Full text of "World Atlas of Biodiversity: earth's living resources in the 21st century"

WORLD ATLAS OF

BIODIVERSITY

EARTH'S LIVING RESOURCES IN THE 21st CENTURY

> (\

X >r

UNEP WCMC

BRIAN GROOMBRIDGE and MARTIN D. JENKINS

World Atlas of Biodiversity addresses the remark-

ible growth in concern at all levels for living things

and the environment and the increased appreciation

' the links between the state of ecosystems and

the state of humankind. Building on a wealth of re-

search and analysis by the conservation community
worldwide, this book provides a comprehensive
and accessible view of key global issues in biodiver-
sity. It outlines some of the broad ecological
relationships between humans and the rest of the

iterial world and summarizes information on the

health of the planet. Opening with an outline of

some fundamental aspects of material cycles and

energy flow in the biosphere, the book goes on to

discuss the expansion of this diversity through geo-
logical time and the pattern of its distribution over

the surface of the Earth, and analyzes trends in the

condition of the main ecosystem types and the

species integral to them.

Digitized by tine Internet Arciiive

in 2010 witii funding from

UNEP-WCIVIC, Cambridge

Iittp://www.arcliive.org/details/worldatlasofbiod02groo

World Atlas of Biodiversity

Published in association witli
UNEP-WCMC by the University of
California Press

University of California Press
Berl<eley and Los Angeles, California
University of California Press, Ltd.
London, England

UNEP-WCMC
219 Huntingdon Road
Cambridge CB3 DDL, UK
Tel; *Uk 101 1223 277 3U
Fax:+a(01 1223 277 136
E-mail: info0unep-wcmc.org
Website: www.unep-wcmc.org

World Atlas of Biodiversity:

Earth's Living Resources in the 21st Century

IS a revised and updated edition of

Global Biodiversity:

Earth's Living Resources in the 21st Century

No part of this book may be reproduced by
any means or transmitted into a machine
language without the written permission of
the publisher

Ihe contents of this volume do not
necessarily reflect the views or policies of
UNEP-WCfvIC, contributory organizations,
editors or publishers. The designations
employed and the presentations do not imply
the expression of any opinion whatsoever on
the part of UNEP-WCIvIC or contributory
organizations, editors or publishers
concerning the legal status of any country,
territory, city or area or its authority, or
concerning the delimitation of its frontiers or
boundaries or the designation of its name or
allegiances.

Cloth edition ISBN

0-520-23668-8

Cataloging-in-publication data is on file with
the Library of Congress

Citation Groombndge B. and Jenkins M.D.
120021 World Atlas of Biodiversity. Prepared
by the UNEP World Conservation Monitoring
Centre. University of California Press,
Berkeley, USA.

UNEP WCMC

World Atlas of Biodiversity

Earth's Living Resources in the 21st Century

Brian Groombridge & Martin D. Jenkins

UNIVERSITY OF CALIFORNIA PRESS

Berkeley Los Angeles London

World Atlas of Biodiversity

Prepared by

UNEP World Conservation

Monitoring Centre

219 Huntingdon Road
Cambridge CB3 ODL, UK
Tel: +M 101 1223 277 3U
Fax: +44 (01 1223 277 136
E-mail: info0unep-wcmc.org
Website: www.unep-wcmc.org

Director

Marl< Collins

Authors

Brian Groombndge
Martin D Jenkins

Additional contributors

Adrian C. Newton (Project manager)

Rachel Cool<

Neil Cox

Victoria Gaillard

Edmund Green

Janina Jakubowska

Thomas Kaissl

Valerie Kapos

Charlotte Lusty

Anna Morton

Mark Spalding

Christoph Zockier

UNEP WCMC

The UNEP World Conservation Monitoring
Centre is the biodiversity information and
assessment arm of the United Nations
Environment Programme, the worlds
foremost intergovernmental environmental
organization. UNEP-WCMC aims to help
decision-makers recognize the value of
biodiversity to people everywhere, and to
apply this knowledge to all that they do. The
Centre's challenge is to transform complex
data into policy-relevant information, to build
tools and systems for analysis and
integration, and to support the needs of
nations and the international community as
they engage in joint programs of action.

Production of maps

Simon Blyth
with the assistance of
Igor Lysenko
Corinna Ravilious
Jonathan Rhind

Layout

Yves Messer

A Benson production

27 Devonshire Road
Cambridge CBl 2BH, UK

Color separations

Swaingrove

Printed in the UK

Acknowledgments

First and foremost we would [ike to express our deepest ttianks to the Aventis Foundation,
without whose generous funding the research and production work for this book could not
have been undertaken. Preparation of the book was also generously supported by the
Department of Environment, Food and Rural Affairs IDEFRAI of the UK Government. The
Owen Family Trust is also acknowledged for financial support to the first edition of this text.

We also acknowledge with thanks the generous assistance extended by the following, listed
approximately in the same sequence as the chapters in which their material appears:

Christopher Field and George tvlerchant. Department of Global Ecology, Carnegie Institution of
Washington for use of data from a global model of net primary production.

Robert Lesslie of the Department of Geography, Australian National University, Canberra, for
allowing us to use data resulting from his global wilderness analysis.

BirdLife International, of Cambridge, UK, for allowing use of spatial data on endemic bird
areas and on threatened bird species.

Gene Carl Feldman, Oceanographer at NASA/Goddard Space Flight Center, Greenbelt,
Maryland, for approving use of material from the SeaWiFS Project of NASA/Goddard Space
Flight Center and ORBIMAGE.

The University of Maryland Global Land Cover Facility, for facilitating use of land cover data.

Professor Wilhelm Barthlott of the Botanisches Institut und Botanischer Garten, Rheinischen
Fnedrich-Wilhelms-Universitat, Bonn, for kindly allowing use of a map showing contours of
global plant species diversity

Jonathan Loh, responsible for the WWF Living Planet Report, for kindly approving use of
global trend indices from the Living Planet Report 2000.

John E.N. Veron, Chief Scientist at the Australian institute of Marine Sciences, Townsville,
Queensland for allowing use of coral generic diversity data.

Several biologists associated with IUCN/S5C specialist groups on fishes, mollusks and inland
water Crustacea, for providing data and expertise on important areas for freshwater
biodiversity collated in an earlier publication: Gerald R. Alien (Western Australian Museum);
the late Denton Belk [TexasI; Philippe Bouchet ILaboratoire de Biologie des invertebres
marins et matacologie. Museum National d'Histoire Naturelle, Paris); Keith Crandall
(Department of Zoology, Brigham Young University); Neil Cumberlidge (Department of

Biology, Northern Michigan University); Olivier Gargominy iLaboratoire de Biologie des
invertebres marins et nnalacologie, Museum National d'Histoire Nalurelle, Pans); Maurice
Kottelat (Cornol, Switzerland); Sven 0. Kullander IDepartment of Vertebrate Zoology, Swedish
Museum ot Natural History, Stockholm); Christian Leveque (ORSTOM. Pans); R. von Sternberg
(Center for Intelligent Systems, State University of New York at Binghamton); Guy Teugels
ILaboratoire Ichthyologie, Musee Royal de I'Afnque Centrale. Tervuren).

Ben ten Brink, Jan Bakkes and Jaap van Woerden for facilitating use of material illustrating
work on scenarios earned out at the Rijksinstituut voor Volksgezondheid en Milieu IRIVM),
BiUhoven, the Netherlands.

Christian Nellemann (Norwegian Institute for Nature Research), Hugo Ahlenius (UNEP GRID-
Arendal) and the Secretariat of GLOBIO (Global methodology for mapping human impacts on
the biosphere), for material applying this approach to scenario development.

Photographs

Pages: 6, L. Olesen/UNEP/Still Pictures; 7, L.L. Hock/UNEP/Topham;

U, B. Groombndge; 15. M. Wakabayashi/UNEP/Topham; 37. Giotto Castelli;

38, M. Fnedlander/UNEP/Topham; 39. UNEP/Topham; 72. G. Bluhm/UNEP/Topham;

73. K. Kaznaki/UNEP/Topham; 86. M. Schneider/UNEP/Topham;

93, M.R. Andrianavalona/UNEP/Still Pictures; 96, H. Mundell/UNEP/Topham;

98, R. Fana/UNEP/Topham; 99. UNEP/Topham; 103 J. Nuab/UNEP/Topham;

110 top. R. delRosanon/UNEP/Topham;110 bottom. Mazinsky/UNEP/Topham;

118. E. Green; U3 top, E. Green; bottom. M. Spalding; U4, M. Garcia Blanco/UNEP/Topham;

151. E. Green; 152. D. Nayak/UNEP/StiU Pictures; 15i. E. Green;

155. D. Seifert/UNEP/StiU Pictures; 165, UNEP/Topham; 169. S.W. Mmg/UNEP/Stilt Pictures;

174, F. Colombini/UNEP/Topham; 178. C.K. Au/UNEP/Still Pictures;

185. K. Lohua/UNEP/Still Pictures; 194. C. Petersen/UNEP/Topham;

212. P Garside/UNEP/Topham; 216, C. Senanunsakul/UNEP/Still Pictures

Contents

Foreword xi

Preface xii

Introduction 1

Chapter 1: The biosphere 3

Maps

1.1 Physical geography of the Earth 4

1.2 Primary production in the
biosphere 8

Figure

1.1 Hypsographic curve 5

Tables

1.1 Global annual net primary
production 10

1.2 Estimated global carbon

budget and biomass totals 11

Chapter 2: The diversity of organisms 13

Figure

2.1 The phylogenetic tree 19

Tables

2.1 Estimated numbers of
described species, and

possible global total 18

2.2 Key features of the major

groups of living organisms 20

Boxes

2.1 New species discoveries 16

2.2 Improving taxonomic
knowledge and capacity 17

Chapter 3: Biodiversity through time

Figures

3.1 The four eons of the
geological timescale

3.2 Periods and eras of the
Phanerozoic

3.3 Animal family diversity
through time

3.4 Plant diversity through time

3.5 Freguency of percent extinction
per million year period 29

3.6 Number of family extinctions
per geological interval through

the Phanerozoic 29

Table
3.1 The principal mass extinctions

in the Phanerozoic fossil

record 30

Chapter 4: Humans, food and

biodiversity

Maps

4,1

Early human dispersal

4.2

Livestock breeds: numbers

and status

4.3

FAO world diet classes

4.4

Human population density

4,5

Terrestrial wilderness

4.6

Vertebrate extinctions since

AD1500

4.7

Threatened mammal species

4.8

Critically endangered

mammals and birds

4.9

Threatened bird species density

Figures

4.1

Human population

4.2

Vertebrate extinctions by

period since ADl 500

Tables

4.1

4.2

4.3

4.4

4.5

Top ten food commodities,
ranked by percentage
contribution to global food
supply 41

World diet classes 44

Examples of diversity in
agricultural systems 45

Number of individuals and
biomass, selected organisms 46
Land converted to cropland 49

A. 6 Estimated large herbivore
numbers and biomass in
Mesolithic and modern Britain

4.7 Late Pleistocene extinct and
living genera of large animals

4.8 Numbers of extinct animal
species according to lUCN

4.9 Island diversity at risk: birds

4.10 Threatened species

4.11 Number of threatened animal
species in major biomes

Boxes

4.1 Loss of diversity in agricultural
genetic resources

4.2 'Lazarus species'

Chapter 5: Terrestrial biodiversity

Maps

5.1 Photosynthetic activity on land

5.2 Global land cover

5.3 Diversity of vascular plant
species

5.4 Biodiversity at country level

5.5 Flowering plant family density

5.6 Terrestrial vertebrate family
density 1

5.7 Current forest distribution 1

5.8 Non-forest terrestrial
ecosystems 1

Figure

5.1 A typical species-area plot

Tables

5.1 Global distribution of land area,
by latitude bands

5.2 Different definitions of forest
cover

5.3 Sample effects on forest area
estimates of different forest
definitions

5.4 Global area of five mam forest
types

5.5 Important families and genera,
and numbers of species, in four
areas of temperate broadleaf
deciduous forest

5.6 Biomass and carbon storage in
the world's major forest types

5.7 Tree species richness in tropical
moist forests

5.8 Estimated annual change in
forest cover 1990-2000

5.9 Global protection of forests
within protected areas in

50 lUCN categories l-VI 97

5.10 Estimated plant species

51 richness in the five regions of
Mediterranean-type climate 105

57 Boxes

60 5.1 Defining ecosystems 74

5.2

Species and energy

5.3

Fire in temperate and boreal

forest

5.4

Temperate forest bird trends

5.5

Grassland bird trends

104

Chapter 6: Marine biodiversity

Maps

117

6.1

Coral reef hotspots

126

Shark family diversity

128

Marine turtle diversity

130

Mangrove diversity

134

Seagrass species diversity

136

Coral diversity

140

Marine fisheries catch and

discards

Figures

6.1

Species contributing most to

global marine fisheries

145

6.2

Marine fisheries landings by

major group

146

6.3

Global trends in the state

of world stocks since 1 974

6.4

Trends in global fisheries

catch since 1970

147

6.5

Marine aquaculture

production

150

6.6

Marine population trends

158

Tables

6.1

Area and maximum depth of

the world's oceans and seas

117

6.2

Relative areas of continental

shelves and open ocean

118

6.3

Marine diversity by phylum

122

6.4

Diversity of craniates in the

sea by class

123

6.5

Diversity of fishes in the

seas by order

124

6.6

Marine tetrapod diversity

125

6.7

Regional distribution of

breeding in seabirds

129

6.8

Diversity of mangroves

132

Current mangrove cover

133

6.10 Diversity of stony corals in
ttie order Scleractmia

6.1 1 Coral reef area

6.12 Taxonomic distribution and
status of tfireatened marine
animals

Boxes

6.1 Life in sediments

6.2 Marine introductions

7.2

7.3

7.4

7.5

7.2

7.3
lA

138
139

156

119
153

Chapter 7: Inland water biodiversity 163

Maps

7.1 Freshwater fisfi family

diversity 170

Major areas of diversity of

inland water fish 176

Major areas of diversity of

inland water mollusks 180

Major areas of diversity of

selected inland water

crustacean groups 182

Priority river basins 190

Figures

7.1 Reported global inland
fisheries production
Inland water fish
introductions 1

Freshwater population trends 1
River basin richness and
vulnerability 1

7.5 Changes in condition of a
sample of freshwater lakes
between 1950s and 198Ds 1

Tables

7.1 Components of the hydrosphere 1

7.2

179

Physical and biodiversity features
of major long-lived lakes

7.3 Partial list of global hotspots
of freshwater biodiversity

7.4 Insects of inland waters

7.5 Fish diversity in inland
waters, by order

7.6 Tetrapod diversity in inland
waters

7.7 Major inland fishery countries

7.8 Numbers of threatened
freshwater fishes in selected
countries

7.9 Taxonomic distribution and
status of threatened inland
water vertebrates

68
72

173

175
179

187

7.10 Thirty high-priority river basins 190
Boxes

7.1 Saline and soda lakes 164

7.2 Wetland loss in Asian drylands 184

Chapter 8: Global biodiversity:

responding to change 195

Maps

8.1 World protected areas 200

8.2 Centers of plant diversity 202

8.3 Major areas of amphibian
diversity 204

8.4 Endemic bird areas 206

8.5 Marine protected areas 208

8.6 International protected area
agreements 210

Figures

8.1 Development of the global
network of protected areas 198

8.2 Possible future scenarios
from GEO 3. evaluated with

RIVM IMAGE 218

8.3 Possible future scenarios,
GLOBIO 220

Table

8.1 Major global conventions

relevant to biodiversity

maintenance 214

Boxes

8.1 Pioneering NGOs in
biodiversity conservation 197

8.2 The precautionary principle 199

8.3 Systematic conservation
planning 203

8.4 Negotiating a multilateral treaty 213

APPENDICES

1 The phyla of living organisms 225

2 Important food crops

3 Domestic livestock

244
271

4 Recent vertebrate extinctions 278

5 Biodiversity at country level 295

6 Important areas for
freshwater biodiversity

INDEX

306
329

Foreword

Klaus Topfer, Executive Director, United Nations Environment Programme

It is a great pleasure for me to introduce tliis important new bool< from tlie UNEP World
Conservation Monitoring Centre. Building on the analyses it carried out for tfie Eartii Summit
in Rio de Janeiro In 1992 and for the new millennium just two years ago, UNEP-WCMC has
once again updated and revised its important overview of life on Earth in time for a major
global event.

In tune with the message of the Johannesburg World Summit on Sustainable Development,
this new atlas places humankind firmly in the context of the species and ecosystems
upon which we all depend for our livelihoods. We are a part of biodiversity and as such
we should treasure it, use It wisely and share its benefits In our own enlightened self-interest.

I commend this bool< to all who seel< a greater appreciation of the Inter-dependency between
our own future and that of global biodiversity.

In closing I should like on behalf of UNEP to thank most warmly the Aventis Foundation and
the UK Department for Environment, Food and Rural Affairs for their support In the
preparation of the World Atlas of Biodiversity.

Preface

Mark Collins, Director, UNEP World Conservation Monitoring Centre

The diversity of life is tfie defining feature of planet Earth. It is unique - as far as we know - in
the infinity of the universe. For 11 000 years since agriculture began, humankind has
increasingly appropriated the biological resources and natural productivity of lands and seas
to support the expansion of civilizations and technologies. Everything that we have achieved
has its origins in living animals, plants and the communities and ecosystems of which they are
a part. But it is only in the past 30 years, since the United Nations Conference on the Human
Environment in Stockholm in 1972, that we have begun to recognize the limits to natures gifts.
We now know that our own success is placing strain on nature's ability to evolve, diversify,
cleanse our air and water and provide us with the raw materials we need for food, fuel, fiber
and health.

Just ten years ago we began to take integrated and holistic action to ensure conservation
and sustainable use ot biological resources. The UN Conference on Environment and
Development lUNCEDI saw the signing of the Convention on Biological Diversity, the first
global agreement on biodiversity that clearly positioned humankind as an integral part of the
complex of life on Earth, rather than a special case somehow separate from nature and
immune from its laws. The 'ecosystem approach' espoused by the Biodiversity Convention
acknowledges that our relations with the rest of the living world are truly interactive, and that
what we do to nature will in turn reflect on nature's ability to respond to our own needs. The
Convention foresaw a careful balance in the management of the Earth's living wealth through
conservation, sustainable use and equitable sharing of costs and benefits.

There could be no better time to launch a fresh assessment of the living world. This World
Atlas of Biodiversity is published to coincide with the World Summit on Sustainable
Development in Johannesburg, Republic of South Africa. The focus of the meeting is once
again on sustainable development, but this time the emphasis is clearly on poverty alleviation.
The message is clear: harmonized economic, social and environmental development will be
but a dream while so many of the world's people have no choices and no opportunities to take
a planned approach to their lives. What is the relevance of this book in the context of the
Johannesburg message?

The reality is that this World Atlas of Biodiversity is of greater relevance now than at any
time in the past. The world's living wealth remains the cornerstone of sustainable livelihoods
and quality lifestyles in both the industrialized and developing worlds. Recognition of this fact
is spreading, and the value of biodiversity in people's lives, socially, economically and
environmentally, has never been more apparent than it is today.

This Is not a textbook, it is a resource pack and a survival kit for the future. I hope that all
who read it will find new insights into the significance of life on Earth to their own lives.
And that they will take steps within their homes, communities and nations to utilize and
enjoy living resources wisely, share' the benefits and hold the capital in trust for
future generations.

Introduction

OBJECTIVES

The past ten years have seen a remarkable growth in concern for wildlife and the
environment, with an increased appreciation of the linl<s between the state of ecosystems
and the state of humankind. Many analysts have concluded that achieving sustainable and
equitable human development will require, among other measures, taking a more effective
approach to managing human impacts on the biosphere. This was reinforced by the 1992
United Nations Conference on Environment and Development Ithe Earth Summit), at which
the Convention on Biological Diversity ICBD) was opened for signature. Many conservation
and management Initiatives worldwide have arisen from efforts to meet the objectives
framed by the CBD text.

In principle, any kind of variation at any level of biological organization - encompassing
genes, populations, species and communities - is biological diversity. The text of the CBD
defines biological diversity as 'the variability among living organisms from all sources
including, inter alia, terrestrial, marine and other aquatic ecosystems and the ecological
complexes of which they are part; this Includes diversity within species, between species
and of ecosystems'. In practice, the term is often contracted to biodiversity', and used to
refer collectively to all such variation: In effect, as a convenient shorthand for the total
complex of life In some given area, or on the Earth as a whole.

In the present volume we aim to use the data now available to provide an overview of
the current state of global biodiversity, using maps where helpful, and to ensure that
this information is accessible to a wide readership. While biodiversity has many dimensions,
attention is here focused on the diversity of living organisms and their populations, and on
major aquatic and terrestrial ecosystem types. Far more space is given to the macro-scale
organisms and landscape elements that may be subject to planning and management
intervention than to microorganisms, despite the Immense metabolic diversity of the
latter, and their pivotal role in driving biosphere cycles.

STRUCTURE OF THIS BOOK

The eight chapters fall informally Into four thematic sections. The first section opens with an
outline of some fundamental aspects of material cycles and energy flow in the biosphere
(Chapter 1). This Is followed by a synopsis of the diversity of living organisms (Chapter 2) and
of change In this diversity through geological time (Chapter 3). The second section (Chapter M
Is largely concerned with relationships between humankind and biodiversity, noting the
Increasing human impact on the environment from early modern humans onward, the use of
biodiversity in human nutrition, and reviewing trends in recent time, focusing on depletion and
extinction of species. The third section alms to characterize communities and biodiversity
trends in the three basic blome types: terrestrial, marine and inland waters (Chapters 5, 6 and
7, respectively). Finally, Chapter 8 introduces some of the management and planning
responses that have been implemented with a view to maintaining ecosystem health and
putting human development on a sustainable foundation.

The biosphere 3

The biosphere

rrr— he BIOSPHERE IS THE THIN AND IRREGULAR ENVELOPE around and including the
Earth's surface that contains all living organisms and the elements they exchange with
the non-living environment. Water makes up about tw/o thirds of an average living cell,
and organic molecules based on hydrogen, carbon, nitrogen and oxygen make up the remaining
one third. These and other elements of living cells cycle repeatedly betw^een the soil, sediment,
air and water of the environment and the transient substance of living organisms.

The energy to maintain the structure of organisms enters the biosphere when sunlight is
used by bacteria, algae and plants to produce organic molecules by photosynthesis, and all
energy eventually leaves the biosphere again in the form of heat. Photosynthetic organisms
themselves use a proportion of the organic material they synthesize; net primary production
Is the amount of energy-rich material left to sustain all other life on Earth.

Humans now appropriate a large proportion of global net primary production, and have
caused planetary-scale perturbations In cycling of carbon, nitrogen and other elements.

THE LIVING PLANET

The defining characteristic of the planet Earth
Is that it supports life, and has done so for at
least 70 percent of its history (see Chapter 31.
The position of the Earth relative to the sun,
its size and composition appear to be the
main factors that have allowed life to develop
here. Most importantly, these factors have
combined to ensure the permanent presence
of a large amount of liquid water on the
planets surface, and this is the fundamental
prerequisite of life as we know it.

The space occupied by living organisms
and the part of the planet that supports them
is called the biosphere'. The non-living
biosphere comprises the hydrosphere (the
waters), the soil and upper part of the
lithosphere (the solid matter that forms the
rocky crust of the Earth], and the lower part of
the atmosphere (the thin layer of gas coating
the planet's surface). These domains interact
in ways critical to the operation of the
biosphere, and are linked in particular by
the properties of water as a solvent and
medium that fosters the chemical reactions
basic to life.

