS2 variants in octocorals. One method is to study in de-
tail the genetics of the different ITS2 variants by crossing
individuals with different intragenomic patterns, which
can provide inheritance information and linkage disequi-
librium configurations. An alternative method includes
techniques such as reverse transcription (RT)-PCR and
quantitative real-time PCR, which can offer more accu-
rate information on the functionality of the different in-
tragenomic ITS2 copies. The RT-PCR technique can filter
copies that are not expressed in the cell, and quantitative

PCR can quantify the amount of ITS transcripts from each
particular copy. These methods could also test if the inten-
sity of bands in DGGE actually corresponds to the number
of copies of a particular intragenomic variant.

ACKNOWLEDGMENTS

This study was partially funded by Facultad de Cien-
cias [Department of Biological Sciences], Universidad
de los Andes, COLCIENCIAS (Grant 120409-16825;
funding to J. A. Sanchez); a Smithsonian postdoctoral
fellowship (NMNH); the MSN Invertebrate Workshop
(2003) at Bocas Research Station, Bocas del Toro, Pan-
ama (STRI); and the Smithsonian Marine Science Net-
work. We are grateful to Rachel Collin, Gabriel Jacome,
Howard Lasker, Klaus Ruetzler, Michael Lang, Stephen

178 e

Gorgonia marizo

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Colombia

FIGURE 4. Maximum-likelihood phylograms show above-node support from the combined sequence-
molecular morphometrics Bayesian analysis (left) and maximum-parsimony bootstrapping (1000 replicates,
right): Pseudopterogorgia bipinnata colonies from Panama (Bocas del Toro; stars) and Belize (Carrie Bow
Cay; circles). The tree at the right is a radial representation of a set of terminal branches corresponding to

Panama and Belize sequences pruned from the left tree.

Cairns, BIOMMAR colleagues, and the Smithsonian
Station at Carrie Bow Cay, Belize. The Minister of En-
vironment, Household and Territorial Development of
Colombia granted access to genetic resources to JAS for
the DNA analyses included in this paper (Contract 007,
resolution 634; 14 March 2007). This work is contribu-
tion number 841 of the Caribbean Coral Reef Ecosystems
Program (CCRE), Smithsonian Institution, supported in
part by the Hunterdon Oceanographic Research Fund.

LITERATURE CITED

Aguilar, C., and J. A. Sanchez. 2007. Phylogenetic Hypothesis of Gorgoniid
Octocorals According to ITS2 and Their Predicted RNA Secondary
Structures. Molecular Phylogenetics and Evolution, 43:774-786.

Arnheim, N., M. Krystal, R. Shmickel, G. Wilson, and O. Ryder. 1980.
Molecular Evidence for Genetic Exchanges Among Ribosomal
Genes on Nonhomologous Chromosomes in Man and Apes. Pro-
ceedings of the National Academy of Sciences of the United States
of America, 77:7323-7327.

Bayer, F. M. 1961. The Shallow Water Octocorallia of the West Indian
Region. Studies of the Fauna of Curagao, 12:1-373.

Chen, Z. J., L. Comai, and C. S. Pikaard. 1998. Gene Dosage and Sto-
chastic Effects Determine the Severity and Direction of Uniparental
Ribosomal RNA Gene Silencing (Nucleolar Dominance) in Arabi-
dopsis Allopolyploids. Proceedings of the National Academy of Sci-
ences of the United States of America, 95:14891-14896.

Coffroth, M. A., H. R. Lasker, M. E. Diamond, J. A. Bruenn, and
E. Bermingham. 1992. DNA Fingerprints of a Gorgonian Coral:

A Method for Detecting Clonal Structure in a Vegetative Species.
Marine Biology, 114:317-325.

Coleman, A. W. 2003. ITS2 Is a Double-Edged Tool for Eukaryote Evo-
lutionary Comparisons. Trends in Genetics, 19:370-375.

Coté, C. A., and B. A. Peculis. 2001. Role of the ITS2-Proximal Stem and
Evidence for Indirect Recognition of Processing Sites in Pre-rRNA
Processing in Yeast. Nucleic Acids Research, 29:2106-2116.

Frieman, M., Z. J. Chen, J. S. Vasquez, L. A. Shen, and C. S. Pikaard.
1999. RNA Polymerase I Transcription in a Brassica Interspecific
Hybrid and Its Progenitors: Test of Transcription Factor Involve-
ment in Nuclear Dominance. Genetics, 152:451-460.

Grajales, A., C. Aguilar, and J. A. Sanchez. 2007. Phylogenetic Recon-
struction Using Secondary Structures of Internal Transcribed Spacer
2 (ITS2, rDNA): Finding the Molecular and Morphological Gap in
Caribbean Gorgonian Corals. BMC Evolutionary Biology, 7:90.

Harpke, D., and A. Peterson. 2006. Non-Concerted ITS Evolution in
Mammillaria (Cactaceae) Molecular Phylogenetics and Evolution,
41:573-593.

Harris, D. J., and K. A. Crandall. 2000. Intragenomic Variation Within
ITS1 and ITS2 of Freshwater Crayfishes (Decapoda: Cambaridae):
Implications for Phylogenetic and Microsatellite Studies. Molecular
Biology and Evolution, 17:284-291.

Hillis, D. M., and S. K. Davis. 1988. Ribosomal DNA: Interspecific Poly-
morphism, Concerted Evolution, and Phylogeny Reconstruction.
Systematic Zoology, 37:63-66.

LaJeunesse, T. C., and J. H. Pinzon. 2007. Screening Intragenomic rDNA
for Dominant Variants Can Provide a Consistent Retrieval of Evo-
lutionarily Persistent ITS (rDNA) Sequences. Molecular Phyloge-
netics and Evolution, 45:417-422.

Liao, D. 2000. Gene Conversion Drives Within Genic Sequences: Con-
verted Evolution of Ribosomal RNA Genes in Bacteria and Ar-
chaea. Journal of Molecular Evolution, 51:305-317.

Marquez, L. M., D. J. Miller, J. B. Mackenzie, and M. J. Van Oppen.
2003. Pseudogenes Contribute to the Extreme Diversity of Nuclear

Ribosomal DNA in the Hard Coral Acropora. Molecular Biology
and Evolution, 20:1077-1086.

Muir, G., C. C. Fleming, and C. Schlotterer. 2001. Three Divergent
rDNA Clusters Predate the Species Divergence in Quercus petrea
(Matt.) Liebl. and Quercus robur L. Molecular Biology and Evolu-
tion, 18:112-119.

Miller, T., N. Philippi, T. Dandekar, J. Schultz, and M. Wolf. 2007. Dis-
tinguishing Species. RNA, 13:1469-1472.

Nei, M., and A. P. Rooney. 2005. Concerted and Birth-and-Death Evolu-
tion of Multigene Families. Annual Review in Genetics, 39:121-152.

Sanchez, J. A., C. Aguilar, D. Dorado, and N. Manrique. 2007. Pheno-
typic Plasticity and Morphological Integration in a Marine Modu-
lar Invertebrate. BMC Evolutionary Biology, 7:122.

Sanchez, J. A., S. Zea, and J. M. Diaz. 1997. Gorgonian Communities
of Two Contrasting Environments from Oceanic Caribbean Atolls.
Bulletin of Marine Science, 61:61-72.

Schultz, J., S. Maisel, D. Gerlach, T. Miller, and M. Wolf. 2005. A Com-
mon Core of Secondary Structure of the Internal Transcribed Spacer
2 (ITS2) Throughout the Eukaryota. RNA, 11:361-364.

NUMBER 38 e¢ 179

Thornhill, D. J., T. C. LaJeunesse, and S. R. Santos. 2007. Measuring
rDNA Diversity in Eukaryotic Microbial Systems: How Intra-
genomic Variation, Pseudogenes, and PCR Artifacts Confound Bio-
diversity Estimates. Molecular Ecology, 16(24):5326-5340.

Van Nues, R. W., M. J. Rientjes, S. A. Morre, E. Molee, R. J. Planta,
J. Venema, and A. H. Raue. 1995. Evolutionary Conserved Ele-
ments Are Critical for Processing of Internal Transcribed Spacer 2
from Saccharomyces cerevisiae Precursor Ribosomal RNA. Journal
of Molecular Evolution, 250:24-36.

Van Oppen, M. J., B. J. McDonald, B. L. Willis, and D. J. Miller. 2001.
The Evolutionary History of the Coral Genus Acropora (Sclerac-
tinea, Cnidaria) Based on a Mitochondrial and Nuclear Marker:
Reticulation, Incomplete Lineage Sorting, or Morphological Con-
vergence? Molecular Biology and Evolution, 18:1315-1329.

Vollmer, S. V., and S. R. Palumbi. 2004. Testing the Utility of Internally
Transcribed Spacer Sequences in Coral Phylogenetics. Molecular
Ecology, 13:2763-2772.

Zuker, M. 2003. Mfold Web Server for Nucleic Acid Folding and Hy-
bridization Prediction. Nucleic Acids Research, 31:3406-3415.

_

Obvious Invaders and Overlooked Infauna:
Unexpected Constituents of the Decapod
Crustacean Fauna at Twin Cays, Belize

Darryl L. Felder, Peter C. Dworschak, Rafael
Robles, Heather D. Bracken, Amanda M. Windsor,
Jennifer M. Felder, and Rafael Lemaitre

Darryl L. Felder, Rafael Robles, Heather D.
Bracken, Amanda M. Windsor, and Jennifer M.
Felder, Department of Biology and Laboratory
for Crustacean Research, University of Louisiana
at Lafayette, Lafayette, Louisiana 70504-2451,
USA. Peter C. Dworschak, Naturhistorisches
Museum in Wien, Burgring 7, A-1014 Wien, Aus-
tria. Rafael Lemaitre, Department of Invertebrate
Zoology, National Museum of Natural History,
MRC 163, Smithsonian Institution, Washington,
D.C. 20013-7012, USA. Corresponding author:
D. Felder (dlf4517@louisiana.edu). Manuscript
received 13 May 2008; accepted 20 April 2009.

ABSTRACT. Decapod crustaceans in the vicinity of Carrie Bow Cay and Twin Cays,
Belize, have been under study for more than 25 years. Large collections have been as-
sembled, and new species have been discovered. The effort has included photographic
documentation of coloration, yielding characters of value in identification of problematic
tropical taxa. Measurements of diversity have been markedly enhanced by extraction
corer (yabby pump) sampling in shallow subtidal sediments, especially at Twin Cays.
This technique revealed species, genera, and families of thalassinidean decapods not pre-
viously known from the region. Studies continue on the ecological roles of these burrow-
ers, dominant bioturbators in seagrass beds where they produce conspicuous mounds
of sediment and constitute a major infaunal biomass at Twin Cays. By contrast, famil-
iar large reptant decapods typically dominate shallow rocky substrates. Within the past
four years, however, the nonindigenous portunid crab Charybdis hellerii has extensively
invaded large portions of hard substrates at Twin Cays. In 2007, it was found to domi-
nate cavities under coral heads in survey areas along the northeastern and southwestern
shorelines, possibly displacing populations of large Mithrax, Menippe, Callinectes, and
Panulirus previously found there in abundance.

INTRODUCTION

Fieldwork centered on Carrie Bow Cay and surrounding habitats, including
a variety of settings at Twin Cays. The effort continues work by the first author
in collaboration with the late Ray Manning in 1983, as well as work by the late
Brian Kensley during the 1980s and early 1990s (Kensley, 1981, 1996). Early
efforts produced abundant grass-bed and reef-crest species generally identifiable
with known Caribbean taxa, along with small cryptic forms obtained by cutting
open sponges, breaking rubble, poisoning in situ, or using several narcotants to
drive out small decapods from rubble isolated in containers. Rich collections
that have accumulated in the the Smithsonian Institution’s National Museum of
Natural History were fixed in formalin, limiting their value in genetic analyses.
Efforts in 2002 and 2007 shifted emphasis to varied intertidal and subtidal habi-
tats of Twin Cays and to resampling the regional fauna to obtain alcohol-fixed
materials for molecular genetic analyses.

182 e

Concerted effort has been made to photographically
document coloration of fresh specimens, given the value
of color in the identifications of tropical species and the
long-term goal of producing a guidebook for the regional
decapod fauna (DLF and RL, in progress). More than 260
decapod species have been enumerated in our collections
from the Carrie Bow Cay region, some yet to be named.
Under U.S. National Science Foundation AToL “Deca-
pod Tree of Life” support, molecular and morphological
systematic studies are under way concerning alpheid and
other caridean shrimps, paguroid hermit crabs, thalassini-
dean shrimps, and panopeid, portunid, grapsoid, pinnoth-
erid, and majoid crabs, as well as family-level relationships
among all major decapod groups. Work incorporating
porcelain crab collections from the region has been pub-
lished (Rodriguez et al., 2005, 2006) as has work by other
investigators on some alpheid shrimp groups (Duffy, 1996;
Duffy and Macdonald, 1999; Duffy et al., 2000, 2002;
Macdonald et al., 2006; Rios and Duffy, 2007). Previous
collections of upogebiid thalassinidean shrimp from Be-
lize were included in Williams (1993). Several descriptions
of new species from our Belize collections have also ap-
peared (Goy and Felder, 1988; Manning and Felder, 1996;
Felder and Manning, 1997), but many species remain to
be described. The second author has been involved in
several ecological studies of the infaunal decapods of the
region (Dworschak and Ott, 1993; Abed-Navandi and
Dworschak, 2005; Dworschak et al., 2006).

Our protracted field sampling program has in some
cases allowed us to observe apparent changes in com-
munity composition. In a striking example, shallow sub-
tidal habitats at Twin Cays have been recently invaded
by the nonindigenous swimming crab Charybdis hellerii
(A. Milne-Edwards, 1867), previously unreported from
Belize. Recurrent trips have also provided opportunities
for shallow subtidal sampling and burrow-casting of fos-
sorial infauna in turtle grass (Thalassia) beds along shore-
lines of Twin Cays, revealing unexpected thalassinidean
diversity. A brief account of these latest efforts is our pres-
ent focus, preliminary to more comprehensive treatment
of the full decapod assemblage.

MATERIALS AND METHODS

Sampling included the breaking of dead coral and
conch shell rubble, netting, extraction of sediments, and
sorting through hard-surface fouling organisms, but sam-
pling of large crabs such as Charybdis hellerii (Brachyura)

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

and its macrocrustacean associates was a targeted effort.
These decapods were captured from under pieces of dead
subtidal coral and debris that were lifted while snorkeling
over and adjacent to seagrass (Thalassia) beds in water 1-2
m deep. Sampling of most thalassinideans and related deca-
pod burrowers was accomplished with a suction extrac-
tor (yabby pump) and bag-sieve while wading, snorkeling,
or SCUBA diving in water 0.5—4 m deep. In addition to
collections of Glypturus acanthochirus (Callianassidae)
by suction extractor, some specimens of this species were
obtained with “weighted line” traps (de Vaugelas, 1985).
Specimens of Axiopsis serratifrons (Axiidae) were ob-
tained by baiting animals to the apertures of their burrows,
where they were captured by cutting off the burrow or by
spearing the specimens. Casts of burrows were made as
described by Dworschak and Ott (1993). Specimens were
immobilized by immersion in chilled seawater or by nar-
cotization with clove oil before photography. Photographs
of specimens immersed in a pan of seawater were made
with a Fuji Fine Pix $1Pro digital camera equipped with a
60 mm macrolens while the subject was lighted by a com-
bination of direct and reflected sunlight or high-intensity
5000°K fluorescent photographic lamps. All specimens
were subsequently preserved in several exchanges of 95%
nondenatured ethanol and then stored in 75% nondena-
tured ethanol. Photographic voucher specimens were ar-
chived in the Zoological Collections of the University of
Louisiana at Lafayette (ULLZ), and most other materials
were deposited in the Smithsonian Institution—National
Museum of Natural History (USNM). Some collections
by the second author (especially thalassinideans) were de-
posited in the Naturhistorisches Museum in Wien, Austria
(NHMW) and the Muséum National d’Histoire Naturelle,
Paris, France (MNHN). For figured specimens, size is indi-
cated as carapace width (cw) or carapace length (cl).

RESULTS AND DISCUSSION
INVASION BY CHARYBDIS HELLERI!

Large bottom debris (waterlogged wood, discarded
building materials, dead coral heads) typically provides
cover for large reptants such as spiny lobsters (Panulirus
spp.), swimming crabs (Callinectes spp.), stone crabs (Me-
nippe spp.), and large spider crabs (Mithrax spp.), espe-
cially in shallow well-lighted waters. Sampling of these
environments at both Carrie Bow Cay and Twin Cays in
October 2002 revealed no large decapods other than these
genera. That same year, however, a small specimen of the

nonindigenous portunid crab Charybdis hellerii was found
in an empty conch shell on the inshore side of Carrie Bow
Cay, the first such occurrence recorded in our sampling
program.

In April 2007, sampling under large pieces of cover at
Twin Cays was undertaken to obtain fresh materials of the
aforementioned resident genera for genetic analyses. Initial
sampling centered in the vicinity of the “Fisheries Camp”
on the southeastern end of Twin Cays, where a storm had
scattered sheets of metal building siding from the shore-
line to depths of nearly 2 m. Inspections beneath 20 such
sheets across this entire range of depths revealed none of
the target species but at least seven variously sized indi-
viduals of the nonindigenous swimming crab C. hellerii.

NUMBER 38 e¢ 183

Sampling was thereafter shifted to dead coral heads
scattered among turtle grass beds on the northeast side of
Twin Cays. A crude survey was there undertaken for coral
heads in 1-1.5 m depths, each head roughly 0.5-0.7 m in
diameter and separated from one another by roughly 6-15
m. Of the 25 coral heads inspected, 13 were uninhabited
by large reptant decapods, 8 harbored large specimens of
C. hellerii (Figure 1b), and four harbored only Menippe
nodifrons Stimpson, 1859 (Figure 1a). Large single indi-
viduals of C. hellerii were found under 6 of the 25 heads
that were lifted, a mating pair of C. hellerii was found
under a single head, and a specimen of C. hellerii together
with a large specimen of M. nodifrons was found under an-
other head. No specimens of Mithrax spp., Callinectes spp.,

FIGURE 1. a, Stone crab Menippe nodifrons, male, 69.7 mm carapace width (cw), Twin Cays, Belize 10
April 2007, ULLZ 8991. b, Invasive Indo-Pacific swimming crab Charybdis hellerii, male, 75.3 mm cw, Twin
Cays, Belize, 10 April 2007, ULLZ $990. c, Callianassid Eucalliax sp., female, 8.3 mm carapace length (cl),
South Water Cay, 22 October 2002, ULLZ 9230. d, Polyester resin burrow cast from Twin Cays, prob-
ably assignable to Axianassa australis, cast length 85 cm, made by PCD, Twin Cays, Belize, August 1989,
NHMW 24001. e, Laomediid Naushonia sp. female, 5.8 mm cl, Carrie Bow Cay, Belize, 3 April 2007, ULLZ
8895. ULLZ, University of Louisiana at Lafayette; NHMW, Naturhistorisches Museum Wien. Photographs

a—c and e by DLF; photograph d by PCD.

184 e

or Panulirus spp. were observed, despite these taxa being
commonly found in such settings during 1983 and 2002.

Small or immature specimens of Charybdis hellerii
are easily confused with Cronius ruber (Lamarck, 1818)
and to a lesser extent with Achelous tumidulus Stimpson,
1871, both of which also occur in Belize and adjacent wa-
ters of the Caribbean, Gulf of Mexico, and other areas
of the warm temperate Atlantic. This similarity led us to
initially question the identity of the single small specimen
collected in 2002, but it was confirmed to be Charybdis
hellerii by 16S mtDNA sequence analysis by comparing to
other sequence data for the species (Robles et al., 2007;
Mantelatto et al., 2009). Widely used diagnostic morpho-
logical characters that apply well to full-sized adults do
not readily facilitate identification of juveniles among these
three species, and records of subadults could easily be in
error if based on presently limited descriptions. At the very
least, A. tumidulus differs from both C. ruber and Charyb-
dis hellerii by lacking a striking posterior or posterodistal
meral spine on the fifth pereiopod (swimming leg) in all
crab stages. Cronius ruber and Charybdis hellerii, how-
ever, share a strongly spined fifth pereiopod, albeit with the
spine usually occupying a relatively more distal position
and being less posteriorly directed on the merus of Cronius
ruber. The relative position of the spine is, however, diffi-
cult to distinguish in small juveniles. These two species also
share the presence of small spinules bordering the posterior
margin of the fifth pereiopod propodus, although these spi-
nules are of relatively larger size in Charybdis hellerti. This
characteristic is readily evident in adults, where setation
obscures small acute granules along the margins of the pro-
podus in Cronius ruber, which are unlikely to be confused
with the well-formed adult spinules of Charybdis hellerii
(see Figure 1b). In juveniles of Cronius ruber, microspina-
tion of this propodal margin is relatively stronger than in
adults, and distinction from juveniles of Charybdis hellerii
is somewhat subjective, especially if one lacks comparative
specimens of similar size. No feature in the carapace of
early crab stages (Dineen et al., 2001: fig. 24) appears to
separate small individuals of these species.

Recent observations have revealed an ongoing inva-
sion of C. hellerii into coastal western Atlantic locations,
and its documented distribution must now include Belize
along with Brazil, Venezuela, Colombia, Cuba, the Yu-
catan shelf of Mexico, both coasts of Florida, and other
northern Atlantic U.S. coastal habitats through at least
the Carolinas (Campos and Tirkay, 1989; Gomez and
Martinez-Iglesias, 1990; Hernandez and Bolafios, 1995;
Lemaitre, 1995; Calado, 1996; Mantelatto and Dias,
1999; Dineen et al., 2001; Mantelatto and Garcia, 2001;

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Mantelatto et al., 2007; Robles et al., 2007; McMillen-
Jackson, 2008; Felder et al., 2009). Clearly, the foregoing
chronology of reports reveals continued western Atlantic
range expansion for C. hellerii, although the potential
trophic impacts of this invader remain poorly docu-
mented. The first author has on two occasions observed
individuals of C. hellerii in the Indian River Lagoon, Flor-
ida, feeding (inside shallow cavities of hard substrates
that they occupied) on soft-shelled, postmolt individu-
als of native species of large decapods (one Callinectes,
one Panulirus), and in another instance feeding on small
mussels. As in the present report, all such observations
and inferences of this invader’s potential competitive and
predatory impacts in the western Atlantic remain very
limited and anecdotal, but they serve to justify a call for
controlled experimental studies.

THALASSINIDEANS

Our collections of cryptic burrowing thalassinideans
from various habitats in the vicinity of Carrie Bow and Twin
Cays, along with the few previously reported records, include
at least 17 species representing the families Callianassidae,
Laomediidae, Thomassiniidae, Axianassidae, Axiidae, and
Upogebiidae. The species of these often overlooked groups
are presented in the following list, with collection sites indi-
cated as TC (Twin Cays), CB (Carrie Bow Cay), SW (South
Water Cay), and SL (shorelines near Dangriga); catalogue
numbers are shown for archived specimens.

INFRAORDER THALASSINIDEA SENSU LATO
CALLIANASSIDAE (Ghost Shrimps)
Corallianassa longiventris (A. Milne-Edwards, 1870)—TC,
CB: NHMW 6774, 6775, 15352-15355; ULLZ 4228-

4230, 6083, 8997.

Eucalliax sp.—TC, SW: ULLZ 9230.

Glypturus acanthochirus Stimpson, 1866—TC, CB: NHMW
6765-6770, 15338-15342; MNHN Th 1181, Th 1185;
ULLZ 8993-8995, 9233; USNM 266241-266244.

Lepidophthalmus richardi Felder and Manning, 1997—
SL [near river mouths]: NHMW 15343-15349: ULLZ
3577, 5186-5188, 8992; USNM 277777-277779.

Neocallichirus grandimana (Gibbes, 1850)—TC, CB,
SW: NHMW 6796-6799, 15356-15367; MNHN Th

1182-1184; ULLZ 8998, 9235-9237, 9239-9241,
9243, 9244.

Neocallichirus maryae Karasawa, 2004—TC: ULLZ
9234, 9238.

LAOMEDIIDAE
Naushonia sp.—CB: ULLZ 8895, 8915.
AXIANASSIDAE

Axianassa australis Rodrigues and Shimizu, 1992—TC
[identified by burrow cast]: NHMW 24001.

‘THOMASSINIIDAE

Mictaxius thalassicola Kensley and Heard, 1991—TC:
ULLZ 9246.

UpoGEBriDaE (Mud Shrimps)
Pomatogebia operculata (Schmitt, 1924—CB: ULLZ 9231.
Upogebia acanthura (Coélho, 1973)—?CB: USNM 251246.

Upogebia omissa Gomes Corréa, 1968—TC, SL: ULLZ
5165.

Upogebia sp.—CB: ULLZ 9232.
AXtIDAE (Lobster Shrimps)

Axiopsis serratifrons (A. Milne-Edwards, 1873)—CB:
NHMW 6771-6773, 15350-15351: ULLZ 4232, 4233,
5827, 8996; USNM 18905, 18907, 18908.

Coralaxius nodulosus (Meinert, 1877)—CB: USNM
170856, 171764-171766, 243431-243434.

Paraxiopsis hispida Kensley, 1996—CB: USNM 211462.

Paraxiopsis spinipleura Kensley, 1996—CB: USNM
211451.

Sediments in lower intertidal to subtidal seagrass beds
at Twin Cays are densely populated by Neocallichirus
grandimana, Glypturus acanthochirus, Corallianassa lon-
giventris, Neocallichirus maryae, Mictaxius thalassicola,
and Eucalliax sp., often burrowing more than 1 m into

NUMBER 38 e¢ 185

sediments. Dworschak and Ott (1993) previously analyzed
burrow morphologies and distributions for three of these
species, as well as for Axiopsis serratifrons and two spe-
cies of pistol shrimp. Their food sources were investigated
by stable isotope studies (Abed-Navandi and Dworschak,
2005). Among the species from Twin Cays, M. thalassi-
cola has not previously been reported from the northern
Caribbean, and Eucalliax sp. (Figure 1c) represents an un-
described taxon presently known only from Belize.

The newly reported Neocallichirus maryae is a replace-
ment name for the more familiar N. rathbunae (Schmitt,
1935), which proved to be a junior primary homonym
of a different fossil species (Karasawa, 2004). Although
Sakai (2005) placed N. raymanningi Blanco Rambla and
Lemaitre, 1999, in synonymy with N. rathbunae (Schmitt,
1935), and N. raymanningi would thus predate recent es-
tablishment of N. maryae, we do not accept the presently
limited evidence for this synonymy.

Ejecta from burrows of these thalassinideans domi-
nates bottom topography in intertidal to shallow subtidal
seagrass beds of this area, but along intertidal muddy shore-
lines at Twin Cays, especially those immediately adjacent
to mangroves, it appears that the axianassid Axianassa
australis also occurs. As no specimens have been captured,
this can be deduced only from highly characteristic ejecta
patterns and spiraled burrow casts (see Dworschak and
Rodrigues, 1997; Felder, 2001), the latter obtained by the
second author in 1989 (Figure 1d).

Neocallichirus grandimana appears to be the most
widely distributed callianassid among sites sampled in the
vicinity, inhabiting both vegetated and nonvegetated sedi-
ments. Together with Eucalliax sp., it densely populates
sparsely vegetated calcareous sands of shallow shoals bor-
dering South Water Cay in addition to sites at Twin Cays.
At South Water Cay, upper reaches of its burrows are com-
monly inhabited by Processa sp. and early juvenile stages
of Callinectes sp., the latter being uniquely pigmented an
opaque bluish-black. Glypturus acanthochirus and Coral-
lianassa longiventris range into deeper grass beds, where
they appear to draw grass blades into their burrows. Dis-
tributions of all the collected thalassinideans depend on
sediment characteristics, depths, vegetation, and water
quality, whereas characteristic burrow architectures are
both diagnostic of species and suggestive of ecological ad-
aptations (Dworschak and Ott, 1993, Abed-Navandi and
Dworschak, 2005; Dworschak et al., 2006). Less conspicu-
ous evidence of sediment ejecta characterizes areas among
seagrasses that are burrowed primarily by nonthalassinidean
decapods such as the Alpheus spp. reported by Dworschak
and Ott (1993). Surface features of these burrows can be all

186 e

but indistinguishable from those made by what appear to
be several species of Upogebia, including U. omissa.

The assemblage of upogebiids in the Carrie Bow Cay
region remains poorly understood. It appears that Upoge-
bia omissa ranges widely here, from the shoreline along the
mainland to offshore cays, and the first author has identi-
fied specimens taken as “pests” from commercial penaeid
shrimp farms on the mainland. General treatment of west-
ern Atlantic upogebiids by Williams (1993) included re-
cords of U. acanthura from a patch reef southwest of Car-
rie Bow Cay and U. brasiliensis Holthuis, 1956 from more
distant shoreline areas of Belize, although our collections
have produced no additional specimens. Two other spe-
cies listed by Williams (1993) from nearby coastal envi-
ronments of Quintana Roo (U. corallifora Williams and
Scott, 1989 and U. vasquezi Ngoc-Ho, 1989) could also
be expected in Belize, although we have yet to find them.
Specimens of this genus from coralline rubble just off the
reef crest at Carrie Bow Cay (ULLZ 9232) and other un-
catalogued specimens from Twin Cays (in areas also bur-
rowed by alpheid shrimp) cannot confidently be assigned
to known species and warrant further study. Generally
found in deeper subtidal habitats (Felder et al., in press),
the upogebiid Pomatogebia operculata ranges into waters
as shallow as 2 m depth off Carrie Bow Cay and likely oc-
curs elsewhere between cays in appropriate deeper calcare-
ous rubble habitats; these have been collected by breaking
open highly eroded pieces of coralline rubble to expose the
muddy interstices and cavities occupied by this upogebiid.

Axiids are also found in association with rubble and
reef structures of outer cays, as, for example, at Carrie
Bow. The widely distributed Coralaxius nodulosus, a
small-sized species inhabiting cavities in subtidal coralline
rubble from the fore-reef (see also Kensley, 1994), is rou-
tinely found along with the upogebiid Pomatogebia oper-
culata in interstices of broken rubble retrieved from depths
greater than 2 m. By contrast, the large and strongly armed
Axiopsis serratifrons is widely distributed between pieces
of coarse coral rubble in back-reef flats of Carrie Bow
(0.5-2 m depths), there positioned to ambush prey from
its somewhat concealed burrow aperture. In addition, two
new species of Paraxiopsis described by Kensley (1996)
both range into reef habitats of Carrie Bow Cay. Although
P. spinipleura was originally found there in shallow (1.5
m) back-reef rubble, we have not encountered additional
specimens. We have also not found additional materials
of P. hispidus, previously collected at the reef drop-off in
depths greater than 20 m.

A remarkable thalassinidean find at Carrie Bow was
the April 2007 discovery of a laomediid assignable to

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Naushonia sp. (Figure le). Two specimens were cap-
tured, both from cavities of empty conch shells in shal-
low (<1.5 m) subtidal waters. These individuals appear
to also represent an undescribed species of a rarely en-
countered genus in the northern Caribbean region. To
date known only from Carrie Bow Cay, they are cur-
rently being described.

The thalassinidean fauna of the general region also
includes an abundant nearshore species, Lepidophthalmus
richardi, adapted to euryhaline waters and muddy sand
shorelines of the Stann Creek District (Felder and Man-
ning, 1997). This species has not been found in habitats
immediately associated with Twin Cays or Carrie Bow
Cay, despite intensive search.

These collections have allowed us to update and ex-
pand the burrow distribution schemes for Belize given by
Dworschak and Ott (1993). We herewith add additional
taxa and habitat distributions (Figure 2) to underscore the
overlooked diversity of infaunal macrocrustaceans, some
of which are dominant bioturbators.

Note ADDED IN PRESS

Additional sampling in Belize was conducted in Feb-
ruary 2009. Observations in shallow waters at Twin Cays
confirmed that populations of Charybdis helleri remained
as seen in 2007. Further sampling for thalassinideans sup-
ported accounts on the preceding pages, with noteworthy
additions. Sampling among shoreline mangrove roots at
Twin Cays produced the first specimens of the Axianas-
sidae, representing new records for Axianassa interme-
dia Schmitt, 1924. Five such specimens were extracted
by yabby pump from beneath a surface area of no more
than 0.25 m? at low tide, but less productive adjacent
sampling suggested heterogeneous patterning. Given the
small size of these specimens, we question whether this
species accounts for burrows provisionally attributed to
A. australis on the basis of castings mentioned on the
preceding pages. From these same habitats at Twin Cays,
the first specimen of the callianassid Biffarius fragilis
(Biffar, 1970) was captured, along with a specimen of the
same Naushonia sp. reported from Carrie Bow Cay on
preceding pages. Finally, the first specimen of the family
Callianideidae, Callianidea laevicauda Grill, 1959, was
taken from intertidal rubble of the exposed reef crest at
Carrie Bow Cay. These latest efforts confirm presence
of at least one species of the family Axianassidae, add
a seventh thalassinidean family to our report, and bring
the documented number of thalassinidean species in our
survey to at least 19.

lagoon

reef

’ Y; Feito

WALA Ti Meta fiiy

Li MEAl sf ft
Hingis gi

turtle

NUMBER 38 ¢ 187

mangrove
channel

mangrove
island

mangrove margins

coral rubble and sand coarse

rubble pavement rubble sand grass peat sand
me --.---- — —S OT"
Coralaxius Axlopsis Corallianassa Neocallichirus 7Axlanassa Upogebia omissa, _Neocallichirus
nodulosus, __ serratifrons longiventris grandimana, australis Upogebia sp., grandimana,
Pomatogebla N. maryae, Alpheus spp., N. maryae,
operculata, Glypturus acanthochirus, Upogebia sp.,
Paraxiopsis spp., Eucalliax sp., Alpheus sp.
Upogebia spp. Mictaxius thalassicola

ee ill

along mainland shorelines (polyhaline)
Lepidophthalmus richardi, Upogebia omissa, Upogebia sp.

FIGURE 2. Schematic of thalassinidean distributions in channel and back-reef environments near Carrie Bow
Cay and Twin Cays, Belize. Modified from Dworschak and Ott (1993:fig. 9).

ACKNOWLEDGMENTS

We are grateful to the late R. Manning and B. Kensley
for assistance in field collections. We thank S. De Grave,
E. Palacios-Theil, and B. Thoma for assistance in recent ef-
forts. Research was supported under several travel grants
to the authors from the Smithsonian Caribbean Coral Reef
Ecosystems (CCRE) Program and from National Science
Foundation grants DEB-0315995 and EF-0531603 to
the first author. This work is contribution number 825 of
the Caribbean Coral Reef Ecosystems Program (CCRE),
Smithsonian Institution, supported in part by the Hunter-
don Oceanographic Research Fund.

LITERATURE CITED

Abed-Navandi, D., and P. C. Dworschak. 2005. Food Sources of Tropi-
cal Thalassinidean Shrimps: A Stable-Isotope Study. Marine Ecol-
ogy Progress Series, 291:159-168.

Calado, T. C. S. 1996. Registro de Charybdis hellerii (Milne Edwards,
1867) em Aguas do Litoral Brasileiro (Decapoda: Portunidae).
[Record of Charybdis hellerii (Milne Edwards, 1867) in waters fo
the Brazilian coast (Decapoda: Portunidae).] Boletim de Estudios
de Ciéncias Mar, 9:175-180.

Campos, N. H., and M. Tiirkay. 1989. On a Record of Charybdis hel-
lerii from the Caribbean Coast of Colombia. Senckenbergiana Mar-
itima, 20:119-123.

de Vaugelas, J. 1985. A New Technique for Collecting Large-Sized Cal-
lianassid Mud-Shrimp (Decapoda, Thalassinidea). Crustaceana,
49:105-109.

Dineen, J. FE, P. K Clark, A. H. Hines, S. A. Reed, and H. P. Walton.
2001. Life History, Larval Description, and Natural History of
Charybdis hellerii (Decapoda, Brachyura, Portunidae), an Inva-
sive Crab in the Western Atlantic. Journal of Crustacean Biology,
21:774-805.

Duffy, E. D. 1996. Species Boundaries, Specialization, and the Radiation
of Sponge-Dwelling Alpheid Shrimp. Biological Journal of the Lin-
nean Society, 58:307-324.

Duffy, E. D., and K. §. Macdonald. 1999. Colony Structure of the Social
Snapping Shrimp Synalpheus filidigitus in Belize. Journal of Crusta-
cean Biology, 19:283-292.

Duffy, E. D., C. L. Morrison, and K. S. Macdonald. 2002. Colony De-
fense and Behavioral Differentiation in the Eusocial Shrimp Synal-
pheus regalis. Behavioral Ecology and Sociobiology, 51:488-495.

Duffy, E. D., C. L. Morrison, and R. Rios. 2000. Multiple Origins of
Eusociality Among Sponge-Dwelling Shrimps (Synalpheus). Evolu-
tion, 54:503-516.

Dworschak, P. C., H. Koller, and D. Abed-Navandi. 2006. Burrow Struc-
ture, Burrowing and Feeding Behaviour of Corallianassa longiventris
and Pestarella tyrrhena (Crustacea, Thalassinidea, Callianassidae).
Marine Biology, 148:1369-1382.

Dworschak, P. C., and J. A. Ott. 1993. Decapod Burrows in Mangrove-
Channel and Back-Reef Environments at the Atlantic Barrier Reef,
Belize. Ichnos, 2:277-290.

Dworschak, P. C., and S. de A. Rodrigues. 1997. A Modern Analogue for
the Trace Fossil Gyrolithes: Burrows of the Thalassinidean Shrimp
Axianassa australis. Lethaia, 30:41-52.

Felder, D. L. 2001. Diversity and Ecological Significance of Deep-
Burrowing Macrocrustaceans in Coastal Tropical Waters of the
Americas (Decapoda: Thalassinidea). Interciencia, 26:2-12.

188 e

Felder, D. L., E. Alvarez, J. W. Goy, and R. Lemaitre. 2009. “Decapoda
(Crustacea) of the Gulf of Mexico, with Comments on the Am-
phionidacea.” In Gulf of Mexico: Its Origin, Waters, and Biota.
Volume I, Biodiversity, ed. D. L. Felder and D. K. Camp, pp. 1019-
1104. College Station: Texas A&M University Press.

Felder, D. L., and R. B. Manning. 1997. Ghost Shrimp of the Genus
Lepidophthalmus Holmes, 1904, from the Caribbean Region, with
Description of L. richardi, New Species, from Belize (Decapoda:
Thalassinidea: Callianassidae). Journal of Crustacean Biology, 17:
309-331.

Gomez, O., and J. C. Martinez-Iglesias. 1990. Reciente Hallazgo de la
Especie Indopacifica Charybdis hellerii (A. Milne Edwards, 1867)
(Crustacea: Decapoda: Portunidae) en Aguas Cubanas. [Record of
the Indopacific Species Charybdis helleri (A. Milne Edwards, 1867)
(Crustacea: Decapoda: Portunidae) in Cuban Waters. Caribbean
Journal of Science, 26:70-72.

Goy, J. W., and D. L. Felder. 1988. Two New Species of Microprosthema
from the Western Atlantic (Crustacea: Decapoda: Stenopodidea).
Journal of Natural History, 22:1277-1292.

Hernandez, G., and J. Bolafios. 1995. “Additions to the Anomuran and
Brachyuran Fauna of Northeastern Venezuela.” Program [abstract
booklet], Summer Meeting 24-27 May 1995, The Crustacean Soci-
ety, Smithsonian Marine Station at Fort Pierce, Florida.

Karasawa, H. 2004. Neocallichirus maryae, a Replacement Name for
Neocallichirus rathbunae (Schmitt, 1935) (Crustacea: Decapoda:
Thalassinidea). Paleontological Research, 8:87.

Kensley, B. 1981. Notes on Axiopsis (Axiopsis) serratifrons (A. Milne
Edwards) (Crustacea: Decapoda: Thalassinidea). Proceedings of the
Biological Society of Washington, 93:1253-1263.

. 1994. The Genus Coralaxius Redefined, with the Description

of Two New Species (Crustacea: Decapoda: Axiidae). Journal of

Natural History, 28:813-828.

. 1996. The Genus Paraxiopsis De Man, with Descriptions of
New Species from the Western Atlantic (Crustacea: Decapoda: Axi-
idae). Bulletin of Marine Science, 58:709-729.

Lemaitre, R. 1995. Charybdis hellerii (Milne Edwards, 1867), a nonin-
digenous portunid crab (Crustacea: Decapoda: Brachyura) discov-
ered in the Indian River lagoon system of Florida. Proceedings of
the Biological Society of Washington, 108(4):643-648.

Macdonald, K. S., R. Rios, and J. E. Duffy. 2006. Biodiversity, Host
Specificity, and Dominance by Eusocial Species Among Sponge-
Dwelling Alpheid Shrimp on the Belize Barrier Reef. Diversity and
Distributions, 12:165-178.

Manning, R. B., and D. L. Felder. 1996. Nannotheres moorei, a Minute
New Genus and Species of Pinnotherid Crab from Belize, Carib-

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

bean Sea (Crustacea: Decapoda: Pinnotheridae). Proceedings of the
Biological Society of Washington, 109:311-317.

Mantelatto, FE. L., and L. L. Dias. 1999. Extension of the Known Distri-
bution of Charybdis hellerii (A. Milne-Edwards, 1867) (Decapoda,
Portunidae) Along the Western Tropical South Atlantic. Crusta-
ceana, 72:617-620.

Mantelatto, F. L., and R. B. Garcia. 2001. Biological Aspects of the Non-
indigenous Portunid Crab Charybdis hellerii in the Western Tropi-
cal South Atlantic. Bulletin of Marine Science, 68:469-477.

Mantelatto, E L., R. Robles, and D. L. Felder. 2007. Molecular Phylogeny
of the Crab Genus Portunus (Crustacea, Portunidae) from the Western
Atlantic. Zoological Journal of the Linnean Society, 150:211-220.

Mantelatto, E L., R. Robles, C. D. Schubart, and D. L. Felder. 2009.
“Molecular Phylogeny of the Genus Cronius Stimpson, 1860, with
Reassignment of C. tumidulus and Several American Species of
Portunus to the Genus Achelous De Haan, 1833 (Brachyura: Por-
tunidae).” In Crustacean Issues: Decapod Crustacean Phylogenet-
ics, ed. J. W. Martin, K. A. Crandall, and D. L. Felder. Boca Raton,
Fla.: Taylor and Francis/CRC Press.

McMillen-Jackson, A. L. 2008. First Record of the Indo-Pacific Swim-
ming Crab, Charybdis hellerii (A. Milne-Edwards, 1867) in the
Gulf of Mexico. Crustaceana 81:889-894.

Rios, R., and J. E. Duffy. 2007. A Review of the Sponge-Dwelling Snap-
ping Shrimp from Carrie Bow Cay, Belize, with Description of
Zuzalpheus, New Genus, and Six New Species (Crustacea: Decap-
oda: Alpheidae). Zootaxa, 1602:1-89.

Robles, R., C. D. Schubart, J. E. Conde, C. Carmona-Sudrez, F. Alva-
rez, J. L. Villalobos, and D. L. Felder. 2007. Molecular Phylogeny
of the American Callinectes Stimpson, 1860 (Brachyura: Portuni-
dae), Based on Two Partial Mitochondrial Genes. Marine Biology,
150:1265-1274.

Rodriguez, I. T., G. Hernandez, and D. L. Felder. 2005. Review of the
Western Atlantic Porcellanidae (Crustacea: Decapoda: Anomura)
with New Records, Systematic Observations, and Comments on
Biogeography. Caribbean Journal of Science, 41:544-582.

. 2006. Phylogenetic Relationships Among Western Atlantic Por-
cellanidae (Decapoda: Anomura), Based on Partial Sequences of the
Mitochondrial 16s rRNA gene, with Comments on Morphology.
Crustacean Research, Special Number 6:151-166.

Sakai, K. 2005. Callianassoidea of the World (Decapoda, Thalassinidea).
Crustaceana Monographs, 4:1-285.

Williams, A. B. 1993. Mud Shrimps, Upogebtidae, from the Western
Atlantic (Crustacea: Decapoda: Thalassinidea). Smithsonian Con-
tributions to Zoology, No. 544. Washington, D.C.: Smithsonian
Institution Press.

Imposex in One of the World’s Busiest

Shipping Zones

Carter Li and Rachel Collin

Carter Li, Department of Biology, McGill Uni-
versity (current address: Department of Surgery,
Montreal General Hospital). Rachel Collin, Smith-
sonian Tropical Research Institute, MRC 0580-08,
Unit 9100, Box 0948, DPO AA 34002, USA.
Corresponding author: R. Collin (collinr@si.edu).
Manuscript received 21 July 2008; accepted 20
April 2009.

ABSTRACT. Tributyltin pollution from antifouling paint is well known to disrupt the en-
docrine system in female marine gastropods. The masculinization of females, including the
aberrant growth of a penis and vas deferens and occlusion of the capsule gland, has been
reported primarily in neogastropods and is particularly well documented in muricids. Com-
pared to temperate areas, few studies of imposex have been undertaken in the tropics, and
there are few studies in general on non-neogastropods. Here we report a high frequency of
imposex near the Pacific mouth of the Panama Canal in two species of muricids and two
species of calyptraeids. The frequency of imposex declined rapidly with distance away from
the canal, and several species appeared to be mostly normal less than 10 km from the en-
trance. This is the first report of imposex in Acanthais brevidentata, Thaisella kiosquiformis,
Bostrycapulus calyptraeformis, Crepidula cf. nivea, and Anachis fluctuata. Because imposex
has not previously been reported for the Calyptraeidae, a family of protandrous gastropods,
a laboratory study was conducted to verify that imposex was not simply retention of the
penis after sex change. The 2007 ratification of the International Maritime Organization’s
convention on antifouling systems should reduce the levels of TBT worldwide, but the persis-
tence of this compound in sediments suggests that imposex may continue to be a problem at
the mouth of the canal as routine dredging and large tides frequently resuspend sediment.

INTRODUCTION

Tributyltin (TBT) is well known to be a highly effective antifouling agent,
used primarily on ship hulls, but it has numerous detrimental effects on a wide
variety of non-target taxa. Despite having demonstrable effects on molluscan
shell growth (Alzieu et al., 1981), embryological development of fish and marine
invertebrates (Hano et al., 2007; Inoue et al., 2006), neurulation in ascidians
(Dolcemascolo et al., 2005), and testosterone metabolism in mysids (Verslycke
et al., 2003), the most well studied and widespread effect is the disruption of the
endocrine system in marine gastropods. Exposure to very low levels (as little as
0.5 ng/L) of TBT causes the masculinization of females, including the aberrant
growth of a penis and occlusion of the capsule gland (Gibbs and Bryan, 1996).
This condition is referred to as imposex, and severe cases can lead to reproduc-
tive failure. For example, an extreme case of population decline as a result of im-
posex has been demonstrated for Nucella lapillus in southwest England (Bryan
et al., 1986; Gibbs and Bryan, 1986).

190 e

In a recent review, Shi et al. (2005) reported that
imposex has been recorded in 170 species of gastropods
from 28 families. The vast majority, 134 species, are
neogastropods. Among the non-neogastropod, caeno-
gastropod families, ampullarids, rissioids, cypreaids, cy-
matids, and tonnids all contain several species for which
imposex has been reported (Shi et al., 2005). Although
the taxonomic coverage is wide, much of the basic in-
formation on imposex in relation to TBT pollution is
centered on muricids, buccinids, and conids (Fioroni et
al., 1991; Shi et al., 2005). On a worldwide scale, it is
necessary to extend the scope of studies to include more
tropical forms and locations (Ellis and Pattisina, 1990)
to get a global picture of the effects of TBT pollution on
gastropods.

The Panama Canal is one of the world’s busiest ship-
ping zones, and commercial transport through the canal
represents about 5% of world trade. About 14,000 vessels
pass through the Canal annually (statistics available from
the Autoridad del Canal de Panama web site http://www
._pancanal.com/), and the most common shipping route is
between the east coast of North America and Asia. Most of
the shipping traffic is composed of large, oceangoing ves-
sels, which have not previously been subject to restrictions
on the use of tributyltin antifouling paint. The entrance to
the Canal, on the Pacific coast adjacent to Panama City,
was the site of Rodman Naval Base (1943-1999), and
is currently the site of the container port of Balboa and
a shipyard. The anchorage for the canal commonly has
more than 30 vessels waiting to transit the Canal. The sub-
strate in this area is primarily a mix of rocky debris and
sandy mud in the intertidal and fine mud in the subtidal.
With the consistently high levels of shipping traffic, fre-
quent dredging, and muddy substrate (which is known to
retain TBT for years, as reviewed in de Mora, 1996), the
local levels of TBT and, therefore, imposex are expected
to be higher around the entrance to the Canal than they
are along the open coast. We conducted a survey of four
common intertidal gastropod species around the mouth of
the canal to document the levels and geographic extent of
imposex in this area.

MATERIALS AND METHODS

Gastropods were collected between February and
April 2005 from four sites along the Pacific coast of
Panama at varying distances from the mouth of the Pan-
ama Canal (Figure 1). The site closest to the mouth of

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

ip

ee Culebra

FIGURE 1. Map of the study area at the entrance to the Panama
Canal. Arrows indicate the sites of sample collection and locations
mentioned in the text.

the canal consisted of rocky outcrops near Farfan beach
(8.93°N, 79.58°W) and the Bridge of the Americas. Pro-
gressively further away to the west were Isla Venado
(8.91°N, 79.63°W), Chumical (8.5°N, 79.66°W), and
Bique (8.90°N, 79.66°W). In November 2007 additional
samples were collected from Punta Culebra (8.91°N,
79.53°W), which faces the entrance to the Canal and is at
the edge of the Canal anchorage.

We collected four species, which were clearly identi-
fiable and abundant at two or more of the sites. Efforts
were made to collect the same species from all sites, but
because of the habitat heterogeneity in the area, we were
not able to collect sufficient numbers of females for sta-
tistical analyses for several sites. Adequate samples were
collected for the muricids Acanthais brevidentata (Wood,
1828) from Farfan and Chumical and Thaisella kiosqui-
formis (Duclos, 1832) from Farfan and Bique, and the
calyptraeids Bostrycapulus calyptraeformis (Deshayes,
1830) and Crepidula cf. nivea from Farfan, Venado, and
Chumical (Table 1).

Shell length was measured with vernier calipers,
and live snails were extracted from their shells. The re-
productive system was immediately examined under a
stereomicroscope, and the sex was determined based on
characteristics of the gonad and presence or absence of
seminal receptacles and seminar vesicles. If the sex was
not easily identified, sex was verified by examining ga-
metes from a smear of gonad. The length of the penis (if
present) was measured using an ocular micrometer on a
stereomicroscope.

NUMBER 38 e¢ 191

TABLE 1. Frequency of imposex in four gastropod species at sites arranged here from nearest to furthest from the
Panama Canal entrance. Frequency at each site was compared to that at Farfan (the site at the entrance to the Canal)
using a Fisher’s exact test. A one-tailed test was used, but two-tailed results did not differ; * P = 0.001; ** P = 0.0001;

*e% P= < (0.0001.
Site
Species Farfan Culebra Venado Chumical Bique
Muricids
Thaisella kiosquiformis 29/53 - ~ 13/52*
Acanthais brevidentata 8/32 - 0/57** -
Calyptraeids
Bostrycapulus calyptraeformis 60/63 22/434*** yer? Ino: -
Crepidula cf. nivea 87/90 19/22 - 0/9 9a -
4 Significantly different from Venado and Chumical <0.0001.
Significant differences in the frequency of imposex be- RESULTS

tween the entrance to the Canal and more distant sites
were tested for using Fisher’s exact test. Analysis of co-
variance (ANCOVA) was used to examine the relation-
ship between penis length in male and imposex females,
with shell length as a covariate for samples collected from
Venado and Farfan. Because samples from Culebra were
preserved in ethanol before examination, the penis length
from these samples could not be directly compared to the
others that were measured fresh.

Experiments to determine if imposex develops in adult
snails after exposure to ambient water levels of TBT were
conducted at STRI’s Naos Marine Laboratories, only a
few hundred meters from the Culebra site. Anachis fluc-
tuata (Sowerby, 1832) and Bostrycapulus calyptraeformis
were both collected from Isla Venado, an area with low
levels of imposex, and maintained in the laboratory. Sixty
adult Anachis fluctuata were kept in a 100 L fiberglass
tank in the outside seawater system and fed frozen com-
mercial clams once a week. After five months the animals
were killed and levels of imposex were determined as al-
ready described. Bostrycapulus calpytraeformis were col-
lected as small males. They were maintained in pairs in the
laboratory in 350 mL plastic cups. The water was changed
every other day and the animals were fed 10 mL Isochrysis
galbana culture every day. Animals were measured every
four weeks, and their sexual state was recorded on the
basis of external features. The experiment was terminated
after 400 days. Both species were cultured using the same
source of seawater (from the side of Isla Naos away from
the Canal entrance), and neither was exposed to local sedi-
ment other than that which settled out of the seawater.

FiELD COLLECTIONS

Imposex was detected in all four species. In the two
muricids, the imposex was almost always in the early
stages with limited penis development and no indication
of any occlusion of the capsule gland. We never observed
imposex that was so far advanced that the females were
found to retain eggs or that an obvious vas deferens had
developed. Imposex in the calyptraeids was more devel-
oped; penes were large in many specimens and could easily
be confused with a normal male penis. Several imposex fe-
males of Bostrycapulus calyptreaformis and Crepidula cf.
nivea were observed brooding egg capsules, showing that
imposex females were not sterile. Near the entrance of the
Canal the frequency of imposex ranged from 25% to 50%
in muricids and was greater than 80% in calyptraeids. The
number of females collected for each species at each site and
the frequency of imposex are given in Table 1. In all cases
the frequency of imposex was significantly higher near the
entrance to the Canal than at farther sites (Table 1).

Acanthais brevidentata: Because there were no imposex
individuals in Bique and because animals from that site
were significantly larger (mean = 28.9 mm) than from
Farfan (mean = 26.9 mm; P < 0.001), comparisons of
imposex females with normal males and females were
conducted for data collected from Farfan only. Imposex
females were significantly larger (length = 30.1 mm) than
non-imposex females (length = 26.4 mm; P < 0.02).
ANCOVA showed that there were significant effects
of shell length (P = 0.01) and imposex (P < 0.001) on

penis length as well as a significant interaction effect
(P < 0.04). Penes of imposex females were smaller than
those of normal males, and male penis length increased
with shell length, although imposex penis length was not
associated with shell length (Figure 2).

192 e© SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

from Bique and Farfan were combined for the analysis.
ANCOVA showed that there were significant effects of
shell length (P < 0.0001) and imposex (P < 0.0001)
on penis length as well as a significant interaction effect
(P = 0.003). Penes of imposex females were smaller

than those of normal males, and penis length increased
with shell length in both sexes (Figure 2). There was a
significant incidence of imposex at Bique, despite it be-
ing the site furthest from the Canal. We attribute this
relatively high frequency of imposex to this site’s prox-

Thaisella kiosquiformis: Animals from Bique and Farfan did
not differ in size, nor did the sexes differ in size. Imposex
females were also the same size as non-imposex females.
The average size for all categories was 26-27 mm. Data

Crepidula cf. nivea

4 Acanthais brevidentata

Penis Length (mm)

6 °

Penis Length (mm)

Shell Length (mm)

Shell Length (mm)

FIGURE 2. Relationship between shell length and penis length for males (white diamonds) and imposex females (black dots) of Acanthais
brevidentata, Thaisella kiosquiformis, Crepidula cf. nivea, and Bostrycapulus calyptraeformis.

imity to a dry dock facility slightly further to the west
in Vacamonte.

Crepidula cf. nivea: Calyptraeids are protandrous hermaph-
rodites (Collin, 2006) and the small animals are almost
always males; therefore, we did not make as much effort
to collect the smallest animals as we did in the other spe-
cies. The size of females differed significantly between
the sites (16.1 mm for Chumical vs. 19.6 mm for Farfan;
P < 0.0001). Imposex females were larger than non-
imposex females. Because all females at Chumical were
normal and virtually all in Farfan had imposex, this size
difference cannot be distinguished from a site effect on
size. An ANCOVA showed that shell length had a sig-
nificant effect on penis length (P = 0.03), that imposex
females and males did not differ in penis length as there
was considerable penis growth in the imposex females,
and that there was no significant interaction between im-
posex status and shell length (see Figure 2). Although
there were significant levels of imposex in samples from
Culebra, in all these cases the penis was very small; they
were not much more developed than a small bump at
the base of the tentacle, whereas those from Farfan were
often as long as or longer than the tentacles.

Bostrycapulus calyptraeformis: The average size of fe-
males differed significantly among the three sites (17.5
mm at Chumical; 16.5 mm at Venado; 20.6 at Farfan;
P < 0.01). Again, because nearly all the females in Far-
fan had imposex but no imposex was detected in the
other locations, the larger size of imposex females may
have been a site effect. ANCOVA analysis of animals
from Farfan showed that there was a significant effect
of imposex on penis length (P < 0.001), and imposex
females had smaller penes than males. Shell length and
the interaction between shell length and imposex had
no significant effect on penis length. The level of im-
posex in animals from Culebra was again very rudi-

NUMBER 38 e¢ 193

mentary, with penes little more than a nub at the base
of the tentacle.

In summary, all four species showed significant higher
rates of imposex near the entrance of the canal. By 20 km
away, rates were generally of the order of 1%-2%. In the
two muricids and one of the calyptraeids, the penes of
imposex females were smaller than those of similar-sized
males. In the two muricids and the other calyptraeid, shell
length was a significant covariate of penis length, and in
two species the penis length of imposex females increased
with shell length.

LABORATORY EXPERIMENTS

After five months in the laboratory, 2 of 29 female
Anachis fluctuata had developed penes, indicating that
this species can develop imposex. However, this was not
statistically significantly different from the frequency of
imposex in the field in Venado (P = 0.09; Table 2). No
comparisons to the entrance to the Canal could be made
because this species could not be found there.

Of the 60 Bostrycapulus calyptraeformis that were
raised in the laboratory, the largest animals in 6 of the 30
cups retained penes throughout the experiment and did
not change to become female. In the remaining 24 cups,
the larger of the 2 animals lost the penis, indicating sex
change from male to female. Of these 24 animals, 10 lost
the penis and then subsequently regained it 1 to 3 months
after sex change. In many cases the penis was not as long
or thick as a normal male penis, but they were fairly large,
and casual observers would be likely to categorize such
animals as males (Figure 3). The largest animals in the re-
maining 14 cups underwent transition to normal females
and did not develop imposex before the end of the expert-
ment. The smaller of the 2 animals in each cup was not
examined, as they usually remain male in the presence of
the larger animals (Collin et al., 2005).

TABLE 2. Frequency of imposex from field-collected and laboratory-reared snails.

Fisher’s exact test,

Species Laboratory Venado Chumical P value
Anachis fluctuata 2/29 1/133 - P=0.091
Bostrycapulus calyptraeformis 10/24 2/79 1/122 P < 0.0001, < 0.0001

194 e

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 3. A, B. Photographs of two individuals with well-developed imposex in Crepidula cf. nivea from
Farfan, with penis (p) and female genital papilla (fgp) indicated in each image. B. The female genital papilla
can be confused with the anus (a), which is usually obscured by the gills; brooded eggs are visible as the light
mass in this image.

DISCUSSION

Imposex was found in all snail species examined
from the mouth of the Panama Canal, and in all cases
the frequency and severity of imposex declined with dis-
tance away from the Canal. The frequency of imposex
differed among the species examined, with calyptraeids
more likely to display imposex than the muricids. Ca-
lyptraeids showed higher frequencies of imposex at the
mouth of the Canal, and the penes of imposex females
were much more fully developed then we ever observed
in muricids. Species differences in sensitivity both to TBT
(Wilson et al., 1993; Tan, 1999) and to its bioaccumula-
tion (Liu et al., 1997) have been demonstrated in other
surveys. Liu et al. (1997) found that imposex was much
more severe in Thais species than Morula, despite similar
organotin burdens, and suggested a genus-specific suscep-
tibility to organotin pollution with the ranking order of
Nucella, Thais, and Morula. The differences in habitat
(high on the intertidal versus low on the intertidal), diet,
and physiology have been suggested as causes of inter-
specific differences in imposex (Tan, 1999). If TBT were
primarily waterborne either in solution as bis(tributyltin)
oxide or adsorbed by suspended solids (de Mora, 1996) at
our study sites, it is possible that filter-feeding calyptraeids
would be exposed to more TBT, by filtering large volumes

of water, than would other gastropods. Suspended parti-
cles may have TBT adhered to them and may be captured
in the mucous net and ingested during filter feeding, thus
increasing the exposure of calyptraeids relative to the mu-
ricids. These scenarios are not in agreement with a number
of laboratory studies (Bryan et al., 1989; see Gibbs and
Bryan, 1996, for review) that show that TBT accumulates
more rapidly from the diet than from the ambient water
and which suggest that carnivores could accumulate more
TBT from their diet than would herbivores. However,
controlled experimental comparisons of bioaccumulation
between carnivorous and suspension-feeding gastropods
have not been made, and the effects of suspended solids
have not been examined.

Another factor that can influence the expression of
imposex is age. Because extended exposure to TBT is nec-
essary to elicit imposex, those species that are longer lived
or slower growing may be more likely to have high lev-
els of TBT and thus exhibit imposex. Studies have also
shown that juvenile snails are more sensitive to TBT than
are adults (Gibbs and Bryan, 1996). Our data for Acan-
thais brevidentata, showing that females with imposex are
larger than normal females, are consistent with either in-
crease in imposex development with long-term exposure
or recent reductions in TBT levels. However, Thaisella
kiosquiformis did not show this pattern. Few data on the

age or lifespans of tropical gastropods are available and so
this possibility is difficult to evaluate. However, Panama-
nian calyptraeids grown in the laboratory generally reach
maturity at sizes similar to animals that matured in the
field, in less than a year (Collin et al., 2005, and personal
observation), and it seems unlikely that TBT in the sedi-
ment, which has a half-life of years, would have changed
drastically in such a short interval.

Imposex has not been previously reported in calyp-
traeid gastropods. Because animals normally change from
males to females and transitional animals may sometimes
retain a penis while also showing well-developed female re-
productive structures, it is possible that imposex individuals
have previously been misidentified as undergoing the nor-
mal transition between the male and females phases. Here
we found, in sites with low expected TBT exposure, that
there are virtually no individuals that display both male and
female characteristics at the same time. In addition, our lab-
oratory studies show that during sex change the penis can
be reabsorbed and that the penes of imposex individuals
can grow following this reabsorption. These results show
that the abundant large females with penes collected at the
entrance to the Canal are indeed imposex females and not
transitional individuals that have yet to lose the penis.

Numerous studies have shown a tight relation-
ship between levels of TBT in the environment, levels of
TBT in gastropod tissues, and the frequency of imposex
within species (Gibbs et al., 1987; Horiguchi et al., 1994;
Minchin et al., 1997; Ruiz et al., 1998). However, the re-
lationships between sites, species, and the different types
of triorganotins are not always simple (Ide et al., 1997).
Imposex has also been shown to be a more sensitive way
to detect TBT than many chemical detection methods, and
imposex has been used as a bio-indicator when TBT lev-
els are too low for easy analytical detection (Gibbs and
Bryan, 1996). Despite an extensive literature on the re-
lationship between TBT and imposex, one study (Nias et
al., 1993) indicates imposex could result from exposure
to paint matrix or copper. However, this result has not
been pursued or elaborated. Although we could not mea-
sure levels of TBT directly at the sites around the Canal,
it can, in the light of this literature, be inferred with some
level of confidence that the exposure of animals to TBT
is higher at the entrance to the Canal than it is in the sur-
rounding areas. Despite the high levels of shipping and
presumably high levels of TBT leaching into the surround-
ing water, the development of imposex was not so severe
as has been reported for areas with high shipping traffic
in Europe and Asia, and TBT does not have an extreme
impact on reproduction by occluding the pallial oviduct

NUMBER 38 ¢ 195

or splitting the bursa copulatrix and capsule gland, as has
been reported from these regions (Oehlmann et al., 1996;
Shi et al., 2005). Less obvious effects on reproduction were
not directly evaluated in this study. The high amount of
flushing in the area, from large volumes of discharge from
the Canal and the 6 m tides, may help to prevent local
buildup of high concentrations of TBT in this partially en-
closed area.

In 2002 the International Maritime Organization
adopted a Convention on Antifouling Systems (AFS) that
called for a global prohibition of the application of or-
ganotin compounds as biocides in antifouling systems on
ships by 1 January 2003 and a complete prohibition by 1
January 2008. However, the prohibition was only to be
implemented 12 months after 25 states representing 25%
of the world’s merchant shipping tonnage ratified it. In
September 2007 this quota was met when Panama rati-
fied the convention, and therefore these regulations went
into effect in September 2008. As the AFS convention ap-
plies to ships flagged in, operated by, or docking in states
that have ratified it, the convention should significantly
reduce the exposure of Panama’s marine habitats to TBT
pollution in the coming years. This regulation is especially
important because the planned expansion of the Canal in
2014 will significantly increase shipping traffic along both
the Pacific and Caribbean coasts of Panama.

ACKNOWLEDGMENTS

We are grateful to L. Weintraub and M. Salazar for as-
sisting with field collections of animals, M. deMaintenon
for verifying the identification of the columbellid samples,
and M. Torchin for sharing his knowledge of gastropod
reproductive systems. We thank the Autoridad Maritima
de Panama for providing collecting permits. This project
was conducted during a McGill University Field Semester
in Panama.

LITERATURE CITED

Alzieu, C., M. Heral, Y. Thibaud, M. J. Dardignac, and M. Feuillet. 1981.
Influence des Peintures Antisalissures a Base d’Organostanniques sur
la Calcification de la Coquille de l’Huitre Crassostrea gigas. Revue
des Travaux de l'Institut des Peches Maritimes, 45:101-116.

Bryan, G. W., P. E. Gibbs, L. G. Hummerstone, and G. R. Burt. 1986.
The Decline of the Gastropod Nucella lapillus around Southwest
England: Evidence for the Effect of Tributyltin from Antifouling
Paints. Journal of the Marine Biological Association of the United
Kingdom, 66:611-640.

. 1989. Uptake and Transformation of '*C-Labelled Tributyltin

Chloride by the Dog-Whelk Nucella lapillus: Importance of Absorp-

tion from the Diet. Marine Environmental Research, 28:241-245.

196 e

Collin, R. 2006. Sex Ratio, Life History Invariants, and Patterns of
Sex Change in a Family of Protandrous Gastropods. Evolution,
60:735-745.

Collin, R., M. McLellan, K. Gruber, and C. Bailey-Jourdain. 2005. Ef-
fects of Conspecific Associations on Size at Sex Change in Three
Species of Calyptraeid Gastropods. Marine Ecology Progress Series,
293:89-97.

de Mora, S. J. 1996. “The Tributyltin Debate: Ocean Transportation
Versus Seafood Harvesting.” In Tributyltin: Case Study of an Envi-
ronmental Contaminant, ed. S. J. de Mora, pp. 1-19. Cambridge:
Cambridge University Press.

Dolcemascolo, G., P. Gianguzza, C. Pellerito, L. Pellerito, and M.
Gianguzza. 2005. Effects of Tri-m-Butyltin (IV) Chloride on Neu-
rulation of Ciona intestinalis (Tunicata, Ascidiacea): An Ultrastruc-
tural Study. Applied Organometallic Chemistry, 19:11-22.

Ellis, D. V., and L. A. Pattisina. 1990. Widespread Neogastropod Im-
posex: A Biological Indicator of Global TBT Contamination. Ma-
rine Pollution Bulletin, 21:248-253.

Fioroni, P., J. Oehlmann, and E. Stroben. 1991. The PseudoHermaph-
roditism of Prosobranchs; Morphological Aspects. Zoologischer
Anzeiger, 226:1-26.

Gibbs, P. E., and G. W. Bryan. 1986. Reproductive Failure in Populations
of the Dog-Whelk, Nucella lapillus, Caused by Imposex Induced by
Tributyltin from Antifouling Paints. Journal of the Marine Biologi-
cal Association of the United Kingdom, 66:767-777.

. 1996. “TBT-Induced Imposex in Neogastropod Snails: Mascu-
linization to Mass Extinction.” In Tributyltin: Case Study of an
Environmental Contaminant ed. S. J. de Mora, pp. 212-236. Cam-
bridge: Cambridge University Press.

Gibbs, P. E., G. W. Bryan, P. L. Pascoe, and G. R. Burt. 1987. The Use
of the Dog-Whelk, Nucella lapillus, as an Indicator of Tributyltin
(TBT) Contamination. Journal of the Marine Biological Associa-
tion of the United Kingdom, 68:507-523.

Hano, T., Y. Oshima, S. G. Kim, H. Satone, Y. Oba, T. Kitano, S. Inoue,
Y. Shimasaki, and T. Honjo. 2007. Tributyltin Causes Abnormal
Development in Embryos of Medaka, Oryzias latipes. Chemo-
sphere, 69(6):927-933.

Horiguchi, T., H. Shiraishi, M. Shimizu, and M. Morita. 1994. Imposex
and Organotin Compounds in Thais claigera and T. bronni in Japan.
Journal of the Marine Biological Association of the United King-
dom, 74:651-669.

Ide, I., E. P. Witten, J. Fischer, W. Kalbfus, A. Zeller, E. Stroben, and
B. Watermann. 1997. The Accumulation of Organotin Compounds

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

in the Common Whelk Buccinum undatum and the Red Whelk
Neptunea antiqua in Association with Imposex. Marine Ecology
Progress Series, 152:197-203.

Inoue, S., Y. Oshima, H. Usuki, M. Hamaguchi, Y. Hanamura, N. Kai,
Y. Shimasaki, and T. Honjo. 2006. Effects of Tributyltin Mater-
nal and/or Waterborne Exposure on the Embryonic Development
of the Manila Clam, Ruditapes philippinarum. Chemosphere,
63(5):881-888.

Liu, L. L., S. J. Chen, W. Y. Peng, and J. J. Hung. 1997. Organotin Con-
centrations in Three Intertidal Neogastropods from the Coastal
Waters of Taiwan. Environmental Pollution, 98:113-118.

Minchin, D., B. Bauer, J. Oehlmann, U. Schulte-Oehlmann, and C. G.
Duggan. 1997. Biological Indicators Used to Map Organotin Con-
tamination from a Fishing Port, Killybegs, Ireland. Marine Pollu-
tion Bulletin, 34:235-243.

Nias, D. J., S. C. McKillup, and K. S. Edyvane. 1993. Imposex in Lep-
stella vinosa from Southern Australia. Marine Pollution Bulletin,
26:380-384.

Oehlmann, J., E. Stroben, and P. Fioroni. 1996. “New Facts About Tribu-
tyltin-Induced Imposex in Prosobranchs: General Aspects.” In Mol-
luscan Reproduction: Malacological Review, Supplement 6, ed. N. H.
Runham, W. H. Heard, and J. B. Burch, pp. 149-156. Ann Arbor,
Mich.: Society for Experimental and Descriptive Malacology.

Ruiz, J. M., M. Quintela, and R. Barreiro. 1998. Ubiquitous Imposex
and Organotin Bioaccumulation in the Gastropod Nucella lapillus
from Galicia (NW Spain): A Possible Effect of Nearshore Shipping.
Marine Ecology Progress Series, 164:237-244.

Shi, H. H., C. J. Huang, S. X. Zhu, X. J. Yu, and W. Y. Xie. 2005.
Generalized System of Imposex and Reproductive Failure in Female
Gastropods of Coastal Waters of Mainland China. Marine Ecology
Progress Series 304:179-189.

Tan, K. S. 1999. Imposex in Thais gradata and Chicoreus capucinus
(Mollusca, Neogastropoda, Muricidae) from the Straights of Johor:
A Case Study Using Penis Length, Area and Weight As Measures of
Imposex Severity. Marine Pollution Bulletin, 39:295-303.

Verslycke, T., S. Poelmans, K. De Wasch, J. Vercauteren, C. DeVos, L.
Moens, P. Sandra, H. De Brabander, and C. Janssen. 2003. Testos-
terone Metabolism in the Estuarine Mysid Neomysis integer (Crus-
tacea; Mysidacea) Following Tributyltin Exposure. Environmental
Toxicology and Chemistry, 22:2030-2036.

Wilson, S. P., M. Ahsanullah, and G. B. Thompson. 1993. Imposex in
Neogastropods: An Indicator of Tributyltin Contamination in East-
ern Australia. Marine Pollution Bulletin, 26:44-48.

Shorefishes of the Tropical Eastern Pacific
Online Information System

D. Ross Robertson

D. Ross Robertson, Smithsonian Tropical Re-
search Institute, Unit 0948, APO AA 34002
(drr@stri.org). Manuscript received 15 August
2008; accepted 20 April 2009.

ABSTRACT. Shorefishes of the Tropical Eastern Pacific Online Information System
(SFTEP) version 1, 2008, provides an online electronic identification guide and in-
formation system for the known fauna of shorefishes found in the Tropical Eastern
Pacific. SFTEP allows users (i) to identify all shorefishes known from the Tropical
Eastern Pacific (TEP) (1,287 species in version 1) and (ii) to analyze and conduct
biogeographic research on the composition of that fish fauna at varying spatial scales.
Tools for identification emphasize the use of color photographs, along with descriptive
text that highlights key morphological features; allow comparison of similar species;
facilitate identification of unfamiliar species using information on location and fish
morphology (shape, color pattern, and color); and incorporate interactive keys to
members of two species-rich families (Gobiidae, Sciaenidae) that have many similar-
looking species. To accommodate nonspecialist users, scientific jargon is minimized;
the interface is intuitive and user-friendly, and searches for species can be made using
common names. The Research Engine, which provides information about the com-
position of local faunas and the regional fauna, allows users to compare geographic
ranges of multiple taxa, to construct faunal lists of taxonomic and functional groups
of species for single and paired sites, and, at varying spatial scales, to determine local
endemism and to display region-wide patterns of species richness of different taxa and
functional groups of fishes. The system is accessible online at www.stri.org/sftep.

INTRODUCTION

Shorefishes of the Tropical Eastern Pacific Online Information System
(SFTEP), version 1, 2008, provides an online electronic identification guide
and information system for the known fauna of shorefishes found in the Trop-
ical Eastern Pacific (TEP). This version represents the latest iteration of a
series that began with the 1994 English-language printed identification guide
of the same name (Allen and Robertson, 1994). That book was followed by a
Spanish-language printed edition in 1998 (Allen and Robertson, 1998). Both
these works were succeeded by a dual-language CD-based information system
in 2002 (Robertson and Allen, 2002), which was revised and expanded in
2006 (Robertson and Allen, 2006).

198 e

SYSTEM FEATURES
DUAL LANGUAGE INTERFACES

The system incorporates separate, full-capability Eng-
lish- and Spanish-language interfaces.

AIDS TO VISUAL IMPAIRMENT
The system incorporates two types of aids:

1. Variable map-color formats are available. Users can se-
lect various color schemes designed to accommodate
different patterns of color blindness, including mono-
chrome or color with the ability to select colors.

2. Page layout structure accommodates variation in font
size. Two page layouts are possible—landscape and
portrait. Page structure is stable over a threefold range
in font size.

SYSTEM MODULES
HOME

The home page provides an overview of the capabili-
ties of the system and access to all major modules through
buttons and/or tabs that act as shortcuts (Figure 1). In ad-
dition there are links to several modules not accessible from
other parts of the system: to the Copyright notice, a switch
to change between English and Spanish interfaces, and to
the websites of the Smithsonian Tropical Research Institute
and Coeus, the company that programmed the system.

Each of the authors and major contributors of infor-
mation directly related to the construction of SFTEP has
an individual contributor page, accessible from the Con-
tributors button and from a link at the top of any screen.
In addition, the major contributors of information pre-
sented on each family are noted on each family page.

GENERAL INFORMATION

General information about Shorefishes of the Tropi-
cal Eastern Pacific Online Information System (SFTEP) is
shown in Figure 2; this module includes three sections.

Introduction

The “Introduction” to the TEP and its shorefish fauna
provides background information on the oceanography of
the region and its marine habitats (geographic and tem-
poral variation in climate, rainfall and salinity, primary

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

production and coastal upwelling systems, ocean current
systems, influences of the El Nifio cycle, shoreline habitats
and rocky and coral reefs in the region); a history of taxo-
nomic fish guides, major modern guides, global online re-
sources, systematic ordering of the fishes, and the scientific
and common names of fishes); the ecology of TEP shore-
fishes (species that occur in the upper 100 m of the water
column over the continental shelf or within ~50 km of the
shore), their use of different environments and habitats,
their depth-distribution patterns, their dietary groupings,
and their modes of reproduction; and the zoogeography of
the fauna—studies of the region’s zoogeography, resident
versus vagrant species, relationships of the fauna to the
faunas of other areas, distribution of the fauna in differ-
ent climate zones, the geography of variation in species
richness and local endemism throughout the region, and
biogeographic subdivisions of the TEP.

Features & User Guide

The “Features & User Guide” section describes system
features, providing information available on taxon pages,
databases on biological and zoogeographic characteristics,
information used to identify fishes, an interactive glossary
of ichthyological terms, the functioning of the zoogeo-
graphic research engine (comparison of taxon ranges, as-
sembly of faunal lists, determination of local endemism,
assembly of maps of species richness and sampling inten-
sity, assembly of lists of species from predefined parts of
the TEP), the functioning of the interactive library, the
database of images, and credits to contributors.

Acknowledgments

The “Acknowledgments” section recognizes support
from STRI, funding, government permissions, logistical
support, assistance collecting fishes, identification of speci-
mens and reviews of section, databases, Spanish transla-
tions, images and illustrations, database management, and
digital image preparation.

THE FISHES
A Page for Each Species, Genus, and Family

Information on the members of the fauna is provided
through interlinked species, genus and family pages. Gen-
era and species are ordered alphabetically within each
family, with families being arranged in “phylogenetic”
order. “The Fishes” module provides access to Species,
Genera, and Families pages by browsing within each taxo-

3% Smithsonian Tropical Research Institute

What Fish is That? The Fishes

Information

Ds Ross Robertson & Gerald R*Allen

Copyright 2008 - Smithsonian Tropical

Prograrnmed

FIGURE 1. Opening screen and “Home” module.

NUMBER 38 e 199

Shorefishes of the Tropical Eastern Pacific
Online Information System

ee
Library Random Images Glossary

Research Engine

| General Information | contributors |

| What Fish is That? } copyright Notice |

E Library | Eeniglages ‘il
English

Research Engine

Programmes By

Cove b)

iSmithsonian Tropical Research Institute

arch Institute
Pty Ltd

nomic level, browsing from within a Systematic Tree (with
optional alphabetic or systematic ordering, and optional
use of common or scientific names), browsing from within
a Book Mode (species within genera within families), or
user-selection of level and taxon from pull-down lists.
Family and genus pages include a brief introduction
to systematics, biology, global geographic distribution,
and an estimate of the number of genera and species
worldwide and present within the TEP; a text description

of distinguishing morphological features—black text in-
dicates the least distinctive features for identification
purposes, red text indicates important features, and red
text with yellow high-lighting shows the most important
features (see Figure 3); a database map of the taxon’s
range limits distribution in the TEP (assembled from the
distributional maps of component species) and a list of
component genera and species with links to their pages;
an image of a representative species that has a key feature

200 © SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

2 P Updated: 08/04/2008
Shorefishes of the Tropical Eastern Pacific Version: 1.0.4.53

Online Information System

26 Smithsonian Tropical Research Institute

Contributors | Glossary | Settings

General

= What Fish is That? The Fishes i Random Images Glossary Research Engine
Information

Introduction - Features & User Guide - Acknowledgments

INTRODUCTION TO THE TROPICAL EASTERN PACIFIC AND ITS SHOREFISH FAUNA

ae ; ; 1. THE TROPICAL EASTERN PACIFIC (TEP)

1. THE TROPICAL EASTERN PACIFIC (TEP) j ; : : ;
We cover the marine biogeographic region known as the Tropical

2. OCEANOGRAPHY AND MARINE HABITATS OF THE TEP Eastern Pacific (TEP), which encompasses the continental shoreline that
2.1 Climatic variation in the region | extends south of Magdalena Bay (~ 25°N) along the outer coast of
2.2 Rainfall and ocean salinity ; | southern Baja California, throughout the Gulf of California, and down the
2.3 Primary production and coastal upwelling systems | continental coastline to about Cabo Blanco (4°S) in northern Peru. This
2.4 oaen See eis one TEP region also includes five offshore islands and groups of islands - the
s A eit cra ties aidan tied nee | Revillagigedos, Clipperton, Cocos, Malpelo and the Galapagos. Politically
47 Rocky Sadicorall rears in the TEP | the region spans all or part of the Pacific coasts of 10 Central and South

: Z American countries: (most of) Mexico, Guatemala, El Salvador, a small

3. THE SHOREFISH FAUNA | part of Honduras in the upper reaches of the Gulf of Fonseca, Nicaragua,
3.1 A short history of taxonomic studies | Costa Rica, Panama, Colombia, Ecuador, and northern Peru, as well as a
3.2 Major modern identification guides } tiny piece of French Polynesia in the form of Clipperton Island. The
3.3 Global Online resources northern and southern continental limits of this region are defined by
3.4 Systematic order in which fishes are arranged in this system cold currents that flow from the poles along the continental coasts
3.5 Names of Fishes | towards the equator and then move away from the coast towards the

3.5.1 Scientific | central Pacific at about these points. The northern quarter of the Gulf of
3.5.2 Common names California also included as part of this tropical region even though it has

a more subtropical to temperate environment and a fish fauna with

2 NEMSNMS INE NACCIEIO) C)e MSE Sale) Sareiles significant affinities to the fauna of the temperate Californian Province.

4.1 Use of environments and habitats

4.2 Reef-associated fishes

4.3 Soft-bottom fishes

.4 Water-column fishes

4.5 Use of environments of differing salinities
4.6 Depth distribution patterns
4.7 Fishes dietary groupings
4.8 Modes of reproduction
4.9 Longevity and size

PAHAHRHHHHHA

5. ZOOGEOGRAPHY OF THE SHOREFISH FAUNA

-1 Sc setifc studies of TEP zoogeography

2 Resident and vagrant species

3 The size of the fauna

4 Relationships of the fauna to the faunas of other areas

-5 Distribution of the fauna in different climate zones

‘6 Variation in species richness and local endemism throughout the TEP
7 Zoogeographic subdivisions of the TEP

S. 7.1 One, two or three continental provinces?

5.7.2 Continental and island components of the regional fauna

5.7.3 An ocean-island province?

UU oO

FIGURE 2. Opening screen from the “General Information” module.

aie Smithsonian Tropical Research Institute

= ae es

Species Information
Heterodontiformes - Heterodontidae - Heterodontus

Heterodontus francisci

Similar Species (2) Y | This Genus species (3)

ci (Girard, 1855)

Hom shatk — Pacific hom-sharh

Head high, conical; snout piglike: mouth small, anterior; a low bony ridge above each eye that
ends abruptly at rear; space between eyes deeply concave; nasal grooves before mouth; front

Shorefishes of the Trop

| Eastern Pacific
Online Information System

Ubrary Random Images

- Heterodontus francisci

High Resolution Map

Map Color Settings

Book mode off

Heterodontus

— =

Habitat Depth

— —_———$—$—$—<—— |

Reproduction Zoogeography Range

teeth on both jaws with 1 large central point and a small pomt at each side on base of tooth: 5

gil sits, fust enlarged, 23 over pectoral fin; 2 dorsal fins, each with spine at front,

finlonign OVer pedioral Base! skin denticles on flank small ( 200/em? in a

Dark to

g brightly colored. with

datk saddles

Size: 122 om,

Habitat: rocky and sandy habitats, and macroaigal beds.

Depth: 1-150 m, usually 2-11 m.

California to the westem and NE Gulf of California; possibly Eousdor and Peru.

FIGURE 3. Example of a species page.

} and smooth.

IUCN Red List
© Listed (S)
© Data deficient (S)

cITes
© Not listed (S)

Feeding

Conservation
status

NUMBER 38

201

202 e

overlay indicating diagnostic features of the taxon that dis-
tinguish it from similar taxa; and comparisons with simi-
lar taxa. To assist in distinguishing look-alike fishes, each
taxon page includes a button-link that allows the user to
compare images (with key feature overlays) of such taxa.
Each page also includes a list of designated similar fishes (at
the same taxonomic level), with links to their taxon pages.

Species data pages are similar to genus and family pages
but also include multiple images (e.g., juvenile, female, male,
color morphs, specific morphological characteristics) and ac-
cess to a downloadable list of species zoogeographic and eco-
logical attribute data. For example, the Zoogeography tab
includes Global endemism, a species global-scale distribution
and its occurrence outside the TEP; Regional endemism, dis-
tributions of species within the TEP, including TEP endemics
(species that occur only in the region or have the great bulk
of their distributions within it), temperate eastern Pacific en-
demics (whose distributions are primarily to the north and
south of the TEP, in the Californian and Peruvian provinces),
eastern Pacific non-endemics (species that have populations
outside the eastern Pacific, for instance circumtropical spe-
cies). Categories relating to the distributions of species within
the TEP include the occurrence of endemic and non-endemic
species at offshore islands and/or the continental shore,
whether TEP endemics are endemic to the offshore islands
(and which islands) or to the mainland, and to which of the
three mainland provinces (or combinations thereof) each
continental TEP endemic is restricted. Attributes for Climate
zone and Residency (whether the species appears to be a resi-
dent or a vagrant in the region) are also included.

Other species ecological attributes that are presented
include the following:

¢ the known maximum total length of each species;

© a species’ maximum and minimum depth of occurrence;

e the salinity of environment(s) in which a species occurs;

e the specific habitat(s) a species uses (as well as habitat cat-
egories as defined by FishBase, see www.FishBase.org);

¢ whether a species is restricted to inshore waters or oc-
curs in offshore, oceanic conditions;

¢ the position in the water column at which a species lives
(e.g., bottom, surface);

© a species’ feeding group (e.g., carnivore);

* items in a species’ diet (e.g., fishes, pelagic crustaceans,
microalgae);

* a species’ reproductive mode (e.g., different types of
eggs, live birth); and

* a species’ CITES and IUCN REDLIST status.

When information is available (e.g., for diet) for a spe-
cies itself, an “S” is given after the value in the database. In

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

cases for which such species-level information is lacking,
the page displays information for the genus (indicated by
“G”), or for the family (“F”) if there is no information for
the genus.

Taxon pages includes direct links to external websites
concerning the same taxon in the following external on-
line sources: William Eschmeyer’s Catalog of Fishes (www
.calacademy.org/research/icthyology/catalog), which pro-
vides comprehensive up-to-date data on the systematics
of fishes; FishBase (www.fishbase.org), which covers a
variety of aspects of the biology of fishes; ITIS, the Inter-
national Taxonomic Information System (http://www..itis
.gov) and WoRMS, the World Register of Marine Species
(http://www.marinespecies.org), both of which focus on
scientific names of fishes; and OBIS, the International Bio-
geographic Information System (http://www.iobis.org),
which aggregates geo-referenced databases of collection
records of fishes.

WHAT FISH IS THAT?

This module facilitates identification of unknown
fishes using four distinct tools (Figure 4).

Find a Fish

This tool allows users who are not scientifically
trained to identify an unfamiliar fish by choosing among
the following in any order or combination, with the abil-
ity to back-up steps: Where was it?—select location and
size of area in question on a database map—and combi-
nations of Body Shape, Color Pattern, and Colors. Each
step narrows the list of possibilities, with each species on
the possibilities list linked to its image, and hence to its
species page.

Identification Keys Search

Illustrated dichotomous keys are provided for the genera
and species in the two families with the largest number of spe-
cies: Gobiidae (88 species in 27 genera) and Sciaenidae (82
species in 26 genera). Search results link to species pages.

Compare Images of Fishes

This function allows simultaneous comparison of im-
ages of any two to six families, genera, or species selected.
The feature enables users to compare “apples” with “or-
anges,” whereas the comparison of designated similar taxa
on taxon pages limits users to comparing only “apples.”
Resultant images are linked to taxon pages.

NUMBER 38 ¢* 203

Shorefishes of the Tropical Eastern Pacific
Online Information System

Se Smithsonian Tropical Research Institute

ed

te cs eN
ee
<

eB ST aa ok zie ae

“General ke memes
Tnfornaton What Fish is That? The Fishes Random Images Glossary Research Engine

What Fish is That? Find a Fish |

Ablennes hians

Aboma etheostoma

Abudefduf concolor

Abudefduf declivifrons
Abudefduf troschelii
Acanthemblemaria atrata
Acanthemblemaria balanorum
Acanthemblemaria castroi
Acanthemblemaria crockeri
Acanthemblemaria exilispinus
Acanthemblemaria hancocki
Acanthemblemaria macrospilus
Acanthemblemaria mangognatt
Acanthemblemaria stephensi
Acanthistius pictus
Acanthocybium solandri
Acanthurus achilles

Acanthurus guttatus
Acanthurus nigricans

Acanthurus triostegus triostegu Selection Criteria nil & = remove criterium
Acanthurus xanthopterus

Achirus klunzingeri Where was it? Body Shape Color Pattern Colors
Achirus mazatlanus -

<
Total: 1287

ey] Map Color Settings

Paintbrush
(1=~1? x 1?)
0.5

1
15

OO0O00000

O0®

OO

Clear Map

Report: includes selection criteria and included-species list.

Sort species list: © Systematic O Alphabetic

FIGURE 4. Screen capture from the “What Fish Is That?” module.

204 e

Common Names Search

Searches can be made for families, genera, and species
from pull-down lists of common names, with results linked
to taxon pages. The systematic tree or taxonomic hierarchy
(see “The Fishes” module) also functions with the use of ei-
ther common names or scientific names. The use of common
names in this hierarchy helps users who are not scientifically
trained to appreciate the relationships among fishes.

GLOSSARY

An interactive glossary of taxonomic terms is provided
that uses a combination of images and text to explain ba-
sic terms relating primarily to morphological characteristics
that are used in the identification of fishes. In addition the
usage of scientific jargon has been reduced as much as pos-
sible throughout the taxon pages by using simple descriptive
phrases from everyday English to replace technical terms.

RESEARCH ENGINE

This module provides a variety of types of zoogeo-
graphical data and the ability to generate maps and site-
specific species-lists based on complex queries constructed
by the user (Figure 5).

Taxon Range Maps

This feature provides overlaid displays of the regional
ranges of up to three taxa (species, genera, families, or a
mixture thereof). In addition, maps can be generated of
the geographic distribution of all species-range centroids
(both paint and point data) and of all geo-referenced sam-
pling points in the system’s database.

Species Richness Displays

This feature provides maps with color-coded overlays
of patterns of variation in species richness throughout
the region. Those patterns include richness of individual
families and richness of species in specified “functional
groups” (e.g., species sharing one or more biogeographic
and ecological attribute). Richness displays can indicate
either absolute richness (number of species) or relative
richness (number of species as a percentage of the local
fauna). A display of relative sampling intensity indicates
the number of species recorded at minimally one site
within each unit of area (1° of latitude X 1° of longitude)
as a percentage of the number of species whose ranges
encompass that unit.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Species — List Assembly

Family and genus lists can be constructed for single lo-
cations. Species lists for “functional groups” of fishes for a
particular location can be constructed using any combina-
tion of biogeographic and ecological attributes. Species lists
include both single-location lists and lists of species found or
not found at two locations. The spatial scale of a location in
such a search varies from a single island to an area of vari-
able shape and size, to the entire TEP or map. Locations are
defined by the user employing a library of approximately
300 preformed templates that include geographic entities
(e.g., shoreline, continental shelf, named gulfs), habitat fea-
tures (e.g., mangroves, rocky shores, upwelling areas), is-
lands (individual and archipelagos), biogeographic entities
(provinces of the TEP), political areas (Exclusive Economic
Zones and parts thereof), and marine reserves (individual
reserves, combine country reserves). In addition, quadrants
of varying sizes (12 groups ranging from 0.5° latitude x
0.5° longitude to 24° X 24°) used by the map of the Find a
Fish tool in the “What Fish Is That?” module provide ap-
proximately 5,000 additional [square] templates.

Unconfirmed/Confirmed Occurrences

Single-area species lists indicate both likely occur-
rences (species whose ranges include the selected area) and
confirmed occurrences (species with at least one collection
record in the same area).

Local Endemism Indicator

This feature provides a list of species found only
within one or two template areas, and nowhere else on the
system’s map.

List and Map Exports

Lists and Maps produced by searches are exportable/
printable. Lists may be arranged alphabetically or system-
atically (genera and species arranged alphabetically within
families arranged in systematic order).

Species Range Maps and Range Data

A database map on each taxon page incorporates two
types of data: a two-dimensional painted representation
of the geographic range based on museum and literature
records of occurrence and range maps, and our own field
surveys in Mexico, El Salvador, Costa Rica, Panama, Co-
lombia, Ecuador, the Revillagigedos, Clipperton, Cocos,

NUMBER 38 e 205

- Taxon - Range Species - Richness
Research Engine Maps Displays

Interactive Mapper

|O Local Endemism: members of regional fauna found
either exclusively inside or exclusively outside A (A = one or
more areas defined on map)

Only inside 4
Only outside A

| © Taxonomic/Functional Group Lists

| O all Species
| © all Genera
| O All Families
Eee © Species Functional Group (define)

Length Max < 100
Feeding - Feeding Group is Carnivore
Reproduction - Egg Type is Benthic

Species Functional Groups Options

- Size

- Habitat

- Depth

- Feeding

- Reproduction

- Zoogeography

- Range

- Conservation status

\Define question to generate list
|O Inside A

| © Combined total in A and B

| © Present in both A and B
| O Present in A but not in B

Define Area(s) on Map using variably-sized quadrats or by
| adding/removing area-templates

Single Quadrat

Template Category
Gulfofpanama | [J
a= B= MB boch=
| O Areaa © areas

Loading (Cetera) Cote) Coteere
| Map Functions ‘
|

36.37, -7.23
Add Coastline

Report: includes map, group type and species list

Sort taxa list
© Systematic
i@) Alphabetic

FIGURE 5S. Screen capture from the “Research Engine” module.

206 e

Malpelo, and the Galapagos; and points indicating site
records from museum collections, the scientific literature,
and our own field surveys. Geographic-range statistics
derived from these two offsets of data and presented on
species pages include: latitudinal and longitudinal limits,
ranges and midpoints based on paint data, and, separately,
site records; data on range characteristics derived from
paint data have been adjusted in species that occur in the
eastern Pacific beyond the limits of the system’s base map;
habitat area, based on number of painted pixels in the spe-
cies range map; and separate range-area polygons with
centroids based on painted data and site record data.

Continental Ranges

Comprehensive faunal lists exist for few locations, and
large sections of the coastline and continental platform of the
TEP remain poorly sampled, a situation that will not be recti-
fied in the near future. Hence the range of most species on the
continental shelf is derived from data on the northern and
southern limits of occurrence. Thus painted areas on taxon
page maps represent the potential range and potential habi-
tat area, and a species is assumed to be present in appropriate
habitat anywhere between those limits. Exceptions include
species that are known to have wide gaps in their distribu-
tions, such as some well-known anti-tropical species. Those
gaps are represented in the range maps of such species.

Habitat Area Calculations

Maps constructed for the determination of habitat areas
incorporate information on habitat usage and depth range as
well as the extent of the geographic range. Continental areas
of range maps were modified to exclude large areas of habi-
tat that was inappropriate for the particular species; for ex-
ample, shorelines composed primarily of sand and mud were
excluded from ranges of reef-fishes and rocky shores were ex-
cluded from ranges of fishes living on beaches or in lagoons
and mangroves. The depth ranges of individual species were
also taken into account: ranges of demersal species restricted
to very shallow water (less than ~20 m depth) are indicated
by lines that follow coastlines. For habitat area calculations
of such species, the coastal strip of habitat was taken to be
1 km wide. Ranges of coastal species found in deeper water
on the continental and insular shelves are divided into three
groups: those occurring down to ~60 m have maps that span
the inner continental shelf; species that are limited to depths
below ~60 m occur on the outer shelf; and the third group
has depth ranges (and maps) that span the entire shelf. Maps
for pelagic species variously include parts of the shelf and/or
open ocean, depending on the biology of the species.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Mercator Projection Distortions and Adjustments
to Habitat Area Calculations

Mercator projection maps, such as that used in this
system, incorporate distortions of both latitude and longi-
tude that affect estimation of habitat area. In such projec-
tions lines of longitude are shown as parallel rather than
converging with increasing latitude, and lines of latitude
diverge with increasing latitude instead of remaining a
fixed distance apart. When calculating the habitat area for
each species those two distortions were taken into account
by making appropriate adjustments to the areas of indi-
vidual pixels in different latitudinal bands. Range polygon
areas were calculated independently using the GIS (Geo-
graphic Information System) ArcInfo system.

Cleanup of the Geo-Referenced Records Database

Both the scientific literature and databases from muse-
ums inevitably include erroneous records as a consequence
of misidentification offishesand sites, as wellas bookkeeping
errors. In addition museum specimens of demersal species
include not only individuals collected in demersal habitats
but also larvae collected in the open sea far from adult
habitat, and, in some cases, far from the known adult geo-
graphic range (Robertson, 2008). Records from the multi-
source database of ~67,000 collection site records included
here that were adjacent to the currently known limits of
the geographic range were used to adjust (by expanding)
those ranges. However, we removed from the systems
database those “suspect” records that were well outside
the known habitat and geographic ranges of the “adult”
phase of each species, based on extensive records from
other sources, the biology of the species, and expert deter-
minations of ranges. This cleanup process reduced the size
of the database by approximately 6%. Points outside the
current range were retained for some species that, because
of overfishing, have had their historical ranges reduced.
For example, historically the mackerel Scomberomus con-
color, which currently occurs only in the northern Gulf
of California, occurred throughout that gulf and also off
California, USA (B. B. Collette, National Marine Fisher-
ies Service Systematics Laboratory, personal communica-
tion, 2008).

LIBRARY

The library database (Figure 6) includes 1,143 cita-
tions. The citation for its original description is included
for each species, along with citations of revisions of genera
and families. Other citations include local and larger scale

3 Smithsonian Tropical Research Institute

NUMBER 38

Shorefishes of the Tropical Eastern Pacific
Online Information System

General
Information

Define search

Keyword Search
|Bussing|

Random Images

Taxon Search
ispeciesiLevel ius

Ablennes hians

te [Title —

Bussing, W.A.

1972|Halichoeres aestuaricola, a replacement name for the tropi ystern Pacific
labrid fish, Iridio bimaculata Wilson, with a redescription based on new
material.

a a

Glossary Research Engine

Acanthemblemaria
stephensi

“~

Brenesia (Nat. Mus. Nac. Costa Rica), Vol. 1,
pp.3-8

Bussing, W.A.

1981/Elacatinus janssi, a new gobiid fish from Costa Rica.

Bussing, W.A.

1983/4 new tropical eastern Pacific labrid fish, Halichoeres discolor endemic to
Isla del Coco, Costa Rica.

Revista de Biologia Tropical, Vol. 29 issue 2,
[pp.251-256
Revista de Biologia Tropical, Vol. 31 issue 1,
ipp.19-23

Bussing, W.A.

1983/Evermannia erici, a new burrowing gobiid fish from the Pacific coast of Costa
Rica.

Revista de Biologia Tropical, Vol, 31 issue 1,
pp.125-131

Bussing, W.A.

1990/New species of gobiid fishes of the genera Lythrypnus, Elacatinus and
Chriolepis from the eastern tropical Pacific.

Revista de Biologia Tropical, Vol. 38 issue 1,
pp.99-118

1991/4 new genus and two new species of tripterygiid fishes from Costa Rica.

Revista de Biologia Tropical, Vol. 39 issue 1,
pp. 77-85

A new species of eastern Pacific moray eel (Pisces: Muraenidae).

fe lRef 1 ID|Author

il Bussing, W.A.

q Bussing, WA.
Bussing, W.A,
Bussing, W.A.
Bussing, W.A.

Bussing, W.A.

lacement name for the tropical eastern Pacific
labrid fish, Iridio bimaculata Wilson, with a redescription based on new
material.

Revista de Biologia Tropical, Vol. 39 issue 1,

‘Nat. Mus. Nac. Costa Rica), Vol. 1,

Elacatinus janssi, a new gobiid fish from Costa Rica.

Revista de Biologia Tropical, Vol. 29 issue 2,
pp.251-256

& new tropical eastern Pacific labrid fish, Halichoeres discolor endemic to
Isla del Coco, Costa Rica.

Evermannia erici, a new burrowing gobiid fish from the Pacific coast of Costa
Rica.

[i224 New species of gobiid fishes of the genera Lythrypnus, Elacatinus and
Chriolepis from the eastern tropical Pacific.

fissile new genus and two new species of tripterygiid fishes from Costa Rica.

36 Bussing, W.A.

j1991 & new species of eastern Pacific moray eel (Pisces: Muraenidae).

| Report On:

| Citation

Species Linked to

| Report Type:

Create Report

FIGURE 6. Screen capture from “L

arch Institute
ems Pty Ltd

© Copyright 2008 - Smithsonian Tropical R
Programmed by Coeus Knowled

2s

ibrary” module.

Revista de Biologia Tropical, Vol. 31 issue 1,
pp.19-23

Revista de Biologia Tropical, Vol. 31 issue 1,
pp-125-131

Revista de Biologia Tropical, Vol. 38 issue 1,
pp.99-118 a
Revista de Biologia Tropical, Vol. 39 issue 1,
pp. 77-85

Revista de Biologia Tropical, Vol. 39 issue 1,

207

208 e

lists of species; identification guides to species; and pub-
lications about the biology and zoogeography of species
inside and outside the TEP. Each citation is linked to the
species it discusses (and hence to appropriate genera and
families).

Exportable lists can be generated for:

e Citations linked to individual families, genera, and
species.

e Species linked to a particular citation.

e Citations linked to a particular author, date, or source
journal.

e The entire bibliography arranged alphabetically by au-
thor name.

RANDOM IMAGES

The image database incorporates 2,927 images. These
include 2,346 color photographs that cover 83% of the
fauna (1,068 of 1,237 species). In comparison, the 1994
book from which this system was developed included color
images of 683 species and treated only 768 species.

This module presents color images in a randomized
order.

Digital Manipulation of Images

The user should assume that all illustrations used in
this system have been digitally manipulated to some ex-
tent, to increase their utility as identification aids. Such ma-
nipulation includes cropping, image sharpening, changes
in lighting and contrast relationships of different parts of
individual subjects, changes of subject-to-background con-
trast, changes of background to enhance subject visibility,

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

the (occasional) combination of multiple images of fishes
in a montage to provide examples of variation in color pat-
terns within the same image, and minor repairs to fin mem-
branes and removal of body blemishes (scratches, minor
cuts, blood spots) that resulted from capture handling.

Image Credits

All images are accompanied by a relevant ownership
credit, copyright notice, and usage notice. Each image is
accompanied by a link to either the owner’s e-mail contact
or website.

ACKNOWLEDGMENTS

Funding throughout the development of this system
(1990-2008) was provided the following Smithsonian In-
stitution (SI) sources: the SI Women’s Committee, the SI
Seidell Fund, the SI Latino Initiatives Fund, the SI Marine
Science Network, and the Smithsonian Tropical Research
Institute.

LITERATURE CITED

Allen, G. R., and D. R. Robertson. 1994. Fishes of the Tropical Eastern
Pacific. Bathurst, Australia: Crawford House Press.

. 1998. Peces del Pacifico oriental tropical. CONABIO, Mexico.

Robertson, D. R. 2008. Global Biogeographic Data Bases on Marine
Fishes: Caveat Emptor. Diversity and Distributions 14:891-892.

Robertson, D. R., and G. R. Allen. 2002. Shorefishes of the Tropical
Eastern Pacific: An Information System. CD-ROM. Balboa, Pan-
ama: Smithsonian Tropical Research Institute.

. 2006. Shorefishes of the Tropical Eastern Pacific: An Informa-

tion System. Version 2. DVD-ROM. Balboa, Panama: Smithsonian

Tropical Research Institute.

Nephasoma pellucidum: A Model
Species for Sipunculan Development?

Anja Schulze and Mary E. Rice

Anja Schulze, Texas AGxM University at Galves-

ton, 5007 Avenue U, Galveston, Texas 77551,
USA. Mary E. Rice, Smithsonian Marine Station
at Fort Pierce, 701 Seaway Drive, Fort Pierce,
Florida 34949, USA. Corresponding author: A.
Schulze (schulzea@tamug.edu). Manuscript re-
ceived 9 June 2008; accepted 20 April 2009.

ABSTRACT. Recent developments in metazoan phylogeny, especially with regard to
the position of the Sipuncula in the annelid clade, have sparked a renewed interest in
sipunculan development. If Sipuncula are annelids, they must have secondarily lost seg-
mentation. By comparison with segmented annelids, they could provide important clues
for the evolution of segmentation. A sipunculan model species is needed to examine
fundamental developmental processes. Here we describe the development of Nephasoma
pellucidum and explore its potential as a model species for sipunculan development.
Like other sipunculans, N. pellucidum produces eggs with a thick, porous, multilayered
egg envelope. Cleavage in N. pellucidum is spiral, holoblastic, and unequal. The spe-
cies shows the most common, and likely ancestral, developmental mode in the group.
Its life cycle includes a lecithotrophic trochophore and a planktotrophic pelagosphera
larva. The trochophore is enclosed in the egg envelope, with cilia growing through the
envelope’s pores. The trochophore larva metamorphoses into the pelagosphera larva at
approximately 60 h. Pelagosphera larvae reached metamorphic competence at about five
weeks. Metamorphosis to the juvenile was induced by supplying sediment that had been
inhabited previously by conspecific adults. Juveniles were observed for several weeks.
We conclude that N. pellucidum is a good model species for sipunculan development,
although rearing conditions in the laboratory still need to be optimized.

INTRODUCTION

During the past two decades, our understanding of metazoan relationships
has changed radically, starting with the first use of ribosomal RNA sequences
for phylogenetic analysis (Field et al., 1988). Many taxa for which evolutionary
origins have long been mysterious or controversial can now be placed with more
certainty into the metazoan tree of life (Dunn et al., 2008; Halanych, 2004).
Among those, two groups that were long regarded as distinct phyla have been
absorbed into the Annelida: the Echiura, or spoon worms, and the Siboglinidae,
previously called Pogonophora and Vestimentifera (McHugh, 1997; Rouse and
Fauchald, 1997).

The Sipuncula, commonly known as peanut worms or star worms, have hada
complex taxonomic history but now appear to be following the same route as the
echiurans and siboglinids. Nearly 50 years after Hyman (1959) affirmed phylum
status for the group, recent authors place them into the annelid clade, based on

210 e

phylogenetic analyses of mitochondrial gene order (Blei-
dorn et al., 2005; Boore and Staton, 2002), sequence data
from several genes (Struck et al., 2007), and expressed se-
quence tags (Dunn et al., 2008). Although there is a grow-
ing consensus on the annelid affinities of sipunculans, it
remains to be determined which of the incredibly diverse
annelids is the sister group to the Sipuncula. With a simple
body, consisting of a trunk and a retractable introvert with
an array of tentacles at the anterior end, they show little
similarity to any other polychaete group. In the molecu-
lar analyses, support for a sister group relationship with
any other polychaete taxon is low. The monophyly of the
Sipuncula is uncontested, and solid hypotheses of within-
group relationships have been published (Maxmen et al.,
2003; Schulze et al., 2005, 2007; Staton, 2003).

Sipuncula are an interesting case in the field of
“EvoDevo,” or the interface of evolution and development.
Embryonic and larval characters have often been cited as
support for phylogenetic hypotheses. Rice (1985) listed sev-
eral similarities between sipunculan and annelid develop-
ment, such as the larval prototroch and metatroch and the
retention of the egg envelope to form the larval cuticle. She
also noted that in some sipunculan larvae the ventral ner-
vous system develops in paired cords, similar to most poly-
chaetes. On the other hand, Scheltema (1993), comparing
embryos and larvae of annelids, mollusks, and sipunculans,
argued that sipunculan development shows more similarity
with that of mollusks. The development of all three taxa
includes spirally cleaving embryos and a trochophore larva.
A long-held view is that annelid and mollusk embryos can
be distinguished at the 64-cell stage by the arrangement of
the micromeres around the animal pole: they form either an
“annelid cross” or a “molluskan cross.” Reproducing Ger-
ould’s (1906) drawing of the embryo of Golfingia vulgaris
with a molluskan cross, Scheltema concluded that sipuncu-
lans and mollusks were sister groups. However, Maslakova
et al. (2004) showed that the annelid and molluskan crosses
are far from universal within the respective taxa and prob-
ably hold no phylogenetic significance.

The primary reason why few past researchers have
recognized sipunculans and echiurans as annelids is that
adults of both taxa show no sign of segmentation, either
externally or internally. It took advanced techniques in im-
munohistochemistry and confocal laser scanning micros-
copy to demonstrate segmentation in the nervous system
of echiuran larvae (Hessling and Westheide, 2002). Similar
techniques initially failed to show segmentation in sipun-
culan larvae (Wanninger et al., 2005) but a recent study
showed a segmental mode of neural patterning in the early
pelagosphera stage (Kristof et al., 2008).

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

If the Sipuncula fall into the annelids, they must have
secondarily lost segmentation in the later larval stages and
the adult. If no morphological segmentation is evident,
what happened with the molecular pathways responsible
for segment formation in other annelids? By comparison
with other species, sipunculans are valuable for the identi-
fication of the genetic and cellular basis of segment forma-
tion in annelids.

The recent developments in metazoan phylogeny have
thus sparked a renewed interest in sipunculan develop-
ment. A model species is needed to study fundamental
developmental processes. A good model species has to
be readily available, be easy and inexpensive to maintain
in the laboratory, lend itself to a variety of examination
techniques, and be representative for its taxonomic group.
Here we describe the development of Nephasoma pel-
lucidum and explore its potential as a model species. N.
pellucidum is a relatively common species that inhabits
cracks and crevices in hard substrates in shallow warm
waters. The species exhibits the most common develop-
mental mode within the Sipuncula, which includes a leci-
thotrophic trochophore stage and a planktotrophic pela-
gosphera larva (Rice 1967, 1975a, 1975b, 1976, 1989).
We have accumulated these data between 1972 and 1984
and, more recently, between 2003 and 2006.

MATERIALS AND METHODS

Specimens of Nephasoma pellucidum were collected
from numerous localities offshore from Fort Pierce, Flor-
ida, extending from Capron Shoal and Pierce Shoal 4 and
6 miles, respectively, southeast of the Fort Pierce Inlet to
the Sebastian Pinnacles approximately 32 miles north
of the Inlet. At the Pinnacles, worms inhabited rubble
of oculinid coral at depths of 70 to 100 m, whereas on
the more southern shoals they occurred in depths of 9 to
15 m in rubble composed of mollusk shells, sand dollar
tests, and rocks. Occasionally specimens were found in the
Fort Pierce Inlet in intertidal clumps of oyster shells. The
worms were carefully removed from the rubble with ham-
mer and chisel. Multiple adults from each collection were
kept in glass dishes in approximately 200 mL seawater
at room temperature. Spawning occurred in the lab, gen-
erally after changing the water. Whenever eggs were ob-
served in the culture dishes, they were pipetted into a clean
dish and observed for development. Larval cultures were
kept until the larvae either died or metamorphosed. Water
was changed at least every two days. The larvae were peri-
odically fed with unicellular algae or diatoms (Isochrysis,

Dunaliella, or Nanochloropsis). To induce metamorpho-
sis, larvae were pipetted into dishes with muddy sediment
previously inhabited by conspecific adults.

For scanning electron microscopy (SEM) and transmis-
sion electron microscopy (TEM), specimens were fixed in
2.5% glutaraldehyde in Millonig’s phosphate buffer (Mil-
lonig, 1964) for at least 1 h and up to several days at 4°C.
Fixation was followed by three washes in a 1:1 mixture
of Millonig’s phosphate buffer and 0.6 M sodium chlo-
ride and postfixation in 1% osmium tetroxide (1:1:2 mix
of 4% OsO, : Millonig’s buffer : 0.75 M NaCl). Samples
were then dehydrated in an ethanol series up to 100%.
For SEM, they were critical point dried and mounted on
SEM stubs using double-sticky tape and viewed in either
a Nova Scan or a JEOL 6400 Visions scanning electron
microscope. Images were either scanned from negatives
or stored digitally. For TEM, the dehydrated specimens
were transferred to propylene oxide and subsequently em-
bedded in Epon resin and sectioned. Thin sections were
stained with uranyl acetate and lead citrate and viewed in
a JEOL 100CX transmission electron microscope.

RESULTS
GAMETES

The spermatozoan of Nephasoma pellucidum is of the
primitive type according to Franzén’s classification (Fran-
zen, 1958). The nuclear region is rounded and capped by
a doughnut-shaped acrosome with a central nipple-like
protuberance. The head, including nucleus and acrosome,
measures 1.5 X 1.7 wm. Posterior to the nucleus, four
mitochondrial spheres are arranged in a circle, from the
center of which extends the flagellum (Figure 1A).

The egg at the time of spawning is spherical, measur-
ing 105 wm in diameter (Figures 1B, 2A). In direct light
the surface appears opalescent, and the color is pale gray.
The egg envelope, up to 6 pm in thickness, is multilayered
and perforated by numerous pores (Figure 3).

SPAWNING

As in most sipunculans, sexes are separate; eggs and
sperm are spawned freely via the nephridipores into the
surrounding water where fertilization occurs. From data
accumulated on spawning in the laboratory, two spawn-
ing peaks are evident: one in the spring (April-May) and
the other in the fall (September-November). Observations
of spawning were carried out on animals in the laboratory,
usually for a period of one month after collection from

NUMBER 38 e¢ 211

the field: 139 spawnings were recorded over a period of 8
years (1972-1980), and spawning occurred every month
of the year except January. Although a few animals were
observed to spawn after maintenance in the laboratory
for as long as 18 months, 88% of the spawnings were re-
corded within 30 days of collection.

CLEAVAGE

The eggs at spawning may be arrested in the first
meiotic metaphase, or they may possess an intact germi-
nal vesicle. In the latter case, the germinal vesicle breaks
down soon (within at least 30 min) after spawning, re-
gardless of whether the egg is fertilized. Within 40 min
after fertilization (23°C), the first polar body is formed
(Figure 2B), and at 55 min the second polar body makes
its appearance. The first cleavage, occurring at 90 min,
is unequal, the CD cell exceeding the AB cell in size (Fig-
ure 2C). The next three cleavages occur at approximately
half-hour intervals, and the 16 cell stage is attained
within 3 h after fertilization. The third cleavage, from
4 to 8 cells, is spiral and unequal. The A, B, and C cells,
all approximately the same size, divide simultaneously,
preceding the initiation of division of the larger D cell by
about 1 min and completing their divisions 5 min before
that of the D cell. In the 8 cell stage, the micromeres
and macromeres of the A, B, and C quadrants are ap-
proximately the same size, the C sometimes being slightly
larger, but all are smaller than the d cell which, in turn, is
smaller than the D cell.

After the first few cleavages, the divisions are more
frequent, and by 7 h after fertilization the egg has devel-
oped into an early blastula; cilia from the prototrochal
cells protrude through the pores of the egg envelope and
the embryo begins to rotate slowly on the bottom of the
container. By 16 h the embryos are swimming throughout
the dish, no longer confined to the bottom. At this time the
stomodaeal invagination is evident, and the embryos show
the first signs of positive phototropism (Figure 2D).

TROCHOPHORE: MORPHOLOGY AND METAMORPHOSIS
TO THE PELAGOSPHERA

By 48 h the embryo has reached the stage of trocho-
phore. The shape has changed from spherical to oval, as
a result of a slight posttrochal elongation (see Figure 1C).
A pair of dorsolateral red eyespots is present in the pre-
trochal hemisphere. Prototrochal cavities are evident to the
inner side of the prototroch cells, and the gut is differenti-
ated into three regions: esophagus, stomach, and intestine.

212 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 1. Scanning electron micrographs showing development of Nephasoma pellucidum (scale bar lengths here in parentheses). A. Sperm
on the surface of the egg (5 pm). B. Egg with sperm on surface (20 wm). C. Trochophore larva; note cilia extending through egg envelope
(20 pm). D. Early pelagosphera larva (20 pm). E. Fully formed pelagosphera larva, ventral view (50 wm). F. Head of a pelagosphera larva,
lateral view (20 ym). G. Terminal organ of the pelagosphera larva (10 wm). H. Metamorphosed juvenile (100 pm). I. Tip of juvenile introvert
with tentacle buds and lobes of nuchal organ (50 jm). Abbreviations: at = apical tuft; h = head; Il = lower lip; m = mouth; mc = metatrochal
collar; mt = metatroch; no = nuchal organ; pt = prototroch; t = tentacles; to = terminal organ; tr = trunk. (Images A, B from Rice, 1989:
fig. 4E,F; used with permission)

FIGURE 2. (facing page) Light micrographs showing development of Nephasoma pellucidum (scale bar lengths here in parentheses).
A. Unfertilized egg (20 pm). B. Egg with polar body (arrow) (20 pm). C. Two-cell stage; note size difference between CD and AB blasto-
meres (20 ym). D. Blastula stage; beginning invagination of stomodaeum (arrow) (20 pm). E. Early trochophore; note eyespots at anterior
end (top) (20 pm). F. Trochophore shortly before metamorphosis to pelagosphera (20 xm). G. Early pelagosphera in the process of elonga-
tion (20 zm). H. Fully metamorphosed pelagosphera larva (20 ym). I. Feeding pelagosphera, 10 d old (50 um). Abbreviations: bo = buccal
organ; es = esophagus; in = intestine; lg = lip gland; mt = metatroch; st = stomach; to = terminal organ.

NUMBER 38

213

FIGURE 3. Transmission electron micrographs showing egg envelopes
of Nephasoma pellucidum (scale bars = 1 wm). A. Section through
multilayered egg envelope. B. Cilium growing through pore in egg en-

velope of trochophore larva. Abbreviations: ci = cilium; po = pore.

The trochophore is lecithotrophic, being completely en-
closed by the egg envelope (Figure 2E,F).

Metamorphosis of the trochophore to the pelago-
sphera larva occurs at 60 to 65 h (23°C) and extends over
a period of 6 to 8 h. The body elongates from 140 ym to
250 wm, by an increase in the length of the posttrochal
hemisphere. The lumen of the gut is completed, and mouth
and anus break through the overlying egg envelope (Fig-
ures 1D, 2G). The ventral ciliated surfaces of the head and
the lower lip are formed apparently by an evagination and
expansion of the anterior stomodaeum. Larval organs of
the lower lip become functional: the buccal organ is pro-
trusible, and the pore of the lip glands opens (Figure 2H).
The metatrochal cilia project from a prominent metatro-
chal collar posterior to the mouth and lower lip. As the
retractor muscles become functional, the entire pretrochal
body is retractable into the posterior or posttrochal re-
gion of the larva. The coelom is considerably expanded,
and the posttrochal body is capable of great extension
and contraction. Whereas the posttrochal egg envelope
is transformed into the larval cuticle, the pretrochal egg
envelope is gradually sloughed off, leaving a thin cuticle
covering the head. The terminal organ appears first as an
evagination of the posterior extremity of the trunk and
within a few hours differentiates into a more discrete elon-
gate structure (40 wm) that is retractable into the trunk

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

and provides a temporary attachment for the larva to the
substratum (Figure 1G).

PELAGOSPHERA: MORPHOLOGY, BEHAVIOR, AND
METAMORPHOSIS TO THE JUVENILE

Four regions of the body can be distinguished: head,
metatrochal collar, trunk, and terminal organ (see Figures
1E,F, 2H). The terminal organ is well developed with an
unusually long neck, terminated by a bulbous posterior
expansion (Figure 1G). The terminal organ of a 10-day
larva may be extended to a length one-third that of the
entire larva.

For approximately two weeks after metamorphosis,
the majority of larvae are attached by their terminal or-
gans; some continue to swim, or else attach and swim
intermittently. At two weeks there is a high rate of mor-
tality and, in the absence of substratum, most larvae die
within two months; the maximum survival time of larvae
reared in culture dishes is 103 days. Surviving larvae of
one month of age attain a maximum size of 1.2 mm. At
this age the body proportions have changed, the head be-
ing relatively smaller than in the younger stages. The ex-
ternal body wall is smooth, glistening in reflected light,
and through the relatively transparent body wall the gut is
apparent as an elongate dark yellow stomach and a lighter
yellow recurved intestine, ending at the dorsal anus in the
anterior trunk. Usually larvae are still attached by the ter-
minal organ at these later stages, although some may lie
on the bottom, relatively quiescent. Swimming occurs only
rarely, although metatrochal cilia are still present.

Attached larvae are observed to feed on the substratum
surrounding their points of attachment (Figure 21). The
body may be bent downward so that the ventral surface of
the head is applied to the bottom of the dish, or the body
may be stretched out from the point of attachment paral-
lel to the substratum. In culture dishes in which there is
an algal growth covering the bottom, the area surrounding
the attached larva is often bare, indicative of larval grazing
activity. The area of attachment is often marked by clumps
of feces on which the larva may graze and ingest. Occa-
sionally larvae release themselves from the attachment and
swim or move along the bottom to a new site. Free larvae
sometimes move with head applied to the substratum and
posterior end directed upward, either exploring or feeding
on the bottom. Frequently the terminal organ is placed in
or near the mouth. Older larvae detach and move to new
locations less often than younger larvae. A larval behavior,
common to all sipunculans but of unknown function, is
placement of the terminal organ in or near the mouth.

Metamorphosis of larvae reared in culture dishes
could be induced at the age of 5 to 6 weeks by exposure
to a fine, muddy sediment that had been occupied previ-
ously by adults. Attempts to induce metamorphosis before
this age were not successful. Before metamorphosis, larvae
buried themselves in the sediment and in 3 d underwent
metamorphic changes to the juvenile stage.

The process of metamorphosis is initiated by the loss
of the metatrochal cilia, reduction in the size of the lower
lip, narrowing of the head, and elongation of the pretro-
chal body. At the end of 3 d, both posttrochal and pre-
trochal regions of the body are narrowed and elongated,
the metatrochal collar is reduced, the terminal organ and
lip are partly regressed, the mouth moves to a terminal
position, and dorsal to the mouth a pair of developing ten-
tacular lobes is apparent (Figure 1H,I). These morphologi-
cal modifications, along with the behavioral changes of
initiation of burrowing and cessation of swimming, mark
the beginning of the juvenile stage. Regions of the body of
the juvenile are reduced from the four found in the larva
to two: (1) the broader and longer posterior trunk, formed
from the posttrochal larva, and (2) the more narrow ante-
rior introvert, which is terminated by mouth and develop-
ing tentacles and formed from the pretrochal larva. Similar
to the pretrochal larval body, the introvert of the juvenile
is retractable into the trunk.

The most immediate modifications are found in the
head and metatrochal regions. As the mouth becomes ter-
minal, the dorsal surface of the head is foreshortened. The
ventral lip is lost, but ciliation of the ventral surface of
the head persists to surround the mouth and the ventral
surface of the developing dorsal lobes. On the dorsal head
two heavily ciliated patches that will give rise to the paired
nuchal organ have moved further anteriorly as the head
foreshortens. In the 7- to 9-day-old juveniles the buccal
organ is no longer apparent. One to four rings of sim-
ple hooks appear in the region of the former metatrochal
band. Papillae, already apparent in older larvae, are more
prominent and numerous. Scattered among the hooks, the
papillae are dome shaped and, as seen in scanning elec-
tron micrographs, have central pores from which several
cilia protrude. Papillae of similar structure, but somewhat
larger, cover the entire trunk (Figure 1H). A vestigial ter-
minal organ may persist for one or two weeks. Within two
to four weeks a second pair of rudimentary tentacles ap-
pear ventral to the mouth.

The body wall of the juvenile thickens, losing its
transparency. Externally circular constrictions, also seen
in late larval stages, are more prominent. Juveniles of one
week also show longitudinal “folds” of the body wall,

NUMBER 38 ¢° 215

resulting in a checkered appearance of the integument in
some regions.

DISCUSSION

Nephasoma pellucidum is one of the few sipunculan
species in which the life cycle has been observed from
spawning to juvenile stage. Other species are Siphono-
soma cumanense (Rice, 1988), Thysanocardia nigra
(Rice, 1967), Themiste pyroides (Rice, 1967), Themiste
lageniformis (Pilger, 1987), Themiste alutacea (Rice,
1975c), Phascolion strombus (Akesson, 1958; Wanninger
et al., 2005), and Phascolion cryptum (Rice, 1975c). Most
of these species show abbreviated development, either
omitting both the trochophore and pelagosphera stage, or
omitting the pelagosphera stage, or having a lecithotro-
phic pelagosphera (Rice, 1976). The oceanic, plankto-
trophic pelagosphera larvae of many aspidosiphonid and
phascolosomatid larvae have been recovered in plankton
tows; however, their complete life cycles are unknown
(Rice, 1981; Hall and Scheltema, 1975).

We argue that a model species should show the ances-
tral developmental mode for the taxon. Cutler (1994) con-
cluded that indirect development with a planktotrophic
pelagosphera was ancestral in Sipuncula. The most recent
phylogenetic analyses (Schulze et al., 2007; Schulze and
Rice, 2009) seem to confirm this view. The genera Sipun-
culus and Siphonosoma, which to our present knowledge
only contain species with planktotrophic pelagosphera
larvae, represent the two basal clades in both analyses.
The remaining three major clades have species of Phas-
colosoma and Apionsoma as their basal branches, two
additional genera in which abbreviated development is
unknown.

Of the species listed above, only the life cycle of Si-
phonosoma cumanense includes a planktotrophic pelago-
sphera as N. pellucidum does. Siphonosoma cumanense is
a large, sand-burrowing species. Like other sipunculans, it
survives well under laboratory conditions, when supplied
with sediment and adequate aeration. However, its poten-
tial for use as a model species is limited by two factors.
First, even though the species has a wide geographic dis-
tribution, it is rarely found in significant numbers, and the
establishment of a viable population would require major
efforts. Second, larvae do not seem to be competent to
metamorphose until about 8 weeks old (Rice, 1988).

Nephasoma pellucidum is geographically widespread,
mostly in shallow warm waters, although it does not seem
to be as abundant in most places as at our collecting station

216 e

near Fort Pierce, Florida. Collection of a significant num-
ber of individuals can be time consuming because they
have to be carefully removed from the cracks and crevices
of rubble; the removal process can damage the animals,
often causing their death. After successful retrieval, how-
ever, adults are easy to maintain in laboratory conditions.
Removed from their shelter, they survive in simple glass
bowls without aeration or food supplement for at least a
year, feeding only on the biofilm at the bottom of the dish.
We assume that, left in their shelter or in sediment, with
proper aeration and occasional food supply, they would
survive for years. This assumption is based on the longev-
ity of other sipunculans: individuals of Apionsoma misaki-
anum have been kept in holding tanks at the Smithsonian
Marine Station for nearly 30 years.

Nephasoma pellucidum spawns frequently during the
warmer months of the year. Spawning can be induced by
changing the seawater in the dish, although this procedure
does not reliably yield the desired results, leaving some
uncertainty as to when the spawning occurs. Embryos and
larvae are easy to observe with different microscopic tech-
niques. Mortality before the first metamorphosis, from
trochophore to pelagosphera, is minor. The pelagosphera
larvae are more transparent than other sipunculan larvae,
facilitating observation by light and confocal laser scan-
ning microscopy. In contrast to the pelagosphera larvae of
some other sipunculan species, they relax relatively well
when temporarily cooled to 4°C and treated with men-
thol, magnesium chloride, or 10% ethanol, leaving their
head and terminal organ exposed.

The increase in mortality during the prolonged pelago-
sphera phase presents some difficulties. By the time meta-
morphic competence is reached, the percentage of surviv-
ing larvae is low. A further reduction in numbers occurs
at metamorphosis, because not all larvae respond to the
settlement cue, that is, adult-conditioned sediment. There-
fore, postmetamorphic juveniles are only rarely observed.
Common metamorphosis-inducing agents such as potas-
sium chloride, cesium chloride, gamma-aminobutyric acid,
3,4-dihydroxy-l-phenylalanine (1-dopa), and isobutylmeth-
ylxanthine (Bryan et al., 1997; Morse et al., 1979; Yool
et al., 1986) seem to have no effect on metamorphosis in
N. pellucidum.

As a conclusion, among the sipunculans for which de-
velopment has been studied, N. pellucidum is a good can-
didate for a model species. Rearing larvae through meta-
morphosis still presents some difficulties, and future work
should focus on optimizing the conditions. Recently the
cold-water species Phascolosoma agassizii from the Sea
of Japan, which also has a planktotrophic pelagosphera,

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

has been reared through metamorphosis (A. S. Maiorova
and A. V. Adrianoy, Institute of Marine Biology, Far East
Branch of the Russian Academy of Sciences, Russia, per-
sonal communication) and might be another appropriate
candidate for a model species, even though metamorpho-
sis could never be observed in individuals of the same spe-
cies collected in the Pacific Northwest and reared at the
University of Washington Friday Harbor Laboratories.

ACKNOWLEDGMENTS

We thank the staff at the Smithsonian Marine Station
at Fort Pierce for their research support (SMSFP Contri-
bution No. 750). We are particularly grateful to Julie Pi-
raino, Hugh Reichardt, and Woody Lee for their support
with specimen collection, microscopy, and imaging. Valu-
able assistance was provided in the earlier phases of this
study by Douglas Putnam and Cindy Hunter. This work
was partially supported by a Smithsonian Marine Station
postdoctoral fellowship granted to AS. We thank the orga-
nizers of the Smithsonian Marine Science Symposium for
the opportunity to present our research.

LITERATURE CITED

Akesson, B. 1958. A Study of the Nervous System of the Sipunculoideae
with Some Remarks on the Development of the Two Species Phas-
colion strombi Montagu and Golfingia minuta Keferstein. Lund,
Sweden: C. W. K. Gleerup.

Bleidorn, C., L. Podsiadlowski, and T. Bartolomaeus. 2005. The Com-
plete Mitochondrial Genome of the Orbiniid Polychaete Orbinia
latreillii (Annelida, Orbiniidae): A Novel Gene Order for the An-
nelida and Implications for Annelid Phylogeny. Gene (Amsterdam),
370:96-103.

Boore, J. L., and J. L. Staton. 2002. The Mitochondrial Genome of the
Sipunculid Phascolopsis gouldii Supports Its Association with An-
nelida Rather than Mollusca. Molecular Biology and Evolution,
19:127-137.

Bryan, P. J., P. Qian, J. L. Kreider, and F. Chia. 1997. Induction of Larval
Settlement and Metamorphosis by Pharmacological and Conspe-
cific Associated Compounds in the Serpulid Polychaete Hydroides
elegans. Marine Ecology Progress Series, 146:81-90.

Cutler, E. B. 1994. The Sipuncula: Their Systematics, Biology and Evolu-
tion. Ithaca, N.Y.: Cornell University Press.

Dunn, C. W., A. Hejnol, D. Q. Matus, K. Pang, W. E. Browne, S. A.
Smith, E. Seaver, G. W. Rouse, M. Obst, G. D. Edgecombe, M. V.
Sorensen, S. H. D. Haddock, A. Schmidt-Rhaesa, A. Okusu, R. M.
Kristensen, W. C. Wheeler, M. Q. Martindale, and G. Giribet. 2008.
Broad Phylogenomic Sampling Improves Resolution of the Animal
Tree of Life. Nature (London), 452:745-749.

Field, K. G., G. J. Olsen, D. J. Lane, S. L. Giovannoni, M. T. Ghiselin,
E. C. Raff, N. R. Pace, and R. A. Raff. 1988. Molecular Phylogeny
of the Animal Kingdom. Science, 239:748-753.

Franzén, A. 1958. On Sperm Morphology and Acrosome Filament For-
mation in Some Annelida, Echiurida, and Tunicata. Zoologiska
Bidrag fran Uppsala, 33:1-28.

Gerould, J. H. 1906. The Development of Phascolosoma. Zoologische
Jabrbiicher Abtheilung fiir Anatomie, 23:77-162.

Halanych, K. M. 2004. The New View of Animal Phylogeny. Annual
Review of Ecology, Evolution, and Systematics, 35:229-256.

Hall, J. R., and R. S. Scheltema. 1975. “Comparative Morphology of
Open-Ocean Pelagosphaera.” In Proceedings of the International
Symposium of Sipuncula and Echiura, ed. M. E. Rice and M. Todo-
rovic, pp. 183-197. Belgrade: Naucno Delo Press.

Hessling, R., and W. Westheide. 2002. Are Echiura Derived from a Seg-
mented Ancestor? Immunohistochemical Analysis of the Nervous
System in Developmental Stages of Bonellia viridis. Journal of Mor-
phology, 252:100-113.

Hyman, L. H. 1959. “Phylum Sipunculida.” In The Invertebrates,
Smaller Coelomate Groups. Volume 5, pp. 610-690. New York:
McGraw-Hill.

Kristof, A., T. Wollesen, and A. Wanninger. 2008. Segmental Mode of
Neural Patterning in Sipuncula. Current Biology, 18:1129-1132.

Maslakova, S. A., M. Q. Martindale, and J. L. Norenburg. 2004. Fun-
damental Properties of the Spiralian Developmental Programs Are
Displayed by the Basal Nemertean Carinoma tremaophoros (Palae-
onemertea, Nemertea). Developmental Biology, 267:342-360.

Maxmen, A. B., B. F. King, E. B. Cutler, and G. Giribet. 2003. Evolu-
tionary Relationships within the Protostome Phylum Sipuncula: A
Molecular Analysis of Ribosomal Genes and Histone H3 Sequence
Data. Molecular Phylogenetics and Evolution, 27:489-503.

McHugh, D. 1997. Molecular Evidence That Echiurans and Pogonopho-
rans Are Derived Annelids. Proceedings of the National Academy
of Sciences of the United States of America, 94:8006-8009.

Millonig, G. 1964. “Study on the Factors Which Influence Preservation
of Fine Structure.” In Symposium on Electron Microscopy, ed. P.
Buffa, p. 347. Rome: Consiglio Nazionale delle Ricerce.

Morse, D. E., N. Hooker, H. Duncan, and L. Jensen. 1979. y-Aminobutyric
Acid, a Neurotransmitter, Induces Planktonic Abalone Larvae to
Settle and Begin Metamorphosis. Science, 204:407-410. _

Pilger, J. F 1987. Reproductive Biology and Development of Themiste
lageniformis, a Parthenogenic Sipunculan. Bulletin of Marine Sci-
ence, 41:59-67.

Rice, M. E. 1967. A Comparative Study of the Development of Phasco-
losoma agassizii, Golfingia pugettensis, and Themiste pyroides with
a Discussion of Developmental Patterns in the Sipuncula. Ophelia,
4:143-171.

. 1975a. “Observations on the Development of Six Species of Ca-

ribbean Sipuncula with a Review of Development in the Phylum.”

In Proceedings of the International Symposium of Sipuncula and

Echtura, ed. M. E. Rice and M. Todorovic, pp. 141-160. Belgrade:

Naucno Delo Press.

NUMBER 38 ¢ 217

. 1975b. “Sipuncula.” In Reproduction of Marine Invertebrates,
ed. K. G. Adiyodi and R. G. Adiyodi, pp. 67-127. New York: Aca-
demic Press.

. 1975c. “Survey of the Sipuncula of the Coral and Beach Rock

Communities of the Caribbean Sea.” In Proceedings of the Interna-

tional Symposium of Sipuncula and Echiura, ed. M. E. Rice and

M. Todorovic, pp. 35-49. Belgrade: Naucno Delo Press.

. 1976. Larval Development and Metamorphosis in Sipuncula.

American Zoologist, 16:563-571.

. 1981. Larvae Adrift: Patterns and Problems in the Life Histories

of Sipunculans. American Zoologist, 21:605-619.

. 1985. “Sipuncula: Developmental Evidence for Phylogenetic In-

ference.” In The Origins and Relationships of Lower Invertebrates,

ed. S. C. Morris, J. D. George, R. Gibson, and H. M. Platt, pp.

274-296. Oxford: Oxford University Press.

. 1988. Observations on Development and Metamorphosis of Sipho-

nosoma cumanense with Comparative Remarks on Sipunculus nudus

(Sipuncula, Sipunculidae). Bulletin of Marine Science, 42:1-15.

. 1989. “Comparative Observations of Gametes, Fertilization,
and Maturation in Sipunculans.” In Reproduction, Genetics and
Distributions of Marine Organisms, ed. J. S. Ryland and P. A. Tyler,
pp. 167-188. Fredensborg, Denmark: Olsen & Olsen.

Rouse, G., and K. Fauchald. 1997. Cladistics and the Polychaetes.
Zoologica Scripta, 26:139-204.

Scheltema, A. H. 1993. Aplacophora as Progenetic Aculiferans and the
Coelomate Origin of Mollusks as the Sister Taxon of Sipuncula.
Biological Bulletin, 184:57-78.

Schulze, A., E. B. Cutler, and G. Giribet. 2005. Reconstructing the Phy-
logeny of the Sipuncula. Hydrobiologia, 535-536:277-296.

. 2007. Phylogeny of Sipunculan Worms: A Combined Analysis of
Four Gene Regions and Morphology. Molecular Phylogenetics and
Evolution, 42:171-192.

Schulze, A., and M. E. Rice. 2009. Musculature in Sipunculan Worms: On-
togeny and Ancestral States. Evolution & Development, 11:97-108.

Staton, J. L. 2003. Phylogenetic Analysis of the Mitochondrial Cytochrome
c Oxidase Subunit 1 Gene from 13 Sipunculan Genera: Intra- and
Interphylum Relationships. Invertebrate Biology, 122:252-264.

Struck, T., N. Schult, T. Kusen, E. Hickman, C. Bleidorn, D. McHugh,
and K. Halanych. 2007. Annelid Phylogeny and the Status of Sipun-
cula and Echiura. BMC Evolutionary Biology, 7:57.

Wanninger, A., D. Koop, L. Bromham, E. Noonan, and B. M. Degnan.
2005. Nervous and Muscle System Development in Phascolion
strombus. Development, Genes and Evolution, 215:508-518.

Yool, A. J., S. M. Graul, M. G. Hadfield, R. A. Jensen, D. A. Markell,
and D. E. Morse. 1986. Excess Potassium Induces Metamorphosis in
Four Marine Invertebrate Species. Biological Bulletin, 170:255-266.

*
DP - ee =
= ut jan
2 )
1 2 .
7 s
: hi
Ae 2
. a
= p
,
1 es
2 ms a
P L
Ere 7 :
ae @ i t
iy i 7 ‘
i { E . , <2 ’
=¥ ‘ = - , = fe. = -
1 e i‘ a bs : a
ie hi ag et ON ee repels alan Bi
: ; f } : 7

2

fe

, : ye a), ras: . ee ae SOF ole (Pe aita. ae

r ‘ = s
t A ‘ th => > ise,

a "5 : ee

Oe - ; fli i 6 pi Soak
: | a
p = Fc :
iy > = fc
if
> 4 t
+ Ss 5

ig? a a)
Met | se +e

t A _ = eo Bs
, 5 5 s { ’ i oll ‘
> a St eae ne ¥
rs iar A tah S a a4 ae ra 4
by 4 i § yy : yee he de eee
e . = 2

Mitochondrial Phylogeography of the
Intertidal Isopod Excirolana braziliensis
on the Two Sides of the Isthmus of Panama

Renate Sponer and Harilaos A. Lessios

Renate Sponer and Harilaos A. Lessios, Smithson-
ian Tropical Research Institute, Box 0843-03092,
Balboa, Panama. Corresponding author: H. Lessios
(Lessiosh@post.harvard.edu). Manuscript received
15 August 2008; accepted 20 April 2009.

ABSTRACT. The intertidal isopod Excirolana braziliensis Richardson possesses limited
means of dispersal; there is no larval stage, and adults remain sedentary under the sand.
It is represented on the two coasts of Panama by three morphs, two in the Pacific (P and
C’) and one in the Atlantic (C). Previous work has quantified morphometric differences
between the morphs, found that there are multiple allozyme differences between them,
and produced indirect evidence that they are reproductively isolated from each other.
Here we report comparisons of 345 bp of 12S and 678 bp of cytochrome oxidase I (COI)
mitochondrial DNA (mtDNA) from three populations of each morph. The mtDNA se-
quences from the three morphs are reciprocally monophyletic, strengthening the case for
recognizing them as separate species. As in morphology and isozymes, the C morph and
the C’ morph are sister clades, and the P morph is an outgroup. In contrast to what was
previously supposed, the C and C’ morphs neither are the result of a recent introduction
from one ocean to the other, nor were separated at the final stages of the completion of
the Isthmus of Panama three million years ago, but rather are anciently separated sister
clades that now exist on separate shores. Patterns of mitochondrial gene flow between
populations of the same morph vary. The C and C’ morphs show large genetic differences
between local populations, as would be expected from an organism with such limited
vagility. In the P morph, on the other hand, populations from localities 5 km apart are
identical in mitochondrial DNA, even though they differ in one allozyme locus, suggest-
ing the possibility of sex-biased migration.

INTRODUCTION

Many marine organisms are capable of dispersing over large distances at some
point of their life cycle. The population genetics of such organisms usually reveal
a genetic neighborhood size in the order of thousands or tens of thousands of ki-
lometers. Some marine species, however, provide a contrast to this picture of wide
dispersal in that they lack any means of transferring their genes by either vagile
adults or free-swimming larvae, yet have wide geographic ranges. How much gene
flow may occur between noncontiguous populations of such species and whether
species cohesion is maintained in the face of limited vagility is of special interest to
population genetics. The tropical isopod Excirolana braziliensis is an example of a
species apparently spread over the tropical seas of Americas, even though it lacks
the means of maintaining genetic contact between distant populations.

220 e

Excirolana braziliensis Richardson is a common iso-
pod of intertidal beaches on both sides of the Americas,
from the Pacific coast of Mexico (30°N) to S. Chile (40°S)
and from the northern Caribbean (31°N) to Uruguay (25°S)
(Cardoso and Defeo, 2003). It is a small (approximately
3-4 mm in length), dioecious species, which reaches its
highest abundance just above the high-tide mark, where
it lives buried in the sand during low tide and rises to the
water column at high tide to feed on live and dead fishes
and invertebrates (Brusca and Iverson, 1985). E. brazilien-
sis has very limited means of dispersal. The female carries
broods of 4 to 17 offspring per reproductive event. Young
are released directly into the adult habitat (Klapow, 1970).
The frequency of reproduction is such that a population
may turn over every four months (Brusca and Iverson,
1985). In Panama, recruitment occurs throughout the year
(Dexter, 1977). Dispersal in E. braziliensis may occur as a
result of feeding events, during which individuals of this
genus have been observed attached to fish or other prey
items for several minutes (Brusca, 1980). This behavior
may represent the only means of transport of this organ-
ism between beaches because free-swimming individuals
are likely to be eaten by fish.

Weinberg and Starczak (1988, 1989) reported the ex-
istence of three morphological variants of E. braziliensis
from Panama. Two similar and presumably closely related
types, termed C and C’, are found on the Caribbean and
Pacific coasts, respectively. The third type (P) is morpho-
logically distinct from C and C’; its distribution overlaps
with that of C’ throughout most of its range (Weinberg
and Starczak, 1989). In general, Pacific beaches contain
only one of the two morphotypes. Nevertheless, 2 of 43
beaches sampled by Weinberg and Starczack (1989) were
found to contain C’ and P morphs in approximately equal
numbers. The geographic patterns of morphotype distri-
bution and genetic composition remain stable over time,
but there are occasional complete replacements of entire
beaches by a different morph, presumably as the result of
extirpation and subsequent recolonization (Lessios et al.,
1994). Morphological and genetic divergence (based on
allozyme data) between morphs are highly correlated and
large enough to suggest that the P morph constitutes a dis-
tinct species (Lessios and Weinberg, 1994). The allozyme
data are also consistent with the hypothesis that the C and
C’ morphotypes are geminate species that resulted from
the rise of the Panamanian Isthmus three million years ago
(Lessios and Weinberg, 1994).

Allozyme analyses indicate that the three morpho-
types of E. braziliensis are probably reproductively iso-
lated, because they form few hybrids even when they

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

co-occur at the same beach. Even within morphotypes,
gene flow between populations from different beaches
is low, as deduced from the predominance of distinct
alleles in one or more loci, even among beaches situated
less than 5 km apart. However, dispersal (as measured
by individuals homozygous for alleles that otherwise oc-
cur on a different beach) is rather high, suggesting that
some form of reproductive isolation prevents them from
mating with individuals from the local population (Les-
sios and Weinberg, 1993).

The purpose of the present study is to investigate the
phylogenetic and phylogeographic relationships within and
between the three morphotypes of Excirolana braziliensis
using sequences of mitochondrial DNA (mtDNA). Specifi-
cally, we are addressing the following questions: (1) Does
mtDNA show patterns of genetic divergence, phylogeny,
and geographic distribution congruent with those suggested
by isozymes and morphology? (2) When did the three lin-
eages diverge? (3) What are the patterns of population
genetic structure? Do mtDNA data show similar levels of
gene flow as isozymes within and between morphotypes?
(4) What processes can explain mtDNA discrepancies be-
tween patterns from mtDNA and allozyme markers?

MATERIALS AND METHODS
SAMPLE COLLECTION

Excirolana braziliensis were obtained from nine loca-
tions along the Pacific and Caribbean coasts of Panama
(Figure 1). Each of the three morphotypes was represented
in our collections by three populations. Isopods were col-
lected on beaches during low tide. The top 10 cm of sand
at haphazard locations above the high-tide mark were
sifted through a 500 wm sieve, and isopods were placed
in plastic bags with wet sand. The collected isopods were
brought alive to the laboratory and frozen at —80°C. The
majority of samples used in this study were from collec-
tions made in 1988, the same collections used to assay
isozymes (Lessios and Weinberg, 1993, 1994). Additional
individuals were collected in 1998 from Isla Culebra.
Specimens of Excirolana mayana were also collected at
Isla Culebra to be used as outgroups.

DNA ExTRACTION, POLYMERASE CHAIN REACTION,
AND mtDNA SEQUENCING

Genomic DNA was extracted using a standard phe-
nol/chloroform protocol (Sambrook et al., 1989) with
ethanol precipitation. For amplification and sequencing

82° 80° 78°

FIGURE 1. Map of Panama, indicating localities in which Exciro-
lana braziliensis was sampled (sample size in parentheses). 1, Bocas
del Toro (14); 2, Shimmey Beach (10); 3, Maria Chiquita (9); 4, Isla
Santelmo (18); 5, Isla Adentro (15); 6, Causeway (6); 7, Lab (8); 8,
Perico (10); 9, Isla Culebra (14).

of 345 base pairs (bp) of the 12S mtDNA gene, we used
the universal primers 12Sa and 12Sb (Simon et al., 1994).
A 678 bp fragment of cytochrome oxidase I (COI) was
amplified and sequenced with combinations of the for-
ward primers BWBK (5’-GAG CTC CAG ATA TAG CAT
TCC-3’) and ISO-F1 (5’-CYC TTT TAT TAG GRA GGG
GG-3’) and the reverse primers BWBJ (5'-CAA TAC CTG
TGA GTC CTC CTA-3’) and ISO-R2 (5’-ACR GCA ATA
ATT ATG GTA GC-3’). The following conditions were
used for polymerase chain reaction (PCR): initial denatur-
ation for 2 min at 94°C, then 30 cycles of denaturation for
30 s at 94°C, annealing for 30 s at 50°-53°C, extension
for 1 min at 72°C, and final extension for 10 min at 72°C.
PCR products were cleaned for sequencing using silica gel
purification columns. Cycle sequencing was carried out in
both directions, with the ABI PRISM d-Rhodamine Ter-
minator Cycle Sequencing Ready Reaction Kit (Applied
Biosystematics). Sequences were obtained on an ABI 377
automated sequencer and verified and aligned by eye in
the program Sequencher (Gene Codes Corporation). 12S
was sequenced in 104 individuals; a subset of 22 individu-
als was also successfully sampled for COI, whereas the
rest failed to amplify for this locus.

PHYLOGENETIC ANALYSIS

For phylogenetic analysis, identical haplotypes from
multiple individuals were collapsed. We applied the pro-

NUMBER 38 e¢ 221

gram Modeltest 3.0 (Posada and Crandall, 1998) to
calculate the goodness of fit of various models of DNA
evolution. The selected model for the 12S data was that
of Tamura and Nei (1993), with equal base frequen-
cies, a gamma distribution with a shape parameter of
0.443, and the following substitution rates: [A—C] =
1.00; [A—G] = 6.00; [A—T] = 1.00; [C—G] = 1.00;
[C—T] = 11.71; and [G—T] = 1.00. The selected model
for the COI sequence was a transversional model (TVM
+ I; Posada and Crandall, 2001), with a proportion of
0.69 of invariable sites and the following substitution
rates: [A—C] = 8908.76; [A—G] = 258211.81; [A—T]
= 26092.37; [C—G] = 0.0001; [C—T] = 258211.82, and
[G—T] = 1.00. A partition homogeneity test, executed in
version 4.0b10 of PAUP* (Swofford, 2000), indicated that
phylogenetic signals in the COI and 12S data were not
significantly different (P = 0.256). The best fitting model
for the combined 12S and COI data was HKY (Hasegawa
et al., 1985) with a transition/transversion ratio of 11.46
and a gamma distribution shape parameter of 0.782.
Employing these parameters, we ran phylogenetic analy-
ses for 12S and COI separately and with the two DNA
regions concatenated. We used the BioNJ algorithm
(Gascuel, 1997) and heuristic searches in maximum par-
simony and maximum likelihood with PAUP* (Swofford,
2000). Bootstrap confidence values for distance and like-
lihood trees were calculated in 5,000 and 500 iterations,
respectively. Bayesian phylogeny inference was carried
out in the program MrBayes v.3.04b (Huelsenbeck and
Ronquist, 2001). Bayesian analyses on the COI and the
combined data sets were run for 800,000 generations, of
which the first 20,000 (2,000 trees) were discarded. For
12S, 2,760,000 generations were run, and 67,500 (6,750
trees) were discarded. Convergence of chains was deter-
mined by average standard deviations of split frequen-
cies less than 0.01 and by potential scale reduction fac-
tors approximately equal to 1.0. The trees were rooted
on sequences of Excirolana mayana. Clock-like evolution
of sequences was tested with likelihood ratio tests. The
tests were carried out in PAUP* 4.0b10 by calculating the
difference in log-likelihood of the neighbor-joining trees
(see above) with and without the enforcement of a mo-
lecular clock and comparing the likelihood ratios to the
x’ distribution.

GEOGRAPHIC DISTRIBUTION OF GENETIC VARIATION
WITHIN MORPHS

Genealogical relationships of haplotypes within spe-
cies may be better represented by networks than trees, as

PAA O

ancestral haplotypes may still be present in the population
(Crandall and Templeton, 1993; Posada and Crandall,
2001). We calculated unrooted parsimony haplotype net-
works based on 12S for each of the three morphotypes
separately, using the computer program TCS (Clement et
al., 2000). In this method the parsimony limit (the maxi-
mum number of differences among haplotypes as a result
of single substitutions) is calculated with 95% statistical
confidence, and haplotypes are connected in order of in-
creasing number of substitutions. To investigate the popu-
lation genetic structure within each morphotype, we ap-
plied analysis of molecular variance (AMOVA; Excoffier
et al., 1992) to the 12S data. Genetic variation for this
analysis was assessed based on the Kimura (1980) two-
parameter distance between haplotypes. The significance
of fixation indices was tested by 10,000 rearrangements of
haplotypes between populations. Calculations were car-
ried out in version 2000 of the computer program ARLE-
QUIN (Schneider et al., 2000).

RESULTS
DESCRIPTIVE STATISTICS AND PHYLOGENETIC ANALYSIS

Although we sampled many more individuals of Ex-
cirolana braziliensis for 12S than for COI, trees based on
the former DNA region (Figure 2) displayed less resolu-
tion than the combined analysis of both genes together.
Despite this, all analyses of the 12S segment alone re-
sulted in three distinct lineages, which correspond to
the previously described C, C’, and P morphs. The 12S
sequences of each morph were monophyletic in all analy-
ses. The node joining C and C’ was well supported by
maximum-parsimony analysis but fairly weakly sup-
ported by neighbor-joining, maximum-likelihood, and
Bayesian analysis. The three main lineages were pres-
ent in more than one beach, but each beach contained
representatives of only one lineage. Although Santelmo
had previously been found to contain a mixture of C’
and P morphotypes and the allozymes corresponding
to these morphs (Weinberg and Starczak, 1989; Lessios
and Weinberg, 1994), all nineteen 12S sequences from
Santelmo, differing from each other by a maximum of
three substitutions, belonged to the C’ morph. The tree
based on fewer sequences of COI (not shown) and the
tree based on the combined data (Figure 3) were well re-
solved and gave strong support to the expected grouping
of the C and C’ lineages as sister groups, irrespective of
the type of phylogenetic algorithm used.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

The 12S Tamura and Nei average distances ranged
between 11% and 18% between morphs (Table 1).
Within morphs, distances varied between 0% and 2.3%.
For COI, average distances (TMV) among lineages were
17.4%-26.1% and within lineages 0%-1.5%. There
were five amino acid changes in the COI fragment, of
which four were substitutions of nonpolar for nonpolar
residues (Met/Ileu; Val/Ileu) and one was a nonpolar for
a polar residue (Ala/Thr). Three of the changes differen-
tiate the C/C’ and P lineage; one groups C and P, versus
C’, and one is shared between C’ and P, compared to
C. Likelihood ratio tests failed to reject the hypothesis
of clock-like evolution of either the 12S or the COI se-
quences (P > 0.05).

GENETIC VARIATION WITHIN MORPHS

Parsimony haplotype networks showed that popula-
tions of the C and C’ morphs, but not the P morph, were
genetically structured (Figure 4). The most common and
(presumably) ancestral haplotype of the P morph was
shared by all three populations. Two derived haplotypes
were also shared, one between all populations and the
other between two populations. In the C’ morph two hap-
lotypes, including the ancestral one, were shared between
Santelmo and Isla Culebra. Although Isla Adentro con-
tained three haplotypes not found in any other population,
the majority of specimens from this island were of a single
haplotype, leading to a low haplotype diversity compared
to other populations (H = 0.14). The population at Bocas
del Toro (C morph), was characterized by high nucleotide
diversity compared to other populations (7 = 0.0077 +
0.0045). Haplotypes from Bocas del Toro were differen-
tiated from Maria Chiquita and Shimmey Beach by one
to eight substitutions whereas the latter two populations
shared the ancestral haplotype.

In the C morph, AMOVA (Table 2) found that 67.44%
of genetic variance was partitioned among populations;
population pairwise F,; comparisons (Table 3) showed that
all populations of this morph were significantly differen-
tiated from each other. In the C’ morph, 35.62% of the
variance was the result of differences between populations.
The population at Adentro had significantly high Fs; values
when compared to both Santelmo and Culebra, whereas
the latter two were not significantly different from each
other (Table 3). In the P morph, all the genetic variance
was contained within populations, a result in stark contrast
with high levels of population subdivision seen in the other
two morphs.

NUMBER 38 ¢ 223

E. mayana

rco4,5,6, csw2,4,14,16,
b6,7,9

prco3
CSW5

99 |F Ib5 Pp
100 PRco13
99 Ib1, Ib3
100 prco1, prco7, Ib4
Ib2, csw15, prco2

prco4

bca2,3,4,6,7,8,9,13,14
bca15
bea5,12
93 bca11
100 mcq3,4,5,7,11, shmy1,3,5

94 mceq2
95 ‘ EC

mcq6
shmy2
bca1,10
mcqi,8

57 shmy4,7,9,11,14

70
62

73
ade1,2,3,4,5,6,10,11,12,13,14,16,17

ade15

bs4,7,9,10, xbs3,6,
— 0.005 substitutions/site selm1,2,7,16,19,20,25
83 selm6
94 selm11 |
E selm17
98 bs11
xbs4,7
bs3,6,13, selm3,10,15,21
bs1
selm23
bs8,12, xbs2,8,9

selm22
selm12,14,18

FIGURE 2. 12S mitochondrial DNA (mtDNA) maximum-likelihood bootstrapped consensus tree relating three morphotypes (C, C’, and P) of
Excirolana braziliensis. Numbers above branches indicate maximum-likelihood bootstrap confidence values; numbers below branches refer to
posterior probabilities (Bayesian analysis), neighbor-joining bootstrap support, and maximum-parsimony bootstrap support, respectively, from
top to bottom. Support values <50% are not shown. Locality codes of specimens: prco = Perico; csw = Causeway; lb = Lab; bca = Bocas
del Toro; meq = Maria Chiquita; shmy = Shimmey Beach; ade = Isla Adentro; bs = Isla Culebra; selm = Santelmo; xbs = Isla Culebra (xbs
specimens were collected in 1998; all other samples were collected in 1988). See Figure 1 for the position of each locality.

224 e

0.05 substitutions/site

FIGURE 3. Combined 12S/cytochrome oxidase I (COI) mtDNA
maximum-likelihood bootstrapped consensus tree relating three
major lineages (C, C’, and P) of Excirolana braziliensis in Panama.
Numbers above branches indicate maximum-likelihood bootstrap
confidence values, numbers below branches refer to posterior prob-
abilities (Bayesian analysis), neighbor-joining bootstrap support,
and maximum-parsimony bootstrap support, respectively, from top
to bottom. Localities: BCA = Bocas del Toro; MCQ = Maria Chi-
quita; SHMY = Shimmey Beach; SELM = Santelmo; LB = Lab;
PRCO = Perico.

DISCUSSION

The mtDNA data presented here confirm the results
from analysis of both morphology (Weinberg and Starczak,
1988, 1989; Lessios and Weinberg, 1994) and allozymes
(Lessios and Weinberg, 1994) that Excirolana brazilien-
sis populations from the Pacific and Caribbean coasts of
Panama consist of three distinct lineages. Allozymes sug-
gest that these lineages are reproductively isolated (Lessios
and Weinberg, 1993) and should, therefore, be considered
separate species. Although mtDNA data agree with mor-
phological and allozyme data on the grouping of the C and

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

TABLE 1. Genetic distances within (along diagonal) and between
(below diagonal) morphs of Excirolana braziliensis, for 12S
(Tamura and Nei, 1993) and cytochrome oxidase I (COI) mito-
chondrial DNA (mtDNA) (in parentheses; TVM + I [Posada and
Crandall, 2001)).

Morph C Gi P
C 2.25 (1.52) e i
C’ 11.03 (17.38) 1.36 (0.49) -
P 16.21 (24.32) 17.84 (26.12) 1.19 (0.84)

C’ lineages as sister groups with respect to P, the relative
magnitude of the measures of differentiation in the three
sets of characters is different. Mahalanobis generalized
distance from morphometric characters and Nei’s D from
allozymes indicate that the P morphotype is three times
more distant from C and C’ than the latter are from each
other (Lessios and Weinberg, 1994). Mitochondrial DNA,
on the other hand, gives a P/(C, C’) distance that is only
1.2 to 1.3 times higher than that between C/C’. It is clear
that each type of character evolves at a different rate.

A review of molecular divergence across the Isthmus
of Panama in 34 lineages likely to have been separated
by the final closure of the Isthmus of Panama (Lessios,
2008) has shown that during 3 million years of indepen-
dent evolution (Coates and Obando, 1996; Coates et al.,
2005), crustacean COI has accumulated genetic distances
ranging from 4.1% to 8.7% (Knowlton and Weigt, 1998;
Schubart et al., 1998; Williams et al., 2001; Morrison et
al., 2004) and 12S ranging from 2% to 3% (Robles et al.,
2007). Based on these calibrations, and given the differ-
ences we determined in COI and 12S, the divergence of
the P morph from the two C morphs occurred between 9
and 25 million years ago and that of C from the C’ morph
between 6 and 17 million years ago. Thus, in contrast to
what was surmised by Lessios and Weinberg (1994) on the
basis of isozymes, mtDNA data do not support the idea
that the C and C’ morphotypes were isolated at the final
stages of the closure of the Panamanian Isthmus, 3 million
years ago, but rather that their populations were separated
well before the final closure. On the basis of molecular
divergence, this appears to be also the case in 73 other
amphi-isthmian sister lineages of crustaceans, sea urchins,
fishes, and mollusks (Lessios, 2008).

The combination of large mitochondrial differences
and evidence for reproductive isolation from allozyme data
(Lessios and Weinberg, 1993, 1994; Lessios, 1998) rules
out the hypothesis that C’ merely represents a recently

Clade P

Clade C' 1

FIGURE 4. Parsimony network of 12S mtDNA haplotypes of the three morphs (clades) of Excirolana
braziliensis. Each large oval represents a unique haplotype, boxes represent ancestral haplotypes, and
small ovals indicate hypothetical, intermediate haplotypes not observed in the populations. The size of
each shape represents the frequency of each haplotype. Numbers within each symbol indicate the num-
ber of individuals bearing each haplotype. Localities: Prco = Perico; Lab = Lab; CSW = Causeway;
ADE = Isla Adentro; BS = Isla Culebra; SELM = Santelmo; MCQ = Maria Chiquita; BCT = Bocas
del Toro; SHMY = Shimmy Beach.

226 e

TABLE 2. Analysis of molecular variance (AMOVA) of Pana-
manian populations of Excirolana braziliensis based on 12S
mtDNA sequences. Partitioning of genetic variance within and
between populations (beaches) was estimated for each morph
(clade) separately. The significance of fixation indices was tested
by 10,000 permutations.

Variance (%)

Between Within
Morph populations populations Pcr P
C 67.44 32.56 0.67 <0.001
Gi 35.62 64.38 0.36 <0.001
P =12),39) 102.39 —0.02 >0.05

introduced population of C from the Caribbean into the
Pacific, as had been suggested by Weinberg and Starczak
(1988, 1989) and strengthens the case that each of the
three lineages represents a distinct species.