While providing the conditions necessary
for life, the structure and composition of the
non-living parts of the biosphere have them-
selves been profoundly affected through time
by living organisms. Most clearly, and from
the human viewpoint most importantly, the
presence of significant quantities of free
oxygen in the atmosphere is entirely the
product of oxygen-releasing photosynthesis
by cyanobacteria starting more than 2 000
million years ago. The idea that living organ-
isms do not merely influence conditions in the
biosphere but in some way regulate them to
maintain the conditions conducive to life has
received considerable attention in recent
years, chiefly in terms of the Gaia hypothesis'
proposed in the 1970s.

The extent of the biosphere

At planetary scale, the biosphere can be
pictured as a thin and irregular envelope
around the Earth's surface, just a few kilo-
meters deep on the globe's 6 371 -kilometer
radius. Because most living organisms depend
directly or indirectly on sunlight, the regions
reached by sunlight form the core of the

k WORLD ATLAS OF BIODIVERSITY

Map 1.1

Physical geography of
the Earth

The relative areas occupied
by dry land and by water,
and the general distribution
of areas of extreme height
or depth.

'■^-

, ,£•■;

.ii*-.- ;

f-'

,'--fc\' -

V^\

>i'.

', V'

' >■■ '■;

•*.

'''

h '--

''r-^y.:

i '-•

■'.■■ . ''-'l

' '■.

5%-

'-mi

■■■•i>^:\.

biosphere: i.e. the land surface, the top few
millimeters of the soil, and the upper waters
of lakes and the ocean through which sunlight
can penetrate.

The biosphere is not homogenous, be-
cause actively metabolizing living organisms
are sparse or absent where liquid water is
absent, such as in the permanent ice at the
poles and on the very highest mountain
peaks, but abundant where conditions are
favorable. Nor are its boundaries sharply
defined, because bacterial spores and other
dormant forms of life passively disperse
virtually everywhere, from polar icecaps to
several tens of kilometers above the surface
of the Earth [approaching the upper limit of
the stratosphere], and living microorganisms
occur^ within rocks more than 3 kilometers
deep in the lithosphere.

The whole of the sea is theoretically
capable of supporting active life and con-
stitutes therefore the vast majority of the
volume of the biosphere [Figure 1.1|.
Depending on water clarity, the sunlit (photic)
zone may reach just a few centimeters to a few
hundred meters in depth, but the marine
biosphere is extended into regions of total
darkness, down to more than 10 000 meters in
the ocean depths, by organisms that subsist
on the rain of organic debris falling from the
upper waters, in addition, there are animal
communities on the sea floor based on
microorganisms deriving their energy from
hydrogen sulfide emitted from hydrothermal
vents. Overall, however, the amount of living
material in most of the sea - that part of the
open ocean below the upper hundred or so
meters - is relatively low.

The biosphere 5

The atmosphere plays a vital role in
the biosphere, not only in providing a source
of essential gases, but also in buffering
conditions at ground level, by regulating
temperature and providing a shield against
excessive ultraviolet radiation. Many organ-
isms, from microscopic bacteria to bats and
birds, spend part of their lives suspended in
the atmosphere; however, no organism is
known that passes its complete life cycle in
the air, and living biomass per unit volume
above the Earth's solid or liquid surface is
extremely low.

Photosynthesis and the biosphere

Life on Earth is based essentially on the
chemistry of water and carbon. Indeed, in
biochemical terms, living organisms are
simply elaborate systems of organic macro-

molecules dispersed in an aqueous medium.
The average cell is about 70 percent water by
weight; the remainder consists very largely of
carbon-containing (organic! compounds com-
posed mainly of the four elements hydrogen,
carbon, nitrogen and oxygen. These com-

8 849 m

Figure 1.1
Hypsographic curve

The horizontal baseline in
this figure represents the
Earth's total surface area
of 510 million km'.
The figure shows that 71%
of this surface is covered by
marine waters and 29% is
dry land. It also shows the
mean land elevation and
mean ocean depth, and the
amount of Earth's surface,
in percentage terms,
standing at any given
elevation or depth.

average elevation 840 m

^"■"■^^^ average depth 3 800 m

29%

71%

6 ~

11 035 m

i WORLD ATLAS OF BIODIVERSITY

Energy from the sun
drives photosynthesis,
responsible for the vast
majority of organic
production.

pounds include four major types of large
organic molecule - proteins, carbohydrates,
lipids and nucleic acids - and about 100
different small organic molecules. A number
of other elements are required in much
smaller, though still vital, quantities. These
include phosphorus, sulfur, iron and
magnesium. All these elements cycle through
the biosphere in a variety of forms, both
organic and inorganic, following complex and
interlinked pathways many of which are yet to
be fully elucidated. Except for some micro-
organisms that use energy derived from
inorganic chemicals, the engine that drives
the organic part of this turnover is photo-
synthesis - the capture by living tissues of
energy from the sun.

Photosynthesis essentially involves the use
of energy from sunlight to reduce carbon
dioxide ICO2I with a source of electrons
(almost invariably hydrogen) to produce
carbohydrates, water IH2OI and, generally, a
by-product from the hydrogen donor. In some
bacteria the hydrogen donor is hydrogen gas,
in others it is hydrogen sulfide; but, in
cyanobacteria, algae and plants, water is the
hydrogen donor and gaseous elemental
oxygen IO2I is the by-product. This is over-
whelmingly the predominant and most impor-
tant form of photosynthesis on the planet, and
is described by the following equation:

2nH20 + nC02 + light -^ nHjO + nCH20 + 2nO

The initial products of photosynthesis in
plants are simple sugars such as glucose

(041-1,2041. Larger carbohydrate molecules
made from glucose include cellulose, the
main component of plant cell walls and
woody tissues, and starch, a key storage
carbohydrate found in roots and tubers.
Energy is needed to make the chemical bonds
within these organic molecules, and energy is
released when the bonds are broken down
again. The controlled breakdown of these
molecules within cells is the mechanism
by which all cells obtain energy to do
useful work.

All organisms can break down sugars very
directly without the need for oxygen. Many
bacteria that live in aerobic conditions use only
this method. Virtually all eukaryotes (see
Chapter 2) have evolved a more complex
additional pathway that requires oxygen but
yields much more energy. This latter pathway
- aerobic respiration - essentially reverses the
basic photosynthetic reaction shown above.

The major cycling process of the biosphere,
therefore, consists of the photosynthetic fixing
of carbon dioxide with water to produce
organic compounds, in which energy is stored,
and oxygen; this is followed by respiration of
these compounds, in which the stored energy
is released and carbon dioxide and water are
produced. Photosynthesis therefore is not only
responsible for the vast majority of organic
production, but also for the maintenance of
free oxygen in the atmosphere, without which
aerobic organisms (the great majority of
eukaryotic organisms, including humans)
could not survive.

Although photosynthesis is the primary
engine of the biosphere, in the sense that it
injects energy into the system and creates
basic organic molecules, production of the
full range of organic molecules on which life
depends requires additional elements. Of the
four key elements, nitrogen is often the one
in limited supply, but it is an essential com-
ponent of nucleic acids and proteins.
Although the atmosphere consists of 79
percent nitrogen, this inert gaseous form of
the element cannot be used by plants or most
other organisms until combined (fixed) with
other elements. In the biosphere, atmos-
pheric nitrogen is fixed by a range of bacteria,
including cyanobacteria, some free-living soil

The biosphere 7

bacteria, and most importantly by specialized
bacteria that live symbiotically in the root
nodules of leguminous plants Ipeas, beans,
etc.) Some nitrogen is also fixed by lightning
in electric storms and, in the modern world,
industrially in the production of fertilizer.
Fixed nitrogen is made available to plant roots
through association with fungi ImychorrhizasI
and as nitrogen-fixing organisms decay.
From plant roots, it is transported to
metabolizing plant cells. On death, these
steps are reversed, and the fixed nitrogen
may be immediately recycled or revert to
elemental nitrogen.

PRODUCTIVITY AND THE CARBON CYCLE

About half of the solar energy reaching the
upper atmosphere of the Earth is immediately
reflected. Most of the remainder interacts with
the atmosphere, ocean or land, where it
evaporates water and heats air, so driving
atmospheric and ocean circulation. Much less
than 1 percent of the incoming energy is
intercepted and absorbed by photosynthetic
organisms. On land these photosynthesizers
are overwhelmingly green plants, although
cyanobacteria and algae are also present, the
latter particularly in the symbiotic associations
with fungi known as lichens. In aquatic
habitats, particularly the sea, virtually all
photosynthesis is carried out by cyanobacteria
and algae, although green plants are also
present in shallow coastal and inland waters.

Photosynthesizers fix carbon and therefore
accumulate organic mass or biomass loften
measured in dry form - that is, the once-living
tissues of an organism with the water
extracted]. These organisms are the primary
producers. The amount of carbon fixed is
referred to as gross primary production and
is typically measured in grams (gl of carbon
ICl per unit of space larea or volume) per unit
of time.

The photosynthetic producers also respire
to meet their own energetic needs. Under
some circumstances, respiration of photo-
synthesizers over a given period may balance
their carbon fixation, so that there is no net
accumulation of organic carbon. More nor-
mally, however, there is a surplus of fixation
over respiration, so that organic matter is

accumulated over time. This accumulation is
referred to as net primary production iNPPl.
The accumulated matter is available to the vast
suite of organisms of all sizes, including
humans, that cannot synthesize their own
organic compounds from an inorganic base or
harness energy from inorganic sources. Such
organisms are referred to as heterotrophs,
while photosynthesizers and the few l<inds of
microorganisms that use other energy sources
to synthesize organic compounds are referred
to as autotrophs.

Organic products pass
through the food chain,
through processes such
as predation.

Food webs

An organic product produced by a photo-
synthesizer may pass through a number of
heterotrophs before finally being brol<en
down again to its inorganic constituents.
Conventionally this can be viewed as a food
chain. At macroscopic level, a green plant
may be eaten by a herbivore - a grasshopper,
say - which is eaten by a lizard, which is
itself eaten by a hawk, which dies and is
disassembled and partially consumed by
animal scavengers, with the remainder
decomposed by bacteria and fungi.

In reality, this is an enormous over-
simplification. The plant will almost certainly
have a complex network of symbiotic fungi
associated with its roots, which make
use of some of the gross production of the
plant but which also provide it with some
essential nutrients. The plant itself may shed
leaves which are directly broken down by
other fungi, protoctists such as slime molds.

8 WORLD ATLAS OF BIODIVERSITY

Map 1.2

Primary production In the

biosphere

Global spatial variation in
annual net primary
production INPPI, in g C
per m" per year, calculated
from an integrated model
of production based on
satellite indices of
absorbed solar radiation.

Source: Map created (rom data
supplied by Chrts Field and George
Merchant, Department of Global
Ecology, Carnegie Institution of
Washington. See

http://jaspenstanford.edu/chrisweb/flab/
flab, html, and Field et 3l. ,

g C per m' per year

1 782

-3 859

1 107

- 1 781

881 -

1 106

671 -

880

487 -

670

3A1 -

486

230-

340

U3-

229

61 -

142

0-60

and many forms of bacteria. The grass-
hopper is likely to be parasitized by a host of
smaller organisms, some of which are
themselves in turn parasitized. It will also
support a host of benign microorganisms in
its intestine that are themselves constantly
growing and reproducing. The lizard may die
and decompose and the hawk may eat the
grasshopper directly. The overall pattern of
feeding relationships thus forms a web of
immense complexity in any but the simplest
ecosystems.

Each organism in the food web respires,
releasing energy which is eventually dissi-
pated in the form of heat, carbon dioxide and
water At each stage, therefore, some carbon
is returned to the inorganic part of the
carbon cycle. In addition, all living organisms
produce waste products, some of which are

incompletely metabolized organic compounds.
Heterotrophic organisms are also not
completely efficient in their appropriation of
the organic material they consume, so that
some proportion of this is excreted as waste
product. These organic wastes are theor-
etically available to other organisms in the
food web. The assimilation efficiency of
heterotrophic organisms may be anything
from 20 percent (in the case of some
terrestrial herbivores! to 90 percent (in
the case of some carnivores), with the
remainder excreted.

Of the amount assimilated, a high
proportion is expended as respiration, with
the remainder available to add biomass, i.e. to
enable the organism to grow and reproduce.
The proportion available to add biomass is
dependent on the organisms involved as well

The biosphere 9

'jr

./ " ^ *

as a range of other factors. It can be as low as
10 percent or less and as high as 50 percent
or more. This proportion is a measure of the
net growth efficiency of the organism.

For purposes of ecological analysis,
particularly involving productivity estimates,
the gross growth efficiency is the most
commonly used measure. This is simply the
product of the assimilation efficiency and
the net growth efficiency of a particular
heterotroph and is a measure of the pro-
portion of food consumed by that organism
that, after excretion and respiration, is
ultimately available for its growth. As a
very coarse generalization, a value of
10 percent is widely used, although it is
acknowledged that in terrestrial herbivores
the figure is likely to be lower and in
planktonic communities and terrestrial

V ,

carnivores it is likely to be higher. Using the
figure of 10 percent in the example above, for
every kilo of plant matter eaten by the
grasshopper, the latter would add 10 grams
to its body weight. When the grasshopper was
eaten by the lizard, this would add 1 gram to
the lizards body weight, and when the lizard
was eaten by the hawk, this would add
0.1 grams to the hawks weight. This explains
why, at the species level, so-called higher
predators are rarer than herbivores and in any
given area have a lower biomass, while the
biomass of primary producers exceeds that of
all heterotrophs combined.

Measures of local and global productivity

Primary productivity varies enormously, both
spatially and temporally, at all scales. Most ob-
viously, under natural conditions productivity

10 WORLD ATLAS OF BIODIVERSITY

Table 1.1

Global annual net primary

production

Note: This represents one
among several attempts to
estimate global production;
see text for further details.
Ocean data averaged 1978-
83, land 1982-90, units in
petagrams II Pg = 10"g).

Source: Adapted Irom Field et al .

Biosphere units

NPP

lx10"gCl

Ocean

48.5

Terrestrial

56.4

Tropical rainforest

17.8

Deciduous broadleaf forest

1.5

Broadleaf and needleleaf forest

3.1

Evergreen needleleaf forest

3.1

Deciduous needleleaf forest

1.4

Savannah

16.8

Perennial grassland

2.4

Broadleaf shrubs with bare soil

1.0

Tundra

0.8

Desert

0.5

Cultivation

8.0

effectively ceases every night. Seasonal
variations In most parts of the world are also
marked. Productivity is, however, difficult to
measure, so that estimates at all scales are
subject to considerable uncertainty

On land one major source of uncertainty is
below ground productivity: In natural eco-
systems, less than 20 percent of plant produc-
tion Is typically consumed by herbivores The
remainder enters the soil system, either
through the plant roots or as leaf litter
Measuring this portion of terrestrial produc-
tivity - probably over 80 percent of the total -
Is particularly problematic. In the past there
has been a marked tendency to under-
estimate it. Similarly, It had long been
assumed that the nutrient-poor waters of the
open ocean were extremely unproductive, but
It Is now known that large populations of
extremely small photosyntheslzing unicellular
organisms - the so-called picoplankton - form
the basis of a surprisingly productive eco-
system in these regions.

One approach to estimating global pro-
ductivity Is based on measurements in
particular ecosystems and extrapolation from
these using estimates of the global extent of
those ecosystems. This suggests that net
primary production on land is of the order of
^5-65 X 10'= g C per year (that is, 45-65
petagrams or thousand million metric tons),
and at sea Is around 51 x 10'^g C per year This

gives a global estimate of annual net primary
production in the order of 100 x 10'^ g C. Gross
primary production is estimated to be about
twice this.

However, global measures using a
somewhat different technique, involving
assessments of relative concentration of
oxygen Isotopes, have Indicated that annual
gross primary production on land may be
greater than 180 x 10'* g C, while that in the
sea Is around 140 X 10'* g C. This would give a
global figure of over 320 x 10'* g C, implying
global net primary production of more than
160 X 10'* g C. This global figure Is 60 percent
higher than those global estimates based
on summation of Individual ecosystem
measurements.

Another approach' has used a com-
prehensive set of satellite indices of
photosynthetic activity In the ocean and on
land, combined with a model of primary
production, to generate a more integrated
global estimate of NPP. There is considerable
spatial and temporal variability, but on
average annual NPP on land amounts
to around 56 x 10'* g C. while that in the sea is
around 48 x 10'* g C Isee Table 1.11.

The carbon cycle and global blomass
estimates

Carbon fixation by photosynthesis forms one
crucial step in the carbon cycle*. Once fixed,
the carbon will remain for a greater or lesser
period within living tissues, that Is, form part
of the planets biomass. As all cells and
individual organisms have a limited lifespan,
eventually the carbon will rejoin the non-
living carbon pool (see Table 1.2). It may,
however, remain as organic carbon com-
pounds for a far greater period than It
remained part of the blomass; e.g. the woody
tissues of Paleozoic forests were formed
several hundred million years ago but remain,
fossilized, as a source of coal and oil.
Eventually all carbon will recycle through the
inorganic pool, as carbon dioxide in the
atmosphere, in the soil or dissolved In the sea,
or as inorganic carbon compounds (car-
bonates) In rocks or dissolved in the sea.

The great majority of carbon at any one
time lies within the lithosphere. around

The biosphere n

80 percent as carbonate and the remainder as
organic carbon compounds. A large propor-
tion of this carbon is effectively inaccessible to
the biosphere in the short term but itself
participates in the overall carbon cycle,
mainly through tectonic activity As seafloor
crust is gradually consumed along subduction
plate margins, carbon sediments are taken
into the Earth's mantle and later released as
carbon dioxide by volcanic and hydrothermal
activity. Although the loss of carbon to the
mantle is extremely slow, without volcanic
activity tectonic processes would eventually
exhaust the available carbon pool.

Most of the carbon incorporated in living
organisms is associated with green plants,
and almost all of this is in the form of
cellulose-rich woody tissues. The total
terrestrial animal biomass appears to be
Insignificant in comparison, probably more
than two orders of magnitude less. The
cyanobacteria and algae that are the primary
producers In the ocean are estimated' to
amount to only 0.2 percent of the biomass of
all primary producers globally, although they
generate in the region of half the global NPP
Ithey cycle organic material much more
rapidly than land plants, which also sequester
large amounts In woody tissues).

The role of diversity In the biosphere

The biota play the pivotal rile In the major
blogeochemlcal cycles, with different groups
of organisms (e.g. nitrogen-fixing bacteria
and photosyntheslzing plants] mediating
different processes. Simplistically, therefore,
at least some biological diversity Is necessary
to maintain the biosphere as It currently
operates, and microbial diversity is partic-
ularly fundamental. However, just how much
diversity Is needed, and how much redun-
dancy. If any, is built Into the system, remains
unclear. Indeed, the relationship between
biological diversity and a whole suite of
ecological measures including stability,
resilience and productivity remains incom-
pletely understood', although there Is an
increasing volume of theoretical and ex-
perimental work" indicating that diversity
may play an Important role In long-term
ecosystem functioning.

Human Influence on the biosphere

There are believed to be more than 6 billion
humans on the planet at present. A significant
proportion of global net primary production is
diverted to support this population. Using a
relatively conservative definition of appro-
priation that takes account of the global
agricultural and natural production used by
humans', that proportion has recently been
estimated at around one third of the
terrestrial global total'". This result is very
similar to that obtained in an earlier study",
but because of uncertainties in almost alt the
figures on which it and the earlier estimate
were based, the margin for error is very high.
Less conservative definitions also attempt to
take into account the loss of overall net
primary production that may result from
human actions (e.g. through severe land
degradation or accumulation of waste] but
which IS not directly used by humans. This Is
done by estimating the net primary pro-
duction of what Is thought would have been
the prevailing biomes in the absence of
humans. A detailed analysis In one European
country using this approach" " suggested that
around 50 percent of NPP there was
appropriated by humans.

Human efforts to appropriate the pro-
ducts of photosynthesis and other actions

Total carbon content on Earth

Amount buried in sedimentary rocl<S: organic
carbonate

Active carbon pool near surface;
of which

Dissolved inorganic carbon in sea

Atmospheric CO2

Organic carbon in soil

Biomass on land

Biomass in the sea

associated with the development of complex
societies have had enormous impacts on
natural biomes and blogeochemlcal cycles.
Over large areas of the Earth's surface,
humans have replaced complex and species-
rich natural habitats with simplified modified
habitats specialized for agricultural production.

1.6xlO"g
6.5xl0"g

40 000xlO'=g

38 000xl0"g

750xl0"g

ISOOxlO'^g

560x10''g

5-10x10'*g

Table 1.2

Estimated global carbon

budget and biomass totals

Note: Biomass figure on land
refers to plants.

Source; Adapted from Schlesinger .

12 WORLD ATLAS OF BIODIVERSITY

Clearance by fire, burning of fuelwood and
charcoal, soil cultivation, and fossil fuel use all
increase movement of organic carbon Into the
atmosphere. Global cycling of nitrogen,
phosphorus and sulfur has also been per-
turbed. Application of industrially produced
fertilizer has doubled the rate at which nitro-
gen in fixed form enters the terrestrial cycle,
and industrial processes have doubled
movement of sulfur from the lithosphere into
the atmosphere. Increasing levels of nitrogen
and phosphorus lead to shifts in nutrient
availability which can cause radical change in
natural communities, and sulfur is a major
contributor to acidification phenomena.

That human activities may have profound
local impacts on natural biota Is Indisputable.
What IS now becoming clear is that these

REFERENCES

activities may also have planet-wide impacts,
particularly on climate. Analysis of atmos-
pheric samples trapped in polar ice cores
indicates that present-day concentrations of
atmospheric carbon dioxide and methane
(CH^l are unprecedented in the past 420 000
years". Although their absolute concentration
In the atmosphere Is low (CO2 around 360
and CH^ at 1,7 parts per million by volume)
these two gases play an extremely important
role in determining atmospheric temperature.
It IS indisputable that the rise In these gases
is a result of human activities, so clearly
these activities are having some Impact on
global climate. The extent of this impact,
particularly when compared with natural
climatic fluctuations, remains a subject of
great controversy.

1 Hutchinson, G.E. 1970. The biosphere. Scientific American 233(31: 45-53.

2 Lenton, T.M. 1998. Gala and natural selection. Nature 394: 439-447.

3 Parkes, R.J. 1999. Oiling the wheels of controversy. Nafure 401: 644.

4 Field, C.B. e( at. 1 998. Primary production of the biosphere: Integrating terrestrial and
oceanic components. Science 282: 237-240.

5 Schleslnger, W.H. 1997. Biogeoctiemistry: An analysis of global cfiange. 2nd edition.
Academic Press, San Diego and London.

6 Schulze, E.D. and Mooney, H.A. (edsl 1993. Biodiversity and ecosystem function. Springer,
Berlin.

7 Grime, J. P. 1997. Biodiversity and ecosystem function: The debate deepens. Science 277:
1260-1261.

8 t«1cGrady-Steed, J., Harris, P.M. and t«1orin, P.J. 1997. Biodiversity regulates ecosystem
predictability Nature390: 162-165.

9 Wright, D.H. 1990. Human impacts on energy flow through natural ecosystems, and
Implications for species endangerment. Ambio 19(4): 189-194.

10 Rojstaczer, S., Sterling, S.M. and Moore, N.J. 2001. Human appropriation of
photosynthetic products. Science 294: 2549-2552.

11 Vitousel<, P.M. et al. 1986. Human appropriation of the products of photosynthesis.
BioScienceZb: 368-373.

12 Haberl, H. et al. 1999. Colonizing landscapes: Human appropriation of net primary
production and Its Influence on standing crop and biomass turnover In Austria. IFF-Social
Ecology Papers No. 57. Institute for Interdisciplinary Research of Austrian Universities,
Vienna. Also see: http://www.cloc.org/conference/presentations/ln4/npp-abstract.htm

13 Schulz, N. 1999. Effects of human land-use on the amount of biologically available
blomass-energy throughout the landscape. An empirical case study about Austria.
Conference paper available online in pdf at:

http://vwvw.univle.ac.at/lffsocec/conference99/htmlfiles/blue.html (accessed January
20021.

14 Petit, J.R. et al. 1999, Climate and atmospheric history of the past 420,000 years from the
Vostok Ice core, Antarctica. Nature399: 429-436.

The diversity of organisms 13

2 T

he diversity of organisms

g-^ YSTEMATICS AIMS TO DEFINE SPECIES and sort them into a hierarchy of named groups
^^^ congruent with the branching pattern of evolution. There is no single operational
definition of what a species is, and taxonomy at all levels is subject to change as a result
of new methods and data, but species diversity of better known organisms can often be
assessed with useful accuracy. Globally, about 1.75 million species have been described and
named, but the total including undescribed species might be up to ten times greater.

All species known are assigned on the basis of shared patterns of form and function to one
of about 100 major groups (phyla). There are marked differences between phyla in overall
morphology, physiology and mode of life. These differences imply the existence of major
genetic diversity, and contribute directly to structural, trophic and other dimensions of
diversity within ecosystems.

The phyla of living organisms fall into three primary lineages: the true bacteria, the
archeans and other organisms. The first two are prokaryotes, the remainder (protoctists,
animals, fungi and plants! are eukaryotes.

EVOLUTION AND SYSTEMATICS

A basic principle in evolution is that just as new/
individuals arise from ancestral individuals, so
new populations arise from existing popu-
lations, and ultimately new species arise from
existing species. The chief mechanism by
which this occurs is believed to be reproductive
isolation. For example, physiographic or
climatic change may divide an existing single
population into two or more separate popu-
lations, or individuals may colonize a new and
geographically separate habitat. The genetic
makeup of these isolated populations will
diverge, mainly through natural selection
acting on them, but probably also through
other mechanisms. This genetic divergence
may be manifested in various ways, physically,
physiologically and behaviorally. If the period of
isolation continues for long enough, the
populations will diverge enough that they can
be regarded as separate species. Each one of
these species may itself in turn give rise to
other species in due course, although some
will die out without giving rise to any progeny.
The surviving descendant species may

themselves give rise to new species and so on
through the long march of evolutionary time.

The result of this process is a branching
tree-like structure - a phylogenetic tree -
rooted in the distant past. In archeans and
bacteria, but probably rarely in other groups,
elements of two separate branches may
combine into one, giving a reticulate rather
than exclusively tree-lii<e pattern. If it is
assumed that all life on Earth had a common
origin in the distant past (see Chapter 3), then
all existing organisms form the topmost
extremities of a vast and unimaginably
complex single phylogenetic tree.

Systematics has two roles' '. The first is
to name the immense variety of different
sorts of organisms that exist. The second is to
try to elucidate the relationships between all
these different organisms, that is to develop
hypotheses of where they are positioned in
the phylogenetic tree. Systematics provides
the basic framework for the whole of biology,
and IS a fundamental discipline for bio-
diversity studies. Taxonomy is the subset of
systematics that deals in particular with the

u WORLD ATLAS OF BIODIVERSITY

Relationships between
organisms have to be
inferred from their
genetic, morphological,
biochemical or behavioral
characteristics.

definition, naming and classification of
species and. in some cases, subspecitic
populations. The traditional output consists of
species descriptions or revisions, or lists of
species in a given group (possibly with
hypotheses of their evolutionary relation-
ships), or checklists of all the species in some
higher taxon in a site or region.

Because the actual evolutionary events that
generated the overall phylogenetic tree are lost
in history, the relationships between organisms
have to be inferred from the evidence to hand.
The most important forms of evidence are the
characters of organisms, both living and fossil.
These characters may be genetic, morpho-
logical, biochemical or behavioral. Methods to
reconstruct phytogeny generally use two
working assumptions: that species sharing a
large number of characters are likely to be
related, and that species sharing some
uniquely complex and specialized feature are
likely to be more closely related than species
not possessing this feature.

Groups and names

In the current system for naming species
(nomenclature), each has a two-part sci-
entific name (binomial], based on Latin or
latinized Greek, comprised of the genus name
(e.g. Vipera] and specific epithet (e.g. berus].
The author of the specific epithet may be
given after the binomial. By convention, both
parts of the binomial are italicized when
printed, and the author name is shown in
parentheses if the species was originally put
into a different genus.

Similar species (e.g. the European adder
Vipera berus and asp viper Vipera aspis\ are
grouped together in the same genus (Wpera),
similar genera in families (Viperidae), families
in orders (Serpentes), orders in classes
(Reptilia) and classes in phyla (Craniata or
Vertebrata) up to the highest level, the
kingdom. An organism can only be assigned to
a single species, genus, family, etc., and the
taxonomic system forms a hierarchy with each
lower taxonomic level being nested entirely
within each increasingly inclusive higher level.
Although the traditional Linnaean hierarchy
includes only the seven obligatory categories
above, intermediate categories are some-

times used, and a further more inclusive
category - the domain.

Groups such as mammals or snakes, that
because of shared unique characters are
considered to contain all the living descen-
dants of a common ancestor, are called
monophyletic groups. Groups with no shared
unique characters but only unspecialized or
non-unique characters in common, are
termed paraphyletic groups. These typically
are groups of related species left over after
one or more clearly monophyletic lineages
that evolved within the group have been
recognized and named. Examples are fishes
(the craniate vertebrates without the unique
features of tetrapods], reptiles (the amniote
vertebrates without the unique features of
birds or mammals), and the entire kingdom
Protoctista. Groups defined on characters
that appear to have evolved more than once
are called polyphyletic groups. Although the
goal of most systematists is to recognize only
monophyletic groups in order to be able to
retrieve evolutionary relationships from a
classification, many paraphyletic groups,
such as the three named above, continue to
be very widely used in practice.

Given a classification congruent with
phytogeny, the taxonomic hierarchy becomes
a device to store information on hypotheses
about evolutionary history. Because hypoth-
eses about relationships are always subject
to revision as new information becomes
available, or existing data are reinterpreted,
the taxonomy of species is not fixed. The fact
that species names are liable to change can
cause confusion, for example when, as is
commonly the case, conservation legislation
uses a name no longer current.

Species concepts and diversity
assessment

Despite the importance of 'the species',
there is no unequivocal and operational
definition of what a species is or how species
can be recognized' I There are several
definitions, differing in mainly theoretical
and often subtle ways, but much of the
existing body of systematic knowledge has
been built up around elements of 'the
biological species concept'. This defines a

The diversity of organisms 15

species as a population of organisms tliat
actually or potentially interbreed in nature,
and tliat are reproductively isolated by
morphological, behavioral or genetic means
from other such groups. It is, however,
applicable only to organisms where sexual
reproduction is the norm.

In most real cases, especially where all
the systematist has to hand is a collection
of preserved specimens, whether criteria
concerning reproductive isolation are met
or not cannot in fact be tested, but an
experienced worker will come to hold some
particular level of morphological or other
difference as deserving of species status.
Where there is good evidence from fieldwork
and geographic data attached to specimens
that two somewhat similar populations
occur in the same locality (i.e. are sympatric)
but maintain their differences, they may be
presumed not to interbreed, and will be
treated and named as species.

Different taxonomists will often use
different criteria for the same group of
organisms, so that one specialist may regard
a group of fundamentally similar populations
as a single species, whereas another will treat
each smaller distinct population as a separate
species. In the latter case, often associated
with the 'phylogenetic species concept', the
assumption is that each distinct lineage once
established on its own evolutionary course is
de facto a separate species. Many populations
that were formerly described at subspecies
level (i.e. somewhat distinct, but regarded as
part of a single polytypic species complex!
have subsequently been elevated to full
species in this way, particularly if geo-
graphically isolated. However, subspecies
have often been named on the basis of just a
few superficial features, not representative of
the overall pattern of variation within species,
and the formal subspecies category is now
much less used than in the past (particularly
among vertebrates).

Different taxonomic characters and
criteria are used to classify species in
different groups of organisms. Those used,
for example, to define species of fungi are
very different from those used to define
species of bird, and their application

demands taxonomists with relevant specia-
lized knowledge. Some organisms are diffi-
cult or logically impossible to accommodate
in any species concept involving criteria that
assume all individuals reproduce by out-
breeding, that is by sexual reproduction with
another individual. Many higher plants are
self-fertile or make use of various forms of
asexual reproduction; in the latter case all
individuals in a lineage are identical clones
(assuming a very low or non-existent
mutation ratel. Among the prokaryotes,
bacteria can readily receive extraneous genes
through direct entry of genetic material from
the fluid environment, or from viruses or
other bacteria', and this horizontal' transfer
is independent of reproduction. Different
kinds of bacteria have traditionally been
defined by the cytological or biochemical
properties of colonies in culture but, more

Different taxonomic
characters and criteria are
used to classify species in
different groups of
organisms.

16 WORLD ATLAS OF BIODIVERSITY

recently, comparison of DNA and RNA
sequences from sample collections has been
used to distinguish one lineage from another
In either case, the biological species concept,
developed with reference to sexually
reproducing, outbreeding animals and plants,
is not appropriate. Such factors mean that 'the
species' cannot provide a standard unit in
which to evaluate all biodiversity because it
does not define a single level in the hierarchy
and its significance is not equivalent across all
groups of organisms; it can, however, serve
this purpose for most plants and animals.

The use of different species concepts by
different systematists can make a very
large difference to the number of species
recognized in a group and to complications
in nomenclature (these will both affect the
outcome of biodiversity inventory, an

Box 2.1 New species discoveries

The discovery of entirely new species of mammals and birds is rare, and often involves small
and obscure forms. Remarkably, two large mammals previously unknown to science were
discovered in one small area, the Vu Quang Nature Reserve in Truong Son, Viet Nam (along
with many new species in other groups!. The Vu Quang ox or soala Pseudoryx nghetinensis
was described in 1993, followed a couple of years later by a giant muntjac deer
Megamuntiacus vuquangensis from the same area. The world's smallest muntjac deer, the
Truong Son muntjac Munt/acus truongsonensis, was recently found in another part of the
same region in Viet Nam. The soala is of particular interest because it does not appear to fit
neatly in any of the main bovid groups as currently recognized. It is now known also to occur
in adjacent parts of Laos. However, claims that another new bovid species existed in
Southeast Asia, described as Pseudonovibos spiralis, were premature because the
distinctive horns described as new were later shown to be domestic cattle horns, apparently
reshaped artificially

important application of systematicsl.
Usually, however, only a small or very
small number of taxonomists is working on
any one group of organisms at any time, so
that while the species level taxonomy of
organisms is in a continual state of flux, it
is generally not subject to radical and
wholesale change. Exceptions arise where
new techniques, chiefly molecular infor-
mation and DNA data in particular, are
applied to previously neglected groups or
ones that have not been revised for many
years. Nevertheless, the key point appears

to be that units corresponding more or less
closely to the biologists' model of the
species' do indeed exist in nature and, in
animals in particular, they define themselves
to an extent through their reproductive
behavior It has thus been possible to reach
some measure of consensus on species-level
classification of well-studied groups of larger
organisms such as terrestrial vertebrates,
and to estimate and compare the number and
kinds of species in different sites, areas or
countries.

NUMBERS OF LIVING SPECIES

From a practical point of view it is more
important to know how many species, and
which ones, occur in some spatially restricted
area, such as a protected area or a country,
than in the world overall. However, proper
evaluation of each local situation requires
some knowledge of the wider context and.
where the goal is maintenance of global
biodiversity in the face of increased risk, it is
clearly important to have a sound appreci-
ation of the full baseline range of diversity.
This requires both an estimate of the number
of known valid species, and an estimate of the
number of unknown species, neither of which
is readily available.

The number of known species can be
estimated by collating data from systematists
and the taxonomic literature. Although many
species names are synonyms (i.e. different
names inadvertently applied to the same
species), this can be done with reasonable
precision for more familiar and well-reviewed
groups of species. Recent calculations of this
kind suggest that around 1.75 million of the
probably far larger number that exist have
been discovered, collected and later named
by systematists".

Any estimate of how many undiscovered
and hence undescribed species are likely to
exist in any given group, and in the biosphere
overall, involves substantial uncertainty''. In
taxonomic groups where individuals are
readily visible, popular or economically
important, and subject to sustained system-
atic attention, e.g. mammals and birds, the
number of known species is certainly very
close to the total number of species in the

The diversity of organisms 17

group that exist. On average around 25 new
species of mammals and and five of birds
have been described annually in recent years'.
Changing systematic opinion on which
populations should be regarded as separate
species and which should not, rather than
completely new discoveries, is the major
source of change in the number of named
species in such groups.

The converse applies to groups whose
individuals are small, difficult to collect,
obscure and of no popular interest, e.g. many
groups of invertebrate animals. Frequently
there are so few systematists actively working
on a group that the number of named species
appears to be limited mainly by the rate
at which collected specimens waiting on
museum shelves can be studied and
described, and changing opinion on which
populations are separate species is insig-
nificant. In some cases, where new sampling
and collection methods have been used,
unexpectedly large numbers of new species
have been found (e.g. tropical forest canopy
insects and marine sediment nematodes). If
findings from such local work are extrapolated
to global level the total number of species
calculated to exist is many orders of mag-
nitude greater than the number actually
known. Some estimates suggest that most
undescribed terrestrial forms are likely to be
tropical forest beetles, but new molecular
techniques are revealing unsuspected divers-
ity among microorganisms'.

Although the goal of systematics is to
recognize and name species, and to maintain
an ordered body of information on names and
associated biological data, there is no master
catalog of all known species. Developing such
a resource has only become feasible with
advances in information technology during
the past ten years. However, while many
systematic data, in the form of checklists and
museum catalogs, are now available in digital
form over the Internet, and more will become
so, a harmonized catalog in this format of all
known species remains a distant prospect.

Recent estimates of the numbers of known
and possibly existing species in the world
biota are given in Table 2.1. These are mostly
large numbers, and the fossil record suggests

that overall diversity has been increasing for
some 600 million years up to the very recent
past, but the numbers themselves are of little
significance except in a wider context.
Currently, much concern is focused on
species numbers in relation to anthropogenic
environmental change.

Box 2.2 Improving taxonomic knowledge and capacity

The Convention on Biological Diversity decided in 1998 to establish a Global Taxonomic
Initiative IGTII", in recognition of a major 'taxonomic impediment' to effective biodiversity
management. The objective is to improve decision-making for conservation, sustainable
use and benefit-sharing by increasing taxonomic knowdedge and the number of trained
taxonomists and curators. Similarly, the Global Biodiversity Information Facility IGBIFj"
aims to develop an interoperable net«/ork of biodiversity databases and information
technology tools that will enable users to access the viforld's stores of biodiversity
information. In the same field. Species 2000'- (a global network based in the UK and
Japan], and the Integrated Taxonomic Information System IITISI" in North America have
joined forces to create a unified Catalogue of Life, planned to cover all known species of
living organisms. Basic reference data on 250 000 species had been collated by mid-2001,
and the plan is to reach 500 000 by 2003.

DIVERSITY AT HIGHER LEVELS

Until van Leeuwenhoek observed micro-
organisms through a primitive microscope in
the late 17th century, humans had been
aware only of organisms visible to the naked
eye (macroscopic! and regarded all living
things as either plants or animals. At the
end of the 19th century, with improved
cytological techniques and new views on
evolution, a third kingdom of organisms
(Protista) was recognized for bacteria and
other unicellular organisms. Around this
time it became widely accepted that the cell
was the fundamental unit of organization of
all living organisms. Subsequent work, in the
mid-20th century, recognized a basic dis-
tinction between two kinds of cellular
organization - prokaryotic and eukaryotic. In
prokaryotic organisms, the genetic material
is free within the cell. In eukaryotes, the
genetic material is linked to proteins and
organized into chromosomes that are packed
within a membrane-bounded cell nucleus.
There are several other profound differences.
In eukaryotes the enzymes needed to extract
energy from organic molecules are organized

18 WORLD ATLAS OF BIODIVERSITY

Table 2.1

Estimated numbers of
described species, and
possible global total

into discrete membrane-bounded organelles
Imitochondna) within the cell, and in the
eukaryotes that photosynthesize (plants,
some protoctistsi the pigments and enzymes
needed to fix solar energy are also in
discrete organelles Ichloroplasts) within the
cell. On the basis of this, it became accepted
that the major taxonomic divide in organisms
was between the prokaryotes, containing
only the bacteria, and eukaryotes, which
included all other organisms, including
many unicellular forms included in the
original kingdom Protista.

Systematics has traditionally generated
information on similarity and hypotheses of
relationship on the basis of characters
restricted to particular sectors of the
phylogenetic tree, but molecular sequencing
has broadened the range of useful evidence.

with the potential for radically dissimilar
groups to be compared and their phylogeny
estimated. Analysis of small subunit ribo-
somal RNA ISSU rRNAl - a molecule that is
universal, functionally constant (central to
protein manufacture in all cells! and very
highly conserved over time - has proved
especially informative. This work has revealed
that there are two kinds of prokaryote: the true
bacteria and the archeans. The archeans were
first known only by extremophiles', i.e. forms
living under exceptional conditions of high
temperature or salt concentration, but repre-
sentatives are now known to be widespread
alongside bacteria in less extreme habitats. At
biochemical level, the bacteria and archeans
are as different from each other as from
eukaryotes, leading to the conclusion that all
organisms can be assigned to three basic

Domain Eukaryote
kingdoms

Arcliaea

Bacteria
Eukarya

Animalia

Craniata (vertebrates!, total

Mammals"

Birds"

Reptiles'*

Amphibians"

Fishes"
Mandibulata (insects and myriapods!
Chelicerata (arachnids, etc.]
Molluscs
Crustacea
Nematoda

No. of

Estimated

described

total

species

175

10 000

1 320 000

52 500

55 000

4 630

9 750

8 002

i950

25 000

963 000

8 000 000

75 000

750 000

70 000

200 000

40 000

150 000

25 000

400 000

Fungi

72 000

1 500 000

Plantae

270 000

320 000

Protoctista

80 000

600 000

TOTAL

1 750 000

U 000 000

Notes: This table presents recent estimates of the
number of species of living organisms in the high-
level groups recognized, and in some selected
groups v^ithin them. Vertebrate classes are
distinguished because of the general interest in
these groups. The described species column refers
to species named by taxonomists. Most groups lack
a formal list of species. All estimates are
approximations. They are inevitably inaccurate
because new species wiill have been described since
publication of any checklist and more are
continually being described, and other names turn
out to be redundant synonyms. In general, the
diversity of microorganisms, small-sized species,
and those from habitats difficult to access, are likely
to be seriously underestimated. Among Archaea and
Bacteria the figures of 175 and 10 000 are very
rough estimates of 'species' defined on features
shov^n in culture"; there appears to be no sound
estimate of the total amount of prokaryote diversity.
The estimated total column includes provisional
vKorking estimates of the number of described
species plus the number of unknown species; the
total figure is highly imprecise. Only a small
selection of animal phyla is shown, but the figure for
Animalia applies to all. Figures in the total row are
for all species in all domains.

Source: Data mainly from United Nations Environment
Programme^ and Hammond^; vertebrates from individual
sources indicated.

The diversity of organisms 19

BACTERIA

ARCHAEA

EUKARYA

Animals

forms of life, or domains: Archaea, Bacteria
and Eukarya, rattier than two lprol<aryotes and
eukaryotesl. The distance between lineages.
In terms of amount of change in rRNA
sequence, and their branching sequence In
evolution, have been represented as a
'universal phylogenetic tree'" ".

Where the root of the tree Is located, and
Its basal branching pattern, i.e. which two
of the three domains are more closely
related than either Is to the third, remains
open to discussion"". It has been argued'"
that this deep branching took place in very
early evolution, possibly before modern cell
types had been established, when genes
were widely spread by horizontal transfer
las persists to a lesser extent among
extant prokaryotesl and in a sense, shared
communally.

Reticulate evolution would have been
common, with distinct lineages of organisms
emerging only when the evolving cell became
sufficiently integrated that horizontal gene
transfer was reduced and vertical transfer
down closed ancestor-descendent series
became the norm. It has also been per-
suasively argued that the Eukarya arose
through the permanent symbiotic fusion of a
number of different prokaryotic organisms In
one cell, with organelles such as mitoc-
hondria and chloroplasts representing the
vestiges of different lineages of formerly
independent prokaryotes' '".

PROTOCTISTS

Within the Eukarya, the fungi, animals and
plants form Insignificant clusters over-
shadowed by the morphological and physio-
logical diversity of the protoctists Iprotistsl.
Protoctista is a name of convenience for the
enormous and very diverse collection of all
the small-to-mlcroscopic eukaryotes that
lack the distinguishing features of fungi,
animals or plants. Further changes of pers-
pective are expected as research on protoctlst
systematics continues. For example, a recent
tendency^' Is to recognize the Chromista as
another major group distinct from remaining
eukaryotes. The chromists are aquatic
species distinguished by structural and
biochemical characters, and include organ-
Isms with the largest linear dimensions
known Ikelpl, as well as microscopic but
ecologically important organisms (e.g.
diatoms, downy mildews).

The only known organisms that are not
cells, or assemblages of cells, are viruses. They
exist on the very boundary of most definitions of
life. Consisting only of nucleic acids and protein,
they are much smaller than the smallest
bacteria, they can only replicate Inside other
living cells, and they are totally inert outside
other cells, when they can survive for years in a
crystallized state. Each type of virus may be
more closely related to the organism in which It
grows than to other viruses'. They are not
discussed elsewhere In this book.

Some characters of the high level group-

Figure 2.1

The phylogenetic tree

Note: This diagram
represents in highly
simplified form the distance
between the three domains
of organisms and the
general branching pattern
within them; because of
conflicting interpretations
the root and branching
sequence of the three
domains are not
represented.

Source: Adapted from Woese .

WORLD ATLAS OF BIODIVERSITY

Table 2.2

Key features of the major

groups of living organisms

Note: The four groups within
the Eukarya are commonly
regarded as kingdoms, the
highest formal category of
the Linnaean system of
taxonomy The Archaea and
Bacteria have been treated as
a single group, or as separate
kingdoms, but a more recent
tendency is to treat Archaea,
Bacteria and Eukarya as
three separate domains in
recognition of the deep
phyletic divergence between
them, and fundamental
differences in RNA
composition.

Source: Principally Margulis and
Schwartz . also University of California
and University of Arizona

Arcfiaea

Prokaryotic. Composition of the cell wall and of lipids in cell membranes differ from those in
Bacteria. Distinctive SSU rRNA, more similar in some respects to Eukarya than to Bacteria.
Reproduce asexually by cell splitting, or produce genetic recombinants without any fusion of cells by
accepting genes from other bacteria, or from the fluid medium, or through viruses, independent of
cell reproduction. Flourish in habitats with radical extremes of temperature or salinity that are
unavailable to other organisms lapart from bacteria in some casesi, but also occur in other
environments.

Bacteria

Prokaryotic. Reproduce asexually by cell splitting, or produce genetic recombinants without any fusion
of cells by accepting genes from other bacteria, or from the fluid medium, or through viruses,
independent of cell reproduction. Metabolically uniquely versatile; key mediators of major
biogeochemical cycles. Permeate the entire biosphere, including other organisms, although dominant
only in exceptional habitats.