POPULATION STRUCTURE AND DISPERSAL

Populations of the C and C’ morphs were character-
ized by population subdivision, as illustrated by high F.7
estimates (overall values of 0.67 and 0.36, respectively),
whereas samples from different localities of the P morph
can be considered as belonging to the same genetic popu-
lation (Fs = —0.02). Populations from Isla Adentro (C’
morph) and from Bocas del Toro (C morph) stand out for
their lack of alleles shared with individuals from other lo-
calities. Maria Chiquita and Shimmey Beach (C morph)
also have significantly different allele frequencies, whereas
the populations at Santelmo and Isla Culebra, as well as
at Perico, Lab and Causeway, are not significantly differ-
entiated. The two populations most divergent from oth-
ers in the same morph, Adentro and Bocas del Toro, are
also the most geographically distant from other localities
containing individuals of their respective morphs, raising
the possibility that dispersal to and from these localities is
restricted as a result of physical distance. With only three
populations per morph, statistical verification of a cor-
relation between geographic and genetic distances is not
meaningful.

We observed several differences in the degree of pop-
ulation subdivision when comparing mtDNA and allo-
zyme markers (Lessios and Weinberg, 1994). Based upon
mtDNA sequence, the populations at Bocas del Toro and

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

TABLE 3. Excirolana braziliensis population pairwise Fs7 values
from 12S mtDNA sequences. Bold values are significant at the
P < 0.01 level.

P Morph
Locality Lab Perico
Perico —0.02 -
Causeway —0.02 —0.04
C Morph
Locality Maria Chiquita Bocas del Toro
Bocas del Toro 0.69 -
Shimmey Beach 0.27 0.71
C’ Morph
Locality Isla Adentro Isla Culebra
Isla Culebra 0.58 -
San Telmo 0.50 —0.02

Shimmey Beach were the most different of all (Fs¢ = 0.71),
whereas their allozyme allele frequencies were rather simi-
lar (Fst = 0.097, as calculated from data in Lessios and
Weinberg, 1993). On the whole, mtDNA data suggest a
higher divergence between the morphs, but a lesser degree
of subdivision between populations of the same morph,
compared to data on allozymes. These results support Les-
sios and Weinberg’s (1993) findings that dispersal among
populations is much higher compared to gene flow, be-
cause even individuals of the same morph show some sort
of reproductive isolation. According to their estimates, up
to 2.5% of individuals in a locality consist of new immi-
grants that do not inject their genes into the host popula-
tion, indicating that some form of reproductive isolation
exists between populations of the same morph, even at
the scale of a few kilometers. The data from Santelmo are
interesting in this connection: This is the only locality in
which two morphs, P and C’, coexist (Lessios and Wein-
berg, 1993, 1994). The number of hybrids between them,
as judged by allozymes, is lower than would be expected
from random mating (Lessios and Weinberg, 1993), but
hybrids do exist. However, all 19 mitochondrial haplo-
types from this locality belong to the mtDNA clade that
corresponds to the C’ morph, despite having been sampled
from the same collections as the allozymes. Barring the
unlikely possibility of a sampling accident, this finding in-
dicates that some individuals with a P nuclear genotype,
as manifested in morphology and isozymes, actually carry

a C’ mitochondrial DNA. This, in turn, suggests that hy-
bridization between the morphs, when it occurs, is suc-
cessful in only in one direction, that is, only if the mother
belongs to the C’ clade.

In conclusion, E. braziliensis in Panama consists of
at least three lineages (C, C’, and P), which diverged well
before the final closure of the Isthmus and warrant sepa-
rate species status. Populations that are more than 30
km distant from each other (C, C’) are genetically diver-
gent, whereas those at less than 5 km (P) are panmictic
in mtDNA, even though they are different in at least one
allozyme locus (Lessios and Weinberg, 1994). It remains
to be seen whether population structure is a result of isola-
tion by physical factors or whether the three species have
inherently different dispersal potential, and whether the
higher degree of gene flow in mtDNA relative to isozymes
is the result of sex-biased migration.

ACKNOWLEDGMENTS

We thank Alison Dwileski, Axel Calderon, and Ligia
Calderon at Smithsonian Tropical Research Institute
(STRI) for extensive laboratory work, and Bailey Kessing
of STRI for expert advice and technical help in this study.

LITERATURE CITED

Brusca, R. C. 1980. Common Intertidal Invertebrates of the Gulf of Cali-
fornia. 2nd ed. Tucson: University of Arizona Press.

Brusca, R. C., and E. W. Iverson. 1985. A Guide to the Marine Iso-
pod Crustacea of Pacific Costa Rica. Revista de Biologia Tropical,
33:1-77.

Cardoso, R. S., and O. Defeo. 2003. Geographical Patterns in Reproduc-
tive Biology of the Pan-American Sandy Beach Isopod Excirolana
braziliensis. Marine Biology, 134:573-581.

Clement, M. D., D. Posada, and K. A. Crandall. 2000. TCS: A Com-
puter Program to Estimate Gene Genealogies. Molecular Ecology,
9:1657-1660.

Coates, A. G., D. EF McNeill, M. P. Aubry, W. A. Berggren, and L. S. Col-
lins. 2005. An Introduction to the Geology of the Bocas del Toro
Archipelago, Panama. Caribbean Journal of Science, 41:374-391.

Coates, A. G., and J. A. Obando. 1996. “The Geologic Evolution of
the Central American Isthmus.” In Evolution and Environment
in Tropical America, ed. J. B. C. Jackson, A. F. Budd, and A. G.
Coates, pp. 21-56. Chicago: University of Chicago Press.

Crandall, K., and A. R. Templeton. 1993. Empirical Tests of Some Pre-
dictions from Coalescent Theory with Applications to Intraspecific
Phylogeny Reconstruction. Genetics, 134:959-969.

Dexter, D. M. 1977. Natural History of the Pan-American Sand Beach
Isopod Excirolana braziliensis (Crustacea: Malacostraca). Journal
of Zoology, 183:103-109.

Excoffier, L., P. Smouse, and J. Quattro. 1992. Analysis of Molecular
Variance Inferred from Metric Distances Among DNA Haplotypes:

NUMBER 38 ¢ 227

Application to Human Mitochondrial DNA Restriction Data.
Genetics, 131:479-491.

Gascuel, O. 1997. BIONJ: An Improved Version of the NJ Algorithm
Based on a Simple Model of Sequence Data. Molecular Biology and
Evolution, 14:685-695.

Hasegawa, M., K. Kishino, and T. Yano. 1985. Dating the Human—Ape
Splitting by a Molecular Clock of Mitochondrial DNA. Journal of
Molecular Evolution, 22:160-174.

Huelsenbeck, J. P., and F. Ronquist. 2001. MRBAYES: Inference of Phy-
logeny. Bioinformatics, 17:754-755.

Kimura, M. 1980. A Simple Method for Estimating Evolutionary Rates
of Base Substitutions Through Comparative Studies of Nucleotide
Sequences. Journal of Molecular Evolution, 16:111-120.

Klapow, L. A. 1970. Ovoviviparity in the Genus Excirolana (Crustacea:
Isopoda). Journal of Zoology (London), 162:359-369.

Knowlton, N., and L. A. Weigt. 1998. New Dates and New Rates for
Divergence Across the Isthmus of Panama. Proceedings of the Royal
Society of London, Series B, 265:2257-2263.

Lessios, H. A. 1998. “The First Stage of Speciation as Seen in Organisms
Separated by the Isthmus of Panama.” In Endless Forms: Species
and Speciation, ed. D. J. Howard and S. Berlocher, pp. 186-201.
Oxford: Oxford University Press.

. 2008. The Great American Schism: Divergence of Marine Or-
ganisms after the Rise of the Central American Isthmus. Annual
Reviews of Ecology, Evolution, and Systematics, 39:63-91.

Lessios, H. A., and J. R. Weinberg. 1993. Migration, Gene Flow and
Reproductive Isolation Between and Within Morphotypes of the
Isopod Excirolana in Two Oceans. Heredity, 71:561-573.

. 1994. Genetic and Morphological Divergence among Morpho-
types of the Isopod Excirolana on the Two Sides of the Isthmus of
Panama. Evolution, 48:530-548.

Lessios, H. A., J. R. Weinberg, and V. R. Starczak. 1994. Tempo-
ral Variation in Populations of the Marine Isopod Excirolana:
How Stable Are Gene Frequencies and Morphology? Evolution,
48:549-563.

Morrison, C. L., R. Rios, and J. E. Duffy. 2004. Phylogenetic Evidence
for an Ancient Rapid Radiation of Caribbean Sponge-Dwelling
Snapping Shrimps (Synalpheus). Molecular Phylogenetics and Evo-
lution, 30:563-581.

Posada, D., and K. A. Crandall. 1998. Modeltest: Testing the Model of
DNA Substitution. Bioinformatics, 14:817-818.

. 2001. Selecting Models of Nucleotide Substitution: An Applica-
tion to Human Immunodeficiency Virus (HIV-1). Molecular Biol-
ogy and Evolution, 18:897-906.

Robles, R., C. D. Schubart, J. E. Conde, C. Carmona-Suarez, F. Alva-
rez, J. L. Villalobos, and D. L. Felder. 2007. Molecular Phylogeny
of the American Callinectes Stimpson, 1860 (Brachyura: Portuni-
dae), Based on Two Partial Mitochondrial Genes. Marine Biology,
150:1265-1274.

Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular Cloning.
A Laboratory Manual. 2nd ed. Cold Spring Harbor Laboratory,
N. Y.: Cold Spring Harbor Laboratory Press.

Schneider, S., D. Roessli, and L. Excoffier. 2000. ARLEQUIN, Version
2000: A Software for Population Genetics Data Analysis. Genetics
and Biometry Laboratory, University of Geneva, Switzerland.

Schubart, C. D., R. Diesel, and S. B. Hedges. 1998. Rapid Evolution
to Terrestrial Life in Jamaican Crabs. Nature (London), 393:
363-365.

Simon, C. L., E Frati, A. Beckenbach, B. Crespi, H. Liu, and P. Flook.
1994. Evolution, Weighting, and Phylogenetic Utility of Mitochon-
drial Gene Sequences and a Compilation of Conserved Polymerase
Chain Reaction Primers. Annals of the Entomological Society of
America, 87:651-701.

228 e

Swofford, D. L. 2000. PAUP*: Phylogenetic Analysis Using Parsimony
(* and Other Methods). Ver. 4.068. Sunderland, Mn.: Sinauer As-
sociates.

Tamura, K., and M. Nei. 1993. Estimation of the Number of Nucleo-
tide Substitutions in the Control Region of Mitochondrial DNA
in Humans and Chimpanzees. Molecular Biology and Evolution,
10:512-526.

Weinberg, J. R., and V. R. Starczak. 1988. Morphological Differences
and Low Dispersal Between Local Populations of the Tropical

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Beach Isopod, Excirolana braziliensis. Bulletin of Marine Science,

42:296-309.

. 1989. Morphological Divergence of Eastern Pacific and Carib-
bean Isopods: Effects of a Land Barrier and the Panama Canal. Ma-
rine Biology, 103:143-152.

Williams, S. T., N. Knowlton, L. A. Weigt, and J. A. Jara. 2001. Evi-
dence for Three Major Clades Within the Snapping Shrimp Genus
Alpheus Inferred from Nuclear and Mitochondrial Gene Sequence
Data. Molecular Phylogenetics and Evolution, 20:375-389.

Stability and Change in the Indian River
Area Bryozoan Fauna over a
Twenty-Four Year Period

Judith E. Winston

Virginia Museum of Natural History, 21 Starling
Avenue, Martinsville, Virginia 24112, USA (judith
.winston@vmnh.virginia.gov). Manuscript received
13 May 2008; accepted 20 April 2009.

ABSTRACT. Two surveys describe changes and stability in bryozoan assemblages at
sites in the temperate to tropical transition zone of the Florida Atlantic coast over a
24-year interval in which seawater temperatures increased. Results of a monthly survey
of the Indian River Area bryozoan fauna carried out in 1974-1975 as part of a postdoc-
toral fellowship at the Smithsonian Marine Station were published in 1982. The exis-
tence of this baseline work made it possible to resurvey some of the same areas during
1998-1999 to determine whether the bryozoan communities at three of the sites in the
original study had changed or remained stable. Results showed that most of the species
that had been abundant at a site still occurred at that site 24 years later, indicating a
high degree of stability. However, there were some important changes. Temperate spe-
cies such as Hippoporina verrilli, Cryptosula pallasiana, and Bugula stolonifera, which
had been abundant in 1974, were rare or absent in 1998. Those species were replaced
by Caribbean species, such as Exechonella antillea and Caulibugula armata. Although
local seawater temperatures during the time period were not available, the Fort Pierce air
temperature records indicated that despite the year-to-year variability in both minimum
and maximum temperatures over the seasons, mean winter air temperatures maintained
a slow increase from 1974 to 1999.

INTRODUCTION

Most ecological research projects are carried out over a very short time
period, the length of a research grant or dissertation project, a few years at
most, and once the researcher moves on to new studies these research efforts are
seldom repeated. Long-term studies are essential to document effects of climate
change in communities over time, but the number of such publications for ma-
rine communities is extremely low compared with the number documenting the
effects of climate change in terrestrial systems (e.g., Richardson and Poloczan-
ska, 2008). This paper describes a repeated survey of coastal and lagoon sites
conducted 24 years after the original survey was completed.

In 1974 and 1975, as part of my research as a postdoctoral fellow at the
Smithsonian Marine Station, I carried out monthly surveys of the bryozoan fauna
at five intertidal sites in the Indian River Lagoon region, both in the lagoon itself
and on the coasts of North and South Hutchinson Islands. Descriptions of the
species found at these sites, together with descriptions of species taken in one-time

230 e

collections at 18 additional localities in the region and
notes on their distribution and ecology, were published in
a taxonomic paper, “Marine Bryozoans (Ectoprocta) of
the Indian River Area (Florida)” (Winston, 1982). Over
the years I returned to the area many times to study vari-
ous aspects of the biology and ecology of the bryozoans
of the region. The apparent persistence of species at par-
ticular sites year after year led me to believe that bryozoan
communities in the area might be very stable. Yet, the
patchiness and limited extent of the hard substrata avail-
able for settlement, combined with the fact that certain
species were found consistently at only a single site, made
me wonder about the potential effect of a man-made or
natural disturbance. If a site were to be destroyed, would
that mean the regional extinction of the bryozoan species
uniquely found there, or did they, in fact, have additional
refuges at other sites in the area? To begin to answer these
questions, 24 years after the first study, I resurveyed three
of the original sites over a one year period in 1998-1999
to learn how stable was the species composition and to
look for additions or losses of species at each site.

STUDY AREA

The Indian River Lagoon system, including Mosquito
Lagoon, extends along about a third of the Atlantic coast
of Florida, from Ponce de Leon Inlet to Jupiter Inlet, a
distance of 295 km. Its western boundary is the Florida
mainland, while a barrier island complex broached by sev-
eral inlets forms its eastern boundary. The Indian River
Lagoon proper is a shallow microtidal lagoon 195 km in
length. It is believed to have the highest biodiversity of
any estuarine system in North America, perhaps in part
because of its location at the transition between two bio-
geographic provinces, the warm temperate Carolinean
and the tropical Caribbean (Swain et al., 1995).

METHODS

The samples taken in the original survey had been
gathered at first only to acquire living colonies of as many
species as possible for behavioral and morphological stud-
ies (Winston, 1978). As I became interested in the life
histories of the species involved, I began collecting at the
most convenient and interesting sites in the south central
part of the Indian River Lagoon area on an approximately
monthly schedule from the fall of 1974 through the sum-
mer of 1975. The sites studied were the inner breakwater

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

at Sebastian Inlet, the Johnson House seagrass bed at
Harbor Branch Oceanographic Institution at Link Port,
the North Beach breakwater at Fort Pierce Inlet, Walton
Rocks, South Hutchinson Island, and Seminole Shores,
South Hutchinson Island. Collections from those sites
were taken in all seasons, an important consideration in
the seasonal environment of the Indian River Lagoon re-
gion. For bryozoans, as for many organisms inhabiting
the area, the highest diversity is achieved and the great-
est amount of reproduction, recruitment, and growth
of colonies of most species take place during the cooler
months (Winston, 1982, 1995). However, tropical species
are more apt to be present or active in summer. It was
not possible to return to Florida monthly in 1998-1999,
but for the best comparison to 1974-1975, the sampling
dates were selected to span the seasons and thus reflect the
known seasonality of the bryozoan fauna.

Collections were made quarterly (in November 1998,
and February, April, and July 1999) at four sites: two
within the lagoon and two on the open coast.

Sites SAMPLED

It was not possible to resurvey all the sites sampled
in the original study, for reasons of time and because
changes such as the development of some sites into of-
ficial county or state parks had increased restrictions on
scientific collecting. The coastal sites sampled in the re-
study were the North Beach breakwater, Fort Pierce Inlet
State Park (by special permit), and the Walton Rocks area,
South Hutchinson Island, plus one site in the Indian River
Lagoon, the Johnson House seagrass bed. One additional
site was chosen for the 1998-1999 survey: the intertidal
bridge pilings on the east side of the Route A1A causeway
to the North Beach in Fort Pierce. This site was added
because it was within the Lagoon, yet was close enough
to the Fort Pierce Inlet, local marinas, and the commercial
port in Fort Pierce to be a likely settlement spot for any
newly arrived bryozoan species.

COLLECTING METHODS

Some bryozoan species have colonies several centime-
ters or more in size and are recognizable in the field, but
in many other species the colonies are microscopic and
cryptic. Therefore collections were made by scraping hard
surfaces and by gathering encrusted substrata: algae, hy-
droid stems, rocks, shells, or trash. As in the original study,
sampling was not quantitative but was thorough. At each
locality all microhabitats available—crevices of break-

waters, surfaces of rocks, shells, wood, algae, hydroids,
octocorals, sabellariid worm tubes, etc.—were examined
carefully for bryozoans. In addition, encrusted examples
of each kind of substratum available were taken back to
the laboratory and examined alive in seawater; attached
bryozoans were identified under a dissecting microscope
at 12-100X. Careful microscopic examination made it
possible to identify the many tiny and/or uncalcified speci-
mens that could not be identified or even detected in the
field. The condition of the colonies and the presence of
reproductive structures and/or embryos were also noted,
as was the relative abundance of each species at a site.
Voucher samples for the project are deposited at the Vir-
ginia Museum of Natural History.

TEMPERATURE AND SALINITY DATA

Seawater temperatures and salinities were recorded at
each census in this study. Temperature ranges are given in
the results for each site. Salinity varied little. All readings
were in the normal ocean range of the area (35-37%o).
The salinity range in the Indian River Lagoon can be more
variable than that recorded at any of the sites during the
resurvey, but low salinities are connected with periods of

NUMBER 38 e¢ 231

heavy rainfall, and 1998-1999 was a drought year. No
temperature or salinity data were collected in the original
study, and no seawater temperature data were available
for the area that covered the entire time period. Air tem-
peratures for Fort Pierce were available (Figure 1) and are
summarized in the Discussion section.

RESULTS

NortH BEACH BREAKWATER, NORTH HUTCHINSON
ISLAND, ForT PIERCE INLET

This site was located at the southern tip of North
Hutchinson Island. Specimens were collected from the in-
tertidal rocks on the north side of the north breakwater
that protects Fort Pierce Inlet. Habitats sampled included
the rocks of the breakwater; sabellariid tubes, hydroids,
octocorals, and algae attached to the rocks; and driftwood
and other debris wedged among the rocks. The bryozoan
diversity at the breakwater is largely dependent on the
presence of the hydroid Thyroscyphus ramosus Allman,
1877 and the soft coral Carijoa riisei (Duchassaing and
Michelloti, 1860), whose colonies provide habitat for
most of the epifaunal invertebrates at the site. The large

A. mean annual temps

71.5
1974 1977 1980 1983 1986 1989 1992 1995 1998

B. mean high temps

84
83.5

83

82.5
82

81.5
81

80.5
1974 1977 1980 1983 1986 1989 1992 1995 1998

C. mean low temps

54 ;
1974 1977 1980 1983 1986 1989 1992 1995 1998

FIGURE 1. Mean annual (A), high (B), and low (C) air temperatures (in degrees Fahrenheit) for Fort Pierce,
Florida, from 1974 to 1998. Note: lowest, 54°F = 12.2°C; highest, 83.6°F = 28.7°C.

232 °é

mounds produced by the sabellariid worm Phragmato-
poma lapidosa Kinberg, 1867 stay clean and unfouled
when the worms are growing actively but break down as
they age, the mounds dissolving or becoming riddled with
holes and channels in which other organisms settle.

At the November census, water temperature was 23.3°C.
New sabellariid tubes were covering the old eroded sabel-
lariid mounds on many of the rocks. The hydroids Thy-
roscyphus ramosus and Eudendrium carneum Clarke, 1882
were in an active phase of growth. The cyclostome Crisia
elongata Milne-Edwards, 1838 was the most common
bryozoan found; masses of short young Crisia colonies
were attached to hydroid roots and branches. Colonies
of the encrusting cheilostome Watersipora subtorquata
d’Orbigny, 1852 were also common, attached directly to
the rocks near the low water mark.

At the February census the water temperature was
23.1°C. Hydroids had proliferated. Thyroscyphus and
Eudendrium colonies were thriving, and Tubularia sp.
and Halocordyle disticha (Goldfuss, 1820) were also pres-
ent along with colonies of the octocoral Carijoa riisei. The
worm reef was extensive and in healthy condition. Water-
sipora was abundant, with some small, recently recruited
colonies present along with large mature colonies. Crisia
was still a dominant, with large mature colonies produc-
ing gonozooids containing yellow embryos.

At the April census water temperature was 25.3°C.
Old fouled colonies of Watersipora were still present, but
the most abundant encrusting cheilostome was Thalamo-
porella floridana Osburn, 1940, which formed thin whit-
ish crusts and bilaminate expansions around the stems of
Thyroscyphus. Crisia colonies with gonozooids were still
abundant.

At the July census the water temperature was 29.2°C.
The worm reef mounds were crumbling in places but still
showed areas of active growth. Carijoa was abundant, and
there were large, well-grown colonies of Thyroscyphus,
still active, with functional polyps and characteristic gar-
licky smell. The most abundant bryozoan at this census
was the primitive cheilostome Aetea sica (Couch, 1844).
This bryozoan has a runner-like growth form, producing
uniserial rows of semierect zooids, and an ephemeral life
history. Species of Aetea occasionally appear in an area
in large numbers, encrusting almost every substratum. At
other times they may be rare or absent at the same locality.
At this census Aetea colonies were attached to sabellariid
worm tubes, sponges, and Codium species of algae, as well
as to hydroid stems. Crisia was still abundant, but colonies
were short and there were very few gonozooids. The other
species common at this census was the branching cteno-

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

stome Amathia distans Busk, 1886, whose colonies form
limp yellow-speckled clumps. They were attached to vari-
ous substrata, including the senescent worm reef mounds.

Overall, the North Beach Breakwater site was re-
markably stable in its bryozoan fauna over the 24-year
interval. Thirty species were recorded at the site during
this study, including 20 of the 31 species originally found
there (Table 1). The species that were dominant in the
original survey—Amathia vidovici (Heller, 1867), Bea-
nia hirtissima (Heller, 1967), Beania klugei Cook, 1968,
Celleporina hassalli (Johnston, 1848), Crisia elongata,
Pasythea tulipifera (Ellis and Solander, 1786), Savignyella
lafontii (Audouin, 1826), Synnotum aegyptiacum (Aud-
ouin, 1826), Thalamoporella floridana, and Watersipora
subtorquata—were still abundant and were present dur-
ing at least three of the four censuses. In addition, two
new species were common at this site. Amathia alternata
Lamouroux, 1816 was present at other Indian River sites
in the past and still occurs at those sites, but it had not
previously been recorded at the North Beach Breakwater.
Caulibugula armata Verrill, 1900 is new to the region
since the original study was carried out.

WALTON ROCKS

This site is located about 13.7 km south of Fort Pierce
Inlet on South Hutchinson Island, on the beach just south
of the Hutchinson Island Nuclear Power Plant (which was
not yet constructed at the time of the original survey). The
habitat consists of exposed sandy beach, with intertidal
beach rock ledges; their upper surfaces are covered year-
round by a macroalgal turf and seasonally by mounds of
sabellariid worm reef. Numerous loose slabs of beach rock
are present in a sandy trough in the surf zone between the
ledges and the low water line. The exposed location makes
collecting at this site difficult or impossible under high surf
or wind conditions, and the full extent of the ledges is re-
vealed only during the lowest tides of the year, and then
only under calm sea conditions. Encrusting bryozoan spe-
cies and branching species such as Scrupocellaria regularis
Osburn, 1940 occur on the undersides of both the beach
rock ledges and loose beach rock slabs and stones. Other
branching and encrusting species grow on the algae and
hydroids attached to the ledges.

At the November census the water temperature was
23.3°C. Most common were the spiny mats of the cheilo-
stome Beania hirtissima, which were found on the under-
side of almost every piece of beach rock. Also common
were the beach rock-encrusting species Exechonella an-
tillea (Osburn, 1927), Schizoporella “unicornis,” and

NUMBER 38 ¢ 233

TABLE 1. Bryozoans found (+) in 1998-1999 four-season resurvey in comparison with those found in

original survey at the North Beach Breakwater, North Hutchinson Island, Fort Pierce, Florida. Domi-
nant species are shown in bold type; a dash (—) indicates species not found during the resurvey.

Species November 1998

February 1999 April 1999 July 1999

Species found in 1974-1975 survey

Aetea sica =
Aeverrillia armata oF
Amathia distans -
Amathia vidovici +
Anguinella palmata -
Antropora leucocypha -
Beania hirtissima +
Beania klugei -
Beania mirabilis -
Bowerbankia imbricata +
Bowerbankia maxima +
Bugula minima -
Bugula turrita -
Celleporina hassalli +
Crisia elongata +
Cryptosula pallasiana -
Exechonella antillea -
Hippoporina verrilli -
Jellyella tuberculata ® ST
Nolella stipata ar
Pasythea tulipifera +
Pourtalesella incrassata * -
Savignyella lafontii +
Scrupocellaria regularis -
Synnotum aegyptiacum +
Thalamoporella floridana +t:
Valkeria atlantica -
Vittaticella contei ae
Vittacella uberrima

Watersipora subtorquata *
Zoobotryon verticillatum -

ar

= - +
+ = —

Additional species found, 1998-1999

Amathia alternata +
Caulibugula armata aP
Caulibugula pearsei -
Biflustra arborescens 4 -
Biflustra denticulata 4 -
Bugula neritina -
Bugula stolonifera =
Parasmittina sp. 3 =
Rhynchozoon sp. -
Schizoporella “unicornis” -

4 Species for which nomenclature has been revised since Winston (1982).

Pourtalesella incrassata (Canu and Bassler, 1928), actively
growing peach or pink colonies with red embryos pres-
ent in ovicells, along with the ctenostome Nolella stipata
Gosse, 1855. Nolella zooids are straight mud-covered
tubes resembling miniature polychaete tubes. They are
connected by a thin stolon, but at this site zooids were so

thickly aggregated that the stolons were invisible and the
colony appeared as a fuzzy mat of tubes.

At the February census the water temperature was
23.0°C. The wind was strong because of a cold front,
and the surf was high, making collection difficult. The
macroalgal turf was thriving and mostly unfouled except

234 °

by epiphytic hydroids. There were few branching bryozo-
ans. Loose rock in the trough was almost all buried under
sand. The colonies of beach rock-encrusting bryozoans
collected were abraded and bleached in color.

At the April census the water temperature was 23.7°C.
The algal turf was growing luxuriantly. Many more beach
rock stones, some freshly broken off the ledges, were uncov-
ered. The undersides of most rocks were completely covered
by a cryptic community that included zooanthids, didemnid
ascidians, sponges, anemones, and branching and encrusting
bryozoans. Beania hirtissima was again dominant, but other
colonies of encrusting bryozoans, including Schizoporella
“ynicornis,” Exechonella antillea, Watersipora subtorquata,
and Cryptosula pallasiana (Moll, 1803), were brightly col-
ored and healthy. Nolella stipata zooids were clean and
translucent, less mud-coated than in February. Colonies
were sexually reproductive, as well; many zooids brooded
two or three yellow-ochre eggs near their distal ends.

At the July census water temperature was 30.9°C.
Surf was moderate, sand had filled in around ledges again,
and a considerable amount of detached beach rock ledge
algae was washed up on the beach. The undersides of large
beach rock slabs still had a healthy cryptic fauna consist-
ing of zooanthids, ascidians, sponges, and bryozoans on
their undersides, despite being buried in sand. Dominant
bryozoans were Exechonella antillea, Biflustra denticulata
(Busk, 1856), and Beania hirtissima, as well as Nolella sti-
pata (which was still reproducing), plus two erect branch-
ing species, the ctenostome Amathia vidovici and the cy-
clostome Crisia elongata, both present as large, old, fouled
colonies.

This site, Walton Rocks, had been the most diverse
intertidal site in the original study, with 36 species re-
corded at that time. Twenty-five of the same species were
found in 1998-1999 (Table 2). Of the dominant species in
the original survey, all were still present in at least two of
the four censuses, and all but one, Parasmittina betamor-
phaea Winston, 2005, was present at three of the four.

The biggest change at this site was a decline in abun-
dance of Cryptosula pallasiana and its apparent replace-
ment in beach rock undersurface habitats by Exechonella
antillea, which in 1974 had been found only once, at the
North Beach Breakwater, and which had not been col-
lected at Walton Rocks.

JOHNSON House SEAGRASS BED, INDIAN RIVER LAGOON
This seagrass bed is located about 9.7 km north of Fort

Pierce Inlet. It lies in a shallow cove just north of the Harbor
Branch Canal, behind the Johnson residence on the campus

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

of Harbor Branch Oceanographic Institution. The grass
bed has been the site of several studies of seagrass and soft
substratum communities (e.g., Mook, 1976; Kulczycki et
al., 1981; Virnstein and Carbonara, 1985; Virnstein and
Howard, 1987) and was one of the bryozoan sites studied
monthly in 1974-1975. The turtle grass, Thalassia testu-
dinum Banks and Soland. ex Koenig, is the most abundant
seagrass at this site, but manatee grass, Syringodium fili-
forme Kuetz., is also common. Drift algae float among the
grass blades.

At the November census water temperature was
24.4°C. Collections were made of all substrata: Thalas-
sia, Syringodium, and drift algae. Most drift algae were
fouled by a colonial ascidian, Lissoclinum fragile (Van
Name, 1902). The stoloniferous ctenostome Bowerbankia
maxima Winston, 1982, and the encrusting cheilostome
Conopeum tenuissimum (Canu, 1908) were the dominant
bryozoans.

At the February census, the water temperature was
15.2°C, with a strong north wind. Masses of drift algae
had been cast up on shore. Bowerbankia maxima, Cono-
peum tenuissimum, and the branching cheilostome Bugula
neritina (Linnaeus, 1758) were the dominant bryozoans
on seagrass and drift algae, respectively.

At the April census the water temperature was 23.5°C.
Drift red algae appeared bleached in color compared with
their February condition; other algal species appeared to
be thriving. There had been a new settlement of spirorbid
polychaetes onto the seagrass since February, and Cono-
peum had decreased in abundance on Thalassia. However,
there were larger numbers and larger sexually reproduc-
tive colonies of Bugula neritina on the Syringodium, along
with small recent recruits.

At the July census the water temperature was 29.7°C.
Thalassia and Syringodium blades were heavily fouled by
filamentous algae and hydroids. Large colonies of Bower-
bankia maxima, clean and healthy in appearance, with
long free-trailing masses of stolons and zooids, occurred
on the drift algae. Conopeum tenuissimum and Schizo-
porella floridana Osburn, 1914, with recently settled re-
cruits and with embryos in mature colonies, were found
on the Thalassia.

In the original study nine species of bryozoans were
recorded from this site. Six of these were collected at
least once in the re-study (Table 3). The dominant spe-
cies, Conopeum tenuissimum, Schizoporella floridana,
and Bowerbankia maxima, remained unchanged. Four
additional species, Aetea sica, Aeverrillia armata (Verrill,
1873), Hippoporina verrilli Maturo and Schopf, 1968,
and Scrupocellaria “bertholletii,” not recorded here in

NUMBER 38

TABLE 2. Bryozoans found (+) in 1998-1999 four-season resurvey in comparison with those found in
original survey at Walton Rocks, South Hutchinson Island, St. Lucie County, Florida. Dominant species
are shown in bold type; a dash (-—) indicates species not found at this location during the resurvey.

Species

November 1998 February 1999

April 1999

July 1999

Aetea sica

Alcyonidium polypylum
Amathia alternata
Amathia distans
Anguinella palmata
Antropora leucocypha
Beania hirtissima

Beania klugei

Biflustra denticulata 4
Bowerbankia gracilis
Bowerbankia imbricata
Bowerbankia maxima
Bugula neritina

Bugula stolonifera
Bugula turrita

Bugula uniserialis
Caulibugula pearsei
Celleporella carolinensis
Crisia elongata
Cryptosula pallasiana
Electra bellula

Jellyella tuberculata @
Microporella umbracula
Nolella stipata
Parasmittina betamorphaea
Pourtalesella incrassata *
Savignyella lafontii
Schizoporella “unicornis”
Scrupocellaria regularis
Sundanella sibogae
Synnotum aegyptiacum
Thalamoporella floridana
Vittaticella contei
Vittacella uberrima
Watersipora subtorquata
Zoobotryon verticillatum

Aimulosia spp

Amathia vidovici
Celleporaria sp. 2
Escharoides costifer
Exechonella antillea
Biflustra arborescens
Lichenopora sp.
Parasmittina sp. 2

Pasythea tulipifera
Scrupocellaria “bertholletii”

Species found in 1974-1975 survey

Additional species found, 1998-1999

+
++)

$+ +1

++ 14

++ 44+

EE

4 Species for which nomenclature has been revised since Winston (1982).

235

236 °

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

TABLE 3. Bryozoans found (+) in 1998-1999 four-season resurvey in comparison with those found

in original survey at the Johnson House Seagrass Bed, Harbor Branch Oceanographic Institution, Link

Port, Fort Pierce, Florida. Dominant species are shown in bold type; a dash (-) indicates species not

found at this location during the resurvey.

Species November 1998 February 1999 April 1999 July 1999
Species found in 1974-1975 survey
Amathia distans - = = =
Beania klugei = = a a
Bugula neritina - + + a
Bowerbankia gracilis - - = a
Bowerbankia maxima + + + aL
Conopeum tenuissimum + + + +
Electra bellula = = = a
Nolella stipata — + st eo
Schizoporella floridana + + A aL
Additional species found in resurvey
Aetea sica = = ae ae
Aeverrillia armata +e = = =
Hippoporina verrilli +r = = =
Scrupocellaria “bertholletii” - = + =

1974-1975, were also found at one or more censuses in
1998-1999. The three species not found during the re-
study, Amathia distans, Bowerbankia gracilis Leidy, 1855,
and Electra bellula (Hincks, 1881), were still present in
the lagoon at other sites.

A1A CAUSEWAY

In addition to the three sites from the original study,
one new site was also surveyed quarterly. The site is a
shaded spot under the east end of the Route A1A causeway
bridge to North Hutchinson Island. This site was chosen
because of its position in the Indian River Lagoon, about
3 km north of the mouth of Fort Pierce Inlet, and close to
Little Jim Island, where in 1989 a Scrupocellaria species
previously unrecorded in the region had first been collected
(Winston, 1995). Material was collected from bridge pil-
ings, from drift algae, and from submerged wood.

At the November census water temperature was
23.3°C. The most abundant species were Bugula neritina,
Caulibugula armata, Bugula stolonifera Ryland, 1960 (the
latter two reproductive), and Zoobotryon verticillatum
(Delle Chiaje, 1828). Medium-sized Zoobotryon colo-
nies had some areas of new growth with actively feeding
polypides.

At the February census the water temperature was
21.1°C with a cold north wind and turbid water condi-
tions. Bugula neritina was again dominant, with large,

bright wine red-colored, sexually reproductive colonies.
Other abundant species were Amathia vidovici (colonies
mostly mud coated, but with clean actively growing branch
tips) and long stalks of Caulibugula armata. Zoobotryon
verticillatum was present only as short, heavily fouled,
and senescent clumps.

At the April census water temperature was 23.2°C,
with almost no wind and extremely clear water. Bugula
neritina was still dominant on bridge pilings, with more
mature and senescent colonies than in February. Zoobo-
tryon verticillatum was still present, as large colonies drift-
ing among seagrasses and short clumps attached to pilings,
all of them heavily fouled, but with some young actively
growing branches. Caulibugula armata was still present,
with large and unfouled colonies. Amathia vidovici was
still abundant, but colonies were heavily fouled.

At the July census water temperature was 28.8°C,
wind calm, with fairly clear water (visibility about 1 m).
Bugula neritina and Zoobotryon verticillatum were absent.
Dominant species were Caulibugula armata (old, fouled
colonies, with many brown bodies in the lower parts of
branches, but with zooids containing feeding polypides
and ovicelled zooids containing creamy white embryos
near branch tips), Savignyella lafontii, a delicate branch-
ing cheilostome, Nolella stipata, and Amathia vidovici (as
small, heavily fouled colonies).

Twelve species were found at this site (Table 4), mak-
ing it less diverse than the open coast sites but more diverse

NUMBER 38 ¢ 237

a ae ee ee eee >
TABLE 4. Bryozoans found (+) in the 1998-1999 four-season survey at the AIA Causeway Bridge,

North Hutchinson Island, Fort Pierce, Florida. Dominant species are shown in bold type; a dash (-)
indicates species not found at this location during the resurvey.

Species November 1998

February 1999

April 1999 July 1999

Aetea sica

Amathia vidovici
Beania klugei
Bowerbankia gracilis
Bowerbankia maxima
Bugula neritina
Bugula stolonifera
Caulibugula armata
Nolella stipata
Savignyella lafontii
Scrupocellaria “bertholletii”

t sre il

t+te¢ 41

Zoobotryon verticillatum

than the Johnson House Seagrass Bed site (10 species) fur-
ther up the lagoon from Fort Pierce Inlet. Species composi-
tion was stable; most species found there were collected in
at least three of the four censuses. Overall dominants were
Amathia vidovici, Bugula neritina, Zoobotryon verticilla-
tum, and Caulibugula armata, a species that had not been
collected in the area until about 1994.

DISCUSSION

In the 1974-1975 study, 55 species were recorded
from all lagoon and shallow coastal sites. Forty-nine spe-
cies were recorded at the three sites later resurveyed. Dur-
ing the 1998-1999 survey, 39 species were found at those
three sites. Thus, 80% of the bryozoan species known
originally from those sites were recollected after a 24-year
interval, despite a smaller sampling effort (4 versus 12 col-
lections). Seventy percent of the species found originally
from all inshore sites combined were also found in the
four-site resurvey, again with a much smaller sampling
effort involved. There has been remarkable stability in
species composition of the bryozoan fauna over the time
period.

Sixteen species had additional localities (that is, they
were present in the area originally, but occurred at a dif-
ferent site in the second study than that from which they
had been recorded in the original survey), indicating that
most species were not restricted to one site and could be
expected at any or all sites provided the appropriate sub-
stratum and environmental conditions were present. Even

though most of the species involved have nonfeeding, rap-
idly settling larvae, there is apparently enough dispersal
and recruitment that disappearance from one site would
not mean that a species would disappear from the region
entirely. Only one species, Schizoporella floridana, was
limited to one site, the Johnson Seagrass Bed, and to one
substratum, Thalassia testudinum, and was not collected
elsewhere in 1998-1999.

Species new for inshore intertidal sites, but known from
offshore hard substrata or algae (Winston and Eisman,
1980; Winston and Hakansson, 1986), included Aimulosia
uvulifera (Osburn, 1914), Aimulosia pusilla (Smitt, 1873),
and Escharoides costifer (Osburn, 1914). Four species
were newly recorded for the region during the study: two
species of Parasmittina, a species of Celleporaria, and a
Lichenopora species.

Although species composition remained very stable,
species abundances changed considerably, not only from
season to season but also between the two studies. The
most notable changes involved the decline in abundance
of the warm temperate species Bugula stolonifera, Cryp-
tosula pallasiana, and Hippoporina verrilli, all of which
have western Atlantic distributions extending northward
to Long Island or Cape Cod. During the original study
period abundant Bugula stolonifera colonies were found
attached to the proximal portions of Bugula neritina colo-
nies. In the re-study only a few colonies were found, and
they were not in association with Bugula neritina. Hip-
poporina verrilli was a common species on Indian River
Lagoon panels (Mook, 1976) and on panels and seagrasses
in 1974-1975, and it was also found at two coastal sites

238 °

at that time. Reproduction and settlement were heaviest in
the cooler months (October—January). In the re-study only
a few small colonies were found at the Johnson Seagrass
Bed. Cryptosula pallasiana is a cosmopolitan temperate
fouling species. In 1974-1975 it occurred at four intertidal
coastal sites. In the re-survey, however, it was found only
at Walton Rocks where it was much less abundant under
beach rock stones than originally. Instead, in the under-
rock habitat the dominant encrusting bryozoans in 1998-
1999 included Exechonella antillea, a Caribbean species
which, in the original study, had been collected only one
time, at the North Beach Breakwater site. That original
record itself may have indicated a range expansion for the
species because a distributional survey by Maturo (1968)
reported the species only from Miami south.

The other new species in the study are similarly
warm-water species. Caulibugula armata was described
by Verrill from Bermuda, and it is known from the Tortu-
gas, Puerto Rico, and Brazil, according to Osburn (1940).
Aimulosia pusilla was described from the Tortugas by
Smitt (1873) and Aimulosia uvulifera and Escharoides
costifer from the same locality by Osburn (1914). The
typical Scrupocellaria bertholletii is a circumtropical spe-
cies, often associated with coral reefs (Winston, 1986),
but Indian River and other western Atlantic specimens
show some morphological differences to those from other
localities, indicating that Scrupocellaria bertholletii is a
species complex rather than a single widespread species. It
was first recorded in the Indian River lagoon in 1989 and
continues to occur at both coastal and lagoon sites. The
genera Celleporaria and Parasmittina contain numerous
species that are extremely successful in both tropical foul-
ing and cryptic coral reef communities (Winston, 1986).
The addition of species in this group is not surprising.

The increase in warm-water species has continued
since the re-study was completed. Nellia tenella (Lamarck,
1816), another circumtropical fouling and reef-associated
species, was first found in the Indian River area in 1999,
in intertidal collections in Fort Pierce Inlet. It has been
found every year since then, although its abundance has
varied. Hippopodina irregularis, a species described form
Guanica Harbor, Puerto Rico, by Osburn (1940), was
first found on Syringodium seagrass in Fort Pierce inlet
in the summer of 2001. Schizoporella pungens (Canu and
Bassler, 1928), the massive dark purple, Caribbean—Gulf
of Mexico Schizoporella, whose colonies are characteristi-
cally found on submerged mangrove roots and in harbor
fouling communities, had been on noted on drift plastic
items washed ashore in the area for several years, always
with an associated fauna of small corals and Millepora

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

species that suggested the debris had been colonized fur-
ther south, perhaps in the Straits of Florida or the Flor-
ida Keys. Schizoporella pungens colonies first recruited
to panels in Indian River Lagoon (Faber Cove), as well
as to numerous benthic substrata in Fort Pierce Inlet be-
tween July 2002 and July 2003. Celleporaria sherryae
Winston, 2005, another Caribbean fouling and shallow
reef-associated species, has also appeared at some coastal
sites (2001) and within the Fort Pierce Inlet (2003).

Reasons for the increase in warm-water species are
harder to identify. One explanation might be global warm-
ing. As noted by many recent studies, the decade of the
1990s was the warmest on record (Levitus et al., 2000). The
effects of warming seawater temperatures on marine organ-
isms, including bryozoans (Kelmo et al., 2004), have been
noted worldwide. In addition to direct effects on growth and
survival of benthic organisms, changes in water temperature
also affect food supply (Menge et al., 2008; Richardson,
2008), as well as producing indirect effects via changes in
ocean chemistry and circulation (Harley et al., 2006).

For these collecting sites no records of seawater tem-
perature exist for the entire time period of the two studies
(1974-1999). However, as these sites are all intertidal, it
seemed reasonable to make use of the published air tem-
perature data that were available for Fort Pierce as a sub-
stitute. Although mean annual temperatures and mean
annual high temperatures (based on monthly averages)
showed no discernible statistically significant pattern (Fig-
ure 1A,B), there is a suggestion in the data that mean an-
nual low temperatures (Figure 1C) have increased over the
time period. If warm-water species are more susceptible
to cold-water shock than high temperatures, as has been
shown in studies of Florida fish kills after freezes in the
region (Gilmore et al., 1978), warmer winter temperatures
might be a factor permitting the invasion and survival
of populations of the more tropical species, as has been
shown to be the case for some introduced marine inverte-
brates in other studies (Stachowicz et al., 2002).

However, other factors are involved. The Indian River
Lagoon is part of the Intracoastal Waterway, a passage
for boat traffic moving up and down the Atlantic coast,
as well as in and out of the Gulf of Mexico and the Ca-
ribbean. Fort Pierce has a small commercial port with
shipping traffic from the Bahamas (especially Freeport,
where containers from China and other distant sources
are transferred for transshipment into the USA), the Gulf
of Mexico, and the Caribbean, as well as U.S. ports along
the east coast. Species could be introduced through ballast
water exchange by larger ships, as well as by hull fouling
of small and large vessels.

Although the stability of the bryozoan fauna over this
time period gives a positive picture of the health and sta-
bility of the lagoon epifauna overall, there is no way to
predict the long-term impact of these factors. The depen-
dence of many bryozoans on living substrata such as sea-
grasses, hydroids, and octocorals also makes it clear that
disturbances affecting substratum organisms would have
a major impact on the bryozoans and would probably be
more destructive to their local diversity than the environ-
mental fluctuations noted so far.

ACKNOWLEDGMENTS

I thank Dr. Mary Rice, Dr. Valerie Paul, and the staff
at the Smithsonian Marine Station (SMS) for logistical sup-
port for this project, as well as for many other projects over
the years. Special thanks are given to Julie Piraino (also of
SMS) for SEM and digital camera assistance. I thank Drs.
Mark and Diane Littler (Department of Botany, National
Museum of Natural History) for collections of bryozoans
from offshore algae. I also thank the Florida Department
of Environmental Protection, Division of Recreation and
Parks, for the permits (5-98-49 and 5-99-24) to collect
bryozoans at the Fort Pierce Inlet State Park breakwater.
This work is Smithsonian Marine Station at Fort Pierce
(SMSFP) Contribution No. 762.

LITERATURE CITED

Gilmore, R. G., Bullock, L. H, and F. H. Berry. 1978. Hypothermal Mor-
tality in Marine Fishes of South-Central Florida. Northeast Gulf
Science, 2:77-97.

Harley, C. D., A. R. Hughes, K. M. Hultgren, B. G. Miner, C. J. B. Sorte,
C. S. Thornber, L. EF Rodriguez, L. Tomanek, and S. L. Williams.
2006. The Impacts of Climate Change in Coastal Marine Systems.
Ecology Letters, 9:228-241.

Kelmo, F., M. J. Atrill, R. C. T. Gomes, and M. B. Jones. 2004. El Nino
Induced Local Extinction of Coral Reef Bryozoan Species from
Northern Bahia, Brazil. Biological Conservation, 118:609-617.

Kulczycki, G. R., R. W. Virnstein, and W. G. Nelson. 1981. The Relationship
Between Fish Abundance and Algal Biomass in a Seagrass-Drift Algae
Community. Estuarine, Coastal and Shelf Science, 12:341-347.

Levitus, S., J. I. Antonov, T. P. Boyer, and C. Stephens. 2000. Warming of
the World Ocean. Science, 287:2225-2229.

Maturo, F. J. S. 1968. The Distributional Pattern of the Bryozoa of the
East Coast of North America, Exclusive of New England. Atti della

NUMBER 38 ¢ 239

Societa Italiana di Scienze Naturali e del Museo Cicivo di Storia
Naturale di Milano, 108:261-284.

Menge, B., F. Chan, and J. Lubchenco. 2008. Response of a Rocky In-
tertidal Ecosystem Engineer and Community Dominant to Climate
Change. Ecology Letters, 11:151-162.

Mook, D. 1976. Studies on Fouling Invertebrates in the Indian River.
Bulletin of Marine Science, 26:610-615.

Osburn, R. C. 1914. Bryozoa of the Tortugas Islands, Florida. Carnegie
Institution of Washington Publication, 182:181-222.

. 1940. “The Bryozoa of Porto Rico with Resumé of the West
Indian Bryozoan Fauna.” In Scientific Survey of Porto Rico and
the Virgin Islands. Volume 16, pp. 321-486. New York: New York
Academy of Sciences.

Richardson, A. J. 2008. In Hot Water: Zooplankton and Climate Change.
ICES Journal of Marine Science, 65:279-295.

Richardson, A. J., and E. S. Poloczanska. 2008. Under-Resourced, Under
Threat. Science, 320:1294-1295.

Smitt, E A. 1873. Floridan Bryozoa Collected by Count L. FE. de Pour-
tales, Pt. 2. Kongliga Svenska Vetenskaps-Akademiens Handlingar,
11:1-83.

Stachowicz, J. J., J. R. Terwin, R. B. Whitlach, and R. W. Osman. 2002.
Linking Climate Change and Biological Invasions: Ocean Warm-
ing Facilitates Nonindigenous Species Invasion. Proceedings of the
National Academy of Sciences of the United States of America,
99:15497-15550.

Swain, H. M., D. R. Breininger, D. S. Busby, K. B. Clark, $. A. Cook,
R. A. Day, D. E. De Freese, R. G. Gilmore, A. W. Hart, C. R. Hin-
kle, D. A. McArdle, P. M. Mikkelsen, W. G. Nelson, and A. J. Za-
horcak. 1995. Introduction to the Biodiversity of the Indian River
Lagoon Conference. Bulletin of Marine Science, 57:1-7.

Virnstein, R. W., and P. A. Carbonara. 1985. Seasonal Abundance and
Distribution of Drift Algae and Seagrasses in the Mid-Indian River
Lagoon, Florida. Aquatic Botany, 3(1):67-82.

Virnstein, R. W., and R. K. Howard. 1987. Motile Epifauna of Marine
Macrophytes in the Indian River Lagoon, Florida: I. Comparisons
among the Three Species of Seagrasses from Adjacent Beds. Bulletin
of Marine Science, 41:1-12.

Winston, J. E. 1978. Polypide Morphology and Feeding Behavior in Ma-
rine Ectoprocts. Bulletin of Marine Science, 28:1-31.

. 1982. Marine Bryozoans (Ectoprocta) of the Indian River Area

(Florida). Bulletin of the American Museum of Natural History,

173:99-176.

. 1986. An Annotated Checklist of Coral-Associated Bryozoans.

American Museum Novitates, 2859:1-39.

. 1995. Ectoproct Diversity of the Indian River Coastal Lagoon.

Bulletin of Marine Science, 57:84-93.

. 2005. Re-Description and Revision of Smitt’s “Floridan Bryo-
zoa” in the Collection of the Museum of Comparative Zoology,
Harvard University. Virginia Museum of Natural History Memoir,
7:1-160.

Winston, J. E., and N. J. Eiseman. 1980. Bryozoan-Algal Associations in
Coastal and Continental Shelf Waters of Eastern Florida. Florida
Scientist, 43:65—74.

Winston, J. E., and E. Hakansson, 1986. The Interstitial Bryozoan
Fauna from Capron Shoal, Florida. American Museum Novitates,
2865:1-5S0.

im,’ rie
,
k >
ip -
‘
'
i yy
t
k
re
te
E
if
)
a
/
7
y
i)
og i
j
1
, -_ =

The Turtles’ Tale: Flagships and

Instruments for Marine Research,
Education, and Conservation

John G. Frazier

John G. Frazier, Conservation and Research Cen-
ter, National Zoological Park, Smithsonian Insti-
tution, 1500 Remount Road, Front Royal, Vir-
ginia 22630, USA (frazierja@si.edu). Manuscript
received 9 June 2008; accepted 20 April 2009.

ABSTRACT. Marine turtles are classic flagship species. Their remarkable natural his-
tory—large body size, dependence on both terrestrial and oceanic environments, delayed
maturity requiring decades to reach adulthood, regular migrations that crisscross ocean
basins, massive reproductive output, mammal-like physiology, and other features—make
them attractive to researchers and the general public alike. This attraction is further en-
hanced by the fact that these reptiles are widely recognized as endangered species. They
are “biomagnets” for people around the world, from various sectors of society; incred-
ible amounts of time, energy, and resources go into diverse types of investigation, public
education, conservation, and international policy directed specifically at these “lowly
reptiles.” Oceanographers, ecologists, geneticists, marine biologists, and specialists from
other related disciplines frequently begin basic research projects on marine turtles. These
activities quickly evolve into large multifaceted programs including conservation activi-
ties, community-based approaches, and public education together with other forms of
development and social projects, and even policy initiatives for promoting regional and
global cooperation in the conservation of these shared resources and the habitats on
which they depend. Besides enhancing better understanding of the biology and ecology of
these animals and nurturing more active and diverse conservation and education initia-
tives, work on marine turtles also promotes much-needed initiatives in interdisciplinary
and international cooperation, which are fundamental challenges to marine work in gen-
eral. This paper provides a summary of the flagship species concept and gives examples of
how work focused on marine turtles has promoted diverse initiatives in marine research,
education, and conservation at multiple scholarly, social, and political levels; it argues
that this approach serves as a critical integrating force to nurture a wider comprehension
and appreciation of the scientific endeavor and its role in society.

FLAGSHIP SPECIES AND THE INCREASE
AND DIFFUSION OF KNOWLEDGE

Scientists, educators, and conservationists who specialize on marine or-
ganisms and marine environments may all be convinced of the fundamen-
tal importance of such things as larval nectophores, pedunculate siphono-
phores, disappearing zooxanthellae, discharged nematocysts, mitochondrial
cytochrome oxidase 1, maximum parsimony, and other indicators of “good
science,” but what of the rest of society? Marine biodiversity is unique yet
poorly understood or appreciated by the general public or decision makers;

242 e

and a central question with which we all must contend
is “How can we promote it?”

Many marine organisms have complex, intriguing life
histories, and marine turtles, comprising just seven living
species, are classic examples. These air-breathing reptiles
are typified by highly complex life cycles: they live with fish
but nest on land, relying on terrestrial, coastal, benthic,
and pelagic environments during different parts of their
life cycle; they can occur in extremely dense concentra-
tions both on land and in the sea; they are “highly migra-
tory,” crossing ocean basins; they take a decade or more
to reach sexual maturity and can live for half a century or
more; and they have highly specialized morphological and
dietary adaptations, including mammal-like physiology. A
single female often lays more than 100 eggs in a nest and
can lay several nests in a season. Their large body size (up
to 1 ton), striking coloration, and primeval appearance all
add to the attractiveness of these marine reptiles. The fact
that marine turtles are globally recognized as endangered
species adds a further level of importance. Hence, these
reptiles are flagship species: ambassadors of the oceans.
The attraction has led to not only enormous interest on
the part of the general public but also disproportionate at-
tention in academic circles (Frazier, 2003a, 2005a, 2005b):
nearly as much research is conducted on just seven species
of marine turtles as is carried out for the remaining 300-
some species of chelonians.

In addition, marine turtles are widely valued as sources
of meat, eggs, oil, skin, and shell, which have been uti-
lized, crafted, and traded for millennia. A global trading
network that supplied elite urbanites of the Mediterranean
with raw materials from the shores of the Indian Ocean
and beyond was well established before the time of Christ,
and the most frequently mentioned commodity was tor-
toise shell (the external keratinous scutes of the hawksbill
turtle, Eretmochelys imbricata Linn.). Intricately fash-
ioned toilet articles, particularly ornamental combs, some
of which were 85 cm wide, as well as a special style of
French furniture luxuriously inlaid with tortoise shell and
metal (“Boulle”), and religious accoutrements have all been
made famous by the tortoise shell used in their creation.
In addition to the tremendous diversity of objects crafted
from turtle parts, these animals have been portrayed for
millennia on a wide variety of media, from cave walls to
carved rocks to delicate ceramics to the cylindrical seals
of ancient Arabia (Frazier, 2003b, 2004a, 2005c). Hence,
they have had very important cultural, social, and spiri-
tual values in many societies. During contemporary times
marine turtles have been celebrated in many and diverse
forms, ranging from symbols of sacred nature and “pris-

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

tine” environments to evidence of the evils committed by
modern society on the environment (Campbell, 2003). All
this conveys upon these animals a wide variety of values,
from cultural and historic to economic and spiritual.

ACTIVITIES FOCUSED ON MARINE TURTLE
RESEARCH AND CONSERVATION

The national marine turtle program in Brazil, which
began as a dedicated study of reproductive biology and
natural history, has evolved into one of the best known
long-term programs in South America and the world in
general, and the attraction of the turtle flagship over the
years has resulted in the incorporation of massive efforts
in public education and community development, includ-
ing alternate livelihoods for community residents, training,
and facilitated interactions between different sectors of gov-
ernment and society, not to mention national counsel for
regional and international policy actions (Marcovaldi et
al., 2005). Similarly, multiyear programs in Uruguay (La-
porta and Miller, 2005), northwestern Mexico (Delgado
and Nichols, 2005), the Caribbean (Eckert and Hemphill,
2005), and Nova Scotia (Martin and James, 2005) con-
duct research on diverse topics such as feeding ecology,
reproductive biology, genetics, migration, and fisheries in-
teractions. All this research, as well as the associated edu-
cational and conservation activities, has been greatly facil-
itated—if indeed not made possible—by the attractiveness
of marine turtles and the ease with which researchers have
been able to make use of these flagship species to promote
interest in collaborating with different research activities.
It is not uncommon for fishermen to go out of their way
not only to inform researchers about sightings and cap-
tures of marine turtles but also to take on extra work,
requiring time, effort, and materials to deliver informa-
tion and specimens to researchers. Frequently this means
allowing, or even inviting, researchers to come onboard
and make free use of the fishermen’s vessels and materials.
Swordfish fishermen in Nova Scotia provide their vessels
as research platforms for the complicated process of cap-
turing, boarding, measuring, instrumenting, and releasing
turtles of half a ton in body weight or more; researchers
are very much aware that the success of their work de-
pends on the altruistic behavior of fishermen (Martin and
James, 2005). Uruguayan fishermen, many of whom live
at a subsistence level, not only invite researchers to make
use of their boats but are active collaborators in the re-
search, attending meetings and participating in presenta-
tions (Laporta and Miller, 2005). A dramatic example of

the level of dedication to, and investment in, marine turtle
projects is Theeram Pakriti Samrakshana Samiti (Coastal
Ecosystem Protection Committee) in Kolavipalam Village,
Kerala, India. A group of artisanal fishermen decided to
protect nesting turtles and their eggs, formed the com-
mittee, built a modest beach station, and now run nightly
beach patrols, maintain an interpretation center with live
turtles, and give public education presentations: all these
activities have been self-organized and self-motivated,
thanks to the attractive power of the turtles (Shanker and
Kutty, 2005). This sort of material and moral support is
difficult to evaluate adequately in simple financial terms,
but it has been absolutely essential in supporting various
aspects of basic research, education, and conservation ac-
tivities. Indeed, many of these activities would not only be
far outside the operational budgets of the organizations
involved but simply impossible to achieve without the full
collaboration of the fishing communities.

Adventure tourism, often referred to as “eco-tourism,”
has been widely promoted around the world with marine
turtles as the central attractants; indeed, there is even an
international travel guidebook that is dedicated specifi-
cally to marine turtle tourism (Devaux and De Wetter,
2000). In addition to paying their travel costs, it is not
uncommon for tourists to actually pay for the privilege
of working as volunteers in turtle research projects, some
of which have been operating for decades (Campbell and
Smith, 2005). In this way the flagship attraction directly
supports research through both funding and the availabil-
ity of trained volunteer assistants.

An incredible diversity of outreach and public educa-
tion has been developed with marine turtles as the center-
piece, a phenomenon common around the world and far
too diverse to summarize easily (Frazier, 2005d). There are
national and regional training programs specific to marine
turtle biology and conservation, and some of these have
been active for more than a decade, during which time they
have seeded well-trained and enthusiastic researchers, ed-
ucators, and conservationists throughout vast areas, such
as India (Shanker and Kutty, 2005), the Caribbean (Eckert
and Hemphill, 2005), and Latin America (Buitrago et al.,
2008; Marcovaldi et al., 2005). In some cases, the activi-
ties and festivals organized by conservationists have been
appropriated by local people, who have completely taken
over what were initially devised to “sensitize” and “moti-
vate” them to collaborate with marine turtle projects. One
of the clearest examples of the rapidly increasing and pow-
erful attraction of marine turtles is the Annual Symposium
on Sea Turtle Biology and Conservation, an event that is
attended by about a thousand people, with representation

NUMBER 38 ¢ 243

from scores of countries and hundreds of presentations
(Frazier, 2003a). By using the turtles as attractants “to get
people in the door,” these activities, events, and projects
clearly transcend the turtles and provide a wide basis of
information on a diversity of marine organisms and en-
vironments, thereby promoting greater interest, research,
and appreciation for these topics.

There is ample evidence that the flagship attraction
can be instrumental for developing popular and political
support to affect local policy decisions, such as the cre-
ation of special protected areas and tourism management
programs (Tisdell and Wilson, 2005). Moreover, interna-
tional maritime and fisheries policies have been directly
affected by international, regional, and national efforts to
conserve marine turtles, particularly through such efforts
as mitigation of fisheries bycatch (Bache, 2005). In fact, an
extraordinary amount of attention has been paid to ma-
rine turtles in the field of international environmental law
(Frazier, 2002). At present there are two bilateral agree-
ments, an incipient trilateral agreement, a program under
a United Nations Environmental Programme (UNEP) Re-
gional Seas convention in the southeast Pacific, a memo-
randum of understanding for the Atlantic coast of Africa,
and another memorandum of understanding for the In-
dian Ocean (both under the United Nations Convention
for Migratory Species), and a “stand-alone” treaty for
the Western Hemisphere, all focused specifically on the
conservation of marine turtles. Every one of these instru-
ments includes measures of habitat protection, and the
term “habitat” is even included in the title of one accord.
Hence, through activities to protect marine turtle habitats
over vast areas, these instruments have direct relevance
to a wide range of marine organisms and environments,
again clearly transcending marine turtles.

In addition to these seven agreements specific to ma-
rine turtles, there are many other international agreements
that are relevant to marine turtle research, conservation,
and education: these include such major global treaties as
the UN Convention on the Law of the Sea (UNCLOS),
Convention on International Trade in Endangered Species
of Wild Fauna and Flora (CITES), Convention on Biologi-
cal Diversity (CBD), and the Convention on the Conserva-
tion of Migratory Species of Wild Animals (CMS) (Wold,
2002). Moreover, intense concern for the conservation of
marine turtles has been instrumental in shaping policy and
management directions in fisheries and other maritime
issues (Bache, 2002). For example, the Inter-American
Tropical Tuna Commission (IATTC), which was consti-
tuted nearly 60 years ago to develop regional manage-
ment of tuna stocks in the Eastern Tropical Pacific, has

244 e

been dealing specifically with accidental capture of marine
turtles since 2003 and has adopted at least eight resolu-
tions to promote mitigation measures on turtle bycatch.
Even the United Nations Food and Agriculture Organiza-
tion (FAQ), originally created to enhance the production
of food, has become intimately involved in marine turtle
conservation. In March 2004 the “Expert Consultation on
Interactions between Sea Turtles and Fisheries within an
Ecosystem Context” was held in Rome (FAO, 2004), fol-
lowed by a technical consultation at which guidelines for
mitigating turtle bycatch in fisheries were proposed (FAO,
2005). These technical considerations and recommen-
dations were then taken up by the political body, FAO’s
Committee on Fisheries, where the proposal was adopted
at a global level (COFI, 2005). The result is a set of rec-
ommendations for all States that are members of FAO
(virtually every country that exists). Some of the specific
actions that States are supposed to carry out include stock
identification and assessment, tagging and genetic studies,
testing mitigation techniques, “pay urgent attention... to
collection of statistics,” collect and share information, and
harmonize conservation and management initiatives.

At an even greater level of political importance was
a dispute brought before the World Trade Organization
(WTO), which challenged the right of a Party to the WTO
to enact unilateral measures that ban certain imports in an
effort to protect marine turtles from capture and mortal-
ity in certain fisheries operations. After several years of
contentious debate and the production and exchange of
thousands of pages of documentation, an WTO Appellate
Body decision released on 22 October 2001 concluded that
because marine turtles are endangered species, countries
can take exceptions to the all-powerful free-trade rules of
the WTO and—following certain procedures—enact uni-
lateral measures to protect turtles, including trade embar-
gos (Bache and Frazier, 2006; Frazier and Bache, 2002).

COMPLEXITIES OF FLAGSHIP PERCEPTIONS

It is important to point out, however, that the inap-
propriate use of a flagship can lead to totally misguided
policies and activities, counterproductive to both environ-
mental and social needs. For example, easy access to highly
attractive hatchling marine turtles led to an explosion of
“sea turtle conservation hatcheries” along the coast of Sri
Lanka, generously funded by unknowing tourists, despite
the fact that these establishments were illegal and had neg-
ative impacts on hatchling recruitment (Tisdell and Wil-
son, 2005). Conservation programs that focus reflexively

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

on an urgent need to do everything possible to protect ma-
rine turtles but ignore local sociopolitical complexities can
create tremendous conflict, for different sectors of society
often have divergent, even conflicting, views on how to
respond to the flagship and what it primarily symbolizes
(Shanker and Kutty, 2005; Frazier, 2008). Although con-
servationists view marine turtles as indisputable symbols
of the need for people to cherish and protect the environ-
ment, other sectors of society—for example, certain eth-
nic groups—see the same turtles in very different ways,
such as symbols of cultural identity and reclamation. This
divergence in perceptions is true both on Pacific islands
(Kinan and Dalzell, 2005) and on a Greek island in the
Mediterranean, where contradictions in perceived value of
the marine turtle flagship have resulted in violence, death
threats, and other forms of intense conflict between differ-
ent sectors of society (Theodossopoulos, 2005).

SHARED RESOURCES—THE ROOT PROBLEM

Because of their life history characteristics (particu-
larly the long lifespan, dependence on a variety of diverse
environments, and dispersal and migration across oceanic
basins), marine turtles provide a classic case of shared re-
sources, or “common property.” Simple, but basic, ques-
tions such as “Who owns turtles?” or “Who has rights to
turtles?” clearly show that many parts of many societies
have direct impacts, rights, and responsibilities relating to
these animals (Frazier, 2004b). This contention is easily il-
lustrated by the fact that more than 2 million reproduc-
tive turtles were taken from the breeding grounds in Pacific
Mexico between 1964 and 1980 (Frazier et al., 2007). Yet,
animals from this population migrate widely throughout
the eastern tropical Pacific, living at different times within
the jurisdiction of different sovereign States or on the high
seas (Morreale et al., 2007). Who had the right to slaughter
so many reproductive animals that are part of the fauna of
a vast region (an action that had enormous implications on
the status of a shared population)? The same question can
be asked of people who pollute the oceans with oil spills,
plastics, or other wastes: What right do they have to con-
taminate a common resource? Similarly, when endangered
species of marine wildlife, such as dolphins, whales, sea-
birds, and marine turtles, are caught and killed in fishing
activities, the question arises: “What right does the fishing
industry have to be killing (even if it is accidentally) wildlife
species that are valued by the citizens of many nations?”

Dealing with shared resources is the root issue for
nearly all questions regarding biological conservation—

particularly in marine environments. Hence, by high-
lighting the importance of this central problem, work on
marine turtles brings even greater attention to this critical
issue, and because these reptiles are regarded globally as
endangered species, their importance is further enhanced.
Investigations on marine turtles that help promote ways to
resolve intractable issues of common property have impli-
cations that go far beyond chelonian biology and natural
history: they bear on the way modern societies interact
with the oceans.

CONCLUSIONS:
PROMOTING MARINE RESEARCH,
EDUCATION, AND CONSERVATION

THROUGH FLAGSHIP SPECIES

The attention given to marine turtles spans the entire
sociopolitical spectrum, from marginalized, politically in-
significant fishing communities to the most politically pow-
erful organizations on the planet. From one extreme of the
political continuum to the other, these animals have been
given extraordinary importance. These local, national, re-
gional, and global policy decisions have enormous impor-
tance in the ways that individuals, governments, and or-
ganizations at various levels assign priorities and allocate
resources. Even if the intent is only to comply superficially
with obligations that are not enforced, the end result is
resources and personnel allotted to some aspect of marine
turtle research, education, and conservation.

Although the scientific enterprise and its practitio-
ners strive to develop and maintain an objective, unbi-
ased view of the world, there is no escaping the fact that
both the enterprise and the practitioners are immersed
within complex social and political systems. The result,
despite the firmest of desires, is that there is close inter-
play and interaction between scientific activities and at-
titudes that dominate in the surrounding society (Rozzi,
1999). In fact, an anthropological study of the scientific
establishment shows not that scientists and their prac-
tices are unique among humanity, but rather that they
are immersed in a world of power struggles, politics, and
myths—little different from the world of the lay public
that is so often demeaned by the scientific community
(Nader, 1996).

There is no inherent reason that information pro-
duced by scientific research will be read, understood,
appreciated, followed, used, or even recognized in the
halls of power; if practitioners of the scientific endeavor
want their information to impact society outside the ivory

NUMBER 38 e¢ 245

towers of academia, it is essential that we learn how to
“package” the information in digestible, understandable,
interesting, and convincing ways (Frazier, 200Se). Flag-
ship species greatly facilitate this exercise for they have
values that are attractive to the general society. Used effi-
ciently and appropriately, such species are powerful tools
for promoting research, education, and conservation of
countless marine issues.

LITERATURE CITED

Bache, S. J. 2002. Turtles, Tuna and Treaties: Strengthening the Links
between International Fisheries Management and Marine Species
Conservation. Journal of International Wildlife Law and Policy,
5:49-64.

Bache, S. J. 2005. Marine Policy Development : The Impact of a Flagship
Species. MAST (Maritime Studies), 3/4:241-271.

Bache, S. J., and J. Frazier. 2006. “International Instruments and Marine
Turtle Conservation.” In Sea Turtles on the Indian Subcontinent,
ed. K. Shanker and B. C. Choudhry, pp. 324-353. Hyderabad, India:
Universities Press.

Buitrago, J., H. J. Guada, and E. Doyle. 2008. Conservation Science in
Developing Countries: An Inside Perspective on the Struggles in Sea
Turtle Research and Conservation in Venezuela. Environmental Sci-
ence and Policy, 11(6):562-578.

Campbell, L. M. 2003. “Contemporary Culture, Use, and Conservation
of Sea Turtles.” In The Biology of Sea Turtles, Volume 2, ed. P. L.
Lutz, J. A. Musick, and J. Wyneken, pp. 307-338. Boca Raton,
Fla.: CRC Press.

Campbell, L. M., and C. Smith. 2005. Volunteering for Sea Turtles?
Characteristics and Motives of Volunteers Working with the Carib-
bean Conservation Corporation in Tortuguero, Costa Rica. MAST
(Maritime Studies), 3/4:169-173.

Committee on Fisheries (COFI). 2005. “Outcome of the Technical
Consultation on Sea Turtles Conservation and Fisheries, Bang-
kok, Thailand, 29 November—2 December 2004, COFI/2005/7.”
Committee on Fisheries, Twenty-Sixth Session, Rome, Italy, 7-11
March 2005. Rome: Food and Agriculture Organization of the
United Nations.

Delgado, S., and W. J. Nichols. 2005. Saving Sea Turtles from the Ground
Up: Awakening Sea Turtle Conservation in Northwestern Mexico.
MAST (Maritime Studies), 3/4:89-104.

Devaux, B., and B. De Wetter. 2000. On the Trail of Sea Turtles.
Hauppauge, N. Y.: Barron’s Educational Series.

Eckert, K. L., and A. H. Hemphill. 2005. Sea Turtles as Flagships for
Protection of the Wider Caribbean Region. MAST (Maritime Stud-
ies), 3/4:119-143.

Food and Agriculture Organization (FAO) of the United Nations. 2004.
Papers Presented at the Expert Consultation on Interactions be-
tween Sea Turtles and Fisheries within an Ecosystem Context,
Rome 9-12 March 2004, FIRM/R738 Suppl. Rome: Food and Ag-
riculture Organization of the United Nations.

. 2005. Report of the Technical Consultation on Sea Turtles
Conservation and Fisheries, Bangkok, Thailand, 29 November-2
December 2004, FIRM/R765 (En). Rome: Food and Agriculture
Organization of the United Nations.

Frazier, J., ed. 2002. International Instruments and the Conservation of
Marine Turtles. Journal of International Wildlife Law and Policy,
5:1-207.

. 2003a. Why Do We Do This? Marine Turtle Newsletter, 100:

9-15.

246 e

. 2003b. “Prehistoric and Ancient Historic Interactions Between
Humans and Marine Turtles.” In The Biology of Sea Turtles, Vol-
ume 2, ed. P. L. Lutz, J. A. Musick, and J. Wyneken, pp. 1-38. Boca
Raton, Fla.: CRC Press.
. 2004a. “Marine Turtles of the Past: A Vision for the Future?” In
The Future from the Past: Archaeozoology in Wildlife Conservation
and Heritage Management, Volume 3, ed. R. C. G. M. Lauwerier
and I. Plug, pp. 103-116. Oxford: Oxbow Books.
. 2004b. “Marine Turtles: Whose Property? Whose Rights?” In
Marine Turtles: A Case Study of ‘Common Property’ from the
‘Global Commons,’ The Tenth Biennial Conference of the Interna-
tional Association for the Study of Common Property (IASCP): The
Commons in an Age of Global Transition: Challenges, Risks and
Opportunities, 9-13 August 2004, Oaxaca, Mexico. Bloomington,
Ind.: Digital Library of the Commons. http://dlc.dlib.indiana.edu/
archive/00001388/ (access date: 2 June 2009).
. 2005a. Marine Turtles: The Role of Flagship Species in Inter-
actions Between People and the Sea. MAST (Maritime Studies),
3/4:5-38.
. 2005b. Flagging the Flagship: Valuing Experiences from Ancient
Depths. MAST (Maritime Studies), 3/4:273-303.
. 2005c. “Marine Turtles: The Ultimate Tool Kit. A Review of
Worked Bones in Marine Turtles.” In From Hooves to Horns, from
Mollusc to Mammoth: Manufacture and Use of Bone Artefacts from
Prehistoric Times to the Present. Proceedings of the 4th Meeting of
the ICAZ Worked Bone Research Group at Tallinn, 26th-31st of
August 2003, Volume 15, ed. H. Luik, A. M. Choyke, C. E. Batey,
and L. Lougas, pp. 359-382. Tallinn, Estonia: Muinasaja Teadus.
. ed. 2005d. Marine Turtles as Flagships. MAST (Maritime Stud-
ies), 3/4:1-303.
. 2005e. Science, Conservation, and Sea Turtles: What’s the Con-
nection? In Proceedings of the 21st Annual Symposium on Sea
Turtle Biology and Conservation. NOAA Technical Memorandum
NMES-SEFSC-528, compilers M. S. Coyne, and R. D. Clark, pp.
27-29. Miami, Fla.: National Oceanographic and Atmospheric Ad-
ministration.
. 2008. Why Do They Do That? Ruminations on the Dhamra
Drama. Marine Turtle Newsletter, 121:28-33.
Frazier, J., R. Arauz, J. Chevalier, A. Formia, J. Fretey, M. H. Godfrey,
R. Marquez-M., and K. Shanker. 2007. “Human-Turtle Interac-
tions at Sea.” In Biology and Conservation of Ridley Sea Turtles,

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

ed. P. T. Plotkin, pp. 253-295. Baltimore, Md.: The Johns Hopkins
University Press.

Frazier, J., and S. Bache. 2002. Sea Turtle Conservation and the “Big
Stick: The Effects of Unilateral US Embargoes on International
Fishing Activities.” In Proceedings of the 20th Annual Sympo-
sium on Sea Turtle Biology and Conservation. NOAA Technical
Memorandum NMFS-SEFSC-477, compilers A. Mosier, A. Foley,
and B. Brost, pp. 118-121. Miami, Fla.: National Oceanographic
and Atmospheric Administration.

Kinan, I., and P. Dalzell. 2005. Sea Turtles as a Flagship Species: Dif-
ferent Perspectives Create Conflicts in the Pacific Islands. MAST
(Maritime Studies), 3/4:195-212.

Laporta, M., and P. Miller. 2005. Sea Turtles in Uruguay: Where Will
They Lead Us... ? MAST (Maritime Studies), 3/4:63-87.

Marcovaldi, M. A., V. Patiri, and J. C. Thomé. 2005. Projecto TAMAR-
IBAMA: Twenty-Five Years Protecting Brazilian Sea Turtles
Through a Community-Based Conservation Programme. MAST
(Maritime Studies), 3/4:39-62.

Martin, K., and M. C. James. 2005. The Need for Altruism: Engendering
a Stewardship Ethic Amongst Fishers for the Conservation of Sea
Turtles in Canada. MAST (Maritime Studies), 3/4:105-118.

Morreale, S. J., P. T. Plotkin, D. J. Shaver, and H. J. Kalb. 2007. “Adult
Migration and Habitat Utilization: Ridley Turtles in Their Ele-
ment.” In Biology and Conservation of Ridley Sea Turtles, ed.
P. T. Plotkin, pp. 213-229. Baltimore, Md.: The Johns Hopkins
University Press.

Nader, L., ed. 1996. Naked Science: Anthropological Inquiry into Bound-
aries, Power, and Knowledge. New York: Routledge.

Rozzi, R. 1999. The Reciprocal Links Between Evolutionary-Ecological
Sciences and Environmental Ethics. BioScience, 49:911-921.
Shanker, K., and R. Kutty. 2005. Sailing the Flagship Fantastic: Different
Approaches to Sea Turtle Conservation in India. MAST (Maritime

Studies), 3/4:213-240.

Theodossopoulos, D. 2005. Troubles with Turtles. New York: Berghahn
Books.

Tisdell, C. A., and C. Wilson. 2005. Does Tourism Contribute to Sea
Turtle Conservation? Is the Flagship Status of Turtles Advanta-
geous? MAST (Maritime Studies), 3/4:145-167.

Wold, C. 2002. The Status of Sea Turtles under International Environ-
mental Law and International Environmental Agreements. Journal
of International Wildlife Law and Policy, 5:11-48.

Latitudinal Gradients in Recruitment
and Community Dynamics in Marine
Epifaunal Communities: Implications
for Invasion Success

Amy L. Freestone, Richard W. Osman,
and Robert B. Whitlatch

Amy L. Freestone and Richard W. Osman,
Smithsonian Environmental Research Center,
647 Contees Wharf Road, Edgewater, Maryland
21037, USA. Robert B. Whitlatch, Department
of Marine Science, University of Connecticut,
1080 Shennacossett Road, Groton, Connecticut
06340, USA. Corresponding author: A. Freestone
(freestonea@si.edu). Manuscript received 25 July
2008; accepted 20 April 2009.

ABSTRACT. Although the latitudinal diversity gradient, where species diversity peaks
at low latitudes, is well documented, much less is known about how species life history
strategies differ among regions and the implications of these differences for community
development trajectories and particularly for invasion dynamics. As a first step in trying
to understand these factors, we contrast spatial and temporal variation in recruitment
rates and resultant community development of epifaunal assemblages in regions along a
latitudinal gradient from the temperate zone to the tropics. We exposed settlement panels
in four regions: Long Island Sound (Connecticut), Chesapeake Bay and Virginia’s Eastern
Shore (Maryland and Virginia), Indian River Lagoon (Florida), and a portion of the Meso-
american reef in Belize. Panels were deployed for either one to two weeks, to evaluate
recruitment patterns, or one year, to monitor community development. We found that
both recruitment and community development rates were inversely correlated with di-
versity, with the highest rates seen in temperate latitudes and the lowest in tropical Belize.
Seasonal variability in recruitment also varied latitudinally, with strong summer pulses
of recruitment in northern latitudes shifting to low and year-round recruitment at low
latitudes. However, species turnover through time in communities becoming established
was highest in Belize. We conclude with predictions regarding the implications these pat-
terns may have on invasion dynamics at different latitudes.

INTRODUCTION

Latitudinal patterns in diversity have remained an important theme in ecology
for more than a century, yet we still continue to debate the relative contributions
of processes that may cause these patterns (Currie et al., 2004; Mittelbach et al.,
2007). There are many environmental variables that change with latitude, and it
is easy to correlate species distribution patterns with these factors. Unfortunately,
it is as easy to find exceptions to these correlations. In addition, with increased
transport of nonnative species (Ruiz et al., 2000), species distributions continue to
be altered. Although latitudinal gradients in native species diversity are well docu-
mented, studies on terrestrial and freshwater systems suggest latitudinal gradients
in invasion success occur as well (Sax, 2001). However, little work to date has
examined this question in marine systems. Therefore, we have been documenting
latitudinal differences in both the recruitment and the community development of

248 e

marine epifaunal invertebrates as a first step in under-
standing latitudinal differences in species invasions.

The mode by which species successfully invade new
habitats is a pressing ecological research issue (Reymanek
and Richardson, 1996; Williamson and Fitter, 1996; Moyle
and Light, 1996). Current theory predicting the attributes
of successful invaders has largely been developed in terres-
trial environments and usually stresses the importance of
life history traits associated with rapid reproduction and
wide dispersal ability (Rejmanek and Richardson, 1996).
In the far more open marine environment, ocean cur-
rents can disperse larvae and adults of many species for
great distances over relatively short time periods (Jokiel,
1984; Scheltema, 1986). Additionally, man inadvertently
transports countless individuals and species between dis-
crete biogeographic provinces (Ruiz et al., 1997; Carl-
ton, 1999). Given the generally good dispersal abilities of
marine species, those attributes of new species that allow
them to coexist with, or even displace, native species will
be as important as dispersal ability to a species invasion
potential.

The sessile invertebrate or epifaunal community is an
excellent system in which to examine rigorously both the
life history attributes that characterize successful invaders
as well as those attributes of native communities that gov-
ern their susceptibility to invasion. Epifaunal communities
occur in all coastal habitats and can be found in all biogeo-
graphic regions. These communities contain species with a
variety of life histories, yet their principal species are usu-
ally permanently attached as adults and are easy to manip-
ulate. Although the species within these communities differ
among regions, they function in similar ways. Most have
planktonic larvae as the main means of dispersal, feed from
the water column, compete for limited available space, and
are preyed on by a variety of mobile vertebrate and inver-
tebrate predators. Because epifaunal species are sessile and
relatively small in size, natural communities can develop
on small discrete substrates, with larval dispersal and re-
cruitment linking communities within a site or habitat as
well as within a region. These attributes make them ideal
systems that can be experimentally manipulated in the field
to test directly hypothetical relationships while maintain-
ing natural levels of abundance, species composition, and
diversity.

Among epifaunal communities, a major difference is
the number of available species that have some reason-
able probability of recruiting to a particular site within a
region. Osman and Dean (1987) found that these regional
pools of species varied by almost an order of magnitude
and that both the mean number of species found on indi-

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

vidual substrates and the correlated richness at each site
varied greatly among the study sites within each region,
with overlap among sites in different regions. These pat-
terns potentially result from (1) the low probability of re-
cruits of many species in the regional species pool actually
reaching a particular site during the course of investiga-
tion and (2) the high probability of local, within-site dis-
persal of species already present at a particular site. Alter-
natively, high predation and local extinction rates at some
sites may prevent certain species in the regional species
pool from colonizing these sites (Osman and Whitlatch,
1996, 1998). As a first step in trying to understand fac-
tors contributing to the local and regional differences in
diversity and how these are likely to influence species in-
vasions, we have been contrasting temporal variation in
recruitment rates and resultant community development
in regions along a latitudinal gradient from tropical to
temperate regions.

METHODS

We deployed experimental panels in four biogeo-
graphic regions along the eastern seaboard of the United
States and in the Caribbean Sea. These regions were Long
Island Sound in Connecticut (LIS; 41°N), Maryland and
Virginia’s Chesapeake Bay and Eastern Shore region (CB;
37°N), the Indian River Lagoon in Florida (IRL; 27°N),
and the vicinity of Carrie Bow Cay in Belize (BEL; 16°N).
Polyvinyl chloride (PVC) panels, 100 cm?, were abraded to
facilitate settlement of invertebrates and were suspended
on racks underneath docks. The panels were held horizon-
tal with the experimental surface facing the seafloor.

RECRUITMENT

To estimate recruitment in all regions, panels were
sampled either weekly (LIS) or biweekly (CB, IRL, BEL).
At the beginning of each sampling period, four clean pan-
els were exposed at each of the field sites. After the one- or
two-week exposure period the panels were collected and
new panels were deployed. In the laboratory, all panels
were examined under a dissecting microscope, and all at-
tached invertebrates were identified to the lowest possible
taxonomic unit (usually species) and counted.

Sample schedules varied by region as necessitated
by recruitment patterns and destructive storm activity.
Weekly sampling at the LIS Avery Point (AP) site began
in 1991 and has continued unabated to the present. In the
years 1991-1996 sampling was suspended during the win-

ter months when almost no settlement occurs. From 1997
to the present, sampling was conducted continuously with
biweekly sampling during the winter. The remaining LIS
sites (Groton Long Point [GLP] and Mystic River [MR])
were added in 2001 and have been sampled on the same
schedule as the AP site. Sampling in CB and IRL was begun
in 2004 with two sites in each region. The CB sites were at
the Smithsonian Environmental Research Center (SERC)
in the upper Bay and at the Virginia Institute of Marine
Science (VIMS) in the lower Bay. The IRL sites were the
Smithsonian Marine Station (SMS) and the Ft. Pierce In-
let (Inlet). Sampling at the VIMS site was discontinued in
2007 after hurricane damage to the dock, and sampling at
both IRL sites was suspended from September 2004 until
March 2005 because of the loss of docks as the result of
two hurricanes. Sampling in BEL began in December 2004
and continued through February 2006.

DATA ANALYSIS

Recruitment differences among sites within and across
regions were compared by matching means for each sam-
pling time and using paired f tests to analyze for significant
differences. Wilcoxon signed-rank tests were also con-
ducted for each pairing to eliminate the possible effects of
large seasonal differences biasing the results. Because of
the species differences among regions, analyses were done
for total recruitment of all species, pooled invasive species,
and pooled native species. Species identified as cryptogenic
were included with the native species. The number of sam-
pling periods varied greatly among the regions, and we
conducted the analysis of each pair of stations using the
maximum number of sampling periods in common based
on the year and week of sampling. Data were corrected
for exposure time to account for the one- and two-week
sampling periods used in different regions.

COMMUNITY DEVELOPMENT

To measure difference in community development,
experimental panels (same as above) were deployed for
at least one year and nondestructively sampled for in-
vertebrate richness. Four panels were deployed at each
site (three per region) between July and August 2006 to
a depth of 0.6 m below LLT and at least 0.5 m above the
bottom. Panels in LIS, CB, and IRL were sampled itera-
tively 1, 3, and 12 months after deployment. Panels in
BEL were sampled 3, 6, and 12 months after deployment.
Panels were sampled with a dissecting microscope, and
attached invertebrates were identified to the lowest pos-

NUMBER 38 e¢ 249

sible taxonomic unit. Taxonomic richness on each panel
was recorded.

RESULTS
RECRUITMENT

Three types of recruitment patterns are evident. Within
sites there are temporal patterns, among sites within re-
gions there are fairly consistent relationships, and together
these produce broader patterns among the regions.

Within-Site Temporal Patterns

Within each site there are temporal patterns in recruit-
ment that result from seasonal cycles in reproduction and
year-to-year variation in recruitment that can result from a
variety of causes. Seasonal variability in recruitment is most
evident in the two northern regions, LIS and CB, which ex-
perience large variations in temperature. In both regions
recruitment is largely absent during the coldest winter
months. The three sites in LIS are consistent in exhibiting
peak recruitment in the late summer (Figure 1). Recruit-
ment at the GLP site begins earlier and remains consis-
tently higher than at the other sites throughout the whole
season. This site is in shallower water and consequently
experiences lower winter temperatures and higher summer
temperatures (Osman and Whitlatch, 2007) and warms
more quickly in the spring. At the CB sites, the major-
ity of recruitment occurs in the spring and early summer,
with a second, much smaller peak period in the autumn
(Figure 1). Most dominant species in this region such as
barnacles, bivalves, and polychaetes are planktotrophic
with feeding larvae dependent on the spring and autumn
plankton blooms. Recruitment in the remaining two re-
gions, although temporally variable, exhibits no consistent
seasonal cycle. Recruitment occurs year round at both IRL
sites, with the inlet having somewhat higher recruitment
in the summer (Figure 1). Recruitment at the SMS dock is
dominated by several species of barnacles and has much
more sporadic peaks. Finally, BEL recruitment was ex-
tremely low and demonstrated no obvious patterns.

Spatial Variability among Sites within Regions

Within the three regions with multiple sites we have
observed fairly consistent differences among the sites. Based
on the paired ¢ tests of weekly differences in total recruit-
ment over the period 2001 through 2007, the three sites
in LIS were significantly different, with GLP > AP > MR

250

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

n land 12000
a ve = 2 Chesapeake Bay
penne oe —a— SERC

—@— MR —@® VIMS
6000

500

Feb Apr Jun Aug Oct Dec Jan Mar May Jul Sep Nov

30007 Florida

—@— Inlet
—@ sms 207 Belize

0 0
Jan Mar May Jul Sep Nov Jan Mar May Jul Sep Nov

FIGURE 1. Comparison of temporal variation in mean recruitment in the four regions. Individual sites within
regions are shown. Means were based on 1-6 years of data depending on region and the periods over which
recruitment was measured (see Methods). Sampling sites are as follows: New England: AP = Avery Point,
GLP = Groton Long Point, MR = Mystic River; Chesapeake Bay: SERC = Smithsonian Environmental
Research Center, VIMS = Virginia Institute of Marine Science; Florida: Inlet = Ft. Pierce Inlet, SMS = Smith-

sonian Marine Station; Belize: Carrie Bow Cay.

(Table 1). Native species recruitment at the three sites showed
the same pattern while the recruitment of invasive species
was not significantly different among the sites (Table 2).

A similar analysis of the two CB sites for 2004 through
2006 found that total recruitment at the VIMS site was sig-
nificantly greater than at the SERC site (Table 1). Recruit-
ment at both sites was dominated by native barnacles, and
invasive species recruitment was very low. Nevertheless
both native and invasive species exhibited the same pattern
as total recruitment (Table 2). Although experiencing simi-
lar variability in temperature, these two sites differ greatly
in their salinity regimes. The SERC site is in the upper, low-
salinity region of CB whereas the VIMS site is in the lower
CB with higher salinities. In general fewer species recruit at
the SERC site, and barnacle recruitment is much less.

Similarly, the two IRL sites differed significantly
(2004-2006) in total recruitment, with SMS greater than
Inlet (Table 1). Native and invasive species exhibited the

same pattern (Table 2). Although there was little differ-
ence between the sites in temperature and salinity, they
did differ in dominant species, which resulted in strong
differences in total recruitment. Barnacle recruitment (six
different species) was consistently much higher at the SMS
site and this contributed greatly to the overall site dif-
ferences. Most species of bryozoans as well as spirorbid
worms had higher recruitment at the Inlet site. Figure 2
illustrates these differences.

The nonparametric paired analyses of the data from
all three regions were almost identical to those above. The
only difference was that in LIS invasive species recruitment
was significantly greater at GLP than at either AP or MR.

Regional Patterns

The regional differences in temporal and spatial pat-
terns in recruitment can be seen in Figure 1. In LIS the strong

NUMBER 38 e¢ 251

TABLE 1. Results of paired analysis of mean recruitment between each pair of sites. Recruitment data were
paired by sampling time. Mean values are for 2-week sampling periods and vary based on the number of sam-
pling dates in common between any two pairs of sites (df = degrees of freedom;). Significant probabilities (Prob)

are in bold.
Total Annual Recruitment
Site 17 Site 24 df Mean 1 Mean 2 t-ratio Prob > |¢| One-sided
Avery Point Mystic River 229 474.5 299.2 6.39 < 0.0001 < 0.0001
Groton LP 252 446.6 1000.4 8.60 < 0.0001 < 0.0001
SERC 21 208.4 550.8 1.44 0.16 0.08
VIMS 21 251.3 2206.2 2.99 0.007 0.004
SMS 38 638.2 1010.0 pp) 0.03 0.02
Inlet 36 607.5 607.5 1.37 0.17 0.09
Mystic River Groton LP 234 296.7 996.8 9.63 < 0.0001 < 0.0001
SERC 23 152.3 506.2 1.65 0.11 0.06
VIMS 23 184.1 2097.8 3.22 0.004 0.002
SMS 39 425.4 1068.1 4.15 0.0002 0.0001
Inlet 38 451.2 459.9 0.09 0.93 0.46
Groton LP SERC 21 698.5 550.8 0.52 0.61 0.31
VIMS 21 761.3 2275.1 2.11 0.05 0.02
SMS 39 1288.4 1050.5 0.91 0.36 0.18
Inlet 37 1192.5 471.3 3.60 0.0009 0.0005
SERC VIMS 17 381.6 1519.7 2.11 0.05 0.03
SMS 10 341.6 685.3 1.18 0.26 0.13
Inlet 10 341.6 242.1 0.42 0.68 0.34
VIMS SMS 15 2694.9 903.9 2.38 0.03 0.02
Inlet 15 2694.9 169.6 3.01 0.009 0.004
SMS Inlet 71 1045.5 399.6 7.20 < 0.0001 < 0.0001

4 Groton LP = Groton Long Point (GLP); SERC = Smithsonian Environmental Research Center; VIMS = Virginia Institute of Marine Science;

SMS = Smithsonian Marine Station.

seasonality produces a relatively normal distribution in re-
cruitment centered on the summer months of peak tempera-
tures. Peak periods are relatively broad, with 1,000 to 2,000
recruits per panel per week. This overall pattern reflects the
concentration of recruitment by most species in the summer
period. Recruitment in CB is also seasonal but generally
dominated by a few species, with sharp peaks in recruitment
of 3,000 to 10,000 individuals per panel. The pattern in IRL
is more diffuse with recruitment occurring throughout the
year and several sharp peaks of 2,000 to 3,000 recruits per
panel (barnacles) over a background of continuous recruit-
ment. Individual species do have peaks in recruitment but
they do not occur at the same time as in the northern re-
gions. Thus, some species recruit in the winter and others in
the summer, and this difference is reflected in the continuous
total recruitment throughout the year. Finally, recruitment
at the BEL site was extremely low, despite the much greater
species diversity in the region.

Given these patterns, we examined whether total an-
nual recruitment was influenced by the regional differences
in variability and peak abundances. Figure 3 shows the to-
tal mean annual recruitment for each of the sites; no gen-
eral pattern is discernible from these data. Except for BEL,
within-region differences in total annual recruitment are as
great as, if not greater than, differences among regions. Fig-
ure 3 also shows the dominance of barnacle recruitment in
both the low-diversity CB and high-diversity IRL regions,
whereas bryozoans and ascidians dominate recruitment in
LIS. Interregional differences in total recruitment, regard-
less of strong differences in temporal patterns, exhibited no
pattern that could be associated with diversity or latitude.

Results from the paired analyses did show some regional
differences (Tables 1, 2; see Figure 3). For total recruitment
the VIMS site in CB had significantly greater recruitment
than all other sites. The GLP in LIS and SMS in IRL were sig-
nificantly greater than the Inlet IRL, SERC CB, and AP LIS

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

252

"OUT VIII “IJ = [UT “J9ANY = -Y p

1000°0 > 1000°0 > 96'S 6 097 L029 1000°0 > 1000°0 > Ils L8El 8 PLE IZ qo] SWS
+¥00°0 600°0 COE O'ErI 6 €897 7c0°0 £0°0 8E°C 99T OTT ST dU]

Z00°0 100 82°C S¥99 6 €89T €00°0 900°0 eGis V6ET OTT ST SWS SWIA
870 95°0 190 veel CIE +0°0 80°0 S61 L8t v0 Or qo]

LT0 bS°0 €9°0 CTS CIPe 900°0 10°0 IVe TEZt v0 Or SWS

£0°0 90°0 SOC Crobl © 08€ 70°0 $00 IV? IST cl LI SWIA OUTS
$000°0 100°0 Soe £087 €°686 860 ZL‘0 0€'0 (GG C EOC LE qu]

£00 rl0 OS T O'SOZ $6901 €0°0 S00 10°C 99TT 6 81T 6€ SWS

100 70°0 LV'C €9SCC LOS 100 70°0 LVT 8°81 LO0CT Ic SWIA

8r'0 96°0 $00 S6rs 8 rES 70°0 £0°0 LET el! LE9t 1¢ OAS dT UOIoIy
9€°0 1Z0 LE0 8 ELT 80ST cv'0 €8°0 170 L981 +F'00C 8€ qo]

1000°0> 1000°0 LEV €CIL VSCC +0°0 L0°0 $8 6 SSE 6 661 6€ SWS

700°0 €00°0 VOE 8° €L0T TSSt 870 Ss0 09°0 Ove 6 87 €C SWIA

+0°0 80°0 98°T 0-S0S POTT 70°0 €0°0 LEG cl SSE €C OS

1000°0 > 1000°0 > L8°6 SLr8 Vest 80 S$Z0 EO €6rl err VET dT 430TH) JOATY SSAT
80°0 STO oV'T JOLT EL8€ Se0 020 6£0 OT6L CENT 9€ WU

+0°0 L0°0 r8'T 9999 0604+ 1 0) CT 0 ETT S SHE 6 817C 8€ SWS

€00°0 $00°0 ve Py ZL81C 9ST £0°0 £00 v6 8°81 8°S6 1c SWIA

+0°0 80°0 18'T S6rs 8 Trl c0'0 40°0 8c el 9°99 1c OS

1000°0 > 1000°0 > 168 SrS8 6 CLE S70 0s°0 £90 6 SHI LEEl CST dT] 40301

1000'0 > 1000'0 > 699 ev 9°67E 05°0 660 00'0 Grrl 6'bbl 677 FaTYONSA UT Aaa
P2prs-ouO |2|< qoig onel-7 ZT uvayy [ uvoyl P2pis-ouO |2| < qo4d onel-7 ZT ues ] uro; dp eT US el us

JATJENT SATSPAUT

"pjoq ul ore (qo1g) sarpiqeqoid yueoyTUsIS “sous Jo ssred Om Aue UddMI9q UOUTWOD UI sa}ep SuI;dures Jo Jaquinu ay} UO paseq AIvA puke spotted Suljduies YIaM-Z IOF

ore sanjeA uvayy ‘our Surdures fq posted oa vyep WUOWINIDIY “says Jo Ted yoes UdaMJoq WOUWIIINIDAI 9ATJLU PUL IATSBAUT UvIUT JO sIsAjeue pasted Jo s]Nsoy *7 ATAV.L

Balanus improvisus

—@— Inlet
15007 —@- SMS

500

150 be
Bugula neritina

—@®— Inlet

ace —@® SMS

50

2004 2005 2006 2007

NUMBER 38 ¢ 253

a Spirorbis spp

400. —®— Inlet
—@ SMS

300

200

100

60
Hippoporina

°°) _@- sms

4o| —@ Inlet

2004 2005 2006 2007

FIGURE 2. Comparison of recruitment at the two sites in Indian River Lagoon, Florida, for the barnacle Balanus improvisus, spirorbid poly-
chaetes (Spirorbis spp.), arborescent bryozoan Bugula neritina, and encrusting bryozoan Hipporina sp.

sites, and all were greater than the MR LIS site. Recruitment
of native species exhibited similar interregional patterns.
However, for invasive species, recruitment was significantly
higher in IRL than CB, with the LIS sites intermediate.

COMMUNITY DEVELOPMENT
Spatial and Temporal Variability

The speed of community development varied dramati-
cally with latitude. The primary limiting resource, space,
was quickly occupied in the temperate and subtropical
regions by three months, compared to the tropical com-
munities, which took close to a year to attain comparable
spatial coverage (Figure 4). In the northernmost region,

LIS, growth rates were particularly high, and at one site
(AP) panels were completely covered after only one month,
which is a striking comparison to comparably aged com-
munities in BEL (Figure 5). These productive communities
in AP were primarily composed of Diplosoma listerianum,
an invasive colonial tunicate, and Mogula manhattensis, a
solitary tunicate. These animals quickly became too heavy
to remain attached to the panel and sloughed off, provid-
ing another flush of open space to recruiting species. This
punctuated seasonal pulse of productivity of Diplosoma
and Mogula did not occur at the other two sites in LIS, but
growth rates remained high throughout the region.
Overall, communities in LIS experienced less temporal
turnover in species composition than more southerly sites,

9
25 Y

oq |

2% |
so
S505

‘o
0

ox
a,
OD
reas

©
O

LA?
OS
—WRAS
29S

CAA AAAS
ras
Xxx

o,
o,

os
O06

A
Ox

\7
a

TO

va
2995
TX KR

IB RBC. BEG GW
V~LLI PREY GI

“4
CG

eo

N24

TZZZZBXr. BCS
xX
ox

6,

AP GLP MR SERC VjJMS INLET SMS BLZ
GH Other Bd Barnacle KXI' Bryozoan Ascidian

Me Native (VIMS>SMS>GLP=SERC>inlet=AP>MR)
Hi Invasive (SMS>Inlet=AP=MR=GLP>VIMS>SERC)
4500 Paired Means

500

AP GLP MR _ SERC VIMS INLET SMS

Ee All (VIMS>SMS=GLP>Inlet=SERC=AP>MR)
Inv (SMS>Inle=AP=MR=GLP>VIMS>SERC)

PairedMeans

500

AP GLP
> _______ LS a a a EE

MR SERC VIMS INLET SMS

FIGURE 3. Comparison of recruitment among regions. Top: Total
annual recruitment at each of the sites within regions. The contribu-
tions in each region of major taxonomic groups are represented by
colored shading and/or scoring within the histogram bars. Middle:
Mean recruitment of invasive (blue) and native (red) species by sites
within region; significant differences are based on paired analyses
(see Table 2). Bottom: Total mean recruitment by site showing the
contribution of invasive (Inv) and native species. Dashed lines sepa-
rate regions in all graphs.

Percent Open Space

Time Interval
DN OR ae a

FIGURE 4. Total percent cover of open space on panels in the four
regions after 3 and 12 months (CT = Connecticut; VA = Virginia;
FL = Florida; BZ = Belize). In Belize, space occupied by algae was
included as covered space, so percent cover of invertebrates was even
lower than shown.

particularly BEL (Figure 6). Communities in LIS were consis-
tently dominated by bryozoans, particularly Bugula turrita,
and both solitary and colonial tunicates (Figure 7; personal
observations). This observation is in contrast to the higher
rates of species turnover that characterized communities in
tropical BEL (Figure 6; Freestone, unpublished data).

Epifaunal communities in CB had low temporal and spa-
tial variability in species composition compared to other re-
gions. Barnacle recruitment occurred soon after deployment,
in July 2006. After the first month, all panels had 99% to
100% cover (see Figure 4) and were almost completely cov-
ered with barnacles, with few other coexisting species. After
three months, community structure still closely resembled
the one-month communities; however, barnacles began to
die and other species, such as Mogula, various hydroids, and
sabellid polychaete worms recruited (see Figure 7). After one
year, the primary layer of barnacles was less visible, having
been covered with a thick layer of sediment tubes, mostly
from amphipods and worms. Anemones were also common
throughout. Panels deployed at the three sites were also very
similar. Porifera were least common in CB when compared
with other regions throughout the experiment.

Overall, communities in IRL retained almost com-
plete phyla representation through time (Figure 7). All
focal phyla were found on all panels in IRL after three
months, and only Porifera had a very modest decline by
one year. Species in these communities coexisted at very
small spatial scales, with the result that IRL had the most
diverse invertebrate assemblages at the panel scale after

NUMBER 38 e¢ 255

FIGURE 5. Growth rates were higher at northern latitudes, a pattern that is clearly visible in this comparison
between (a) a 3-week-old community from Carrie Bow Cay, Belize and (b) a 4-week-old community from
Avery Point, Long Island Sound (100-cm? panels are shown).

one year (Freestone, personal observation). More species
turnover was apparent over the course of the year at IRL
than in LIS or CB, but not as much as at BEL.

Communities in BEL were characterized by significant
temporal and spatial variability in species composition.
Communities developed much more slowly than did more
northern communities (see Figure 4), and community
composition clearly changed with time. These communi-
ties also varied at very small spatial scales, as panels that
were deployed within a meter of each other harbored very
distinct community assemblages with differing amounts
of open space. In contrast to more northern communities,
BEL communities were more consistently dominated by
polychaetes, Cnidaria (sea anemones, hydroids, coral),
and Porifera. After one year, Porifera clearly dominated
the panels (Freestone, personal observation). Compared to
the bushy and common bryozoan colonies that occur in
LIS, bryozoans in BEL were generally very small, delicate,
and rare.

Similar to the recruitment study, the largest difference
in taxonomic composition of developing communities
across all regions was the presence of barnacles. Barnacles
were common in temperate and subtropical zones but were
completely absent in BEL at 3 months. After 12 months,
their dominance was still seen in CB and IRL, but barnacles
were less common in LIS. However, only one barnacle on
one panel was found in BEL. Although barnacles in tem-
perate and subtropical zones are both intertidal and sub-
tidal, barnacles are almost exclusively intertidal in BEL.

DISCUSSION

Based on our preliminary examination of these on-
going studies, it is clear that there are both strong intra-
regional and interregional patterns in both the recruitment
and the development of epifaunal communities. Seasonality
in recruitment clearly varies with latitude. Strong summer
peaks coupled with the almost complete absence of any
recruitment in the winter were found for most species in
the temperate regions (LIS and CB). The strong dominance
of barnacles in CB resulted in a bimodal pattern gener-
ally associated with the spring—fall plankton blooms upon
which barnacle larvae feed. In IRL there was also a strong
temporal variability in recruitment but neither a consistent
seasonal pattern nor any similarity among species. Finally,
in BEL recruitment was too low to discern any pattern.
There were also fairly distinct patterns among sites within
regions. In the temperate regions, sites showed consistent
differences in numbers of recruits but little difference in the
species recruiting at any one time or in the relative abun-
dances of these species. In the subtropical IRL, there were
greater differences between the two sites in the composi-
tion of the fauna recruiting at any one time. Based on the
low recruitment and greater community variability among
sites in BEL, it would appear that site differences in recruit-
ment in the tropics are likely to be even greater.

The interregional variation in recruitment for native
species was influenced by the variation in barnacle domi-
nance, with the CB and IRL sites showing significantly

256

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 6. Time-series comparison of community development in (a) Long Island Sound after 1, 3, and 12
months and (b) Belize after 3, 6, 9, 12, and 20 months (100-cm? panels are shown). Greater species turnover
occurred in Belize than in Long Island Sound.

higher recruitment than the LIS sites. However, the re-
cruitment of invasive species varied in a similar manner
to patterns described by Sax (2001). The tropical BEL
site had little overall recruitment and no recruitment of
invasive species, the subtropical IRL sites had the highest
recruitment of invasive species, and the higher latitude
temperate sites had lower recruitment of invasive species.
The combined patterns of native and invasive species re-
sulted in invasive species representing a much higher pro-
portion of overall recruitment in IRL than in all other re-
gions. In the low-diversity estuarine CB, invasive species
contributed a very small proportion of total recruitment
(see Figure 3).

Finally, even with the strong differences in recruitment
reflected in the paired analyses, similarities in total annual
recruitment were found between some sites in the three
northern regions. This result suggests that the cumulative
recruitment in more seasonal regions with strong peaks
and in regions with little or no seasonality in recruitment
can be similar.

In BEL, the epifaunal communities were quite spatially
variable in community composition even at the smallest
scale of panels at each site. This pattern is consistent with
the spatial heterogeneity hypothesis for the latitudinal diver-
sity gradient, which states that abiotic variability in tropical
systems allows more species to coexist (see Davidowitz and

; ]
2 |
6 Y Y |
2
sg ] @ Polychaeta
q ] | @ Cnidaria
& ] | a Porifera
@ ] | ™ Bryozoa
8 ] | @ Tunicata
© 7 | m@ Arthropoda
S y
8 j
g |
é |
Lib
Connecticut Virginia Florida Belize
Region
a
€
rf
o
a Z
g Z |
£ Z | G Polychaeta
s Z | @ Cnidaria
2 ] | @Porifera
e j | Bryozoa
2 7 | Tunicata
2 7 | Arthropoda
Z
a Y
8 Z
: ]
e Z
Connecticut Virginia Florida Belize
Region
b

FIGURE 7. Proportion of panels that had listed taxa present at (a) 3
months and (b) 12 months by region.

Rosenzweig, 1998). Interestingly, in contrast to the abiotic
variation that characterizes other systems, such as terrestrial
plant-soil relationships, the settlement panels were identi-
cal in size and material, so substrate composition was not a
source of variability. While it is possible that differences in
subtle small-scale variation in currents or eddies could drive
community variability, a more parsimonious explanation is
that propagule supply is very low and sporadic. Commu-
nity developmental trajectories may therefore be more a re-
sult of random recruitment from a limited larval pool rather
than spatial variability in abiotic conditions.

Another possible explanation that has strong theoreti-
cal underpinnings is that biotic interactions, such as preda-
tion, are also strong and spatially variable in the tropics
(Schemske, 2002). Although this hypothesis has not been
empirically tested in a comprehensive experiment, the idea
that biotic interactions are stronger in the tropics has re-
ceived much theoretical attention (Mittelbach et al., 2007).
Visual observations of the communities in BEL support this

NUMBER 38 ¢ 257

hypothesis. For example, we commonly observed grazing
or saw indirect evidence of grazing (i.e., abrasions) on the
panels from indiscriminate consumers, including gastro-
pods, crabs, and fish. While predation undoubtedly occurs
in temperate environments (Osman and Whitlatch, 1995,
1998, 2004), overall interaction strengths may be weaker
and more spatially predictable in northern latitudes. Spo-
radic and low larval recruitment, spatially variable preda-
tion, and low growth rates in areas of low productivity are
all potential contributors to the spatial and temporal vari-
ability of tropical epifaunal community development.

Our main goal has been to document and contrast
recruitment and community development patterns among
regions along a latitudinal gradient to ascertain potential
differences in the ability of nonnative species to invade
these systems. Except for our sites in BEL, all the regions
we have been studying have nonnative species present, and
such species are often dominant within these epifaunal
communities. In LIS we have found that early recruitment
of invasive ascidians in years with warm winters (Stacho-
wicz et al., 2002a) and their dominance at harbor sites
without native predators that prey on their recruits (Os-
man and Whitlatch, 1995, 1998, 2004) have contributed
to their successful invasion. The strong and consistent tim-
ing of recruitment of native species certainly can create an
opening for invaders that can recruit outside this window.
In CB recruitment is even more constrained temporally
with much higher numbers of recruits, again creating po-
tential temporal windows for invasion. However, as our
community development data have shown, the communi-
ties in both these temperate systems develop rapidly and
thus quickly limit resources for new species. Studies in LIS
(Stachowicz et al., 2002b) have shown that as community
diversity within these systems increases, the communities
become more resistant to invasions, mostly by increasing
the likelihood of limiting open space.

In IRL, space was also rapidly occupied and the
amount of open space remained low after three months.
In addition, the diversity within this system is higher, and
some species are recruiting at any time of the year. Based
on the results of the LIS study (Stachowicz et al., 2002a)
these factors should increase the resistance of this system
to invaders. Although we have found several invasive spe-
cies at our study sites, none of these species appears to be
particularly abundant or dominant. Finally, in BEL we have
observed much more diverse and spatially variable com-
munities. Recruitment and the rates of community devel-
opment are low, and this situation certainly allows spatial
resource to be available for much longer periods of time,
which should create a greater window for species invasion.
However, the extremely high diversity of both epifaunal

258 e

species and predators may inhibit invasion success. It is
also likely that, given the spatial variability in communi-
ties, invaders will face completely different communities at
each site as well as temporal variability in communities
within sites. Although it is much too early in our studies
to link latitudinal variation in recruitment or community
development to invasion success, our preliminary results
do suggest that there is a correlation between decreasing
invasion success and increasing diversity, increasing com-
munity variability, and the reduction in recruitment win-
dows in less seasonal environments, with species varying
greatly in the timing of recruitment.

ACKNOWLEDGMENTS

We thank the Smithsonian Marine Science Network
for support for this research. Funding from the National
Science Foundation (NSF), U.S. Environmental Protection
Agency (EPA), and Connecticut Sea Grant supported the
long-term recruitment studies in Long Island Sound (LIS).
The research at Carrie Bow Cay could not have been done
without the support of Klaus Ruetzler, Michael Carpen-
ter, and the many extremely helpful station managers. The
staff of the Smithsonian Marine Station, Sherry Reed in
particular, was instrumental in the studies being conducted
in Indian River Lagoon (IRL). None of this research could
have been done without their generous support. We offer
special thanks to Gregory Ruiz and Tuck Hines for their
helpful conversations and ideas. This work is Smithson-
ian Marine Station at Fort Pierce Contribution number
786 and contribution number 843 of the Caribbean Coral
Reef Ecosystems Program (CCRE), Smithsonian Institu-
tion, supported in part by the Hunterdon Oceanographic
Research Fund.

LITERATURE CITED

Carlton, J. T. 1999. Molluscan Invasions in Marine and Estuarine Com-
munities. Malacologia, 41:439-454.

Currie, D. J., G. G. Mittelbach, H. V. Cornell, R. Field, J. E Guegan, B. A.
Hawkins, D. M. Kaufman, J. T. Kerr, T. Oberdorff, E. O’Brien, and
J. R. G. Turner. 2004. Predictions and Tests of Climate-Based Hy-
potheses of Broad-Scale Variation in Taxonomic Richness. Ecology
Letters, 7:1121-1134.

Davidowitz, G., and M. L. Rosenzweig. 1998. The Latitudinal Gradient
of Species Diversity among North American Grasshoppers (Acridi-
dae) Within a Single Habitat: A Test of the Spatial Heterogeneity
Hypothesis. Journal of Biogeography, 25:553-560.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Jokiel, P. L. 1984. Long-Distance Dispersal of Reef Corals by Rafting.
Coral Reefs, 3:113-116.

Mittelbach, G. G., D. W. Schemske, H. V. Cornell, A. P. Allen, J. M.
Brown, M. B. Bush, S. P. Harrison, A. H. Hurlbert, N. Knowlton,
H. A. Lessios, C. M. McCain, A. R. McCune, L. A. McDade, M. A.
McPeek, T. J. Near, T. D. Price, R. E. Ricklefs, K. Roy, D. FE. Sax,
D. Schluter, J. M. Sobel, and M. Turelli. 2007. Evolution and the
Latitudinal Diversity Gradient: Speciation, Extinction and Biogeog-
raphy. Ecology Letters, 10:315-331.

Moyle, P. B., and T. Light. 1996. Fish Invasions in California: Do Abiotic
Factors Determine Success. Ecology, 77:1666-1670.

Osman, R. W., and T. A. Dean. 1987. Intra- and Interregional Com-
parisons of Numbers of Species on Marine Hard Substrate Islands.
Journal of Biogeography, 14:53-68.

Osman, R. W., and R. B. Whitlatch. 1995. Predation on Early Ontoge-
netic Life-Stages and Its Effect on Recruitment into a Marine Com-
munity. Marine Ecology Progress Series, 117:111-126.

. 1996. Processes Affecting Newly-Settled Juveniles and the Con-

sequences to Subsequent Community Development. Invertebrate

Reproduction and Development, 30:217-225.

. 1998. Local Control of Recruitment in an Epifaunal Community

and the Consequences to Colonization Processes. Hydrobiologia,

375/376:113-123.

. 2004. The Control of the Development of a Marine Benthic

Community by Predation on Recruits. Journal of Experimental

Marine Biology and Ecology, 311:117-145.

. 2007. Variation in the Ability of Didemnum sp. to Invade Estab-
lished Communities. Journal of Experimental Marine Biology and
Ecology, 342:40-53.

Rejmanek, M., and D. D. Richardson. 1996. What Attributes Make
Some Plant Species More Invasive? Ecology, 77:1655-1661.

Ruiz, G. M., J. T. Carlton, E.D. Grosholz, and A.H. Hines. 1997. Global
Invasions of Marine and Estuarine Habitats by Non-indigenous
Species: Mechanisms, Extent, and Consequences. American Zoolo-
gist, 37:621-632.

Ruiz, G. M., P. W. Fofonoff, J. T. Carlton, M. J. Wonham, and A. H.
Hines. 2000. Invasion of Coastal Marine Communities in North
America: Apparent Patterns, Processes, and Biases. Annual Review
of Ecology and Systematics, 31:481-531.

Sax, D. F. 2001. Latitudinal Gradients and Geographic Ranges of Exotic
Species: Implications for Biogeography. Journal of Biogeography,
28:139-150.

Scheltema, R. S. 1986. On Dispersal and Planktonic Larvae of Marine
Benthic Invertebrates: An Eclectic Overview and Summary of Prob-
lems. Bulletin of Marine Science, 39:290-322.

Schemske, D. W. 2002. “Ecological and Evolutionary Perspectives on
the Origins of Tropical Diversity.” In Foundations of Tropical For-
est Biology: Classic Papers with Commentaries, ed. R. L. Chazdon
and T. C. Whitmore, pp. 163-173. Chicago: University of Chicago
Press.

Stachowicz, J. J., H. Fried, R. W. Osman, and R. B. Whitlatch. 2002a.
Reconciling Pattern and Process in Marine Bioinvasions: How Im-
portant Is Diversity in Determining Community Invisibility? Ecol-
ogy, 83:2575-2590.

Stachowicz, J. J., J. R. Terwin, R. B. Whitlatch, and R. W. Osman.
2002b. Linking Climate Change and Biological Invasions: Ocean
Warming Facilitates Nonindigenous Species Invasions. Proceed-
ings of the National Academy of Sciences of the United States of
America, 99:15497-15500.

Williamson, M., and A. Fitter. 1996. The varying success of invaders.
Ecology, 77:1661-1666.

Ex Situ Culture of Caribbean and Pacific

Coral Larvae Comparing Various
Flow-Through Chambers

Mary Hagedorn, Virginia L. Carter, Lea
Hollingsworth, JoAnne C. Leong, Roland
Kanno, Eric H. Borneman, Dirk Petersen,
Michael Laterveer, Michael Brittsan,

and Mark Schick

Mary Hagedorn and Virginia L. Carter, Depart-
ment of Reproductive Sciences, National Zoo-
logical Park, Smithsonian Institution, Washing-
ton, D.C. 20008, USA, and Hawaii’ Institute
of Marine Biology, University of Hawaii, P.O.
Box 1346, Kaneohe, Hawaii 96744, USA. Lea
Hollingsworth, JoAnne C. Leong, and Roland
Kanno, Hawaii’ Institute of Marine Biology,
University of Hawaii, P.O. Box 1346, Kaneohe,
Hawaii 96744, USA. Eric H. Borneman, Univer-
sity of Houston, Department of Biology and Bio-
chemistry, Houston, Texas 77204-5001, USA.
Dirk Petersen and Michael Laterveer, Rotterdam
Zoo, P.O. Box 532, 3000 AM Rotterdam, The
Netherlands. Michael Brittsan, Columbus Zoo
and Aquarium, 990 Riverside Drive, Powell,
Ohio 43065, USA. Mark Schick, John G. Shedd
Aquarium, 1200 South Lake Shore Drive, Chi-
cago, Illinois 60605, USA. Corresponding author:
M. Hagedorn (hagedornm@si.edu). Manuscript
received 21 July 2008; accepted 20 April 2009.

ABSTRACT. Coral reefs are some of the oldest and most diverse ecosystems on our
planet, yet throughout their range coral reefs are declining precipitously, mainly as the
consequence of human activities. In situ conservation practices, such as habitat pres-
ervation, are an important way to protect coral reefs. However, reefs now face global
threats in addition to local impacts. It is therefore critical that ex situ conservation ac-
tivities are incorporated into conservation practices for coral reefs. Many coral species
reproduce sexually during a limited yearly breeding season. If the resulting larvae are
cultured, their husbandry can be very time consuming: time that is often taken away
from larval research. Three different types of flow-through larval rearing systems were
designed and tested during breeding seasons of the elkhorn coral Acropora palmata, the
mushroom coral Fungia scutaria, and the cauliflower coral Pocillopora meandrina. The
flow-through systems were tested against static bowl rearing, and no difference was
observed in the survival of the larvae in two of the species: P = 0.12 for A. palmata
and P = 0.99 for F. scutaria. These results suggested that these chambers may result in
significant savings of limited research time during a coral spawning event. However,
P. meandrina larval survival was better in bowls than in the flow-through chamber (P =
0.03). Rearing the maximum number of larvae possible with minimal maintenance will
enhance opportunities for larval research, settlement, and growth. This is especially im-
portant for species that are now threatened, for which time and information are critical
during the breeding season.

INTRODUCTION

Coral reefs are some of the oldest and most diverse ecosystems on our
planet. They are essential nurseries and feeding grounds for fish and inverte-
brates, act as natural storm barriers for coastlines, and are a potential source
for novel pharmaceuticals (Colin, 1998). Throughout their range, coral reefs
are declining precipitously, mainly because of human activities. These negative
influences induce stress and can increase diseases in corals. Even in the most

260 °

remote marine bioreserves, such as the northwestern
Hawaiian Islands (Maragos et al., 2004), human activi-
ties are damaging fragile coral ecosystems (Bellwood et al.,
2004). Additionally, other environmental pressures, such
as El Nifio-Southern Oscillation events, result in bleach-
ing and coral mortality (Glynn and D’Croz, 1990; Glynn,
1996). As greenhouse gases increase, atmospheric and sea-
surface temperatures and ocean acidification are also ex-
pected to increase (Kleypas et al., 1999; Hoegh-Guldberg
et al., 2007). When these effects are coupled with human-
induced stresses, reefs will remain in crisis, their existence
worldwide increasingly threatened (Hoegh-Guldberg,
1999; Hughes et al., 2003).

Scientists speculate that unless committed efforts are
made to remedy this situation functional coral ecosystems
may disappear in less than 50 years (World Wildlife Re-
port, 2004; Hoegh-Guldberg et al., 2007). Although all
the oceans in the world have corals, reef-building corals
in the Caribbean are showing the greatest signs of disease-
related mortality, and these corals may have far less than
50 years left to survive (Hoegh-Guldberg et al., 2007). The
massive elkhorn coral, Acropora palmata, has historically
been the most ecologically important reef-building coral
in the Caribbean, but its populations have declined 90%
to 99% since the mid-1980s, primarily because of disease
(Aronson and Precht, 2001). Because of this decline and
its critical role for Caribbean reefs, A. palmata has been
one of the first two corals listed as “threatened” under
the Endangered Species Act (Acropora Biological Review
Team, 2005). As stony corals continue to die, they are be-
ing replaced with sponges, gorgonians, and algae (Hughes,
1994; McClanahan and Muthiga, 1998), altering the com-
position of Caribbean ecosystems.

In situ conservation practices, such as establishment
of marine protected areas, are an important way to protect
coral reefs. However, reefs now face global rather than just
local threats. Therefore it is critical that ex situ conserva-
tion techniques are incorporated into conservation actions
for coral reefs. Ex situ conservation techniques, defined as
protecting organisms outside their native habitat, such as
rearing sexually produced larvae in seminatural enclosures
for future restoration purposes, hold strong promise for
improvements in preserving species and genetic diversity
within ecosystems. This stage is particularly needed to
help diversify some of the declining endangered popula-
tions in Florida where many of the stands of A. palmata
are genetically identical (Baums et al., 2005).

To address the ex situ conservation needs for coral
reefs, SECORE (www.secore.org) was initiated by the
Rotterdam Zoo in 2001 with the primary goals of study-

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

ing sexual coral reproduction, specifically developing ex
situ breeding techniques, disseminating techniques among
aquarium and research communities through workshops
and publications, developing a cooperative international
network of public aquariums and research institutions,
and establishing breeding programs to help sustain ex situ
and field populations. In 2006 and 2007 SECORE mem-
bers representing several national and international insti-
tutions held workshops in Puerto Rico with goals to suc-
cessfully rear elkhorn coral from spawn produced during
the annual mass spawning at Rincon and Bajo Gallardo
sites. Gametes were collected and fertilized, producing
close to a million larvae, of which hundreds of thousands
were raised in the field laboratory and more than 400,000
were brought into captivity, resulting in approximately
2,300 juvenile larval recruits now living in public aquaria
around the world (Petersen et al., 2007). These larvae
were the first juveniles of this species ever reared in captiv-
ity, constituting a major step that will help with the con-
servation of their genome and restoration of this species
in the wild.

Although ex situ conservation practices have yet to be
applied to coral populations in conjunction with restora-
tion, extensive work has been conducted in the zoological
community on maintaining gene diversity in populations
with ex situ techniques (Ballou, 1992; Harnal et al., 2002;
Pukazhenthi et al., 2006). In particular, the black-footed
ferret was rescued from the brink of extinction, with only
18 individuals remaining in the population, using ex situ
conservation practices in parallel with restoration prac-
tices (Howard et al., 2003). Enhancing reproductive suc-
cess of endangered coral through ex situ practices may
be key to their future restoration and preservation (Rich-
mond and Hunter, 1990). There are a number of ex situ
techniques that have enhanced larval survival and settle-
ment. Heyward et al. (2002) used a seminatural enhance-
ment procedure for maintaining acroporid corals in open
floating pools in the ocean. Water was pumped into the
pools throughout the larval growth period, and then the
contents were pumped into an enclosed area on the sea
bottom with conditioned ceramic tiles. Heyward et al.
started with ~10.5 X 10° larvae/pool and after 144 h post-
fertilization had ~7.5 X 10° larvae/pool (~0.7% survival),
resulting in ~1,500 settled recruits in the best treatments
versus 0 on the control tiles. Although this settlement rate
was relatively low, it was far greater than the natural settle-
ment rate and indicated a robust enhancement of recruits
for this area.

Most current coral larval husbandry practices are
low-cost efforts, such as bowls or aquaria filled with fil-

tered seawater, and these methods are very successful at
rearing larvae (Babcock and Heyward, 1986; Schwartz et
al., 1999; Petersen et al., 2007). The problem is that these
time-consuming and labor-intensive husbandry practices
compete with the limited time available for research dur-
ing a coral breeding season, especially if the coral spe-
cies is limited to a single annual breeding, as is Acropora
palmata. For coral in need of replenishment, rearing the
maximum number of larvae possible with the least time in-
vested in husbandry would enhance opportunities for lar-
val growth and settlement (Richmond and Hunter, 1990;
Petersen and Tollrian, 2001; Borneman, 2006). The goal
of this paper was to design and test simple flow-through
systems in the field that would minimize husbandry and
yet successfully rear large numbers of coral larvae with-
out compromising survival.

Three species of coral larvae were tested in three differ-
ent types of rearing chambers. These larvae were selected
because they represented a good cross section of coral lar-
val types with different buoyancies, swimming behaviors,
and rates of development that might benefit from these
chambers. Acropora palmata are large lipid-filled float-
ing larvae (Figure 1a) that develop slow swimming ability
in the water column after 48 h. Fungia scutaria are small
negatively buoyant larvae containing modest lipid stores.
These larvae develop rapid swimming behavior in the wa-
ter column within 12 to 24 h (Figure 1b). Pocillopora me-
andrina are negatively buoyant larvae with modest lipid
stores (Figure 1c); these larvae develop slow swimming be-
havior along the bottom after 24 h. In designing and con-
structing these low-tech chambers, we made an effort to
use materials for their components that are affordable and
available in most hardware stores throughout the world.

NUMBER 38 ¢ 261

MATERIALS AND METHODS
LARVAL COLLECTION AND REARING

Acropora palmata eggs and sperm were collected dur-
ing the annual spawn from Tres Palmas Reserve (Rincon,
Puerto Rico) and the offshore submerged bank Bajo Gal-
lardo (Boqueron, Puerto Rico) in August 2007. Egg/sperm
bundles were collected in the water over the spawning coral
with 1 L plastic Nalgene bottles attached to fine mesh nets.
The egg/sperm bundles were brought to shore in the plastic
bottles, separated by gentle agitation, and then combined
with the eggs and sperm from at least three to four individ-
uals to yield a sperm concentration of approximately 10°
cells/mL (final concentration in water). The eggs and sperm
were gently agitated for 2 h, cleaned with 1 wm-filtered
seawater, assessed for fertilization rates, and released into
rearing chambers for subsequent development.

Fungia scutaria eggs and sperm were collected from
captive animals held in flowing seawater tanks from June
through October 2006 at Coconut Island, Hawaii. Ani-
mals were prepared for spawning following the methods
of Krupp (1983). Briefly, as a female spawned, these eggs
were gently moved into a plastic bowl and fertilized with
~150 mL sperm (10° cells/mL, final concentration in wa-
ter) from four or five males. The embryos, resulting from
several male and female gametes, were kept in a single plas-
tic bowl (8 L) and left overnight to develop. In the morning
the developing larvae were cleaned with four changes of
0.5 m-filtered seawater and then released into their rear-
ing chambers for subsequent development.

Egg/sperm bundles were collected from Pocillopora
meandrina fragments in April and May 2008 from Co-
conut Island, Hawaii. The eggs and sperm were separated

FIGURE 1. Three species of coral were reared in this study. a, Acropora palmata, elkhorn coral; inset: larvae at 24 h postfertilization. b, Fungia
scutaria, mushroom coral; inset: larvae at 96 h postfertilization. c, Pocillopora meandrina, cauliflower coral; inset: larvae at 96 h postfertiliza-
tion. All scales = 50 pm.

262 e

by gentle agitation, and then combined with the eggs and
sperm from at least three or four individuals to yield a
sperm concentration of approximately 10° cells/mL. The
eggs and sperm were gently agitated for 0.5 h, cleaned with
0.5 pm-filtered seawater, and left overnight to develop.
The next morning the developing larvae were cleaned with
0.5 ym-filtered seawater and released into their rearing
chambers for subsequent development.

Digital images of the larvae from all three species were
captured with an Olympus BX41 microscope with an at-
tached digital camera Sony DFWV300, and the major and
minor axes were measured with NIH Image software.

CONSTRUCTION OF REARING TANKS
AND MEASUREMENT OF DENSITIES

Larval corals were reared in three different designs
of flow-through chambers (Figures 2-4), as well as static
bowls that required daily cleaning and water changes. The
names of these chambers were chosen to describe the ma-
jor water movement they provided to the larvae. All de-
velopmental times reported throughout the paper are in
hours postfertilization.

Up-Flowing Tanks

These tanks were made from 20 L heavy-walled plastic
pans (U.S. Plastics Corp., Lima, Ohio) modified by cover-
ing the handles in a buoyant foam and removing four pan-
els from the bottom and replacing them with nylon screen-
ing (240 ym mesh). A central cross-shaped area was left
intact to create an inlet for upward-directed water flow;
then additional shear flow was added with four additional
adjustable water inlets around the edges approximately
16 cm above the bottom, yielding a final volume in the
chambers of ~23 L (see Figure 2). All flow was regulated
by valves to optimize the slow tumbling movement of the
larvae in the chamber. The floating chambers were im-
mersed in large 2,400 L pools to stabilize their temperature
(28°-31°C) and mimic natural temperature cycles through-
out a 24 h period. To maintain water quality close to that
which the larvae would experience in open water, the cham-
bers were attached to a filtered (1 zm) flow-through system
with seawater pumped from the reserve, so that water was
completely exchanged in the chambers several times each
day. Flow rates through the chambers were maintained
at approximately 2 L/min, and the bath of fresh seawater
surrounding the chambers was turned over about one to
two times per hour. Salinity, temperature, and pH closely
mimicked natural conditions without additional effort.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

)
=
a
S]
S
a
z
5
a
£
a
mal
ft) 24 48 72 96
Developmental Time (h)

FIGURE 2. Larval rearing of Acropora palmata larvae. a, Up-flowing
tank used to rear the A. palmata larvae. b, There was no signifi-
cant difference in survival between the up-flowing tanks and static
bins (ANCOVA, P = 0.12). Points show mean counts for each trial.
Larvae in the upward-flow tanks were developing rapidly (~28°C)
while those in the static bins were maintained at a slightly lower tem-
perature (~25°C). Larvae in the flow chambers were removed from
the experiment one day earlier than those in the bins to ship them to
an aquarium for settlement and rearing.

Approximately 1,000 to 1,500 larvae/L were fertilized in
50 mL conical plastic tubes, then placed into either the up-
flowing tanks (7 = 2) or static bins. Counts were taken im-
mediately, and then daily for all groups. Bin density began
at ~900 larvae/L, and the two flow chambers contained
either 1,100 or 1,500 larvae/L. The fertilization rate for A.
palmata spawn used for these tests was ~90%.

The static treatments used for comparison with the
up-flowing tanks were plastic rectangular bins (length [L]
x width [W] = 51 cm X 36 cm) with water depth of 12
cm, yielding a volume of 22 L. These tanks were main-
tained in an air-conditioned room; the water was main-

s

Bowls

100000

Larval Survival (Density/L)

50000

0 48 96 144

Developmental Time (Days)

NUMBER 38 °¢ 263

tained at 25° to 26°C and changed twice daily. To keep
the floating A. palmata larvae from clumping and forming
an anoxic layer, the water and floating larvae were stirred
every hour with a bubble-wand (2-mm-diameter rigid air
line attached to a small air pump) throughout the rearing
period. The previous year, a stocking density of approxi-
mately 1,000 A. palmata larvae/L was used successfully in
each bin and we used this same level for these tests. Larvae
from the same spawn and bulk fertilization as were used
in the chambers were placed in the static bins (7 = 2) 2h
after fertilization.

To determine larval survival, the chamber and static
containers were stirred to suspend the larvae evenly in the
water column, and five 15 mL samples were taken and
the number of larvae counted each day. The number of
larvae/mL was multiplied by 1,000 to determine the den-
sity per liter (density/L). The larvae in both systems were
only allowed to develop for two to three days, and then
they were packaged for shipment. Approximately 4,000
larvae were placed into a 2 L Nalgene bottle with filtered
seawater (FSW); the bottles were filled to the top with
FSW and capped leaving no bubbles, taped for security,
placed horizontally in a cooler (8-12 bottles were placed
in a single cooler), and sent by express mail to aquaria
throughout the USA.

Spiral-Flowing Tanks

Small conical fiberglass tanks (~75 L) were fitted with
a central standpipe covered with 40 .m nylon mesh to rear
Fungia scutaria larvae (see Figure 3). To maintain the wa-
ter quality close to that which the larvae would experience
under natural conditions, the conical tanks were attached
to a filtered (0.5 um) flow-through system with seawater

FIGURE 3. Larval rearing of Fungia scutaria. a, Drawing of assem-
bled spiral-flowing tank and its flow, mixing, and position of dye ex-
periments. Curved lines with arrowheads indicate direction of water
flow spiraling upward from one inlet of a Loc-Line (both inlets had
flow, but for simplicity flow from only one is drawn). Double arrows
indicate water freely flows into and back out of the mesh areas on
the standpipe. Asterisks (*) indicate locations in the water column
where dye was injected for dye tests. b, Survival rate of F. scutaria
larvae maintained in the spiral-flowing tank (upper graph, “Flow”)
and the static bowls (lower graph, “Bowls”) between 24 and 144 h
postfertilization. Each point shows mean and standard error. There
was no difference in survival of larvae from flow chambers and static

bowls (ANCOVA, P = 0.99).

264 e

—O— Bowl
—- Flow

<)
=
F
e 15000.0
a
<
g
-}
NH
a
g
a
ral

5000.0
0.0

o

24 48 72 96

Developmental Time (h)

SSS a ee a ES ED)
FIGURE 4. Larval rearing of Pocillopora meandrina. a, Down-
flowing tank used to rear P. meandrina. b, Curves for larval survival
in the static (“Bowl”) and down-flowing (“Flow”) tanks did not
have the same slope (ANCOVA, P = 0.03); bowl rearing for this
species produced substantially better survival than the down-flowing
tank. Points show mean counts for each trial and standard error of
the mean.

pumped from the reef. Flowing filtered seawater entered
the top of the central tube and moved through nozzles
at the tank base to produce gentle circular movement
throughout the water column, and the wastewater exited
the tank through the mesh-covered outflow. The flow rate
was 150 to 300 mL/min, producing a complete turnover
in the tanks approximately every 4 to 8 h. To test whether
the conical tank could support the growth and develop-
ment of FE scutaria, approximately 10,000 larvae/L were
stocked in the conical tanks (7 = 4) tanks. To reduce mor-
tality of the early fragile stages (0-24 h postfertilization)
from excessive motion in the flow-through chambers, the
spiral-flow tanks were tested with 24 h postfertilization

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

larvae. Therefore fertilization rate was not an issue, be-
cause all the larvae used for these tests were intact and
developing. A comparison of the static bowl method (with
daily cleaning) and the flow-through method was then per-
formed to determine survival rate over time (up to 144 h
postfertilization).

The static treatments used for comparison to the coni-
cal tanks were 3 L plastic bowls that had been the stan-
dard for successfully rearing FE scutaria larvae for many
years (Schwarz et al., 1999). At 24 h postfertilization, F
scutaria larvae from 10 bowls were combined and evenly
distributed into 2 larger bowls. Counts (see below) were
taken for each bowl, and a standard rearing density was
redistributed into 8 separate smaller bowls at a larval den-
sity of 100,000 larvae/L filtered seawater.

Fungia scutaria larvae were counted each day by gen-
tly stirring to homogeneously suspend them in the water
column. Ten 1 mL samples were taken midwater from each
conical flow-through tank and placed into a Sedgewick-
Rafter counting chamber; the larvae were counted with
a dissecting microscope and their numbers averaged. In
addition, 10 samples (20 wL each) were taken midwater
from each bowl, the number of larvae counted under a
dissecting microscope, and their numbers averaged. These
smaller 20 wL samples were assessed from the E scutaria
because of their high densities in the chambers.

Down-Flowing Tank

In April 2008 we attempted to rear Pocillopora me-
andrina larvae in the spiral-flowing tanks, but because of
the negatively buoyant nature of the larvae, this method
resulted in 100% mortality. Therefore, we developed a
down-flowing tank for rearing P. meandrina larvae in May
2008 (see Figure 4). The tank was constructed of a 1.65 L
glass bowl with a plastic lid. The center of the lid was re-
moved and replaced with 40 wm mesh, leaving a 2 cm ring
around the outside in which a hole was made to insert a 2
mm plastic rigid air line attached with air-line tubing to a
manifold for controlling water flow. To maintain the wa-
ter quality close to that which the larvae would experience
in the open water, the down-flowing tanks were attached
to a filtered (1 pm) flow-through system with seawater
pumped from the reef. Flow to the tanks was maintained
at 120 mL/min.

The larvae were fertilized in 50 mL tubes, rinsed with
sterile filtered seawater, and placed into two bowls at
28°C to develop overnight at a density of approximately
80,000/L. After 24 h postfertilization, larvae were counted
and then cleaned using a 40 wm mesh and 0.5 pm-filtered

seawater. One group was placed into the downward-flow
chamber at a flow rate of 120 mL/min with a 40 um mesh
top to allow the water to flow out. The other bowl re-
mained static. P. meandrina larvae were cleaned (static
bowl only) and counted daily from the flow tanks and
static bowl (maintained as described for F. scutaria above)
for comparison.

STATISTICS

To determine the differences between survival in
flow chamber versus static treatment, the data from all
experiments was normalized and the y-values linearly
transformed; analysis of covariance (ANCOVA) was then
performed to determine whether the slopes were signifi-
cantly different, using GraphPad Prism 5 software for the
Macintosh GraphPad Software (San Diego, CA).

RESULTS

Rearing chambers were designed for three different
coral species exhibiting different larval swimming behav-
iors, buoyancy, and sizes.

Up-Flowing Tank

Acropora palmata was the largest of the larvae studied
(~700 x 500 wm depending on the developmental stage)
and had the slowest rate of development (see Figure 1a). A.
palmata larvae float at or near the surface of the water ap-
proximately 48 to 60 h postfertilization (depending on the
temperature) until they began swimming. Even once they
had begun swimming, they swam at or near the surface
and were considered positively buoyant for most of their
larval development before metamorphosis and settlement
(~144 h). Clearly, all the larvae must become negatively
buoyant before settlement; therefore, these categories only
apply to the early larval periods (up to ~120 h, depending
on the species).

During the first 24 h of development, the larvae devel-
oped asymmetrical, small protrusions of cells that could
have easily be damaged in the up-flowing tank, but the
chamber produced normally developed A. palmata larvae
(as compared to the bins) because it simulated the gentle
tumbling that the larvae would experience in the natural
water (see Figure 2). Larval survival in the up-flowing
tanks was similar to that in the static bins (P = 0.12) (Figure
2a). However, the up-flowing tank did not produce viable
larvae for E. scutaria and P. meandrina. This tank produced

NUMBER 38 °¢ 265

100% mortality in FE scutaria larvae within 24 h, even if
the larvae were slightly more developed when placed in
the chambers.

Spiral-Flowing Tank

Dye injection studies were performed on the spiral-
flowing tank, using food coloring to examine the mixing
properties of the vessel with an inlet flow of 150 mL/min.
Figure 3 illustrates the mixing pattern in the flow-through
vessel. Flow into and out of the system was equal, but the
open area of the slits in the 10 cm central polyvinyl! chlo-
ride (PVC) tube dictated the velocity through the screens.
The nozzles were angled slightly downward to promote
turbulence at the bottom to keep the larvae well mixed.
The 180° positions of the nozzles provided rotation within
the water column and encouraged mixing. Dye studies
with separate injections were made at positions noted by
asterisks (*) in Figure 3. At a flow of 150 mL/min, full
vertical mixing occurred within minutes.

Developing FE scutaria larvae were fairly small (~200
x 100 pm) and fragile during the first 12 h of develop-
ment and were just negatively buoyant during their early
embryonic period (0-12 h postfertilization) (Figure 1b).
However, once they began swimming, they were evenly
distributed in the water column, and we considered this
species to be neutrally buoyant. There was no difference
in the survival between the spiral-flowing tank and bowls
(P = 0.99). Both rearing systems produced similar survival
rates in which the densities remained relatively steady
through 96 h postfertilization and then dropped off at 120
to 144 h postfertilization (see Figure 3). This decrease in
densities may reflect the complete absorption of stored fats
(M. Hagedorn, unpublished data), as these larvae did not
have zooxanthellae. P. meandrina larvae were tested in the
spiral-flow tank, but 100% mortality was observed after
48 h postfertilization.

Down-Flowing Tank

Pocillopora meandrina larvae were the smallest of the
larvae tested (~120 X 40 ym); they began slow swimming
at 24 h but were negatively buoyant for the remainder of
their larval development, remaining on or near the bot-
tom (Figure 1c). Similar to E scutaria, P. meandrina larvae
were relatively susceptible to damage within the first 24
h, so they were reared in 3 L bowls for the first 24 h. The
down-flowing tank was used for rearing P. meandrina;
however, the static bowls appeared superior to the down-
flowing tank for this species (P = 0.03) (see Figure 4).

266 e

DISCUSSION

For many years, large numbers of coral larvae have
been reared successfully using simple husbandry methods
such as static bowls and tanks. We have demonstrated
that species of buoyant and neutrally buoyant coral lar-
vae have similar survival in either static or flow-through
chambers (see Figures 2, 3). These devices have proven to
be very useful in improving culture conditions to reduce
husbandry labor because neither embryos nor fresh water
needed to be constantly transferred.

Modified examples of the up-flowing tank have al-
ready been used successfully by coral restoration bi-
ologists in the field (Margaret Miller, NOAA Southwest
Fisheries Center, personal communication). Montastraea
faveolata and Diploria strigosa were reared successfully
in the up-flowing tanks and shipped to Columbus Zoo
and Aquarium for settlement with 3-month survival as
high as 65% and 45% for each species, respectively.
Thus, the up-flowing tank has proven to be both practi-
cal, in that it can be adapted to the researcher’s needs, and
valuable, because it reduced husbandry time and facili-
tated restoration science under field conditions.

In weighing the benefits of each rearing system, one
of the biggest factors to consider is time. For species that
have only a single breeding season consisting of a few days,
time available for conservation and restoration research
is precious, and any time savings is a benefit. Moreover,
the time remaining for some species that are threatened
has become critical, and restoration practices need to be
improved. Acropora palmata (elkhorn coral) and Acro-
pora cervicornis (staghorn coral) were the first corals to
be listed as threatened species under the U.S. Endangered
Species Act. These major reef-building species once formed
dense thickets and stands in the Caribbean. Today, these
two species are currently at 1% to 20% of their historical
levels throughout their range (Bruckner, 2003). Here we
describe only one aspect of an ex situ conservation pro-
cess, namely improved rearing associated with yielding
better time management.

However, both the static and flow-through methods
described here have their strengths and weaknesses. The
static method was inexpensive to set up in terms of equip-
ment and space. For example, 60 bowls can be maintained
in two double-tiered flowing water tables taking up only
about 2.5 m?; however, this method was very expensive in
terms of labor needed for cleaning (~5,000 h year~!). The
flow-through system was more costly to set up because
it required a filtered flow-through water system and spe-
cially constructed rearing chambers. The amount of salary

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

needed to pay one person for a season cleaning larvae,
however, far exceeds the cost of the filtered seawater sys-
tem and rearing chambers. The flow-through chambers re-
quired more space than the bowls, but each flow-through
vessel could maintain almost three times the density (in
~0.25 m*) than was ordinarily maintained in a static bowl
and with little maintenance time required.

One of the major issues facing biologists in rearing
coral larvae is how to keep them cool (28°-30°C) under
field conditions. During daylight hours, static bins left
outside without any cooling mechanism can easily reach
31° to 33°C, which is lethal for most species. The rearing
data in Figure 3 reflect some of these issues. These data
were not exactly comparable, because they did not have
the same developmental temperatures. Had the static bins
been maintained at 28° to 30°C (as were the flow cham-
bers), possibly their survival would have been far worse,
because their water quality would decay so rapidly. Be-
cause A. palmata is an endangered species, our goal was to
produce the most larvae for captive maintenance in public
zoos and aquaria (Petersen et al., 2007), this required hav-
ing static “backup” bins maintained at a slightly cooler
temperature to provide the larvae sufficient development
time in transit to reach their respective sites before settle-
ment. However, without an air-conditioned room to cool
the bins, this would not have been possible, making this
impractical under some field conditions.

Within the first 24 h of development, many coral lar-
vae are susceptible to fragmentation by mechanical disrup-
tion. However, the water movement within the up-flowing
tank and potential contact with the walls did not cause
substantial fragmentation of A. palmata during early de-
velopment, even when the A. palmata larvae were placed
in the chambers within the first few hours after fertiliza-
tion. In contrast, P. meandrina was far more delicate, did
not develop strong swimming behaviors, and could not
withstand the water movements in the flow chambers.
E scutaria larvae are negatively/neutrally buoyant larvae
that develop strong swimming behaviors within the first
12 to 24 h, and the spiral-flow system shown in Figure 3
functioned well for them, because the water flow is up-
ward and any disintegrating unfertilized oocytes and larvae
passed through the mesh, allowing for the maintenance of
excellent water quality in the rearing chambers. However,
no one type of rearing chamber can be applied universally
across species. Instead, the type of water flow within the
chamber must be matched with the buoyancy and early
swimming behavior of the larvae. Regardless, these readily
built and easily maintained flow-through chambers may be
a substantial aid to coral conservation and restoration.

ACKNOWLEDGMENTS

This work was supported by NOAA (grant #
NA07NMF4630109), the Columbus Zoo and Aquarium
Conservation Fund, the Rotterdam Zoo, the Green Foun-
dation, The Clyde and Connie Woodburn Foundation, the
Smithsonian Institution, Louisiana State University Agri-
cultural Center, the University of California, the Omaha
Zoo, Lou Nesslar, and the John G. Shedd Aquarium. We
thank the anonymous reviewers whose comments have
greatly improved this paper.

LITERATURE CITED

Acropora Biological Review Team. 2005. Atlantic Acropora Status Re-
view. Report to National Marine Fisheries Service, Southeast Re-
gional Office. March 3, 2005. St. Petersburg, Fla.: NOAA Fisheries
Service Southeast Region.

Aronson, R. B., and W. F. Precht. 2001. White-Band Disease and the
Changing Face of Caribbean Coral Reefs. Hydrobiologia, 460:
25-38.

Babcock, R. C., and A. J. Heyward. 1986. Larval Development of Certain
Gamete-Spawning Scleractinian Corals. Coral Reefs, 5:111-116.

Ballou J. D. 1992. Potential Contribution of Cryopreserved Germplasm
to the Preservation of Genetic Diversity and Conservation of En-
dangered Species in Captivity. Cryobiology, 29:19-25.

Baums, I. B., M. W. Miller, and M. E. Hellberg. 2005. Regionally Iso-
lated Populations of an Imperiled Caribbean Coral, Acropora pal-
mata. Molecular Ecology, 14(5):1377-1390.

Bellwood, D. R., T. P. Hughes, C. Folke, and M. Nystrom. 2004. Con-
fronting the Coral Reef Crisis. Nature (London), 429:827-833.

Borneman, E. H. 2006. Reproduction in Aquarium Corals. Proceedings
of the 10th International Coral Reef Symposium, 1:50-60.

Bruckner, A. W. 2003. Proceedings of the Caribbean Acropora Work-
shop: Potential Application of the U.S. Endangered Species Act as
a Conservation Strategy. NOAA Technical Memorandum NMFS-
OPR-24. Silver Spring, Md.: National Oceanographic and Atmo-
spheric Administration.

Colin, P. L. 1998. “Marine Pharmaceuticals from the Reef: A View
from the Field.” In Coral Reefs: Challenges and Opportunities for
Sustainable Management, ed. M. E. Hatziolos, A. J. Hooten, and
M. Fodor, pp. 76-79. Washington, D.C.: The World Bank.

Glynn, P. W. 1996. Coral Reef Bleaching: Facts, Hypotheses and Implica-
tions. Global Change Biology, 2(6):495-509.

Glynn, P. W., and L. D’Croz. 1990. Experimental Evidence for High
Temperature Stress as the Cause of El Nifio-Coincident Coral Mor-
tality. Coral Reefs, 8:181-191.

Harnal, V. K., D. E. Wildt, D. M. Bird, S. L. Monfort, and J. D. Ballou.
2002. Computer Simulations to Determine the Efficacy of Differ-
ent Genome Resource Banking Strategies for Maintaining Genetic
Diversity. Cryobiology, 44:122-131.

NUMBER 38 ¢ 267

Heyward, A. J., L. D. Smith, M. Rees, and S. N. Field. 2002. Enhance-
ment of Coral Recruitment by In Situ Mass Culture of Coral Lar-
vae. Marine Ecology Progress Series, 230:113-118.

Hoegh-Guldberg, O. 1999. Climate Change, Coral Bleaching and the
Future of the World’s Coral Reefs. Marine and Freshwater Research,
50:839-866.

Hoegh-Guldberg, O., P. J. Mumby, A. J. Hooten, R. S. Steneck, P. Green-
field, E. Gomez, C. D. Harvell, P. E. Sale, A. J. Edwards, K. Cal-
deira, N. Knowlton, C. M. Eakin, R. Iglesias-Prieto, N. Muthiga,
R. H. Bradbury, A. Dubi, and M. E. Hatziolos. 2007. Coral Reefs
under Rapid Climate Change and Ocean Acidification. Science,
318:1737-1742.

Howard, J., P. E. Marinari, and D. E. Wildt. 2003. Black-Footed Fer-
ret: Model for Assisted Reproductive Technologies Contributing
to In Situ Conservation. Conservation Biology Series-Cambridge,
8:249-266.

Hughes, T. P. 1994. Catastrophes, Phase Shifts, and Large-Scale Degra-
dation of a Caribbean Coral Reef. Science, 265:1547-1551.

Hughes, T. P., A. H. Baird, D. R. Bellwood, M. Card, S. R. Connolly,
C. Folke, R. Grosberg, O. Hoegh-Guldberg, J. B. C. Jackson,
J. Kleypas, J. M. Lough, P. Marshall, M. Nystrom, S. R. Palumbi,
J. M. Pandolfi, B. Rosen, and J. Roughgarden. 2003. Climate
Change, Human Impacts, and the Resilience of Coral Reefs. Sci-
ence, 301:929-933.

Kleypas, J. A., R. W. Budemeier, D. Archer, J.-P. Gattuso, C. Lang-
don, and B. N. Opdyke. 1999. Geochemical Consequences of
Increased Atmospheric Carbon Dioxide on Coral Reefs. Science,
284:118-120.

Krupp, D. A. 1983. Sexual Reproduction and Early Development of
the Solitary Coral Fungia scutaria (Anthozoa: Scleractinia). Coral
Reefs, 2:159-164.

Maragos, J. E., D. C. Potts, G. Abey, D. Gulko, J. Kenyon, D. Siciliano,
and D. van Ravenswaay. 2004. 2000-2003 Rapid Ecological As-
sessment of Corals (Anthozoa) on Shallow Reefs of the Northwest-
ern Hawaiian Islands. Part 1: Species and Distribution. Pacific Sci-
ence, 58(2):211-230.

McClanahan, T. R., and N. A. Muthiga. 1998. An Ecological Shift
Among Patch Reefs of the Glovers Reef Atoll, Belize over 25 Years.
Environmental Conservation, 25:122-130.

Petersen, D., E. Borneman, M. Brittsan, M. Hagedorn, and M. Laterveer.
2007. Conserving the Threatened Elkhorn Coral, Acropora pal-
mata. AZA Connect, 1:22-25.

Petersen, D., and R. Tollrian. 2001. Methods to Enhance Sexual Re-
cruitment for Restoration of Damaged Reefs. Bulletin of Marine
Science, 69(2):989-1000.

Pukazhenthi B., P. Comizzoli, A. J. Travis, and D. E. Wildt. 2006. Ap-
plications of Emerging Technologies to the Study and Conservation
of Threatened and Endangered Species. Reproduction, Fertility and
Development, 18:77-90.

Richmond, R. D., and C. L. Hunter. 1990. Reproduction and Recruit-
ment of Corals: Comparisons Among the Caribbean, the Tropi-
cal Pacific and the Red Sea. Marine Ecology Progress Series, 60:
185-203.

Schwarz, J. A., D. A. Krupp, and V. M. Weis. 1999. Late Larval Devel-
opment and Onset of Symbiosis in the Scleractinian Coral Fungia
scutaria. Biological Bulletin, 196:70-79.

ae

oe yh ah ANE, aia
—s ’

2] . * i ae | pie sae OMS ceive

Worldwide Diving Discoveries
of Living Fossil Animals from the
Depths of Anchialine and Marine Caves

Thomas M. Iliffe and Louis S. Kornicker

Thomas M. Iliffe, Department of Marine Biology,
Texas AGM University at Galveston, Galveston,
Texas 77553-1675, USA. Louis S. Kornicker,
Department of Invertebrate Zoology, National
Museum of Natural History, Smithsonian Institu-
tion, Washington, D.C. 20560-0163, USA. Cor-
responding author: T. Iliffe (iliffet@tamug.edu).
Manuscript received 9 June 2008; accepted 20
April 2009.

ABSTRACT. Inland (anchialine) and offshore submarine caves in limestone and vol-
canic bedrock are extreme environments inhabited by endemic, cave-adapted (typically
eye- and pigment-reduced) fauna. Specialized cave diving technology is essential for in-
vestigating this habitat. A number of new higher taxa are represented herein, including
closely related species inhabiting caves on opposite sides of the Earth, thus suggesting an
ancient common ancestry. Because many of these species are known from only a single
cave, pollution or destruction of caves will result in their extinction.

INTRODUCTION
DEFINITION OF ANCHIALINE AND MARINE CAVES

Anchialine caves are partially or totally submerged caves situated within a
few kilometers inland from the coast in volcanic or karstic limestone terrain.
Tidal marine waters in these caves have a long residence time, of months to years.
Such caves are locally termed “cenotes” in the Yucatan Peninsula of Mexico,
“blue holes” in the Bahamas and Belize, and “grietas” in the Galapagos Islands.
The caves typically possess a highly stratified water column, with surface layers
of freshwater or brackish water, separated by a thermo—chemocline from under-
lying fully marine waters low in dissolved oxygen (Iliffe, 2000). Animals that
are restricted to the anchialine habitat and show pronounced morphological,
physiological, biochemical, and behavioral adaptations are termed stygofauna
or stygobites. In some areas such as Yucatan, freshwater and marine stygobites
inhabit their respective water masses within the same caves.

In contrast to anchialine caves, marine caves are located either directly on
the coastline (e.g., tidal springs) or are wholly submerged beneath the seafloor
(e.g., offshore blue holes) and contain marine waters that freely exchange with
the sea on each tidal cycle. The stygophilic fauna of marine caves can also be
found in suitable and similar habitats outside of caves (e.g., under rocks or in
crevices within the reef) and lack specialized adaptations for subterranean life.

Moderate to strong tidal currents are present in many marine caves. As a
result, encrusting and low-growing, filter-feeding animals such as sponges, hy-
droids, anemones, tube worms, and even some corals may completely cover all

270 e

hard surfaces. Other organisms are swept into caves by
tidal currents but can only survive there for short periods
of time and are termed accidentals. Some species of fishes,
lobster, and mysidaceans seek shelter within marine caves
but must venture out into open waters to feed and are clas-
sified as stygoxenes.

Some extensive marine caves extend far or deep
enough so that a more or less gradual transition to long
water residence times takes place and conversion to a
true anchialine habitat occurs. Similarly, a number of in-
land anchialine systems have submerged entrances in the
sea, with significant water exchange occurring in the en-
trance sections but with a transition to anchialine char-
acteristics and fauna taking place as distance from the
sea increases and the magnitude and impact of tidally
exchanging water decline.

Biological Significance

Anchialine caves contain a rich and diverse, endemic
stygobitic fauna (Sket, 1996; Iliffe, 2000, 2004) but, be-
cause of the specialized technological demands and poten-
tial dangers of cave diving, are relatively unstudied. These
habitats serve as refuges to “living fossil” organisms, for
instance, members of remiped crustaceans, and to animals
closely related to deep-sea species, such as the galatheid
crab Munidopsis polymorpha. Such stygobites typically
possess regressed features including loss of eyes and body
pigmentation. For reasons that remain unclear, the inver-
tebrate fauna is dominated by crustaceans and includes
the new class Remipedia, plus three new orders, nine new
families, more than 75 new genera, and 300 new species.
This extraordinary degree of novelty qualifies anchialine
habitats as uniquely important. Because anchialine spe-
cies commonly have a highly restricted distribution, often
being found only in a single cave system on one island,
pollution or destruction of the caves will result in their
extinction.

Stygobitic anchialine fauna often have highly disjunct
biogeographic distributions, inhabiting caves in isolated
locations on opposite sides of the Atlantic and Pacific
Oceans, as well as in the Mediterranean, and are consid-
ered Tethyan relicts. Various hypotheses have been pro-
posed to explain the origin of anchialine fauna. In general,
these theories invoke either vicariance (geological) or dis-
persal (biological) processes. Recently initiated molecular
genetic comparisons of cave populations from distant lo-
cations may help provide data for determining the age and
dispersal sequence of anchialine stygobites (Zakéek et al.,
2007; Hunter et al., 2008).

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Lifestyle Adaptations

The extreme environmental conditions in anchialine
caves, such as the absence of light, hypoxia, and limited
food reserves, present a unique set of challenges for the
organisms that reside there. The lack of light precludes
photosynthetic (primary) production of oxygen and food.
Without light, organisms receive no visual information for
orientation or communication and must function without
diurnal timing mechanisms.

Adaptations to anchialine and marine caves can be
morphological, behavioral, and physiological (Iliffe and
Bishop, 2007). As a result of both food scarcity and hy-
poxia, there is a high selective advantage for economy of
energy observed in many taxa, with possible adaptations
including enhanced chemo-mechano-receptors for im-
proved food finding capability, starvation resistance, and
reduction in energy demand via reduced metabolism.

METHODS
Diving Investigations

Because anchialine stygobites are commonly found only
at significant depths or distances from the water surface,
cave diving is an essential component of the collection and
study of anchialine fauna (Iliffe and Bowen, 2001). Cave
diving requires specialized training, equipment, and tech-
niques because a direct assent to the surface is not possible
and divers may be hundreds of meters from outside access.
In case of equipment failure or loss of air supply, cave divers
must have readily available backups. Special techniques for
cave diving may include the use of side-mounted, instead of
back-mounted, scuba tanks to allow divers to pass through
low bedding plane passages. Closed circuit rebreathers,
which recycle the diver’s exhaled gases, reduce the amount
of percolation, that is, of silt dislodged from cave ceiling or
walls by the exhaust bubbles produced in conventional open
circuit scuba, and lessen contamination of the cave waters,
which are low in dissolved oxygen (Figure 1). Rebreathers
allow for much longer dives and generally less decompres-
sion time. Deep dives, depths below 40 m, require the use of
special breathing gas mixtures that replace part or all of the
nitrogen with helium to reduce the effect of nitrogen narco-
sis. As many cave dives are for longer durations and/or to
deeper depths, they frequently involve long decompression.

Sampling and Fixation

The exceptionally clear waters of anchialine caves
facilitate visual observation and collection of stygobitic

FIGURE 1. A diver uses a Megalodon closed-circuit rebreather with
full face mask to collect a small shrimp, Typhlatya sp., from a cave

in Yucatan. Rebreathers recycle expired gas so that no bubbles are
produced.

species. Collectors generally lead the dive to have undis-
turbed water in front of them. As they slowly sweep their
dive lights back and forth in an arc, observing the water
column illuminated by the light beam, animals as small as
a few millimeters can be distinguished as white pinpoints,
sharply contrasting with the black background of the cave.
Specimens recognized in this manner can be collected ei-
ther individually in clear glass vials or plastic tubes or in
larger numbers using a type of suction device known as the
“Sket bottle” (Chevaldonné et al., 2008). Plankton nets,
of 93 wm mesh with a 30 cm mouth diameter and 1 m
length, can be used to collect smaller animals, such as co-
pepods, from the water column. When collecting animals
from the surface layer of sediments, divers can gently fan
up the sediments with a hand and then sweep the plankton
net through the disturbed water. This agitation should be
done with care so as not to obscure overall visibility, which
could cause the dive team to lose sight of their guideline
leading back to the surface. Larger amounts of sediment
can be collected in sealable plastic bags for later sorting
in the laboratory. Finally, minnow traps or similar funnel-
shaped traps made from plastic bottles (Manning, 1986)
can be baited with a small amount of fish, crab, or other
attractant and left within the cave for 6 to 24 h. If the trap

NUMBER 38 @ 271

is carefully placed inside a sealed plastic bag when it is
recovered, even small invertebrates can be collected.