Eukarya

Animalia

Fungi

Multicellular, mainly macroscopic, eukaryotes. Reproduce through fertilization cf an egg by a
sperm, the fertilized egg Inow diploid, i.e. a duplicate set of chromosomes] is called a zygote
and (except spongesl this forms a characteristic hollow multicelled blastula from which the
embryo develops. All heterotrophic.

Mainly multicellular, micro- to macroscopic eukaryotes. Fungi develop directly without an
embryo stage from resistant non-motile haploid (one set of chromosomes] spores that can be
produced by a single parent. Sexual reproduction also results in haploid spores. Most consist
of a network of threadlike hyphae. Heterotrophs, vital to decomposition processes; form
mycorrhizal symbioses with plants, facilitating exchange of soil nutrients.

Plantae

Multicellular macroscopic eukaryotes. The fertilized egg develops into a multicelled embryo
different from blastula of animals. Alternate spore-producing generations and egg or sperm-
producing generations. Virtually all are terrestrial photosynthetic autotrophs.

Protoctista

Mainly microorganisms. Possess the features of eukaryotes, but lack the characteristics
of fungi, animals or plants. Extraordinary variation in life cycle and morphology Early
evolution probably based on symbiotic relationships between different kinds of bacteria
forming lineages of composite organisms resulting in the protoctist grade of
organization. Include photosynthetic algae (formerly classed as plants] and heterotrophs
(formerly called 'protozoa'].

ings of organisms (domains and kingdoms] is presented in Appendix 1. The objective of

are outlined in Table 2.2. A synopsis of this material is to provide a convenient

information on eacfi of the 96 phyla recognized overview of global organismal diversity, in

in the most recent comprehensive synthesis' terms of higher taxon diversity.

The diversity of organisms 21

REFERENCES

1 Minelli, A. 1993. Biological systematics: The state of the art, Cliapman and Hall, London.

2 Vane-Wright, R.I. 1992. Systennatics and diversity. In: World Conservation Monitoring
Centre. Groombridge, B. led.) Global biodiversity: Status of the Earth's living resources,
pp. 7-12. Chapman and Hall, London.

3 Mayr, E. and Ashlock, P.D. 1991 . Principles of systematic zoology. McGraw-Hill Inc., New York.
A Gaston, K.J. led.] 1996. Biodiversity: A biology of numbers and difference. Blackwell

Science Ltd, Oxford.

5 Vane-Wright, R.I. 1992. Species concepts. In: World Conservation Monitoring Centre.
Groombridge, B. led.) Global biodiversity: Status of the Earth's living resources, pp. 13-16.
Chapman and Hall, London.

6 Margulis, L. and Schwartz, K.V. 1998. Five kingdoms. An illustrated guide to the phyla of
life on earth. 3rd edition. W.H. Freeman and Company, New York.

7 United Nations Environment Programme 1995, Heywood, V. (ed.l Global biodiversity
assessment. Cambridge University Press, Cambridge.

8 Hammond, P. 1992. Species inventory. In: World Conservation Monitoring Centre.
Groombridge, 8. (ed.l Global biodiversity: Status of the Earth's living resources, pp. 17-39.
Chapman and Hall, London.

9 Williams, D.M. and Embley, T.M. 1996. Microbial diversity: Domains and kingdoms.
Annu. Rev Ecol. Syst. 27: 569-595.

10 GTI. http://www.biodiv.org/programmes/cross-cutting/taxonomy/ (accessed January 20021.

1 1 GBIF. http://wviw.gbif.org/ (accessed January 20021.

12 Species 2000. http://vww.sp2000.org/ (accessed January 2002).

13 ITIS. http://www.itis.usda.gov/ (accessed January 2GG2I.

14 Woese, C.R., Kandler, 0. and Wheelis, M.L. 1990. Towards a natural system of organisms:
Proposal for the domains Archaea, Bacteria and Eukarya. Proceedings of the National
Academy of Sciences 87: 4576-4579.

15 Woese, C.R. 2000. Interpreting the universal phylogenetic tree. Proceedings of the
National Academy of Sciences 97: 8392-8396.

16 University of California, Berkeley, Museum of paleontology.
http://www.ucmp.berkeley.edu/alllife/threedomains.html (accessed January 20021.

17 University of Arizona, Tree of Life project.
http://phylogeny.arizona.edu/tree/phylogeny.html (accessed January 2002].

18 Margulis, L. 1998. The symfa/of/c p(anef; /\ new (oofc af ei/o/uf/on. Weidenfeld and
Nicolson, London.

19 Philippe, H. and Forterre, P. 1999. The rooting of the universal tree of life is not reliable.
Journal of Molecular Evolution 49: 509-523.

20 Woese, C.R. 1998. The universal ancestor Proceedings of the National Academy of
Sciences 9^3: 6854-6859.

21 http://www.ucmp.berkeley.edu/chromista/chromi5ta.html (accessed January 20021.

22 Woese, C.R. 1998. Default taxonomy. Proceedings of the National Academy of Sciences
95: 11043-11046.

23 Wilson, D.E. and Reeder, D.M. ledsl 1993. Mammal species of the world: A taxonomic and
geographic reference. 2nd edition. Smithsonian Institution Press, Washington DC and
London. Available (accessed January 2002] in searchable format at
http://nmnhwww.si.edu/msw/

Covers existing or recently extinct species, as of book publication year.

24 Clements, J.F. 2000. Birds of the world: A checklist. Pica Press, Sussex. Other lists are
available online, e.g. a list based on Sibley, C.G. and Monroe, Jr, B.L. 1990. Distribution
and taxonomy of birds of the world. Yale University Press, New Haven and London is
available (accessed January 20021 at http://www.itc.nl/~deby/SM/SMorg/sm.html

22 WORLD ATLAS OF BIODIVERSITY

25 http;//www.embl-heidelberg.de/-uetz/LivingReptiles.html (accessed January 2002).

26 Estimate is sum of species numbers for Anura 14 701 1, Caudata (505) and Gymnophiona
(caeciliansl (1591 retrieved from ontine resource Amphibia Web at
http://elib.cs.berkeley.edu/aw/lists/index.stitml (accessed January 20021. Also see
Duellman, W.E. 1993. Amphibian species of the world: Additions and corrections. Special
Publication No. 21, University of Kansas Museum of Natural History, Lawrence, Kansas,
and database with later revisions available online in searchable format at
http://research.amnh.org/herpetology/amphibia/ (accessed January 20021.

27 Eschmeyer, W.N. et ai 1998. A catalog of the species of fe/ies. Vo[s 1-3. California
Academy of Sciences, San Francisco. Available online in searchable format at
http://www.calacademy.org/research/ichthyology/catalog/fishcatsearch.html (accessed
January 20021.

Biodiversity through time 23

Biodiversity through time

Two FUNDAMENTAL PATTERNS CAN BE DISTINGUISHED in the fossil record. On one
hand, new groups of organisms appear, diversify and generally persist for very long
periods of time; on the other, most such groups and their Included species eventually
cease to exist. Most analyses of the fossil record show an erratic rise In overall biodiversity,
increasing through the Mesozolc and Cenozolc and reaching a peak around the end of the
Tertiary. However, diversity has been greatly reduced during several periods of radical
environmental change during each of which more than half the multicellular species then
living became extinct. Such mass extinction phases have provided Important new
opportunities for diversification in remaining lineages, and the spread of new communities.
Evidently, the large number of species existing now on Earth Is the result of a modest net
excess of originations over extinctions during the 3 800 million years of evolution of life.

THE FOSSIL RECORD

Knowledge of the history of diversity through
geological time is based on analysis of the
fossil record. When fossils, as the inert
mineralized parts or casts or imprints of dead
organisms, are interpreted in a biological
context they provide the only direct evidence
of the history of life on the planet.

The fossils discovered and described by
paleontologists represent more than a quarter
of a million species, virtually all of them now
extinct', but these are believed to make up only
a very small fraction of all the species that have
ever existed. For example, the fossil record of
marine animals is far more comprehensive
than that of terrestrial forms, but the marine
sample is estimated to represent only about
2 percent of all the marine animals that have
lived'. The fossil record overall may represent
as little as 1 percent, or less, of all the species
that have existed'. Clearly, statements about
broad patterns in the evolution of life, and the
ascendancy or extinction of groups of organ-
isms, thus rest on a very narrow base of
tangible evidence.

Macroscopic animals with hard skeletons
that lived in shallow marine environments,
where their remains could be buried by

sediment, petrified, and later exposed in
uplifted rock strata, are by far the most likely to
be both preserved and found. The fossil record
from the past 600 million years is thus
dominated by mollusks, brachiopods and
corals'. For many other kinds of organism,
exceptional circumstances are required if a
dead individual is to become preserved and
found. With some exceptions, microscopic
or soft-bodied organisms rarely leave discern-
able traces of their existence, and larger
organisms are usually decomposed, dis-
assembled and never discovered. Terrestrial
vertebrate fossils are often of individuals that
must have been preserved by the smallest of
chances, perhaps sudden burial during a
natural disaster, or as a result of the body falling
into a rock crevice out of reach of scavengers.

All else being equal, the probability of
preservation will rise greatly the more
widespread and abundant the species is, and
the longer it persists through time. Conversely,
there is a very low probability that any individual
of a numerically rare or restricted-range
species with a short persistence time will die in
circumstances conducive to fossilization and
subsequently be found. Factors such as these
mean that even the relatively better known

24 WORLD ATLAS OF BIODIVERSITY

Figure 3.1

The four eons of the

geological timescale

Arrows indicate
approximate age of oldest
confirmed fossils of the
groups named.

Source: Adapted from Mdrgulis and
Schwartz .

groups are certain to be very incompletely
l<nown. The plant and animal species now living
include a substantial number of rare or local
species and it is difficult to imagine many of
them being represented in the fossil record of
our time as recovered in the future^

Because the fossil record gives an
incomplete and biased view of the past history
of life, the reconstruction of that history has
been the subject of great debate. It has been
generally accepted that the record can give a
reasonable insight into past diversity in terms of
taxonomic richness", particularly at higher
taxonomic levels. On the other hand, recent
analysis of a new database of fossil occurrence
has indicated that sampling effects (reflecting
variation in the nature of the fossil record and
the way it is reported! can make a very
significant contribution to the shape of global
diversity curves. Preliminary results of this
analysis have suggested, for example, that the
widely accepted post-Paleozoic increase in
marine diversity Isee belowl may to some extent
be an artifact of the analytic methods used^

PATTERNS OF DIVERSIFICATION
The early history of life
In its earliest history, there was no life on the
Earth. Now, there are about 1.75 million
different species of all kinds known, with
perhaps many times that number still
unknown. Self-evidently, life has both arisen
and diversified.

The planet Earth is between 4 500 million
and 5 000 million years old. The oldest known
rocks are about 4 000 million years IMyl in
age, and mineral grains IzirconI of apparently
greater age have been reported. Carbon

chemofossils' suggest that life had evolved
by 3 800 My ago' but periodic intense ocean
heating caused by extraterrestrial impacts is
suspected to have restricted early organisms
to thermophilic non-photosynthetic forms,
possibly associated with hydrothermal
vents". Most of the principal biochemical
pathways key to the modern biosphere were
probably in place by 3 500 My ago'. The first
tangible evidence of cellular organisms
themselves in the fossil record consists of
filaments and spheroids, believed to be the
remains of prokaryote microorganisms, in
rocks of about this age, including traces from
an apparent Precambrian submarine thermal
spring system of about 3 235 My in age'.

Stromatolites [rock domes formed in
shallow waters from multiple layers of sedi-
ment and bacterial also first appear at around
3 500 My ago and are the most abundant fossils
so far known during the 3 000 My up to the start
of the Phanerozoic eon, marked by the start of
the Cambrian period around 5^5 My ago. For
much of this time, bacteria-like organisms
were the only known life forms; gross
morphological diversity was very low and many
kinds apparently persisted for hundreds of
millions of years, some outwardly indis-
tinguishable from existing forms. Probable
microbial mats, similar to stromatolites but
developed on an exposed soil surface, have
been interpreted as the first evidence of
terrestrial ecosystems around 2 600 My ago'".

The next major step in the evolution of life
was the development of eukaryotic organisms.
Biochemicals characteristic of eukaryotes
have been found in shales 2 700 My old", far
earlier than the first fossils, reported from

Eon

Millions of
years ago

HADEAN

4 500

origin

of Earth

4 000 3 900

3 500

3 000

t t t

oldest

'chemofossils'

oldest

Earth

possible

microfossils

rocks

evidence
of life

of bacteria

evidence of
eukaryotes

Biodiversity through time 25

shales of around 1 500 My age". These earliest
eukaryote microtossils Include a number of
'acritarchs' (widely interpreted as the cysts or
resting stages of marine algael some of which
show cytological features characteristic of
protoctists (see previous chapter]' . Diversity in
acritarchs, and the rate at which different
forms replaced one another in the record,
were low until around 1 000 My ago when
both species number and species turnover
increased markedly".

Radiations around the early
Phanerozoic boundary

For many years it was assumed that animals
originated in the Cambrian period at the base
of the Phanerozoic Ithe Phanerozoic is the eon
of time characterized by presence of animal
fossils; it includes the Paleozoic, Mesozoic and
Cenozoic). This is now known not to be the
case, as a wide range of fossil animals,
including recognizable arthropods and possibly
echinoderms, is now known from about 1 00 My
before the Cambrian. Most fossils from this
time, however, appear completely unrelated to
extant forms, and consist mainly of enigmatic
frond- and disc-shaped soft-bodied animals:
the so-called Ediacaran fauna.

The lower Cambrian marks a dramatic
change from this early fauna, with the sudden
appearance in the fossil record (e.g. the
Chengjiang and Burgess Shale faunas! of a
wide range of animals, many with calcareous
skeletons. It is generally accepted that this
represents a genuine explosion of diversity that
took place over only a few million years, and is
not an artifact of the fossil record. The lower
Cambrian thus represents the most important

period of high-level diversification In the his-
tory of animal life on Earth. These archaic
Invertebrates had by the end of the Cambrian
period, around 500 My ago, established all the
basic body plans seen in extant animals, and
many others besides. Each such basic lineage
Is recognized taxonomically at phylum level,
and the range of morphological diversity was
higher at 500 My ago than at any time before
or since. As many as 100 different animal
phyla may have existed during the Cambrian"
including every well-skeletalized animal phy-
lum living today (except perhaps the Bryozoa).
whereas in the latest synthesis all extant
animals are placed in 37 phyla'^

Plants and animals began to extend into
terrestrial habitats during the first half of the
Paleozoic, with the first fossil material known
from the late Silurian, around iOO My ago. At
this point approximately 90 percent of the
history of life to the present had already
passed. Fossils suggest low diversity tor the
next 100 My until the later Devonian period.
No new animal phyla appeared with the
colonization of land, millions of years after
the initial Cambrian radiation of animal phyla.

Diversity of marine animals in the Phanerozoic

The overall pattern of diversity [assessed as
numbers of families! appears to show a
possible early peak around the start of
Phanerozoic time, followed by a plateau of
somewhat higher diversity extending through
most of the Paleozoic era, and then, after the
end-Permian mass extinction (see below] a
steady increase in diversity over remaining
geological time'. Some recent methods of
analysis' giving special attention to sampling

2 000

PROTEROZOIC

1 500

oldest

protcctist

fossils

PHANEROZOIC

1 000

545 500

t ft

oldest oldest oldest

animal fungus plant

fossils fossils fossils

24 WORLD ATLAS OF BIODIVERSITY

Figure 3.2

Periods and eras of the

Phanerozoic

This IS an expanded version
of the most recent segment
of the geological timescale
shown in Figure 3.1 .
The Phanerozoic is the eon
of time extending from the
base of the Cambrian,
some 545 million years ago,
to the present, and to which
the entire fossil record was
formerly thought to be
restricted.

Source: Adapted from Margulis and
Schwartz , Cambrian base date from
International Subcommission on
Cambrian Stratigrapfiy website .

procedures suggest this broad view may be
subject to significant revision.

Although the number of phyla has
decreased markedly since the Cambrian,
diversity at all lower taxonomic levels appears
to have either increased overall or, in a few
cases, remained more or less level. The
number of orders of marine animals present
in the fossil record climbed steadily through
the Cambrian and Ordovician, slowing
towards the end of the Ordovician to a figure
of between 125 and 1A0, which has been
maintained throughout the Phanerozoic.

The diversity of marine families repre-
sented in the fossil record shows a similar
pattern of increase through the Cambrian
(possibly falling during the latter half of the
period! and Ordovician, leveling off at around
500, a figure which was maintained until the
late Permian mass extinction. This extinction
event resulted in the loss of around 200
families, but diversity increased subsequently
to the modern level of around 1 100 families,
with a number of temporary reversals during
minor extinction events. The trend in number
of species in the fossil record is even more
extreme. From the early Cambrian until the
mid-Cretaceous, the number of marine
species remained low; since then, that is
during the past 100 million years, it appears to
have increased dramatically, perhaps by a
factor of 10. (See Figure 3.3.)

Diversity of terrestrial organisms
in the Phanerozoic

Although low to moderate peaks and troughs
are evident in the record, the overall pattern
of family diversity in terrestrial organisms
shows a continuing rise from the Silurian to
the present, thus differing somewhat from

the pattern shown by many analyses of
marine animals.

It is generally accepted that vascular
terrestrial plants first arose in the Silurian,
although some paleobotanists argue for a late
Ordovician origin, a time when microfossils
suggest that bryophyte-like (non-vascularl
plants already existed". Diversity increased
during the Silurian, and then more rapidly
during the Devonian, owing to the first appear-
ance of seed-bearing plants, leading to a peak
of more than AO genera during the late
Devonian. Diversity then declined slightly, but
started to increase markedly during the
Carboniferous, with 20 families and more than
250 species in the mid-Carboniferous record of
the northern hemisphere. Following this,
diversity increased only slowly until the end of
the Permian. There was a marked decrease in
diversity at the end of the Permian, coinciding
with or preceding the mass extinction of
animal species. Subsequent increase in diver-
sity was slow, reaching around 400 species in
the early Cretaceous, but apparently more
rapid from the mid-Cretaceous.

This overall pattern masks important
changes with time in the composition of the
flora, most notably in the relative importance of
the three main groups of vascular plant: the
pteridophytes, gymnosperms and angio-
sperms (see Figure 3.4|. The Silurian and early
Devonian are marked by a radiation of primitive
pteridophytes. During the Carboniferous, more
advanced pteridophytes and gymnosperms
developed and underwent extensive diversifi-
cation. Following the late Permian extinction
event, pteridophytes were largely replaced by
gymnosperms (although ferns remained
abundant! and these became the dominant
group until the mid-Cretaceous. The dramatic

Period Cambrian

Era

Ordovician

Silurian

PALEOZOIC

Devonian

Carboniferous

Millions of
545 years ago 490

438

408

360

Biodiversity through time

600

300 200

Geological time I lO' years!

100

increase in plant diversity since then is entirely
due to the radiation of the angiosperms, fossils
of which first appear in the lower Cretaceous,
although an early to mid-Jurassic origin,
around 1 70 My ago, has been argued".

Colonization of land by animals has
occurred many times; although the oldest
body fossils of terrestrial animals date from
the early Devonian, it is generally accepted
that the primary period of land invasion by
animals was the Silurian.

The overwhelming number of described
extant species of terrestrial animals are
insects. Their fossil record is more extensive
than might be expected, but had been little
studied until recently. Data on insect diversity
at family level have been collated, based on
nearly 1 300 families'". This analysis shows a
very slow increase in families from the first
appearance of insects in the Devonian, a rise

in numbers during the Carboniferous, and a
steeply increasing rise throughout the
Mesozoic to the Tertiary. The apparent
explosion in insect diversity had previously
been attributed to ecological opportunities
provided by the expansion of flowering plants,
but the insects are now known to have begun
their diversification some 150 fvly before the
flowering plants".

The fossil record of vertebrates includes
around 1 iOO families, with tetrapods (amph-
ibians, reptiles, birds, mammals) somewhat
outnumbering fishes. The bird record is much
less substantial than that for other groups,
probably because their light skeletons have
been less frequently preserved.

Terrestrial vertebrates first appear in the
fossil record in the late Devonian. Diversity
remained relatively low during the Paleozoic,
with around 50 families, and may have

Figure 3.3

Animal family diversity

through time

The lines plotted represent
the number of families in
the fossil record.

Notes: The blue line
essentially represents marine
invertebrate animals.
Although a small number of
vertebrate groups, notably
fishes and a few tetrapod
species, are included, these
make up a tiny proportion of
the total marine family
diversity shown. The curve for
fishes includes an increasing
proportion of freshwater
forms through the Cenozoic.
Tetrapods are amphibians,
reptiles, birds and mammals.
The extent to which such
results, based on sampling
and interpretation of the
fossil record, represent
actual diversity has been
subject to discussion.

Source Marine animals, adapted Irom
Sepkoski ; fishes and tetrapods,
adapted after Benton : insects, adapted
from Labandeira and Sepkoski .

rboniferous Pernnlan Triassic

Jurassic

Cretaceous

Tertiary Quaternary -

^V PALE(

PALEOZOIC

MESOZOIC

KI-IH,-U«.1INU

286

248

213

CENOZOIC

WORLD ATLAS OF BIODIVERSITY

800

700

ui 600

500

400

300

200

100

i..„ Pteridophytes
Gymnosperms
^H Angiosperms

400

300

200

100

Geological time 110° yearsi

Figure 3.4

Plant diversity through time

Notes- Pteridophytes are
ferns Filicinophyta and allies,
gymnospernns are conifers
Coniferophyta and allies,
angiosperms are flowering
plants Anthophyta. Note
changing numerical
dominance of each group
over time.

This generalized diagram is
to illustrate changing
abundance of major groups
over time; it does not
represent short-term
extinction events.

Source: Adapted from Kemp . after
Niklas.

declined during the early Mesozoic. From the
mid-Cretaceous the number of families
started to increase rapidly, reaching a recent
peak of around 340. Diversity of genera follows
this overall pattern in a more exaggerated
form. It appears that periodic increase in the
number of tetrapod families is mainly a result
of lineages becoming adapted to modes of life
not already followed by other organisms, i.e.
by adopting new diets or new habitats".

PATTERNS OF EXTINCTION

If living species represent between 2 and 4
percent of all species that have ever lived™,
almost all species that have lived are extinct,
and extinction can be presumed to be the
ultimate fate of all species.

Numerous estimates have been made of the
lifespan of species in the fossil record; these
range from 0.5 My to 13 My for groups as varied
as mammals and microscopic protoctists.
Analysis of 17 500 genera of extinct marine
microorganisms, invertebrates and verte-
brates, suggests an average lifespan of 4 My in
these groups'. Given this average lifespan, at a
very gross estimate, the mean extinction rate
would be 2.5 species per year if there were
around 10 million species in total. However,
because of bias inherent in the fossil record,
such lifespan estimates are likely to relate to

widespread, abundant and geologically longer-
lived species, in effect, the extinction-resistant
forms; and so do not represent the biota as a
whole-'. If most species survived for less than
4 My, the overall extinction rate at any given
time would have been correspondingly higher

Major extinctions in animals
In general the Precambrian fossil record is too
incomplete to allow detailed analysis of extinc-
tion rates. However, there is good evidence of a
major loss of diversity during the Vendian
period in latest Precambrian times, around
550 My ago, when the entire Ediacaran fauna
disappeared lalong with many acritarchsl.
Another wave of extinction affected archaeo-
cyathid sponges, mollusks and trilobites during
the lower Cambrian, some 530 My ago.