If temperatures are kept close to cave temperature af-
ter collecting, specimens will remain alive for up to 24 h.
Photographic documentation of color pattern and natural
body position in live specimens is highly desirable. Smaller
animals can be photographed using a phototube attach-
ment on a dissecting microscope and larger specimens with
the macro setting found on many digital cameras. If animals
are too active to be photographed easily, they can be chilled
in a small dish placed in a refrigerator or an ice bath until
they stop moving. Digital video segments showing swim-
ming and other behaviors can be made in the same manner.
Specimens are sorted under a dissecting microscope using
small pipettes to transfer them to individual dishes for each
taxon. Depending upon the type of animal and its intended
use, various fixatives can be used. Most animals are best
preserved in 70% to 95% pharmaceutical grade ethanol,
which allows them to be used for either morphological
or molecular investigations. Specimens for confocal laser
scanning microscopy can be fixed in 4% paraformaldehyde
in phosphate-buffered saline (PBS) buffer (1:1 in seawater),
while those intended for scanning or transmission electron
microscopy are fixed in 2% glutaraldehyde in seawater.

GEOLOGICAL ORIGINS, AGE, AND
DISTRIBUTION OF ANCHIALINE HABITATS

Anchialine caves occur in both volcanic bedrock and
karstic limestone. Lava tube caves form during volcanic
eruptions of basaltic lava. They typically occur close to the
earth’s surface and are thus relatively short lived (thousands
to a few tens of thousands of years). Anchialine lava tubes
may originate on land and extend out under the coastline
and beneath the seafloor or can form from submarine erup-
tions. Anchialine lava tube caves are known from the Ca-
nary Islands, Galapagos Islands, Hawaii, and Western Sa-
moa. The longest of these is the Jameos del Agua (Atlantida
Tunnel) on Lanzarote in the Canary Islands, the submerged
portion of which extends 1.6 km beyond the coastline,
reaching a depth of 50 m (Iliffe et al., 2000).

The most extensive of known anchialine habitats
are solutionally developed limestone caves that typically
contain both freshwater and marine waters. Such caves
are sometimes referred to as flank margin caves and were
formed by mixing dissolution in a fresh groundwater lens
(Mylroie and Carew, 1990). The largest anchialine cave
is Sistema Ox Bel Ha located on the Caribbean coast of
the Yucatan Peninsula in Mexico; it contains 180 km of

PT/7%

surveyed underwater passages interconnecting 130 cenote
entrances. Extensive anchialine limestone caves are also
known from the Bahamas, Bermuda, Belize, Dominican
Republic, and Bonaire in the Caribbean, plus the Balearic
Islands and Sardinia in the Mediterranean. Smaller anchi-
aline caves are present on many islands in the Indo-South
Pacific and in Western Australia.

Limestone caves last much longer than lava tubes and
can be hundreds of thousands to many millions of years
old. Commonly, massive stalactites and stalagmites occur
underwater to depths in excess of 50 m in coastal lime-
stone caves. Because speleothems form very slowly and
only in air, these caves must have been dry and filled with
air for long periods of time when glacial sea levels were as
much as 130 m lower than today. The last low stand of Ice
Age sea level occurred only 18,000 years ago.

Coastal tectonic faults that extend below sea level
constitute another form of anchialine habitat. On Santa
Cruz in the Galapagos Islands, vertical faults in coastal
volcanic rock are locally called “grietas” (Iliffe, 1991).
Wedged breakdown blocks have partially roofed over
submerged portions of grietas so that they are in total
darkness. Similar faults are present in Iceland. Fault caves
also occur in uplifted reef limestone on the island of Niue
in the Central Pacific, producing deep chasms containing
anchialine pools. The Ras Muhammad Crack in the Sinai
Peninsula consists of a water-filled crack in an elevated
fossil reef formed by a 1968 earthquake (Por and Tsur-
namal, 1973). Many of the offshore ocean blue holes of
the Bahamas consist of submarine faults running parallel
to the platform edge. Ocean blue holes typically exhibit
exceptionally strong, reversing tidal currents created by an
imbalance between tides on opposite sides of the islands.

ANCHIALINE CAVE ECOLOGY
PHYSICAL AND CHEMICAL CHARACTERISTICS

The water column in most anchialine caves is highly
stratified (Iliffe, 2000). The largest changes in chemical and
physical parameters typically occur at the halocline where
freshwater or brackish water is separated from underly-
ing fully marine waters (Figure 2). It is not uncommon for
caves to possess multiple haloclines. On larger islands and
in continental regions such as Yucatan and Western Austra-
lia, freshwater occurs in the shape of a lens with thickness
increasing in a direct relationship with distance inland from
the coast. In Yucatan, the depth of the halocline and cor-
responding thickness of the freshwater lens increases from
10 m at 2 km distance inland to 20 m at 10 km inland.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Water temperature in Yucatan caves generally in-
creases with depth, although in the Bahamas the inverse
occurs and water below the halocline is generally cooler
than surface water. Warmer waters below the halocline
could be caused by geothermal heating at depth or evapo-
rative cooling at the surface.

In the lightless interior of caves, there are no plants
and hence no photosynthetic oxygen production; stable
and stratified water masses also restrict vertical mixing and
exchange of oxygen with surface waters. Thus, cave waters
are typically hypoxic to anoxic. Where deeper, water-filled
vertical shafts extend to the surface, such as in many ceno-
tes and inland blue holes, input of leaves and other organic
detritus has caused the total depletion of dissolved oxygen
with resulting anoxia and hydrogen sulfide production. A
cloud-like layer of hydrogen sulfide several meters thick oc-
curs just below the halocline and may reduce underwater
visibility to near zero, but water clarity improves consider-
ably below the H)S layer. In some caves, dissolved oxygen
levels can recover to 1 mg/L or less and populations of sty-
gobitic animals occur.

A pH minimum generally occurs at the halocline, pos-
sibly arising from microbial oxidation of organic matter
suspended at the density interface and resulting CO) pro-
duction. Increased acidity at the halocline may explain the
dissolution of limestone and the resulting development of
cave passages at this depth.

TROPHIC RELATIONSHIPS

Determination of stable carbon and nitrogen isotopes
values from animals, sediments, and other sources of or-
ganic matter in Yucatan caves has been used to examine
the trophic ecology of these systems (Pohlman et al., 1997,
2000). Four potential sources of organic matter were iden-
tified in Yucatan caves: the soil from the surrounding jun-
gle, algae from the cenote pool, chemoautotrophic bacte-
ria, and, to a lesser extent, organic matter originating from
marine waters. Stable nitrogen isotope data determined
that the food web comprised 2 to 2.5 trophic levels.

The paucity of food in anchialine caves drives organ-
isms toward a generalist diet. Mysids and isopods tend to-
ward omnivory, while ostracods and thermosbaenaceans
occupy the roles of detritivores. The thermosbaenacean Tu-
lumella and atyid shrimp Typhlatya have modified append-
ages that allow them to filter out even the tiniest particles.
Remipedes, fishes, and some amphipods, operating either
as top-level predators or as scavengers, feed on ostracods,
thermosbaenaceans, copepods, isopods, amphipods, and
shrimps.

Salinity (ppt)

(0) 10 20 30 40

Depth (m)

Dissolved Oxygen (mg/l)
0 1 2

Depth (m)

NUMBER 38 e¢ 273

Temperature (deg. C)
24.5 24.9 25.3 25.7 26.1

Depth (m)

Depth (m)

FIGURE 2. Depth profiles of salinity, temperature, dissolved oxygen, and pH from an anchialine cave, Cenote 27 Steps, Akumal, Mexico,
7 December 2003, recorded with a YSI 600 XLM multiparameter water quality monitor. Individual measurements (diamond symbols) were

taken at 4 s intervals between the surface and 26 m water depth.

BIODIVERSITY
FISHES

Stygobitic anchialine fishes (Figure 3a) are represented
in the families Bythidae (eight species in two genera from
the Bahamas, Cuba, Yucatan, and Galapagos Islands), El-
eotridae (one species from Northwestern Australia), Gobi-
idae (three species in two genera from the Philippines and
Japan), and Synbranchidae (two species in one genus from
Northwestern Australia and Yucatan) (Romero, 2001).

NON-CRUSTACEAN INVERTEBRATES

Although most stygobitic anchialine invertebrates are
crustaceans, a variety of non-crustacean invertebrate stygo-
faunal species have been described. Anchialine species in-
clude four sponges, one turbellarian, five gastropods, ten
annelids, four chaetognaths, one tantulocarid, and three
water mites. Although some of these species are question-
able stygobites, several are clearly cave adapted. The poly-

chaetes Gesiella jameensis from the Canary Islands and
Pelagomacellicephala iliffei from the Caicos Islands and Ba-
hamas (Figure 3b) conserve energy by slowly swimming in
the cave water column, while the chaetognath Paraspadella
anops from the Bahamas lacks eyes and pigment.

CRUSTACEANS

Crustaceans are the most abundant and diverse group
present in both freshwater and anchialine cave habitats.
Among the anchialine Crustacea, the largest numbers of
species are represented by amphipods, copepods, deca-
pods, ostracods, isopods, mysids, and thermosbaenaceans,
approximately in that order.

Remipedia
Remipedes are a class of Crustacea originally described

from Bahamian caves by Yager (1981). Although their
multi-segmented trunk and paired swimming appendages

274 e* SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

‘ LWA,

TD OTTAE

a) Fish: Typhliasina pearsei b) Polychaete: Pelagomacellicephala iliffei

d) Thermosbaenacean: Tulumella unidens

e) Mictacean: Mictocaris halope f) Ostracode: Spelaeoecia capax

g) Stygiomysid: Siygiomysis sp. h) Isopod: Bahalana caicosana

FIGURE 3. Characteristic anchialine cave animals include the (a) Yucatan cave fish Typhliasina pearset; (b) polynoid polychaete worm
Pelagomacellicephala iliffei from the Bahamas; (c) remipede Cryptocorynetes longulus from the Bahamas; (d) thermosbaenacean Tulumella
unidens from Yucatan; (e) mictacean Mictocaris halope from Bermuda; (f) halocyprid ostracod Spelaeoecia capax from the Bahamas; (g) stygio-
mysid Stygiomysis sp. from Yucatan; and (h) cirolanid isopod Bahalana caicosana from the Caicos Islands.

appear primitive, their head and mouth parts are highly
specialized (Figure 3c). Remipedes have paired hollow
fangs for capturing prey and are among the top preda-
tors in anchialine habitats. They are up to 4.5 cm in
length, usually colorless and blind, with elongate, cen-
tipede-like bodies. Twenty species of remipedes inhabit
fully marine, oxygen-deficient waters in caves from the
Bahamas, Caicos Islands, Cuba, Yucatan Peninsula, Do-
minican Republic, Canary Islands, and Western Austra-
lia (Koenemann et al., 2008b; Daenekas et al., 2009).
The recent discovery of free-living, nonfeeding remipede
larvae promises to yield information on the reproduction
and development as well as the evolutionary affinities of
this enigmatic group (Koenemann et al., 2007, 2009).

Thermosbaenacea

Thermosbaenaceans (Figure 3d) are small (5 mm or
less), eyeless or eye-reduced, anchialine and freshwater
peracarid crustaceans with a dorsal brood pouch in fe-
males (Wagner, 1994; Jaume, 2008). They include at least
34 species with a wide distribution in caves and thermal
springs around the Mediterranean and Caribbean, as well
as in Australia and Cambodia.

Mictacea

Mictaceans (Figure 3e) are small (3-3.5 mm), eyeless
and depigmented, nonpredatory crustaceans. This peracarid
order is represented by only a single species that inhabits
anchialine caves in Bermuda (Bowman and Iliffe, 1985).

Bochusacea

Bochusaceans are very small (1.2-1.6 mm), semi-
transparent, and eyeless peracarid crustaceans that in-
clude two anchialine species from the Bahamas and Cay-
man Islands, plus two deep-sea species (Gutu and Iliffe,
1998; Jaume et al., 2006).

Copepoda

Platycopioid, misophrioid, cyclopoid, harpacticoid,
and calanoid (especially epacteriscid and ridgewayiid)
copepods inhabit anchialine caves in tropical regions
around the globe. They are small (typically 1-2 mm long)
and have a short, cylindrical body with head and thorax
fused into a cephalothorax. Most are planktonic filter
feeders, but some, such as the harpacticoids and cyclo-
poids, are benthic, while epacteriscids are predators on
other copepods.

NUMBER 38 ¢ 275

Ostracoda

Halocyprid ostracods (Figure 3f) include anchialine
species with a distribution and co-occurrence similar to
that of remipedes (Kornicker et al., 2007). Danielopolina
is the most widely distributed stygobitic genus with species
on opposite sides of both the Atlantic and Pacific, inhabit-
ing caves in the Bahamas, Cuba, Yucatan, Jamaica, Canary
Islands, Galapagos, Western Australia, and Christmas Is-
land. More than 300 species of podocopid ostracods have
been found in springs, caves, and anchialine habitats.

Mysidacea

Stygobitic mysids are found in freshwater and anchia-
line habitats in Africa, the Caribbean, Mediterranean, and
India. Their distribution suggests that they were stranded in
caves by lowering of the sea level in the Tethys and Mediter-
ranean. Recent molecular phylogenies of the mysids suggest
that a new order is justified for the stygiomysids (Figure 3g),
which inhabit caves in the Caribbean and Italy (Meland and
Willassen, 2007).

Isopoda

Stygobitic isopods (Figure 3h) range from several mil-
limeters to several centimeters in length. Anthurid isopods
occur in anchialine and freshwater caves in the Canary
Islands, Caribbean and Indian Ocean islands, Mexico,
and South America. Asellot isopods inhabit anchialine and
freshwater caves in the Caribbean, Europe, Galapagos, In-
dia, Indonesia, Japan, Malaysia, North and Central Amer-
ica, and Polynesia. Cirolanid isopods have been found in
freshwater and anchialine caves clustered in Mexico and
the Caribbean (Iliffe and Botosaneanu, 2006), as well as in
Europe and the Mediterranean.

Amphipoda

Amphipods occur in freshwater and marine cave habi-
tats. Stygobitic representatives are present in the bogidiel-
lid, crangonyctid, hadziid, and niphargid families of the
amphipod suborder Gammaridea. They are very widely
dispersed, with large numbers of species inhabiting caves
in Central and Southern Europe, the Mediterranean, east-
ern and southern North America, and the Caribbean.

Decapoda

The anomuran galatheid crab Munidopsis polymor-
pha inhabits an anchialine lava tube in the Canary Islands

276 e

(Wilkens et al., 1990). Brachyuran crabs are widely dis-
tributed in caves of the tropics and subtropics. Anchialine
stygobitic shrimp include representatives from the carid-
ean families Agostocarididae, Alpheidae, Atyidae, Hippo-
lytidae, Palaemonidae, and Procarididae; the stenopodid
family Macromaxillocarididae; and the thalassinid family
Laomediidae.

Other Crustacean Stygofauna

One tantulocarid, an exceptionally tiny ectoparasite
on anchialine harpacticoid copepods, occurs in the Canary
Islands (Boxshall and Huys, 1989). A species of stygobitic
nebaliacean inhabits anchialine caves in the Bahamas and
Caicos Islands (Bowman et al., 1985). Several species of
cumaceans and tanaidaceans have been collected from an-
chialine caves in Bermuda and the Bahamas, but it is not
clear whether they belong to the stygofauna.

BIOGEOGRAPHY

Upon examining the biogeography of anchialine fauna,
some extraordinary patterns are evident. A number of an-
chialine genera, including the remipede Lasionectes, ostra-
cod Danielopolina, thermosbaenacean Halosbaena, and
misophrioid Speleophria, inhabit caves on opposite sides of
the Earth and are believed to be relicts whose ancestors in-
habited the Tethys Sea during the Mesozoic (Humphreys,
2000). Some anchialine taxa are represented in the Mediter-
ranean, but others, notably remipedes and Halosbaena, are
absent. The presence of anchialine taxa at all in the Mediter-
ranean is remarkable considering that this basin was com-
pletely dry for long periods of time during the Miocene. The
aytid shrimp Typhlatya shows an especially interesting dis-
tribution with 17 known species inhabiting freshwater and
anchialine caves in the Mediterranean region, Bermuda, As-
cension Island, Caribbean locations including Cuba and Yu-
catan, and the Galapagos Islands (Alvarez et al., 2005). The
shrimp family Procaridae contains one genus with species in
the mid-Atlantic and Caribbean, as well as Hawaii.

Based on numbers of stygobitic species, the Bahamian
archipelago appears to have been a possible center of ori-
gin for anchialine fauna. Among the Remipedia, 15 of
20 described species inhabit caves in the Bahamas (Koen-
emann et al., 2008b; Daenekas et al., 2009), whereas
among anchialine halocyprid ostracods, Bahamian spe-
cies account for 4 of 11 in the genus Danielopolina, 6 of
11 in Spelaeoecia, and all 8 species of Deeveya (Kornicker
et al., 2007).

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

The Bahamas archipelago consists of a series of broad,
shallow-water, highly karstified, carbonate platforms rising
abruptly from the deep sea. The islands and cays consist of
Pleistocene limestone covered by a thin veneer of Holocene
carbonate reefs and sediments. Underlying these younger
limestones is a continuous section of Tertiary and Cretaceous
limestones and dolomites exceeding 11 km in thickness. If
the position of the tectonic plates before the development of
the Atlantic Ocean is reconstructed, virtually all the Baha-
mas overlap the African continent and its continental shelf.
This finding suggests that the Bahama platform developed
over oceanic crust during the earliest phase of the creation of
the Atlantic. The extended shallow-water history of the Ba-
hamas, coupled with the cavernous nature of the limestone,
may help to explain its rich and diverse anchialine fauna.

ORIGINS OF ANCHIALINE BIOTA

A number of theories have been proposed to explain
the trans-oceanic distribution of many anchialine taxa. The
vicariance model suggests that plate tectonics served as a
mechanism for the dispersal of anchialine fauna (Rosen,
1976; Wiley, 1988). This model mainly describes the Teth-
yan track of ancient taxa that were rafted on the drifting
continents to their present positions (Stock, 1993; Jaume
et al., 2001). However, the existence of anchialine fauna on
mid-ocean islands such as Bermuda, Ascension, and Hawaii
that have never been part of or closer to a continent cannot
be explained by this mechanism (Iliffe, 2000).

The regression model suggests that sea-level regres-
sions, caused by tectonic uplift or eustatic glacial lower-
ing of sea levels, stranded crevicular or interstitial marine
littoral species that subsequently adapted to brackish or
freshwater conditions (Stock, 1980). This model is sup-
ported by the observed correlation between the distribu-
tion patterns of numerous, marine-derived cave organisms
and the position of shorelines during the Late Mesozoic
or Tertiary seas. Nevertheless, the presence of anchialine
fauna in caves that were completely dry and air filled (as
evidenced by their now-submarine speleothems) less than
10,000 years ago indicates that these animals can migrate
vertically with raising postglacial sea levels (Iliffe, 2000).
Also, small islands such as Bermuda offer little chance for
marine species to be stranded.

A deep-sea origin has been proposed for some an-
chialine species having close relatives that inhabit bathyal
depths (Hart et al., 1985). Both caves and the deep sea
are old, climatically stable, lightless, and nonrigorous en-
vironments. Anchialine habitats on islands and continental

margins could be connected via a continuum of crevicular
corridors extending from shallow depths to the deep sea
(Iliffe, 1990). However, evidence against a deep-sea origin
of cave faunas includes the questionable ability of deep-
sea species to cross the oceanic thermocline, the relatively
recent nature of deep-sea species (resulting from the lack
of oxygen in Atlantic bathyal waters during the late Oligo-
cene), and phylogenetic analyses of morphological charac-
ters supporting independent colonization of deep-sea and
anchialine habitats (Stock, 1986; Danielopol, 1990).

The active migration model involves the inland dis-
persal and colonization of subterranean habitats by ex-
pansionistic marine species with a high degree of salinity
tolerance (Rouch and Danielopol, 1987). This process is
independent of climatological and geological variations.

Passive oceanic dispersal of larval or postlarval stages
of anchialine species by currents could explain the wide
distribution of some anchialine shrimp species within the
Indo-Pacific. Rafting on floating objects, such as wood,
algae, kelp, and coconuts, or on mobile and migratory
animals, for instance, sea turtles, fishes, and larger arthro-
pods, could disperse anchialine species, even those without
a free larval stage. However, oceanic dispersal is unlikely
for many anchialine groups that produce few offspring or
have narrow physiological tolerances.

ADAPTATION TO LIFE
IN ANCHIALINE CAVES

BEHAVIORAL ADAPTATIONS

Behavioral adaptations are the most immediate adap-
tations for survival and colonization in cave systems. Cave
organisms, in particular amblyopsid cave fishes, use a glide-
and-rest technique to conserve energy in their search for
food. Remipede locomotion is also designed for the economy
of movement. Remipedes swim slowly, using less energy for
the same distance than if they swam at higher speeds (Koen-
emann et al., 2008a). The power stroke produces drag by
individual legs, but the recovery stroke is completed with the
legs folded with other legs to reduce water resistance.

The stygobitic galatheid crab Munidopsis polymor-
pha, inhabiting an anchialine lava tube in the Canary Is-
lands, has a number of specialized behaviors (Parzefall,
1992, 2000). These small crabs are most abundant in a
dimly illuminated pool where they hide in rock crevices
during the day but come out at night to feed on diatoms.
Because of the large numbers of individuals in this pool,
they spread in an almost regular pattern determined by the
length of the second antennae.

NUMBER 38 ¢* 277

Munidopsis crabs remain aggressive throughout the
year. They detect intruders from water movements and at-
tack with extended chelipeds. Male crabs are attracted by
a molting hormone released by females. To prevent the
females from fleeing, males rhythmically move their che-
lipeds as they approach, until the female responds by vi-
brating one of her chelipeds. The male then seems to turn
the female over on her back to initiate insemination.

MORPHOLOGICAL ADAPTATIONS
Regressive Features

The loss of features that in cave environments no longer
have a function, such as eyes and pigmentation, is regarded as
regressive evolution. There are two main theories explaining
the driving force for regressive evolution. In an environment
with a depauperate food supply, natural selection should
favor reallocating energy from developing unused features,
such as eyes and pigment, to growth and survival. A second
explanation is that regressive evolution may be the result of
nonselective processes such as neutral mutation and genetic
drift. Features such as eyes and pigment that abruptly lose
their biological function when animals enter caves are free to
be turned off by now non-lethal mutations.

Unfortunately, the theory of energy economy by char-
acter reduction in stygobites is not well tested, especially
with anchialine stygobites, yet the anchialine environment
is dominated by blind, depigmented organisms.

Constructive Features

In the case of constructive features, priority is given to
life history, metabolism, development, and starvation re-
sistance, with sensory development such as mechano- and
chemoreceptors being subordinate. For troglomorphy to
occur, two factors must be present: (1) selective pressure
in favor of the development and (2) genetic, physiological,
or behavioral ability of the organism to respond to the
selective pressure. A prerequisite for constructive traits is
their genetic availability in epigean forms: if traits are not
present in epigean ancestors, they will not be present in
hypogean descendants.

There are several areas of the body where constructive
features occur. In crustaceans, appendages may be elon-
gated, in particular, the antennae, and in fish, the head
may become enlarged or flattened. Corresponding to the
morphological changes, there is an increased sensitivity to
chemical and mechanical stimulants. As a result of com-
pensatory enhancement of extraocular senses, the signal-
processing structures in the brain are altered.

278 e

PHYSIOLOGICAL ADAPTATIONS
Adaptations to a Food-Poor Environment

Food in the stygobitic environment may be in general
scarce or at best patchy; therefore, the stygofauna need
to cope with temporal periodicity of food availability and
potentially tolerate long periods of starvation. This adap-
tation occurs through lipid accumulation or energy econ-
omy. In comparison with pelagic crustaceans, anchialine
crustaceans sacrifice protein mass for increased lipid stores
(Iliffe and Bishop, 2007). Lipids provide neutral buoyancy
without energy expenditure, while also serving as an en-
ergy reserve when food is limiting. Anchialine stygobites
also tend to be smaller than their epigean counterparts.
Their small size is a mechanism for energy economy.

Adaptation to Hypoxia and Anoxia

As mentioned previously, the anchialine environment,
especially at or below the halocline, is commonly hypoxic
or even anoxic. As a result, hypogean organisms tend to
have substantially lower oxygen consumption rates than
their epigean relatives (Bishop et al., 2004). Many organ-
isms are capable of obtaining energy when faced with a re-
duction or absence of oxygen, but few are able to survive
indefinitely without a return to oxygen. When the oxygen
supply becomes inadequate, organisms switch to anaero-
biosis to compensate for adenosine triphosphate (ATP)
demand.

During periods of anaerobiosis, organisms conserve
their energy stores by a loss in physiological functions
such as motility, ingestion, and digestion, combined with a
dramatic depression of their energy (ATP) demand. When
oxygen is temporarily unavailable, many organisms switch
to anaerobic glycolysis. Anaerobic glycolysis is, however, a
fundamentally inefficient metabolic strategy and thus not
an attractive solution for anchialine organisms.

By examining the activities of enzymes critical to me-
tabolism and energy conversion, it is possible to deter-
mine the rate at which food is converted to cellular energy
(Bishop et al., 2004). Citrate synthase (CS) is an indicator
of an organism’s maximum aerobic potential, or how fast
an organism can aerobically convert glucose to energy.
Malate dehydrogenase (MDH) functions in the presence
as well as absence of oxygen, whereas lactate dehydrog-
enase (LDH) contributes to both aerobic and anaerobic
metabolic pathways and serves as an indicator of glyco-
lytic potential.

Anchialine organisms are anaerobically poised with
both LDH:CS and MDH:LDH ratios tending to be greater

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

than one. The higher the MDH:LDH ratio, the greater
is the tolerance to hypoxia. Such high ratios indicate
an evolutionary adaptation to the anaerobic anchialine
environment.

CONSERVATION

Over the past 25 years, more than 400 new species of
anchialine stygobites have been discovered and described.
A high percentage of these species are known only from a
single cave or cave system. Even within caves, species are
characteristically found only at specific depths or locations
as defined by a narrow range of environmental parame-
ters. In many parts of the world, tourism development,
limestone quarries, and groundwater pollution are either
destroying or grossly polluting numerous caves, resulting
in extinction of untold numbers of species.

Anchialine species qualify for inclusion on endangered
species lists for reasons of their limited distribution and
the declining environmental quality of their habitat. In
Bermuda, 25 cave species are on the IUCN (International
Union for Conservation of Nature) Red List of endangered
species. Other cave species from the Yucatan Peninsula are
on the official Mexican list of threatened and endangered
species.

Maintaining groundwater quality is essential to the
environmental health of the subterranean environment.
For example, the small oceanic island of Bermuda is the
third most densely populated country in the world and has
the largest number of private cesspits per capita. Disposal
of sewage and other wastewater into cesspits or by pump-
ing down boreholes is contaminating the groundwater
and cave water with nitrates, detergents, toxic metals, and
pharmaceuticals; depleting the very limited amounts of
dissolved oxygen in cave water; and generating toxic levels
of hydrogen sulfide.

Some ocean caves such as the Blue Holes of the Baha-
mas have strong tidal currents sweeping through them for
very considerable distances. In one such cave, plastic bot-
tles and other trash have been observed littering the floor
of the cave nearly a mile back into previously unexplored
passages. Far too many caves and sinkholes are viewed as
preferred locations for the dumping of garbage and other
waste products.

Another serious environmental problem concerns the
destruction of caves by limestone quarries or construction
activities. Half a dozen or more Bermuda caves have been
totally destroyed by two limestone quarries that produce
crushed aggregate for construction purposes. Untold other

caves have been lost to enormous limestone quarries in the
Yucatan Peninsula. Many caves have been filled in and built
over by golf courses, hotels, and housing developments in
Bermuda. Recently, a series of luxury town homes was
built directly on top of the largest cave lake in Bermuda.

Sometimes even seemingly innocent activities can
threaten caves and cave animals. Along the Caribbean
coast of the Yucatan Peninsula, many open water cenote
pools are inhabited by the freshwater fish Astyanax fascia-
tus. Some of these fish frequently follow divers into caves,
moving in front of the dive team and voraciously darting
in to consume any cave fish or crustaceans that are illumi-
nated by the beam of a dive light. Considering the many
thousands of cave divers who use these systems each year,
it is not surprising that the caves most heavily visited by
tourist divers are now essentially devoid of life.

Even the gas exhaled by divers may have adverse ef-
fects on cave animals. Because anchialine cave waters
typically contain extremely low levels of dissolved oxy-
gen, exhaust bubbles from open circuit scuba could have
profound effects on the cave ecosystem. Several anchialine
caves in Western Australia with unique fauna are currently
off limits to open circuit divers and may only be visited by
those using rebreathers (Humphreys et al., 1999).

Some anchialine caves in Bermuda, the Canary Islands,
and Mallorca have been developed into commercial tourist
attractions. Unfortunately, many of the tourists visiting these
sites have viewed the deep clear water cave pools as natural
wishing wells in which to throw a coin or two. Copper coins
tend to rapidly deteriorate and dissolve in saltwater, pro-
ducing high levels of toxic copper ions in the cave waters.
In one such cave in the Canary Islands, the endemic crab
Munidopsis polymorpha has shown a marked decline in
abundance during the past decade or longer, probably in
response to high levels of copper in the cave water.

ACKNOWLEDGMENTS

Investigations of the anchialine cave fauna of the Baha-
mas were supported by awards DEB-9870219 and DEB-
0315903 from the U.S. National Science Foundation’s Bio-
diversity Surveys and Inventories Program and by grants
from the NOAA Caribbean Marine Research Center to
T. Iliffe. Collection of specimens was carried out under a
permit from the Bahamas Department of Fisheries. This re-
search would not have been possible without the generous
collaboration of numerous scientists, graduate students,
and cave divers from around the world. Renee Bishop
(Penn State University at Worthington Scranton) gener-
ously provided much useful information on physiological

NUMBER 38 e¢ 279

adaptations of anchialine stygobites. This is contribution
number 845 of the Caribbean Coral Reef Ecosystems Pro-
gram (CCRE), Smithsonian Institution, supported in part
by the Hunterdon Oceanographic Research Fund.

LITERATURE CITED

Alvarez, E, T. M. Iliffe, and J. L. Villalobos. 2005. New Species of the
Genus Typhlatya (Decapoda: Atyidae) from Anchialine Caves in
Mexico, the Bahamas, and Honduras. Journal of Crustacean Biol-
ogy, 25(1):81-94.

Bishop, R. E., B. Kakuk, and J. J. Tones. 2004. Life in the Hypoxic and
Anoxic Zones: Metabolism and Proximate Composition of Carib-
bean Troglobitic Crustaceans with Observations on the Water
Chemistry of Two Anchialine Caves. Journal of Crustacean Biol-
ogy, 24:379-392.

Bowman, T. E., and T. M. Iliffe. 1985. Mictocaris halope, a New Unusual
Peracaridan Crustacean from Marine Caves on Bermuda. Journal
of Crustacean Biology, 5:58-73.

Bowman, T. E., J. Yager, and T. M. Iliffe. 1985. Speonebalia cannoni,
n. gen., n. sp., from the Caicos Islands, the First Hypogean Lep-
tostracan (Nebaliacea, Nebaliidae). Proceedings of the Biological
Society of Washington, 98:439-446.

Boxshall, G. A., and R. Huys. 1989. New Tantulocarid, Stygotantulus
stocki, Parasitic on Harpacticoid Copepods, with an Analysis of
the Phylogenetic Relationships Within the Maxillopoda. Journal of
Crustacean Biology, 9(1):126-140.

Chevaldonné, P., B. Sket, C. Marschal, C. Lejeusne, and R. Calado. 2008.
Improvements to the “Sket Bottle”: A Simple Manual Device for
Sampling Small Crustaceans from Marine Caves and Other Cryptic
Habitats. Journal of Crustacean Biology, 29(1):185-188.

Daenekas, J., T. M. Iliffe, J. Yager, and S. Koenemann. 2009. Speleonectes
kakuki, a New Species of Remipedia (Crustacea) from Anchialine
and Sub-seafloor Caves on Andros and Cat Island, Bahamas. Zoo-
taxa, 2016:51-66.

Danielopol, D. L. 1990. The Origin of the Anchialine Cave Fauna—the
“Deep Sea” Versus the “Shallow Water” Hypothesis Tested Against
the Empirical Evidence of the Thaumatocyprididae (Ostracoda).
Bijdragen tot de Dierkunde, 60(3/4):137-143.

Gutu, M., and T. M. Iliffe. 1998. Description of a New Hirsutiid (n. g.,
n. sp.) and Reassignment of This Family from Order Mictacea to the
New Order, Bochusacea (Crustacea, Peracarida). Travaux du Mu-
see National d’Histoire Naturelle “Grigore Antipa”, 40:93-120.

Hart, C. W., Jr, R. B. Manning, and T. M. Iliffe. 1985. The Fauna of
Atlantic Marine Caves: Evidence of Dispersal by Sea Floor Spread-
ing While Maintaining Ties to Deep Waters. Proceedings of the Bio-
logical Society of Washington, 98:288-292.

Humphreys, W. F. 2000. “The Hypogean Fauna of the Cape Range Pen-
insula and Barrow Island, Northwestern Australia.” In Ecosystems
of the World. 30. Subterranean Ecosystems, ed. H. Wilkens, D. C.
Culver, and W. EF Humphreys, pp. 581-601. Amsterdam: Elsevier
Science.

Humphreys, W. F., A. Poole, S. M. Eberhard, and D. Warren. 1999. Ef-
fects of Research Diving on the Physico-Chemical Profile of Bun-
dera Sinkhole, an Anchialine Remiped Habitat at Cape Range,
Western Australia. Journal of the Royal Society of Western Aus-
tralia, 82:99-108.

Hunter, R. L., M. S. Webb, T. M. Iliffe, and J. R. Alvarado Bremer.
2008. Phylogeny and Historical Biogeography of the Cave-Adapted
Shrimp Genus Typhlatya (Atyidae) in the Caribbean Sea and West-
ern Atlantic. Journal of Biogeography, 35(1):65-75.

280 e

lliffe, T. M. 1990. Crevicular Dispersal of Marine Cave Faunas. Mémoires
de Biospéologie, 17:93-96.

. 1991. “Anchialine Cave Fauna of the Galapagos Islands.” In

Galapagos Marine Invertebrates, ed. M. J. James, pp. 209-231.

New York: Plenum Press.

. 2000. “Anchialine Cave Ecology.” In Ecosystems of the World.

30. Subterranean Ecosystems, ed. H. Wilkens, D. C. Culver, and

W. F. Humphreys, pp. 59-76. Amsterdam: Elsevier Science.

. 2004. “Anchialine Caves, Biodiversity in.” In Encyclopedia of
Caves, ed. D. C. Culver and W. B. White, pp. 24-30. Burlington,
Mass:.: Elsevier.

Iliffe, T. M., and R. E. Bishop. 2007. “Adaptations to Life in Marine
Caves.” In Fisheries and Aquaculture: Towards Sustainable Aquatic
Living Resources Management, ed. P. Safran, pp. 1-26. In Encyclo-
pedia of Life Support Systems (EOLSS). Oxford: UNESCO-EOLSS
Publishers.

Iliffe, T. M., and L. Botosaneanu. 2006. The Remarkable Diversity of Sub-
terranean Cirolanidae (Crustacea: Isopoda) in the Peri-Caribbean
and Mexican Realm. Bulletin de L’Institute Royal des Sciences
Naturelles de Belgique, Biologie, 76:5-26.

Iliffe, T. M., and C. Bowen. 2001. Scientific Cave Diving. Marine Tech-
nology Society Journal, 35(2):36-41.

Iliffe, T. M., J. Parzefall, and H. Wilkens. 2000. “Ecology and Species
Distribution of the Monte Corona Lava Tunnel on Lanzarote, Ca-
nary Islands.” In Ecosystems of the World. 30. Subterranean Eco-
systems, ed. H. Wilkens, D. C. Culver, and W. F Humphreys, pp.
633-644. Amsterdam: Elsevier Science.

Jaume, D. 2008. Global Diversity of Spelaeogriphaceans & Thermos-
baenaceans (Crustacea; Spelaeogriphacea & Thermosbaenacea) in
Freshwater. Hydrobiologia, 595:219-224.

Jaume, D., G. A. Boxshall, and R. N. Bamber. 2006. A New Genus from
the Continental Slope Off Brazil and the Discovery of the First
Males in the Hirsutiidae (Crustacea: Peracarida: Bochusacea). Zoo-
logical Journal of the Linnaean Society, 148(2):169-208.

Jaume, D., G. A. Boxshall, and W. F Humphreys. 2001. New Stygobiont
Copepods (Calanoida; Misophrioida) from Bundera Sinkhole, an
Anchialine Cenote in North-Western Australia. Zoological Journal
of the Linnean Society, 133:1-24.

Koenemann, S., F. R. Schram, A. Bloechl, T. M. Iliffe, M. Hoenemann,
and C. Held. 2007. Post-embryonic Development of Remipede
Crustaceans. Evolution & Development, 9(2): 117-121.

Koenemann, S., F. R. Schram, T. M. Iliffe, L. M. Hinderstein, and A.
Bloechl. 2008a. Behavior of Remipedia in the Laboratory, with
Supporting Field Observations. Journal of Crustacean Biology,
27(4):534-542.

Koenemann, S., M. Ziegler, and T. M. Iliffe. 2008b. Pleomothra fragilis
n. sp. (Remipedia) from the Bahamas, with Remarks on Morpho-
logic Reductions and Postnaupliar Development. Journal of Crus-
tacean Biology, 28(1):128-136.

Koenemann, S., J. Olesen, F. Alwes, T. M. Iliffe, M. Hoenemann, P. Un-
gerer, C. Wolff, and G. Scholtz. 2009. The Post-embryonic Devel-
opment of Remipedia (Crustacea)—Additional Results and New
Insights. Development Genes and Evolution, 9(2):117-21.

Kornicker, L. S., T. M. Iliffe, and E. Harrison-Nelson. 2007. Ostracoda
(Myodocopa) from Anchialine Caves and Ocean Blue Holes. Zoo-
taxa, 1565:1-151.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Manning, R. B. 1986. A Small Trap for Collecting Crustaceans in Shal-
low Water. Proceedings of the Biological Society of Washington,
99(2):266-268.

Meland, K., and E. Willassen. 2007. The Disunity of “Mysidacea”
(Crustacea). Molecular Phylogenetics and Evolution, 44(3):1083-
1104.

Mylroie, J. E., and J. L. Carew. 1990. The Flank Margin Model for Dis-
solution Cave Development in Carbonate Platform. Earth Surface
Processes and Landforms, 15:413-424.

Parzefall, J. 1992. “Behavioural Aspects in Animals Living in Caves.” In
The Natural History of Biospeleology, ed. A. T. Camacho, pp. 325-
376. Madrid: Museo Nacional de Ciencias Naturales.

. 2000. “Ecological Role of Aggressiveness in the Dark.” In Eco-
systems of the World. 30. Subterranean Ecosystems, ed. H. Wilkens,
D. C. Culver, and W. F. Humphreys, pp. 221-228. Amsterdam: Else-
vier Science.

Pohlman, J. W., L. A. Cifuentes, and T. M. Iliffe. 2000. “Food Web Dy-
namics and Biogeochemistry of Anchialine Caves: A Stable Isotope
Approach.” In Ecosystems of the World. 30. Subterranean Eco-
systems, ed. H. Wilkens, D. C. Culver, and W. EF Humphreys, pp.
345-357. Amsterdam: Elsevier Science.

Pohlman, J. W., T. M. Iliffe, and L. A. Cifuentes. 1997. A Stable Isotope
Study of Organic Cycling and the Ecology of an Anchialine Cave
Ecosystem. Marine Ecology Progress Series, 155:17-27.

Por, EF D., and M. Tsurnamal. 1973. Ecology of the Ras Muhammad
Crack in Sinai. Nature (London), 241:43-44.

Rouch, R., and D. L. Danielopol. 1987. VOrigine de la Faune Aquatique
Souterraine, entre le Paradigme du Refuge et le Modéle de la Colo-
nization Active. [The Origin of the Subterranean Aquatic Fauna,
Between the Refuge Paradigm and the Active Colonization Model.]
Stygologia, 3:345-372.

Romero, A., ed. 2001. The Biology of Hypogean Fishes. Dordrecht, The
Netherlands: Kluwer Academic.

Rosen, D. E. 1976. A Vicariance Model of Caribbean Biogeography. Sys-
tematic Zoology, 24:431-464.

Sket, B. 1996. The Ecology of the Anchialine Caves. Trends in Ecology
and Evolution, 11:221-225.

Stock, J. H. 1980. Regression Model Evolution as Exemplified by the
Genus Pseudoniphargus. Bijdragen tot de Dierkunde, 50:105-144.

. 1986. Deep Sea Origin of Cave Faunas, an Unlikely Supposition.

Stygologia, 2:105-111.

. 1993. Some Remarkable Distribution Patterns in Stygobiont
Amphipoda. Journal of Natural History, 27:807-819.

Wagner, H. P. 1994. A Monographic Review of the Thermosbaenacea
(Crustacea: Peracarida). Zoologische Verhandelingen, 291:1-338.

Wiley, E. O. 1988. Vicariance Biogeography. Annual Review of Ecology
and Systematics, 19:513-542.

Wilkens, H., J. Parzefall, and A. Ribowski. 1990. Population Biology
and Larvae of the Anchialine Crab Munidopsis polymorpha (Gal-
atheidae) from Lanzarote (Canary Islands). Journal of Crustacean
Biology, 10(4):667-675.

Yager, J. 1981. A New Class of Crustacea from a Marine Cave in the
Bahamas. Journal of Crustacean Biology, 1:328-333.

Zakések, V., B. Sket, and P. Trontelj. 2007. Phylogeny of the Cave Shrimp
Troglocaris: Evidence of a Young Connection Between Balkans and
Caucasus. Molecular Phylogenetics and Evolution, 42(1):223-235.

Decimating Mangrove Forests
for Commercial Development
in the Pelican Cays, Belize: Long-Term
Ecological Loss for Short-Term Gain?

lan G. Macintyre, Marguerite A. Toscano,
Ilka C. Feller, and Maria A. Faust

Ian G. Macintyre and Marguerite A. Toscano,
Department of Paleobiology, National Museum
of Natural History, Smithsonian Institution,
P.O. Box 37012, MRC 125, Washington, D.C.
20013-7012, USA. Ilka C. Feller, Smithsonian
Environmental Research Center, 647 Contees
Wharf Road, Edgewater, Maryland 21037, USA.
Maria A. Faust, Department of Botany, National
Museum of Natural History, Smithsonian Institu-
tion, 4210 Silver Hill Road, Suitland, Maryland
20746, USA. Corresponding author: I. Macintyre
(macintyr @si.edu). Manuscript received 13 May
2008; accepted 20 April 2009.

ABSTRACT. The unique, biologically diverse ecosystems of Pelican Cays, Belize, are in
serious danger from sediment suffocation related to the recent clear-cutting of mangroves
for commercial development in what is currently designated Southwater Cay Marine
Reserve. Field observations in the Pelican Cays in March 2007 revealed extensive clear-
cutting of mangroves and covering of exposed peat surfaces with sediment dredged from
the adjacent seafloor to create false sand cays. On Manatee Cay, introduction of dredge
spoils taken from the nearby seabed resulted in fine sediment plumes spilling into the
adjacent ponds, smothering the attached benthic communities on mangrove roots and
burying Thalassia bottom communities. In addition, comparative studies of microalgal
(phytoplankton) assemblages in a Manatee Cay pond before and after mangrove clearing
indicate a dramatic loss in this group. This change, related to high turbidity observed in
the water column, signals a serious impact to this aquatic ecosystem. In March 2007,
clear-cutting, burning, and dredge and fill operations were taking place on Fisherman’s
Cay, with additional survey lines visible on Fisherman’s, Manatee, and Cat Cays. We used
a series of aerial photographic surveys from 2003 to 2007 to document the extensive loss
of mangroves on both Manatee and Fisherman’s Cays. To date, additional clearing of
mangroves has occurred on Northeast Cay, Bird Cays, and Ridge Cay, resulting in a total
of 15.3 ha or more than 29% of the mangrove community that have been destroyed in
the Pelican Cays. Furthermore, several survey lines through still-forested areas on these
islands indicated that additional clearing of mangroves was planned. The Pelican Cays
ponds contain unique, biologically diverse ecosystems dominated by delicate sessile pho-
tosynthetic and filter-feeding populations; these rare communities will be lost as a result
of sediment suffocation caused by the clearing and filling of these islands. However, the
conversion of mangrove ecosystems for residential, tourism, and commercial uses is both
widespread and accelerating in Belize and throughout the global tropics. This pressure
is having an adverse effect on the health of coral reefs and the biomass and viability of
commercial fisheries, which are essential for both tourism and local livelihoods.

INTRODUCTION

The Pelican Cays group is an oceanic coral reef boundary environment (Mac-
intyre et al., 2000a), containing a network of coral ridges and semi-enclosed or
enclosed ponds (Figure 1) where shallow mangrove cays are immediately adja-
cent to channels approximately 20 to 30 m deep. The lagoon-like ponds, which

282 e

—E |
Oo km

Mexico

Caribbean Sea

Guatemala

Honduras
88°W

Co Cat Cay:

6

Avicennia Cay

Little Cat Cay :

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

: Northeast Cay

: Ridge Cay

Fisherman’s Cay

FIGURE 1. Index map showing the major mangrove islands of the Pelican Cays and their ponds.

may be 10 to 12 m deep, harbor high-diversity, low-energy
environments dominated by populations of photosynthetic
and filter-feeding organisms. These ponds were formed by
differential coral accumulations on polygonal karst pat-
terns of the underlying Pleistocene limestone (Macintyre et
al., 2000b). The scleractinian coral Acropora cervicornis,
which is capable of rapid accretion in response to sea-level

rise (catch up), exaggerated the underlying karst topogra-
phy, resulting in greater relief than that of the antecedent
karst surface, and created steep-sided ridges that form the
ponds. Red mangrove (Rhizophora mangle) communities
have established on the reef ridges of this area, forming
intertidal cays encircling or partially encircling the ponds.
The mangrove prop roots extending into the ponds pro-

vide substrate for rich communities of sponges, ascidians,
algae, corals, bryozoans, mollusks, and other organisms
(Macintyre and Ritzler, 2000).

Pelican Cays ponds and associated ridges support a
very high level of shallow marine biodiversity within the
Belize Lagoon. Examples include 70 species in 30 genera
of ascidians (Figure 2A), representing 60% of all known
shallow-water species in the Caribbean (Goodbody, 2000).
Ten of 52 species of echinoderms found in the ponds and
associated space had never been previously reported from
Belize waters (Hendler and Pawson, 2000). Of 187 sponge
species (Figure 2B) reported for several mangrove island
groups along Belize Mesoamerican Barrier Reef, Ritzler
et al. (2000) found the “most diverse sponge fauna” at
Pelican Cays. Of the 147 sponge species at Pelican Cays,
45% were new species or variants special to the pond en-
vironments. Manatee Cay had 95 species, Cat Cay had 77
species, and Fisherman’s Cay had 90 species. Wulff (2000)
attributed sponge community differences to the fact that
the Pelican Cays’ mangrove roots are embedded in coral
reefs rather than thick peat sections as at Twin Cays (Mac-
intyre et al., 2004) and Tobacco Range (Macintyre et al.,
1995). The reef substrate may be a preferred environment
for spongivorous fishes that determine the distinctive spe-
cies composition of the Pelican Cays sponge community
(Wulff, 2000). Richardson (2000) reported a total of 7
species of epiphytic foraminifera living on turtle grass
(Thalassia testudinum) blades, of which 2 were new spe-
cies. Littler et al. (2000) reported 152 species of marine
macrophytes, of which 148 were algae and 4 were vas-
cular plants. A total of 31 bryozoan species were found
in the Pelican Cays, forming extensive colonies on the
mangrove roots (Winston, 2007). Coral species on ridges
and in deeper or more open areas of the ponds included
Porites furcata, P. divaricata, P. porites, P. astreoides, Acro-
pora cervicornis, Siderastrea siderea, Agaricia tenuifolia,
Millepora complanata, and Montastrea annularis (mainly
at the opening of Pond E, Fisherman’s Cay; see Figure 1).
Barnacles and mollusks also inhabit the ponds in signifi-
cant numbers.

Faust (2000) identified 110 species in 33 genera of
planktonic, oceanic, red tide-forming, benthic, and coastal
dinoflagellate species from six of the Pelican Cays of great
typological diversity. Approximately 50% of these ap-
peared to be new species. Manatee Cay had 93 species,
Douglas Cay, 47 species, and Cat Cay, 32 species. Waters
in the Pelican Cays allow dinoflagellates to proliferate in
a naturally nutrient-enriched environment, protected from
prevailing winds by the surrounding mangroves and coral
ridge (Faust, 2000).

NUMBER 38 e¢ 283

In March 2000, a special issue of the Atoll Research
Bulletin on the biology and physical characteristics of the
Pelican Cays ponds was published to assist the efforts of
the Government of Belize (GOB) to determine if this area
should be included in the South Water Cay Marine Re-
serve (SWCMR). It was hoped that by bringing attention
to the unique characteristics of these fragile communities
they would be preserved. Based on those studies, the Peli-
can Cays were incorporated into the SWCMR that extends
from Tobacco Cay in the north to Cat Cay in the south.
The SWCMR is part of the Belize Barrier Reef Reserve
System, which was inscribed on the UNESCO World Heri-
tage List in 1996 (http://whc.unesco.org/). At that time,
it was recognized by the World Heritage Committee and
the GOB that, except for privately owned cays and those
with preexisting leases, the cays of the SWCMR would
be protected from development. In the Pelican Cays, such
an exclusion would apply to a small area (<1 ha) at the
southern tip of Northeast Cay. However, since 1996 most
of the mangrove cays within the SWCMR have been leased
for proposed resort developments. Based on our recent
aerial surveys in April 2008, in most of the islands in the
Pelican Cays archipelago, large sections of the mangrove
forests have been cut down and covered with dredged ma-
rine sediment from the adjacent seafloor. Runoff of the
fine fraction of the covering sediment has entered the in-
terior ponds and smothered both the prop root-based and
benthic seagrass communities. The extensive land clear-
ing and filling is apparently an attempt to convert these
mangrove islands into sandy cays in preparation for new
tourist resorts.

METHODS AND RESULTS
FIELD OBSERVATIONS, MARCH 2007

We visited Manatee Cay (16°39.97'N, 88°11.53'W)
in mid-March 2007 to conduct reconnaissance of seagrass
and prop root-based benthic communities along the perim-
eter of Pond C. At a point on the southeast side of Pond C
(see Figure 1), we encountered an area of dead mangrove
roots and a thick sediment drape covering the steep slope
into the center of the pond (see Figure 2C). The seagrass
communities lining the pond slope had been effectively
buried by the sediment drape. At the top of the slope in
this area, the ubiquitous submerged mangrove root epiben-
thos was conspicuously absent. At the surface, we noted
a fringe of dead red mangrove trees behind which was
recently cleared land covered in white marine sediment
containing numerous coral fragments and mollusk shells.

Manatee Cay

Dredged area

FIGURE 2. Marine communities of the Pelican Cay before (A, B) and after (C-F) dredging operations. A. Pre-dredging example of a rich
encrusting community on a mangrove root that is dominated by the purple ascidian Clavelina puertosecensis. B. Healthy sponge community
dominated by Mycale sp. and Scopalina sp. on a root before dredging. C. Spillover sediment surrounding bare dead mangrove roots in Pond
C, Manatee Cay, after the deposition of dredged lagoonal sediment on clear-cut mangrove peat, March 2007. D. Aerial photograph showing
the clear-cut sedimented areas of Fisherman’s and Manatee Cays. Note the dredge site in the lagoon at the right of Manatee Cay. E. Dredged
Thalassia—Porites bottom community in the lagoon east of Manatee Cay. F. Harmful toxic alga Dictyota caribaea overgrowing the seagrass
Thalassia testudinum at the bottom of Pond C, August 2007.

At that time, the dredging work on Manatee Cay had
been completed, but on Fisherman’s Cay (16°40.25’N,
88°11.40’W), it appeared to be under way. Here, we ob-
served similar clear-cutting and numerous large-diameter
plastic pipes laid across the cleared area (Figure 3). The
pipes are used to spread the sediment—water slurry over
the mangrove peat and stumps. A dredging vessel was an-
chored just offshore of Fisherman’s Cay (Figure 4).
Dredged marine sediment covered the exposed man-
grove peat substrate and created the illusion of a sandy
cay. However, the introduction of loose, water-laden sedi-
ment over peat resulted in runoff, which smothered and
killed the mangrove root and lagoonal bottom communi-
ties along the edges of the islands. Sediment collected on
the mud-covered slope in Pond C (see Figure 2C) was very
fine grained (rich in the clay-size fraction), which caused
turbidity when it washed off the island. Along the outer
edges of the island, runoff of the sediment slurry carried
these fine sediments into nearshore waters. In one case, it
smothered shallow-water head coral communities along
the windward side of Fisherman’s Cay and created a muddy

NUMBER 38 e¢ 285

plume that extended into the pass between Fisherman’s
and Manatee Cays. More detailed studies are needed to
establish the changes that have occurred in the pond com-
munities as a result of the dredging activity. A brief visit in
August 2007 indicated a marked increase in macroalgae,
most notably Dictyota caribaea, a bushy and toxic brown
alga (Littler et al., 2006; see Figure 2F). In addition, further
studies of the dredged lagoonal seafloor areas should under-
taken to assess the damage. A seagrass—Porites community
to the east of Manatee Cay had been dredged down to
2 to 3 m below the shallow surface (Figure 2E). We esti-
mate that such destruction will take decades to recover, as
indicated by the still bare but shallower seismic line de-
pressions in seagrass beds surveyed in the 1960s between
Carrie Bow Cay and Twin Cays.

AERIAL PHOTO SurRVEYS, APRIL 2007

Aerial photos taken in April 2007 indicated extensive
clearing of mangroves and dredge-spoil filling over the ex-
posed mangrove peat on Manatee and Fisherman’s Cays

FIGURE 3. Stumps of clear-cut mangroves being covered with lagoonal sediment transported by the black
pipes (arrow) in the background, March 2007.

286 °

FIGURE 4. Dredge boat operating off Fisherman’s Cay, March 2007.

(see Figure 2D). An area of disturbed, bare seabed (see
Figure 2E) was visible where the dredge removed the sedi-
ment near Manatee Cay. The time series of aerial photos
since 2003 indicated that Northeast Cay had been partially
cleared and had buildings on it before March 2003. Mana-
tee, Fisherman’s, Ridge, Bird, and Cat Cays had not yet been
cleared in March 2003, but clearing had begun on all but
one of the Bird Cays by April 2006. With the pumping of
dredge spoil over the exposed mangrove peat and truncated
roots, sediment spillover and runoff has resulted in exten-
sive nearshore turbidity, which was also visible from the air
along the edges of the ponds and the outer shorelines.

As of April 2007, approximately 15.4 ha (of 53.3 ha,
or 29% of the total) of mangrove forests in the Pelican Cays
had already been cleared, burned, and filled. Additional sur-
vey lines, which are typically the first evidence of develop-
ment activity, were visible on both Fisherman and Manatee
Cays in the aerial photographs taken in April 2007.

PHYTOPLANKTON SAMPLING ALONG
THE NorTH SIDE OF POND C

In May 2007, we conducted a preliminary survey at
Manatee Cay to determine the effect of mangrove clear-
ing and dredging in the Pelican Cays on the phytoplankton
populations dominated by dinoflagellates, which may form
red tides and visibly discolor the water (Morton and Vil-
lareal, 2001). Manatee Cay Pond C (see Figure 1) is large,
semi-enclosed, separated from open water by coral ridges
(Figures 1, 2D), and has distinct hydrographic, chemical,
and biological characteristics. With little water exchange

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

from the ocean side, the pond is warmer and more saline
than the surrounding waters. This environmental setting al-
lows microplankton, filter feeders, and corals to proliferate
(Villareal et al., 2000). We observed that dying mangrove
trees edged the pond and the water surface was highly tur-
bid. Dinoflagellates and associated microplankton were col-
lected in the center of Pond C via a vertical tow with a 20
zm pore size plankton net. Specimens were examined in the
laboratory with an Axiophot Carl Zeiss light microscope,
and dinoflagellate species were identified (Faust, 2000). The
water sample included a total of 14 species representing six
genera. Oceanic species included Ceratium (2), Protoperi-
dinium (5), and Pyrophacus (2). Coastal planktonic species
included Gymnodinium (3) and Peridinium (2). Benthic
species included Prorocentrum (3) (Table 1).

TABLE 1. Number of species in the dinoflagellate genera re-
corded in Manatee Cay, Pond C, Pelican Cays, collected in May
1996 and May 2007.

Number of species

Dinoflagellate genus 1996 2007

Amphidinium
Bepharocysta
Bysmatrum
Ceratium
Cochlodinium

a

Coolia
Corythodinium
Dinophysis
Diplopelta
Diplopsalis

Diplopsalopsis
Gambierdiscus
Gonyaulax
Gymnodinium
Heteraulacus

Lingolidinium
Ostreopsis
Peridiniella
Peridinium
Phaeopolykrikos

Plagodinium
Prorocentrum
Protoceratium
Protoperidinium
Pyrodinium

pA
PNM NOANMRPRP PWR HRP RPWDHWR WRWRHE FP ORF Bb
|

Pyrophacus
Scrippsiella
Zygabikonidium sp.

N
n
aN
N

Total genera: 28

Live dinoflagellate cells were fewer than expected.
Table 1 provides a comparison of the biodiversity and
species associations of dinoflagellate assemblages before
(May 1996) and after (May 2007) the clearing. In 1996,
dinoflagellates included 28 genera and 83 species. In con-
trast, in 2007 only 6 genera and 14 species were present,
and all were reduced in numbers of individuals present.

DISCUSSION

Despite repeated demonstration of their ecologi-
cal and economic importance, mangroves are one of the
world’s most threatened ecosystems (Valiela et al., 2001;
Alongi, 2002; Rivera-Monroy et al., 2004). Overall, 50%
of the world’s mangrove forests have been lost in the past
50 years, with at least 35% lost in just the past two de-
cades (Valiela et al., 2001). Duke et al. (2007) predicted
the current rate of loss would lead to mangrove extinction
in 100 years. Loss of mangroves is occurring faster in some
areas. For example, 29% of Guatemala’s mangroves were
lost in just 6 years between 1992 and 1998 (Abt Associ-
ates Inc., 2003). Most of that loss is directly attributed to
unfettered clear-cutting for shrimp farm aquaculture, ag-
riculture, mining, and development (Alongi, 2002). Based
on growing evidence from around the world, the clearing
and filling of mangrove forests for waterfront property to
meet the demands of the leisure and tourism market for
seaside resorts and retirement homes are also contributing
significantly to this loss (Ellison and Farnsworth, 1996;
Curran and Cruz, 2002; Barbier and Cox, 2003; Naylor et
al., 2002; Choong, 2005). In the Gulf of Honduras along
the Caribbean coast of Central America, mangroves have
been destroyed to make way for hotels and other tourism
infrastructure (CZAI, 2000). In these low-lying areas, such
development requires fill material that is dredged from the
seabed of nearby subtidal habitats. This action not only
destroys corals and seagrass directly but also causes the
suspension of sediments (turbidity), reduces light penetra-
tion, smothers seagrass and corals, increases nutrient levels,
and releases contaminants (Rambell Consulting Engineers,
2000). In addition, the inadequately disposed solid waste
and untreated sewage associated with this coastal devel-
opment enter waterways and increase unwanted nutri-
ents, thus decreasing water quality. Similar building efforts
have been observed all along the Belize mainland coastline
where many acres of mangroves have been cut down to
make room for numerous large private homes and resorts.
In addition, many acres of mangroves have been removed
to create extensive areas of shrimp ponds.

NUMBER 38 ¢ 287

Although tourism is the second largest foreign ex-
change earner for the countries of Belize, Honduras, and
Guatemala in the region, this type of development is coun-
terproductive because loss of mangroves leads to a reduc-
tion in income from tourism and fisheries, changes in em-
ployment, loss of aesthetic value, loss of cultural heritage,
conflicts between user groups, and loss of recreational op-
portunities (Abt Associates Inc., 2003).

Although mangrove forests are apparently considered
of little value, recent studies have demonstrated the vital
role of mangroves as nursery habitat for several species of
reef fish (Mumby et al., 2004; Mumby, 2006). As stated
by Mumby et al. (2004; specifically concerning Belize
and Mexico), “Current rates of mangrove deforestation
are likely to have severe deleterious consequences for the
ecosystem function, fisheries productivity and resilience of
coral reefs.” In particular, the availability of mangroves
for fish nursery habitat (intermediate between seagrass
beds and coral reefs) is highly correlated to the numbers
of reproducing adults and even the continued existence
of certain species. According to these authors, parrotfish,
which are important herbivores on reefs, have become lo-
cally extinct as a consequence of mangrove removal. Com-
mercial species biomass has been effectively halved in ar-
eas of mangrove removal. Thus, the health of coral reefs
and of fisheries, both essential for both tourism and local
livelihood, are deleteriously affected by mangrove loss.

Despite the legislated restrictions on leasing govern-
ment-owned lands within the Belize Barrier Reef World
Heritage Site and the SWCMR, most of these cays have
been leased or sold to foreign developers since 1996 by
following standard procedures. These procedures involve
locating a suitable site and having it surveyed; this requires
a permit from the Lands & Surveys Office in Belmopan,
Belize, which has apparently ignored the legal protection
status afforded areas in marine protected areas. Separate
permits from government departments are required to
clear the leased areas delineated by survey lines. After the
areas have been cleared, a developer must obtain another
permit from the government to dredge material from the
adjacent seabed to fill the leased areas.

The areas of mangrove clearing and filling with la-
goonal sediments on Manatee and Fisherman’s Cays (and
other islands in the group) are extensive, exposing most of
the available island surfaces. As such, sufficiently large areas
have been cleared on these cays to account for potentially
extensive development of seaside resorts. Disturbingly, the
thin veneer of sediment laid over mangrove peat, especially
where the mangroves themselves have been cleared, will
not prevent significant subsidence as the underlying peat

288 °

decomposes and compacts. The substrate will then further
subside because of the pressure of any load placed upon
it, pilings notwithstanding, which will prove to be a sig-
nificant long-term problem for construction on mangrove
substrate.

In addition to these obvious problems, construction
on and habitation of these islands will ensure perpetual
pollution of the ponds from continued sediment runoff
combined with the eventual addition of sewage outflow
and solid waste. Turbidity in the water column from
runoff of dredge spoil will continue to deleteriously af-
fect marine communities adjacent to the affected islands.
Along the outer shorelines of Manatee and Fisherman’s
Cays, numerous nearshore coral heads and patch reefs will
eventually be smothered by the sediment load noted along
the shorelines and in the passes between the Cays. In the
Ponds, the rich benthic communities inhabiting the ridges
upon which the islands are built and in the seagrass beds
lining the slopes of the ponds have been locally decimated
by direct sediment runoff. The water column in Pond C is
now generally turbid, and it is anticipated that further sed-
iment pumping, runoff, construction waste, and eventu-
ally untreated sewage outfall will further impact the pond
habitats and cause eutrophication of the water column,
which could lead to further losses of species, particularly
photosynthetic organisms.

The Pelican Cays, although small in geographic scale,
are characterized by great topological diversity in coral
reef-mangrove habitats. Biological communities within
this system vary markedly from one pond to another. Be-
cause of this complexity, some important details about the
associations of dinoflagellate species in this ecologically
diverse environment have come to light as a result of long-
term studies in Pelican Cays (Faust, 2000). Dinoflagellates
and microalgae are the primary food source for zooplank-
ton, their primary consumer, including filter feeders and
juvenile fish (Frenchel, 1988). Dinoflagellates and zoo-
plankton proliferate in response to their unique physical,
chemical, and biological needs (Villareal et al., 2000).

Species associations of dinoflagellates is another impor-
tant indicator of certain stability in mangrove communities
that are constantly threatened. Studies targeting processes in
the Caribbean have examined benthic and epiphytic dinofla-
gellates in the coral reefs of the Virgin Islands (Tindall and
Morton, 1998). Mangrove detritus, a unique microcosm,
maintains a reservoir of diverse microalgae and meiofauna at
Twin Cays, Belize (Faust, 1996). Most species tend to show
preference for one habitat, either on sessile macroalgae or
free-floating in the water column (Faust, 2004), although
some species are found in a wide range of habitats.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Microbial communities can be damaged and species
driven to local extinction by external factors; however,
the damage is not immediately apparent to the human
eye. Recent field observations of microscopic microalgae
and zooplankton signaled significantly altered dinoflag-
ellate populations, dead cells, and a greatly changed mi-
croscopic food web in Manatee Cay Pond C (see Table
1). This is yet another example of the continuing trend
in the Belizean coral reef-mangrove ecosystem observed
over 25 years (beginning in 1982) indicating declining
abundances of dinoflagellate and zooplankton in the
microbial food web caused by human activities (Faust,
2004). This finding in itself has important implications
for the ecology and economy of the Belizean Barrier
Reef, in that dinoflagellate populations are the primary
food source of zooplankton, including fish larvae and
juvenile fish.

CONCLUSIONS

Despite the location of Pelican Cays within the
SWCMR and the World Heritage Site, development has
been accomplished by following a sequence of procedures
involving several separate jurisdictions. Investigations be-
gan recently into the process of mangrove cutting, clear-
ing, and filling in the Pelican Cays, and it appears that
some of this activity was illegal (Melanie McField, Smith-
sonian Marine Station at Fort Pierce, personal communi-
cation, April 2008).

It is highly questionable that the proposed highly vul-
nerable tourist resort on the Pelican Cays will survive the
subsidence related to rotting peat or storms. Indeed, these
sea-level structures will be readily destroyed by severe
storms, leaving abandoned communities both on land and
in the sea in an area originally noted for its unique and
unusually high biological diversity.

The future of the unique ecology of the mangrove
and seagrass communities in the Pelican Cays appears to
be very bleak. The dredging barge is no longer operating
off Fisherman’s Cay, and a Caribbean Island Brokers web-
site is now offering 37 acres of cleared mangrove on this
cay for US $1,750,000. Given this situation, all mangrove
islands in this area will be cleared and developed, so that
the ponds adjacent to cleared areas will likely suffer the same
fate as Pond C of Manatee Cay, as will nearshore marine
communities along the outer perimeters of these islands.
Lack of foresight, which is disrupting the connectivity be-
tween mangroves and the health of the nearshore marine
realm, will result in economic losses following the reduc-

tion of commercial and recreational fisheries that rely on
mangroves. Much of this collapse is related to the dramatic
loss of dinoflagellate assemblages, which provide the base
of food webs supporting fisheries in Belize.

Tourism losses will subsequently occur as coral reefs
decline without mangroves to support the mangrove-
dependent fish species essential to reef herbivory and
commerce. Thus, the short-term economic gains from
construction will lead to long-term environmental disrup-
tion, ecological degradation, local species extinction, and
the consequent economic collapse of the tourism and fish-
ing industries all along the Belizean coast and similarly
affected areas of Mesoamerican reefs. The government of
Belize has instituted a mangrove clearing moratorium to
evaluate the situation.

ACKNOWLEDGMENTS

We thank Gary Peresta for field assistance; LightHawk
(Michele Gangaware) for providing the plane and pilot for
aerial surveys; Molly Ryan of the Smithsonian National
Museum of Natural History (NMNH) for preparing the
index map; Chip Clark of the Smithsonian NMNH for
photographs (Figure 2A,B); and finally Klaus Riutzler, Janet
Gibson, Isaias Majil, Melanie McField, and John Tschirky
for providing advice and information for this manuscript.
This is contribution number 831 of the Caribbean Coral
Reef Ecosystems Program (CCRE), Smithsonian Institu-
tion, supported in part by the Hunterdon Oceanographic
Research Fund.

LITERATURE CITED

Abt Associates Inc./Woods Hole Group. 2003. Gulf of Honduras
Preliminary Transboundary Diagnostic Analysis. Report for the
Global Environment Facility and the Inter-American Develop-
ment Bank Project Development Facility.

Alongi, D. M. 2002. Present State and Future of the World’s Mangrove
Forests. Environmental Conservation, 29:331-349.

Barbier, E. B., and M. Cox. 2003. Does Economic Development Lead
to Mangrove Loss? A Cross-Country Analysis. Contemporary Eco-
nomic Policy, 21:418-432.

Choong, J. 2005. Protective Values of Mangrove and Coral Ecosystems:
A Review of Methods and Evidence. Gland, Switzerland: IUCN
(The World Conservation Union).

Coastal Zone Management Authority and Institute (CZAI). 2000. State
of the Coast Report Belize, 1999. Belize City, Belize: CZAI.

Curran, S. R., and M. C. Cruz. 2002. Markets, Population Dynamics,
and Coastal Ecosystems. Ambio, 31:373-376.

Duke, N. C., J. O. Meynecke, S. Dittmann, A. M. Ellison, K. Anger, U.
Berger, S. Cannicci, K. Diele, K. C. Ewel, C. D. Field, N. Koedam,
S. Y. Lee, C. Marchand, I. Nordhaus, and FE. Dahdouh-Guebas.
2007. A World Without Mangroves. Science, 317:41—42.

NUMBER 38 e¢ 289

Ellison, A. E., and E. J. Farnsworth. 1996. Spatial and Temporal Vari-
ability in Growth of Rhizophora mangle Saplings on Coral Cays:
Links with Variation in Insolation, Herbivory, and Local Sedimen-
tation Rate. Journal of Ecology, 84:717-731.

Faust, M. A. 1996. Dinoflagellates in Mangrove Ecosystems, Twin Cays,
Belize. Nova Hedwigia, 112:447-460.

. 2000. “Dinoflagellate Associations in a Coral-Reef Mangrove

Ecosystem: Pelican and Associated Cays, Belize.” In Natural His-

tory of the Pelican Cays, Belize, ed. I. G. Macintyre and K. Ritzler,

pp. 133-150. Atoll Research Bulletin, No. 473.

. 2004. “The Dinoflagellates of Twin Cays: Biodiversity, Distri-
bution, and Vulnerability.” In The Twin Cays Mangrove Ecosys-
tem, Belize: Biodiversity, Geological History, and Two Decades of
Change, ed. K. Riitzler, I. C. Feller, and I. G. Macintyre, pp. 1-20.
Atoll Research Bulletin, No. 515.

Frenchel, T. 1988. Marine Plankton Food Chain. Annual Review of
Ecology and Systematics 18:19-38.

Goodbody, I. 2000. “Diversity and Distribution of Ascidians (Tunicata)
in the Pelican Cays, Belize.” In Natural History of the Pelican Cays,
Belize, ed. I. G. Macintyre and K. Riitzler, pp. 303-326. Atoll Re-
search Bulletin, No. 480.

Hendler, G., and D. L. Pawson. 2000. “Echinoderms of the Rhomboidal
Cays, Belize: Biodiversity, Distribution and Ecology.” In Natural
History of the Pelican Cays, Belize, ed. 1. G. Macintyre and K. Rutz-
ler, pp. 275-302. Atoll Research Bulletin, No. 479.

Littler, D.S., M. M. Littler, and B. L. Brooks. 2000. “Checklist of Marine
Algae and Seagrasses from the Ponds of the Pelican Cays, Belize.”
In Natural History of the Pelican Cays, Belize, ed. I. G. Macintyre
and K. Ritzler, pp. 153-208. Atoll Research Bulletin, No. 474.

Littler, M. M., D. S. Littler, and B. L. Brooks. 2006. Harmful Algae on
Tropical Coral Reefs: Bottom-up Eutrophication and Top-down
Herbivory. Harmful Algae, 5:565-585.

Macintyre, I. G., I. Goodbody, K. Riitzler, D. S. Littler, and M. M. Littler.
2000a. “A General Biological and Geological Survey of the Rims of
Ponds in the Major Mangrove Islands of the Pelican Cays, Belize.”
In Natural History of the Pelican Cays, Belize, ed. 1. G. Macintyre
and K. Riitzler, pp. 14-34. Atoll Research Bulletin, No. 467.

Macintyre, I. G., M. M. Littler, and D. S. Littler. 1995. “Holocene His-
tory of Tobacco Range, Belize, Central America.” In Biology and
Geology of Tobacco Range, Belize, Central America, ed. M. M.
Littler, I. G. Macintyre, and D. S. Littler, pp. 1-18. Atoll Research
Bulletin, No. 430.

Macintyre, I. G., W. F. Precht, and R. B. Aronson. 2000b. “Origin of
the Pelican Cays Ponds, Belize.” In Natural History of the Pelican
Cays, Belize, ed. I. G. Macintyre and K. Riutzler, pp. 1-12. Atoll
Research Bulletin, No. 466.

Macintyre, I. G., and K. Riitzler, eds. 2000. Natural History of the Peli-
can Cays, Belize. Atoll Research Bulletin, Nos. 466-480.

Macintyre, I. G., K. Riitzler, and I. C. Feller, eds. 2004. The Twin Cays
Mangrove Ecosystem, Belize: Biodiversity, Geological History, and
Two Decades of Change. Atoll Research Bulletin, Nos. 509-530.

Morton, S. L., and T. A. Villareal. 2001. Bloom of Gonyaulx poly-
gramma Stein (Dinophyceae) in a Coral Reef Mangrove Lagoon,
Douglas Cay, Belize. Bulletin of Marine Science, 63:1-4.

Mumby, P. J. 2006. Connectivity of Reef Fish Between Mangroves and
Coral Reefs: Algorithms for the Design of Marine Reserves at Sea-
scape Scales. Biological Conservation, 128:215-222.

Mumby, P. J., A. J. Edwards, J. E. Arias-Gonzalez, K. C. Lindeman, P. G.
Blackwell, A. Gall, M. I. Gorezynska, A. R. Harbone, C. L. Pescod,
H. Renken, C. C. C. Wabnitz, and G. Llewellyn. 2004. Mangroves
Enhance the Biomass of Coral Reef Fish Communities in the Carib-
bean. Nature (London), 427:533-536.

Naylor, R. L., K. M. Bonine, K. C. Ewel, and E. E. Waguk. 2002. Migra-
tion, Markets, and Mangrove Resource Use on Kosrae, Federated
States of Micronesia. Ambio, 31:340-350.

290 e

Ramboll Consulting Engineers. 2000. Environmental Efficiency of Mari-
time Transport Operations in the Gulf of Honduras. Final Report.
Funded by Inter-American Development Bank (IADB) Denmark.

Richardson, S. L. 2000. “Epiphytic Foraminifera of the Pelican Cays,
Belize: Diversity and Distribution.” In Natural History of the Peli-
can Cays, Belize, ed. J. G. Macintyre and K. Ritzler, pp. 209-230.
Atoll Research Bulletin, No. 467.

Rivera-Monroy, V. H., R. R. Twilley, D. Bone, D. L. Childers, C. Coro-
nado-Molina, I. C. Feller, J. Herrera-Silveira, R. Jaffe, E. Mancera,
E. Rejmankova, J. E. Salisbury, and E. Weil. 2004. A Biogeochemi-
cal Conceptual Framework to Develop Long Term Ecological Re-
search and Sustain Coastal Management in the Wider Caribbean
Region. BioScience, 54:843-856.

Riutzler, K, M. C. Diaz, R. W. M. Van Soest, S. Zea, K. P. Smith, B. Alva-
rez, and J. Wulff. 2000. “Diversity of Sponge Fauna in Mangrove
Ponds, Pelican Cays, Belize.” In Natural History of the Pelican
Cays, Belize, ed. 1. G. Macintyre and K. Ritzler, pp. 231-250. Atoll
Research Bulletin, No. 476.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Tindall, D.R., and S. L. Morton. 1998. “Community Dynamics and
Physiology of Epiphytic/Benthic Dinoflagellates Associated with
Ciguatera.” In Physiological Ecology of Harmful Algae Blooms,
ed. D. M.Anderson, A. D. Cembella, and G. M. Hallegraeff, pp. 293-
313. Berlin: Springer-Verlag.

Valiela, I., J. L. Bowen, and J. K. York. 2001. Mangrove Forests: One of
the World’s Threatened Major Tropical Environments. BioScience,
51:807-815.

Villareal, T. A., S. L. Morton, and G. B. Gardner. 2000. “Hydrography
of a Semi-enclosed Mangrove Pond, Manatee Cay, Belize.” In Natu-
ral History of the Pelican Cays, Belize, ed. I. G. Macintyre and K.
Ritzler, pp. 86-103. Atoll Research Bulletin, No. 470.

Winston, J. E. 2007. Diversity and Distribution of Bryozoans in the Pelican
Cays, Belize, Central America. Atoll Research Bulletin, No. 546:1-24.

Wulff, J. 2000. “Sponge Predators May Determine Differences in Sponge
Fauna Between Two Sets of Mangrove Cays, Belize Barrier Reef.”
In Natural History of the Pelican Cays, Belize, ed. I. G. Macintyre
and K. Ritzler, pp. 251-266. Atoll Research Bulletin, No. 467.

Using the Panama Canal to Test Predictions
about Tropical Marine Invasions

Gregory M. Ruiz, Mark E. Torchin,
and Katharine Grant

Gregory M. Ruiz and Katharine Grant, Smithson-
ian Environmental Research Center, 647 Contees
Wharf Road, Edgewater, Maryland 21037, USA.
Mark E. Torchin, Smithsonian Tropical Research
Institute, Apartado 0843-03092, Balboa, Ancon,
Panama, Republic of Panama. Corresponding au-
thor: G. Ruiz (ruizg@si.edu). Manuscript received
29 August 2008; accepted 20 April 2009.

ABSTRACT. As humans alter the landscape of the Earth and economic globalization
expands, biological invasions increasingly homogenize the world’s biota. In temper-
ate marine systems, invasions are occurring at a rapid pace, driven by the transfer of
organisms by vessels and live trade (including aquaculture and fisheries activities). In
contrast, little is known about patterns and processes of tropical marine invasions,
although the same species transfer mechanisms are in operation. This disparity may
be the result of limited studies of invasions in the tropics relative to temperate regions.
Alternatively, the tropics may be less susceptible to invasion than temperate regions for
reasons of environmental unsuitability and biotic interactions. This paper provides a
brief summary of the current but limited information of marine invasions across lati-
tudes, focusing particular attention on the eastern Pacific north of the Equator. Within
this latitudinal framework, the Panama Canal provides an especially important model
system for testing predictions about marine invasions in the tropics for reasons of (a)
the high level of shipping traffic since the Canal opened in 1914; (b) the permeability
of the Canal as a conduit for marine invaders, despite the apparent freshwater barrier;
and (c) the current expansion of the Canal that is expected to increase the size and
number of ships visiting the region.

INTRODUCTION

Biological invasions are common in coastal marine ecosystems around the
world (Cohen and Carlton, 1995; Orensanz et al., 2002; Fofonoff et al., 2008).
In fact, reports of new invasions are increasing exponentially in many well-
studied regions (Cohen and Carlton, 1998; Ruiz et al., 2000; Hewitt et al.,
2004). Although invasions can result from natural dispersal, most contempo-
rary invasions derive from human-mediated transfer associated with a variety
of activities. As economic globalization continues to expand, creating a high
degree of connectivity through the movement of commodities and people, op-
portunities for new invasions also increase. Bays and estuaries have been the
most invaded marine systems, probably because they are hubs for shipping,
aquaculture, and other human endeavors known to transfer organisms (Ruiz et

al., 1997; Wasson et al., 2005).

ZA)7} O

To date, most human-mediated invasions (hereafter
introduced species) in marine habitats have been reported
in temperate latitudes (Ruiz and Hewitt, 2008, and refer-
ences therein). Relatively few introduced species have been
reported from tropical or polar regions. This difference
across latitudes may result partly from historical research
effort and taxonomic knowledge, which are greatest in the
temperate zone. However, a small but growing literature
for high latitudes suggests that marine invasions may be
limited in polar regions by a combination of current low
temperatures and low propagule supply (Barnes et al.,
2006; Aronson et al., 2007; Ruiz and Hewitt, 2008).

It is evident that marine invasions can occur in tropical
marine systems (Agard et al., 1992; Guerrero and Franco,
2008), but the extent to which they occur remains largely
unexplored. Few studies have evaluated marine invasions
in the tropics. The exceptions are extensive analyses of
introduced species on the Hawaiian Islands and Guam
(Eldredge and Carlton, 2002; Paulay et al., 2002). It is un-
certain whether these island ecosystems are broadly repre-
sentative of the tropics, including especially mainland sites
that may differ from islands in susceptibility to invasion
(Elton, 1958; MacArthur and Wilson, 1967; Sax, 2001).

In a preliminary analysis of marine invasion pat-
terns for mainland Australia, Hewitt (2002) reported an
increase in introduced species richness with increasing
latitude. The study included four tropical and four tem-
perate sites, spanning 13°-38°S latitude. Despite a signifi-
cant relationship with latitude, there is uncertainty about
the taxonomic identification and biogeographic origin of
many tropical species, resulting from limited information
and relative lack of study for low latitude biotas. For this
reason, Hewitt urges some caution and underscores the
need for further analyses to interpret the observed pattern.
It is nonetheless intriguing that this preliminary analy-
sis provides results similar to those reported for tropical
terrestrial systems, where relatively few exotic species of
birds, mammals, and plants are established (Sax, 2001).

We have begun to explore latitudinal patterns of
marine invasions for the mainland (continental) habitats
within the Americas. To date, most of our analyses have
focused on bays and estuaries within the United States,
particularly on the Pacific Coast. We are currently initi-
ating a research program to compare the number of in-
troduced species, scale of vector operations (propagule
supply), and ecology of invasions across temperate and
tropical latitudes. Here, we briefly review the current state
of knowledge about invasions and invasion processes
along the Pacific Coast of the Central and North Amer-
ica and discuss the potential significance of Panama as a

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

model system to evaluate regional and latitudinal patterns
of marine invasion.

LATITUDINAL PATTERN OF INVASIONS
ALONG THE NORTHEASTERN PACIFIC

Outside of the tropics, there is a clear increase in the
number of nonnative species reported with decreasing
latitude, from Alaska to California, 61°-32°N (Ruiz et
al., 2006a). An extensive review and synthesis of the lit-
erature indicate that more than 250 nonnative species of
invertebrates and algae are established in coastal waters
of California (NEMESIS, 2008). Most of these invasions
are attributed to commercial shipping and live ship-
ments of organisms, especially oysters and their associ-
ated biota (Cohen and Carlton, 1995; Miller, 2000; Ruiz
et al., unpublished data). Some of the California inva-
sions have spread northward through natural dispersal,
and other species have been introduced independently to
the north. However, compared to California, far fewer
nonnative species are known from Oregon, Washington,
and Alaska (Cohen et al., 1998; Wonham and Carlton,
2005; Ruiz et al., 2006a).

Although this latitudinal pattern of invasion could re-
sult from reporting biases in the literature, particularly in
the level of research (search effort) among regions, recent
surveys suggest that the pattern is robust for sessile inver-
tebrates in hard substrate fouling communities. Using stan-
dardized surveys to sample sessile invertebrates, deRivera et
al. (2005) and Ruiz et al. (2006a) found that the number
of introduced species increased with decreasing latitude
from Alaska to southern California. It appears that the
northern spread of many nonnative species from Califor-
nia may have been limited by dispersal as a result of the
relatively low level of human activities (and, thus, species
transfer opportunities) that have been present historically
(Ruiz and Hewitt, 2008).

Similar analyses are not yet available to extend this
comparison to lower latitudes along the eastern Pacific.
Although there have been some studies reporting intro-
duced marine species in Central America (Rubinoff and
Rubinoff, 1969; Lambert and Lambert, 2003; Wysor,
2004, Roche and Torchin, 2007; Roche et al., 2009; Bas-
tida-Zavala, 2008), standardized, quantitative commu-
nity-level comparisons are lacking. In particular, synthetic
studies focused within bays and estuaries of Central Amer-
ica targeting those taxonomic groups for which invasions
are often most prevalent do not exist. Even where synthe-
ses from the literature have been attempted, the paucity

of available data limits conclusions about the scope of
invasions. For example, Cohen (2006) provides a useful
summary of available information on invasions surround-
ing the Panama Canal, which has received considerable
attention for a tropical system. Despite the historical in-
terest on biotic exchange in Panama, Cohen characterizes
the current state of knowledge as follows: “The Panama
Canal lies in a region of the world where the marine biota
is both diverse and relatively poorly known, and there has
been remarkably little investigation of the effect that the
Canal has had on the distribution of that biota.”

With a broad goal to evaluate patterns and processes
in marine invasions using a latitudinal framework, we have
initiated a research program in Central America (a) to com-
pile available data from the literature on nonnative marine
species, as part of our database (NEMESIS, 2008), and
(b) to conduct standardized surveys at multiple sites. Our
approach will allow direct comparisons with more than
two dozen sites surveyed on the Pacific and Atlantic coasts
of the USA. Our initial effort is focused primarily on ses-
sile invertebrates (including ascidians, barnacles, bryozo-
ans, hydroids, mussels, and sponges), which comprise a
large proportion of marine introductions, are relatively
well studied, and are conducive to standardized, quantita-
tive field surveys.

Alaska A

Oregon and
Washington

California

Panama

0 1 2 3 4 5

Number of Non-Native Barnacle Species

Oregon and
Washington

NUMBER 38 e¢ 293

A preliminary review of the literature for barnacles
suggests the number of introduced species increases from
Alaska to Panama (Figure 1A), consistent with an increase
in the magnitude of shipping (see next section). At least
four nonnative species of barnacles are reported to occur
on the Pacific coast of Panama, including Amphibalanus
amphitrite, A. reticulatus, Balanus trigonus, and Fistulo-
balanus pallidus (Matsui et al., 1964; Jones and Dawson,
1973; McCosker and Dawson, 1975; Laguna, 1985).
Three introduced barnacles are known from California:
Amphibalanus amphitrite, A. eburneus, and A. improvi-
sus (Carlton, 1979; Carlton and Zullo, 1969; Cohen and
Carlton, 1995; Cohen et al., 2002). Amphibalanus reticu-
latus has also been detected in recent surveys in south-
ern California, but it is not yet known to be established
(Ruiz, unpublished data). Only one introduced barnacle,
A. improvisus, is reported in Oregon and Washington
(Carlton, 1979; Wonham and Carlton, 2005), and there
are no introduced barnacles known from Alaska (Ruiz et
al., 2006a). It is noteworthy that the reported number of
nonnative barnacle species in Panama exceeds that along
the western USA, considering the latter is relatively well
surveyed. Thus, we expect that strength of this inverse
relationship with latitude may increase with additional
information.

Alaska | B

California

Foe pe Le AL
0 10000 20000 30000
Number of Ship Arrivals (2004 - 2005)

FIGURE 1. A. Number of nonnative barnacle species established by geographic region. Shown are the numbers of nonnative barnacle species
reported to be established from Alaska to Panama (see text). B. Number of vessel arrivals by geographic region. Shown are the numbers of com-
mercial vessel arrivals from overseas to different geographic regions, from Alaska to Panama, over a two-year period (2004-2005). Coastwise
domestic traffic is excluded from arrivals to U.S. locations. (Data from Miller et al., 2007; ACP, 2008b.)

294 e

At the present time, the relationship between intro-
duced species richness and latitude is poorly resolved for
the northeastern Pacific and other global regions. The pat-
tern presented in Figure 1A should be considered as pre-
liminary, and it may change with further research. We also
caution that these data are restricted to barnacles, a very
small subset of species present in the fouling community.

PANAMA: A TEST CASE FOR
TROPICAL MARINE INVASIONS

Panama is a potential hotspot for tropical marine in-
vasions, because of the country’s historic significance as
a hub of world trade since the fifteenth century, expand-
ing greatly since construction of the Panama Canal. The
Canal created a new shipping route between the Atlantic
and Pacific basins, resulting in a large influx of commercial
ships, which have been an important source of introduced
species in North America (Cohen and Carlton, 1995; Co-
hen et al., 1998, 2002; Ruiz et al., 2000; Wonham and
Carlton, 2005; see discussion below). Figure 1B compares
the magnitude of commercial shipping to several major
port systems, indicating that ship arrivals to Panama ex-
ceed those to major port systems in the western United
States by a large margin. Over the two-year period 2004-
2005, nearly twice as many vessels arrived to Panama as
overseas vessels arrived to California. In fact, Panama re-
ceives more ship arrivals than any of the largest ports in
the United States (Ruiz et al., 2006b; Miller et al., 2007).

Since its opening in 1914, the number of Canal tran-
sits increased rapidly, with the exception of a brief inter-
ruption during WW II, until reaching capacity in 1970
(ACP, 2008a; Figure 2). Currently, the Canal is operat-
ing at 90% of its theoretical maximum capacity, servic-
ing 12,000 to 14,000 vessels and carrying approximately
5% of the world’s cargo annually (Reagan, 2007). More
than 800,000 ocean-going commercial vessels have
passed through the Canal since its completion (Ruiz et
al., 2006b).

While the number of transits has leveled off, the av-
erage size of ships transiting the Canal has continued to
increase, allowing for a continued increase in the volume
of cargo passing through the Canal (ACP, 2008a; see Fig-
ure 2). The average tonnage (based on CPSUAB, a uni-
versal system of tonnage for the Panama Canal, or Canal
ton, which is equivalent to approximately 100 cubic feet
of cargo) per transit has increased from 4,832 in 1955 to
21,963 in 2005 (ACP, 2008a). This change in cargo ca-
pacity reflects an increase in the size of vessels over time;

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

50 300
45 S
S 40 oat)
[—]
—) °
% 35 =
= 200 %
2 30
c <x
§ 25 150 3
-_
© 20 o
i=
Boop 100
E D
10 %
“= 50mce
5 2
0 0
moonMmnonaonaonwnaonwnaononwnaownndeow
TCT NN OMT TWHONOOORKR DODD CO
PEDO DEON COO OREM ORO ORO
Year

a a EET SE TE EO a |
FIGURE 2. Number of commercial vessel transits (black line)
through the Panama Canal and associated cargo tonnage (gray line);
CPSUAB is a universal system of tonnage for the Panama Canal, or
Canal ton, which is equivalent to approximately 100 cubic feet of
cargo. (Figure modified from ACP, 2008a.)

these changes are the topic of a future analysis that will
characterize changes both in vessel size and in under-
water surface area available for colonization by organ-
isms. In recent years, however, the size of vessels has been
constrained by the lock dimensions and has been relatively
static, with the Panamax ships designed specifically as the
largest vessels able to transit the locks (see next section).

Likely consequences of the Panama Canal and ports
located at both entrances are an increase in (a) the global
transfer of marine organisms, as the canal provides a con-
duit for worldwide shipping, and (b) regional biological
invasions in Central America. Commercial shipping is a
major pathway for the movement of species and appears
largely responsible for a dramatic increase in the rate of
known invasions for many regions in recent time (Ruiz
et al., 2000; Fofonoff et al., 2003; Hewitt et al., 2004).
Ships move organisms associated primarily with hull
and sea chest fouling and with ballasted materials, as an
unintended result of normal operations (Carlton, 1985;
Minchin and Gollasch, 2003). In general, the likelihood
of invasions increases with increasing propagule supply,
including the magnitude and frequency of organisms de-
livered (Ruiz and Carlton, 2003; Lockwood et al., 2005).
Thus, the chance of colonization by introduced species
in Panama is likely to have increased over time with the
high frequency of vessels arriving to Panama from around
the globe.

Given the high number of vessel arrivals, we might
also expect the relative magnitude of propagule sup-
ply and invasions to be high in Panama. However, this
remains to be tested, and there are several reasons why
this may not be the case. First, different ship types and
operational behaviors vary in their potential to transfer
marine organisms (Verling et al., 2005; Miller et al., 2007;
NBIC, 2008). Second, independent of propagule supply,
some sites are less susceptible to invasion for reasons of
either environmental conditions or biological interactions
(Lonsdale, 1999; Ruiz et al., 2000; Roche et al., 2009).

Past studies have certainly highlighted the potential
significance of vessels as a source of invasions to the Pan-
ama Canal and surrounding waters (see Cohen, 2006,
and references therein for recent review). For example,
Chesher (1968) discusses the potential importance of
ballast water. Menzies (1968) considers the capacity of
vessels to transfer fouling organisms. Hay and Gaines
(1984) suggest that small pleasure boats may be espe-
cially important in the transfer or organisms across the
Isthmus of Panama. A few studies also test the capacity
of marine organisms to survive freshwater exposure for
the duration of a transit through the Canal (Chesher,
1968; Hay and Gaines, 1984). Despite the long inter-
est and recognition in ship-mediated transfer, the esti-
mates given above are limited to few (if any) data on
species composition or direct quantitative estimates of
propagule supply (abundance) on vessels. Surprisingly
few data exist on biota associated with ballast water
or hulls of vessels associated with the Canal. Instead,
there are only coarse data available on general opera-
tional aspects of vessels that may affect species transport
opportunities.

Most commercial ships arriving to Panama will tran-
sit the Canal, but some will have considerable time at an-
chorage before entering the Canal. From 2000 to 2005,

NUMBER 38 ¢* 295

the average service time (from arrival to complete transit)
of ships passing through the Canal was 16 hours when
holding reservations. However, many ships have not had
reservations, and average service times for these ships can
reach 57 hours (Table 1). Although the proportion of ships
holding reservations has increased in recent years, half of
all ships still experienced some delay. Such increased resi-
dence time is likely to also increase the opportunity for re-
production and colonization of organisms associated with
ships’ hulls (Minchin and Gollasch, 2003; Davidson et
al., 2008), relative to shorter residence times. It is evident
that some organisms arrive to Panama on the hulls of ves-
sels (Figure 3). However, a lack of quantitative informa-
tion on the biota associated with outer surfaces of vessels
transiting the Panama Canal and surrounding ports limits
any detailed analyses.

For ballast water, we are not aware of any reliable
estimates of the historical patterns of ballast water man-
agement and discharge of vessels arriving to Panama, in-
cluding those ships delivering cargo to the terminals and
those simply transiting the Canal. Even a coarse estimate
of volume is challenging, given large differences in opera-
tions among vessels (Verling et al., 2005; but see Chesher,
1968). Presumably, ballast water discharge today is rather
limited because many vessels conduct ballast operations to
compensate for loading or off-loading cargo. In addition,
Panama prohibits ballasting operations in the Canal under
most circumstances (ACP, 2008b).

Despite the limited information available, we surmise
that propagule supply has been relatively high in Panama,
compared to many other temperate and tropical sites. Based
solely on the large number of vessel arrivals and their rela-
tively long residence times (see Figure 2, Table 1), it is likely
that Panama has received large inocula of nonnative organ-
isms associated with the vessels’ hulls and sea chests, which
have been historically important sources of invasions in

TABLE 1. Comparison of service time for ships with and without reservations transiting the Panama
Canal; 7 = number of ships. (Source: Modified from ACP, 2008a.)

Mean transit time (hours) through canal

Year Reservation (7) No reservation (7) Could not get reservation (7)
2000 16.7 (1,944) 35.7 (6,864) 42.1 (121)
2001 15.7 (5,008) 26.3 (6,590) 43.7 (306)
2002 16.1 (5,692) 29.0 (5,134) 57.1 (1,062)
2003 16.2 (5,527) 24.9 (4,596) 45.1 (1,361)
2004 16.4 (6,419) 30.5 (3,568) 49.8 (2,531)
2005 16.5 (6,972) B32) (3,406) 45.8 (2,270)

296 e

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 3. Photograph of a vessel hull upon arrival to Panama showing associated biofouling organisms.
Inset: Close up of bow with barnacles.

other regions (Coutts, 1999; Coutts et al., 2003; Coutts
and Taylor, 2004; Hewitt et al., 2004).

As a result of its shipping history, Panama provides
a unique opportunity to test hypotheses about patterns
and processes of invasions to tropical marine systems. If
propagule supply drives invasion patterns, we predict that
Panama may be a hotspot for invasions. If tropical systems
are inherently less susceptible to invasions (Elton, 1958;
Sax, 2001), we would expect to see low introduced species
richness despite high historical propagule supply. Our cur-
rent research seeks to estimate nonnative species richness
and advance our understanding of historical propagule
supply in Panama, in the context of a broader latitudinal
comparison as discussed above.

EVALUATING FUTURE
CHANGES IN PANAMA

In October 2006, the Republic of Panama passed a
referendum to expand the capacity of the existing Canal.
The modernization will include (a) two new sets of locks,

one at the Pacific entrance and one at the Atlantic; (b) two
new navigational channels to connect the locks to existing
channels; and (c) deeper and wider shipping lanes (Rea-
gan, 2007). The expansion project is now under way and
is scheduled to be completed by 2015 (Reagan, 2007).

When the expansion is completed, the Panama Canal
Authority estimates that Canal transits will most likely
increase from 12,700 per year in 2005 to approximately
19,600 in 2025, with an optimistic forecast as high as
22,100 transits per year (Figure 4). Further, the largest ves-
sels currently capable of transiting the Canal are Panamax
ships reaching 320 m in length that can carry 65,000 tons
of cargo. After the completion of the new locks, the Canal
will accommodate vessels up to 425 m long, carrying about
twice the amount of cargo of today’s ships (Gawrylewski,
2007; Reagan, 2007).

While efforts have been made to evaluate potential en-
vironmental effects of the Panama Canal expansion (ACP,
2008a), the possible effects of this expansion on invasion
dynamics have not received much attention to date. One
might expect an increase in propagule supply associated
with the increased number and size of vessels transiting the

25000

20000

15000

Number of transits/ year

-

10000
WOR DDOrTNMTNORDDAOYTANMY WY
ORORON OMG ea Sa Se Sa SSeS See ONT ON CON CNU CNT
SLOLOTOLOROnOlOeOnOuOnOnoOnOnOtONlO OO! ©)
NANNNNNNNNNNNNNNNNNANN SN
Year

ni a
FIGURE 4. Forecast of demand for Canal transits. Solid black line
= probable demand, dashed line = high (optimistic) demand, and
shaded gray line = low (pessimistic) forecast demand. In 2005, there
were 12,647 recorded Canal transits (solid circle). (Figure modified
from ACP, 2008a.)

Canal. There may also be shifts in trade routes that could
expand the species pool associated with ships’ arrivals, re-
sulting from either new markets or previous constraints
on the size of vessels that could previously use this corri-
dor. Alternatively, the service time of vessels may decrease
as the capacity to accommodate more transits increases.
This decrease could reduce the establishment probability
of organisms attached to the hulls of arriving vessels, as
residence time and likelihood of invasion are thought to be
positively correlated (Davidson et al., 2008).

Potential changes in environmental conditions as-
sociated with both the ships and the Canal entrances
could also influence future invasions. With the interna-
tional ban on tributyl tin as an antifouling coating now
coming into force, some have suggested that biofouling
of ships’ hulls, and hence ship-mediated propagule sup-
ply, may increase (Nehring, 2001). Additionally, changes
in the salinity regimes will probably occur at both Pa-
cific and Atlantic entrances to the Canal, as well as in
areas within the Canal near the lockages, as a result of
increased freshwater discharges into the oceans and po-
tential seawater intrusion into the Canal. Such changes
in salinities could alter the susceptibility to invasion for
arriving organisms. However, any predictions about di-
rectional changes in propagule supply and susceptibility

NUMBER 38 ¢ 297

are currently speculative at best, as sufficient information
presently is not available.

There is also a regional context for the Panama Canal
that deserves consideration. Although the Canal provides
a critical corridor across the Isthmus of Panama for global
trade, Panama’s ports are becoming increasingly important
hubs for the regional distribution of commodities. More
specifically, cargo that is delivered to Panama’s ports is of-
ten transferred secondarily by other vessels to surrounding
countries in the region. As Panama is a distribution center,
any increase in introduced species increases the chances for
ship-mediated dispersal to surrounding ports. Conversely,
increased commerce with the other countries in the region
also enhances the opportunity for delivery of organisms to
Panama. The potential significance of such regional dis-
persal through this hub-and-spoke system of shipping has
not been evaluated for the past, present, or future.

We are currently working with the Panama Canal Au-
thority and the University of Panama to evaluate the role of
the Panama Canal in regional and global marine invasions.
Although the major focus of our efforts is to evaluate past
and current levels of invasion, as well as to obtain some
coarse estimates of propagule supply to the region, we
hope to provide the baseline needed to forecast and evalu-
ate potential impacts of future changes on invasion risks.

CONCLUSIONS

Panama provides exceptional opportunities to test
hypotheses about invasions in tropical marine systems.
The presence of the Canal and the magnitude of shipping
to the region have undoubtedly increased the supply of
nonnative species delivered to the shores of Panama.
While there is limited information on actual propagule
delivery, the Panama Canal Authority has maintained
historical records on the number and characteristics of
transiting vessels. This information provides a unique
view of the magnitude of shipping and changes through
time and could be used as an initial coarse proxy for
propagule supply. We predict that invasions are common
in Panama relative to surrounding regions as a result of
the intensity of shipping in the area. If propagule supply
is positively correlated to introduced species richness, as
the literature suggests, we predict a relatively high num-
ber of invasions have occurred. However, if relatively
few introduced species are detected in Panama, this sug-
gests that some combination of environmental condi-
tions and biotic resistance may limit invasions in this
tropical region.

298 e

We have focused attention on Panama as a model sys-
tem to understand marine invasion dynamics, but a robust
analysis must also include comparisons to other locations
that differ in the intensity of shipping and other transfer
mechanisms. Ideally, such comparisons should be repli-
cated across latitudes. Such a comparative approach is key
to untangling patterns of marine invasions in tropical and
temperate regions and, ultimately, in determining the pro-
cesses that drive these patterns.

ACKNOWLEDGMENTS

We thank Michael Lang for organizing the Smithson-
ian Marine Science Symposium and for providing us the
opportunity to participate. For discussions on the topic,
we thank James Carlton, Ian Davidson, Richard Everett,
Paul Fofonoff, Chad Hewitt, Anson Hines, Julio Lorda,
Valentine Lynch, Whitman Miller, Ira Rubinoff, and Mark
Sytsma. For comments and improvements to the manu-
script, we thank James Carlton, Daniel Muschett, John
Pearse, and Brian Wysor. We also acknowledge assistance
from the Panama Canal Authority (ACP) and funding from
the National Sea Grant Program, Smithsonian Institution,
Secretaria Nacional de Ciencia Tecnologia y Innovacion
de Panama (SENACYT), and the U.S. Coast Guard.

LITERATURE CITED

ACP (Autoridad del Canal de Panama ). 2008a. Master Plan 2005-2025.
http://www.pancanal.com/eng/plan/temas/plan-maestro/ (date ac-
cessed: 14 August 2008).

. 2008b. http://www.pancanal.com. (date accessed: 14 August
2008).

Agard, J., R. Kishore, and B. Bayne. 1992. Perna viridis (Linnaeus, 1758):
First Record of the Indo-Pacific Green Mussel (Mollusca: Bivalvia)
in the Caribbean. Caribbean Marine Studies, 3:59-60.

Aronson, R., S. Thatje, A. Clarke, L. Peck, D. Blake, C. Silga, and B. Sei-
bel. 2007. Climate Change and Invisibility of the Antarctic Benthos.
Annual Review of Ecology and Systematics, 38:129-154.

Barnes, D. K. A., D. A. Hodgson, P. Convey, C. S. Allen, and A. Clarke.
2006. Incursion and Excursion of Antarctic Biota: Past, Present,
and Future. Global Ecology and Biogeography, 15:121-142.

Bastida-Zavala, J. R. 2008. Serpulids (Annelida: Polychaeta) from the
Eastern Pacific, Including a Brief Mention of Hawaiian Serpulids.
Zootaxa, 1722:1-61.

Carlton, J. T. 1979. History, Biogeography, and Ecology of the Intro-
duced Marine and Estuarine Invertebrates of the Pacific Coast of
North America. Ph.D. diss., University of California, Davis.

. 1985. Transoceanic and Interoceanic Dispersal of Coastal Ma-
rine Organisms: The Biology of Ballast Water. Oceanography and
Marine Biology: An Annual Review, 23:313-371.

Carlton, J. T., and V. A. Zullo. 1969. Early Records of the Barnacle
Balanus improvisus Darwin from the Pacific Coast of North
America. Occasional Papers of the California Academy of Sci-
ences, 75:1-6.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Chesher, R. H. 1968. Transport of Marine Plankton Through the Pan-
ama Canal. Limnology and Oceanography, 13:387-388.

Cohen, A. N. 2006. “Species Introductions and the Panama Canal.” In
Bridging Divides: Maritime Canals as Invasion Corridors, ed. S.
Gollasch, B. S. Galil, and A. N. Cohen, pp. 127-206. Dordrecht,
Netherlands: Springer.

Cohen, A. N., and J. T. Carlton. 1995. Biological Study: Non-indigenous
Aquatic Species in a United States Estuary: A Case Study of the Bio-
logical Invasions of the San Francisco Bay and Delta. NTIS Report
PB96-166525. U.S. Fisheries and Wildlife and National Sea Grant
College Program, Springfield, Va.

. 1998. Accelerating Invasion Rate in a Highly Invaded Estuary.
Science, 279:555-558.

Cohen, A. N., L. H. Harris, B. L. Bingham, J. T. Carlton, J. W. Chapman,
C. C. Lambert, G. Lamber, J. C. Ljubenkoy, S. N. Murray, L. C.
Rao, K. Reardon, and E. Schwindt. 2002. “Project Report for the
Southern California Exotics Expedition 2000. A Rapid Assessment
Survey of Exotic Species in Sheltered Coastal Waters.” In A Sur-
vey of Nonindigenous Aquatic Species in the Coastal and Estuarine
Waters of California, Appendix C. Sacramento, Calif.: California
Department of Fish and Game.

Cohen, A. N., C. Mills, H. Berry, M. Wonham, B. Bingham, B. Bookheim,
J. Carlton, J. Chapman, J. Cordell, L. Harris, T. Klinger, A. Kohn,
C. Lambert, G. Lambert, K. Li, D. Secord, and J. Toft. 1998. Puget
Sound Expedition. A Rapid Assessment Survey of Non-Indigenous
Species in the Shallow Waters of Puget Sound. Final Report. Olym-
pia, Wash.: Washington State Department of Natural Resources.

Coutts, A. D. M. 1999. Hull Fouling as a Modern Vector for Marine
Biological Invasions: Investigation of Merchant Vessels Visiting
Northern Tasmania. Master’s thesis, Australian Maritime College,
Launceston, Tasmania.

Coutts, A. D. M., K. M. Moore, and C. L. Hewitt. 2003. Ships’ Sea-
chests: An Overlooked Transfer Mechanism for Non-indigenous
Marine Species? Marine Pollution Bulletin, 46:1510-1513.

Coutts, A. D. M., and M. D. Taylor. 2004. A Preliminary Investigation of
Biosecurity Risks Associated with Biofouling on Merchant Vessels
in New Zealand. New Zealand Journal of Marine and Freshwater
Research, 38:215-229.

Davidson, I. C., L. D. McCann, P. W. Fofonoff, M. D. Sytsma, and G. M.
Ruiz. 2008. The Potential for Hull-Mediated Species Transfers by
Obsolete Ships on Their Final Voyages. Diversity and Distributions,
14:518-529.

deRivera, C. E., G. M. Ruiz, J. A. Crooks, K. Wasson, S. I. Lonhart,
P. Fofonoff, B. P. Steves, S. S. Rumrill, M. S. Brancato, W. S. Pegau,
D. A. Bulthuis, R. K. Preisler, G. C. Schoch, E. Bowlby, A. DeVo-
gelaere, M. K. Crawford, S. R. Gittings, A. H. Hines, L. Takata, K.
Larson, T. Huber, A. M. Leyman, E. Collinetti, T. Pasco, S. Shull, M.
Anderson, and S. Powell. 2005. Broad-Scale Nonindigenous Species
Monitoring Along the West Coast in National Marine Sanctuaries
and National Estuarine Research Reserves. Report to the National
Fish and Wildlife Foundation, San Francisco, Calif.

Eldredge, L. G., and J. T. Carlton. 2002. Hawaiian Marine Bioinvasions:
A Preliminary Assessment. Pacific Science, 56:211-212.

Elton, C. S. 1958. The Ecology of Invasions by Animals and Plants. Lon-
don: Methueun.

Fofonoff, P. W., G. M. Ruiz, A. H. Hines, B. D. Steves, and J. T. Carl-
ton. 2008. “Four Centuries of Estuarine Biological Invasions in
the Chesapeake Bay Region.” In Marine Bioinvasions: Ecology,
Conservation, and Management Perspectives, ed. G. Rilov and J.
Crooks. New York: Springer-Verlag.

Fofonoff, P. W., G. M. Ruiz, B. Steves, and J. T. Carlton. 2003. “In Ships
or on Ships? Mechanisms of Transfer and Invasion for Non-native
Species to the Coasts of North America.” In Invasive Species: Vec-
tors and Management Strategies, ed. G. M. Ruiz and J. T. Carlton,
pp. 152-182. Washington, D.C.: Island Press.

Gawrylewski, A. 2007. Opening Pandora’s Locks. The Scientist, 21:47-53.

Guerrero, K. A., and A. L. Franco. 2008. First Record of the Indo-Pacific
Red Lionfish Pterois volitans (Linnaeus, 1758) for the Dominican
Republic. Aquatic Invasions, 3:255-256.

Hay, M. E., and S. D. Gaines. 1984. Geographic Differences in Herbi-
vore Impact: Do Pacific Herbivores Prevent Caribbean Seaweeds
from Colonizing via the Panama Canal? Biotropica, 16:24-30.

Hewitt, C. L. 2002. The Distribution and Diversity of Tropical Austra-
lian Marine Bio-invasions. Pacific Science, 56:213-222.

Hewitt, C. L., M. L. Campbell, R. E. Thresher, R. B. Martin, S. Boyd,
B. FE Cohen, D. R. Currie, M. F Gomon, M. J. Keogh, J. A. Lewis,
M. M. Lockett, N. Mays, M. A. McArthur, T. D. O’Hara, G. C. B.
Poore, D. J. Ross, M. J. Storey, J. E. Watson, and R. S. Wilson.
2004. Introduced and Cryptogenic Species in Port Phillip Bay, Vic-
toria, Australia. Marine Biology, 144:183-202.

Jones, M. L. and C. E. Dawson. 1973. Salinity-Temperature Profiles in
the Panama Canal Locks. Marine Biology, 21:86-90.

Laguna, G. J. 1985. Systematics, Ecology and Distribution of Barnacles
(Cirripedia: Thoracica) of Panama. Master’s thesis, University of
California, San Diego.

Lambert, C. C., and G. Lambert. 2003. Persistence and Differential Dis-
tribution of Nonindigenous Ascidians in Harbors of the Southern
California Bight. Marine Ecology Progress Series, 259:145-161.

Lockwood, J. L., P. Cassey, and T. Blackburn. 2005. The Role of Propa-
gule Pressure in Explaining Species Invasions. Trends in Ecology
and Evolution, 20:223-228.

Lonsdale, W. M. 1999. Global Patterns of Plant Invasions and the Con-
cept of Invasibility. Ecology, 80:1522-1536.

MacArthur, R. H., and E. O. Wilson. 1967. The Theory of Island Bio-
geography. Princeton, N. J.: Princeton University Press.

Matsui, T., G. Shane, and W. Newman. 1964. On Balanus eburneus (Cir-
ripedia, Thoracica) in Hawaii. Crustaceana, 7:141-145.

McCosker, J. E., and C. E. Dawson. 1975. Biotic Passage Through the
Panama Canal, with Particular Reference to Fishes. Marine Biol-
ogy, 30:343-351.

Menzies, R. J. 1968. Transport of Marine Life Between Oceans Through
the Panama Canal. Nature (London), 220:802-803.

Miller, A. 2000. Assessing the Importance of Biological Attributes for
Invasion Success: Eastern Oyster (Crassostrea virginica) Introduc-
tions and Associated Molluscan Invasions of Pacific and Atlantic
Coastal Systems. Ph.D. diss., University of California at Los Angeles,
Los Angeles.

Miller, A. W., K. Lion, M. S. Minton, and G. M. Ruiz. 2007. Status and
Trends of Ballast Water Management in the United States. Third
Biennial Report of the National Ballast Information Clearinghouse.
Washington, D.C.: U.S. Coast Guard.

Minchin, D., and S. Gollasch. 2003. Fouling and Ships’ Hulls: How
Changing Circumstances and Spawning Events May Result in the
Spread of Exotic Species. Biofouling, 19:111-122.

NEMESIS. 2008. National Exotic Marine and Estuarine Species Infor-
mation System. http://invasions.si.edu/nemesis/index.html (date ac-
cessed: 1 July 2008).

NBIC. 2008. National Ballast Information Clearinghouse. http://invasions
.si.edu/nbic/ (date accessed: 1 July 2008).

Nehring, S. 2001. After the TBT Era: Alternative Anti-fouling Paints and
Their Ecological Risks. Senckenbergiana Maritima, 31:341-351.

Orensanz, J. M., E. Schwindt, G. Pastorino, A. Bortolus, G. Casas, G. Dar-
rigran, R. Elias, J. J. Lopez Gappa, S. Obenat, M. Pascual, P. Pen-
chaszadeh, M. L. Piriz, FE. Scarabino, E. D. Spivak, and E. A. Vallarino.
2002. No Longer the Pristine Confines of the World Ocean: A Survey

NUMBER 38 ¢ 299

of Exotic Marine Species in the Southwestern Atlantic. Biological In-
vasions, 4:115-143.

Paulay, G., L. Kirkendale, G. Lambert, and C. Meyer. 2002. Anthropo-
genic Biotic Interchange in a Coral Reef Ecosystem: A Case Study
from Guam. Pacific Science, 56:403-422.

Reagan, B. 2007. “The Panama Canal’s Ultimate Upgrade.” Popular Me-
chanics, February 2007.

Roche, D. G., and M. E. Torchin. 2007. Established Population of the
North American Harris Mud Crab, Rhithropanopeus harrisii
(Gould 1841) (Crustacea: Brachyura: Xanthidae) in the Panama
Canal. Aquatic Invasions, 2:155-161.

Roche, D. G., M. E. Torchin, B. Leung, and S. A. Binning. 2009. Local-
ized Invasion of the North American Harris Mud Crab, Rhithro-
panopeus harrisii, in the Panama Canal: Implications for Eradica-
tion and Spread. Biological Invasions, 11:983-993.

Rubinoff, R. W., and I. Rubinoff. 1969. Observations on Migration of a
Marine Goby Through the Panama Canal. Copeia, 1969:395-397.

Ruiz, G. M., and J. T. Carlton. 2003. “Invasion Vectors: A Conceptual
Framework for Management.” In Invasive Species: Vectors and
Management Strategies, ed. G. M. Ruiz and J. T. Carlton, pp. 459-
504. Washington, D.C.: Island Press.

Ruiz, G. M., J. T. Carlton, E. D. Grosholz, and A. H. Hines. 1997. Global
Invasions of Marine and Estuarine Habitats by Non-indigenous Spe-
cies: Mechanisms, Extent, and Consequences. American Zoologist,
37:621-632.

Ruiz, G. M., P. W. Fofonoff, J. T. Carlton, M. J. Wonham, and A. H.
Hines. 2000. Invasion of Coastal Marine Communities in North
America: Apparent Patterns, Processes, and Biases. Annual Review
of Ecology and Systematics, 31:481-531.

Ruiz, G. M., and C. L. Hewitt. 2008. “Latitudinal Patterns of Biological
Invasions in Marine Ecosystems: A Polar Perspective. In Smithson-
ian at the Poles: Contributions to International Polar Year Science,
ed. I. Krupnik, M. A. Lang, and S. E. Miller. Washington, D.C.:
Smithsonian Institution Scholarly Press.

Ruiz, G. M., T. Huber, K. Larson, L. McCann, B. Steves, P. W. Fofonoff,
and A. H. Hines. 2006a. Biological Invasions in Alaska’s Coastal
Marine Ecosystems: Establishing a Baseline. Final Report. Prince
William Sound Regional Citizens’ Advisory Council and U.S. Fish
and Wildlife Service, Anchorage, Alaska.

Ruiz, G. M., J. Lorda, A. Arnwine, and K. Lion. 2006b. “Shipping
Patterns Associated with the Panama Canal: Effects on Biotic Ex-
change?” In Bridging Divides: Maritime Canals as Invasion Cor-
ridors, ed. S. Gollasch, B. S. Galil, and A. N. Cohen, pp. 113-126.
Dordrecht: Springer.

Sax, D. FE 2001. Latitudinal Gradients and Geographic Ranges of Exotic
Species: Implications for Biogeography. Journal of Biogeography,
28:139-150.

Verling, E., G. M. Ruiz, L. D. Smith, B. Galil, A. W. Miller, and K. Mur-
phy. 2005. Supply-Side Invasion Ecology: Characterizing Propagule
Pressure in Coastal Ecosystems. Proceedings of the Royal Society of
London, Series B, 272:1249-1256.

Wasson, K., K. Fenn, and J. S. Pearse. 2005. Habitat Differences in
Marine Invasions of Central California. Biological Invasions, 7:
935-948.

Wonham, M., and J. T. Carlton. 2005. Cool-Temperate Marine Inva-
sions at Local and Regional Scales: The Northeast Pacific Ocean as
a Model System. Biological Invasions, 7(3):369-392.

Wysor, B. 2004. An Annotated List of Marine Chlorophyta from the
Pacific Coast of the Republic of Panama with Comparison to Carib-
bean Panama Species. Nova Hedwigia 78:209-241.

i Petes ie uA

oa - 3. eae ery
> mts wee att iea th
Pent biag

Ciguatera Fish Poisoning in the Caribbean

Patricia A. Tester, Rebecca L. Feldman,
Amy W. Nau, Maria A. Faust,
and R. Wayne Litaker

Patricia A. Tester, Amy W. Nau, and R. Wayne Lita-
ker, National Ocean Service, NOAA, 101 Pivers Is-
land Road, Beaufort, North Carolina 28516, USA.
Rebecca L. Feldman, RLF Environmental, 2220
Shefflin Court, Baltimore, Maryland 21209, USA.
Maria A. Faust, Department of Botany, National
Museum of Natural History, Smithsonian Institu-
tion, 4210 Silver Hill Road, Suitland, Maryland
20746, USA. Corresponding author: P. Tester
(pat.tester@noaa.gov). Manuscript received 9 June
2008; accepted 20 April 2009.

ABSTRACT. Ciguatera fish poisoning (CFP) is a significant illness in the Caribbean. Lo-
cal fishers and natives attempt to avoid CFP by applying traditional knowledge concern-
ing where and when certain fish species are likely to be ciguatoxic, but this knowledge
is incomplete. Evidence gathered over the past decades indicates that CFP events are
increasing and becoming more unpredictable, thereby posing a greater threat to local
inhabitants as well as tourists. The current understanding of CFP distribution is from
studies nearly a decade old and generated largely by self-reported CFP incidents to a
call-in “hotline” in Miami, Florida. To better guide resource allocation and focus future
research, an active survey method was used to uniformly query public health profession-
als and fisheries officials on the occurrence of CFP. Points of contact from each of these
two groups were compiled for the 24 Caribbean island countries and territories and 9
mainland countries bordering the Caribbean. An outcome of this project will be to pro-
vide public health agencies, resource managers, and others with information they can use
in developing CFP tracking systems and effective public education programs. The long-
term goal of associated efforts is to provide accurate and affordable monitoring tools for
predicting the onset of CFP events.

PREFACE

Ciguatera fish poisoning (CFP) occurs in tropical regions worldwide and is
globally the most common nonbacterial food-borne illness (Tester, 1994; CDC,
2007; Figure 1A). The toxic organisms most commonly associated with CFP
are benthic dinoflagellates reported to produce ciguatoxins or maitotoxins (Ya-
sumoto et al., 1977; Durand-Clement, 1987; Satake, 2007). Ciguatoxins bio-
concentrate in the food chain and reach their highest levels in top predators
such as barracuda or other tropical reef fish. These toxins have been found in
more than 400 fish species, including groupers, snappers, jacks, mackerels, trig-
gerfish, and surgeonfish (Bagnis et al., 1970). Consumption of tainted fish can
lead to gastrointestinal distress followed by neurological (perioral numbness,
tingling, temperature sensory reversal) and cardiovascular (arrhythmia, brady-
cardia, tachycardia, reduced blood pressure) symptoms and, in rare cases, death.
The chronic phase of CFP can persist for weeks, months, or years (Freudenthal,
1990), and repeated exposure to ciguatoxins exacerbates the symptoms.

302 e

FIGURE 1. A, Potential global distribution of ciguatera fish poison-
ing (CFP). Red areas indicate regions with a high CFP prevalence,
yellow indicates moderate potential exposure, and green indicates
regions where the dinoflagellates responsible for the disease are
found and represent a potential problem. This map represents a
composite of the data obtained from an aquatic biotoxins review by
Huss et al. (2003), the CFP distribution map maintained by the jour-
nal Harmful Algae (WHOI, 2008), and Lewis (2006). B, Potential
distribution map of CFP in the Caribbean, as modified from Stinn et
al. (2000), combined with some recent incident reports showing the
presumed distribution of ciguatera fish poisoning in the Caribbean,
mostly collected by passive means; that is, a self-reporting CFP hot-
line in Miami (“Cigualine” at 1-888-232-8635). Red areas indicate
high frequency of CFP reports; yellow indicates regions where CFP
is reported less frequently; green indicates infrequent reports of CFP.
These maps may not accurately portray the actual CFP distribution
because many cases go unreported.

This paper provides the justification for and an over-
view of our recent efforts to conduct an active survey of
public health officials and fishery management profes-
sionals on the incidence of CFP in the Caribbean. We cur-
rently lack an accurate picture of CFP in the Caribbean

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

because of the difficulty in diagnosing CFP and the ab-
sence of uniform reporting criteria or any entity respon-
sible for maintaining this information. Previous informa-
tion gathered on the incidence of CFP in the Caribbean
has relied heavily on self-reporting mechanisms, such as
calls to a “hot-line” in Miami, Florida. Because people
living closer to Miami are more likely to know about the
hot-line, the reported incidence rates could reflect a geo-
graphic bias (Figure 1B). Another important aspect of this
research has been to focus the joint research efforts of
the National Oceanic and Atmospheric Administration
(NOAA) and Smithsonian Institution scientists who are
working on the molecular and morphological character-
ization of the toxic dinoflagellates responsible for CFP.
Both groups have strong interests in understanding how
changes in the distribution and abundance of ciguatera-
associated dinoflagellate species relate to the occurrence
and severity of CFP.

An important outcome of this project will be to pro-
vide public health agencies, resource managers, and oth-
ers with information that they can use in developing CFP
tracking systems and effective public education programs.
The long-term goal of associated efforts is to provide ac-
curate and affordable monitoring tools for predicting the
onset of CFP events.

INTRODUCTION AND BACKGROUND

Ciguatera fish poisoning is a common disease in the
Caribbean, caused by the ingestion of a wide variety of
fishes that contain toxins accumulated from the marine
food web (Lewis and Holmes, 1993) (Figure 1B). The ul-
timate sources of these toxins (ciguatoxins and maitotox-
ins) are small benthic microalgae belonging to the dino-
flagellate genera Gambierdiscus, Coolia, Ostreopsis, and
Prorocentrum (Figure 2) (Steidinger and Baden, 1984). Al-
though ciguatera fish poisoning (CFP) is a threat to public
health throughout the Caribbean, it is generally managed
by local, traditional knowledge of the native fishers. How-
ever, their knowledge of the seasonality of occurrence and
locations of ciguatoxic reefs may no longer be accurate
because of changing environmental conditions (Tester,
1994; Tosteson, 2004). These environmental changes in
turn alter the distribution and abundance patterns of the
cells that cause CFP. Some evidence exists that ciguatoxic-
ity may vary seasonally, but not all studies support this
view (de Fouw et al., 2001). Tosteson (2004) argued that
seasonality of CFP and the correlation of dinoflagellate
abundance with CFP intoxications evident before 1990

NUMBER 38 ¢ 303

FIGURE 2. Scanning electron micrographs of ciguatera-associated dinoflagellates: A, Gambierdiscus;
B, Coolia; C, Ostreopsis; D, Prorocentrum.

was not observed in data from 1990-2000. He suggested
these changes appeared to be associated with increasing
periods of elevated sea-surface temperatures in the Carib-
bean. Further, the potential for a greater number of people
to be exposed to CFP has increased because of more in-
tense exploitation of fisheries and the depersonalization
of markets (Olsen et al., 1984). Both trends have been ac-
celerated by tourism and rapidly growing resident popula-
tions (CIA, 2008).

The average number of tourist days (excluding ships’
passengers) in the Caribbean, 174 million, dwarfs the 38.8
million residents and represents a significant exposure of
a naive population to CFP. The most common route of
exposure is through consumption of locally harvested fish.
Currently, the annual total Caribbean fishery landings ex-
ceed 1.6 million metric tons (CRFM, 2008; FAO, 2005,

2008; WRI, 2007), making a strong argument for focused
studies on CFP occurrence and on the environmental fac-
tors that affect the distribution and abundance of CFP-
associated organisms.

As part of its commitment to understand and char-
acterize the diversity, distribution, and abundance of
organisms throughout the Caribbean, the Smithsonian
Institution has carried out extensive studies on dinoflagel-
lates over the past 20 years (Faust and Gulledge, 2002).
Because of this pioneering work, much of the background
information and expertise needed to characterize the di-
versity of ciguatera-causing dinoflagellates are already in
place. During the past five years, NOAA (National Oce-
anic and Atmospheric Administration) and Smithsonian
scientists have collaborated to isolate, identify, and genet-
ically characterize the ciguatera-causing dinoflagellates

304 e

of the Caribbean, as well as to develop species-specific
molecular assays for assessing their abundance. As part
of this work, four new Gambierdiscus species have been
discovered and are being described (Tester et al., 2008;
Litaker et al., in press).

We are now in a position to begin systematic studies
of the incidence of CFP and distribution and abundance
of CFP-causing dinoflagellates throughout the Caribbean.
To identify areas of concern from both public health and
marine resource perspectives, and to focus the effective-
ness of environmental sampling, we needed to identify ar-
eas where CFP was most common. Consequently, we ini-
tiated active surveys of local fishery managers and public
health officials. By examining the CFP incidences among
the 24 islands and the 9 mainland countries surrounding

Venezuela

U.S. Virgin Islands
Turks and Caicos
Trinidad and Tobago
St. Vincent and the Grenadines
St. Lucia

St. Kitts and Nevis
Puerto Rico

Panama

Nicaragua
Netherlands Antilles
Montserrat

Mexico

Martinique

Jamaica

Honduras

Haiti

Guatemala
Guadeloupe
Grenada
Dominican Re public
Dominica

Cuba

Costa Rica
Colombia

Cayman Islands
British Virgin Islands
Belize

Barbados

Bahamas

Aruba
Antigua and Barbuda
Anguilla

0 10 20

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

the Caribbean, additional insights can be gained into fac-
tors that govern the spatial and temporal variations in the
prevalence of CFP.

A second objective of this study was to determine
how CFP was being monitored and reported throughout
the Caribbean, where more than 46% of the tourists are
from the United States (United Nations Statistics Division,
2004; CTO, 2008; Figure 3) and the average length of
stay is 8.7 days (United Nations Statistics Division, 2004;
CTO, 2008; Figure 4). This project represents the first
steps toward an assessment of community vulnerability by
the identification of susceptible populations and serves as
a framework for developing human dimensions research
as a cross-cutting priority of ecosystem science supporting
marine resource management (Bauer, 2006).

40 50 60 70 80 90 100

Percent

FIGURE 3. Average percentage of American tourists visiting the Caribbean by country (1996-2005). Only
data from Cancun and Cozumel were used for Mexico. On average, 46% of tourists who visited all Carib-
bean countries came from the United States. On average, not counting visits from passengers on cruise ships,
tourists spend over 174 million tourist days in this region each year (OAS, 1997; ACS-AEC, 2003; UNSD,

2004; CIA, 2008; CTO, 2008).

Venezuela

U.S. Virgin Islands
Turks and Caicos
Trinidad and Tobago
St. Vincent and the Grenadines
St. Lucia

St. Kitts and Nevis
Puerto Rico

Panama

Nicaragua
Netherlands Antilles
Montserrat

Mexico

Martinique

Jamaica

Honduras

Haiti

Guatemala
Guadeloupe
Grenada

Dominican Republic
Dominica

Cuba

Costa Rica
Colombia

Cayman Islands
British Virgin Islands
Belize

Barbados

Bahamas

Aruba

Antigua and Barbuda
Anguilla

NUMBER 38 °* 305

Days
a I a ea ae RS ES

FIGURE 4. Average length of stay for all tourists visiting the Caribbean during 1996-2005, by country. Only
data from Cancun and Cozumel were used for Mexico. The average length of stay for all tourists was 8.7
days (OAS, 1997; ACS-AEC, 2003; UNSD, 2004; CIA, 2008; CTO, 2008). ND = no data.

Our ultimate goal is a comprehensive assessment of
the environmental, sociocultural, and economic impacts
of CFP in the Caribbean and the development of effec-
tive detection and monitoring tools to support manage-
ment decisions and improve inter-island communications
among public health officials, marine resource managers,
Caribbean residents, and tourists.

METHODS

Based on published cases and self-reporting, it ap-
pears that CFP is more prevalent in the eastern Carib-
bean than the western Caribbean (Stinn et al., 2000; see
Figure 1B). To assess whether this is the case or whether
the pattern derives from reporting bias, we used an ac-

tive method to query public health officials and fisheries
managers about the occurrence of CFP from 1996 through
2006 in 24 Caribbean island nations and territories and
9 mainland countries bordering the Caribbean. Fisheries
and public health officials were contacted separately. One
or both agencies could be involved in the surveillance of
and response to CFP, although often within different ad-
ministrative units. Querying two separate agencies was
intended to allow corroboration of the data and to mea-
sure information-sharing between agencies. The question-
naires used in this study were vetted by a panel of experts
with experience in designing human health surveys (see
Acknowledgments).

Initial contact was made with public health and fish-
eries department staff persons by telephone. Introduc-
tory conversations were conducted in English, Spanish,
or French, depending on the preference of the official

306 °«

contacted. The following preliminary information and
questions were provided during these telephone calls be-
fore sending the survey:

e The focus of the project is to gather information about where
and how many people are poisoned by eating fish contami-
nated with ciguatera toxin (ciguatoxin) throughout the Ca-
ribbean, including in (name of country). People
who eat fish carrying this toxin can develop ciguatera fish
poisoning, an illness that affects primarily the digestive and
nervous systems.

¢ Does your office compile information about fish that transmit
ciguatera or cases of ciguatera fish poisoning in
(name of country)?

(If No) Do you know of another office that does?

(If Yes) What is the name of that office?

(If No) Do you know anyone who might be able to help me
locate an office that compiles information about ciguatera fish
poisoning?

A long-term goal of the research project is to better under-
stand where ciguatera fish poisoning occurs, which could improve
the use of resources to monitor and respond to it. The results of
our research throughout the Caribbean will be summarized in a
report, documented in a database, and displayed on maps that
will be available to you and others interested in the project. We
will not be collecting names, addresses, or other personal infor-

mation from people who have ciguatera fish poisoning.

e Are you the best person in your office to provide information
about ciguatera in (name of country)?
(If referred to another person or agency) Do you have any con-

tact information for the person you recommend I speak with?
We are asking for your voluntary assistance with our research.

e¢ Would you be willing to answer a few questions?
(If Yes) Thank you! I would greatly appreciate being able to
e-mail you some specific questions I have. May I do so?
(If No) Why not?

Once an appropriate contact was identified, a writ-
ten copy of the questionnaire was provided in the ap-
propriate language (or languages, as some participants
received the questionnaire in both Spanish and English
or French and English). Both questionnaires (the fisher-
ies department version and the public health department
version) included the 11 core questions listed in Appen-
dix I, as well as 4 questions that applied to only the fish-
eries department (Appendix II) or the public health de-
partment (Appendix III). Efforts were made, in designing

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

the questionnaire, to allow respondents to qualify how
confident they were of the completeness of the data they
were providing.

RESULTS AND DISCUSSION

To date, results are preliminary, as not all the ques-
tionnaires have been returned. However, some trends
have begun to emerge, and it is possible to provide a brief
synopsis of these. One of the most striking results was the
wide range of concern and knowledge about CFP. Some
government agencies have simply asserted that CFP cases
do not occur in their jurisdictions and declined to receive
or complete the questionnaire. Other agencies acknowl-
edged that a potential problem exists but have been ham-
pered by insufficient resources to institute an organized
monitoring system. Still other governments reported
making progress toward bringing CFP surveillance pro-
grams online, sometimes in response to a recent outbreak
of CFP cases.

Some countries had a well-developed mandatory
protocol for reporting CFP, including information on
the name of the patients, symptoms, and diagnosis. In
some instances, public health officials have a high de-
gree of confidence that they are finding 90% or more
of the cases, but most public health officials who have
responded to date are less confident in their statistics.
In some countries, when clusters of CFP cases are ob-
served, the health department issues a press release. At
the same time, the department may do a public service
announcement for radio and TV about the risk of con-
suming barracuda.

A wide range of opinions were offered about how
aware and concerned local populations and fishers are
about the risk that eating certain types of locally caught
fish could result in developing CFP. These responses
ranged from “Not aware” or “Not concerned,” to
“Somewhat aware” or “Somewhat concerned,” to “Very
aware” or “Very concerned.” One respondent com-
mented that native-born citizens had a higher level of
awareness and concern than people who recently moved
to the region. Perceived levels of risk might depend more
on being educated about the problem rather than an ac-
tual risk of exposure. At least in some regions where CFP
is well known, most people seem to understand that if
they feel tingling or prickling on their tongues when they
are eating fish, they should stop eating it to minimize the
risk of becoming sick.

In some countries, the data also suggest a trend to-
ward increasing numbers of CFP cases with time. Public
health officials on a few islands attributed this not to en-
vironmental change, but to population growth, in some
cases as rapid as a doubling of the population in 20 years.
As the population has grown, so too has the demand for
fish, which could result in an increase in the number of
people exposed to CFP. It is generally agreed that CFP is
underreported and that this lack could be attributable to
a variety of reasons (e.g., because its symptoms resemble
those of other diseases when the poisoning is mild). This
apparent increase may also be attributable to increased re-
porting because of heightened awareness, or it may reflect
an actual increase in new cases of CFP.

Several public health departments have compiled and
reported the months and years when people ate ciguatoxic
fish and were diagnosed with CFP. From these limited data
it appears that the number of CFP episodes was distrib-
uted evenly throughout the year but that the number of
cases (people diagnosed with CFP) per episode was greater
in September and October (Figure 5).

Overall, public health and fisheries officials indi-
cated that consumption of contaminated barracuda was
the most common cause of CFP. Other species frequently
identified with CFP include jack, grouper, snapper, hog-
fish, and mackerel. Some fishermen discard barracuda that
do not “put up a fight” when caught, believing that if a
fish does not fight, it is sick. However, it should be noted
that ciguatoxic status cannot be discerned visually; seem-

250

@ Total Cases
B Total Episodes

Number of cases/episodes (1996-2007)

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month
EE
FIGURE 5. Ciguatera episodes and cases in the Caribbean by month
from 1996 to 2007. Episodes indicate multiple cases (usually defined
by zip code) during the same week.

NUMBER 38 ¢ 307

ingly healthy fish can be quite toxic. One positive outcome
of this research was that some countries provided data in-
dicating geographic locations where ciguatoxic fish were
frequently found. This information will guide future sam-
pling efforts.

CONCLUSIONS

The data currently available from Caribbean countries
suggest there is wide variability in the amount of attention
given to CFP. This variability is probably not entirely at-
tributable to how prevalent CFP is in various areas. The
reasons for this include differences in (1) how significant
a problem CFP is thought to pose, (2) awareness of the
risk of CFP, (3) whether central reporting of CFP cases is
mandatory, and (4) resources available for CFP monitor-
ing and education.

Active surveys, such as the one described in this
study, can help countries quantify potential risks and
establish training and monitoring systems for CFP. This
study also provides unique insights into human dimen-
sions of CFP, including perceptions of how significant the
risks are in different areas and how frequently health and
fisheries departments exchange information concerning
CFP. The data from this study were also detailed enough,
in some cases, to suggest specific regions in the Caribbean
where CFP occurrences are elevated or are relatively rare.
This information will facilitate identification of specific
sampling sites for future investigations of the factors that
affect the temporal and spatial variability in exposure to
CFP. The fruitful partnership between the Smithsonian In-
stitution and NOAA continues the Smithsonian’s tradition
of documenting the diversity of life on earth and NOAA’s
mission to bring state-of-the-art management tools to the
marine community.

ACKNOWLEDGMENTS

We thank Drs. Nathalie Valette-Silver, David John-
son, and Michael Dowgiallo and Mr. Joseph Schittone of
the National Ocean Service (NOS), NOAA, for facilitat-
ing this study. Dr. Richard Appledorn, Director of the Ca-
ribbean Coral Reef Institute, provided guidance for field
sampling. Mr. Michael Carpenter, National Museum of
Natural History (NMNH), kindly provided logistic sup-
port. Mr. Scott Whittaker (Head) and Mr. Don Hurlbert
of the Smithsonian’s NMNH Imaging and Photography
Department generously gave us technical assistance. We

308 e

appreciate the guidance from Leonard Young, Public
Health Scientist, Eastern Research Group; Dr. MaryBeth
Bauer, NOAA National Centers for Coastal Ocean Sci-
ence; Dr. Chris Ellis and Ms. Lauren Smith, Human Di-
mensions Program, NOAA Coastal Services Center; and
Dr. Janna Shakeroff, Duke University Marine Labora-
tory, in the development of the questionnaire used in this
study. We are especially grateful to Dr. Klaus Ruetzler
and the Caribbean Coral Reef Ecosystem Program and
the NOS International Program Office for funding. This
is contribution number 851 of the Caribbean Coral Reef
Ecosystems Program (CCRE), Smithsonian Institution,
supported in part by the Hunterdon Oceanographic Re-
search Fund.

LITERATURE CITED

Association of Caribbean States (ACS-AEC). 2003. Tourism Statistics
1995-2002. http://www.acs-aec.org/Documents/Tourism/Projects/
ACS_ST_000/Tourism_Stats0603.pdf (accessed 16 July 2008).

Bagnis, R., FE. Berglund, P. S. Elias, J. G. Van Esch, B. W. Halstead, and
K. Kojima. 1970. Problems of Toxicants in Marine Food Products:
1. Marine Biotoxins. Bulletin of the World Heath Organization,
42:69-88.

Bauer, M. 2006. Harmful Algal Research and Response: A Human
Dimensions Strategy. Woods Hole, Mass.: Woods Hole Oceano-
graphic Institution.

Caribbean Regional Fisheries Mechanism (CRFM). 2008. Members.
http://www.caricom-fisheries.com/main/members.asp (accessed 16
July 2008).

Caribbean Tourism Organization (CTO). 2008. Individual Country
Statistics. http://www.onecaribbean.org/statistics/countrystats/ (ac-
cessed 16 July 2008).

Central Intelligence Agency (CIA). 2008. The World Factbook. https://
www.cia.gov/library/publications/the-world-factbook/ (accessed 16
July 2008).

Centers for Disease Control (CDC). 2007. “Non Infectious Risks During
Travel.” In CDC Health Information for International Travel 2008.
http://wwwn.cdc.gov/travel/yellowBookCh6-FoodPoisoningMarine
aspx (accessed 16 July 2008).

de Fouw, J. C., H. P. van Egmond, and G. J. A. Speijers. 2001. Ciguatera
Fish Poisoning: A Review. RIVM Report 388802 021. Bilthoven,
Netherlands: National Institute of Public Health and the Environ-
ment. http://www.rivm.nl/bibliotheek/rapporten/388802021.pdf (ac-
cessed 16 July 2008).

Durand-Clement, D. 1987. Study of Production and Toxicity of Cultured
Gambierdiscus toxicus. Biological Bulletin, 172:108-121.

Food and Agriculture Organization of the United Nations (FAO). 2005.
“Fish and Fishery Products—Apparent Consumption.” In Yearbook of
Fishery Statistics Summary Tables. Rome: FAO Fisheries and Aqua-
culture Department. ftp://ftp.fao.org/fi/stat/summary/applybc.pdf (ac-
cessed 16 July 2008).

. 2008. Fishery and Aquaculture Country Profiles. http://www
.fao.org/fishery/countryprofiles/search (accessed 16 July 2008).
Faust, M. A., and R. A. Gulledge. 2002. Identifying Harmful Marine
Dinoflagellates. Contributions from the United States National

Herbarium, 42:1-144.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Freudenthal, F. R. 1990. Public Health Aspects of Ciguatera Poisoning
Contracted on Tropical Vacations by North American Tourists.
In Toxic Marine Phytoplankton, ed. E. Graneli, B. Sundstrom, L.
Edler, and D. A. Anderson, pp. 463-468. New York: Elsevier.

Huss, H. H., Ababouch, L., and L. Gram. 2003. Assessment and Man-
agement of Seafood Safety and Quality. FAO Fisheries Technical
Paper 444. Rome: FAO Fisheries and Aquaculture Department.
http://www.fao.org/docrep/006/y4743e/y4743e00.htm (accessed 17
July 2008).

Lewis, R. J. 2006. Ciguatera: Australian Perspectives on a Global Prob-
lem. Toxicon, 48:799-809.

Lewis, R. J., and M. J. Holmes. 1993. Origin and Transfer of Toxins
Involved in Ciguatera. Comparative Biochemistry and Physiology,
106C:615-628.

Litaker, R. W., M. W. Vandersea, M. A. Faust, S. R. Kibler, M. Chinain,
M. J. Holmes, W. C. Holland, and P. A. Tester. In press. Taxonomy
of Gambierdiscus: Including Four New Species, Gambierdiscus
caribaeus sp. nov., Gambierdiscus carolinianus sp. nov., Gam-
bierdiscus carpenteri sp. nov. and Gambierdiscus ruetzleri sp. nov.
(Gonyaulacales, Dinophyceae). Phycologia.

Olsen, D. A., D. W. Nellis, and R. S. Wood. 1984. Ciguatera in the East-
ern Caribbean. Marine Fisheries Review, 46:13-18.

Organization of American States Tourism Section (OAS). 1997. Tourism
Statistics and Market Information. http://www.oas.org/TOURISM/
stats.htm (accessed 17 July 2008).

Satake, M. 2007. “Chemistry of Maitotoxin.” In Phycotoxins: Chem-
istry and Biochemistry, ed. L. M. Botana, pp. 47-53. Ames, Iowa:
Blackwell Publishing.

Steidinger, K. A., and D. G. Baden. 1984. “Marine Dinoflagellates.” In
Dinoflagellates, ed. D. L. Spector, pp. 201-261. Orlando, Fla.: Aca-
demic Press.

Stinn, J. F., D. P. de Sylva, L. E. Fleming, and E. Hack. 2000. Geographic
Information Systems and Ciguatera Fish Poisoning in the Tropical
Western Atlantic Region. Geographic Information Systems in Pub-
lic Health: Proceedings of the Third National Conference. http://
www.atsdr.cdce.gov/gis/conference9 8/proceedings/html/stinn.html
(accessed 17 July 2008).

Tester, P. A. 1994. “Harmful Marine Phytoplankton and Shellfish Toxic-
ity: Potential Consequences of Climate Change.” In Disease in Evo-
lution: Global Changes and Emergence of Infectious Diseases, ed.
M. E. Wilson, R. Levins, and A. Spielman, pp. 65-70. New York:
The New York Academy of Sciences.

Tester, P. A., M. A. Faust, M. W. Vandersea, S. R. Kibler, M. Chinain,
M. Holmes, C. Holland, and R. W. Litaker. 2008. “Gambierdiscus
toxicus: Does Clonal Type Material Exist?” In 12th International
Conference on Harmful Algae, ed. @. Moestrup, and H. Enevold-
son, pp. 269-271. Paris: IOC of UNESCO.

Tosteson, T. R. 2004. Caribbean Ciguatera: A Changing Paradigm. Re-
vista de Biologia Tropical, 52:109-113.

United Nations Statistics Division (UNSD). 2004. National Accounts.
http://unstats.un.org/unsd/nationalaccount/amasample.pdf (accessed
16 June 2008).

Woods Hole Oceanographic Institute (WHOI). 2008. Maps: Ciguatera
Fish Poisoning. Harmful Algae, http://www.whoi.edu/redtide/page
.do?pid=18103&tid=542&cid=47588&c=3 (accessed 17 July
2008).

World Resources Institute (WRI). 2007. Coastal and Marine Ecosystems:
Country Profiles. Earth Trends: The Environmental Information Por-
tal, http://earthtrends.wri.org/country_profiles/index.php?theme=1
(accessed 16 July 2008).

Yasumoto, T., I. Nakajima, Y. Oshima, and R. Bagnis. 1977. A New
Toxic Dinoflagellate Found in Association with Ciguatera. In Toxic
Dinoflagellate Blooms, ed. D. L. Taylor, and H. H. Seliger, pp. 65-70.
New York: Elsevier.

NUMBER 38 °¢ 309

APPENDIX 1

The following core questions were used in both fisheries and health department questionnaires:

1. a. What information does your office compile about cases of ciguatera fish poisoning (for example, number of people
diagnosed, locations of people diagnosed, locations where fish were caught, etc.)?
b. If your office does not compile such information, is there another governmental office or agency that does, and
what is its name? Be (CSe eee NOs ann eminotsure
Name of other office or agency:
c. If yes, what types of information do you think that office might have?
d. Please provide contact information for someone in that office, if possible (contact name, e-mail address, phone

number, and fax number).

2. If you receive reports of ciguatera fish poisoning, from whom do the reports come? (Please check ALL that apply.)

___ Doctors __ Clinics and Hospitals ___ Fisheries Department [for the health department survey] ___ Fishermen
___ Health Department [for the fisheries department survey]/Other Health Agencies [for the health department
survey] (please specify jurisdiction represented and contact person, if available) __ Restaurants __ Hotels

___ Individual Citizens ___ Other Sources (please list)

3. Please indicate the total number of reported ciguatera fish poisoning cases per calendar year from January 1, 1996,
through December 31, 2006.
[A table containing one line for each year, a column for the number of cases reported, and a column for any com-
ments was provided here. |

4. To the extent available, please provide the following information for each episode of ciguatera fish poisoning. (For
this study, an episode is defined as an occasion when one or more people were poisoned on the same day by one or
multiple fish of the same variety, caught in the same place.)

. Number of people poisoned

. Date of episode (list season and year if date or month is not known)

. Date of diagnosis, if date of poisoning (B) is not known

. City where fish with ciguatera was eaten

. Home city of patient(s), if city where fish was eaten (D) is not known

. Type of fish with ciguatera (common name or scientific name)

. Describe where fish with ciguatera was caught, in as much detail as possible (with latitudes and longitudes, if

available)

mmoanoep

5. At this time, is reporting any information about fish transmitting ciguatera or cases of ciguatera fish poisoning volun-
tary or mandatory? ___ Voluntary __ Mandatory _ [Tm not sure
If reporting is mandatory:
a. What information must be reported?
b. When did it become mandatory?
c. What agency receives these reports initially?

6. a. What percentage of ciguatera fish poisoning cases diagnosed each year in (name of country or terri-
tory) do you think are reported to your office?
b. How confident are you of this estimate? ___ Very confident _ Somewhat confident ___ Slightly confident
___ Not at all confident _ I’m not sure

continued

310 © SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Appendix 1 continued

Te

10.

Jkil,

a. To your knowledge, has your agency or another governmental agency issued any advisory warnings related
to consuming fish that might carry ciguatera, such as barracuda or large reef fish?
—__Yes __No _I’mnotsure

b. If yes, please indicate which office issued the advisory.

c. If applicable, please include or attach the wording of each such advisory and indicate when it was issued. Attach
additional pages, if necessary.

d. If your agency has not issued an advisory, who or what agency would be most appropriate to consult for
information on advisories? (Please list the agency name and the following, if available: a contact name, e-mail

address, phone number, and fax number.)

How often do your department and fisheries department officials [for the health department survey]/health depart-
ment officials [for the fisheries department survey] exchange information about episodes of ciguatera fish poisoning?
___ As cases occur __ Every month __ Every 3 months _ Every 6 months __ Every year __ Never

____ Other (please specify):

How aware do you think local citizens are of the risk that eating certain types of fish could cause them to develop
ciguatera fish poisoning? Very aware Somewhat aware Not very aware Not aware
I’m not sure

To what extent do you think local citizens are concerned about ciguatera fish poisoning?
__ Very concerned _ Somewhat concerned __ Slightly concerned _ _Notconcerned __ I’m not sure

Please provide your contact information for future reference. Thanks again for your assistance!
Government represented:

Agency and office:

Name and title of person completing questionnaire:

Telephone number, with city code:

Fax number, with city code:

e-mail address:

Date information provided:

Would you like to receive notification of the results of the study? YES No

APPENDIX 2

The following questions were directed only to officials representing fisheries departments:

ily

a. Is information usually communicated to you about where fish suspected of carrying ciguatera were caught?

Yes No
b. Is information usually communicated to you about what types of fish have carried ciguatera?
Yes No

c. If yes to either (a) or (b), and if you do not have information in the format provided in Question 4, please provide
any information you have about the types of fish, and the locations involved in episodes of ciguatera fish poisoning
reported to you, for the years 1996 to 2006. [A table was provided with the following headings: Year, Common or
scientific names of fish reported, Locations of fish reported (latitudes/longitudes, if possible, or place names, in as
much detail as possible).]

NUMBER 38 e¢ 311

2. Please provide any information you have on economic losses resulting from ciguatera fish poisoning, either quantita-

tive or qualitative (for example, if fishing had ceased at a particular reef because of the suspected presence of cigua-
tera toxins, there might be an annual loss of $10,000 to the fishing industry). Please include the year(s) your data
reflects and note your data sources.

. To what extent do you think fishermen are aware of the risk of catching certain types of fish that could cause people

to develop ciguatera fish poisoning? Very aware Somewhat aware Not very aware
Not aware I’m not sure

. To what extent do you think fishermen are concerned about catching certain types of fish that could cause ciguatera

fish poisoning? Very concerned Somewhat concerned Slightly concerned Not concerned
I’m not sure

APPENDIX 3

The following questions were directed only to officials representing health departments:

. Is any information available to you on the cost per year to your government of monitoring or documenting the inci-

dence of ciguatera fish poisoning? Yes No
If yes, please provide the information below and note your data sources.

. Is any information available to you on the cost per year of medical treatments in (name of country or

territory) for ciguatera fish poisoning, as an average per person affected by ciguatera fish poisoning and/or annually
for (name of country or territory)? SESS No

If yes, please provide it below, specify whether it reflects a total or an average per person, and note your data sources.

. Is any information available to you related to the number of days people have been unable to work due to ciguatera

fish poisoning per year in (name of country or territory)? Yes No

If yes, please provide it below, specify whether it reflects a total or an average per person, and note your data sources.

. Would you rank ciguatera fish poisoning as one of the 10 most severe food-borne illnesses in (name of

country or territory)? —_ Yes __No _Tmnot sure
If yes, would it rank in the top__ 1 to 5 or ____6 to 10?

: ; ; E MRSA.

L
i
1

De: Le

Kies od na wy all Fel
3 f : 2 5 A Vicvairng :

ae Tlelac

, E eu iz eo din) yeni ot
. peas 1 Ser:

—*
:

Dal
ry
}

Mee
;
ag

.
t

~ wit neat SY
“th isape anole ata

ev

fd sicdelteree

|
.
}
Bi,
'
7
=
; ‘y
Bay Z ss =
2 i
: 2
i = es
an ve ? ai

History of Reef Coral Assemblages
on the Rhomboid Shoals of Belize

Richard B. Aronson, lan G. Macintyre,
Anke M. Moesinger, William E. Precht,
and Michael R. Dardeau

Richard B. Aronson, Department of Biological
Sciences, Florida Institute of Technology, 150
West University Boulevard, Melbourne, Florida
32901, USA. Ian G. Macintyre, Department
of Paleobiology, National Museum of Natural
History, Smithsonian Institution, P.O. Box 37012,
MRC 125, Washington, D.C. 20013-7012, USA.
Anke M. Moesinger and Michael R. Dardeau,
Dauphin Island Sea Lab, 101 Bienville Boulevard,
Dauphin Island, Alabama 36528, USA. William
E, Precht, Florida Keys National Marine Sanctu-
ary, National Oceanic and Atmospheric Admin-
istration, 95230 Overseas Highway, Key Largo,
Florida 33037, USA. Corresponding author: R.
Aronson (raronson@fit.edu). Manuscript received
25 July 2008; accepted 20 April 2009.

ABSTRACT. Coral assemblages of the rhomboid shoals of the Belizean barrier reef have
undergone dramatic, historically unprecedented changes over the past several decades.
Before the late 1980s, the flanks of the shoals exhibited a distinct biological zonation,
with branching Porites spp. dominant in a shallow zone (0-3 m water depth); the stag-
horn coral Acropora cervicornis dominant in an intermediate zone (3-15 m depth); and
large, plating agariciids and the lettuce coral Agaricia tenuifolia dominant in a deep zone
(15-30 m depth). Acropora cervicornis died off catastrophically from white-band disease
after 1986 and was replaced by Agaricia tenuifolia in the intermediate zone. Push-cores
extracted from intermediate depths in previous studies showed that Acropora cervicornis
was the dominant space occupant and primary framework builder for millennia before
the phase shift to Agaricia tenuifolia. Cores extracted from the shallow zone showed
that Acropora cervicornis dominated until several centuries ago, when the tops of the
reefs reached approximately 2 m water depth and branching Porites spp. replaced it.
In contrast, three cores extracted from the deep zone in the present study showed that
for millennia the subsurface coral assemblage, like the assemblage on the modern deep-
reef surface, was dominated by large, plating agariciids and Agaricia tenuifolia. Because
white-band disease only affects acroporid corals, the unprecedented phase shift that fol-
lowed the outbreak was confined to the intermediate zone. High sea temperatures in
the summer of 1998 caused coral bleaching and mortality, especially of agariciids in the
intermediate and deep zones, but to date this event has not left a geologic signature in
the Holocene record.

INTRODUCTION

Coral reef ecosystems are collapsing at an accelerating rate, jeopardizing
the ecosystem services that they provide (Hughes et al., 2003; Wilkinson, 2006;
Carpenter et al., 2008). The common presumption that mortality of hard corals
(Scleractinia and Milleporina) commenced earlier and was more severe in the
Caribbean and eastern Pacific than in other tropical and subtropical regions
may not be correct (Bruno and Selig, 2007). Nevertheless, the causes and conse-
quences are best understood for the Caribbean and eastern Pacific.

Coral mortality has been elevated in the Caribbean since the late 1970s
(Gardner et al., 2003). The impacts of global change, including increasing sea
temperatures, increasing cyclone intensity, and declining aragonite saturation

314 e

state (Kleypas et al., 1999; Buddemeier et al., 2004; Hoegh-
Guldberg et al., 2007), are sources of grave concern, but
coral assemblages throughout the Caribbean have already
been severely affected by outbreaks of infectious marine
diseases (Aronson and Precht, 2001b; Sutherland et al.,
2004; Weil et al., 2006). In particular, white-band disease
(WBD), a bacterial infection that is specific to acroporid
corals, decimated Acropora palmata (elkhorn coral) and
Acropora cervicornis (staghorn coral) on reefs throughout
the western Atlantic from the late 1970s through the early
1990s (Aronson and Precht, 2001a, 2001b). Acroporid
populations have been reduced so drastically that the two
species are now listed as threatened under the U.S. Endan-
gered Species Act (Hogarth, 2006) and are classified as
critically endangered according to the Red List criteria of
IUCN, the International Union for Conservation of Nature
(Carpenter et al., 2008). Hurricanes, temperature-induced
bleaching, declining herbivory, nutrient loading, and preda-
tion by corallivores have had additional, interacting impacts
on coral mortality and the scope for population recovery
(Aronson and Precht, 2006). Emergent diseases, for exam-
ple, could be related to or exacerbated by global warming
and nutrient loading (Harvell et al., 2002; Rosenberg and
Ben-Haim, 2002; Bruno et al., 2003, 2007; Sutherland et
al., 2004; Kline et al., 2006). Recent changes on Carib-
bean reefs were novel events in at least the last 3,000 to
4,000 years (Aronson et al., 2002a, 2004, 2005a; Wap-
nick et al., 2004; Hubbard et al., 2005; Greer et al., 2009),
and Pandolfi et al. (2006) drew a similar conclusion about
Holocene reef dynamics in Papua New Guinea.

Aronson and Precht (2001a, 2001b, 2006; Precht
and Aronson, 2006) argued that because WBD was the
primary cause of recent mortality of Acropora palmata
and Acropora cervicornis in the Caribbean, and because
the two species were the dominant space occupants at
depth ranges of 0-5 and 5-25 m, WBD was clearly one
of the most important causes of recent coral mortality
in the region. Mass mortality of the acroporids was fol-
lowed by two types of phase shifts. Where coral mortality
exceeded the capacity of herbivores to respond to algal
growth on the space that had been opened, macroalgae
rose to dominance (Ostrander et al., 2000; Aronson and
Precht, 2001a, 2006; Williams et al., 2001; Rogers and
Miller, 2006). Where herbivory was sufficient to control
the algae, brooding, self-fertilizing corals, primarily of the
families Agariciidae and Poritidae, replaced the acropo-
rids (Aronson and Precht, 1997; Greenstein et al., 1998;
Bythell et al., 2000; Knowlton, 2001; Green et al., 2008).
The shift to macroalgal dominance has not been as wide-
spread as previously supposed (Bruno et al., 2009).

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

An important exception to the overall Caribbean
trend is the Flower Garden Banks (FGB) in the northwest-
ern Gulf of Mexico, where coral cover has held steady at
40%-60% at depths of 17-26 m from the 1970s to the
present. Aronson et al. (2005c) explained the persistently
high coverage of living corals based on the historical ab-
sence of the cold-sensitive acroporids. Coral mortality has
been far lower at the FGB than elsewhere in the Western
Atlantic region because no acroporids were present to die
of WBD. The appearance of Acropora palmata at the FGB
in the past few years could be related to global warming
(Precht and Aronson, 2004).

An ecosystem-level version of this biogeographic ar-
gument is that reef zones historically not dominated by
acroporids should not have undergone phase shifts at the
same time as the adjacent Acropora-dominated zones. In
this study we examined the millennial-scale history of the
coral assemblage near the bases of the rhomboid shoals
in the central shelf lagoon of the Belizean barrier reef. We
cored the deep-reef framework of two of the shoals, recon-
structed the history of the coral assemblage during the late
Holocene, and completed a model of reef development
over the last several thousand years from present sea level
down to the bases of the shoals. Although Acropora cervi-
cornis dominated at intermediate depths for millennia un-
til the late 1980s, acroporids apparently did not dominate
the deep zone for at least the past 1,500 to 2,000 years,
providing an opportunity to test our hypothesis of the oc-
currence and timing of phase shifts.

ZONATION AND PALEOECOLOGY
OF THE RHOMBOID SHOALS

The rhomboid shoals are uncemented, atoll-like reefs
lying within the central shelf lagoon of the Belizean bar-
rier reef. The sloping outer flanks of the rhomboid shoals
displayed a clear pattern of coral zonation from at least as
far back as the early 1970s, when the first rigorous ecologi-
cal observations were made, until 1986 (Westphall, 1986;
Aronson and Precht, 1997; Aronson et al., 1998). A shal-
low zone (0-3 m water depth) was dominated by branching
Porites spp., primarily Porites furcata and Porites divaricata,
mixed with the hydrocoral Millepora alcicornis. Acropora
cervicornis dominated an intermediate-depth zone (3-15 m
depth), with the blade-forming lettuce coral Agaricia tenui-
folia as the subdominant. (Agaricia tenuifolia recently has
been revised to Undaria tenuifolia; however, we will retain
Agaricia as the generic designation in this paper.) A deep
zone, extending from 15 m to the lagoon floor at 22-30 m

depth, was dominated by large colonies of plating agari-
ciids (Agaricia lamarcki, Agaricia grahamae, Agaricia un-
data, and Leptoseris cucullata) and Agaricia tenuifolia,
with scattered massive corals. The total hard-coral fauna
consisted of approximately 25 species, most of which were
rare (Aronson and Precht, 1997).

In the decade following 1986, the dominant coral at
intermediate depths, Acropora cervicornis, succumbed
to WBD and was replaced by Agaricia tenuifolia. This
phase shift was mediated by an abundant, herbivorous
sea urchin, Echinometra viridis, which limited macroalgal
growth and promoted the recruitment and opportunistic
growth of agariciids on the dead skeletons of Acropora
cervicornis (Aronson and Precht, 1997). Agaricia tenuifo-
lia was the fastest growing of the agariciids that recruited
and, therefore, it became the new dominant.

To determine whether the transition was historically
unique, Aronson et al. (2002a) extracted push-cores at
5-10 m water depth from stations distributed over a 375
km? area of the lagoon (Figure 1). Analysis and radiocar-
bon dating of the cores revealed continuous dominance
of Acropora cervicornis and upward growth of the reef
for at least 3,000 years before the late 1980s. Spines of
Echinometra viridis were present throughout the cores,
indicating continuously high herbivory. During the past
three millennia Agaricia tenuifolia grew in small patches
(of the order of square meters), which appeared as subsur-
face layers of skeletal plates that were isolated in time and
space (Aronson et al., 2002a). The recent, area-wide phase
shift, in contrast, was preserved at the tops of the cores as
a layer of Agaricia tenuifolia plates overlying a thin layer
of taphonomically degraded Acropora cervicornis. This
signature persisted in the Holocene record despite sub-
sequent hurricanes and bleaching events (Aronson et al.,
2000, 2002b, 2005b). Coring in a lagoonal habitat at Dis-
covery Bay, Jamaica, showed that a more common phase
shift, in which Acropora cervicornis was killed by WBD
and replaced by macroalgae as the result of limited her-
bivory, was similarly unprecedented on a millennial time
scale (Wapnick et al., 2004).

Cores extracted from the rims and ridges of the shoals
near the present sea level revealed that Acropora cervicor-
nis dominated the shallowest portions of these reefs for at
least several millennia until approximately 500 years ago
(Westphall, 1986; Aronson et al., 1998, 2005a; Macintyre
et al., 2000). At that time the reef tops grew to within
2 m of sea level, and branching Porites spp. replaced
Acropora cervicornis as the dominant coral taxon.
Since then, the Porites-dominated assemblage has kept
up with the slowly rising sea level, forming the shallow

NUMBER 38 e¢ 315

zone. The shallowing-upward, successional sequence in
the shallow zone contrasts with the post-1986, disease-
induced replacement of Acropora cervicornis by Agaricia
tenuifolia at intermediate depths.

As part of the worldwide reef-bleaching event of
1997-1998 (Wilkinson, 2000), which was related to the
El] Nino-Southern Oscillation and probably augmented by
global warming, a high-temperature anomaly in the sum-
mer of 1998 bleached almost all corals in the intermedi-
ate and deep zones of the rhomboid shoals (Aronson et
al., 2000, 2002b). Agaricia tenuifolia is particularly prone
to temperature-induced bleaching (Robbart et al., 2004),
and populations of this coral at intermediate and deeper
depths experienced nearly complete mortality. Mortality
rates were lower, but still very high, for plating agariciids.
The dead coral skeletons were colonized primarily by thin
algal turfs and the sponge Chondrilla aft. nucula (Aron-
son et al., 2002b), which Ritzler et al. (2007) have now
described as Chondrilla caribensis. Agariciid populations
had not recovered as of December 2008 (W. F. Precht, per-
sonal observation). Branching Porites corals in the shal-
low zone were less affected by the 1998 thermal anomaly.
These corals did not bleach to the extent the agariciids did,
and asa result they did not experience large-scale mortality
(W. F. Precht and R. B. Aronson, personal observation).

MATERIALS AND METHODS

In April 2008, we extracted six push-cores in water
depths of 14.0-19.5 m from the reefs at Channel and El-
bow Cays, in the center of our 375 km? study area (see
Figure 1). Push-coring requires less equipment than me-
chanical techniques such as rotary drilling and percus-
sion vibracoring. By eliminating the need for tripods and
other heavy equipment, push-coring offers easier logistics,
greater mobility, and a much lower cost per core. Penetra-
tion and recovery of cores dominated by branching and
foliose corals have been excellent in the shallow and in-
termediate-depth zones of the rhomboid shoals, as well
as on uncemented lagoonal reefs in Panama and Jamaica
(Dardeau et al., 2000; Aronson et al., 2004; Wapnick et
al., 2004). Rotary drilling is not an option because branch-
ing and foliose corals generally are broken up and flushed
out of the core barrel. As a result, recoveries are poor in
lagoonal and fore-reef environments dominated by fragile
corals (Glynn and Macintyre, 1977; Halley et al., 1977;
Macintyre et al., 1981; Shinn et al., 1982).

Dardeau et al. (2000) described the push-coring
method in detail. Briefly, aluminum tubes, 5 m long and

316 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

© — |

(2 km

Og Wee Wee ~ ©
A Cay

4

Teardrop®
Shoal @*

@
A, &
a a) & ‘
Sey ro Honduras

o °

Peter pouolas The Needle 88°W
Q
¢.
BZ08-6 Elbow
N Sd Cays ® ‘
Northeast Cay
~@

Quamino

Mexico

Caribbean Sea

Guatemala

Slasher @
Sand net

Crawl Cay

Victoria
Channel
Carbonate Shoal

(<2 m depth)

|| Mangrove Cay

® Coring Station

Bakers Rendezvous

FIGURE 1. Map of the central shelf lagoon of the Belizean barrier reef, showing the rhomboid shoals and the locations of
the two coring stations. Three of the six cores extracted were analyzed for this study, as noted on the map. (Modified from
Aronson et al., 2002a.)

7.6 cm (3 in.) in diameter with teeth cut into their leading
ends, were driven by hand into the uncemented reef frame-
work. The tubes were rotated in using adjustable core slips
with handles and tapped with a sliding hammer-weight,
sleeved over the top, to aid in penetration. The tubes cut
through, penetrated, and captured the loose framework
of branching and foliose coral skeletons, and they cored
through most massive coral heads as well. Although head
corals are rare in the subsurface at shallow and intermedi-
ate depths, they are more common in the deep zone. The
cores were sealed with plastic caps and electrical tape,
extracted from the reef, and transported to the laboratory
for analysis.

Estimates of recovery were obtained at intervals dur-
ing the coring process by dropping a weighted fiberglass
measuring tape down the open core barrel. In previous
studies these measurements, along with simultaneous
measurements of penetration, confirmed that material
entered the tubes continuously as they were forced into
the reef. In some cases in the present study the tube cored
through a massive coral and was plugged by it, preventing
further recovery as the tube was forced deeper into the
uncemented framework. We used the penetration depth
at the point at which the tube was plugged to calculate
percent recovery. Comparison of final recoveries measured
before extraction with recoveries measured after extrusion
in the laboratory showed that little or no material was
lost from the bottoms of the tubes during extraction. Of
course no material was lost from cores that were plugged
at their bases by massive corals.

There were no indications of significant voids in the
reef framework. In no case did the tube suddenly drop
vertically while we were driving it into the reef. We also
saw no reversals in the in situ estimates of recovery, which
would have indicated episodic compaction during coring.

Three of the extruded cores were analyzed at inter-
vals of 5 cm. The constituents of each interval retained
on a 5 mm sieve were cleaned of matrix, sorted to spe-
cies, dried to a constant mass, and weighed to the nearest
milligram. In earlier studies, we showed from regression
analysis that, for the coral constituents, log(mass) was a
strong predictor of log(volume), as measured by water
displacement.

In the manner described previously by Wapnick et al.
(2004), we assessed the degree of taphonomic degradation
of the Acropora cervicornis material—encrustation, sur-
ficial erosion, and internal boring—using a modified ver-
sion of the rank scales of Greenstein and Moffat (1996).
The average taphonomic condition of each coral fragment
was rated as good, intermediate, or poor. The good rating

NUMBER 38 e¢ 317

was applied to fresh-looking pieces that had little or no
encrustation, retained essentially all their surface sculp-
ture, and showed little to no evidence of internal boring.
Poor fragments were those with extensive encrustation,
surficial erosion, and/or boring; degradation was extensive
enough that the structure of the corallites was completely
obscured. Fragments were rated as intermediate if their
condition, averaged over the three categories, fell between
good and poor. A coral taxon, or a taxon in a particular
taphonomic condition, was considered dominant in a 5
cm interval if its mass exceeded the mass of each of the
other taxa/conditions in that interval.

A coral sample from the bottom of each core was ra-
diocarbon dated by Beta Analytic, Inc. (Miami, Florida),
using standard techniques. Measured dates were corrected
for isotopic fractionation to generate conventional dates,
which are expressed as radiocarbon years before 1950
(14C year). Conventional dates were calibrated to calendar
years before 1950 (CalBP).

RESULTS

The cores captured the framework of loose coral skel-
etons surrounded by a light gray watery matrix of sandy
mud. The matrix was almost entirely carbonate, with only
a trace of noncarbonate material. It was less compact
than the matrix in cores collected from the shallow and
intermediate zones (Aronson et al., 1998; Macintyre et al.,
2000; Aronson et al., 2002a). X-ray diffraction analysis
of sediment samples revealed a notable lack of high mag-
nesium calcite in the sand and silt fractions. The majority
of high magnesium calcite was found in the clay fraction,
corroborating our earlier conclusion of active precipita-
tion of micritic high magnesium calcite without significant
cementation (Macintyre and Aronson, 2006). Spines of
Echinometra viridis in the matrix indicated that those her-
bivores were present during the time interval represented
by the cores.

Three of the six cores we collected provided penetra-
tions, recoveries, and bottom dates sufficient to analyze
temporal trends in the coral assemblage of the deep zone
(Table 1). Cores BZ08-3 and BZ08-5 from Channel Cay
were both plugged by heads of Porites astreoides at pen-
etration depths of 2-3 m. Core BZ08-6 from Elbow Cays
penetrated nearly 3.5 m. Bottom samples consisting of
Porites astreoides from the bases of cores BZ08-3 and
BZ08-5, and plating Agaricia from the base of BZ08-6,
were radiocarbon dated. The remaining three cores yielded
recoveries of 65 cm or less and were not analyzed.

318 e

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

TABLE 1. Summary statistics for the three cores.

Basal radiocarbon dates *

Water Percent Conventional date Calibrated date
Core depth (m) Site Penetration (cm) Recovery (cm) recovery (14C year + SE) b (CalBP) ©
BZ08-3 14.5 Channel 256 31.6% 2,730 + 50 2,420 (1,860-1,560)
BZ08-5 16.2 Channel 216 36.1% 2,130 + 60 1,710 (2,650—2,320)
BZ08-6 15.3 Elbow 347 31.4% 1,290 + 60 840 (940-700)

4 Radiocarbon dates are of coral samples from the bases of the cores.

Conventional dates are measured dates corrected for isotopic fractionation, expressed as radiocarbon years before 1950 (14C year) and accompanied by standard errors (SE).
© Calibrated dates (CalBP) are expressed as calendar years before 1950, with 95% confidence intervals in parentheses.

The mean recovery for the three cores analyzed was
33.0% of penetration depth (£1.53 SD). This figure is
slightly lower than the mean of 35.9% obtained for cores
from intermediate depths on the rhomboid shoals and con-
siderably lower than the mean of 62.3% for cores from
intermediate depths in Bahia Almirante, a coastal lagoon
in Panama (Dardeau et al., 2000). The low recoveries in
the present study probably reflect the open reef framework
of the rhomboid shoals (compared to Bahia Almirante),
combined with the low sediment content of the matrix in
the deep zone (compared to intermediate depths on the
rhomboid shoals).

All three cores were dominated by agariciid cor-
als (Figure 2). These corals were primarily large, plating
forms, which characterized the living community until
1998 and that now characterize the modern, postbleach-
ing death assemblage in the deep zone. Agaricia tenuifo-
lia was more common near the tops of cores BZ08-5 and
BZ08-6 than lower in those cores. The agariciids were in
mixed taphonomic condition, with most intervals contain-
ing both intermediate and poor material. The skeletons
from the top 20 cm of the cores were in neither better nor
worse condition than those further down.

Slope angles in the vicinity of the coring sites, mea-
sured with an inclinometer (Aronson et al., 2002a), were
36°-39°. Those slopes were less steep than the critical an-
gle of 45°, above which Agaricia tenuifolia skeletons are
transported downslope (Aronson et al., 2002a). The criti-
cal angle of 45° probably applies to the dead, fragmented
skeletons of plating agariciids as well.

Core BZ08-3 contained a layer of Acropora cervicor-
nis branch fragments in poor taphonomic condition. This
layer could have been the result of downward transport
from intermediate depths. On the other hand, Acropora
cervicornis is less sensitive to slope angle than Agaricia

tenuifolia (Aronson et al., 2002b), so an autochthonous
layer cannot be ruled out. Other coral taxa, including
branching Porites spp. and Porites astreoides, Montas-
traea annularis species complex, Colpophyllia natans,
Madracis auretenra (formerly Madracis mirabilis; Locke

BZ08-5 BZ08-6

SS =

o
(—)

L
i—)

Depth in Core (cm)
o (3)

“I
So

QD Plating Agaricia

FS) Ag. tenuifolia

co
i=)

©
So

Lightly altered
Ac. cervicornis
100
Degraded Ac.
cervicornis

Porites spp.

GED) P. astreoides
ES Montastraea sp.

FIGURE 2. Schematic diagrams of the three extruded cores. The
lengths of the cores depart slightly from recoveries estimated in the
field (Table 1). Gray shading indicates a matrix of watery, sandy
mud; Ag. = Agaricia; Ac. = Acropora; P. = Porites.

et al., 2007), and Stephanocoenia intersepta, were rare in
the cores. None of the cores recorded millennial-scale in-
tervals of actively accreting Acropora cervicornis frame-
work, which were represented in the intermediate-zone
cores by thick accumulations of Acropora cervicornis in
good taphonomic condition.

DISCUSSION

Shinn et al. (1979; see also Westphall, 1986) extracted
cores from the flanks of the Channel Cay shoal, including
one from the base of the reef near our coring station (see
Figure 1). Their general statement, that the cores were
dominated by Acropora cervicornis with agariciids as the
subdominants, did not draw distinctions between cores
extracted from the different zones. We found that agari-
ciids were the dominant framework constituents in the
deep zone.

Core BZ08-3, which contained a layer of taphonomi-
cally degraded Acropora cervicornis underlying a thick
uppermost layer of agariciids, could represent a deepening-
upward sequence. This scenario seems unlikely, however,
considering that sea level has risen only approximately
2 m during the past 3,000 years (Toscano and Mac-
intyre, 2003). Furthermore, the other two cores showed
no such Acropora cervicornis-dominated layer. Regard-
less, none of the three cores suggests a recent transition
from millennia of fast-growing and rapidly accumulat-
ing Acropora cervicornis framework to dominance by
agariciid corals, as was observed in the cores from in-
termediate depths. The layer of Acropora cervicornis in
BZ08-3 is more likely derived from material that was
transported downslope, forming debris fans at the bases
of the shoals.

Aronson et al. (2005a) compared late Holocene reef
development between the rhomboid shoals and the un-
cemented reefs of Bahia Almirante in Panama. The shal-
low and intermediate zones had been cored extensively in
both locations, providing an accurate picture of stasis and
change in the dominant coral taxa. In both cases, however,
the deep zones were poorly characterized. The dearth of
push-cores from the bases of the reefs has been primarily
a consequence of the greater densities of core-occluding
massive corals in the subsurface, compared to the subsur-
face of the shallow and intermediate zones.

The cores analyzed in this study allow us to present
a more comprehensive model of the history of the coral
assemblages of the rhomboid shoals (Figure 3). In the
shallow zone, catch-up dynamics gave way to keep-up

NUMBER 38 e¢ 319

™) Plating Agaricia

SS Ag. tenuifolia

Lightly altered
Ac. cervicornis

= Degraded Ac.
cervicornis

yoN Massive corals

] Porites spp.

Depth (m)

20

FIGURE 3. Model of reef development on the rhomboid shoals of
Belize over the last several thousand years. Gray shading indicates
that the coral assemblages at intermediate water depths experienced
a recent transition. Black fill represents earlier Holocene and ante-
cedent Pleistocene reef framework at depths not penetrated by the
cores. Horizontally oriented, subsurface ellipses indicate spatially
isolated layers of Agaricia tenuifolia and taphonomically degraded
Acropora cervicornis. (Modified from Aronson et al., 2005a.)

dynamics: the Acropora cervicornis that had dominated
for millennia during the catch-up phase was replaced cen-
turies ago by branching Porites spp. during the keep-up
phase. Acropora cervicornis was also dominant for mil-
lennia at intermediate depths, but in the late 1980s it
was nearly extirpated by white-band disease and then re-
placed by Agaricia tenuifolia. The deep zone, in contrast,
appears to have been dominated by agariciids for at least
1,500 to 2,000 years. No recent transitions were evident
in the deep zone, a result consistent with the hypothesis
that such shifts were predicated on the prior dominance
and subsequent mortality of acroporids.

Thus, only the intermediate zone was affected when
Acropora cervicornis died off regionally in the late 1980s
to the early 1990s. The subsequent bleaching event in
1998 killed most of the agariciids on the rhomboid shoals.
Cores extracted from the intermediate zone in 2004 did
not display a taphonomic signature of that mass mortality
event, which would have appeared as a discrete, upper-
most layer of taphonomically degraded agariciid skeletons
(Aronson et al., 2005b). Similarly, because of the mixed
taphonomic character of the subfossil agariciid material
in the deep zone, the expected signature of the 1998 event
had not been observed in the Holocene record of that zone
as of April 2008.

320 e

ACKNOWLEDGMENTS

We thank M. E. Parrish for assistance with the field-
work and R. M. Moody and T. J. T. Murdoch for draft-
ing the illustrations. Fieldwork was supported by a grant
from the Smithsonian Institution’s Caribbean Coral Reef
Ecosystems (CCRE) program, and laboratory analysis
was funded by the Dauphin Island Sea Lab (DISL). K.
Riitzler and M. Carpenter of the CCRE facilitated our
fieldwork, which was based at the Smithsonian’s Car-
rie Bow Cay research station and carried out under a
research permit from the Belize Department of Fisher-
ies. This is contribution number 833 of the Caribbean
Coral Reef Ecosystems Program (CCRE), Smithsonian
Institution, supported in part by the Hunterdon Oceano-
graphic Research Fund; DISL contribution number 393;
and contribution number 8 from the Institute for Adap-
tation to Global Climate Change at the Florida Institute
of Technology.

LITERATURE CITED

Aronson, R. B., I. G. Macintyre, S. A. Lewis, and N. L. Hilbun. 200Sa.
Emergent Zonation and Geographic Convergence of Coral Reefs.
Ecology, 86:2586-2600.

Aronson, R. B., I. G. Macintyre, and W. F. Precht. 2005b. Event Preser-
vation in Lagoonal Reef Systems. Geology, 33:717-720.

Aronson, R. B., I. G. Macintyre, W. F. Precht, T. J. T. Murdoch, and C. M.
Wapnick. 2002a. The Expanding Scale of Species Turnover Events
on Coral Reefs in Belize. Ecological Monographs, 72:233-249.

Aronson, R. B., I. G. Macintyre, C. M. Wapnick, and M. W. O’Neill.
2004. Phase Shifts, Alternative States, and the Unprecedented Con-
vergence of Two Reef Systems. Ecology, 85:1876-1891.

Aronson, R. B., and W. F. Precht. 1997. Stasis, Biological Disturbance,
and Community Structure of a Holocene Coral Reef. Paleobiology,
23:326-346.

. 2001a. “Evolutionary Paleoecology of Caribbean Coral Reefs.”

In Evolutionary Paleoecology: The Ecological Context of Macro-

evolutionary Change, ed. W. D. Allmon and D. J. Bottjer, pp. 171-

233. New York: Columbia University Press.

. 2001b. White-Band Disease and the Changing Face of Carib-

bean Coral Reefs. Hydrobiologia, 460:25-38.

. 2006. Conservation, Precaution, and Coral Reefs. Coral Reefs,
25:441-450.

Aronson, R. B., W. F. Precht, and I. G. Macintyre. 1998. Extrinsic Con-
trol of Species Replacement on a Holocene Reef in Belize: The Role
of Coral Disease. Coral Reefs, 17:223-230.

Aronson, R. B., W. F. Precht, I. G. Macintyre, and T. J. T. Murdoch.
2000. Coral Bleach-Out in Belize. Nature (London), 405:36.
Aronson, R. B., W. F. Precht, T. J. T. Murdoch, and M. L. Robbart.
200Sc. Long-Term Persistence of Coral Assemblages on the Flower
Garden Banks, Northwestern Gulf of Mexico: Implications for Sci-

ence and Management. Gulf of Mexico Science, 23:84-94.

Aronson, R. B., W. F. Precht, M. A. Toscano, and K. H. Koltes. 2002b.
The 1998 Bleaching Event and Its Aftermath on a Coral Reef in
Belize. Marine Biology, 141:435-447.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Bruno, J. F., L. E. Petes, C. D. Harvell, and A. Hettinger. 2003. Nutrient
Enrichment Can Increase the Severity of Coral Diseases. Ecology
Letters, 6:1056-1061.

Bruno, J. F, and E. R. Selig. 2007. Regional Decline of Coral Cover
in the Indo-Pacific: Timing, Extent, and Subregional Comparisons.
PloS ONE, 2(8):e711, doi:10.1371/journal.pone.0000711.

Bruno, J. F., E. R. Selig, K. S. Casey, C. A. Page, B. L. Willis, C. D.
Harvell, H. Sweatman, and A. Melendy. 2007. Thermal Stress and
Coral Cover as Drivers of Coral Disease Outbreaks. PloS Biology,
5(6):e124, doi:10.1371/journal.pbio.0050124.

Bruno, J. F., H. Sweatman, W. FE. Precht, E. R. Selig, and V. G. W. Schutte.
2009. Assessing Evidence of Phase Shifts from Coral to Macroalgal
Dominance on Coral Reefs. Ecology, 90:1478-1484.

Buddemeier, R. W., J. A. Kleypas, and R. B. Aronson. 2004. Coral Reefs
and Global Climate Change: Potential Contributions of Climate
Change to Stresses on Coral Reef Ecosystems. Arlington, Va.: Pew
Center on Global Climate Change.

Bythell, J. C., Z. Starr-Hillis, and C. S. Rogers. 2000. Local Variabil-
ity but Landscape Stability in Coral Reef Communities Follow-
ing Repeated Hurricane Impacts. Marine Ecology Progress Series,
204:93-100.

Carpenter, K. E., M. Abrar, G. Aeby, R. B. Aronson, S. Banks, A. Bruck-
ner, A. Chiriboga, J. Cortés, J. C. Delbeek, L. DeVantier, G. J. Ed-
gar, A. J. Edwards, D. Fenner, H. M. Guzman, B. W. Hoeksema, G.
Hodgson, O. Johan, W. Y. Licuanan, S. R. Livingstone, E. R. Lovell,
J. A. Moore, D. O. Obura, D. Ochavillo, B. A. Polidoro, W. E
Precht, M. C. Quibilan, C. Reboton, Z. T. Richards, A. D. Rog-
ers, J. Sanciangco, A. Sheppard, C. Sheppard, J. Smith, S. Stuart,
E. Turak, J. E. N. Veron, C. Wallace, E. Weil, and E. Wood. 2008.
One-Third of Reef-Building Corals Face Elevated Extinction Risk
from Climate Change and Local Impacts. Science, 321:560-563.

Dardeau, M. R., R. B. Aronson, W. F. Precht, and I. G. Macintyre. 2000.
“Use of a Hand-Operated, Open-Barrel Corer to Sample Unce-
mented Holocene Coral Reefs.” In Diving for Science in the 21st
Century: 20th Annual Symposium of the American Academy of Un-
derwater Sciences, ed. P. Hallock and L. French, pp. 6-9. St. Peters-
burg: University of South Florida.

Gardner, T. A., I. M. Cété, J. A. Gill, A. Grant, and A. R. Watkinson.
2003. Long-Term Region-Wide Declines in Caribbean Corals. Sci-
ence, 301:958-960.

Glynn, P. W., and I. G. Macintyre. 1977. Growth Rate and Age of Coral
Reefs on the Pacific Coast of Panama. Proceedings of the Third
International Coral Reef Symposium, Miami, 2:251-259.

Green, D. H., P. J. Edmunds, and R. C. Carpenter. 2008. Increasing Rela-
tive Abundance of Porites astreoides on Caribbean Reefs Mediated
by an Overall Decline in Coral Cover. Marine Ecology Progress
Series, 359:1-10.

Greenstein, B. J., H. A. Curran, and J. M. Pandolfi. 1998. Shifting Eco-
logical Baselines and the Demise of Acropora cervicornis in the
Western Atlantic and Caribbean Province: A Pleistocene Perspec-
tive. Coral Reefs, 17:249-261.

Greenstein, B. J., and H. A. Moffat. 1996. Comparative Taphonomy of
Modern and Pleistocene Corals, San Salvador, Bahamas. Palaios,
11:57-63.

Greer, L., J. E. Jackson, H. A. Curran, T. P. Guilderson, and L. Teneva.
2009. How Vulnerable is Acropora cervicornis to Environmental
Change? Lessons from the Early to Middle Holocene. Geology,
37:263-266.

Halley, R. B., E. A. Shinn, J. H. Hudson, and B. Lidz. 1977. Recent and
Relict Topography of Boo Bee Patch Reef, Belize. Proceedings of the
Third International Coral Reef Symposium, Miami, 2:29-35.

Harvell, C. D., C. E. Mitchell, J. R. Ward, S. Altizer, A. P. Dobson, R. S.
Ostfeld, and M. D. Samuel. 2002. Climate Warming and Disease
Risks for Terrestrial and Marine Biota. Science, 296:2158-2162.

Hoegh-Guldberg, O., P. J. Mumby, A. J. Hooten, R. S. Steneck, P. Green-
field, E. Gomez, C. D. Harvell, P. F. Sale, A. J. Edwards, K. Caldiera,
N. Knowlton, C. M. Eakin, R. Iglesias-Prieto, N. Muthiga, R. H.
Bradbury, A. Dui, and M. E. Hatziolos. 2007. Coral Reefs un-
der Rapid Climate Change and Ocean Acidification. Science,
318:1737-1742.

Hogarth, W. T. 2006. Endangered and Threatened Species: Final List-
ing Determinations for Elkhorn Coral and Staghorn Coral. Federal
Register, 71:26852-26861.

Hubbard, D. K., H. Zankl, I. van Heerden, and I. P. Gill. 2005. Ho-
locene Reef Development along the Northeastern St. Croix Shelf,
Buck Island, U.S. Virgin Islands. Journal of Sedimentary Research,
75:97-113.

Hughes, T. P., A. H. Baird, D. R. Bellwood, M. Card, S. R. Connolly,
C. Folke, R. Grosberg, O. Hoegh-Guldberg, J. B. C. Jackson,
J. Kleypas, J. M. Lough, P. Marshall, N. Nystrom, S. R. Palumbi,
J. R. Pandolfi, B. Rosen, and J. Roughgarden. 2003. Climate
Change, Human Impacts, and the Resilience of Coral Reefs. Sci-
ence, 301:929-933.

Kleypas, J. A., R. W. Buddemeier, D. Archer, J.-P. Gattuso, C. Langdon,
and B. N. Opdyke. 1999. Geochemical Consequences of Increased
Atmospheric CO) on Coral Reefs. Science, 284:118-120.

Kline, D. I., N. M. Kuntz, M. Breitbart, N. Knowlton, and F. Rohwer.
2006. Role of Elevated Organic Carbon Levels and Microbial
Activity in Coral Mortality. Marine Ecology Progress Series, 314:
119-125.

Knowlton, N. 2001. The Future of Coral Reefs. Proceedings of the
National Academy of Sciences of the United States of America,
98:5419-5425.

Locke, J. M., E. Weil, and K. A. Coates. 2007. A Newly Documented
Species of Madracis (Scleractinia: Pocilloporidae) from the Carib-
bean. Proceedings of the Biological Society of Washington, 120:
214-226.

Macintyre, I. G., and R. B. Aronson. 2006. Lithified and Unlithified Mg-
Calcite Precipitates in Tropical Reef Environments. Journal of Sedi-
mentary Research, 76:81-90.

Macintyre, I. G., R. B. Burke, and R. Stuckenrath. 1981. Core Holes
in the Outer Fore Reef Off Carrie Bow Cay, Belize: A Key to the
Holocene History of the Belizean Barrier Reef Complex. Proceed-
ings of the Fourth International Coral Reef Symposium, Manila,
1:567-5S74.

Macintyre, I. G., W. FE. Precht, and R. B. Aronson. 2000. Origin of the
Pelican Cays Ponds, Belize. Atoll Research Bulletin, 466:1-12.
Ostrander, G. K., K. M. Armstrong, E. T. Knobbe, D. Gerace, and E. P.
Scully. 2000. Rapid Transition in the Structure of a Coral Reef
Community: The Effects of Coral Bleaching and Physical Distur-
bance. Proceedings of the National Academy of Sciences of the

United States of America, 97:5297-5302.

Pandolfi, J. M., A. W. Tudhope, G. Burr, J. Chappell, E. Edinger, M.
Frey, R. Steneck, C. Sharma, A. Yeates, M. Jennions, H. Lescinsky,
and A. Newton. 2006. Mass Mortality Following Disturbance
in Holocene Coral Reefs from Papua New Guinea. Geology,
34:949-952.

Precht, W. FE, and R. B. Aronson. 2004. Climate Flickers and Range
Shifts of Reef Corals. Frontiers in Ecology and the Environment,
2:307-314.

NUMBER 38 e¢ 321

. 2006. “Death and Resurrection of Caribbean Coral Reefs: A
Paleoecological Perspective.” In Coral Reef Conservation, ed. I. M.
Coté and J. D. Reynolds, pp. 40-77. Cambridge, UK: Cambridge
University Press.

Robbart, M. L., P. Peckol, S. P. Scordilis, H. A. Curran, and J. Brown-
Saracino. 2004. Population Dynamics and Differential Heat Shock
Protein Expression for the Corals Agaricia agaricites and A. tenui-
folia in Belize: Implications for Future ENSO-Related Thermal
Anomalies. Marine Ecology Progress Series, 283:150-160.

Rogers, C. S., and J. Miller. 2006. Permanent “Phase Shifts” or Revers-
ible Declines in Coral Cover? Lack of Recovery of Two Coral Reefs
in St. John, US Virgin Islands. Marine Ecology Progress Series,
306:103-114.

Rosenberg, E., and Y. Ben-Haim. 2002. Microbial Diseases of Corals and
Global Warming. Environmental Microbiology, 4:318-326.

Rutzler, K., S$. Duran, and C. Piantoni. 2007. Adaptation of Reef and
Mangrove Sponges to Stress: Evidence for Ecological Speciation
Exemplified by Chondrilla caribensis New Species (Demospongiae,
Chondrosida). Marine Ecology, 28(Suppl. 1):95-111.

Shinn, E. A., R. B. Halley, J. H. Hudson, B. Lidz, and D. M. Robbin.
1979. Three-Dimensional Aspects of Belize Patch Reefs (abstract).
American Association of Petroleum Geologists Bulletin, 63:528.

Shinn, E. A., J. H. Hudson, R. B. Halley, B. Lidz, D. M. Robbin, and
I. G. Macintyre. 1982. “Geology and Sediment Accumulation Rates
at Carrie Bow Cay, Belize.” In The Atlantic Barrier Reef System
at Carrie Bow Cay, Belize, I. Structure and Communities, ed. K.
Riitzler and I. G. Macintyre, pp. 63-75. Smithsonian Contributions
to the Marine Sciences, No. 12. Washington, D.C.: Smithsonian
Institution Press.

Sutherland, K. P., J. W. Porter, and C. Torres. 2004. Disease and Immu-
nity in Caribbean and Indo-Pacific Zooxanthellate Corals. Marine
Ecology Progress Series, 266:273-302.

Toscano, M. A., and I. G. Macintyre. 2003. Corrected Western-Atlantic
Sea-Level Curve for the Last 11,000 Years Based on Calibrated 'C
Dates from Acropora palmata Framework and Intertidal Mangrove
Peat. Coral Reefs, 22:257-270.

Wapnick, C. M., W. F. Precht, and R. B. Aronson. 2004. Millennial-Scale
Dynamics of Staghorn Coral in Discovery Bay, Jamaica. Ecology
Letters, 7:354-361.

Weil, E., G. Smith, and D. Gil-Agudelo. 2006. Status and Progress in
Coral Reef Disease Research. Diseases of Aquatic Organisms,
69:1-7.

Westphall, M. J. 1986. Anatomy and History of a Ringed-Reef Complex,
Belize, Central America. Master’s thesis, University of Miami, Coral
Gables, Fla.

Wilkinson, C., ed. 2000. Status of Coral Reefs of the World: 2000. Cape
Ferguson and Dampier, Australia: Australian Institute of Marine
Science.

. 2006. “Status of Coral Reefs of the World: Summary of Threats
and Remedial Action.” In Coral Reef Conservation, ed. 1. M. Coté
and J. D. Reynolds, pp. 3-39. Cambridge, UK: Cambridge Univer-
sity Press.

Williams, I. D., N. V. C. Polunin, and V. J. Hendrick. 2001. Limits to
Grazing by Herbivorous Fishes and the Impact of Low Coral Cover
on Macroalgal Abundance on a Coral Reef in Belize. Marine Ecol-
ogy Progress Series, 222:187-196.

blanc eee ee) at. +4 = | >!
i) 3 = = iG

evar oeetieec:
a) 2
ae) ithaca |

_RlemaeBs t ar tak a
Shi peste ae

7 ee: 4 We

eer ‘
a4)? Ohiian —
<.toet eeen err

5 7 shew cos

See ti iawiea fi ia
yA ver P| ea nati :

ena
acieled
soy 8 -
r 4 ge Rigo ;
Weve Rt it cm aati dis
Jha tech Cai
nen Pa 7 Uk ea
AES
ah Reg ce | ‘ om bzcd Svein Ae
bt icone. le A ete

a a” hon 5 a he wad bers =) p

> hii) oF wie iT ETA F

Climate and Hydrological Factors
Affecting Variation in Chlorophyll
Concentration and Water Clarity
in the Bahia Almirante, Panama

Rachel Collin, Luis D’Croz, Plinio Gondola,
and Juan B. Del Rosario

Rachel Collin, Luis D’Croz, Plinio Gondola, and
Juan B. Del Rosario, Smithsonian Tropical Re-
search Institute, MRC 0580-08, Unit 9100, Box
0948, DPO AA 34002, USA. Corresponding
author: R. Collin (collinr@si.edu). Manuscript re-
ceived 29 August 2008; accepted 20 April 2009.

ABSTRACT. Water clarity and productivity are fundamentally important for the dis-
tribution of tropical marine organisms. In the Caribbean, changes in nutrient loading
that result from rapid development are thought to have caused increased planktonic pro-
ductivity, reduced water clarity, and reduced reef and seagrass health. Here we analyze
chlorophyll a concentration and water clarity from eight years of environmental moni-
toring in Bocas del Toro, Panama. Chlorophyll a concentrations did not vary significantly
among the six sampled sites and showed no significant temporal changes, despite the
recent rapid development in the region, accompanied by scant wastewater treatment. In
contrast, water clarity increased significantly during the study period. Because chloro-
phyll a does not vary closely with water clarity, Secchi depths are likely to reflect changes
in suspended particulate matter rather than in phytoplankton biomass. Secchi depths
decreased with rainfall and wind speed but increased with solar radiation, supporting the
idea that clarity was not tightly linked to phytoplankton biomass. The decrease in annual
rainfall, but not wind speed, over the past eight years suggests that the long-term trend in
Secchi readings is the result of changes in rainfall patterns.

INTRODUCTION

Water clarity and productivity are fundamentally important to the distribu-
tion of tropical marine organisms, especially corals. Ocean primary productivity
is also important for global geochemistry and carbon sequestration (Falkowski et
al., 1998). Global warming and increase in atmospheric CO) are expected to influ-
ence the distribution of the biota, as well as its abundance, and the photosynthetic
activity of phytoplankton (Falkowski et al., 1998). SeaWiFS satellite imagery
shows that worldwide oceanic chlorophyll a concentrations are about 0.2 mg/m?
(Yoder et al., 1993) and can reach 5 mg/m? in coastal upwelling zones (Falkowski
et al., 1991; Walsh et al., 1978). It is difficult to use this method to obtain infor-
mation on chlorophyll a concentrations for many onshore tropical areas because
accurate remote sensing is difficult in coastal areas with large sediment input and
because many tropical regions have high frequencies of cloud cover. In such areas
field measurements of water clarity and chlorophyll a concentrations are vital for
assessing short-term variation and ground-truthing remote measurements.

324 e

Coral reef environments are particularly sensitive to
changes in water quality, especially changes in nutrients,
sediment load, and productivity. The paradigm of coral
reef biology is that reef development and coral health are
greatest in areas with low sedimentation, low primary pro-
ductivity, low abundances of zooplankton, and high water
clarity. These habitats are most conspicuous in the Indo-
Pacific and the offshore islands in the Caribbean. In many
locations these habitats are suffering from reduction of wa-
ter quality associated with coastal development (Bell, 1992;
Lapointe, 1992). In the Caribbean, most studies of reefs and
their waters are conducted in the Bahamas, Puerto Rico,
Netherlands Antilles, and other offshore islands (Gilbes et
al., 1996; Otero and Carbery, 2005; van Duy] et al., 2002;
Webber et al., 2003). In addition there have been some stud-
ies of the unusual upwelling sites along the coast of Ven-
ezuela and Colombia (Franco-Herrera et al., 2006) and the
strongly freshwater-influenced regions around the Yucatan

N

a ee
W Ef Se

Bahia
Almirante

9°20'0"N

Cristobal
. >

Ty

Pastores
@

. Caribbean Sea

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

(Herrera-Silveira et al., 2002). However, few studies have
examined heavily terrestrially influenced systems without
these unusual features in the Caribbean. Here we report the
results of eight years of physical climatic and water quality
monitoring in Bahia Almirante, an enclosed Caribbean ar-
chipelago that is highly terrestrially influenced.

STUDY LOCATION:
BOCAS DEL TORO, PANAMA

Three bodies of water surround the Bocas del Toro
Archipelago on the Caribbean coast of Panama: the Bahia
Almirante and the Laguna de Chiriqui on the landward
side, and the Caribbean Sea on the exposed coastal side
(Figure 1). The mainland surrounding the region is largely
forested, although the completion of a road linking Costa
Rica to Bocas del Toro and the rest of Panama in the year

82100"W]

kilometers

@
Bottomwood

Zapatilla
&

FIGURE 1. Map of the Bahia Almirante region with sampling sites indicated. BIRS = Bocas del Toro Re-
search Station; CSS = CARICOMP seagrass site; MT = instrument platform; STRI = Smithsonian Tropical

Research Institute.

2000 has resulted in increased lowland deforestation, as
well as land development for small farms and tourism.
The landward sides of the islands are fringed with largely
intact red mangrove forests, although these are also being
cleared from the landward side. The Laguna de Chiriqui,
which receives twice the freshwater runoff of Bahia Almi-
rante, has higher nutrient loads and limited coral reef
development (D’Croz et al., 2005) and is not discussed
further here.

The Bahia Almirante has significantly more oceanic
influence than the Laguna de Chiriqui and supports well-
developed coral reefs. It receives runoff from only two riv-
ers of any note. High rainfall (about 3 m/year) and run-
off from the San San Pond Sac peat swamp forest often
result in pronounced haloclines with low-salinity waters
(which are often cold) overlying full-salinity bottom waters
(Kaufmann and Thompson, 2005). Surface salinities are
generally 30-34 PSS (practical salinity scale) but can drop
as low as 20 PSS after heavy rains (Kaufmann and Thomp-
son, 2005). Other conditions are relatively aseasonal, with
mean sea surface temperatures varying from 27.5°C
(January-February) to 29.7°C (September—October). Aver-
age wind speed is about 7 km/h but occasionally exceeds
20 km/h (Kaufmann and Thompson, 2005). Despite the
high levels of runoff, N:P was always recorded to be below
Redfield ratios in a previous survey of the area (D’Croz
et al., 2005). This observation suggests that primary pro-
ductivity could be nutrient limited and that inputs from
freshwater runoff or wind mixing could fertilize primary
productivity in the Bahia Almirante.

The Smithsonian Tropical Research Institute’s Bocas
del Toro Research Station is on the Bahia Almirante side
of Isla Colon. As part of the development of the scientific
knowledge base of the station, and as part of the CARI-
COMP program, various physical and biological features
of the surrounding environment have been monitored
since 1999. Here we examine these data (1) to develop a
baseline to which future studies can be compared, (2) to
determine if the recent rapid development of the region
has had an effect on water clarity and phytoplankton bio-
mass, and (3) to explore the physical data to understand
what factors influence the variation in these parameters.

MATERIALS AND METHODS
MONITORING HIStory

Isla Colon is the site of the Smithsonian Tropical
Research Institute’s Bocas del Toro Research Station. At
its inception in 1998 a long-term physical and biological

NUMBER 38 ¢ 325

monitoring program was initiated. Physical records of air
and water temperature, rainfall, salinity, solar radiation,
and wind speed have been kept since 1999 (reviewed in
Kaufmann and Thompson, 2005). Monitoring of Secchi
depths and chlorophyll a concentrations was conducted
approximately biweekly at five sites (see Figure 1) from
1999 until 2001. The sampling intervals were not equal
(ranging from 7 to 28 days), so these data were not ap-
propriate for time-series analyses.

The Secchi depths and chlorophyll a monitoring was
reinitiated at three of these sites and at an additional site
in 2006 and continues to be measured weekly. At one of
these, the CARICOMP reef monitoring site (described
in Guzman et al., 2005), the Secchi depths have been re-
corded weekly since 2000. At the CARICOMP seagrass
site, horizontal Secchi readings have been taken weekly
since 1999. During the entire period, measurements were
made by the same three-person team.

SAMPLING LOCALITIES

Sampling sites (Figure 1; Table 1) were chosen in 1999
to include a range of environments. In 2006 sites were
chosen to include an onshore-offshore gradient, in which
we expected more oceanic conditions on one end and ter-
restrially influenced conditions on the other end.

e Colon, 6.3 km northeast of Bocas del Toro Town, is
the most exposed site. The bottom at 20.5 m is muddy.
Rough conditions occasionally made it impossible to
take measurements in this location.

e Cristobal, in the middle of the Almirante Bay, is a site
7 km from the mainland and surrounded by patch reefs.
The bottom at 25 m is muddy.

e Pastores is a semienclosed bay, 500 m from the main-
land. It is more heavily influenced by continental runoff
and creek discharge than the other sites. Depth at the
sampling site is 26 m but a nearby coral reef slopes from
5 to 16 m. Jellyfishes are abundant at this site.

e Smithsonian Tropical Research Institute (STRI) is the site
closest to the Bocas del Toro Research Station, 500 m
from the shore. This site serves as the water monitor-
ing site for the CARICOMP reef site, which is onshore
of this location, over a reef that slopes from 5 to 20 m.
The bottom is muddy and sandy with isolated patches
of coral.

¢ Bottomwood, between Solarte and San Cristobal Islands,
is protected from oceanic influence. The sampling site is
near mangrove islets, sand cays, and a shallow coral reef.
The reef slopes to a fine sand bottom at 16 m.

326 * SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

e Zapatillas has the highest diversity and abundance of
coral and octocoral species of any of our sampling sites.
The bottom at about 15 m is mostly covered by patch
reefs and fine sand.

e The CARICOMP seagrass site is several hundred meters
along the shore to the northwest of the Bocas del Toro
Research Station. This shallow (2 m depth) location has
extensive Thalassia cover, and the small bay is fringed
by red mangroves.

HYDROLOGICAL MEASURES

Water temperature and salinity were recorded with
an YSI 85 multiparameter probe (Yellow Springs Instru-
ments, Yellow Springs, Ohio, USA) at the same time and
depth as the seawater was sampled. Measurements were
taken at approximately 50 cm. Dissolved oxygen was
also measured in 2006-2008. Salinity is expressed in the
practical salinity scale (PSS) and dissolved oxygen in mil-
ligrams per liter.

CLIMATE RECORDS

Rainfall, solar radiation, and wind speed are moni-
tored continuously at the Bocas del Toro Research Sta-
tion, as described by Kaufmann and Thompson (2005).
These measurements are taken close to the STRI site (see
Figure 1). For the purposes of this study average rain-
fall, solar radiation, and wind speed were calculated for
3 days and 6 days before each sampling day. We chose

these periods because Beman et al. (2005) showed that
phytoplankton blooms can peak 3 to 5 days after nutrient
input from terrestrial runoff.

Annual rainfall was obtained in two ways. First, an
hourly tipping bucket measured rainfall from 2002. Be-
cause data are incomplete for three of the years (including
2008), we calculated the average daily rainfall to stan-
dardize across the years. The second estimates were from
the Bocas del Toro airport. These records extend to 1999
and were also converted to annual daily averages.

SECCH! DEPTHS

Water clarity was measured by lowering a 30 cm di-
ameter Secchi disk into the water until it was no longer
seen and then raised until it reappeared. The Secchi depth
was measured according to the length of the submerged
rope. This operation was repeated three times at each site
during each measurement. At the seagrass site the Secchi
was measured horizontally, underwater at 0.5 m depth,
and was read with a dive mask.

CHLOROPHYLL A

Three replicate water samples were collected by hand
at 50 cm below the surface in polyethylene bottles and
placed in a cooler for the return to the laboratory. Two
liters of each replicate were vacuum filtered on Whatman
GF/F (0.7 um pore size). Filters were wrapped in aluminum
foil and stored frozen (—20°C). A Teflon pestle was used

TABLE 1. Study site locations and the data available for each site.

Site Location Secchi depths Chlorophyll a
Colon SI BYVEN| 1999-2001 1999-2001
82°12'37'W 2006-2008 2006-2008
Cristobal 9°18'1S"N 2006-2008 2006-2008
82°17'55"W - =
Pastores 9°12'36"N 1999-2001 1999-2001
82°19'37'W 2006-2008 2006-2008
STRI 2 9°20'40"N 2000-2008 1999-2001
82°16'39'W - 2006-2008
Bottomwood 9°17'47'N 1999-2001 1999-2001
82°13'25"W - =
Zapatilla OPUS QIN | 1999-2001 1999-2001
82°06'19"W - =
CARICOMP seagrass 9°21'06"N 1999-2008 -
82°15'29"W Horizontal

4 STRI, Smithsonian Tropical Research Institute.

to grind the filters in 5 mL 90% aqueous acetone solution.
The slurry was transferred to 15 mL polypropylene screw-
cap centrifuge tubes and filled to 10 mL with acetone. The
tubes were kept in the dark at —20°C for 24 h. Extracts
were centrifuged at 3,000 rpm, and the supernatant was
analyzed for chlorophyll a following the nonacidification
fluorometric method (Welschmeyer, 1994).

STATISTICAL ANALYSIS

Correlation analyses and multiple regression analyses
were used to describe the relationships between the vari-
ables of interest (Secchi depth and chlorophyll a concen-
tration) and the hydrological data and the climate data.
Student’s ¢ test, analysis of variance (ANOVA), and analysis
of covariance (ANCOVA) were used to test for differences
between sites and sampling periods. Because it is likely that
there is a lag in the response of phytoplankton to the input
of nutrients from river runoff or turbulence, we looked for
correlations between Secchi depths or chlorophyll a con-
centrations and the average of rainfall, solar radiation, and
wind speed over the previous 3 and 6 days. Because the 3
day and 6 day results did not differ substantively, only the
results using the 3 day average are reported here. Because
rainfall and cloud cover are patchy on a local scale, we only
examined climatic variables for the STRI and CARICOMP
seagrass sites. Time-series autocorrelation analyses were ap-
plied to Secchi depth and chlorophyll a data. The few weeks
of missing data were filled with the averages values for the
time series under analysis.

NUMBER 38 ¢ 327

RESULTS

A number of complex relationships were demonstrated
between the hydrological parameters, Secchi depths, and
chlorophyll a concentrations at the different sites. Several
of these relationships vary among the sites, and there are
a number of interactions between factors; however, the
following generalizations can be made. (1) Secchi depths
increased with temperature, salinity, and solar radiation,
and decreased with rainfall, wind speed, and chlorophyll a
concentration. (2) Correlations between any hydrologi-
cal characteristic and Secchi depths or chlorophyll a were
low (r? rarely exceeding 0.10) but were higher for all the
climatic variables (rain and solar radiation had r* up to
0.22). (3) Chlorophyll a concentrations showed no con-
sistent temporal or spatial patterns. (4) Secchi depths were
not tightly correlated with chlorophyll a concentrations.
(5) Secchi depths increased and rainfall decreased through-
out the study.

HYDROLOGICAL CONDITIONS AT EACH SITE

Hydrological parameters varied somewhat among the
six sites (Table 2, Figure 2). Salinity was significantly differ-
ent at all sites (ANOVA with post hoc t test; Table 2), with
the lowest average salinity in Pastores, the most inland site,
and the highest average salinity in Colon, the most oceanic
site. For 1999-2001 the average temperature at Pastores
was significantly higher than the other sites and the tem-
perature at Colon was significantly lower. The temperature

TABLE 2. Summary of physical and biological data from 1999-2001 and 2006-2008.

Dissolved Significant changes
Temperature, Salinity,* oxygen, Secchi depth, Chlorophyll a, between periods
Site Years °C (SD) PSS (SD) mg/L (SD) m (SD) mg/m? (SD) (t test)
Colon 1999-2001 28.1 (1.26) 33.0 (1.70) - 9.1 (3.6) 0.44 (0.19) Temperature increased
2006-2008 28.6 (0.89) Ss)o9) ((le3)i))) 5.80 (0.43) 10.0 (3.6) 0.47 (0.23)
Cristobal 2006-2008 28.8 (0.95) 32.9 (1.46) 5.83 (0.49) 12.4 (3.2) 0.46 (0.25) NA
Pastores 1999-2001 28.6 (1.29) 31.9 (2.15) - 11.0 (3.3) 0.46 (0.24) Temperature and Secchi
2006-2008 29.2 (0.98) 32.4 (1.76) §.81 (0.51) 13.2 (3.0) 0.49 (0.28) depth increased
STRI 1999-2001 28.3 (1.16) 32.9 (1.33) - 10.9 (3.9) 0.37 (0.24) Temperature and Secchi
2006-2008 28.7 (0.91) 33.0 (1.34) 5.78 (0.49) 13.2 (3.6) 0.43 (0.23) depth increased
Bottomwood 1999-2001 28.3 (1.24) 32.7 (1.43) AEST) 0.36 (0.19) NA
Zapatillas 1999-2001 28.3 (1.28) 33.0 (2.78) 11.3 (2.6) 0.46 (0.22) NA

4 PSS = practical salinity scale.

328 °

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

C

a 18
2) —
= £
5 £12
8 o
® mo}
= = 6l|A BA Y BEY Y Ble B
s FA C C
72)
Colon Cristobal Pastores STRI Bottomwood Zapatilla L Colon Cristobal Pastores STRI Bottomwood Zapatilla
2000-2002
B Ml 2006-2008
36 D
— 0.8
E
2% Pos
BB E
2 28 iS
eA Re Y Bie z w Be A = 04
B 24 C & 9o/|A A A e A
2 | IB
=
20 0

Colon Cristobal Pastores STRI Bottomwood Zapatilla

Colon Cristobal Pastores STRI Bottomwood Zapatilla

FIGURE 2. Averages of temperature (A), salinity (B), Secchi depths (C), and chlorophyll concentration (D) from the two sampling periods (1999
data excluded). Student’s ¢ tests showed significant increases in temperature between periods at Colon, Pastores, and STRI and increases in Secchi
depth at Pastores and STRI. Single-factor analysis of variance (ANOVA) detected significant (P < 0.01) site effects. Significant differences between
groups of sites within either sampling period are indicated with letters, so that bars both labeled with “A” are not significantly different from each

other but are different from those otherwise labeled with post hoc tests. Specific letters were assigned arbitrarily, but A-C refer to 2000-2002 data
and W-Z refer to 2006-2008 data. Bars = one standard deviation (1 SD) of the mean. (Salinity is expressed in the practical salinity scale, PSS.)

at Pastores was also significantly higher than at the other
sites in 2006-2008, but there were no significant differ-
ences between the remaining sites. Dissolved oxygen did
not differ between sites.

Temperature increased significantly between the two
time periods at the three sites for which data were avail-
able over both periods (t test, P < 0.002 for each site),
despite an overall temperature decrease during the 2006-
2008 period. Salinity did not show a significant temporal
trend during either time period nor did it differ between
the two periods. Dissolved oxygen was only measured for
the 2006-2008 period, where it showed no temporal trend.
Eight years of data from climatic monitoring at the Bocas
Research Station instrument platform shows a downward
trend in rainfall, but little change in average solar radia-
tion or average wind speed (Figure 3).

FACTORS AFFECTING SECCHI DEPTHS

Secchi depths ranged from 2 to 22 m, and the depths
varied substantially from week to week (Figure 4). Sec-

chi depths showed significant effects of site, and signifi-
cant associations with temperature, salinity and chloro-
phyll a concentrations during both the 1999-2001 and
2006-2008 periods (ANCOVA; Table 3). The correla-

tion of any one variable with water clarity was low, with

155 15 iS
=
oO
= 124 12 12s
5 ole e
Zz 94E9 98
c = ; . ——— ®
S =z @ Airport rain = a
& 64 £6 4 6 =
5 & @ STRI rain z
+ amin (hy «abe | Hee RRM WER ay Gente 7 eer rere ee ee
e A Wind speed are ia ET A ars ee =
& 3 3 3 5
8 @ Solar radiation

0 0 1 - 0

T
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
SS TE

FIGURE 3. Long-term trends in climatic variables. Yearly aver-
ages for rainfall, solar radiation, and wind speed during the past
8 years show the decline in average daily rainfall. Daily averages
are used because missing data prevent the use of cumulative data.
(Rainfall is mm/d.)

AWWA
SANT STRI

E RIV UW

A\_[N\_ aA
a 2 Vi w vil y PAR ial ANN A \\ A Pastores
OD I Wha nA. ANG Waa AaN Gristobal
20 VOV TW a
8 si nes :
5 1 ANG r\ : J f. Z\ Nl. Col
BOTW V A a ae

\

Chlorophyll @2 (mg/m?)

0
5/25/06 9/25/06 1/25/07 5/25/07 9/25/07 1/25/08 5/25/08

FIGURE 4. Variation in Secchi depths (top graph) and chlorophyll a
(bottom graph) concentrations over time for the 2006-2008 dataset.
Straight line is the trend line of depth or concentration on that date.
Superimposed lines show individual variation.

the highest r* value for temperature at 0.13-0.17. Sec-
chi depths increased with temperature (Ordinary Least
Squares regression [OLS]: 1999-2001, r* = 0.13, P <
0.001; and 2006-2008, r* = 0.17, P < 0.001), and
salinity (OLS: 1999-2001, r* = 0.10, P < 0.001; and
2006-2008, r* = 0.06, P < 0.001) and decreased with
chlorophyll a concentration (OLS: 1999-2001, r? = 0.09,
P < 0.0001; and 2006-2008, r= 0.06, P < 0.001). The
average Secchi depth was significantly lower for the ex-
posed Colon site than for the other sites in both time
periods (see Table 2, Figure 2). Analysis of the data from
the two different periods showed different combinations
of interaction effects (see Table 3).

Climatic variables were more tightly correlated with
Secchi depth than were the seawater variables. Using the
six years of complete climate data from the STRI site, we
found that Secchi depths at the STRI site show significant
effect of year (P < 0.0001), a marginal effect of rainfall
(P = 0.056), and significant effects of solar radiation (P
= 0.002) and wind speed (P < 0.001), but no significant
interactions between these factors. For the CARICOMP
seagrass site there was no effect of year, but rainfall and
wind speed over the prior 3 days were significant (P <
0.01), as well as the interaction between rainfall and so-

NUMBER 38 °° 329

lar radiation (P < 0.0008). Secchi depth decreased with
the amount of rainfall (r* = 0.22 and 0.23, respectively,
with P < 0.001) and wind speed (r* = 0.21; P < 0.001)
at both sites and increased with solar radiation (r* =
0.21 and 0.15, respectively, with P < 0.0001) at the STRI
site but was only significant by its interaction with rain-
fall in the seagrass site. The interaction at the seagrass
site showed that Secchi distances decreased more quickly
with rainfall at high levels of solar radiation than at low
solar radiation.

Secchi depths increased over the long term: they in-
creased from 1999-2001 to 2006-2008 at Pastores and
STRI (¢ test, P < 0.0001 for both) but not at Colon. Least
squares regression showed a significant increase in Secchi
depths (r* = 0.02; nm = 381; P < 0.0002; slope = 0.3)
with date over the 8 years of weekly sampling at STRI.
The horizontal Secchi data from the nearby CARICOMP

TABLE 3. Analysis of covariance (ANCOVA) effects of physi-
cal variables on chlorophyll a concentration and Secchi depth
in 1999-2001 and 2006-2008 data after stepwise removal of
nonsignificant variables.

Sum of
Source df* squares F ratio P

Secchi depth, 1999-2001

Site 4 235.03 6.78 <0.0001
Temperature 1 100.97 11.65 0.0008
Salinity 1 290.63 33.53 <0.0001
Chlorophyll a concentration 1 150.82 17.40 <0.0001
Site* salinity > 4 104.02 3.00 0.02
Salinity * temperature > 1 67.31 7.76 0.006
Secchi depth, 2006-2008
Site 3 556.41 21.77. <0.0001
Temperature 1 454.35 $3.33 <0.0001
Salinity 1 216.91 25.46 <0.0001
Oxygen 1 2.86 0.33 0.56
Chlorophyll a concentration 1 47.15 5.53 0.02
Site* temperature > 3 70.34 DPS) 0.04
Temperature* oxygen > 1 49.58 5.82 0.02
Chlorophyll a, 1999-2001
Site 4 0.38 2.30 0.06
Temperature 1 1.24 30.11 <0.0001
Salinity 1 0.06 1.43 0.23
Site* salinity > 4 0.55 3.34 0.01
Salinity * temperature > 1 0.31 7.58 0.006
Chlorophyll a, 2006-2008
Temperature 1 0.26 4.90 0.03
Salinity 1 2.41 45.35 <0.0001
Oxygen 1 0.25 4.67 0.03
a df = Degrees of freedom.
b* — Run with only two-way interactions.

330 °

seagrass site show no long-term trend. When these values
are binned by month, there is a marginal effect of month
on the Secchi depths in the seagrass site (P = 0.07) and a
significant effect of month at the reef site (P = 0.0007).
Greater Secchi depths were recorded from drier months
and sunnier months (Figure 5), a result also found by
Kaufmann and Thompson (2005).

FACTORS AFFECTING CHLOROPHYLL A CONCENTRATIONS

Chlorophyll a concentration varied between 0.04
and 1.66 mg/m’. Similar to Secchi depths, concentrations
varied substantially from week to week and with no clear
seasonal component to the variation (see Figure 4). Dur-
ing the first 14 sampling dates of the 1999-2001 study
period, chlorophyll a concentrations were measured with
a spectrophotometer. An ANOVA testing for effects of
site and method showed that the results from the spectro-
photometer were significantly higher (site: F = 3.96; df
= 4; P < 0.005; method: F = 26.8; df = 1; P < 0.0001;
n = 328). Therefore values obtained from the spectro-
photometer were excluded from the subsequent analyses
and this dataset included only data from 2000-2001 or
2006-2008.

Although the average chlorophyll a concentrations
did not differ between the two periods (¢ test, P > 0.05),
there were different patterns for the two sampling pe-
riods. The only common results were that chlorophyll a
concentration decreased with temperature, and that the
variables examined explained no more than 10% of the
variance in chlorophyll a concentrations. Data from
2000-2001 showed a significant effect of temperature
and a marginal effect of site, but no effect of salinity
(ANCOVA; see Table 3). There were significant inter-
actions between site and salinity and between site and tem-
perature (Table 3). Overall, chlorophyll a concentrations
decreased with temperature (OLS: n = 328; r* = 0.06,
P < (0.0001). Results from 2006-2008 were different:
there were significant effects of temperature and salinity,
but not of site or oxygen concentration, nor were there
significant interactions (Table 3). Chlorophyll a concen-
trations decreased with temperature (OLS: 7 = 402; r* =
0.03, P = 0.002) and salinity (r? = 0.11, P < 0.0001) but
these factors explained very little of the variation.

Climate data were poorly linked to chlorophyll a
concentrations. Average wind speed for the 3 days before
sampling was positively correlated with chlorophyll con-
centration (r* = 0.19, P < 0.005) but this appeared to be
caused by a few periods with extremely high winds. Chlo-

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

STRI Secchi

=12
240
28
£
8 6
a 4
2
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Seagrass Secchi
16
14
E12
= 10
o 8
mo}
= 6
(5)
o 4
no
2
0

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Rainfall

E
=
3
‘E 200
oO
i
100
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Solar Radiation
6000
5
3% 5000
*E
= 4000
=
5 3000
8
@ 2000
&
> 1000
7)
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Wind Speed
14
2
=
—£ 10
—
3 8
o
a 6
= 4
= 2
0

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

FIGURE 5. Bar graph with monthly averages of Secchi depths, rain-
fall, solar radiation, and wind speed for the 8-year dataset showing
that months with higher average Secchi depths also had lower aver-
age rainfall. Bars = 1 SD of the mean.

rophyll a concentration was independent of both rainfall
over 3 days prior and average solar radiation over the 3
days before the measurements.

TIME SERIES

Both the Secchi depths and chlorophyll a concentrations
varied considerably from week to week (see Figure 4). To
determine if this variation has a temporal autocorrelation,
we conducted a time-series analysis. For 1999-2008 Sec-
chi depth is temporally auto-correlated at both the CARI-
COMP seagrass and the STRI sites (seagrass: Fisher’s
kappa = 9.9, P < 0.01; for coral: Fisher’s kappa = 12.2,
P < 0.001). Over the shorter period, 2006-2008, Secchi
depth was temporally auto-correlated at Colon (Fisher’s
kappa = 8.59, P < 0.01), Cristobal (Fisher’s kappa = 7.39,
P < 0.02), and STRI (Fisher’s kappa = 6.91, P < 0.04)
but not Pastores (P > 0.05). Chlorophyll a concentrations,

Coral Secchi (2000-2008)

1.0
0.8
0.6
0.4
0.2
0.0
em

11 16 21 26 31 36 41 46 51 56

Autocorrelation

Lag (weeks)

Seagrass Site (2000-2008)

1.0
0.8
0.6
0.4
0.2
0.0
-0.2

11 16 21 26 31 36 41 46 51

Autocorrelation

Lag (weeks)

NUMBER 38 °¢ 331

on the other hand, showed an autocorrelation only for Co-
lon (P < 0.005). Examination of the autocorrelation func-
tion shows that, over the short term (lag of up to several
months), the autocorrelation function appears stationary
(Figure 6). However, a peak around the 52 week lag (Fig-
ure 6) is evidence of seasonal externally driven periodicity
and suggests an annual cycle that is not obvious from plots
of the raw data (see Figure 4).

DISCUSSION

The overall values for the data presented here are
similar to those reported for the Bahia Almirante by
D’Croz et al. (2005), Kaufmann and Thompson (2005),
and Carruthers et al. (2005). We report some long-
term trends that were not detected by Kaufmann and
Thompson, who closely examined the patterns of daily

Chlorophyll a at STRI (2006-2008)

Cc

[=
2
—
&
2
‘=
(eo)
[S)
fe)
—_
=>
og
6 11 16 21 26 31 36 41 46 51 56
Lag (weeks)
Chlorophyll a at Colon (2006-2008)

1.0 D
¢
6 0.8
S 0.6
(3)
= 0.4
8
eo 0.2
3
q 0.0

-0.2

Ape Atte 160021) 26031) 36) 41) 146). (51156

Lag (weeks)

FIGURE 6. Autocorrelation function for 8 years of Secchi data for STRI and CARICOMP seagrass sites and chlorophyll a concentrations for 2
years for Colon (significant autocorrelation) and STRI (no significant autocorrelation) sites.

332 °¢

and monthly variation in physical parameters. Compar-
ing the YSI 85 probe measures of temperature and sa-
linity between the 2000-2002 sampling period and the
2006-2008 period, we found significant increases in wa-
ter temperature and salinity. This finding appears to be
associated with the recent trend toward lower rainfall in
the region.

Annual rainfall is typically high, in excess of 3,000
mm, in the Bahia Almirante, and the mean freshwater run-
off is approximately 1,600 mm per year (IGNTG, 1988).
However, average daily rainfall per year dropped from
12.48 mm in 2002 to 7.91 mm in 2007 (see Figure 3).
Reduced rainfall likely affected the hydrological condi-
tions in Bahia Almirante, which result from the interaction
between river discharge and ocean intrusion (D’Croz et
al., 2005). During the 1999-2001 sampling period, rain-
fall was high and salinity showed the typical increasing
trend from Pastores, the site nearest the mainland, to the
ocean-exposed sites at Colon and Zapatilla. This pattern
in surface salinity is consistent with the expected high dilu-
tion at nearshore sites resulting from river discharge into
the bay. The inshore-to-offshore salinity gradient was not
apparent during the 2006-2008 sampling period, presum-
ably because of the reduction of river discharge and conse-
quent greater influence of salinity from open ocean waters
(see Figure 2).

LONG-TERM TREND IN SECCHI DEPTHS

The most striking long-term trend we detected was
the surprising increase in Secchi disk depths. During the 8
years of monitoring, visibility has increased by 2 m (a rate
of 0.25 m/year) for several of the sites. Long-term trends
of decreased Secchi depths have been reported for moni-
toring in other areas. For example, in a dataset from the
Baltic Sea spanning 77 years, Secchi depths have decreased
0.05 m/year (Sandén and Hakansson, 1996), and a de-
crease of 0.03 m/year was reported in the Menai Strait in
Wales (Kratzer et al., 2003). The few reports of increased
Secchi depths were associated with bioremediation or ef-
forts to reduce untreated sewage outfall. For example, Sec-
chi depths increased at 0.05 m/year in Narragansett Bay,
Rhode Island, coincident with reductions in anthropogenic
total suspended solids (Borkman and Smayda, 1998), as
at one of several sampled sites in the Southern Califor-
nia Bight (Convers and McGowan, 1994). Our measures
show a much more rapid change in Secchi depths than
these previous studies.

The observed changes in Secchi depths were not in the
expected direction. A number of complicated, interacting

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

factors can influence water clarity, as measured by Secchi
disk, but many of them would indicate a decrease in Sec-
chi depth. The ongoing rapid development of tourism in
Bocas del Toro, particularly on Isla Colon, is accompanied
by an increase in wastewater input to the Bahia Almirante.
Changes in Secchi readings can reflect changes in particu-
late matter (from runoff or wind-induced turbidity) but
can also be caused by changes in phytoplankton biomass
or yellow pigments (mostly humic and fulvic acids) in the
water. Deforestation and coastal development can affect
all three of these factors. Inputs from untreated waste-
water as well as runoff from deforested areas can increase
the nutrients, particulate matter, and yellow pigments in
the water. In addition, increased nutrients often lead to
increased primary productivity, which can result in higher
standing phytoplankton biomass. These anthropogenic ef-
fects have been increasingly affecting coral reef habitats
throughout the Caribbean, where wastewater disposal is
the leading cause for eutrophication and decreased water
clarity (Szmant, 2002). We had expected to see a long-term
reduction in water clarity as a result of similar changes in
Bocas del Toro.

Secchi depth is often strongly correlated with chloro-
phyll a concentration and has been used as a proxy for pro-
ductivity in highly seasonal upwelling zones or temperate
lakes. This method is often favored because it is cheaper,
faster, and easier to obtain than a quantification of chloro-
phyll a concentration. Sandén and Hakansson (1996) re-
viewed four studies as well as their own data that showed
a relationship between Secchi depths and chlorophyll a
concentrations. The relationships are reported as power
functions and show chlorophyll a to scale with Secchi
depth to the 1.47-2.6 power. Megard and Berman (1989)
showed that the proportion of light attenuation caused
by chlorophyll concentration differed between neritic and
pelagic seawater, but there were clear relationships none-
theless. Here we found that chlorophyll a concentration
explained only 6%-9% of the variance in Secchi depths.
In addition, mean chlorophyll a concentrations were rela-
tively low in the Almirante Bay, near 0.5 mg m~?, which is
the suggested threshold value for oligotrophic conditions
required for coral reef development (Bell, 1992). There-
fore, these measures are not consistent with the presence
of phytoplankton blooms resulting from anthropogenic
nutrient enrichment. They do, however, suggest that even
small increases in nutrients or chlorophyll a concentrations
in this region could result in a shift from coral-dominated
to algal-dominated benthic communities.

It seems unlikely to us that there has been a drop in the
load of anthropogenic suspended solids and/or nutrients

during the past eight years, despite the probable decrease
in the volume of runoff. In fact, it appears that, if any-
thing, these inputs have increased. So, what is the cause
of the long-term trend in Secchi depth? The correlation
analysis and the monthly trends (see Figure 5) both sug-
gest that rainfall and solar radiation are the most closely
associated with Secchi depth. However, rainfall is the only
variable showing a strong annual trend consistent with the
increased Secchi depths, and rainfall over the three days
before the measurements was the variable most highly cor-
related with Secchi depths of any hydrological or climate
variable examined. Solar radiation, although it is positively
correlated with Secchi depths on the reef, does not show
the pronounced long-term trend that rainfall does. The ef-
fect of wind on Secchi depth similarly does not explain the
long-term trend. It could, however, explain the fact that
Colon, the most exposed site, with the highest winds had
consistently lower Secchi depths than the other sites. Wind-
induced turbidity, which resuspends bottom sediment, is
the likely cause of the limited water clarity at this site.

CONCLUSION

The baseline data reported here will be useful for fu-
ture studies of anthropogenic effects in the unique Bocas
del Toro archipelago. Rapid local development is pro-
gressing in the face of little information on the impact of
such development and the factors affecting such impacts
(e.g., water residence time or currents in the Bahia Almi-
rante). It is likely that anthropogenic inputs of nutrients
and suspended particulate matter will contribute to eutro-
phication of some areas. This study suggests that despite
the impact of development, patterns of water clarity and
chlorophyll a concentrations in the region are currently
driven mainly by large-scale climate patterns. There is lit-
tle evidence of a tight relationship between these measures
and features of the local water mass, nor is there evidence
of eutrophication at the sites we sampled. Future sampling
closer to highly developed areas is necessary to document
and monitor the impact of development on water quality.

ACKNOWLEDGMENTS

We thank Arcadio Castillo for help in the field, James
Roper for comments on the manuscript, the Hunterdon
and Johnson Oceanographic Research Endowments for
financial support, and Empresa de Transmisi6n Eléctrica
Panama for supplying the rainfall data from the Bocas del
Toro airport.

NUMBER 38 ¢ 333

LITERATURE CITED

Bell, P. 1992. Eutrophication and Coral Reefs: Some Examples in the
Great Barrier Reef Lagoon. Water Research, 26:553-568.

Beman, J. M., K. R. Arrigo, and P. A. Matson. 2005. Agricultural Run-
off Fuels Large Phytoplankton Blooms in Vulnerable Areas of the
Ocean. Nature (London), 434:211-214.

Borkman D. G., and T. J. Smayda. 1998. Long-Term Trends in Water
Clarity Revealed by Secchi-Disk Measurements in Lower Narra-
gansett Bay. ICES Journal of Marine Science, 55:668-679.

Carruthers, T. J. B., P. A. G. Barnes, G. E. Jacome, and J. W. Fourqurean.
2005. Lagoon Scale Processes in a Coastally Influenced Caribbean
System: Implications for the Seagrass Thalassia testudinum. Carib-
bean Journal of Science, 41:441-455.

Convers, A., and J. A. McGowan. 1994. Natural Versus Human-Caused
Variability of Water Clarity in the Southern California Bight. Lim-
nology and Oceanography, 39:632-648.

D’Croz, L., J. B. Del Rosario, and P. Gondola. 2005. The Effect of Fresh
Water Runoff on the Distribution of Dissolved Inorganic Nutrients
and Plankton in the Bocas del Toro Archipelago, Caribbean Pan-
ama. Caribbean Journal of Science, 41:414-429.

Falkowski, P. G., R. T. Barber, and V. Smetacek. 1998. Biogeochemi-
cal Controls and Feedbacks on Ocean Primary Production. Science,
281:200-206.

Falkowski, P. G., D. Ziemann, Z. Kolber, and P. K. Bienfang. 1991. Eddy
Pumping in Enhancing Primary Production in the Ocean. Nature
(London), 352:55-58.

Franco-Herrera, A., A. Castro, and P. Tigreros. 2006. Plankton Dy-
namics in the South-Central Caribbean Sea: Strong Seasonal
Changes in a Coastal Tropical System. Caribbean Journal of Sci-
ence, 42:24-38.

Gilbes, E, J. M. Lopez, and P. M. Yoshioka. 1996. Spatial and Tempo-
ral Variations of Phytoplankton Chlorophyll a and Suspended Par-
ticulate Matter in Mayagiiez Bay, Puerto Rico. Journal of Plankton
Research, 18:29-43.

Guzman, H. M., P. A. G. Barnes, C. E. Lovelock, and I. C. Feller. 2005. A
Site Description of the CARICOMP Mangrove, Seagrass and Coral
Reef Sites in Bocas del Toro, Panama. Caribbean Journal of Science,
41:430-440.

Herrera-Silveira, J. A., I. Medina-Gomez, and R. Collin. 2002. Tro-
phic Status Based on Nutrient Concentration Scales and Primary
Producers Community of Tropical Coastal Lagoons Influenced by
Groundwater Discharges. Hydrobiologia, 475/476:91-98.

Instituto Geografico Nacional Tomy Guardia (IGNTG). 1988. Atlas na-
cional de la Republica de Panama. Balboa, Panama: Repulica de
Panama: Ministerio Obras Publicas.

Kaufmann, K. W., and R. C. Thompson. 2005. Water Temperature Varia-
tion and the Meteorological and Hydrographic Environment of Bo-
cas del Toro, Panama. Caribbean Journal of Science, 41:392-413.

Kratzer, S., S. Buchan, and D. G. Bowers. 2003. Testing Long-Term
Trends in Turbidity in the Menai Strait, North Wales Estuarine.
Coastal and Shelf Science, 56:221-226.

Lapointe, B. 1992. “Eutrophication Thresholds for Macroalgal Over-
growth of Coral Reefs.” In Protecting Jamaica’s Coral Reefs: Water
Quality Issues, ed. K. Thacker, pp. 105-112. Negril, Jamaica: Ne-
gril Coral Reef Preservation Society.

Megard, R. O., and T. Berman. 1989. Effects of Algae on the Secchi
Transparency of the Southeastern Mediterranean Sea. Limnology
and Oceanography, 34:1640-1655.

Otero, E., and K. K. Carbery. 2005. Chlorophyll a and Turbidity Patterns
over Coral Reefs Systems of La Parguera Natural Reserve, Puerto
Rico. Revista Biologia Tropical (Suppl. 1), 53:25-32.

Sandén, P., and B. Hakansson. 1996. Long-Term Trends in Secchi Depth
in the Baltic Sea. Limnology and Oceanography, 41:346-351.

334 e

Szmant, A. M. 2002. Nutrient Enrichment on Coral Reefs: Is It a Major
Cause of Coral Reef Decline? Estuaries, 25:743-766.

van Duyl, F. C., G. J. Gast, W. Steinhoff, S. Kloff, M. J. W. Veldhuis, and
R. P. M. Bak. 2002. Factors Influencing the Short-Term Variation
in Phytoplankton Composition and Biomass in Coral Reef Waters.
Coral Reefs, 21:293-306.

Walsh, J. J., T. E. Whitledge, EF W. Barvenik, C. D. Wirick, $. O. Howe,
W. E. Esaias, and J. T. Scott. 1978. Wind Events and Food Chain
Dynamics within the New York Bight. Limnology and Oceanogra-
phy, 23:659-683.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Webber, M. K., D. F Webber, and E. R. Ranston. 2003. Changes in Wa-
ter Quality and Plankton of Kingston Harbour, Jamaica, After 200
Years of Continued Eutrophication. Bulletin of Marine Science,
73:361-378.

Welschmeyer, N. A. 1994. Fluorometric Analysis of Chlorophyll a in
the Presence of Chlorophyll b and Pheopigments. Limnology and
Oceanography, 39:1985-1992.

Yoder, J. A., C. R. McClain, G. C. Feldman, and W. E. Esaias. 1993. Annual
Cycles of Phytoplankton Chlorophyll Concentrations in the Global
Ocean: A Satellite View. Global Biogeochemical Cycles, 7:181-193.

Nutrient and Chlorophyll Dynamics
in Pacific Central America (Panama)

Luis D’Croz and Aaron O’Dea

Luis D’Croz and Aaron O’Dea, Smithsonian
Tropical Research Institute, Box 0843-03092,
Panama, Republic of Panama. Corresponding
author: L. D’Croz (dcrozl@si.edu). Manuscript
received 13 May 2008; accepted 20 April 2009.

ABSTRACT. Strong wind jets from the Caribbean and the Gulf of Mexico cross Cen-
tral America through topographic depressions in the cordillera during the boreal winter,
pushing Pacific coastal waters offshore, lowering sea levels at the coast, and causing
coastal upwelling. Where high mountains impede the winds, this phenomenon does not
occur. The Panamanian Pacific shelf is an excellent example of this variability. The coast
is divided into two large areas, the Gulf of Panama and the Gulf of Chiriqui. To investi-
gate hydrological conditions between the two gulfs, we sampled the water column during
upwelling and non-upwelling seasons in each region. In both gulfs during non-upwelling
conditions, surface-level nutrients are poor, and the chlorophyll maximum occurs around
30 m where the thermocline intersects the euphotic zone. Oxygen-poor waters (<2 ppm)
commonly occurred below the thermocline. During the dry season, wind strength in-
creased and strong upwelling was observed in the Gulf of Panama. The thermocline rose
and surface waters became nutrient enriched and chlorophyll a levels increased. Well-
oxygenated waters were compressed to shallow depths. In the Gulf of Chiriqui, wind
strength was weaker, surface waters did not become enriched with nutrients, and surface
chlorophyll a remained low. We did observe a shallowing of the thermocline in the Gulf
of Chiriqui, but in contrast to the Gulf of Panama, wind mixing was not strong enough
to result in sea-surface cooling and nutrient enrichment. We postulate that the conver-
gence of a shallow thermocline and internal waves in the Gulf of Chiriqui is the likely
mechanism that causes pockets of deep water to occasionally migrate into surface waters,
leading to restricted and ephemeral upwelling-like conditions. Although its effects upon
shallow-water communities remain to be studied, we propose that the process may be
more likely to occur during the boreal winter when the thermocline is shallower.

INTRODUCTION

One of the most pervasive hydrological events to influence the shelf wa-
ters of Pacific Central America is upwelling. Intermittent or seasonal upwell-
ing develops in the gulfs of Tehuantepec (Mexico), Papagayo (Costa Rica), and
Panama (Legeckis, 1988; McCreary et al., 1989; Xie et al., 2005), driving exten-
sive planktonic productivity and shaping the secondary production of biological
communities (Jackson and D’Croz, 1997; O’Dea and Jackson, 2002).

The shelf waters along the Pacific coast of Panama are among the most dy-
namic in the region. Here, the coastal shelf is naturally divided into two large gulfs
by the Azuero Peninsula: the Gulf of Panama (shelf area, 27,175 km”) and the Gulf

336 °

of Chiriqui (shelf area, 13,119 km?) (Figure 1). The Gulf
of Panama experiences strong seasonal upwelling while the
Gulf of Chiriqui exemplifies a non-upwelling environment
(Dana, 1975; Kwiecinski and Chial, 1983). This distinc-
tion is customarily explained using geographic differences
between the two gulfs. Seasonal upwelling in the Gulf of
Panama develops during Panama’s dry season, correspond-
ing to the boreal winter, when northeast trade winds cross
to the Pacific over low areas in the isthmian mountain range,
pushing warm and nutrient-poor coastal surface water off-
shore, lowering the nearshore sea level, and causing the up-
ward movement of colder and nutrient-richer deep water
(Smayda, 1966; Forsbergh, 1969; Kwiecinski et al., 1975;
D’Croz et al., 1991; D’Croz and Robertson, 1997). The es-
tablished model proposes that because western Panama has
higher mountain ranges that block the winds, surface waters
in the Gulf of Chiriqui are not displaced out to the Pacific,
and no upwelling as such occurs there.

The structure of shallow biological communities be-
tween the two regions supports this inference. Coral reefs,
which respond poorly to upwelling conditions, are more
extensive in size in the Gulf of Chiriqui than in the Gulf of
Panama (Glynn, 1977; Glynn and Maté, 1997), whereas
small pelagic fish species from the Gulf of Panama repre-
sent a large proportion of the total estimated fishery re-
source in the country (NORAD, 1988). Satellite imagery

PP
—

CA
oS

St fo

———

Azuero “SN

azure

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

shows both wind speeds and chlorophyll content of surface
waters to be lower in the Gulf of Chiriqui than the Gulf of
Panama during the dry seasons (Pennington et al., 2006).

However, the statement that upwelling does not occur
in the Gulf of Chiriqui is supported by sea-surface data
derived from satellite imagery analysis or from the mea-
surement of properties in the shallow section of the water
column. Hydrological profiles of the water column have
documented the shoaling of the thermocline in the Gulf
of Chiriqui, yet there appears to be no clear association
between the physical forcing of this event with the wind-
induced upwelling in the Gulf of Panama. Nevertheless,
the movement of pockets of cool water that bring nutri-
ents into the upper layer may be a more common occur-
rence in the Gulf of Chiriqui than previously suspected
(D’Croz and O’Dea, 2007).

It is therefore essential that we obtain detailed and
comparable hydrological data from both gulfs if we wish
to explain variability in biological communities along the
Pacific coast of the Isthmus of Panama today and through
geologic time (O’Dea et al., 2007). In this paper we ex-
pand the information presented in our previous study
(D’Croz and O’Dea, 2007), adding new hydrological and
biological data from the Gulf of Chiriqui and the Gulf of
Panama, and we further discuss the issue of whether up-
welling takes place in the Gulf of Chiriqui.

FIGURE 1. Map of the Republic of Panama showing sampling sites. Red dots represent the location of the

rainy season samplings in the Gulf of Panama (a = 18 December 2004) and in the Gulf of Chiriqui (a = 13
July 2003; b = 17 December 2004). Blue dots represent the location of the dry season samplings in the Gulf
of Panama (a = 29 February 2000) and in the Gulf of Chiriqui (a = 1 March 2000; b = 13 April 2007).
Yellow squares indicate the location of the meteorological stations.

MATERIALS AND METHODS
STUDY AREA

Panama’s Pacific shelf is located from 07°30’ to
09°01’N and 78°10’ to 82°52'W. The shelf is predomi-
nantly occupied by low-salinity surface water, similar to
the water mass found over the center of the tropical Pacific
Ocean at about 10°N (Wyrtki, 1967; Fiedler and Talley,
2006). The climatology is governed by the Inter-Tropical
Convergence Zone (ITCZ), the position of which defines
the seasonal pattern of rainfall and winds. The rainy season
develops between May and December when the ITCZ is
located over or slightly to the north of Panama and winds
are light and variable in direction. The dry season devel-
ops between January and March when the ITCZ moves
south of Panama, a time period characterized by pre-
dominating intense northeast trade winds. The mean an-
nual rainfall recorded at meteorological stations near the
coast (1999-2004) was 2,760 mm in the Gulf of Chiriqui
(David) and 1,880 mm in the Gulf of Panama (Tocumen).
Approximately 94% of the annual rainfall in both areas
corresponded to the rainy season, the months of Septem-
ber and October being the rainiest in both regions. The
estimated sizes of the drainage basins are 11,846 km? in
the Gulf of Chiriqui and 33,828 km? in the Gulf of Pan-
ama. River discharges into both gulfs typically follow the
seasonal trend described for rainfall. Detailed discussions
on wind-stress, rainfall, and river discharge patterns are
presented in D’Croz and O’Dea (2007). The tidal regime
is semidiurnal, and the sea-level difference during spring
tides is 6 m (Glynn, 1972).

SAMPLING PROCEDURES

Sampling research cruises were conducted in the gulfs
of Panama and Chiriqui using the Smithsonian Tropical
Research Institute’s R/V Urraca (see Figure 1). Samplings
were scheduled to correspond with different times of the
year, representing contrasting hydrological conditions (up-
welling and non-upwelling). Surface-to-bottom profiles for
salinity, temperature, dissolved oxygen, and chlorophyll a
were recorded with a CTD (conductivity, temperature,
depth) multiparameter profiler (Ocean Seven 316, Idro-
naut Srl, Milano, Italy). Hydrological casts with the CTD
corresponding to the dry season were carried out in both
gulfs on 29 February 2000 and 1 March 2000 and in the
Gulf of Chiriqui on 13 April 2007. Rainy season CTD
casts were carried out in the Gulf of Chiriqui on 13 July
2003 and in both gulfs during 17 and 18 December 2004.
The water column was sampled at discrete levels to study

NUMBER 38 °¢ 337

nutrient and chlorophyll a concentrations. Water samples
were collected using Niskin bottles during the dry season
of the year 2000 (29 February to 1 March) and during
the rainy season of the year 2004 (17 and 18 December).
Three replicate water samples per selected depth were col-
lected at each site. Two liters of each individual replicate
water sample were immediately sieved through Nitex
(350 jm) to exclude zooplankton and vacuum filtered on
Whatman GF/F filter (0.7 ym pore size) for chlorophyll a
analysis. An aliquot from each filtrate was set apart for the
determination of dissolved inorganic nutrients. Filters and
water samples were stored frozen (—20°C) until analysis.
Salinity is expressed using the Practical Salinity Scale (pss)
indicated by UNESCO (1981). Results from the chloro-
phyll a analyses were used to check the calibration of the
CTD’s fluorometer. The depth of the euphotic zone (1%
incident radiation) was estimated from Secchi disk read-
ings (Parsons et al., 1984). The light attenuation coeffi-
cient was calculated as Ky = f/zs where zs is the Secchi
depth and f = 1.4.

ANALYSIS OF SAMPLES

Not later than two weeks after sampling, filters hold-
ing the phytoplankton were analyzed for chlorophyll a
using the non-acidification fluorometric method (Welsch-
meyer, 1994). Water samples were analyzed for NO3;— +
NO, (nitrate + nitrite), Si(OH), (silicate), and PO,3-
(phosphate) by colorimetric methods using an Alpkem
Flow Solution IV automated analyzer. Minimum detection
limits were 0.02 wM for nitrate, 0.01 M for nitrite, 0.12
uM for silicate, and 0.02 1M for phosphate.

ANALYSIS AND PRESENTATION OF DATA

Water quality variables, namely temperature, salin-
ity, dissolved oxygen, dissolved inorganic nutrients, and
chlorophyll a, are presented graphically as profiles of the
samplings. Overall differences in between the two gulfs
were assessed with the Mann-Whitney test (U) by taking
the median of each variable from samples collected in the
top 30 m of the ocean where the highest hydrological vari-
ability occurred (Table 1). Water transparency data were
compared using the paired ¢ test. We followed the practice
of taking the position of the 20°C isotherm to represents
the depth of the center of the permanent thermocline in
the eastern Pacific Ocean (Wyrtki, 1964; Fiedler et al.,
1991; Xie et al., 2005). Pearson correlations with Bonfer-
roni adjustment were used to test statistical relationships
among variables.

338 °

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

TABLE 1. Average value of hydrological variables in the top water column (30 m) in the gulfs of Panama (GP) and Chiriqui (GC); SE =
standard error of the mean. Statistical tests were either Mann-Whitney U test or paired t test (*P < 0.05, **P < 0.01, ***P < 0.001,

ns = nonsignificant).

Dry season values

Rainy season values

Hydrological GP GC Statistical GP GC Statistical
variables (Mean = SE) (Mean = SE) value 2» (Mean = SE) (Mean + SE) value >
Temperature (°C) W/O == 02 BIN = 02 16.0% 4 26.75 + 0.54 28.61 + 0.05 18.0 ns 4
Salinity (pss) © 34.18 + 0.29 32.98 + 0.29 IDO? & 31.67 + 0.64 30.48 + 0.38 3.0 ns
Chlorophyll a (ug L~!) 1.82 + 0.65 0.83 + 0.65 4.0% a 0.23 = 0.13 0.18 + 0.06 8.5 ns 2
Dissolved oxygen (ppm) 3.45 + 0.27 4.78 + 0.27 IO? 2 3.98 + 0.16 4.38 = 0.01 4.0 ns #
NO3~ (wM) 14.37 + 2.48 3.72 + 2.48 = 0.99 + 0.34 0.36 + 0.02 2.5 ns 4
PO,3- (wM) 1.08 + 0.21 0.39 + 0.21 0.43 + 0.07 0.24 + 0.03 4.0 ns 4
N:P ratio 12.82 + 1.10 7.77 = 1.10 2.11 + 0.36 1.49 + 0.10 3.0 ns 4
Si(OH),4 (uM) 8.99 + 1.03 4.40 + 1.03 5.40 + 0.71 4.87 + 0.47 13.0 ns 4
Secchi depth (m) 4.20 + 0.00 14.80 + 0.00 20.00 + 0.00 19.00 + 0.00 2.0 ns >
Euphotic zone (m) 13.8 + 0.00 48.63 + 0.00 65.71 + 0.00 62.43 + 0.00 188.4 ns >
4 Mann-Whitney U test.
Paired ¢ test.

© pss = practical salinity scale.

RESULTS season, both regions experienced high freshwater dilu-

THERMOHALINE STRUCTURE

Both the Gulf of Panama and the Gulf of Chiriqui
exhibit the typical tropical coastal ocean water struc-
ture of cool deep waters leading upward to a shallow
thermocline topped by warm surface waters. However,
significant differences occur between the two gulfs with
respect to climatic variability. During the rainy season,
the thermal structure in both gulfs is remarkably similar
(see Table 1). Sea-surface temperatures (SSTs) are invari-
ably warm (27°-28°C), and the thermocline sits at ap-
proximately 60 m (Figure 2).

During the dry season, thermal conditions become dis-
similar between the two regions (Table 1). In our observa-
tions, the thermocline in the Gulf of Panama rose sharply
and nearly broke at the surface, resulting in a significant
cooling of surface waters to 22°C (Figure 3a). Simultane-
ously, the thermocline in Gulf of Chiriqui rose to around
30 m, compressing warm SSTs into shallow waters (Figure
3b). However, the shoaling of the thermocline in Chiriqui
was not as intense as that seen in the Gulf of Panama and
did not result in SST cooling.

In general, salinity profiles in both regions revealed
a sharp gradient from high-salinity deep water to fresher
surface waters. Seasonal variability in surface salinities
in both gulfs was very similar (Table 1). During the rainy

_ tion in the upper-layer waters, with surface salinities

below 30 on the pss (see Figure 2). The halocline was
located at 60 m depth, coinciding with the thermocline.
During the dry season, lower rainfall led to increased
salinities in the surface waters of both gulfs (Figure 3).
However, the effect was more striking in the Gulf of
Panama as the halocline shoaled and salinity in surface
waters reached 34.

In April 2007, the thermohaline structure in the
Gulf of Chiriqui departed drastically from the typical
condition as the thermocline/halocline shoaled to 20 m.
Despite this condition, however, SSTs remained warm
(Figure 3c).

CHLOROPHYLL

Concentrations of surface chlorophyll were always be-
low 0.30 g/L in both gulfs during the rainy season (Table
1), but a deep chlorophyll maximum developed from 30 m
to 50 m, lying above the thermocline (Figure 2). The deep
chlorophyll maximum contained most of the chlorophyll a
in the water column in both gulfs, concentrations reaching
1 we/L during the rainy season. The dry season upwelling
changed this pattern in the Gulf of Panama, as the chloro-
phyll maximum moved into shallower waters, where con-
centrations surpassed 4 wg/L (Figure 3a). Surface chloro-

Dissolved oxygen (ppm)
FY Ao s 4b. SG G

Chlorophyll a (ug L*)
3

0 1 2 4 5 0 1

Dissolved oxygen (ppm)
0 2 7

SS SS ee ee Ree ey eee eee See es Se Os fe Ce OQ |
Chlorophyll a (ug L")

NUMBER 38 ¢ 339

Dissolved oxygen (ppm)
eh &) Wo aya GG
Chlorophyll a (ug L")

3 4 5 0 1 2 3 4 5

EL te ee a | Re ve a ey AE SE |) al (ae UR (ane en

Salinity (pss)

29 30 31 32 33 34 35 3629 30 31
a ee es ee ee ee ee

Temperature (°C)
215) 18) 21) 24 278 30812) 415

Depth (m)
iS
oa

225 Gulf of Panama
18/December/2004
250 (Rainy season)

Salinity (pss)
32 33 34 #35 #36291 30 31 32 33 34 #35 36
ee ES EE

Temperature (°C)
21 247727, 30012). 15; 18) 24 2AM 2 oO,

Gulf of Chiriqui
13/July/2003
(Rainy season)

Salinity (pss)

Temperature (°C)

Gulf of Chiriqui
17/December/2004
(Rainy season)

FIGURE 2. Profiles of dissolved oxygen, chlorophyll a, salinity, and temperature in the Gulf of Panama and
the Gulf of Chiriqui during the rainy season. a = Gulf of Panama, 18 December 2004; b = Gulf of Chiriqui,

13 July 2003; c = Gulf of Chiriqui, 17 December 2004.

phyll a remained at very low values in the Gulf of Chiriqui
during the dry season, but the deep chlorophyll maximum
became remarkably intense at 30 m where concentration
reached 3 pg/L (Figure 3b).

DISSOLVED OXYGEN

Dissolved oxygen profiles followed the typical pat-
tern of well-oxygenated surface waters lying on top of
deeper oxygen-poor waters. During the rainy season, se-
vere hypoxic conditions (<2 ppm) were recorded below
the strong oxycline, at 50 m and nearly coincident with
the thermocline (see Figure 2). Oxygen concentrations in
waters above the thermocline were strongly correlated
with temperature in both the Gulf of Panama (r = 0.91;
P < 0.001) and the Gulf of Chiriqui (r = 0.89; P <
0.001) during the rainy season. This arrangement, how-
ever, had strong seasonal variation in the Gulf of Panama
during the dry season, as the oxycline rose to 25 m and

compressed the oxygenated waters into shallow depths
(Figure 3). Dissolved oxygen below this depth rapidly de-
clined to less than 1 ppm (Figure 3a), whereas waters in
the Gulf of Chiriqui only became hypoxic below the 50 m
oxycline (Figure 3b). No correlations were confirmed be-
tween dissolved oxygen and temperature in any of these
regions during the dry season.

DISSOLVED NUTRIENTS

Both gulfs exhibit a strong vertical gradient of upwardly
decreasing nutrient concentrations. Nitrate in surface waters
was depleted in both gulfs during the rainy season, with val-
ues below 0.5 M (Figure 4). During the dry season, nitrate
concentrations at the surface were observed to increase 10
fold in the Gulf of Panama when the nutricline shoaled to
around 10 m (Figure 5a). No similar surface enrichment
was detected in the Gulf of Chiriqui, where a strong nutri-
cline was developed at 60 m (Figure 5b).

340 e

Dissolved oxygen (ppm)
OM Say 2 Sie 45 eG

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Dissolved oxygen (ppm)
ie 10 eel 2

Ch Sy 7

es ee ee ee ee ee ee eee ee ee ee eee

Chlorophyll a (ug L”)
0 1 2 3 4 5 0 1

Chlorophyll a (ug L”)

Chlorophyll a (ug L")
3) AW 5ia0) Bt | WE Sees

bet nt tt nn nt

Salinity (pss)

29 30 31 32 33 34 35 3629 30 31
es ee ee ee ee ee

Temperature (°C) Te’
12 #15 18 #219 24 27 3042 15 4

Depth (m)

225 =| Gulf of Panama
29/February/2000
250 4 (Dry season)

Salinity (pss)
32 33 34 35 36 29 30 31 32 33 34 35 36
——— a on onl

mperature (°C)
21 24 27 3012 15

Gulf of Chiriqui
01/March/2000
(Dry season)

Salinity (pss)

Temperature (°C)
18 21 24 27 30

Gulf of Chiriqui
13/April/2007
(Dry season)

FIGURE 3. Profiles of dissolved oxygen, chlorophyll a, salinity, and temperature in the Gulf of Panama and
the Gulf of Chiriqui during the dry season. a = Gulf of Panama, 29 February 2000; b = Gulf of Chiriqui, 1

March 2000; c = Gulf of Chiriqui, 13 April 2007.

Overall, the patterns of phosphate resembled those of
nitrate, but concentrations were lower by an order of mag-
nitude. Concentrations of phosphate in excess of 1 ~M were
usually found below 30 m depth. Phosphate concentrations
in surface waters remained relatively low (<0.3 wM) in the
Gulf of Chiriqui during both climatic seasons (Figures 4b,
5b). However, phosphate enrichment of surface waters
clearly occurred in the Gulf of Panama during the dry sea-
son when the nutricline shoaled and phosphate concentra-
tions in the top of the water column reached about 1.0 ~.M
(Figure 5a).

Silicate profiles followed similar trends to that of the
nitrate and phosphate (Figures 4, 5). Although silicate
concentrations were similar in surface waters in both gulfs
during the rainy season, they doubled in the Gulf of Pan-
ama during the dry season (Table 1).

Dissolved nutrients in the upper 50 m had a high de-
gree of relationship with temperature and salinity. During
the rainy season, nitrate concentrations were inversely cor-

related to temperature in both the Gulf of Chiriqui (r =
—0.78; P < 0.001) and the Gulf of Panama (r = -0.97; P <
0.002). In the dry season, nitrate in the Gulf of Panama was
negatively correlated to temperature (r = —0.98; P < 0.044)
and directly related to salinity (r = 0.98; P < 0.049). Nir-
trate was negatively correlated to temperature in the Gulf of
Chiriqui during the dry season (r = —0.89; P < 0.016), but
not to salinity (r = 0.67; P > 0.159). Phosphate was nega-
tively correlated to temperature during the dry season in the
Gulf of Panama (r = —0.98; P < 0.038) and in the Gulf of
Chiriqui (r = -0.97; P < 0.036). Dry season phosphate was
also correlated to salinity in the Gulf of Chiriqui (r = 0.98;
IP & 05).

The extremely low nitrate to phosphate ratios (N:P)
suggest that phytoplankton growth in both regions was
under severe nitrogen limitation during the rainy season
(Figure 6). The N:P ratio was below 2:1 in surface water
and increased with depth, surpassing the value of 10:1
below the depth of 50 m. During the dry season, N:P ra-

Si-Silicate (uM)
0 8 16 24 32

P-Phosphate (uM)
(OG) (0s) 4-0)

N-Nitrate (uM)

0} ey 0)

100

Depth (m)

120
140
160

180 “| Gulf of Panama
18/December/200:
(Rainy season)

200

as) PL) 72s) KOO) (0) 4/0)

(Sie Z0ee 25306 350 Ol 5) 10

NUMBER 38 ¢ 341

b
Si-Silicate (uM)
40 0 8 16 24 32 40
We

P-Phosphate (uM)
ds) ZA) Ze) S10

N-Nitrate (uM)
15eeeZ20e 253005

Gulf of Chiriqui
17/December/2004
(Rainy season)

FIGURE 4. Mean profiles of silicate (Si), phosphate (P), and nitrate (N) in the Gulf of Panama and the Gulf
of Chiriqui during the rainy season. a = Gulf of Panama, 18 December 2004; b = Gulf of Chiriqui, 17 De-

cember 2004.

tios within the euphotic zone largely increased in both re-
gions, becoming closer to the N:P ratio of 16:1 suggested
as favorable for phytoplankton growth (Redfield, 1958).

WATER TRANSPARENCY

Water transparency was seasonably stable in the Gulf
of Chiriqui but varied considerably in the Gulf of Panama
(see Table 1). Water transparency in both gulfs was higher
during the rainy season when the euphotic zone was ap-
proximately 60 m deep, in contrast to the limited trans-

parency and shallow euphotic zone (14 m) observed in the
Gulf of Panama during the dry season upwelling.

DISCUSSION

Our data on bottom-to-surface profiles reveal the dy-
namics of hydrological conditions along the Pacific coast of
Panama during times of both upwelling and non-upwelling.
During the non-upwelling rainy season, both gulfs exhibit
extremely similar hydrological structures dominated by the

342 e

Si-Silicate (uM)
0 Gi-- 4G 24 2

P-Phosphate (uM)
O10) O55 120

N-Nitrate (uM)
Oe Damn 0

100

Depth (m)

120
140

160

180— Gulf of Panama

200 (Dry season)

1:5) 62:0) 922558 3:0010) G'S) a0

159620772557 308 350m Om Om 0

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

b
Si-Silicate (uM)

40 0 8 16 24 32 40
en

P-Phosphate (uM)
1:5) 2107 O25 0

N-Nitrate (uM)
15020) 255 S0mesS

Gulf of Chiriqui
01/March/2000
(Dry season)

FIGURE 5. Mean profiles of silicate (Si), phosphate (P), and nitrate (N) in the Gulf of Panama and the Gulf
of Chiriqui during the dry season. a = Gulf of Panama, 29 February 2000; b = Gulf of Chiriqui, 1 March

2000.

development of an intense thermocline at approximately
60 m. Surface waters tend to have low salinities and are
warm and nutrient depleted. Low N:P ratios in surface
waters during the rainy season suggest that phytoplank-
ton growth is strongly nitrogen limited. Consequently, the
standing stock of chlorophyll a is maintained at relatively
low levels in surface waters. Phytoplankton does however
peak at subsurface levels as the nutrient-rich thermocline
waters intersect the euphotic zone, increasing N:P ratios
and favoring algal growth. The strong inverse correlation
between nutrients and sea temperature is consistent with
the coincidence of a shallow thermocline and strong nutri-

cline typical of the eastern tropical Pacific Ocean (Enfield,
2001). As such, the seasonal movement of the thermocline
represents a key source of nutrients for phytoplankton.
Our sampling sites were far offshore and therefore silicate
concentrations were not as high as previously reported for
the inner shelf (D’Croz and O’Dea, 2007) even though the
concentration of silicate in the Gulf of Panama is reported
to be the highest in the eastern Pacific as a consequence of
the intense runoff in the area (Pennington et al., 2006).
During the dry season, the hydrological patterns of the
two gulfs become dissimilar. In the Gulf of Panama strong
upwelling of cold deep waters into coastal and surface wa-

Gulf of Panama Gulf of Chiriqui
N:P ratio N:P ratio
0 3 6 S12 15310 3 6 9 12 15

Depth (m)
°
ro)

4 Rainy season
@ Dry season

FIGURE 6. Profiles of average nitrate to phosphate ratios (N:P) in
rainy (triangles) and dry (circles) seasons: a = Gulf of Panama; b =
Gulf of Chiriqui.

ters drives significant changes in the hydrological properties
of the water column. The thermocline migrates vertically
upward, leading to cooling, increased salinity, and nutrient
enrichment on surface waters. Surface N:P ratios become
closer to the Redfield value and, as a result, phytoplankton
growth intensifies, leading to a reduction in water clarity.
A shallow oxycline also develops and oxygen concentra-
tion below the oxycline is reduced, often leading to severe
hypoxic conditions. In contrast, the oxycline in the Gulf of
Chiriqui is deeper and deep water remains hypoxic. Low
oxygen minima are nonetheless typical in the eastern tropi-
cal Pacific as a combination of high algal growth at the
surface, a strong pycnocline that impedes the ventilation of
waters below, and the sluggish circulation of deep waters
(Fiedler and Talley, 2006). The report of large filamentous
Thioploca-like sulfur bacteria on shallow sediments in both
regions strongly suggests that the inner shelf is exposed to
episodes of reduced oxygen (Gallardo and Espinoza, 2007).

A significant relationship between wind-stress index
(calculated from the sum of northerly winds) and sea level
provides an explanatory mechanism for upwelling in the
Gulf of Panama (Schaefer et al., 1958; Legeckis, 1988;
Xie et al., 2005). Surface waters are displaced into open
ocean by strong northerly winds during the dry season,
and deep waters move vertically upward to replace them
(Fleming, 1940; Smayda, 1966; Forsbergh, 1969). Conse-
quently, wind stress is inversely related to SST in the Gulf

NUMBER 38 ¢ 343

of Panama during the dry season but not during the rainy
season (D’Croz and O’Dea, 2007).

Data from the Gulf of Chiriqui are scant but did sug-
gest that upwelling does not occur, because wind stress
during the dry season is normally one-third of that of
the Gulf of Panama (Kwiecinski and Chial, 1983) and it
does not displace surface waters offshore. High mountain
ranges running along western Panama impede the flow of
northerly winds across to the Gulf of Chiriqui (see Fig-
ure 1), whereas mountain ranges in central Panama are
low, allowing strong wind jets to form toward the Gulf
of Panama. Despite this clear distinction, our data show
that similar hydrological changes to those that occur in the
Gulf of Panama do take place in the Gulf of Chiriqui. Dur-
ing the dry season, and concurrent with strong upwelling
in the Gulf of Panama, we observed deeper waters rise to-
ward shallower depths in the Gulf of Chiriqui. This move-
ment led to a substantial compression of the mixed layer
and the corresponding rise of available nutrients within
the euphotic zone, shifting the chlorophyll maximum
above the shallow thermocline. Although direct evidence
of prolonged surface water cooling was not observed, we
postulate that cooling and nutrient-enrichment episodes
in the Gulf of Chiriqui may occur and that their inten-
sity is dependent upon the depth to which the thermocline
reaches in the eastern Pacific during the boreal winter.
Nonetheless, the process is clearly much less intense than
that in the Gulf of Panama. Despite substantial shifts in
deeper water conditions in the Gulf of Chiriqui, surface
waters remain warm and nutrient poor, presumably be-
cause wind stress is not strong enough to cause the advec-
tion of deep, cool, and nutrient-rich waters to the surface
(D’Croz and O’Dea, 2007). However, ocean forces such as
internal waves might change the oceanographic structure
in the Gulf of Chiriqui, causing brief periods of advection
of deep cold water to the surface layer (Dana, 1975). Long-
term records from data loggers deployed in coral reefs give
evidence of such brief SST drops in the Gulf of Chiriqui
that are possibly related to internal waves (D’Croz and
O’Dea, 2007). This effect might be more evident as the
internal waves approach the shallow coasts around the is-
lands in the Gulf of Chiriqui and may be more likely to
occur during times of thermocline shallowing.

In conclusion, although the Gulf of Chiriqui does not
experience the intense seasonal upwelling characteristic
of the Gulf of Panama, deeper waters do migrate upward
in synchrony with Gulf of Panama upwelling. This move-
ment is probably caused by an overall shallowing of the
thermocline across Central America. The difference in in-
tensity of upward movement of the thermocline between

344 e

the two gulfs strongly influences the phytoplankton com-
munity, with seasonal blooms occurring in the Gulf of
Panama but not in the Gulf of Chiriqui. Deeper waters do
nonetheless experience similar patterns of seasonal hydro-
graphic change, and shallow waters of the Gulf of Chiriqui
can be exposed to brief pulses of cold and nutrient-rich
waters by advection. However, the effects of thermocline
migration and advection on the shallow-water communi-
ties of the Gulf of Chiriqui remain to be studied in detail.

ACKNOWLEDGMENTS

Juan B. Del Rosario, Plinio Gondola, and Dayanara
Macias assisted in the collection and processing of the
samples. Sebastien Tilmans and Juan L. Maté reviewed the
manuscript. Rainfall data were kindly provided by Empresa
de Transmision Eléctrica $.A., Panama. We acknowledge the
participants, skipper, and crew of the R/V Urraca for their
assistance during the cruises. The Government of the Re-
public of Panama granted the permits for the collections.

LITERATURE CITED

Dana, T. F. 1975. Development of Contemporary Eastern Pacific Coral
Reefs. Marine Biology, 33:355-374.

D’Croz, L., J. B. Del Rosario, and J. A. Gémez. 1991. Upwelling and
Phytoplankton in the Bay of Panama. Revista de Biologia Tropical,
39:233-241.

D’Croz, L., and A. O’Dea. 2007. Variability in Upwelling Along the
Pacific Shelf of Panama and Implications for the Distribution of
Nutrients and Chlorophyll. Estuarine, Coastal, and Shelf Science,
73:325-340.

D’Croz, L., and D. R. Robertson. 1997. Coastal Oceanographic Condi-
tions Affecting Coral Reefs on Both Sides of the Isthmus of Pan-
ama. Proceedings of the 8th International Coral Reef Symposium,
2:2053-2058.

Enfield, D. B. 2001. Evolution and Historical Perspective of the 1997-
1998 El Nifto-Southern Oscillation Event. Bulletin of Marine Sci-
ence, 69:7-25.

Fiedler, P. C., V. Philbrick, and F. T. Chavez. 1991. Oceanic Upwelling
and Productivity in the Eastern Tropical Pacific. Limnology and
Oceanography, 36:1834-1850.

Fiedler, P. C., and L. D. Talley. 2006. Hydrography of the Eastern Tropi-
cal Pacific: A Review. Progress in Oceanography, 69:143-180.
Fleming, R. H. 1940. A Contribution to the Oceanography of the Cen-
tral American Region. Proceedings of the 6th Pacific Science Con-

gress, 3:167-175.

Forsbergh, E. D. 1969. On the Climatology, Oceanography and Fisheries
of the Panama Bight. Inter-American Tropical Tuna Commission
Bulletin, 14:49-259.

Gallardo, V. A., and C. Espinoza. 2007. New Communities of Large Fila-
mentous Sulfur Bacteria in the Eastern South Pacific. International
Microbiology, 10:97-102.

Glynn, P. W. 1972. “Observations on the Ecology of the Caribbean and
Pacific Coast of Panama.” In The Panamic Biota: Some Observa-
tions Prior to a Sea-Level Canal, ed. M. L. Jones. Bulletin of the
Biological Society of Washington, 2:13-30.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

. 1977. Coral Growth in Upwelling and Non-upwelling Areas
off the Pacific Coast of Panama. Journal of Marine Research,
35:567-585.

Glynn, P. W., and J. L. Maté. 1997. Field Guide to the Pacific Coral
Reefs of Panama. Proceedings of the 8th International Coral Reef
Symposium, 1:145-166.

Jackson J. B. C., and L. D’Croz. 1997. “The Ocean Divided.” In Central
America: A Natural and Cultural History, ed. A. G. Coates, pp.
38-71. New Haven and London: Yale University Press.

Kwiecinski, B., and B. Chial. 1983. Algunos Aspectos de la Oceanografia
del Golfo de Chiriqui, su Comparacion con el Golfo de Panama.
Revista de Biologia Tropical, 31:323-325.

Kwiecinski, B., A. L. Jaén, and A. M. Muschett. 1975. Afloramiento en
el Golfo de Panama durante la Temporada de 1973. Universidad
Autonoma de México. Anales Centro de Ciencias del Mar y Lim-
nologia, 2:73-80.

Legeckis, R. 1988. Upwelling off the Gulfs of Panama and Papagayo in
the Tropical Pacific during March 1985. Journal of Geophysical
Research, 93:15485-15489.

McCreary, J. P., Jr, H. S. Lee, and D. B. Enfield. 1989. The Response
of the Coastal Ocean to Strong Offshore Winds: With Applications
to Circulation in the Gulf of Tehuantepec and Papagayo. Journal of
Marine Research, 47:81-109.

NORAD. 1988. Surveys off the Pacific Coast of Central America. Re-
ports on Surveys with R/V Dr. Fridtjof Nansen. Institute of Marine
Research, Bergen, Norway: NORAD/UNDP/FAO.

O’Dea, A., and J. B. C. Jackson. 2002. Bryozoan Growth Mirrors Con-
trasting Seasonal Regimes Across the Isthmus of Panama. Palaeo,
185:77-94.

O'Dea, A., J. B. C. Jackson, H. Fortunato, J. Travis-Smith, L. D’Croz,
K. G. Johnson, and J. A. Todd. 2007. Environmental Change Pre-
ceded Caribbean Extinction by 2 Million Years. Proceedings of the
National Academy of Sciences of the United States of America,
104(13):5501-5506.

Parsons, T. R., M. Takahashi, and B. Hargrave. 1984. Biological Ocean-
ographic Processes, 3rd ed. Oxford: Pergamon Press.

Pennington, J. T., K. L. Mahoney, V. S. Kuwahara, D. D. Kolber, R.
Calienes, and F. P. Chavez. 2006. Primary Production in the
Eastern Tropical Pacific: A Review. Progress in Oceanography,
69:285-317.

Redfield, A. C. 1958. The Biological Control of Chemical Factors in the
Environment. American Scientist, 46:205-221.

Schaefer, M. B., Y. M. Bishop, and G. V. Howard. 1958. Some Aspects
of Upwelling in the Gulf of Panama. Inter-American Tropical Tuna
Commission Bulletin, 3:77-132.

Smayda, T. J. 1966. A Quantitative Analysis of the Phytoplankton of the
Gulf of Panama. III. General Ecological Conditions, and the Phyto-
plankton Dynamics at 8°45'N, 79°23’W from November 1954 to
May 1957. Inter-American Tropical Tuna Commission Bulletin,
11:353-612.

UNESCO. 1981. The Practical Salinity Scale 1978 and the International
Equation of State of Seawater 1980. UNESCO Technical Papers in
Marine Science No. 36. Paris: UNESCO.

Welschmeyer, N. A. 1994. Fluorometric Analysis of Chlorophyll a in
the Presence of Chlorophyll 6 and Pheopigments. Limnology and
Oceanography, 39:1985-1992.

Wyrtki, K. 1964. Upwelling in the Costa Rica Dome. Fishery Bulletin,
63:355-372.

. 1967. Circulation and Water Masses in the Eastern Equatorial
Pacific Ocean. International Journal of Oceanology and Limnology,
1:117-147.

Xie, S.-P., H. Xu, W. S. Kessler, and M. Nonaka. 2005. Air—Sea Inter-
action over the Eastern Pacific Warm Pool: Gap Winds, Thermo-
cline Dome, and Atmospheric Convection. Journal of Climate,
18:5-20.

Growth and Nutrient Conservation
in Rhizophora mangle in Response
to Fertilization along Latitudinal

and Tidal Gradients
Ilka C. Feller, Catherine E. Lovelock,

and Cyril Piou

Ilka C. Feller, Smithsonian Environmental Re-
search Center, 647 Contees Wharf Road, Edge-
water, Maryland 21037, USA. Catherine E.
Lovelock, Centre for Marine Studies/School of
Integrative Biology, University of Queensland, St.
Lucia, Queensland 4072, Australia. Cyril Piou,
Institute of Forest Growth and Computer Science,
Technical University Dresden, Germany; currently
at Institut Nacional de la Recherche Agronomique,
Unité Mixte de Recherche-ECOBIOP, F-64310
Saint-Pée sur Nivelle, France. Corresponding
author: I. Feller (felleri@si.edu). Manuscript re-
ceived 21 July 2008; accepted 20 April 2009.

ABSTRACT. Mangroves form heterogeneous marine ecosystems with spatial differences
in structural complexity, biodiversity, biogeochemistry, and hydrology that vary at local
and regional scales. Although mangroves provide critical ecosystem goods and services,
they are threatened globally by human activities, including nutrient over-enrichment.
Our goal was to determine if enrichment with nitrogen (N) or phosphorus (P) interacts
with forest structure and latitude to alter growth and nutrient dynamics. We established
a series fertilization experiments across more than 2,000 km and 18° of latitude from the
Indian River Lagoon (IRL), Florida, to Twin Cays, Belize, to Bocas del Toro, Panama. At
each site, we fertilized individual trees with one of three treatment levels (control, +N,
+P) in two intertidal zones (fringe, scrub) and measured their responses for four years.
We tested the effects of nutrient over-enrichment on growth, resorption efficiency, and
resorption proficiency of the red mangrove Rhizophora mangle. All sites were nutrient
limited, but patterns of nutrient limitation varied by zone and latitude. At IRL, growth
was N limited; at Twin Cays, the fringe was N limited, but the scrub forest was P lim-
ited; at Bocas del Toro, the fringe was N limited, but the scrub forest was both N- and P
limited. Nutrient enrichment had dramatic and complex effects on nutrient conservation.
Adding nutrients to mangrove ecosystems affected growth and the nutrient recycling, but
the pattern depended on location, site characteristics, and the nature of nutrient limita-
tion. Predicting how forests will respond to nutrient over-enrichment requires an assess-
ment of spatial heterogeneity at multiple scales of response.

INTRODUCTION

Mangrove ecosystems are coastal wetlands dominated by woody plants that
span gradients in latitude (30°N to 37°S), tidal height (<1 m to >4 m), geomor-
phology (oceanic islands to riverine systems), sedimentary environment (peat to al-
luvial), climate (warm temperate to both arid and wet tropics), and nutrient loading
(oligotrophic to eutrophic). Throughout their distribution, mangroves are critical
not only for sustaining biodiversity in these intertidal forests but also for their di-
rect and indirect benefit to human activities. As a detritus-based ecosystem, the leaf
litter from these trees provides the basis for adjacent aquatic and terrestrial food
webs (Odum and Heald, 1975). Mangroves function as nurseries for many of the
sport and commercial fishes found in deeper waters and provide feeding grounds

346 °

for large reef fishes (Nagelkerken et al., 2000; Mumby et
al., 2004). As a result, mangrove-assimilated energy and
nutrients are exported to surrounding coral reefs (Dittmar
and Lara, 2001). Besides supporting and renewing coastal
fishing stock, mangroves also benefit human economic de-
velopment by stabilizing shorelines. This stabilization is a
critical function in tropical coastal areas that may be bat-
tered periodically by tropical storms, hurricanes, and tsuna-
mis (Danielson et al., 2005; Barbier, 2006).

Despite repeated demonstration of their ecological and
economic importance, mangroves are one of the world’s
most threatened ecosystems (Valiela et al., 2001; Alongi,
2002; Barbier and Cox, 2003; Rivera-Monroy et al., 2004;
Duke et al., 2007). In addition to direct destruction, increas-
ing input of human-caused nutrient pollution is widely rec-
ognized as one of the major threats to mangroves and other
marine environments worldwide (NRC, 1995, 2000, 2001;
Duce et al., 2008). However, system-specific attributes may
lead to large differences among coastal and estuarine sys-
tems in their sensitivity and susceptibility to these increas-
ing nutrient levels (Cloern, 2001). The complex suite of
direct and indirect responses in coastal systems to nutrient
over-enrichment include changes in water chemistry, distri-
bution and biomass of plants, sediment biogeochemistry,
decomposition processes, nutrient cycling, nutrient ratios,
phytoplankton communities, habitat quality for metazo-
ans, and ecosystem functions.

Relatively little is known about how the structure and
function of mangrove ecosystems are altered by nutrient
enrichment. In temperate salt marshes and mangroves, eco-
logical processes have been shown to be nitrogen- (N) lim-
ited (Valiela and Teal, 1979; Feller et al., 2003b). The few
tropical and subtropical mangrove wetlands that have been
studied were shown to be both phosphorus- (P) and N lim-
ited (Boto and Wellington, 1984; Feller, 1995; Feller et al.,
1999, 2003a, 2003b; Lovelock and Feller, 2003; Lovelock
et al., 2004). Because mangroves are responsive to processes
operating at multiple spatial scales, comparisons along a
broad latitudinal gradient in climate and across narrow
tidal gradients will improve our understanding of the rela-
tive impacts of global versus local factors on the structure
and function of these ecosystems. In this study, we focused
on the mangrove Rhizophora mangle (red mangrove), an
evergreen tree that has a large geographic range throughout
the Atlantic-East Pacific region (Duke, 1992). Along the At-
lantic coasts of North and South America, its distribution is
continuous and spans almost 60° of latitude from its north-
ern limit along the coast of Florida at 29°42.94'N (Zomle-
fer et al., 2006) to its southern limit along the coast of Bra-
zil at 27°53’S (Shaeffer-Novelli et al., 1990). In this study,

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

our goals were to determine how nutrient availability varies
among R. mangle forests spanning a temperate to tropical
gradient and how nutrient over-enrichment affects plant
growth and nutrient conservation. We manipulated nutri-
ent availability and measured responses of trees fertilized
with nitrogen (+N) or phosphorus (+P) growing along in-
tertidal gradients in similar habitats at three locations along
this latitudinal gradient to test the following hypotheses.

1. Nutrient availability varies along a latitudinal gradient
with a decreasing supply of P relative to N toward the
tropics (Vitousek, 1984; Vitousek and Sanford, 1986;
Crews et al., 1995). This hypothesis predicts increas-
ing P limitation in mangrove forests at lower latitudes
and N limitation at higher latitudes (Gisewell, 2004;
McGroddy et al., 2004; Reich and Oleksyn, 2004;
Kerkhoff et al., 2005).

2. Delivery, uptake, or assimilation of P is more strongly
affected by tidal flushing and concomitant factors that
vary spatially than is that of N (Smith, 1984; McKee
et al., 2002). This hypothesis predicts differences in N
versus P limitation within mangrove forests at different
intertidal elevations (Ross et al., 2006). Specifically, N
limitation is predicted for the low intertidal where tidal
flushing is greater (residence time is shorter) than in the
high intertidal where P limitation is predicted.

3. Because of difference in growth rates along climatic
gradients, the mechanisms used by plants to recycle and
conserve nutrients will be more efficient at higher lati-
tudes (Oleksyn et al., 2003). This hypothesis predicts
increased nutrient conservation by mangroves growing
near their temperate limit (Lovelock et al., 2007).

4. As nutrient availability increases, nutrient conservation
mechanisms become less efficient (Shaver and Melillo,
1984; Vitousek, 1984; Schlesinger et al., 1989; Escu-
dero et al., 1992). This hypothesis predicts that the ef-
fects of nutrient loading on mangrove forests will differ
depending on whether a system is N- or P limited, with
the expectation that the limiting nutrient will be more
efficiently and tightly conserved (Feller et al., 1999).

MATERIALS AND METHODS
Site DESCRIPTIONS

We compared the effects of nutrient over-enrichment
on plant growth and nutrient dynamics in Rhizophora man-
gle L. at three locations along the Atlantic and Caribbean
coasts from Florida to Panama spanning a climatic gradi-
ent of more than 2,000 km and 18° of latitude (Figure 1):

BRL,
Florida

Twin Cays,
Belize

Bocas del Toro,
Panama

FIGURE 1. The three study sites used in this study span more than
18° of latitude and extend from the Indian River Lagoon (IRL),
Florida, in the north, to Twin Cays, Belize, and to Bocas del Toro,
Panama, in the south.

(1) Indian River Lagoon (referred to hereafter as IRL),
Florida; (2) Twin Cays, Belize (referred to hereafter as Twin
Cays); and (3) Bocas del Toro, Republic of Panama (referred
to hereafter as Bocas) (Table 1). Table 2 provides a sum-
mary of the characteristics for the three locations (Koltes et
al., 1998; McKee et al., 2002; Feller et al., 2003a; Feller and
Chamberlain, 2007; Lovelock et al., 2005). Forest structure
at the three locations was heterogeneous and characterized
by complex gradients in tree height that included a narrow
seaward fringe of uniformly tall (~4 m) trees dominated

NUMBER 38 ¢ 347

by R. mangle, varying in width from 5 to 20 m (Figure 2).
Tree height decreased rapidly to landward with interior
areas dominated by old-growth stands of low stature, or
“scrub,” trees (~1.5 m) (Table 3). The black mangrove
(Avicennia germinans L.) and the white mangrove (Lagun-
cularia racemosa (L.) Gaertn. f.) were also present in each
of these locations, typically near the landward ecotone. The
hydrogeomorphic settings were variable among the three
locations. IRL and Bocas were continental in contrast with
Twin Cays, which is a low oceanic island. However, Twin
Cays and Bocas were more similar in mineralogy (Phillips et
al., 1997; Macintyre et al., 2004; Coates et al., 2005), with
mangrove forests atop a carbonate platform and deep peat
deposits. All sites were microtidal with mixed semidiurnal
tides (Kjerfve et al., 1982; Kaufmann and Thompson,
2005). The fringe zones at the three locations were similarly
well flushed, but the hydrological conditions of the scrub
zones varied. At Twin Cays, these interior portions of the
forest were completely inundated and waterlogged (McKee
et al., 2007). In contrast, the Bocas scrub zone drained com-
pletely at low tide (Lovelock et al., 2005). At IRL, the scrub
zone drained completely at low tide during the summer but
remained inundated for days during the winter (Feller et al.,
2003b).

In the IRL, our experimental sites were situated on the
lagoonal side of two barrier islands. The fringe site was in
Avalon State Park on North Hutchinson Island, St. Lucie
County; the scrub site was in the Hobe Sound National
Wildlife Refuge on Jupiter Island, Martin County. In
this area, soil was composed primarily of marine sand
with mangrove forests adjacent to coastal strand veg-
etation and maritime hammocks. Descriptions of forest

TABLE 1. Hydrogeomorphic characteristics of the study sites along a latitudinal gradient from the In-
dian River Lagoon (IRL), Florida, to Twin Cays, Belize, to Bocas del Toro (Bocas), Panama.

Characteristic IRL Twin Cays Bocas
Latitude 27°33'N, 80°13'W 16°50'N, 88°06'W 9°09'N, 82°15'W
Freshwater inflow Medium Low High

Type of landscape Continental Oceanic Continental
Topographic relief Medium Low High
Nutrient flux High Low Medium
Mineralogy Siliclastic/carbonate Peat/limestone Peat/limestone
Annual rainfall 1.3m 2.8m 3.5m

Mean temperature range #

Mean tidal range
Major disturbances

4w = winter; s = summer.

12.4°-23.6°C (w)
22.6°-31.9°C (s)
37 cm
Anthropogenic, hurricanes

18.3°-29.9°C (w)
22.2°-31.3°C (s)
34 cm
Anthropogenic, hurricanes

20.1°-31.1°C (w)
21.9°-31.8°C (s)
19 cm
Anthropogenic, flooding

348 © SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

TABLE 2. Characteristics of the mangrove forest structure in the fringe and scrub zones at the Indian River Lagoon (IRL), Florida, to
Twin Cays, Belize, to Bocas del Toro (Bocas), Republic of Panama. Data are from Koltes et al. (1998), McKee et al. (2002), Feller et al.
(2003a), Lovelock et al. (2005), and Feller and Chamberlain (2007).

Salinity (%o) Tree height (m) DBH (cm) Stem density Basal area

Location Zone (mean +SE) Species (mean + 1 SE) (mean +1SE) (stems-0.1 ha7!) (m?-0.1 ha~*)
IRL Fringe S27 32 0.7 Rhizophora mangle 39) = il 4.5 3,9536.4
Laguncularia racemosa 3 2 3} 6.1 1,3433.9
Avicennia germinans 3.8 + 0.3 4.8 6711.2
Scrub 32.4 + 0.5 Rhizophora mangle 17 2 OzAl Dns) 22 OL 2,3861.5
Laguncularia racemosa 4.5 + 0.4 4.6 + 0.7 4771.0
Avicennia germinans 1.6 + 0.0 1.2 + 0.0 730.01
Twin Cays Fringe 36.9 + 1.2 Rhizophora mangle 3) se OW 7.3 + 0.4 4012.1
Laguncularia racemosa 7).j)2 Dg? Se0s
Avicennia germinans Dd? 4.04 320.01 4
Scrub 394412 Rhizophora mangle 0.8 + 0.1 Di a= OD 8970.4
Bocas Fringe 34.4 + 0.6 Rhizophora mangle 39) se 0,11 5.3 + 0.6 8501.6
Scrub 33.3) 22 LY) Rhizophora mangle 0.7 += 0.1 1.5 22 O-il 3,3570.7

4 Based on occurrence of a single tree in each zone.

structure, hydro-edaphic conditions, growth, nutrient dy-
namics, and photosynthesis at the Avalon State Park site
were previously reported (Feller et al., 2003a; Lovelock
and Feller, 2003).

At Twin Cays, our fringe and scrub sites were located
on the two largest islands of this 92-ha mangrove archi-
pelago, 10 km offshore. Descriptions of forest structure,
biogeochemistry, ecophysiology, growth, and nutrient dy-
namics were previously reported (Rutzler and Feller, 1996;
McKee et al., 2002; Feller et al., 2003b, 2007; Lovelock

——Tree-height gradient =>

Shoreline
Fringe Zone

Interior
Scrub Zone

ah ay
ABM RIRAAR

o
om
ES
co)
jos
ie)

pueyowuiy
vy

=== [ntertidal gradient ->

FIGURE 2. Mangrove forests at each of our study sites are charac-
terized by a distinctive tree-height gradient with tall trees fringing the
shoreline and scrub trees in the interior.

et al., 2006a, 2006b, 2006c, 2006d). These oceanic man-
groves islands are underlain by deep deposits of mangrove
peat 8 to 12 m thick (Macintyre et al., 2004; McKee et
al., 2007).

At Bocas, fringe and scrub sites were located on three
islands (San Cristobal, Solarte, Isla Popa) in Almirante Bay
and the Chiriqui Lagoon in a vast network of mangrove
islands and mainland peninsulas covering approximately
2,885 km? (De Croz, 1993; Guzman and Guevara, 1998;
Guzman et al., 2005; Lovelock et al., 2004, 2005). Here,
mangroves occurred adjacent to tropical rainforests and
grew on peat approximately 5 m deep atop ancient coral
reef limestone (Phillips and Bustin, 1996; Phillips et al.,
1997). This location was outside the hurricane belt, but
flooding was common. Earthquakes are episodic (Phillips
et al., 1994, 1997; Phillips and Bustin, 1996) and are likely
to be the major nonanthropogenic disturbance regime in-
fluencing these forests.

EXPERIMENTAL DESIGN

Fertilization experiments were set up at IRL in Janu-
ary 1997, at Twin Cays in January 1995, and at Bocas in
January 1999. To compare responses, we used a three-way
factorial analysis of variance (ANOVA) design (i.e., 3 nu-
trient enrichment treatment levels [Control, +N, +P] xX 2
zones [fringe, scrub] X 3 locations [IRL, Twin Cays, Bo-
cas] X 3 sites per location X 3 replicate trees per site, for
a total of 162 trees). Nutrient treatment was randomly as-

NUMBER 38 ¢ 349

TABLE 3. Three-way factorial analysis of variance (ANOVA) results on the seven response variables: shoot elongation (Growth), N-, P-,
and K-resorption efficiencies (NRE, PRE, KRE), and N-, P-, and K-resorption proficiencies (NRP, PRP, KRP). The kind of transforma-
tion conducted on response variables for normalization and homogeneity of variances is given in the second line of column headings.
Results are in the form of F statistical values for each effect and the corresponding level of significance: ***P < 0.001; **P < 0.01;
*P < 0.05; and ~ for P < 0.1.

Growth NRE PRE KRE NRP PRP KRP
Log(x) Exp(x) Exp(4x) Exp(x) Log(x) Log(1000x) Log(x)
Factor df F P F P F P F P F P F P F P
Location (L) 2D; 24.2 en 28.6 ae 126 Fa 16.5 ee 86.1 asa 373 ae 18.9
Zone (Z) Ost 652, RO AST O05 TS 2 MAS) 5.18
Treatment (T) D; 23.4 Si 10.4 ae 75.4 vie 14.6 ee 40.3 ch 108 aia 36.8
eXeZ 2 12.0 seg 3.21 i 1.21 0.78 2.88 ~ Tell) ae 1.98
LxXT 4 12.9 BACT 4.37 aie 12.5 eit 5.60 ie 2.16 ~ 23.6 pee 8.13
ZXT 2 17.8 ag 3.27 HS 4.36 # 8.97 He 4.77 ae 13.4 pie 25.0
LT 4 A AS. 82 ABS 2 aaG BIT aE ATOR is ER AI Ga A SG
Residuals 140
signed within each zone and site. Trees were amended with LEAF NUTRIENT DYNAMICS
150 g N as NHg (45:0:0), or P fertilizer as P30; (0:45:0),
per centimeter diameter breast height, as described in To determine the relative effects of nutrient over-
Feller (1995). Doses (150 g) of fertilizer were sealed in enrichment on the ability of R. mangle to conserve nu-
dialysis tubing and placed in each of two holes 30 cm trients invested in foliage, we measured N, P, and potas-
deep, cored into the substrate on opposing sides of a tree sium (K) concentrations in green and senescent leaves.
beneath the outermost margin of its canopy, and sealed. For green leaves, we sampled the youngest, fully mature
Experiments at IRL and Twin Cays were fertilized twice green leaves from penapical stem positions in sunlit por-
per year. Because of limited access, the Bocas experiment tions of the canopy. Fully senescent yellow leaves with
was fertilized once per year. Thus, growth responses were well-developed abscission layers were taken directly
normalized to the annual rate of fertilizer application. For from the trees. Leaf area was determined with a Li-
controls, holes were cored and sealed but no fertilizer was Cor 3000 Leaf Area Meter (Lincoln, Neb., USA). Leaf
added. Direct fertilizer application to the root zone of our samples were dried at 70°C in a convection oven and
target trees was used because all sites were flooded at high ground in a Wiley Mill to pass through a 40 mesh (0.38
tides and fertilizer broadcasted on the surface would have mm) screen. Concentrations of carbon (C) and N were
washed away. determined with a Model 440 CHN Elemental Analyzer
(Exeter Analytical, North Chelmsford, Mass., USA) at
TREE GROWTH the Smithsonian Environmental Research Center, Edge-
water, Md. Concentrations of P and K were determined
To quantify growth, we measured the length of five using an inductively coupled plasma spectrophotometer
initially unbranched shoots in sunlit positions in the outer by Analytical Services, Pennsylvania State University, Pa.
part of the canopy of each tree at the three locations. To Nutrient concentrations expressed on a leaf area basis
compare growth responses among the three locations, (mg - cm 7) were used to calculate N, P, and K resorption
we calculated the annual shoot elongation based on the efficiencies (NRE, PRE, KRE), as below (Chapin and Van
amount fertilizer added per location (cm - year! kg~!). Cleve, 1989):

N, P, or K (mg - Cin) leaf ~ N, P, or K (mg - cm~)

: senescent leaf x 100
N, P, or K (mg - cm ~)

resorption efficiency =

green leaf

350 °

The absolute levels to which N, P, and K were reduced (%
dry mass) in senesced leaves (indicated as %Nenesced leafs
%P senesced leafy ANd YK senesced leafy respectively) were used di-
rectly as indices of N, P, and K resorption proficiencies
(NRP, PRP, KRP), as below (Killingbeck, 1996):

resorption proficiency = the level to which N, P, or K
has been reduced in senescent leaves (% dry mass).

Note that low levels for %oNenesced leafy 70Psenesced leafy and
%K senesced leaf are indicative of high resorption proficiency
whereas high levels indicate low resorption proficiency.
Concentrations less than 0.7% are considered complete
resorption for N and concentrations less than 0.04% are
considered complete resorption for P (Killingbeck, 1996).
Higher values indicate incomplete resorption. In this study,
we considered values less than 0.3% N and less than 0.01%
P as the ultimate resorption potential for R. mangle, as
proposed by Killingbeck (1996). Comparable values for K
resorption potential have not been determined.

STATISTICS

Our data were grouped by nutrient treatment (Con-
trol, +N, +P) X zone (fringe, scrub) X location (IRL,

: IRL
2 140 4
= 1204
§ 100 +
6 807
Say 4
S
TH 40 5
2 ADs
iS
” 0
Control +N +P Control

Twin Cays

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Twin Cays, Bocas), to compare seven response variables
of R. mangle, including growth responses, N-, P-, and
K-resorption efficiencies, and N-, P-, and K-resorption
proficiencies. Three-way factorial analyses of variance
(ANOVA) were applied for each response variable. When
an ANOVA found significant effects, Tukey’s honestly sig-
nificant difference (HSD) tests were applied to examine
pairwise differences within and among the treatment lev-
els. To respect the assumptions of heterogeneity of vari-
ances and normality, the response variables were trans-
formed using logarithms and exponentials. To investigate
relationships between nutrient content of green and senes-
cent leaves as well as among nutrient resorption proficien-
cies, we used the Spearman rho (p) correlation test on the
ranked row values. These analyses were conducted using
the R software 2.7.0 (R Development Core Team, 2008).

RESULTS
TREE GROWTH

There was a significant three-way interaction of nutri-
ent enrichment X location X zone on growth rates of R.
mangle trees (see Table 3; Figure 3). For control trees in the
fringe zone, the rate of shoot elongation at IRL was signifi-

Bocas del Toro

Control +N +P

Nutrient Treatment

FIGURE 3. Rhizophora mangle growth (cm - year !- kg!) measured as elongation of individual shoots per year (normalized to fertilizer ap-
plication at each site) at Indian River Lagoon (IRL), Twin Cays, and Bocas del Toro, in two zones (fringe, scrub), and in response to nutrient
enrichment with nitrogen (+N) or phosphorus (+P). (IRL and Twin Cays data from Feller et al., 2003a, 2003b).

cantly lower than at Bocas (HSD adjusted P < 0.001) but
similar to those at Twin Cays (HSD adjusted P = 0.070),
which had similar values. There were no significant dif-
ferences in shoot elongation rates for control trees in the
scrub zone among all the locations. +N caused significant
increases in shoot elongation rates for fringe and scrub
trees at IRL, but only for fringe trees at Twin Cays and Bo-
cas. However, shoot elongation for +N fringe trees in the
IRL was lower than observed at Bocas (HSD adjusted P =
0.089). +N caused similar increases in shoot elongation
in the fringe at Bocas and Twin Cays. In the scrub zone,
+P increased growth at Twin Cays (HSD adjusted P <
0.001) and Bocas (HSD adjusted P = 0.095), although
the rates were much higher for Twin Cays (HSD adjusted
P = 0.047). +P had no effect on growth in either fringe
or scrub zones at IRL. The +N treatment had no effect on
growth rates in the scrub zones at Twin Cays and Bocas.

NUTRIENT CONSERVATION

The impact of fertilization on N-, P-, and K-resorption
efficiencies varied by location and zone (Figure 4a—c). For
N-resorption efficiency (NRE), there was a significant
three-way interaction among location, zone, and nutrient
enrichment (see Table 3; Figure 4a). Values ranged from
26% to 68%. In control trees at all locations, NRE was
consistently highest for the fringe. At IRL, +N caused a
slight decline in values for fringe but not scrub trees. At
Twin Cays, +N had no effect on NRE in the fringe where
growth was N limited. However, +P caused an approxi-
mately 40% increase in NRE for the P-limited scrub trees
(HSD adjusted P < 0.001). Although +N had no effect
on the growth of scrub trees at Twin Cays, it did result in
a slight increase in NRE. Overall, values for NRE were
lowest at Bocas.

There were significant two-way interactions among
nutrient enrichment X location and nutrient enrichment
x zone on P-resorption efficiencies. However, the three-
Way interaction among nutrient enrichment X location X
zone was not significant (see Table 3, PRE; Figure 4b).
PRE values ranged from 36% to 80%. Overall, IRL had
the lowest PRE. Here, values for control fringe and scrub
trees were approximately half those at Twin Cays and Bo-
cas where values were similar. +N caused a slight increase
in PRE for IRL fringe and scrub trees. At Twin Cays and
Bocas, +N had no effect in either zone, but +P caused an
approximately 50% decrease in PRE for scrub trees and
an approximately 25% decrease for fringe trees.

For K-resorption efficiency (KRE), we found a signifi-
cant three-way interaction of nutrient enrichment X loca-

NUMBER 38 °¢ 351

tion X zone (see Table 3; Figure 4c). In the IRL, values
were uniformly low but positive in both zones, and nutri-
ent enrichment had no effect. For control fringe trees at all
locations, KRE was consistently positive. Overall, the low-
est KRE values occurred at Twin Cays. The negative val-
ues for senescent foliage from control scrub trees at Twin
Cays and Bocas indicated that K accumulated in leaves
rather than being resorbed by the plant during senescence.
At Twin Cays and Bocas, +P caused a significant increase
in KRE by scrub trees, but had little effect on fringe trees.
However, +N had no significant effect on KRE in either
zone.

Fertilization also had striking and complex effects on
resorption proficiencies, measured as the %Nenesced leafs
%P senesced leafy ANd %K senesced leafy that varied by location and
zone (Table 3; Figure Sa—c). Concentrations of N, P, and
K in senesced leaves were positively associated with their
concentrations in green leaves (Spearman p values for N, P,
and K = 0.52, 0.87, and 0.65, respectively, all significantly
different than 0 with P < 0.0001). There was no relation-
ship between YoNenesced leaf ANd Y%Prenesced leat (Spearman
p = 0.03, P = 0.66), but %K senesced leat Was significantly
correlated with %Neenesced leaf (Spearman p = 0.19, P =
0.02) and with %Prenesced leat (Spearman p = —0.43, P <
0.0001). For NRP, there was a significant three-way in-
teraction among location, zone, and nutrient enrichment
(see Table 3; Figure 5a). The %Noenesced leat ranged from a
low of 0.28% for +P scrub trees at Twin Cays to a high
of 0.91% in +N fringe trees at Bocas. For control trees
from the fringe and scrub zones, values were similar at
IRL and Bocas but were significantly lower at Twin Cays,
which indicated increased NRP. +N caused an increase of
20% in YNenesced leaf from the fringe at IRL but had little
effect on fringe trees at the other locations. In the scrub
zone, +N had no effect on %Neenesced leaf at IRL and Twin
Cays, but significantly higher values at Bocas resulted in
a decrease in NRP. +P had little effect on either fringe or
scrub zones at IRL and Bocas, but it caused a dramatic
decrease in %Nenesced leat aNd a corresponding increase in
NRP in scrub trees at Twin Cays.

We found the highest levels of %Peenesced leaf (~0-06 %)
in the control trees in both zones at IRL, which indicated
low PRP compared to Twin Cays and Bocas. Fertilization
with +N or +P had no detectable effect on these levels at
IRL (Figure 4b; all HSD adjusted P > 0.5). Very low lev-
els (~0.01%) of YP senesced leaf in both zones at Twin Cays
and Bocas indicated high PRP in the range of maximal P
resorption (Figure 6). +N had no effect on values in either
zone at Twin Cays or Bocas. +P caused the most dramatic
increase in %Penesced leafy With a concomitant decrease in

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

352 e

(%)SeAe9| JUBDSSUSS

N

(%) SSAPBD] JUBDSOUSS

(%) SSABD| JUBISSUDSy,

Nutrient Enrichment Treatment

sium (K) at Indian River Lagoon (IRL), Twin Cays, and

FIGURE 4. Resorption

se to nutrient enrichment with nitrogen (+N) or

nd in respon

),a

hatched bars

(uo) —

). (IRL and Twin Cays data from Feller et al., 2003a, 2003b).

two zones

) in

del Toro (Bocas

Bocas

e 353

NUMBER 38

Adusaloiyj4 uondiosay N

Adualoijj3 uondiosey qd

Adusldijj4 uondiosey y

+N +P Control +N +P Control +N +P

Control

ee Eninge

Nutrient Enrichment Treatment

Scrub

LLL

s, and (c) potassium at Indian River Lagoon (IRL), Twin Cays, and Bocas

to nutrient enrichment with nitrogen

proficienc

FIGURE 5. Resorption

del Toro

(+P).

phosphorus

(+N) or

d in response

(Bocas

354 e

Ultirnate potential resorption of P

Test resorption of P
eee resorption of P

1.0

S =
n co

N ITOGEN senesced leaves (%)
a

=
iu

0.00 0.01 0.02 0.03 0.04 0.05

Phosphorus

senesced kaves (

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Ultimate potential resorption of N

0.06 0.07 0.08

FIGURE 6. N-resorption proficiency (%Nenesced leaf) Versus P-resorption proficiency (%Penesced leat) for Rhi-
zophora mangle by location (Indian River Lagoon [IRL], Twin Cays, Bocas del Toro), nutrient enrichment
treatment level (control [0], +N [oJ], +P [A]), and intertidal zone (fringe [closed symbols] or scrub [open

symbols]).

PRP, for both the fringe and scrub zones at Twin Cays
(both HSD adjusted P < 0.001). At Bocas, there was a
similar increase in %P senesced leaf With +P in the scrub zone
(HSD adjusted P < 0.001), whereas the response in the
fringe zone was comparatively smaller and not significant
(HSD adjusted P = 0.694).

There was also a significant three-way interaction
among location, zone, and nutrient enrichment on KRP
(see Table 3; Figure Sc). Values for control trees in both
zones at IRL and Bocas were similar with no differences
between zones. Neither +N nor +P had any effect at IRL,
but +P caused a significant decrease in %K cenesced leaf IN
both zones at Bocas. The %K senesced leat ranged from a low
of 0.26% for +P scrub trees at Bocas to a high of 1.56%
in control scrub trees at Twin Cays, which was more than
double the K concentrations in senescent foliage of fringe

trees. In the Twin Cays scrub zone, +P caused a fourfold
decrease in %K cenesced leafy resulting in an associated increase
in KRP.

DISCUSSION

Long-term fertilization experiments at IRL, Twin
Cays, and Bocas del Toro demonstrated that these three
locations, which were arrayed along a latitudinal gradient,
were nutrient limited. However, system-specific attributes
resulted in significant differences in patterns of nutrient
limitation and responses to fertilization. Although the
mangrove ecosystems at these locations exhibited similar
tree-height gradients dominated by Rhizophora mangle,
they differed in several hydrogeomorphic and structural

features (see Tables 1, 2). The locations also differed in
substrate types; that is, the soil at IRL was composed of
Pleistocene marine sands while the soils at Twin Cays and
Bocas site were deep deposits (6-12 m) of mangrove peat
formed during the Holocene (Phillips and Bustin, 1996;
Lovejoy, 1998; Macintyre et al., 2004). Our experimental
site in the IRL was in a young forest, less than 40 years
old, in an abandoned mosquito impoundment (Rey et
al., 1986). In contrast, the experiments at Twin Cays and
Bocas were in old-growth forests. Although no data are
available for a direct comparison, it is likely that the for-
ests at Bocas are older than at Twin Cays because of dif-
ferences in their exposures to hurricanes (Stoddart, 1963;
Carruthers et al., 2005). Overall, stem density was lowest
at Twin Cays. Stem density in the IRL fringe was approxi-
mately 10 times greater than at Twin Cays and 4 times
greater than at Bocas. On the other hand, the density of
trees in the scrub forest was highest at Bocas.

Growth of R. mangle stems, which we used as a bio-
assay of nutrient limitation in our fertilization experiments,
varied among IRL, Twin Cays, and Bocas. However, the
responses did not support Hypothesis 1 of increasing P
limitation toward the tropics (Vitousek, 1984; Vitousek
and Sanford, 1986; Crews et al., 1995). This hypothesis
predicted that P limitation would be greatest at Bocas,
which was located at the lowest of the three latitudes com-
pared in this study. Instead, shoot elongation indicated an
order that ranged from N limitation in both fringe and
scrub zones at IRL, to N limitation in fringe and scrub as
well as P limitation in scrub at Bocas, and to N limitation
in fringe and P limitation in scrub at Twin Cays. The mag-
nitude of the growth responses to fertilization with the
limiting nutrient at each location was also consistent with
this order, that is, IRL < Bocas < Twin Cays, with the
most severe P limitation and the greatest growth response
to P fertilization in the scrub zone at Twin Cays.

The differences in growth responses that we observed
at the three locations suggest that nutrient limitation
within and among mangrove ecosystems is likely deter-
mined by several features of their geomorphology, includ-
ing sediment/nutrient flux, tidal range, and substrate type.
These findings contrast with other studies that attribute P
limitation in the tropics mainly to differences in the age
of soils between tropical and temperate regions, with the
most P-limited forests on the oldest soils (Vitousek, 1984;
Vitousek and Sanford, 1986; Crews et al., 1995; Giisewell,
2004; McGroddy et al., 2004; Reich and Oleksyn, 2004;
Kerkhoff et al., 2005).

Based on findings from Twin Cays (Feller, 1995; Feller
et al., 2003a, 2007), McKee et al. (2002) hypothesized

NUMBER 38 e¢ 355

that the shift from N limitation in fringe zone around the
periphery of the island to severe P limitation in scrub zone
in the interior was the result of differences in factors as-
sociated with tidal flushing. Our results from the other
two locations compared in this study partially support
this hypothesis. Although all locations were N limited in
the fringe, growth in the scrub zone at Bocas was limited
by both N and P. This finding again differs from the IRL
where growth was N limited in both zones (Feller et al.,
2003b). These patterns along tidal gradients indicate that
differences in nutrient limitation among the three loca-
tions are the result of variations in tidal flushing, external
nutrient supply, substrate, and endogenous biological pro-
cesses. The scrub forests in the interior areas have a low
tidal exchange and a low supply of exogenous nutrients,
whereas the fringe zones are well flushed with a higher net
exchange of nutrients. Mangroves at the IRL and Bocas
locations are in continental settings with medium to high
relief, freshwater inflow, and nutrient flux. However, their
tidal regimes and underlying soils differ dramatically. In
contrast with IRL where mangroves are growing on sandy
soils, mangroves at Twin Cays and Bocas are growing on
peat. Although both of these locations are associated with
low-nutrient coral reef ecosystems, Twin Cays receives
negligible terrigenous inputs of freshwater or sediments
whereas Bocas mangroves experiences a high flux of nu-
trients from several rivers draining into the archipelago. In
addition, patterns of nutrient limitation in these systems
may be affected by local patterns of N> fixation (Joye and
Lee, 2004; Borgatti, 2008).

Resorption of phloem-mobile nutrients from leaves
during senescence is an important nutrient conservation
strategy for plants that influences many ecological pro-
cesses, including primary production, nutrient uptake,
competition, and nutrient cycling (Chapin, 1980). To
resolve the relative degree to which latitude and nutri-
ent enrichment affect the ability of R. mangle to conserve
nutrients invested in foliage, we examined resorption of
N, P, and K. Across location, zone, and nutrient treat-
ment levels, our results indicate that a major control of
the nutrient concentrations in senesced leaves was nutrient
concentration in green leaves, which is consistent with a
global dataset compiled by Kobe et al. (2005). Specifically,
concentrations of N, P, and K in senesced leaves were posi-
tively associated with their concentrations in green leaves.
In contrast to Oleksyn et al. (2003), who predicted that
nutrient resorption efficiencies should increase with lati-
tude, we found the lowest efficiencies at IRL, our north-
ernmost location, consistent with Lovelock et al. (2007).
We also found the most efficient nutrient conservation for

356 °

N and P at Twin Cays, the location positioned at the in-
termediate latitude. Although the levels to which nutrients
were conserved varied by nutrient, location, and zone, the
patterns did not fall clearly along a latitudinal gradient.
All experimental trees at the three locations, except for the
+N trees in the scrub zone at Bocas, had less than 0.7%
N concentrations in their senescent leaves, which is within
the range of complete resorption in the model proposed by
Killingbeck (1996) (see Figure 6). In the Twin Cays fringe
and the +P-fertilized scrub trees, the N concentration in
senesced leaves was less than 0.3%, which was found to
be the maximal level to which N can be reduced in senes-
cent leaves of evergreen species and is regarded by Killing-
beck (1996) as the ultimate potential resorption for N. In
Killingbeck’s model, less than 0.04% Prenesced leaf represents
complete resorption of P for evergreens. All experimental
trees at Twin Cays and Bocas had values below this thresh-
old and thus exhibited complete P resorption. Moreover,
control and +N trees in the scrub and fringe zones at Twin
Cays and Bocas had 0.01% Prenesced leaf OF less, which is the
maximal level to which P can be reduced in senescent leaves
in evergreens representing the ultimate potential resorption
of P. Comparable levels of %Peenesced leat have been reported
for mangroves elsewhere (Alongi et al., 2005). In contrast,
all the trees at IRL had values for Prenesced eat greater than
0.04%, which represents incomplete resorption. In contrast
to suggestions by Aerts and Chapin (2000), the results pre-
sented here indicate there are nutritional controls on nutrient
resorption in R. mangle. Nutrient enrichment clearly altered
resorption of N and P at Twin Cays and Bocas but had no
effect at IRL. Enrichment with +P resulted in increases in N
and K resorption efficiency and proficiency at Twin Cays and
Bocas but had the opposite effect on P resorption. Similarly,
+N decreased N resorption, but only in the N-fertilized
trees in the scrub zone at Bocas. These findings suggest that
P enrichment may have either increased the requirements for
N and K in R. mangle or it may have increased its physi-
ological capacity to conserve these nutrients during leaf se-
nescence. Increased resorption of N and K in response to
+P may also indicate that under P-limiting conditions these
nutrients become limiting when P is added to the system.
Although we found no relationships between growth and N
or P concentrations in green leaves, we did observe a weak
but significant relationship between %Kegreen teas and growth
rates (r = 0.230, F = 8.723, P < 0.01). These results indi-
cate that K availability may be important to the structure
and function of some mangrove forests (Kathiresan et al.,
1994), which warrants further study.

In conclusion, our results indicate that nutrient over-
enrichment of the coastal zone will alter forest structure and

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

nutrient dynamics in mangrove ecosystems. We showed that
fertilization altered growth and nutrient conservation in R.
mangle, but the patterns did not correspond with a latitu-
dinal gradient. Growth was consistently N limited for trees
in fringing forests, which have higher water exchange rates
compared to scrub forests, supporting the hypothesis of
Smith (1984) and McKee et al. (2002) that open systems are
more likely to be N limited than P limited. In the IRL, scrub
trees in the interior of the forest were also N limited. Pat-
terns of nutrient limitation became more complex at lower
latitudes. Phosphorus limitations characterized the scrub
zone at Twin Cays whereas both N and P limitations were
widespread in the scrub zone at Bocas. Our results clearly in-
dicated that the phenotypic potential of R. mangle to resorb
N, P, and K from senescing leaves varied as a function of
nutrient availability, which was driven by differences in hy-
drology and substrate along latitudinal and tidal gradients.

ACKNOWLEDGMENTS

We thank the Governments of Belize and Panama for
permission to use study sites at Twin Cays and at Bocas
del Toro; Klaus Riitzler, Valerie Paul, and Rachel Collin
for support and permission to work at the Smithsonian
marine laboratories at Carrie Bow Cay, Fort Pierce, and
Bocas del Toro; Michael Carpenter, Woody Lee, and Ga-
briel Jacome for logistical arrangements; Anne Chamber-
lain for assistance in the laboratory and in the field. Finan-
cial support was provided by the Caribbean Coral Reef
Ecosystems Program (CCRE), Smithsonian Institution; the
Smithsonian Marine Station at Fort Pierce (SMSFP); and
the National Science Foundation DEB-9981535. This is
contribution number 842 of the CCRE, supported in part
by the Hunterdon Oceanographic Research Fund, and
SMSFP contribution no. 757.

LITERATURE CITED

Aerts, R., and FE. S. Chapin III. 2000. The Mineral Nutrition of Wild
Plants Revisited: A Re-Evaluation of Processes and Patterns. Ad-
vances in Ecological Research, 30:1-67.

Alongi, D. M. 2002. Present State and Future of the World’s Mangrove
Forests. Environmental Conservation, 29:331-349.

Alongi, D. M., B. FE. Clough, and A. I. Robertson. 2005. Nutrient-Use
Efficiency in Arid-Zone Forests of the Mangroves Rhizophora sty-
losa and Avicennia marina. Aquatic Botany, 82:121-131.

Barbier, E. B. 2006. Natural Barriers to Natural Disasters: Replanting
Mangroves after the Tsunami. Frontiers in Ecology and the Envi-
ronment, 4:124-131.

Barbier, E. B., and M. Cox. 2003. Does Economic Development Lead
to Mangrove Loss? A Cross-Country Analysis. Contemporary Eco-
nomic Policy, 21:418-432.

Borgatti, A. R. 2008. Nutrient Enrichment Effects on Mangrove Sedi-
ments in Differing Tree Height Zones in Bocas del Toro, Panama.
Master’s thesis, University of Maryland, College Park, Md.

Boto, K. G., and J. T. Wellington. 1984. Soil Characteristics and Nutri-
ent Status in a Northern Australian Mangrove Forest. Estuaries,
71:61-69.

Carruthers, T. J. B., P. A. G. Barnes, G. Jacome, and J. W. Fourqurean.
2005. Lagoon Scale Processes in a Coastally Influenced Caribbean
System: Implications for the Seagrass Thalassia testudinum. Carib-
bean Journal of Science, 41:441-455.

Chapin, EF. S., II. 1980. The Mineral Nutrition of Wild Plants. Annual
Review of Ecology and Systematics, 11:233-260.

Chapin, F. S., III, and K. Van Cleve. 1989. “Approaches to Studying
Nutrient Uptake, Use and Loss in Plants.” In Plant Physiological
Ecology: Field Methods and Instrumentation, ed. R. W. Pearcy,
J. R. Ehleringer, H. A. Mooney, and P. W. Rundel, pp. 185-207.
New York: Chapman & Hall.

Cloern, J. E. 2001. Our Evolving Conceptual Model of the Coastal Eutro-
phication Problem. Marine Ecology Progress Series, 210:223-253.

Coates, A. G., D. F McNeill, M.-P. Aubry, W. A. Berggren, and L. S. Col-
lins. 2005. An Introduction to the Geology of the Bocas del Toro
Archipelago, Panama. Caribbean Journal of Science, 41:374-391.

Crews, T. E., K. Kitayama, J. H. Fownes, R. H. Riley, D. A. Herbert, D.
Mueller-Dombois, and P. M. Vitousek. 1995. Changes in Soil Phos-
phorus Fractions and Ecosystem Dynamics Across a Long Chrono-
sequence in Hawaii. Ecology, 76:1407-1424.

Danielson, F., M. K. Sorensen, M. F. Olwig, V. Selvam, F. Parish, N. D.
Burgess, T. Hiraishi, V. M. Karunagaran, M. S. Rasmussen, L. B.
Hansen, A. Quarto, and N. Suryadiputra. 2005. The Asian Tsu-
nami: A Protective Role for Coastal Vegetation. Science, 310:643.

D’Croz, L. 1993. “Status and Uses of Mangroves in the Republic of Pan-
ama.” In Proceedings of a Workshop on Conservation and Sustain-
able Utilization of Mangrove Forests in Latin America and Africa
Regions. Part 1. Latin America, ed. L. D. Lacerda, pp. 115-127.
Okinawa, Japan: International Society for Mangrove Ecosystems
and International Tropical Timber Organization.

Dittmar, T., and R. J. Lara. 2001. Do Mangroves Rather Than Rivers
Provide Nutrients to Coastal Environments South of the Amazon
River? Evidence from Long-Term Flux Measurements. Marine
Ecology Progress Series, 213:67-77.

Duce, R. A., J. LaRoche, K. Altieri, K. R. Arrigo, A. R. Baker, D. G.
Capone, S. Cornell, F. Dentener, J. Gallow, R. S. Ganeshram, R. J.
Geider, T. Jickells, M. M. Kuypers, R. Langlois, P. S. Liss, S. M.
Liu, J. J. Middelburg, C. M. Moore, S. Nickovic, A. Oschlies, T.
Pedersen, J. Prospero, R. Schlitzer, L. L. Sorensen, M. Uematsu,
O. Ulloa, M. Voss, B. Ward, and L. Zamora. 2008. Impacts of At-
mospheric Anthropogenic Nitrogen on the Open Ocean. Science,
320:893-897.

Duke, N. C. 1992. “Mangrove Floristics and Biogeography.” In Tropi-
cal Mangrove Ecosystems, ed. A. I. Robertson and D. M Alongi,
pp- 63-100. Washington, D.C.: American Geophysical Union.

Duke, N. C., J.-O. Meynecke, S. Dittmann, A. M. Ellison, K. Anger,
U. Berger, S. Cannicci, K. Diele, K. C. Ewel, C. D. Field, N.
Koedam, S. Y. Lee, C. Marchand, I. Nordhaus, and F. Dahdouh-
Guebas. 2007. A World Without Mangroves? Science, 317:41-42.

Escudero, A., J. M. del Arco, I. C. Sanz, and J. Ayala. 1992. Effects
of Leaf Longevity and Retranslocation Efficiency on the Retention
Time of Nutrients in the Leaf Biomass of Different Wood Species.
Oecologia (Berlin), 90:80-87.

Feller, I. C. 1995. Effects of Nutrient Enrichment on Growth and Her-
bivory in Dwarf Red Mangrove (Rhizophora mangle). Ecological
Monographs, 65:477-505.

Feller, I. C., and A. H. Chamberlain. 2007. Herbivore Responses to Nu-
trient Enrichment and Landscape Heterogeneity in a Mangrove
Ecosystem. Oecologia (Berlin), 153:607-616.

NUMBER 38 °¢ 357

Feller, I. C., C. E. Lovelock, and K. L. McKee. 2007. Nutrient Addition
Differentially Affects Ecological Processes of Avicennia germinans
in Nitrogen vs. Phosphorus Limited Mangrove Ecosystems. Ecosys-
tems, 10:347-359.

Feller, I. C., K. L. McKee, D. F Whigham, and J. P. O’Neill. 2003a. Ni-
trogen vs. Phosphorus Limitation across an Ecotonal Gradient in a
Mangrove Forest. Biogeochemistry, 62:145-175.

Feller, I. C., D. E Whigham, K. L. McKee, and C. E. Lovelock. 2003b.
Nitrogen Limitation of Growth and Nutrient Dynamics in a Dis-
turbed Mangrove Forest, Indian River Lagoon, Florida. Oecologia
(Berlin), 134:405-414.

Feller, I. C., D. EF Whigham, J. P. O’Neill, and K. M. McKee. 1999. Ef-
fects of Nutrient Enrichment on Within-Stand Nutrient Cycling in
Mangrove Ecosystems in Belize. Ecology, 80:2193-2205.

Giisewell, $. 2004. N:P Ratios in Terrestrial Plants: Variation and Func-
tional Significance. New Phytologist, 164:243-266.

Guzman, H. M., P. A. G. Barnes, C. E. Lovelock, and I. C. Feller. 2005.
CARICOMP: Mangrove, Seagrass and Coral Reef Sites in Bocas del
Toro, Panama. Caribbean Journal of Science, 41:430-440.

Guzman, H. M., and C. A. Guevara. 1998. Bocas del Toro, Panama
Coral Reefs: I. Distribution, Structure, and Conservation State of
Continental Reefs in the Laguna de Chiriqui and Bahia Almirante.
Revista de Biologia Tropical, 46:601-623.

Joye, S. B., and R. Y. Lee. 2004. Benthic Microbial Mats: Important
Sources of Fixed Nitrogen and Carbon to the Twin Cays, Belize
Ecosystem. Atoll Research Bulletin, 528:1-24.

Kathiresan, K., P. Moorthy, and N. Rajendran. 1994. Seedling Perfor-
mance of Mangrove Rhizophora apiculata (Rhizophorales: Rhi-
zophoraceae) in Different Environs. Indian Journal of Marine Sci-
ences, 23:168-169.

Kaufmann, K. W., and R. C. Thompson. 2005. Water Temperature Varia-
tion and the Meteorological and Hydrographic Environment of Bocas
del Toro, Panama. Caribbean Journal of Science, 41:392-413.

Kerkhoff, A. J., B. J. Enquist, W. F. Fagan, and J. J. Elser. 2005. Plant Al-
lometry, Ecological Stoichiometry and the Temperature-Dependence
of Terrestrial Primary Productivity. Global Ecology and Biogeogra-
phy, 14:585-598.

Killingbeck, K. T. 1996. Nutrients in Senesced Leaves: Keys to the Search
for Potential Resorption and Resorption Proficiency. Ecology,
77:1716-1727.

Kjerfve, B., K. Riitzler, and G. H. Kierspe. 1982. “Tides at Carrie Bow
Cay, Belize.” In The Atlantic Barrier Reef Ecosystem at Carrie
Bow Cay, Belize, I. Structure and Communities, ed. K. Riitzler and
I. G. Macintyre, pp. 47-51. Smithsonian Contributions to the Marine
Sciences, No. 12. Washington, D.C.: Smithsonian Institution Press.

Kobe, R. K., C. A. Lepcyzk, and M. Iyer. 2005. Resorption Efficiency
Decreases with Increasing Green Leaf Nutrients in Global Dataset.
Ecology, 86:2780-2792.