By far the most severe marine invertebrate
mass extinction was in the late Permian
1250 My ago). At that time, the number of
families of marine animals recorded in the
fossil record declined by 54 percent and
the number of genera by 78-84 percent.
Extrapolation from these figures indicates
that species diversity may have dropped by as
much as 95 percent. Other major extinctions
in marine invertebrates occurred at the end of
the Ordovician 1440 My ago) (see Figure 3.31,
when around 22 percent of families were lost,
and during the late Devonian and late Triassic
121 percent and 20 percent, respectively!.
Around 15 percent of marine families
disappeared at the end of the Cretaceous.

The vertebrate fossil record, especially for
terrestrial tetrapods, is much less amenable to
analysis of extinction rates than the inverte-
brate record, chiefly because it is less complete
and less diverse. However, studies indicate
that fishes have been subject to at least eight
important extinction events since their recorded
origin in the Silurian, while tetrapods have
experienced at least six such events since their
appearance in the late Devonian. Some of these
events coincide with each other and with those
recorded for marine invertebrates; in particular,
the five major mass extinction events outlined
above are paralleled by losses in vertebrate
diversity. The most significant is the late
Permian event, which, in terms of percentage
loss, IS the largest recorded extinction both for

iiodiversity through time 29

fishes (44 percent of families disappearing from
the fossil record) and tetrapods (58 percent of
families disappearing). The late Cretaceous
event was more significant for tetrapods than
for fishes, with at least 30 of the 80-90
families then in the fossil record disappearing
at this time. These families were, however,
virtually confined to three major groups which
suffered complete extirpation - the dinosaurs,
plesiosaurs and pterosaurs. Most other
vertebrates were almost completely unaffected.

Major extinctions in vascular plants

Fewer major extinction events have to date
been distinguished in the plant fossil record
than in the animal record. Periods of elevated
plant extinction appear in some cases more
protracted than animal extinction events and
not usually coincident with them". However,
there is good evidence for extensive reduction
in woody vegetation at the end of the Permian,
with widespread loss of peat forests in humid
areas and of conifers in some semi-arid
regions. A fourfold increase in atmospheric
carbon dioxide around the Triassic-Jurassic
boundary is correlated with a more than
95 percent turnover in the megaflora (i.e. leaf
fossils, etc., as opposed to pollen or spores]".
The end-Cretaceous catastrophe appears to
have had a major influence on the structure and
composition of terrestrial vegetation and on the
survival of species. Data from fossil leaves
suggest that perhaps 75 percent of late

0.20

0.15

0.05

Figure 3.5

Frequency of percent
extinction per million year
period

See text for explanation.
Source- Adapted from Raup',

40 60

% extinction

Cretaceous species became extinct, although
data from fossil pollens indicate a lower though
still significant level of extinction. During the
Tertiary there are two other periods of
widespread heightened extinction rates, during
the late Eocene and from the late Miocene to
the Quaternary, although in the latter extinction
of taxa at generic level and above appears to
have been mainly regional rather than global.

Mass extinctions

The very many species extinctions represented
in the fossil record are not distributed evenly
through time, nor do they occur randomly In
paleontology much attention has been devoted
to mass extinction periods, during which some
75-95 percent of species then living became

Figure 3.6
Number of family
extinctions per geological
interval through the
Phanerozoic

Note: 76 geological intervals
represented, average
duration around 7 million
years.

Source: Benton .

350

■2 300

S 250

^ 200

i 150

z 100

kill

500

Paleozoic .

400

iiiil.liilaiLi..liljLi

300 200

Geological time (1 0' years!

Mesozoic

dilluk

Cenozoic

100

30 WORLD ATLAS OF BIODIVERSITY

Table 3.1

The principal mass
extinctions in the
Phanerozoic fossil record

Source: Summarized from Kemp ,
Hallam and Wignalt ,

extinct during geologically very short periods of
time, in some cases possibly as little as a few
hundred thousand years or even less. Five such
phases (based chiefly on extinction of marine
species] are recognized during Phanerozoic
time, late in each of the Ordovician, Devonian,
Permian, Triassic and Cretaceous periods
(Table 3.1).

Although each of the 'Big Five' mass
extinctions had a very profound effect on then
contemporary life, they are not isolated peaks

standing out from a constant, very low,
background rate. Rather, the extinction rate
has varied continuously throughout the
Phanerozoic with periods of more or less
elevated rates of extinction, of which the most
extreme can be characterized as mass extinc-
tion events. A frequency plot (Figure 3.5| of the
percentage of species becoming extinct in each
1 -My interval of the 600-My record of animal life
provides an indication of this variation in
extinction intensity'. The plot takes the form of

D3TB species

(l^lyagol loss (%1

Late Ordovician 4A0

□loiic cnange

The last but largest of several extinction events during the Ordovician. More than
25% of marine invertebrate families lost. Entire class Graptolithina reduced to a
few species; acritarchs, brachiopods, conodonts, corals, echinoderms, trilobites,
all much reduced.

Cooling; Warming
Marine regression
Marine transgression
and anoxia

Late Devonian 365

End Permian 250 V5

End Triassic 205

End Cretaceous 66 75

Mass extinctions came at the end of prolonged period of diversity reduction. Marine transgression

Rugose corals lost >95% of shallow water species; stromatoporoid corals and anoxia

reduced by half and reefs disappeared; brachiopods lost 33 families; ammonoids

and trilobites severely affected. Fishes suffer only major mass extinction; all

early jawed fishes (placodermsl disappear and most agnatha. First major crisis

in plants; diversity greatly reduced, but spread of first tree, the gymnosperm

Archaeopteris.

The most severe extinction cnsis: metazoan life came within a few percent of total Volcanism

extinction.Tabulate and rugose corals terminated, complex reefs disappear (return Warming

after 8-My gap); echinoderms almost wiped out; worst crisis in history of Marine transgression

foraminifera; severe extinction in ammonites, brachiopods, bryozoa, mollusks.

Some losses in early ray-fin fishes. Major loss of terrestrial vertebrates (75% of

families! and insects (8 of 27 insect orders extinctl. Mass extinction in plants: large

plants including peat-forming trees lost (spread of small conifers, lycopods and

guillwortsi; sudden unprecedented abundance of fungal spores at end of period.

Mass extinction in marine invertebrates, especially brachiopods, cephalopods Marine regression

and mollusks; also mass disappearance of scleractinian corals and sponges.
Several seed fern families lost; some land vertebrates lost, but evidence for
mass extinction questionable.

Radical change in planktonic foraminifera; 85% of calcareous nanoplankton lost. Impact of large meteor
also all ammonites, belemnites and many bivalves; losses in echinoids and Volcanism

corals. Many marine reptiles extinct (ichthyosaurs, plesiosaurs, mosasaurs); Cooling

significant losses in freshwater and terrestrial vertebrates, including last Marine regression

dinosaurs (high turnover throughout dinosaur history - end Cretaceous unusual
in that no replacements emerged]. Mass extinctions in plants: highest (possibly
60% species loss! in angiosperms, lowest in ferns.

Biodiversity through time 31

an even [eft-sl<ewed curve, with no substantial
discontinuity between rare periods of high
extinction rate Imore than 60 percent of species
extinct per one million years) and the most
frequent rate (10-15 percent); the mean
intensity is 25 percent of species extinct per
1-f^y interval. This suggests that mass extinc-
tions may arise from causes not qualitatively
different from those associated with extinctions
at other times. It should also be remembered
that because of the extremely long duration of
the Phanerozoic during which species have
been constantly becoming extinct, and the very
short duration of the mass extinction events
themselves, the latter account for only a small
percentage (estimated at around 4 percent) of
all extinctions.

The precise causes and timespans of each
of the mass extinctions have been the subject of
much debate and study"". It is now widely
accepted that the late Permian mass extinction
was a long-term event, lasting for 5-8 million
years. It appears to have been associated with
geologically rapid global physical changes
(including the formation of the supercontinent
Pangaea], climate change, and extensive,
tectonicatly-induced marine transgression and
increased volcanic activity. The late Cretaceous
extinction is probably the best known, but in
terms of overall loss of diversity is also the least
important. There is strong evidence that this
extinction event was associated with climate
change following an extra-terrestrial impact,
although this remains somewhat controversial.
The late Ordovician event appears to be
correlated with the global Hirnantian glaciation,

REFERENCES

1 Raup, D.M. 1994. The role of extinction in evolution. Proceedings of the National
/Academy o^Sc/ences 91:6758-6763.

2 Sepkoski, J.J. 1992. Phylogenetic and ecologic patterns in the Phanerozoic history of
marine biodiversity. In: Eldredge, N. (ed.) Systematics, ecology and the biodiversity
crisis, pp. 77-100. Columbia University Press, New York.

3 Jablonski, D. 1991. Extinctions: A paleontological perspective. Science 253: 754-757.

4 Benton, M.J. 1995. Diversification and extinction in the history of life. Science 268: 52-58.

5 Alroy, J. et al. 2001. Effects of sampling standardization on estimates of Phanerozoic marine
diversification. Proceedings of the National Academy of Sciences 98: 6261-6266.

6 Mojzsis, S.J. et ai 1996. Evidence tor life on Earth before 3,800 million years ago. Nature
384: 55-59.

7 Nisbet, E.G. and Sleep, N.H. 2001. The habitat and nature of early life. Nature A09: 1083-1091.

8 Farmer, J.D. 2000. Hydrothermal systems: Doorways to early biosphere evolution. GSA

with three separate episodes of extinction
spread over only 500 000 years. In all cases,
however, the ability to determine accurately the
timing and periodicity of extinction is heavily
dependent on the completeness of the fossil
record and the reliability and precision of
stratigraphic analysis.

A mass extinction period is typically
followed by a phase of 5-10 My of very low
diversity, with a handful of species dominant in
fossil faunas and floras. When diversity again
increases the biota may be very different in
composition from those preceding. In several
instances, groups previously showing low
diversity have radiated and spread widely
following the demise of groups previously
dominant; e.g. ray-finned fishes diversified
following loss of placoderms; quillworts and
seed ferns diversified for a time after loss of
glossopterids; and the mammals radiated
after loss of many terrestrial reptile groups.
Reef organisms provide a classic example:
reef communities have been lost several
times during the Phanerozoic but with each
reappearance the main reef-building forms
have been different'^ Important new evo-
lutionary opportunities arise for the lineages
that survive mass extinction events, which can
radically redirect the course of evolution'"'.

The present-day diversity of living
organisms is the result of a net excess of
originations over extinctions through geological
time. Figure 3.6 shows the number of family
extinctions in each interval of geological time,
and the cumulative diversity of families, both
marine and terrestrial.

32 WORLD ATLAS OF BIODIVERSITY

Today 10. Available online (accessed January 2002] at http://www.geosoclety.org

9 Rasmussen, B. 2000. Filamentous microfossils in a 3,235-mi[licn-year-old
vulcanogenic massive sulphide deposit. Nature AOb: 676-678.

10 Watanabe, Y., Martin, J.E.J, and Ohmoto, H. 2000. Geochemical evidence for terrestrial
ecosystems 2.6 billion years ago. Nature ^08: 574-578.

1 1 Brocks. J.J. ef al. 1999. Archean molecular fossils and the early rise of eukaryotes.
Sc/ence 285: 1033-1036.

12 Javaux, E.J., Knoll. A.H. and Walter. M.R. 2001. Morphological and ecological
complexity In early eukaryotic ecosystems. Nature k\l: bb-k'^ .

13 Knoll. A.H. 1995. Proterozolc and early Cambrian protlsts: Evidence for accelerating
evolutionary tempo. In: Fitch. W.M. and Ayala. F.J. ledsl. Tempo and mode in
evolution: Genetics and paleontology 50 years after Simpson, pp. 63-68. National
Academy Press, Washington DC.

U McMenamin. M.A.S. 1989. The origins and radiation of the early metazoa. In: Allen. K.C.
and Bnggs. D.E.G. ledsl. Evolution and the fossil record, pp. 73-98. Belhaven Press, a
division of Pinter Publishers Ltd, London.

15 Margulls, L. and Schwartz, K.V. 1998. Five kingdoms. An illustrated guide to the phyla
of life on earth. 3rd edition. W.H. Freeman and Company, New York.

16 Wellman, C.H. and Gray, J. 2000. The microfossil record of early land plants.
Philosophical Transactions of the Royal Society of London. Series B. Biological
Sciences 255: 717-731.

17 Wikstrom, N., Savolalnen, V. and Chase. M.W. 2001. Evolution of the angiosperms:
Calibrating the family tree. Proceedings of the Royal Society of London. Series B.
Biological Sciences 268: 221 1 -2220.

18 Labandeira. C.C. and Sepkoski. J.J. 1993. Insect diversity in the fossil record. Science
261: 310-315.

19 Benton, M.J. 1999, The history of life: Large databases In palaeontology. In: Harper,
DAT. (ed.l Numerical palaeontology: Computer-based modelling and analysis of
fossils and their distributions. John Wiley and Sons, Chichester and New York.

20 May, R.M., Lawton, J.H. and Stork, N.E. 1995. Assessing extinction rates. In: Lawton,
J.H. and May, R.M. ledsl. Extinction rates, pp. 1-24. Oxford University Press, Oxford.

21 Jablonski, D. 1995. Extinctions In the fossil record. In: Lawton, J.H. and May, R.M.
ledsl. Extinction rates, pp. 25-44. Oxford University Press, Oxford.

22 Crane, PR. 1989. Patterns of evolution and extinction in vascular plants. In: Allen, K.C.
and Bnggs, D.E.G. ledsl. Evolution and the fossil record, pp. 153-187. Belhaven Press,
a division of Pinter Publishers Ltd, London.

23 McElwain, J.C, Beerling, D.J. and Woodward, Rl. 1999. Fossil plants and global
warming at the Triassic-Jurassic boundary. Science 285: 1386-1390.

24 Kemp, T.5. 1999. Fossils and evolution. Oxford University Press, Oxford.

25 Hallam, A. and Wignall, PB. 1997. Mass extinctions and their aftermath. Oxford
University Press, Oxford.

26 Erwin, D.H. 2001. Lessons from the past: Biotic recoveries from mass extinctions.
Proceedings of the National Academy of Sciences 98: 5399-5403.

27 International Subcommlssion on Cambrian Stratigraphy, http://www.unl-
wuerzburg.de/palaeontologie/Stuff/casu6.htm laccessed January 20021.

28 Benton, M.J. 1989. Patterns of evolution and extinction In vertebrates. In: Allen, K.C.
and Briggs, D.E.G. ledsl. Evolution and the fossil record, pp. 218-241. Belhaven Press,
a division of Pinter Publishers Ltd, London.

29 Benton, M.J. led. I 1993. The fossil record 2. Chapman and Hall, London. Data available
online In interactive format at http://lbs.uel.ac.uk/ibs/palaeo/benton/ laccessed
January 20021; also see http://palaeo.glybrls.ac.uk/frwhole/fr2.familles

Humans, food and biodiversity 33

HumanSp food

and biodiversity

THE LINEAGE LEADING TO THE HUMAN SPECIES EMERGED from an ancestry among the
apes, almost certainly in Africa around 5 million years ago. Fossils usually attributed
to the genus Homo itself date from the late Pliocene, perhaps 2 million years ago, and
anatomically modern humans appeared some 100 000 years or more before the present.

Agriculture developed independently in several regions around 10 500 years ago, in
association with increased population density. The global human population continued to grow
only slowly until the 19th century, when revolutionary developments in agriculture, industry
and public health triggered an exponential rise that has continued to the present day.

Agriculture is a means to channel the Earth's resources into production of human bodies.
Humans have converted large areas of terrestrial habitat and use more than one third of net
primary production on land. They are strongly implicated in the extinction of many large
terrestrial mammal and bird species in prehistory, and are responsible for habitat change
and exploitation that have caused the decline and extinction of many species in recent times.

HUMAN ORIGINS

The human species evolved as a natural
element of diversity in the living world, and it
is a simple ecological imperative that humans
depend on other species and communities to
supply the basic requirements of existence
and to maintain biosphere function.

The creation of organic compounds by
photosynthesizing organisms is the point at
which the sun's energy enters the biosphere
(Chapter II; humans and other animals are
unable to capture energy in this way and must
consume and digest either primary producers
or other organisms that are themselves
dependent on primary producers, in order to
obtain these energy-rich organic compounds
for their own activities.

While humans are doing nothing funda-
mentally different from other animals, with
the benefits of society and technology, which
serve to increase the rates of resource
extraction, they are uniquely successful at it.
Self-evidently, humans have not arrived at
their extraordinarily dominant position on
the planet overnight. The growth of their
influence can be traced back several million

years, to well before the Pleistocene when a
stone-tool-wielding hominid first emerged
somewhere in eastern Africa.

Climate during the past 2 million years
The Earth's climate appears always to have
been in a state of flux. Generally the degree
of accuracy in our understanding of global
climate, and certainly the degree of resolution
in the timescale of climate change, decreases
the further back in time we go. It is therefore
difficult to compare periods that in geological
terms are recent with more distant times.
However, it does seem that during the past
2 million years there have been numerous,
intense climate changes that were at least as
severe, or perhaps more severe, than any
recorded earlier in the Earth's history'. It is
during this period that hominids very similar
to modern humans first appear in the fossil
record. By the early Holocene, some 10 000
years ago, technologically sophisticated
humans had spread to all the major land
masses except Antarctica, and had evidently
started to exert a major, and ever-growing,
impact on the biosphere. Indeed there is

34 WORLD ATLAS OF BIODIVERSITY

Map 4.1

Early human dispersal

A highly generalized view
of the colonization of the
world by advanced humans
from the early Pleistocene
onwards. Coastline is shown
150 m lower than at present
and. with northern
hemisphere ice cover,
represents an approximation
of that at periods of glacial
maximum. Dates represent
earliest well-established
time of arrival; arrows
indicate general net
direction of dispersal, not
actual routes. Presence and
dispersal of early humans
are not represented: the
lineage is believed to have
arisen in Africa and later
fossil material dated to
between 1.8 and 1.4 million
years before the present is
known from several sites in
Eurasia.

Ice

Human dispersal

Hawaiil

ilCuba,"'
HIspani&la).

x^<^

-y^

000 Years before the present

evidence that such impacts - for example in
the extinction of large mammal and bird
species - were already being felt con-
siderably earlier than this. However, because
the rise of humans coincides with a period of
major climatic and ecological fluctuations, it
is often difficult to disentangle the effects
of the former from the latter, and so the
precise nature of these impacts remains
controversial.

The Tertiary period, which began some
65 million years ago and ended with the start
of the Quaternary (the Pleistocene and
Holocenel 1.8 million years ago, is charac-
terized overall by a gradual decrease in
global temperature and increase in aridity
Superimposed on this general pattern were
many oscillations, occurring on a timescale
of thousands of years'. These oscillations are

1 500

" ■ Vv , ■

-1 500

-^^^,^.J6.. ::\ I

believed to be linked to cyclic variations in the
Earth's position in its orbit around the sun,
known as Milankovitch cycles. These cycles
notwithstanding, the climate during virtually
the whole of the Tertiary was notably warmer
than at present.

Around 1.8 million years ago, at the
very start of the Pleistocene^ there was
apparently rapid global cooling leading to the
start of a period dominated by marked
climate cycles of around 100 000 years'
duration. For long periods of each cycle
global temperatures were significantly lower
than they are today, and extensive areas of
the northern hemisphere land masses were
covered in ice sheets. Detailed analysis of
climate changes over the past AOO 000 years
or so, particularly through examination of
Antarctic ice cores', indicate that each cycle

Humans, food and biodiversity 35

V-.\ ■ • ^<^ t ■ -"i' ■! ' ^''''^-^'

:.>V.

■Jl^'

^'•■\^^J^-

56 000

}i' (M,

' 3 500

(New
Zealand!

ii-..^.

over this period Inas been broadly similar,
vi/ith a gradual decline in average temp-
erature, although w/ith many minor and some
large oscillations (relatively colder periods
being referred to as stadials and warmer
ones as interstadialsl, followed by a short
period of intense warming in which temp-
eratures rose from those of fully glacial
conditions to those characterizing a warm
interglacial state, perhaps sometimes over
only a few decades.

Mean global temperature during glacial
periods was around 6''C cooler than during
interglacials, with cooling more pronounced at
the poles than the equator. The mid-latitudes
and equatorial regions were probably some-
what more arid than at present. The large
amount of water locked up in the greatly
expanded polar ice caps meant that mean sea

level during glacial periods was probably
around 100-150 meters lower than at present.
Until around 11000 years ago, each
temperature peal< was apparently followed
by an almost immediate decline as the
next glacial cycle began'. Overall, it appears
that during the past half million years the
Earth's climate has been as warm as, or
warmer than, today's for only around
2 percent of the time', it seems lil<ely
that this also holds true for the early
Pleistocene. The end of the last glacial cycle,
around 1 1 000-12 000 years ago, marked the
start of the Holocene. Temperatures similar
to today's have prevailed throughout the
(Holocene, making it by far the longest true
interglacial period for at least the past half
million years and probably for the last
1.8 million years.

36 WORLD ATLAS OF BIODIVERSITY

Human origins and dispersal

The origins and early history of humans are
among the most controversial subjects in
paleontology. Remains are generally scarce
and often open to varying interpretation.
Nevertheless, consensus has emerged over
the broad outlines. It is likely that the direct
ancestors of humans - the homlnid line -
diverged from the apes In Africa during a cool,
dry phase of the late Miocene, around
6 million to 5.5 million years ago. The primary
evidence for this divergence is in the form of
'molecular clocks' calculated from comparison
of human and ape genetic materiall Relevant
fossil material from this time horizon Is
sparse and often of uncertain status, but
material from Ethiopia dated to 5.8 million to
5.2 million years ago' is believed to represent
Ardipethecus, generally accepted as the
earliest known member of the homlnid
branch. Other species of this genus are
known from before A million years ago, with
various species of Australopithecus and
Paranthropus, all from eastern Africa, some-
what younger in age.

Sometime during the middle to late
Pliocene, 2.5 million to 1.8 million years ago,
the genus Homo is thought to have evolved
from Australopithecus stock^'. Until com-
paratively recently it had been assumed that
early man then remained confined to Africa
until less than 1 million years ago. However,
two well-preserved skulls recently found In
the Caucasus (southern Georgia) have been
dated to about 1.8 million years' (and variously
attributed to l-lomo ergaster, otherwise
known from Africa, or H. erectus]. Elsewhere,
the earliest stone tools from Turkey and
southwest Asia have been dated to about 1.5
million years ago, and from northeast Asia to
nearly 1.A million years', while tools and
fossils from East and Southeast Asia might be
as old as 1.9 million years. These finds
suggest that populations of early forms of
l-iomo spread through the southeastern
fringes of Europe as well as into Asia within at
most a few hundred thousand years of the
genus originating in Africa^

The earliest hominid remains elsewhere in
Europe date from 780 000 years ago in Spain'",
followed by somewhat different material

known from several sites widely distributed in
northern Europe of about 500 000 years in age
or less". Fossil material from these two
periods may be attributed to H. antecessor
and H. heidelbergensis. respectively. The
diversity of homlnid material and the range
of dates suggest that the spread of early
hominids was complex and multidirectional,
with possibly three lineages moving into
Eurasia from Africa.

Ivlodern humans, i-iomo sapiens, are
believed to have appeared sometime after
200 000 years before the present. Anatomically
modern human remains are known from
around 100 000 in Africa and the Middle East,
with more recent dates In East and Southeast
Asia. The first appearance in Europe dates
from about AQ 000 years ago"", with
expansion apparently from the east toward
the west, where in places premodern
(Neanderthal) humans remained until about
28 000 years ago. One interpretation of the
several lines of genetic evidence suggests, on
the basis of mathematical models, that the
modern human population Is derived from a
small founding population of perhaps 10 000
breeding Individuals that existed sometime
between 130 000 and 30 000 years before
the present'-' ".