Koltes, K., J. Tschirky, and I. C. Feller. 1998. “Carrie Bow Cay, Belize.”
In CARICOMP: Caribbean Coral Reef, Seagrass and Mangrove
Sites. Coastal Region and Small Island Papers 3, ed. B. Kjerfve,
pp. 79-94. Paris: UNESCO.

Lovejoy, D. W. 1998. Classic Exposures of the Anastasia Formation in
Martin and Palm Beach Counties, Florida. Miami, Fla.: Miami
Geological Society.

Lovelock, C. E., M. C. Ball, B. Choat, B. M. J. Engelbrecht, N. M. Hol-
brook, and I. C. Feller. 2006a. Linking Physiological Processes with
Mangrove Forest Structure: Phosphorus Deficiency Limits Canopy
Development, Hydraulic Conductivity and Photosynthetic Carbon
Gain in Dwarf Rhizophora mangle. Plant, Cell and Environment,
29:793-802.

Lovelock, C. E., M. C. Ball, I. C. Feller, B. M. J. Engelbrecht, and M. L.
Ewe. 2006b. Variation in Hydraulic Conductivity of Mangroves:
Influence of Species, Salinity, and Nitrogen and Phosphorus Avail-
ability. Physiologia Plantarum, 127:457-464.

358 °

Lovelock, C.E., and I. C. Feller. 2003. Photosynthetic Performance and
Resource Utilization of Two Mangrove Species Coexisting in Hy-
persaline Scrub Forest. Oecologia (Berlin), 134:455—462.

Lovelock, C. E., I. C. Feller, M. C. Ball, J. Ellis, and B. Sorrell. 2007. Test-
ing the Growth Rate vs. Geochemical Hypothesis for Latitudinal
Variation in Plant Nutrients. Ecology Letters, 10:1154-1163.

Lovelock, C. E., I. C. Feller, M. C. Ball, B. M. J. Engelbrecht, and M. L.
Ewe. 2006c. Differences in Plant Function in Phosphorus and Nitro-
gen Limited Mangrove Ecosystems. New Phytologist, 172:514-522.

Lovelock, C. E., I. C. Feller, K. L. McKee, B. M. J. Engelbrecht, and
M. C. Ball. 2004. The Effect of Nutrient Enrichment on Growth,
Photosynthesis and Hydraulic Conductance of Dwarf Mangroves
in Panama. Functional Ecology, 18:25-33.

Lovelock, C. E., I. C. Feller, K. L. McKee, and R. Thompson. 2005. For-
est Structure of the Extensive Caribbean Mangrove Forests of Bocas
del Toro, Panama. Caribbean Journal of Science, 41:456-464.

Lovelock, C. E., R. W. Ruess, and I. C. Feller. 2006d. Fine Root Respira-
tion in the Mangrove Rhizophora mangle over Variation in Forest
Stature and Nutrient Availability. Tree Physiology, 26:1601-1606.

Macintyre, I. G., M. A. Toscano, R. G. Lighty, and G. B. Bond. 2004.
Holocene History of the Mangrove Islands of Twin Cays, Belize,
Central America. Atoll Research Bulletin, 510:1-16.

McGroddy, M. E., T. Daufresne, and L. O. Hedin. 2004. Scaling of C:N:P
Stoichiometry in Forests Worldwide: Implications of Terrestrial
Redfield-Type Ratios. Ecology, 85:2390-2401.

McKee, K. L, D. Cahoon, and I. C. Feller. 2007. Caribbean Mangroves
Adjust to Rising Sea-Level Through Biotic Controls on Soil Eleva-
tion Change. Global Ecology and Biogeography, 16:546-556.

McKee, K. L., I. C. Feller, M. Popp, and W. Wanek. 2002. Mangrove
Isotopic Fractionation (8;;N and 6;3C) across a Nitrogen Versus
Phosphorus Limitation Gradient. Ecology, 83:1065-1075.

Mumby, P. J., A. J. Edwards, J. E. Arias-Gonzalez, K. C. Lindeman, P. G.
Blackwell, A. Gall, M. I. Gorczynska, A. R. Harborne, C. L. Pescod,
H. Renken, C. C. C. Wabnitz, and G. Llewellyn. 2004. Mangroves
Enhance the Biomass of Coral Reef Fish Communities in the Carib-
bean. Nature (London), 427:533-536.

Nagelkerken, I., G. van der Velde, M. W. Gorissen, G. J. Meijer, T. van’t
Hof, and C. den Hartog. 2000. Importance of Mangroves, Seagrass
Beds and the Shallow Coral Reef as a Nursery for Important Coral
Reef Fishes, Using a Visual Census Technique. Estuarine, Coastal
and Shelf Science, 51:31-44.

National Research Council (NRC). 1995. Priorities for Coastal Ecosys-
tem Science. Washington, D.C.: National Academy Press.

. 2000. Clean Coastal Waters: Understanding and Reducing the

Effects of Nutrient Pollution. Washington, D.C.: National Acad-

emy Press.

. 2001. Grand Challenges in Environmental Science. Washington,
D.C.: National Academy Press.

Odum, W. E., and E. J. Heald. 1975. “The Detritus-Based Food Web of
an Estuarine Mangrove Community.” In Estuarine Research, ed. L. E.
Cronin, pp. 265-286. New York: Academic Press.

Oleksyn, J., P. B. Reich, R. Zytkowiak, P. Karolewski, and M. G. Tjo-
elker. 2003. Nutrient Conservation Increases with Latitude of Ori-
gin in European Pinus sylvestris. Oecologia (Berlin), 136:220-235.

Phillips, S., and R. M. Bustin. 1996. Sedimentology of the Changuinola
Peat Deposit: Organic and Clastic Sedimentary Response to Punc-
tuated Coastal Subsidence. Geological Society of America Bulletin,
108:794-814.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Phillips, S., R. M. Bustin, and L. E. Lowe. 1994. Earthquake Induced
Flooding of a Tropical Coastal Peat Swamp: A Modern Analogue
for High-Sulfur Coals? Geology, 22:929-932.

Phillips, S., G. E. Rouse, and R. M. Bustin. 1997. Vegetation Zones
and Diagnostic Pollen Profiles of a Coastal Peat Swamp, Bocas del
Toro, Panama. Paleogeography, Paleoclimatology, Paleoecology,
128:301-338.

R Development Core Team. 2008. An Introduction to R: Notes on
R: A Programming Environment for Data Analysis and Graph-
ics. Version 2.7.1. Vienna, Austria: R Foundation for Statistical
Computing.

Reich, P. B., and J. Oleksyn. 2004. Global Patterns of Plant Leaf N and
P in Relation to Temperature and Latitude. Proceedings of the
National Academy of Sciences of the United States of America,
101:11001-11006.

Rey, J. R., R. Crossman, T. R. Kain, and D. S. Taylor. 1986. “An Over-
view of Impounded Mangrove Forests along a Subtropical Lagoon
in East-Central Florida, U.S.A.” In The Mangrove, ed. L. Bhosale,
pp. 396-402. Kolhapur, India: Shivaji University Press.

Rivera-Monroy, V. H., R. R. Twilley, D. Bone, D. L. Childers, C. Coronado-
Molina, I. C. Feller, J. Herrera-Silveira, R. Jaffe, E. Mancera, E.
Rejmankova, J. E. Salisbury, and E. Weil. 2004. A Biogeochemical
Conceptual Framework to Develop Long Term Ecological Research
and Sustain Coastal Management in the Wider Caribbean Region.
BioScience, 54:843-856.

Ross, M., S. Mitchell-Bruker, J. Sah, S. Stothoff, P. Ruiz, D. Reed, K.
Jayachandran, and C. Coultas. 2006. Interaction of Hydrology
and Nutrient Limitation in the Ridge and Slough Landscape of the
Southern Everglades. Hydrobiologia, 569:37-59.

Riitzler, K., and I. C. Feller. 1996. Caribbean Swamp Communities. Sci-
entific American, 274(3):70-75.

Schlesinger, W. H., E. H. DeLucia, and W. D. Billings. 1989. Nutrient-
Use Efficiency of Woody Plants on Contrasting Soils in the Western
Great Basin, Nevada. Ecology, 70:105-113.

Shaeffer-Novelli, Y., G. Cintrén-Molero, R. R. Adaime, and T. M. Ca-
margo. 1990. Variability of Mangrove Ecosystems along the Brazil-
ian Coast. Estuaries, 13:204-218.

Shaver, G. R., and J. M. Melillo. 1984. Nutrient Budgets of Marsh
Plants: Efficiency Concepts and Relation to Availability. Ecology,
65:1491-1510.

Smith, S. V. 1984. Phosphorus Versus Nitrogen Limitation in the Marine
Environment. Limnology and Oceanography, 29:1149-1160.
Stoddart, D. R. 1963. Effects of Hurricane Hattie on the British Hondu-
ras Reefs and Cays, October 30-31, 1961. Atoll Research Bulletin,

95:1-76.

Valiela, I., J. L. Bowen, and J. K. York. 2001. Mangrove Forests: One of
the World’s Threatened Major Tropical Environments. BioScience,
51:807-815.

Valiela, I., and J. M. Teal. 1979. The Nitrogen Budget of a Salt Marsh
Ecosystem. Nature (London), 280:652-656.

Vitousek, P. M. 1984. Litterfall, Nutrient Cycling, and Nutrient Limita-
tions in Tropical Forests. Ecology, 65:285-298.

Vitousek, P. M., and R. L. Sanford. 1986. Nutrient Cycling in Moist
Tropical Forests. Annual Review of Ecology and Systematics, 17:
137-167.

Zomlefer, W. B., W. S. Judd, and D. E. Giannasi. 2006. Northernmost
Limit of Rhizophora mangle (Red Mangrove; Rhizophoraceae) in
St. Johns County, Florida. Castanea, 71:239-244.

Underwater Spectral Energy Distribution
and Seagrass Depth Limits along an Optical
Water Quality Gradient

Charles L. Gallegos, W. Judson Kenworthy,
Patrick D. Biber, and Bret S. Wolfe

Charles L. Gallegos, Smithsonian Environmental
Research Center, 647 Contees Wharf Road, Edge-
water, Maryland 21037, USA. W. Judson Ken-
worthy, Center for Coastal Fisheries and Habitat
Research, NCCOS, National Oceanic and Atmo-
spheric Administration, 101 Pivers Island Road,
Beaufort, North Carolina 28516, USA. Patrick
D. Biber, Department of Coastal Sciences, The
University of Southern Mississippi, Gulf Coast Re-
search Laboratory, 703 East Beach Drive, Ocean
Springs, Mississippi 39564, USA. Bret S. Wolfe,
Department of Environmental Science, Univer-
sity of Virginia, Clark Hall, 291 McCormick
Rd., Charlottesville, Virginia 22904, USA. Corre-
sponding author: C. Gallegos (gallegosc@si.edu).
Manuscript received 23 July 2008; accepted 20
April 2009.

ABSTRACT. We measured in situ inherent optical properties and seagrass maximum
depth distribution in widely differing optical water types, including turbid green waters
of the Indian River Lagoon (IRL, Florida, USA), a mix of turbid and clear waters in Pan-
ama, and very clear waters in Belize. We used Hydrolight to model in situ spectral energy
distributions and measured leaf absorbance spectra (Thalassia testudinum) to distinguish
between photosynthetically available radiation (PAR) and photosynthetically usable ra-
diation (PUR). Attenuation coefficients for PAR and PUR were nearly indistinguishable
in Belize and Panama and differed only slightly in the IRL. Grass grew to depths of pen-
etration of 33% of PAR in the IRL, 14% in Panama, and approximately 5% in Belize,
although we expect the value for Belize is an underestimate because conditions more
turbid than are typical were prevailing at the time of the measurements. Corresponding
percentages for PUR were 27%, 12%, and 5% for IRL, Panama, and Belize, respectively.
These regional differences in light requirements were striking, and less than half of the
difference could be attributed to latitudinal variations in incident light. We conclude that
factors other than spectral energy distribution that covary with water clarity control
site-specific light requirements of seagrasses. Possibilities include epiphytes and sediment
quality.

INTRODUCTION

Seagrasses are important primary producers that play a role in the stability,
nursery function, biogeochemical cycling, and trophodynamics of many coastal
and estuarine ecosystems and as such are important for sustaining a broad spec-
trum of organisms (Hemminga and Duarte, 2000). Seagrasses are potentially
sensitive indicators of declining water quality because of their high light re-
quirements (11%-37% surface irradiance) compared to those of other aquatic
primary producers with much lower light requirements (<1%) (Dennison et
al., 1993; Zimmerman, 2003). Seagrass communities have declined in coastal
regions worldwide (Orth et al., 2006), which is usually attributed to reductions
in water clarity brought about, at least initially, by accelerated eutrophication in
the coastal zone (Krause-Jensen et al., 2008).

Management efforts aimed at preserving and restoring seagrass systems gen-
erally focus on improving water clarity (Batiuk et al., 2000; Kenworthy and
Haunert, 1991; Steward and Green, 2007), based on the high light requirements

360 °

of seagrasses and the reduction in light penetration associ-
ated with eutrophication (Ralph et al., 2007). Deciding
on the extent of water quality improvements (or limit of
allowable deterioration) requires more detailed knowl-
edge of the wavelength-specific light requirements of sea-
grasses. Based on a survey of available literature, Carter
et al. (2000) determined that mesohaline and polyhaline
submerged grass communities in Chesapeake Bay require
a long-term average of 22% of surface irradiance at the
deep edge of the grass meadow for survival. Gallegos and
Kenworthy (1996) determined a similar requirement for
mixed beds of Thalassia testudinum, Halodule wrightii,
and Syringodium filiforme in the Indian River Lagoon
(IRL) near Ft. Pierce, Florida. In contrast, Steward et al.
(2005) found 20% to be near the minimum for the IRL,
while the average light requirement was 33% of annual
incident irradiance, similar to the wide range (24%-37%)
reported for the southern Indian River Lagoon (Kenworthy
and Fonseca, 1996). More recently, Duarte et al. (2007)
analyzed 424 reports of seagrass colonization depths and
light attenuation and found generally higher light require-
ments for plant communities growing in shallow, turbid
waters than in clear, deep waters. The authors suggested
that large differences in light requirements between shal-
low- and deep-growing seagrasses may be partially attrib-
uted to differences in the quality of light. Seagrasses may
grow deeper in clear water because there is more high-
energy blue light available for photosynthesis, whereas
in shallow turbid water the shorter blue wavelengths are
rapidly attenuated.

The wavelength specificity of light absorption by
seagrasses has implications for setting water quality re-
quirements needed to protect or restore these plants in
eutrophic waters that are dominated by inefficient green
wavelengths. The absorption of light by the complement of
pigments (chlorophyll a and chlorophyll b) in seagrasses is
highly wavelength selective, with absorption peaks in the
blue (centered around 450 nm) and red (centered around
670 nm) regions of the visible spectrum, and a broad ab-
sorption minimum in the green between 500 and 600 nm
(Drake et al., 2003; Zimmerman, 2003). Wavelengths of
light that are poorly absorbed by the plant are relatively
inefficient at driving photosynthesis (Drake et al., 2003;
Falkowski and Raven, 2007).

Light requirements of seagrasses that have been de-
termined to date (Batiuk et al., 2000; Kenworthy and
Fonseca, 1996) have been based on photosynthetically
available radiation (PAR, 400-700 nm) because of the
widespread availability of underwater quantum sensors.
PAR measurements weight quanta of all visible wave-

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

lengths equally. By contrast, measurements of photosyn-
thetically usable radiation (i.e., PUR; see Morel, 1978)
weight quanta in proportion to the efficiency with which
they are absorbed. There are no sensors for direct mea-
surement of PUR; it must be calculated from the under-
water spectrum (measured or modeled) weighted by the
relative absorption spectrum of the plant of interest.

Using a bio-optical model of light penetration in the
mesohaline Chesapeake Bay, Gallegos (1994) determined
that the 22% surface PAR requirement for seagrasses oc-
curred at the same depth as the penetration of 16% of sur-
face PUR. The distinction is potentially important because
the penetration of PUR is more sensitive to the concentra-
tion of phytoplankton chlorophyll (i.e., eutrophication)
than is the penetration of PAR, for the reason that phyto-
plankton chlorophyll absorption selectively removes those
same wavelengths most efficiently used in photosynthesis
by seagrass. Thus, by basing light requirements on PUR
rather than on PAR, we would predict greater restoration
benefit from chlorophyll reduction, and greater seagrass
losses from chlorophyll increases, than by light require-
ments based on PAR (Gallegos, 1994).

The objective of this work was to determine whether
the distinction between PAR and PUR requirements
could be determined from in situ depth distributions of
seagrass communities. The distinction cannot be drawn
from depth distributions at a single site such as Chesa-
peake Bay or the IRL, because within these systems the
underwater spectrum is peaked in the green, and thus
there is insufficient spectral variability in available light
to differentiate between depth limits based on PAR com-
pared with PUR. The gradient of optical water quality
types across locations of the Smithsonian Marine Science
Network, however, offers a potentially ideal scenario
for making this determination. All three of the domi-
nant seagrass species found in the IRL also occur in the
tropical waters of Carrie Bow Cay, Belize, and Bocas del
Toro, Panama. In optically clear waters, the underwater
spectrum peaks in the blue, near an absorption peak of
chlorophyll a or b. In blue water, therefore, PUR pen-
etrates deeper than PAR, and plants should grow to rela-
tively deeper depths in blue tropical waters if PUR rather
than PAR is the determining factor. To investigate this
distinction, we surveyed seagrass distributions and mea-
sured inherent optical properties (IOPs), from which we
calculated underwater light spectra at the deep edges of
grass beds, to test the hypothesis that across the optical
water quality gradient seagrass would grow to a consis-
tent depth of penetration of PUR but a variable percent-
age of PAR.

METHODS
STUDY SITES

Station locations are shown in Figure 1. We occupied
stations in the clear tropical waters off Carrie Bow Cay,
Belize (station Blue Ground Range, BGR), and in Bahia
Almirante, Panama (station STRI [Smithsonian Tropical
Research Institute]), a station receiving colored-water dis-
charge from a nearby creek in Panama (station SNO3), and
the more eutrophic waters of the Indian River Lagoon,
Florida (ICW194; see Figure 1). Detailed characteristics
of these sites are given by Lang (2009) in the Introduction
to this volume.

OPTICAL PROPERTIES

We measured in situ profiles of IOPs, the spectral ab-
sorption and beam attenuation coefficients, at nine wave-

lengths (412, 440, 488, 510, 532, 555, 650, and 715 nm)

80°20'0"W

Kilometers

27°30'0"N

30°0'0"N

10°0'0"N

Kilometers

0 250 500 1,000 1,500

90°0'0"W

70°0'0"W

NUMBER 38 e¢ 361

using a WETLabs ac-9 instrument with a 0.1 m path-
length, equipped with a pressure sensor to measure depth.
A Seabird SBE-5T pump provided water flow to the ac-9
and a WETLabs MPAK unit that controlled pump and in-
struments and logged data.

Measured absorption and beam attenuation coef-
ficients were corrected for temperature according to the
manufacturer’s protocols. We corrected absorption coeffi-
cients for scattering errors (Kirk, 1992) by the Zaneveld et
al. (1994) algorithm that subtracts a fraction of measured
scattering coefficient from absorption (Equation 1):

t-w (A) a ain (A) ra e(cray (A) i any (A)) (1)

where d;,,() is the scattering-corrected absorption coef-
ficient less pure water absorption at wavelength i, a, is
the measured non-water absorption coefficient subject to
scattering error, c;,, 1s the measured non-water beam at-
tenuation coefficient, and « is a coefficient that accounts

88°10'0"W
Kilometers

OF aie Ain i Wl 6

0
Blue Ground Range

16°50'0"N

{Carrie Bow Cay

9°20'0"N

82°20'0"W 82°0'0"W

FIGURE 1. Locations of stations in the Indian River Lagoon, Florida (upper left), Belize (upper right), and
Panama (lower right). Lower left panel shows overview of Caribbean. Light gray shading in Belize panel

indicates coral reef habitat.

362 e

for overall errors with the reflective tube absorption meter
of the ac-9 that result from a failure to collect all scattered
light (Kirk, 1992). In this work we verified the assumption
that non-water absorption at the longest ac-9 wavelength
(715 nm) was not measurable in the laboratory (Tzortziou
et al., 2006). Thus, we calculated ¢ by Equation 2:

A a,,(715) a
e_ (5) =a, (75)

ne

We measured the absorption spectrum of Thalassia
testudinum leaves in an integrating sphere (LICOR 1800-
12S) interfaced to an Ocean Optics USB2000 spectrom-
eter. A clean segment of leaf was placed on a microscope
slide over the opening to the sphere and illuminated with
a fiberoptic microscope light source. Black tape on the
slide obscured the portion of the opening not covered
by the leaf. Percent transmittance (%T) of the leaf was
calculated referenced to the slide and tape without a leaf
in place. Absorbance was calculated as —In(%T), and
the spectrum was normalized to the value at the absorp-
tion peak at 675 nm. Measurements on eight leaves col-
lected from the deep edge at the site in Belize were aver-
aged. Similar measurements made in Panama had similar
results.

RADIATIVE TRANSFER MODELING

To calculate spectral diffuse attenuation coefficients
and underwater light spectra, we used the commercially
available radiative transfer model, Hydrolight 4.2, which
is extensively documented by Mobley (1994). User in-
put consists of specifications for IOPs, boundary condi-
tions, and assumptions on inelastic scattering processes.
We used the pure-water absorption coefficients of Pope
and Fry (1997) and pure-water scattering coefficients for
freshwater from Buiteveld et al. (1994). We used in situ
estimates of absorption, attenuation, and scattering coef-
ficients binned at 0.5 m intervals. Following Tzortziou et
al. (2006), we used the Fournier—Forand scattering-phase
function, the shape of which was shown by Mobley et al.
(2002) to be well specified by the backscattering ratio. We
omitted inelastic scattering processes because our interest
is in downwelling irradiance, and these processes primar-
ily affect only calculations of upwelling radiance. For in-
cident irradiance and the distribution of total irradiance
between direct and sky irradiance we used the built-in
RADTRAN routine for the time, location, and estimate of
approximate cloud cover.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

From the simulations of spectral downwelling ir-
radiance we calculated PAR according to its definition
(Equation 3):

700 700
PAR(z)= { O(d,z)dr= | a (3)

400 400

where O is the quantum flux, E, is the spectral down-
welling irradiance in energy units, / is Planck’s constant, \
is the wavelength and y = 27c/) is the frequency of light,
and c is the speed of light in vacuum. PUR was calculated
in an analogous manner, weighted by the plant absorption
spectrum, measured at the deep edge of the Belize site:

700
PUR(z) = | Q(A,z)aq,(d) dd (4)

400

where G7), (\) is the absorption spectrum of T: testudinum
normalized to its peak at 675 nm and to unit sum. For
comparison of attenuation rates, PAR and PUR were both
normalized to their values at the surface.

SEAGRASS SURVEYS

At each sampling site a pair of scuba divers entered
the water to visually confirm the seagrass bed (T. testudi-
num) deep edge, defined as the visible transition between
vegetated and unvegetated bottom. Once the physical
boundaries of the meadow edges were identified under-
water, the divers laid out two 10 m long transects par-
allel to the edge of the seagrass bed. At 1.0 m intervals
along each transect, the divers visually estimated seagrass
cover in a 0.25 m* quadrat using the Braun-Blanquet scale
(1965). The Braun-Blanquet cover abundance scale is a vi-
sual assessment technique for estimating the canopy cover.
Values are 0.1 = solitary shoot, with small cover; 0.5 =
few shoots, with small cover; 1 = numerous, but less than
5% cover, 2 = 5%-25% cover, 3 = 25%-50% cover, 4 =
50%-75% cover, and 5 = more than 75% cover.

At the same location each diver counted the number
of seagrass short shoots in either a 0.25 m? or 0.0625 m?
quadrat, depending on the shoot density. Short shoot
counts were multiplied by the appropriate scaling factor
and averaged for the 10 quadrats to obtain an estimate
of the number of short shoots per square meter. For com-
parison of deep edge seagrass characteristics, we also sur-
veyed relatively shallow sites at the Blue Ground Range
station in Belize (2.4 m) and the STRI station in Panama
(1.8 m). At SNO3 in Panama we only surveyed at the deep

edge. Deep edge data for the IRL are from annual sur-
veys by the South Florida Water Management District
(http://my.sfwmd.gov/gisapps/sfwmdxwebdc/dataview
.asp?query=unq_id=1797).

RESULTS
SEAGRASS DEPTH LIMITS

At the Blue Ground Range station in Belize, the deep
edge of the Thalassia testudinum meadow was located at
10-11 m. The deep edge was a distinct transition from
a sparse cover of T: testudinum to unvegetated, fine car-
bonate mud. Recently germinated seedlings of the small
opportunistic species Halophila decipiens were observed
just outside of the deep edge of the T. testudinum meadow.
Braun-Blanquet cover values ranged from 0.5 (a few in-
dividual short shoots) to 1 (<5%). Thalassia testudinum
short shoot densities ranged from 0 to 48 shoots m7’,
averaging 22.4 shoots m. At the shallow Blue Ground
Range transect, T. testudinum Braun-Blanquet scores
ranged from 3 to 4, indicating that cover generally varied
from 25% to 75%, while densities ranged from 176 to
416 shoots m7, averaging 310 shoots m~’. At the shal-
low station T: testudinum was 14 times more dense than
at the deep edge. No other seagrass species were observed
at this station.

At the STRI station in Panama we located the deep
edge of the T: testudinum at 8.5 m. The transition edge
of the T: testudinum meadow was distinct; however, there
was considerably more H. decipiens just downslope of the
edge than there was at the Blue Ground Station in Belize.
Thalassia testudinum short shoot densities ranged from
0 to 56 shoots m~’, averaging 18 shoots m~, similar to
the deep edge at the Blue Ground Range Station in Be-
lize. Braun-Blanquet values ranged from 0 to 1, indicating
that cover was generally less than 5%. We also observed
three quadrats with a relatively sparse cover of Halodule
wrightii. At the shallow STRI station (1.8 m), T. testudi-
num densities ranged from 160 to 528 shoots m7 with
an average of 465, 25 times the density at the deep edge
and more dense than the shallow station at Blue Ground
Range in Belize. Braun-Blanquet values ranged from 3 to
4, similar to the shallow station at Blue Ground Range
(BGR) in Belize.

At the SNO3 site in Panama, the deep edge of the T:
testudinum bed was located at 2.4 m. Short shoot densi-
ties ranged from 0 to 288 m *, with a mean value of 114.
The deep edge of the T. testudinum meadow was marked

NUMBER 38 © 363

by a transition from T. testudinum to unvegetated sedi-
ment. Braun-Blanquet scores ranged from 0 to 3, indicat-
ing cover values less than 50%.

Seagrass depth limits in the IRL at the site where op-
tical measurements were made in 2001 were reported as
0.92 m for beds described as continuous and dense, with a
lower limit of 50% to 60% cover.

OPTICAL PROPERTIES

A wide range of optical properties was observed
among the four sites (Figure 2a). Based on absorption
spectra, Belize had the clearest water while the most tur-
bid water occurred in the IRL. The two sites in Panama
were intermediate. The rank order of sites was different
for scattering coefficients (Figure 2b), with scattering coef-
ficients at the Panama shallow site (SN03) being the high-
est and the Panama deep site (STRI) the lowest.

ABSORPTION SPECTRUM

Normalized absorption by T. testudinum was simi-
lar to measurements by other investigators (Zimmerman,
2003), having peaks in the red wavelengths (~680 nm), a
broad maximum at blue wavelengths (400-490 nm), and
a trough at green wavelengths (~525-625 nm) (Figure 3,
solid line). This spectrum was used to calculate PUR from
simulated downwelling spectral irradiance according to
Equation 4. However, even at the local minimum at 555
nm, measured absorption was still 37% of the red peak.
On considering that T: testudinum has no chlorophyll pig-
ments that absorb green wavelengths (Zimmerman, 2003),
we also constructed a hypothetical photosynthetic action
spectrum based on chlorophyll absorption alone, consisting
of Gaussian curves with peaks at 410, 430, 455, 642, and
680 nm for an alternate calculation of PUR (see Figure 3,
dashed line). The hypothetical action spectrum is expected
to produce the maximal separation between PAR and PUR,
especially in turbid green water, because the trough in the
hypothetical chlorophyll absorption spectrum at green
wavelengths is much more pronounced compared with the
measured absorption spectrum, which includes an unquan-
tified contribution by photosynthetic carotenoids. This hy-
pothetical chlorophyll-based action spectrum serves as a
site-independent sensitivity test for the greatest possible
difference between PAR and PUR for a higher plant. We
did not measure absorption spectra in the IRL, so they are
unknown. The hypothetical spectrum allows a comparison
among sites in the absence of measurements at all sites.

364 e

a. Surface Absorption Spectra

AOA — BGR, Belize
1.8 ‘ ----STRI, Panama
i aes arc eal i saie a Musee eo SNO3, Panama

a 16 Ne) ieee opal ICW194, Florida

San Water

=

40)

oO

[=

o

je)

O

1

2

2

fo)

172)

xe}

<

Wavelength (nm)

b. Surface Scattering Spectra

-1

Scattering Coefficient (m_)

Wavelength (nm)

FIGURE 2. Surface water absorption spectra (a) and surface water
scattering spectra (b) at sites sampled in Belize, Panama, and Florida.

PAR AND PUR PROFILES

Profiles of normalized downwelling PAR and PUR
based on the measured absorption spectrum (PURm) and
PUR based on the hypothetical action spectrum (PURh)
are shown for the stations having the least and the most
separation between PAR and PUR in Figure 4. The dif-
fuse attenuation coefficients for each of the three quanti-
ties are reported for all stations in Table 1. At the Blue

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Ground Range station in Belize, diffuse attenuation coef-
ficients for PAR and PURm were indistinguishable, while
that for PURh was only 7% higher than for PAR (Table 1).
The largest differences among the three attenuation coef-
ficients occurred at the IRL. The relative differences be-
tween attenuation coefficients for PAR and PURm (13%)
and between PAR and PURh (31%) were similar for the
IRL and SNO3 site in Panama, although the absolute coef-
ficients were smaller at SNO3 (Table 1).

The percentages of surface light remaining at the deep
edges of the seagrass beds varied widely among the loca-
tions, from about 5% at the Blue Ground Range site in
Belize to about 30% at the IRL (see Table 1). The percent-
ages based on PUR were, as expected, lower than those
based on PAR, but the differences among sites was still
large (Table 1). Because of the extremely large differences
among sites in the percentage of light at the seagrass bed
deep edge, the calculation of PUR did not yield a consis-
tent value across sites. The overall range was, however,
somewhat smaller for PUR than for PAR (Table 1). Spec-
tra of downwelling irradiance at the deep edges calculated
by Hydrolight are shown in Figure 5. The overall frac-
tion of surface irradiance remaining at the deep edges at
the different locations follows the percentages in Table 1.
Qualitative differences in the spectra of light remaining at

Measured
- - - - Hypothetical

Normalized absorption
(=)
(o>)

400 450 500 550 600 650 700
Wavelength (nm)

FIGURE 3. Normalized absorption spectra used for calculating pho-
tosynthetically usable radiation, based on absorption spectrum mea-
sured on Thalassia testudinum leaves (solid line), and a hypothetical
action spectrum derived by assuming only light absorbed by chloro-
phylls a and b drive photosynthesis in Thalassia (dashed line).

the deep edges also occur. Because of absorption by water,
virtually no light is present at wavelengths greater than
600 nm at the BGR location in Belize and very little at
STRI in Panama. Increasing amounts of red wavelengths
are present at the SNO3 and IRL sites as a result of the
shallower depths of the deep edges. The peaks of the in
situ spectra shift progressively toward green wavelengths
along the progression from BGR to IRL, and the greatest
similarities are at 400 to 410 nm, where the percentage of
surface irradiance remaining ranges from 2% to 6%.

Percent surface irradiance
0 20 40 60 80 100

E
s
[ok
®
Q
144. a. Belize
16
Percent surface irradiance
0 20 40 60 80 100
E
s
[oe
®
Q

b. Florida

FIGURE 4. Vertical profiles of photosynthetically active radiation
(PAR, solid line), and photosynthetically usable radiation (PUR)
based on measured absorption spectrum (PURm; dashed line) and
hypothetical action spectrum (PURh; dotted line) in (a) Belize and
(b) the Indian River Lagoon (IRL), Florida. Profiles were normal-
ized to the irradiance incident at the surface (100%).

NUMBER 38 e¢ 365

TABLE 1. Depths of seagrass deep edge (Zax) and attenuation
coefficients for photosynthetically active radiation (Kpap) and
photosynthetically usable radiation (PUR) weighted by measured
absorption spectrum of Thalassia testudinum leaves (Kpypm) or
weighted by a hypothetical action spectrum (Kpypn; see Figure 2).
Percentage of surface light penetrating to the seagrass deep edge
is given in parentheses.

Vlawew Kpar Kpurm Kpurh

Site 2 (m) (m7) (m~') (m~')
BGR, Belize 10 0.293 0.293 0.314
(5.2%) (5.4%) (4.2%)

STRI, Panama 8.5 0.232 0.247 0.304
(13.6%) (12.0%) (7.4%)

SNO3, Panama 2.4 0.836 0.945 1.098
(14.1%) (11.0%) (7.7%)

IRL, Florida 0.92 1.157 1.301 1.52
(32.7%) (27.1%) (21.8%)

4BGR, Blue Ground Range; STRI, Smithsonian Tropical Research Institute; SN03,
Panama creek station; IRL, Indian River Lagoon.

DISCUSSION

At all three study sites we were able to locate a distinct
deep edge of the Thalassia testudinum meadows, charac-
terized by a transition from moderate and sparsely veg-
etated seagrass to either unvegetated substrate or patches
of the smaller, low light adapted seagrass Halophila de-
cipiens. Where we were able to sample shallower sites in
Belize and Panama, there were substantially higher densi-
ties of T. testudinum. The presence of H. decipiens at the
Blue Ground Range (BGR) station in Belize and the STRI
site in Panama further confirmed that we were sampling
at light-limiting edges of the T. testudinum distribution.
Halophila decipiens is a small, ruderal species of seagrass
commonly found growing in deep or turbid water and has
lower light requirements than T. testudinum (Kenwor-
thy, 2000; Gallegos and Kenworthy, 1996; Kenworthy
et al., 1989). The presence of H. decipiens at these two
stations was a good indication of light-limiting conditions
for Thalassia. Although we did not record H. decipiens at
SNO3 in Panama, a thorough visual examination by divers
at deeper depths than the observed T. testudinum distribu-
tion confirmed there were no seagrasses growing beyond
2.4 m depth.

Attenuation coefficients for PAR and PUR were nearly
indistinguishable in Belize and Panama and differed only
slightly in the IRL. Based on these one-time profiles, we
calculated that seagrass grew to depths of penetration of

366 °¢

50

NO w ‘=
oO Oo o

Percent surface PAR(Z___)
=)

Wavelength (nm)

a a a
FIGURE 5. Spectra of photosynthetically active radiation (PAR) at
the depth of the seagrass deep edge (Zax) in Belize (BGR, solid line),
Panama (STRI, dashed line, and SN03, dotted line), and Florida
(IRL [ICW194], dot-dashed line).

33% of PAR in the IRL, 14% in Panama, and approxi-
mately 5% in Belize. Corresponding percentages for PUR
were 27%, 12%, and 5% for IRL, Panama, and Belize,
respectively. The accuracy of these estimates depends on
the degree to which the profiles were measured under
conditions that are typical for their respective growing
seasons. We are fairly certain this was not the case in
Belize, where strong northerly winds, atypical for the
season, blew for several days before and on the day of
sampling. Horizontally sighted Secchi disk visibility at a
seagrass bed near Twin Cays was 5.5 m during the time
of our measurements, compared with annual means of
10.1 m (+0.38 m SE) for 2004 and 8.9 m (+0.25 m SE)
for 2005 (see Koltes and Opishinski, 2009: fig. 6, this
volume). If the water column were more strongly stirred
with higher than typical concentrations of particulate
matter, then our estimates for Belize would be biased low,
as we suspect they are. The estimated PAR light require-
ments for the IRL are, however, based on more frequent
visits and are in agreement with other published estimates
(Kenworthy and Fonseca, 1996; Steward et al., 2005). The
limitation of our approach was the inability to determine
the integral of light requirements for the whole growing
season from only a few days of measurements. Because of
this limitation, it is unlikely that the observed depth distri-
bution of the seagrasses is fully captured by PAR and PUR

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

percentages calculated, and repeating this study during an-
other season could yield different percentages.
Nonetheless, assuming that the light requirements for
seagrasses at Belize are similar to those in Panama, the
regional differences in light requirements between the IRL
and the two tropical sites remain striking. Qualitatively,
the differences are consistent with the observations of
Duarte et al. (2007) that seagrasses growing in shallow,
turbid waters (e.g., IRL) have higher light requirements
than those growing in clear, deep water (Panama, Belize).
Calculation of PUR closed the gap only slightly, leading
us to conclude that factors other than spectral energy dis-
tribution contribute substantially to site-specific light re-
quirements of seagrasses, especially at the deep edges. An
extended growing season in the more tropical locations of
Belize and Panama could possibly account for some of the
difference. The tropical sites receive about 7% more inci-
dent radiation annually than the IRL site, most of which
occurs during winter months (November through Febru-
ary) when temperatures are also more favorable in the
tropics. Other possible differences between sites include
leaf-shading epiphytes, sediment quality (e.g., grain size
or organic matter content), and possible periods of low
oxygen in thermally stratified deeper waters. These latter
factors have management implications because they are all
affected by coastal eutrophication. Improved understand-
ing of the factors accounting for site-specific differences in
seagrass light requirements is, therefore, urgently needed.

ACKNOWLEDGMENTS

We gratefully acknowledge the Smithsonian Marine Sci-
ence Network for Pilot Project funds to conduct these stud-
ies. We thank Sam Benson for diving assistance in Belize and
Amy Lewis for help with optical data reduction and analy-
sis. This is contribution no. 844 of the Caribbean Coral Reef
Ecosystem Program (CCRE), supported in part by the Hunt-
erdon Oceanographic Research Fund, and Smithsonian Ma-
rine Station at Fort Pierce contribution no. 787.

LITERATURE CITED

Batiuk, R. A., P. Bergstrom, M. Kemp, E. Koch, L. Murray, J. C. Steven-
son, R. Bartleson, V. Carter, N. B. Rybicki, J. M. Landwehr, C. L.
Gallegos, L. Karrh, M. Naylor, D. Wilcox, K. A. Moore, S. Ailstock,
and M. Teichberg. 2000. Chesapeake Bay Submerged Aquatic Veg-
etation Water Quality and Habitat-Based Requirements and Resto-
ration Targets: A Second Technical synthesis. U. S. Environmental
Protection Agency, Chesapeake Bay Program.

Braun-Blanquet, J. 1965. Plant Sociology: The Study of Plant Communi-
ties. London: Haffner.

Buiteveld, H., J. H. M. Hakvoort, and M. Donze. 1994. “The Optical
Properties of Pure Water.” In Ocean Optics XII, ed. J. F. Jaffe,
pp. 174-183. Bellingham, Wash.: Society of Photo-Optical Instru-
mentation Engineers.

Carter, V., N. B. Rybicki, J. M. Landwehr, and M. Naylor. 2000. “Light
Requirements for SAV Survival and Growth.” In Chesapeake Bay
Submerged Aquatic Vegetation Water Quality and Habitat-Based
Requirements and Restoration Targets: A Second Technical Syn-
thesis, ed. R. A. Batiuk, P. Bergstrom, W. M. Kemp, E. W. Koch,
L. Murray, J. C. Stevenson, R. Bartleson, V. Carter, N. B. Rybicki,
J. M. Landwehr, C. L. Gallegos, L. Karrh, M. Naylor, D. J. Wilcox,
K. A. Moore, S. Ailstock, and M. Teichberg, pp. 4-15. U. S. Envi-
ronmental Protection Agency, Chesapeake Bay Program.

Dennison, W. C., R. J. Orth, K. A. Moore, J. C. Stevenson, V. Carter,
S. Kollar, P. W. Bergstrom, and R. A. Batiuk. 1993. Assessing Water
Quality with Submersed Aquatic Vegetation. BioScience, 43:86-94.

Drake, L. A., F. C. Dobbs, and R. C. Zimmerman. 2003. Effects of Epi-
phyte Load on Optical Properties and Photosynthetic Potential of
the Seagrasses Thalassia testudinum Banks ex K6nig and Zostera
marina L. Limnology and Oceanography, 48:456-463.

Duarte, C. M., N. Marba, D. Krause-Jensen, and M. Sanchez-Camacho.
2007. Testing the Predictive Power of Seagrass Depth Limit Models.
Estuaries and Coasts, 30:652-656.

Falkowski, P. G., and J. A. Raven. 2007. Aquatic Photosynthesis, 2nd ed.
Princeton, N. J.: Princeton University Press.

Gallegos, C. L. 1994. Refining Habitat Requirements of Submersed
Aquatic Vegetation: Role of Optical Models. Estuaries, 17:198-219.

Gallegos, C. L., and W. J. Kenworthy. 1996. Seagrass Depth Limits in
the Indian River Lagoon (Florida, U.S.A.): Application of an Op-
tical Water Quality Model. Estuarine Coastal and Shelf Science,
42:267-288.

Hemminga, M. A., and C. M. Duarte. 2000. Seagrass Ecology. Cam-
bridge: Cambridge University Press.

Kenworthy, W. J. 2000. The Role of Sexual Reproduction in Maintain-
ing Populations of Halophila decipiens: Implications for the Biodi-
versity and Conservation of Tropical Seagrass Ecosystems. Pacific
Conservation Biology, 5:251-259.

Kenworthy, W. J., C. A. Currin, M. S. Fonseca, and G. Smith. 1989.
Production, Decomposition, and Heterotrophic Utilization of the
Seagrass, Halophila decipiens, in a Submarine Canyon. Marine
Ecology Progress Series, 51:277-290.

Kenworthy, W. J., and M. S. Fonseca. 1996. Light Requirements of Sea-
grasses Halodule wrightii and Syringodium filiforme Derived from
the Relationship Between Diffuse Light Attenuation and Maximum
Depth Distribution. Estuaries, 19:740-750.

Kenworthy, W. J., and D. E. Haunert. 1991. The Light Requirements of Sea-
grasses: Proceedings of a Workshop to Examine the Capability of Wa-
ter Quality Criteria, Standards and Monitoring Programs to Protect
Seagrasses. National Oceanic and Atmospheric Administration Tech-
nical Memorandum NMFS-SEFC-287. Washington, D.C.: NOAA.

NUMBER 38 ¢ 367

Kirk, J. T. O. 1992. Monte Carlo Modeling of the Performance of a Re-
flective Tube Absorption Meter. Applied Optics, 31:6463-6468.

Koltes, K. H., and T. B. Opishinski. 2009. “Patterns of Water Quality
and Movement in the Vicinity of Carrie Bow Cay, Belize.” In Pro-
ceedings of the Smithsonian Marine Science Symposium, ed. M. A.
Lang, I. G. Macintyre, and K. Ritzler, pp. 379-390. Smithsonian
Contributions to the Marine Sciences, No. 38. Washington, D.C.:
Smithsonian Institution Scholarly Press.

Krause-Jensen, D., S. Sagert, H. Schubert, and C. Bostrom. 2008.
Empirical Relationships Linking Distribution and Abundance
of Marine Vegetation to Eutrophication. Ecological Indicators,
8:515-529.

Lang, M. A. 2009. “Introduction to the Smithsonian Marine Science
Network.” In Proceedings of the Smithsonian Marine Science Sym-
posium, ed. M. A. Lang, I. G. Macintyre, and K. Riitzler, pp. 1-9.
Smithsonian Contributions to the Marine Sciences, No. 38. Wash-
ington, D.C.: Smithsonian Institution Scholarly Press.

Mobley, C. D. 1994. Light and Water. Radiative Transfer in Natural
Waters. San Diego: Academic Press.

Mobley, C. D., L. K. Sundman, and E. Boss. 2002. Phase Function Ef-
fects on Oceanic Light Fields. Applied Optics, 41:1035-1050.
Morel, A. 1978. Available, Usable, and Stored Radiant Energy in Rela-

tion to Marine Photosynthesis. Deep-Sea Research, 25:673-688.

Orth, R. J., T. J. B. Carruthers, W. C. Dennison, C. M. Duarte, J. W.
Fourqurean, K. I. Heck Jr, A. R. Hughes, G. A. Kendrick, W. J.
Kenworthy, S. Olyarnik, FE. T. Short, M. Waycott, and S. L. Wil-
liams. 2006. A Global Crisis for Seagrass Ecosystems. BioScience,
56:987-996.

Pope, R. M., and E. S. Fry. 1997. Absorption Spectrum (380-700) nm of
Pure Water. II. Integrating Cavity Measurements. Applied Optics,
36:8710-8723.

Ralph, P. J., M. J. Durako, S. Enriquez, C. J. Collier, and M. A. Doblin.
2007. Impact of Light Limitation on Seagrasses. Journal of Experi-
mental Marine Biology and Ecology, 350:176-193.

Steward, J.S., and W.C. Green. 2007. Setting Load Limits for Nutrients
and Suspended Solids Based upon Seagrass Depth-Limit Targets.
Estuaries and Coasts, 30:657-670.

Steward, J. S., R. W. Virnstein, L. J. Morris, and E. F. Lowe. 2005S. Setting
Seagrass Depth, Coverage, and Light Targets for the Indian River
Lagoon System, Florida. Estuaries, 28:923-935.

Tzortziou, M., J. R. Herman, C. L. Gallegos, P. J. Neale, A. Subrama-
niam, L. W. Harding Jr., and Z. Ahmad. 2006. Bio-optics of the
Chesapeake Bay from Measurements and Radiative Transfer Clo-
sure. Estuarine Coastal and Shelf Science, 68:348-362.

Zaneveld, J. R. V., J. C. Kitchen, and C. Moore. 1994. “The Scattering Er-
ror Correction of Reflecting-Tube Absorption Meters.” pp. 44-55.
In Ocean Optics XII, ed. S. Ackleson. Bellingham, Wash.: SPIE.

Zimmerman, R. C. 2003. A Biooptical Model of Irradiance Distribution
and Photosynthesis in Seagrass Canopies. Limnology and Ocean-
ography, 48:568-585.

Laat arty iy,
cannes Hie gh
yi =
Al =
k
ry
Y
ee

Interannual Variation in Gelatinous
Zooplankton and Their Prey
in the Rhode River, Maryland

Eileen §. Graham, Danielle M. Tuzzolino,
Rebecca B. Burrell, and Denise L. Breitburg

Eileen S. Graham, Danielle M. Tuzzolino, Re-
becca B. Burrell, and Denise L. Breitburg, Smith-
sonian Environmental Research Center, 647 Con-
tees Wharf Road, Edgewater, Maryland 21037,
USA. (Current addresses: E. Graham, Applied
Science Associates, Inc., 55 Village Square Drive,
Wakefield, Rhode Island 02879, USA. D. Tuzzo-
lino, University of Delaware, College of Marine
and Earth Studies, 111 Robinson Hall, Newark,
Delaware 19716, USA.) Corresponding author:
D. Breitburg (breitburgd@si.edu). Manuscript re-
ceived 29 August 2008; accepted 20 June 2009.

ABSTRACT. The lobate ctenophore Mnemiopsis leidyi is an important predator of
zooplankton and ichthyoplankton both within and outside its native range, and it is a
dominant consumer within the Chesapeake Bay food web. We sampled the Rhode River,
a subestuary of Chesapeake Bay, during 2004 and 2005 to quantify the abundances of
M. leidyi, its scyphomedusan predators, and its mesozooplankton prey, and conducted
ctenophore egg production experiments in 2004. Despite low mesozooplankton densities,
ctenophores produced up to 9,380 eggs individual”! day~!. Temporal patterns, as well as
peak abundances, of copepods, ctenophores, and sea nettles (Chrysaora quinquecirrha;
the major predator of M. leidyi) varied considerably between years. This interannual vari-
ation may have been caused by direct and indirect effects of physical factors, especially
low salinities during 2004, on all components of the food web. In 2004, zooplankton
abundances peaked in June, M. /eidyi abundances steadily increased throughout the sum-
mer, and C. quinquecirrha was rare. In contrast, during 2005, C. quinquecirrha density
increased during midsummer. As this medusa increased in abundance, M. leidyi numbers
declined and copepod abundances increased. Shallow systems with salinities near the
minimum threshold for C. quinquecirrha ephyra production may exhibit more extreme
interannual variability than deeper, higher-salinity systems and may serve as models to
provide insight into factors controlling gelatinous zooplankton dynamics.

INTRODUCTION

The lobate ctenophore Mnemiopsis leidyi is native to Atlantic and Carib-
bean estuaries and coastal waters from Massachusetts to southern Argentina
and has been introduced to several Eurasian systems including the Black, Cas-
pian, Baltic, and North Seas (Purcell et al., 2001; Kube et al., 2007). Mnemiu-
opsis leidyi can tolerate a wide range of temperatures, salinities, and dissolved
oxygen (DO) concentrations. It occurs in waters with salinities ranging from less
than 5 to more than 36 (Purcell et al., 2001; Purcell and Decker, 2005) and can
survive exposure to DO concentrations of 0.5 mg L“! for at least 4 d (Decker et
al., 2004). Optimal temperatures for M. leidyi reproduction are approximately
18°-20°C (Costello et al., 2006).

In late spring and early summer, M. leidyi can be abundant in Chesapeake
Bay and its tributaries, where it is a dominant consumer, potentially capable of
clearing much of the daily standing stock of zooplankton and ichthyoplankton

370 °

(Cowan et al., 1992; Cowan and Houde, 1993; Purcell et
al., 1994; Purcell and Decker, 2005). In mesohaline por-
tions of the Chesapeake Bay system, the major predator
of M. leidyi, the scyphomedusa Chrysaora quinquecir-
rha, usually becomes abundant in early July and persists
through the end of summer (Cargo and King, 1990). As
C. quinquecirrha population densities increase, M. leidyi
abundances typically decline and zooplankton popula-
tions rebound (Purcell and Cowan, 1995). However, in
years when C. quinquecirrha populations are low, M.
leidyi may exert much greater and prolonged control
within the food web. Chrysaora quinquecirrha polyps
are generally found in salinities of 7 to 20 and strobi-
late when temperatures exceed 17°C (Cargo and Schultz,
1967; Cargo and King, 1990). Medusae are most abun-
dant at salinities of 10-16 (using the Practical Salinity
Scale) and temperatures of 26°-30°C (Decker et al.,
2007). Thus, interannual variation in salinity and tem-
perature can strongly affect the timing and spatial distri-
bution of C. guinquecirrha and its control of M. leidyi.
The Rhode River is a small, shallow subestuary on
the western shore of Chesapeake Bay (Figure 1) charac-
terized by summer salinities that vary interannually in
both absolute maxima and timing of these maxima. Simi-
lar to other tributaries in the Chesapeake Bay system,

Upper Chesapeake Bay

es Kilometers
0510 20 30 40

area of

10.5.0 1 2 3 4

eee Kilometers

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

this estuary supports a gelatinous zooplankton food web
throughout late spring and summer months. The most
abundant gelatinous species are the zooplanktivorous
M. leidyi and its scyphomedusan predator and competi-
tor C. quinquecirrha. Average spring—summer salinity
in the Rhode River is near the lower limit required for
strobilation by C. quinquecirrha. In addition, interan-
nual variation in water temperature has the potential
to cause variation in the timing of initial and peak oc-
currences of these gelatinous species and their prey. As a
result, the Rhode River can have two distinct gelatinous
food webs: one in which the top predator (C. quinquecir-
rha) exerts control over the intermediate consumer (M.
leidyi) and one in which the intermediate consumer is not
controlled by predation.

The objectives of this study were to examine tempo-
ral and spatial patterns in abundances of M. leidyi and
C. quinquecirrha within and near the Rhode River and
to examine how those patterns varied in relationship to
water temperature, salinity, and the abundance of meso-
zooplankton prey. We also examined temporal and spatial
variation in egg production by M. leidyi. This study was
conducted during the summers of 2004 and 2005, years
with very different temporal patterns of M. leidyi and C.
quinquecirrha densities.

FIGURE 1. The Rhode River and its location in the Chesapeake Bay. Dots indicate location of sampling sites;
the SERC dock is located directly inshore (northwest) of site 4B.

METHODS

We sampled seven sites: six within the Rhode River
and one just beyond the mouth of the river in the main-
stem Chesapeake Bay. Sites were chosen based on prior
research conducted in the Rhode River and designed to
cover its entire length. At each site, weather conditions
were noted and temperature, DO, and salinity were re-
corded at the surface and subsequent 1 m depth intervals
with a YSI 600QS meter. Additional temperature, DO,
and salinity data were available from the monitoring sta-
tion located at the dock of the Smithsonian Environmental
Research Center (SERC) in the Rhode River, which was
equipped with a YSI 6600 meter (C. Gallegos, SERC, un-
published data).

Gelatinous zooplankton samples were collected in du-
plicate 3 min stepped oblique tows using a 0.5 m diameter,
202 wm mesh hoop plankton net towed at approximately
2 knots and equipped with a General Oceanics flowmeter
(model 2030). Excess water was strained from the sample,
total volume of gelatinous zooplankton was measured, and
all individuals were identified to species and enumerated.
Bell diameters of C. quinquecirrha and the oral to aboral
lengths of up to 15 M. leidyi were recorded. Remaining
specimens of M. Jeidyi were classified as either larger than
or equal to or less than 3.0 cm.

Mesozooplankton samples were collected using 0.3 m
diameter, 202 wm mesh paired hoop nets. Samples were
rinsed through a 2 mm sieve to remove gelatinous zoo-
plankton and preserved with 10% buffered formalin;
mesozooplankton species were subsequently identified
and enumerated.

Whole water column chlorophyll data were collected
by another research group (C. Gallegos, SERC) at the
four central Rhode River sites (1A, 2A, 3A, 4B; see Figure
1) on different days during each sampling week. Chloro-
phyll a (chl a) was measured with a Spectronics Genesis
5 spectrophotometer and converted into micrograms per
liter (ug L~').

Mnemiopsis leidyi egg production assays were con-
ducted in 2004 using established methodology (Kremer,
1976; Grove and Breitburg, 2005). Undamaged individu-
als covering the size range from each site (3-8 cm) were
randomly assigned to jars containing 3 L filtered Rhode
River water and left overnight at ambient water tempera-
tures. At approximately 0900 the following morning,
adult ctenophores were removed and lengths and volumes
recorded. Water from each jar was strained through a
35 \xm sieve, preserved with 10% acid Lugol’s solution
(Sullivan and Gifford, unpublished data; Grove and Bre-

NUMBER 38 e¢ 371

itburg, 2005), and eggs were enumerated. Egg production
was normalized by ctenophore volume to facilitate com-
parisons among individuals.

Data were analyzed using analysis of variance (Proc
GLM: SAS v. 9.1) on rank-transformed data. Student—
Newman-Keuls tests were used for a posteriori compari-
sons. Regression models were used to examine the effects
of ctenophore volume, site, date, and interactions between
these factors on egg production. Nonsignificant inter-
action terms with P = 0.25 were dropped from statistical
models.

RESULTS
PHYSICAL PARAMETERS

Temperature, salinity, and DO all varied among sites
and between years (Table 1; Figure 2; two-way analysis
of variance [ANOVA]). Surface water temperature varied
among sites (F = 38.21, P < 0.01), and was cooler adja-
cent to, and near the mouth of, the Rhode River and at
the deeper sites. Surface salinity also varied significantly
among sites (F = 3.55, P < 0.01), and was generally high-
est at the Bay site (Site 73) and at sites near the mouth
of the Rhode River. Minimum DO concentration varied
among sites (F = 7.33, P < 0.01) and was significantly
lower at the Bay site than elsewhere.

Measurements at the SERC dock indicated that sur-
face water temperatures reached 25°C more than 3 weeks
earlier in 2004 than in 2005 but exceeded 30°C only dur-
ing 2005. Salinity remained below 8 except for a brief
period in 2004 but exceeded 8 for most of the summer in
2005. Daytime low DO concentrations (<2 mg L~!) were
occasionally recorded in the bottom waters during cruises;
all low daytime DO measurements in 2004 and all but
one in 2005 were recorded at the Bay site. The continu-
ous YSI 6600 monitor at the SERC dock indicated that
low DO concentrations occurred near the surface within
the Rhode River in the early morning hours of both years
(C. Gallegos, SERC, personal communication, 2004).
Analysis of our weekly sampling data indicated that tem-
perature (F = 5.38, P = 0.02), salinity (F = 135.18, P <
0.01), and DO concentrations (F = 6.39, P = 0.01) were
all significantly higher in 2005 than in 2004.

2004 Biota

Chlorophyll a concentrations peaked in early June,
declined, and then rose continually during the period sam-
pled from mid-June through early September 2004 (see

372 * SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES
ee TT Te

TABLE 1. Mean environmental conditions measured at each site sampled for 2004 and 2005. See Figure 1 for site locations. Chlorophyll
a concentrations are whole-water integrated values (C. Gallegos, SERC); minimum dissolved oxygen (DO) values are based on near-

bottom measurements; temperature and salinity are from surface waters (<1 m depth); NA = site not sampled.

2004 2005
Chlorophylla Minimum DO _ Temperature Chlorophyll a Minimum DO Temperature
Site (pg L~?) (mg L~!) (°C) Salinity (ug L~) (mg L~!) (°C) Salinity
W332 NA 2.31 25.40 7.76 NA 3.46 26.26 9.79
1A 24.55 5.20 ASS) 8.20 21.37 6.61 26.64 9.26
2A 24.76 DAY 26.66 8.04 37.02 6.88 27.58 9.44
CC NA 4.70 29.14 7.35 NA 5.14 28.91 9.10
3A 32.43 4.73 27.54 7.94 28.20 5.21 28.00 8.87
SC NA 4.23 28.09 7.78 NA 5.31 MND] 9.38
4B 44.34 5.22 28.41 7.29 32.29 SolY 28.62 9.08

4 Because of sea conditions, site 73 was not sampled during mid- to late summer 2004 as frequently as other sites; thus, averages are not necessarily representative of physical

conditions at site 73 relative to other sites measured on the same dates.

Figure 2). Mesozooplankton samples in both years were
dominated (>95% of individuals) by the calanoid cope-
pod Acartia tonsa. During 2004, mesozooplankton densi-
ties varied significantly among dates (F = 6.28, P < 0.01).
Peak densities of 4-7 individuals L~! occurred on 21 June
and 7 July and then declined to approximately 1.0 indi-
viduals L! for the rest of the season (see Figure 2).

Mnemiopsis leidyi volumes also varied significantly
among dates (one-way ANOVA on ranks; F = 6.08, P <
0.01). Numerical densities and volumes were lowest in
mid-June ($0.62 + 0.25 individuals m~* and <2.3 +
0.77 mL m~’, respectively), and then gradually increased
to a maximum of 51 + 30.2 individuals m~? and 58 +
33.5 mL m™? on 19 August (see Figure 2), the date that
coincided with highest densities of “recruits” (individu-
als =1 cm in length). Regression analyses indicated a
significant relationship between the zooplankton den-
sity of the prior week and both M. leidyi volume (r2 =
0.13, P < 0.01) and the density of recruits (r7 = 0.21,
P < 0.01). However, the previous week’s chl a concen-
tration explained a greater percentage of the variation
in both these measures of M. leidyi abundance for the
sites at which chl a data were available (1A, 2A, 3A, 4B)
(volume: r? = 0.33, P < 0.01; density of new recruits:
r> = 0.25, P < 0.01). Chrysaora quinquecirrha abun-
dances were low during 2004. A few medusae were seen
in the field during August and early September but were
never caught with either the 0.5 m diameter hoop net
or the larger 1 m* neuston net, which was deployed in
an attempt to more accurately sample the low-density C.
quinquecirrha population.

2005 BIoTA

Temporal patterns and peak abundances of most bi-
ota in 2005 differed from those in 2004 (see Figure 2).
Mid-June chl a concentrations in 2005 were similar to
those in the corresponding time period in 2004, and as
in 2004 generally increased during the remainder of the
season. However, sampling did not detect an early June
chl a peak in 2005, and maximum chl a concentrations in
late summer 2005 reached only about two-thirds the con-
centrations reached in 2004 (Figure 2). Mesozooplankton
densities varied among dates (one-way ANOVA on ranks,
F = 4.87, P < 0.01). The 21 July peak density of 2.3 indi-
viduals L~! was both later and lower than peak densities
in 2004. Early June through early July mesozooplankton
densities remained below 1 individual L~! and were simi-
lar to mid-July—early September densities in 2004.

The timing of the increase in mesozooplankton den-
sities in 2005 corresponded to a decrease in M. leidyi
densities and the appearance of C. quinquecirrha. M.
leidyi densities varied significantly among dates (one-way
ANOVA on ranks: F = 13.98, P < 0.01). Peak M. leidyi

FIGURE 2. (facing page) Weekly mean temperature (°C) and salinity
at the SERC dock (C. Gallegos, unpublished data), and river-wide
mean (+SE) chlorophyll a concentration (ug L~'), mesozooplankton
abundance (number L~!), and Mnemiopsis (M.) leidyi and Chrysa-
ora (C.) quinquecirrha abundance (volume, mL m~°; density, num-
ber m~?) for 2004 (left) and 2005 (right).

373

NUMBER 38

2005

2004

jo) ite) (j=)
oO N WN

(D0) ainjyesoduiay

Ww
—

Ayuijes

Sep

Aug

Lo

COO O aOR Ont MONT Oe
no ww
3 a S
=F
co) 2.
) ®
N ”
=
at
Sts
wo <x
(3)
Wes o
op ® =
=
53 es Oo =
ac Eo ®
oo | J {2
>a Oo oO ©
>a | ® c
os ¢
Cc
EF ae >
s
o2oo0cC OOOO moan ontON ro)
Sy OM (Or OO) MOR OF KONI Nt Om on Oo oe ool e BB
ON N TK

(,-7 61) e |AydosojyD (,-1 “pul) (g- wi ou) Ajisuaq

uo}Ue|doozosa|

(¢-w -ou) Aj1suaq

pue (¢-W |W) sUIN|oA pue (¢-W jw) aWIN|OA

Apia] ‘iW eywizanbuinb ‘9

374 e

densities were higher and occurred earlier in 2005 than in
2004. Volumes peaked on 16 June (279 + 205 mL m7),
declined substantially by the 21 July sample date, and then
remained low throughout the rest of the season (Figure 2).
Medusae of Chrysaora quinquecirrha were first caught in
our sample nets on 18 July 2005, and numbers continu-
ally increased over the season, reaching a maximum on
the last sample date, 7 September (Figure 2). Mnemiopsis
leidyi densities declined as C. quinquecirrha abundances
increased. Regression analysis was run on C. quinquecir-
rha density and the prior week zooplankton density and
chl a concentrations. Partial r* values indicated that C.
quinquecirrha number explained 41% of the variation in
the number of M. /eidyi recruits whereas prior week zoo-
plankton explained only 17% and 13% of the variation in
number of recruits and M. Jeidyi volume, respectively.

MNEMIOPSIS LEIDY| EGG PRODUCTION

Egg production assays were performed on three dates
in July 2004. Muemiopsis leidyi produced between 0 and
668 eggs mL”! of ctenophore. There was a significant
positive correlation between M. leidyi volume and the
number of eggs produced, both on each date and for the
three dates combined (Figure 3). Egg production on each
date in 2004 differed significantly from all others. Egg
production was highest on 7 July, 355 + 28.2 eggs mL“!
(n = 36); lower on 1 July, 274 + 25.7 eggs mL“! (n =
33); and lowest on 22 July, 50 + 8.71 eggs mL“! (m = 35)
(see Table 2).

10000
y = 281.94x - 303.54 e

r= 0.4851 °
8000 +

6000 5

4000 5

Total Eggs Produced

2000 5

M. leidyi Volume (mL)

FIGURE 3. Total number of eggs produced and volume (mL) of
each M. leidyi in all three reproduction experiments.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

See ee aaa aes er are ae ee ee ee ee
TABLE 2. Mean egg production rate (eggs mL! ctenophore +

SE) and 2-week mean mesozooplankton density (number L™!
+ SE) for each reproduction study date.

Date Egg production Zooplankton density
1 July 274 (+ 25.7) 2.72 (+ 0.56)
7 July 355 (+ 28.2) 4.13 (+ 1.03)
22 July 50 (+ 8.7) 1.03 (+ 0.34)

Mesozooplankton prey density during the week
leading up to the reproduction assays was estimated by
averaging mesozooplankton densities measured in sam-
ples collected on the day of the egg production assay
and from the previous week to include food immediately
available as well as prey quantity potentially affecting
prior growth and reproduction. Two week average zoo-
plankton densities were 4.13 + 1.03 (m = 6), 2.72 +
0.56 (x = 7), and 1.03 + 0.34 (n = 7) individuals L~
for 7 July, 1 July, and 22 July, respectively (see Table 2).
These zooplankton densities corresponded directly with
the ranked egg production rates on these dates. ANOVA
indicated that the total number of eggs produced per in-
dividual increased significantly with ctenophore volume
(F = 201.24, P < 0.01) and average zooplankton den-
sity (F = 34.87, P < 0.01), and varied among sites (F =
5.70, 2 =< 0:01), dates (F = 1378252 10301) Nandithe
interaction between sites and dates (F = 3.39, P < 0.01);
the model r2 was 0.80 (P < 0.01).

DISCUSSION

Temporal patterns of mesozooplankton, M. leidyi,
and C. quinquecirrha in the Rhode River differed strongly
between 2004 and 2005. In 2004 mesozooplankton
abundances peaked in early summer and then declined as
ctenophores gradually increased throughout the season.
C. quinquecirrha medusae were rare, and their appear-
ance did not result in a decline in ctenophore density or
biomass. In contrast, in 2005, late spring through early
summer mesozooplankton densities were low and cteno-
phore density and biomass were high. As C. quinquecirrha
abundances increased in late summer, M. leidyi decreased
and mesozooplankton densities increased. Peak densities
of M. leidyi measured during this study in the Rhode River
(approximately 200 individuals m~? and nearly 300 mL
m °) are higher than those reported in the Pamlico River,

North Carolina (just over 60 mL m~?; Miller, 1974) or the
mid-Chesapeake Bay (Purcell et al. 2001), but similar to
abundances reported for Narragansett Bay, Rhode Island
(Deason, 1982; Sullivan et al., 2001). Peak Rhode River
densities measured in this study were lower, however, than
those reported for systems such as the Black and Caspian
Seas to which M. leidyi has been introduced (Kideys and
Romanovya, 2001; Bilio and Niermann, 2004).

Interannual variation in salinity likely contributed to
observed interannual differences in gelatinous zooplank-
ton densities and food web interactions, but the effect of
interannual variation in water temperatures is less clear.
Low salinities in 2004 likely resulted in the low densities
of C. quinquecirrha in that year. Chrysaora quinquecirrha
polyps are generally not found in salinities less than 7 and
become more abundant as salinities increase to between
7 and 10 (Cargo and King, 1990). During 2004, surface
salinity did not reach 5 until mid-June, or 7 until July, and
never reached 10. In contrast, surface salinity reached 7
by mid-June and 10 by early August in 2005. We suggest
that salinities below 5 in May and early June also delayed
or reduced early-season M. leidyi reproduction in Rhode
River in 2004 (Purcell et al., 2001). We were unable to
find published studies that report M. leidyi reproductive
rates at salinities below 5. However, if this hypothesis is
correct, there is a very narrow margin between salinities
that prevent recruitment of C. quinquecirrha and allow M.
leidyi populations to grow unchecked by predation and
salinities that hinder M. leidyi populations by limiting re-
production. The combined effects of salinity on these two
gelatinous species in Rhode River in 2004 appears to have
resulted in a persistent M. leidyi population that did not
become abundant until mid- to late July but then remained
abundant at least through early September.

Although surface waters warmed earlier in the season
during 2004 than during 2005, the effect of this warm-
ing on gelatinous zooplankton seasonal abundances is
not clear and may have been overwhelmed by other fac-
tors. Spring temperatures were 5°C higher in 2004 than
in 2005. By early May of both years, however, tempera-
tures exceeded the 9°-13°C minimum temperature re-
quired for M. leidyi reproduction (P. Kremer, University of
Connecticut, unpublished data), and by mid-May of both
years, temperatures exceeded the 17°C threshold required
for strobilation by C. quinquecirrha (Cargo and King,
1990; Purcell and Decker, 2005). In addition, there are
no data to suggest that temperatures that occurred dur-
ing the warmer 2004 spring should have reduced growth
or reproduction of either gelatinous species. By late July
2005, surface water temperatures exceeded 30°C, the tem-

NUMBER 38 ¢ 375

perature at which M. l/eidyi suffers mortality in laboratory
experiments (D. Breitburg, unpublished data). However,
M. leidyi could have avoided high midday surface tem-
peratures by moving lower in the water column, and the
appearance of predatory C. quinquecirrha is a more par-
simonious explanation as the major cause of the seasonal
ctenophore decline during 2005, given the high percentage
of M. leidyi with damage indicative of encounters with
medusae (Purcell and Cowan, 1995; Kreps et al., 1997).

With a mean depth of 2 m, the shallow bathymetry of
the Rhode River may limit the potential for coexistence of
M. leidyi and C. quinquecirrha. In the Rhode River, densi-
ties of M. leidyi averaged less than 2 mL m~? in August and
September 2005 when C. quinquecirrha densities reached
an average of 2-6 mL m °. In contrast, Keister et al.
(2000) found 26.6 mL M. leidyi m > in the Patuxent River,
Maryland, when C. quinquecirrha density averaged 11.8
mL m3. The deeper water column of the Patuxent, which
includes a bottom layer with variable and sometimes se-
verely hypoxic DO concentrations (Breitburg et al., 2003),
may provide greater opportunity for spatial separation of
M. leidyi and C. quinquecirrha and increase survival of M.
leidyi at moderate C. quinquecirrha densities.

Prey availability could limit M. leidyi abundance and
production, but our data do not suggest that low meso-
zooplankton densities were likely to have caused the large
interannual variation in ctenophore abundances. Meso-
zooplankton densities were higher in 2004 than in 2005,
and the temporal pattern of mesozooplankton and cteno-
phore abundances was more suggestive of ctenophore
control of mesozooplankton than the reverse. An inverse
relationship between copepod densities and ctenophore
abundance has been noted previously in both Chesapeake
Bay (Feigenbaum and Kelly, 1984; Purcell and Cowan,
1995) and Narragansett Bay (Sullivan et al., 2001). In both
years of our sampling, high densities of M. leidyi recruits
were found in the Rhode River during periods of lowest
mesozooplankton densities. We did not sample microzoo-
plankton, however, and cannot rule out their potential in-
fluence on ctenophore abundance.

The maximum egg production we measured in the
Rhode River (9,000 eggs individual”! M. leidyi day~')
was lower than the maximum reported value of 14,000
eggs individual"! day! (Kremer, 1976; Reeve et al.,
1989) but well within the range of values reported else-
where. Munemiopsis leidyi egg production in the Rhode
River was similar to that of field-collected ctenophores
from elsewhere in Chesapeake Bay (Purcell et al.,
2001), including the Patuxent River (D. Breitburg and
R. Burrell, unpublished data). Mnemiopsis leidyi from

376 °

the Patuxent produced a maximum of 610 eggs mL! of
ctenophores at mesozooplankton abundance of 1 indi-
vidual L~!, which is very close to the rate found in this
study of 668 eggs mL“! at 2.2 mesozooplankton individ-
uals L~!. Variation among dates in the relationship be-
tween zooplankton density and egg production suggests
an interesting pattern of trade-offs in energy allocation
to somatic growth versus reproduction, or nutritional
constraints.

Predicted changes in sea-surface temperatures and
rainfall throughout the world may lead to changes in
the geographic ranges of many aquatic organisms. The
Rhode River provides an interesting model that may aid
predictions of climate change-related shifts in ranges and
predator-prey dynamics because it is often near the thresh-
old of salinity tolerances and the dynamics of the system can
fluctuate markedly from year to year. These characteristics
of the Rhode River allowed us to examine the gelatinous
zooplankton food web within the river during two distinct
years: one with, and one without, strong influence by a top
predator. Differences in species abundances and food web
interactions observed here may help to predict dynamics in
other systems as environmental conditions, and the range
of C. quinquecirrha, change. Although generally consid-
ered a nuisance species by swimmers and fishermen, C.
quinquecirrha may benefit fisheries and habitat by control-
ling densities of M. leidyi, which is an important predator
of oyster larvae—a prey not utilized by C. quinquecirrha
(Purcell et al., 1991; Breitburg and Fulford, 2006).

ACKNOWLEDGMENTS

We thank Charles Gallegos and Sam Benson (SERC)
for the chlorophyll, temperature, and salinity data from
their cruises and monitoring station. We also thank the
Smithsonian Environmental Research Center NSF-REU
program, the Smithsonian Institution Women’s Commit-
tee, and the Smithsonian Institution Marine Science Net-
work for funding.

LITERATURE CITED

Bilio, M., and U. Niermann. 2004. Is the Comb Jelly Really to Blame for
It All? Mnemiopsis leidyi and the Ecological Concerns about the
Caspian Sea. Marine Ecology Progress Series, 269:173-183.

Breitburg, D. L., A. Adamack, S. E. Kolesar, M. B. Decker, K. A. Rose,
J. E. Purcell, J. E. Keister, and J. H. Cowan Jr. 2003. The Pattern
and Influence of Low Dissolved Oxygen in the Patuxent River, a
Seasonally Hypoxic Estuary. Estuaries, 26:280-297.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Breitburg, D. L., and R. S. Fulford. 2006. Oyster-Sea Nettle Interdepen-
dence and Altered Control within the Chesapeake Bay Ecosystem.
Estuaries and Coasts, 29:776-784.

Cargo, D. G., and D. R. King. 1990. Forecasting the Abundance of the
Sea Nettle, Chrysaora quinquecirrha, in the Chesapeake Bay. Estu-
aries, 13:486-491.

Cargo, D. G., and L. Schultz. 1967. Further Observations on the Biology
of the Sea Nettle and Jellyfishes in Chesapeake Bay. Chesapeake
Science, 8:209-220.

Costello, J. H., B. K. Sullivan, D. J. Gifford, D. Van Keuren, and L. J. Sul-
livan. 2006. Seasonal Refugia, Shoreward Thermal Amplification
and Metapopulation Dynamics of the Ctenophore. Limnology and
Oceanography, 51:1819-1831.

Cowan, J. H., Jr., R. S. Birdsong, E. D. Houde, J. S. Priest, W. C. Sharp,
and G. B. Mateja. 1992. Enclosure Experiments on Survival and
Growth of Black Drum Eggs and Larvae in Lower Chesapeake Bay.
Estuaries, 15:392-402.

Cowan, J. H., Jr, and E. D. Houde. 1993. Relative Predation Potentials
of Scyphomedusae, Ctenophores and Planktivorous Fish on Ich-
thyoplankton in Chesapeake Bay. Marine Ecology Progress Series,
95:55-65.

Deason, E. E. 1982. Mnemiopsis leidyi (Ctenophora) in Narragansett
Bay, 1975-1979: Abundance, Size Composition and Estimation of
Grazing. Estuarine Coastal and Shelf Science, 15:121-134.

Decker, M. B., D. L. Breitburg, and J. E. Purcell. 2004. Effects of
Low Dissolved Oxygen on Zooplankton Predation by the Cteno-
phore Muemiopsis leidyi. Marine Ecology Progress Series, 280:
163-172.

Decker, M. B., C. W. Brown, R. R. Hood, J. E. Purcell, T. F. Gross,
J. C. Matanoski, R. O. Bannon, and E. M. Setzler-Hamilton. 2007.
Predicting the Distribution of the Scyphomedusa Chrysaora quin-
quecirrha in Chesapeake Bay. Marine Ecology Progress Series,
329:99-113.

Feigenbaum, D., and M. Kelly. 1984. Changes in the Lower Chesapeake
Bay Food Chain in Presence of the Sea Nettle Chrysaora quinquecir-
rha (Scyphomedusa). Marine Ecology Progress Series, 19:39-41.

Grove, M., and D. L. Breitburg. 2005. Growth and Reproduction of Ge-
latinous Zooplankton Exposed to Low Dissolved Oxygen. Marine
Ecology Progress Series, 301:185-198.

Keister, J. E., E. D. Houde, and D. L. Breitburg. 2000. Effects of Bottom-
Layer Hypoxia on Abundances and Depth Distributions of Organ-
isms in Patuxent River, Chesapeake Bay. Marine Ecology Progress
Series, 205:43-59.

Kideys, A.E., and Z. Romanova. 2001. Distribution of Gelatinous Macro-
zooplankton in the Southern Black Sea During 1996-1999. Marine
Biology, 139:535-547.

Kremer, P. 1976. “Population Dynamics and Ecological Energetics of a
Pulsed Zooplankton Predator, the Ctenophore, Mnemiopsis leidyi.”
In Estuarine Processes, ed. M. L. Wiley, pp. 197-215. New York:
Academic Press.

Kreps, T. A., J. E. Purcell, and K. B. Heidelberg. 1997. Escape of the
Ctenophore Mnemiopsis leidyi from the Scyphomedusa Predator
Chrysaora quinquecirrha. Marine Biology, 128:441-446.

Kube, S., L. Postel, C. Honnef, and C. B. Augustin. 2007. Mnemiopsis
leidyi in the Baltic Sea—Distribution and Overwintering between
Autumn 2006 and Spring 2007. Aquatic Invasions, 2:137-145.

Miller, R. J. 1974. Distribution and Biomass of an Estuarine Ctenophore
Population, Mnemiopsis leidyi (A. Agassiz). Chesapeake Science,
15:1-8.

Purcell, J. E., and J. H. Cowan Jr. 1995. Predation by the Scyphomedu-
san Chrysaora quinquecirrha on Mnemiopsis leidyi Ctenophores.
Marine Ecology Progress Series, 129:63-70.

Purcell, J. E., F. P. Cresswell, D. G. Cargo, and V. S. Kennedy. 1991. Differ-
ential Ingestion and Digestion of Bivalve Larvae by the Scyphozoan

Chrysaora quinquecirrha and the Ctenophore Munemiopsis leidyi.
Biological Bulletin, 180:103-111.

Purcell, J. E., and M. B. Decker. 2005. Effects of Climate on Relative
Predation by Scyphomedusae and Ctenophores on Copepods in
Chesapeake Bay during 1987-2000. Limnology and Oceanogra-
phy, 50:376-387.

Purcell, J. E., D. A. Nemazie, S. E. Dorsey, E. D. Houde, and J. C. Gam-
ble. 1994. Predation Mortality of Bay Anchovy Anchoa mitchilli
Eggs and Larvae Due to Scyphomedusae and Ctenophores in Chesa-
peake Bay. Marine Ecology Progress Series, 114:47-58.

NUMBER 38 °¢ 377

Purcell, J. E., T. A. Shiganova, M. B. Decker, and E. D. Houde. 2001. The
Ctenophore Mnemiopsis in Native and Exotic Habitats: U.S. Estu-
aries Versus the Black Sea Basin. Hydrobiologia, 451:145-176.

Reeve, M. R., M. A. Syms, and P. Kremer. 1989. Growth Dynamics of a
Ctenophore (Mnemiopsis) in Relation to Variable Food Supply.
I. Carbon Biomass, Feeding, Egg Production, Growth and Assimila-
tion Efficiency. Journal of Plankton Research, 11:535-552.

Sullivan, B. K., D. Van Keuren, and M. Clancy. 2001. Timing and Size of
Blooms of the Ctenophore Mnemiopsis leidyi in Relation to Tem-
perature in Narragansett Bay, RI. Hydrobiologia, 451:113-120.

is

aa Trap sig

i

a

a a an abe
sean a ens

i

2 . ses 7 : : 7 S Gig i ky

iy
md

Patterns of Water Quality and Movement
in the Vicinity of Carrie Bow Cay, Belize

Karen H. Koltes and Thomas B. Opishinski

Karen H. Koltes, U.S. Department of the Interior,
Washington, D.C. USA; Thomas B. Opishinski,
Interactive Oceanographics; East Greenwich,
Rhode Island. Corresponding author: K. Koltes
(Karen_Koltes@ios.doi.gov). Manuscript received
29 August 2008; accepted 20 April 2009

ABSTRACT. Meteorological and oceanographic conditions have been monitored at
the Smithsonian Field Station at Carrie Bow Cay, Belize, since 1993 through the Carib-
bean Coastal Marine Productivity (CARICOMP) program, and since 1997 through an
automated monitoring system operated by the Caribbean Coral Reef Ecosystems Pro-
gram (CCRE). Collectively, the two datasets represent a unique resource that provides
a mechanism to improve our understanding of changing environmental conditions on
the Mesoamerican Barrier Reef and, particularly, the conditions governing water qual-
ity and movement around Carrie Bow Cay. Especially evident is the broad influence on
water quality of seasonal climate patterns as well as short-term events such as cold fronts
and major storms. Among several variables examined, wind direction appears to be a
good indicator of water quality conditions. From March to June, prevailing northeast-
erly airflow and limited rainfall result in higher water quality along this portion of the
Belize Barrier Reef. Under decreased trade or increasing westerly winds, especially during
periods of higher rainfall from October to January, turbid coastal water moves (drifts or
is pushed) out onto the reef from the lagoon. The most significant finding, however, has
been a dramatic loss of water quality along this portion of the Belize Barrier Reef since
monitoring began at Carrie Bow Cay in 1993.

INTRODUCTION

Carrie Bow Cay, Belize, has been the site of extensive biological, geological,
and ecological study as part of the Smithsonian Institution’s Caribbean Coral
Reef Ecosystem (CCRE) program (Rutzler and Macintyre, 1982). Despite more
than three decades of multidisciplinary research, however, relatively little is
known about the complex interaction of physical factors that influence the reef
environment, including winds, tides, temperature, and rainfall. These physical
parameters provide the context for understanding and predicting relationships
between reef organisms and their environment, but such parameters require ac-
curate and consistent measurement over long time periods to establish a reliable
description of baselines and trends. This is particularly true of water quality and
movement, as the constant mixing and motion of water masses, bathymetry, and
proximity to sediment inputs lead to significant spatial and temporal variability.

Establishing baselines and trends for environmental conditions has also become
increasingly important as Belize, as well as neighboring countries, experience rapid

380 «

urban and economic growth from the recent expansion
of agriculture, aquaculture, and tourism. As the coastline
and mangrove cays of Belize have experienced accelerated
development over the past few decades, the barrier reef
environment has been subjected increasingly to chronic
and acute disturbances from terrigenous inputs. Poorly
managed exploitation of coastal and offshore natural re-
sources, extensive land modifications from dredging, land
reclamation, deforestation, and conversion, and effluents
from sewage and agriculture/aquaculture are delivering
increased loads of sediments, nutrients, pesticides, herbi-
cides, and other man-made chemicals to the central lagoon
(Gibson and Carter, 2003). Eroded sediments and the resi-
dues of fertilizers and pesticides are also entering Belize’s
coastal waters from the more than 300,000 hectares of
banana, oil palm, sugar cane, citrus, and pineapple crops
cultivated across the wider Mesoamerican region (Burke
and Sugg, 2006).

Early qualitative observations suggested that a period
of heavy rainfall in the central portion of Belize was fol-
lowed within 1 to 2 days by the appearance of a plume of
low-salinity, turbid water over the fore-reef at Carrie Bow
Cay (CBC). For major storms, the lens could be significant
in duration and thickness. Rainfall from Hurricane Mitch,
a Category 5 storm, approached 2 m over Central Amer-
ica between 29 October and 1 November 1998, causing
severe flooding, landslides, and mudflows. Much of the
storm discharge entered the Gulf of Honduras, where it
flowed north as a highly turbid, plankton-enriched water
mass, reaching the Belize shelf on 3 November 1998 (An-
dréfouét et al., 2002; Sheng et al., 2007). On 15 November
1998, K. Koltes observed this surface lens to be about 15-
20 m in thickness in the waters adjacent to CBC. Further in
situ characterization of the lens was not possible because of
the closure of the field station, but recent numerical mod-
eling of satellite images of terrestrial runoff plumes in the
Gulf of Honduras confirms that influxes of sediments and
nutrients are reaching the central reefs from local and more
distant origins (Andréfouét et al., 2002; Tang et al., 2006;
Chérubin et al., 2008; Paris and Chérubin, 2008).

Consistent monitoring of environmental variables be-
gan at Carrie Bow Cay in 1993 as part of the Caribbean
Coastal Marine Productivity Program (CARICOMP).
CARICOMP is a regional scientific effort to study land-sea
interaction processes, to monitor for change on a local and
regional scale, and to provide appropriate scientific infor-
mation for management (Kjerfve et al., 1999; http://www
.ccdc.org.jm/frontpage.html). In 1997, an automated envi-
ronmental monitoring system (EMS) was installed at Car-
rie Bow Cay that provides an independent set of weather

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

and water quality measurements. It was one of the earliest
monitoring systems to process and transfer real-time data
from a remote geographic location to a website for public
access (http://nmnhmp.riocean.com). To our knowledge it
is still the only automated system continuously monitor-
ing oceanographic and meteorological conditions on the
outer Mesoamerican Barrier Reef. In December 1997, a
catastrophic fire destroyed the field station, suspending
all monitoring except the automated water temperature
measurements under the CARICOMP program. The full
complement of CARICOMP measurements resumed af-
ter the station reopened in late 1999, and the automated
measurements resumed after a new EMS was installed in
September 2000.

Together, the CARICOMP and EMS datasets have
provided a nearly continuous record of climatic and
oceanographic conditions along the central portion of the
Belize Barrier Reef. The 15 years of water temperature
and Secchi disk measurements likely now constitute the
longest continuous records for the Belize Barrier Reef and
are among the longest for the entire Mesoamerican region.
The data serve as an important resource for researchers to
examine long-term trends, episodic events, and short-term
and seasonal cycles (McField and Kramer, 2004). The data
also allow comparative studies with other reef ecosystems
to assess biodiversity and correlate environmental factors
with biological phenomena. Data have been used to char-
acterize prevailing conditions (Koltes et al., 1998) as well
as the anomalous conditions that occur during extreme
climatic events such as the El Nino-Southern Oscillation
(ENSO) of 1997-1998 (Aronson et al., 2002) and power-
ful storms such as the Category 5 hurricanes Mitch, Keith,
Iris, Dean, and Wilma.

These long-term datasets are beginning to yield reli-
able descriptions of meteorological and oceanographic
conditions around Carrie Bow Cay. We report here on
preliminary analyses of the patterns, trends, and relation-
ships that are emerging from these long-term records, with
special reference to factors that control water quality and
movement around Carrie Bow Cay.

METHODS

Carrie Bow Cay is a small island (0.7 acres) located
in the central province of the Belize Barrier Reef (Burke,
1982) about 18 km from the mainland (16°48’N and
88°05'W; Figure 1). Carrie Bow Cay lies on the barrier
reef proper between two tidal passes, a relatively rare oc-
currence along the otherwise nearly continuous barrier

Legend
| Reef Crest

tas os
ey i
A

b

v
Ls
3

i

*

FIGURE 1. Map of the central province of Belize showing the lo-
cation of the Smithsonian Institution’s Field Station (16°48’N and
88°05'W) at Carrie Bow Cay. Note the Lagoon Channel that sep-
arates the outer lagoon platform from the mainland. Inset. upper
right: Location of the CARICOMP permanent monitoring sites in
the seagrass beds near Twin Cays (A = “lagoon”) and Carrie Bow
Cay (B), on the inner fore-reef slope (D) and seaward of the bar-
rier reef (E = “drop-off”), and (C) the Environmental Monitoring
System on Carrie Bow Cay. Inset lower left: The Belize National
Meteorological Service stations at Melinda Forest (MF), Maya King
(MK), Savannah Forest (SF), and Punta Gorda (PG).

reef. It is in close proximity to deep ocean water (>300 m)
to the east. A line of cays to the west that includes Twin
Cays is part of a fault-block ridge that separates the shal-
lower back-reef lagoon from the deeper Lagoon Channel
that parallels Belize’s shoreline. Although shoal formations
are the outcome of a variety of factors such as currents and
sea level changes, the present-day patterns of parallel shoals,
reefs, and mangrove islands in this area resulted principally
from faulting along a NNE trend during the Pliocene era
(Dillon and Vedder, 1973; Macintyre and Aronson, 1997).

CARICOMP. MEASUREMENTS

Scientific monitoring of meteorological and oceano-
graphic conditions under the CARICOMP program has
included daily measurement of precipitation (mm) and air

NUMBER 38 e¢ 381

temperature (°C) at Carrie Bow Cay (“C” in Figure 1) and
weekly measurement of surface water temperature (°C) and
salinity (%o) at a permanent monitoring station in the seagrass
beds adjacent to Twin Cays (“lagoon”; “A” in Figure 1; wa-
ter depth ~1.2 m) and in the ocean seaward of the drop-off
(“drop-off”; “E” in Figure 1; water depth >300 m). Bot-
tom water temperatures have been recorded continuously
at intervals of 15-48 min at the permanent CARICOMP
monitoring sites in the lagoon (“A” in Figure 1), in the sea-
grass beds adjacent to Carrie Bow Cay (“B” in Figure 1;
water depth ~2m), and on the inner fore-reef (“D” in Figure
1; water depth ~13.5m) using Onset Corporation’s model
HOBO, StowAway, and TidbiT data loggers (+0.2°C).

Water quality characteristics that are associated with
water transparency have been measured by Secchi disk as
horizontal distance (m) taken 0.5 m below the surface in
the lagoon (“A” in Figure 1) and vertical distance (m) at the
drop-off (“E” in Figure 1). From 1993 to 1997, water trans-
parency was measured once a week between 1000 and 1200
using a 30 cm diameter disk with black and white quad-
rants. In 1999, the CARICOMP protocol was modified to
take advantage of the 20 cm diameter black and white disk
that is more available commercially. The difference in the
diameters of the two Secchi disks has little effect on the mea-
surements, particularly compared to other sources of error
such as sun angle, cloud cover, sea state, wind, currents, and
observer difference (Steel and Neuhausser, 2002; Hou et al.,
2007). In 2002, the Secchi disk measurements were increased
to two times per week to more accurately characterize water
quality trends. No water transparency measurements were
made during adverse weather conditions, particularly over
the drop-off, during closure of the station in the fall of 1993
and 1994, and for approximately two years following de-
struction of the field station from the fire in 1997.

To better characterize water quality, especially dur-
ing storm events, light intensity loggers were mounted on
a cinder block (water depth ~13.5 m) at the permanent
CARICOMP monitoring site on the inner fore-reef (note
“TD” in Figure 1). Onset Corp.’s model StowAway LI was
used from 2002 until 2005 when Onset ceased manufac-
turing this model; Onset’s model HOBO UA-002-64 Pen-
dant Temp/Light has been used since 2005. Light intensity
(lumens/ft?) was recorded at intervals from 5 s to 15 min
over periods ranging from days to weeks from 2002 to
2008. While deployed, the light logger was kept free of
sediment and epibionts by periodically wiping the surface
of the housing. The StowAway LI was designed to mea-
sure relative light levels (e.g., sun versus shade) and was
calibrated for incandescent sources (spectral response,
~200-1,100 nm; range, ~0.001-1,000 lumens/ft*). The

382 e

HOBO UA-002-64 Pendant Temp/Light was designed to
measure relative light levels indoors or outdoors (spectral
response, ~200-1,200 nm; range, ~0-30,000 lumens/ft”).
Our objective was to establish patterns in relative light lev-
els, and hence water transparency, by comparing in situ
irradiance on the inner fore-reef (“reef irradiance”) to that
at the surface (“incident irradiance”). No attempt was
made to relate these measurements to biologically active
wavelengths.

ENVIRONMENTAL MONITORING SYSTEM

The EMS continuously records meteorological and
oceanographic conditions at Carrie Bow Cay. A marine-
grade wind monitor (RM Young model 05106), LI-COR
model LI-200 pyranometer, temperature/relative humid-
ity sensor (Vaisala model HMPS0O), and barometric pres-
sure sensors monitor meteorological conditions. The
weather sensors, mounted on an aluminum tower above
the main laboratory, are approximately 13 m above
ground level (“C” in Figure 1). Rain is measured with
a Texas Electronics solid-state tipping bucket rain gauge
(model 525USW). The pyranometer features a silicon
photovoltaic detector designed to measure solar radiation
under conditions of unobstructed natural daylight. Mea-
surements of wind speed (mph) and direction (0°-360°),
solar radiation (W/m7), rain accumulation (mm) and rate
(cm/h), barometric pressure (mbar), air temperature (°C),
and relative humidity (%) are recorded every 10 min.

Oceanographic conditions are monitored via a YSI
model 6600EDS multiparameter water quality sonde,
mounted inside a stilling tube. The sonde is mounted on
the dock on the west side of Carrie Bow Cay about 0.6 m
below the surface of the water (“C” in Figure 1). Every
10 min, measurements are taken of water level (m), water
temperature (°C), salinity (%o), dissolved oxygen (% satu-
ration), pH, and turbidity (NTU).

Data acquisition of both oceanographic and meteo-
rological systems is managed automatically by a data-
logger and control system and transmitted by radio to
a server on the mainland. A regular program of mainte-
nance, including calibration of sensors to manufactur-
er’s standards (Eaton et al., 2005) and minimization of
fouling, is followed to maintain the best possible data
quality. Data are subjected to a quality assurance/quality
control (QA/QC) process to remove outliers and invalid
and suspect data before they are included in the histori-
cal archives. The QA/QC process also incorporates es-
tablished criteria and procedures to correct for sensor
and fouling drift (Wagner et al., 2006).

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

The data have contributed to new and existing re-
search studies, publications, and management programs
for both Smithsonian researchers and an increasing num-
ber of organizations from the region (e.g., Renken and
Mumby, 2009).

ANALYSIS

The annual means and standard deviation of Secchi
disk measurements from 1993 to 2008 for the lagoon (” =
787) and drop-off (1 = 727) were calculated, and the 15-
year trend in water quality was determined by least-squares
regression analysis. To evaluate seasonal patterns in water
quality, monthly means were calculated for the Secchi mea-
surements from the lagoon because horizontal measurements
are generally more reliable than vertical measurements (Steel
and Neuhausser, 2002) and were more often taken during
adverse weather conditions.

To examine annual variation in wind patterns, monthly
wind rose plots were generated for Carrie Bow Cay using
approximately 350,000 wind measurements collected be-
tween 2003 and 2008. A wind rose plot is a combination
of a polar plot and a histogram that depicts the distribu-
tion of wind speed and the frequency of occurrences that
wind blows from 1 of 16 cardinal directions (N, NNE,
NE, etc.). Seasonal and geographic patterns of rainfall
were derived by calculating monthly mean precipitation
for Carrie Bow Cay and four coastal stations (Melinda
Forest, Mayaking, Savannah Forest, and Punta Gorda; see
Figure 1, lower inset) maintained by the Belize National
Meteorological Service (2003-2008).

Previous oceanographic studies at Carrie Bow Cay
suggested that tides play a major role in controlling water
movement, and hence water quality, around Carrie Bow
Cay (Greer and Kjerfve, 1982; Kjerfve et al., 1982). To
examine the role of tides, reef irradiance was compared to
meteorological and oceanographic variables using a har-
monic regression analysis model based on linear regres-
sion algorithms to compute constituents. Specifically, we
compared reef irradiance to incident irradiance and then
wind direction and intensity, tidal stage, and precipitation
from all stations. The objective of the analyses was to ex-
tract harmonic constituents to judge the relative strength
of the tidal forcing on the light patterns. The analyses were
inconclusive because tidal components were small relative
to the strong solar forcing inherent to the measurements of
light and from the lack of a strong diurnal or semidiurnal
signal in the “microtides” of Belize (Kjerfve et al., 1982).

To establish annual changes in light, monthly means
were computed for reef and incident irradiance from 2003

to 2008. The analyses included more than a half million
reef irradiance samples and approximately 350,000 mea-
surements of incident irradiance. The monthly means were
normalized with respect to the maximum monthly mean
observed in each set of light measurements; this established
a common datum reference and allowed comparisons of
monthly and seasonal variations and patterns. Additional
qualitative analyses of the various time series suggested
that a consistent correlation existed between patterns of
reef irradiance and both seasonal and short-term changes
in wind direction. A comparative analysis of conditions
during periods of normal weather conditions and specific
weather “events” was undertaken to further define the lo-
calized influence of wind direction and other parameters
on water quality over the fore-reef and to identify external
(regional) events that are observed in the Carrie Bow Cay
measurements.

To establish annual changes in light, monthly means
were computed for reef (7 ~ 500,000) and incident ir-
radiance (1 ~ 350,000) from 2003 to 2008. The monthly
means were normalized with respect to the maximum
monthly mean computed for each set of light measure-
ments; this established the maximum mean for each set
as a common reference point and allowed comparisons of
monthly and seasonal variations and patterns. Additional
qualitative analyses of the various time series suggested
that a consistent correlation existed between patterns of
reef irradiance and both seasonal and short-term changes
in wind direction. A comparative analysis of conditions
during periods of normal weather conditions and specific
weather “events” was undertaken to further define the lo-
calized influence of wind direction and other parameters
on water quality over the fore-reef and to identify exter-
nal (regional) events that are observed in the Carrie Bow
measurements.

RESULTS

Significant temporal and spatial variability exists in
the records of the meteorological and oceanographic vari-
ables. Particularly evident are longer-term seasonal pat-
terns as well as signatures of short-term events such as
cold fronts and major storms. The most significant find-
ing, however, has been a dramatic decline in water clarity
along this portion of the Belize Barrier Reef since monitor-
ing began at Carrie Bow Cay in 1993 (Figure 2). Mean
annual Secchi distance (horizontal) declined from 12.8 m
in the lagoon in 1993 to 8.7 m by 2008, a loss of almost
0.3 m/year. During the same period, mean annual Secchi

NUMBER 38 °¢ 383

16 Lagoon

y = -0.2898x + 590,25
R? =0.8172

a
ay

ra
N

10 }

Horizontal Distance (m)

4
1992 1994 1996 1998 2000 2002 2004 2006 2008 2010

35

Drop-off

30 y = -0.512x + 1043.9
R? = 0.6358

25

20

Vertical Distance (m)

10 a 1

1992 1994 1996 1998 2000 2002 2004 2006 2008 2010

SSS a IEE POS ET WE EE
FIGURE 2. Annual mean and standard deviation of Secchi disk
distance (horizontal, m) in the lagoon (top) and over the drop-off
seaward of Carrie Bow Cay (vertical, m; bottom) from 1993 to
2008. There has been a dramatic and significant (P < 0.001) loss
of transparency in the waters around Carrie Bow Cay during the
past 15 years.

distance (vertical) declined from 23.8 m to 15.6 m over the
drop-off or by about 0.5 m/year.

Comparisons of monthly means of water transpar-
ency, wind direction, and precipitation show that seasonal
weather patterns strongly influence water quality (Figure 3).
From February through May, the prevailing northeasterly
airflow (Figure 3, left) is associated with a uniformly dry
pattern across all stations (Figure 3, right). Monthly rainfall
amounts average less than 75 mm. By June, a divergence can
be seen among the stations, with those to the south receiving
increasingly greater amounts of rain relative to the stations
to the north, including Carrie Bow Cay. The sharp onset of
the rainy season in the south is partly the result of the intru-
sion of the Inter-Tropical Convergence Zone as it migrates
northward (http://hydromet.govy.bz/Climate_Summary.htm).

SSW

February

December Npeatucedrerctsssineneawetnectvanasea

1000
C— Carrie Bow Cay Rainfall
900 ----| [== Melinda Forest Rainfall
> Maya King Rainfall
800 |- --| @ZZaSavannah Forest Rainfall
Wg Punta Gorda Rainfall
700 ......| —fLagoon - Secchi Distance
“o™
€
€ 600
= a
Oo
= coon a a Oo = y
& 400 }------ Z
® Y)
= O Z
300 ==> Z
AG
1G
a Z
200 . = Z
Z
Y
% Z
100 ; y | j j
Z A Z
7 YA Z Y
» LEELA eval | I) Z NEL?
= = § a 2
oO 9 pes Ss 3
3 3 0 =
= & Ss
us @
my ir

NUMBER 38 ¢ 385

15

11

10

(1) 99UNBSIG IyDDeg Je]UOZOY UPS)!

SJ

SS SSS AMA
RS MQAAAQAY

AQ a oOii inodnaon

SMA MOON

SAH

[eocsomeees]
December ——

ay & Sea
August fereceseracmnecene =
Sep tember ——— ay

October
November

FIGURE 3. Left (facing page), Monthly means of wind speed and direction at Carrie Bow Cay for 2003-2008.
The wind rose depicts the distribution of wind speed and direction: the length of each “spoke” indicates the
percentage of time that winds blow from 1 of 16 cardinal directions (N, NNE, NE, etc.); the categories within
the spoke represent the speed. Trade winds are the dominant pattern from March to September. Wind pat-
terns are most variable during the transitional months of October/January/February. Right (above), Monthly
means of rainfall at Carrie Bow Cay and the four Belize Nationai Meteorological Service stations (left axis),
and water transparency in the lagoon (horizontal Secchi disk distance [m], right axis) from 2003 to 2008.

Rainfall maxima also diverge in terms of timing.
Mean monthly rainfall peaks in June-July in southern
Belize, with approximately 900 mm at Punta Gorda; the
maximum at Carrie Bow Cay (about 290 mm) occurs in
October. The marked shift in rainfall patterns beginning
in June is accompanied by consistent northeasterly winds
and tropical waves moving westward from June to No-
vember (http://hydromet.gov.bz/Climate_Summary.htm).
Cold fronts, or “northers,” occur frequently from De-
cember to February and are associated with the southerly
extension of the North American high-pressure system.
During the peak season of December and January, cold
fronts pass through Belize approximately every 10 days,
the signatures of which appear in the temperature profiles
on the fore-reef (Figure 4).

Monthly Secchi disk measurements in the lagoon (see
Figure 3, right) correlate with the seasonal shift in climate
patterns. Water transparency peaks in August (12.6 m),
while incident irradiance is high (Figure 5) and the winds
are still predominantly from the NE (see Figure 3, left). This
peak also corresponds to a break in the rainy season on the

mainland known as the “Mauga” (http://hydromet.gov.bz/
Climate_Summary.htm). In contrast, the break in the rainy
season occurs in July at Carrie Bow Cay (see Figure 3,
right). Water transparency reaches a minimum in January
(8.4 m) around the winter solstice and the period when
cold fronts, characterized by strong NW winds, reach a
peak (Figure 3, left). This period of increased storm ac-
tivity and resulting high sea states also interrupts routine
Secchi disk measurements such that the winter means may
be biased upward.

The seasonal pattern of reef irradiance is similar to
that observed for water transparency measured by Secchi
disk (see Figure 5) and is largely governed by changes in
incident irradiance. Maximum irradiance occurs around
the solstice, with reef irradiance showing a peak in June
compared to August for Secchi disk distance (see Figure 2)
and July for incident irradiance. However, reef irradiance
attenuates more rapidly toward the winter solstice rela-
tive to incident irradiance. By December, reef irradiance
has fallen to less than 40% of its summer peak whereas
incident irradiance has fallen only to 60% of its summer

386 * SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

32

— Inner forereef at 13.5 m water depth

31

30

29

28

27

26

Water temperature (°C)

25

24 n — 1 n 4 a
Jul 2004 Oct 2004 Jan2005 Apr2005 Jul 2005 Oct 2005 Jan2006 Apr2006 Jul 2006 Oct 2006

FIGURE 4. Water temperature on the inner fore-reef at 13.5 m from 29 July 2004 to 11 December 2006.
Note the signatures of cold fronts in the temperature profile between October and January and a late season
cold front that passed 1-3 April 2005.

i122
[ @ Incident (surface) Irradiance
O Reef (inner forereef; 13.5 m) Irradiance
i1,(0) |r
0.8 F -

Normalized Irradiance

March
April
June

August

0.6

0.4

0.2

0.0 : :
iS re a =>
s 5 = =
5 | 6g
ay LL

FIGURE 5. Monthly mean of reef irradiance (inner fore-reef, 13.5 m) and incident (surface) irradiance

September
October
November
December

(2003-2008), normalized to allow comparisons among the light recording instruments. Reef irradiance at-
tenuates at an accelerated rate relative to incident irradiance approaching the winter solstice.

peak. Decreased light levels around the winter solstice
may be caused by lower sun angle, increased turbidity
and/or higher sea states associated with the winter climate
pattern.

Flow of turbid water into the area can also be seen
in finer-scale comparisons of reef and incident irradiance
(Figure 6) that suggest turbidity is transported from more
distant locations. Following two days of light winds that
had shifted from NNE first to the south and eventually to
the NW, turbid water flowed on to the fore-reef, driving
light levels down by about 50% despite nearly full incident
irradiance. At least some of the flow may be related to
tides (Figure 7), but as previously mentioned, the mixed,
semidiurnal microtide of this region has made it difficult
to analyze the role of tides.

O Carrie Bow Cay Rainfall
© Melinda Forest Rainfall
@ Mayaking Rainfall
Savanna Forest Rainfall
@ Punta Gorda Rainfall

Daily Rainfall (mm)

Incident Irradiance (W/sq m)

§-Jul-05

6-Jul-05 7-Jul-05 8-Jul-05

9-Jul-05

NUMBER 38 ¢ 387

DISCUSSION

The constant mixing and motion of water masses, lo-
calized inputs of sediments, and changing meteorological
conditions produce strong spatial and temporal variations
in conditions along the Belize Barrier Reef. These factors
present a challenging analysis, especially when combined
with Carrie Bow Cay’s complex geomorphology that in-
cludes cuts in the reef to the north and south where dis-
parate bodies of water mix together. Nevertheless, initial
analyses have shown significant correlations between
certain weather conditions, both episodic and seasonal,
and water properties. Particularly evident are longer-term
seasonal patterns, as well as signatures of discrete events
such as the passage of cold fronts and other changes from

300
—— Incident Surface Irradiance

Reef (inner forereef; 13.5 m) Irradiance

a
Oo
(y bsyq) soueipesy jaay

10-Jul-05 11-Jul-05 12-Jul-05 13-Jul-05

FIGURE 6. Progression of a turbidity event: Early in July 2005, light levels on the reef were among the highest recorded for reef irradiance (bot-
tom, right axis). Beginning on 7 July, winds that had been blowing from the northeast at 15 mph began to taper off and remained calm through
9 July, as shown in the daily wind rose (top; see Figure 3 for description of wind rose). On 10 July, the winds increased at Carrie Bow Cay and
heavy rainfall occurred on the mainland, particularly to the south (middle; stations as in Figure 3, right). Incident irradiance (lumens[L]/ft?; bot-
tom, left axis) remained high, but light levels on the reef dropped by nearly 50%. The drop in reef irradiance is attributed to a sediment plume,
possibly from the south, that drifted over the fore-reef under conditions of little or no wind. Winds shifted around again to the northeast by
11 July, and subsequent mixing of water returned light levels over the reef to near maximum by 13 July.

388 °

Tidal Elevation (m)

—— Hourly Means
— Weekly Means
O Monthly Means

32

“-

O

i=)

—

@ 30

=

=]

~

o

o

Qa 28

£

o

ke

oD H

co

= ;
24 fF —— Hourly Means aan j

— Weekly Means '
O Monthly Means '

22
Apr 2003 Oct 2003 Apr 2004

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Oct 2005

Oct 2004 Apr 2005

FIGURE 7. Hourly, weekly, and monthly means of tidal elevation (m) and water temperature (°C) measured at the dock at Carrie Bow Cay.
Tidal elevation (top) reveals a complex pattern with a seasonal increase in elevation in October, when water temperature (bottom) reaches a

maximum, that is typical of the Caribbean.

the typical weather patterns. Strong correlations are also
apparent during less prominent variations in weather and
particularly changes in wind direction such as those ob-
served in July 2005. We found several other instances in
which light levels on the reef dropped significantly when
winds simply tapered off completely during the 24 to
48 hour “calm” period that often preceded a cold front.
We also document a substantial decline of water quality
around Carrie Bow Cay over the past 15 years.

Patterns of water transparency are driven largely by in-
cident solar radiation, reflecting the strong seasonal cycle of
solar radiation. However, variations in those patterns are
useful at tracking short- and long-term changes in water
transparency. The pattern of decreased light intensity on the

fore-reef as a function of incident solar radiation in Decem-
ber versus June likely reflects the higher rainfall, consistent
northwesterly winds, and frequent “northers” of the win-
ter months and is consistent with recent simulation models
showing locally higher turbidity during the winter months
(Paris and Chérubin, 2008) from increased precipitation.
This seasonal cycle is also apparent in the record of Secchi
disk measurements. The difference observed between the
timing of the maximum quality measured by Secchi disk and
those measured by light logger may reflect the differences in
the sampling period (1993-1998 versus 2002-2008). The
disparity may also reflect the fact that the light logger data
have not been collected as often during the spring-summer
months as they have been during fall-winter months.

Although additional modeling and analysis are re-
quired, particularly of tidal currents, preliminary results
suggest a strong link between climate patterns and water
quality and movement along the barrier reef in the cen-
tral province of Belize. We propose that, under the typical
pattern of prevailing trade winds, sediment-laden riverine
input is pushed shoreward and held along the coast where
it flows south in the Lagoon Channel. A southerly flow
of water immediately adjacent to the shoreline is consis-
tent with circulation models that have been developed for
the Gulf of Honduras (Ezer et al., 2005). We also suggest
that the fault-block ridge along the eastern boundary of
the Lagoon Channel forms a natural “sill” or “dam,” fa-
cilitating the segregation of turbid coastal water driven
shoreward by wind forcing. Under the influence of “oce-
anic” water, turbidity over the fore-reef at Carrie Bow
Cay is generally lower.

Localized turbidity events are triggered when the north-
easterly flow of the trade winds changes in either speed or
direction. Turbid water that is normally contained along
the coast spills over the submarine ridge of the Lagoon
Channel and drifts or is pushed out across the outer (back-
reef) lagoon and onto the fore-reef. Changes in the prevail-
ing weather patterns are also frequently accompanied by
periods of rain on the mainland that discharge additional
sediment and freshwater into the lagoon, increasing the
volume and degree of turbid coastal water and spillover
onto the reef platform.

The dramatic loss of water quality at Carrie Bow Cay
over the past 15 years also suggests longer-term effects of
increasing sediment and nutrient loads to the lagoon from
the rapid modifications of the Belizean coastline and that
of neighboring countries. Recent hydrological models of
the Mesoamerican Barrier Reef estimated that runoff and
associated river discharge have doubled and sediment de-
livery has increased 20 fold under present-day land use
changes compared to a hypothetical “natural” (unaltered)
state (Burke and Sugg, 2006). Although the modeling sug-
gested that Belize contributed only about 10%-15% of the
sediment load to the region, the Belize River was identified
as a significant contributor of sediments and nutrients to
the Mesoamerican Barrier Reef. In the central and south-
ern portions of Belize, large amounts of fertilizers used to
cultivate citrus and bananas and the direct discharges of
domestic sewage produce high nutrient levels in several ar-
eas along the coast (Gibson and Carter, 2003). Shrimp and
other aquaculture operations are also discharging effluents
directly to the lagoon. These sediments appear to become
entrained in the Gulf of Honduras gyre and, over the long
term, are driving down water quality across the region.

NUMBER 38 ¢ 389

Increasing turbidity also appears to be related to a
wider regional increase in sedimentation and nutrient en-
richment in the Gulf of Honduras (Burke and Sugg, 2006).
Hydrological models indicate that sediment delivery in-
creases southward along the coastline of Central America
with Honduras contributing an estimated 80% of the
sediment and half of the nutrients to the region. Circula-
tion models of the Gulf of Honduras suggest that these
sediments and nutrients are carried north along the Meso-
american Barrier Reef by the Caribbean Current (Thattai
et al., 2003). Recent modeling of satellite images indicate
that runoff from watersheds in northern Honduras can
extend as far north as Glovers Reef atoll (Andréfouét,
2002; Paris and Chérubin, 2008). Based on modeling of
satellite images and the hydrological models of Burke and
Sugg (2006), Chérubin et al. (2008) concluded that con-
centrations of buoyant matter from terrestrial runoff into
the Gulf of Honduras were high from October to January.
Plumes were transported by a cyclonic gyre toward the
Yucatan, creating seasonal variation in the concentration
of runoff loads along the Mesoamerican Barrier Reef. The
influence of terrestrial runoff was maximal from October
to January and minimal from March to April.

Long-term in situ measurements of the sort presented
here are relatively rare. Although the results of our analy-
ses are preliminary, they already demonstrate the value of
these measurements for advancing our understanding of
the range and complexity of interactions of natural and
human-induced variables governing the conditions around
Carrie Bow Cay and across the region. These in situ data
are also critical to ground-truthing remotely sensed data
such as the satellite-generated sea-surface temperature
(SST) records used to calculate bleaching thresholds dur-
ing the 1998 ENSO (Aronson et al., 2002). Finally, our
data are beginning to yield reliable descriptions of water
quality conditions in the central portion of the Belize Bar-
rier Reef, including those conditions that have accompa-
nied the rapid modifications of the Belizean coastal zone
during the past few decades. Most significantly, the dra-
matic loss of water quality documented by these long-term
records has significant biological, management, and eco-
nomic implications for Belize and the other countries of
the Mesoamerican Barrier Reef.

ACKNOWLEDGMENTS

This research was funded by grants from the Smithso-
nian Institution’s Caribbean Coral Reef Ecosystems Pro-
gram. We thank Dr. Klaus Ruetzler for his support and
encouragement of the monitoring programs and Mike

390 «

Carpenter for exceptional support to the field effort. We
thank the many station managers who have assisted with
data collection, especially noting the dedicated efforts of
Dan Miller and Claudette DeCourley over many years.
Dan has been largely responsible for maintaining the light
meters on the reef. Dr. Randi Rotjan also generously as-
sisted with the light loggers. We thank John Tschirky who
has been part of the CARICOMP program since its incep-
tion. We thank two reviewers for helpful criticisms that
improved the manuscript. This is contribution number
846 of the Caribbean Coral Reef Ecosystems Program
(CCRE), Smithsonian Institution, supported in part by the
Hunterdon Oceanographic Research Fund.

LITERATURE CITED

Andréfouét, S., P. J. Mumby, M. McField, C. Hu, and F. E. Muller-Karger.
2002. Revisiting Coral Reef Connectivity. Coral Reefs, 21:43-48.

Aronson, R. B, W. F. Precht, M. A. Toscano, and K. H. Koltes. 2002. The
1998 Bleaching Event and Its Aftermath on a Coral Reef in Belize.
Marine Biology, 144:435-447.

Burke, L., and Z. Sugg. 2006. Hydrologic Modeling of Watersheds
Discharging Adjacent to the Mesoamerican Reef: Analysis Sum-
mary. World Resources Institute http://www.wri.org/publication/
watershed-analysis-mesoamerican-reef (accessed 25 May 2009).

Burke, R. B. 1982. “Reconnaissance Study of the Geomorphology and
Benthic Communities of the Outer Barrier Reef Platform, Belize.”
In The Atlantic Barrier Reef Ecosystem at Carrie Bow Cay, Be-
lize, ed. K. Riitzler and I. G. Macintyre, pp. 509-526. Smithsonian
Contributions to the Marine Sciences, No. 12. Washington, D.C.:
Smithsonian Institution Press.

Chérubin, L. M., C. P. Kuchinke, and C. B. Paris. 2008. Ocean Circula-
tion and Terrestrial Runoff Dynamics in the Mesoamerican Region
from Spectral Optimization of SeaWiFS Data and a High Resolu-
tion Simulation. Coral Reefs, 27(3):503-519.

Dillon, W. P., and J. G. Vedder. 1973. Structure and Development of the
Continental Margin of British Honduras. Bulletin of the Geological
Society of America, 84:2713-2732.

Eaton, A. D., L. S. Leonore, E. W. Rice, and A. E. Greenberg, eds.
2005. Standard Methods for the Examination of Water and
Wastewater. 21st ed. Washington, D.C.: American Public Health
Association, Water Environment Federation, and American Wa-
ter Works Association.

Ezer, T., D. V. Thattai, B. Kjerfve, and W. D. Heyman. 2005. On the Vari-
ability of the Flow Along the Meso-American Barrier Reef System:
A Numerical Model Study of the Influence of the Caribbean Cur-
rent and Eddies. Ocean Dynamics, 55:458-475.

Gibson, J., and J. Carter. 2003. “The Reefs of Belize.” In Latin Ameri-
can Coral Reefs, ed. J. Cortes, pp. 171-202. Amsterdam: Elsevier
Press.

Greer, J. E., and B. Kjerfve. 1982. “Water Currents Adjacent to Carrie
Bow Cay, Belize.” In The Atlantic Barrier Reef Ecosystem at Carrie
Bow Cay, Belize, ed. K. Ritzler and J. G. Macintyre, pp. 53-58.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Smithsonian Contributions to the Marine Sciences, No. 12. Wash-
ington, D.C.: Smithsonian Institution Press.

Hou, W., Z. Lee, and A. D. Weidemann. 2007. Why Does the Secchi Disk
Disappear? An Imaging Perspective. Optics Express, 15:2791-2802.

Kjerfve, B., J. Ogden, J. Garzon-Ferreira, E. Jordan-Dahlgren, K. De
Meyer, P. Penchaszadeh, W. Wiebe, J. Woodley, and J. Zieman.
1999. “CARICOMP: A Caribbean Network of Marine Laborato-
ries, Parks, and Reserves for Coastal Monitoring and Scientific Col-
laboration.” In CARICOMP: Caribbean Coral Reef, Seagrass and
Mangrove Sites, ed. B. Kjerfve, pp. 1-16. Coastal Region and Small
Island Papers 3. Paris: UNESCO.

Kjerfve, B., K. Riitzler, and G. H. Kierspe. 1982. “Tides at Carrie Bow
Cay, Belize.” In The Atlantic Barrier Reef Ecosystem at Carrie
Bow Cay, Belize, ed. K. Riitzler and I. G. Macintyre. Smithsonian
Contributions to the Marine Sciences, 12:47-51.

Koltes, K. H., J. J. Tschirky, and I. C. Feller. 1998. “Carrie Bow Cay,
Belize.” In CARICOMP: Caribbean Coral Reef, Seagrass and Man-
grove Sites, ed. B. Kjerfve, pp. 74-99. Coastal Region and Small
Island Papers 3. Paris: UNESCO.

Macintyre, I. G., and R. B. Aronson. 1997. “Field Guidebook to the
Reefs of Belize.” In Proceedings of the Eighth International Coral
Reef Symposium, Volume 1, ed. H. A. Lessios and I. G. Macintyre,
pp. 203-222. Balboa: Smithsonian Tropical Research Institute.

McField, M., and P. Kramer. 2004. Healthy Mesoamerican Reef Initiative.
http://healthyreefs.org/Book_launch/Healthy_Reef_Engl_Final.pdf
(accessed 25 May 2009.)

Paris, C. B., and L. M. Chérubin. 2008. River-Reef Connectivity in the
Meso-American Region. Coral Reefs, 27(4):773-781 (doi:10.1007/
s00338-008-0396-1).

Renken, H., and P. J. Mumby. 2009. Modelling the Dynamics of Coral
Reef Macroalgae Using a Bayesian Belief Network Approach. Eco-
logical Modelling, 220:1305-1314.

Riitzler, K., and I. G. Macintyre, eds. 1982. The Atlantic Barrier Reef
Ecosystem at Carrie Bow Cay, Belize. I. Structure and Communt-
ties. Smithsonian Contributions to the Marine Sciences, No. 12.
Washington, D.C.: Smithsonian Institution Press.

Sheng, J., L. Wang, S. Andréfouét, C. Hu, B. G. Hatcher, F E. Muller-
Karger, B. Kjerfve, W. D. Heyman, and B. Yang. 2007. Upper Ocean
Response of the Mesoamerican Barrier Reef System to Hurricane
Mitch and Coastal Freshwater Inputs: A Study Using Sea-Viewing
Wide Field-of-View Sensor (SeaWiFS) Ocean Color Data and a
Nested-Grid Ocean Circulation Model. Journal of Geophysical Re-
search, 112, C07016, doi:10.1029/2006JC003900.

Steel, E. A., and S. Neuhausser. 2002. Comparison of Methods for Mea-
suring Visual Water Clarity. Journal of the North American Ben-
thological Society, 21(2):326-335.

Tang, L., J. Sheng, B. G. Hatcher, and P. F. Sale. 2006. Numerical Study
of Circulation. Dispersion and Hydrodynamic Connectivity of Sur-
face Waters on the Belize Shelf. Journal of Geophysical Research,
111, C01003, doi:10.1029/2005JC002930.

Thattai, D., B. Kjerfve, and W. D. Heyman. 2003. Hydrometeorology
and Variability of Water Discharge and Sediment Load in the Inner
Gulf of Honduras, Western Caribbean. Journal of Hydrometeorol-
ogy, 4:985-995.

Wagner, R. J., R.W. Boulger Jr., C. J. Oblinger, and B. A. Smith. 2006.
Guidelines and Standard Procedures for Continuous Water-Quality
Monitors: Station Operation, Record Computation, and Data Re-
porting: U.S. Geological Survey Techniques and Methods 1—D3,
version 1.0, 51 p. + 8 attachments. Posted April 2006, at http://
pubs.water.usgs.gov/tm1d3.

Global Change and Marsh Elevation

Dynamics: Experimenting Where Land
Meets Sea and Biology Meets Geology

J. Adam Langley, Marc V. Sigrist, James Duls,
Donald R. Cahoon, James C. Lynch,
and J. Patrick Megonigal

J. Adam Langley, Marc V. Sigrist, James Duls, and
J. Patrick Megonigal, Smithsonian Environmen-
tal Research Center, 647 Contees Wharf Road,
Edgewater, Maryland 21037, USA. Donald R.
Cahoon and James C. Lynch, U.S. Geological
Survey, Patuxent Wildlife Research Center, 10300
Baltimore Avenue, BARC-East Bldg. 308, Belts-
ville, Maryland 20705, USA. Corresponding au-
thor: J. A. Langley (langleya@si.edu). Manuscript
received 13 May 2008; accepted 20 April 2009.

ABSTRACT. Coastal marshes must accumulate soil to keep up with rising sea levels. It is
unknown how the response of these ecosystems to global change will influence their abil-
ity to continue to keep up with sea-level rise. Here, we describe an in situ experimental
chamber approach for manipulating key environmental variables, such as atmospheric
CO, and soil N availability, in a brackish marsh. We outfitted each chamber with sur-
face elevation tables (SETs) to closely monitor soil elevation change, a sensitive indicator
of marsh vulnerability to sea-level rise. Further, the design facilitates measurements of
ecosystem exchange of CO3, plant productivity, porewater chemistry, and other environ-
mental parameters.

INTRODUCTION

Projecting the impacts of climate change, eutrophication, and other pertur-
bations on ecosystems requires experimental manipulations. Large experimental
facilities have been built and operated in all types of ecosystems over the past
decades to provide such data. There are at least six characteristics that compli-
cate experimental manipulations in tidal wetlands. First, such ecosystems can
be quickly and irreversibly damaged by heavy foot traffic, so boardwalks must
be built to minimize long-term impacts on vegetation and soils. Second, because
many wetlands have deep, low-density, peaty soils, the permanent infrastruc-
ture, such as boardwalks and chambers, must be well anchored for stability.
Third, tidal wetlands are often inundated by tides and can be under more than
a meter of water during storm surges, which dictates that all buoyant equip-
ment must be soundly fixed in place. All electrical service and sensitive equipment
must be positioned high and be easy to remove during extreme flooding events.
Further, emergency shutoff systems must be in place to cut off the electrical
supply and gas exchange equipment during flood events. Fourth, the water that
floods brackish marshes is saline and corrodes most metals. Fifth, the lack of
shade means that UV-sensitive materials will degrade. Care must be taken to
select UV-resistant materials, and even those must be monitored and frequently
replaced. Sixth, high-latitude marshes may experience cold winters. Ice forma-
tion can severely damage even rigid and well-anchored infrastructure. Here we

392 °e

describe a global change experiment in a brackish marsh
that was designed to overcome these substantial technical
challenges.

SITE DESCRIPTION

This study took place at Kirkpatrick Marsh, which is
located on the Rhode River, a subestuary of Chesapeake
Bay at the Smithsonian Environmental Research Center
in Edgewater, Maryland. The site is dominated by the C;
sedge, Schoenoplectus americanus (formerly Scirpus ol-
neyi), and less so by two Cy grasses, Spartina patens and
Distichlis spicata. The soils at this site are organic (80%
organic matter) to a depth of approximately 5 m. Mean
tidal range is 30 cm. The high marsh zone is 40-60 cm
above mean low water level and is inundated by 2% of
high tides. Salinity averages 10 parts per thousand (ppt)
and ranges from 4 to 15 ppt seasonally. Average daily low
air temperature is —4°C in January, and the average daily
high is 31°C in July.

To examine the interactive effects of elevated CO, and
nitrogen addition, we identified 20 plots of similar plant
composition in summer 2005. Each plot consisted of one
octagon (2 m across) that would be enclosed in an experi-
mental chamber to allow for atmosphere manipulation
and an adjacent rectangular portion (2 X 1 m) that served
as a reference plot to account for spatial variation and to
gauge potential chamber effects.

CONSTRUCTION
WALKWAYS AND EQUIPMENT HOUSING

A main boardwalk and a series of thinner, lighter “cat-
walks” were built to access each plot without continually
walking on the marsh (Figure 1; see also Figures 4, 5). The
main boardwalk, built perpendicular to shore, roughly bi-
sected the experimental plots. Most of the horizontal sur-
faces of the boardwalks were fiberglass grating (50% open),
which allowed light to penetrate through the boardwalks,
sustaining plant life and providing excellent traction. The
supports for the main boardwalk were built of 10 x 10
cm posts sunk 2 m into the ground. The catwalks departed
from the main boardwalk, forming a perimeter around each
experimental chamber (Figure 2). These catwalks were less
than 30 cm above the ground to avoid shading the plots.
They were built of fiberglass grating (30 cm wide planks)
laid flat on supports built of 2.5 cm polyvinyl chloride
(PVC) that were anchored more than 1 m into the marsh

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 1. An overhead image of the entire CO) site. The main
boardwalk connects to 20 experimental plots by smaller catwalks,
the paths of which are visible. Gas samples are pumped via the pump
houses to the analytical shed on the bank.

using segments of 2.5 cm PVC pipe. After all walks were in
place, the marsh surface was rarely stepped upon directly.

Three small pump houses, which housed air sample
pumps and remote datalogging equipment, were built
alongside the main boardwalk. An analytical shed was
constructed on the bank 5 m above sea level to house sen-
sitive analytical equipment.

OPpeN-TOP CHAMBERS

The chamber design followed that of a previous open-
top chamber study in the same marsh (Drake et al., 1989),
but with several major innovations to enhance durability
and plot accessibility (Figures 2-5). In 2006, the cham-
bers consisted of four major components: base, manifold,
chamber skeleton, and chamber panels. The octagonal
shape of the chamber was a compromise between two de-
sign goals. It approximated a cylinder, which was ideal

for uniform air mixing inside the chamber and minimiz-
ing dead spots. The flat surface of each side allowed us
to enclose the chamber with eight flat panels that can be
removed easily for access to the inside of the chamber.

The base of the chamber was an aluminum octagon
(0.5 cm thick, with an L-shaped cross section) implanted
10 cm into the marsh surface. In the portion of the base
that was implanted into the soil, 2 cm holes were cut to
allow root growth to further stabilize the base. A hollow
octagonal manifold (cross section, 30 cm high X 6.35 cm
wide) was attached to the base to distribute inflowing air
equally around the chamber (see Figure 2). Manifolds were
built from welded aluminum (grade 6061-T5) covered with
transparent acrylic panels that allowed light transmittance.

Mounted to the top of the manifold was the “skel-
eton,” consisting of eight vertical legs supporting an oc-
tagonal ring oriented horizontally at the top. The skeleton
was built from 2.5 cm diameter PVC pipe. The only cus-
tom pieces in the skeleton were three-way fittings on the
octagonal ring that join two PVC pipes in the ring with
one leg. The joint was made by tapping female thread into
the side of a 45° elbow. The legs of the skeleton sat in
welded supports on the top of the manifold.

Solenoid Valve Bank 1 t

\ | Solenoid Valve Bank2 |_

Environmental

Blower }
sensors intake aA

Removable
side panels

ae pjoyiueyy
Tre Jequeyg

delivery

Manifold

FIGURE 2. Schematic of an experimental chamber and gas sam-
pling system. The ambient CO, chambers are the same except there
is no CO, delivered into the blower stream. Solid (black) arrows
represent air flow; dotted arrows (appearing light gray) represent
information flow.

NUMBER 38 ¢ 393

-Unchambered

Experimental chamber
eference plot

Shallow
Benchmarks

SET Rod

Peat
Mineral Clay

|
Total
Elevation |
|

FIGURE 3. Schematic of the surface elevation table (SET) design.
The SET arm is periodically connected to the SET rod benchmark,
which has been anchored into the mineral clay underlying the peat
profile. Pins are placed through holes in the SET arm to the soil sur-

face to measure changes in elevation occurring over the entire pro-
file. Any change in elevation of the shallow benchmarks must occur
as a result of processes occurring beneath the root zone. Root zone
changes are calculated by subtracting deep zone elevation change
from total elevation change.

Removable rectangular panels were made from alu-
minum (grade 6063-T52), covered with infrared (IR)-
transparent film (Aclar 22A, Honeywell) that was taped
on using transparent UV-resistant tape (3M, 851). The
film does not absorb IR radiation as do other films and
therefore heat is not allowed to accumulate. The panels
were attached to the PVC frame with custom-fitted PVC
snaps so that any of the eight panels could be removed
to access any portion of the plot. Further, panels were re-
moved in the winter to prevent damage while CO, fumiga-
tion was terminated.

Finally, after the 2006 growing season, to conserve CO)
and achieve a more stable CO) concentration by reducing
wind incursions, an octagonal frustum, or wind foil, was
added to the chamber design (see Figure 5). The frustum
was constructed of 1.9 cm PVC, angled inward at 45°, and
covered with the same film to reduce wind incursions. The
final dimensions of the chamber were 1.5 m in height, 2 m
in diameter, and with 1 m sides; the volume is 6.5 m°.

FIGURE 4. Photograph of SET measurements being made in sum-
mer 2006. Each pin is gently lowered to the soil surface, while apply-
ing minimal pressure so that the pin does not depress the soil surface.
(Photograph by M. V. Sigrist.)

To move ambient air through the open-top chamber,
a blower (Dayton, 5CO95) was mounted on a stand 1.5
m above the marsh surface to avoid high tides. The blow-
ers were placed at least 6 m away from each chamber and
oriented to avoid shading the study area (see Figures 1,
2, 4). PVC chimneys (two 15 cm pipes per blower) were
affixed vertically to the top of the blower intake so that the
blowers would take in air from 4 m above the ground that
was not influenced by biological activity on the ground
and thus had relatively stable [CO ]. The chimneys were
capped to prevent rainwater from entering the blower. A
20 cm diameter duct fed air from the blower to the cham-
ber manifold. Two hundred fifty-two 1 cm diameter holes
(the same total area as the intact pipes) were drilled on the
inside of the manifold, so that air would flow in the cham-
bers equally from each side of the octagon. The blowers
forced 12.5 m? per minute through the chambers, resulting
in an approximate chamber air turnover rate of 2 min=!.

SURFACE ELEVATION TABLES

To take repeatable measurements of soil elevation,
each plot was outfitted with a rod surface elevation table
(SET; see Figures 3, 4) (Cahoon et al., 2002) modified to
accommodate plot dimensions. Outside each experimental
chamber, a posthole was dug roughly 15 cm in diameter
and 20 cm deep. A 30 cm long PVC pipe (15 cm diameter)
was placed vertically into the hole. In the center of the

PVC pipe, a series of attachable stainless steel rods was
driven with an electric hammer through the entire profile
of organic matter (4-5 m depth) and anchored to the point
of refusal (6-7 m) into the subsurface mineral clay under-
lying the marsh. Concrete was poured into the PVC pipe
to secure the top of the SET rod.

To isolate the influence of root zone processes on el-
evation, we implanted “shallow benchmarks” to a depth
of 30 cm. The vertical movement of these benchmarks
results from processes that occur below the top 30 cm
of soil. The benchmarks were made of aluminum pipe
(5 cm diameter by 40 cm long). Several 1 cm diameter
holes were drilled into the sides of the lower 10 cm of
the pipe to allow roots to grow through and anchor the
benchmarks in place. Six benchmarks were implanted to
a depth of 30 cm under the path of the SET arm in each
chamber, three inside the chamber and three outside. After

FIGURE 5. Photograph showing a chamber with a frustum that was
added to all chambers before the 2007 growing season. The tubing
leading to one set of porewater wells, the gas sampling tube, and the
catwalk is also visible. (Photograph courtesy J. A. Langley.)

placement, solid caps were placed on the top of each pipe.
All these perturbations, as well as boardwalks to service
each plot, were completed in the summer of 2005, at least
9 months before the beginning of the experiment.

At intervals ranging from 1 to 3 months, the modi-
fied horizontal aluminum SET arm (4 m long compared
to less than 0.5 m long for the original rod SET design)
was attached to the top of the SET rod benchmark, leveled
precisely, and affixed to an aluminum post at the other
end. The arm provided a horizontal reference of known
elevation across the soil surface; changes in the distance
from this reference surface to the soil surface were a sensi-
tive measure of changes in soil elevation. Fiberglass pins
(3 mm in diameter), all exactly 91.0 cm in length, were
placed through precision-drilled holes in the SET arm at
1 cm increments. Approximately 40 individual measure-
ments were made in each chamber and 40 in each adja-
cent, unchambered reference plot. Each pin was carefully
lowered to the soil surface and gently placed so that no
litter or live plant obstructed the pin. The height from the
SET arm to the top of each pin was measured to the near-
est millimeter (mm), providing a measurement of total
elevation. Changes in absolute soil elevation were parti-
tioned into either the root zone (top 30 cm of soil) or the
deep zone (below 30 cm). To measure elevation changes
occurring in the deep zone, we lowered 2 to 4 pins to the
surface of each of the six shallow benchmarks (three in-
side and three outside each chamber) and measured in the
same manner. We calculated the change in elevation result-
ing from processes occurring in the root zone, AE, from
the two measured variables following the equation AER
= AE; — AEp, where AE; represented the change in total
elevation and AEp represented elevation change attribut-
able to change in thickness of the deep zone. Surface ac-
cretion was also measured using feldspar marker horizons
in each plot (Cahoon et al., 1995). To eliminate compac-
tion during coring, the deposition rate was estimated by
taking cryocores (Cahoon et al., 1996) and measuring the
amount of soil deposited on top of the marker horizon.

Total soil elevation was strongly related to innate spa-
tial and temporal variability of deep zone dynamics. Spe-
cifically, changes in the thickness of the deep zone followed
mean monthly sea level through time, and distance from the
bank predicted the amplitude of that oscillation. To isolate
treatment effects on various soil elevation parameters, we
accounted for variation by referencing SET measurements
in experimental chambers to those in the adjacent, uncham-
bered reference plots, so that relativized AE = experimental
plot AE — reference plot AE, where E = the elevation pa-
rameter of interest. We used a repeated-measures multivari-

NUMBER 38 ¢ 395

ate analysis of variance (MANOVA) to test for changes
in elevation through time; we used ¢ tests to liberally de-
tect chamber effects on elevation parameters at individual
dates and a two-way analysis of variance (ANOVA) to test
for treatment differences in surface accretion.

TREATMENT APPLICATION
CO, DELIVERY AND SAMPLING SYSTEM

Carbon dioxide was delivered to each of the 10 elevated
CO, chambers at a rate of approximately 6 L min™! to
achieve a target concentration of 720 ppm, which is nearly
double the current ambient concentration of 380 ppm. Each
CO) delivery line was controlled with metered valves and
fed into the intake chimney on the blower for each re-
spective elevated chamber. Adding the CO, upstream of
the blower ensured sufficient mixing before air entered the
chamber through the manifold.

Two sample lines continuously pumped air from each
of 20 chambers to instruments located in a nearby shed:
one line sampled manifold air and the other sampled the
chamber atmosphere. To achieve a representative sample
of the chamber atmosphere, air was sampled with a 2 m
long pipe oriented horizontally across each chamber. The
pipe was 1.3 cm diameter PVC with caps on both ends and
a series of 2 mm diameter holes at geometrically increas-
ing intervals away from the center of the pipe. The geom-
etry allowed air drawn from the center of the pipe to be
a composite sample representing each point on a transect
through the chamber equally. The sampling pipe was posi-
tioned horizontally and adjusted to roughly half the green
canopy height to best represent the air that photosynthetic
tissue experienced.

Air was pulled under negative pressure from each
chamber a short distance to a Teflon-coated double dia-
phragm pump (Thomas Industries, 2107-CA14-TFE),
from which it was pushed under positive pressure to the
analytical shed (see Figure 1). To avoid drawing water into
the pumps, they were plugged into normally closed float
switches (Dayton 3BY75) that cut the power supply when
the water level approached the height of the gas sampling
lines. Each of the 40 lines entered a bank of solenoid
valves (model 3V1, Sizto Tech Corporation), then flowed
into a common line, one for manifold lines and one for
chamber lines. The two solenoid valves controlling each
chamber opened simultaneously, one with manifold air
and the other with chamber air; the other solenoid valves
remained closed so that the contents of only one cham-
ber at a time passed through the common lines to the gas

396 °

analyzers. Each chamber was sampled for 2 min to allow
ample time for air in the common portion of the system
to be flushed out before measurements were logged, which
meant that each chamber was sampled at least once every
40 min.

One infrared gas analyzer (IRGA) measured the dif-
ference between a chamber’s manifold air (1.e., incoming
air) and a dry, zero-CO) reference gas. A second IRGA
measured the difference between manifold air and cham-
ber air. This configuration maximized our ability to pre-
cisely measure absolute CO, concentration and to accu-
rately measure the CO) concentration difference between
two locations. A LI-6262 (Licor, Lincoln, NE) had dry,
zero-CQ) air cycling through the reference cell and the
manifold air passing through the sample cell. A Li-7000
(Licor) had the manifold air passing through the reference
cell and chamber air passing through the sample cell. Cell
A of the LI-7000 was referenced to an analog signal from
the LI-6262 as the absolute concentration of CO, and
H,O in the manifold air.

We monitored the manifold line to determine how
much CQO, was being delivered to each manifold. The
chamber air sampling line allowed us to monitor the ac-
tual chamber atmosphere and to fine tune the CO) deliv-
ery rate to achieve our target concentration in the cham-
ber atmosphere, accounting for photosynthetic drawdown
and wind incursions.

NITROGEN FERTILIZATION

A total of 25 g N year ! was applied to each high-N
plot. Ammonium chloride was dissolved in 5 L brackish
water from the nearby Rhode River, the subestuary ad-
jacent to the site. At five dates (approximately monthly,
avoiding high tides) throughout the growing season we
used backpack sprayers to deliver the fertilizer (equivalent
to 5 g N) solution to 10 plots. Then, the fertilizer solution
was rinsed from standing vegetation with another 5 L un-
amended river water applied with backpack sprayers. Each
fertilization treatment simulated 5 g N m ? in the equiva-
lent of 0.5 cm river water. The 10 unfertilized chambers
received 10 L unamended river water applied in the same
manner. The river water was taken from the tidal fetch
area adjacent to the marsh. Mean annual [NH,4*] in that
water ranges from 32 to 82 pg L~!, with a mean of 52,
and salinity has ranged from 4.0 to 10.6 ppt, with a mean
of 6.7, over the past 20 years (growing season means from
biweekly sampling; Thomas Jordan, unpublished data).
Assuming the added NHC] integrated into the top 40 cm
of porewater (as our sampling indicates), and excluding

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

losses from the ecosystem or plant uptake, we estimated
that this fertilization would have increased porewater sa-
linity by a maximum of 0.05 ppt, less than 1% of normal
salinity.

MEASUREMENTS

The chambered experimental plots consisted of two
halves, one-half geological and one-half biogeochemical.
All sampling that involved disturbance of soil was per-
formed on the biogeochemical half. All elevation measure-
ments, which were considered to be more sensitive to soil
disturbance, were performed on the geological half.

ABOVEGROUND BIOMASS

We estimated peak aboveground biomass with a com-
bination of allometry and harvested subplots (Erickson et
al., 2007). At the end of July of each year, eight 30 X 30
cm quadrats were placed in prescribed locations in each
plot, six inside the chamber and two in an adjacent un-
chambered control plot. In the quadrats, each Schoeno-
plectus americanus stem was counted and nondestruc-
tively measured for total height, green height, and width
at half-height. In the corner of each quadrat, we clipped
and removed all vegetation and litter in a 5 X 5 cm area.
Vegetation was sorted according to species. We measured
the clipped S. americanus stems for total height and width.
Clippings were dried for 72 h at 60°C and weighed. We
measured length and width on a subset of freshly clipped
stems. We used the calculated relationship between linear
dimensions and dry mass (r* > 0.9) to estimate the mass
of each live S. americanus stem. To estimate Spartina pat-
ens and Distichlis spicata mass, we scaled up from mass in
the clipped areas to total chamber area.

ROOT PRODUCTIVITY

Three soil cores (30 cm depth X 5 cm diameter) were
taken from each plot and replaced with cylindrical in-
growth bags (30 cm height X 5 cm diameter). The bags
were constructed from 1 cm mesh and filled with milled,
moistened peat so as to achieve the bulk density of in situ
peat, 0.12 g cm 3. Bags were implanted in winter and re-
moved in November the following year. Contents were
washed over a 1 mm sieve. Large organic fragments were
picked out by hand. Root mass was separated into fine
(<2 mm diameter) and coarse (>2 mm) categories, dried
for 72 h at 60°C, and weighed.

POREWATER WELLS

We implanted nine porewater wells (three replicates
at each of three depths: 20, 40 and 80 cm) in each ex-
perimental plot. We built wells from 0.6 cm internal di-
ameter rigid Teflon tubing (GE Polymershapes) plugged at
the bottom with silicon caulk and open at the top, which
extended 10 cm above ground. Sixteen holes (1 mm di-
ameter) were drilled into the bottom 10 cm of the Teflon
tube to allow ample conductance of porewater into the
well. A vinyl hose (6 mm [OD], 3 mm inner diameter [ID])
was fastened to the top of each well and draped over the
chamber for easy access from outside the chamber. Wells
were flushed with 60 mL, equivalent to more than total
well volume, and sampled monthly for a suite of chemical
parameters using syringes.

Net Ecosystem EXCHANGE (NEE)

The chambers were also designed to allow for measure-
ment of net ecosystem exchange (NEE) of CO, between the
atmosphere and the enclosed ecosystem (Figure 6). Periodi-

—_ =
ey ~~ oS

'
N
o

as
>

'
Wa
So

——amb -&- ambN
—@— elev -@ elevN

NEE (;zmol COz m* s”)

-70

NUMBER 38 °¢ 397

cally throughout the growing season, octagonal caps were
placed on a subset of chambers. The purpose of the caps
was not to render the chamber airtight, but to eliminate
wind incursions and generate a consistent, predictable
pattern of air flow through the chambers. The caps were
octagons with a crossbeam built from 1.9 cm PVC pipe
covered with the same IR-transparent film for the chamber
panels. The film was perforated with 2 cm diameter holes.
The gas sampling pipe, described above, was raised to a
height roughly 30 cm below the cap and aligned with the
cap perforations. This arrangement allowed us to measure
the [CO,] of air exiting each chamber after it had been
influenced by soil and vegetation.

To estimate flow rate of air through the chamber, we
cut a slit in the air delivery ducts from each blower and
measured air velocity using a handheld anemometer (AM
4822, Mastech; www.p-mastech.com). We initially mea-
sured the velocity at a range of distances from the duct
wall and determined that the mean of two measurements
(centered at 4 cm and 9 cm from the duct wall) adequately
estimated the average velocity for the entire cross section.
Multiplying velocity (cms!) and cross-sectional area (cm*)

H20/07 F20/07 F2U07? PW2Uu07 P2207 P2207 F23/07 F23/07 F24/07

FIGURE 6. Net ecosystem exchange (NEE) of CO) over three days in July 2007. Negative values represent
net uptake by the ecosystem. Generally, values are negative in summer daytimes when photosynthetic rate
surpasses respiration rate. Each point represents the means of approximately 12 individual measurements
from each of two replicate chambers binned into 2.4-h intervals. amb = ambient; elev = elevated; ambN =

ambient N; elevN = elevated N.

398 °

yielded volumetric flow rate (cm? s~'). The volumetric flow
rate was converted to mass flow using air temperatures
from the site. We calculated NEE as ([COz]in — [CO>]our)
x flow rate.

Because we did not want to incur chamber effects such
as warming or rain exclusion, we measured NEE on a rotat-
ing subset of chambers balanced by treatment, for variable
intervals from 3 to 7 days. These data will be used to calcu-
late NEE light-response curves for net CO, uptake during
the day and NEE temperature-response curves for net CO)
release during the night. The response curve models will be
driven with continuous measurements of soil temperature
and photosynthetically active radiation to extrapolate up to
integrated NEE for a complete growing season (Rasse et al.,
2003). The gas sampling program was adjusted to increase
the frequency with which NEE chambers were sampled to
increase resolution for these low signal-to-noise NEE mea-
surements, compared to the relatively stable absolute atmo-
spheric [CO] data when all chambers are sampled equally.

ENVIRONMENTAL VARIABLES

Soil temperature was measured at 5 and 15 cm depth
using type-I thermocouples. Wind speed was monitored
with an anemometer (O14A-L, Campbell Scientific, Logan,
UT). Water level was recorded using a differential pres-
sure transducer (PS-9805, Northwest Technologies) placed
at the bottom of a 0.5 m well. All environmental data
were logged on a combination of a multiplexor (AM32T,
Campbell Scientific) for temperature and a datalogger
(CR10X, Campbell Scientific), which were positioned
remotely in the marsh to minimize analog signal degra-
dation. Information was then relayed digitally between
the marsh and main datalogger (CR1000, Campbell Sci-
entific) using multidrop interfaces (MD485, Campbell
Scientific).

RESULTS AND DISCUSSION
TREATMENT APPLICATION

Average daily mean CO, + SE was 394 + 1.2 ppm in
ambient and 707 + 6.0 ppm in elevated chamber atmosphere
in 2007, a treatment difference of 313 ppm. The standard
deviation among daily means for individual chambers avy-
eraged 21.9 and 59.0 ppm for individual ambient and el-
evated chambers. The variation in means between days was
driven by differences in wind speed (Figure 7). High winds
resulted in incursions of ambient air into elevated cham-

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

900 Still day Windy day _ 49
pee pee

800} - 9

700 . 8
600 Lie
= 6 2
Es :
S 58
4 =
3 43
a
300 Windy night A=

s ri A ‘|

200 a pr vigee ‘es

100 : 4,

0

8/18/07 8/18/07 8/19/07 8/19/07 8/20/07 8/20/07 8/21/07 8/21/07 8/22/07
0:00 12:00 0:00 12:00 0:00 12:00 0:00 12:00 0:00

FIGURE 7. The CO) treatment and wind speed from four varying
days in August 2007. The shaded areas represent hours of dark-
ness when the CO, delivery was shut down. During those off hours,
all chambers were at ambient [CO,]. During still nights, ambient
[CO ] approached concentrations much higher than well-mixed at-
mospheric [CO ] as respired CO, accumulated relatively near the
ground.

bers, thereby diluting the elevated [CO ]. On the other
hand, stillness allowed respired CO, to accumulate over-
night, which increased background [CO,] in ambient and
elevated chambers. Because this buildup affects each treat-
ment equally, the difference between ambient and elevated
chambers persisted. However, wind incursions drove down
concentrations in elevated chambers only, which decreased
the difference between ambient and elevated [CO)].

In 2006, before chambers were equipped with frusta,
ambient and elevated chambers [CO] were 395 and 669,
a difference of 274 ppm. Although the mean [CO;] could
have been elevated in the chamber without adding frusta,
the fluctuations with wind would have been extreme, and
the expense of the additional CO) was deemed prohibitive.

The [NH,] in porewater was successfully increased by
the N addition by a factor of 2.9, from 17 to 64 umol L“!
averaged over the growing season in 2006. The factor by
which N addition increased porewater [NH,4] was much
higher early in the season and declined as growing plants
took up N.

MEASUREMENT VALIDATION: CHAMBER EFFECTS

Elevation

To examine the possibility of chamber effects on eleva-
tion, we examined the measurements in the ambient CO),

low-N (no added N) treatment (Figure 8). The in-chamber
measurements were very similar to those in the reference
plots. Both sets of data revealed significant changes in el-
evation through time (repeated-measures MANOVA, P <
0.05). Most notably, all plots experienced a dip of roughly
0.8 cm in total elevation during March 2007, followed
by a strong recovery. This dip was driven entirely by dy-
namics in the thickness of the deep zone. Compared to
absolute changes in elevation (range, >1.2 cm), the dif-
ferences between in-chamber and reference elevation were
relatively small (range, <0.2 cm). There was a trend of
a chamber effect on total elevation driven by deep zone
dynamics. This effect was significant in summer 2007 but
has vanished since then. The relativized root zone thick-
ness in ambient CO>, low-N chambers never differed from
zero (t test, P > 0.40 for all dates), which indicated that

In Chamber

oS ©
KO

2) ©
on’

-0.2

-~@- Total elevation
—v Deep zone thickness
-0.8 -4&- Root zone thickness

2 9
Oe

Cumulative change in
elevation or thickness (cm)

So
fo

Reference Plot

Cumulative change in
oO 2 ©
oN A

elevation or thickness (cm)

oo9989
Dn oO RW

Relativized

22 ©
On SB

-0.2

o 9
oe

Cumulative change in
elevation or thickness (cm)

-0.8

1/1/06 5/1/06 9/1/06 1/1/07 5/1/07 9/1/07 1/1/08
SS ee ee ea Se ed)
FIGURE 8. Elevation data from the five ambient CO , low-nitrogen
(N) plots. Change in total elevation is partitioned between changes
in thickness of either the deep zone or root zone. Top panel: eleva-
tion and thicknesses from inside the experimental chambers; middle
panel: from the adjacent reference plots; bottom panel: difference
between the in-chamber and reference measurements (relativized).
There was a slight chamber effect on total elevation, driven by con-
traction of deep zone thickness. Root zone did not differ from zero.

NUMBER 38 ¢ 399

there was not a detectable chamber effect in this stratum
where we expected treatment effects to be manifested.

Surface Accretion

One criticism of the design was that the chambers, by
enclosing plots, may have excluded sediments from being
deposited on the marsh surface. The difference between
in-chamber and reference accretion measured with cryo-
cores in November 2007 was small (0.058 cm) and did
not differ significantly from zero (95% confidence inter-
val: —0.18 to 0.07, 2 = 20). The treatment means also did
not differ from each other (two-way ANOVA: CO;, P >
0.10; N, P > 0.10) or from the reference plots (chamber
effect: P > 0.10; Figure 9).

CONCLUSIONS

The design of our field experiment proved robust to a
number of challenges unique to tidal salt marsh environ-
ments, including saltwater corrosion and deep tides. More
importantly the chamber design allowed us to consistently
elevate atmospheric CO. The frustum was a key feature
of the chamber because it average-stabilized and raised the
[CO,] in the elevated treatment, likely resulting in saved

12 Surface accretion
10 = In chamber
a ‘~ |] Reference plot
= 08 |
0.8
5 i
o
i 0.6 4
14)
Fs)
o 0.4 4
te
D
0.24
0.0 +

Elev+N

Amb Amb+N Elev

FIGURE 9. Surface accretion measured as the accumulation of mat-
ter on top of the marker horizon inside the chamber versus that
outside the chamber. It was thought the chambers may exclude ex-
ogenous sediment, but there was no difference between in-chamber
(black bars) and outside-chamber (reference plot, white bars) accre-
tion rates. Amb = ambient; Elev = elevation.

400 e

CO ,. N addition yielded higher porewater N concen-
trations as expected, but further chemical analyses are
needed for a more precise estimate of the magnitude of
the N treatment. The SET design allowed for sensitive
measures of soil elevation change. The chambers, perhaps
by virtue of their mass, appeared to slightly depress soil
elevation. However, there was no chamber effect in the
root zone where the most important treatment effects are
expected to occur. Further, the size of the chamber effect
on elevation was small (0.2 cm; Figure 8, bottom panel)
relative to the natural range of variation in those elevation
parameters (1.2 cm; Figure 8, middle panel).

ACKNOWLEDGMENTS

We thank J. Keller, G. Peresta, B. Drake, E. Sage,
A. Martin, D. McKinley, S. Chapman, K. White, and N.
Mudd from the Smithsonian Environmental Research
Center for construction and maintenance of the field ex-
periment. The field study was supported by the U.S. Geo-
logical Survey Global Change Research Program, U.S.
Department of Energy (grant DE-FG02-97ER62458), the
U.S. Department of Energy’s Office of Biological and En-
vironmental Research through the Coastal Center of the

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

National Institutes for Climatic Change Research at Tu-
lane University, and the Smithsonian Institution.

LITERATURE CITED

Cahoon, D. R., J. C. Lynch, and R. M. Knaus. 1996. Improved Cryo-
genic Coring Device for Sampling Wetland Soils. Journal of Sedi-
mentary Research, 66:1025-1027.

Cahoon, D. R., J. C. Lynch, B. C. Perez, B. Segura, R. D. Holland, C. Stelly,
G. Stephenson, and P. Hensel. 2002. High-Precision Measurements
of Wetland Sediment Elevation: II. The Rod Surface Elevation Ta-
ble. Journal of Sedimentary Research, 72:734-739.

Cahoon, D. R., D. J. Reed, and J. W. Day. 1995. Estimating Shal-
low Subsidence in Microtidal Salt Marshes of the Southeastern
United States—Kaye and Barghoorn Revisited. Marine Geology,
128:1-9.

Drake, B. G., P. W. Leadley, W. J. Arp, D. Nassiry, and P. S. Curtis. 1989.
An Open Top Chamber for Field Studies of Elevated Atmospheric
CO, Concentration on Saltmarsh Vegetation. Functional Ecology,
3:363-371.

Erickson, J. E., J. P. Megonigal, G. Peresta, and B. G. Drake. 2007. Salin-
ity and Sea Level Mediate Elevated CO, Effects on C3—C, Plant In-
teractions and Tissue Nitrogen in a Chesapeake Bay Tidal Wetland.
Global Change Biology, 13:202-215.

Rasse, D. P., J. H. Li, and B. G. Drake. 2003. Carbon Dioxide Assimila-
tion by a Wetland Sedge Canopy Exposed to Ambient and Elevated
CO): Measurements and Model Analysis. Functional Ecology,
17:222-230.

Herbivory, Nutrients, Stochastic Events,
and Relative Dominances of Benthic
Indicator Groups on Coral Reefs:

A Review and Recommendations

Mark M. Littler, Diane S. Littler,
and Barrett L. Brooks

Mark M. Littler, Diane S. Littler, and Barrett L.

Brooks, Department of Botany, National Mu-
seum of Natural History, Smithsonian Institution,
P.O. Box 37012, MRC 166, Washington, D.C.
20013-7012, USA. Corresponding author: M.
Littler (littlerm@si.edu). Manuscript received 13
May 2008; accepted 20 April 2009.

ABSTRACT. Threshold levels (i.e., tipping points where the probability of community
phase shifts is increased and the potential for recoverability is reduced) for critical bottom-
up interactions of productivity (e.g., nutrients) and those for top-down disturbances (e.g.,
herbivory) must be known to manage the competitive interactions determining the health
of coral-dominated reefs. We further posit that latent trajectories (reduced resiliencies/
recoverability from phase shifts) are often activated or accelerated by large-scale sto-
chastic disturbances such as tropical storms, cold fronts, warming events, diseases, and
predator outbreaks. In highly diverse and productive reef ecosystems, much of the overall
diversity at the benthic primary producer level is afforded by the interaction of opposing
nutrient-limiting/nutrient-enhancing and herbivory controls with the local physical and
spatial variability, such that a mosaic of environmental conditions typically occur in close
proximity. Although the relative dominance model (RDM) appears straightforwardly
simple, because of the nature of direct/indirect and stimulating/limiting factors and their
interactions it is extremely complex. For example, insufficient nutrients may act directly
to limit fleshy algal domination (via physiological stress); conversely, abundant nutrients
enhance fleshy algal growth, with the opposite effect on reef-building corals (via toxic
inhibition or increased diseases). Furthermore, the effects of controls can be indirect,
by influencing competition. Even this seemingly indirect control can have further levels
of complexity because competition between algae and corals can be direct (e.g., over-
growth) or indirect (e.g., preemption of substrate). High herbivory (via physical removal)
also acts indirectly on fleshy algae through reduced competitive ability, whereas lowered
herbivory and elevated nutrients also indirectly inhibit or control corals and coralline al-
gae by enhancing fleshy algal competition. Other ecologically important bottom-up fac-
tors, such as reduced light, abrasion, allelopathy, disease vectoring, and sediment smoth-
ering, also result from indirect side effects of fleshy algal competition. These factors tend
to selectively eliminate the !ong-lived organisms in favor of weedy fast-growing species,
thereby reducing desirable complexity and biodiversity.

INTRODUCTION

There has been an exhaustive debate in the coral reef literature over the rela-
tive importance of forces that regulate community structure and resilience (i.e.,
the potential to resist stresses and to recover following disturbances). The ex-
pansion of the human population and associated increases in destructive fishing

402 e

(Hughes, 1994) and nutrient loading (Lapointe, 1999), com-
pounded with ocean warming (Hoegh-Guldberg, 1999) and
stochastic environmental effects (Precht et al., 2005), have
been broadly debated to explain the increasing degrada-
tion of coral reefs worldwide. Because human population
growth is not expected to abate, discriminating among
various stressors is critical to determine conservation
strategies and to eventually ameliorate the accelerating
degradation of coral reefs. What has been lacking is the
ability to rigorously test and differentiate among the pos-
sible acute versus chronic stressors—leading to ongoing
controversy. In an attempt to address this problem, sev-
eral workers (Mora, 2008; Burkepile and Hay, 2006) have
conducted broad correlative and statistical assessments
of communities over large regional scales. These studies
have suggested a clear interaction between eutrophication
in conjunction with declining herbivorous organisms as
direct causes for maintaining present undesirable phase
shifts on coral reefs.

Such phase shifts have been devastating to the many
uniquely specialized benthic photosynthetic symbionts
dominating tropical reefs, which are responsible for some
of the most productive natural ecosystems known. Four
major space-occupying groups of benthic primary produc-
ers combine to create high coral-reef primary productivity:
reef-building corals (containing symbiotic algae), crustose
coralline algae, algal turfs (fleshy filamentous and low-
growing prostrate forms, and frondose macroalgae. Of
these, photosynthetic corals create much of the structural
heterogeneity and complexity and, with coralline algae,
are primarily responsible for accretion of CaCQ; into the
reef matrix—making them the most desirable functional
groups from a management perspective.

A basic objective in management ecology is to deter-
mine the mechanisms by which natural and anthropogenic
factors maintain or alter structure and interactions in bi-
otic communities. Anthropogenic eutrophication and de-
structive overfishing (i.e., herbivore removal by trapping,
netting, poisoning, blasting) are the most tractable fac-
tors correlated with the marked global decline in tropical
reef communities over the past two decades (see reviews
in Ginsburg, 1994; Birkeland, 1997; papers in Szmant,
2001). The theoretical framework involving “top-down”
regulation by predators and “bottom-up” control by re-
source availability in terrestrial systems was first proposed
by Hairston et al. (1960), concepts that were later used
(Atkinson and Grigg, 1984) to describe mechanisms that
regulate the structure of coral-reef communities. These
factors provide a valuable perspective (Figure 1) to assess
and manage the human activities that affect the interac-

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

tive mechanisms controlling stable states, tipping points,
phase shifts, and recovery among the dominant functional
groups of primary producers on tropical reefs.

In healthy coral-dominated reefs, nutrient concentra-
tions are extremely low and attachment space is occupied
by a broad diversity of three-dimensional overgrowing or-
ganisms. Given these conditions, the major tenets of the
management model proposed by Littler and Littler (2006:
fig. 1, relative dominance model [RDM)]) are (1) that com-
petition for space and light is crucial in determining the rela-
tive abundances of major benthic photosynthetic organisms,
and (2) that the outcome of competition for these resources
is most often, but not exclusively, controlled by the complex
interactions of biological factors (top-down controls such
as grazing) and environmental factors (bottom-up controls
such as nutrient levels). As suggested by Grime (1979)
for terrestrial plants and expanded for marine macroal-
gae (Littler and Littler, 1984; Steneck and Dethier, 1994),
primary producer abundance and evolutionary strategies
are controlled by physical disturbances (i.e., factors that
remove biomass) coupled with physiological stresses (i.e.,
factors that limit metabolic production). In the conceptual
relative dominance model (RDM; see Figure 1), grazing
physically reduces biomass (top-down) and nutrients con-
trol production (bottom-up). The complex natural inter-
actions between herbivory and nutrients are most dramat-
ically impacted by large-scale catastrophic disturbances
such as tropical storms (Done, 1992), warming events
(Macintyre and Glynn, 1990; Lough, 1994), cold fronts
(Precht and Miller, 2007), diseases (Santavy and Peters,
1997), and predator outbreaks (Cameron, 1977). These
events serve to trigger or accelerate the ultimate long-term
phase shifts postulated in the RDM. Such stochastic events
selectively eliminate the longer-lived organisms in favor of
faster-growing fleshy macroalgae, which are often compet-
itively superior (Birkeland, 1977). However, nutrients and
herbivory, in the absence of large-scale disturbances, are
both sufficient to maintain phase shifts independently or
in concert (Smith et al., 2001; Armitage and Fong, 2004;
Littler et al., 2006a).

On undisturbed oligotrophic coral-reef habitats, the
effects of well-documented top-down physical controls via
intense herbivory prevail, where changes in grazing inten-
sity often show acutely rapid effects. Conversely, bottom-up
stimulatory controls are more chronic, the result of lack of
nutrient availability, overcompensation by grazers, and a
slower growth response, compared with acute physical de-
struction by herbivory. However, under persistent elevated
nutrient conditions (relative to low [near-undetectable]
concentrations), consistent coral declines can occur, con-

INCREASING DECLINING RESILIENCE/RECOVERABILITY
rea SR EASRSaia aa anneeneni nS eNIEREIIN

HUMAN IMPACT
GRAZING ACTIVITY (TOP DOWN CONTROL)
HIGH REDUCED

LOW -GROWING
ALGAE AND TURF

LOW

FRONDOSE
MACROALGAE

DECLINING RESILIENCE/RECOVERABILITY
NUTRIENT LEVELS (BOTTOM UP CONTROL)

ELEVATED

FIGURE 1. The competition-based relative dominance model
(RDM). All the functional indicator groups occur under the condi-
tions of every compartment of the model; however, the RDM pre-
dicts which group will most often dominate. Light to dark shad-
ing indicates declining desirability of each functional group from
a management perspective. Crustose coralline algae are posited to
be competitively inferior and dominate mainly by default, that is,
where fleshy algae are removed by herbivores and some corals are
inhibited by nutrients. The dashed lines approximate tipping points
where declining herbivory and increasing nutrients reach critical lev-
els that begin to reduce resilience to and recoverability from phase
shifts. One vector can partially offset the other; for example, high
herbivory can delay the impact of elevated nutrients, or low nutrients
may offset the impact of reduced herbivory. As a baseline for healthy
coral-reef habitats, herbivore population abundances and diversity
should be high and palatable test plants should show at least greater
than 50% loss 6 h“! (i.e., <6 h half-life) during a replicated series of
midday in situ exposures. Hypothetical nutrient tipping points (i.e.,
thresholds that sustain algal growth) are thus far indicated to be quite
low (i.e., ~0.1 4M soluble reactive phosphorus [SRP], ~1.0 1M dis-
solved inorganic nitrogen [DIN]), as suggested by laboratory growth
experiments, case studies for macroalgal overgrowth of coral reef
communities, and in situ experimental nutrient enrichment research
(Bell, 1992; Bell and Elmetri, 1995; Lapointe et al., 1993, Bell et al.,
2007). We further posit that latent trajectories (reduced resilience/
recoverability from phase shifts) are often triggered or accelerated
by large-scale stochastic disturbances such as tropical storms, cold
fronts, warming events, diseases, and predator outbreaks. Although
these are events from which coral reefs have recovered for millions
of years in the absence of humans, when tipping points remain sur-
passed, less-desirable stable states can persist.

NUMBER 38 ¢ 403

comitant with algal increases that may lead to enduring
states throughout all combinations of herbivory (Littler
et al., 2006a). Changes in bottom-up controls and their
interactions not only alter the dominance patterns of the
major benthic functional groups on coral reefs but, hy-
pothetically, could have profound long-term consequences
mediated through structural transformations and chemi-
cal modifications to reef systems and their herbivorous
fish populations. In other words, excessive nutrient enrich-
ment not only increases the productivity and biomass of
weedy macroalgae via bottom-up controls that alter pat-
terns of competitive dominance (Littler et al., 1993) but,
over the long term, may lead to coral habitat degradation
through (1) reduced spatial heterogeneity by overgrowth
(Johannes, 1975; Pastorok and Bilyard, 1985; Szmant,
1997) and (2) nighttime anoxic conditions (tolerated by
macroalgae, but not by coral competitors and herbivorous
predators; Lapointe and Matzie, 1996) that could indi-
rectly reduce top-down grazer effects. Furthermore, fleshy
macroalgal blooms, irrespective of how they are induced,
decrease the growth and reproductive capacity of the more
structurally complex reef-building corals (Tanner, 1995;
Miller and Hay, 1996; Bellwood et al., 2006; Hughes et
al., 2007), as well as inhibit coral larval recruitment (Birke-
land, 1977; Tomascik, 1991; Ward and Harrison, 1997)
and survival (Lewis, 1986; Hughes et al., 1987; Hughes,
1989; Wittenberg and Hunte, 1992). Such complicated
feedback loops following eutrophication (e.g., anoxia) are
known to occur in seagrass meadows (Sand-Jensen and
Borum, 1991; Duarte, 1995) and could also explain de-
creases in fish populations on coral reefs with long-term
histories of eutrophication.

CORAL-REEF MANAGEMENT

The data relevant for long-term reef management con-
sist of (1) many important short-term caging and feeding
experiments (in the case of exceedingly well-documented
top-down herbivory effects), (2) circumstantial evidence
(Hallock et al., 1993), (3) correlative biogeographic sur-
veys contrasting oligotrophic versus eutrophic systems
(Littler et al., 1991; Verheij, 1993; Mora, 2008), (4) com-
parative experiments on systems containing natural nutri-
ent gradients (Lapointe et al., 2004, 2005b; Vroom et al.,
2005), (5) physiological assays (Littler and Littler, 1990;
Lapointe et al., 1997), and (6) logistically complicated, in
situ, long-term, experimental/causality studies, in the case
of bottom-up nutrient controls (Smith et al., 2001; Littler
et al., 2006a). Top-down control by abundant populations

404 e

of large mobile herbivores is particularly well studied for
coral reefs, beginning nearly five decades ago with the cag-
ing study of Stephenson and Searles (1960 As examples,
Sammarco et al. (1974), Ogden and Lobel (1978), Sam-
marco (1980), Carpenter (1986), Lewis (1986), Morris-
son (1988), and numerous other workers (see review by
McCook et al., 2001) have demonstrated that lowering
herbivory in low-nutrient habitats (usually assumed) often
results in rapid increases in low-growing stages of fleshy
macrophytes.

In the study of Lewis (1986) on the same reef flat stud-
ied by Littler et al. (2006a, 2006b), increases in a domi-
nant vegetative algal turf form (Vaughaniella stage) with
its upright fertile Padina blades, not blooms of mixed
macroalgae, followed short-term (11 week) reductions of
herbivorous fish grazing under conditions of low nutrient
levels. Lewis’ (1986) table 4 (although pseudoreplicated)
shows statistically significant, but relatively small, in-
creases (26%) in the above Vaughaniella-turf stage and its
reproductive Padina blades; however, in contrast to sev-
eral literature citations, no significant increases occurred
in any of the abundant upright macroalgal dominants such
as Turbinaria turbinata and Halimeda spp. Such low mats
are unique in containing an abundance of nitrogen-fixing
blue-green algae that can enrich other members within
the low-growing algal community (Adey and Goertmiller,
1987; Adey, 1998). In presumably higher-nutrient envi-
ronments, herbivore removals usually result in dramatic
blooms of larger frondose macroalgae (Bellwood et al.,
2006: fig. 4; Hughes et al., 2007).

Throughout the past decade, many biologists and
managers have not recognized the importance of chronic
nutrient enrichment and associated eutrophication prob-
lems facing coral reefs. A recent study (Littler et al., 2006b)
provided a detailed review and discussion of the misinter-
pretations, misunderstandings, and suboptimal experimen-
tal designs that pervade the literature in regard to nutrient
enrichment and the health of coral reefs. Overgrown 0.5 L
porous clay-pot diffusers (“mini-reefs,” following a de-
cade of recruitment, colonization, and competition) were
utilized (Littler et al., 2006b) to evaluate protocols for
studies of controlled nutrient enrichment on coral reefs.
A commonly used nutrient source, Tree Food Stakes con-
taining up to 6% chlorine, resulted in a significant 11-
fold and 20-fold decrease of fleshy algae and calcareous
coralline algae, respectively, relative to the control treat-
ments, while blue-green algae (Cyanobacteria) became
significantly (6 fold) more abundant. Osmocote-filled
mini-reefs showed no significant differences from the
controls for any of the indicator groups. By avoiding the

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

pitfalls of suboptimal study areas, insufficient duration of
colonization/competition studies, inadequate nutrient de- ©
tection limits, and inappropriate sources of enrichment in
future research, the potential to provide new insights into
the nutrient status of coral reefs will be greatly improved.
Nutrient research is logistically difficult and, because the
growth responses are relatively slow (i.e., chronic), re-
quires more emphasis on multifaceted approaches carried
out over sufficiently long time periods. Optimally, studies
should include in situ enrichment experiments that test the
long-term competitive interactions of functional indicator
groups on healthy coral-dominated reefs, in addition to
precisely monitoring water column nutrient levels, tissue
C:N:P ratios, and algal physiological response assays.

Although nutrient data are typically lacking in coral-
reef herbivory studies, natural background levels in con-
junction with ample water motion are usually assumed to
exceed levels that are limiting to macroalgal growth (Fong
et al., 2003). As pointed out by Lewis (1986), large fron-
dose macroalgae such as Sargassum and Turbinaria do oc-
cur in oligotrophic reef areas adjacent to coral colonies (see
also Littler et al., 1986; McCook et al, 2001; Vroom at
al., 2005); however, many of these frondose forms occupy
microhabitats that generate increased current acceleration,
such as the reef crest and tops of patch reef rocks, impli-
cating higher nutrient fluxes (Atkinson et al., 2001). Also,
large biomass/standing stocks of slow-growing perennial
macroalgae (e.g., rockweeds) can develop over time under
low inorganic nutrient concentrations; rainforests are good
illustrations of this as well. Furthermore, Sargassum spp.
can coexist with corals in oligotrophic waters by utilizing
particulate organic sources of nutrients (Schaffelke, 1999);
therefore, in this particular situation, large plant biomasses
of low diversity do not necessarily indicate detrimentally
abundant dissolved nutrients. Tissue analyses of mid-shelf
Sargassum transplants on the great barrier reef (McCook,
1999) revealed a C:N ratio of 32:1 and a C:P ratio of
1261:1, exceeding values for pelagic Sargassum in the nu-
trient-impoverished Sargasso Sea (C:P = 877:1; Lapointe,
1995), which are compelling for substantial N limitation
and severe P limitation. A further consideration is the now-
ubiquitous presence of significant anthropogenic nitrogen
sources (from burning fossil fuels) in rainfall worldwide
(Vitousek et al., 1997), making the term “pristine” relative,
at best. The demise of copious coral cover (Pollock, 1928)
and concomitant rise in frondose algae (Doty, 1971) and
coralline algae (Littler, 1971) on the reef flat at Waikiki,
Hawaii, was the first phase shift from coral to macroalgal
domination that was postulated (Littler, 1973) as caused
by increases in eutrophication (bottom-up control).

Eutrophication affects coral reefs to different degrees
and on varying scales. Several studies (Atkinson et al.,
1995; Grigg, 1995; Steven and Broadbent, 1997; McCook,
1999; Bongiorni et al., 2003) indicated no substantial ad-
verse responses of coral species to elevated nutrients. How-
ever, other laboratory and field experiments (Pastorok and
Bilyard, 1985; Tomascik and Sander, 1987; Muscatine et
al., 1989; Stambler et al., 1991; Jokiel et al., 1994; Koop
et al., 2001) have concluded that corals are negatively af-
fected by increased levels of nutrients and that diversity
suffers. Numerous in situ observations exemplify the types
of shifts from coral dominance to algal dominance that
suggest linkages with chronic nutrient loading, including
case studies in Hawaii (Littler, 1973; Banner, 1974; Smith
et al., 1981; Maragos et al., 1985; Grigg, 1995), Venezu-
ela (Weiss and Goddard, 1977), the Red Sea (Mergener,
1981; Walker and Ormond, 1982), Barbados (Tomascik
and Sander, 1985, 1987), American Samoa (Green et al.,
1997), Reunion Island (Cuet et al., 1988; Naim, 1993),
Bermuda (Lapointe and O’Connell, 1989), the Great
Barrier Reef (Bell, 1992), the Florida Keys (Lapointe et
al., 1994), Martinique (Littler et al., 1993), and Jamaica
(Goreau et al., 1997; Lapointe et al., 1997).

In a number of cases, herbivory patterns alone (simi-
lar to nutrient levels) do not explain the distribution and
abundance of benthic algae on coral reefs (Adey et al.,
1977; Hay, 1981; Hatcher, 1983; Hatcher and Larkum,
1983; Carpenter, 1986). Several studies (Hatcher, 1981;
Schmitt, 1997; Lirman and Biber, 2000) found no sig-
nificant correlation between grazing intensity and algal
biomass. A dramatic increase in algal biomass resulting
from eutrophication, without any simultaneous reduction
in herbivore populations, was reported (Fishelson, 1973).
The importance of the very low nutrient levels involved in
eutrophication (i.e., nutrient threshold hypothesis, NTH),
either natural or anthropogenic, has only recently come to
light (Bell, 1992; Lapointe et al., 1997; Small and Adey,
2001; Bell et al., 2007) regarding the potential for phase
shifts from corals toward macroalgal dominance. These
kinds of biotic phase shifts also have been attributed to
overfishing of herbivore stocks (see Hughes, 1994 on
Jamaican reef trends), in concert with cultural eutrophi-
cation (Goreau et al., 1997; Lapointe et al., 1997). It is
now clear (Burkepile and Hay, 2006; Mora, 2008) that
both herbivory and nutrient levels interact on large scales
as major factors in maintaining or degrading coral-reef
health.

We hasten to point out that individuals of all the func-
tional indicator groups can and do occur under the condi-
tions of every compartment of the RDM (see Figure 1);

NUMBER 38 ¢ 405

however, the model predicts which group most often will
dominate (as does the very similar fig. 2a in Bellwood et
al., 2004). Such apparent presence/absence anomalies,
on closer inspection, are often scientifically logical but
have led to different perspectives. Following large coral
bleaching events and die-offs in Belize, we have observed
dramatic increases in chemically defended sponges (e.g.,
Chondrilla) and Cyanobacteria (blue-green algae) under
high levels of grazing by sea urchins and fishes. Other ob-
servations that appear counterintuitive include some corals
growing in high-nutrient habitats, some large fleshy macro-
algae growing under low nutrients, certain turf algae ex-
posed to high herbivory, and the frequent coexistence of
crustose corallines and the other functional groups. We
agree with these observations and have addressed such
anomalies herein.

The general applicability as well as the limitations of
the RDM can be demonstrated further in relationship to
a number of recent studies. For example, nutrients and
herbivory are not independent, and the positive effects of
nutrients on marine plant productivity and growth can ac-
tually make plants more palatable and susceptible to graz-
ers (McGlathery, 1995; Boyer et al., 2004). Furthermore,
nutrient increases are sometimes associated with coral
inhibition (Koop et al., 2001) as well as coral diseases
(Harvell et al., 1999, 2002; Bruno et al., 2003), and algal
blooms can serve as disease vectors (Nugues et al., 2004).
The sophisticated enrichment study (ENCORE) on a large
and carefully controlled scale (Larkum and Koop, 1997;
Encore Group, 2001) did not produce supportive results
because (1) ambient nutrient levels within the lagoon at
One Tree Island are well above tipping-point concentra-
tions that may be inhibitory to some corals, while being
more than sufficient to support luxuriant frondose macro-
algal growth (Bell, 1992; Larkum and Koop, 1997; Bell
et al., 2007) and (2) the test organisms were isolated on
raised grids to measure growth rates, precluding natural
encroachment, overgrowth, or other competitive inter-
actions crucial to testing the RDM. However, all increases
in nutrient levels did adversely affect coral reproduction
(Koop et al., 2001). Additionally, several short-term (<4
months) studies (Thacker et al., 2001; Belliveau and Paul,
2002; Miller et al., 1999; McClanahan et al., 2002) re-
ported lack of algal stimulation following nutrient enrich-
ment, further documenting the low ambient nutrient con-
centrations sustaining ample algal growth.

In contrast, two in situ experimental studies con-
ducted over longer time scales in healthy coral-reef settings
(Smith et al., 2001; Littler et al., 2006a), in conjunction
with natural successional and competitive interactions,

406 e

provided the most relevant causality data demonstrating
the importance of both nutrient and herbivory influences;
the present review builds on these findings. The paper by
Lapointe (1997) was the first to put forth a convincing
case for the effectiveness of the RDM in addressing harm-
ful algal bloom issues on coral reefs. Additionally, highly
diverse living model systems of coral-reef communities
(i.e., mesocosms), operated for decades (Small and Adey,
2001), clearly have demonstrated that minute increases
in nitrogen and phosphorus reduce coral growth (some-
times causing substantial die-backs). Such self-contained
systems require continuous removal of nutrients by algal-
turf scrubbers or protein skimmers in combination with
an abundance of fish and invertebrate grazers to maintain
a high coral and algal diversity. The burgeoning aware-
ness of coral-reef degradation worldwide (see Ginsburg,
1994; chapters in Birkeland, 1997; Gardner et al., 2003),
particularly from coastal eutrophication (Bell, 1992; Win-
dom, 1992; Nixon, 1995; Lapointe, 1997, 1999) and de-
structive overfishing (Hughes, 1994; Jackson et al., 2001),
makes this management perspective relevant and oppor-
tune (see Figure 1).

Although harmful macroalgal blooms on coral reefs
have long been attributed to nutrient enrichment and
eutrophication (Littler, 1973; Banner, 1974; Johannes,
1975; Smith et al., 1981; Lapointe, 1997; Lapointe et al.,
2005a, 200Sb), some reef biologists have countered that
such changes in benthic community structure routinely
result primarily from natural stochastic events (Precht et
al., 2005), overfishing of herbivorous fish stocks (Hughes,
1994; Pandolfi et al., 2003; Lesser, 2004), or loss of key-
stone grazers, such as the long-spined sea urchin Diadema
antillarum (Jackson et al., 2001). Although generally sup-
ported, these last observations are not typical of the major-
ity of grazer reduction experiments in extreme oligotrophic
environments (see Lapointe, 1999), most of which have
reported an expansion of small low-growing algal forms
rather than macroalgal blooms (as predicted in Figure 1).
It is encouraging that the critical role of excess nutrients
on coral reefs has begun to receive attention in recent re-
view papers (Scheffer et al., 2001; Hughes et al. 2003;
Bellwood et al., 2004; Pandolfi et al., 2005; Burkepile and
Hay, 2006; Mora, 2008). Some scientists (e.g., Precht et
al., 2005) downplay declining resilience issues, instead em-
phasizing fundamental stochastic factors such as upwell-
ings, hurricanes, and cold fronts (see caption, Figure 1).
These occurrences represent unmanageable events from
which coral reefs have recovered for millions of years, but
not in the presence of modern human influences such as
destructive overfishing and nutrient pollution (see Mora,

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

2008). There are strong interactions between catastrophic
stochastic factors and the roles of herbivores and nutrients
that strongly impact reefs. For example, coral mortality
following hurricanes and coral bleaching events opens up
large amounts of new two-dimensional space readily colo-
nized by fast-growing algae. Such increases in productiv-
ity and the area available for grazing hypothetically sati-
ate the herbivore pressure over large areas, assuming that
natural herbivore populations have an upper limit in the
amount of reef area that they can graze effectively (Wil-
liams et al., 2001; Mumby, 2006). This diluted grazing
pressure and reduction in suitable shelter could in turn
lead to further increases in algal cover and a decline in the
recovery capacity (i.e., resilience) of coral communities.
Thus, stochastic processes are unquestionably important
factors in determining the trajectories of reef health and
interact with the processes discussed herein.

To establish the baseline conditions and detect subse-
quent changes, a combination of environmental, survey, in-
ventory, and bioassay data are essential to characterize and
monitor the ambient nutrient and herbivory environments
and antecedent nutrient history of a given management
area. Valid and reliable data are the cornerstone needed
to prioritize among different management strategies and
motivate the local populace and politicians/lawmakers to
support and implement the goals necessary for responsible
management. The RDM provides a clear visual depiction
that is easily understood and, therefore, can serve as a
convincing illustrative aid. It is essential that assessment
and monitoring methods should be both simple and rapid
to use. Chlorophyll a concentration (determined by fluo-
rometric or spectrophotometric methods; see Bell and El-
metri, 1995) is an especially useful ancillary indicator of
water column enrichment because phytoplankton blooms
can rapidly attenuate critical light energy while buffer-
ing inorganic nutrient pulses. Along with nutrient levels,
chlorophyll a serves as a valuable tipping-point indicator,
where levels in excess of 0.2-0.3 wg L~! indicate approach-
ing overabundances of nutrients (Bell et al., 2007).

Water column nutrient concentrations represent the
net sum of internal cycling, algal assimilation, and external
inputs, relative to macroalgal growth demands (Lapointe,
1997), and therefore offer the most direct method to assess
nutrient excesses on any given coral reef. Consequently, a
nutrient threshold model based on nutrient concentrations
(rather than on nutrient fluxes) is not only valid but is
likely the best index of nutrient status. Low-nutrient tip-
ping points, where increasing nutrients reach hypotheti-
cally critical levels that begin to reduce recoverability from
phase shifts (i.e., ~1.0 uM dissolved inorganic nitrogen

[DIN] = nitrogen: 0.014 ppm N or 0.040 ppm NO; and
~0.10 uM soluble reactive phosphorus [SRP] = phospho-
rus: 0.003 ppm P or 0.007 ppm POx,), have been broadly
corroborated (in developing the nutrient threshold hypoth-
esis [NTH]; Bell, 1992; Lapointe et al., 1993; Bell et al.,
2007) for sustaining macroalgal overgrowth of seagrass
beds and coral reefs. The physiological/kinetic basis for
such low-nutrient tipping points is the hyperbolic Monod
relation (Droop, 1985; Bell et al., 2007), which is also sup-
ported by controlled, high-flux, continuous-culture labo-
ratory experiments (Caperon et al., 1971 DeBoer et al.,
1978; Lapointe and Tenore, 1981). In our experience, if
modern analytical instruments can detect measurable nu-
trient levels, so can growth-limited macroalgae.

Additionally, a wealth of in situ coral-reef studies car-
ried out in areas characterized by nutrient levels only mod-
erately above the putative 0.1 wM SRP and 1.0 hM DIN
tipping points (Larkum and Koop, 1997; Miller et al.,
1999; Thacker et al., 2001) have reported minimal algal
stimulation following experimental nutrient enrichment,
further documenting the low natural nutrient concentra-
tions required for ample algal growth and their widespread
applicability. Some corals can tolerate high levels of DIN
and SRP; however, nutrient tipping points not much above
the present analytical limits of detection represent levels
of resource availability at which resilience begins to be re-
duced (Scheffer et al., 2001), such that stochastic or other
disturbances and stresses can trigger coral-reef ecosystem
shifts toward sustained dominance by macroalgal stable
states. Moreover, the macroalgal overgrowth experimen-
tally stimulated (Smith et al., 2001; Littler et al., 2006b)
in reduced-grazing/elevated-nutrient treatments demon-
strates that ambient nutrient concentrations inhibitory to
growth under the natural turbulence levels found on coral
reefs are similar to those reported above for other tropi-
cal marine algae. It should be noted that the remote reef
in the northwestern Hawaiian Islands studied by Smith et
al. (2001) had nutrient levels at or above the hypothetical
levels needed to sustain macroalgal growth (i.e., 1.1 ~M
DIN and 0.2 uM SRP). This system, with its present lack
of macroalgae and dominance by unbroken thickets of
three branching and one massive coral species, may be the
result of overcompensation by intense grazing and, con-
sequently, could be susceptible to a future relative domi-
nance reversal.

Littler et al. (2006a: tbl. 1) give typical baseline her-
bivorous fish assay and population density data con-
trasting natural Belize Barrier Reef sites of low and high
herbivory. Based on similar experiments conducted world-
wide on coral reefs by a range of workers (Hay, 1984;

NUMBER 38 ¢ 407

Lewis and Wainwright, 1985; Paul et al., 1987; Sluka
and Miller, 2001; Littler et al., 2006a), Littler and Littler
(2006 posited that less than a six hour half-life (>50%
mean loss per 6 h for palatable algae) during a series of
in situ, midday, assay periods is indicative of a healthy
level of herbivory for the particular habitat(s) tested. Her-
bivore abundances also should be enumerated by counting
numbers of individuals (by species), from midmorning to
midafternoon throughout a typical day for weather (Lit-
tler et al., 2006a, see their table 1), at fixed distances on
either side of random replicates of standardized transect
lines. Video transects are quick; enumeration can be done
later in the laboratory, and the videos provide a perma-
nent record of the target species (Littler et al., 1986).

FUNCTIONAL INDICATOR GROUPS

The fast growth and turnover rates of fleshy algae
compared to other reef organisms suggest their value as
early-warning indicators of reef degradation. Represen-
tatives of ubiquitous algal form/function groups (from
Littler and Littler, 2006) are increasingly encountered as
dominants on reefs, particularly those subjected to human
activities (see Littler and Littler, 2006: fig. 2).

REEF-BUILDING CORALS (CNIDARIA)

A predominance of diverse corals and calcareous
coralline algae are universally accepted as the most de-
sirable components of biotic reefs because of (1) their
three-dimensional architecture, which provides habitats
for a myriad of other reef organisms (largely responsible
for much of the heterogeneity/high biodiversity), (2) their
roles in producing the massive carbonate structure of
reefs, and (3) their aesthetic qualities. The vertical struc-
ture and horizontal canopies of branching forms allow
abundant populations of shade-dwelling crustose coralline
algae to co-occur. Reef-building corals, while preyed upon
by a few omnivorous fishes and specialist invertebrates
(e.g., crown-of-thorns sea star), generally achieve domi-
nance under the top-down control of intense herbivory
(Lewis, 1986; Lirman, 2001) and extremely low nutrient
concentrations (Bell, 1992; Lapointe et al., 1993). Mas-
sive corals are resistant to grazing at the higher levels of
herbivory (Littler et al., 1989). Hard mound-shaped forms
show relatively little colony mortality under high grazing
pressure, even though occasionally rasped by parrotfishes.
Contrastingly, some delicately branched corals such as
Porites porites are quite palatable and readily eaten by

408 e

parrotfishes (e.g., Sparisoma viride; Littler et al., 1989;
Miller and Hay, 1998). Nutrient increases are sometimes
associated with coral diseases (Harvell et al., 1999, 2002;
Bruno et al., 2003). As mentioned earlier, numerous cor-
als tolerate elevated nutrient levels (Atkinson et al., 1995;
Steven and Broadbent, 1997; Bongiorni et al., 2003), but
their diversity suffers. Conversely, others are physiologi-
cally inhibited by increases in nitrate (e.g., Montastrea an-
nularis and Porites porites: Marubini and Davies, 1996),
ammonium (e.g., Pocillopora damicornis: Stambler et al.,
1991; Muller-Parker et al., 1994), and orthophosphate
(e.g., Porites compressa: Townsley, cited in Doty, 1969;
P. damicornis and Stylophora pistillata: Hoegh-Guldberg
et al., 1997). Nutrient inhibition of coral larval settlement
also has been shown for Acropora longicyathis (Ward and
Harrison, 1997). During the extensive ENCORE program
on Heron Island, all increases in nutrient levels adversely
affected coral reproduction (Koop et al., 2001).

MACROALGAE

With an increase in nutrients, the growth of harm-
ful fleshy algae is favored over that of the slower-growing
but highly desirable corals (Genin et al., 1995; Miller and
Hay, 1996; Lapointe et al., 1997), and the latter become
inhibited by competition for space and light, increased
diseases, and physiological inhibition. On healthy oligo-
trophic coral reefs, even very low nutrient increases may
exceed critical levels that can shift relative dominances by
stimulating macroalgal production while inhibiting corals.
As indicated earlier, large biomass, or standing stocks, of
slow-growing perennial macroalgae (e.g., rockweeds) can
develop over time under low inorganic nutrient concentra-
tions (McCook, 1999), and Sargassum spp. can coexist
with corals in oligotrophic waters by utilizing particulate
organic sources of nutrients (Schaffelke, 1999). Therefore,
in this particular situation, large plant biomasses do not
necessarily indicate detrimentally abundant dissolved nu-
trients. Filamentous and frondose algae can outcompete
corals (Birkeland, 1977; but see McCook et al., 2001),
many of which are inhibited under elevated nutrient levels
(reviewed in Marubini and Davies, 1996). Fast-growing al-
gae are not just opportunists that depend on disturbances
to release space resources from established longer-lived
populations but become the superior competitors (Birke-
land, 1977) when provided with sufficient nutrients. As a
result, frondose macroalgae as a group are now generally
recognized as harmful to the longevity of coral reefs be-
cause of the linkage between excessive blooms and coastal
eutrophication (ECOHAB, 1997). Potential competitive

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

dominance of fast-growing macroalgae is inferred from
their overshadowing canopy heights, as well as from in- ~
verse correlations in abundances between algae and the
other benthic producer groups (Lewis, 1986), particularly
at elevated nutrient concentrations (Littler et al., 1993;
Lapointe et al., 1997). Macroalgae, such as Halimeda
spp., also gain competitive advantage by serving as car-
riers of coral diseases (Nugues et al., 2004). The fleshy
macroalgal form-group has proven to be particularly at-
tractive to herbivores (see Hay, 1981; Littler et al., 1983a,
1983b) and only becomes abundant where grazing is de-
creased or swamped by excessive algal growth (chemically
defended forms, e.g., Cyanobacteria, are exceptions). Such
overcompensation by herbivory may explain some of the
reported cases (Crossland et al., 1984; Szmant, 1997;
Smith et al., 2001) of specific corals surviving high-nutri-
ent reef environments.

CRUSTOSE CORALLINE ALGAE

The predominant members of this indicator group,
the coralline algae, tend to be slow-growing, competitively
inferior taxa abundant in most reef systems (Littler, 1972).
However, they span a spectrum of morphotypes from thin
sheet-like crusts to thick massive pavements to upright
branched and columnar coral-like heads that contribute
to both cementation and bulk. This functional group is
highly resilient and is able to recover or restore the coral-
reef system relatively more quickly, given that some crus-
tose coralline algae chemically attract and facilitate the
survival of coral larvae (Harrington et al., 2004) whereas
the other two algal functional groups inhibit larval settle-
ment. Because crustose corallines continually slough up-
per surface layers, they play a key role, as do filter-feeding
corals, in physically preventing the settlement and colo-
nization of many undesirable fleshy fouling organisms on
coral reefs (Littler and Littler, 1997). Crustose corallines,
because of their slow growth rates, tolerate low nutrient
levels and generally are conspicuous, but not dominant,
under low concentrations of nutrients and high levels of
herbivory (Littler et al., 1991). Accordingly, they do well
under both low and elevated nutrients; that is, most are
not inhibited by nutrient stress and many are maintained
competitor free by surface cell layer shedding (Johnson
and Mann, 1986), even at lower levels of grazing (Littler
and Littler, 1997). Therefore, crustose coralline algae do
not require elevated nutrients, as might be inferred from
the RDM (Figure 1); instead, their rise to dominance is
largely controlled indirectly by the factors influencing
the abundances of the other groups, primarily corals and

fleshy macroalgae. The key point is that crustose corallines
predominate mainly by default (i.e., under conditions of
minimal competition), where either corals are inhibited by
elevated nutrients or fleshy algae are removed by intense
herbivory. In independent corroboration of the herbivory
portion of the RDM, a gradient of frondose- to turf- to
coralline algal groups was closely correlated with escalat-
ing herbivory on coral reefs (Steneck, 1989).

LOw-GROWING AND TURF ALGAE

The turf algae are mostly dense filamentous and low-
growing frondose members of all four algal phyla and tend
to become dominant under minimal inhibitory top-down
and stimulatory bottom-up controls. Domination by low-
growing algae suggests desirably low nutrient levels but an
inadequate herbivory component. Their relatively small
size and rapid perennation results in moderate losses to
herbivory at low grazing pressures. They have opportunis-
tic life history characteristics, including the ability to main-
tain substantial nutrient uptake and growth rates under
low-nutrient conditions (Rosenberg and Ramus, 1984),
and also contain an abundance of nitrogen-fixing Cyano-
bacteria (Adey and Goertemiller, 1987; Adey, 1998) that
can enrich other low-growing members of the dense turf
community. Algal turfs have been shown to be favored
under reduced nutrient-loading rates (Fong et al., 1987)
or infrequent nutrient pulses (Fujita et al., 1988) and can
form extensive horizontal mats.

DISCUSSION

This paper directly addresses the goals of an impera-
tive research agenda (ECOHAB, 1997) by providing a
management perspective and assessment strategies for the
mechanisms that initiate and sustain harmful blooms of
algae that degrade coral-reef ecosystems. The complex in-
teractions of herbivory and nutrients can change gradually
with no apparent effects to induce subtle declines in resil-
iency and recoverability of coral/coralline-dominated reef
systems (Scheffer et al., 2001). As mentioned, these sys-
tems then become vulnerable to catastrophic impacts by
large-scale stochastic disturbances that typically trigger or
accelerate such low-resilience reef systems (Scheffer et al.,
2001; Bellwood et al., 2004). Most importantly, recovery
to coral domination cannot occur unless tipping points are
returned to healthy levels, and even then alternative stable
states may persist. For example, when catastrophic events
selectively eliminate the longer-lived organisms in favor of

NUMBER 38 ¢ 409

early-successional fleshy algae (Littler and Littler, 1984),
the settlement of coral planulae is prevented and the algae
persist as competitively superior states (Birkeland, 1977;
Lewis, 1986). For completeness, we also point out the ob-
vious devastating effects of toxic spills, carbonate mining,
land-fill, and sediment inundation, some of which also are
associated with nutrient pollution and algal blooms.
Because of global-scale degradation of coral-reef eco-
systems (Ginsburg, 1994; Wilkinson, 1999), it is impor-
tant to obtain relevant information on tipping points for
both top-down herbivory (relatively fast acting, acute) and
bottom-up nutrient controls (slower acting, chronic), both
of which are reemphasized. As the first approximation, we
posit that on a healthy reef system, herbivore abundances
and diversity should be high, and palatable test plants
should show at least a 50% mean loss per six hours (i.e.,
<6 h half-life) during a series of midday in situ assays.
Table 1 in Littler and Littler (2006 summarizes baseline
assay and critical fish population data of this sort for two
natural coral-reef zones of low and high herbivory.
Nutrient threshold points (where increasing water
column nutrients reach critical resilience levels such that
they reduce recovery from phase shifts) have been widely
postulated (as ~1.0 »M DIN and ~0.10 uM SRP [NTH];
Bell, 1992; Lapointe et al., 1993; Bell and Elmetri, 1995)
for potential macroalgal overgrowth of coral-reef commu-
nities. As mentioned earlier, a further useful tipping-point
indicator is water column chlorophyll a, where levels in
excess of 0.2-0.3 wg L~! also indicate detrimental over-
abundances of nutrients (Bell and Elmetri, 1995).

CONCLUSIONS

Assessment protocols for determining and monitoring
the status of any given coral reef are suggested: these in-
clude (a) herbivore population assessments, (b) herbivory
assays, (c) water column nutrient levels, and (d) standing
stocks of functional indicator groups. These measure-
ments can reveal quantitative tipping-point levels beyond
which resilience to and recovery from undesirable phase
shifts begin to become critically reduced. Tipping-point
approximations are reviewed and posited both for inor-
ganic nutrients and for herbivory.

This review specifically addresses the relatively acute
top-down effects of herbivory and the more chronic
bottom-up effects of nutrient enrichment on critical indi-
cator groups of benthic primary producers: reef-building
corals, crustose coralline algae, dense turf algae, frondose
macroalgae, and herbivore associates.

410 e

A predominance of massive corals and calcareous coral-
line algae relative to frondose macroalgae and low-growing
algae indicates a healthy spatially heterogeneous condition
reflecting low nutrients and high herbivory. With a few ex-
ceptions, an abundance of frondose macroalgae illustrates
the least desirable condition of elevated nutrient levels and
reduced herbivory, possibly reflecting eutrophication in
concert with destructive herbivore fishing practices. A high
coverage of coralline algae suggests healthy high herbivory
levels but also suggests problems with elevated nutrients that
may be inhibitory to some corals. Domination by dense low-
growing and turf algae indicates desirably low nutrient lev-
els but also suggests an inadequate herbivory component.

From a management perspective, levels of herbivory
and herbivore populations and of nutrients rank among
the most useful quantitative indicators of coral-reef re-
silience and recoverability, whereas the degree of health,
degradation, and mortality are inferred by the relative
abundances of functional indicator groups.

The bioassay and indictor group monitoring ap-
proaches provide powerful perspectives and essential
measurement criteria to enable resource managers to pro-
tect coral reefs and similar coastal systems from eutro-
phication, destructive overfishing, and initiation of harm-
ful algal blooms. Human population growth has always
been accompanied by changes in land and sea use and
by increased exploitation of natural resources, attitudes
that continue to cause broad alterations in the structure
of coral-reef communities. Unless curbed, anthropogeni-
cally induced phase shifts will expand geographically at
an accelerated pace. However, solutions are available,
which include the use of Marine Protected Areas, banning
of destructive fishing practices (e.g., trapping, poisoning,
blasting, netting), and regulations protecting keystone her-
bivorous fish species (e.g., parrotfish, surgeonfish) from
market exploitation. Fisheries controls must be backed up
by strategies to regulate the effects of pollution along with
an international commitment to reduce the emission of
greenhouse gases and, finally, the implementation of long-
term strategies to reduce or stabilize the ultimate cause of
all these stressors, the world’s human population growth.

ACKNOWLEDGMENTS

Support for this work came from a Scholarly Studies
Grant (Smithsonian Institution, Office of Fellowships and
Grants), the Caribbean Coral Reef Ecosystems (CCRE)
Program (administered by Klaus Ruetzler, the Smithson-
ian Marine Station at Ft. Pierce (Valerie Paul, Head Scien-

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

tist), and the National Museum of Natural History. This is
CCRE contribution number 847, supported in part by the
Hunterdon Oceanographic Research Fund, and SMSFP
contribution number 763.

LITERATURE CITED

Adey, W. H. 1998. Coral Reefs: Algal Structured and Mediated Ecosys-
tems in Shallow, Turbulent Alkaline Waters. Journal of Phycology,
34:393-406.

Adey, W. H., P. Adey, R. Burke, and L. Kaufman. 1977. The Holocene
Reef Systems of Eastern Martinique, French West Indies. Azoll
Research Bulletin, No. 218:1-40.

Adey, W. H., and T. Goertemiller. 1987. Coral Reef Algal Turfs—Master
Producers in Nutrient Poor Seas. Phycologia, 26:374-386.

Armitage, A. R., and P. Fong. 2004. Upward Cascading Effects of Nu-
trients: Shifts in a Benthic Microalgal Community and a Negative
Herbivore Response. Oecologia (Berlin), 139:560-567.

Atkinson, M. J., B. Carlson, and G. L. Crow. 1995. Coral Growth in High-
Nutrient, Low-pH Seawater: A Case Study of Corals Cultured at the
Waikiki Aquarium, Honolulu, Hawaii. Coral Reefs, 14:215-223.

Atkinson, M. J., J. L. Falter, and C. J. Hearn. 2001. Nutrient Dynamics
in the Biosphere 2 Coral Reef Mesocosm: Water Velocity Controls
NHsg and PO, Uptake. Coral Reefs, 20:341-346.

Atkinson, M. J., and R. W. Grigg. 1984. Model of a Coral Reef Ecosys-
tem. II. Gross and Net Benthic Primary Production at French Frig-
ate Shoals, Hawaii. Coral Reefs, 3:13-22.

Banner, A. H. 1974. Kaneohe Bay, Hawaii: Urban Pollution and a Coral
Reef Ecosystem. Proceedings of the Second International Coral Reef
Symposium, 2:685-702.

Bell, P. R. E 1992. Eutrophication and Coral Reefs: Some Examples in
the Great Barrier Reef Lagoon. Water Resources, 26:555-568.

Bell, P. R. F., and I. Elmetri. 1995. Ecological Indicators of Large-
Scale Eutrophication in the Great Barrier Reef Lagoon. Ambio,
24:208-215.

Bell, P. R. E, B. E. Lapointe, and I. Elmetri. 2007. Reevaluation of EN-
CORE: Support for the Eutrophication Threshold Model for Coral
Reefs. Ambio, 6:416-424.

Belliveau, S. A., and V. J. Paul. 2002. Effects of Herbivory and Nutrients
on the Early Colonization on Crustose Corallines and Fleshy Algae.
Marine Ecology Progress Series, 232:105-114.

Bellwood, D. R., T. P. Hughes, and A. S. Hoey. 2006. Sleeping Func-
tional Group Drives Coral Reef Recovery. Current Biology, 16:
2434-2439.

Bellwood, D. R., T. P. Hughes, and M. Nystrom. 2004. Confronting the
Coral Reef Crisis. Nature (London), 429:827-833.

Birkeland, C. 1977. The Importance of Rate of Biomass Accumulation in
Early Successional Stages of Benthic Communities to the Survival of
Coral Recruits. Proceedings of the Third International Coral Reef
Symposium, 1:15-21.

. 1997. Life and Death of Coral Reefs. New York: Chapman
& Hall.

Bongiorni, L., S. Shafir, D. Angel, and B. Rinkevich. 2003. Survival,
Growth and Gonad Development of Two Hermatypic Corals Sub-
jected to In Situ Fish-Farm Nutrient Enrichment. Marine Ecology
Progress Series, 253:137-144.

Boyer, K. E., P. Fong, A. R. Armitage, and R. A. Cohen. 2004. Elevated
Nutrient Content of Tropical Macroalgae Increases Rates of Her-
bivory in Coral, Seagrass, and Mangrove Habitats. Coral Reefs,
23:530-538.

Bruno, J. E, L. E. Petes, C. E. Harvell, and A. Hettinger. 2003. Nutrient
Enrichment Can Increase the Severity of Coral Diseases. Ecology
Letters, 6:1056-1061.

Burkepile, D. E., and M. E. Hay. 2006. Herbivore Versus Nutrient Con-
trol of Marine Primary Producers: Context-Dependent Effects.
Ecology, 87:3128-3139.

Cameron, A. M. 1977. Acanthaster and Coral Reefs; Population Out-
breaks of a Rare and Specialized Carnivore in a Complex High-
Diversity System. Proceedings of the Third International Coral Reef
Symposium, 1:193-200.

Caperon, J., S. A. Cattel, and G. Krasnick. 1971. Phytoplankton Kinetics
in a Subtropical Estuary: Eutrophication. Limnology and Oceanog-
raphy, 16:599-607.

Carpenter, R. C. 1986. Partitioning Herbivory and Its Effects on Coral
Reef Algal Communities. Ecological Monographs, 56:345-363.

Crossland, C. J., B. G. Hatcher, M. J. Atkinson, and S. V. Smith. 1984.
Dissolved Nutrients of a High-Latitude Coral Reef, Houtman
Abrolhos Island, Western Australia. Marine Ecology Progress
Series, 14:159-163.

Cuet, P., O. Naim, G. Faure, and J. Y. Conan. 1988. Nutrient-Rich
Groundwater Impact on Benthic Communities of La Saline Fring-
ing Reef (Reunion Island, Indian Ocean): Preliminary Results.
Proceedings of the Sixth International Coral Reef Symposium,
2:1335-1340.

DeBoer, J. A., H. J. Guigli, T. L. Israel, and C. F. D’Elia. 1978. Nutri-
tional Studies of Two Red Algae. I. Growth Rate as a Function
of Nitrogen Source and Concentration. Journal of Phycology, 14:
261-266.

Done, T. J. 1992. Effects of Tropical Cyclone Waves on Ecological and
Geomorphological Structures on the Great Barrier Reef. Continen-
tal Shelf Research, 12:859-872.

Doty, M. S. 1969. The Ecology of Honaunau Bay, Hawaii. Hawaii Bo-
tanical Science Paper No. 14:1-221.

. 1971. Antecedent Event Influence on Benthic Marine Algal
Standing Crops in Hawaii. Journal of Experimental Marine Biol-
ogy and Ecology, 6:161-166.

Droop, M. R. 1985. 25 Years of Algal Growth Kinetics—A Personal
View. Botanica Marina, 26:99-112.

Duarte, C. M. 1995. Submerged Aquatic Vegetation in Relation to Dif-
ferent Nutrient Regimes. Ophelia, 41:87-112.

ECOHAB. 1997. “The Ecology and Oceanography of Harmful Algal
Blooms: A National Research Agenda.” In Proceedings of the Na-
tional Workshop, ed. D. M. Anderson, pp. 1-66. Woods Hole,
Mass.: WHOI.

Encore Group. 2001. The Effect of Nutrient Enrichment on Coral Reefs.
Synthesis of Results and Conclusions. Marine Pollution Bulletin,
42:91-120.

Fishelson, L. 1973. Ecology of Coral Reefs in the Gulf of Aqaba (Red
Sea) Influenced by Pollution. Oecologia (Berlin), 12:55-67.

Fong, P., K. E. Boyer, K. Kamer, and K. A. Boyle. 2003. Influence of
Initial Tissue Nutrient Status of Tropical Marine Algae in Response
to Nitrogen and Phosphorus Additions. Marine Ecology Progress
Series, 262:111-123.

Fong, P., R. Rudnicki, and J. B. Zedler. 1987. Algal Community Response
to Nitrogen and Phosphorus Loading in Experimental Mesocosms:
Management Recommendations for Southern California Lagoons.
San Diego, Calif.: San Diego Association of Governments.

Fujita, R. M., P. A. Wheeler, and R. L. Edwards. 1988. Metabolic Regu-
lation of Ammonium Uptake by Ulva rigida (Chlorophyta): A Com-
partmental Analysis of the Rate-Limiting Step for Uptake. Journal
of Phycology, 24:560-566.

Gardner, T. A., I. M. Cote, J. A. Gill, A. Grant, and A. R. Watkinson.
2003. Long-Term Region-Wide Declines in Caribbean Corals. Sci-
ence, 301:958-960.

NUMBER 38 e¢ 411

Genin, A., B. Lazar, and S. Brenner. 1995. Vertical Mixing and Coral
Death in the Red Sea Following the Eruption of Mount Pinatubo.
Nature (London), 377:507-510.

Ginsburg, R. N., compiler. 1994. Proceedings of the Colloquium on
Global Aspects of Coral Reefs: Health, Hazards and History. Mi-
ami, Fla.: University of Miami.

Goreau, T. J., L. Dalay, S. Ciappara, J. Brown, $. Dourke, and K.
Thacker. 1997. Community-Based Whole-Watershed and Coastal
Zone Management in Jamaica. Proceedings of the Eighth Interna-
tional Coral Reef Symposium, 2:2093-2096.

Green, A. L, R. H. Birkeland, R. H. Randall, B. D. Smith, and S. Wilkins.
1997. 78 Years of Coral Reef Degradation in Pago Pago Harbour:
A Quantitative Record. Proceedings of the Eighth International
Coral Reef Symposium, Panama, 2:1883-1888.

Grigg, R. W. 1995. Coral Reefs in an Urban Embayment in Hawaii: A
Complex Case History Controlled by Natural and Anthropogenic
Stress. Coral Reefs, 14:253-266.

Grime, J. P. 1979. Plant Strategies and Vegetation Processes. New York:
John Wiley & Sons.

Hairston, N. G., E E. Smith, and L. B. Slobodkin. 1960. Community
Structure, Population Control, and Competition. American Natu-
ralist, 94:421-425.

Hallock, P., F E. Miiller-Karger, and J. C. Halas. 1993. Coral Reef
Decline. National Geographic Research and Exploration, 9:358-
378.

Harrington, L., K. Fabricius, G. De’ath, and A. Negri. 2004. Recognition
and Selection of Settlement Substrata Determine Post-settlement
Survival in Corals. Ecology, 85:3428-3437.

Harvell, C. D., K. Kim, J. M. Burkholder, R. R. Collwell, P. R. Epstein,
D. J. Grimes, E. E. Hofmann, E. K. Lipp, A. D. M. E. Osterhaus,
R. M. Overstreet, J. W. Porter, G. W. Smith, and G. R. Vasta. 1999.
Emerging Marine Diseases—Climate Links and Anthropogenic
Factors. Science, 301:1505-1510.

Harvell, C. D., C. E. Mitchell, J. R. Ward, S. Altizer, A. P. Dobson, R. S.
Ostfeld, and M. D. Samuel. 2002. Climate Warming and Disease
Risks for Terrestrial and Marine Biota. Science, 296:2158-2162.

Hatcher, B. G. 1981. The Interaction Between Grazing Organisms and
the Epilithic Algal Community of a Coral Reef: A Quantitative As-
sessment. Proceedings of the Fourth International Coral Reef Sym-
posium, Manila, 2:515-524.

. 1983. Grazing in Coral Reef Ecosystems. In Perspectives on
Coral Reefs, ed. D. J. Barnes, pp. 164-179. Townsville, Australia:
Australian Institution of Marine Science.

Hatcher, B., and A. W. D. Larkum. 1983. An Experimental Analysis of
Factors Controlling the Standing Crop of the Epilithic Algal Com-
munity on a Coral Reef. Journal of Experimental Marine Biology
and Ecology, 69:61-84.

Hay, M. E. 1981. Spatial Patterns of Grazing Intensity on a Caribbean
Barrier Reef: Herbivory and Algal Distribution. Aquatic Botany,
11:97-109.

. 1984. Predictable Escapes from Herbivory: How Do These
Affect the Evolution of Herbivore Resistance in Tropical Marine
Communities? Oecologia (Berlin), 64:396-407.

Heegh-Guldberg, O. 1999. Climate Change, Coral Bleaching and the Fu-
ture of the World’s Coral Reefs. Journal of Marine and Freshwater
Research, 50:839-866.

Hgegh-Guldberg, O., M. Takabayashi, and G. Moreno. 1997. The Im-
pact of Long-Term Nutrient Enrichment on Coral Calcification and
Growth. Proceedings of the Eighth International Coral Reef Sym-
posium, 1:861-866.

Hughes, T. P. 1989. Community Structure and the Diversity of Coral
Reefs: The Role of History. Ecology, 78:275-279.

. 1994. Catastrophes, Phase Shifts, and Large-Scale Degradation

of a Caribbean Coral Reef. Science, 265:1547-1551.

412 e

Hughes, T. P., A. H. Baird, D. R. Bellwood, M. Card, S. R. Connolly,
C. Folkes, R. Grosberg, O. Hgegh-Guldberg, J. B. C. Jackson,
J. Kleypas, J. M. Lough, P. Marshall, M. Nystrom, S. R. Palumbi,
J. M. Pandolfi, B. Rosen, and J. Roughgarden. 2003. Climate
Change, Human Impacts, and the Resilience of Coral Reefs. Sci-
ence, 301:929-933.

Hughes, T. P., D. C. Reed, and M. Boyle. 1987. Herbivory on Coral
Reefs: Community Structure Following Mass Mortalities of Sea
Urchins. Journal of Experimental Marine Biology and Ecology,
113:39-59.

Hughes, T. P., M. J. Rodrigues, D. R. Bellwood, D. Ceccarelli, O. Hoegh-
Guldberg, L. McCook, N. Moltschaniwskyj, M. S. Pratchett, R. S. Ste-
neck, and B. Willis. 2007. Phase Shifts, Herbivory, and the Resilience
of Coral Reefs to Climate Change. Current Biology, 17:360-365.

Jackson, J. B. C., M. X. Kirby, W. H. Berger, K. A. Bjorndal, L. W.
Botsford, B. J. Bourque, R. Bradbury, R. Cooke, J. Erlandson,
J. A. Estes, T. P. Hughes, S. Kidwell, C. B. Lange, H. S. Lenihan,
J. M. Pandolfi, C. H. Peterson, R. S. Steneck, M. J. Tegner, and R.
Warner. 2001. Historical Over-Fishing and the Recent Collapse of
Coastal Ecosystems. Science, 293:629-638.

Johannes, R. E. 1975. Pollution and the Degradation of Coral Reef Com-
munities. In Tropical Marine Pollution, ed. E. J. F. Wood and R. E.
Johannes, pp. 13-51. New York: Elsevier.

Johnson, C. R., and K. H. Mann. 1986. The Crustose Coralline Alga,
Phymatolithon Foslie, Inhibits the Overgrowth of Seaweeds With-
out Relying on Herbivores. Journal of Experimental Marine Biol-
ogy and Ecology, 96:127-146.

Jokiel, P., Z. Dubinsky, and N. Stambler. 1994. Nutrient Limitation in
the Symbiotic Association Between Zooxanthellae and Reef Build-
ing Corals. Pacific Science, 48:215-218.

Koop, K., D. Booth, A. Broadbent, J. Brodie, D. Bucher, D. Capone, J.
Coll, W. Dennison, M. Erdmann, P. Harrison, O. Haegh-Guldberg,
P. Hutchings, G. B. Jones, A. W. D. Larkum, J. O’Neill, A. Steven,
E. Tentori, S$. Ward, J. Williamson, and D. Yellowlees. 2001. EN-
CORE: The Effect of Nutrient Enrichment on Coral Reefs. Synthesis
of Results and Conclusions. Marine Pollution Bulletin, 42:91-120.

Lapointe, B. E. 1995. A Comparison of Nutrient-Limited Productivity in
Sargassum natans from Neritic vs. Oceanic Waters of the Western
North Atlantic Ocean. Limnology and Oceanography, 40:625—-
633.

. 1997. Nutrient Thresholds for Bottom-up Control of Macroalgal

Blooms on Coral Reefs in Jamaica and Southeast Florida. Limnol-

ogy and Oceanography, 42:1119-1131.

. 1999. Simultaneous Top-down and Bottom-up Forces Control
Macroalgal Blooms on Coral Reefs (Reply to the Comment by
Hughes et al.). Limnology and Oceanography, 44:1586-1592.

Lapointe, B. E., P. J. Barile, M. M. Littler, D. S. Littler, B. J. Bedford, and
C. Gasque. 2005a. Macroalgal Blooms on Southeast Florida Coral
Reefs: I. Nutrient Stoichiometry of the Invasive Green Alga Codium
isthmocladum in the Wider Caribbean Indicates Nutrient Enrich-
ment. Harmful Algae, 4(6):1092-1105.

Lapointe, B. E., P. J. Barile, M. M. Littler, and D. S. Littler. 2005b. Macro-
algal Blooms on Southeast Florida Coral Reefs: II. Cross-Shelf Dis-
crimination of Nitrogen Sources Indicates Widespread Assimilation
of Sewage Nitrogen. Harmful Algae, 4(6):1106-1122.

Lapointe, B. E., P. J. Barile, C. S. Yentsch, M. M. Littler, D. S. Littler,
and B. Kakuk. 2004. The Relative Importance of Nutrient Enrich-
ment and Herbivory on Macroalgal Communities near Norman’s
Pond Cay, Exumas Cays, Bahamas: A “Natural” Enrichment Ex-
periment. Journal of Experimental Marine Biology and Ecology,
298:275-301.

Lapointe, B. E., M. M. Littler, and D. S. Littler. 1993. Modification of
Benthic Community Structure by Natural Eutrophication: The Be-
lize Barrier Reef. Proceedings of the Seventh International Coral
Reef Symposium, 1:323-334.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Lapointe, B. E., M. M. Littler, and D. S. Littler. 1997. Macroalgal Over-
growth of Fringing Coral Reefs at Discovery Bay, Jamaica: Bottom- '
up Versus Top-down Control. Proceedings of the Eighth Interna-
tional Coral Reef Symposium, 1:927-932.

Lapointe, B. E., and W. R. Matzie. 1996. Effects of Stormwater Nutrient
Discharges on Eutrophication Processes in Nearshore Waters of the
Florida Keys. Estuaries, 19:422-435.

Lapointe, B. E., W. R. Matzie, and M. W. Clark. 1994. Phosphorus In-
puts and Eutrophication on the Florida Reef Tract. In Proceedings
of the Colloquium on Global Aspects of Coral Reefs: Health, Haz-
ards and History, compiler R. N. Ginsburg, pp. 106-112. Miami,
Fla.: University of Miami.

Lapointe, B. E., and J. D. O’Connell. 1989. Nutrient-Enhanced Produc-
tivity of Cladophora prolifera in Harrington Sound, Bermuda: Eu-
trophication of a Confined Phosphorus-Limited Marine Ecosystem.
Estuarine, Coastal and Shelf Science, 28:347-360.

Lapointe, B. E., and K. R. Tenore. 1981. Experimental Outdoor Studies
with Ulva fasciata Delile. J. Interaction of Light and Nitrogen on
Nutrient Uptake, Growth, and Biochemical Composition. Journal
of Experimental Marine Biology and Ecology, 53:135-152.

Larkum, A. W. D., and K. Koop. 1997. ENCORE, Algal Productivity
and Possible Paradigm Shifts. Proceedings of the Eighth Interna-
tional Coral Reef Symposium, 1:881-884.

Lesser, M. P. 2004. Experimental Biology of Coral Reef Ecosystems. Jour-
nal of Experimental Marine Biology and Ecology, 300:217-252.

Lewis, S. M. 1986. The Role of Herbivorous Fishes in the Organiza-
tion of a Caribbean Reef Community. Ecological Monographs,
56:183-200.

Lewis, S. M., and P. C. Wainwright. 1985. Herbivore Abundance and
Grazing Intensity on a Caribbean Coral Reef. Journal of Experi-
mental Marine Biology and Ecology, 87:215-228.

Lirman, D. 2001. Competition Between Macroalgae and Corals: Effects
of Herbivore Exclusion and Increased Algal Biomass on Coral Sur-
vivorship and Growth. Coral Reefs, 19:392-399.

Lirman, D., and P. Biber. 2000. Seasonal Dynamics of Macroalgal Com-
munities of the Northern Florida Reef Tract. Botanica Marina,
43:305-314.

Littler, M. M. 1971. Standing Stock Measurements of Crustose Coralline
Algae (Rhodophyta) and Other Saxicolous Organisms. Journal of
Experimental Marine Biology and Ecology, 6:91-99.

. 1972. The Crustose Corallinaceae. Oceanography and Marine

Biology, Annual Review, 10:311-347.

. 1973. The Population and Community Structure of Hawaiian
Fringing-Reef Crustose Corallinaceae (Rhodophyta, Cryptonemiales).
Journal of Experimental Marine Biology and Ecology, 11:103-120.

Littler, M. M., and D. S. Littler. 1984. Relationships Between Macroalgal
Functional Form Groups and Substrata Stability in a Subtropical
Rocky-Intertidal System. Journal of Experimental Marine Biology
and Ecology, 74:13-34.

. 1990. Productivity and Nutrient Relationships in Psammophytic

Versus Epilithic Forms of Bryopsidales (Chlorophyta): Compari-

sons Based on a Short-Term Physiological Assay. Hydrobiologia,

204/205:49-S5.

. 1997. Disease-Induced Mass Mortality of Crustose Coralline

Algae on Coral Reefs Provides Rationale for the Conservation of

Herbivorous Fish Stocks. Proceedings of the Eighth International

Coral Reef Symposium, 1:719-724.

. 2006. Assessment of Coral Reefs Using Herbivory/Nutrient As-
says and Indicator Groups of Benthic Primary Producers: A Critical
Synthesis, Proposed Protocols, and Critique of Management Strat-
egies. Aquatic Conservation: Marine and Freshwater Ecosystems,
17:195-215.

Littler, M. M., D. S. Littler, and B. L. Brooks. 2006a. Harmful Algae
on Tropical Coral Reefs: Bottom-up Eutrophication and Top-down
Herbivory. Harmful Algae, 5:1-23.

Littler, M. M., D. S. Littler, B. L. Brooks, and B. E. Lapointe. 2006b.
Nutrient Manipulation Methods for Coral Reef Studies: A Criti-
cal Review and Experimental Field Data. Journal of Experimental
Marine Biology and Ecology, 336:242-253.

Littler, M. M., D. S. Littler, and B. E. Lapointe. 1993. Modification of
Tropical Reef Community Structure due to Cultural Eutrophica-
tion: The Southwest Coast of Martinique. Proceedings of the Sev-
enth International Coral Reef Symposium, 1:335-343.

Littler, M. M., D. S. Littler, and P. R. Taylor. 1983a. Evolutionary Strate-
gies in a Tropical Barrier Reef System: Functional-Form Groups of
Marine Macroalgae. Journal of Phycology, 19:229-237.

Littler, M. M., D. S. Littler, and E. A. Titlyanov. 1991. Comparisons of
N- and P-Limited Productivity between High Granitic Islands vs.
Low Carbonate Atolls in the Seychelles Archipelago: A Test of the
Relative-Dominance Paradigm. Coral Reefs, 10:199-209.

Littler, M. M., P. R. Taylor, and D. S. Littler. 1983b. Algal Resistance to
Herbivory on a Caribbean Barrier Reef. Coral Reefs, 2:111-118.

. 1986. Plant Defense Associations in the Marine Environment.

Coral Reefs, 5:63-71.

. 1989. Complex Interactions in the Control of Coral Zonation on
a Caribbean Reef Flat. Oecologia (Berlin), 80:331-340.

Lough, J. M. 1994. Climate Variation and El Nifio-Southern Oscilla-
tion Events on the Great Barrier Reef: 1958-1987. Coral Reefs,
13:181-195.

Macintyre, I. G., and P. W. Glynn. 1990. Upper Limit of El Nifio Killoff.
Coral Reefs, 9:92.

Maragos, J. E., C. Evans, and P. Holthuis. 1985. Reef Corals in Kaneohe
Bay Six Years before and after Termination of Sewage Discharges
(Oahu, Hawaiian Archipelago). Proceedings of the Fifth Interna-
tional Coral Reef Symposium, Tahiti, 4:189-194.

Marubini, F., and P. S. Davies. 1996. Nitrate Increases Zooxanthellae
Population Density and Reduces Skeletogenesis in Corals. Marine
Biology, 127:319-328.

McClanahan, T., B. A. Cocos, and E. Sala. 2002. Algal Growth and Spe-
cies Composition under Experimental Control of Herbivory, Phos-
phorus and Coral Abundance in Glovers Reef, Belize. Marine Pol-
lution Bulletin, 44:441-451.

McCook, L. J. 1999. Macroalgae, Nutrients and Phase Shifts on Coral
Reefs: Scientific Issues and Management Consequences for the
Great Barrier Reef. Coral Reefs, 18:357-367.

McCook, L. J., J. Jompa, and G. Diaz-Pulido. 2001. Competition be-
tween Corals and Algae on Coral Reefs: A Review of Evidence and
Mechanisms. Coral Reefs, 19:400-417.

McGlathery, K. FE. 1995. Nutrient and Grazing Influences on a Subtropi-
cal Seagrass Community. Marine Ecology Progress Series, 122:
239-252.

Mergener, H. 1981. Man-made Influences on and Natural Changes in the
Settlement of the Aqaba Reefs (Red Sea). Proceedings of the Fourth
International Coral Reef Symposium, 1:193-207.

Miller, M. W., and M. E. Hay. 1996. Coral-Seaweed-Grazer-Nutrient
Interactions on Temperate Reefs. Ecological Monographs, 66:
323-344.

. 1998. Effects of Fish Predation and Seaweed Competition on the
Survival and Growth of Corals. Oecologia (Berlin), 113:231-238.

Miller, M. W., M. E. Hay., S. L. Miller, D. Malone, E. E. Sotka, and
A. Szmant. 1999. Effects of Nutrients versus Herbivores on Reef
Algae: A New Method for Manipulating Nutrients on Coral Reefs.
Limnology and Oceanography, 44:1847-1861.

Mora, C. 2008. A Clear Human Footprint in the Coral Reefs of the
Caribbean. Proceedings of the Royal Society of London, B, 275:
767-773.

Morrisson, D. 1988. Comparing Fish and Urchin Grazing in Shallow and
Deeper Coral Reef Algal Communities. Ecology, 69:1367-1382.

Muller-Parker, G., L. R. McCloskey, O. Hoegh-Guldberg, and P. J.
McAuley. 1994. Effects of Ammonium Enrichment on Animal and

NUMBER 38 e¢ 413

Algal Biomass of the Coral Pocillopora damicornis. Pacific Science,
48:273-283.

Mumby, P. J. 2006. The Impact of Exploiting Grazer (Scaridae) on the Dy-
namics of Caribbean Coral Reefs. Applied Ecology, 16:747-769.

Muscatine, L., P. G. Falkowski, Z. Dubinsky, P. Cook, and L. McClosky.
1989. The Effect of External Nutrient Resources on the Population
Dynamics of Zooxanthellae in a Reef-Building Coral. Proceedings
of the Royal Society of London B, 236:311-324.

Naim, O. 1993. Seasonal Responses of a Fringing Community to Eutro-
phication (Reunion Island, Western Indian Ocean). Marine Ecology
Progress Series, 99:137-151.

Nixon, S. W. 1995. Coastal Marine Eutrophication: A Definition, Social
Causes, and Future Concerns. Ophelia, 41:199-219.

Nugues, M. M., G. W. Smith, R. J. van Hooidonk, M. I. Seabra, and
R. P. M. Bak. 2004. Algal Contact as a Trigger for Coral Disease.
Ecology Letters, 7:919-923.

Ogden, J. C., and P. S. Lobel. 1978. The Role of Herbivorous Fishes
and Urchins in Coral Reef Communities. Environmental Biology
of Fishes, 3:49-63.

Pandolfi, J. M., R. H. Bradbury, E. Sala, T. P. Hughes, K. A. Bjorndal,
R. G. Cooke, D. McArdle, L. McClenachan, M. J. H. Newman, G.
Paredes, R. R. Warner, and J. B. C. Jackson. 2003. Global Trajec-
tories of the Long-Term Decline of Coral Reef Ecosystems. Science,
301:955-958.

Pandolfi, J. M., J. B. C. Jackson, N. Baron, R. H. Bradbury, H. M. Guz-
man, T. P. Hughes, C. V. Kappel, F. Micheli, J. C. Ogden, H. P. Pos-
singham, and E. Sala. 2005. Are U.S. Coral Reefs on the Slippery
Slope to Slime? Science, 307:1725-1726.

Pastorok, R. A., and G. R. Bilyard. 1985. Effects of Sewage Pollution
on Coral-Reef Communities. Marine Ecology Progress Series, 21:
175-189.

Paul, V., M. M. Littler, D. S. Littler, and W. Fenical. 1987. Evidence for
Chemical Defense in the Tropical Green Alga Caulerpa ashmeadii
(Caulerpaceae: Chlorophyta): Isolation of New Bioactive Sesquiter-
penoids. Journal of Chemical Ecology, 13:1171-1185.

Pollock, J. B. 1928. Fringing and Fossil Coral Reefs of Oahu. Bulletin of
the Bernice P. Bishop Museum, 55:1-56.

Precht, W. FE, and S. L. Miller. 2007. Ecological Shifts along the Florida
Reef Tract: The Past as a Key to the Future. In Geological Ap-
proaches to Coral Reef Ecology, ed. R. B. Aronson, pp. 237-312.
New York: Springer-Verlag.

Precht, W. E, S. L. Miller, R. B. Aronson, J. F Bruno, and L. Kaufmann.
2005. Reassessing U.S. Coral Reefs. Science, 308:1741.

Rosenberg, G., and J. Ramus. 1984. Uptake of Inorganic Nitrogen and
Seaweed Surface Area:Volume Ratios. Aquatic Botany, 19:65-72.

Sammarco, P. W. 1980. Diadema and Its Relationship to Coral Spat Mor-
tality: Grazing, Competition and Biological Disturbance. Journal of
Experimental Marine Biology and Ecology, 45:245-272.

Sammarco, P. W., J. S. Levinton, and J. C. Ogden. 1974. Grazing and
Control of Coral Reef Community Structure by Diadema antil-
larum Phillipi (Echinodermata: Echinoidea): A Preliminary Study.
Journal of Marine Research, 32:47-53.

Sand-Jensen, K., and J. Borum. 1991. Interactions among Phytoplank-
ton, Periphyton, and Macrophytes in Temperate Freshwaters and
Estuaries. Aquatic Botany, 41:137-175.

Santavy, D. L., and E. C. Peters. 1997. Microbial Pests: Coral Disease in
the Western Atlantic. Proceedings of the Eighth International Coral
Reef Symposium, 1:607-612.

Schaffelke, B. 1999. Particulate Nutrients as a Novel Nutrient Source
for Tropical Sargassum species. Journal of Phycology, 35:1150-
1157.

Scheffer, M. S., S. Carpenter, J. A. Foley, C. Folke, and B. Walker. 2001.
Catastrophic Shifts in Ecosystems. Nature (London), 43:591-596.

Schmitt, E. F. 1997. The Influence of Herbivorous Fishes on Coral Reef
Communities with Low Sea Urchin Abundance: A Study among

414 e

Reef Community Types and Seasons in the Florida Keys. Ph.D.
diss., University of Miami, Miami, Fla.

Sluka, R. D., and M. W. Miller. 2001. Herbivorous Fish Assemblages and
Herbivory Pressure on Laamu Atoll, Republic of Maldives. Coral
Reefs, 20:255-262.

Small, A., and W. H. Adey. 2001. Reef Corals, Zooxanthellae and Free-
Living Algae: A Microcosm That Demonstrates Synergy between
Calcification and Primary Production. Ecological Engineering,
16:443-457.

Smith, J. E., C. M. Smith, and C. L. Hunter. 2001. An Experimental
Analysis of the Effects of Herbivory and Nutrient Enrichment on
Benthic Community Dynamics on a Hawaiian Reef. Coral Reefs,
19:332-342.

Smith, S. V., W. J. Kimmerer, E. A. Laws, R. E. Brock, and T. W. Walsh.
1981. Kaneohe Bay Sewage Diversion Experiment: Perspectives on
Ecosystem Responses to Nutritional Perturbation. Pacific Science,
35:279-397.

Stambler, N., N. Popper, Z. Dubinsky, and J. Stimson. 1991. Effects of
Nutrient Enrichment and Water Motion on the Coral Pocillopora
damicornis. Pacific Science, 45:299-307.

Steneck, R. S. 1989. Herbivory on Coral Reefs: A Synthesis. Proceedings
of the Sixth International Coral Reef Symposium, Australia, 1988,
1:37-49.

Steneck, R. S., and M. N. Dethier. 1994. A Functional Group Ap-
proach to the Structure of Algal-Dominated Communities. Oikos,
69:476-498.

Stephenson, W., and R. B. Searles. 1960. Experimental Studies on the
Ecology of Intertidal Environments at Heron Island. I. Exclusion
of Fish from Beach Rock. Australian Journal of Marine and Fresh-
water Research, 2:241-267.

Steven, A. D. L., and A. D. Broadbent. 1997. Growth and Metabolic Re-
sponses of Acropora palifera to Long-Term Nutrient Enrichment.
Proceedings of the Eighth International Coral Reef Symposium,
1:867-872.

Szmant, A. M. 1997. Nutrient Effects on Coral Reefs: A Hypothesis on
the Importance of Topographic and Trophic Complexity to Reef
Nutrient Dynamics. Proceedings of the Eighth International Coral
Reef Symposium, 2:1527-1532.

Szmant, A. M. 2001. Introduction to the Special Issue of Coral Reefs
on “Coral Reef Algal Community Dynamics.” Coral Reefs, 19:
299-302.

Tanner, J. E. 1995. Competition between Hard Corals and Macroalgae:
An Experimental Analysis of Growth, Survival, and Reproduc-
tion. Journal of Experimental Marine Biology and Ecology,
190:51-168.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Thacker, R. W., D. W. Ginsburg, and V. J. Paul. 2001. Effects of Herbi-
vore Exclusion and Nutrient Enrichment on Coral Reef Macroalgae }
and Cyanobacteria. Coral Reefs, 19:318-329.

Tomascik, T. 1991. Settlement Patterns of Caribbean Scleractinian Cor-
als on Artificial Substrata along an Eutrophication Gradient, Barba-
dos, West Indies. Marine Ecology Progress Series, 77:261-263.

Tomascik, T., and FE. Sander. 1985. Effects of Eutrophication on Reef-
Building Corals. I. Growth Rate of the Reef-Building Coral Monas-
trea annularis. Marine Biology, 87:143-155.

. 1987. Effects of Eutrophication on Reef-Building Corals: Repro-
duction of the Reef-Building Coral Porites porites. Marine Biology,
94:77-94.

Verheij, E. 1993. Marine Plants of the Spermonde Archipelago, SW Su-
lawesi Indonesia: Aspects of Taxonomy, Floristics, and Ecology.
Leiden: Rijksherbarium

Vitousek, P. M., J. Aber, R. W. Howarth, G. E. Likens, P. A. Matson,
D. W. Schindler, W. H. Schlesinger, and D. G. Tilman. 1997. Hu-
man Alteration of the Global Nitrogen Cycle: Sources and Conse-
quences. Ecological Adaptation, 7:737-750.

Vroom, P. S., K. N. Page, K. A. Peyton, and J. K. Kukea-Shultz. 2005.
Spatial Heterogeneity of Benthic Community Assemblages with an
Emphasis on Reef Algae at French Frigate Shoals, Northwestern
Hawai’ian Islands. Coral Reefs, 24:574-581.

Walker, D.I., and R. F. G. Ormond. 1982. Coral Death from Sewage and
Phosphate Pollution at Aqaba, Red Sea. Marine Pollution Bulletin,
13:21-25S.

Ward, S., and P. L. Harrison. 1997. The Effects of Elevated Nutrient
Levels on Settlement of Coral Larvae During the Encore Experi-
ment, Great Barrier Reef, Australia. I. Proceedings of the Eighth
International Coral Reef Symposium, 1:891-896.

Weiss, M. P., and D. A. Goddard. 1977. Man’s Impact on Coastal Reefs:
An Example from Venezuela. American Association of Petroleum
Geologists, Studies in Geology, 4:111-124.

Wilkinson, C. 1999. Global and Local Threats to Coral Reef Function-
ing and Existence: Review and Predictions. Marine and Freshwater
Research, 50:867-878.

Williams, I. D., N. V. C. Polunin, and V. J. Hendrick. 2001. Limits to
Grazing by Herbivorous Fishes and the Impact of Low Coral Cover
on Macroalgal Abundance on a Coral Reef in Belize. Marine Ecol-
ogy, Progress Series, 222:187-196.

Windom, H. 1992. Contamination of the Marine Environment from
Land-Based Sources. Marine Pollution Bulletin, 25:1-4.

Wittenberg, M., and W. Hunte. 1992. Effects of Eutrophication and Sed-
imentation on Juvenile Corals. I. Abundance, Mortality and Com-
munity Structure. Marine Biology (Berlin), 112:131-138.

Impacts of Human Disturbance on Soil
Erosion Potential and Habitat Stability of
Mangrove-Dominated Islands in the Pelican
Cays and Twin Cays Ranges, Belize

Karen L. McKee and William C. Vervaeke

Karen L. McKee and William C. Vervaeke, U.S.

Geological Survey, National Wetlands Research
Center, 700 Cajundome Boulevard, Lafayette,
Louisiana 70506, USA. Corresponding author:
K. McKee (karen_mckee@usgs.gov). Manuscript
received 9 June 2008; accepted 20 April 2009.

ABSTRACT. The Mesoamerican Barrier Reef System (MBRS) is the longest unbroken
reef in the Western Hemisphere and contains hundreds of mangrove-dominated islands.
These islands provide critical habitat supporting marine biodiversity and create a self-
sustaining system that counterbalances sea-level rise. Undisturbed mangrove islands
build vertically through accumulation of organic matter (peat), which forms a strong,
erosion-resistant matrix. Clear-cutting and dredging activities for development of tourist
resorts, fishing camps, and “improved land” for resale, however, threaten mangrove-
dominated islands and adjacent seagrass and coral reef assemblages. Effects of mangrove
disturbance were examined on four islands in the designated marine preserves of Twin
Cays and the Pelican Cays, Belize. Mangroves were clear cut (1.0-6.2 ha), and marine
sediment was dredged from nearby reef flats and seagrass beds to raise land elevations
to support beach vegetation and buildings. Removal of mangroves and especially addi-
tion of dredged fill significantly altered soil characteristics and decreased shear strength
and aggregate stability of soil surfaces. Deep cores collected at both island ranges also
revealed underlying deposits of peat (1.5-10.8 m thick), which influence local land sub-
sidence. Although infilling with dredged material temporarily raised elevations, the in-
exorable subsidence of peat through natural processes of compaction and decomposition
and sea-level rise will ultimately submerge such areas. Our findings thus show that soil
erosion potential is increased and that long-term stability of islands may be compromised
by mangrove clearing and dredging activities. Degradation of key biophysical compo-
nents and critical habitat will ultimately impact ecotourism activities that depend on a
healthy, natural environment.

INTRODUCTION

Persistence of oceanic mangrove islands is dependent upon maintenance of
soil elevations relative to sea level. Mangrove-dominated islands can counter-
balance rising seas by accumulating organic matter (mangrove-derived peat),
which gradually builds land vertically (McKee et al., 2007a). In addition, biodi-
versity of intertidal and subtidal ecosystems in the Caribbean Region is dependent
upon the presence of mangroves because a number of marine species are exclusively
associated with the mangrove habitat (Ellison and Farnsworth, 1992; Goodbody,
2000; Taylor, 2000; Rocha et al., 2005). Mangroves also serve as nurseries for
many reef fish and other marine organisms (Mumby et al., 2004). Consequently,

416 e

changes in mangrove extent may have a cascading effect
on habitat stability and capacity to keep pace with sea-
level rise, as well as on marine biodiversity, in regions with
mangrove-dominated islands and adjacent seagrass and
coral reef assemblages.

The Mesoamerican Barrier Reef System (MBRS) off
the coast of Belize, Central America, contains hundreds
of mangrove islands in association with extensive seagrass
beds and coral reefs (www.mbrs.org.bz; accessed 11 June
2008). Unfortunately, clear-cutting of insular mangroves
has greatly accelerated in Belize and other locations in
recent years for development of tourist resorts, fishing
camps, and “improved land” for resale (K. L. McKee,
personal observation). Even if such areas are ultimately
abandoned and allowed to recover, recolonization by
mangroves may be extremely slow, if it occurs at all. For
example, a site on Twin Cays, an island range that was
clear cut in 1992, is slowly recovering, but regenerating
mangroves are still sparse and have taken 15 years to
reach sapling size (~1-2 m tall) (McKee et al., 2007b).
In addition to removal of mangroves, bottom sediments
are dredged from adjacent seagrass beds and reef flats and
pumped onto cleared mangrove areas to raise elevations
sufficiently to support beach vegetation and buildings.
This type of disturbance thus destroys multiple ecosys-
tems, which require many years for recovery. In the in-
terim, there may be additional consequences from the loss
of habitat stability and increased erosion.

The specific objectives of this study were to assess the
potential for changes in soil erosion and habitat stability
on mangrove islands subjected to clear-cutting and dredg-
ing activities. The work focused on the designated ma-
rine preserves of Twin Cays and the Pelican Cays ranges,
which have been highlighted as critical habitat for marine
biodiversity in the region (Macintyre and Riitzler, 2000;
Macintyre et al., 2004a).

STUDY SITE

The MBRS is the longest unbroken reef in the Western
Hemisphere and extends 220 km from the southern part
of the Yucatan Peninsula to the Bay Islands of Honduras.
Two main areas where clear-cutting and filling of mangrove
islands has occurred were studied: Twin Cays and the Peli-
can Cays. Twin Cays, which consists of two larger and two
smaller islands, is located in the central part of the barrier
reef system and about 2 km west of the reef crest (Figure 1).
The Pelican Cays archipelago is located 21 km south of Twin
Cays and contains multiple mangrove-dominated islands

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

(Figure 1). Twin Cays and the Pelican Cays ranges have been
a major focus of mangrove research by the Smithsonian In- —
stitution (Macintyre and Riutzler, 2000; Macintyre et al.,
2004a). These mangrove islands are far from the mainland,
and peat cores contain no terrigenous sediment (Cameron
and Palmer, 1995; McKee and Faulkner, 2000; Purdy and
Gischler, 2003; Macintyre et al., 2004b). The only source of
freshwater is rainfall, and the entire landform is intertidal
(mean tide range [neap] = 0.2 m). The vegetation on undis-
turbed islands is dominated by Rhizophora mangle L. (red
mangrove), which is the most common mangrove species
in the Caribbean Region. The area has been impacted by
numerous hurricanes and tropical storms.

Twin Cays occurs in the Central Province of the bar-
rier reef (16°50’N, 88°06’W) (see Figure 1). Mangrove
communities were established at Twin Cays about 8,000
years ago on the Pleistocene surface of the Belize Barrier
Reef Platform when it was flooded during the Holocene
Transgression (Macintyre et al., 2004b; Purdy and Gisch-
ler, 2003). Deep deposits of peat (as much as 11 m thick)
have developed as Twin Cays accreted vertically with
rising sea level (Macintyre et al., 2004b; McKee et al.,
2007a). Cores collected through these deposits indicate
that the primary means of vertical land movement is accu-
mulation of mangrove organic matter (mostly root matter)
(McKee et al., 2007a). At Twin Cays, five areas ranging in
size from 0.1 to 6.2 ha have been cleared and elevations
raised by infilling with dredged material, beginning in the
early 1990s and continuing until the present; small fish-
ing camps consisting of one or more buildings have been
established. Our study targeted the largest site on East Is-
land that was clear cut and filled in 1995.

The Pelican Cays occur in the south central lagoon
of the reef system (16°39.8’N, 88°11.5'W). Here, man-
groves have developed as part of a complex network of
coral ridges that surround deep circular ponds (Macintyre
et al., 2000). There are several mangrove cays, including
Northeast Cay, Fisherman’s Cay, Manatee Cay, Cat Cay,
Ridge Cay, Avicennia Cay, Co Cat Cay, the Bird Cays, and
several unnamed smaller cays. Mangroves established at
the Pelican Cays only within the past 1,000 years (Mac-
intyre et al., 2000; McKee et al., 2007a) but have accumu-
lated deposits of peat as much as 1.5 m thick (McKee and
Faulkner, 2000). Disturbance at the Pelican Cays began in
the mid-1990s with mangrove clearing on Northeast Cay.
Aerial surveys conducted in April 2006 showed disturbed
areas on Northeast Cay, Co Cat Cay, and Ridge Cay, and
a follow-up survey in April 2007 showed new areas of
mangrove clearing and dredging at Manatee Cay and Fish-
erman’s Cay (I. C. Feller, personal communication).

NUMBER 38

Carrie Bow Research Area

with most Cays named revised 8/2004

“Columbus Cay
-; QMosquito Cay 7 ‘00'N
s:oSandfly Cay

HutsonCay A Cross Sie
y\Pansriga Fach Garoutt Cay
Columbus Reef

ty
oy
i,
Belize City ly

Dangriga i

oe 16°55'N
Tobacco Range e 2 g
Gt ere "Tobacco Cay
Guatemala Coco Plum Cay ae i {, J: &
: enn -o'-War f i Pea

88°W \\

: .
.

Ragged Cay ‘8

Blue Ground :

Range a : cs
© o 42South Water Cay
yy Sittee Point iF : “ss Carrie Bow Cay
Wes? oR,

on Curlew Bank

Sapodilla ors 5
Dagoen Stewart me os Patch Reefs bas

” Wee Wee Cay’ & Sand Bores *,

Spruce Cay

a:
Douglas Cay. Ss

Fisherman’s Cay
Pelican Cays
Ridge Cay :
Manatee Cay ae 3

of Riversdale

lonathan Point

c 2, ween” Patch Reefs. S,
@ False Cay 3. eS) 9. . “i Ss & Sand Bores», °o.
Crawl Cay". ° oe wae % % :
False Point rie So shee ice % bi, 16°35'N
Be Gon af Harpum ‘Gay x :
Bakers Rendezvous AG we CERISE tare © ;
i] fs | ety On 5 Cea. @o' *.
5km 3% 2 “085 Gladden Cays *-,

“8 Rendezvous Cay % )
£-.[88°05'W

88°15'W

FIGURE 1. Map of research area. The Twin Cays and Pelican Cays ranges are circled.

417

418 e

MATERIALS AND METHODS
EXPERIMENTAL DESIGN

Sampling of Pelican Cays and Twin Cays was con-
ducted during May 2008. The sampling design was a split-
plot in which the main plot was disturbance type (fixed
effect) and the subplot was spatial position relative to the
shoreline (fixed effect). Two disturbance levels were desig-
nated: undisturbed and disturbed. Undisturbed, reference
areas exhibited no visible signs of human activity. Disturbed
areas were characterized by removal of mangroves by clear-
cutting followed by deposition of dredged marine sediment.
Islands with a disturbed area of this type were identified
initially by aerial photography. Four islands were selected
for soil sampling based on disturbance history and acces-
sibility: one in the Twin Cays range (East Island) and three
in the Pelican Cays range (Manatee, Fisherman’s, and Ridge
Cays). Extent of natural and disturbed areas at each study
location was estimated from satellite imagery (Landsat 7,
May 2008; http://landsat.usgs.gov) and aerial photographs
(April 2006, 2007; I. C. Feller, unpublished) and confirmed
by ground-truthing (May 2008).

At each island, two transects were established per-
pendicular to the shoreline and traversing the island to a
distance of 100 m inland. One transect was located in the
disturbed area and the other in an adjacent undisturbed
forest. Fourteen to sixteen sampling stations were prese-
lected along each transect in a stratified-random design;
that is, two to four stations occurred within each of five
intervals (0-10, 11-30, 31-50, 51-70, and 71-100 m
from the shoreline). In the disturbed areas, some portions
were not clear cut (but may have been buried by dredged
fill) or were clear cut and remained free of dredged fill;
these zones were sampled as well. The reference transect
originated along the same shoreline and was oriented in
the same direction as the disturbed transect, although the
length was not always the same (as a consequence of vari-
ation in island configuration). Each island was considered
to be a replicate block.

At each sampling station, percent cover of herbaceous
vegetation and mangrove canopy was estimated visually.
The following soil variables were measured: bulk density,
texture, shear strength, organic matter content, particle
size distribution, and aggregate stability (as described be-
low). A surface core (30 cm depth) was also collected at
each disturbed island to determine the thickness of dredged
material. Also, deeper cores were collected at one island at
Twin Cays (West Island) and one undisturbed island in the
Pelican Cays (Cat Cay) to determine the stratigraphy and
composition of deposits beneath these islands.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

ANALYSES
Soil Shear Strength

Soil shear strength was determined with a Torvane de-
vice (H-4212 1, Humbolt Manufacturing Company, Dur-
ham Geo-Enterprises), which measures the torque required
to shear or deform the soil (McGinnis, 1997). Soil strength
was measured at the soil surface, and the only selective
criterion was flatness, because the Torvane required a flat
or nearly flat surface for accurate measurements. Five rep-
licate measurements were made at each sampling station
and averaged.

Soil Aggregate Stability

Duplicate soil cores (2 cm diameter X 10 cm long)
were collected at each station. The cores were carefully
extruded onto a board, and the upper 1 cm was severed
with a knife, providing a total soil volume of 3.14 cm? per
sample. One core was used for stability testing and the
other was placed into a Ziplock bag for determination of
soil bulk density and texture (described below). —

Soil aggregate stability was determined based on a
modification of standard methods (Angers and Mehuys,
1993; Herrick et al., 2001) to better assess the substrates
(peat, marine sediment) and the types of erosive forces
(waves, currents) typical of the mangrove habitat. In the
field, cores were placed in collection boxes, which pro-
tected them from disturbance until processing. Because
the soils in this study were naturally moist to saturated,
samples were not dried before measurement. Each core
was transferred to a sieve (#20 mesh, 850 pm) and gen-
tly lowered into a container of water, then scored as to
initial structural integrity. The core was then gently agi-
tated by repeated dipping (five times) in water and again
assessed. Initial stability (based on slaking or disintegra-
tion of the core) was scored as 0 (soil too unstable to
sample), 1 (50% of structural integrity lost upon im-
mersion or less than 10% of soil remained on sieve after
five dips), 2 (10%-25% of soil remained), 3 (25%-75%
of soil remained), or 4 (75%-100% of soil remained).
After initial assessment, 200 mL water was poured over
the sample; the material remaining on the sieve and that
washed through the sieve (including the portion from
the initial assessment) was transferred to separate bags
for drying and weighing. Samples were oven dried at
60°C for 24 h and weighed. The percent by weight of
material retained on the sieve was calculated. These two
measures were designated as Stability Index 1 and 2,
respectively.

Soil Water Content, Bulk Density, and Texture

At the laboratory, the soil was weighed wet, dried at
60°C to constant mass, and reweighed to determine mois-
ture content (percent water in soil sample). Dry bulk den-
sity was calculated by dividing the dry mass by the vol-
ume (g cm 3). The dried soil was ashed at 550°C for 6 h
to determine mineral mass after organic loss on ignition.
Percent organic matter content was calculated as 100 mi-
nus the percent ash. Particle size distribution (PSD) was
determined (only for mineral sediments) based on a mi-
cropipette method (Burt et al., 1993). Subsamples (three
or four) collected within each zone along a transect were
combined to provide sufficient mass for PSD.

Peat Coring

Deep cores were collected to the point of refusal with
a Russian peat corer, which extracts uncompressed cores
in sections 0.5 m long (McKee et al., 2007a). At Twin
Cays, seven cores were collected across a transect travers-
ing West Island (west to east) to depths to 10.8 m. At Cat
Cay, nine cores were similarly collected, but the maximum
depth was 1.5 m because the peat layer was thinner. A
deep core was also collected at Manatee Cay (disturbed
area). Each core was extracted and transferred to a half-
section of polyvinyl chloride (PVC) pipe, wrapped with
plastic wrap, and refrigerated until processing. At the field
station, each core section was logged, photographed, and
thicknesses of major strata measured. Subsamples were
taken at intervals of approximately 10 cm and washed

NUMBER 38 ¢ 419

on a | mm mesh sieve; plant fragments were identified to
species using a key as described previously (McKee and
Faulkner, 2000). At all disturbed sites, shallow cores (5 cm
diameter X 30 cm deep) were collected with a piston corer
to determine the thickness of the dredged fill.

RESULTS
GENERAL OBSERVATIONS

Large areas of Twin Cays, Fisherman’s Cay, Manatee
Cay, and Ridge Cay were clear cut (Table 1), and marine
sediment had been dredged from a nearby reef flat and
pumped to the island interior. Scars on the seafloor were
visible from the air (Figure 2), showing that large areas
(0.6 to 1.0 ha) of reef flat had been disturbed. The man-
grove areas disturbed at these four study sites varied from
1.0 to 6.2 ha, accounting for up to 34% of the mangrove
area per island (Table 1). In most cases, the woody de-
bris from clear-cutting had been burned and only stumps
remained to mark the past presence of mangrove trees.
All four disturbed sites had received varying amounts of
dredged marine sediment that had created a relatively flat,
homogeneous landscape of dry, highly reflective, inorganic
substrate (Figure 3). The dredged material varied in depth
from 12 to 25 cm and contained coral, shells, and sand
that indicated the marine origin of the materials. All the
sites examined at the Pelican Cays were filled with material
dredged from nearby reef flats, as evidenced by the pres-
ence of carbonate sand (Halimeda spp.), shells, and coral
fragments. At Twin Cays, the dredged fill was composed of

TABLE 1. Summary of natural and disturbed mangrove areas at two island ranges: Twin Cays (East
and West) and Pelican Cays (Manatee Cay, Fisherman’s Cay, and Ridge Cay). A dash (-) indicates data

were not obtained.

Twin Cays Pelican Cays

Measurement East Island WestIsland Manatee Cay Fisherman’s Cay Ridge Cay
Total island area (ha) 41.4 17.6 12.6 24.3 3.4
Undisturbed area (ha)

Natural ponds 9.8 2.0 4.3 6.2 0.1

Mangrove 28.7 13.0 6.3 11.9 Me9)
Disturbed area (ha)

Clear-cut only 1.2 1.4 - = -

Clear-cut + filled ilo 1.2 2.0 6.2 1.0
Percent disturbed area

Island 7% 14% 16% 26% 29%

Mangrove only 9% 17% 24% 34% 30%

420 e

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 2. Aerial photograph of Fisherman’s Cay and Manatee Cay in April 2007 showing mangrove areas
that were clear cut and filled with marine sediment dredged from an adjacent reef flat (indicated by red circle

and arrow pointing to white filled area on the cay). (Photograph by I. C. Feller.)

quartz sand, indicating a mainland origin. With the excep-
tion of a narrow fringe of uncut trees and occasionally a
lone tree, no mangrove canopy remained in these disturbed
areas. In some cases, dredged material had buried the aer-
ial roots of intact trees along some shorelines, and these
trees had subsequently died (Figure 4). At all disturbed
sites, mangrove associates (Conocarpus erectus) and com-
mon coastal beach species were present, but total cover
was low (<10%). The most common herbaceous species
included Batis maritima L., Sesuvium portulacastrum (L.)
L., Distichlis spicata (L.) Greene, Paspalum distichum L.,
Salicornia virginica L., Spartina spartinae (Trin.) Merr. ex
A. S. Hitche., Rhabdadenia biflora (Jacq.) Muell.-Arg.,
Cyperus spp., [Ipomoea pescaprae ssp. Brasiliensis (L.) van
Ooststr., Ageratum littorale Gray, and Typha sp. In unveg-
etated areas, a biological crust of unidentified composition
had sometimes formed a thin surface layer on top of the
dredged fill.

In contrast, the reference sites contained intact man-
grove canopy (dominant species = R. mangle with sub-
dominants = Avicennia germinans (L.) L. and Lagun-

cularia racemosa (L.) Gaertn. f.), low herbaceous cover
(<5%; most commonly B. maritima), and undisturbed
substrate that was dark in color, saturated with water, and
composed of live and dead mangrove roots and other or-
ganic matter. Abundant aerial roots of intact mangrove
vegetation (prop roots and pneumatophores) formed an
interlacing network that contributed to the overall struc-
tural integrity of the reference areas and also served as sub-
strate for a variety of epiphytes and epibionts. The forest
floor was usually covered by algal-microbial mats (Rho-
dophyta, Chlorophyta, Cyanophycota, Bacillariophyta),
typical of mangrove forests in this region (K. L. McKee,
unpublished data).

SURFACE SOIL CHARACTERISTICS AND EROSION POTENTIAL

Major differences in surface soil characteristics and
potential for erosion occurred between reference and
disturbed areas (Table 2). The surface soil of reference
forests was peat composed of a matrix of live and dead
mangrove roots, filamentous algal-microbial mats, and

NUMBER 38 ¢ 421

FIGURE 3. Views of disturbed (upper panels, A and B) and reference (lower panels, C and D) areas showing
detail of soil surfaces. A circular quadrat in panel D (0.1 m?) provides scale.

trapped organic matter that retained its structural integ-
rity even when disturbed by sampling. Bulk density was
low (0.17 g cm), and water (68%) and organic contents
(60%) were high. Despite its organic nature, the reference
soil (peat) had high shear strength (overall mean, 0.084
kg cm?), which varied little spatially (Figure 5). Core sam-
ples retained their shape and showed little or no slaking
upon immersion in water (Stability Index 1 = 3.98) and
little loss of material upon repeated agitation (Stability
Index 2 = 87%). When stability indices were plotted as
x-y coordinates, the reference sites grouped together, in-
dicating little difference among islands (Figure 6).

In contrast, the surface soil in disturbed areas was
composed of inorganic carbonate particles derived pri-
marily from calcareous algae (Halimeda spp.), coral frag-
ments, and shells (PSD showed that >90% of the mass
was sand or larger particles). This material had a high bulk
density (0.72 g cm?) and low water content (30%) and
organic content (9%). Soil shear strength (0.044 kg cm?)
in disturbed areas was lower overall compared to refer-
ence areas and varied spatially (see Figure 5). Aggregate
stability was lower overall (Stability Index 1 = 2.14, Sta-

bility Index 2 = 47%), and differed among island loca-
tions (see Figure 6). Many cores were friable and readily
disintegrated when disturbed mechanically. Where vegeta-
tion or biological crusts had developed on the dredged
material, the shear strength was higher, but aggregate sta-
bility remained low; that is, cores typically did not retain
their integrity and exhibited a high degree of slaking in
water. The mass retained on the sieve was composed of
particles greater than 1 mm in diameter (Halimeda chips,
coral fragments, shells). Shear strength increased overall
with increasing distance along some disturbed transects
because of the absence of dredged fill at interior stations
where the old peat surface remained exposed (e.g., Twin
Cays). In such cases, the exposed peat substrate retained
high shear strength and high aggregate stability despite the
removal of the mangroves.

PEAT STRATIGRAPHY
Mangrove islands in the Twin Cays and Pelican Cays

ranges were underlain by deposits of peat, varying in
maximum thickness from 1.5 m (Cat Cay and Manatee

422 e

Live Fringe

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Se >

— Roots smothered by:
~_ dredge material

EGET HEL. Shs SS a RS

Dead Fringe

FIGURE 4. Red mangrove trees along the shoreline of disturbed areas: some trees died as a result of burial

with dredged fill (right).

Cay) to more than 10 m (Twin Cays) (Figure 7). At Cat
Cay, a series of cores traversing the island showed that peat
thickness was greatest on the southern, leeward side and
decreased toward the northern, windward shore. Botanical
matter in the peat consisted predominately of mangrove
roots with fragments of leaves and wood. Beneath the peat

layer was sand and/or coral. Deeper peat layers were domi-
nated by R. mangle, whereas upper layers (0-50 cm) in
the island interior contained remains of A. germinans. At
Twin Cays, cores across an east—west transect were 7.5 to
10.8 m thick, with two of the cores reaching the limestone
platform underlying this range. These cores also consisted

TABLE 2. Summary of analysis of variance (ANOVA) results for soil characteristics. The main plot factor (disturbance) was tested with
within-subject error (island); subplot factor (spatial position) and interactions were tested with residual error. Values are the F ratio;
significance: *P = 0.05, **P = 0.01, ***P = 0.001, ****P = 0.0001, ns = not significant.

Aggregate stability Vegetative cover
Source Shear Bulk Organic Water
of error strength Index 1 Index 2 density matter content Herbaceous Mangrove
Disturbance 50.24% %** 180.0% *** 86.4% % 184.0% *** 353.2% % 350.0% *** 6.78% 191.3%***
Island (block) 623mm TQM 2% E9328 DDFs 5.92"? DAS Tisrr er” TDS os
Position 1.36 4.85** Joa)? il 223} 2 0.68 BS 3.02* 1837/8 1, QQs
Disturbance X position Dorr Seine 1a Gooey 2.82* 8.67**** 1.708 1.59ms

cl = Disturbed
Reference

Shear Strength (kg cm”)

0-10 11-30 31-50 51-70 71-100

Organic Matter (%)

0-10 11-30 31-50 51-70 71-100

NUMBER 38 e¢ 423

Bulk Density (g cm’)

0-10 11-30 31-50 51-70 71-100

Water Content (%)

0-10 11-30 31-50 51-70 71-100

Distance from Shoreline (m)

FIGURE 5S. Spatial variation in soil shear strength (A), organic matter (B), bulk density (C), and water content
(D) in reference (squares) and disturbed (circles) mangrove areas. Values are the mean + SE (note some SE

bars are smaller than the symbols).

of mangrove peat, predominately R. mangle. In the is-
land interior, surface layers of A. germinans peat varied
in thickness from 50 to 100 cm, similar to the pattern ob-
served at Cat Cay.

DISCUSSION

The MBRS is a unique and valuable resource to the
Central American countries of Belize, Honduras, and
Guatemala (www.mbrs.org.bz; accessed 11 June 2008).
Destruction of mangrove islands has rapidly accelerated
in this system as a result of attempts to transform these
sensitive and fragile habitats into environments more at-
tractive to tourists. Because there are few sand-based
islands underlain by shallow limestone platforms and

suitable for development, mangrove-dominated islands
have been targeted for conversion. In addition to the two
ranges included in this study (Pelican Cays, Twin Cays),
other mangrove ranges in the vicinity also have undergone
similar mangrove clearing and filling (e.g., Blue Ground
Range, Tobacco Range, Coco Plum) (K. L. McKee, per-
sonal observation). Survey lines found during this study
indicate plans for further development at many of these
island ranges.

EFFECTS OF MANGROVE REMOVAL
AND DREDGED FILL ON EROSION

Although the direct and indirect effects of mangrove
clear-cutting and marine dredging are many and varied, our
study focused on the specific consequences for erosion and

were encountered during surveys (e.g., at Twin Cays), and
here shear strength was equal to that in reference areas;
these were all areas that had been previously occupied by
R. mangle and had only been altered by removal of trees.
Similarly, some mangrove areas in the Bay Islands of Hon-
duras killed by Hurricane Mitch retained shear strength
up to two years following mortality because of the strong
matrix of R. mangle roots forming the peat substrate (Ca-
hoon et al., 2003). However, those areas that had been
dominated by Avicennia germinans lost soil integrity and
collapsed following mortality of the trees. Eventually,
however, the lack of live roots and algal mats may lead
to loss of shear strength wherever mangroves have been

A review of sediment burial effects on mangroves sug-
gests that some species are more sensitive and may suffer

424 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES
Sy =
ji Reference Sites —__
od
=
Ss 3F
= ns
| DE |
are
Nn
l= : :
= ~-—Disturbed Sites removed.
0 | L | | J
0 20 40 60 80 100

Stability Index 2

FIGURE 6. Aggregate stability of soils from reference sites (closed
symbols) and disturbed mangrove areas (open symbols) at Twin
Cays (diamond), Fisherman’s Cay (square), Manatee Cay (triangle),
and Ridge Cay (circle). Stability indices (mean + SE) are based on
maintenance of soil structure upon immersion in water (Index 1) and
percent of soil retained on a 850 «um sieve (Index 2). Probability el-
lipses (90% confidence curves) are drawn for each group.

long-term loss of elevation. Removal of mangroves and
alteration of the soil surface by dredged fill significantly
altered the potential for erosion and substrate stability at
Twin Cays and the Pelican Cays. The natural substrate in
undisturbed mangroves comprised a strong matrix com-
posed of living and dead fibrous roots as well as filamen-
tous algae, which formed mats on the soil surface. This
material was extremely resistant to shearing and retained
its integrity even when repeatedly agitated by submersion
in water (see Figure 6). Although some interior areas con-
tained natural deposits of flocculent material (e.g., micro-
bial mats) that were soft and friable, they were underlain by
solid peat. Work in other locations, such as the Bay Islands
of Honduras, found similarly high resistance of mangrove
peat soils to shearing (McKee and McGinnis, 2002; Cahoon
et al., 2003). These results demonstrate the high resistance
to soil erosion afforded by intact mangrove peat.

Removal of mangroves by clear-cutting did not by itself
appear to have an immediate effect on soil shear strength
or aggregate stability in the areas sampled. In a few cases,
clear-cut areas that were not covered by dredged material

mortality when subjected to excessive rates of sedimen-
tation (Ellison, 1998). We also found that live trees (R.
mangle) exposed to dredged fill often died—presumably
the result of smothering of aerial roots. This outcome was
particularly evident where a narrow (<10 m wide) fringe
of trees was left intact along the shoreline and the dredged
material overflowed into this zone (see Figure 4). In cases
where the shoreline tree zone was wider (20-30 m), there
was a higher survival. Without a protective mangrove buf-
fer along the shoreline, these islands may rapidly erode.
Observations at older sites (Twin Cays) showed rapid
shoreline retreat (up to 0.3 m per year) where mangroves
had been removed in 1992 (McKee et al., 2007b).

LONG-TERM CONSEQUENCES FOR ISLAND STABILITY

Oceanic islands are generally vulnerable to distur-
bance because of their low-lying position and potential for
submergence as well as exposure to tropical storms, hur-
ricanes, and tsunamis that generate strong erosive forces.
Mangrove islands in the MBRS have developed and built
vertically over thousands of years through deposition of
peat derived from mangrove organic matter (McKee and
Faulkner, 2000; McKee et al., 2007a). This process occurs
in the intertidal zone where abundant mangrove roots are
produced and accumulate biomass because of their slow
decomposition in the anaerobic environment (Middle-
ton and McKee, 2001). Other biogenic processes include
formation of algal and microbial mats on the soil sur-
face (intertidal and subtidal) and carbonate sand formed
from calcareous algae (subtidal) in mangrove ecosystems
(McKee et al., 2007a; McKee, unpublished data). Growth
of mangrove root-algal mats and other biofilms not only
contributes to vertical accretion but also stabilizes islands

NUMBER 38 e¢ 425

Twin Cays

Tilt

BH Avicennia® f
Rhizophora
Cal YBP aa ;
cal = es mat Reconstniction oF
= an ; Vegetation History
Depth (m) ] Pleistocene limestone
Cat Cay

~520 foo} fos

0.9
ene pal [| o Se |

fal i BH Avicennia® ‘

i Rhizophora > 2

21 t Sand ie
Depth (m) O Coral

Cal YBP

Reconstruction of
Vegetation History

FIGURE 7. Peat stratigraphy across Twin Cays (top) and Cat Cay (Pelican Cays) (bottom) and recon-
struction of vegetation history. Radiocarbon dates (calendar years before present [Cal YBP]) are based
on previous work (McKee et al., 2007a).

426 «e

by trapping and consolidating organic and inorganic sedi-
ment that is deposited.

To persist, such oceanic islands must accrete verti-
cally to counterbalance both sea-level rise and local rates
of subsidence, which vary depending on geomorphology,
isostacy, and tectonic movements. Mangrove-dominated
islands have the capacity to self-adjust to subsidence and
sea-level rise through peat formation. Previous work has
shown that mangrove islands in Belize and other Caribbean
areas have kept up with changing sea level for thousands
of years through the slow accumulation of mangrove roots
and other organic material and that vertical building rates
are determined by the health and productivity of the man-
grove community (McKee and Faulkner, 2000; Middleton
and McKee, 2001; McKee et al., 2007a). Peat subsidence
rates determined at Twin Cays in vegetated areas averaged
7 mm year ! (McKee et al., 2007a). Similar rates of sub-
sidence were found in the Bay Islands, Honduras (Cahoon
et al., 2003). Island subsidence combined with eustatic rise
in sea level (3.5 mm year !; Rahmstorf, 2007) means that
the relative rise in sea level in this area is at least 10.5 mm
year | (assuming negligible deep subsidence). On undis-
turbed cays with intact mangroves, vertical building from
peat accumulation should maintain surface elevations
within the intertidal zone, unless sea-level rise accelerates
beyond the capacity of the system to compensate.

Removal of mangroves by clear-cutting eliminates the
main mechanism of peat formation and also may alter the
environmental conditions necessary for the survival of al-
gal and microbial mats that contribute to sediment trap-
ping and resistance to erosion. Cays disturbed by man-
grove clearing and dredged fill deposition will continue to
subside, but peat formation will cease. Even if these areas
become revegetated with coastal beach vegetation, peat
cannot form because of oxidizing conditions (caused by the
higher elevations) and lack of the primary peat builder—R.
mangle. Although disturbed island surfaces have been tem-
porarily raised by dredged fill, the inexorable subsidence of
underlying peat and rising seas will lead to submergence.
At current rates of peat subsidence and sea-level rise, the el-
evation gain from dredging will be offset within 20 years.

IMPLICATIONS FOR SUSTAINABLE ECOTOURISM

The Caribbean Region, and in particular the MBRS, is
a major destination for “eco-tourists,” who are attracted
to the tropical climate, clear waters, and abundant marine
life (Uyarra et al., 2005; Diedrich, 2007). A prerequisite
for sustainable ecotourism is maintenance of a pristine
natural environment and protection of all biophysical
components necessary for healthy ecosystems and habitat

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

stability (Casagrandi and Rinaldi, 2002). Unregulated eco-
tourism enterprises threaten the very features that under- —
pin this industry and, in addition, lead to the degradation
of natural resources essential to the livelihood of citizens
(e.g., sport and commercial fisheries) (Burger, 2000; Hall,
2001). Although “charismatic” ecosystems such as coral
reefs receive much attention by conservationists and tour-
ism regulators (Diedrich, 2007), less emphasis is placed
on mangroves. The mangrove destruction occurring in
the Belize reef system likely reflects a general mispercep-
tion that the land beneath mangrove-dominated islands is
stable, as well as a failure to recognize mangroves as es-
sential components contributing to habitat stability and
marine biodiversity—which is what attracts eco-tourists
in the first place (Uyarra et al., 2005).

Our work suggests that the alterations occurring on
mangrove islands in the MBRS are inconsistent with sus-
tainable ecotourism. In their natural state, mangroves
build a peat substrate that is resistant to erosion and
counterbalances subsidence and sea-level rise. In fact, this
dynamic peat-building process has allowed mangrove is-
lands such as Twin Cays to persist for the past 8,000 years
(Macintyre et al., 2004a, 2004b; McKee et al., 2007a).
Attempts to convert mangrove islands to sand islands,
with white beaches and coconut palms, will ultimately
fail because the underlying peat subsidence and rising seas
will eventually prevail. Filling with marine sediment tem-
porarily raises elevations, but without repeated dredging,
eventually these cleared areas will become submerged,
ultimately reducing the total land area of islands in the
MBRS.

Aerial roots of mangroves additionally provide one of
the few natural hard substrates for growth of many marine
organisms in the MBRS (Ellison et al., 1996; Goodbody,
2000; Macintyre et al., 2000) and also create a permeable
barrier that dampens wave energy, decreasing shoreline
erosion (Alongi, 2008). Loss of mangrove fringes directly
decreases the abundance of marine organisms dependent
on mangrove roots for substrate as well as that of reef
species dependent on the mangroves as nurseries (Mumby
et al., 2004). Although the direct and indirect effects of
dredging on reef flats and seagrass beds were not exam-
ined in this study, these effects are likely to be substantial,
given the sensitivity of such systems to disturbance and
sedimentation (Nugues and Roberts, 2003; Erftemeijer
and Lewis, 2006).

Future work should examine the long-term conse-
quences of human activities on the resilience of mangrove
islands to global change and the contribution of man-
groves to terrestrial and marine biodiversity and fishery
productivity. In addition, cost-benefit analyses of man-

grove clearing and dredging and the consequences for
sustainable ecotourism should be conducted to provide
economic rationales for conservation and management of
mangroves and associated habitats.

ACKNOWLEDGMENTS

We thank the government of Belize for permission to
work at our study sites, Klaus Ruetzler (Smithsonian In-
stitution (SI)) for support and permission to work at the
Smithsonian Institution Marine Field Station at Carrie Bow
Cay (CBC) field station, and Mike Carpenter (SI) for logis-
tical arrangements at CBC. We also thank Amy Bunch, Ada
Diz, and Heather Dyer of IAP World Services for assistance
with sample processing and Carole MclIvor (USGS), Candy
Feller (SI), and two anonymous reviewers for comments on
the manuscript. Any use of trade, product, or firm names is
for descriptive purposes only and does not imply endorse-
ment by the U.S. government. This is contribution num-
ber 848 of the Caribbean Coral Reef Ecosystems Program
(CCRE), Smithsonian Institution, supported in part by the
Hunterdon Oceanographic Research Fund.

LITERATURE CITED

Alongi, D. M. 2008. Mangrove Forests: Resilience, Protection from Tsu-
namis, and Responses to Global Climate Change. Estuarine Coastal
and Shelf Science, 76:1-13.

Angers, D. A., and G. R. Mehuys. 1993. “Aggregate Stability to Water.”
In Soil Sampling and Methods of Analysis, ed. M. R. Carter, pp.
651-657. Boca Raton, Fla.: Lewis Publishers.

Burger, J. 2000. Landscapes, Tourism, and Conservation. Science of the
Total Environment, 249:39-49.

Burt, R., T. G. Reinsch, and W. P. Miller. 1993. A Micro-Pipette Method
for Water Dispersible Clay. Communications in Soil Science and
Plant Analysis, 24:2531-2544.

Cahoon, D. R., P. Hensel, J. Rybczyk, K. L. McKee, C. E. Proffitt, and
B.C. Perez. 2003. Mass Tree Mortality Leads to Mangrove Peat
Collapse at Bay Islands, Honduras after Hurricane Mitch. Journal
of Ecology, 91:1093-1105.

Cameron, C. C., and C. A. Palmer. 1995. The Mangrove Peat of the
Tobacco Range Islands, Belize Barrier Reef, Central America. Atoll
Research Bulletin, 431:1-32.

Casagrandi, R., and S. Rinaldi. 2002. A Theoretical Approach to Tour-
ism Sustainability. Conservation Ecology, 6:13.

Diedrich, A. 2007. The Impacts of Tourism on Coral Reef Conservation
Awareness and Support in Coastal Communities in Belize. Coral
Reefs, 26:985-996.

Ellison, J. C. 1998. Impacts of Sediment Burial on Mangroves. Marine
Pollution Bulletin, 37:420-426.

Ellison, A. M., and E. J. Farnsworth. 1992. The Ecology of Belizean
Mangrove-Root Fouling Communities: Patterns of Epibiont Distri-
bution and Abundance, and Effects on Root Growth. Hydrobiolo-
gia, 247:87-98.

Ellison, A. M., E. J. Farnsworth, and R. R. Twilley. 1996. Facultative
Mutualism between Red Mangroves and Root-Fouling Sponges in
Belizean Mangal. Ecology, 77:2431-2444.

NUMBER 38 ¢ 427

Erftemeijer, P. L. A., and R. R. R. Lewis. 2006. Environmental Impacts
of Dredging on Seagrasses: A Review. Marine Pollution Bulletin,
52:1553-1572.

Goodbody, I. 2000. Diversity and Distribution of Ascidians (Tunicata) in
the Pelican Cays, Belize. Atoll Research Bulletin, 480:301-333.

Hall, C. M. 2001. Trends in Ocean and Coastal Tourism: The End of the
Last Frontier? Ocean and Coastal Management, 44:601-618.

Herrick, J. E., W. G. Whitford, A. G. de Soyza, J. W. Van Zee, K. M.
Havstad, C. A. Seybold, and M. Walton. 2001. Field Soil Aggregate
Stability Kit for Soil Quality and Rangeland Health Evaluations.
Catena, 44:27-35.

Macintyre, I. G., I. Goodbody, K. Riitzler, D. S. Littler, and M. M. Lit-
tler. 2000. A General Biological and Geological Survey of the Rims
of Ponds in the Major Islands of the Pelican Cays, Belize. Atoll
Research Bulletin, 467:13-43.

Macintyre, I. G., and K. Ritzler, eds. 2000. Natural History of the Peli-
can Cays, Belize. Atoll Research Bulletin, 467:466-480.

Macintyre, I. G., K. Riitzler, and I. C. Feller, eds. 2004a. The Twin Cays
Mangrove Ecosystem, Belize: Biodiversity, Geological History, and
Two Decades of Change. Atoll Research Bulletin, Nos. 509-530.

Macintyre, I. G., M. A. Toscano, R. G. Lightly, and G. B. Bond. 2004b.
Holocene History of the Mangrove Islands of Twin Cays, Belize,
Central America. Atoll Research Bulletin, 510:1-16.

McGinnis, T. E. I. 1997. Shoreline Movement and Soil Strength in a Lou-
isiana Coastal Marsh. Lafayette, La.: University of Southwestern
Louisiana.

McKee, K. L., D. R. Cahoon, and I. C. Feller. 2007a. Caribbean Man-
groves Adjust to Rising Sea Level through Biotic Controls on
Change in Soil Elevation. Global Ecology and Biogeography,
16:545-556.

McKee, K. L., and P. L. Faulkner. 2000. Mangrove Peat Analysis and
Reconstruction of Vegetation History at the Pelican Cays, Belize.
Atoll Research Bulletin, 468:46-58.

McKee, K. L., and T. E. McGinnis. 2002. Hurricane Mitch: Effects
on Mangrove Soil Characteristics and Root Contributions to Soil
Stabilization. USGS Open File Report OFR-02-178. Lafayette,
La.: United States Geological Survey National Wetlands Research
Center.

McKee, K. L., J. E. Rooth, and I. C. Feller. 2007b. Mangrove Recruit-
ment after Forest Disturbance is Facilitated by Herbaceous Species
in the Caribbean. Ecological Applications, 17:1678-1693.

Middleton, B. A., and K. L. McKee. 2001. Degradation of Mangrove
Tissues and Implications for Peat Formation in Belizean Island For-
ests. Journal of Ecology, 89:818-828.

Mumby, P. J., A. J. Edwards, J. E. Arias-Gonzalez, K. C. Lindeman, P. G.
Blackwell, A. Gall, M. I. Gorezynska, A. R. Harborne, C.L. Pescod,
H. Renken, C. C. Wabnitz, and G. Llewellyn. 2004. Mangroves
Enhance the Biomass of Coral Reef Fish Communities in the Carib-
bean. Nature (London), 427:533-536.

Nugues, M. M., and C. M. Roberts. 2003. Coral Mortality and In-
teraction with Algae in Relation to Sedimentation. Coral Reefs,
22:507-516.

Purdy, E. G., and E. Gischler. 2003. The Belize Margin Revisited: 1.
Holocene Marine Facies. International Journal of Earth Sciences,
92:532-551.

Rahmstorf, $. 2007. A Semi-Empirical Approach to Projecting Future
Sea-Level Rise. Science, 315:368-370.

Rocha, R. M., S. B. Faria, and T. R. Moreno. 2005. Ascidians from Bo-
cas del Toro, Panama. I. Biodiversity. Caribbean Journal of Science,
41:600-612.

Taylor, D. S. 2000. Biology and Ecology of Rivulus marmoratus: New
Insights and a Review. Biological Sciences, 63:242-255.

Uyarra, M. C., I. M. Coté, J. A. Gill, R. T. Tinch, D. Viner, and A. R.
Watkinson. 2005. Island-Specific Preferences of Tourists for Envi-
ronmental Features: Implications of Climate Change for Tourism-
Dependent States. Environmental Conservation, 32:11-19.

: . Sie a :
< = “ 2 ; + a rr
- < es _ 4°"
= x ; as
Pay
ke = P Crees sae otf Sig sawn
| b Kepeacel
| S.ice an cokiomeala
aig a ocak
: = ; 2 Tan mae ; ,
; a el CSOs 0
. ee re betes <wrigt
‘ pa. ny Ani el
{ = e a ? 1 owe ahem “Gs erik
‘@ iv ow Sie CASA cB
| any SEaseAL bat,
a oe epee <
5 Aiiite ’
a-
i - ;
¢: ae
; ipa?
Jide a> © ms
; ie
ih Paes les ae

i one pe 4 :

3 ete — pen inh wealac ai

ay a a
ys = ras

An Overview of Symbiont Bleaching
in the Epiphytic Foraminiferan
Sorites dominicensis

Susan L. Richardson

Susan L. Richardson, Smithsonian Marine Sta-
tion at Fort Pierce, 701 Seaway Drive, Fort
Pierce, Florida 34949, USA, and Harriet L. Wil-
kes Honors College, Florida Atlantic University,
5353 Parkside Drive, Jupiter, Florida 33458, USA
(richards@fau.edu). Manuscript received 15 Au-
gust 2008; accepted 20 April 2009.

ABSTRACT. Populations of Sorites dominicensis, an epiphytic foraminiferan that pos-
sesses dinoflagellate endosymbionts (Symbiodinium), were sampled from seagrass mead-
ows located in Florida and Belize and surveyed for evidence of bleaching. Symbiont
bleaching was first documented in S$. dominicensis populations in the Indian River La-
goon, Florida, in August 2003. Subsequent surveys indicated high rates of bleaching in
August 2004, followed by a near eradication of the epiphytic foraminiferal population as
a result of the 2004-2005 hurricane seasons. Two contrasting sites in Belize, seagrass beds
on the reef flat at Carrie Bow Cay and in Boston Bay, Twin Cays, were surveyed in 2005
and 2006. High rates of bleaching characterize the S. dominicensis populations living on
turtle grass on the reef flat off Carrie Bow Cay, although freshwater runoff from summer
storms during the rainy season may trigger localized bleaching events. Moderate rates of
bleaching were also observed in S. dominicensis populations in Florida Bay in July 2007.
Symbiont bleaching in S. dominicensis appears to be triggered by multiple environmental
factors: increased water temperatures, high levels of irradiance, and influx of freshwater
during storm events. Seasonal summer bleaching events may leave already compromised
S. dominicensis populations vulnerable to periodic disturbance by hurricanes.

INTRODUCTION

Sorites dominicensis Ehrenberg, 1839, is one of several living foraminiferal
species that are host to algal endosymbionts (Hallock, 1999; Lee et al., 1979).
Benthic foraminiferans with algal symbionts occur in several different clades
(Soritacea, Alveolinacea, Nummulitacea, Calcarinidae, and Amphisteginidae)
and are widely distributed in shallow-water, tropical to subtropical reef-associated
marine ecosystems (Langer and Hottinger, 2000). As a group, foraminiferans
host a diverse array of endosymbionts, most of which are microbial eukary-
otic taxa, including stramenopiles (diatoms and chrysophytes), unicellular rho-
dophytes, unicellular chlorophytes, and alveolates (dinoflagellates) (Lee, 2006;
Hallock, 1999). Cyanobacterial endosymbionts have also been isolated from
two different soritid taxa collected from the Red Sea and the Great Barrier Reef
(Lee, 2006). Foraminiferans with photosymbionts possess enhanced calcifica-
tion rates, as well as endogenous sources of nutrition (algal photosynthates)
that allow them to allocate more of their energy resources to cell growth and

430 e

maintenance (Lee, 2006; Hallock, 1999; Duguay, 1983;
Kremer et al., 1980; Lee and Bock, 1976). The algal endo-
symbionts presumably benefit from the mutualism as well,
gaining access to nutrients that are scarce in oligotrophic
environments and to refuge from predation (Lee, 2006;
Hallock, 1999).

The mutalistic association of Sorites and other taxa in
the more inclusive foraminiferal clade Soritida, with dino-
flagellate endosymbionts in the Symbiodinium clade, is
of particular interest to the marine biological community
because this clade comprises the zooxanthellae in stony
corals, soft corals, gorgonians, anemones, jellyfish, bivalve
mollusks, nudibranchs, sponges, and ciliates (Baker, 2003;
Douglas, 2003; Glynn, 1996). Originally considered to be
a single pandemic species that was symbiotic with a broad
range of marine taxa, Symbiodinium microadriaticum is
now known to be part of a more inclusive and genetically
diverse clade composed of eight major subclades, identified
by the letters A-H (Pochon and Pawlowski, 2006; Coffroth
and Santos, 2005; Baker, 2003; Rowan, 1998; Rowan and
Powers, 1991, 1992). Symbiodinium symbionts from fo-
raminiferal hosts are found in clades C, D, EK, G, and H,
with clades F and H being composed almost exclusively of
Symbiodinium isolated from soritid foraminiferans (Garcia-
Cuetos et al., 2005; Pochon and Pawlowski, 2006; Paw-
lowski et al., 2001; Pochon et al., 2001, 2004, 2006; Ro-
driguez-Lanetty, 2003). Although there is relatively high
specificity between Symbiodinium clades F, G, and H and
Foraminifera, there appears to be very little congruence
between host and symbiont phylogenies, indicating that
coevolution has not taken place, at least not at the taxo-
nomic levels sampled to date (Garcia-Cuetos et al., 2005;
Pochon and Pawlowski, 2006; Pawlowski et al., 2001; Po-
chon et al., 2001, 2004, 2006). Although DNA sequences
have not yet been obtained from the endosymbionts of ei-
ther the Belizean or Indian River Lagoon populations of
Sorites dominicensis, Symbiodinium sequences from Flor-
ida Keys specimens fall within either clade F (subclade F4)
or H (Garcia-Cuetos et al., 2005; Pochon and Pawlowski,
2006; Pochon et al., 2006). In all phylogenies published
to date, clade H, the dominant phylotype isolated from
the Florida Keys, branches as the sister group to clade C,
a clade that is widely distributed in the Indo-Pacific, and
exhibits more sensitivity to bleaching than the other Sym-
biodinium clades (Garcia-Cuetos et al., 2005; Pochon and
Pawlowski, 2006; Pawlowski et al. 2001; Pochon et al.,
2001, 2004, 2006; Rowan, 1998, 2004).

The morphological characteristics of Symbiodinium
symbionts isolated in culture from specimens of Sorites
dominicensis collected from the Florida Keys have been

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

described by Lee et al. (1979, 1997). Symbionts are dis-
tributed throughout the foraminiferal cytoplasm, with the
highest densities occurring in the intermediate chambers
and the lowest densities occurring in the outer chambers
where the digestive vacuoles are concentrated (Richard-
son, 2006; Miiller-Merz and Lee, 1976). Similar to other
species of foraminiferans, S. dominicensis is multinucle-
ate and possesses two different types of nuclei: generative
nuclei that participate in reproduction only, and vegeta-
tive nuclei that are transcriptionally active and coordinate
the day-to-day activities of the cell (Miller-Merz and Lee,
1976). In S. dominicensis, the generative nuclei are local-
ized in the central initial chambers of the test (external
shell), which are the chambers with the lowest densities of
dinoflagellates, whereas the transcriptionally active foram-
iniferal nuclei are distributed throughout the cytoplasm
in regions with the high symbiont densities (Miller-Merz
and Lee, 1976).

Estimates of symbiont population size per cell vary
depending on the methodology employed (Richardson,
2006; Doyle and Doyle, 1940). Doyle and Doyle (1940)
estimated the population of dinoflagellates in a 2-mm
sized individual of S. dominicensis to be approximately
1.6 X 10* using light microscopy. In contrast, confocal
microscopy of a 2-mm sized individual of S. dominicensis
collected from Jupiter Sound yielded an estimated 4 X
10° dinoflagellates, equivalent to a density of 1.27 * 10°
endosymbionts cm? of cytoplasm (Richardson, 2006)
(Figure 1). Hemacytometer estimates of endosymbiont
densities in live individuals of S. dominicensis collected
from Jupiter Sound indicate that symbiont densities range
from 6.1 X 10? to 4.8 X 10° dinoflagellates cm~7, with an
average of 6.5 X 104 dinoflagellates cm~* (1 = 85, 0 = 7.9
x 104, o? = 6.2 X 109) (Ross and Richardson, unpub-
lished data). Endosymbiont populations linearly increase
with test size: the average number of symbionts per foram-
iniferal cell is estimated to be 1,469 (7 = 85, s = 2,919,
s? = 8,523,907) for an individual with a test diameter of
1.42 mm (z = 85, s = 0.62, s? = 0.39) (Ross and Rich-
ardson, unpublished data).

Live individuals possess a dark yellowish-brown col-
oration to their cytoplasm as a result of the dense popula-
tions of Symbiodinium in each cell (Figure 2). In healthy
individuals, the coloration is evenly distributed through-
out the test, except for the outer chambers, which appear
colorless because of the low density or absence of endo-
symbionts from the zone of cytoplasm where digestion
takes place (Figure 2). The distinctive coloration of the fo-
raminiferal cytoplasm makes it easy to recognize bleached
or mottled individuals, as described below.

FIGURE 1. Confocal image of live individual of Sorites domini-
censis from Jupiter Sound, Florida. The foraminiferal test is sub-

divided into hexagonal chamberlets. The dinoflagellate endosym-
bionts are most densely packed into the intermediate chambers.
Scale bar = 200 um.

FIELD OBSERVATIONS OF BLEACHING
IN SORITES DOMINICENSIS

Symbiont bleaching has been observed in field surveys
of epiphytic foraminiferal populations from Florida (In-
dian River Lagoon and Long Key, Florida Keys) and Be-
lize (Carrie Bow Cay and Twin Cays). Bleaching in Sorites
dominicensis was first documented in epiphytic popula-
tions attached to Thalassia testudinum (turtle grass) grow-
ing in Jupiter Sound in August 2003 and August 2004,
followed by field surveys of populations in Belize in July
2005 and July 2006. Bleaching was also observed in epi-
phytic populations of S. dominicensis surveyed from the
Florida Keys in July 2007 (Richardson, unpublished data).
Although each of the collecting sites studied hosts seagrass
meadows dominated by T: testudinum, each locality is
subject to different physical factors (salinity, temperature,
water clarity, and subaerial exposure), as well as differing
levels of anthropogenic impact. Detailed descriptions of
the field sites in Florida and Belize are given by Richardson
(2006). Although experimental studies of bleaching in S.
dominicensis have yet to be carried out, field observations

NUMBER 38 e¢ 431

indicate that symbiont bleaching may occur in response to
a number of environmental stressors, including increased
water temperature, freshwater influx, subaerial exposure
during extreme low tides, and periodic disturbance by
hurricanes.

FIELD METHODS

Only epiphytic specimens of the foraminiferan Sorites
dominicensis that were attached to blades of the seagrass
Thalassia testudinum were examined in the studies described
below. Blades of T: testudinum were harvested by wading or
snorkeling. Seagrass leaves were removed at the base of the
blade, submerged in seawater in a Ziploc bag, and stored ina
cooler until return from the field. Both sides of each seagrass
blade were examined for the presence of epiphytic forami-
niferans using a binocular dissecting microscope (Leica MS).
All specimens of the species $. dominicensis were removed
from the blade using a fine paintbrush or dental pick, mea-
sured, and stored on cardboard microslides for additional
study and reference material. The cytoplasmic condition
(healthy, pale, mottled, totally bleached) and reproductive
state (nonreproductive, presence of brood chambers, pres-
ence of embryos in brood chambers, or postreproductive)

FIGURE 2. Live specimens of Sorites dominicensis from Belize, Cen-
tral America. The two individuals in the upper part of the image
show patches of bleached cytoplasm. Note that all specimens, except
for the individual in the lower left, possess few, if any, endosymbi-
onts in the outer two or three chambers. The specimen in the lower
left is a reproductive individual preparing to undergo multiple fis-

sion. The specimen on the upper right is approximately 2 mm in
diameter.

432 e

of each specimen were noted. Specimens were measured
using an optical micrometer calibrated to a stage microm-
eter. Micrographs of representative individuals (healthy,
mottled, and bleached) were taken using a Nikon Coolpix
camera with an MCool (Martin Optics) phototube.

Live individuals were recognized by their distinctive
cytoplasmic coloration as described below, and/or by the
presence of pseudopodial arrays emanating from around the
periphery of the protist’s test. Bundles of bifurcating pseu-
dopodia in live individuals are usually covered with a light
dusting of fine-grained sediment, giving the specimens a star-
shaped appearance. Individuals were recorded as having
healthy cytoplasm if the cytoplasm possessed an evenly dis-
tributed, yellowish-brown coloration (see Figure 2). Individ-
uals were recorded as having a mottled cytoplasm if the cy-
toplasm contained white-colored patches, interspersed with
yellowish-brown sections of cytoplasm (Figure 2). Mottled
individuals contained patches of white cytoplasm that were
visible on both sides of the disk-shaped test. Specimens were
recorded as being totally bleached if the test was completely
white. The tests of postreproductive individuals, that is, in-
dividuals that had undergone reproduction by multiple fis-
sion, were not included in the tallies of bleached specimens.
Postreproductive tests are easily distinguished from bleached
tests by the presence of fragmented brood chambers, undis-
seminated embryos, and clusters of dispersed juveniles in
close proximity to the parental test. It is assumed that few, if
any, of the totally bleached tests had undergone gametogen-
esis, as microspheric tests (tests formed by syngamy) have
never been observed in any of the populations of this species
surveyed by the author.

WATER TEMPERATURE AND BLEACHING

Studies conducted at both the Jupiter Sound and Belize
sites indicate that elevated water temperature, or a combina-
tion of elevated water temperature and subaerial exposure,
can induce symbiont bleaching in S. dominicensis. Bleach-
ing was first observed in the Jupiter Sound populations in
2003 during August (Table 1), when water temperatures
are typically at their maximum, often reaching extremes as
high as 31°C (RiverKeeper Data, Loxahatchee River Dis-
trict). A relatively low abundance of bleached individuals
was recorded in late July 2004; however, a resampling of
the site a few weeks later in August indicated that the in-
cidence of bleaching had risen 14 fold (Table 2). In July
2004, water temperatures recorded at the Jupiter Sound site
ranged from 30° to 31°C between 1:00 pm and 3:30 pm dur-
ing an extremely low spring tide that resulted in the subaer-

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

TABLE 1. Relative abundance of bleached individuals of Sorites —
dominicensis from Jupiter Sound, Florida, during August 2003
(n = total number of tests examined).

Percent of tests

2 Aug 2003 12 Aug 2003
Test condition (x = 580) (n = 147)
Mottled cytoplasm 1.0% 1.0%
White cytoplasm 15% 12%
Total bleached 16% 13%

ial exposure of major portions of the seagrass bed. No water
temperature data are available for Jupiter Sound in August
2004, although the water was uncomfortably hot to the
touch at the time of collection (Richardson, unpublished).
Bleaching was undetectable in surveys of the S. dominicensis
populations conducted at other times of the year in both
2003 and 2004 (Richardson, unpublished data).

In Belize, water temperatures were recorded using
HOBO Tidbit (Onset) submersible temperature loggers
deployed for three days in July 2005. One logger was de-
ployed on the reef flat at Carrie Bow Cay and the other in
Boston Bay, Twin Cays. The range of water temperatures
recorded for both sites are listed in Table 3 and Figure 3.
Although the overall mean temperatures were identical for
both sites (s = 32°C), the reef flat off Carrie Bow Cay
experienced a wider range of temperatures (29°—40°C),
with higher maximum temperatures recorded during the
late afternoon and lower minimum temperatures recorded
at night (Figure 3; Table 3). Correspondingly, the rate of
bleaching recorded from the reef flat at Carrie Bow Cay
was almost five times higher than that observed in Bos-

a aaa SET Se FRE ES
TABLE 2. Relative abundance of bleached individuals of Sorites

dominicensis from Jupiter Sound, Florida during July and Au-
gust 2004 (7 = total number of tests examined).

Percent of tests

29 Jul 2004 19 Aug 2004
Test condition (n = 446) (n = 14)
Mottled cytoplasm 2.0% 29%
White cytoplasm 0% 0%
Total bleached 2.0% 29%

NUMBER 38 ¢ 433

TABLE 3. Characteristics of two collecting sites in Belize.

Characteristic Carrie Bow Cay Twin Cays

Water depth <0.5m 1.0m

Exposure Exposed during low tides Subtidal

Water clarity Very clear High tannins and mangrove detritus
Water movement Swift current Sheltered with slower current
Temperature range (1—4 July 2005) 30°-35°C 29°-40°C

ton Bay, Twin Cays (Table 4). Although the water tem-
peratures recorded in Boston Bay, Twin Cays, were not
as extreme as those recorded off Carrie Bow Cay, they
still were higher than the HotSpot (28.9°C) and bleaching
(HotSpot + 1°C) thresholds derived by NOAA/NESDIS
for Glovers Reef (Opishinski, 2006). The same sites were
resurveyed in July 2006, and the incidence of bleaching
on the reef flat at Carrie Bow Cay was observed to be 11
times higher than the incidence of bleaching recorded in
Boston Bay, Twin Cays, which exhibited almost negligible
levels of bleaching (Table 5).

40.00

FRESHWATER INFLUX AND BLEACHING

In July and August 2006, continued sampling of the
Carrie Bow Cay and Twin Cays field sites in Belize yielded
results that indicate that symbiont bleaching in S. domi-
nicensis can also be triggered by an influx of freshwater
during storm events. In July 2006, field collections were
suspended during a three-day period of intense rain then
restarted after the storms subsided. After the rainstorms,
the incidence of bleaching recorded at both sites rose in
all three categories (pale cytoplasm, mottled cytoplasm,

36.25
je
O
QU
©
p=
5
=)
©
—
® 32.50
5
fe
—
3)
L
s
28.75
25.00
SSSSSSSSssSsssss ss ss ssSsSsSssgsssgsssgssgssgsgssgss
N+¥TOdDONSONTOHDONTHODONDONTODONYTODONSONYTOBDON
a N a | a
1 July 2005 2July 2008 Date & Time 3 July 2005 4 July 2005

FIGURE 3. Water temperature variations on the reef flat at Carrie Bow Cay and in Boston Bay, Twin Cays,

Belize, as measured at noon and every 2.5 hours thereafter during 14 July 2005.

434 e

TABLE 4. Relative abundance of bleached individuals of Sorites
dominicensis from two localities in Belize during July 2005 (x =
total number of tests examined).

Percent of tests

Carrie Bow Cay _ Boston Bay, Twin Cays

Test condition (1 = 797) (n = 685)
Mottled cytoplasm 4.3% 2.5%
White cytoplasm 14% 1.5%
Total bleached 19% 3.9%

TABLE 5. Relative abundance of bleached individuals of Sorites
dominicensis from two localities in Belize during July 2006 (n =
total number of tests examined). All specimens were collected
before a three-day period of intense rain.

Percent of tests

Carrie Bow Cay, Boston Bay, Twin Cays,

21 Jul 2006 23 Jul 2006
Test condition (1 = 62) (n = 349)
Pale cytoplasm 0% 0%
Mottled cytoplasm 3.2% 0.29%
White cytoplasm 4.8% 0%
Total bleached 8.1% 0.29%

CS a a Ee a SY TE
TABLE 6. Relative abundance of bleached individuals of Sorites
dominicensis from two localities in Belize during July and Au-
gust 2006 (m = total number of tests examined). All specimens
were collected after a three-day period of intense rain.

Percent of tests

Carrie Bow Cay, Boston Bay, Twin Cays,

1 Aug 2006 27 Jul 2006
Test condition (2 = 132) (n = 369)
Pale cytoplasm 1.5% 0.27%
Mottled cytoplasm 0.76% 2.4%
White cytoplasm 8.3% 16% 4
Total bleached 11% 19%

4 Of 60 individuals, 34 were juveniles from the same brood.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

and white cytoplasm) (Table 6). Although bleaching on
the reef flat at Carrie Bow Cay was slightly higher than ~
the prestorm levels (11% vs. 8.1%), the total poststorm
incidence of bleaching in Boston Bay was observed to be
more than 65 times higher than that observed just a few
days earlier (Table 6). Although the waters in Boston Bay
are normally of open ocean marine salinities, during heavy
rains and slack tides cold, brackish water drains off Hid-
den Lake in the Twin Cays and empties into Boston Bay
through Hidden Creek (Rutzler et al., 2004). Interestingly,
juveniles were disproportionately impacted by the bleach-
ing event: 34 of 60 of the tests with white cytoplasm ap-
peared to be individuals from the same brood (Table 6).

IMPACT OF HURRICANES AND RECOVERY

Seasonal bleaching events cause increased mortality in
S. dominicensis, resulting in compromised populations that
are more sensitive to periodic disturbance by hurricanes.
Monthly surveys in 2001, 2003, and 2004 indicate that
S. dominicensis populations normally plummet in the late
summer, stay low throughout the winter, and eventually
recover and bloom the following spring in late April and
May (Richardson, unpublished data). In September 2004,
the Jupiter Sound site was traversed by two hurricanes,
Jeanne and Frances (Beven, 2005; Lawrence and Cobb,
2005). The Jupiter site was situated in the south eyewall
for both storms, and experienced high winds and storm
surges and extensive freshwater inundation. Dark, cloudy,
turbid water continued to characterize the site for several
months following the hurricanes. Other impacts included
loss of shading because of downed trees and overgrowth
of the seagrass by cyanobacterial blooms. The entire epi-
phytic foraminiferal community at the Jupiter Sound site
was impacted by the 2004 hurricane season (Richardson,
unpublished data). Initially, a dramatic reduction in species
diversity and abundance was observed, with two species
comprising 92% of the community in April and May 2005.
By August 2005 the community had rebounded to 2001
levels of species diversity and density, with the exception
of the apparent local eradication of S. dominicensis (Rich-
ardson, unpublished data). Sorites dominicensis is the only
species at this site to possess photosynthetic endosymbi-
onts and thus is sensitive to the reduced transmission of
light in the water column that resulted from the months of
increased turbidity following the 2004 hurricanes.

In October 2005, Jupiter Sound was impacted by
Hurricane Wilma (Pasch et al., 2006), although this time
the region experienced the high winds of the north eye-

SSE a ae a oe EE TS EE en Sen
TABLE 7. Relative abundance of bleached individuals of Sorites
dominicensis from Jupiter Sound, Florida, 4 April 2008, as de-
termined from examination of 446 tests.

Test condition Percent of tests (1 = 446)

Pale cytoplasm 5.9%
Mottled cytoplasm 7.8%
White cytoplasm 9.8%
Total bleached 24%

wall of the storm. Individuals of S$. dominicensis were not
recovered from the Jupiter Sound site until the summer of
2007 and did not reach their pre-hurricane densities until
April 2008 (Richardson, unpublished). A survey of 446
individuals of S. dominicensis, collected in April 2008,
yielded a high incidence of bleached individuals (24% to-
tal), an unusual event for the spring (Table 7). The trigger
for this event is unknown; the rainfall during this period
was below average as the region was experiencing an ex-
tended seasonal drought. It is also not known whether the
population recovered through the reproduction of relict
populations of S. dominicensis that survived the. hurri-
canes of 2004 and 2005 or whether the site was repopu-
lated through immigrants transported by the Gulf Stream
from the Florida Keys and/or the Caribbean.

DISCUSSION

The results from the field studies described above doc-
ument the occurrence of bleaching in Sorites dominicensis,
a dinoflagellate-bearing foraminiferan, and delineate some
of the environmental stressors that trigger bleaching. As
has been observed in corals, bleaching in epiphytic speci-
mens of S. dominicensis may be triggered by multiple en-
vironmental factors, such as increased irradiance during
subaerial exposure at low tide, increased water tempera-
tures, influx of freshwater runoff during storm events, and
catastrophic disturbance during hurricanes. The symp-
toms of bleaching in S. dominicensis include decrease in
intensity of coloration (pale appearance), the patchy loss
of cytoplasmic coloration (mottled appearance), and the
total loss of cytoplasmic coloration (white tests). Symbi-
ont bleaching in S. dominicensis can be distinguished from
the loss of cytoplasmic coloration that occurs during the
process of reproduction through multiple fission as the
symbiont-rich cytoplasm moves from the central region of
the test to the periphery where the brood chambers and

NUMBER 38 e¢ 435

embryos will form. Studies are currently underway to link
qualitative observations of bleaching in S. dominicensis to
quantitative studies of symbiont density in bleached speci-
mens using staining techniques that differentiate necrotic
or apoptotic algal cells.

The relatively high water temperatures recorded on the
reef flat at Carrie Bow Cay in July 2005 are not unusual
for tropical seagrasses, which may experience annual fluc-
tuations in seawater temperatures ranging from 19.8° to
41°C (Campbell et al., 2006). Unusually high water daily
temperatures (40°-43°C) have been recorded in seagrass
beds growing in shallow water off Papua New Guinea
(Fred Short, University of New Hampshire, personal com-
munication, January 2006). In addition to high tempera-
tures, tropical seagrasses growing in shallow-water pools
in the intertidal zone are subject to desiccation, extremely
high levels of photosynthetically active radiation, and high
levels of ultraviolet radiation (Campbell et al., 2006; Du-
rako and Kunzelman, 2002).

Although the underlying mechanisms of bleaching in
S. dominicensis are unknown, it is hypothesized that sev-
eral of the proposed mechanisms for bleaching in corals
may function in foraminiferans as well, such as reduced
efficiency of photosystem II resulting from increased ir-
radiance (Venn et al., 2008; Smith et al., 2005), and the
production of damaging reactive oxygen species via several
different pathways (Lesser, 2006; Smith et al., 2005).

Soritid foraminiferans have the potential to serve as a
model system for bleaching, the need of which was recently
emphasized by Weis et al. (2008). Not only do S. domini-
censis and other soritids possess Symbiodinium endosym-
bionts that are closely related to the zooxanthellae in corals
and other metazoans, but the small size of §. dominicensis
facilitates investigation of symbiont bleaching in hospite,
using methods such as in situ hybridization, immunofluo-
rescence, and other imaging techniques. Future research
will focus on developing culture methods for S$. domini-
censis and on exploring cytological methods that will fa-
cilitate the visualization of the cell processes underlying the
bleaching response in foraminiferans.

ACKNOWLEDGMENTS

The field research for this study was funded through
grants from the Caribbean Coral Reef Ecosystems (CCRE)
Program, National Museum of Natural History (NMNH),
Washington, D.C., and the Smithsonian Marine Station at
Fort Pierce (SMSFP), Florida. I am grateful to Jon, Kath-
leen, and Scott Moore (Florida Atlantic University, New
College of Florida, and Indian River Charter High School,

436 °

respectively) for their invaluable assistance in the field at
the Jupiter Island site in Florida since 2001. Mike Car-
penter (NMNH) provided assistance in the field in Belize
in 2005. This is contribution number 839 of CCRE, sup-
ported in part by the Hunterdon Oceanographic Research
Fund, and SMSFP contribution number 758.

LITERATURE CITED

Baker, A. C. 2003. Flexibility and Specificity in Coral-Algal Symbiosis:
Diversity, Ecology, and Biogeography of Symbiodinium. Annual
Review of Ecology and Systematics, 34:661-689.

Beven, J. L. 2005. Tropical Cyclone Report. Hurricane Frances: 25
August-8 September 2004. NOAA/National Weather Service, Na-
tional Hurricane Center, Miami, Fla. Available from: http://www
.nhe.noaa.gov/2004/frances.shtml.

Campbell, S. J., L. J. McKenzie, and S. P. Kerville. 2006. Photosynthetic
Responses of Seven Tropical Seagrasses to Elevated Seawater Tem-
perature. Journal of Experimental Marine Biology and Ecology,
330:455-408.

Coffroth, M. A., and S. R. Santos. 2005. Genetic Diversity of Symbiotic
Dinoflagellates in the Genus Symbiodinium. Protist, 156:19-34.

Douglas, A. E. 2003. Coral Bleaching—How and Why? Marine Pollu-
tion Bulletin, 46:385-392.

Doyle, W. L., and M. M. Doyle. 1940. The Structure of Zooxanthellae.
Papers from the Tortugas Laboratory of the Carnegie Institution of
Washington, 32:127-142.

Duguay, L. E. 1983. Comparative Laboratory and Field Studies on Cal-
cification and Carbon Fixation in Foraminiferal-Algal Associations.
Journal of Foraminiferal Research, 13:252-261.

Durako, M. J., and J. I. Kunzelman. 2002. Photosynthetic characteristics
of Thalassia testudinum measured in situ by pulse-amplitude modu-
lated (PAM) fluorometry: methodological and scale-based consider-
ations. Aquatic Botany, 73:173-185.

Garcia-Cuetos, L., X. Pochon, and J. Pawlowski. 2005. Molecular Evi-
dence for Host-Symbiont Specificity in Soritid Foraminifera. Pro-
tist, 156:399-412.

Glynn, P. 1996. Coral Reef Bleaching: Facts, Hypotheses and Implica-
tions. Global Change Biology, 2:495-509.

Hallock, P. 1999. “Symbiont-Bearing Foraminifera.” In Modern Foram-
inifera, ed. B. Sen Gupta, pp. 123-139. Dordrecht and Boston,
Mass.: Kluwer Academic Publishers.

Kremer, B. P., R. Schmaljohann, and R. Rottger. 1980. Features and Nu-
tritional Significance of Photosynthates Produced by Unicellular Al-
gae Symbiotic with Larger Foraminifera. Marine Ecology Progress
Series, 2:225-228.

Langer, M.R., and L. Hottinger. 2000. Biogeography of Selected “Larger”
Foraminifera. Micropaleontology, 46 (Suppl. 1):105-106.

Lawrence, M. B., and H. D. Cobb. 2005. Tropical Cyclone Report. Hur-
ricane Jeanne: 13-28 September 2004. NOAA/National Weather
Service, National Hurricane Center, Miami, Fla. Available from:
http://www.nhc.noaa.gov/2004/jeanne.shtml.

Lee, J. J. 2006. Algal Symbiosis in Larger Foraminiferan. Symbiosis,
42:63-75.

Lee, J. J., and W. Bock. 1976. The Relative Importance of Feeding in Two
Species of Soritid Foraminifera with Algal Symbionts. Bulletin of
Marine Science, 26:530—537.

Lee, J. J., M. E. McEnery, E. G. Kahn, and F. L. Schuster. 1979. Symbio-
sis and the Evolution of Larger Foraminifera. Micropaleontology,
25:118-140.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Lee, J. J., J. Morales, S. Bacus, A. Diamont, P. Hallock, J. Pawlowski,
and J. Thorpe. 1997. Progress in Characterizing the Endosymbi-
otic Dinoflagellates of Soritid Foraminifera, and Related Studies on
Some Stages in the Life Cycle of Marginopora vertebralis. Journal
of Foraminiferal Research, 27:254-263.

Lesser, M. 2006. Oxidative Stress in Marine Environments: Biochem-
istry and Physiological Ecology. Annual Review of Physiology,
68:253-278.

Loxahatchee River District. Research Reports. Available from: http://
www.loxahatcheeriver.org/reports.php (accessed 5 August 2008).

Miiller-Merz, E., and J. J. Lee. 1976. Symbiosis in the Larger Foraminif-
eran Sorites marginalis (with Notes on Archaias spp.). Journal of
Protozoology, 23(3):390-396.

Opishinski, T. 2006. Carrie Bow Cay Environmental Monitoring System.
http://cbc.riocean.com/cbc2005.htm (accessed 14 June 2006).
Pasch, R. J., E. S. Blake, H. D. Cobb, and D. P. Roberts. 2006. Tropical
Cyclone Report. Hurricane Wilma: 15-25 October 2005. NOAA/
National Weather Service, National Hurricane Center, Miami, Fla.
Available from: http://www.nhc.noaa.gov/HAW2/english/history

.shtml#wilma.

Pawlowski, J., M. Holzmann, J. F. Fahrni, X. Pochon, and J. J. Lee. 2001.
Molecular Identification of Algal Endosymbionts in Large Miliolid
Foraminifera: 2. Dinoflagellates. Journal of Eukaryotic Microbiol-
ogy, 48:368-373.

Pochon, X., T. C. LaJeunesse, and J. Pawlowski. 2004. Biogeographic Par-
titioning and Host Specialization among Foraminiferan Dinoflagellate
Symbionts (Symbiodinium; Dinophyta). Marine Biology, 146:17-27.

Pochon, X., J. I. Montoya-Burgos, B. Stadelmann, and J. Pawlowski.
2006. Molecular Phylogeny, Evolutionary Rates, and Divergence
Timing of the Symbiotic Dinoflagellate Genus Symbiodinium.
Molecular Phylogenetics and Evolution, 38:20-30.

Pochon, X., and J. Pawlowski. 2006. Evolution of the Soritids—Symbio-
dinium Symbiosis. Symbiosis, 42:77-88.

Pochon, X., J. Pawlowski, L. Zaninetti, and R. Rowan. 2001. High Ge-
netic Diversity and Relative Specificity among Symbiodinium-like
Endosymbiotic Dinoflagellates in Soritid Foraminiferans. Marine
Biology, 139:1069-1078.

Richardson, S. L. 2006. Endosymbiont-Bleaching in Epiphytic Popula-
tions of Sorites dominicensis. Symbiosis, 42:103-117.

Rodriguez-Lanetty, M. 2003. Evolving Lineages of Symbiodinium-like
Dinoflagellates Based on ITS1 rDNA. Molecular Phylogenetics and
Evolution, 28:152-168.

Rowan, R. 1998. Diversity and Ecology of Zooxanthellae on Coral
Reefs. Journal of Phycology, 34:407-417.

. 2004. Thermal Adaptation in Reef Coral Symbionts. Nature
(London), 430:742-743.

Rowan, R., and D. A. Powers. 1991. A Molecular Genetic Classification
of Zooxanthellae and the Evolution of Animal—Algal Symbioses.
Science, 251:1348-1351.

. 1992. Ribosomal RNA Sequences and the Diversity of Symbiotic
Dinoflagellates (Zooxanthellae). Proceedings of the National Acad-
emy of Sciences of the United States of America, 89:3639-3643.

Rutzler, K., I. Goodbody, M. C. Diaz, I. C. Feller, and I. G. Macintyre.
2004. The Aquatic Environment of Twin Cays, Belize. Atoll Re-
search Bulletin, 512:1-35.

Smith, D. J., D. J. Suggett, and N. R. Baker. 2005. Is Photoinhibition
of Zooxanthellae Photosynthesis the Primary Cause of Thermal
Bleaching in Corals? Global Change Biology, 11:1-11.

Venn, A. A., J. E. Loram, and A. E. Douglas. 2008. Photosynthetic Symbio-
ses in Animals. Journal of Experimental Botany, 59(5):1069-1080.

Weis, V. M., S. K. Davy, O. Hoegh-Guldberg, M. Rodriguez-Lanetty,
and J. R. Pringle. 2008. Cell Biology in Model Systems as the
Key to Understanding Corals. Trends in Ecology and Evolution,
23(7):369-376.

New Perspectives on Ecological Mechanisms
Affecting Coral Recruitment on Reefs

Raphael Ritson- Williams, Suzanne N. Arnold,
Nicole D. Fogarty, Robert S. Steneck,
Mark J. A. Vermeij, and Valerie J. Paul

Raphael Ritson-Williams and Valerie J. Paul,
Smithsonian Marine Station at Fort Pierce, 701
Seaway Drive, Fort Pierce, Florida 34949, USA.
Suzanne N. Arnold and Robert S. Steneck, Uni-
versity of Maine, School of Marine Sciences,
Darling Marine Center, Walpole, Maine 04573,
USA. Nicole D. Fogarty, Department of Biologi-
cal Science, Florida State University, Tallahassee,
Florida 32306-4295, USA. Mark J. A. Vermeij,
CARMABI, Piscaderabaai z/n, Curacao, Nether-
lands Antilles. Corresponding author: R. Ritson-
Williams (williams@si.edu). Manuscript received
9 June 2008; accepted 20 April 2009.

ABSTRACT. Coral mortality has increased in recent decades, making coral recruitment
more important than ever in sustaining coral reef ecosystems and contributing to their
resilience. This review summarizes existing information on ecological factors affecting
scleractinian coral recruitment. Successful recruitment requires the survival of coral off-
spring through sequential life history stages. Larval availability, successful settlement,
and post-settlement survival and growth are all necessary for the addition of new coral
individuals to a reef and ultimately maintenance or recovery of coral reef ecosystems. As
environmental conditions continue to become more hostile to corals on a global scale,
further research on fertilization ecology, connectivity, larval condition, positive and nega-
tive cues influencing substrate selection, and post-settlement ecology will be critical to
our ability to manage these diverse ecosystems for recovery. A better understanding of
the ecological factors influencing coral recruitment is fundamental to coral reef ecology
and management.

INTRODUCTION

Coral reefs are facing unprecedented human impacts and continuing acute
and chronic threats that can impact community structure (Nystrom et al.,
2000). Their ability to resist such changes or to recover from them defines their
“resilience” (sensu Holling, 1973). Unfortunately, coral reef ecosystems can be
resilient in either the more desirable coral-dominated phase or in the less desir-
able algal-dominated phase (Hughes et al., 2005). Although we know much
about what causes undesirable “phase shifts” (Done, 1992; Hughes, 1994;
Pandolfi et al., 2005), we know relatively little about what drives coral com-
munity recovery (Connell, 1997).

Scleractinian corals are uniquely important to coral reef ecosystems as
ecosystem engineers that structure the habitat (Jones et al., 1994, 1997). The
abundance of live coral drives key ecological processes in the wider coral reef
community, such as providing recruitment habitat for reef fish, lobsters, and
sea urchins (Lee, 2006; Mumby and Steneck, 2008). In the past 30 years, the
percent cover of live coral has decreased on a global scale (Gardner et al., 2003;
Bruno and Selig, 2007), raising the question: How can we increase the number
of corals in these ecosystems for recovery?

438 e

Larval settlement (when they first attach to the ben-
thos) and subsequent survival (recruitment) are processes
that can control marine population dynamics (Gaines and
Roughgarden, 1985; Doherty and Fowler, 1994; Palma et
al., 1999). Although corals can reproduce clonally (Fau-
tin, 2003; Baums et al., 2006), recruitment resulting from
sexual reproduction is the primary means of recoloniza-
tion for most species (Connell et al., 1997) and adds ge-
netic variation to coral populations, which may increase
survival of a species. Coral settlement followed by subse-
quent recruit survival and growth maintains coral popula-
tions and is necessary for coral reef recovery. For this cycle
to occur on any given reef, larval survival and recruitment
are dependent on a sequence of three phases: (1) larval
availability, which integrates gamete production, fertil-
ization success, and connectivity; (2) settlement ecology,
which relates to larval condition and substrate selection
behavior; and (3) post-settlement ecology, including sub-
strate-specific survival and growth (Figure 1).

This review summarizes existing information on eco-
logical factors affecting scleractinian corals during these
first three phases of their life, covering the period from

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

gamete release to juvenile coral colonies (typically de-
scribed as <40 mm). We discuss factors that are critical for
coral recruitment success, and where insufficient data exist,
we draw parallels to concepts that have been developed for
other marine larvae or adult corals and briefly discuss their
relevance for the early life history stages of corals.

LARVAL AVAILABILITY

Larval supply to a reef depends on sequential pro-
cesses of gamete production, fertilization success, and lar-
val transport (i.e., larval dispersal and connectivity). Basic
life history traits of corals can greatly influence the range of
strategies that are used to ensure larval availability. Scler-
actinians have two main reproductive modes: brooding,
where sperm are released into the water column and taken
in by conspecifics for internal fertilization, and broadcast
spawning, wherein both egg and sperm are released into
the environment so that fertilization occurs externally, that
is, in the water column (Figure 2; Fadlallah, 1983; Szmant,
1986; Richmond and Hunter, 1990; Richmond, 1997).

Fertilization

occurs

Gamete bundles
break apart

Gamete bundles 4
float to sea C

LARVAL
AVAILABILITY

er:

3-7 years

many cycles per year
few planulae per cycle

ita,/

Juvenile

POSTSETTLEMENT
ECOLOGY

=)

Planula larva
PE te

wp

Recruit

FIGURE 1. Three sequential phases necessary for successful coral recruitment starting with larval availability,
progressing to settlement ecology, and ending with post-settlement ecology. (Drawn by Mark Vermeij.)

NUMBER 38 ¢ 439

FIGURE 2. Different modes of reproduction influence larval supply in coral species. a, Female

Stephanocoenia intersepta, a gonochoric spawner, releases eggs. b, A male S. intersepta releases
sperm. c, The hermaphrodite Montastraea faveolata releases eggs and sperm as bundles that
float to the surface, where they break apart for fertilization. d, For Acropora palmata (and other
spawners), fertilization of coral eggs occurs in the water column. e, A larva of Acropora pal-
mata completes development in the water column. f, In contrast, a larva of Porites astreoides
(a brooder) is fully developed when it is released from its parent and contains zooxanthellae.
(Photographs a, b, by Mark Vermeij; c, e, f, by Raphael Ritson-Williams; d, by Nicole Fogarty.)

440 e

A minority of reef-building coral species worldwide are
brooders, but brooding is the dominant reproductive mode
found in the Caribbean Sea (Szmant, 1986; Richmond and
Hunter, 1990; Smith, 1992). Broadcast spawning is a more
common reproductive mode in coral species, and in Aus-
tralia more than 100 coral species may spawn on a single
night (Harrison et al., 1984; Willis et al., 1985; Babcock
and Heyward, 1986). Species representing these modes
differ in colony size, gametic cycles, larval competency,
dispersal distance, and zooxanthellae transmission (Rich-
mond and Hunter, 1990). Brooders are typically smaller
than spawning corals and have multiple planulating cycles
per year, as opposed to one or two cycles in broadcast
spawners (Szmant, 1986).

FECUNDITY

Reproductive mode determines the frequency of lar-
val release; however, both abiotic and biotic factors can
influence the amount of gametes produced in corals. The
production of gametes is only possible when a coral has
reached an age, and perhaps more importantly a size, ca-
pable of reproduction (Hughes, 1984; Szmant, 1986). It is
difficult to measure the impact of stressors on gamete pro-
duction because it is naturally variable both temporally
and between individuals within a species (Chorneskey and
Peters, 1987). As coral cover declines in both the Caribbean
Sea and the Pacific Ocean (Gardner et al., 2003; Bruno
and Selig, 2007) there are fewer and often smaller adult
colonies. This change could reduce coral fecundity be-
cause small body size reduces gamete production (Szmant,
1986) and low population densities reduce fertilization
success (see Fertilization section, below). Even with rela-
tively high adult coral densities the fecundity of individual
colonies can be decreased by many stressors before and
during gametogenesis.

Coral bleaching has been observed to stop gameto-
genesis (Szmant and Gassman, 1990), reduce the number
of gametes produced (Fine et al., 2001), and decrease fer-
tilization rates in Acropora corals (Omori et al., 2001).
Nutrients added to the water column decreased the num-
ber of successfully developed embryos that were formed in
the corals Acropora longicyathus and A. aspera (Koop et
al., 2001). Changes in salinity and sedimentation can also
reduce gamete production and fertilization success in cor-
als (Richmond, 1993a, 1993b). Guzman et al. (1994) sug-
gested that the increase in injury levels and slower growth
in corals exposed to an oil spill further reduced gamete
size, viability, and fecundity. The presence of macroalgae
adjacent to coral colonies can decrease fecundity (includ-

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

ing the number and size of eggs) in the corals Montastraea
annularis and Montipora digitata (Hughes et al., 2007;
Foster et al., 2008). Impacts on fecundity are perhaps best
summarized by Rinkevich and Loya (1987), who sug-
gested that because reproductive activity involves such
high energy expenditure, any stress that diminishes energy
reserves will have an effect on adult fecundity.

FERTILIZATION ECOLOGY

Because broadcast spawners only have one or two
planulating cycles a year, it is imperative that fertiliza-
tion be successful. In any broadcast species, fertilization
success is highly variable and largely depends on the
synchronization of gamete release, gamete compatibility
(Palumbi, 1994; Levitan et al., 2004), gamete age (Oliver
and Babcock, 1992; Levitan et al., 2004), and abundance
of spawning adults (Levitan et al., 1992, 2004). However,
the health of the spawning colony and environmental con-
ditions during the spawning event also affect fertilization
success (Richmond, 1997; Humphrey et al., 2008).

During multispecies spawning events, synchronized
gamete release and species-specific gamete recognition
are critical for fertilization success and reducing the prob-
ability of interspecific fertilization (hybridization), which
may result in reduced offspring fitness (Mayr, 1963);
however, Willis et al. (2006) suggest a role for hybrid-
ization in range expansion and adaptation to a changing
environment. Species with overlapping spawning times
typically display low interspecific fertilization success in
laboratory crosses (Willis et al., 1997; Hatta et al., 1999;
Levitan et al., 2004). Interspecific fertilization success is
usually higher among morphologically similar species,
suggesting they are more closely related or possibly the
same species (Willis et al., 1997; Hatta et al., 1999; Wol-
stenholme, 2004), but interspecific fertilization can also
occur between Acropora species that have very different
branching morphologies (Hatta et al., 1999). Fertiliza-
tion success during a mass spawning event could be the
result of sperm attractant molecules produced by coral
eggs (Coll et al., 1994; Babcock, 1995) but could also be
regulated by gamete recognition proteins, such as those
that ensure species-specific fertilization in spawning sea
urchins (Zigler et al., 2005).