The earliest evidence of hominids outside
Africa, Europe and Asia Is much more recent.
The age of the human skeleton found In 197i
at Lake Mungo in Australia, evidently the
earliest human remains known on the
continent, has been estimated at 62 000 years
before the present'', but a date close to 56 000
years Is now widely accepted for human
arrival on the continent".

in the Americas, the oldest good evidence of
human presence Is that of a coastal settlement
in Chile dated around U 000-15 000 years
ago'"', although even this Is far from universally
accepted amongst archeologlsts. There are
very controversial and now widely questioned
claims for evidence of human settlement much
earlier, most notably that dating from 32 000
years before the present from Pedra Furada in
northeast Brazil". The earliest unequivocal
evidence of widespread occupation in the
Americas comes from the so-called Clovis
hunting culture whose oldest remains are

Humans, food and biodiversity 37

generally dated at around 12 000 years before
the present". Settlement of the Caribbean
islands including Cuba and Hispaniola appears
to have taken place considerably later, around
6 000 years ago'",

Archeological evidence indicates that
colonization of the Pacific Islands east of New
Guinea began around it 000 years ago, when
much of Melanesia and Micronesia was
settled. Fiji and Samoa were probably colon-
ized around 3 500 years ago. and the outliers
of Hawaii, Easter Island and New Zealand
within the last 1 500 years-'. In the Indian
Ocean. Madagascar was probably first settled
around 1 500 years ago'*.

Technology

The earliest evidence of tool manufacture by
hominids is from the Gona River drainage in
northern Ethiopia where stone tools have
been dated to at least 2.5 million years ago.
These tools are small (generally less than 10
centimeters long) and simple, but are alteady
of relatively sophisticated manufacture,
suggesting that even older artifacts will
eventually be found". They are of essentially
the same design as those dated from 2.3
million to 1.5 million years ago from other
sites in East Africa (e.g. Lokalalei in the Lake
Turkana basin in northern Kenya and the
Olduvai gorge in Tanzania) indicating effective
technological stasis tor at least 1 million
years (from 2.5 million to roughly 1.5 million
years agol". Collectively these tools are
referred to as products of the Oldowan stone
tool industry".

Around 1.5 million years ago much
more sophisticated and often larger tools
including hand axes and cleavers suddenly
appear in the archeological record in East
Africa. These are referred to as Acheulian
tools. The Acheulian tool industry spread into
Europe, the Near East and India, and remained
apparently relatively unchanged until around
200 000 years ago, showing similar temporal
persistence to the Oldowan industry.

Evidence of tools and artifacts made from
organic materials from the Paleolithic is
understandably extremely scarce. In 1995,
however, three large wooden implements
very similar in design to modern-day javelins

and dated to around AOO 000 years ago were
discovered at Schdnmgen in Germany. These
were found in association with a smaller
wooden implement (probably a throwing
stick), stone tools and the butchered remains
of more than ten horses, and can be per-
suasively interpreted as throwing spears used
in systematic, organized hunting".

There is also intriguing although indirect
evidence from Southeast Asia dating to
around 900 000 years ago that hominids at
this time, at least in this region, were capable
of repeated water crossings using watercraft.
The evidence is in the form of stone tools
dated to this age from the island of Flores in
eastern Indonesia. Even at the time of the last
glacial maximum (when global sea levels
would have been at their lowest), reaching
Flores from the Asian mainland would have
required crossing three deep-water straits
with a total distance of at least 19 kilometers.
The impoverished - and typically island -
nature of the Paleolithic fauna of Flores would
appear to preclude the existence of any now
submerged land bridge". The next oldest,
again indirect, evidence for the use of
watercraft is the colonization of Sahul,
probably sometime around ^0 000 to 60 000
years ago. Even at times of lowest sea level,
this would have necessitated crossing some
100 kilometers of open sea.

Fire

At some point in their evolutionary history,
hominids clearly learnt to control, manipu-
late and, presumably later, to start fires.
Determining even very approximately when
this may have happened is difficult, and as

■4

Watercraft may have played
an important role in human
dispersal for millennia.

WORLD ATLAS OF BIODIVERSITY

Large-scale burning of
vegetation is one of the
major fiuman impacts on
the biosphere.

with all else to do with human evolution,
controversial. This is chiefly because the
existence of natural fires caused by lightning
strikes and volcanic activity greatly compli-
cates the interpretation of the archeological
and geological record - an association
between artifacts and evidence of burning
does not necessarily indicate a direct link
between the two'°.

The earliest dated associations between
artifacts and burning that could be construed
as deliberate use of fire are from Africa. Stone
tools and splintered bones, which can be
interpreted as evidence of butchery, have
been found associated with clay baked at
several hundred degrees for several hours at
Chesowanja in Kenya, in deposits around
1.4 million years old''". The characteristics of
the clay are consistent with formation
beneath a campfire, but are also consistent
with formation around a slow-burning tree
stump that could be associated with a natural
bushfire". Charred animal bones and other
evidence of human occupation from just over
1 million years ago have been found at
Swartkrans cave in South Africa, although
similar problems of interpretation apply.

Numerous sites in Europe and Asia provide
evidence of human occupation and associated
fire from the mid-Pleistocene, some 400 000
years ago. The best known of these is at
Zhoukoudien in China, although even here
the evidence for deliberate use of fire is
widely considered equivocal"™. Others are
at Torralba-Ambrona (Spain), Terra Amata
IFrancel, Westbury-sub-Mendip (England!
and Vertesszollos (Hungary]''.

Large-scale burning of terrestrial vege-
tation is undoubtedly one of the major
present-day impacts by humans on the
biosphere, and is believed to constitute
around one third of current annual anthropo-
genic carbon dioxide emissions. Evidence of
earlier impact is invariably circumstantial. It
is difficult to demonstrate widespread
biomass burning in the fossil or subfossil
record and even harder to demonstrate a link
with human activities - natural fires may be
caused by storms or volcanic activity and may
be expected to vary in extent, frequency and
intensity according to prevailing climatic
and ecological conditions.

However, the abundance of elemental
carbon in marine sediments off Sierra Leone
in Africa can be persuasively interpreted as a
measure of intensity of biomass burning in
sub-Saharan Africa. Analysis of a core from
these sediments covering the past million
years or so indicates that inferred fire
incidence in the region was low until around
400 000 years ago. Since then, five episodes of
intense burning of vegetation can be inferred,
all except the most recent coinciding with
periods when the global climate was changing
from interglacial to glacial. The current peak
is unique in that it is occurring during an
interglacial period. The change from inter-
glacial to glacial climate is generally
associated with increased aridity, so that
vegetation may be expected to be more
vulnerable to fire. It would also be expected
that a substantial fuel base in the form of
woody biomass would have accumulated
during the warmer, wetter, interglacial. It may
well be merely coincidental, but it is intriguing
that the period of increased fire incidence in
the Sierra Leone core - around 400 000 years
ago - coincides with the timing of the first
widespread evidence for hominid use of fire.

Food resources

The early australopithecine members of the
hominid lineage appear, on craniodental
evidence, to have been well adapted to
consume hard but brittle items, such as nuts
and seeds, and soft items, such as fruits'^
Use and development of stone tools by later
hominids is usually thought of as associated

Humans, food and biodiversity 39

with meat consumption, and cut and hammer
marks on large mammal bones from hominid
sites are evidence of this, but microscopic
wear patterns found on some tools indicate
they were used to scrape wood and cut coarse
plants such as reeds or grass". Oldowan
and Acheulian stone tools are rarely found
more than 10-20 kilometers from their rock
sources, suggesting that their owners ranged
over a relatively small area", so if hominid
populations at the time were small in size, the
environmental impacts of targeted resource
extraction would have been minimal.

The ratio of different forms of carbon and
nitrogen incorporated in human bone colla-
gen can provide information on the relative
importance of food from terrestrial, coastal or
freshwater habitats. Such 'stable isotope
analysis' of collagen extracted from archeo-
logical remains, coupled with other material
evidence, suggests that pre-modern humans
(Neanderthals] in Europe, although omni-
vorous and opportunistic, behaved as preda-
tory carnivores^'. Similar evidence for slightly
later human remains suggests a significant
broadening of the resource base, owing
particularly to increased use of freshwater
resources". Remains found at Ivlediterranean
sites suggest further diversification of human
diet and the means of food gathering".

associated with an apparent rise in human
numbers during the later Paleolithic and
approaching the start of the Neolithic. Here,
there was increasing use of agile small game,
such as partridges and hares, and apparently
decreasing reliance on slow-maturing, easily
collected forms such as tortoises and shell-
fish, the average size of which decreased over
time at a rate consistent with the effects of
excess harvesting, possibly indicative of
increasing human density^'.

Origins of agriculture

Until the end of the Pleistocene, humans
evidently depended on hunting and gathering
of wild resources for their sustenance.
Around the end of this period a radical
change was initiated - the emergence of
crop-based agriculture and domestication of
livestock - phenomena which appear to have
arisen independently in Africa, Eurasia and
the Americas.

The earliest direct evidence for animal
domestication, of the dog Canis. dates from
around 14 000 years ago in Oberkassel in
Germany". However, analysis of molecular
clocks' appears to show that the dog diverged
from its wild ancestor, the wolf Canis lupus,
far earlier than this - perhaps more than
100 000 years ago". It is quite possible that

The dog diverged from
its wild ancestor, the
wolf, more than 100 000
years ago.

40 WORLD ATLAS OF BIODIVERSITY

domestication of the dog as a guard animal
and aid in hunting predates domestication of
other animats and plants as food sources
by many tens of thousands of years, but In
the absence of archeological evidence this
remains speculative.

The study of plant and animal domestication
for agricultural purposes is a rapidly expand-

Box iA Loss of diversity in agricultural genetic resources

Cfilna

Wheat varieties

Korea

Garden tandraces

IRep.l

Mexico

Maize varieties

With the rise of Industrial-scale agriculture and commercial breeding, many local
agricultural genetic resources, both crops and livestock, have been lost and replaced by
modern, genetically uniform types specialized for superior production in higher Input
systems. There is much evidence that many local varieties possessed features of adaptive
value in a particular environment and cultural context, and the precautionary principle
argues that the diversity of domesticated forms and their wild relatives should be conserved
where possible in order to maintain options for future breeding Improvements. A complete
picture of the global reduction In local genetic resources is not available, but there is
abundant evidence at national level of the enormous scale of genetic erosion in crop plants.

About 1 000 110 percent) of 10 000 vaneties used
in 1949 remained in 1970s

About 25 percent of landraces of U crops grown
in home gardens In 1985 remained in 1993

Only 20 percent of maize varieties planted in
1930s remain

USA Varieties of apple, cabbage. Only 15-20 percent of varieties grown in 1804-

field maize, pea, tomato 1904 available at present

A growing number of countries have documented and evaluated livestock resources, see
Appendix 3 for summary of mammal breed status.

ing field, based on study of fossil pollen
records, archeological remains, the genetics
of present-day crops and their close relatives,
and remaining indigenous agricultural
systems". The development of agriculture
began independently in different continents
and proceeded at different rates, while early
cultivators undoubtedly continued to rely
heavily on hunting and gathering from the wild.
The first development of agriculture as an
integrated system for food production was
based in the Fertile Crescent, composed of
the uplands of Anatolia and western Iran and
the arid lowlands to the south, with the first
tangible evidence approaching 1 1 000 years in

age. Wheat Triticum. barley Hordeum, rye
Secale, pea Pisum and lentil Lens, cattle 80s,
sheep Ovis. goat Capra and pig Sus were all
domesticated in this region, and formed the
basis of the Neolithic peasant economy that
spread steadily through much of western
Eurasia into surrounding areas after about
10 000 years before the present. The system
integrated the use of food plants, cereals
especially, and domesticated animals for
fertilizer and power as well as food.

Although early plant domesticates are
known elsewhere in the world. Integrated
agricultural systems appear to have taken
longer to develop, perhaps in part because of
the absence of domesticated animals. For
example, in Middle America seeds of domes-
ticated squash Cucurbita are known from
about 10 000 years ago, maize from 6 300 years
and beans from 2 300, with village-based
farming economies evidently taking several
thousand years to develop'". Farming systems
based on rice cultivation In Asia appear to date
from 7 500 years ago.

Analysis of mitochondrial DNA has
indicated that a multiple maternal origin is
general among domestic livestock species,
i.e. female animals from more than one wild
stock have contributed, at different times and
places, to the present genetic diversity among
each breed". For each of the four major
breeds (cattle, sheep, goats, pigs), this
evidence is consistent with archeological
findings in suggesting a primary center of
origin around the Fertile Crescent region, but
also suggests additional centers in Asia.

Ivlost of the major crops on which humans
presently depend have been grown con-
tinuously since the early or middle Holocene.
They have been constantly selected over this
period and have developed large amounts of
useful genetic variation. Indeed the success
of Individual crop species over wide geo-
graphical areas is partly determined by
their flexibility in evolving and sustaining
genotypes suitable for local environments.
Conventional breeding Involves the selection
and crossing of desirable phenotypes within
a crop in order to create more productive
genotypes. The process of harvest, storage
and sowing alone may have assisted In the

Humans, food and biodiversity 41

selection of traits sucfi as non-shattering seed
fieads, uniform ripening of seeds, uniform
germination, targe fruits and seeds, and easy
storage. Breeding methods have increased the
rate of introgression of desired genetic traits
into new cuttivars; genetic modification is now
possible at the level of incorporating individual
genes directly into genomes.

Domesticated crops and livestock have
been transported around the world probably
since full-scale agriculture began, e.g.
wheats are recorded in areas outside their
presumed center of origin at least 8 000
years before the present. Some crops became
increasingly widely distributed after the 1 500s
when European colonists moved out of their
home continent.

CURRENT FOOD AND NUTRITION

Perhaps as many as 7 000 of the 270 000
described plant species have been collected
or cultivated for consumption'"', but very few,
some 200 or so, have been domesticated, and
only a handful are crops of major economic
importance at global level". The variety of
species used is limited more by production
and cultural factors, such as tradition and
palatability, than by nutritional value. Twelve
crop plants together provide about 75 percent
of the world's calorie intake. These comprise
lin alphabetical order]: bananas/plantains,
beans, cassava, maize, millet, potatoes, rice,
sorghum, soybean, sugar cane, sweet potatoes
and wheat''. More than 40 mammal and bird
species have been domesticated, around 12
of which are important to global agricultural
production (see Appendix 3). Some, such as
cattle, pigs, goats, sheep and chickens, are
fundamental to many non-industrial agri-
cultural systems, and provide a wide range of
products in addition to foodstuffs.

Although the global consumption of wild
terrestrial species cannot be assessed acc-
urately, the amount is likely to be insig-
nificant in comparison to products from just
three domestic forms: pigs, cattle and
chickens. On the other hand, fishes from wild
sources are an important nutrient source,
and the amount harvested and consumed
is known to be greatly under-reported in
official statistics.

At world level, cereals are the most
important single class of food commodities,
providing around 50 percent of daily calorie
intake and 45 percent of protein. In contrast,
meat provides around 15 percent of the
protein, and fishery products only some
6 percent, or 15 percent of all animal protein.
In each case, when comparing sources of
calories, protein and fat in the global human
diet, just two or three commodities stand out
from a large number of commodities of lesser
importance. Rice and wheat together provide
around 40 percent of the world supply of both
calories and protein, while milk and pigmeat
are key sources of fat. calories and protein
(see Table 4.11.

At country level, a much wider variety of
plant species are important in that they make
a significant contribution to human nutrition".
Around 22 species and groups of species
provide more than 5 percent of the per capita
supply of either calories, protein or fat in at
least ten countries of the world. Notes on
the origin, uses and genetic resources in
these 22 crops are tabulated in Appendix 2,
together with briefer information on the
remaining 50 or so crops that are also
nutritionally important but to a smaller
number of countries.

Although the vast bulk of human food
supply on a global level is derived from

Table 4.1

Top ten food commodities,
ranked by percentage
contribution to global
food supply

Note: This information is
partly determined by the way
food commodities are
aggregated for reporting
purposes; for example,
fishery products collectively
provide about 6% of the
hypothetical global protein
supply, but no individual
fishery commodity is
important enough to appear
in the table.

Source: FAO food balance sheets
1997 data.

^^^HJ^^^^HJH^^HHH

Rice (milled equiv.l

Wheat

Pigmeat

Wheal

Rice (milled equiv.]

Soybean oil

Sugar (raw equiv.l

Milk (excl. butter]

Maize

Pigmeat

Rape and mustard oi

Milk (excl. butter]

Bovine meat

Sunflowerseed oil

Pigmeat

Poultry meat

Palm oil

Soybean oil

Maize

Fats, animal, raw

Potatoes

Vegetables (other than

Bovine meat

Vegetables (other than

tomatoes, onions]

Poultry meat

tomatoes, onions]

Eggs

Butter, ghee

Cassava

Pulses (other than
peas, beans]

Other

Global mean: 2 745

Global mean: 73.4

Global mean: 70.1

calories/person/day

grams/person/day

grams/person/day
J

42 WORLD ATLAS OF BIODIVERSITY

Map 4.2

Livestock breeds:
numbers and status

Color represents the
number of mammal breeds
in each country. Recording
of breeds is incomplete
globally. Pie charts
represent the proportion of
all mammal breeds
associated with each FAO
region assessed as
threatened Igrayl or extinct
Iblackl.

Source; Charts calculated from FAO
World Watch List for Domestic Animal
Diversity \3rti edition. 2000); country
data derived from FAO Domestic Animal
Diversity Information System IDAD-ISI
database, available online at
tittp://www fao org/dad-is/ laccessed
February 20021-

Status

% threatened
% extinct

farming, a large number of people worldwide,
including many who are principally agricul-
turalists or pastoratists, make extensive use of
wild resources. In parts of Africa, bushmeat
(meat from wild animalsl supplies most of the
protein intake; similarly some 80 percent of
people in sub-Saharan Africa are believed to
rely largely or wholly on traditional medicines
derived almost exclusively from wild sources.
Indeed, traditional medicine continues to be
the source of health care for the majority of
people living in developing countries, and is
widely incorporated in primary health care
systems. Wild plants in many areas are
extremely important as famine foods when
crops fail, or may provide important dietary
supplements, and use of fuelwood and
charcoal from wild sources is almost universal
in the developing world.

A small number of people continue to
derive most of their requirements from wild
sources by hunting, fishing and gathering. In a
large sample 12201 of such societies, nearly AO
percent have a high dependence on fishing,
around one third are highly dependent on
gathering and 28 percent on hunting of
terrestrial resources". Fishing and gathering
tend to be alternative activities, both of them
complementary to terrestrial hunting. In
contemporary conditions, fishing tends to be
more important where temperature is lower
(especially in high northern latitudes! while
the converse applies to gathering. People in
such communities also derive shelter, medi-
cine, fuelwood and esthetic or spiritual fulfill-
ment from wild species, and the immense
variety of plant and animal species used has
been well documented.

Humans, food and biodiversity 43

Global food supply

National data on reported food commodity
supplies are collated by the Food and
Agriculture Organization of the United
Nations (FAO)". Given information on the food
value of these commodities, and the size of
the human population, it is possible to
estimate the average national food supply per
person. Dietary food value can be broadly
assessed in terms of energy or materials. The
former is conventionally measured in calories
Icalories/person/dayl, the latter in terms of
weight of protein or fat Igrams/person/dayl.
These standard measures take no account of
the vitamins and minerals (micronutrientsi
that are required for maintenance of full
health. The nutritional value of the human diet
varies geographically and over time, and to
a great extent according to the state of

development and purchasing power of the
people concerned.

The human population of the world
doubled between 1950 and 1990, reached
around 6 billion in the late 1990s, and will
continue to grow for decades, albeit more
slowly because of decreased reproduction
rates. At global level, there has been enough
food available in recent years to supply the
entire human population with a very basic
diet, largely vegetarian, providing 2 350 kcal
per day. In 1992 the average food supply was
estimated at 2 718 kcal daily lafter losses
during storage and cooking], made up of 2 290
kcal from plants and ^28 kcal from livestock
products". Thus the global food supply was
nominally sufficient for the calorific needs of
a population 15 percent larger than the
estimated population; a similar small annual

w WORLD ATLAS OF BIODIVERSITY

Diet class Class 1

Class 2

Class 3

Class i

Class 5

Class 6

Rice

Maize

Wlieat

Milk, meat.

Millet,

Cassava, plantain

wheat

sorghum

taro, yams

iHain Asia

Central

N.Africa,

Europe,

Sahel,

Central Africa, '

geographic

and

West Asia,

N.America,

Namibia

Madagascar

distribution

S. America

InorthI

S. America
(south!

Australia

^^^

Human

population

Ithousandsl 2 920 923

5U911

664 507

9i2 924

61867

357 361

% of world

population 54

Increase in

supply needed

by 2050 X 2.37

xL96

X2.84

xl.13

X4.82

x7.17

Table U.l

World diet classes

Notes: Population calculation
based on 1997 data, from
FAO website. Excludes Japan
and Malaysia, in anomalous
position using six-part
classification. Total
population in 1 1 7 countries in
sample: 5 462 ^93 000. Tiie
penultimate row in fact
shows the percent eacti class
forms of ttiis total population
In the sample countries; this ■
is assumed to be an
acceptable surrogate for the
total world population.

surplus has existed since the 1970s". Talking
these aggregated global data at face value, It
appears that a more than sufficient amount of
food has been produced annually during the
past two decades to maintain the world's
human population. Global food supply has
doubled since the 19i0s; the Increases in
supply over the last two decades are attri-
buted mainly to Increased productivity (69
percent] and secondarily to an increase in
production area 131 percent).

Each year during the past two decades,
between 850 million and 900 million people
have been undernourished". Given that there
has been sufficient food available in the world
overall, undernourishment must be an effect
of unequal access to the appropriate amount
and type of food. Unequal distribution is
evident at macro and micro scales: some
countries are more productive and richer in
resources than others and, whether at nation-
al or village level, high-status social groups
secure better diets than others, compounded
in some cases by gender differences. Poverty
Is a key cause of undernutrition, and often
also an effect of it, forming a self-reinforcing
cycle from which It is difficult to escape
without appropriate outside Intervention. Put
in different terms 'food insecurity is a problem
of lack of access resulting from either
inadequate purchasing power or an Inade-

quate endowment with the productive
resources that are needed for subsistence'".
Although the 1996 World Food Summit in
Rome called for a 50 percent reduction In the
global number of chronically malnourished
people by 2015, the absolute number of
people affected remains high, and if trends
continue this target will not be met".

Regional variation

The global average diet' Is a simplifying
abstraction that ignores an enormous amount
of regional, national and local variation in food
sources and in supply. Nor does it take account
of micronutrlent availability (vitamins, miner-
als, trace elements). I.e. substances that do
not directly contribute to energy or protein
Intake but which are nevertheless essential
elements of a healthy diet.

The human diet can be assessed in
several ways. The FAO devised a simple
classification to serve as the basis for an
analysis of food requirements In relation to
population growth". In this scheme, diet is
assessed In terms of calorie sources as
recorded at national level In the FAO food
balance database, and countries with similar
diet structures are clustered together In one
of six diet classes (Table 4.2, Map 4.3). Each
class is named after the food product that
best characterizes the diet (although this is
not necessarily the principal calorie source).