If coral colonies spawn asynchronously or encoun-
tered gametes are not compatible, eggs may go unfertilized
for extended periods of time or sperm may lose its viabil-
ity. The effect of age on gamete viability and fertilization
success differs among coral species; Platygyra sinensis
showed reduced fertilization after three hours (Oliver and

Babcock, 1992), but in Acropora spp. reduced fertiliza-
tion success occurred after seven to eight hours (Willis et
al., 1997; Omori et al., 2001). With increasing gamete
age, fertilization success is reduced in conspecific crosses,
but aging effects on gamete viability differ between sperm
and eggs. Montastraea spp. sperm lose viability after two
hours but eggs stay viable for more than three hours (Levi-
tan et al., 2004). Another consequence of gamete aging is
an increase in the likelihood of interspecific fertilization.
Hybridization rates between Montastraea faveolata eggs
and M. annularis and M. franksi sperm increased when
eggs had aged at least 75 minutes (Levitan et al., 2004).
Increased interspecific fertilization may be caused by a
breakdown in gamete recognition proteins, but the specific
mechanisms remain to be determined.

The density of spawning individuals plays a critical
role in fertilization success. If reproductive individual
densities are too low, fertilization success will be limited
(also referred to as the allee effect) (Levitan and McGov-
ern, 2005). Coma and Lasker (1997) found that fertiliza-
tion success in gorgonians was influenced by the density
of gametes, which was determined by nearest neighbor
distances (approximately 10 m), synchronous gamete re-
lease, or hydrodynamic processes. These factors probably
influence scleractinian fertilization success; however, it is
difficult to directly measure species-specific sperm concen-
trations in situ because a number of coral species spawn
synchronously. Field studies examining sperm concentra-
tions have used either of two methods: (1) measuring the
percent of fertilized eggs collected at different times and
locations on the reef or (2) determining the fertilization
potential of collected surface water samples by adding
them to unfertilized eggs and recording the proportion of
eggs fertilized (Oliver and Babcock, 1992; Levitan et al.,
2004). When lower production or dilution resulted in
locally lower than normal sperm concentrations, fertil-
ization success was reduced (Oliver and Babcock, 1992;
Willis et al., 1997; Omori et al., 2001; Levitan et al.,
2004). These studies showed peak fertilization potential
during or shortly after coral species spawn (Oliver and
Babcock, 1992; Levitan et al., 2004). Hence, synchronized
gamete release is a mechanism for the high gamete density
needed to ensure fertilization success.

High gamete concentration brings with it a potential
risk as well; as sperm densities increase so does the prob-
ability of polyspermy, whereby eggs become fertilized by
more than one sperm cell, which results in lowered fertil-
ization rates and developmental failure (Styan, 1998; To-
maiuolo et al., 2007). Reduced fertilization success at high
sperm concentrations has been described for several coral

NUMBER 38 e¢ 441

species (Oliver and Babcock, 1992; Willis et al., 1997;
Levitan et al., 2004), suggesting polyspermic fertilization
can occur in scleractinian corals. These findings suggest a
trade-off between spawning synchronously (i.e., high gam-
ete density) with other conspecifics to increase fertilization
and the potential risk of polyspermy. Polyspermy may
therefore act as a negative density-dependent mechanism.
Despite the evidence for polyspermy in coral laboratory
crosses, field fertilization rates never reached 100% dur-
ing mass spawning events (97% maximum; Levitan et al.,
2004), suggesting that polyspermic conditions are unlikely
to occur in nature. In light of recent decreases in adult
coral populations, reduced adult density and gamete aging
are perhaps the greatest threats to larval production.

LARVAL TRANSPORT: DISPERSAL AND CONNECTIVITY

After gamete fertilization, developing planula larvae
transport typically away from reproductive populations
(called “dispersal”) and to reefs where they recruit (called
“connectivity”) (Levin, 2006). The density of planulae ar-
riving to a reef determines recruitment strength. Larval
survival during dispersal varies by means of a combina-
tion of hydrodynamic processes, larval energetics, preda-
tion pressure (Fabricius and Metzner, 2004), and water
quality (Richmond et al., 2007).

Reproductive modes can provide insight into dispersal
potential, even though the planktonic duration of coral
species can be highly variable and remains undocumented
for the majority of scleractinian species. For example,
brooders generally settle within hours after release (Carlon
and Olson, 1993), whereas broadcast spawners such as
Acropora spp., Goniastrea spp., Platygyra spp., and Mon-
tastraea spp. have planktonic period of 4 to 7 days before
they are competent to settle and metamorphose (Babcock
and Heyward, 1986; Szmant, 1986). Larvae of the broad-
cast spawners Acropora muricata and A. valida settled
within 9 to 10 days (Nozawa and Harrison, 2008), but
larvae of the spawning corals Platygyra daedalea and Go-
niastrea favulus can settle between 2 and 3 days after fer-
tilization, which is sooner than some brooding corals, sug-
gesting that dispersal of these species might be of shorter
duration than has been assumed from survival estimates
(Miller and Mundy, 2003). In the absence of settlement
substrate, a small percentage of Acropora latistella, Favia
pallida, Pectinia paeonia, Goniastrea aspera, and Montas-
traea magnistellata larvae survived for 195 to 244 days in
the water column (Graham et al., 2008). Planulae larvae
can probably survive drifting in the plankton for long du-
rations until they encounter suitable settlement substrate;

442 e

however, the length of the planktonic period will partially
depend on whether the larvae have acquired zooxanthel-
lae, which give them additional energy reserves, from the
parent colony (Richmond, 1987).

The frequency of recruitment as a function of dis-
tance from a reproductive source population is called a
“dispersal kernel” (Steneck, 2006). For most planktonic
larvae it was assumed that relatively long larval survival
potential in combination with oceanographic transport
would generally prevent settlement close to a reproductive
source (Cowen et al., 2006). Recent reviews suggest that
even though many marine invertebrate larvae have the po-
tential (energy reserves) for long-distance dispersal, they
often settle locally because of a combination of oceano-
graphic conditions, larval behavior, and increasing mortal-
ity associated with planktonic conditions (Cowen et al.,
2000; Strathmann et al., 2002; Levin, 2006). The shape
of most dispersal kernels is now thought to be skewed to-
ward the reproductive source, that is, increased rates of lo-
cal recruitment (Figure 3; Steneck, 2006). Most dispersal
and connectivity research to date has focused on fishes;

A. Dispersal Kernel for Ecological and Evolutionary Connectivity

Ecologically important
(to sustain local populations)

Number of recruits to counteract local mortali

Evolutionarily important
(to prevent local extinction)

Number of successful recruits

Distance from larval source

B. Coral Mortality Shrinks Connectivity

a mortality (fewer reproductive adults)

ae ; BS Shrinks ecological connectivity distance

Distance from larval source

Number of successful recruits

FIGURE 3. Dispersal kernels determine potential connectivity dis-
tance between reproductive populations and offspring. A, Distinc-
tion between ecologically important recruitment necessary to balance
against local mortality and evolutionarily important recruitment to
balance against local extinction. B, Shrinking dispersal kernels result-
ing from adult coral mortality. (After Steneck, 2006.)

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

however, one study measured ecological connectivity of
coral larvae via a field experiment conducted around the
isolated Helix Reef in Australia (Sammarco and Andrews,
1988). They reported that 70% of coral recruitment oc-
curred within 300 m of the larval source and that rates
of recruitment declined with distance downstream from
the reef (Sammarco and Andrews, 1988). Further, as ex-
pected, broadcasters dispersed farther than did species of
brooding corals, but the estimated ecologically relevant
dispersal kernel for both species was remarkably local. A
recent review discusses the limited dispersal kernel of coral
planulae (Steneck, 2006); however, there is little experi-
mental evidence for the mechanisms that determine coral
ecological connectivity.

Recruitment rates must equal or exceed rates of adult
mortality to sustain a local population. Most dispersal ker-
nels show high rates of recruitment near the reproductive
source, with recruitment decreasing as distance increases
(Figure 3). Although that tail is important for gene flow,
that low density of settlement is not sufficient to sustain
populations. That is, the ecologically relevant portion of
a dispersal kernel reflects the sustained rate of recruitment
necessary to compensate for rates of mortality. The critical
level of settlement to sustain populations (i.e., horizontal
line above each shaded half of Figure 3) is not known; how-
ever, colonization rates of the introduced orange cup coral
Tubastraea coccina can provide some real-world insights
into the scale of ecological and evolutionary connectivity.
This brooding species was first introduced to the Nether-
lands Antilles in 1943 and then spread from island to is-
land through the Caribbean, taking 50 years to reach the
Bahamas and 60 years to reach Florida (Fenner and Banks,
2004). Once in a region, local populations grew rapidly.
This finding is consistent with the concept that the biogeo-
graphic spread results from the evolutionarily important
“long tail” of the dispersal kernel, whereas the ecologically
and demographically significant portion of the dispersal
kernel controlling local colonization is much smaller and
more local (Figure 3A). Observations of the spread of T.
coccina are conservative because some of the spread of
this species probably resulted from colonized ships moving
among the regions (Fenner and Banks, 2004).

Ecological connectivity necessary to sustain popula-
tions against chronic mortality is much more difficult to
measure than is evolutionary connectivity. Evolutionary
or genetic connectivity can be directly measured using a
variety of molecular genetic techniques (reviewed in Hell-
berg, 2007). In Japan, gene flow between islands 30 to
150 km apart was determined to be consistently higher
for the spawner Acropora tenuis than for the brooding

species Stylophora pistillata, but both coral species had
unique genotypes across islands separated by 500 km (Ni-
shikawa et al., 2003). In the Caribbean, a genetic break
was detected for Acropora palmata, roughly dividing pop-
ulations from the Greater Antilles and western Caribbean
from populations in the Lesser Antilles and the southern
and eastern Caribbean (Baums et al., 2005). On the rela-
tively contiguous Great Barrier Reef (GBR), high rates of
genetic connectivity were observed for both brooders and
spawners. For example, gene flow was detected in all the
spawners and three of the five brooders despite being sepa-
rated by 500 to 1,200 km (Ayre and Hughes, 2000). How-
ever, the same species of corals were genetically distinct
on Lord Howe Island, which is separated from the GBR
by 700 km (Ayre and Hughes, 2004). This observation
suggests that coral larvae can use islands within the evolu-
tionarily important tail of the dispersal kernel as “stepping
stones” to maintain genetic connectivity between distant
reefs separated by long distances (Steneck, 2006).

Although dispersal kernels are useful for visualiz-
ing how larval availability declines with distance from a
source, their ecological effect can be variable. For example,
without changing the shape of the kernel but reducing the
number of recruits as a consequence of reduced reproduc-
tive output following an adult mortality event (Figure 3B),
the range of both the ecological and evolutionary parts of
the kernel can shrink. If this happens, connectivity among
distant reefs could sever, making recovery following an
acute disturbance difficult or impossible.

SETTLEMENT ECOLOGY

As local and global threats continue to decrease coral
cover, it is likely that fewer coral larvae will be supplied
to reefs that may or may not have appropriate settlement
habitat. For corals, the transitional stage from planktonic
planula larvae to sessile benthic juveniles involves a two-
step process of settlement and metamorphosis. Settlement
is the behavioral response of a larva when it stops disper-
sal and selects substrate for recruitment. Metamorphosis
includes the subsequent morphological and physiological
changes that pelagic larvae undergo to become benthic
juveniles. Settlement of coral larvae can be influenced by
habitat qualities that facilitate or inhibit settlement and
metamorphosis of larvae supplied to a reef (Figure 4).
Larval settlement behavior can be determined by the con-
ditions the larvae experienced in the plankton or by the
presence of positive or negative cues on the benthos or in
the water overlying the reef.

NUMBER 38 ¢ 443

LARVAL CONDITION UPON ARRIVAL

As coral larvae disperse in the plankton they are ex-
posed to water quality conditions that may affect larval
health, behavior, survival, and settlement success (Vermeij
et al., 2006). Experiences during early life stages (i.e., de-
pleted energy reserves, nutritional stress, environmental
stressors, and pollutant exposure) have latent effects on
later life stages in numerous marine larvae across differ-
ent phyla (reviewed in Pechenik, 2006). Even short-term
exposure to stressors or a slight delay in metamorphosis
can reduce fitness in juveniles and adults (i.e., decrease
growth rate, lower competitive ability, reduce survival,
and decrease fecundity) (Pechenik, 2006). Although the
mechanisms through which latent effects are mediated are
not known, it is suspected that transcriptional or trans-
lational processes or direct DNA or key enzyme dam-
age are responsible (Pechenik et al., 1998; Heintz et al.,
2000; Pechenik, 2006). As very few studies have tested
latent effects in coral larvae, we describe some of the pat-
terns found in other marine organisms to highlight how
pre-settlement stress might impact post-settlement coral
growth and survival.

Marine invertebrate larvae often rely on external cues
to trigger metamorphosis. Without these cues, the larval
period can be prolonged (reviewed in Pechenik, 1990),
and post-settlement fitness may be reduced (Pechenik,
2006). For some invertebrates, including abalones, tu-
nicates, and bryozoans, delayed metamorphosis slowed
post-metamorphic development (Wendt, 1998; Roberts
and Lapworth, 2001; Marshall et al., 2003). Depleted
energy resources during the larval stage may also be an
important contributor to post-settlement growth and sur-
vival. Bennett and Marshall (2005) found that depleted
energy reserves caused by increased activity in larvae of
the ascidian Diplosoma listerianum were more costly en-
ergetically than extending the larval period or completing
metamorphosis. Food limitation during the larval period
can reduce size, total organic content, energy reserves of
metamorphosed animals, juvenile growth rates, and sur-
vival (Miller, 1993; Pechenik, 2002; Thiyagarajan et al.,
2003; Chiu et al., 2007, 2008).

Water quality conditions can directly reduce coral
larval survival and settlement but also may cause latent
effects for new recruits. Salinity reductions during pre-
settlement periods can reduce post-metamorphic growth
rates and survival for various marine invertebrates (Pech-
enik et al., 2001; Thiyagarajan et al., 2008). Vermeij et
al. (2006) tested salinity stress on Montastraea faveolata
larvae and how that influenced subsequent post-settlement

444 e¢ SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 4. Coral larval substrate selection is critical to post-settlement survival. a, Favia fragum
larvae explore the benthos for a suitable settlement site; some larvae have already attached and
are beginning to metamorphose. b, Acropora cervicornis settlers are attached to Titanoderma
prototypum and have metamorphosed. c, A new recruit of Montastraea faveolata has settled on
coralline algae, which has started to slough its outer layer of tissue, knocking the coral recruit
off the substrate. d, Montastraea faveolata recruits are being overgrown by a coralline alga. e,
Montipora capitata \arvae have settled on Ulva sp., an ephemeral substrate. f, A Siderastrea
radians recruit has settled in a high-sedimentation environment. (Photographs a, b, by Raphael
Ritson-Williams; c, d, by Nicole Fogarty; e, f, by Mark Vermeij.)

performance. Lower than normal seawater salinity caused
increased pre- and post-settlement mortality and increased
the mobility of coral planulae. It was suggested that the
increased activity of the larvae in the lower salinities was
an attempt to escape the unfavorable conditions. With
increased activity, energy reserves were depleted, which
was suggested to be the cause of pre-settlement mortal-
ity, smaller post-settlement size, and lower post-settlement
survival. Planulae in the lower-salinity treatments settled
on a greater range of substrate types. This study empha-
sized the importance of planktonic conditions on the per-
formance of settling coral larvae, which could then influ-
ence post-settlement ecology.

LARVAL BEHAVIOR IN OVERLYING WATER

Coral larvae possess a wide array of behaviors that
allow them to enhance the likelihood of successful settle-
ment, including, but not limited to, sensitivity to light
(Lewis, 1974; Mundy and Babcock, 1998), depth (Car-
lon, 2001, 2002; Baird et al., 2003; Suzuki et al., 2008),
and chemical cues (Morse et al., 1994). One field study
showed that multiple behavioral choices determined the
larval settlement site of the Caribbean brooder Agaricia
humilis (Raimondi and Morse, 2000). The larvae swam
down when restricted to 3 and 8 m but swam toward the
surface when restricted to 25 m. In further field experi-
ments, larvae settled in response to the coralline alga Hy-
drolithon boergesenii but would only settle directly on the
coralline alga when it was on the underside of a settlement
tile. This study showed that coral larvae are capable of
complex behaviors, which are determined to some extent
by their ability to detect and discriminate between positive
and negative settlement cues in their habitat.

POsITIVE SETTLEMENT CUES

Many marine invertebrate larvae use chemical cues to
determine the appropriate habitat for settlement (Pawlik,
1992; Hadfield and Paul, 2001). Chemical cues are impli-
cated for both settlement and metamorphosis of corals and
may be released by conspecifics and other organisms that
indicate appropriate habitat for survival and growth. Re-
search in the Caribbean showed that a membrane-bound
carbohydrate complex from the coralline red alga Hydro-
lithon boergesenii induced settlement and metamorphosis in
the brooded larvae of Agaricia humilis (Morse and Morse,
1991; Morse et al., 1994). It was suggested that many corals
require an algal cue for the induction of settlement, indicat-
ing a common chemosensory mechanism for settlement and
metamorphosis among coral larvae (Morse et al., 1996).

NUMBER 38 ¢ 445

Both the larvae of Acropora millepora, a common
Indo-Pacific coral species, and coral larvae collected from
natural slicks after mass spawning events used coralline
algae for settlement and metamorphosis (Heyward and
Negri, 1999). Four species of crustose coralline algae,
one non-coralline crustose alga, two branching coralline
algae, and the skeleton of the massive coral Goniastrea
retiformis induced metamorphosis. Chemical extracts
from both the crustose red alga Peyssonnelia sp. and the
coral skeleton were highly active, inducing up to 80%
larval metamorphosis. Coral larvae can also distinguish
among species of coralline algae. The Australian spawning
coral Acropora tenuis had different rates of settlement in
response to different species of coralline algae (Harrington
et al., 2004). Settlement choice resulted in higher rates of
post-settlement survival on the preferred coralline algae,
illustrating the recruitment consequences of larval selectiv-
ity. Chemical cues appeared to be involved in this selective
behavior, because methanol extracts of the coralline red al-
gae Titanoderma prototypum and Hydrolithon reinboldii
both induced metamorphosis of A. tenuis.

Comparative studies have revealed that settlement
and metamorphosis in response to crustose coralline algae
is not an obligate trait of all coral species. Two brood-
ing Australian corals were compared for their settlement
selectivity (Baird and Morse, 2004). Acropora palifera lar-
vae only metamorphosed in the presence of coralline red
algae, but Stylophora pistillata larvae showed some meta-
morphosis in unfiltered seawater and also metamorphosed
onto glass coverslips. A study in Guam found that larvae
of the spawning species Goniastrea retiformis preferred
substrate covered with crustose coralline algae (CCA), but
the reef-flat brooding coral Stylaraea punctata preferred
biofilmed rubble (Golbuu and Richmond, 2007).

Coralline algae have been identified as a positive set-
tlement cue for some corals, but it is unclear if the biofilms
present on these algae or the algae themselves are respon-
sible for the observed settlement behavior (Johnson et al.,
1991; Webster et al., 2004). Biofilms were isolated from
the coralline alga Hydrolithon onkodes, and one strain
of bacteria alone was enough to induce settlement and
metamorphosis of Acropora millepora larvae (Negri et al.,
2001). When H. onkodes was sterilized in an autoclave
and treated with antibiotics, it still induced significantly
more settlement and metamorphosis than seawater or ter-
racotta tiles. Additionally, coral larvae can distinguish be-
tween tiles conditioned at different depths, which could be
related to depth-related differences in bacterial commu-
nity composition of biofilms that formed on tiles (Webster
et al., 2004). Whether the coralline algae or its biofilm is
producing the inductive compound(s) may depend on the

446 «°

coral and the coralline algae species tested. The specificity
of bacterial communities to different coralline algal species
has rarely been investigated (Johnson et al., 1991). With
the recent development of more refined genetic techniques
it is possible to compare different microbial communities,
which might enable the identification of the microbe(s) that
can induce coral larval settlement and metamorphosis.

NEGATIVE SETTLEMENT CUES

Water quality and substrate conditions impact fertiliza-
tion rates and also may inhibit some coral larvae from nor-
mal settlement and metamorphosis. Low coral recruitment
is commonly documented in the field, yet surprisingly few
studies have experimentally tested which substrate char-
acteristics might deter coral larval settlement. Coral larval
survival and settlement can be reduced by many environ-
mental stresses, such as elevated temperatures (Edmunds
et al., 2001), variation in salinity (Vermeij et al., 2006),
sedimentation (Hodgson, 1990; Gilmour, 1999), and UVB
radiation (Kuffner, 2001; Gleason et al., 2006). Survival
and settlement are reasonable ecological metrics for the ef-
fects of stress, but an important gap in our knowledge is
how sublethal stress influences larval behavior and post-
settlement health and success (Downs et al., 2005). New
techniques including cellular biomarkers and differential
gene expression using microarrays should provide impor-
tant techniques to measure sublethal stress in coral larvae.

Water quality conditions that are known to impact
adult corals also have dramatic effects on larval supply
and settlement. Of the physical conditions that negatively
influence larval settlement, elevated temperature has re-
ceived the most attention and has the potential to increase
in frequency and duration as ocean temperatures continue
to warm. Larvae of the Caribbean brooding coral Porites
astreoides were killed and had low densities of zooxan-
thellae when exposed to elevated temperatures for 24
hours (Edmunds et al., 2001, 2005). High temperatures
(36°C) killed Acropora muricata \arvae within 40 hours
(Baird et al., 2006), and temperatures of 32°C killed Dip-
loria strigosa larvae and reduced their settlement (Bassim
and Sammarco, 2003). However, at elevated temperatures
(29°C) larvae of Stylophora pistillata had the same settle-
ment as at 25°C (Putman et al., 2008), and more larvae
settled on the CCA in 25°C than in 23°C. Many of these
studies used different experimental conditions, making it
difficult to compare the effects of temperature on different
species of coral larvae. Temperature is one stress that is
relatively well studied, but more research is necessary to
understand other physical stressors, such as ocean acidifi-

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

cation (Albright et al., 2008), that will affect coral larvae
in the future.

Larval interactions with the biological inhabitants
of reef communities can also reduce larval settlement.
Algal turfs, macroalgae, and benthic cyanobacteria can
negatively impact the settlement of coral larvae (Kuffner
and Paul, 2004; Birrell et al., 2005; Kuffner et al., 2006;
Birrell et al., 2008a). In the Florida Keys, two brown al-
gae, Dictyota pulchella and Lobophora variegata, reduced
the total number of Porites astreoides settlers (Kuffner et
al., 2006). In the Philippines, the algae Sargassum polycys-
tum and Laurencia papillosa decreased larval settlement
of Pocilloproa damicornis, but water conditioned with
these algae increased settlement over the seawater controls
(Maypa and Raymundo, 2004). In Australia, water con-
ditioned with the foliose brown alga Padina sp. reduced
larval settlement of Acropora millepora; however, water
conditioned with the brown alga Lobophora variegata
increased settlement (Birrell et al., 2008a). The cyanobac-
terium Lyngbya majuscula reduced the survivorship of
Acropora surculosa \arvae and settlement and metamor-
phosis of Pocillopora damicornis in studies conducted on
Guam (Kuffner and Paul, 2004), and in Florida, the cya-
nobacterium Lyngbya polychroa caused Porites astreoides
to avoid settling adjacent to it on settlement tiles (Kuffner
et al., 2006). Some macroalgae and cyanobacteria can act
as settlement inhibitors for coral larvae, but this was not
true for all the algae tested. A surprising contrast was ob-
served for Favia fragum larvae, which had high rates of
settlement and metamorphosis onto live Halimeda opun-
tia when offered with coral rubble (Nugues and Szmant,
2006). Coral larvae of Montipora capitata were observed
to settle onto Ulva sp. (Figure 4e; Vermeij et al., 2009).
Why these larvae would settle directly onto blades of algae
is unclear as this substrate is ephemeral, thus probably in-
creasing post-settlement mortality. Little research has been
done on the mechanisms that algae use to inhibit settle-
ment, but algal qualities such as natural products, shading
and abrasion, serving as vectors of bacteria, and releasing
dissolved organic matter may contribute to the negative
impacts of algae on larval settlement.

Competition from other members of coral reef com-
munities also influences larval behavior. Tissue of the scler-
actinian coral Goniopora tenuidens suspended in seawater
inhibited metamorphosis of Pocillopora damicornis larvae
and reduced the growth of new recruits over seven days
(Fearon and Cameron, 1996). The tissue from Goniopora
tenuidens also caused increased mortality of larvae from
P. damicornis, Platygyra daedalea, Fungia fungites, and
Oxypora lacera. Increased research on the types of benthic

organisms and the mechanisms they use for competition
with coral larvae is an important area for further study.
An integrated approach to larval stress, physiology, and
the physical and biological characteristics of settlement
substrata will reveal the impact of benthic organisms on
coral larval behavior, settlement, and post-settlement sur-
vival. Determining what benthic habitat characteristics are
necessary for increased settlement will be a critical step for
managing reef habitats for increased coral recruitment.

POST-SETTLEMENT ECOLOGY

Corals, and most benthic marine organisms, suffer
high rates of mortality soon after settlement because they
are small and vulnerable. Post-settlement processes from
the time corals settle (i.e., attach to the benthos) to recruit-
ment (1.e., survive to some later phase) determines much of
coral demography (Vermeij and Sandin, 2008). This con-
cept is consistent with the tenet of clonal population biol-
ogy that states as clonal organisms grow the probability of
their death declines but the probability of injury increases
(Hughes and Jackson, 1985). Thus, the two rates of early
post-settlement mortality and growth can strongly. influ-
ence the local abundance of corals.

Post-SETTLEMENT MORTALITY

Coral recruits can die from a myriad of causes includ-
ing chronic disturbances such as competition and preda-
tion and pulse disturbances such as bleaching and disease.
However, the chronic disturbances probably drive most
post-settlement mortality and thus are serious impedi-
ments to reef recovery. Caribbean reefs are a case in point,
with incidences of recovery much lower than Indo-Pacific
reefs as a result of setbacks from chronic disturbances
(Connell et al., 1997).

Algae, encrusting invertebrates, and sediment have all
been shown to have deleterious effects on newly settled cor-
als (Figure 5; Rylaarsdam, 1983). Settling corals, with lim-
ited stores of energy to invest in competitive interactions, are
particularly vulnerable when faced with a well-developed
benthic community structure and limited space (Jackson and
Buss, 1975; Sebens, 1982; Connell et al., 1997). However,
the mechanisms, or causes, of reduced growth and mortality
of newly settled larvae, recruits, and juveniles have, for the
most part, only recently been investigated.

Encrusting invertebrates (particularly sponges) can be
especially inhospitable for newly settled corals. In cryp-
tic habitats, newly settled corals are likely to lose out by

NUMBER 38 ¢ 447

overgrowth of fast-growing heterotrophic groups such as
sponges, bryozoans, and bivalves (Vermeij, 2005). Aerts
and van Soest (1997) determined the impact of sponges on
coral survival to be greatly species specific. Physical, chem-
ical, and biological properties of benthic invertebrates
may inhibit coral growth and survival. Some studies used
chemical extracts of sponges (Sullivan et al., 1983; Pawlik
et al., 2007) to show that allelopathy can negatively im-
pact adult corals. Coral recruits are even more susceptible
to stress, yet surprisingly few studies have examined sec-
ondary metabolites for their impact on the early life history
stages of corals. A field study by Maida et al. (1995) sug-
gested that allelopathy reduced recruitment of corals adja-
cent to the octocorals Sinularia flexibilis and Sarcophyton
glaucum, and both the live octocorals and settlement plates
with dichloromethane extracts of S. flexibilis inhibited coral
settlement and survival. More long-term, small spatial scale
(millimeters to centimeters) studies are needed to determine
the effect of benthic invertebrates on post-settlement sur-
vival (Edmunds et al., 2004; Vermeij, 2006).

Areas of high algal biomass are known to be poor nurs-
ery habitats for settling corals (Birkeland, 1977; Bak and
Engel, 1979; Harriott, 1983; Birrell et al., 2008b; Vermeij
and Sandin, 2008; Vermeij et al., 2009). There are several
mechanisms by which algae may be deleterious to corals.
Algae may interfere with larval settlement by simply pre-
empting available settlement space (Mumby et al., 2006;
Box and Mumby, 2007). At least one species of turf algae
alone (without sediment) has reduced settlement of corals
in laboratory experiments (Birrell et al., 2005). More direct
physical interactions including algal shading, abrasion, or
basal encroachment can result in reduced coral growth or
increased mortality (Lirman, 2001; McCook et al., 2001).
Shading by the encrusting brown alga Lobophora variegata
over six months caused a 50% increase in morality of ju-
venile Agaricia agaricites (less than 20 mm diameter), and
the mere presence of L. variegata around the coral reduced
colony growth by 60% (Box and Mumby, 2007). However,
shading by Dictyota pulchella resulted in no direct mortality
but caused a 99% decrease in coral growth. Other studies
have determined that Lobophora variegata (in the absence
of grazing) is a superior competitor to Caribbean corals,
including A. agaricites, A. lamarcki, Meandrina meandrites,
Mycetophyllia aliciae, and Stephanocoenia intersepta, and
to at least one species of Pacific coral, Porites cylindrica (De
Ruyter van Steveninck et al., 1988; Jompa and McCook,
2003). Thus, it is likely that community phase shifts to high
algal biomass decrease recruitment by reducing larval set-
tlement and post-settlement survival (Hughes and Tanner,
2000; Kuffner et al., 2006).

448 e

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 5. Macroalgae can be a dominant space occupier on degraded reefs and can inhibit coral recruit-

ment at multiple life history stages. a, The macroalgae Dictyota spp. and Halimeda opuntia covered most

of the benthos on this Belize reef, potentially inhibiting coral settlement. b, Recruits of Acropora palmata

surrounded by Dictyota sp. c, A new recruit of Diploria sp. surrounded by Gelidiella, Jania, Dictyota, and

the cyanobacterium Dichothrix sp. d, Montastraea annularis overgrown by Halimeda sp. (All photographs

by Raphael Ritson-Williams.)

Reduced coral recruitment in algal-dominated reefs (Ed-
munds and Carpenter, 2001; Birrell et al., 2005) is thought to
be in part the consequence of chemically induced mortality
or the increased biomass of fleshy algae actually functioning
as a reservoir for coral pathogens (Littler and Littler, 1997;
Nugues et al., 2004). Bak and Borsboom (1984) proposed
that the reduction in water flow adjacent to macroalgae
could cause increased coral mortality through changes in the
flow regime and increased allelochemical concentrations.
Most recently, enhanced microbial activity caused by algal

exudates has been proposed as a mechanism of competition
(Smith et al., 2006; Vermeij et al., 2009). Kline et al. (2006)
determined that elevated levels of dissolved organic carbon,
which can occur in areas of high algal biomass, increased the
growth rate of microbes living in the mucopolysaccharide
layer of corals. These studies all suggest that the detrimental
effect of algae on corals could be mediated by several prop-
erties of macrophytes.

On modern reefs, algal-related post-settlement mor-
tality probably decreases the population density of coral

recruits. Vermeij (2006) compared his recruitment study
in Curacao from 1998 to 2004 to that of Van Moorsel
(1989) from 1979 to 1981, using the same method in the
same location. Recruit densities on the topsides of settle-
ment panels in the more recent study were 5.16 times
lower and recruitment on the undersides was 1.14 times
lower than the 1979-1981 study. Macroalgae had re-
placed CCA as the dominant topside space occupier, cre-
ating a less-suitable habitat for coral recruitment com-
pared to the crustose algae that had dominated the same
site roughly 20 years earlier. In places where Diadema ur-
chin recovery and grazing have reduced algal abundance,
the population density of juvenile corals has increased
(Edmunds and Carpenter, 2001; Aronson et al., 2004;
Macintyre et al., 2005).

While herbivory can improve the recruitment poten-
tial by keeping reefs relatively free of algae, it can also be a
potential cause of mortality for newly settled corals. Graz-
ing rates on exposed outer surfaces of shallow reefs are ex-
tremely high, exceeding thousands of bites per square me-
ter per day (Carpenter, 1986; Steneck and Dethier, 1994;
Steneck and Lang, 2003). Bites, especially from parrotfish
that graze deeply into carbonate substrates, would easily
kill a newly settled coral. Few studies have documented
recruit mortality resulting from fish grazing (Mumby et
al., 2006), although it has been suggested as the cause of
the low number of recruits observed on the top surface
of settlement plates (Adjeroud et al., 2007). The herbiv-
orous sea urchin Diadema antillarum was shown to be
a significant agent of mortality for newly settled corals
(Sammarco and Carleton, 1981). The highest mortality of
newly settled corals is likely to occur on outer exposed
surfaces where algal growth rates and herbivore grazing
rates are greatest and rates of sedimentation are highest.
In shallow reef habitats where algal growth and herbivory
rates are greatest, coral recruitment is greater in subcryp-
tic microhabitats (Bak and Engel, 1979). However, which
microhabitats increase post-settlement survival has rarely
been tested (but see Babcock and Mundy, 1996).

Post-SETTLEMENT GROWTH RATES

Given the vulnerability of small size classes, the adap-
tive advantages of rapid growth rates are obvious. Coral
recruit survival is not merely a function of the attributes
of the settlement substrate but also of the coral’s ability
to resist overgrowth by neighboring encrusting inverte-
brates and algae (Richmond, 1997). As new corals grow,
their mortality rates decline (Vermeij and Sandin, 2008),
and they are less likely to be overgrown by competitors

NUMBER 38 ¢ 449

(Hughes and Jackson, 1985). Often, however, the slow
growth rates of newly settled corals make this a losing
battle, and early post-settlement mortality is generally
high (Figure 6; Bak and Engel, 1979; Edmunds, 2000; Ver-
meij and Sandin, 2008). Even in a controlled environment,
laboratory studies showed that a coral that remains less
than 3 mm in diameter for two or three months has only a
20% chance of survival (Rylaarsdam, 1983). Field studies
report a huge amount of variance in early post-settlement
mortality. Babcock (1985) found post-settlement survivor-
ship over the first three to six months ranged from 16%
to 71%, whereas more recently Box and Mumby (2007)
determined a monthly estimated mortality rate for Agari-
cia agaricites to be 3.5% per month. Annual juvenile coral
survivorship estimates range from 0% to 77% (Smith,
1992; Wittenberg and Hunte, 1992; Maida et al., 1994;
Smith, 1997; Edmunds, 2000).

Different species of corals have distinctly different
rates of growth and ability to recover following a distur-
bance (Wakeford et al., 2008). Specifically, some of the
Indo-Pacific acroporid corals (e.g., Acropora tenuis) are
extremely “weedy” and are capable of growing nearly
6 cm in 1.5 years (Omori et al., 2008); this translates to
an average growth rate of 3.2 mm/month compared to
the much slower growth rates reported for Oxypora sp.
as ranging between 0.2 and 0.5 mm/month (Babcock and
Mundy, 1996).

Settlement habitat also influences growth rates of
newly settled corals. Subcryptic habitats protect coral re-
cruits from stresses and disturbances common on outer reef
surfaces, but they will invariably have lower productivity
potential. Diameters of Platygyra sp. and Oxypora sp. set-
tlers increased one-quarter to one-half as fast in cryptic
undersides than they did on upper exposed surfaces for the
two species, respectively (Babcock and Mundy, 1996). Im-
portantly, however, new recruits that selected subcryptic
microhabitats had higher survivorship despite their slower
growth rates (Babcock and Mundy, 1996).

VARIABILITY OF POST-SETTLEMENT
SURVIVAL AND GROWTH:
THE ROLES OF BIODIVERSITY
AND LIFE HISTORY STRATEGIES

Before the disease-induced Acropora spp. decline in the
Caribbean, fundamental differences existed between acro-
porid-dominated reefs of the Caribbean and Indo-Pacific
regions. Caribbean reefs are largely built by two species of
Acropora. Both species recruit rarely (Rylaarsdam, 1983;

450 °

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 6. A time series of the growth of Agaricia sp. settled on a terracotta tile in Bonaire over 3.75 years.
After March 2007 (e), this recruit is being overgrown by the coralline alga Titanoderma prototypum, a
known settlement-facilitating species, illustrating just how hazardous the settlement environment can be.
a, June 2004, recruit diameter is 1.3 mm; b, March 2005, 3.4 mm; c, July 2005, 8.4 mm; d, June 2006,
15.2 mm; e, March 2007, 16.0 mm; f, March 2008, 12.0 mm. One segment on the scale bar = 1 mm. (All
photographs by Suzanne Arnold.)

Sammarco, 1985), but their clonal growth created massive
monocultures of rapidly growing reefs capable of keeping
up with rising sea level (Adey, 1978). In contrast, there
are two orders of magnitude more species of Acropora on
Indo-Pacific coral reefs, and the population density of their
recruits are also orders of magnitude greater on Indo-Pacific
reefs than on Caribbean reefs (Hughes et al., 1999).
Although the high diversity of acroporid corals in
the Pacific spans the spectrum of life history characteris-
tics from weeds (i.e., high reproductive output and rapid

growth rate; Omori et al., 2008) to trees (i.e., competitively
dominant, large colonies; Baird and Hughes, 2000), the two
acroporid species comprising Caribbean reefs require long
adult lives and considerable clonal propagation. However,
since the acroporid die-off in the early 1980s, Caribbean
reefs have fundamentally changed. Because of the resul-
tant algal phase shift (Hughes, 1994), acroporid reefs have
become hostile to the rare acroporid recruits, and they
have lost their receptivity for reattachment of encrusting
fragments (Williams et al., 2008). These changes on many

Caribbean reefs may be the primary reason why they ap-
pear less capable of recovering from widespread distur-
bances such as coral disease and bleaching.

The massive, slow-growing coral Montastraea annu-
laris is also a broadcast spawner and framework builder
in the Caribbean. It also has very low rates of recruitment
(Hughes and Tanner, 2000) and thus requires long adult
life to establish its dominance. Although it dominates Ca-
ribbean reefs today (Kramer, 2003) and is relatively hardy,
it too has shown elevated levels of disease in recent years
(Pantos et al., 2003) and has increased susceptibility to dis-
ease after bleaching (Miller et al., 2006). Again, the long-
term prognosis for this Caribbean reef builder is poor.

Weedy, brooding species such as Agaricia spp. and
Porites spp. are the thrust behind the current rates of coral
recruitment in the Caribbean. The Caribbean brooder
Agaricia agaricites is often the most abundant recruit on
Caribbean reefs in recent times (Bak and Engel, 1979).
This species has well-documented high rates of recruit-
ment and adequate sediment-rejection capabilities yet re-
generates poorly from lesions and is often outcompeted by
other corals (Bak and Engel, 1979). In the past 30 years
Agaricia tenuifolia has replaced other corals to dominate
the community on two reefs that had historically differ-
ent community compositions (Aronson et al., 2004).The
increasing community dominance observed for Porites as-
treoides at six sites in the Caribbean is being driven by a
constant recruitment rate coinciding with reduced percent
cover of other coral species (Green et al., 2008).

It is possible that these life history-related differences
are fundamentally changing Caribbean reefs. Are Carib-
bean reefs today following the paths of forests and other
marine ecosystems in their shift to weedy, stress-tolerant
species? (see Knowlton, 2001). A recovery such as seen in
Palau following the 1998 bleaching event, where sexual
recruitment and remnant regrowth were equal contribu-
tors (Golbuu et al., 2007), has yet to be recorded in the
Caribbean. Success stories of Caribbean recoveries led by
broadcast spawning species are scarce (but see Idjadi et
al., 2006). Thus, the relative importance of sexual versus
asexual reproduction to recovery in the Caribbean needs
to be addressed by long-term observations with particular
focus on recovery following large-scale disturbances such
as major storms and bleaching events.

Thus, it seems that Caribbean reefs were built by
corals that have been successful since the Pleistocene
(Pandolfi and Jackson, 2006) with a strategy of low re-
cruitment, considerable clonal growth, and low post-
settlement mortality. However, that strategy may not be
broadly viable today, given the global climate trajectory

NUMBER 38 e¢ 451

(Hoegh-Guldberg et al., 2007) and patterns of human
activities. While Indo-Pacific reefs are not immune to de-
clines in rates of coral recruitment in recent years (Wake-
ford et al., 2008), the higher biodiversity and range of re-
cruitment and post-recruitment strategies (e.g., high rates
of growth) allow reefs there to be more resilient.

CONCLUSIONS

Coral mortality has increased in recent decades, mak-
ing coral recruitment more important than ever before in
sustaining coral reef ecosystems and contributing to their
resilience. We identified three critical sequential phases to
the recruitment process of corals: larval availability, larval
settlement, and post-settlement ecology. All three factors
are necessary for coral recruitment and, ultimately, for
maintenance or recovery of coral reef ecosystems.

Most coral planulae available for recruitment are
probably from relatively local reproduction and relatively
short-distance connectivity. As adult coral abundance de-
clines, both fertilization success and the effective disper-
sal distance of corals (see Figure 3B) will likely decline as
well. Physiological stress on reproducing corals might also
result in fewer and possibly weaker coral larvae arriving,
thereby reducing the per capita rate of settlement success.

Once in the vicinity of a coral reef, settling corals re-
spond to a hierarchy of environmental cues both in the
water and from the reef. Several studies have identified or-
ganisms that facilitate or inhibit the settlement and meta-
morphosis of corals. Crustose coralline algae can facilitate
coral settlement but, disturbingly, this group of algae is
becoming rarer on coral reefs as macroalgae become in-
creasingly dominant. Macroalgae are known inhibitors of
settlement, which may result from their ability to rapidly
occupy settlement habitat, their suite of secondary metab-
olites, their microbial communities, or a combination of
some or all of these mechanisms.

Stressors that impact multiple life history stages of corals
have the most potential to greatly reduce coral recruitment.
Poor water quality (such as sedimentation and increased
temperatures) and the increased abundance of macroalgae
are known to decrease coral recruitment and negatively
impact corals at many different life history stages. Human
impacts on the water quality of marine systems continue to
grow, and few locations remain untouched (Halpern et al.,
2008). These and other stressors may decrease the reproduc-
tive output of corals, physiologically stress the larvae, block
subcryptic nursery habitats, create negative settlement cues,
and result in increased post-settlement mortality.

452 e

Globally, many Indo-Pacific reefs have higher rates of
settlement, recruitment, and recovery from disturbances,
which could be the result of higher biodiversity in the region.
In contrast, Caribbean reefs may have evolved a strategy of
low recruitment and considerable clonal growth, with low
post-settlement mortality for its few reef-building acropo-
rid corals. Unfortunately, that strategy may be ineffective
in the future given the global climate trajectory of higher
ocean temperatures, acidification, and greater disturbance
from tropical storms, which will continue to physiologically
stress corals. Because Indo-Pacific reefs have two orders of
magnitude more acroporid species, weedy and potentially
resilient strategies could succeed. If current trends continue
on modern reefs, it is possible that reefs in the future will
differ from those of the recent past.

ACKNOWLEDGMENTS

Funding for our research on these various topics was
provided, in part, by the Smithsonian Marine Science
Network. Special thanks to Klaus Ruetzler and Michael
Carpenter for facilitating research for RRW, VP, NF, SA,
and RS at the Carrie Bow Field Station in Belize. SA and
RS thank the Coral Reef Targeted Research Project (Con-
nectivity), National Fish and Wildlife Fund, Wildlife Con-
servation Society and the Bonaire Marine National Park
(STINAPA). VP and RRW thank the Mote “Protect Our
Reefs” grant program. This is contribution number 849 of
the Caribbean Coral Reef Ecosystems Program (CCRE),
Smithsonian Institution, supported in part by the Hunt-
erdon Oceanographic Research Fund, and Smithsonian
Marine Station at Fort Pierce (SMSFP) contribution num-
ber 765.

LITERATURE CITED

Adey, W. 1978. Coral Reef Morphogenesis: A Multidimensional Model.
Science, 202:831-837.

Adjeroud, M., L. Penin, and A. Carroll. 2007. Spatio-temporal Hetero-
geneity in Coral Recruitment around Moorea, French Polynesia:
Implications for Population Maintenance. Journal of Experimental
Marine Biology and Ecology, 341:204-218.

Aerts, L.A.M., and R. W. M. van Soest. 1997. Quantification of Sponge/
Coral Interactions in a Physically Stressed Reef Community, NE
Colombia. Marine Ecology Progress Series, 148:125-134.

Albright, R., B. Mason, and C. Langdon. 2008. Effect of Aragonite Satu-
ration State on Settlement and Post-settlement Growth of Porites
astreoides Larvae. Coral Reefs, 27:485-490.

Aronson, R.B., I. G. Macintyre, C. M. Wapnick, and W. O. O’Neill.
2004. Phase Shifts, Alternative States, and the Unprecedented Con-
vergence of Two Reef Systems. Ecology, 85:1876-1891.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Ayre, D. J., and T. P. Hughes. 2000. Genotypic Diversity and Gene Flow
in Brooding and Spawning Corals along the Great Barrier Reef,
Australia. Evolution, 54:1590-1605.

. 2004. Climate Change, Genotypic Diversity and Gene Flow in
Reef-Building Corals. Ecology Letters, 7:273-278.

Babcock, R. 1985. Growth and Mortality in Juvenile Corals: The First
Year. Proceedings of the 8th International Coral Reef Symposium,
2:1197-1202.

. 1995. Synchronous Multispecific Spawning on Coral Reefs: Po-
tential for Hybridization and Roles of Gamete Recognition. Repro-
duction, Fertility and Development, 7:943-950.

Babcock, R., and C. Mundy. 1996. Coral Recruitment: Consequences
of Settlement Choice for Early Growth and Survivorship in Two
Scleractinians. Journal of Experimental Marine Biology and Ecol-
ogy, 206:179-201.

Babcock, R. C. and A. J. Heyward. 1986. Larval Development of
Certain Gamete-Spawning Scleractinian Corals. Coral Reefs,
5:111-116.

Baird, A. H., R.C. Babcock, and C. P. Mundy. 2003. Habitat Selection
by Larvae Influences the Depth Distribution of Six Common Coral
Species. Marine Ecology Progress Series, 252:289-293.

Baird, A. H., J. Gilmour, T. M. Kamiki, M. Nonada, M. S. Pratchett,
H. H. Yamamoto, and H. Yamasaki. 2006. Temperature Tolerance
of Symbiotic and Non-symbiotic Coral Larvae. Proceedings of the
10th International Coral Reef Symposium, 1:351-358.

Baird, A. H., and T. P. Hughes. 2000. Competitive Dominance by Tabu-
lar Corals: An Experimental Analysis of Recruitment and Survival
of Understory Assemblages. Journal of Experimental Marine Biol-
ogy and Ecology, 251:117-132.

Baird, A. H., and A. N. C. Morse. 2004. Induction of Metamorphosis
in Larvae of the Brooding Corals Acropora palifera and Stylophora
pistillata. Marine and Freshwater Research, 55:469-472.

Bak, R. P. M., and J. L. A. Borsboom. 1984. Allelopathic Interaction
between a Reef Coelenterate and Benthic Algae. Oecologia (Berlin),
63:194-198.

Bak, R. P. M., and M. S. Engel. 1979. Distribution, Abundance and Sur-
vival of Juvenile Hermatypic Corals (Scleractinia) and the Impor-
tance of Life History Strategies in the Parent Coral Community.
Marine Biology, 54:341-352.

Bassim, K. M., and P. W. Sammarco. 2003. Effects of Temperature and
Ammonium on Larval Development and Survivorship in a Sclerac-
tinian Coral (Diploria strigosa). Marine Biology, 142:241-252.

Baums, I. B., M. W. Miller, and M. E. Hellberg. 2005. Regionally Iso-
lated Populations of an Imperiled Caribbean Coral, Acropora pal-
mata. Molecular Ecology, 14:1377-1390.

. 2006. Geographic Variation in Clonal Structure in a Reef-
Building Caribbean Coral, Acropora palmata. Ecological Mono-
graphs, 76:503-519.

Bennett, C. E., and D. J. Marshall. 2005. The Relative Energetic Costs
of the Larval Period, Larval Swimming and Metamorphosis for the
Ascidian Diplosoma listerianum. Marine and Freshwater Behav-
iour and Physiology, 38:21-29.

Birkeland, C. 1977. The Importance of Rate of Biomass Accumulation in
Early Successional Stages of Benthic Communities to the Survival of
Coral Recruits. In Proceedings of the 3rd International Coral Reef
Symposium, Miami, 1:15-21.

Birrell, C. L., L. J. McCook, and B. L. Willis. 2005. Effects of Algal
Turfs and Sediment on Coral Settlement. Marine Pollution Bulletin,
51:408-414.

Birrell, C. L., L. J. McCook, B. L. Willis, and L. Harrington. 2008a.
Chemical Effects of Macroalgae on Larval Settlement of the Broad-
cast Spawning Coral Acropora millepora. Marine Ecology Progress
Series, 362:129-137.

Birrell, C. L., L. J. McCook, B. L. Willis, and G. A. Diaz-Pulido. 2008b.
Effects of Benthic Algae on the Replenishment of Corals and the
Implications for the Resilience of Coral Reefs. Oceanography and
Marine Biology: An Annual Review, 46:25-63.

Box, S. J., and P. J. Mumby. 2007. Effect of Macroalgal Competition on
Growth and Survival of Juvenile Caribbean corals. Marine Ecology
Progress Series, 342:139-149.

Bruno, J. FE, and E. R. Selig. 2007. Regional Decline of Coral Cover
in the Indo-Pacific: Timing, Extent, and Subregional Comparisons.
PLOS One, 8:1-8.

Carlon, D. B. 2001. Depth-Related Patterns of Coral Recruitment and
Cryptic Suspension-Feeding Invertebrates on Guana Island, British
Virgin Islands. Bulletin of Marine Science, 68:525-541.

. 2002. Production and Supply of Larvae as Determinants of Zo-
nation in a Brooding Tropical Coral. Journal of Experimental Ma-
rine Biology and Ecology, 268:33-46.

Carlon, D. B., and R. R. Olson. 1993. Larval Dispersal Distance as
an Explanation for Adult Spatial Pattern in Two Caribbean Reef
Corals. Journal of Experimental Marine Biology and Ecology,
173:247-263.

Carpenter, R. C. 1986. Partitioning Herbivory and Its Effects on Coral
Reef Algal Communities. Ecological Monographs, 56:343-363.

Chiu, J. M. Y., T. Y. T. Ng, W. X. Wang, V. Thiyagarajan, and P. Y. Qian.
2007. Latent Effects of Larval Food Limitation on Filtration Rate,
Carbon Assimilation and Growth in Juvenile Gastropod Crepidula
onyx. Marine Ecology Progress Series, 343:173-182.

Chiu, J. M. Y., H. Wang, V. Thiyagarajan, and P. Y. Qian. 2008. Dif-
ferential Timing of Larval Starvation Effects on Filtration Rate and
Growth in Juvenile Crepidula onyx. Marine Biology, 154:91-98.

Chornesky, E. A., and E. C. Peters. 1987. Sexual Reproduction and Col-
ony Growth in the Scleractinian Coral Porites astreoides. Biological
Bulletin, 172:161-177.

Coll, J. C., B. F Bowden, G. V. Meehan, G. M. Konig, A. R. Carroll,
D. M. Tapiolas, P. M. Alifo, A. Heaton, R. De Nys, P. A. Leone,
M. Maida, T. L. Aceret, R. H. Willis, R. C. Babcock, B. L. Willis,
Z. Florian, M. N. Clayton, and R. L. Miller. 1994. Chemical As-
pects of Mass Spawning in Corals. I. Sperm-Attractant Molecules
in the Eggs of the Scleractinian Coral Montipora digitata. Marine
Biology, 118:117-182.

Coma, R., and H. R. Lasker. 1997. Effects of Spatial Distribution and
Reproductive Biology on In Situ Fertilization Rates of a Broadcast-
Spawning Invertebrate. Biological Bulletin, 193:20-29.

Connell, J. H. 1997. Disturbance and Recovery of Coral Assemblages.
Coral Reefs, 16:5101-S113.

Connell, J. H., T. P. Hughes, and C. C. Wallace. 1997. A 30-Year Study of
Coral Abundance, Recruitment, and Disturbance at Several Scales
in Space and Time. Ecological Monographs, 67:461-488.

Cowen, R. K., K. M. M. Lwiza, S. Sponaugle, C. B. Paris, and D. B. Olson.
2000. Connectivity of Marine Populations: Open or Closed? Sci-
ence, 287:857-859.

Cowen, R. K., C. B. Paris, and A. Srinivasan. 2006. Scaling of Connectiv-
ity in Marine Populations. Science, 311:522-527.

de Ruyter van Steveninck, E. D., L. L. Van Mulekom, and A. M. Bree-
man. 1988. Growth Inhibition of Lobophora variegata (Lamour-
oux) Womersley by Scleractinian Corals. Journal of Experimental
Marine Biology and Ecology, 115:169-178.

Doherty, P., and T. Fowler. 1994. An Empirical Test of Recruitment Limi-
tation in a Coral Reef Fish. Science, 263:935-939.

Done, T. 1992. Phase Shifts in Coral Reef Communities and Their Eco-
logical Significance. Hydrobiologia, 247:121-132.

Downs, C. A., C. M. Woodley, R. H. Richmond, L. L. Lanning, and R.
Owen. 2005. Shifting the Paradigm of Coral-Reef ‘Health’ Assess-
ment. Marine Pollution Bulletin, 51:486-494.

NUMBER 38 e¢ 453

Edmunds, P. J. 2000. Patterns in the Distribution of Juvenile Corals and
Coral Reef Community Structure in St. John, US Virgin Islands.
Marine Ecology Progress Series, 202:113-124.

Edmunds, P. J., J. F Bruno, and D. B. Carlon. 2004. Effects of Depth and
Microhabitat on Growth and Survivorship of Juvenile Corals in the
Florida Keys. Marine Ecology Progress Series, 278:115-124.

Edmunds, P. J., and R. C. Carpenter. 2001. Recovery of Diadema an-
tillarum Reduces Macroalgal Cover and Increases Abundance
of Juvenile Corals on a Caribbean Reef. Proceedings of the Na-
tional Academy of Sciences of the United States of America, 98:
5067-5071.

Edmunds, P. J., R. D. Gates, and D. FE. Gleason. 2001. The Biology of
Larvae from the Reef Coral Porites astreoides, and Their Response
to Temperature Disturbances. Marine Biology, 139:981-989.

Edmunds, P. J., R. D. Gates, W. Leggat, O. Hoegh-Guldberg, and L. Allen-
Requa. 2005. The Effect of Temperature on the Size and Popula-
tion Density of Dinoflagellates in Larvae of the Reef Coral Porites
astreoides. Invertebrate Biology, 124:185-193.

Fabricius, K. E., and J. Metzner. 2004. Scleractinian Walls of Mouths:
Predation on Coral Larvae by Corals. Coral Reefs, 23:245-248.

Fadlallah, Y. H. 1983. Sexual Reproduction, Development and Larval
Biology in Scleractinian Corals. Coral Reefs, 2:129-150.

Fautin, D. G. 2003. Reproduction of Cnidaria. Canadian Journal of
Zoology, 80:1735-1754.

Fearon, R. J., and A. M. Cameron. 1996. Larvotoxic Extracts of the
Hard Coral Goniopora tenuidens: Allelochemicals That Limit Set-
tlement of Potential Competitors? Toxicon, 34:361-367.

Fenner, D., and K. Banks. 2004. Orange Cup Coral Tubastraea coccinea
Invades Florida and the Flower Garden Banks, Northwestern Gulf
of Mexico. Coral Reefs, 23:505-507.

Fine, M., H. Zibrowius, and Y. Loya. 2001. Oculina patagonica: A Non-
Lessepsian Scleractinian Coral Invading the Mediterranean Sea.
Marine Biology, 138:1195-1203.

Foster, N. L., S. J. Box, and P. J. Mumby. 2008. Competitive Effects of
Macroalgae on the Fecundity of the Reef-Building Coral Montas-
traea annularis. Marine Ecology Progress Series, 367:143-152.

Gaines, S., and J. Roughgarden. 1985. Larval Settlement Rate: A Lead-
ing Determinant of Structure in an Ecological Community of the
Marine Intertidal Zone. Proceedings of the National Academy of
Sciences of the United States of America, 82:3707-3711.

Gardner, T. A., I. M. Cote, J. A. Gill, A. Grant, and A. R. Watkinson.
2003. Long-term Region-wide Declines in Caribbean Corals. Sci-
ence, 301:958-960.

Gilmour, J. 1999. Experimental Investigation into the Effects of Sus-
pended Sediment on Fertilization, Larval Survival and Settlement in
a Scleractinian Coral. Marine Biology, 135:451-462.

Gleason, D. E, P. J. Edmunds, and R. D. Gates. 2006. Ultraviolet Radia-
tion Effects on the Behavior and Recruitment of Larvae from the
Reef Coral Porites astreoides. Marine Biology, 148:503-512.

Golbuu, Y., and R. H. Richmond. 2007. Substratum Preferences in Plan-
ula Larvae of Two Species of Scleractinian Corals, Goniastrea reti-
formis and Stylaraea punctata. Marine Biology, 152:639-644.

Golbuu, Y., S. Victor, L. Penland, D. Idip, C. Emaurois, K. Okaji, H.
Yukihira, A. Iwase, and R. van Woesik. 2007. Palau’s Coral Reefs
Show Differential Habitat Recovery Following the 1998 Bleaching
Event. Coral Reefs, 26:319-332.

Graham, E. M., A. H. Baird, and S. R. Connolly. 2008. Survival Dynam-
ics of Scleractinian Coral Larvae and Implications for Dispersal.
Coral Reefs, 27:529-539.

Green, D. H., P. J. Edmunds, and R. C. Carpenter. 2008. Increasing Rela-
tive Abundance of Porites astreoides on Caribbean Reefs Mediated
by an Overall Decline in Coral Cover. Marine Ecology Progress
Series, 359:1-10.

454 e

Guzman, H. M., K. A. Burns, and J. B. C. Jackson. 1994. Injury, Regen-
eration and Growth of Caribbean Reef Corals after a Major Oil
Spill in Panama. Marine Ecology Progress Series, 105:231-241.

Hadfield, M. G., and V. J. Paul. 2001. Natural Chemical Cues for Settle-
ment and Metamorphosis of Marine Invertebrate Larvae. In Marine
Chemical Ecology, ed. J. McClintock and B. Baker, pp. 431-462.
Boca Raton: CRC Press.

Halpern, B. S., S. Walbridge, K. A. Selkoe, C. V. Kappel, F. Micheli, C.
D’Agrosa, J. F Bruno, K. S. Casey, C. Ebert, H. E. Fox, R. Fujita, D.
Heinemann, H. S. Lenihan, E. M. P. Madin, M. T. Perry, E. R. Selig,
M. Spalding, R. Steneck, and R. Watson. 2008. A Global Map of
Human Impact on Marine Ecosystems. Science, 319:948-952.

Harrington, L., K. Fabricius, G. De’ath, and A. Negri. 2004. Recognition
and Selection of Settlement Substrata Determine Post-settlement
Survival in Corals. Ecology, 85:3428-3437.

Harriott, V. J. 1983. Reproductive Seasonality, Settlement, and Post-
settlement Mortality of Pocillopora damicornis (Linnaeus), at Liz-
ard Island, Great Barrier Reef. Coral Reefs, 2:151-157.

Harrison, P. L., R. C. Babcock, G. D. Bull, J. K. Oliver, C. C. Wallace,
and B. L. Willis. 1984. Mass Spawning in Tropical Reef Corals.
Science, 223:1186-1189.

Hatta, M., H. Fukami, W. Q. Wang, M. Omori, K. Shimoike, T. Hay-
ashibara, Y. Ina, and T. Sugiyama. 1999. Reproductive and Genetic
Evidence for a Reticulate Evolutionary History of Mass-Spawning
Corals. Molecular Biology and Evolution, 16:1607-1613.

Heintz, R. A.,S. D. Rice, A. C. Wertheimer, R. F. Bradshaw, F. P. Thrower,
J. E. Joyce, and J. W. Short. 2000. Delayed Effects on Growth and
Marine Survival of Pink Salmon Oncorhynchus gorbuscha after
Exposure to Crude Oil During Embryonic Development. Marine
Ecology Progress Series, 208:205-216.

Hellberg, M. E. 2007. Footprints on Water: The Genetic Wake of Disper-
sal among Reefs. Coral Reefs, 26:463-473.

Heyward, A. J., and A. P. Negri. 1999. Natural Inducers for Coral Larval
Metamorphosis. Coral Reefs, 18:273-279.

Hodgson, G. 1990. Sediment and the Settlement of Larvae of the Reef
Coral Pocillopora damicornis. Coral Reefs, 9:41-43.

Hoegh-Guldberg, O., P. J. Mumby, A. J. Hooten, R. S. Steneck, P. Green-
field, E. Gomez, C. D. Harvell, P. E Sale, A. J. Edwards, K. Cal-
deira, N. Knowlton, C. M. Eakin, R. Iglesias-Prieto, N. Muthiga,
R. H. Bradbury, A. Dubi, and M. E. Hatziolos. 2007. Coral Reefs
under Rapid Climate Change and Ocean Acidification. Science,
318:1737-1742.

Holling, C. S. 1973. Resilience and Stability of Ecological Systems. An-
nual Review of Ecology and Systematics, 4:1-23.

Hughes, T. 1984. Population Dynamics Based on Individual Size Rather
Than Age: A General Model with a Reef Coral Example. American
Naturalist, 123:778-795.

Hughes, T., and J. B. C. Jackson. 1985. Population Dynamics and Life
Histories of Foliaceous Corals. Ecological Monographs, 55:141-
166.

Hughes, T. P. 1994. Catastrophes, Phase-Shifts, and Large-Scale Degra-
dation of a Caribbean Coral Reef. Science, 265:1547-1551.

Hughes, T. P., A. H. Baird, E. A. Dinsdale, N. A. Moltschaniwkyj, M. S.
Pratchett, J. E. Tanner, and B. L. Willis. 1999. Patterns of Recruit-
ment and Abundance of Corals along the Great Barrier Reef.
Nature (London), 397:59-63.

Hughes, T. P., D. R. Bellwood, C. Folke, R. Steneck, and J. Wilson. 2005.
New Paradigms for Supporting the Resilience of Marine Ecosys-
tems. Trends in Ecology and Evolution, 20:380-386.

Hughes, T. P., M. J. Rodrigues, D. R. Bellwood, D. Ceccarelli, O.
Hoegh-Guldberg, L. McCook, N. Moltschaniwskyj, M. S. Pratch-
ett, R. S. Steneck, and B. Willis. 2007. Phase Shifts, Herbivory, and
the Resilience of Coral Reefs to Climate Change. Current Biology,
17:360-365.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Hughes, T. P., and J. E. Tanner. 2000. Recruitment Failure, Life His-
tories, and Long-Term Decline of Caribbean Corals. Ecology,
81:2250-2263.

Humphrey, C., M. Weber, C. Lott, T. Cooper, and K. Fabricius. 2008.
Effects of Suspended Sediments, Dissolved Inorganic Nutrients and
Salinity on Fertilisation and Embryo Development in the Coral
Acropora millepora (Ehrenberg, 1834). Coral Reefs, 27:837-850.

Idjadi, J. A., S. C. Lee, J. F Bruno, W. EF. Precht, L. Allen-Requa, and P. J.
Edmunds. 2006. Rapid Phase-Shift Reversal on a Jamaican Coral
Reef. Coral Reefs, 25:209-211.

Jackson, J. B. C., and L. Buss. 1975. Allelopathy and Spatial Competition
among Coral Reef Invertebrates. Proceedings of the National Acad-
emy of Sciences of the United States of America, 72:5160-5163.

Johnson, C. R., C. D. Muir, and A.-L. Reysenback. 1991. Character-
istic Bacteria Associated with the Surfaces of Coralline Algae: A
Hypothesis for Bacterial Induction of Marine Invertebrate Larvae.
Marine Ecology Progress Series, 74:281-294.

Jompa, J., and L. J. McCook. 2003. Contrasting Effects of Turf Algae
on Corals: Massive Porites spp. Are Unaffected by Mixed-Species
Turfs, But Killed by the Red Alga Anotrichium tenue. Marine Ecol-
ogy Progress Series, 258:79-86.

Jones, C. G., J. H. Lawton, and M. Shachak. 1994. Organisms as Ecosys-
tem Engineers. Oikos, 69:373-386.

. 1997. Positive and Negative Effects of Organisms as Physical
Ecosystem Engineers. Ecology, 78:1946-1957.

Kline, D. I., N. M. Kuntz, M. Breitbart, N. Knowlton, and EF. Rohwer.
2006. Role of Elevated Organic Carbon Levels and Microbial
Activity in Coral Mortality. Marine Ecology Progress Series,
314:119-125.

Knowlton, N. 2001. The Future of Coral Reefs. Proceedings of the
National Academy of Sciences of the United States of America,
98:5419-5425.

Koop, K., D. Booth, A. Broadbent, J. Brodie, D. Bucher, D. Capone,
J. Coll, W. Dennison, M. Erdmann, P. Harrison, O. Hoegh-Guldberg,
P. Hutchings, G. B. Jones, A. W. D. Larkum, J. O’Neil, A. Ste-
ven, E. Tentori, S. Ward, J. Williamson, and D. Yellowlees. 2001.
ENCORE: The Effect of Nutrient Enrichment on Coral Reefs.
Synthesis of Results and Conclusions. Marine Pollution Bulletin,
42:91-120.

Kramer, P. A. 2003. Synthesis of Coral Reef Health Indicators for the
Western Atlantic: Results of the AGRRA Program (1997-2000).
Atoll Research Bulletin, 496:1-58.

Kuffner, I. B. 2001. Effects of Ultraviolet Radiation on Larval Recruit-
ment of the Reef Coral, Pocillopora damicornis. Marine Ecology
Progress Series, 217:251-261.

Kuffner, I. B., and V. J. Paul. 2004. Effects of the Benthic Cyanobacte-
rium Lyngbya majuscula on the Larval Settlement of the Reef Cor-
als Acropora surculosa and Pocillopora damicornis. Coral Reefs,
23:455-458.

Kuffner, I. B., L. J. Walters, M. A. Becerro, V. J. Paul, R. Ritson-Williams,
and K. Beach. 2006. Inhibition of Coral Recruitment by Macroalgae
and Cyanobacteria. Marine Ecology Progress Series, 323:107-117.

Lee, S. C. 2006. Habitat Complexity and Consumer-Mediated Positive
Feedbacks on a Caribbean Coral Reef. Oikos, 112:442-447.

Levin, L. A. 2006. Recent Progress in Understanding Larval Dispersal:
New Directions and Digressions. Integrative and Comparative Bi-
ology, 46:282-297.

Levitan, D. R., H. Fukami, J. Jara, D. Kline, T. M. McGovern, K. E.
McGhee, C. A. Swanson, and N. Knowlton. 2004. Mechanisms
of Reproductive Isolation among Sympatric Broadcast-Spawning
Corals of the Montastraea annularis Species Complex. Evolution,
58:308-323.

Levitan, D. R., and T. M. McGovern. 2005. The Allee Effect in the Sea.
Washington, D.C.: Island Press.

Levitan, D. R., M. A. Sewell, and E. S. Chia. 1992. How Distribution
and Abundance Influence Fertilization Success in the Sea Urchin
Strongylocentrotus franciscanus. Ecology, 73:248-254.

Lewis, J. B. 1974. The Settlement Behavior of Planulae Larvae of the
Hermatypic Coral Favia fragum (Esper). Journal of Experimental
Marine Biology and Ecology, 15:165-172.

Lirman, D. 2001. Competition Between Macroalgae and Corals: Effects
of Herbivore Exclusion and Increased Algal Biomass on Coral Sur-
vivorship and Growth. Coral Reefs, 19:392-399.

Littler, M. M., and D. S. Littler. 1997. Disease-Induced Mass Mortal-
ity of Crustose Algae on Coral Reefs Provides Rationale for the
Conservation of Herbivorous Fish Stocks. Proceedings of the 8th
International Coral Reef Symposium, 1:719-724.

Macintyre, I. G., P. W. Glynn, and F. Hinds. 2005. Evidence of the Role
of Diadema antillarum in the Promotion of Coral Settlement and
Survivorship. Coral Reefs, 24:273.

Maida, M., J. C. Coll, and P. W. Sammarco. 1994. Shedding New Light
on Scleractinian Coral Recruitment. Journal of Experimental Ma-
rine Biology and Ecology, 180:189-202.

Maida, M., P. W. Sammarco, and J. C. Coll. 1995. Effects of Soft Corals on
Scleractinian Coral Recruitment. I: Directional Allelopathy and Inhi-
bition of Settlement. Marine Ecology Progress Series, 121:191-202.

Marshall, D. J., J. A. Pechenik, and M. J. Keough. 2003. Larval Activity
Levels and Delayed Metamorphosis Affect Post-larval Performance
in the Colonial Ascidian Diplosoma listerianum. Marine Ecology
Progress Series, 246:153-162.

Maypa, A. P., and L. J. Raymundo. 2004. Algae-Coral Interactions. Me-
diation of Coral Settlement, Early Survival, and Growth by Macro-
algae. Silliman Journal, 45:76-95.

Mayr, E. 1963. Animal Species and Evolution. Cambridge, Maine:
Belknap Press.

McCook, L. J., J. Jompa, and G. Diaz-Pulido. 2001. Competition Be-
tween Corals and Algae on Coral Reefs: A Review of Evidence and
Mechanisms. Coral Reefs, 19:400-417.

Miller, J., R. Waara, E. Muller, and C. Rogers. 2006. Coral Bleaching
and Disease Combine to Cause Extensive Mortality on Reefs in the
US Virgin Islands. Coral Reefs, 25:418.

Miller, K., and C. Mundy. 2003. Rapid Settlement in Broadcast Spawning
Corals: Implications for Larval Dispersal. Coral Reefs, 22:99-106.

Miller, S. E. 1993. Larval Period and Its Influence on Postlarval Life
History: Comparison of Lecithotrophy and Facultative Plankto-
trophy in the Aeolid Nudibranch Phestilla sibogae. Marine Biol-
ogy, 117:635-645.

Morse, A. N. C., K. Iwao, M. Baba, K. Shimoike, T. Hayashibara, and
M. Omori. 1996. An Ancient Chemosensory Mechanism Brings
New Life to Coral Reefs. Biological Bulletin, 191:149-154.

Morse, D. E., and A. N. C. Morse. 1991. Enzymatic Characterization
of the Morphogen Recognized by Agaricia humilis (Scleractinian
Coral) Larvae. Biological Bulletin, 181:104-122.

Morse, D. E., A. N. C. Morse, P. T. Raimondi, and N. Hooker. 1994.
Morphogen-Based Chemical Flypaper for Agaricia humilis Coral
Larvae. Biological Bulletin, 186:172-181.

Mumby, P. J., C. P. Dahlgren, A. R. Harborne, C. V. Kappel, FE. Micheli,
D. R. Brumbaugh, K. E. Holmes, J. M. Mendes, K. Broad, J. N.
Sanchirico, K. Buch, S. Box, R. W. Stoffle, and A. B. Gill. 2006.
Fishing, Trophic Cascades, and the Process of Grazing on Coral
Reefs. Science, 311:98-101.

Mumby, P. J., and R. Steneck. 2008. Coral Reef Management and Con-
servation in the Light of Rapidly Evolving Ecological Paradigms.
Trends in Ecology and Evolution, 23:555-563.

Mundy, C. N., and R. C. Babcock. 1998. Role of Light Intensity and Spec-
tral Quality in Coral Settlement: Implications for Depth-Dependent
Settlement? Journal of Experimental Marine Biology and Ecology,
223:235-255.

NUMBER 38 ¢ 455

Negri, A. P., N. S$. Webster, R. T. Hill, and A. J. Heyward. 2001. Meta-
morphosis of Broadcast Spawning Corals in Response to Bacteria
Isolated from Crustose Algae. Marine Ecology Progress Series,
223:121-131.

Nishikawa, A., M. Katoh, and K. Sakai. 2003. Larval Settlement Rates
and Gene Flow of Broadcast-Spawning (Acropora tenuis) and Plan-
ula-Brooding (Stylophora pistillata) Corals. Marine Ecology Prog-
ress Series, 256:87-97.

Nozawa, Y., and P. L. Harrison. 2008. Temporal Patterns of Larval Set-

tlement and Survivorship of Two Broadcast-Spawning Acroporid

Corals. Marine Biology 155:347-351.

Nugues, M. M., G. W. Smith, R. J. Van Hooindonk, M. I. Seabra, and

R. P. M. Bak. 2004. Algal Contact as a Trigger for Coral Disease.

Ecology Letters, 7:919-923.

Nugues, M. M., and A. M. Szmant. 2006. Coral Settlement onto Hali-

meda opuntia: A Fatal Attraction to an Ephemeral Substrate? Coral

Reefs, 25:585-591.

Nystrom, M., C. Folke, and EF. Moberg. 2000. Coral Reef Disturbance
and Resilience in a Human-Dominated Environment. Trends in
Ecology and Evolution, 15:413-417.

Oliver, J., and R. Babcock. 1992. Aspects of the Fertilization Ecol-
ogy of Broadcast Spawning Corals: Sperm Dilution Effects and
In Situ Measurements of Fertilization. Biological Bulletin, 183:
409-417.

Omori, M., H. Fukami, H. Kobinata, and M. Hatta. 2001. Significant
Drop of Fertilization of Acropora Corals in 1999. An After Ef-
fect of Heavy Coral Bleaching? Limnology and Oceanography,
46:704-706.

Omori, M., K. Iwao, and M. Tamura. 2008. Growth of Transplanted
Acropora tenuis 2 Years after Egg Culture. Coral Reefs, 27:165.

Palma, A. T., R. S. Steneck, and C. J. Wilson. 1999. Settlement-Driven,
Multiscale Demographic Patterns of Large Benthic Decapods in the
Gulf of Maine. Journal of Experimental Marine Biology and Ecol-
ogy, 241:107-136.

Palumbi, S. R. 1994. Genetic Divergence, Reproductive Isolation, and
Marine Speciation. Annual Review of Ecology and Systematics,
25:547-572.

Pandolf, J. M., and J. B. C. Jackson. 2006. Ecological Persistence Inter-
rupted in Caribbean Coral Reefs. Ecology Letters, 9:818-826.
Pandolfi, J. M., J. B. C. Jackson, N. Baron, R. H. Bradbury, H. M. Guz-
man, T. P. Hughes, C. V. Kappel, F. Micheli, J. C. Ogden, H. P. Pos-
singham, and E. Sala. 2005. Are U.S. Coral Reefs on the Slippery

Slope to Slime? Science, 307:1725-1726.

Pantos, O., R. P. Cooney, M. D. A. Le Tissier, M. R. Barer, A. G.
O’Donnell, and J. C. Bythell. 2003. The Bacterial Ecology of a
Plague-like Disease Affecting the Caribbean Coral Montastraea an-
nularis. Environmental Microbiology, 5:370-382.

Pawlik, J. R. 1992. Chemical Ecology of the Settlement of Benthic Ma-
rine Invertebrates. Oceanography and Marine Biology: An Annual
Review, 30:273-335.

Pawlik, J. R., L. Steindler, T. P. Henkel, S. Beer, and M. Ilan. 2007.
Chemical Warfare on Coral Reefs: Sponge Metabolites Differen-
tially Affect Coral Symbiosis In Situ. Limnology and Oceanogra-
phy, 52:907-911.

Pechenik. J. A. 1990. Delayed Metamorphosis by Larvae of Benthic Ma-
rine Invertebrates. Does It Occur? Is There a Price to Pay? Ophelia,
32:63-94.

. 2002. Relationships Between Larval Nutritional Experience,

Larval Growth Rates, Juvenile Growth Rates, and Juvenile Feeding

Rates in the Prosobranch Gastropod Crepidula fornicata. Journal

of Experimental Marine Biology and Ecology, 280:63-78.

. 2006. Larval Experience and Latent Effects: Metamorphosis

Is Not a New Beginning. Integrative and Comparative Biology,

46:323-333.

456 e

Pechenik, J. A., R. Berard, D. D. Gleason, T. R. Gleason, and D. Cham-
plin. 2001. Influence of Lowered Salinity and Elevated Cadmium
on the Survival and Metamorphosis of Trochophores in Capitella
sp. Invertebrate Biology, 120:142-148.

Pechenik, J. A., D. E- Wendt, and J. N. Jarrett. 1998. Metamorphosis Is
Not a New Beginning. Bioscience, 48:901-910.

Putnam, H. M., P. J. Edmunds, and T.-Y. Fan. 2008. Effect of Tem-
perature on the Settlement Choice and Photophysiology of Lar-
vae from the Reef Coral Stylophora pistillata. Biological Bulletin,
215:135-142.

Raimondi, P. T., and A. N. C. Morse. 2000. The Consequences of Com-
plex Larval Behavior in a Coral. Ecology, 81:3193-3211.

Richmond, R. H. 1987. Energetics, Competency, and Long-Distance
Dispersal of Planula Larvae of the Coral Pocillopora damicornis.
Marine Biology, 93:527-533.

. 1993a. Coral Reefs: Present Problems and Future Concerns

Resulting from Anthropogenic Disturbance. American Zoologist,

33:524-536.

. 1993b. Effects of Coastal Runoff on Coral Reproduction. In

Proceedings of the Colloquium and Forum on Global Aspects of

Coral Reefs: Health, Hazard, and History, ed. R. N. Ginsburg,

pp. 360-364. University of Miami, Miami, Fla.

. 1997. Reproduction and Recruitment in Corals: Critical Links in
the Persistence of Reefs. In Life and Death of Coral Reefs, ed. C. E.
Birkeland, pp. 175-198. New York: Chapman & Hall.

Richmond, R. H., and C. L. Hunter. 1990. Reproduction and Recruitment
of Corals: Comparisons among the Caribbean, the Tropical Pacific,
and the Red Sea. Marine Ecology Progress Series, 60:185-203.

Richmond, R. H., T. Rongo, Y. Golbuu, S. Victor, N. Idechong, G. Davis,
W. Kostka, L. Neth, M. Hamnett, and E. Wolanski. 2007. Water-
sheds and Coral Reefs: Conservation Science, Policy, and Imple-
mentation. Bioscience, 57:598-607.

Rinkevich, B., and Y. Loya. 1987. Variability in the Pattern of Sexual
Reproduction of the Coral Stylophora pistillata at Eilat, Red-Sea: A
Long-Term Study. Biological Bulletin, 173:335-344.

Roberts, R. D., and C. Lapworth. 2001. Effect of Delayed Metamorpho-
sis on Larval Competence, and Post-Larval Survival and Growth, in
the Abalone Haliotis iris Gmelin. Journal of Experimental Marine
Biology and Ecology, 258:1-13.

Rylaarsdam, K.W. 1983. Life Histories and Abundance Patterns of Co-
lonial Corals on Jamaican Reefs. Marine Ecology Progress Series,
13:249-260.

Sammarco, P. W. 1985. The Great Barrier Reef Versus the Caribbean:
Comparisons of Grazers, Coral Recruitment Patterns and Reef Re-
covery. In Proceedings of the Sth International Coral Reef Sympo-
sium, 4:391-398.

Sammarco, P. W., and J. C. Andrews. 1988. Localized Dispersal and Re-
cruitment in Great Barrier Reef Corals: The Helix Experiment. Sci-
ence, 239:1422-1424.

Sammarco, P. W., and J. H. Carleton. 1981. Damselfish Territoriality and
Coral Community Structure: Reduced Grazing, Coral Recruitment,
and Effects on Coral Spat. In: Proceedings of the 4th International
Coral Reef Symposium, 2:525-535.

Sebens, K. P. 1982. Competition for Space: Growth Rate, Reproductive
Output, and Escape in Size. American Naturalist, 120:189-197.

Smith, J. E., M. Shaw, R. A. Edwards, D. Obura, O. Pantos, E. Sala,
S. A. Sandin, S. Smriga, M. Hatay, and F. L. Rohwer. 2006. Indirect
Effects of Algae on Coral: Algae-Mediated, Microbe-Induced Coral
Mortality. Ecology Letters, 9:835-845.

Smith, S. R. 1992. Patterns of Coral Recruitment and Post-settlement
Mortality on Bermuda’s Reefs: Comparisons to Caribbean and Pa-
cific Reefs. American Zoologist, 32:663-673.

. 1997. Patterns of Coral Settlement, Recruitment and Juvenile

Mortality with Depth at Conch Reef, Florida. Proceedings of the

8th International Coral Reef Symposium, 2:1197-1202.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Steneck, R. S. 2006. Staying Connected in a Turbulent World. Science,
311:480-481.

Steneck, R. S., and J. C. Lang. 2003. Rapid Assessment of Mexico’s Yu-
catan Reef in 1997 and 1999: Pre- and Post-Mass Bleaching and
Hurricane Mitch (Stony Corals, Algae and Fish). Atoll Research
Bulletin, 496:294-317.

Steneck, R. S., and M. N. Dethier. 1994. A Functional Group Ap-
proach to the Structure of Algal-Dominated Communities. Oikos,
69:476-498.

Strathmann, R. R., T. R. Hughes, A. M. Kuris, K. C. Lindeman, S. G.
Morgan, J. M. Pandolfi, and R. R. Warner. 2002. Evolution of
Local Recruitment and Its Consequences for Marine Populations.
Bulletin of Marine Science, 70:377-396.

Styan, C. A. 1998. Polyspermy, Egg Size, and the Fertilization Kinet-
ics of Free-Spawning Marine Invertebrates. American Naturalist,
152:290-297.

Sullivan, B., D. J. Faulkner, and L. Webb. 1983. Siphonodictidine, a Me-
tabolite of the Burrowing Sponge Siphonodictyon sp. That Inhibits
Coral Growth. Science, 221:1175-1176.

Suzuki, G., T. Hayashibara, Y. Shirayama, and H. Fukami. 2008. Evi-
dence of Species-Specific Habitat Selectivity of Acropora Corals
Based on Identification of New Recruits by Two Molecular Mark-
ers. Marine Ecology Progress Series, 355:149-159.

Szmant, A. M. 1986. Reproductive Ecology of Caribbean Reef Corals.
Coral Reefs, 5:43-53.

Szmant, A. M., and N. J. Gassman. 1990. The Effects of Prolonged
Bleaching on the Tissue Biomass and Reproduction of the Reef
Coral Montastraea annularis. Coral Reefs, 8:217-224.

Thiyagarajan, V., L. A. Gosselin, and P. Y. Qian. 2008. Juvenile Growth
in Barnacles: Combined Effect of Delayed Metamorphosis and Sub-
lethal Exposure of Cyprids to Low-Salinity Stress. Marine Ecology
Progress Series, 344:173-184.

Thiyagarajan, V., T. Harder, J. W. Qiu, and P. Y. Qian. 2003. Energy
Content at Metamorphosis and Growth Rate of the Early Juvenile
Barnacle Balanus amphitrite. Marine Biology, 143:543-554.

Tomaiuolo, M., T. F. Hansen, and D. R. Levitan. 2007. A Theoretical
Investigation of Sympatric Evolution of Temporal Reproductive
Isolation as Illustrated by Marine Broadcast Spawners. Evolution,
61:2584-2595.

Van Moorsel, G. W. N. M. 1989. Juvenile Ecology and Reproductive
Strategies of Reef Corals. Leiden: Backhuys Publishers.

Vermeij, M. J. A. 2005. Substrate Composition and Adult Distribution
Determine Recruitment Patterns in a Caribbean Brooding Coral.
Marine Ecology Progress Series, 295:123-133.

. 2006. Early Life-History Dynamics of Caribbean Coral Species
on Artificial Substratum: The Importance of Competition, Growth
and Variation in Life-History Strategy. Coral Reefs, 25:59-71.

Vermeij, M. J. A., N. D. Fogarty, and M. W. Miller. 2006. Pelagic Condi-
tions Affect Larval Behavior, Survival, and Settlement Patterns in
the Caribbean Coral Montastraea faveolata. Marine Ecology Prog-
ress Series, 310:119-128.

Vermeij, M. J. A., and S. A. Sandin. 2008. Density-Dependent Settlement
and Mortality Structure: The Earliest Life Phases of a Coral Popula-
tion. Ecology, 89:1994-2004.

Vermeij, M. J. A., J. E. Smith, C. M. Smith, R. V. Thurber, and S. A.
Sandin. 2009. Survival and Settlement Success of Coral Planulae:
Independent and Synergistic Effects of Macroalgae and Microbes.
Oecologia, 159:325-336.

Wakeford, M., T. J. Done, and C. R. Johnson. 2008. Decadal Trends in a
Coral Community and Evidence of Changed Disturbance Regime.
Coral Reefs, 27:1-16.

Webster, N. S., L. D. Smith, A. J. Heyward, J. E. M. Watts, R. I. Webb,
L. L. Blackall, and A. P. Negri. 2004. Metamorphosis of a Sclerac-
tinian Coral in Response to Microbial Biofilms. Applied and Envi-
ronmental Microbiology, 70:1213-1221.

Wendt, D. E. 1998. Effect of Larval Swimming Duration on Growth and
Reproduction of Bugula neritina (Bryozoa) under Field Conditions.
Biological Bulletin, 195:126-135.

Williams, D. E., M. W. Miller, and K. L. Kramer. 2008. Recruitment
Failure in Florida Keys Acropora palmata, a Threatened Caribbean
Coral. Coral Reefs, 27:697-705.

Willis, B. L., R. Babcock, P. L. Harrison, and T. K. Oliver. 1985. Pat-
terns in Mass Spawning of Corals on the Great Barrier Reef from
1981 to 1984. In Proceedings of the Sth International Coral Reef
Symposium, 4:343-348.

Willis, B. L., R. C. Babcock, P. L. Harrison, and C. C. Wallace. 1997.
Experimental Hybridization and Breeding Incompatibilities within
the Mating Systems of Mass Spawning Reef Corals. Coral Reefs,
16:S53-S65.

NUMBER 38 °¢ 457

Willis, B. L., M. van Oppen, D. J. Miller, S. V. Vollmer, and D. J.
Ayre. 2006. The Role of Hybridization in the Evolution of Reef
Corals. Annual Review of Ecology, Evolution, and Systematics,
37:489-517.

Wittenberg, M., and W. Hunte. 1992. Effects of Eutrophication and Sed-
imentation on Juvenile Corals I. Abundance, Mortality and Com-
munity Structure. Marine Biology, 112:131-138.

Wolstenholme, J. K. 2004. Temporal Reproductive Isolation and Ga-
metic Compatibility Are Evolutionary Mechanisms in the Acropora
humilis Species Group (Cnidaria; Scleractinia). Marine Biology,
144:567-582.

Zigler, K. S., M. A. McCartney, D. R. Levitan, and H. A. Lessios. 2005.
Sea Urchin Bindin Divergence Predicts Gamete Compatibility. Evo-
lution, 59:2399-2404.

LJ Me RE oa 4 7

ef es
i aa
Se - se : ee
. 4 |: ly Bat 2 eg
mead iain uly 2 lee al lm
. Se unin ake my lees
oY abe om ae .
; } ~) i Ji = Dba eS
i hainian or)
7 2 a v |
fe ‘6
es Fe:
: 5 ous > iu ieee :
i ibaa, ag
oy ; :
,
t u
im a
7 = hae i ee)
ole
i ae
: £ i a . as
> ve :

Do Indian River Lagoon Wetland
Impoundments (Eastern Florida)
Negatively Impact Fiddler Crab
(Genus Uca) Populations?

Bjorn G. Tunberg

Bjorn G. Tunberg, Smithsonian Marine Station,
701 Seaway Drive, Fort Pierce, Florida 34949,
USA (Tunberg@si.edu). Manuscript received 13
May 2008; accepted 20 April 2009.

ABSTRACT. Quantitative sampling of fiddler crabs was performed in June-July be-
tween 1992 and 1994 along transects at three St. Lucie County mosquito impoundments,
Florida, running from the Indian River Lagoon (IRL) shore and across the impoundment
perimeter dikes, and in one impoundment across the perimeter ditch. A total of 929 spec-
imens representing four species were found: Uca pugilator, Uca rapax, Uca speciosa, and
Uca thayeri. The quantitative sampling showed that there was no correlation between the
number of Uca burrow openings on the sediment surface and the actual number of crabs
in the sediment. Differences were recorded in abundance and distributional patterns be-
tween impoundments, but no correlation was recorded between substrate organic con-
tent and species distributional patterns. The male/female ratio was close to 1 for all spe-
cies, except for U. thayeri; the males dominated for this species (ratio, 1.8:1). High water
temperatures potentially lethal to fiddler crabs occurred in the impounded marsh in the
summer. U. pugilator and U. rapax were unlikely to be impacted by the impoundment
flooding as they are highly motile and not very site specific. U. speciosa and U. thayeri
were more restricted to the very soft, dark, and wet substrate along perimeter ditch banks
and may therefore be impacted during periods of flooding because they are dependent on
nonflooded areas for feeding and reproduction.

INTRODUCTION

Burrowing crustaceans, such as fiddler crabs, impact the ecology of associ-
ated infaunal communities and, consequently, the ecosystem as a whole (Crane,
1975; Montague, 1982; Dittman, 1996). According to Montague (1980), fid-
dler crabs are the most abundant macrobenthic crustacean inhabitants of North
American estuaries. Their impacts on bioturbation activity and oxygenation of
the substrate are considerable (Bertness, 1985). Fiddler crabs may also play an
important role in recycling nutrients (Macintosh, 1982; Bertness, 1985). They
feed by scraping up and ingesting surface sediment (Crane, 1975; Kraeuter, 1976;
Heard, 1982; Macintosh, 1982; Weis and Weis, 2004) and are in that respect
very important in overturn of substrates. Fiddler crabs are also an important
food source for birds, fish, and mammals (Peterson and Peterson, 1979; Mon-
tague, 1980; Grimes et al., 1989; Gilmore et al., 1990). There is a relatively di-
verse Uca species assemblage within the Indian River Lagoon (IRL) region, with
seven species reported in the IRL (Salmon, 1967; Kerr, 1976; M. Salmon, Florida

460 °

Atlantic University, personal communication, 1992), four
tropical species, Uca rapax (Smith), Uca thayeri Rathbun,
Uca speciosa (Ives), and Uca mordax (Smith), and three
temperate species, Uca pugnax (Smith), Uca pugilator
(Bosc), and Uca minax (Le Conte). Only four species were
found during these studies: Uca pugilator, U. rapax, U.
speciosa, and U. thayeri.

That the impoundment of 90% of the marginal wet-
lands (primarily for mosquito control) of the IRL has a
potential negative impact on regional Uca populations has
been a controversial issue for many years. Each impound-
ment and the management procedures are described in
detail in Rey and Kain (1991). Preliminary studies by Gil-
more et al. (1991) revealed that no Uca spp. were observed
from marsh-mangrove habitats in flooded (short-term and
long-term) impoundments, while they were present in large
numbers at unimpounded sites adjacent to impoundments.
This difference could be associated with a number of fac-
tors, because many aspects of the reproduction of Uca (in-
cluding courtship, female receptivity, egg maturation, and
hatching) are closely synchronized with the semidiurnal
and semilunar tidal cycles (Fingerman, 1957; Barnwell,
1968; Wheeler, 1978; Zucker, 1978; Montague, 1980; De-
Corsey, 1983; Salmon et al., 1986). However, according
to Fingerman (1957), the tidal rhythm differs between
species (U. pugilator and U. speciosa). The exclusion of
natural tidal cycles within several impoundments may
therefore have serious impacts on populations of Uca
spp. In addition, prolonged periods of inundation that
usually occur from May to September (management for
mosquito control) may displace Uca spp., which need
periods of exposure of the burrow entrances for sur-
vival. Periods of heavy precipitation, mainly during the
summer, may also drastically reduce the salinity within
these impoundments.

The main objectives of this study were (a) to evaluate
survival and adaptation of Uca populations to manipu-
lated ecological conditions along the impoundment perim-
eter ditches (compared with the natural IRL conditions),
(b) to determine if these potential adaptations differed
among species, and (c) to elucidate zonation patterns of
each species (from the IRL shore, across the dike road,
down to the impoundment ditch).

METHODS

Figure 1 shows the location of the three studied im-
poundments, which are described in detail in Rey and
Kain (1991). Blue Hole Point (impoundment [Imp.] #23)

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

INDIAN RIVER |
LAGOON

80° 18’

FIGURE 1. The three sampling impoundments (Imp) within St. Lu-
cie County, Florida; SMS = Smithsonian Marine Station.

is a 122 ha breached impoundment. The 20 m breach in
the western dike allowed natural tidal access between the
impoundment and the IRL. This breach was a result of a
severe 1981 winter storm and was left open to the natural
tidal cycles of the IRL. The main reason for not repairing
the breach was that this impoundment could be used as a
reference/control site for numerous impoundment studies
(James R. David, St. Lucie County Mosquito Control Dis-
trict, personal communication, 1992). Jack Island State
Preserve (Imp. #16A) is a 161 ha impoundment divided
into four cells. This impoundment was open via culverts
to the IRL during the winter months but was artificially
flooded during the summer months (early May through
August). Bear Point (Imp. #1) is a 255 ha impoundment.
Since August 1993, the culverts here were left open to tidal
exchange.

Quantitative sampling was performed in these im-
poundments, Imp. #23 and Imp. #16A in June-July 1992,

1993, and 1994 and Imp. #1 in July 1994 along a portion
of transect lines previously established for burrow counts
(Gilmore et al., 1991). One transect line had been estab-
lished in each impoundment. These transects ranged from
the edge of the IRL (0 m), continued across the artificial
dike, and ended at the impoundment perimeter ditch (Fig-
ure 2). Four permanent metal stakes indicated the sam-
pling sites (see below). Because of the very hard substrate
on top of the actual dike (the road), it was impossible to
sample these sites (10 m and 15 m) quantitatively (see
below). The 0 m stake was placed at the waterline (low
tide) on the IRL side, and the other three stakes (markers)
were placed at 5 m intervals across the dike, with the 0 m
stake as the starting point. The 15+ m site was between
the 15 m stake and the upper bank of the perimeter ditch
(Gilmore et al., 1991; see Figure 2). Additional sites were
also established for the studies: site A was at the edge of
the water (low tide) on the dike side of the perimeter ditch
and site B in the corresponding area of the impoundment
marsh side of the ditch (Figure 2). It was not possible to
establish a site B in Imp. #16A, because of the summer
artificial flooding, or in Imp. #1, because it was flooded
naturally. The width of the perimeter ditches was about 5
to 6 m. The ditch shores in all impoundments had a very
dense (but only about 1.5 to 2 m wide) mangrove vegeta-
tion (primarily Rhizophora mangle). Two additional sites
were established in Imp. #23: site C, about 2 m into the
impoundment marsh from site B, immediately behind the
dense mangrove vegetation along the ditch shore (see Fig-
ure 2), and site D on the sand flat within the marsh (25 m
from the ditch). The distance from site 0 m to site C was
about 25 m and to site D about 47 m. The sampling sites
were 2 X 2 m permanent squares situated at each marker
(sites 0 m, 15+ m, C, and D), within which four repli-
cate samples were randomly collected on each sampling

FIGURE 2. Cross section of a typical mosquito impoundment in the
Indian River Lagoon (IRL), showing locations of the fiddler crab
sampling sites. The dike road was approximately 1.5 m above low
tide level in the IRL at three investigated sites (site D is not shown).

NUMBER 38 e¢ 461

date. Sites A and B were sampled the same way (on the
exposed substrate at low tide) close to the dense man-
grove vegetation at each side of the perimeter ditch. The
Imp. #23 impoundment marsh was never flooded dur-
ing my studies. The random sampling was performed by
means of a stainless steel cylinder (0.1 m*, 40 cm high)
with a sharpened bottom edge. Sampling was always per-
formed at low tide and when no, or very few, specimens
were observed on the sediment surface. Sampling was
never performed when many crabs were observed out of
their burrows. Sampling at such times would have re-
sulted in erroneous quantitative results because Uca spp.,
when disturbed on the sediment surface, seek shelter in
the closest burrow or even migrate out into the water.
The cylinder was forced down to a sediment depth of
at least 25 cm. The number of Uca burrows within the
cylinder area was recorded, and then the sediment was
removed with a shovel (with a straight edge). The up-
permost fraction (0-10 cm) was sieved (in the field) in
seawater through a 2 mm stainless steel mesh sieve. In the
remaining fraction (10-25 cm) the crabs were removed by
hand in the field. This procedure was deemed acceptable
as small crabs only occurred in the uppermost layer of
the sediment. The specimens were transferred to plastic
bags and kept in a cooler in the field. In the laboratory,
the samples were either processed immediately or stored
in a freezer for later processing. The crabs were sorted by
hand in a tray filled with seawater. They were then placed
in labeled glass jars in a solution of 5% borax-neutral-
ized formalin, diluted in seawater. After 4 to 5 days the
formalin was replaced with 70% ethanol. All specimens
larger than 5 mm carapace width (CW) were later iden-
tified and weighed (wet weight) and have been archived
for possible future studies. All individuals smaller than 5
mm were regarded as “juveniles.” It was not possible to
identify these to species level with certainty. The litera-
ture sources used for species determination were Tashian
and Vernberg (1958), Salmon (1967), and Crane (1975).
A total of 140 quantitative samples were collected dur-
ing the entire study period: 84 in Imp. #23, 40 in Imp.
#16A, and 16 in Imp. #1.

Water temperature was measured midafternoon on 26
July 1993 and 1 August 1994 within the marsh of Imp. #16A
(which was artificially flooded), in the middle of the adjacent
perimeter ditch, and in the IRL (about 5 m from the shore).
The measurements were taken at 5 cm water depth.

Because many impoundments are closed for natural
tidal exchange to the estuarine waters of the IRL during
the artificial flooding periods (impoundment pumps), sa-
linity may drop rapidly during periods of heavy rainfall.

462 °

An experiment was therefore performed to investigate toler-
ance to rapid salinity changes among the four Uca species.
The laboratory setup consisted of twenty 2 L round plas-
tic containers equipped with a lid. A separate air supply
was provided to each container. Four treatments and one
control (four replicates per treatment) were established:
100%, 75%, 50%, 25%, and 0% seawater. Laboratory-
supplied seawater was diluted with distilled water. The
salinities of the different treatments were 100% = 36-37
ppt (parts per thousand), 75% = 27-29 ppt, 50% = 19
ppt, 25% = 9-10 ppt, and 0% = 0 ppt, measured with an
ocular refractometer. The water temperature was very sta-
ble during the experimental period, 24.0°-26.0°C. Each
experiment lasted for seven days. The crabs were collected
48 hours before each experiment and acclimated in 100%
aerated seawater during this period. Seven randomly se-
lected female crabs of each species were placed in each
experimental container. It was not possible to find enough
specimens of U. thayeri during the period for these stud-
ies. Therefore only 25% and 0% seawater were used as
treatments, and each replicate contained five crabs. The
experiments were monitored twice a day, and any dead
crabs were removed. Water was changed only in the con-
tainers where dead crabs were found. These experiments
were performed between 4 July and 27 July 1994.

Sediment samples for analysis of organic content (loss
on ignition) were collected in 1994 along the three tran-
sects. Three sediment cores (inner diameter, 30 mm) were
collected to a depth of 5 cm at randomly chosen points at
each site. As stated above it was not possible to establish
a site B in Imp. #16A or in Imp. #1. The sediment was
treated in the laboratory according to the procedures de-
scribed in Holme and McIntyre (1971).

A one-way analysis of variance (ANOVA) (Holm-Sidak
method) was performed to compare the respective moni-
tored sites in the three impoundments regarding organic
content (LOI) in the sediment (Table 1).

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

RESULTS
ABUNDANCE

Abundance data from the three transects sampled in
1992, 1993, and 1994 at Imp. #23 and Imp. #16A are pre-
sented in Figures 3 and 4, and the one transect sampled in
1994 at Imp. #1 in Figure 5. High water levels prevented
sampling 0 m (IRL) at Imp. #16A in 1992 and site A (ditch
shore) in 1994.

The results from Imp. #23 were similar the three sam-
pling years (Figure 3). U. pugilator and U. rapax were
relatively evenly distributed across the transect, and a few
specimens of U. rapax were sometimes observed on the
dike road (DIKE; see Figure 2). U. speciosa, the dominant
species, was found only at site 0 m, and in very high den-
sities in the wet, soft, and dark mud on both sides of the
perimeter ditch (sites A and B). U. thayeri was also found
on both sides of the perimeter ditch (sites A and B), in ad-
dition to a few specimens at site 0 min 1992.

In contrast, at Imp. #16A (Figure 4), U. pugilator
dominated in abundance at site 5 m whereas U. rapax was
most abundant at site A (perimeter ditch shore). U. spe-
ciosa was almost exclusively found at site A and U. thayeri
at site 0 m.

At Imp. #1 the distributional patterns were similar
to the other impoundments. However, U. pugilator was
found in comparatively low densities, whereas U. rapax
was abundant at both sites 0 m and 5 m. U. speciosa was
found in high densities in the wet muddy areas at site 0 m
and at site A. U. thayeri was found at the 0 m site and to
even a greater extent at site A (dike side of the ditch).

No statistical tests were performed to elucidate any
potential difference between years at each site, but it was of
higher interest to statistically compare abundance patterns
between impoundments. Therefore, correlation analyses
(Pearson product moment correlation) were performed on
the mean abundance data (1992, 1993, 1994) for sites 0 m

TABLE 1. One-way analyses of variance (Holm-Sidak method) concerning differences in organic con-

tent (LOI) between the different impoundment and sampling sites. Significant differences (P values) are

in bold italic.

Site
Impoundment no. 0m 5m 15+ A C D
23 vs. 16A 0.025 0.068 0.158 0.0002 0.0001 0.001
23 vs. 1 0.148 0.044 0.123 0.124
16A vs. 1 0.004 0.003 0.014 0.00006

NUMBER 38 °¢ 463

IMPOUNDMENT #23

100
ME VU. pugilator

ZZ -U. rapax JL 80

RSSSSS U. speciosa N
REY U. thayeri N 60
40
20
0
60

)

—_
©
Xe)
Ww

2

Bes
>)
|

BIOMASS (g / m
No
oO
|

INDIVIDUALS / m2

i=)

SITE
an a a I I

FIGURE 3. Abundance and biomass (wet weight) between 1992 and 1994 of the four fiddler crab species along the impoundment #23
transect. Error bars represent + standard error values (N = 4). Note the different scaling on the y-axes.

464

e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

160
140
120
100

INDIVIDUALS / m2

IMPOUNDMENT #16A

1992

NO DATA

150

125

100

BIOMASS (g / m*)

SIE

NO DATA

15+ A

HE U. pugilator
LZ, 'U. rapax
SSSSS U. speciosa
RS U. thayeri

NO DATA

15+ A

FIGURE 4. Abundance and biomass (wet weight) between 1992 and 1994 of the four fiddler crab species along the transect within
impoundment #16A. Error bars represent + standard error values (N = 4). Note different scaling on y-axes.

NUMBER 38 ¢ 465

IMPOUNDMENT #1 (1994)

INDIVIDUALS / m7?

0 5 15+ A
SITE

oO
c—)

Ge U. pugilator
“ZZZZA VU. rapax

SS
(=)

U. speciosa
RRR U. thayeri

ie)
i=)

BIOMASS (g / m*)
So

FIGURE 5. Abundance and biomass in 1994 of the four fiddler crab species along the transect within impoundment #1. Error bars represent

+ standard error values (N = 4).

to A for each of the four species separately between Imp.
#23 and Imp. #16A. However, no correlation (P > 0.05)
could be found for any of the species. The same analyses
were performed for the 1994 data from Imp. #23, Imp.
#16A, and Imp. #1. The only correlation (positive) found
was for U. thayeri between Imp. #23 and Imp. #1 (correla-
tion coefficient, 0.968; P value, 0.031).

BIOMASS

The biomass (g wet weight) measurements are pre-
sented in Figures 3-5. No significance tests were performed
concerning the biomass difference among the three years
for each species.

However, the biomass calculations for the three years
in Imp. #23 (Figure 3) indicate that changes took place, but
these changes are based on subjective observations. High
biomass values were recorded for U. pugilator in 1992 at
site C and in 1993 and 1994 at site 0 m. High values were
recorded for U. rapax throughout the entire transect, espe-
cially in 1993 and 1994, except at site D. The highest bio-
mass values for U. speciosa were recorded on both sides of
the perimeter ditch (sites A and B), especially on the marsh
side of the ditch (site B). High U. thayeri biomass values
were recorded along the perimeter ditch (sites A and B).

At Imp. #16A the biomass values for U. pugilator
were high at site 5 m all three years (see Figure 4). Rela-
tively high biomass values were recorded for U. rapax at
sites 5 mand A in 1992. However, data for site A in 1994
are not available. Low biomass values were recorded for
U. speciosa at site A in 1992 and 1993. U. thayeri was
only recorded at low biomass values at site 0 m in 1993
and 1994 and at site A in 1993 and at site 0 m in 1994.

At Imp. #1 low values were observed for U. pugilator
throughout the transect (see Figure 5), but U. rapax was,
by far, the most dominant (biomass) species across the en-
tire transect. The only exception was site A, where the val-
ues for U. speciosa and U. thayeri were somewhat higher.

REPRODUCTION AND Sex DISTRIBUTION

The percentage of “juveniles” found in 1993 and
1994 at the different sites within Imp. #23 and Imp. #16A
is presented in Figure 6. More juveniles were found at
site 0 m at Imp. #16A compared with Imp. #23. Many
juveniles were also recorded along these impoundment
ditch shores (sites A and B in Imp. #23 and site A in Imp.
#16A). The sex distribution among adults of the four spe-
cies from the 1992, 1993, and 1994 (combined) collec-
tions (June-July) is presented in Figure 7 with the number

TEMPERATURE AND SALINITY TOLERANCE

466 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES
JUVENILES
#23 #16A
16 = 16

Anal 1993 14 1993

AZ 1 |
10 4 10 +
i 8 7
(5 >
4-4 4-4
Di Pa 25
0 pail etal 0

0 5 15+ A B 5 15+ A

@ —NO DATA

1994 1994

40 - 40 >

30 | a) = |
20 5 20

10 + Ls 10 -

07> 0 | |

Sota 0 5 15+
SITE SITE

PERCENT JUVENILES

> —|NO DATA
w {NO DATA

FIGURE 6. Percentage specimens having a carapace width (CW) less
than 5 mm (juveniles) of all collected individuals of Uca spp. from
each site in 1993 and 1994. Note different scaling on y-axes.

of ovigerous females. As shown, the sex ratio was near
to 1.0 among all four species except for U. thayeri where
the male/female ratio was approximately 1.8:1. The high-
est ovigerous rate was found among U. pugilator (22.0%)
and the lowest among U. rapax (3.4%). The correspond-
ing figures for U. speciosa and U. thayeri were 6.9% and
10.0%, respectively.

BURROWS

The correlation between the number of burrows and
the actual number of crabs found within each sample
in 1992 and 1993 is presented in Figure 8. A Wilcoxon
signed-rank test showed that there was no correlation
between these two parameters: P = 0.097 (linear regres-
0.02, P [analysis of variance] = 0.23). This
finding has also been reported by Colby and Fonseca
(1984). The same lack of correlation was also found by
the author in a larger and more detailed multiyear study
at Merritt Island impoundments (close to Cape Canav-
eral, eastern Florida.).

sion: R* =

The summer water temperatures within Imp. #16A, the
perimeter ditch, and in the IRL is presented in Table 2. The
water temperature was higher within the impoundment
marsh than in the perimeter ditch and in the IRL.

The laboratory experiment showed that no species
showed any disturbance or mortality in 100%-25% seawater.
However, the reaction toward 0% seawater was severe (Fig-
ure 9). U. speciosa and U. thayeri showed very low tolerance
toward 0% seawater while U. pugilator showed the highest
tolerance. The reaction from U. rapax was intermediate.

SEDIMENT
The results of the sediment analyses are presented

in Figure 10. The loss on ignition (organic content) was
higher in Imp. #16A than in Imp. #23 and Imp. #1 (see

SEX DISTRIBUTION

[| males
| females
ez) ovig fem

NUMBER OF INDIVIDUALS

PUG RAP SPE _ THA
SPECIES

FIGURE 7. Sex distribution of all fiddler crab individuals larger than
5 mm collected during 1992-1994 combined within impoundments
#23, #16A, and #1. The bars show sex distribution for each species
found throughout the study period; PUG = Uca pugilator; RAP =
U. rapax; SPE = U. speciosa; THA = U. thayeri.

NUMBER OF BURROWS
fos)

| | ] I ]

0 10 20 30 40 50
NUMBER OF CRABS

FIGURE 8. Relationship between number of burrows and number
of fiddler crabs found in each quantitative sample.

Table 1). The lowest values on the IRL side (site 0 m) were
recorded from Imp. #1 and the highest from Imp. #16A.
At the ditch (site A) the highest organic value was recorded
at Imp. #16A and the lowest at Imp. #1. Within the marsh
(site C) (Imp. #23 and Imp. #16A only), the loss on igni-
tion was very high within Imp. #16A and very low within
Imp. #23. As indicated in Table 1, Imp. #16A deviated
significantly the most from the other two impoundments,
with generally the highest organic content (LOI).

TABLE 2. Water temperatures (°C, 5 cm water depth) at im-
poundment site #16A, measured in midafternoon during July
1993 and July 1994.

Location 4 1993 1994 Mean
Marsh 44.3 42.4 43.4
Ditch 37.1 36.1 36.6
IRL 37.5 35.2 36.4
4 Marsh = impoundment marsh; ditch = impoundment perimeter ditch; IRL =

Indian River Lagoon.

NUMBER 38 ¢ 467

Uca pugilator Uca speciosa

100 @— 100 -@ - ———
80 — \ 80 — |
| \
\ | \
60 4 ® 60 4 \
Te | \
40 + bs 40 \
20 — ] 2 @
\
iat LS =
Sie arpa alla lees Oil eres ee, eel el el
0 12 24 36 48 60 72 84 0 12 24 36 48 60 72 84
=)
Yn
5 Uca thayeri Uca rapax
WwW 100 @—— =I 100 @ =
6) pS
ow
Ww |
o 807 \ 80 —
\
60 — \ 60 4 TT
| 4 nN
40 4 \ 40 4 \
\
| \ | @
20 4 \ 20 4 I
| \ ;
OS © a cal a 0 imaian alas alee
0 12 24 36 48 60 72 84 0 12 24 36 48 60 72 84
HOURS

FIGURE 9. Percent survival in fresh (distilled) water of the four fid-
dler crab species. Error bars represent standard error values (N = 4).

IMP. #1

IMP. #23
Ez eeeE ar 15 5 =

=
Ny
——

12 5
9 | 9
| | |
Tina. | ohne
04 | | ita 0 | | Sarr |e
0

5 15+ A B C D 0 5 15+ A B C D

IMP. #16A SITE

PERCENT LOSS ON IGNITION

<x
Ee
<
fa}
o}
“=

=a
0 5 15+ A B C D
SITE

FIGURE 10. Sediment organic content (percent weight loss on igni-
tion) at the sampling sites within the three impoundments in 1994.
The error bars represent standard error values (N = 3). Note differ-
ent scaling on y-axes.

468 e

Pearson product moment correlation tests were used
to elucidate any potential correlation between the organic
content of the sediment (LOI) and the distributional pat-
tern (along the transects) of the four different species. All
tests were performed on the mean values of each param-
eter. Only transects where a significant amount of infor-
mation was available were used for these tests: Imp. #23
in 1993 and 1994, and Imp. #16A in 1994 (see expla-
nation above). No significant correlation (P > 0.05) was
recorded for any species between abundance and organic
content, except for U. pugilator at Imp. #23 in 1994, with
a correlation coefficient of 0.797 and a P value of 0.032.

DISCUSSION

Different species of fiddler crabs prefer different sub-
strates and salinities for reasons of their specific physiolog-
ical tolerances and environmental preferences (Teal, 1958;
Vernberg et al., 1973). However, the artificial environment
of the impoundments in the IRL, with the perimeter ditch
and dike, poses a completely new, different type of envi-
ronment for the Uca populations. The occurrence and dis-
tributional patterns of the different species of fiddler crabs
in these impoundment dikes and perimeter ditches have
not been investigated earlier, so this is a first basic study
on these populations. The results from this study indicate
that more detailed studies are needed in the future in these
very extensive artificial environments in the IRL.

Uca speciosa and U. thayeri were the most “site- and
substrate-specific” species within these environments,
whereas U. pugilator, and in some cases also U. rapax,
were more “generalists.” Uca speciosa and U. thayeri were
almost exclusively found in the very soft, black, and wet
substrate close to the water (primarily sites A and B),
which is clearly demonstrated by the data presented from
Imp. #23. The highest abundances of U. speciosa were here
recorded on the perimeter ditch “shores” (sites A and B).
Given the rich mangrove vegetation, it was expected that
the very fine, wet, and muddy sediment on the ditch pe-
rimeter shores (sites A and B) would have a comparatively
high organic content, but this was not the case (see Figure
10). Because no correlation between the organic content
of the substrate and the abundance pattern of the different
species were found, the grain-size distribution, water con-
tent, the chemical content of the sediment, root mat den-
sity, physiological tolerances, and interspecific interactions
may be more important factors in fiddler crab distribu-
tional patterns (Teal, 1958; Ringold, 1979; Bertness and
Miller, 1984). No root mat areas were investigated in this

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

study. The large differences in substrate organic content
(LOI) between Imp. #23 and Imp. #16A are noteworthy
(see Table 1).

Large numbers of U. rapax (and to some extent also
U. pugilator) were quite often observed on the dike roads,
but U. speciosa and U. thayeri were never seen there.
Thompson et al. (1989) have also demonstrated that some
species of desiccated fiddler crabs, among these U. pugila-
tor, can rehydrate on damp sand.

When the impoundments are being flooded, it appears
that U. rapax, and most likely also U. pugilator, are able
to migrate under water, across the perimeter ditch (often
anoxic and with HS in the sediment), to more suitable
areas. It is, however, important to note that this has so far
been confirmed only for U. rapax. Therefore, the ability
to relocate to more suitable habitat may be the decisive
factor in survivorship among Uca species. The banks of
the perimeter dike (immediately above site A) may there-
fore act as a temporary “refuge” for some species dur-
ing periods of impoundment flooding. It is also possible
that further migration takes place toward the IRL shores
(U. rapax, U. pugilator). However, this question does not
apply to the rim and road of the dike because of the un-
suitable substrate. Furthermore, the distance to the water
table is also too great (at least 1 m).

The two species U. rapax and U. pugilator are probably
not adversely affected by impoundment management. Visual
observations, and also in situ experiments, have revealed
that these species are highly motile within the impoundment
areas. According to Thurman (2003) U. rapax is typically
collected in brackish water. Yoder et al. (2005) have also
found that the “herding behavior” in U. pugilator is a water-
conserving group effect, and this behavior makes them less
vulnerable to desiccation. Many specimens of U. pugilator
and U. rapax have been observed (by the author) to migrate
over long distances within and outside the impoundments
(marked individuals, not reported here). However, further
studies need to be performed to clarify these patterns.

Although U. pugilator and U. rapax thrive in these
areas, the fate of the other two species is more uncertain.
According to the quantitative sampling results and inten-
sive visual in situ studies, U. speciosa and U. thayeri are
confined to substrate-specific areas of the impoundments,
and this may have a negative effect on the populations of
these species when the impoundments are being managed
(flooded). However, the results from Imp. #16A, which
was flooded frequently for mosquito control, seem to con-
tradict this assumption. In spite of this management, a rich
community of U. speciosa was recorded on the ditch shore
(site A), but with low densities of U. thayeri.

Even though the data on the occurrence of juveniles
are limited, they indicate that fiddler crab reproduction
(species unknown) occurs also in the impoundment perim-
eter ditch (site A and B). As shown in Figure 6, juveniles
were, as expected, mainly found close to the water (sites
0 m, A, and B).

The dilution experiments indicate that none of the
four species is sensitive to low salinities, a situation that
rarely occurs within the impoundments. During this ex-
periment U. pugilator was the most tolerant species.
Thurman, (2003) investigated the osmoregulation of eight
Uca species and found that U. speciosa and especially U.
pugilator are able to withstand high “osmotic challenge.”
Additionally, Thurman (2005) reported that U. rapax is
best equipped for living in brackish habitats and that U.
thayeri and U. speciosa are best suited physiologically to
inhabit low and moderately saline habitats. This obser-
vation may explain why the latter two species are able
to successfully inhabit the impoundment ditch “shores”
(sites A and B). U. rapax, and most likely also U. pugi-
lator are, as discussed earlier, able to migrate over long
distances, for example, across the perimeter ditches (when
the impoundments are being flooded) and dikes (studies
by the author on ~1,200 of marked U. rapax individuals).
However, this is most likely not the case with U. speciosa
and U. thayeri. As U. speciosa is a comparatively small
species, it may therefore be more vulnerable to desiccation
than the other three species (Pellegrino, 1984).

High summer temperatures in the shallow impound-
ment water (see Table 2) pose a threat to the fiddler crab
populations. Even though the temperature measurements
only were performed twice within Imp. #16A (Table 2),
they show that the temperature in the shallow (flooded)
areas in the impoundment marsh may reach at least 44°C,
which is significantly higher than in the nearby IRL. Large
numbers of dead individuals were observed in very shal-
low water during these high temperature periods (within
the marsh of Imp. #16A), but never at lower temperatures,
and it was assumed that death was the result of short-term
hyperthermia. Replicated laboratory experiments on U.
pugilator and U. rapax collected inside Imp. #23 showed
that lethal water temperatures (LDs509) on individuals from
this area are 41°-42°C. Teal (1958) reported a lethal tem-
perature (LD59) between 39.5° and 40.0°C for U. pugila-
tor, U. minax, and U. pugnax, and Vernberg and Tashian
(1959) found that U. rapax was more resistant to tem-
peratures of 42°-44°C than was U. pugnax. Wilkens and
Fingerman (1965) performed a thorough study on lethal
temperatures for U. pugilator in both saturated and dry
air. LDs9 in saturated air was reached at 40.7°C, which

NUMBER 38 ¢ 469

corresponds well to the results from my observations.
Powers and Cole (1976) have also demonstrated that bur-
row temperature decreased rapidly with depth, proving
the major heat refuge for U. panacea on open sand flats
during a study on Mustang Island, Texas. Edney (1961),
in a study on a number of fiddler crabs at Inhaca Island,
Mozambique, found that the temperature within the bur-
rows during the warmer months was considerably cooler
than the sand on the surface. Preliminary results within
this study (not presented) also indicate that the tempera-
ture drops significantly with sediment depth in the mos-
quito impoundments.

Genoni (1985), on a study on U. rapax in Florida,
reported that there were more burrows than fiddler crabs
in the sediment. Even if there was no correlation between
fiddler crabs (all species) and burrows in the present study,
the results were often the opposite from the results by
Genoni (1985). Mouton and Felder (1996) investigated
the quantitative distribution of U. spinocarpa and U. lon-
gisinalis by quantitatively counting the number of Uca
burrows along transects in a Gulf of Mexico salt marsh.
However, no studies were performed regarding the num-
ber of individuals (and species) living in these burrows.
Excavating the substrate is a very labor-intensive proce-
dure but obviously necessary to be able to evaluate the
actual fiddler crab species distribution and abundance
within specific areas (see Methods, above). The studies
performed by the author in the three St. Lucie County
impoundments and at Merritt Island (Cape Canaveral)
impoundments (not reported here) did not produce any
correlation between burrows and number of Uca speci-
mens. Actually, in several cases when no burrows at all
were found on the sediment surface within the 0.1 m?
sampling area, large amounts of fiddler crabs were found
in deeper areas when excavating the substrate within the
sampling area according to the description above. There-
fore, only counting Uca burrows does not seem to give
correct data regarding Uca population abundance and
species distributional pattern. Further detailed studies are
therefore needed to elucidate this relationship.

In conclusion, these studies in the St. Lucie impound-
ments do not indicate that the construction and manage-
ment of IRL mosquito impoundments pose a serious threat
to fiddler crab populations. However, the impoundments
may change the distributional patterns of the different
species. It is important to note that new, highly suitable
habitats were created when the impoundments were con-
structed, such as the perimeter ditch margins (sites A and
B), especially preferred by U. speciosa and U. thayeri in
impoundments with tidal access to the IRL. However, the

470 e

fate of these two species at the marsh side of the perimeter
ditch (site B) during the prolonged artificial summer flood-
ing is still unknown.

ACKNOWLEDGMENTS

I wish to thank James R. David (Director, St. Lucie
County Mosquito Control District) and Frank Evans for
excellent cooperation and financial support. I am very
grateful to R. Grant Gilmore for his support during
these studies. Thanks are due to Tor Carlsson for great
assistance in the field and laboratory. I am also grateful
to Professor L. B. Holthuis, Nationaal Natuurhistorish
Museum, Leiden, the Netherlands, for help identify-
ing some of the specimens. Special thanks are due to
Johan Tunberg for valuable assistance in the very hot
and humid impoundment habitats during these summer
studies. I am also very grateful to Sherry Reed for the
critical reading and valuable comments on the manu-
script. Finally, I am very grateful to the two referees,
whose comments and criticism significantly improved
the quality of this manuscript. This study was financed
by The St. Lucie County Mosquito Control District,
The Royal Swedish Academy of Sciences, The Harbor
Branch Oceanographic Institution, and private Swedish
funds. This is Smithsonian Marine Station at Fort Pierce
Contribution No. 789.

LITERATURE CITED

Barnwell, F H. 1968. The Role of Rhythmic Systems in the Adapta-
tion of Fiddler Crabs to the Intertidal Zone. American Zoologist,
8:569-583.

Bertness, M. D. 1985. Fiddler Crab Regulation of Spartina alterniflora
Production on a New England Salt Marsh. Ecology, 66:1042-
1055.

Bertness, M. D., and T. Miller. 1984. The Distribution and Dynamics of
Uca pugnax (Smith) Burrows in a New England Salt Marsh. Jour-
nal of Experimental Marine Biology and Ecology, 83:211-237.

Colby, D. R., and M. S. Fonseca. 1984. Population Dynamics, Spatial
Distribution and Somatic Growth of the Sand Fiddler Crab Uca
pugilator. Marine Ecology Progress Series, 16:269-279.

Crane, J. 1975. Fiddler Crabs of the World (Ocypodidae). Princeton,
N.J.: Princeton University Press.

DeCorsey, P. J. 1983. “Biological Timing.” In The Biology of Crustacea,
ed. D. E. Bliss, pp. 107-162. New York: Academic Press.

Dittman, S. 1996. Effects of Macrobenthic Burrows on Infaunal Com-
munities in Tropical Tidal Flats. Marine Ecology Progress Series,
134:119-130.

Edney, E. B. 1961. The Water and Heat Relationships of Fiddler Crabs
(Uca spp.). Transactions of the Royal Society of South Africa,
36:71-91.

Fingerman, M. 1957. Relation Between Position of Burrows and Tidal
Rhythm of Uca. Biological Bulletin, 112:7-20.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Genoni, G. P. 1985. Increased Burrowing by Fiddler Crabs Uca rapax
(Smith) (Decapoda: Ocypodidae) in Response to Low Food Supply.
Journal of Experimental Marine Biology and Ecology, 87:97-110.

Gilmore, R. G., R. E. Brockmeyer Jr., and D. M. Scheidt. 1991. A Pre-
liminary Report: Spatial and Temporal Dynamics of Uca Popula-
tions in High Marsh Habitats Vegetated with Algae, Herbaceous
and Woody Flora under Managed Hydrological Cycles. Ft. Pierce,
Fla.: Harbor Branch Oceanographic Institute.

Gilmore, R. G., D. M. Scheidt, R. E. Brockmeyer Jr., and S. Vader Kooy.
1990. Spatial and Temporal Dynamics of Secondary Productivity
in High Marsh Habitats Vegetated with Algae, Herbaceous and
Woody Flora under Managed Hydrological Cycles. Final Report,
Coastal Zone Management 258. Tallahassee, Fla.: Florida Depart-
ment of Environmental Regulation.

Grimes, B. H., E. T. Huish, J. H. Kerby, and D. Xoran. 1989. Species
Profiles: Life Histories and Environmental Requirements of Coastal
Fishes and Invertebrates (Mid-Atlantic)-Atlantic Marsh Fiddler.
U.S. Fish and Wildlife Service Biology Report 82. Vicksburg, Miss.:
U.S. Army Corps of Engineers Coastal Ecology Group.

Heard, R. W. 1982. Guide to the Common Tidal Marsh Invertebrates
of the Northeastern Gulf of Mexico. Ocean Springs, Miss.: Missis-
sippi-Alabama Sea Grant Consortium.

Holme, N. A., and A .D. McIntyre, eds. 1971. Methods for the Study of
Marine Benthos. Oxford and Edinburgh: Blackwell Scientific Pub-
lications.

Kerr, G. A. 1976. Indian River Coastal Zone Study, Inventory 1975. Annual
Report, Volume II. Fort Pierce, Fla.: Harbor Branch Consortium.

Kraeuter, J. N. 1976. Biodeposition by Salt-Marsh Invertebrates. Marine
Biology, 35:215-223.

Macintosh, D. J. 1982. “Ecological Comparison of Mangrove Swamp
and Salt Marsh Fiddler Crabs.” In Wetlands Ecology and Man-
agement. Proceedings of the First International Wetlands Confer-
ence (New Delhi, India, 10-17 September 1980), ed. B. Gobal,
R. E. Turner, R. G. Wetzel, and D. EF Whigham, pp.243-257.
Jaipur, India: National Institute of Ecology & International Sci-
entific Publications.

Montague, C. L. 1980. A Natural History of Temperate Western Atlantic
Fiddler Crabs (Genus Uca) with Reference to Their Impact on the
Salt Marsh. Contributions in Marine Science, 23:25-55.

. 1982. “The Influence of Fiddler Crab Burrows on Metabolic
Processes in Salt Marsh Sediments.” In Estuarine Comparisons, ed.
V. S. Kennedy, pp. 283-301. New York: Academic Press.

Mouton, E. C., and D. L. Felder. 1996. Burrow Distribution and Popula-
tion Estimates for the Fiddler Crabs Uca spinacrapa and Uca longi-
signalis in a Gulf of Mexico Salt Marsh. Estuaries, 19:51-61.

Peterson, C. H., and N. M. Peterson.1979. The Ecology of Intertidal
Flats of North Carolina: A Community Profile. Biological Services
Program Report FWS/OBS-79/39. Sidell, La.: U.S. Fish and Wild-
life Service.

Pellegrino, C. R. 1984. The Role of Desiccation Pressure and Sur-
face Area/Volume Relationships on Seasonal Zonation and Size
Distribution of Four Intertidal Decapod Crustacea from New
Zealand: Implications for Adaptation to Land. Crustaceana,
47(3):251-268.

Powers, L. W., and J. E Cole. 1976. Temperature Variation in Fiddler
Crab Microhabitats. Journal of Experimental Marine Biology and
Ecology, 21:141-157.

Rey, J .R., and T. Kain. 1991. A Guide to the Salt Marsh Impoundments
of Florida. Vero Beach, Fla.: The Florida Entomology Laboratory.

Ringold, P. 1979. Burrowing, Root Mat Density, and the Distribution of
Fiddler Crabs in the Eastern United States. Journal of Experimental
Marine Biology and Ecology, 21:141-157.

Salmon, M. 1967. Coastal Distribution, Display and Sound Production by
Florida Fiddler Crabs (Genus Uca). Animal Behaviour, 15:449-459.

Salmon, M., W. H. Seiple, and S. G. Morgan.1986. Hatching Rhythms of
Fiddler Crabs and Associated Species at Beaufort, North Carolina.
Journal of Crustacean Biology, 6:24-36.

Tashian, R. E., and F. J. Vernberg. 1958. The Specific Distinctness of the
Fiddler Crabs Uca pugnax (Smith) and Uca rapax (Smith) at their
Zones of Overlap in Northeastern Florida. Zoologica, 43:89-92.

Teal, J. M. 1958. Distribution of Fiddler Crabs in Georgia Salt Marshes.
Ecology, 39:185-193.

Thompson, W. E., P. J. Molinaro, T. M. Greco, J. B. Tedeschi, and C. W.
Holliday. 1989. Regulation of Hemolymph Volume by Uptake
of Sand Capillary Water in Desiccated Fiddler Crabs Uca pugila-
tor and Uca pugnax. Comparative Biochemistry and Physiology,
94A:531-538.

Thurman, C. L. 2003. Osmoregulation in Fiddler Crabs (Uca) from Tem-
perate Atlantic and Gulf of Mexico Coasts of North America. Ma-
rine Biology, 142:77-92.

. 2005. Comparison of Osmoregulation among Subtropical Fid-

dler Crabs (Uca) from Southern Florida and California. Bulletin of

Marine Science, 77:83-100.

NUMBER 38 e¢ 471

Vernberg, F. J., and R. E. Tashian. 1959. Studies on the Physiological
Variation between Tropical and Temperate Zone Fiddler Crabs of
the Genus Uca. I. Thermal Death Limits. Ecology, 40:589-593.

Vernberg, W. B., P. J. DeCoursey, and W. J. Padgett. 1973. Synergistic
Effects of Environmental Variables on Larvae of Uca pugilator. Ma-
rine Biology, 22:307-312.

Weis, J. S., and P. Weis. 2004. Behavior of Four Fiddler Crabs, Genus
Uca, in Southern Sulawesi, Indonesia. Hydrobiologica, 523:47-58.

Wheeler, D. E. 1978. Semilunar Hatching Periodicity in the Mud Fiddler
Crab Uca pugnax (Smith). Estuaries, 1:268-269.

Wilkens, J. L., and M. Fingerman. 1965. Heat Tolerance and Tempera-
ture Relationships of the Fiddler Crab, Uca pugilator, with Refer-
ence to Body Coloration. Biological Bulletin, 128:133-141.

Yoder, J.A., K. A. Reinsel, J. M. Welch, D. M. Clifford, and E. J. Rel-
linger. 2005. Herding Limits Water Loss in the Sand Fiddler Crab,
Uca pugilator. Journal of Crustacean Biology, 25:141-145.

Zucker, N. 1978. Monthly Reproductive Cycles in Three Sympatric
Hood-Building Tropical Fiddler Crabs (Genus Uca). Biological Bul-
letin, 155:410-424.

x
2 lj 7
3 =
<a
- ' =

: 2 Ab gat So) a eo
EL eee = = _ ie
: ‘ 1a = “a ent:
me oo tk AS cll 2: eeteaiiee are inva
: : ; _ 4 cg 2 ees ; 1 oe “ed oma
Sie Bee : Se cemetieratitellt al
| 3 _ rm wa ae en
ae at i
; ; 1
:
ie '
j a
us ' oy ts 7

Dynamic Hydrology of a Mangrove Island:
Twin Cays, Belize

Daniel W. Urish, Raymond M. Wright,
Ilka C. Feller, and Wilfrid Rodriguez

Daniel W. Urish, Raymond M. Wright, and Wil-
frid Rodriguez, University of Rhode Island, Engi-
neering Department, Bliss Hall, 1 Lippitt Road,
Kingston, Rhode Island 02881, USA. Ilka C.
Feller, Smithsonian Environmental Research Cen-
ter, 647 Contees Wharf Road, Edgewater, Mary-
land 21037, USA. Corresponding author: D. Urish
(barbourrose@cox.net). Manuscript received 13
May 2008; accepted 20 April 2009.

ABSTRACT. The hydrology of an overwashed mangrove island is shown to be both
complex and dynamic, with a strong interaction between tide-induced flow and the resi-
dent red mangrove (Rhizophora mangle L.) root system. A topographic map of the tid-
ally flooded area of the island was made and related to the tide-induced water levels. The
flooded area approximately doubled during the usual tidal event. The bottom topogra-
phy is highly irregular with a maximum channel water depth of about 1.5 m, but much
of the flooded area experiences a water depth of less than 0.5 m. Water elevations were
recorded by automatic water level loggers for periods of time up to 9 months. The usual
symmetrical parabolic tide signal was transformed into a highly asymmetrical form as it
moved landward through the tangled root system of the red mangrove forest. A normal
tide range of 13 cm at the island margin attenuated to 3 cm at a distance of 200 m
landward, with a lag time of 2 h for highs and 6 h for lows. Maximum flow velocities
of 5 cm/s were measured in the main channels with marked reduction in regions of
dense mangrove root and shallow water depth. The combined frictional resistance of the
bottom and associated mangrove roots is characterized by a Manning’s roughness coef-
ficient, n, that ranged from 0.084 to 0.445. The changing flow pattern within the flooded
mangrove swamp was mapped during a 7 h high-to-low tide period using aerial photog-
raphy to track the movement of slugs of visible dye placed at three locations. Analysis
of the sequential time-related photos showed limited lateral dispersion in the tortuous
main channel but strong tidally controlled flow direction changes and dispersion along
the channel axis. A strong circulatory pattern is observed in a shallow pond at the south
central terminus of the tidally affected flow system. This large shallow pond is sparsely
populated by dwarf red mangrove and is some 350 m from a primary connection with
the surrounding lagoon. Poor flushing of the pond creates water temperatures ranging
from 25°C in the winter to 40°C in the summer. High surface water evaporation creates
a hypersaline condition of 45 ppt salinity in summer. In winter, with the infusion of fresh
rainwater, salinity of surface water in the pond can be less than 5 ppt. Because of its role
in the transport of nutrients and detritus, and its flushing action, the dynamic hydrological
system of the mangrove island is a highly important ecological feature of the overwashed
mangrove island.

INTRODUCTION

Mangrove forests are tropical wetlands with a specialized vegetation adapted
to waterlogged and saline conditions (Lugo and Snedaker, 1974; Hutchings and
Saenger, 1987; Ball, 1988). These forests provide energy-absorbing buffers from

474 e

hurricane-driven seas, prevent coastal erosion, provide
a protective habitat for many fish juveniles, and are a
nutrient source for the surrounding waters (Odum and
Heald, 1972, 1975; Twilley, 1988; Danielson et al., 2005;
Barbier, 2006, Constanza et al., 2008), as well as a filter-
ing mechanism for sediments and pollution (Alongi and
McKinnon, 2005). Under natural conditions mangroves
live in a highly dynamic environment and in synergistic
balance with their natural neighbors. Mangroves have
evolved features that enable them to cope with an ever-
changing regime of tidal water ranges, variable salinity and
temperature, and anoxic soil conditions, but within limits
(Tomlinson, 1986). The biggest enemy appears to be man,
who can directly or indirectly destroy, in days, whole man-
grove communities that have taken thousands of years to
develop (Alongi, 2002; Macintyre et al., 2004; Rodriguez
and Feller, 2004; Taylor et al. 2007; Duke et al., 2007).

It is widely recognized that hydrological patterns de-
termine mangrove structure and function at the ecosystem
scale (Lugo and Snedaker, 1974; Forman and Godron,
1986; Twilley, 1995), and general models of mangrove hy-
drodynamics have been developed (Wolanski et al., 1992).
In these coastal wetlands, tidal flooding and surface drain-
age influence many ecological processes, including habitat
quality, water movement, filtration, and nutrient cycling
(Forman and Gordon, 1986). Water flow also influences
the dispersal and establishment of mangrove propagules
(Mazda et al., 1999).

The significant role of vegetation and the effect of in-
tertidal root density on tidal movement in mangrove chan-
nels has been described by Wolanski et al. (1980) and over
the broader mangrove swamp environment by Wolanski
et al. (1992), Furukawa and Wolanski (1996), Mazda et
al. (1997), and Mazda et al. (2005). Thus, there is a syn-
ergistic relationship for the development and growth of a
mangrove forest that depends on the dynamics and mag-
nitude of tidal inundation into the swamp. Concurrently,
the frictional resistance of the mangrove roots controls the
degree of tidal inundation and patterns of movement in
the mangrove swamp (Wright et al., 1991).

Based on long-term experiments on offshore man-
grove islands in Belize, hydrodynamics have been linked
to distinct patterns of nitrogen (N) and phosphorus (P)
limitation across the intertidal flow system (Feller et al.,
2003). Lovelock (2008) suggested that differences in tidal
inundation also influence soil respiration and below-
ground carbon sequestion via root production, which is
the source of the deep peat deposits underlying these is-
lands. McKee et al. (2007) predicted that the ability of
islands such as Twin Cays to keep pace with rising sea

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

levels is dependent on the tight coupling between peat for-
mation and hydrology.

Although these and other studies based at Twin Cays
have identified tidal flooding as an important drive of eco-
logical processes, there is limited knowledge on the spe-
cific pattern of water movement across these islands. Thus
the objective of this research was to conduct a detailed
analysis of tidal characteristics and flushing patterns of
West Island, the smaller of the two main islands in the
Twin Cays Archipelago.

LOCATION

The Twin Cays Archipelago lies some 22 km off the
coast of Belize (Figure 1) on the edge of the Belizean Bar-
rier Reef (16°50'N, 88°06’W). Islands of the Barrier Reef
and its surrounding waters have been the locations for
scientific ecosystem studies by the Smithsonian Institu-
tion since 1972 (Rutzler and Macintyre, 1982). Because of
their pristine condition and relative isolation from anthro-
pogenic effects, the islands and contiguous waters of Twin
Cays were selected for detailed scientific research of oce-
anic mangroves and associated marine ecosystems (Ritz-
ler and Feller, 1996). Field studies of the dynamic hydrol-
ogy of the Twin Cays mangrove ecosystems were begun in
1986 and have continued since that time. This particular
study focuses on the surface hydrology of West Island of
Twin Cays (Figure 1), a 21.5 ha kidney-shaped landmass
approximately 900 m long and 400 m wide. According to
the classification of Lugo and Snedaker (1974), the island
is an “overwashed mangrove island,” one frequently over-
washed by tides and with high organic export.

ISLAND CHARACTERISTICS

The land cover on West Island and effect of man are
shown in Figure 2, which depicts the natural mangrove
growth and the man-made clear-cut and dredge-fill as
mapped by I. C. Feller of the Smithsonian Institution
in 2002. Since then even more mangrove destruction
has occurred on the east side of the island. The island
is dominated by the red mangrove, Rhizophora mangle
L., with black mangrove (Avecennia germinans L.) on
somewhat higher topography in the intertidal zone and
white mangrove (Laguncularia racemosea L.) above
the intertidal zone (Rutzler and Feller, 1988; Rodriguez
and Feller, 2004). It is to be noted that the density of
the mangrove is far from uniform, with sparse dwarf
red mangrove dominating the interior, and much more
vigorous red mangrove growth on the island perimeter

Cay

Bie gO 990 ore rr)
ri.: Lagoon. :."°.-..":
£7 Patch Reefs: .. 2.0

oe

‘(| South Water

7,6 2 Carrie Bow
ah : Cay

Barrier Reef
Ae al Cays
Shoals

NUMBER 38 ¢ 475

88° HONDURAS

Sand Bores &
Patch Reefs

FIGURE 1. Location map of Twin Cays, Belize, Central America. (Adapted from Ritzler and Macintyre,

1982.)

and in areas of greater tidal movement. Figure 3 is a
photograph taken from the island interior showing the
dwarf red mangrove in the foreground and the distant
background of taller dense red mangrove growth that
characterizes the island perimeter. Figure 4 provides a
botanical rendering of the cross section of the scrub red
mangrove, showing the relationship between mangrove
foliage, stem and root structure, average tidal range, and

hydrogeologic strata. It is to be noted that the typical
low tide level is near the top of the organic ooze.

The mangroves of Twin Cays have developed on
an ancient limestone plateau over the past 8,000 years
(Macintyre et al., 2004). During this time 9-12 m of
Holocene mangrove deposits have accumulated on the
underlying limestone substrate and kept pace with ris-
ing sea level (Toscano and Macintyre, 2003; Macintyre

476

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

North Point

—)

BIG DIPPER

LEGEND

Clearcut areas

Dwarf red mangrove
Dense mangrove growth
Dredged & filled areas
Red mangrove fringe

dredged and filled ’95| 4s.

palm trees planted,
ES:
houke buitt 98

Zs

N
is) 100
ee
m

FIGURE 2. Land cover characteristics of West Island, Twin Cays, Belize, based on aerial photographs taken
in 2002 that show mangrove density and clear-cut areas. (Drawn by Molly K. Ryan of the Smithsonian
Institution in 2002.)

FIGURE 3. Photograph of West Island, Twin Cays, showing dwarf
red mangrove of island interior with much more vigorous fringe red

mangrove growth in distance along periphery of the island.

et al., 2004; McKee et al., 2007). Macintyre and Toscano
(2004) found Pleistocene limestone at depths of 8.3 to
10.8 m below mean sea level in cores at West Island. As
sea level rose to cover a subaerially eroded limestone
plateau fringing the coastline, mangrove peat appears in
the stratigraphic record. The highest topography of the
island is on the seaward side where deposited sand is no
more than 1 m above mean sea level. The limestone is
now found at depths of 9 to 10 m below present sea level
(Macintyre et al., 2004). The swamp bottom is com-
posed largely of soft silty organic detritus. Exceptions of
harder bottoms are found in the nearshore swamp chan-
nels where stronger tide-induced velocities have scoured
the channel bottoms.

The climate of Twin Cays is marine tropical with air
temperatures ranging during the year from 24°C in Janu-
ary to 29°C in June; humidity averages about 78% (Rutz-
ler and Ferraris, 1982). Lagoon water temperatures range
from 23°C in the winter to 31°C in the summer. The mi-
croclimate of West Island, particularly that of the interior
water, has a much greater range. The estimated annual
precipitation at Twin Cays is about 1,885 mm, based on 4
years of complete records at the climatological station on
Carrie Bow Cay, 4 km away. This precipitation is about
80% of the annual precipitation of the nearest mainland
climatological station, the Melinda Forest Station, 30 km
to the northwest. The monthly pattern is much the same
for both stations. Hurricanes cause seawater to completely

NUMBER 38 °¢ 477

overwash the low-lying island. However, the natural man-
grove ecosystem seems resilient and well suited to survival
from natural events, viz. hurricanes. No significant adverse
effects have been observed on Twin Cays; the same can-
not be said for the response to man-made features, where
mangrove clearing results in severe coastal erosion.

Tides in the lagoon area surrounding the island are
microtidal with an average range of about 15 cm and
are of the mixed semidiurnal type (Kjerfre et al., 1982).
The tides exhibit semidiurnal high and lows with a tidal
cycle periodicity of approximately 12 h and 25 min, but
display a marked asymmetry with a large tide range fol-
lowing a smaller one. In some cases the larger range is as
much as 40 cm, followed by a range of only 10 cm. At
times the smaller range is so small as to appear nonex-
istent. In other cases certain components of the tide oc-
cur simultaneously and create a range as great as 50 cm.
Once the tidal signal enters the tangled root system of
the mangrove, the signal changes from a form that is
approximately parabolic to a highly asymmetrical pat-
tern, in which the rising limb of the flood tide is much
steeper than the falling limb. Concurrently the amplitude
is attenuated, and the highs and lows of the tidal signal
lag the open lagoon tide. The spring—neap tidal cycle is
about 29.5 days and can cause monthly tidal ranges that
completely “dry up” the interior of the island.

¥

TW)
US

z

FA
ft

ORGANIC
OOZE

PEAT

(Ger l
SCALE 0 1 2 3 meters

FIGURE 4. Botanical rendering of cross section of dwarf red man-
grove showing relationship between mangrove foliage, stem and
root structure, average tide range, and hydrogeologic strata. (Drawn
by Molly K. Ryan of the Smithsonian Institution in 1989.)

478 e

METHODOLOGY

The information required for a study of the dynamic
hydrology of West Cay encompassed both spatial and
temporal data and a wide variety of methods. These meth-
ods included field surveying techniques for obtaining the
island topography and bathymetry, automated water level
recorders for water levels, automated temperature loggers,
electromagnetic water current meters, conductivity meters
for determination of water salinity, and aircraft for pho-
tographic recording of dye flows, among a host of lesser
equipment and measuring devices that were employed
over the study period of 18 years (1988-2006).

TOPOGRAPHY

Topography was determined for the tidal flood region
extending from open lagoon water at the west side of the
island along the 350 m long channel and the southern in-
terior pond (Urish et al., 2003; Wright et al., 1991). Some
36 semipermanent monitor locations were established in
1988 in the intertidal swamp to obtain water level and wa-
ter quality measurements. The locations were marked with
2 cm diameter polyvinyl chloride (PVC) pipes driven into
the ground in a grid pattern. Horizontal control was es-
tablished by field measurement with a 35 m long tape and
conventional level and transit surveying techniques (Wolf
and Ghilani, 2006), later located with Global Positioning
System (GPS) technology using a Garmin GPS 76. These
data were later used for georeferencing of all island features
(Rodriguez and Feller, 2004). Vertical control for land and
water measurements was determined from a primary da-
tum reference point on the east side of the island to which
an arbitrary datum was assigned. The initial elevation as-
signed to this reference point was 3.05 m with all readings
later adjusted to an approximate mean lower low water
(MLLW) after several years of time segments of about 2
weeks; one long record of 9 months of tidal data was ob-
tained. A datum lower than the typical terrestrial datum of
“mean sea level” was used to maintain both topography
and water level values positive to the extent possible.

Two principal surveying transects across the island
were established: (1) from the lagoon to the bend in the
channel along an east-to-west run including 6 points (F1
to Al) and (2) from the bend in the channel to the south
pond along a north-to-south run of 12 more points (A1
to A12). In addition, 3 to 5 points were determined per-
pendicular to each transect point. These secondary points
were spaced approximately 15 m apart. These established

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

points, as located on Figure 5, were the primary location
references for all subsequent data collection.

Automated pressure transducer water level loggers
(In-Situ Environmental Data Logger Model SE 1000c
with pressure transducer probes) were employed at five
locations for short-term (1-2 weeks) measurements.
These units were vented to automatically compensate for
ambient atmospheric pressure. Later in the study period
12 of these locations became long-term monitoring sta-
tions with automated self-contained water level loggers
(Remote Data Systems, Navassa, N. C.) that remained in
place for as long as 12 months to record data at 30 min
intervals with an accuracy of about 3 mm. Self-contained
automated temperature loggers (Optic Stowaway by On-
set Computer Corporation) were also deployed to record
temperatures at 30 min intervals for as long as 9 months.
In addition to the monitor locations, stilling wells consist-
ing of slotted 15 cm diameter PVC pipe for both manual
and instrumented tide measurements were established at
both shorelines of the island, and later in the study these
were correlated with a primary oceanographic/climatolog-
ical data collection station established at the Smithsonian
Research Station on Carrie Bow Cay, 4 km southeast. The
tops of the stilling wells were initially assigned an eleva-
tion based on the same arbitrary datum as used at the key
datum reference points. Elevations were established on
the tops of all reference station pipes using survey level-
ing techniques with a Topcon Automatic Level (model
ATF-1A). The coordination of tides at West Island and
Carrie Bow Cay was accomplished by comparison of a
series of six separate short-term tidal cycle measurements
taken concurrently at both stations.

HYDROLOGY

Water flow direction and velocities during various posi-
tions of the tide cycle were determined using conventional
stream gauging techniques along channel cross-sections, or
“reaches:” section A—-A’ was defined between survey points
A1 and D1 and reach B between survey points D1 and E1.
The measurements were taken at various times during the
tidal cycle using a Marsh-McBirney electromagnetic current
velocity meter and standard stream channel cross-sectioning
methods (Watson and Burnett, 1995). Velocities and water
depths were measured at 0.6- to 1.5 m intervals perpendicu-
lar to the flow to provide 25 to 50 individual measurements
at each cross section. These measurements were then plot-
ted to determine flow volumes and flow friction factors and
to examine trends.

NUMBER 38

CONTOUR INTERVAL: 6cm

LEGEND

\, ~~ SHORELINE

\\\, © ~KEY MONITOR STATIONS
A

HIGH TIDE
9S OVERFLOW

CHANNEL

Sy
Nas
‘ Na

HIGH TIDE
Hf OVERFLOW

-
ae s
3 fae
px ne
a w~

a

FIGURE 5. Topographic contour map of interior of West Island showing locations of key monitoring sta-
tions and representative channel cross section A-A’. Elevation datum = mean lower low water (MLLW).
Contour interval = 6 cm.

479

480 e

Water salinity (+0.1 ppt), conductivity (+0.1 uS), and
temperature (+0.1°C) were measured with a YSI 30 S-C-T
(Yellow Springs, OH) in the field and in the laboratory.
Salinity measurements were also made in the field using
a refractometer (model 366ATC; Vista) with an accuracy
of +1 ppt.

Flow patterns were also observed and evaluated by
use of dye studies, both at the surface water level and by
helium balloon low-level photography in 1990 and 1991,
and later by high-level aircraft photography during a tidal
cycle in 1993. Although the balloon photography was only
of limited value because upper air wind currents caused the
balloon to drift off the island, aircraft photography was
highly successful. Large targets, approximately 1 X 2 m
in size, were marked and placed at each station for dye
movement referencing. Continuing runs at 0.5 h intervals
were made across the island on the same flight path at an
altitude of 150 m. Photographs were taken during each
run with a SLR camera with AF Zoom 35-70 mm lens
(Minolta 5000 MAXXUM), thus enabling both the flow
directions and dispersion within the mangrove system to
be observed. Slugs of Rhodamine fluorescent dye, a highly
visible but nontoxic dye, were placed at three stations
at the start of the observation period. The dye remained
highly visible during one tidal cycle. Continuing, but di-
minishing, levels of the fluorescence were measured in the
laboratory on water samples taken during three subse-
quent tidal cycles. The series of photographs taken from
the aircraft runs were reduced to a time sequence of plots
and then used by George L. Venable of the Smithsonian
Institution to produce an animated video of the dye move-
ment for further study.

RESULTS
TOPOGRAPHY

A topographic map of the intertidal flood zone region
of West Island is shown as Figure 5. The highly irregular
nature of the bottom is evident by inspection of the con-
tour pattern. The region of tidal flow and flooded area is
characterized by relatively flat areas with water depths fre-
quently only about 25 cm over much of the flow system,
but highlighted by sections more than 1 m in depth, such
as occur between stations A4 and AS. Such deep holes are
not necessarily coherent with the main flow channel. A
cross-section (A—A’) plot (Figure 6) at station AS depicts
the extreme changes in channel bottom that exist at this
location. In contrast, the other significant feature of the
system is a very large shallow pond of about 2.2 ha at sta-

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

tions A10 to A12 at the south central part of the island.
This region contains only sparse dwarf red mangrove with
a flooded depth of about 0.25 m. Additionally, examina-
tion of the topographic map shows ground level at the east
shoreline of this pond is about 6 cm lower than the rest
of the island periphery. At high tides these limited lower
topographic zones allow lagoon water from outside to en-
ter the internal swamp flow system. In particular, high tide
waters overtop the island perimeter at two other locations
on the east side into the central region, causing short-term
hydrological anomalies of temperature and salinity, as well
as a somewhat irregular tidal signal, in the system.

HYDROLOGY

Figure 7 is a five-day plot showing the typical tidal
signal as it enters the swamp system at TG2. As the signal
enters the mangrove system, the frictional resistance of the
roots cause attenuation in tide amplitude as well as a time
lag in the highs and lows.

The unique hydrological nature of the flow in the
mangrove ecosystem is characterized by a very shallow
water depth that averages only about 0.5 m at low tide
to 0.67 m at high tide, although great variations exist.
However, because of the very flat topography, even this
low tidal fluctuation causes an extensive and significant
hydroperiod of wetting and drying with important impli-
cations to the mangrove ecosystem. As depicted on Fig-
ure 8, the area typically covered by water during high tide
is about double that covered by water during low tide.
Doyle (2003) states that his controlled field experiments

(A) ‘

STATION

ELEVATION (METERS)

0 20 40 60 80 100 (m)
DISTANCE

FIGURE 6. West Island cross section A—A’ at station AS showing
relative relationship of high and low tides to bottom contours. Green
= range of flooding from the tide; blue = low tide flow. Elevation
datum is arbitrary.

2 1day

Elevation (m)

Days

FIGURE 7. Plot of five-day tide sequence showing relative ranges and
asymmetry of the tide at West Island. Elevation datum = approxi-
mate mean sea level.

“suggest that the hydroperiod—the rate and level of tidal
exchange—plays a much more important role in determin-
ing mangrove growth and success than previously docu-
mented.” Figure 9 presents a conceptual plot showing
the wetting—drying cycle during the sequential phases of
the tide. Perhaps even more important, the tide range is
sufficient to cause reversal of flow direction and velocity
throughout the system during each cycle.

Figure 10 shows the changing characteristics of the
tidal signal at three stations as it moves inland in a tortu-
ous path through the mangrove ecosystem. A tide range
of 13 cm at the island margin is attenuated to 8 cm at a
location 50 m landward and to 3 cm at a location 200 m
landward in the main flow channel. Concurrently, there is
a lag time of 1 h for high tide and 2 h for low tide at 50 m
landward, and of 2 h for high tide and 6 h for low tide at
200 m landward. The great difference in lag time between
highs and lows is caused by the much greater influence of
root density during a receding tide; this is also illustrated
by the asymmetrical characteristic of the tidal signal as it
transposes landward.

The seasonal climatic variations had a profound effect
on the monthly hydrological budget, especially when the
high evapotranspiration was considered. Figure 11 shows
the approximate seasonal relationships of precipitation,
surface water evaporation, and vegetation transpiration
(evapotranspiration), assuming a total annual rainfall of
1,885 mm. This value and the estimated monthly values
are based on limited (about 5 years) available data that

NUMBER 38 e¢ 481

mag

FIGURE 8. Plan of West Island showing aerial extent of tidal flood-
ing between average daily low and high tides. Stippling on map
shows relative density of vegetation. (Adapted from drawing by
Molly K. Ryan of the Smithsonian Institution in 2002.)

have been collected at Carrie Bow Cay and correlated
with the longer-term record at the mainland Melinda
Forest Station. The potential evapotranspiration values
for each month were calculated from the Thornthwaite
equation (Dunne and Leopold, 1978; Thornwaithe and
Mather, 1987) using a partial record of temperature and
solar radiation available for Carrie Bow Cay. Examina-
tion of the water budget shows a deficit of precipitation as

482 e

Wet period | Wet period II

Inundation
Phase II

Drawdown
Phase II

Drawdown
Phase |

WATER LEVEL

Dry period

TIME

FIGURE 9. Conceptual plot of tidal phases during mixed semidiur-
nal hydroperiod event. (Adapted from Boulton and Brock, 1999.)

compared with evapotranspiration February through June
during the “dry” season and a surplus during the “wet”
season months from July through January. This “dry” sea-
son water deficit has an extreme effect on the surface wa-
ter in the semienclosed interior swamp system, especially
in the poorly flushed shallow pond at the south end of the
island. Figure 12 is a composite plot of water temperature
and salinity measured along the main flood channel over
several years during the “dry” and “wet” seasons. Near
the inflowing/outflowing location at station F1 the values
approach those of lagoon water at the periphery of the

Tidal Period 2/09 - 2/11, 2003

=
iS 5)
= 15)
[ay
>

Oe
rao)
Ox
v= 0)
Or
t
=<

Sequential Time (h)

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

island, but in the shallow pond the value ranges are much
more extreme, with salinity ranging from 5 to 45 ppt and
temperatures ranging from 25°C to 40°C. Some tempera-
ture and salinities even exceed these values in particularly
isolated locations.

The elevations of the water surfaces and velocities at
three stations within the first 150 m inland from the shore,
as compared with the primary tide signal at shore station
TG2, are shown in Figure 13. It is to be noted that at all
stations the maximum velocities occur during the middle
of the falling tide, with the highest velocities found nearest
shore. This pattern is comparable with the tidal asymme-
try and velocity patterns found by Bryce et al. (2003) in
mangrove creek systems in Australia. The distribution of
flow and variations of velocity for a typical channel sec-
tion are illustrated in Figure 14 for a cross section at sta-
tion A4. The data were acquired during a mid-tide flood
tide at a time of maximum velocity. The upper part of Fig-
ure 14 shows depths of water across the section at the spe-
cific time of the velocity measurements shown in the lower
part. The velocities shown are an average determined from
a series of velocities measured at a series of depths over
the shortest time period possible. As indicated the velocity
changes dramatically, from 2.0 to 0 cm/s, across the sec-
tion, although there is a general pattern of greater velocity
at the deeper parts. However, this is contradictory to the
observation that the deeper part on the right side does not

50 METERS

FIGURE 10. Tidal fluctuation plots at three interior monitoring stations—TG2, A, and B, as shown on pho-
tograph at right—during maximum velocity of flood tide on 27 May 1988. Yellow dots on the photograph
are locations of monitor stations. Elevation datum is arbitrary.

300

PRECIPITATION

© 200
E
—
z
5
oO
S 100
<
0
lero Ae Mad dk SOON D
31
30
ae) 29
Ww
c 28
5
=
eT.
Ww
s
= 26
=
25
24

oe OM SAM ad: «J oA Ss © IN D

MONTHS

FIGURE 11. Hydrological budget for West Island showing an-
nual pattern of precipitation, evapotranspiration, and temperature.
Shaded orange area = the “dry season” with net deficit of water
from evapotranspiration; blue shaded area = the “wet season” with
a surplus of precipitation.

have a high velocity. Inspection of the topographic plan (see
Figure 5) provides the explanation: The deeper right-hand
feature is a part of a closed depression, whereas the deeper
part on the left is continuous with the main channel flow.
The average flow velocity between points in the swamp
is reflective of both the tortuous path of flow through the
mangroves and the frictional resistance of the mangrove
root system and the channel bottom. This resistance can

NUMBER 38 ¢ 483

be quantified by inverse calculation of the Manning equa-
tion for stream flow (Watson and Burnett, 1995). Al-
though the Manning equation was originally developed
for stream flow, it has a logical deterministic relationship
that has been used successfully by other researchers for
mangrove flow characterization and modeling (Wolanski
et al., 1980). The Manning equation in MKS unit format
(Lindsley and Franzini, 1979) is

Vieiin RS,

where V is the average velocity, n is the Manning rough-
ness coefficient, R is the hydraulic radius (cross-section
area divided by wetted perimeter), and S is the slope, or
hydraulic gradient, of the water surface.

Manning’s roughness coefficient, n, was determined at
various locations in the flow system and at various times
in highly fluctuating stream depth and current direction.
The determined values of n for these measurements ranged

MONITOR STATION
Fi Ei Di C1 A4 AS A6 A7 A7 AQ P A111

TEMPERATURE (°C)

WET SEASON (JANUARY)

SALINITY (%o)

SIDE CHANNEL
CONNECTION

i) 1 2 3 4 (m)
DISTANCE FROM LAGOON (x100)

a a ae RE
FIGURE 12. Plot showing ranges of salinity (black dots) and tempera-
ture (circles) during “dry” and “wet” season conditions at the south-
ern shallow pond on West Island. Monitoring stations (see Figure 5)
with distances from the open water lagoon at the west periphery of
the island are indicated.

484 e

ELEVATION
eo (mm)
& FO
SSS.
& So
= ~“* —-——@64 5 x9

VELOCITY x

8 ) 10 14 12 13
TIME (h)

FIGURE 13. Plot of relative elevations (arbitrary datum) and veloci-
ties during falling limb of tide at monitor stations C1, E1, and F1
(see Figure 5).

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

5 yyy Uff
My Y 7)

= 20 yf Yyf |
= -20 ]
E 2 N Nin

= NIN

: N NN

8 N ANN Na

3 NN NON NNN NANA N NANN

é (oN NNNeNANSNANNoo Na NNNN

Ss 0 10 20 30 40

WIDTH OF CROSS-SECTION (m)

FIGURE 14. Water depths and velocities at cross-section A—A’, at
monitoring station A4 (see Figure 5) during a flood tide with maxi-
mum velocities, on 27 May 1988.

from 0.084 to 0.445, far higher than the typical 0.035
found even for natural channels with stones and weeds.
These values however are comparable to those ranges of
0.2 to 0.7 determined by Wolanski et al. (1992) for flow in
southern Japan mangrove systems. Based on earlier stud-
ies, Wolanski et al. (1980) had earlier suggested that n is of
the order of 0.2-0.4 in mangrove swamps. Table 1 shows
the parameters and results for calculation of the Manning
coefficient for section A-A’, with depth and velocity char-
acteristics as depicted in Figure 14. It is to be noted in
Table 1 that the value of n is greatest at shallow water
depth and lower velocities. Again, this is comparable to
the findings of Wolanski et al. (1992).

TABLE 1. Summary of hydraulic parameters for cross-section A—A’ at Station A4 (see Figure 5).

Average water

Hydraulic slope,

Average Manning’s

Observation 4 depth (cm) S (m/m) velocity (cm/s) Flow (m3/s) coefficient, n
1 11.9 0.000117 0.6 0.034 0.415
2 13.4 0.000110 0.6 0.037 0.445
3 14.3 0.000102 1.3 0.085 0.206
4 14.6 0.000098 1.1 0.069 0.261
5 16.4 0.000086 1.3 0.097 0.210
6 20.3 0.000055 1.6 0.145 0.159

* Fach observation with the associated calculations is based on 30 measurements across cross-section A—A’ as detailed in Figure 14.

A dye flow study was accomplished during a high tide
period on 5 June 1993. Single slugs of Rhodamine fluo-
rescent dye, which produced a distinctive red color, were
placed at three locations (monitor stations D1, A6, and
A10; see Figure 5) along the main channel early in the
morning. Large visual targets were placed at the monitor
stations in the mangrove swamp to enable referencing the
dye positions during movement.

The movement of dye was documented by aerial pho-
tography from an aircraft flown in a fixed flight pattern,
and at a fixed altitude of 150 m (500 ft.), Runs were made
at 0.5 h intervals. Figure 15 is a series of drawings made
from 11 of these runs, depicting the leading edge of the dye
with time. The 7 h period of measurement relative to the
position of the tide at the exterior lagoon is depicted on
the inset tide plot of Figure 15. Figure 16 is a high oblique
photograph of West Island taken from an altitude of about
600 m (2,000 feet) showing the position of the dye at 9:10
AM, shortly after high tide. The aircraft run pattern starts
over monitor station TG 2 and progresses east, turning

Run

NUMBER 38 ¢ 485

south to proceed over the large pond at the south end of
the island. The series of photographs have been converted
to an animated visual program by George L. Venable of
the Smithsonian Institution (URL http://www.uri.edu/cve/
dye.moy) that clearly shows the oscillation of the water
of the mangrove swamp water as the dye at station D1,
some 70 m from the lagoon, first went to the east, then re-
versed to finally discharge into the lagoon. The dye flow at
station A6 also oscillated, then merged with the outgoing
dye from station A10 approximately 120 m south of A6,
toward the pond. Interestingly, the dye placed at A10, at
the north margin of the pond, also moved into the pond
and then flowed in a circulatory pattern. This pattern may
be caused by new lagoon water overflowing the rim of
the pond to the east because the tide during the period of
observation was a relatively high spring tide. Finally the
dye from D1 discharges into the lagoon, and the merged
A6/A10 dye moves north. Previous water level studies with
measured levels of fluorescent dye indicated that this dye
persists at continuing reduced levels in the central locations

o
cy
s
8
~
7

RELATIVE WATER LEVEL (cm.)