The countries In each food class tend to
share broad demographic features. Rice
countries (class 1) have high population
densities, higher than average mortality rates
and little diet diversification. Maize countries
(class 2) generally have population densities
near the world average and low mortality,
especially Infant mortality. Wheat countries
on average have low population density, but
this masks serious land and water shortages
in many. Class U countries, with a diet
described as mixed 'mllk-meat-wheaf,
include the worlds most highly developed
nations, with fertility, mortality and popu-
lation growth rates well below the global
average. Millet countries (class 5) In general
tend to have high population growth, high
fertility and low life expectancy; the diet
provides only a marginal surplus of energy

Humans, food and biodiversity 45

supplies over requirements. Diet class 6
countries contain the hunnan populations
most at risk from food Insecurity. The diet is
characterized by roots and tubers, and on
average does not provide basic energy
requirements. These populations show high
fertility, high mortality and a rapid grovirth in
numbers. Poverty is vi^ldespread and the
infrastructure weak, but there are significant
reserves of under-exploited arable land.

These major diet patterns have been used
to help characterize the improvements to
production and food supply that may be
needed in order to meet the per capita energy
needs of the population projected to exist
In 2050. In terms of diet class, availability In
class k could remain little changed, supply
in classes 1, 2 and 3 would need to approxi-
mately double, and classes 5 and 6 would
need at least a fourfold increase.

In some cases, food security may be
better ensured by diversification rather than
Increasing yields. The limitations of agri-
culture based on uniform varieties and a high
Input of fertilizers and pesticides become
more acute for farmers who rely on poor
resources or marginal environments. By
growing diverse and locally adapted crops
farmers can bring about greater security in
food production and more efficient use of
limiting resources. Many traditional agri-
cultural systems manage to varying degrees
a high diversity of both cultivated and wild
food species (Table A. 31, and in a sense reduce
the distinction between wild plants and
domesticated crops. Strong selection pres-
sures exerted by natural processes and
humans over several millennia and wide geo-
graphical areas have resulted In thousands of
varieties within most crop species.

High agricultural diversity can not only
provide insurance against crop failure in
difficult agricultural environments, but tends
to have nutritional benefits. Transition to
more uniform diets, with high Intake of fats
and sugar, has resulted In declining nutri-
tional status among numerous indigenous
groups*"'^ Low dietary diversity Is associated
with micronutrlent deficiency, a problem far
more common than general protein-energy
malnutrition, and particularly prevalent In

children, pregnant women and breastfeeding
mothers. Diverse cropping systems In back-
yard and home gardens, whether rural or
urban, can lead to direct Improvement in
family nutrition and in some cases provide
cash Income. Even a small mixed vegetable
garden is capable" of providing 10-20 percent
of the recommended dally allowance of

Table 4.3

Examples of diversity in

agricultural systems

Source; Multiple sources

Brazil

Kayapo

Over 45 tree species planted for food or to attract game;
86 varieties of food plants grown

Ecuador

Slona-Secoya

Major staples: 15 varieties of manioc, 15 of plantain,
9 of maize; pre-1978 traditional gardens yield 12 300 kilos of
food or 8.8 million calories; 72% calories and U.8% protein,
22.2% fats, 90.9% carbohydrates; po5t-1978 horticulture
provides 67.8% calories, 10.2%, purchased

Indonesia

Java

500 species in home gardens In a single village; i^^m
ability to support 1 000 people per km' ^^H

Indonesia

West Sumatra

6 main tree crops, many vegetables and fruit;

53 species of wild plants also protected and harvested <

Kenya

Bungoma

100 species of fruit and vegetables; 47% households '
collect wild plants, 49% maintain wild plants in gardens i

Kenya

Chagga

Over 100 species produced in gardens

l«lexlco

Huastec

Over 300 useful species found In managed forest plots \
called te'lom, 81 food species '

Mexico

Migrant rural
community
In southeast

338 species of plants and animals In home gardens, '
including 62 species of wild plants

Papua New
Guinea

Gidra

Approximately 54% calories and 82% of protein come
from non-purchased sources (wild, sago, coconut gardens!

Peru

Bora

22 varieties of manioc and 37 tree species are planted;
118 useful species found in fallow fields

Peru

Santa Rosa

168 species Identified in 21 home gardens

Philippines

Hanunoo

System of intercropping with 40 crops in a single field;
over 1 500 plants considered useful, of which 430 may be
grown in swiddens

Sierra
Leone

Gola forest

Of food items 14% are hunted, 25% are from fallow land,
8% are from plantations, 19% are from farm, swamp or
garden, 21% are from streams and rivers, 13% is bought
or given

Ttialiand

Lua

1 1 varieties of food plants and 27 wild food plants are
found in swidden fallows

46 WORLD ATLAS OF BIODIVERSITY

^^^^BNd. of

No. of individuals

BlomaHlkUosT^^H

Per km' Globally J
of habitat -I^^H

Per km'
of habitat

Globally

Bacteria" 400 000

7.84x10-'
-1.18x10"

4x10"
-6x10"

1.39x10'
-2.14x10'

7.06x10"
-1.09x10=

Coliembola" b 500

2.4x10'

2x10'=

5x10'

Termites"" 2 760

2.3x10'

2.4x10"

1400

1.44x10"

Antarctic krill" 1

1.43x10'

5x10"

4.29x10=

1.5x10"

Birds" 9 946

1 600 - 3 200

2x10"
-4x10"

Eleptiants"" 2

0.07-0.1

4.26x10=
-6.31x10=

85-126

5.34x10'
-7.29x10'

Great whales"" 10

<0.01

2.83x10'
-3.6x10*

52-65

1.89x10"
-2.33x10"

Domestic livestock
(excluding pets)" ca 1 5

7.3x10"

Humans 1

6x10'

3.9x10"

Humans plus
livestock

11.2x10"

Table 4.4

Number of individuals and

biomass, selected organisms

Notes; Calculations based on
data in sources cited after
group names. Biomass is
estimated dry mass
standardized on 30% wet
mass. Number of birds per
km- estimated by dividing
global bird population
(2-4 X 10") by land area minus
area of extreme desert, rock,
sand and ice lapprox.
125 million km'l. Estimates of
abundance for Coliembola
represent minimum global
totals. Whale estimates
presume a sex ratio of 1:1.
Mean Asian elephant mass
estimated at 3 500 kilos; range
estimated at 500 000 km'.
African elephant mass
estimated at 4 250 kilos.

protein, 20 percent of iron, 20 percent of
calcium, 80 percent of vitamin A and 100
percent of vitamin C.

HUMAN NUMBERS AND IMPACTS
Human population size

Information on early human population
numbers is based heavily on inference from
circumstantial evidence, and remains on an
uncertain footing even when written historical
material becomes available in some abun-
dance for the past few hundred years.

Highly speculative estimates based on
extrapolations of population densities of great
apes=' and on studies of contemporary human

hunter-gatherers== indicate that the global
late Pleistocene human population may have
been between 5 million and 10 million. It
seems likely that any increase in human
population up to then had been a result of
increasing the total area occupied by the
species, rather than by any major increase in
population density in already occupied areas.
Information on human population size in
historic times is fragmentary, and populations
lacking written records are likely to be
inadequately represented.

One analysis'"' distinguishes three main
phases of population change. First is a cycle
of primary increase in Europe, Asia and the
Mediterranean brought about by the spread
and further development of Neolithic
agriculture, which appears to have allowed a
great increase in population density. At the
start of the Iron Age in Europe and the Near
East, some 3 000 years ago, the world
population may have been doubling every 500
years, and the total probably reached 100
million around this time or soon after. Growth
appears to have slowed to reach near zero by
around year 400, possibly because the limits
of then current technology had been reached.
After the Dark Ages in Europe, a second
growth cycle began around the 10th century in
Europe and Asia, with numbers rising to a
peak of around 360 million during the 13th
century, followed by a slight fall. The global
population then increased slowly until the
19th century, when an increasingly rapid
rise began as a result of revolutionary
developments in agriculture, industry and
public health [Figure 4.1|.

Crucially, the rate of global population
growth peaked during the late 1960s: it was
then at just over 2 percent per annum, but
is now about 1.7 percent. The absolute
increase per annum has also peaked; it was
around 85 million more people per annum in
the late 1980s and is now about 80 million.

12 000

11 000

10 000

9 000

iOOO

7 000 Years ago

Humans, food and biodiversity 47

Such trends suggest that the present global
total of some 6 billion may not Itself double,
as all previous totals have done. The medium
variant of the current UN long-range forecast
suggests the total in 2050 may be 9.3 billion".
Although several countries in Africa have
yet to shift to lower reproduction rates, thus
making a further doubling quite possible, it
may be that 'children born today may be
thinking about their retirement at a time
when the global population count will have
stabilized - or even begun to decline".

The exponential rise in abundance of a
single species, to a position of global eco-
logical dominance, in the sense of using a
disproportionate share of natural resources,
is without known precedent in the history of
the biosphere. This was not achieved without
significant, often adverse, effects on the
environment, many of which stem from the
agricultural activities required to maintain
human numbers. There is no single species in
which so many individuals, of such large body
size, are distributed so widely over the planet.
There appear to be few macroscopic species,
the Antarctic krill Euphausia suberba being
one of these few, in which the number of
individuals approaches the size of the present
human population. No animal of comparable
size has a population remotely similar in
number to that of humans. Biomass provides
a measure of the way in which global net
primary production or NPP Isee Chapter 11 is
partitioned. If the standing crop of domestic
livestock is added to the human biomass,
amounting to around 1 1.2 x 10" kilos in total,
then only the global biomass of bacteria as a
whole Is higher (Table A.A].

The current human population Is unevenly
spread across the land surface of the Earth.
While some areas, such as most of Antarctica,
the interior of Greenland, and hyperarid hot
deserts, have no permanent human presence,
In others human densities may locally reach

an extreme of 1 000 Inhabitants per hectare
le.g. In Calcutta and Shanghai!. Map i.A
shows the current density of human popu-
lations over the Earth, based on census
counts within administrative units of varying
size. Among the most striking features of
this map are the large areas of very high
population density in parts of China and the
Ganges-Brahmaputra lowlands, also on Java,
and the large area of high population density
extending over most of Europe.

Human activities have now made them-
selves felt throughout the biosphere, but it
might be expected that the degree of trans-
formation of terrestrial landscapes would be
related to ease of access and proximity to
population centers. Settlements, ranging in
size from villages to cities of many million
Inhabitants, are connected by networks of
paths, railways and waterways that allow the
influence of humans to diffuse far beyond
these settlements.

Map 4.5 shows the results of a GiS-based
analysis of the relative distance of points on
the Earths surface from all such human
constructions. Those points most remote can
be given a high naturalness' value II. e.
Ignoring other possible Impacts, they are likely
to be least disturbed] and, conversely, points
surrounded by a high density of human
structures can be given a low value. Although
human population density was not part of the
analysis, there are, unsurprisingly, strong
similarities between Maps A.A and 4.5.

- 5

5 000

4 000

3 000

2 000

1 000

Figure 4.1
Human population

This graph shows the long
period of many thousands
of years during which the
world human population
remained small, followed by
an exceptionally rapid rise
to a total of 6 billion at
present.

Source: Data from McEvedy and Jones ;
FAOSTAT database".

48 WORLD ATLAS OF BIODIVERSITY

Map 4.3

FAO world diet classes

A classification of country
dietary patterns, based
mainly on calorie sources.
Eacti class is named after
the foodstuff that best
characterizes the diet. The
classification of a few
countries (e.g. Japan,
Ivlalaysial appears
anomalous, and some
countries are not included
in the classification.

Source: Analysis by FAO .

Diet class

Class 1 Rice \

Class 2 Maize

Class 3 Wtieat

Class A fvliLk, meat, wheat

Class 5 Millet, sorghium

Class 6 Cassava, plantain, tare, yams

Not assigned to a class

The human share of global resources

Agriculture is a set of activities designed to
secure a greater and more reliable share of
the energy and materials in the biosphere for
the benefit of the human species, i.e. to divert
an increased amount of available energy
toward production of human bodies. Naturally
occurring plants and animals are replaced by
specially cultivated or bred varieties that can
produce nutrients efficiently from available
resources, and in a form that humans can
conveniently use. The growth and persistence
of these selected species is subsidized by
humans. In less-developed agricultural sys-
tems, this subsidy may be very small, perhaps
just the removal of competition for light or
grazing, i.e. plots are cleared or weeded and
wild herbivores discouraged. Efficiency can be
high but output is usually low. In western

J!5>.

industrialized agriculture the subsidy is
enormous. Competitors, pests and predators
are removed from vast areas (through the use
of herbicides and pesticides!; fossil fuel is
consumed to process, transport and apply any
nutrients that limit production (nitrogenous
fertilizer) and to store produce. Efficiency is
high in some respects, e.g. use of space and
labor, but much lower if all hidden costs of
fossil fuel use and waste impact are
considered. Output can be very high.

The area of land devoted to agricultural
production is now a significant proportion of
the global land surface. Five thousand years
ago, the amount of agricultural land in the
world was negligible. There is no direct
evidence from the greater part of this period
on the rate of expansion of agricultural land.
Useful historical data relate to the past few

Humans, food and biodiversity 49

hundred years, and this evidence suggests
that about 250 to 300 million hectares (ha) of
land globally were devoted to crops in 1700. At
present, arable and permanent cropland
covers approximately 1 500 million ha of land,
with some 3 400 million ha of additional land
classed as permanent pasture (this figure
includes rangeland and vi^ooded land used
for grazingl. This represents a nearly sixfold
increase in cropped land over the past three
centuries (Table i.5|. The current extent of
cropland is represented in Map 5.2. All the
cropland is used to produce domestic plant
material, and much of the land classed as
pasture, together with large parts of the
grassland and open shrubland landcover
classes in Map 5.2 (of which 'permanent
pasture' is a subset] is used to produce
domestic herbivore biomass.

Most domestic herbivores are destined to
become human biomass or to meet other
human requirements, so can be counted as
surrogate humans. The rise in livestock
numbers has been accompanied by a
decline in wild herbivores. For example, the
extinct wild ox Bos primigenius of Eurasia
and North Africa has been replaced in its
former range by domestic cattle, of which
approximately 1 360 million head exist in the
world". Similarly, the American bison Bison
bison was reduced from perhaps 50 million
before European arrival on the continent to

Cropland area
(million hal

Table 4.5

Land converted to

cropland

Source: 1700-1950 estimates from
Richards^'; 1980 and 1999 data from
FAOSTAT database Icomplete time
series from 1961 available from tfiis
sourcel.

265

537 913

m il. — Tft-ifc' -oc- , ■■

1 170

1431

1501

50 WORLD ATLAS OF BIODIVERSITY

Table 4.6

Estimated large herbivore

numbers and biomass in

Mesolithic and modern

Britain

Source: After Yalden .

under 1 000 at the end of the 19th century;
although there has been significant recovery
under management, it has been replaced
over much of its former range by domestic
cattle, of which 100 million head are present
in the United States.

Data on the fauna of the Bialowieza forest in
Poland la forest remnant with populations of
large wild herbivores and predators) have been
used'' to estimate the possible number and
biomass of wild herbivores in heavily forested
Mesolithic Britain. These estimates were
compared with current herbivore populations
in Britain. Indications are that there are now ^40
times more domestic cattle than there were
wild ox and 20 times more domestic sheep
than there were wild deer; the overall large
herbivore biomass has increased by a factor of
ten Isee Table A.6]. Similar values are likely
to apply to other heavily populated industria-
lized countries. Not only are domesticated
mammals far more abundant than their wild

mSi«kS4)^aWi86rii£^i»tti&^fiUrWi^ir.V/.7\ .

><«;;is"-^*ij<v..

.'v..:~'">, '4H^^^H

Mesolithic Britain

Modem Britain ^^^^|

No.

Kilos

No. Kilos S

Sheep

20 364 600 916 407 000

Wild boar

Sus scrota

1 357 740

108 619 200

Pigs

853 000 127 950 000

Wild ox

Bos primigenius

99 250

39 700 000

Cattle

3 908 900 2149 895 000

Red deer

Cervus elepbus

1 472 870

147 287 000

Red deer

360 000 54 000 000

Roe deer

Capreolus capreolus

1083 810

21 676 200

Roe deer

500 000 10 500 000

Moose

Akes alces

67 490

13 498 000

Introduced deer 111500 5 017 500

Feral sheep

and goats

5 700 256 500

^^^H

Domestic herbivore

^^^M

total

25126 500 3194 252 000

^^^*

Wild herbivore

'^^

^^^■a

total

977 200 69 774 000

4 as; 1M^^

^^00 3 264 02400^^

relatives ever were, the latter are in many
cases extinct or near extinction.

As mentioned in Chapter 1, calculations at
global level and in a study of one country
[Austria! estimate the proportion of terrestrial
global production appropriated or diverted for
human use at 30-50 percent (depending on
how much change beyond direct harvesting is
incorporated). One approach to assessing the
effects of energy appropriation has used the
empirical relationship between energy and
species number". There is much evidence
suggesting that at a range of spatial scales
and within different taxonomic groups, the
diversity of species present in an ecosystem
tends to be positively correlated with the
amount of energy available [see Chapter 5).
Accepting the empirical evidence for this
relationship, it has been argued that the
number of species present will decline if the
amount of energy available for use declines.
At global level, a conservative estimate using
this relationship predicted that 3-9 percent of
terrestrial species would be extinct or
endangered by the year 2000". The evidence
from the Austrian study'" is consistent with
the species-energy relationship: the curve
developed by Wright" predicts that with 41
percent of potential NPP at country level now
being appropriated by humans, 5-13 percent
of species should have been extirpated from
the country; in fact, 8 percent of birds and 7-
14 percent of reptiles have been lost.

The significance human resource appro-
priation has for other species, and for
biosphere function, in part depends on the
extent to which the resources are limited in
availability. Space clearly is limited, and it may
be that the main effects of appropriation will
be exerted through direct changes to land-
cover, and diminished availability of material
resources for wild species, rather than
through a diversion in energy flow.

SPECIES EXTINCTION AND HUMANS
Tracking extinction

Change in biological diversity has principally
been assessed in terms of declining popu-
lations, species, communities and habitats.
There has always been special concern about
extinction, because this is a threshold from

Humans, food and biodiversity 51

whicfi tfiere is no turning back. As discussed
in Chapter 3, ttie extinction of species is
natural and expected, and self-evidently ttiere
fiave always been species at risk of extinction.
It seems very likely, however, that recent and
current extinction rates are considerably
higher than would be expected without the
influence of humans.

For several reasons it is difficult to record
contemporary extinction events with precision.
The species involved may well be unknown.
Even if they have been discovered and named,
they may be too small to be noticed without
special sampling procedures. The entire
process of decline and eventual extinction
may take place over many years or even
centuries in the case of particularly long-lived
organisms such as many trees. The near-
terminal stages in the process of species
extinction are unlikely to be observed. Where
such observations have been possible, it is
because the species has been destroyed in
part by unusually extreme hunting pressure
(e.g. the passenger pigeon Ectopistes
migratorius] or extreme ecological events
le.g. extinction of many native land snails in
French Polynesia and Hawaii following
introduction of the predatory snail Eugiandina
rosea], and has been the subject of sufficient
interest to be closely monitored.

In other cases, positive evidence of
extinction is lacking. Typically, many years
elapse before sightings of a species become
sparse enough to generate concern, and
many more years are likely to pass before
negative evidence li.e. failure to find the
species despite repeated searches] accumu-
lates to the point where extinction is the most
probable explanation. In other words, unless
circumstances are exceptional, monitoring of
recent extinction events has a resolution limit
measured in years or decades. This is why it is
not possible to state with precision how many
species became extinct in a given month, year
or even decade, nor to predict exactly how
many species, let alone which ones, are going
to become extinct this year or decade or
century. The search effort and chance both
play a part in determining whether a species
not seen for decades is rediscovered, as
shown by the occasional new encounter with

Standing out as some good news against a background of widespread species depletion, a
small but significant number of species have been considered extinct but rediscovered after
a gap of several decades. In some cases, prolonged directed searches have been carried out,
but others have emerged by chance. Sometimes a 'rediscovery' is in part a consequence of
new systematic work confirming the species status of long-neglected populations, e.g.
differentiation of the pygmy mouse lemur Microcebus myoxinus from other mouse lemurs
on Madagascar The coelacanth Latimeria is an extreme 'Lazarus taxon', being the living
representative of an entire order (CoelacanthiformesI of early fishes thought to have become
extinct some 80 million years ago. There are at least two population groups, one off
southeast Africa [L. chalumnae] first discovered in 1938, and one in Indonesia (named
L menadoensis] discovered in 1998.

• Bavarian pine vole Microtus bavaricus: an alpine species believed lost when a hotel was
built near its single known locality, but recently rediscovered nearby in the Austrian Alps.

• Fiji petrel Pterodroma macgiltivrayi: known by one specimen collected in 1855, but
regarded as extinct until a bird flew into a researcher's headlight one night in 1984.

• Jerdon's courser Rbinoptilus bitorquatus: known by just two museum skins and last
recorded in 1900, until rediscovered in 1986 in a patch of scrub forest in southern India.

• Jamaican iguana Cyclura collei: believed to have gone extinct during the 19'lOs but
rediscovered in 1990 in the remote Hellshire Hills.

• The Cranbrook pea Gastroiobium lehmannii: endemic to Western Australia where last
collected from the wild in 1918 and since listed as extinct; rediscovered in 1995.

species once feared extinct (see Box 4.21.
Hidden survivors are even more likely in plants,
which may have propagules that can remain
viable but unseen for extremely long periods.
Extinctions in the recent past are likely to be
recorded with significant accuracy either where
circumstances favor preservation of hard
remains in good number (e.g. in caves, potholes
or kitchen middensi or where early naturalists
recorded the fauna or flora with sufficient care
that they set a firm baseline against which the
composition of the modern biota may be
assessed. The detailed record of bird extinc-
tion in Hawaii is a result of the former
circumstance, and the record of mollusk
extinction on many islands a result of the latter.

Table i.7

Late Pleistocene extinct

and living genera of large

animals

Source: Adapted irom Martin .

^H[ Extinct wRhin Still

Total

^^^; last 100 000 extant

genera

of maximum 1

Africa 7 42
North Amenca 33 12

49
45

14 No peak

73 11 000-13 000 years ago

South America 46 12

79 11 000-13 000 years ago

Australia 19 3

86 ca 50 000 years ago

52 WORLD ATLAS OF BIODIVERSITY

Map U.U

Human population density

The relative density of
human population based on
census data relating to
administrative units of
various sizes.

Source- CIESIN, gndded population of
the world, version 2. data available
online at

http://sedac.cie5in, columbia.edu/
plue/gpw/ [accessed April 20021

Population density

High

Low

Early human impacts on biodiversity

A wide range of factors affects the frequency
of occurrence of particular species in the
fossil record, of which the abundance of that
species in life is only one. There is not
necessarily therefore any direct relationship
between the former and the latter, so that
deducing past changes in abundance of
species from the fossil record is a problematic
exercise. Cataloging, though not dating,
extinctions is rather less contentious, al-
though even this may be problematic as
evinced by the existence of so-called 'Lazarus'
taxa Ithose presumed extinct that are
rediscovered alive.

One of the unusual features of the
Quaternary period (the Pleistocene and
Holocenel has been the disproportionately
high extinction rates in the largest ter-

'-W'

'^^■J^-

restrial species, particularly mammals and
birds (Table 4.71. These species are gen-
erally referred to as the 'megafauna', often
defined as those with an adult mass of UU
kilos or more, although the term has not
been used consistently.