FIGURE 15. Sequential plots of dye flow patterns over a 7 h period from high tide through low tide, during
dye flow test on 5 June 1993. The plot in the lower right-hand corner shows the relative time position of runs

with tide levels at station TG2.

486 °

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 16. Aerial photograph of West Island looking to the northwest from an altitude of 600 m during the
5 June 1993 dye flow test, corresponding approximately to the time of run 4 of Figure 15. The dye is evident
as the red configuration in the south pond.

for several tidal cycles before finally being flushed and dis-
sipated into the lagoon.

DISCUSSION

An intimate knowledge of the topography of the flow
and flood area is essential to an understanding of the eco-
system of a semiclosed mangrove hydrosystem such as
West Island. The interior water system of West Island is
primarily tide-induced flow, with modifications caused by
precipitation and seasonal climatic change. Hence, it is
also very important that the climatic factors be considered
in conjunction with the hydrographic characteristics.

The tidally induced hydrodynamics of the water flow
in tidal channels and ponds of an overwashed mangrove
island, in conjunction with the topography, greatly affect
the ecosystem and vitality of the resident mangrove sys-
tems. Analysis of the temporal and spatial characteristics

of a 21.5 ha island hydrological flow system shows that
where flooding and current flow becomes more vigorous,
the growth of the red mangrove is enhanced. It was found,
during the 18-year period of this study, that hydrological
changes such as increased tidal flow, from anthropogenic
as well as natural causes, enhanced the growth of the red
mangrove. On the other hand, observations of relict tree
remains indicate that historically the interior of the island
experienced a transition in mangrove species from black
mangrove to dwarf red mangrove, possibly because of
higher water levels and with poor flushing in the island
interior. This concept is in concurrence with findings of
Knight et al. (2008) in studies of patterns of tidal flooding
within a mangrove forest in Southeast Queensland, Aus-
tralia, that “mangrove basin types represent a succession
in mangrove forest development that corresponds with in-
creasing water depth and tree maturation over time.”
Detailed mapping of the topography of the intertidal
interior region of the island reveals poorly defined flow

channels that vary greatly in depth and width. Within this
system anomalous deep sections exist, further contribut-
ing to the complexity of flow. The root structure of the
mangroves and the irregular bottom result in a frictional
resistance to the flow, quantified by Manning’s n as being
as much as 10 times greater than that of conventional ter-
restrial stream channels. In studies in mangrove swamps
on Iriomote Island, Japan, Kobashi and Mazda (2005)
stress the importance of the hydraulic resistance of man-
grove vegetation in determining the flow patterns, espe-
cially in reducing the velocity component perpendicular
to the main channel. Accordingly, the interaction of the
mangrove itself is a determinate factor in stream flow and
the resulting flushing action, important to the vitality of
the mangrove. It appears to be particularly relevant to the
transport of nutrients and other physicochemical condi-
tions important to the growth of the mangrove. The driv-
ing force for the flow within the mangrove hydrosystem is
the ever-changing hydraulic gradient induced by the tide.
Accordingly, the flow moves, at varying velocities, in and
out of the interior mangrove swamp with the tide. As a
result the seawater entering the mangroves not only fol-
lows a constantly changing path, but is regularly reversed
in direction, and consequently it takes at least several tidal
cycles for flushing of the island interior. There are indica-
tions that the central part of West Island is flooding more
over a span of years, causing commensurate changes in
the mangrove types capable of surviving in the changed
regime, a process described by Knight et al. (2008). In
other areas the geomorphology of the land is changing
consequent to sediment transport, detrius deposition, and
subsidence from peat compaction. In this regard Bryce et
al. (2003), in studies of a small mangrove creek system
near Townsville, Australia, evaluated the role of hydro-
dynamics in the sediment transport process. Importantly
they observed that sediment transport appears to be a
seasonal phenomenon, with net flux going either land-
ward or seaward, but they conclude that the net sedi-
ment transport for the overall system may be close to
long-term equilibrium. They do state that mangrove
swamp areas (in the tidal overflow regions) are most
likely to be places of sediment accumulation; if so, the
more shallow areas of West Island, when flooded at high
tide, may experience accumulation from redistribution
of sediment within the system as well as from direct leaf
drop and in situ detritus accumulation.

The data showed that the annual pattern of precipita-
tion and temperature greatly affects temperature and sa-
linity in the poorly flushed interior pond. On an annual
basis there was a net discharge of water from island to the

NUMBER 38 ¢ 487

exterior lagoon because of precipitation. However, when
the monthly climatic factors are considered it is apparent
that during the “dry period” of February to May there is a
net loss of freshwater in the island water budget, with high
evaporation creating high-temperature hypersaline water
in the interior. When the island is in the rainy season, July
through December, the reverse is true, with the interior
water becoming cooler and fresher from the rains (see
Figure 12). In the extreme case, as described by Wolanski
et al. (1992) for tropical mangrove systems on the coast
of northern Australia, “The balance between rainfall and
evaporation, in conjunction with tidal variations, is the
key factor in determining if the upper levels of the swamp
are (tidally flushed) swamp or (hypersaline) tidal flat.” A
further important implication for the shallow pond in the
south part of West Island is, as stated by Wolanski et al.
(1992), “rainfall significantly affects porewater salinity
and it is likely that it also affects nutrient levels within
the swamp substrate, particularly in areas where regular
flooding by the tides does not occur.” On West Island,
during the “dry period” of February through May, evapo-
transpiration is approximately three times that of precipi-
tation. However, during the “wet period” of June through
December, conditions are reversed with evapotranspira-
tion being approximately one-half that of precipitation
(see Figure 11). Thus, the net effect on the poorly flushed
interior areas of the West Island mangrove system is that
of greatly increased salinity during the “dry season” and
short periods of nearly fresh water from rain storms in the
wet season (see Figure 12).

The effects of human intrusion into the natural eco-
system are illustrated by Figure 17, an aerial photograph
showing the survey lines newly cut in 1993. At that time
the strongest flow from the coastal seawater was some
25 m south of the west-east running survey cut, as identi-
fied by the darker, more vigorous vegetation. By 2003 the
main flow had moved north to the survey cut itself as the
cutting as well as foot traffic in the cut had deepened that
area. During the course of the investigations, it was ob-
served that the previous dwarf red mangrove trees along-
side these survey lines initiated signs of vigorous growth
as a result of the increased flushing, as illustrated by the
photograph in Figure 18.

Although previous studies have shown that for the
past 8,000 years the mangrove growth has managed to
keep up with rising sea level because of peat accretion, the
future may be in doubt because of anticipated greatly in-
creasing sea-level rise rates (Mckee et al., 2007). The abil-
ity of the island to adjust to rising sea level has important
implications for a future that will include sea levels rising

488 e

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

FIGURE 17. Aerial photograph taken in 1993 (looking south) showing survey lines that were newly cut in
1993. The original principal tidal flow path is evident as the darker green vegetation approximately 30 m
south of the 1993 east-west survey line.

at a rate much greater than that experienced over the past
8,000 years when mangroves first appeared and flourished
on Twin Cays. As McKee et al. (2007) have stated, “Rates
of subsurface plus subsurface (root) accretion in fringe,
transition and interior zones at Twin Cay were 10.4, 6.3,
and 2.0 mm/year. Fringe mangroves have kept up and could
accommodate eustatic sea level rise of 4 mm/year if current
rates of accretion were maintained. If eustatic rates exceed 5
mm/year then these mangrove islands would not be likely to
persist, assuming all other conditions remain unchanged.”
The islands of Twin Cays, with a history of compre-
hensive ecological research, remain an important location
for measuring and evaluating changes in the mangrove
and associated ecosystems because they occur in a world
of dramatic coastal change. Much analytical work re-
mains to link the dynamic hydrology of the mangrove is-
land to the physiological parameters essential to mangrove
growth. The research site of Twin Cays, with three decades
of baseline data and research, is a very important asset for
better understanding the ecosystem of the mangrove. It is
very important that this work continue and build on the
substantial foundation of information that now exists.

CONCLUSIONS

Overwashed mangrove islands are extremely complex
ecosystems. They are essentially self-dependent, and the vi-
tality of the resident mangrove species is primarily a result
of the tide that produces the essential hydrological func-
tions of flushing and nutrient transport. The topography,
the geomorphology, and even the existence of a mangrove
island are products of the island vegetation itself. This in-
teraction affecting the island configuration is constantly
changing as the mangrove forest with its multiple species
adjusts to higher sea levels and the resultant changes in
hydrological flow and flooding parameters.

The interior of the island is subject to extremes of tem-
perature (20°-40°C) and salinity (5-45 ppt) with limited
flushing that may adversely affect the vitality and existence
of the mangrove, as well as the natural selection of man-
grove species. A comparison of the hydrological parameters
and flow regimes in the regions of vigorously growing red
mangrove with that of dwarf red mangrove strongly suggests
that enhanced communication with external lagoon water is
best for the vitality of the red mangrove on Twin Cays.

Post-
channel

Pre-
channel

FIGURE 18. Photograph of red mangrove branch taken in 2003
near monitoring station D1 (see Figure 5) showing progressive in-
crease of growth between sequential growth rings after survey lines
were cut approximately 10 years previously. Note that the tape is
marked decimally in feet.

The flow within the island is strongly influenced by
the substantial frictional resistance of the mangrove root
system. This dense root system serves to greatly attenuate
the tidal amplitude as it progresses into the island, creat-
ing a reduced hydraulic gradient for water movement. The
resultant reduced flow creates a poorly flushed island inte-
rior with poor mangrove growth.

Extensive land clearing, especially along the coastal
margins, has long-term continuing effects of mangrove
loss from which the island may never recover (Macintyre
et al., 2009). In contrast, limited incursions such as the ob-
served survey line cutting may shift, but enhance, channel
flow, promoting more vigorous red mangrove growth. In
extensive field research (Feller et al., 2003), it was found
that the patterns of nutrient availability within and among
mangrove ecosystems are complex. Feller et al. (1999)
showed the dramatic effects of nutrient enrichment on
mangrove growth as well the changes in nutrient limita-
tions that can take place within relatively short distances
in swamp ecosystems. At least in the case of the nutrient-
poor (P-limitation) condition of the sparse red mangrove
in the interior of the island, the cause of nutrient limitation
seems to be poor flushing, which limits the refreshing of

NUMBER 38 ¢ 489

the system with phosphorus-rich lagoon water from tidal
flooding.

A further concern is that of the effect of rising sea
level on the ability of the mangrove to survive. The hydro-
dynamics of the mangrove system greatly influences the
mangrove ecosystem both by transport of nutrients and
sediment and by the direct ability of the geomorphology of
the island to develop to keep pace with rising sea level as it
has in the past (McKee et al., 2007). At the least it appears
that differential growth of mangroves will occur as flood-
ing occurs and the hydrodynamics of the system changes.

ACKNOWLEDGMENTS

The authors are grateful to the Smithsonian Institu-
tion for the opportunity and research support provided
over the past twenty years, to the University of Rhode Is-
land for its provision of faculty and staff, especially equip-
ment engineer Brian Gray, in conducting this research, and
to the many fellow island researchers who have enriched
our professional lives as a part of a multidisciplinary team.
We especially thank Klaus Ruetzler, Director of the Smith-
sonian Caribbean Coral Reef Ecosystem Program, for his
encouragement in pursuing the hydrological aspect of the
mangrove ecosystem; Michael Carpenter, Smithsonian In-
stitution (SI), for logistic support that made the fieldwork
possible; Molly Ryan (SI) for her most useful and beauti-
ful artwork; and George Venable (SI) for his computer-
animated graphics that greatly aided our interpretation of
the complex water movement in the mangrove swamps.
We are grateful for financial support that was provided by
the Exxon Corporation, the National Science Foundation
Biocomplexity Grant to I. C. Feller (DEB-9981535), and
the Smithsonian Institution. This is contribution number
852 of the Caribbean Coral Reef Ecosystems Program,
Smithsonian Institution, and was supported in part by the
Hunterdon Oceanographic Research Fund.

LITERATURE CITED

Alongi, D. M. 2002. Present State and Future of the World’s Mangrove
Forests. Environmental Conservation, 24:331-349.

Alongi, D. M., and A. D. McKinnon. 2005. The Cycling and Fate of
Terrestrially-Derived Sediments and Nutrients in the Coastal Zone of
the Great Barrier Reef Shelf. Marine Pollution Bulletin, 51:239-252.

Ball, M. C. 1988. Salinity Tolerance in the Mangroves Aegiceras cornicu-
latum and Avicennia maria. 1. Water Use in Relation to Growth,
Carbon Partitioning, and Salt Balance. Australian Journal of Plant
Physiology, 15:447-464.

Barbier, E. B. 2006. Natural Barriers to Natural Disasters: Replanting
Mangroves After the Tsunami. Frontiers in Ecology and the Envi-
ronment, 4(3):124-131.

490 e

Boulton, A. J., and M. A. Brock. 1999. Australian Freshwater Ecology: Pro-
cesses and Management. Adelaide, Australia: Gleneagles Publishing.

Bryce, S., P. Larcombe, and P. V. Ridd. 2003. Hydrodynamics and Geo-
morphological Controls on Suspended Sediment Transport in Man-
grove Creek Systems, a Case Study: Cocoa Creek, Townsville, Aus-
tralia. Estuarine, Coastal and Shelf Science, 56:415-431.

Costanza, R., O. Perez-Maquco, M. L. Martinez, P. Sutton, S. L. Anderson,
and K. Mulder. 2008. The Value of Coastal Wetlands for Hurricane
Protection. Royal Swedish Academy of Sciences, 37(4):241-248.

Danielson, FE, M. K. Sorenson, M. F. Olwig, V. Selvam, F. Parish, N. D.
Burgess, T. Hiraishi, V. M. Karunagaran, M. S. Rasmussen, L. B.
Hansen, A. Quarto, and N. Suryadiputra. 2005. The Asian Tsu-
nami: A Protective Role for Coastal Vegetation. Science, 310:643.

Doyle, T. W. 2003. Effects of Hydrology on Red Mangrove Recruits. U. S.
Geological Survey Report USGS 023-03. Lafayette, La.: National
Wetlands Research Center.

Duke, N. C., J. O. Meynecke, S. Dittman, A. M. Ellison, K. Anger, U.
Berger, C. Cannicci, K. Diele, K. C. Ewell, C. D. Field, N. Koedam,
S. Y. Lee, C. Marchand, I. Nordhaus, and F. Dahdouh-Guebes.
2007. A World Without Mangroves? Science, 317:41-42.

Dunne, T., and L. B. Leopold. 1978. Water in Environmental Planning,
pp. 135-138. San Francisco: W. H. Freeman and Company.

Feller, I. C., K. McKee, D. F. Whigham, and J. P. O'Neill. 2003. Nitrogen
vs. Phosphorous Limitation across an Ecotonal Gradient in Man-
grove Forest. Biogeochemisty, 62:145-175.

Feller, I. C., D. EK Whigham, J. P. O9Neil, and K. M. Mckee. 1999. Effect
of Nutrient Enrichment Within-Stand Nutrient Cycling in Man-
grove Ecosystems in Belize. Ecology, 80:2193-2205.

Forman, R. T. T., and M. Godron. 1986. Landscape Ecology. New York:
John Wiley and Sons.

Furukawa, K., and E. Wolanski. 1996. Sedimentation in Mangrove For-
ests. Mangroves and Salt Marshes, 1:3-10.

Hutchings, P. A., and P. Saenger. 1987. Ecology of Mangroves. St. Lucia:
University of Queensland Press.

Kjerfve, B., K. Riitzler, and G. H. Kierspe. 1982. “Tides at Carrie Bow
Cay, Belize.” In The Atlantic Barrier Reef Ecosystem at Carrie Bow
Cay, Belize, ed. K. Riitzler and I. G. Macintyre. Smithsonian Con-
tributions to the Marine Sciences, 12:47-5S1.

Knight, J. M., P. E. R. Dale, R. J. K. Dunn, G. J. Broadbent, and C. J.
Lemckert. 2008. Patterns of Tidal Flooding within a Mangrove
Forest: Coombabah, Southeast Queensland, Australia. Estaurine,
Coastal and Shelf Science, 76:580-593.

Kobashi, D., and Y. Mazda. 2005. Tidal Flow in Riverine-type Man-
groves. Wetlands Ecology and Management, 13:315-619.

Lindsley, R. K., and J. B. Franzini. 1979. Water Resources Engineering
(Open Channel). New York: McGraw Hill, pp. 251-256.

Lovelock, C. E. 2008. Soil Respiration and Belowground Carbon Alloca-
tion in Mangrove Forests. Ecosystems, 11:342-354.

Lugo, A. E., and S. C. Snedaker. 1974. The Ecology of Mangroves. An-
nual Review of Ecology and Systematics, 5:39-64.

Macintyre, I. G., and M. A. Toscano. 2004. The Pleistocene Limestone
Foundation below Twin Cays, Belize, Central America. Atoll Re-
search Bulletin, 511:1-16.

Macintyre, I. G., M. A. Toscano, I. C. Feller, and M. Faust. 2009. Decimat-
ing Mangrove Forests for Commercial Development in the Pelican
Cays, Belize: Long-Term Ecological Loss for Short-Term Gain? Smith-
sonian Contributions to the Marine Sciences, No. 38:281-289.

Macintyre, I. G., M. A. Tocsano, R. G. Lightly, and G. B. Bond. 2004.
Holocene History of the Mangrove Islands of Twin Cays, Belize,
Central America. Atoll Research Bulletin, 510:1-16.

Mazda, Y., N. Knazawa, and T. Kurokawa. 1999. Dependence of Dis-
persion on Vegetation Density in a Tidal Creek-Mangrove Swamp
System. Mangrove and Salt Marshes, 3(1):59-66.

Mazda, Y., D. Kobashi, and S. Okada. 2005. Tidal Scale Hydrodynamics
Within Mangrove Swamps. Wetlands Ecology and Management,
13:647-655.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Mazda, Y., E. Wolanski, B. King, A Sase, O. Otsuka, and M. Magi. 1997.
Drag Forces due to Vegetation in Mangrove Swamps. Mangroves
and Salt Marshes, 1:193-199.

McKee, K. L., D. Cahoon, and I. C. Feller. 2007. Caribbean Mangroves
Adjust to Rising Sea Level Through Biotic Controls on Soil Eleva-
tion Change. Global Ecology and Biogeography, 16:546-556.

Odum, W. E., and E. J. Heald. 1972. Trophic Analyses of an Estuarine
Mangrove Community. Bulletin of Marine Science, 22:671-738.

. 1975. “The Detritus-Based Food Web of an Estaurine Mangrove
Community.” In Estaurine Research, ed. L. E. Cronin, pp. 265—
286. New York: Academic Press.

Rodriguez, W., and I. C. Feller. 2004. Mangrove Landscape Character-
ization and Change in Twin Cays, Belize Using Aerial Photography
and IKONOS Satellite Data. Atoll Research Bulletin, 513:1-22.

Ritzler, K., and I. C. Feller. 1988. Mangrove Swamp Communities.
Oceanus, 30(4):16-24.

. 1996. Caribbean Mangrove Swamps. Scientific American,
274(3):94-99.

Ritzler, K., and J. D. Ferrais. 1982. “Terrestrial Environment and
Climate, Carrie Bow Cay, Belize.” In The Atlantic Barrier Reef
Ecosystem at Carrie Bow Cay, Belize, ed. K. Ritzler and I. G.
Macintyre. Smithsonian Contributions to the Marine Sciences,
12:77-91.

Riutzler, K., and I. G. Macintyre. 1982. “Introduction.” In The Atlantic
Barrier Reef Ecosystem at Carrie Bow Cay, Belize, ed. K. Ritzler
and I. G. Macintyre. Smithsonian Contributions to the Marine Sci-
ences, 12:1-7.

Taylor, D. S., E. A. Reyier, W. P. Davis, and C. C. MclIvor. 2007. Man-
grove Removal in the Belize Cays: Effects on Mangrove-Associated
Fish Assemblages in the Intertidal and Subtidal. Bulletin of Marine
Science, 80(3):879-890.

Thornwaithe, C. W., and J. R. Mather. 1987. Instructions and Tables for
Computing Potential Evapo-transpiration and the Water Balance,
pp. 185-311. Centerton, N.J.: Laboratory of Climatology.

Tomlinson, P. B. 1986. The Botany of Mangroves. London: Cambridge
University Press.

Toscano, M. A., and I. G. Macintyre. 2003. Corrected Western Atlantic
Sea-Level Curve for the Last 11,000 Years Based on Calibrated !4C
Dates from Acropora palmate Framework and Intertidal Mangrove
Peat. Coral Reefs, 22:257-270.

Twilley, R. R. 1988. “Coupling of Mangroves to the Productivity of Es-
tuarine and Coastal Waters.” In Coastal Offshore Biosystem In-
teractions, ed. B. O. Jansson, pp. 155-180. Berlin: Springer-Verlag.

. 1995. “Properties of Mangrove Ecosystems in Relation to the En-
ergy Signature of Coastal Environments.” In Maximum Power, ed.
G. A. S. Hall, pp. 43-62. Boulder: University Press of Colorado.

Urish, D. W., R. M. Wright, and O. Viator. 2003. Topographic and Hy-
drographic Study in the Overwash Mangrove Islands of Twin Cays,
Belize. CCRE Report for 2002-2003. Washington, D.C.: Smithson-
ian Institution, National Museum of Natural History.

Watson, I., and A. D. Burnett. 1995. Hydrology: An Environmental Ap-
proach, pp. 460-465. Boca Raton, Fla.: CRC Press.

Wolanski, E., M. Jones, and J. S. Bunt. 1980. Hydrodynamics of a Tidal
Creek-Mangrove Swamp System. Australian Journal of Marine and
Freshwater Research, 31:431-450.

Wolanski, E., Y. Mazda, and P. Ridd. 1992. Mangrove Hydrodynamics.
Coastal and Estuarine Studies (Tropical Mangrove Ecosystems),
41:43-62.

Wolf, P. R., and C. D. Ghilani. 2006. Elementary Surveying: An Intro-
duction to Geomatics, pp. 72-144. Upper Saddle River, N. J.: Pear-
son Prentice and Hall.

Wright, R. W., D. W. Urish and I. Runge. 1991. The Hydrology of
a Caribbean Mangrove Island. In Coastlines of the Caribbean,
Coastal Zone 91 Conference-ASCE, Long Beach, CA July 1991,
ed. G. Cambers, pp. 171-184. New York: American Society of
Civil Engineers.

Ecological Characteristics
of Batis maritima in Florida and Belize

Dennis EF Whigham, Michael C. Whigham,
Ilka C. Feller, Wilfrid Rodriguez,
and Ryan S. King

Dennis EF. Whigham and Ilka C. Feller, Smithson-
ian Environmental Research Center, 647 Contees
Wharf Road, Edgewater, Maryland 21037, USA.
Michael C. Whigham, Antioch University New
England (now at University of New Hampshire,
Natural Resources and Environmental Science,
Durham, New Hampshire 03824, USA). Wilfrid
Rodriguez, University of Rhode Island, Kingston,
Rhode Island 02881, USA. Ryan S. King, Depart-
ment of Biology, Baylor University, Waco, Texas
76798, USA. Corresponding author: D. Whigham
(whighamd@si.edu). Manuscript received 15 Au-
gust 2008; accepted 20 April 2009.

ABSTRACT. Batis maritima, a low-growing perennial species with woody stems and
succulent leaves, occurs in mangroves and, to a lesser degree, in salt marshes in the
Neotropics. It spreads by clonal growth, occurs in a wide range of habitats, and at times
forms monotypic stands. Sites that are permanently flooded or are flooded regularly by
tides and salt pans are the only mangrove habitats in which B. maritima does not occur
or occurs as a few scattered plants. On mangrove-dominated islands in Belize, the cover-
age and height of B. maritima were highest in open habitats, including sites disturbed by
human activities. In a mangrove-dominated mosquito impoundment in Florida, B. mari-
tima occurred in all habitats sampled and, similar to observations in Belize, coverage and
height were greatest in the most open habitats. The abundance and, at times, dominance
of B. maritima suggests that it may play an important role in the dynamics of mangrove
ecosystems, especially in the recruitment and establishment of mangrove seedlings. Man-
grove seedlings and saplings were present in most of the plots that were sampled in Belize
and Florida, but there was no relationship between the percent cover of B. maritima and
the density of seedlings and saplings.

INTRODUCTION

Batis, the only genus in the family Bataceae, has two species. Batis maritima
L. occurs in the Neotropics in coastal salt marshes and mangroves from Georgia
and Brazil on the Atlantic coast and California to Peru on the Pacific Coast of
North and South America. The species is widely distributed in the Caribbean
basin. The second species, Batis agrillicola P. Royan, is endemic to coastal areas
of northern Australia.

Batis maritima, a low-growing C; perennial species with woody stems and
succulent leaves, is associated with saline soils and has been described as a spe-
cies that responds to disturbance in mangroves and salt marshes (Rey et al.,
1990; Pennings and Richards, 1998; Pennings and Callaway, 2000). An im-
portant ecophysiological characteristic of B. maritima is the ability to adjust
photosynthetic rates to increasing soil salinity by making adjustments to leaf sap
osmolalities (Liittge et al., 1989). The ability to propagate clonally (Pennings
and Callaway, 2000) is another characteristic that enables it to respond rapidly
to altered environmental conditions.

492 e

Despite its widespread distribution, there have been
relatively few ecological studies of B. maritima. In Georgia
salt marshes, Pennings and Richards (1998) found a posi-
tive relationship between the presence of wrack (accumu-
lated litter) and the abundance of B. maritima. Pennings
and Callaway (2000) found that clonal integration was an
important factor in the ability of the species to colonize
bare salt pans. The responses of B. maritima to altered
hydrological conditions appear to vary with differing en-
vironmental settings. In hypersaline coastal wetlands in
Texas, B. maritima cover expanded following inundation
with freshwater (Alexander and Dunton, 2002). Con-
versely, the opposite occurred in Baja California (México)
where its cover increased following the construction of a
dike that eliminated tidal flooding and increased soil sa-
linity (Ibarra-Obando and Poumian-Tapia, 1991). Batis
maritima cover was also dynamic in mangroves in Florida
that were impounded for mosquito control. The cover of
all herbaceous halophytic species, including B. maritima,
decreased following the construction of dikes and the sub-
sequent impoundment and flooding of mangroves in the
Indian River (Rey et al., 1990). Several years later, when
tidal exchange between the impoundment and estuary
was restored, B. maritima recolonized areas that were no
longer flooded continuously. Another important ecologi-
cal feature of B. maritima is its inability to tolerate pro-
longed periods of shade in mangrove-dominated wetlands
(Lopez-Portillo and Ezcurra, 1989). Along Florida’s Gulf
Coast, Milbrandt and Tinsley (2006) observed a greater
number of black mangrove (Avicennia germinans (L.)
Stearn) seedlings in existing B. maritima patches compared
to surrounding mudflats. They hypothesized that this im-
proved seedling success was the result of a slight increase
in elevation provided by the B. maritima root system. In
contrast, McKee et al. (2007) found that on offshore is-
lands in Belize B. maritima did not appear to have an ef-
fect on recruitment of red mangrove (Rhizophora mangle
L.) seedlings.

Other than the experimental research on coastal salt
marshes (Pennings and Richards, 1998; Pennings and Cal-
laway, 2000), little is known about the ecological role of
B. maritima in coastal wetlands, especially in mangroves
where it most frequently occurs. Is it a fugitive species
that only persists because it is capable of responding to
changing environmental conditions? Alternatively, is it an
important species in mangroves because of its impact on
patterns of nutrient cycling or its ability to influence the
establishment of mangrove trees (i.e., R. mangle, A. germi-
nans, Laguncularia racemosa (L.) Gaertn. f. [white man-
grove], Conocarpus erectus L. [buttonwood])? Although

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

B. maritima is a common component of mangrove forests
throughout the Neotropics, there is limited knowledge on
distribution patterns within the intertidal landscape or on
the ecological roles of this species across a range of man-
grove habitats.

Our objective was to describe the distribution of B.
maritima in Florida and Belize as part of our overall goal
to determine its ecological role in mangrove ecosystems.
Here we describe our initial efforts to characterize the
ecology of B. maritima at two of the Smithsonian’s long-
term research sites (Figure 1) that also represent the range
of conditions (subtropical and tropical) where this species
associates with mangroves. For Florida (subtropical), we
focus on B. maritima in four habitats in a mangrove-
dominated impoundment along the Indian River Lagoon
(IRL) that has a history of intervention for purposes of
mosquito control (Rey et al., 1990). For Belize (tropical),
we focus on B. maritima in disturbed and undisturbed
sites on offshore mangrove islands. For both sites, we also
present data on the relationships between percent cover of
B. maritima and the density of mangrove seedlings.

STUDY SITES
BELIZE

Twin Cays is the focus of our B. maritima studies
in Belize. Twin Cays (91.5 ha) is an archipelago of peat-
based mangrove islands (Figure 1) located near the crest
of the barrier reef of central Belize. These islands are lo-
cated approximately 17 km east of the mainland, and the
only source of freshwater is precipitation. Vegetation on
Twin Cays is dominated by the mangroves R. mangle, A.
germinans, and L. racemosa. The forest structure is het-
erogeneous and characterized by gradients in hydrology
and tree height that include a seaward fringe of R. mangle
around the periphery of the islands, along tidal creeks, and
in perennially flooded ponds (Feller et al., 1999). Avicen-
nia germinans and L. racemosa primarily occur in habitats
that are not water covered at low tide. Vegetation patterns
on Twin Cays are complex, and the dynamics have been
the focus of many studies (Feller, 1995; Feller and McKee,
1999; Rodriguez and Feller, 2004; Lovelock et al., 2006a).
However, none of the previous research has focused on
the distribution or ecology of B. maritima even though it
occurs in almost all habitats except those that do not expe-
rience prolonged flooding (D. Whigham, personal obser-
vation). Human activities have altered parts of Twin Cays
(Rodriguez and Feller, 2004; McKee et al., 2007), and
the primary anthropogenic activity has been the clearing

NUMBER 38 e¢ 493

Mangrove-dominated wetland (SLC 24)
Indian River Lagoon, Florida
IKONOS 2005 (UTM Zone 17N)

95°0'0"W = 90°0'0"WSs 85°0'0"W Ss 80°0'0"W_Ss 75°0'0'

25°0'0"

15°0'0"

95°0'0"W = 90°0'0"WSs 85°0'0"W_~—Ss 80°0'0"W

Twin Cays Archipelago
Belize, Central America
IKONOS 2003 (UTM Zone 16N)

Meters

FIGURE 1. Approximate locations of SLC24 and Twin Cays (inset map) and IKONOS images of the two study sites. For SLC 24, the white line
that is seen around the impoundment is a dike. Dark areas within the impoundment are dredged from adjacent subtidal habitats. Darker areas
on the two large islands are internal tidally influenced ponds that are most often shallowly water covered.

494 e

of mangroves with or without the addition of sediments
dredged from nearby subtidal habitats. In this study, we
compared the distribution of B. maritima in disturbed and
undisturbed mangrove habitats at Twin Cays (described in
further detail below).

FLORIDA

An impounded, mangrove-dominated wetland (SLC
24) in St. Lucie County in the IRL is the focus of the
Florida studies (see Figure 1). SLC 24 has been managed
in a variety of ways since it was diked in 1970. Rey et
al. (1990) describe management activities and patterns
of vegetation change in SLC 24 between 1970 and 1987.
SLC 24 was hydrologically isolated from the IRL by a dike
(Figure 1) until 1985 when a culvert was installed to re-
move excess water deposited during two tropical storms.
Once water levels were lowered, the culvert was sealed,
and the impoundment remained isolated until 1987 when
the culvert was reopened and other culverts were installed.
The cover of all vegetation decreased from 75% to near
30% following construction of the diked impoundment
in 1970. Over subsequent years, the cover of herbaceous
halophytes, including B. maritima, changed in response to
variations in the timing and duration of flooding and the
establishment and growth of mangroves. Rey et al. (1990)
concluded that a steady decline in the cover of herbaceous
halophytes after 1984 was primarily caused by shading
as the canopies of mangroves developed. Vegetation pat-
terns are also complex in the numerous impoundments
that have been established in the IRL, and they have been
the focus of several studies focused primarily on nutrient
limitation within mangroves (Feller et al., 2003; Lovelock
and Feller, 2003; Lovelock et al., 2006b). We sampled B.
maritima in three mangrove-dominated habitats and areas
associated with salt pans where dwarf A. germinans (sensu
Feller et al., 2003) occurs as scattered individuals or in
patches with almost continuous cover. Details of sampling
locations and methods are given below.

METHODS
BELIZE

We sampled B. maritima in two disturbed sites and six
undisturbed sites on Twin Cays. One disturbed site is a 2
ha area on West Island that was cleared of mangroves and
burned in 1991 and covered with material dredged from
the adjacent subtidal area in 1995 (Rodriguez and Feller,
2004; McKee et al., 2007). The other disturbed site was
clear cut in 2004, but no dredged material was added. In

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

both disturbed sites, we sampled B. maritima in 10 ran-
domly located plots (each 1 X 1 m) in which we made
visual estimates of its cover, measured its height at five
randomly chosen locations in each plot, and identified
and counted all mangrove seedlings and saplings. Seed-
lings of A. germinans and L. racemosa had cotyledons
present. Seedlings of R. mangle were individual, with no
more than one pair of true leaves. Saplings were defined
as individuals less than 50 cm in height with no cotyle-
dons present, or with more than one pair of true leaves in
the case of R. mangle.

For our undisturbed sites at Twin Cays, we sampled
B. maritima in three forested habitat types (Fringe, Transi-
tion, Interior), which were located at different distances
from the ecotone between the mangrove forest and open
water. Fringe habitats, which were dominated by trees 4 to
5 m tall, were at the outer boundary between mangroves
and open water, either along ponds located in the interior
of Twin Cays or along the ocean. Avicennia germinans was
the dominant tree in the three Fringe habitats adjacent to
interior ponds. Rhizophora mangle was the dominant tree
in the three Fringe habitats adjacent to the ocean. Transi-
tion and Interior habitats were all dominated by A. ger-
minans. Transition habitats were located approximately
15 m further into the mangrove forest from the Fringe
habitats, and Interior habitats were located approximately
an additional 15 m beyond the Transition habitats. We
sampled 5 randomly located plots (same procedures as de-
scribed above) in each of the 90 plots (5 plots X 3 habitat
types X 6 sites) in undisturbed mangrove.

FLORIDA

We sampled B. maritima in SLC 24 in four habitat
types (Fringe = R. mangle, Dense = A. germinans, Sparse
= A. germinans, Dwarf = A. germinans). The Fringe habi-
tats, dominated by R. mangle 4 to 6 m tall with scattered
A. germinans, were located at the boundary between man-
groves and open water. The two habitats dominated by
taller (3-6 m) A. germinans (Dense, Sparse) differed in the
size and spatial configuration of the dominant trees. The
Dense A. germinans habitat had trees that were mostly 4
to 6 m tall and formed a continuous canopy dominated
by A. germinans. The Sparse A. germinans habitat was
also dominated by A. germinans but the trees were usually
shorter (3-5 m) and were more widely spaced, resulting
in a more open canopy. The Dwarf A. germinans habitat
was always adjacent to salt pans that were mostly unveg-
etated or only had a few scattered dwarf trees (usually less
than 1 m tall). We sampled B. maritima in one randomly
located plot in each of the replicate sites for each habitat

type. In each 1 X 1 m plot, we made the same set of mea-
surements as described above for Belize.

DATA ANALYSIS

Because of the different sampling regimes, we made
separate statistical comparisons for the Belize and Florida
data sets. Based on initial screening of the data (Proc Uni-
variate; SAS Institute, 1990), we determined that none of
the data were normally distributed either in their original
form or any of the possible transformations. We used the
nonparametric PROC NPARIWAY (SAS Institute, 1990)
to make comparisons of B. maritima data (percent cover,
height) and the number of mangrove seedlings + saplings
for the different habitat types at both locations.

RESULTS
BELIZE

Percent cover of B. maritima differed (Figure 2a)
significantly (df = 4, chi-square for Kruskal-Wallis test
= 27.9272, P < 0.0001) among the sites on Twin Cays.
Mean percent cover ranged from 50% to 53% for the two
disturbed sites and the undisturbed Fringe habitat. Percent
cover decreased from the Fringe to the Transition ([mean
+ 1 SE] = 35.5% + 4.7%) and Interior (16.9% + 2.9%)
undisturbed sites. The average height of B. maritima also
differed significantly between sites (Figure 2b; df = 4, chi-
square for Kruskal-Wallis test = 29.0273, P < 0.0001).
Heights were similar at the two disturbed sites (24.4 + 2.3
cm = clear-cut + fill; 26.4 + 1.0 cm = clear-cut). At the
undisturbed sites, height was greatest at the Fringe habitat
(61.7 + 18.5 cm) and decreased toward the interior of the
mangrove forest (40.4 + 2.2 = Transition; 34.7 + 2.0 =
Interior).

The number of mangrove saplings + seedlings also
differed across sites (Figure 2c), and there were significant
differences for all three species and for the total of all spe-
cies (df = 4, chi-square for Kruskal-Wallis test = 38.9958,
IP & OMOOiIls VAISS, JP OWUS/S WSIS, 12 < OL 7zs2e
15.5953, P < 0.0036 for R. mangle, A. germinans, L. rac-
emosa, and total mangroves, respectively).

The total number of mangrove saplings + seedlings
was higher at the clear-cut and filled site (24.6 + 13.8 m’)
compared to the clear-cut site (2.5 + 0.8 m7) and undis-
turbed mangrove habitats (mean for all three undisturbed
sites was 7.3 + 1.2 m “*). Avicennia germinans saplings +
seedlings at the clear-cut and filled site were less than 1 m
(Figure 2c). Rhizophora mangle was the most abundant
species at the Fringe habitat, whereas A. germinans was

NUMBER 38 e¢ 495

100-

80>

Percent Cover

60>

405

Mean Height (cm)

20-

40

1a R. mangle
(al L. racemosa

oo
o
|

A. germinans
Total

Number of Seedlings
8
iT

=
o

Habitat

FIGURE 2. Cover (a), height (b), and sapling + seedling (c) data
for two disturbed and three undisturbed habitats on Twin Cays
(Belize). Values are means + 1 SE.

496 e

the most abundant species at the Transition and Interior
habitats.

FLORIDA

Percent cover (df = 3, chi-square for Kruskal-Wallis
test = 38.9252, P < 0.0001) and height (df = 3, chi-
square for Kruskal-Wallis test = 33.0923, P < 0.0001)
of B. maritima differed between the four habitat types in
SLC 24 (Figure 3a). There was no B. maritima in the plots
that were sampled in the Fringe R. mangle habitat, and
the cover (2.8% + 1.2%) was very low in the Dense A.
germinans habitat. Percent cover was 42.9 + 8.1 and 27.8
+ 4.6 in the Sparse and Dwarf A. germinans habitats, re-
spectively. Height differences (Figure 3b) among the four
habitats had the same pattern with the tallest plants occur-
ring in the Sparse A. germinans habitat (48.9 + 4.5) and
shortest in the Dense A. germinans habitat (13.1 + 5.4).
The total number of saplings + seedlings and the means
for each mangrove species also differed significantly (Fig-
ure 3c) among the four habitat types (df = 3, chi-square
for Kruskal-Wallis test = 11.5483, P < 0.0091; 12.7678,
P < 0.0052; 16.4377, P < 0.0009; 13.4660, P < 0.0037
for R. mangle, A. germinans, L. racemosa, and total man-
groves, respectively).

DISCUSSION

The objective of this initial investigation of Batis mari-
tima was to quantify aspects of its distribution in a va-
riety of habitats in mangroves at long-term Smithsonian
study sites in Belize and Florida. The impetus for the re-
search was the observation that B. maritima is widespread
in mangroves and, in some habitats, its high abundance
and cover suggest that it potentially plays an important
role in these systems. There have, however, been few stud-
ies that shed light on its possible ecological importance in
mangroves. Studies in salt marshes near its northern limit
found that it was not a dominant species and did not com-
pete well with other marsh plants (Zedler, 1977). There is
some suggestion that B. maritima may be a fugitive species
because it is common in disturbed sites (Milbrandt and
Tinsley, 2006). Pennings and Richards (1998), for exam-
ple, found that stands of B. maritima were associated with
areas that were disturbed by wracks of litter in a Georgia
salt marsh.

Batis maritima has been described as a species that
does not do well in shaded conditions or under conditions
of continuous flooding (Rey et al., 1990; Alexander and
Dunton, 2002). However, Keer and Zedler (2002) found

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

100

80

60

40

Percent Cover

100

©
(—)

fo2)
i=)

Mean Height (cm)
pS
(—)

N
(—)

a
(—)

| HR. mangle

| LJ L. racemosa

ow
i=)

| E.] A. germinans

Total

Number of Seedlings
NO
=)

=
So

Habitat

FIGURE 3. Cover (a), height (b), and sapling + seedling (c) data for
four habitats in mosquito impoundment SCL 24 in the Indian River
Lagoon, Florida. Values are means + 1 SE.

that it can tolerate prolonged flooding, and it does well in
waterlogged conditions when light is not limiting (Zedler,
1980). It also responds positively to increasing salinity
(Ibarra-Obando and Poumian-Tapia, 1991) but is elimi-
nated under hypersaline conditions (Zedler et al., 1986;
Dunton et al., 2001).

The only long-term study of B. maritima in chang-
ing environmental conditions occurred in one of our
study sites, SCL 24 in Florida. Rey et al. (1990) exam-
ined a sequence of aerial photographs taken over a period
of time when hydrological conditions varied from years
when there was continuously flooding, to years when the
impoundment was drained, and to years when there was
partial tidal exchange with the IRL. Vegetation almost
completely disappeared when the impoundment was con-
tinuously flooded. Once the impoundment was opened to
limited tidal exchange, herbaceous halophytes increased in
abundance and cover and B. maritima eventually became
the dominant species. Over time, mangroves recruited and
eventually dominated the vegetation in most parts of the
impoundment. As the abundance and size of mangroves
increased, B. maritima declined along with other herba-
ceous halophytes in response to increased shading by man-
groves (Rey et al., 1990).

Results of our surveys support several of the earlier
studies and suggest that light levels, regular tidal flood-
ing, and soil salinity are three important factors that de-
termine where B. maritima occurs and how abundant it is.
There have been at least two studies (Pennings and Rich-
ards, 1998; Milbrandt and Tinsley, 2006) suggesting that
B. maritima is a fugitive species that colonizes high-light
disturbed sites. In Belize, the highest percent cover was at
the two disturbed sites and the Fringe habitat in the undis-
turbed mangroves (see Figure 2a). Even though the mean
cover of B. maritima in the undisturbed sites on Twin Cays
was lower (34.3% + 3.1.9%) than the disturbed sites, the
highest B. maritima cover (80.3% + 2.6%) of any of the
habitats sampled was in the three Fringe habitats that
were associated with interior ponds. Edge habitats asso-
ciated with interior ponds are mostly in full sun and are
exposed to tidal flooding, but the flooding is rarely more
than a few centimeters deep (D. Whigham, personal obser-
vations). The substrates are almost always waterlogged,
and the sediments are soft, mostly composed of floc that
accumulates on the downwind side of the interior ponds.
The highest B. maritima cover in SLC 24 in Florida also
occurred at sites that had no overhead mangrove canopy
or only a discontinuous canopy.

The mean cover of B. maritima was least in the shaded
habitats in Florida and Belize, supporting the suggestions
of Lopez-Portillo and Ezcurra (1989) that low light levels

NUMBER 38 ¢ 497

can limit its abundance and distribution. The absence of
B. maritima in the sample plots at the Fringe habitat asso-
ciated with SLC 24 and the lower cover in the Fringe habi-
tats closest to the ocean on Twin Cays (20.9% + 2.9%)
also support the suggestions that regular inundation by
tidal flooding has a negative effect on the species (Alexan-
der and Dunton, 2002).

The mean height of the B. maritima canopy also var-
ied among habitats, and the patterns are most likely the re-
sult of variations in light and salinity (Zedler et al., 1986;
Dunton et al., 2001). In SLC 24, mean height decreased
from the more open Dwarf and Sparse A. germinans habi-
tats to the shadier Dense A. germinans habitat (Figure 3b).
At the undisturbed sites in Belize, mean height decreased
from the Fringe to the Interior, most likely in response to
decreasing light. The height of the plants was greatest in
the Fringe habitat associated with the Edge sites that were
closest to the Interior ponds. Mean height at the Fringe
habitats associated with the ponds was 85.4 + 27.1 cm
compared to 40.4 + 2.2 at the more shaded Fringe habitats
closest to the ocean. Taller average height associated with
the Edge habitats may be the result of higher phosphorus
concentrations in the sediments. In a separate fertilization
experiment, we found that B. maritima responded signifi-
cantly to the addition of phosphorus at all the undisturbed
sites on Twin Cays, but the smallest response was at the
Edge habitat associated with interior ponds, suggesting
that phosphorus was more available in those sediments
(D. Whigham, unpublished data). Compared to the Fringe
habitats on Twin Cays and the Sparse A. germinans habitat
in the SLC 24, mean height decreased toward the sites with
no mangrove canopy (Dwarf A. germinans habitat in SLC
24 and the two disturbed sites at Twin Cays). Lower mean
height at the open sites is likely the result of increased salin-
ity as the Dwarf A. germinans site in SLC 24 is hypersaline
(i.e., soil salinity as high as 100%; D. Whigham unpub-
lished data). In addition, soil salinity at the clear-cut and
filled site on Twin Cays, while variable during an annual
cycle, can be more than 60%o (McKee et al., 2007).

Mangrove seedlings are widely dispersed, and their oc-
currence varies spatially in response to light levels and their
ability to withstand flooding, salinity, and attacks from
herbivores (Ellison and Farnsworth, 1993; Olusegun and
Creese, 1997). If B. maritima facilitates the establishment
of mangrove seedlings, we would expect a positive rela-
tionship between percent cover and the number of seed-
lings + saplings for one or more of the mangrove species.
Milbrandt and Tinsley (2006) found that the presence of
B. maritima had a positive effect on the survival of A. ger-
minans seedlings. McKee et al. (2007), however, found that
B. maritima had no effect on the recruitment and survival

498 e

of mangrove seedlings, even though mangrove seedlings
benefited by the presence of other herbaceous species (i.e.,
Distichlis spicata, Sesuvium portulacastrum) at clear-cut
and filled sites sampled in this study. Although there were
habitat differences in the number of seedlings + saplings,
the presence of seedlings + saplings in 88% of the plots
sampled in Belize and 55% of the plots in Florida indi-
cated that mangrove establishment may have been facili-
tated by B. maritima. We found no relationship, however,
between the amount of B. maritima cover and the density
of seedlings + saplings for any of the mangrove species
(Figure 4). The potential for B. maritima to influence the
distribution and growth of mangroves trees and other
mangrove plants and animals remains unknown. But,
given the abundance of the species across a range of habi-

1000
| a @ R. mangle
() L. racemosa
a A A. germinans
£
e
4 100 a
a A
> A
3 A
o
7) 2 A e @ e
+ A C )
° AA
L C on )
o A ® e
A e
5, 0 : 0 0 A
5 oe, $8 a4, ae A e
2 @ () A
a @ _ @ A
@@ e ca ra
( ) ee A® 0 @
c a A AC rx ) @
@aM ii AAO A A@A @

Number of seedlings+saplings

@
a) o
a} oO
A A
A A
A ®
Oo A A
1 #8 A—_® e+
0 20 40 60 80 100

Percent Cover
SSS a a a a a
FIGURE 4. Number of mangrove saplings + seedlings (# m~) plot-
ted against percent cover of Batis maritima for 1 X 1 m plots sam-
pled in Belize (a) and Florida (b) study sites.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

tats, the potential seems high, especially in areas where it
is the dominant species.

In summary, B. maritima was widespread in most
mangrove habitats at both our study locations, and there
were significant interhabitat differences in all the variables
measured. Mangrove seedlings and saplings were com-
mon in areas occupied by B. maritima, but we found no
evidence that the establishment of mangroves benefited
by increasing cover of this common halophytic species.
The ubiquitous distribution of B. maritima at all the sites
sampled, however, indicates that its role in mangrove eco-
systems deserves further consideration.

ACKNOWLEDGMENTS

The research was supported by the Smithsonian Ma-
rine Science Network. We acknowledge logistical support
of the station staff at Carrie Bow Cay (Belize) and the
Smithsonian Marine Station at Ft. Pierce (Florida). Brian
Miller, project volunteer, assisted with fieldwork on Twin
Cays in 2004. This is contribution number 853 of the Ca-
ribbean Coral Reef Ecosystems Program (CCRE), Smith-
sonian Institution, supported in part by the Hunterdon
Oceanographic Research Fund.

LITERATURE CITED

Alexander, H. D., and K. H. Dunton. 2002. Freshwater Inundation Ef-
fects on Emergent Vegetation of a Hypersaline Salt Marsh. Estu-
artes, 25:142-1435.

Dunton, K. H., B. Hardegree, and T. E. Whitledge. 2001. Response of
Estuarine Marsh Vegetation to Interannual Variations in Precipita-
tion. Estuaries, 24:851-861.

Ellison, A. M., and E. J. Farnsworth. 1993. Seedling Establishment,
Growth, and Response to Disturbance in Belizean Mangal. Ameri-
can Journal of Botany, 80:1137-1145.

Feller, I. C. 1995. Effects of Nutrient Enrichment on Growth and Her-
bivory in Dwarf Red Mangrove (Rhizophora mangle). Ecological
Monographs, 65:477-S05.

Feller, I. C., and K. L. McKee. 1999. Light-Gap Creation in a Beliz-
ean Red-Mangrove Forest by a Wood-Boring Insect. Biotropica,
31:607-617.

Feller, I. C., D. F Whigham, K. L. McKee, and C. E. Lovelock. 2003.
Nitrogen Limitation of Growth and Nutrient Dynamics in a Dis-
turbed Mangrove Forest, Indian River Lagoon, Florida. Oecologia
(Berlin), 134:405-414.

Feller, I. C., D. E Whigham, J. P. O’Neill, and K. M. McKee. 1999. Ef-
fects of Nutrient Enrichment on Within-stand Nutrient Cycling in
Mangrove Ecosystems in Belize. Ecology, 80: 2193-2205.

Ibarra-Obando, S. E., and M. Poumian-Tapia. 1991. The Effect of Tidal
Exclusion on Salt Marsh Vegetation in Baja California, México.
Wetlands Ecology and Management, 1:131-148.

Keer, G. H., and J. B. Zedler. 2002. Salt Marsh Canopy Architecture
Differs with the Number and Composition of Species. Ecological
Applications, 12:456-473.

Lépez-Portillo, J., and E. Ezcurra. 1989. Zonation in Mangrove and Salt
Marsh Vegetation at Laguna de Mecoacan, México. Biotropica,
21:107-114.

Lovelock, C. E., and I. C. Feller. 2003. Photosynthetic Performance and
Resource Utilization of Two Mangrove Species Coexisting in Hy-
persaline Scrub Forest. Oecologia (Berlin), 134:455-462.

Lovelock, C. E., M. C. Ball, B. Choat, B. M. J. Engelbrecht, N. M. Hol-
brook, and I. C. Feller. 2006a. Linking Physiological Processes with
Mangrove Forest Structure: Phosphorus Deficiency Limits Canopy
Development, Hydraulic Conductivity and Photosynthetic Carbon
Gain in Dwarf Rhizophora mangle. Plant, Cell and Environment,
29:793-802.

Lovelock, C. E., I. C. Feller, M. C. Ball, B. M. J. Engelbrecht, and
M. L. Ewe. 2006b. Differences in Plant Function in Phosphorus
and Nitrogen Limited Mangrove Ecosystems. New Phytologist,
172:514-522.

Liittge, U., M. Popp, E. Medina, W. J. Cram, M. Diaz, H. Griffiths, H. S. J.
Lee, C. Schafer, J. A. C. Smith, and K. H. Stimmel. 1989. Ecophysi-
ology of Xerophytic and Halophytic Vegetation of a Coastal Allu-
vial Plain in Northern Venezuela. New Phytologist, 111:283-291.

McKee, K. L., J. E. Rooth, and I. C. Feller. 2007. Mangrove Recruitment
After Forest Disturbance is Facilitated by Herbaceous Species in the
Caribbean. Ecological Applications, 17:1678-1693.

Milbrandt, E. C., and M. N. Tinsley. 2006. The role of Batis maritima
in regeneration of degraded mangrove forests. Hydrobiologia,
568:369-377.

NUMBER 38 ¢ 499

Olusegun, O. O., and R. G. Creese. 1997. Population Structure, Spatial
Pattern and Seedlings Establishment of the Grey Mangrove, Avicen-
nia marina vat. australasica, in New Zealand. Australian Journal of
Botany, 45:707-725S.

Pennings, S. C., and R. M. Callaway. 2000. The Advantage of Clonal
Integration under Different Ecological Conditions: A Community-
Wide Test. Ecology, 81:709-716.

Penning, S. C., and C. L. Richards. 1998. Effects of Wrack Burial in
Salt-Stressed Habitats: Batis maritima in a Southwest Atlantic Salt
Marsh. Ecography, 21:630-638.

Rey, J. R., R. A. Crossman, and T. R. Kain. 1990. Vegetation Dynam-
ics in Impounded Marshes along the Indian River Lagoon, Florida,
USA. Environmental Management, 14:397-409.

Rodriguez, W., and I. C. Feller. 2004. Mangrove Landscape Character-
ization and Change in Twin Cays, Belize Using Aerial Photography
and IKONOS Satellite Data. Atoll Research Bulletin, 513:1-22.

SAS Institute. 1990. SAS Language: Reference. Version 6. First Edition.
Cary, N. C.: SAS Institute, Inc.

Zedler, J. B. 1977. Salt Marsh Community Structure in the Tijuana
Estuary California USA. Estuarine and Coastal Marine Sciences,
5:39-53.

. 1980. Algal Mat Productivity: Comparisons in a Salt Marsh.
Estuaries, 3:122-131.

Zedler, J. B., J. Covin, C. Nordby, P. Williams, and J. Boland. 1986.
Catastrophic Events Reveal the Dynamic Nature of Salt Marsh Veg-
etation in Southern California. Estuaries, 9:75-80.

at

¢
1
A
4 -

5 %
f) ‘3!

> ( BaD,

t ia

ey

ia

"Jes wid “iy

ao gn

Sponge Community Dynamics
on Caribbean Mangrove Roots:
Significance of Species Idiosyncrasies

Janie L. Wulff

Janie L. Wulff, Department of Biological Sci-
ence, Florida State University, Tallahassee, Florida
32306-4295, USA (wulff@bio.fsu.edu). Manuscript
received 29 August 2008; accepted 20 April 2009.

ABSTRACT. Descriptions of the rich sponge faunas inhabiting mangrove roots at various
Caribbean sites are unanimous in pointing out the heterogeneity of species distribution
and abundance patterns at all scales, from different portions of a single root to geographic
subregions. Abiotic factors have often been implicated by correlation, but ecological in-
teractions, and the life history and morphological characteristics of the sponge species,
may also play key roles. Published studies vary widely in methods used, hampering direct
comparisons of results, and raising the possibility that conclusions might be influenced
by methods. I have been exploring the processes underlying distribution and abundance
patterns by applying identical methods to studying community composition and dynam-
ics at two sites in Belize (Twin Cays) and one site in Panama (Bocas del Toro). Established
communities on roots have been fully censused, by volume and numbers of individuals,
yearly for three years (i.e., four censuses). Community composition, when evaluated in
terms of total volume of component species, is very similar at these three sites, although
abiotic factors differ and geographic distances between sites range from 330 m to 1,200
km. The nine species found on censused roots at all three sites constituted a total of 89%,
84%, and 73%, respectively, of the total sponge volume at these sites. In general, species
exhibited similar patterns of growth, size decrease, and mortality at all sites where they
were found, suggesting that these are species-level characteristics. Numbers of individu-
als and volume provide very different assessments of the relative importance of different
species in these communities. Community change over time appeared to be substantial,
when measured in terms of shifts in total numbers of individuals or total sponge volume.
However, taking into account dynamics of individual species provides a very different
view, as most large changes in numbers or volume were not community wide but tended
to reflect life history characteristics typical of early successional stage species or idiosyn-
cratic responses of one or a few species to particular environmental circumstances.

INTRODUCTION

Organisms that live in habitats consisting of discrete patches within an unin-
habitable matrix have fascinated biologists who are simultaneously attracted to
community ecology and to life history evolution. A rich set of conceptual frame-
works has developed to explain the dynamics of community assembly and de-
velopment within each patch in the context of interconnections among patches.
Theories of, for example, island biogeography (MacArthur and Wilson, 1967),
multiple stable points (Sutherland, 1974), competitive networks (Jackson and

502 e

Buss, 1975), the intermediate disturbance hypothesis (e.g.,
Connell, 1978), and meta-communities (e.g., Mouquet
and Loreau, 2002) have helped us understand community
dynamics in patchy habitats ranging from oceanic islands
and tropical mountaintops to badger mounds, holes in
mussel beds, and ponds.

Prop roots of Caribbean red mangrove (Rhizophora
mangle) trees are easily accessible, experimentally trac-
table, and extraordinarily colorful examples of inherently
patchy communities. The sessile inhabitants that cover the
root surfaces colonize in the form of water-borne propa-
gules, and most of them are thereafter confined to the root
on which they landed. Although post-recruitment interac-
tions with neighbors and consumers may have relatively
deterministic outcomes, a stochastic element, contributed
by the uncertainty that any particular species will land on
a particular root, is ever present. Throughout the wider
Caribbean region sponges are prominent members of the
prop root communities, and their abundance and diver-
sity of species, color, and forms have inspired time-series
monitoring and experimental manipulations as well as
comparative faunal studies (e.g., Sutherland, 1980; Elli-
son and Farnsworth, 1992; Bingham and Young, 1995;
Riutzler, 1995; Farnsworth and Ellison, 1996; Rutzler and
Feller, 1996; Ritzler et al., 2000; Diaz et al., 2004; Diaz,
2005; Wulff, 2000, 2004, 2005; Engel and Pawlik, 2005).
Specific conclusions relating to how dynamic these sponge
communities are, and what processes drive the dynamics
and influence distribution and abundance patterns, have
differed widely among studies, but a consistent theme is
that the resulting distribution and abundance patterns are
highly heterogeneous on scales ranging from within indi-
vidual roots to between geographic subregions.

Sutherland (1980) complemented repeated monitor-
ing of natural communities on prop roots with a study
of community development on flat settlement panels sus-
pended among the roots in Venezuela. He concluded that
these sponge-dominated communities are relatively stable
over time, and that high diversity could be maintained
by a trade-off between competitive ability and coloniza-
tion efficiency, combined with the continued addition of
fresh roots that provide refuges for inferior competitors.
Farnsworth and Ellison (1996) surveyed prop root com-
munities of mangroves in a variety of abiotic settings in
Belize, focusing on spatial scales of distribution patterns.
They were able to identify scales of heterogeneity that in-
cluded backs versus fronts of individual roots, leeward
versus windward shores, and coastal versus island man-
gal. At the 11 sites where they sampled twice, their data
corroborated Sutherland’s (1980) conclusions that com-

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

munity change is minimal. Bingham and Young (1995)
concluded very differently, from their work in the Florida
Keys, that dynamics of sponges on mangrove roots can
be extreme, influenced by perturbations from physical
disturbance, predators, and asexual recruitment. They at-
tributed differences in community dynamics between sites
in the Florida Keys and Venezuela to differences in sea-
sonality (subtropical versus tropical) and abiotic stressors,
and suggested that the differences between their study and
Sutherland’s (1980) study could be explained by equilib-
rium versus non-equilibrium situations, with the Venezu-
elan mangrove communities primarily structured by com-
petitive interactions.

Disentangling the effects of biogeography and differ-
ent suites of abiotic factors by making direct comparisons
among studies is hampered by the wide variety of ap-
proaches that have been applied. Published studies differ
with respect to units of study, time course and frequency
of monitoring, and metrics for evaluating abundance. To
control for technique, I used identical methods to evaluate
sponge community composition and dynamics for three
years On mangrove prop roots at three sites in Belize and
Panama. Following the fates of individual sponges was a
priority, because my chief interest was in how the morpho-
logical and life history strategies of the different sponge
species constrain or enhance their ability to coexist on the
prop roots. Rather than focusing on community-level met-
rics, such as species diversity or primary space occupancy,
I recorded survival and changes in volume of the same in-
dividuals over time and attempted to identify the causes of
size decrease, fragmentation, or mortality. Two sites near
each other in Belize differed in abiotic conditions, and a site
in Panama provided a geographic comparison. My goals
included (1) assessing the similarity of species composition
among sites differing in abiotic factors and geographic dis-
tance, (2) comparing community dynamics among sites,
with respect to both numbers of individuals and volume,
and (3) exploring the possibility that each mangrove sponge
species adheres to a characteristic approach for maintain-
ing its representation in this community, regardless of the
specific abiotic context and other species present.

METHODS

Three sites characterized by well-developed mangrove
prop root epiphytic communities were chosen for yearly
censuses. The three sites were chosen primarily because
experiments had been established at each several years be-
fore, and so regular visits were already required for moni-

toring. Top priorities in initial site choice had been easy
access and sufficient sponge individuals for experimental
manipulations; species composition was secondary. The
two Belize sites, both at Twin Cays, near the Smithson-
ian Institution’s Carrie Bow Cay research station (map
and further site descriptions in Rutzler et al., 2004; Diaz
et al., 2004), allow comparison of a main channel versus
a tidal creek near each other (330 m). The Panama site,
directly across the channel from the Smithsonian Tropi-
cal Research Institute marine laboratory on Isla Colon in
Bocas del Toro (map coordinates and description of the
overall area are found in Diaz, 2005), adds a geographic
comparison (1,200 km distant) between two main chan-
nel sites. The submerged portions of the prop roots (i.e.,
the portion on which sponges could grow) were from 24
to 143 cm long, with the majority between 40 and 80 cm
in length.

At each site, mangrove roots or root clusters were
chosen that appeared to be healthy (i.e., no signs of rot
or incipient breakage) and on which it was possible to
identify and measure all sponges on all sides of each root.
Root clusters were added to the initial census at each
site until species accumulation curves had leveled off for
sponges, and at least 163 sponge individuals (the num-
ber of sponges in the first census at the first site) were in-
cluded: a total of 10 clusters, 1 to 5 roots each (24 roots
initially) at Hidden Creek; 13 clusters, 1 to 4 roots each
(37 roots initially), at Sponge Haven; and 15 clusters, 1 to
3 roots each (42 roots initially), at the Bocas del Toro site.
Roots were labeled with small plastic tags, coded by color
and shape, on narrow (1 mm) beaded nylon cable ties. Full
censuses were made at approximately 1 year intervals, for
a total of 3 years (i.e., four censuses at each site, except for
Bocas del Toro, where the 2nd year census was skipped),
beginning in March 2004 at both Belize sites and in June
2003 at the Panama site. At each census, every root or
root cluster was drawn and root lengths measured. Every
sponge was drawn to scale, in place on the root draw-
ings, and sufficient dimensions measured to accurately
estimate volume by conglomerations of appropriate geo-
metric solids. In this way, every sponge could be followed
for survival, growth, decrease in size, and fragmentation.
New recruits were added to the root maps as they were
discovered (recruitment data will be reported in a sepa-
rate publication), and notes were made on interactions be-
tween neighboring sponges and other sessile organisms, as
well as damage caused by physical disturbance, predation,
and disease. Some roots at each site were lost by breakage
during the 3 years. To be able to interpret the time-series
data clearly, only roots for which at least some portion

NUMBER 38 °¢ 503

persisted throughout the study were included in the time-
series data analysis, and the only roots added to the study
were those that branched directly off subtidal portions of
the originally censused roots.

RESULTS

SPECIES COMPOSITION AND RELATIVE ABUNDANCE,
BY VOLUME AND NUMBER OF INDIVIDUALS

A total of 21 sponge species were represented by at
least 0.1% of the total sponge volume on censused roots at
one or more sites (Table 1). These species represent the de-
mosponge orders Poecilosclerida (8 species), Haplosclerida
(6 species), Halichondrida (4 species), and Dictyoceratida
(3 species), in a variety of colors, and with growth forms
ranging from thinly encrusting to irregularly branching to
clusters of volcanoes (Figure 1). Of these most abundant
21 species, 9 were found on censused roots at all three
sites, and another 6 were found on censused roots at two
of the three sites (Figure 2). Many of the sponge species
are relatively rare, and so were present at a site but not on
a censused root. Adding three cases in which species were
found at a second or third site on roots directly adjacent to
at least one censused root increases the number of species
shared by all three sites to 10, with an additional 7 spe-
cies shared by two of the three sites. Geographic distance
was not a strong predictor of the percent of species shared.
Sponge Haven and Hidden Creek, only 330 m apart, shared
74% (14/19) of their most common species, and Sponge
Haven and Bocas del Toro, 1,200 km apart, shared 60%
(12/20) of their most common species (comparisons not
significantly different by the G test: 0.1 < P < 0.5). The
Hidden Creek and Bocas sites, geographically distant from
each other and also differing in abiotic factors, shared 55%
(11/20) of their most common species (comparison with
the proportion of species shared by Hidden Creek—-Sponge
Haven by the G test: 0.05 < P < 0.1).

Census data from all years at each site were added
together for an average relative representation of species,
with respect to both volume and number of individu-
als (Figure 3). At all three sites the most abundant spe-
cies by volume, Tedania ignis, accounted for about half
(49%-57%) of the total sponge volume. The nine species
found on censused roots at all three sites contributed a
total of 89%, 84%, and 73% of the total volume at, re-
spectively, Hidden Creek (HC), Sponge Haven (SH), and
Bocas del Toro (BT). Similarity of species representation
at these sites is also borne out by Morisita’s index of com-
munity similarity (using volume as abundance measure),

504 e

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

TABLE 1. Sponge species on censused roots at Hidden Creek and Sponge Haven, both at Twin Cays, Belize; and at Isla Colon, Bocas
del Toro, Panama. A total of 21 sponge species were represented by at least 0.1% of the total sponge volume on censused roots at one
or more sites. Species that rank in the top half of the species on censused roots at a site, with respect to volume, are indicated by “XX”,
and those that rank in the bottom half are indicated by “X”. Species that occurred on censused roots at one site but were only seen on

“ ”

a root or roots directly adjacent to at least one censused root at another site are indicated by “x”. A dash (—) indicates species was not

found at a site.

Location
Sponge taxon Hidden Creek Sponge Haven Bocas del Toro
Order Dictyoceratida
Dysidea etheria de Laubenfels, 1936 x x x
Spongia tubulifera Lamarck, 1814, and S. obscura Hyatt, 1877 XX XX XX
Order Halichondrida
Amorphinopsis sp. XX - -
Halichondria magniconulosa Hechtel, 1965 XX XX XX
Halichondria sp. = XX -
Scopalina ruetzleri (Wiedenmayer, 1977) xX x =
Order Haplosclerida
Chalinula molitba (de Laubenfels, 1949) XX - Xx
Haliclona curacaoensis (van Soest, 1980) x Xx -
Haliclona implexiformis (Hechtel, 1965) XX xX x
Haliclona manglaris Alcolado, 1984 x xX x
Haliclona sp. a x x =
Haliclona sp. b - - XX
Order Poecilosclerida
Biemna caribea Pulitzer-Finali, 1986 XX Xx x
Clathria campecheae Hooper, 1996 Xx Xx =
Clathria schoenus (de Laubenfels, 1936) = = XxX
Clathria venosa (Alcolado, 1984) x xX x
Lissodendoryx isodictyalis (Carter, 1882) XX XX x
Mycale microsigmatosa Arndt, 1927 - XX XX
Tedania ignis (Duchassaing and Michelotti, 1864) XX XX xX
Tedania klausi Wulff, 2006 - XX XX

with similarities of HC-SH = 0.977, SH-BT = 0.971, and
HC-BT = 0.957. These index values are strongly influ-
enced by the similar dominance of T. ignis at all three sites,
but other species were also consistently either relatively
abundant or rare at all sites. When sponge species at each
site are divided into those that rank in the top half by vol-
ume versus those that rank in the bottom half, pairwise
comparisons between sites (see data in Table 1) yield 23
site pairs in which a species was ranked in either the top
half or bottom half at both sites and only 7 pairs in which
a species was ranked in the top half at one site and in the
bottom half at the other site (significantly different from
an even distribution by the G test at P < 0.005).

At each site species were present on censused roots
that were represented by volumes of less than 0.1% of
the total. Among these species were Scopalina ruetzleri
(Wiedenmayer, 1977) at Hidden Creek, Clathrina coriacea

(Montagu, 1818) at Hidden Creek and Sponge Haven, and
Mycale magnirhaphidifera (van Soest, 1984) at Sponge
Haven; and Hyrtios violaceus (Duchassaing and Miche-
lotti, 1864), Haliclona vansoesti (de Weerdt, de Kluijver,
and Gomez, 1999), Haliclona caerulea (Hechtel, 1965),
and Tethya actinia (de Laubenfels, 1950), all at Bocas del
Toro, as well as several as yet unidentified species.

In general, number of individuals and total volume
provide very different views of the relative importance
of the species in these communities (see Figure 3). This
discrepancy is strikingly illustrated by the high represen-
tation by numbers of individuals of Haliclona manglaris
(15.2%, 27.9%, and 50.1% at HC, SH, and BT, respec-
tively), which also consistently contributed minimal vol-
ume (0.06%, 0.08%, and 0.37%). By contrast, Tedania
ignis, which contributed half the volume at each site, con-
tributed only 8.4% to 20.4% of the individuals.

FIGURE 1. Photographs of some of the most common sponge species inhabiting mangrove prop roots at Hidden Creek and Sponge Haven,
Twin Cays, Belize, and across the channel from the STRI Bocas del Toro Marine Station, Isla Colon, Panama. Top row, from left to right:
Clathria venosa, Haliclona curacaoensis, Haliclona manglaris (turquoise) and Haliclona sp. b (purple), Dysidea etheria (ethereal blue). Second
row: Chalinula molitba, Mycale microsigmatosa, Biemna caribea, Tedania ignis (three individuals). Third row: Lissodendoryx isodictyalis,
Tedania klausi, Haliclona sp. a. Bottom row, from left: Halichondria magniconulosa, Haliclona implexiformis (purple). Authors of species are
given in Table 1.

506 °

Sponge species on
mangrove prop roots
Sponge

330 m Haven
1200 km
Hidden
Creek
Bocas
1200 km del Toro

FIGURE 2. Diagram showing sponge species shared among two sites
in Belize (Hidden Creek and Sponge Haven) and one site in Panama
(Bocas del Toro). Only the 21 species that each constituted at least
0.1% of the total sponge volume on censused roots at a minimum of
one of the three sites are included.

Mean density of numbers of sponge individuals per
unit length of subtidal prop root was similar for all three
sites (15, 11, and 15 individuals per meter length of root
for Hidden Creek, Sponge Haven, and Bocas del Toro, re-
spectively). Reflecting a relative preponderance of small
individuals at the Bocas del Toro site, sponge density mea-
sured as volume per unit length of subtidal prop root was
only 5.7 cm?/cm at Bocas, compared with 20.8 cm?/cm
at Hidden Creek and 15.7 cm?/cm at Sponge Haven. Al-
though variation in root diameter renders root length im-
precise as a measure of substratum area monitored, length
was deemed a better measure than number of roots be-
cause of the sixfold variation in root lengths (i.e., from 24
to 143 cm).

SPONGE COMMUNITY DYNAMICS COMPARED WITHIN AND
BETWEEN SITES, BY VOLUME AND NUMBER OF INDIVIDUALS

During the three years of monitoring, the largest dif-
ference between the highest and lowest total sponge vol-
ume was 12%, 35%, and 27% at, respectively, Hidden
Creek, Sponge Haven, and Bocas del Toro; and the larg-
est difference between the highest and lowest number of
sponge individuals was 50%, 34%, and 39%, respectively
(Figure 4). Based on these total abundance values, com-

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

munity-wide change appears to be substantial. However,
comparison over time of abundance of individual species,
with respect to both volume and numbers of sponge in-
dividuals, sheds light on the components of change and
provides a very different picture. In many cases, large
overall changes in total volume or numbers in the course
of a particular year reflect changes in just one or a few
species. For example, the drop in total sponge volume be-
tween 2005 and 2006 at Sponge Haven (see Figure 4) was
mostly caused by losses from Halichondria sp., Halichon-
dria magniconulosa (almost to the point of elimination),

® other spp
= Hal mangl
= Scop ruetz |
Hal spa
= Hal sp b
Amorph
= Chal mol
Dys ether
@ Ted klausi
| & Myc micr
| @ Clath schoen
© Halich sp
_ Biem car
@ Hal cur
| @ Halich magn |
= Lis isod prp
Lis isod
| Spong spp
© Hal imp!
= Ted ign

Species composition by volume

100%
80% 5

60% |

Total volume

40%

20% +

0%
Bocas del

Hidden
Creek Haven Toro

Sponge

Species composition by # individuals

§ other spp
© Hal mangl
= Scop ruetz
Hal sp a
@ Hal sp b
Amorph
| @ Chal mol
Dys ether
| @ Ted klausi
@ Myc micr
= Clath schoen
| © Halich sp
_ Biem car
® Hal cur
| | @ Halich magn
= Lis isod prp
Lis isod
m™ Spong spp
= Hal imp!
® Ted ign

Individual sponges

Sponge Bocas del
Haven Toro

FIGURE 3. Sponge species assemblage composition, with respect to
total volume contributed by each species, and with respect to the to-
tal number of individuals of each species, on mangrove prop roots at
three Caribbean sites. These average relative abundances were calcu-
lated by adding together the volume or numbers of individuals for all
four yearly census dates (three dates in the case of Bocas del Toro).
See Table 1 for complete spelling of species names.

NUMBER 38 °

Hidden Creek census - Volume Mather spp Hidden Creek census - Individuals Mother spp
Hal mangl | Hal mang!
20000 Scop ruetz 180 ae Ei) Sw ae | |i Scop ruetz
(1 Clath ven 160 | |) Clath ven
_ |Clath camp 140 || Clath camp |
E 15000 \DHal n sp o ‘DHalspa
a “J Amorph G 120 | Amorph |
€ || Chal mol 2 100 ||BChal mol |
3 Heo || A Dys eth S 80 || Dys eth |
2 || Biem car § ||Biem car |
8 Hal cur + 60 | | Wi Hal cur |
fe 5000 @Halich magn 40 - GiHalich magn |
|| Lis isod prp 20 - || Lis isod prp |
|| Lis isod | J Lis isod
0 | Spong spp 0 ! | B Spong spp
March0O4 JuneO5 JuneO06 MarchO7 |GHal impl March04 JuneOS June06 MarchO7 |GHal impi
t=0 t=15 t=27 t=36 [BTedign | t=0 t=15 t=27 t=36 (|BTedign
Sponge Haven census - Volume Sponge Haven census - Individuals ees
|i other spp 200 - |B other spp
Hal mangl || Hal mangl
| Clath ven 180 || Clath ven
1) Clath camp 160 | Clath camp
2 |i Ted klausi 3 140 | BiTed Klausi
cS) |i Myc micr = 120 | a Myc micr
E OHalich sp > 100 - | |Halich sp
re) OBiem car 3 ||) Biem car
iz. Bal cur i | Hal cur
8 | Halich magn + 60 || Halich magn
Fe Lis isod prp 40 | ia Lis isod prp
| Lis isod 20 || Lis isod
(3 Spong spp 0 + ! | |iSpong spp
| i | Hal impl
March0O4 JuneO5 June06 March07 = March0O4 June05 June06 MarchO7 a ee ae
[0 (eI5 fey GG oe t=0 t=15 t=27.—s«t=36 oO _d
Bocas census - Volume (Mother spp mecca a
12000 z 240 epee
bapa | Hal mang!
| ve \Clath ven |
ee 10000 GHal sp b 200 [Mal spb |
Oo (|@chal mol 4 160 |iChal mol |
i 8000 ee ae | 3 | Dys ether |
= Mi Ted klausi | 3 \iTed klausi_ |
5 > | |
3 6000 Myc micr | = 120 \|iMyc micr |
Ee Ne @iClath schoen| =) 50 Wi Clath schoen|
8 Fe [0 ae ‘OBiem car
Oo [Bihalich magn | | Halich magn |
a ue ElU))
2000 OLis isod | 40 la Lis isod |
| Hal impl
June03 May04 June06 Ted ign | June03 May04 Juned6 |i Ted ign |
t=0 t=11 t=36 t=0 t=11 t=36 SOONG:

507

FIGURE 4. Community dynamics for sponges on mangrove prop roots at three Caribbean sites. Relative abundance of the species is represented
by both total volume and total numbers of individuals. See Table 1 for complete spelling of species names.

508 e

and T. ignis. For the second and third species these losses
primarily consisted of size decreases and fragmentations
of sponges that were subsequently able to regenerate; and
thus for both of these species, abrupt and dramatic changes
in volume between the second and third censuses are not
reflected in tandem changes in numbers of individuals
(e.g., H. magniconulosa was represented by 12 individuals
of total volume 5,518 cm? in 2005, 14 individuals of total
volume 187 cm? in 2006, and 9 individuals of total vol-
ume 225 cm? in 2007). Large differences in maximum size
achieved by sponges of different species further promote
asynchronous changes in overall volume and numbers of
individuals. For example, during this same year in which
total sponge volume at Sponge Haven decreased by 35%,
the number of individual sponges there increased by 13%,
largely the result of a doubling of the number of Haliclona
manglaris individuals. Yet each H. manglaris individual is
so small that, even in the aggregate, they scarcely register
in the overall volume tally (0.2% for the June 2006 cen-
sus; see Figure 4).

Similarly, progressive loss of individuals of Biemna
caribea and Haliclona curacaoensis at Hidden Creek re-
sulted in decreases in total numbers of individuals by more
than half in the course of three years (Figure 5). If these
species are removed from the “Hidden Creek census — In-
dividuals” graph in Figure 4 (along with the very small
bodied H. manglaris), the community can be seen to oth-
erwise remain very similar throughout the three years with
respect to relative representation of the component species
by numbers of individuals. During this same time period,
the total volume of all sponges at this site remained very
similar, although there were large volume changes for indi-
vidual species (see Figure 4). The Sponge Haven data show
the same pattern of progressive loss of H. curacaoensis
(see Figure 5) and also B. caribea, although the latter spe-
cies was not as abundant to begin with at this site.

Not all changes in abundance of particular species
were abrupt or negative. Volume of Spongia spp. steadily
increased at all sites (see Figure 5), with little increase in
numbers, reflecting high survival of the individuals that
were present at the first census. Illustrating a third pattern
of dynamics, the volume of T. ignis fluctuated at all three
sites, but at the end of the three years the total volume of
this species at each site was similar to what it was at the
start of the study (Figure 5).

Portions of many roots were lost during the three
years of the study, but new roots sprouting from subtidal
portions of censused roots nearly balanced the losses dur-
ing some time periods. Thus the total length of prop roots

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Spongia spp

6000 =——$ $$$ ___—_——
fsa)
£
- HC
cd)
= SH
= “BT
>
t=0 t=1 yr t=2 yr t=3 yr
Census year
Haliclona curacaoensis
800 SSS SE <=
isa)
£
Oo
o HC
E OSH
$
t=0 t=1 yr t=2 yr t=3 yr
Census year
Tedania ignis
18000 ————_—_—_—_—_—____—_——
15000

12000

Volume, cm?
(ep) ito)
(=) (=)
(=) (en)
(o) [@)

3000

t=0 t=1 yr

t=2 yr
Census year

t=3 yr

FIGURE 5. Representative population dynamics graphs for three
sponge species inhabiting mangrove prop roots at sites where yearly
censuses were made (HC = Hidden Creek; SH = Sponge Haven;
BC = Bocas del Toro). Total volume of Spongia spp. consistently
increased between monitoring periods at all three sites; volume of
Haliclona curacaoensis decreased between monitoring periods at
both sites in which it was found; Tedania ignis total volume fluctu-
ated over time but ended up very similar to what it had been at all
three sites at the start of the study three years earlier.

included in the census was very similar for the first three
censuses at Hidden Creek, and for the first two censuses
at Bocas del Toro and Sponge Haven, and then, after a
decrease, also for the final two censuses at Sponge Haven.
Total length (in cm) for the four censuses at Hidden Creek
was 968, 896, 995, and 562; for Sponge Haven, 1,847,
1,787, 1,162, and 1,179; and for Bocas del Toro, 1,483,
1,583, and (after two years) 1,203. Substratum available
was not necessarily related to sponge abundance with
respect to either numbers of individuals or total volume
(compare abundance measures reported in Figure 4 with
total root lengths censused); for example, sponge volume
and total root length were inversely associated over the
three years at the Bocas del Toro site.

VARIATION AMONG SPONGE SPECIES
IN INDIVIDUAL PERSISTENCE

Because individual sponges were mapped and mea-
sured, their fates from one census to the next could be
recorded as (a) increased in size, (b) fragmented, (c) de-
creased in size, or (d) disappeared. To characterize each
species at each site independently of environmental cir-
cumstances during a particular time interval, data from all
1 year intervals between censuses (and one 2 year interval
in the case of the Bocas del Toro site) were added together
in Figure 6. Three patterns emerge from these graphs. First,
fragmentation and size decrease are important aspects of
persistence for many of these species. The only species rep-
resented entirely by individuals that increased in size or
vanished (i.e., none decreased in size or fragmented) be-
tween censuses were Spongia spp. and Amorphinopsis sp.
Second, at each site variation among species in the degree
to which individuals persisted was clear. Yearly rates of
loss ranged from 0% (e.g., Spongia spp.) to 100% (e.g.,
Chalinula molitba). Third, many species exhibited the
same characteristics at each site where they occurred. For
example, a set of species characterized by at least 40% of
the individuals increasing in size from one yearly census to
the next were evident at each site: Tedania ignis, Haliclona
implexiformis, Spongia spp., Lissodendoryx isodictyalis,
and Halichondria magniconulosa. The only exceptions
were Haliclona implexiformis in Bocas del Toro and Hali-
chondria magniconulosa at Sponge Haven. The reason for
the H. implexiformis difference at the Bocas site was not
obvious, but individuals of this species were always very
small there. At Sponge Haven, both H. magniconulosa
and T. ignis suffered high rates of size decrease and frag-
mentation between the second and third censuses. These

NUMBER 38 °¢ 509

Hidden Creek

Miincreased || fragmented || decreased || disappeared

100% ->-7-
80% +
oO
3
DZ 60%
2
EZ 40%
Le |
= 20%
0% :
oe 40 S Qe US SD oO UA
se co es a) Res ae cy & ak “
RRO ESE ONO ON
gS s x
‘+
Sponge Haven
a
©
5
3S
2
aS)
S
&
3d
we s
OMEN
4e ye
Bocas del Toro
increased || fragmented |_| decreased {| disappeared
100%
YL 80%
©
3
S 60%
2
E 40%
= 20%
0%
\ 6 AY aes 1A \
: Q Oe 2 2 Xo \)
x OY KP ae eS e
C
R XS Ks ACES) w
RE aE eee Ee ON es SS ee

FIGURE 6. Proportions of sponge individuals of each species that in-
creased in volume, fragmented, decreased in volume, or disappeared
entirely between yearly census dates at three Caribbean mangrove
sites. Although the fate of each individual was recorded in a single
category at the time of observation, the distinction between “frag-
mented” and “decreased” can be fuzzy, as it depends on whether or
not fragments generated survive until the time of observation. See
Table 1 for complete spelling of species names.

510 e

departures from their usual pattern (i.e., most individuals
increasing in size during a year) coincided with the un-
usual and fleeting presence of a couple of French angelfish
just before the third census. French angelfish have been
demonstrated to readily consume these two species (Wulff,
2005) and Halichondria sp. (in preparation), which also
decreased in volume during the same time interval. Spe-
cies that only occurred at two sites also exhibited similar
characteristics with respect to persistence at both sites. For
example, no Chalinula molitba individuals increased in size
from year to year at either Hidden Creek or Bocas del Toro,
and Haliclona curacaoensis, H. manglaris, and Biemna ca-
ribea had consistently low survival at each site. In contrast,
Tedania klausi survived well at Sponge Haven, but suffered
an extreme decline on censused roots at the Bocas del Toro
site, coinciding with observed high losses of only this spe-
cies throughout the site to disease (Wulff, 2006).

DISCUSSION
COMMUNITY COMPOSITION

These three sites that differ in abiotic factors and in
geographic distance from each other have very similar
species compositions, not only with respect to species
present but also with respect to their relative abundance.
The 9 species that are shared by all three sites constituted
89%, 84%, and 73% of the total sponge volume at, re-
spectively, Hidden Creek, Sponge Haven, and Bocas del
Toro. The ubiquity and consistent local dominance of the
fire sponge Tedania ignis contributes heavily to similar-
ity among these sites. Previous studies also concur that T.
ignis is a signature species for this ecosystem throughout
the wider Caribbean region, and it was the sole species,
of 23, that was recorded in all eight prop root faunal sur-
veys compiled by Wulff (2000) and in four of five of the
studies compared by Diaz et al. (2004). Not only does it
occur at most sites, it tends to be among the most abun-
dant species by any metric. At five sites in Belize (includ-
ing Hidden Creek and Sponge Haven), Diaz et al. (2004)
recorded T. ignis as present on 11% to 34% of the roots,
in the top 3 species ranked by frequency of occurrence.
By measuring area covered from photographs, Bingham
and Young (1995) estimated that 16.7% of the root area
at their Florida Keys site was covered by T. ignis. Using
line transects, along which the length of root covered by
each sponge was measured, Sutherland (1980) estimated
5%-12% coverage by T. ignis in Bahia de Buche, Venezu-
ela. Evaluating abundance by volume boosts the propor-
tional representation of this species because of its massive

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

growth form, and thus in this study T. ignis constituted
from 49% to 57% of the total sponge volume on censused
roots. This species not only holds a large proportion of
the primary substratum space, but it also participates in a
mutualism with the mangroves, enhancing the persistence
and health of the entire ecosystem by protecting the roots
from attacks by boring isopods (Ellison and Farnsworth,
1990; Ellison et al., 1996).

Tedania ignis is not the only species that is both nearly
ubiquitous and locally abundant, although it stands out as
the most extreme. Consistently Lissodendoryx isodictya-
lis, Halichondria magniconulosa, Spongia spp., Haliclona
implexiformis, Haliclona manglaris, and Dysidea etheria
appear on faunal lists and, where authors indicate relative
abundance, by whatever metric, they rank highly (Wulff,
2000; Diaz et al., 2004).

Although the three sites in this study are similar with
respect to these widespread typical mangrove root sponge
species, there are two types of differences among the sites:
(1) a few species that are abundant at one site but do not
occur elsewhere (e.g., Amorphinopsis sp. at Hidden Creek),
and (2) many rare species that appear to differ among sites.
The virtual lack of overlap of these rarer species on spe-
cies lists from different sites does not necessarily indicate
constrained distribution but may simply reflect their rare-
ness. Diaz et al. (2004) discuss this sampling issue with
the highly diverse Caribbean mangrove root sponge fauna
and illustrate it well with their data. Diaz et al. (2004)
also point out the great degree to which community com-
position can vary along a particular mangrove fringe. The
three sites in the present study are known to share addi-
tional sponge species if entire contiguous stretches of man-
grove are included (Ritzler et al., 2000). For example, at
Sponge Haven, Clathria schoenus is not found near the
censused roots but appears on roots at this site that are
further toward the mouth of the main channel.

Sponge species composition differences among sites
characterized by very different abiotic circumstances have
been well documented, and some sites are sufficiently ex-
treme in abiotic factors that sponges are scarcely present
(Farnsworth and Ellison, 1996) or succumbed to unfavor-
able conditions while being studied (Pawlik et al., 2007).
At least some of the differences in species composition
between Hidden Creek and Sponge Haven, only 330 m
apart, have already been ascribed to less hospitable abiotic
factors in the narrow, tidal Hidden Creek. Transplants of
5 species that are conspicuous at Sponge Haven thrived
initially in Hidden Creek, but nearly all (61/63) died over
the course of one year (Wulff, 2004), possibly implicating
episodically wide fluctuations in temperature and salinity.

The similarity between the three sites in this study is espe-
cially interesting, considering that they were chosen for ac-
cessibility and overall sponge abundance, rather than for
species composition, and that they have demonstrated dif-
ferences in abiotic factors and span a geographic distance
of 1,200 km.

METHODS FOR STUDYING SPONGES ON MANGROVE
Prop Roots CAN INFLUENCE EVALUATIONS
OF COMMUNITY SIMILARITY BETWEEN SITES
AND COMMUNITY STABILITY OVER TIME

Methods of studying composition and dynamics of
sponge communities on mangrove roots have varied with
respect to metrics for evaluating abundance, sampling unit,
choice of which units to sample, time interval of sampling,
and materials, size, and shape of recruitment surfaces.
This variety reflects the many different questions posed by
researchers, and the difficulty of quantifying sponges; but
methods may also influence conclusions.

Methods for evaluating abundance have included
analysis of photographs for percent cover, line transects
down roots with distance covered by each species recorded,
point counts through acetate sheets, and presence/absence
on each root, as well as the total numbers of individuals
and volume of each individual. The advantages and dis-
advantages of evaluating sponge abundance with respect
to volume, area covered, or numbers of individuals have
been previously compared in the context of coral reefs
(Ritzler, 1978; Wulff, 2000, 2001). Choice of metric is
influenced by expediency in the field, and also by whether
functional roles, life histories, species diversity, or some
other aspect of these communities is the central focus of
a study. One advantage to measuring sponge abundance
by volume is that growth rates can then be calculated if
the same sponge individuals are followed over time. As
well, functional roles related to trophic interactions, such
as filtering food particles from the water column and pro-
vision of food to spongivores, probably scale with volume.
Unfortunately, sponge volume is time consuming to mea-
sure nondestructively in the field, decreasing the number
of individuals that can be monitored.

Area can be a problematic measure of sponge abun-
dance, as the amount of sponge tissue under a particular
point can range over orders of magnitude. At these three
sites, for example, sponges on prop roots varied in thick-
ness 150 fold, from 0.1 to 15 cm. Evaluating mangrove
sponge abundance in terms of area is further complicated
by the prevalence of epizooism, which results in points
falling simultaneously over more than one sponge species.

NUMBER 38 e¢ 511

At least one functional role of sponges in mangroves may
be related to substratum area covered: protection of man-
grove roots from boring isopods (Ellison and Farnsworth,
1920992)

Numbers of individuals are difficult to interpret in the
contexts of sponge population dynamics and functional
roles, as numbers can increase either by recruitment or
fragmentation, and individual size can vary over many
orders of magnitude. The lack of concordance between
population dynamics of individual species measured
in terms of numbers of sponges versus total volume on
the same roots (see Figure 4) underscores how divergent
conclusions can be when different metrics are chosen for
evaluating sponge abundance. Evaluating abundance us-
ing two or more metrics at the same site can strengthen
understanding of processes underlying the dynamics. For
example, data indicating a small increase in numbers of
individuals of Halichondria magniconulosa at Sponge Ha-
ven allowed the coincident large decrease in volume to be
interpreted as extensive partial mortality and some frag-
mentation, rather than heavy losses of individuals.

An abundance measure that lends itself well to bio-
diversity surveys in this inherently fragmented habitat is
presence/absence of a species on each root. Diaz et al.
(2004) evaluated relative abundance of sponge species at
Hidden Creek and Sponge Haven by prevalence on roots.
Specific ranks of the species by prevalence were differ-
ent from ranks assigned by volume in this study, but the
match between the 10 most abundant species with respect
to percent of prop roots inhabited (Diaz et al. 2004) and
the 10 most abundant species with respect to volume (this
study) is 80% at Hidden Creek and 60% at Sponge Ha-
ven. Resolution of systematic challenges may increase the
match; for example, a second species of Tedania was only
formally identified (Wulff, 2006) at Sponge Haven after
the study by Diaz et al. (2004) was published.

Evaluating abundance by presence/absence can also
address an important community assembly issue: the
probability that the community on a root will include a
particular species. Sutherland (1980) pointed out the great
importance of habitat division into small discrete patches
by explicitly comparing the course of community develop-
ment on prop roots versus on the 20 x 122 cm asbestos
panels he deployed for evaluating recruitment. The larger
area of the panels increased the probability that the com-
petitively dominant, but inefficiently recruiting, Tedania
ignis recruited onto every physically separated substratum
patch. Once settled on a panel, this species was able to
continue its growth in every direction, and each panel be-
came quickly dominated by it. Each root had much less

512 e

surface area, providing a smaller target for settling larvae
of competitively dominant species. As predicted, if the
roots are therefore more reliable refuges for competitively
inferior species, the species composition on the roots was
far more heterogeneous (Sutherland, 1980).

For ranking species by relative abundance, the great-
est discrepancies between abundance measures (i.e., vol-
ume, area, number of individuals, and percent of roots)
emerge when applied to thinly encrusting species, as their
volume can be trivial even when they cover large areas
(e.g., see Wulff, 2001, for an explicit comparison in a coral
reef sponge community). The possibility that encrusting
species may be relatively ephemeral because they are eas-
ily overgrown is supported by a comparison between the
pattern of recruitment of the thinly encrusting species
Clathria campecheae onto initially bare polyvinyl! chloride
(PVC) pipes at Hidden Creek (Wulff, 2004) and its abun-
dance in the established community on prop roots. This
species was described from coral reefs and had not been
reported from mangroves, and yet it distinguished itself by
occurring on more pipes (7/8) than any other species at 20
months after they were suspended among the mangrove
roots. Once the possibility of its occurrence on mangrove
roots was raised, it was discovered at a very low level on
prop roots at Hidden Creek and Sponge Haven.

This finding raises the question of how the succes-
sional stage of communities on censused roots might
influence the evaluation of similarity of assemblages be-
tween sites and over time. Sutherland (1980) labeled 116
roots that had not yet entered the water, in addition to
260 roots that had already been colonized below the wa-
ter surface. Sponge species that specialize on colonizing
fresh roots would have therefore been included in his as-
sessment of the total fauna. Because I followed roots with
already established sponge faunas, and only added new
roots that sprouted from subtidal portions of previously
included roots (i.e., new roots that could be colonized by
sponge growth from already censused portions), the earli-
est successional stages were not included in my assessment
of community dynamics. Clathria campecheae, mentioned
above, was not the only species that was disproportion-
ately well represented on PVC pipes deployed for recruit-
ment at Hidden Creek 20 months earlier. Haliclona cura-
caoensis, Biemna caribea, and Haliclona manglaris were
also conspicuous with respect to numbers of individuals,
percent of pipes colonized, and (for H. curacaoensis and
B. caribea) volume, in this relatively early stage of com-
munity development on initially bare pipes (Wulff, 2004:
fig. 3). The pattern of loss of these species from one census

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

to the next (see Figures 4—6) is consistent with the possibil-
ity that these are early succession species that are progres-
sively lost from roots as sponge species that are superior
competitors accumulate over time. These data support
Sutherland’s (1980) suggestion that the mangrove root in-
habitants illustrate a trade-off between colonization rate
and ability to persist in the community, and raise the pos-
sibility that stability of these communities, if measured as
change over time, will depend on the successional stage
on the monitored roots. The earlier in succession the as-
semblage on a root is, the more likely that subsequent cen-
suses will reveal changes in species composition. Apparent
instability will be further magnified if percent cover is the
metric chosen for abundance, as thinly encrusting species
that are efficient recruiters, but may be eliminated as supe-
rior competitors recruit, will initially have very high abun-
dances with respect to area covered.

Observational units in previous studies have ranged
from camera framer-length segments of roots to root clus-
ters. Bingham and Young (1995) monitored 21 cm long
root segments at 1 and 2 month intervals. Their analysis re-
vealed how changes in abundance appear at different mon-
itoring intervals, providing insight into the complex and
rapidly changing dynamics of these communities at par-
ticular locations on roots. Their spatial position-focused
analysis is complementary to the individual organism-
focused analysis in the present study. Because the position
of sponge individuals can shift along the prop roots as
they increase and decrease in size, it is possible for them
to move into a particular root segment without a recruit-
ment event and to move out of a root segment while still
persisting on the root. Thus a sponge assemblage within a
root segment may appear less stable than the assemblage
on that entire root. Differences in conclusions of Suther-
land and Bingham and Young were attributed by the latter
authors to greater influence of physical disturbance and
seasonality on a subtropical site (Florida Keys) relative to
a tropical site (Venezuela), but it is possible that difference
in choice of observational unit might have also influenced
evaluations of stability.

The balance between numbers of individuals moni-
tored, frequency of monitoring, and method of evaluating
abundance must be struck with the ultimate aim in mind.
Following individual sponges over time and evaluating
their size with respect to volume were essential to the aims
of this study, which were to understand the life history
and morphological strategies employed by each species.
Inevitably the number of individuals and roots that could
be followed in such detail suffered, as did the frequency of

monitoring. Some compensation for these failings is made
by the detail of the time-series drawings of entire roots.
Detailed maps of the location of each sponge and com-
ments about its shape and size at each census allowed frag-
mentation, size decrease, and addition of new recruits to
be unambiguously distinguished, even when the causes of
size change were not obvious. It is likely that new sponges
recruited and vanished, and resident sponges changed in
size in multiple ways, during the year-long intervals be-
tween censuses, and so my data only indicate the net result
of months of unmonitored dynamics.

SIGNIFICANCE OF ECOLOGICAL
CHARACTERISTICS OF SPONGE SPECIES

Proportions of sponges that increased, decreased,
fragmented, or disappeared were similar for given species
among sites, suggesting that these may be species-level
characteristics. With a few exceptions, the set of species
that reliably exhibited 40% to 100% of individuals in-
creased between censuses were the same at all three sites
(Tedania ignis, Haliclona implexiformis, Lissodendoryx
isodictyalis, Spongia spp., Halichondria magniconulosa),
and constituted large proportions of the total sponge vol-
ume (85%, 82%, and 72% at HC, SH, and BT, respec-
tively) at each site. Numbers of individuals of these species
found on eight initially bare PVC pipes 20 months after
they were suspended among prop roots at Hidden Creek,
ranged from 0 (T. ignis, H. implexiformis, and Spongia
spp.) to 7 (L. isodictyalis) (Wulff, 2004). By contrast, the
set of species for which only 0% to 30% of the individuals
increased in size between censuses (i.e., B. caribea, H. cu-
racaoensis, H. manglaris, Clathria campecheae) were each
represented on the recruitment pipes by 11 to 14 individu-
als (Wulff, 2006). These patterns hint at the possibility
of integrated sets of ecological characteristics that help to
maintain all these species in the mangrove prop root sys-
tem. Population dynamics of at least some of the typical
mangrove root sponge species may be tied to their each be-
ing most suited to a particular time period in community
development.

Overall community change, measured by total biomass,
species diversity, numbers of individuals, and space occu-
pied, can be functionally of great importance on an ecosys-
tem level. However, an exclusive focus on these community-
level metrics can obscure the components of community
change—that is, changes in the component species—and
therefore hamper our understanding of underlying pro-
cesses. Consideration of the characteristics of individual

NUMBER 38 ¢ 513

species, such as their probability of persisting from year to
year; their efficiency at recruiting; susceptibility to particu-
lar biotic mortality sources such as predators, competitors,
or pathogens; and the frequency with which they fragment
or suffer partial mortality, may explain much of the com-
munity dynamics. Combining these new data on persistence
with previously reported recruitment data (Wulff, 2004) in-
dicates that some of the heterogeneity in space and time
among mangrove prop root communities may be the re-
sult of the community on each root progressing indepen-
dently through a successional sequence that is mediated,
at least in part, by an inverse relationship between ability
to hold space on mangrove roots and recruitment into the
community that was first suggested by Sutherland (1980).
Adding to these life cycle-mediated patterns the observed
idiosyncratic responses of particular species at a particular
site, such as Tedania klausi succumbing to disease at the Bo-
cas del Toro site or Halichondria magniconulosa targeted
for consumption by a pair of French angelfish at Sponge
Haven, allows community dynamics to be understood as
the result of a complex set of interactions among individual
sponges representing species that are characterized by spe-
cific physiological tolerances and morphological and life
history traits.

ACKNOWLEDGMENTS

Iam grateful to the Caribbean Coral Reef Ecosystems
Program (CCRE), of the National Museum of Natural
History (NMNH), Smithsonian Institution, for the privi-
lege of being able to do fieldwork at the Carrie Bow Cay
field station in Belize; and to Mike Carpenter, Klaus Ru-
etzler, and the volunteer station managers who make the
Carrie Bow station an unparalleled resource, supporting
comprehensive research on tropical marine ecosystems.
I also thank Gabriel Jacome for always facilitating my
fieldwork at the Bocas del Toro station of the Smithso-
nian Tropical Research Institute. | thank two reviewers
for helpful comments, and Carla Piantoni (CCRE) and
Molly K. Ryan (Department of Invertebrate Zoology,
NMNBH) for help with the figures. My field research was
supported by the National Science Foundation under
Grant No. 0550599, and by the Marine Science Network
of the Smithsonian Institution, supported in part by the
Hunterdon Oceanographic Research Endowment. This
is contribution number 854 of the Caribbean Coral Reef
Ecosystems Program (CCRE), Smithsonian Institution,
supported in part by the Hunterdon Oceanographic Re-
search Fund.

514 e

LITERATURE CITED

Bingham, B. L., and C. M. Young. 1995. Stochastic Events and Dynam-
ics of a Mangrove Root Epifaunal Community. P.S.Z.N.I.: Marine
Ecology, 16:145-163.

Connell, J. H. 1978. Diversity in Tropical Rain Forests and Coral Reefs.
Science, 199:1302-1310.

Diaz, M. C. 2005. Common Sponges from Shallow Marine Habitats
from Bocas del Toro Region, Panama. Caribbean Journal of Sci-
ence, 41:465-475.

Diaz, M. C., K. P. Smith, and K. Rutzler. 2004. Sponge Species Richness
and Abundance as Indicators of Mangrove Epibenthic Community
Health. Atoll Research Bulletin, 518:1-11.

Ellison, A. M., and E. J. Farnsworth. 1990. The Ecology of Belizean
Mangrove-Root Fouling Communities. I. Epibenthic Fauna Are
Barriers to Isopod Attack of Red-Mangroves. Journal of Expert-
mental Marine Biology and Ecology, 142:91-104.

. 1992. The Ecology of Belizean Mangrove-Root Fouling Com-
munities: Patterns of Epibiont Distribution and Abundance, and
Effects on Root Growth. Hydrobiologia, 247:87-98.

Ellison, A. M., E. J. Farnsworth, and R. R. Twilley. 1996. Facultative
Mutualism Between Red Mangroves and Root-Fouling Sponges in
Belizean Mangal. Ecology, 77:2431-2444.

Engel, S., and J. R. Pawlik. 2005. Interactions among Florida Sponges. II.
Mangrove Habitats. Marine Ecology Progress Series, 303:145-152.

Farnsworth, E. J., and A. M. Ellison. 1996. Scale-Dependent Spatial and
Temporal Variability in Biogeography of Mangrove Root Epibiont
Communities. Ecological Monographs, 66:45-66.

Jackson, J.B. C., and L. W. Buss. 1975. Allelopathy and Spatial Competition
among Coral Reef Invertebrates. Proceedings of the National Acad-
emy of Sciences of the United States of America, 72:5160-5163.

MacArthur, R. H., and E. O. Wilson. 1967. The Theory of Island Bio-
geography. Princeton, N. J.: Princeton University Press.

Mouquet, M., and N. Loreau. 2002. Coexistence in Metacommunities: The
Regional Similarity Hypothesis. American Naturalist, 159:420-426.

SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Pawlik, J. R., S. E. McMurray, and T. P. Henkel. 2007. Abiotic Factors
Control Sponge Ecology in Florida Mangroves. Marine Ecology
Progress Series, 339:93-98.

Rutzler, K. 1978. “Sponges in Coral Reefs.” In Coral Reefs: Research
Methods. Monographs on Oceanographic Methodology, Volume
5, ed. D. R. Stoddart and R. E. Johannes, pp. 299-313. UNESCO:
Paris.

. 1995. Low-Tide Exposure of Sponges in a Caribbean Mangrove
Community. P.S.Z.N.I.: Marine Ecology, 16:165-179.

Rutzler, K., M. C. Diaz, R. W. M. van Soest, S. Zea, K. P. Smith, B. Al-
varez, and J. Wulff. 2000. Diversity of Sponge Fauna in Mangrove
Ponds, Pelican Cays, Belize. Atoll Research Bulletin, 476:230-248.

Riutzler, K., and I. C. Feller. 1996. Caribbean Mangrove Swamps. Scien-
tific American, 274(3):94-99.

Ritzler, K., I. Goodbody, M. C. Diaz, I. C. Feller, and I. G. Macintyre.
2004. The Aquatic Environment of Twin Cays, Belize. Atoll Re-
search Bulletin, 512:1-49.

Sutherland, J. P. 1974. Multiple Stable Points in Natural Communities.
American Naturalist, 108:859-873.

. 1980. Dynamics of the Epibenthic Community on Roots of the
Mangrove Rhizophora mangle, at Bahia de Buche, Venezuela. Ma-
rine Biology, 58:75-84.

Wulff, J. L. 2000. Sponge Predators May Determine Differences in
Sponge Fauna Between Two Sets of Mangrove Cays, Belize Barrier
Reef. Atoll Research Bulletin, 477:250-263.

. 2001. Assessing and Monitoring Coral Reef Sponges: Why and

How? Bulletin of Marine Science, 69:831-846.

. 2004. Sponges on Mangrove Roots, Twin Cays, Belize: Early

Stages of Community Assembly. Atoll Research Bulletin, 519:1-10.

. 2005. Trade-Offs in Resistance to Competitors and Predators,

and Their Effects on the Diversity of Tropical Marine Sponges.

Journal of Animal Ecology, 74:313-321.

. 2006. Sponge Systematics by Starfish: Predators Distinguish

Cryptic Sympatric Species of Caribbean Fire Sponges, Tedania ignis

and Tedania klausi n. sp. (Demospongiae, Poecilosclerida). Biologi-

cal Bulletin, 211:83-94.

Index

Page numbers for tables are in italics and for figures are in bold.

absorption spectra
photosynthetically usable radiation and, 364 (see also photosynthetically usable
radiation)
seawater and, 363
Acanthais brevidentata, 189-191, 191, 192, 194
Acartia tonsa, 372. See also copepods
Achatina fulica, 108. See also snails
Acropora, 50-51, 142, 260-262, 265-266, 282-283, 314-215, 317, 318, 319, 408,
440-441, 449
aspera, 440
cervicornis (staghorn coral), 266, 282, 314, 315, 317, 318, 319, 444
latistella, 441
longicyathus, 408, 440
millepora, 445, 446
muricata, 441
palifera, 445
palmata, 51, 260-263, 262, 265-266, 314, 439, 443, 448
tenuis, 442, 445, 449
valida, 441
See also coral
active migration model, 277
adventure tourism, 243
Aetea sica, 232, 233, 235, 236, 237. See also bryozoans
Aeverrillia armata, 233, 234, 236. See also bryozoans
Agaricia, 60, 63, 314-315, 317, 318, 319, 451
agaricites, 447, 449, 451
growth of, 450
humilis, 445
lamarcki, 447
tenuifolia, 283, 314-315, 318, 451
undata, 315
See also sponge(s)

516 e* SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Ageratum littorale, 420
Aimulosia
pusilla, 238
uvulifera, 237-238
See also bryozoans
Aiptasia pallida, 165. See also cnidarian worms; anenomes
algae
benthic, 405
blooms, 4, 16, 32, 35, 63, 66, 170, 405, 409-410
brown, 285, 446-447
chemical defense in, 53
coralline, 402, 403, 407, 408, 409, 444-445, 450, 451
coral mortality and, 447
drift, 234, 236
endosymbionts, 429
fleshy, 403, 404, 407-409, 448
green, 31
growth of, 403
indicator species of, 407-409

macro-, 48, 154-155, 164, 167-170, 285, 288, 314-315,

402-405, 407-410, 440, 446, 448, 451
micro-, 281, 288
red, 445
turf, 405, 409-410, 446-447
allelopathy, 447-448
Amathia
alternata, 232, 233, 235
distans, 232, 233, 235-236
vidovici, 232, 233, 234, 235, 236-237, 237
See also bryozoans
American Samoa, 405
Amorphinopsis, 160, 167, 504, 509, 510
atlantica, 167
See also sponge(s)
Amphibalanus
amphitrite 293
eburneus, 293
improvisus barnacle, 293
reticulatus, 293
See also barnacles
Amphimedon erina, 167. See also sponge(s)
amphipods, 275
Anachis fluctuata, 191, 193. See also snails
anchialine cave(s)
adaptation to, 270, 277-278
animals, 274, 276, 278-279
biological significance of, 270
characteristics, 269
chemical characteristics of, 273
ecology of, 270, 272-273
geology of, 271-272
lava tubes and, 271, 275, 277
trophic relationships in, 272
See also caves
ANCOVA, 191-193, 262-265, 327-328, 329, 330

anemones, 96-98, 101, 106, 144, 152, 155, 164-167, 234,
255, 430
annelids, 33, 210, 273
Annual Symposium on Sea Turtle Biology and Conservation,
243. See also marine turtle(s)
ANOVA, 327, 328, 330, 348, 349, 350, 371-372, 374, 395,
399, 462
anoxia, 272, 403
Anthozoa, 140, 142, 143.
Antilles. See Netherlands Antilles
Apionsoma, 215-216. See also sipunculans
Aratus, 66. See also crabs
ARLEQUIN, 222
ascidians, 54, 60, 62, 159, 164, 234, 284, 443
Astyanax fasciatus, 279. See also fish
atmosphere
change, 4
CO» fin, 12, 323
enrichment studies of, 399
Australia, 57, 105, 272-273, 275, 279, 292, 440, 442, 446,
482, 487, 491
Avicennia germinans, 347, 348, 420, 422-424, 474, 492,
494-497. See also mangrove forest(s)
Axianassa intermedia, 186
Axiopsis serratifrons, 182, 185, 186. See also crabs

Bahalana caicosana, 274. See also isopods
Bahamas, 152
archipelago of, 276
Blue Holes of, 269, 278
coral reefs in, 324
Bahia Almirante, 324-325, 331-333, 361
Balanus
improvisus, 253
trigonus, 293
See also barnacles
ballast water, 18, 238, 295
Barbados, 405
barnacles, 249-250, 251, 254, 255, 293, 293, 296
barrier reef. See coral reef(s)
Batis
agrillicola, 491
distribution, 491
maritima, 420, 491, 492, 496, 498
BayesTraits, 98
Bay Islands of Honduras, 416, 424
Beania
hirtissima, 232, 233, 234, 235
klugei, 232, 233, 235-237
behavior
caves and, 277
intertidal animals and, 81
Belize

barrier reef of, 6, 45, 47, 51, 53, 111, 283, 287, 288, 316,

380, 383, 387, 389, 407, 416, 426, 448, 474

blue holes in, 269
central province of, 381
coastal, 46, 389
coral bleaching in, 405, 434
development in, 287, 288
fisheries in, 287, 289
government, 7, 62, 283
hydrogeomorphic characteristics of, 347
mangrove forest in, 153, 347, 348, 474
microtides of, 382
oceanographic conditions, 153, 156
PAR, 365, 366
seagrass in, 365
seasonal rainfall of, 385
sediment loads and, 389
sponges in, 151-152, 159, 166, 170, 506
study sites in, 361
surface water absorption spectra of, 364
tourism in, 287
urban and economic growth in, 379
water clarity in, 363
Belize Coastal Zone Management and Fisheries, 62
Belize Lagoon, 283
Belize National Meteorological Service, 381-382, 385
Bermuda, 60, 135, 136, 238, 272, 274-276, 278-279,
405
Biemna
caribea, 160, 167, 504, 505, 508, 510, 512-513
maritima 492, 495, 497
See also sponge(s)
Biffarius fragilis, 186
Biflustra denticulata, 233, 234, 235
biodiversity, 1, 2, 4, 6-8, 18, 25, 28, 30, 32-33, 35-36, 53, 59,
74, 87, 230, 241, 287, 452
anchialine cave, 273
in Belize, 283
education programs in, 87
intertidal and subtidal ecosystems, 415
mangroves and, 152, 169, 345
marine, 38, 79, 85, 426
reef, 53, 380, 407
surveys, 511
biofilms, 424, 445
biological invasions, 291. See also invasive species
black mangrove, 36, 58, 64, 347, 474, 486, 492. See also
mangrove forest(s)
black sea urchin (Diadema antillarum), 78
bleached coral
abundance of, 432, 434, 435
bleached cytoplasm of, 431
environmental stressors and, 435
mechanisms, 79, 82, 88, 260, 314-315, 319, 389, 430-435,
440, 447, 451
rate, 432
seasonal, 434

NUMBER 38 e¢ 517

symbiotic, 435
symptoms of, 435
See also coral
blue crabs, 17, 18. See also crabs
Bocas del Toro, Panama, 75, 151, 154, 170, 347-348
CARICOMP and 88
research station (BRS), 74, 76, 79, 324, 325, 326
Bochusacea, 275
Bonaire, 450
Bostrycapulus calyptraeformis, 190, 191, 191, 192, 193, 193.
See also snails
Bostrychietum, 58. See also algae
Bowerbankia, 233, 234, 235-237
gracilis, 235, 236, 236-237
maxima, 234
See also bryozoans
box jellyfishes, 145, 147
Brasiliensis, 420
Brazil, 34, 123, 136
Briareum, 142.
broadcast spawners, 440-441
bryozoans
branching, 234
colonies, 230
communities, 230
distribution, 233, 235-236, 237
Buddenbrockia, 140. See also cnidarian worms
Bugula
neritina, 233, 234, 235, 236, 237, 253
stolonifera, 233, 235, 236, 237
turrita, 254
See also bryozoans

Callianidea laevicauda, 186
Callinectes, 182, 183, 184, 185. See also crabs
Callyspongia, 67, 191, 193
Canary Islands, 271, 273, 275-277, 279
carbon dioxide enrichment studies, 392-393, 398
carbon production, 17
Caribbean
characteristics, 7, 74
corals in, 57, 59-60, 62-63, 66, 314 (see also coral)
epifaunal taxa from, 152 (see also epifauna)
mangroves of (see mangrove forest(s))
tourism, 303, 304, 305
Caribbean Coastal Marine Productivity Program
(CARICOMP)
Bocas del Toro research station and, 88
Carrie Bow Cay and 380
common methodologies of, 59
seagrass sites, 324, 324, 325-327, 329-331, 381
Caribbean Coral Reef Ecosystems program (CCRE), 2, 6-7, 43,
152, 379
Carijoa riisei, 231-232. See also coral

518 ¢ SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Carrie Bow Cay, Belize
bleached coral in, 434

Caribbean coral reef ecosystems program and, 6-7, 57 (see
also Caribbean Coral Reef Ecosystems program)

environmental monitoring system of, 380-382
facilities, 47, 64, 65
fieldwork in, 181-184
habitats, 50, 55, 187, 387, 387
history, 44-45
infrastructure, 59
location, 3, 381
Marine Field Station, 63, 64, 66
meteorological conditions, 381, 383
oceanographic conditions in, 380, 433, 433, 435
physical description of, 380
reef fishes of, 111
seasonal rainfall of, 385
water clarity, 383, 389, 433
water quality of, 380, 385, 389, 433
water temperature and tidal elevation, 388
wind conditions, 385
Carybdea, 143.
Cassiopea, 143.
Caulerpa verticillata, 156. See also sponge(s)
Caulibugula armata, 232, 233, 236-237, 237, 238. See also
bryozoans
cave fish Typhliasina pearsei, 274
caves
adaptations to, 277-278
anchialine (see anchialine cave(s))
in Bermuda, 278
limestone, 272
marine, 269
Yucatan, 269, 271, 272, 275, 276
Celleporaria, 235, 237-238
sherryae, 238
See also bryozoans
Celleporina hassalli, 232, 233
Census of Marine Life (COML), 85
Chalinula molitba, 155, 157, 161, 164, 165, 168, 504, 505,
509-510. See also sponge(s)
Charybdis hellerii 182, 183, 184. See also crabs
Chelonibia patula, 108. See also barnacles

Chesapeake Bay, 2, 4, 6, 12, 13, 16, 16-18, 19-20, 33, 66, 248,

250, 360, 369, 370, 371, 375, 392. See also Rhode
River, Chesapeake Bay

Chironex, 143.

chironomid insect larvae, 53

chlorophyll a, 53, 323, 325-333, 335, 337-338, 339, 340, 342,

360, 364, 371, 372, 372-373, 406, 409
Chondrilla, 60, 68, 68, 159, 167, 315, 405
caribensis, 167
aff. nucula, 315
See also sponge(s)

Chrysaora, 143
quinquecirrha, 370-372, 372, 374-376
See also jellyfish
ciguatera fish poisoning (CFP), 301, 302, 306, 307
ciguatoxins, 301
citrate synthase (CS), 278
Cladonema, 143.
Clathria
campecheae, 504, 512-513
microchela, 160, 167
schoenus, 157, 160, 163, 164, 165, 166, 167, 168, 169,
504, 510
venosa, 160, 179, 504, 505
See also sponge(s)
Clathrina
coriacea, 504
primigenia, 167
venosa, 167, 505
See also sponge(s)
Clavelina, 60
puertosecensis, 159, 284
See also ascidians
climate change. See global warming
Cliona, 67. See also sponge(s)
closed-circuit rebreather, 271
cnidarian worms
current classification of, 146
diversity, 139
evolutionary relationships in, 140, 141, 143-146
subclades of, 140
See also coral; specific species
coastal area(s), 11-12, 19, 30, 356, 359
erosion of, 45, 51, 63, 477
mangrove forests and, 474 (see also mangrove forest(s))
overfishing and, 16
pollution and, 16
tectonic faults in, 272
Coastal Ecosystem Protection Committee, 243
Codium, 232. See also algae
COI
analysis, 119
sequence, 113, 123, 129, 132, 221
Coiba National Park, 83, 87-88
Colombia, 53, 87-88, 152, 174, 176
colonial ciliates, 67
Colpophyllia natans, 318. See also coral
competition-based relative dominance model, 403
Comprehensive Everglades Restoration Plan (CERP), 35
confocal laser scanning microscopy (CLSM), 28, 31
Conocarpus erectus, 420, 492
Conopeum tenuissimum, 234, 236, 294. See also bryozoans
conservation
of cave species, 278-279
coastal, 81

marine, 4, 25, 86-88, 260

sea turtle, 8, 242-244
Consortium for the Barcode of Life (CBOL), 85
Convention on Biological Diversity (CBD), 243
Convention on International Trade in Endangered Species of

Wild Fauna and Flora (CITES), 243
Convention on the Conservation of Migratory Species of Wild
Animals (CMS), 243

Coolia, 286, 302, 303. See also dinoflagellates
copepods, 53, 271, 273, 275-276
coral

bleaching (see bleached coral)

broadcast spawning by, 440-441

brooding, 445

clonal growth of, 438, 447, 450-452

Diploria, 67

diseases, 405, 408

egg/sperm, 261

elkhorn, 50, 51, 260, 261, 266, 314

fecundity of, 440

fertilization ecology of, 36, 348, 439, 440

fire, 50, 51

gametogenesis in, 33, 432, 440

growth, 443

hard, 313

larvae, 35, 260-261, 263, 266, 408, 438, 442-443, 444,

445-447, 451

metamorphosis, 445

mortality, 263, 313-314, 443, 449, 451

pathogens, 448

planulae, 409, 442, 445, 451

Porites, 67

post-settlement biology of, 445-451

recruitment, 438, 440-441, 451

reefs (see coral reef(s))

reproduction, 438, 439

self-fertilizing, 314

settlement, 438, 443-447, 450

skeletons, 315, 317

soft, 231-232

staghorn, 266, 282, 314-315, 317, 318, 319, 444

white-band disease (WBD) of, 314-315
Coralaxius nodulosus, 185, 186
Corallianassa longiventris, 184-185. See also crabs
coral reef(s)

assessment protocols for, 409

biodiversity, 451 (see also biodiversity)

challenges to, 437

characteristics of, 6, 44, 259

cores, 318

cyanobacteria and 446

disturbances of, 380

ecosystems, 44, 313, 402, 409

eutrophication and, 404

NUMBER 38 e¢ 519

Florida, 35
irradiance, 382, 383, 386, 387
management, 403-407, 410
nutrient concentrations in, 402
recovery, 35
reproduction, 439
rhomboid shoals and, 314-315, 319, 319
sponges in, 159, 511, 512
structure, 44
upwelling and, 336
water quality and, 324
Coral Reef Development in the Tropical Eastern Pacific (TEP),
81-82
Coryphopterus, 116, 121
alloides, 112-113, 125, 130, 138
bol, 112-113, 115, 123
dicrus, 125, 126, 127, 130
eidolon, 112-113, 126, 128, 129-130, 137
glaucofraenum, 112-113, 115, 119, 120, 121, 125, 126,
129-130, 132, 135
hyalinus, 112-113, 115, 121, 129, 134
kuna, 112-113, 125, 129-130, 131, 138
lipernes, 112-113, 115, 117, 130, 134
personatus, 112-113, 115, 117, 121, 129, 134
punctipectophorus, 125, 129, 130, 131
thrix, 125, 126, 127, 130
tortugae, 112-113, 115, 118, 119, 121, 123, 125, 126,
129-130, 132, 135
venezuelae, 112-113, 115, 119, 121, 123, 123, 124, 125,
129-130, 132-133, 136
western Atlantic, 114, 119, 125, 129, 133
western Caribbean, 112
See also fish
Costa Rica, 83, 87-88
crabs
abundance of, 466
fisheries, 17, 18
intertidal (see Uca)
nursery habitats for, 16
nutrient recycling by, 459
swimming, 182, 183, 183
Crangon crangon, 96. See also crabs
Crepidula cf. nivea, 191, 191, 192, 193, 194, 190
Crisia, 232, 233, 234, 235
elongata, 232, 233, 234
See also bryozoans
Cronius
ruber, 184
tumidulus, 184
See also crabs
crustaceans, 273, 275-276. See also crabs; isopods; shrimps
Cryptocorynetes longulus, 274. See also remipedes
Cryptosula pallasiana, 233, 234, 235, 237
Cuba, 152, 159

520 e* SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Cubozoa, 145, 147.

Culebra Island Education Center, 86

Curacao, 113, 119, 123, 125-126, 128-129, 449

cyanobacteria, 35, 152, 404-405, 408-409
endosymbionts, 429
mats, 57, 154, 155

Cycloporus variegatus, 31

Cyperus spp, 420

cytochrome c oxidase, 113, 116, 134

Danielopolina, 276. See also ostracods
data loggers, 343, 381
decapod crustaceans, 95, 275-276. See also crabs; shrimps
Deeveya, 276. See also ostracods
deforestation, 325
denaturing gradient gel electrophoresis (DGGE), 174-175, 175,
176
Dendronephthya, 142.
Diadema antillarum, 78-79, 385, 406, 449. See also fish
Dichothrix, 448. See also cyanobacteria
Dictyoceratida, 503, 504
Dictyosphaeria cavernosa, 31
Dictyota
caribaea, 284, 448
pulchella, 447
See also algae
dinoflagellates, 5-6, 33, 63, 283, 286-288, 301-302, 302, 303,
304, 429-430, 431, 435
Diploria, 448
strigosa, 266, 446
See also coral
Diplosoma listerianum, 253, 443. See also ascidians
dispersal kernels, 442-443
dissolved oxygen, 4, 16-17, 269-270, 272-273, 278-279, 326,
327, 337, 339, 369, 371, 372, 375, 382
Distichlis spicata, 392, 396, 420, 498
DNA
analysis, 34
barcoding, 34, 111, 112, 132
extraction, 220
mitochondrial, 176, 184, 220-221, 223, 224, 225, 226,
Dy,
Dominican Republic, 83
dredging operations, 284, 286
dwarf red mangrove, 474, 477, 486, 488. See also mangrove
forest(s)
Dysidea etheria, 161, 162, 163, 164, 165, 167, 168, 504, 505,
510. See also sponge(s)

Echinometra viridis, 315, 317
ecology
analysis of, 48, 52
connectivity measurement, 442
diversity patterns in, 247

long-term studies of, 229
recruitment and, 442
eco-tourism, 243, 426. See also tourism
Ecuador, 83, 88
Electra bellula, 235, 236, 236. See also bryozoans
El Nifio-Southern Oscillation (ENSO), 7, 74, 260, 380
ENCORE, 405, 408
Encyclopedia of Life (EOL), 86
Endangered Species Act, 260, 266, 314
endemism, 198, 202
endosymbiont densities, 430
Environmental Protection Agency, 16, 28, 76
environmental research, grand challenges in, 12-19
epifauna
community development, 248, 253, 256, 257
growth rates, 255
recruitment of, 248-249, 252, 254
spatial variability in, 249
temporal patterns of, 249
Eretmochelys imbricata, 242. See also turtles
Escharoides costifer, 235, 237-238. See also bryozoans
Eucalliax, 183, 184-185
Eudendrium carneum, 232
eutrophication, 2, 16-17, 66, 288, 332-333, 359-360, 366,
391, 402-406, 408, 410
evapotranspiration, 36, 481, 483, 487
EvoDevo, 210
evolutionary connectivity, 442
Excirolana
antillea, 232, 233, 234, 235, 238
braziliensis, 219-220, 221, 222, 223, 224, 225, 226, 227
mayana, 221
oplophoroides, 104, 106
See also isopods
Exechonella antillea, 232, 233, 234, 235, 238. See also
bryozoans
Exhippolysmata, 96-98, 101-102, 104, 105, 106, 108. See also
shrimps
external transcribed spacer (ETS), 173

Favia
fragum, 444, 446
pallida, 441
See also coral
fertilizer, 349-350, 380, 396
fiddler crab. See crabs; Uca
fire coral, 50, 51. See also coral
fish
nursery habitats for, 16
of Panama, 80
Shorefishes of theTropical Eastern Pacific Online
Information System (SFTEP) and
stygobitic anchialine, 269-270, 273, 277-278
FishBase, 112, 202

fisheries bycatch, 243
Fistulobalanus pallidus, 293. See also barnacles
flagship species, 241-242, 244-245
flatworms, 31, 33
fleshy algae, 403, 404, 407-409, 448
Florida, 446

Atlantic coast of, 36, 230

coral reefs in, 35

impoundments, 460

PAR of, 366

seagrass in, 365 (see also seagrass)

surface water absorption spectra, 364
Florida Bay, 28

Florida Keys, 2, 28, 36, 405, 430-431, 435, 446, 502, 510, 512

Flower Garden Banks (FGB), 314
Foraminifera. See sponge(s)
Fungia

fungites, 446

larvae of, 263

scutaria, 261, 263, 264

See also coral

Galapagos Islands, 269, 271-273, 276
Galeta Point Marine Laboratory, 76, 86
Gambierdiscus, 286, 302, 303, 304. See also dinoflagellates
gametes, 211
gametogenesis, 33, 432, 440
gastropods, 191

salinity and, 375

temperature and, 375

See also snails
Gelidiella, 448
gene flow, 219, 220, 226-227, 442
Gesiella jameensis, 273. See also polychaete worms
global warming, 2, 44, 238, 314, 315, 323
Glovers Reef, Belize, 45, 389, 433
Glypturus acanthochirus, 182, 184-185
Golfingia vulgaris, 210. See also sipunculans
Goniastrea

aspera, 441

favulus, 441

retiformis, 445
Gonionemus, 143.
Goniopora tenuidens, 446. See also coral
Gorgonia mariaei, 174-175. See also coral
Great Barrier Reef, 405, 429, 443
greenhouse gases, 260, 410
Guadalupe, 152, 159
Guam, 292, 445-446
Guatemala, 287
Gulf of Chiriqui

chlorophyll a values in, 339

habitat, 7, 337

location, 324, 336

NUMBER 38 e¢ 521

NP ratios in, 343
oxycline in, 343
seasonal nutrients in, 341, 342
SST in, 343
thermocline in, 336
thermohaline structure of, 338
upwelling in, 336, 343
water characteristics of, 338, 339-340, 340, 341
See also Gulf of Panama; Panama
Gulf of Honduras, 287, 380, 389
Gulf of Mexico, 8, 105, 125, 129, 131, 238, 314, 469
Gulf of Panama, 74, 82, 87, 335-336, 342-343
dry seasonal nutrients in, 341, 342, 342
NP ratios in, 343
rainfall patterns in, 337
thermohaline structure of, 338
upwelling in, 342-343
water characteristics of, 338-339, 340
water clarity in, 341
See also Panama
Gulf Stream, 6-7, 34, 73, 435

Halammohydra, 143.
Halichondria magniconulosa, 160, 165, 167, 503, 504, 505,
506, 508-511, 513
Haliclona, 55, 152, 161, 165, 168, 504, 505, 508
caerulea, 161, 164, 504

curacaoensis, 159, 161, 162, 164-165, 167, 504, 505, 508,

510, 512-513
implexiformis, 161, 165, 167, 513, 504, 505, 509, 510, 513
manglaris, 157, 161, 163-165, 167, 504, 505, 508, 510,

512-513
tubifera, 161, 164, 167
vermeuleni, 157, 161, 162, 164, 165
See also sponge(s)

Haliclystus, 142
Halimeda, 156, 404, 408, 419, 421, 446, 448
opuntia, 446, 448
See also algae: macro-
Halisarca, 155, 167
caerulea, 162, 168
See also sponge(s)
Halocordyle disticha, 232
Halodule wrightii, 360, 363
Halophila decipiens, 363, 365. See also seagrass
Halosbaena, 276
Haplosclerida, 152, 157, 503, 504
Harbor Branch Oceanographic Institute (HBOI), 26, 33, 35
Hawaii, 60, 260-261, 271, 276, 292, 404-405, 407
hawksbill turtle, 242
herbicides, 380
herbivory
coral reef community and, 35
positive effects on reefs of, 449

522 ¢* SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

hermaphroditism, 96, 98, 105, 106, 108
in shrimps, 96, 106, 108
simultaneous, 96, 106, 108

Hexacorallia, 140, 142, 144.

Hippolyte, 106
inermis, 98
obliquimanus, 95
williamsi, 95, 98
See also shirmps

Hippopodina, 234
irregularis, 238
verrilli, 233, 236, 237
See also bryozoans

Hipporina, 253. See also bryozoans

historical contingency hypothesis, 98

Honduras, 51, 287

human disease, 35

hurricanes
Belize and, 51
coral mortality and, 406
damage from, 380
effects, 36
Honduras and, 51

Hydra, 146

hydrogen sulfide, 59, 66-67, 272, 278

Hydrolithon
boergesenii, 445
onkodes, 445
See also algae: coralline

Hydrozoa, 143, 145.

Hyrtios proteus, 154, 161, 164, 165, 167, 168. See also

sponge(s)

imposex

definition of, 189, 194

distribution of, 190-191, 194

TBT detection with, 195
incident irradiance, 360, 362, 382-383, 385, 386, 387-388
Indian River Lagoon (IRL)

ecosystems, 38

fiddler crab species of, 459 (see also Uca)

habitat, 347

hydrogeomorphic characteristics of, 347, 467

impoundment, 461, 461, 468, 496

location, 230, 361

mangroves in, 347, 348, 350, 352, 354 (see also mangrove

forest(s))

nutrients in, 352, 353

PAR in, 365

resorption efficiences, 352

seagrasses in, 234, 236, 360, 361 (see also seagrass)

sediments, 35

Smithsonian Marine Station at Fort Pierce and 6, 28, 35
Indian River Lagoon Species Inventory, 36

Inter-American Tropical Tuna Commission (IATTC), 243
intermediate disturbance hypothesis, 502
internal transcribed spacer (ITS), 173-174, 175, 176, 177, 202
International Biogeographic Information System, 202
International Taxonomic Information System, 202
International Union for Conservation of Nature (IUCN),
87-88, 202, 278, 314

International Year of the Reef, 7, 62, 74
interspecific fertilization, 440-441
inter-tropical convergence zone, 74, 337, 383
Intracoastal Waterway, 238
invasive species

Chesapeake Bay and, 19

marine, 8

Northeastern Pacific and, 292

Panama Canal and, 293

recruitment of, 249-250, 253, 256-257

research challenges and 18
Iotrochota birotulata, 159. See also sponge(s)
Ipomoea pescaprae, 420
island biogeography, 501
Isochrysis galbana, 191, 219, 220, 274. See also isopods
isopods, 53, 101, 220-222, 226, 273, 275, 510-511
Isthmus of Panama, 2, 7-8, 30, 78-79, 83-84, 295, 297, 336.

See also Panama

Jamaica, 7, 54, 83, 152, 159, 275, 315, 405
Jania, 448

Japan, 142-143, 216, 273, 275, 442, 484, 487
jellyfish, 145, 370

Johnson, Seward Sr, 26

keystone fish, 410
KRE (potassium resorption efficiency), 349, 351, 354

lactate dehydrogenase (LDH), 278
Laguna de Chiriqui, 324
Laguncularia racemosa (white mangrove), 347, 348, 420, 474,
492. See also mangrove forest(s)

larva(1)

behavior, 445

brachiolarian, 33

brachiopod, 33

coral reefs and, 438

dispersa! of 277, 441, 443

fish, 67

rearing, 262

settlement, 53, 408, 438, 443-447, 450

ocean acidification and, 446

oceanic, 33

survival and recruitment, 438

temperature effects on, 446
Lasionectes, 276. See also remipedes
Laurencia papillosa, 446. See also algae

Lepidophthalmus richardi, 184, 186
Lichenopora, 235, 237. See also bryozoans
Lindra thalassiae, 31
Link, Edwin, 26
Lissoclinum fragile, 234. See also ascidians
Lissodendoryx, 152
colombiensis, 160, 164
isodictyalis, 152, 160, 165-167, 504, 505, 509, 510,
513
See also sponge(s)
Lobophora variegata, 446-447. See also algae
Lyngbya
majuscula, 446
polychroa, 446
See also cyanobacteria
Lysmata
amboinensis, 96, 98, 104
anchisteus, 101, 104
ankeri, 98, 104, 106
argentopunctata, 101, 104
boggessi, 97, 99, 101, 102, 104, 106
californica, 96, 98, 104, 105
chica, 101, 103, 104
galapaguensis, 97, 99, 101, 102, 104
grabhami, 96, 98, 104, 106
kuekenthali, 101, 103, 104
moorei, 101, 103, 104
pederseni, 96, 98, 105
philippinensis, 101, 105
racemosa, 492, 494-496
rathbunae, 101, 103, 105
seticaudata 98, 105
phylogeny, 107
trisetacea, 101, 103, 105
vittata, 101, 103, 105
wurdemanni, 96, 105, 106, 108
See also shrimps
Lytocarpia, 143.

Madracis auretenra, 318. See also coral
malate dehydrogenase (MDH), 278
Manania, 142.
manatee grass, 234, 360
mangrove crab, 58
mangrove forest(s)
in Belize, 36, 164-165
black, 36, 474, 347, 348, 420, 424, 492, 494, 495
channel bottom, 58
characteristics, 348, 473
clear cutting, 285, 287, 416, 420, 423, 426-427, 477
coastal erosion and, 423, 474
dwarf, 36, 474, 477, 486, 488
ecosystems, 345, 480
epifauna of, 168 (see also epifauna)

NUMBER 38

flooding and, 487

forests (see mangrove forest(s))
growth, 350, 355, 356

human effects on, 474, 487
hurricane damage to, 36

loss of, 287

nutrient conservation and, 351, 354
in Panama, 36, 156-157

523

peat, 66-67, 284-287, 348, 355, 416, 419, 421, 423-424,

425,477
Pelican Cays and, 419
pollution and, 474

prop roots and prop root communities of, 58, 155, 156,
159, 169, 282, 396, 502, 505, 506, 507, 508,

513

red, 38, 50, 54, 57, 58, 155, 159, 283, 325, 346, 416, 422,

474, 477, 480, 486-489, 492, 502

as reef fish nurseries, 345, 415, 426, 474

resorption efficiences of, 352, 353

root-algal mats, 424

saplings, 495, 497-498, 498

sea level and, 487, 489

senesced leaves of, 355

sensitivity to sediments, 424

sponge faunas and, 162, 166, 510, 512

stability, 288

surveys, 492, 496

swamp communities, 58

Twin Cays and, 282, 419, 475

white, 347, 348, 420, 474, 492
mangrove oyster, 58
Manning equation, 483
Manning roughness coefficient, 483-484, 484
marine

activities, 87-88

caves, 269 (see also caves)

ecology, 35-56

education and outreach, 85-87

natural products, 32, 35

turtles (see marine turtle(s))

See also ocean; sea
Marine Environmental Sciences Program (MESP) 76
Marine Protected Areas, 410

Marine Science Network (MSN), 1, 2, 4-6, 8, 19, 30, 36, 66,

360
marine turtle(s)
conservation, 242-244
endangered, 242
as food source, 242
outreach and public education, 243
program in Brazil, 242
research, 242-244
marine zooarchaeology, 80, 87
Markov chain Monte Carlo (MCMC) methods, 98

524 e* SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Meandrina meandrites, 447
Medusozoa, 140, 143-145
Melinda Forest Station, 477, 481
Menippe, 182, 183
nodifrons, 183
See also crabs
Merguia, 98, 108
rizophorae, 98
See also shrimps
Merhippolyte, 108. See also shrimps
Mesoamerican Barrier Reef System (MBRS), 2, 6, 416, 423-
424, 426
mesozooplankton, 371, 372, 374-375
metabolism, 59, 189, 270, 277-278
meta-communities, 502
metamorphosis, 33-34, 211-212, 214-216, 265, 443, 445-
446, 451
Mexico, 131, 184, 204, 220, 238, 242, 244, 293, 304, 305
Mictaxius thalassicola, 185. See also crabs
Mictocaris halope, 274, 275
Millepora, 50, 51, 238, 314
complanata, 51, 283
See also coral
Mimocaris, 108. See also shrimps
Mithrax, 182-183. See also crabs
Mnemiuopsis leidyi, 369, 371-372, 372, 374-375
Mogula manhattensis, 253. See also tunicates
molecular evolution, 8, 80, 82, 145, 173
Monanchora, 60. See also sponge(s)
Montastrea, 80, 266, 318, 440, 441
annularis, 318, 408, 440-441, 448, 451
faveolata, 266, 439, 441, 443, 444
franksi, 441
leidyi, 369-374, 375-376
Montipora
capitata, 444, 446
digitata, 440
Morone americana, 21. See also fish
Morula, 194
Mozambique, 469
Munidopsis polymorpha, 270, 275, 277, 279.
See also crabs
muricids, 189-190, 191, 192, 192, 194
Mycale
carmigropila, 160, 162, 164
laevis, 167
magnirhaphidifera, 167, 504
muicrosigmatosa, 157, 160, 162-163, 164, 165, 167, 168,
504, 505
See also sponge(s)
Mycetophyllia aliciae, 447
Myxobolus, 143.
Myxosporea classification, 144
Myxozoa, 140, 143, 143, 145.

Naos Island Laboratory, 76, 80
National Museum of Natural History (NMNH), 2, 3, 4-6, 26,
30, 32, 34, 38, 57, 98, 103, 125, 134, 138
National Zoological Park (NZP), 5
Nature Conservancy, 87
Naushonia, 183, 185-186. See also crabs
nematocyst, 139
nemerteans, 34
Neocallichirus
grandimana, 184-185
maryae, 184-185
rathbunae, 185
raymanningi, 185
See also crabs
Nephasoma pellucidum, 210-211, 212, 214, 215-216
net ecosystem exchange (NEE), 397
Netherlands Antilles, 152, 324, 442
nitrogen
cycling, 36
discharge, 18, 21
limitation, 346, 355, 404
resorption efficiency (NRE), 349, 349, 351
resorption proficiency of, 349
sources, 404
See also nutrient
NOAA (National Oceanic and Atmospheric Administration),
302-303
Nolella stipata, 233, 234, 235, 236, 236, 237. See also
bryozoans
non-governmental organizations (NGOs), 86
Nova Scotia, 242
N-resorption efficiency, 349, 351
N-resorption proficiency, 354
Nucella lapillus, 189, 194
nutrient
concentrations, 156
pollution from, 346, 406, 409
recycling by crabs, 459
threshold hypothesis, 405, 407

OBIS, 202
ocean
acidification, 260, 446
chlorophyll a concentrations, 323 (see also chlorophyll a)
currents, 66, 248
primary productivity, 323
upwelling, 335, 341
See also marine; sea
Octocorallia, 140, 142, 144, 147.
Oculina, 6, 38
Online BRS Bilingual Biodiversity Database, 87
Online Information System on Tropical Eastern Pacific
Shorefishes, 87
Ophryotrocha puerulis, 108

Oreaster, 68. See also starfish
ostracods, 272, 273, 274, 275-276
Ostreopsis, 286, 302, 303. See also dinoflagellates
Ototyphlonemertes, 34
overfishing, 4, 12, 16, 19, 83, 206, 402, 405-406, 410
oxygen-deficient waters, 275
Oxypora lacera, 446, 449
oyster
fishery, 16
larvae, 376

Padina, 404
Palmer Long-Term Ecological Research Program, 5
Palythoa, 142
Panama
Bocas del Toro Archipelago of (see Bocas del Toro,
Panama)
Canal (see Panama Canal)
coastal communities of, 74
fishery resources of, 336
geography of, 73, 74-76, 77, 336, 337-338, 361
Gulf of (see Gulf of Panama)
history and ecology of, 7
intertidal community ecology of, 81
invasions in, 80-81, 295, 298
mangrove forests and, 347 (see also mangrove forest(s))
map of, 77
oceanographic conditions in, 156, 336, 338-343, 339,
340, 341, 342
Pacific shelf of, 74
Paleontology Project (see Panama Paleontology Project)
seagrass in, 365 (see also seagrass)
shipping history of, 296 (see also Panama Canal)
Smithsonian Tropical Research Institute in (see
Smithsonian Tropical Research Institute (STRI),
Panama)
sponges in, 159, 166, 506
surface water absorption spectra in, 364, 366
water quality in, 363
Panama Canal, 7, 74, 80-81, 83, 88, 190
barnacles and, 296
construction of, 294
expansion, 296
marine invasions and, 291, 297
shorefishes and, 80
transiting, 295, 297
vessel transits, 80, 294
See also Panama
Panama Canal Authority, 296-297
Panama Canal Zone, 74
Panama Paleontology Project (PPP), 8, 83-85
Panulirus, 182
Papua New Guinea, 314, 435
Parahippolyte, 108. See also shrimps

NUMBER 38 e¢ 525

Parasmittina, 233, 234, 235, 237-238. See also bryozoans
Paraspadella anops, 273
Paraxiopsi spinipleura, 185-186
parrotfish, 34, 407, 408. See also fish
Paspalum distichum, 420
Pasythea tulipifera, 232, 233, 235
peanut worms, 209. See also sipunculans
peat. See mangrove forest(s): peat
Pectinia paeonia, 441
Pelagomacellicephala iliffei, 273, 274. See also polychaete
worms
pelagosphera, 33-34, 210, 212, 214-216
Pelican Cays, 60, 157, 281, 417
periwinkle, 58
pesticides, 380
Peyssonnelia, 445
Phascolion
cryptum, 33, 215
strombus, 215
Phascolosoma, 215-216. See also sipunculans
phosphorous limitation, 346, 355, 404
photosynthetically available radiation (PAR), 360, 364-365,
365, 365
photosynthetically usable radiation (PUR), 364-365, 365
Phragmatopoma lapidosa, 232. See also worms
Phragmites australis, 14, 18
physiological ecology, 80-81
phytoplankton, 62, 286, 323, 325-327, 332, 360, 406
biomass, 325, 332
dissolved nutrients and, 340
growth, 342, 343
thermocline influence on, 344
See also plankton; zooplankton
plankton
ciliates, 33
larvae, 248, 441-442
mesozoo-, 371, 372, 374-375
phyto- (see phytoplankton)
productivity, 16
zoo- (see zooplankton)
plate tectonics, 276
Platygyra, 440, 441, 449
daedalea, 441, 446
sinensis, 440
Plaxaura, 142
Pocillopora
damicornis, 408, 446
meandrina, 261, 264-265, 264
See also coral
Poecilosclerida, 152, 157, 503, 504
pollution, 474
polychaete worms, 53-54, 210, 234, 249, 253, 255
polychlorinated biphenyls (PCBs), 19, 21
polymerase chain reaction (PCR), 112, 174-175, 177, 220-221

526 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

Polypodium, 142, 14S, 147.
Pomatogebia operculata, 185-186
Pontonia margarita, 95. See also shrimps
Porifera. See sponge(s)
Porites, 285
astreoides, 283, 317, 318, 439, 446, 451
compressa, 408
cylindrica, 447
damicornis, 408
divaricata, 283, 314
furcata, 283, 314
porites, 283, 407
See also coral
potassium resorption efficiency (KRE), 349, 351, 354
Pourtalesella incrassata, 233, 235
Processa, 185. See also crabs
Prorocentrum, 286, 302, 303. See also dinoflagellates
protandric simultaneous hermaphroditism (PSH), 96-98,
102-106, 108
protogyny, 96
Pseudopterogorgia bipinnata, 174-176, 174, 175, 176, 177,
178. See also coral
public health 307
Puerto Rico, 119, 123, 136, 152, 238, 260-261, 324
Punta Culebra Nature Center (PCNC), 86
Punta Galeta Laboratory, 76
pyranometer, 382
Pyura chilensis, 108

radiative transfer modeling, 362

rainfall, temporal changes in, 330
Rancheria Island field station, 76, 83
rDNA, 66, 140, 143, 144, 145, 173-174
reactive oxygen species, 31, 435
rebreathers, 279

Red Sea, 104-105, 405, 429

red tide, 2, 35, 283, 286

relative dominance model (RDM), 402, 403, 405, 406, 408
remipedes, 273, 274, 275, 276

Reunion Island, 405

Rhabdadenia biflora, 420

Rhizophora mangle, 54, 58, 151, 282, 346-347, 348, 349-350,

350, 354, 355-356, 416, 420, 422, 422, 424, 426,
461, 474, 492, 494-496, 502

rhodamine fluorescent dye, 480, 485

Rhode River, Chesapeake Bay, 4, 12-19, 13, 14, 15, 18,
370-371, 374-376, 392, 396

Rhynchocinetes typus, 95. See also shrimps

Rypticus, 67. See also fish

Salicornia virginica, 420. See also saltmarsh
salt marshes, 12, 14, 15, 16, 18, 346, 399, 491-492, 496
San Blas Archipelago, 7, 75

San Blas islands, 76, 78
sand dollars, 32
Sarcophyton glaucum, 447
Sargassum, 404, 408
polycystum, 446
See also algae: macro-
Savignyella lafontii, 232, 233, 235, 236, 237. See also
bryozoans
scanning electron microscopy, 30, 51, 211
Schizoporella
floridana, 234, 236, 237
pungens, 238
unicornis, 232, 233, 234, 235
See also bryozoans
Schoenoplectus americanus (formerly Scirpus olneyi), 12, 15,
392, 396
Scleractinian corals, 437-438. See also coral
Scopalina ruetzleri, 160, 284, 504. See also sponge(s)
Scripps Institution of Oceanography, 51, 138
Scrupocellaria
bertholletti, 234, 235, 236, 237
regularis, 232, 233, 235
See also bryozoans
sea
grass (see seagrass)
level, 14, 315, 416, 450, 475, 487, 489
seasonal upwelling in, 335
-surface temperatures (SSTs), 260, 303, 338, 343, 376,
389
warming, 74
See also marine; ocean
seagrass
depth limits, 363, 365
importance of, 359
light absorption by, 360
light requirements of, 360, 366
temperature, 313
as water quality indicator, 359
sea urchin, 78, 406, 449
Secchi depth, 326, 327, 330-331, 333, 338
chlorophyll a concentration and, 332
climatic variables and, 329
factors affecting, 332
physical variables and, 330
temporal changes in, 328, 329, 330, 331, 331-332
See also water clarity
Secchi disc, 16
SECORE, 260
sediment(s)
characteristics of, 466-468
Gulf of Honduras and, 389
in mangroves, 474, 497
nutrient loads, 389

organic content, 462, 467
reefs and, 387
senescence, 350-351, 355
sequential hermaphroditism, 106
Sesuvium portulacastrum, 420, 498
Shorefishes of theTropical Eastern Pacific Online Information
System (SFTEP), 197, 198, 204, 206, 206, 207, 208
shrimps
abundance of, 101
geographic distribution of, 104
hermaphroditism in, 96, 106, 108
ovigerous, 97
population structure of, 101
reproductive organs of, 99
Siderastrea siderea, 283
Sinularia flexibilis, 447
Siphonosoma cumanense, 215
sipunculans, 33-34, 60, 209-211, 214-216. See also worms
Sipunculus, 215. See also peanut worms
Smithsonian Contributions to the Marine Sciences, 2
Smithsonian Environmental Research Center (SERC), 3, 4,
12, 14, 16-19, 30, 32, 60, 249, 250, 251-252,
370), 37
grand research challenges of, 12-19
Smithsonian Foundation of Panama, 86
Smithsonian Institution, 2, 5, 8, 25-26, 44, 79, 125, 138,
302-303, 307, 474, 476-477, 480
Smithsonian Institution Field Station (Belize), 381
Smithsonian Institution Tropical Environmental Sciences
Program, 76
Smithsonian Marine Science Network, 3
Smithsonian Marine Science Symposium, 1, 2
Smithsonian Marine Station (SMS), 249
equipment at, 28
at Fort Pierce (SMSEP), 3, 5-6, 25, 28, 30-36
at Link Port, 26
marine biodiversity surrounding, 32
Smithsonian Tropical Research Institute (STRI), Panama
bioinformatics office of, 87
coral reef research at, 74
educational programs in, 2, 8
facilities, 8, 76, 77
location, 3, 7, 324
marine invasion studies and, 80
Punta Culebra Nature Center of, 86
research vessels, 83
snails, 108, 193
software, 28, 63, 102, 175, 262, 265, 350
soil
characteristics of, 422
elevation, 394, 393, 394, 399
erosion, 416, 420
measurement, 418-419

NUMBER 38 ¢ 527

shear strength, 418, 423
stability, 424
surfaces, 421
solar radiation, 330
Sorites dominicensis, 429-430, 431, 432, 434-435. See also
sponge(s)
South Water Cay Marine Reserve (SWCMR), 283, 287, 288
Sparisoma viride, 408. See also fish
Spartina
patens, 12, 392, 396
spartinae, 420
Spelaeoecia, 276. See also ostracods
Speleophria, 276
spermatophore, 97, 101
spiny lobsters, 182. See also crustaceans
Spirastrella, 60
mollis, 159, 162-163, 164, 165, 168
See also sponge(s)
sponge(s)
abundance of, 164, 164, 167, 502, 503, 506,
508-512
disease, 67
distribution of, 159, 165, 166, 167
dredging and, 284
dynamics, 506, 508-509, 509
ecological characteristics of, 513
mangroves and, 50, 151, 164, 504, 505
methods for studying, 502, 511
population dynamics of, 508
species composition and relative abundance of, 502-506,
506, 507, 510-512
volume, 502-504, 506-513
Spongia, 154, 157, 161, 163-165, 167-168, 508-510, 513. See
also sponge(s)
Sri Lanka, 244
staghorn coral, 266, 282, 314-315, 317, 318, 319, 444
stalactites, 272. See also caves
stalagmites, 272. See also caves
starfish, 33, 68
crown of thorns, 407
statistics. See ANCOVA; ANOVA
Stephanocoenia intersepta, 319, 439, 447
Stygiomysis, 274
stygofauna, 269-270, 273-278
stygoxenes, 270
Stylaraea punctata, 445. See also coral
Stylophora pistillata, 408, 443, 446. See also coral
surface elevation measurement, 394
Symbiodinium microadriaticum, 430
Synnotum aegyptiacum, 232, 233, 235
Syringodium, 234, 238
filiforme, 234, 360
See also seagrass

528 e SMITHSONIAN CONTRIBUTIONS TO THE MARINE SCIENCES

TBT, 189, 190, 194, 195
Tedania
ignis, 157, 159, 160, 162-163, 164, 165-166, 167,
168-169, 503, 504, 505, 508-510, 513
(see also sponge(s))
klausi, 504, 505, 510, 513
temperature loggers, 432, 478
Thais, 194
Thaisella kiosquiformis, 191. See also gastropods
Thalamoporella floridana, 232-233, 235
Thalassia, 182, 326
testudinum, 237, 283, 284, 360, 362-365, 431
See also seagrass
Thalassianthus, 142, 184.
Themiste
alutacea, 215
lageniformis, 215
pyroides, 215
See also sipunculans
Thermosbaenacea, 275. See also crustaceans
Thor, 106. See also shrimps
Thyroscyphus ramosus, 231, 232
Thysanocardia nigra, 215. See also sipunculans
tide
currents, 153, 269, 272, 278, 389
cycle, 269, 477-478, 480
elevation, 388
fluctuation, 482, 484, 488
flushing, 346, 355
phases, 482
range, 416, 477, 481
wetlands, 391
See also sea
tourism, 11, 86-87, 243, 287, 289, 303, 325, 332, 380, 426
transmission electron microscopy (TEM), 28, 211, 271
Tributyltin (TBT), 189-190, 194-195
Trichoplax, 63
Trinidad, 152
Tubastraea coccina, 442. See also coral
Tulumella, 272, 274. See also shrimps
tunicates, 108, 253
Turbinaria, 404
turbinata, 404
See also coral
turtle grass, 182-183, 234, 283, 431
turtles
harvesting of, 244
hawksbill, 242
marine (see marine turtle(s))
Twin Cays, Belize
archipelago, 474
bleached coral in, 434
B. maritima studies at, 492, 494
climate, 477

hydrology 480-486

location, 157, 416, 417, 475, 493
mangroves on, 477

oceanographic conditions, 433, 484
research at, 56, 488

seasonal climatic variations, 481
sponges of, 505

topography, 476, 480

West Island, 418-419, 474, 476-478, 479, 480, 481, 483,

485, 486, 487, 494
two-way analysis of variance, 371, 395. See also ANCOVA;
ANOVA
Typha, 420
Typhlatya, 272, 276. See also shrimps
Typhliasina pearsei, 274

Uca, 81, 459, 468, 469
abundance and biomass, 462, 463, 464, 465
biomass of, 465minax, 460
burrows, 466, 467, 467
fresh water tolerance of, 467
longisinalis, 469
mordax, 460
pugilator, 460, 466
pugnax, 460
rapax, 460, 462, 465, 466, 468, 469
sex distribution of, 465, 466
size distribution of, 466
speciosa, 460, 462, 465, 466, 468, 469
spinocarpa, 469
substrate and salinity preferences, 468
temperature tolerance of, 469
thayeri, 460, 462, 463, 465, 466, 468, 469
See also crabs
Ulva, 444, 446
UNESCO World Heritage List, 283
United Nations
Convention on the Law of the Sea (UNCLOS), 243
Educational, Scientific and Cultural Organization
(UNESCO), 51
Environmental Programme (UNEP) Regional Seas
convention, 243
Food and Agriculture Organization (FAO), 244, 303
United States
Antarctic Program (USAP), 5
Army Corps of Engineers, 35
Convention for Migratory Species, 243
Environmental Protection Agency, 16, 28, 76
Geological Survey, 48, 86
National Oceanic and Atmospheric Administration
(NOAA), 302-303
National Science Foundation, 45, 182
University National Oceanographic Laboratory System
(UNOLS), 7

Upogebia, 185-186
acanthura, 186
corallifera, 186
missa, 186
vasquezi, 186
See also crabs

Uruguay, 220, 242

Vaughaniella, 404. See also algae: turf

Venezuela, 83, 118-119, 121, 123, 137-138, 151, 152, 159,
166, 324, 405, 502, 510, 512

Virginia Institute of Marine Science (VIMS), 249

water
absorption spectra, 364
clarity (see water clarity)
depths and velocities, 484
mangrove forests and, 485
temperature and bleaching, 432
water clarity, 16, 17, 272, 325, 328, 333, 381, 383, 431
factors affecting, 332, 333
incident solar radiation and, 388
measurement, 361
phytoplankton and, 343
seagrass and, 359
significance of, 323
turbidity and, 387
watershed nutrient discharge, 4, 16
Watersipora, 232
subtorquata, 232, 233, 234, 235
See also bryozoans
wetlands
coastal, 474, 492
mangrove, 346
marginal, 460

NUMBER 38

restoration 35
tidal, 391
tropical, 473
white mangrove. See mangrove forest(s): white
wind speed, 330
World Heritage Site, 6, 83, 87, 287, 288
World Register of Marine Species (WoRMS), 202
World Trade Organization (WTO), 244
worms
cnidarian (see cnidarian worms)
flat-, 80
nemertine, 44
peanut, 209
sabellariid, 58, 231-232, 269
sipuculan, 33-34, 60, 209-211, 214-216
tube (see worms: sabellariid)
WoRMS (World Register of Marine Species), 202

Xestospongia bocatorensis, 152

Yucatan, 324

Yucatan Peninsula, Mexico, 269, 271, 275, 278-279, 416

zooarchaeology, 82
Zoobotryon verticillatum, 233, 235, 236, 237. See also
bryozoans
zooplankton
egg production, 374-375
environmental conditions for, 372
gelatinous, 370-371, 376
polyps, 375
zooplanktivorous, 370
Zoothamnium, 67

529

zooxanthellae, 6, 82, 241, 265, 430, 435, 439, 440, 442, 446

|
|
}
{

REQUIREMENTS FOR SMITHSONIAN SERIES PUBLICATION

ALL MANUSCRIPTS ARE REVIEWED FOR ADHER-
ENCE TO THE SISP MANUSCRIPT PREPARATION AND
STYLE GUIDE FOR AUTHORS (available on the “Submis-
sions” page at www.scholarlypress.si.edu). Manuscripts not
in compliance will be returned to the author. Manuscripts in-
tended for publication in the Contributions Series are evalu-
ated by a content review board and undergo substantive peer
review. Accepted manuscripts are submitted fer funding ap-
proval and scheduling to the Publications Oversight Board.

MINIMUM MANUSCRIPT LENGTH is thirty manuscript
pages. If a manuscript is longer than average, an appropriate
length will be determined during peer review and evaluation
by the Content Review Board. Authors may be asked to edit
manuscripts that are determined to be too long.

TEXT must be prepared in a recent version of Microsoft
Word; use a Times font in 12 point for regular text; be double
spaced; and have 1” margins. Each chapter/section must be
saved in a separate file.

REQUIRED ELEMENTS are title page, abstract page, table
of contents, main text, and reference section. See the SISP
Manuscript Preparation and Style Guide for Authors for the
order of all elements.

HEADINGS should be styled so different levels of headings are
distinct from each other and so the organization of the manu-
script is clear. Insert one line space above and one line space
below all headings.

FRONT MATTER should include title page, abstract page,
and table of contents. All other sections are optional. Ab-
stracts must not exceed 250 words. Table of contents should
include A-, B-, and C-level headings.

TABLES (numbered, with captions, stubs, rules) should be
submitted in separate MS Word files; should include foot-
notes, if appropriate; should have rules only at top, bottom,
and beneath column heads. Print outs of each table should
accompany the manuscript to ensure correct layout of data.
Tabulations within running text should not be numbered or
formatted like formal tables, and should be included in the
text of the manuscript.

FIGURE CAPTIONS should be provided in a separate MS
Word file.

FIGURES (e.g., photographs, line art, maps) should be num-
bered sequentially (1, 2, 3, etc.) in the order called out; be
placed throughout text, not at end of manuscript; have all
components of composites lettered with lowercase letters and
described in the caption; include a scale bar or scale descrip-
tion, if appropriate; include any legends in or on the figure
rather than in a caption.

ART must not be embedded in the main text.

Figures must be original and submitted as individual TIFF
or EPS files. Resolution for art files must be at least 300 dpi
for grayscale and color images and at least 1200 dpi for line
art. Electronic images should measure no more than 100%
and no less than 75% of final size when published. JPG files
will not be accepted. Color images significantly increase
costs so should be included only if required. Funding for
color art is subject to approval by SISP and the Publications
Oversight Board.

TAXONOMIC KEYS in natural history papers should use
the aligned-couplet form for zoology. If cross referencing is
required between key and text, do not include page references
within the key but number the keyed-out taxa, using the same
numbers with their corresponding heads in the text.

SYNONOMY IN ZOOLOGY must use the short form (tax-
on, author, year:page), with full reference at the end of the
paper under “References.”

IN-TEXT REFERENCES should be used rather than biblio-
graphic notes and should follow the author-date system in the
following format: “(author last name, year)” or “. . . author
(year); ~
author (year:page).” A full citation should be included in a

“

(author, year:page used within the text)” or

“References” section.

ENDNOTES are to be used in lieu of footnotes and should
be keyed manually into a separate MS Word file, in a sec-
tion titled “Notes”. Notes should not contain bibliographic
information. Manually type superscript numerals in text and
use full-sized numerals at the beginning of each note in the
“Notes” section. SISP will determine the best placement of the
notes section, either at the end of each chapter or at the end
of the main text.

REFERENCES should be in alphabetical order, and in chron-
ological order for same-author entries. Each reference should
be cited at least once in main text. Complete bibliographic
information must be included in all citations (e.g., author/edi-
tor, title, subtitle, edition, volume, issue, pages, figures). For
books, place of publication and publisher are required. For
journals, use the parentheses system for volume(number):pag-
ination [e.g., “10(2):5-9”]. Do not use “et al.”; all authors/
editors should be included in reference citations. In titles,
capitalize first word, last word, first word after colon, and all
other words except articles, conjunctions, and prepositions.
Examples of the most common types of citations are provided
in the SISP Manuscript Preparation and Author Style Guide.

For questions regarding the guidelines, please email SISP at
schol.press@si.edu.