The extinct American fauna include such
well-known genera as the sabretooth cats
Smilodon. giant ground sloths Eremotherium,
glyptodonts Glyptotherium and mammoths
Mammuthus, as well as a number of
scavenging and raptorial birds, including the
giant Teratornis and Cathartornis. Those in
Australia include the marsupial equivalents
of rhinoceroses (family Diprotodontidael
and lions \Thylacoleo\, giant wombats
(Phascolonus, Ramsayia and Phascolomis],
the large emu-like Genyornis and the giant
monitor lizard Megatania. In all, some 40

Humans, food and biodiversity 53

•'^^^

■.yv

~^-

species of tfie larger Australian land mam-
mals, reptiles and birds became extinct
across the entire continent about 46 000 years
ago", approximately 10 000 years after the
first known human colonists.

These extinctions have been followed by a
similar series on islands during the Holocene.
On New Zealand, the moas (giant flightless
ratite birds in the family Anomalopterygidael
became extinct after the first humans came
to the islands. On tvladagascar, two endemic
hippopotamus species Hippopotamus lemerlei
and H. madagascariensis, the elephantbird
Aepyornis maximus, and a number of large to
very large lemur species all appear to have
died out 500 to 900 years ago. Similarly on
the Caribbean islands, a number of large
mammals, including several ground sloths
(order Xenarthra, family Megalonychidael

appeared to have survived until human
occupation but to have died out at some point
since then.

The precise causes of all these extinctions
have been the subject of endless debate,
which centers chiefly on the role of humans.
At one extreme lies the 'blitzkrieg' hypo-
thesis, applied particularly to the apparently
sudden collapse (i.e. over a few hundred
years! of the North American megafauna at
the hand of humans. At the other are those
who maintain that in most, if not all, cases
the impact of early humans was negligible
and climate change, particularly increasing
aridity, was the cause.

Several 'eatures of the phenomenon seem
to point persuasively to humans having played
a pivotal role in most, if not all, of these
extinctions. The most compelling is their

54 WORLD ATLAS OF BIODIVERSITY

Map 4.5

Terrestrial wilderness

The wilderness value of any
given point is essentially a
measure of remoteness
from human influence,
assessed on the basis of
distance from settlement,
access routes and
permanent manmade
structures.

Source: GIS analysis by R, Lesslie lANUI.
method developed for the Australian
Heritage Commission.

Wilderness level

High

Low

timing. In each case the arrival of humans
seems to have preceded the major spate of
extinctions (in as much as these can be
dated], with no or very few such extinctions
having been recorded prior to human arrival
(compare fvjap 4.1 and Table 4.7). The
cumulative weight of this coincidence is
difficult to counter, and is supported by
population modeling. For example, new
archeological evidence suggests that the first
Polynesian settlements in New Zealand date
from the late 13th century, and moas were
becoming scarce by the end of the Hth. This
evidence, coupled with a mathematical model
of population and predation, indicates that all
11 moa species were driven to extinction
within 100 years of human arrival". Similarly,
a computer simulation" of population
dynamics of humans and large herbivores in

North America accurately models mega-
faunal extinction in accord with archeological
evidence for significant human arrival around
13 400 years ago and the first wave of
extinction starting some 1 000 years later

The situation is somewhat different in Africa
and Eurasia, where megataunal extinctions
were relatively few, and were spread out over
the whole of the Pleistocene. In Africa the peal<
was in the lower Pleistocene |21 genera
extirpated from the region between 1.8 million
and 700 000 years ago compared with nine
between 700 000 and 130 000 years ago, and
seven later than 130 000 years ago]. It is
noteworthy that this more gradual, earlier and
less extreme pattern of extinctions has typified
the region where humans evolved.

It is difficult to formulate an entirely
climate-based model of extinction that can

Humans, food and biodiversity 55

account for tfiis asynctironicity - outside Africa
and Eurasia, ttiese species survived a series of
climatic cfianges at least as extreme as those
they faced at the start of the Holocene. The
recent studies of Genyornis newtoni and other
megafauna in Australia"" seem to indicate a
widespread and largely synchronous dis-
appearance from a wide range of habitat types
during a period of relative climatic stability,
strongly implying that some other agent was
responsible. It is also noteworthy that the
fossil and archeological evidence indicates
that, on continents at least, these extinctions
were not matched by parallel extinctions of
smaller species (for example, as far as Is
known no Insect species at all became extinct
in Europe In the entire Pleistocene'!. If climate
change were the cause, it would be expected
that these species would be at least as affected

as larger ones as In general their opportunities
for long-range dispersal and migration are
much more limited.

Even if humans are accepted as the major
agents in these extinctions, evidence for the
mechanisms Involved in most cases remains
elusive. It seems likely that extensive use of
fire, and direct hunting of a fauna that had
evolved In the absence of humans and was
therefore unlikely to recognize them as
potential predators, may have been sufficient
cause to exterminate the large herbivores.
The large carnivores and scavengers may
then have suffered population collapses
owing to the disappearance of their prey base.

Extinction in modern times

From the relatively sparse evidence that is
available, it appears that amongst animals

56 WORLD ATLAS OF BIODIVERSITY

Map 4.6

Vertebrate extinctions

since AD1500

An indication of the number
and former occurrence of
recently extinct mammal,
bird and freshwater fish
species. Each of these three
vertebrate classes is
represented by a differently
colored symbol, sized
according to number of
extinct species. Only
extinctions that are fully
resolved according to firm
Committee on Recently Extinct
Organisms ICREO) criteria
are covered. The exception is
Lake Victoria, indicated by a
blue circle: some reports
have suggested that up to a
third of the 500 or so
endemic fishes are extinct,
but others note that
available evidence is
inconclusive, in many cases,
including most islands and
lakes, the position of the
symbol indicates last record
or core of former range.
Where several species
ranged more widely over a
country, the symbol is
positioned at the centre of
that country.

Source; Based on several sources; see
Appendix k.

P^^^'^'

Vertebrates

•

Birds

•

Fishes

•

Mammals

Possibly-extinct

Lake Victoria fishes

Number

of species

10-25

•

6-9

•

3-5

•

1-2

more than 300 vertebrates, including at least 60
and possibly more than 80 mammals, more
than 120 birds and around 375 invertebrates,
have become extinct during the past AOO years
(see numerical summary in Table UZ and list of
extinct vertebrates in Appendix 41. Data for
plants are much more equivocal, owing in part
to the uncertain taxonomic status of many
extinct plant populations. One source" lists
some 380 extinct plant taxa and a further 370 or
so classified as extinct or endangered (these
include a number of infraspecific taxa and
a number that although believed extinct in
the wild are extant, and sometimes abundant,
in cultivation].

Because mammals and birds tend to be
relatively well recorded, and because they leave
recognizable macroscopic skeletal remains, it
is principally among these groups that known

ij^N)

extinctions may be reasonably representative of
actual extinctions. In these groups the known
extinction rate over the past 400 years, based
on data in Appendix 4, averages out at around
20-25 species per 100 years.

A crucial question then, is how this
observed extinction rate compares with some
hypothetical or expected background extinc-
tion rate. It is of course impossible to derive
such a rate from observation of the modern
world, as this has already been highly
modified by human activity The only reason-
able comparison is thus with historical
records, for which we must turn to the fossil
record, discussed in Chapter 3. Although
extinction rates have evidently been highly
variable during the history of life on Earth, it
seems that the average persistence time of
species in the fossil record is around 4 million

Humans, food and biodiversity 57

'Ml

••

years. If 10 million species exist at any one
time, on tliis basis the extinction rate would
amount to around 2.5 species annually.
However, it is unclear fiov*/ representative the
fossil record is of species as a whole. It is
likely that many rare species or those with
very restricted distribution never appear in it
at all. These rare species may almost by
definition be expected to be inherently more
prone to extinction than the species that are
recorded and may therefore be expected to
have a lower persistence time. This would
mean that average species duration was less
- perhaps much less - than A million years
and the actual extinction rate in geological
time considerably higher than the rate
observed in the fossil record.

Applying a mean persistence time of
A million years to birds and mammals land

assuming some 10 000 species of the former
and around 5 000 of the latter], the back-
ground extinction rate would be around one
species every 500 years and 1 000 years,
respectively, so that current rates would be
some 100 or 200 higher than background.
Even if background rates in these groups were

Mammals

Birds

Reptiles

Amphibians

Fisfies

Insects

Mollusks

Other invertebrates

83
128

21
5

72
291

Table 4.8

Numbers of extinct animal

species according to lUCN

Note: Alternative criteria used
by the Committee on Recently
Extinct Organisms ICREOI
result in a lower number of
mammals and fishes being
regarded as certainly extinct
than IS given in this table (see
full list in Appendix 4).

Source; Hilton-Taylor .

58 WORLD ATLAS OF BIODIVERSITY

Map 4.7

Threatened mammal
species

Color represents number
of globally threatened
mammal species in each
country in 2000. Pie charts
represent the proportion
of the mammal fauna
assessed as threatened at
the national level in a small
sample of countries. This
IS a highly generalized
comparison because of
differences in status
assessment methods.

Note: To reduce ambiguity
Alaska (United States] has for
the purposes of this map
been assigned to the same
class as adjacent Canada
rather than the conterminous
United States.

Source: Global data from Hilton-
Taylor , country data (rom selection ol
national Red Data books.

National status

( A — % threatened

ten times this, the currently observed
extinction rates would still be 10-20 times
those expected. It seems therefore that, even
if a high background extinction rate is
postulated, recent extinctions are still much
higher than might be expected (the alternative
explanation is that the background rate was
higher still). This elevated rate is particularly
noteworthy as the Holocene appears to have
been a period of relative climatic and
geological stability, in which extinction rates
might have been expected to be low.

Most known extinctions have occurred on
islands, including 42 of the 61 resolved
mammal extinctions 168 percent], and 105 of
the 128 bird extinctions (82 percent]. Reasons
for the former are probably twofold. First,
island species do appear to be particularly
extinction-prone, by virtue of their limited

ranges and usually small population sizes,
and also because they have often evolved in
the absence of certain pressures (e.g. ter-
restrial predators, grazing ungulates!; if faced
with these pressures (usually through human
intervention] their populations may collapse
completely. Second, it is much easier to arrive
at some certainty that a given species is no
longer present on an island of limited extent
than that it has disappeared completely from
a continental range, where the limits of its
range were probably uncertain to start with.

Most known or probable continental
extinctions have been among freshwater
organisms, particularly fishes and moUusks.
Ivlany freshwater biota appear to have the
characteristics of island organisms, in that
they have limited and highly circumscribed
ranges and they are similarly often sensitive to

Humans, food and biodiversity 59

ao Tome
d Principe

■^f^m-f

Soutln Africa

Australia

external pressures (e.g. introduction of preda-
tory fisti species, complete fiabitat destruction
througfi drainage or dam construction).

Figure 4.2 represents the number of
accepted extinction events among mammals,
birds and fisfies in eacfi third of the five
centuries since AD1500 (see also Map 4.6|.
The information in Appendix 4 allows some
elements of this broad global picture to be
disentangled:

• early extinctions among the remaining
archaic fauna of large islands, notably mam-
mals on Hispaniola, Cuba and Madagascar;

• somewhat later extinctions, especially
among ground birds, on many isolated
small islands, such as St Helena, the
Mascarenes and others, during European
colonial consolidation;

I 20-

Fishes

Birds

Mammals

20 - -m

Figure 4.2

Vertebrate extinctions by

period since AD1500

This graphic is derived from
information in the 'Period'
column of the list of extinct
vertebrates given in
Appendix 4. 'Early' refers to
the first four decades of a
century, 'mid' to the next
three decades, 'late' to the
final three decades. Only
the extinction events
regarded by Committee on
Recently Extinct Organisms
(CREOI criteria as 'resolved'
are shown, and only the
three groups recently
assessed are represented.

Source: Derived from Appendix 4.

early mid late early mid late
16th century I 17th century

early mid late early mid late early mid late
18th century I 19th century I 20th century

60 WORLD ATLAS OF BIODIVERSITY

^^^■l^^^pl

f No. of

nSH

|P^% breeding

breeding

globally

species

1^^^^^^^

bird species

threatened

threatened ^^

Continental countries

Brazil

U92

113

China

1100

India

923

Colombia

1695

■ Peru

Small islands

1538

French Polynesia

Solomon Islands

163

Mauritius (and Rodrigues)

Sao Tome and Principe

L Saint Helena and dependencies

Table A.9

Island diversity at risk:

birds

Note: The first five rows are
the continental countries
(Indonesia and Philippines
excluded! with most
threatened bird species; the
other five rows are the small
island groups with most
threatened bird species.

Source: BirdLife International .

• increasing evidence of 20th century
extinctions among fisfies in many countries,
and among marsupial mammals on the
Australian mainland, with continuing losses
of island species everywhere.

The apparent reduction in extinction rate from
the late 19th century onwards may be in part
attributed to management action designed to
maintain highly threatened species, and there
is indeed good evidence that populations of
a small number of target species have
recovered significantly, but it is sure also to
reflect the difficulties of observing and
documenting extinction events. As mentioned
above, many years may pass before a
possibly extinct species can be treated with
certainty as extinct, and the declining
extinction rate toward the end of the 20th
century is most probably in part an artifact
of this monitoring process.

Clearly, in view of our very incomplete
knowledge of the worlds species, and the fact
that only a minute proportion of living species
are being actively monitored at any one time,
it is extremely likely that more extinctions are
occurring than are currently known. Indeed,
most predictions of present and near-future
extinctions suggest extremely high rates.
Most are based on combining estimates of
species richness in tropical forest with
estimates of rate of loss of these forests, and

predict species extinction on the basis of the
general species-area relationship (which
predicts a decline in species richness as area
declines, see figure 5.1). It is widely believed
that the great majority of all terrestrial
species occur in tropical forests, and most of
these species will be undescribed arthropods
(notably beetles). At present rates of forest
loss, it has been predicted that between 2 and
8 percent of forest species will become
extinct, or committed to extinction, between
1990 and 2015'"'. Depending on whether
higher or lower estimates of tropical species
richness are used, extinction at this rate could
entail loss of up to 100 000 species annually.
As a cautionary note, it should be observed
that few extinctions to date have actually been
recorded in continental tropical moist forests,
although monitoring species in these habitats
presents great difficulty.

Threatened species assessment

Except for species lost as a result of random
environmental factors, extinct species must
largely be drawn from a pool of species that
could be assessed as in decline or at risk, all
of which face eventual extinction if negative
trends or threats to their populations are not
reversed. Various national and other programs
have developed methods to assess the relative
severity of risks faced by species, and to label
species with an indicative category name.
Conservation activities can then be prioritized
on the basis of relative risk, taking account of
other relevant factors, such as feasibility, cost
and benefits, as appropriate.

The system" developed by the Species
Survival Commission of lUCN-the World
Conservation Union (lUCN/SSC) and collabo-
rators in conjunction with its Red Data Book"
and Red List Programme" has been designed
to provide an explicit and objective frame-
work for assessment of extinction risk, and to
be applicable to any taxonomic unit at or
below the species level, and within any
specified geographical or political area. To be
categorized as threatened, any species has to
meet one of five sets of criteria formulated to
permit evaluation of all kinds of species, with
a wide range of biological characteristics.
The criteria are defined on population

Humans, food and biodiversity 61

"W/j:

reduction, population size, geographic area
and pattern of occurrence, and quantitative
population analysis.

The size and connectedness of different
populations of a species influence the like-
lihood of its survival. In general, small Isolated
populations will be more sensitive than larger
connected ones to demographic factors (e.g.
random events affecting the survival and
reproduction of individuals! or environmental
factors (e.g. hurricanes, spread of disease,
changes in the availability of foodl. Islands tend
to have a much higher proportion of their biota
at risk than continental countries because they
start with many fewer species, all or many of
which face the risks associated with a small
range size (see Table 4.9 for an example using
bird data). Biogeographic theory based around
the species-area relationship, supported by
much empirical evidence, predicts that each
'habitat island' created by fragmentation
of a continuous habitat area will come at
equilibrium to contain fewer species than
previously. Human activities everywhere tend
to promote fragmentation of natural and often
species-rich habitats (e.g. primary tropical
forest or temperate meadow grassland] and
the spread of highly managed species-poor
habitats (e.g. eucalypt plantations or cereal
croplands!. As a result, many species occur
in just the kind of fragmented pattern that
Increases the risk of extinction.

Recent declines

Reduction in population numbers, or com-
plete loss of a species from a site or an
Individual country (often termed extirpation!
are far easier to observe than global species
extinction, and appear liable to occur
wherever humankind has modified the en-
vironment for its own ends. The conservation
status of most species is not known in detail,
and this certainly applies to the many million
as yet undescrlbed species, but two large
animal groups - the mammals and birds -
have been comprehensively assessed and
may be representative of the status of
biodiversity in general. Approximately lU
percent (1 130] of the world's mammals and
12 percent (1 183] of the worlds bird species
are regarded on the basis of lUCN/SSC

criteria as threatened (see Table 4.10].
Proportions are a great deal lower in other
vertebrates, but none of these has been
assessed fully. Empirical observations such
as these give sufficient grounds for serious
concern for biodiversity maintenance, regard-
less of any hypotheses that have been pro-
posed regarding the future rate of extinction.
Interestingly, the ratio of threatened mam-
mals to threatened birds is near 1:1, as Is
the ratio of recorded, recently extinct mam-
mals to birds, giving some indication that
threatened species categories may be a
reasonably reliable indicator of proneness to
extinction, and also that mammals may as a
group be somewhat more susceptible than
birds to extinction.

Countless other species, although not yet
globally threatened, now exist in reduced num-
bers and fragmented populations, and many of
these are threatened with extinction at national
level. The significance of loss of diversity at
gene level Implied by loss of local populations
of species Is not clear, although It has been
argued that loss of resilience in response to
environmental change is Inevitable.

Table 4.10
Threatened species

Notes: Includes species
assessed as globally
threatened and assigned to
categories 'critically
endangered', 'endangered'
or vulnerable' under lUCN
criteria. Ivlosses not included
here; most plant taxa listed
were assessed for the World
list of threatened trees'" in
1998.

Source: Hilton-Taylor ; BirdLife
Internalional .

^^^r

No. of

Approx.

Threatened

%oftotafl||

^^^^^K

species In

%of

species

In group ' 1

group

group
assessed

threatened |

Vertebrates

Ivlammals

U30

100

1 130

Birds

9 750

100

1186

Reptiles

8 002

<15

296

Amphibians

4 950

<15

Fisties

25 000

<10

752

Invertebrates

Insects

950 000

<0.01

555

0.06

Mollusl<s

70 000

938

Crustaceans

ilOOOO

iOB

Others

Plants

Gymnosperms

(Conlferoptiyta, Cycadopfiy

Ginkgophytal

876

141

Angiospernis

(Anthopfiyta, flowenng

plants]

250 000

5 390

62 WORLD ATLAS OF BIODIVERSITY

Map 4.8

Critically endangered

mannmals and birds

The general distribution of
almost all the 362 mammal
and bird species categorized
as 'critically endangered',
the highest risk category, in
2000. Each circle represents
one distribution record;
some species are known
from a single point locality,
others are represented by a
cluster of localities. On this
map, a high density of
symbols can represent many
records of a single species,
or single records of many
separate species. Red and
blue symbols represent
mammals and birds,
respectively

Source: Mammals; species selection
based on Hilton-Taylor* . tjtstrtbution
researcti and mapping by UNEP-WCf^lC;
birds: spatial data provided by BirdLife
International (February 20021, further
information in BirdLife International .

^^!>

Critically endangered
o Mammals

o Birds

At global level, most of the species
assessed as threatened are terrestrial forms
(Table 4.111. The preponderance of terrestrial
species is because the great majority of
mammals and birds are terrestrial, and
little or nothing is known of the population
status of most aquatic species in most
groups. Where significant numbers of aquatic
species have been assessed, e.g. among
crustaceans and mollusks, the proportion of
threatened aquatic species rises markedly.
Among fishes, the high number of freshwater
species doubtless in part reflects the general
lack of data on marine species, but to some
extent indicates relative risk - many fresh-
water species being restricted to small and
isolated habitat patches.

Evidence of the vulnerable nature of
freshwater habitats and the risk faced by

t' \

■■»

'^ -^ I,

many aquatic groups is accumulating. For
example, in the United States, freshwater
groups are considerably more threatened
than terrestrial groups (specifically, nearly 70
percent of the mussels, 50 percent of the
crayfish and 37 percent of the fishesi".

Forest is an important habitat for a high pro-
portion of the threatened terrestrial verte-
brates. Among birds, for example, about 70
percent of the 1 186 species assessed as
threatened in the year 2000 occur in forest, and
25 percent in grassland, savannah and scrub
habitats'". Of the threatened forest birds, 41
percent occur in lowland moist forest and about
35 percent in montane moist forest™. Among
the 515 mammals regarded as threatened in
2000 that were assigned to a habitat category,
33 percent occur in lowland moist forest and 22
percent in montane formations".

Humans, food and biodiversity 63

:- %.

Numbers of globally threatened species
can be mapped at country level le.g. Map 4.7,
mammals). Maps of tfiis kind provide at best a
broad overview of ttie occurrence of sucti
species, shown as direct numbers or as a
proportion of the total number in that group in
the country. As with other country-level
biodiversity analyses, this information could
be used to help focus efforts to slow or
reverse biodiversity decline. Data with
improved spatial resolution (although plotted
in a highly simplified way! are shown in Map
i.8, representing the individual ranges of all
the mammals and birds assessed as 'critically
endangered'. BirdLife International has gen-
erated individual distribution maps of all
threatened birds. Among other applications,
these are being used to develop the first-ever
world map of all the threatened species in

an entire major group of organisms. A
preliminary version of this, as a global density
surface, is shown in Map 4.9. Being geo-
referenced rather than country based, and
at relatively fine spatial resolution, this has
considerable potential to focus bird con-
servation efforts effectively.

Table 4.11

Number of threatened

animal species in major

biomes

Notes: Counts include globally
threatened species tabulated
in each biome (lUCN
categories 'critically
endangered', 'endangered' or
vulnerable'!. Only mammals
and birds have been
comprehensively assessed.
Some species, e.g
amphibians and migratory
fishes, are counted in more
than one biome row. Plants
not tabulated; almost all
assessed as globally
threatened are terrestrial.
Inland water includes saline
wetlands, cave waters, etc., as
well as freshwaters.

Source: Hilton-Taylor .

■

Animal gn

■

■■

E
E

-0

'1.

lA
01

^»..me.ype

Q£

Marine

315

105

163

Inland water

1932

111

131

627

409

125

420

Terrestrial

3 627

1 111

1 U4

283

438

508

■'-■ 1

44 WORLD ATLAS OF BIODIVERSITY

Map 4.9

Threatened bird species

density

BirdLite International tias
developed digital range
maps of all threatened bird
species. This map is based
on distribution data for the
1 186 species assessed as
threatened in 2000. The
data are plotted as a global
density surface, representing
the number of species
potentially present in each
location. The map is the
first such treatment of any
large group of threatened
species.

Source; Density surface provided by
BirdLife Inlernational, further
information on Iflreatened birds in
BirdLife International .

Threatened bird species
density

High

Low

Proximate causes of recent declines

The continuing conversion of natural habitats
to cropland'' and other uses typically entails
the replacement of systems rich in bio-
diversity with monocultures or systems poor
in biodiversity. Habitat modification, from
agricultural conversion and a variety of other
causes, is in general the most important
factor acting to increase species' risk of
extinction. Among species assessed as
globally threatened in 2000", habitat
modification is the principal threat affecting
more than 80 percent of the mammals, birds
and plants; it is similarly predominant in
several other major groups, notably in 95
percent of threatened bivalve mollusks, and is
the main cause of loss in 75 percent of extinct
freshwater fishes".

A second major source of biodiversity loss

"'^^

(

IS the widespread introduction of species
outside their natural range where they
typically induce change at the community and
ecosystem level. The effects of alien species
are especially pronounced in closed systems
such as lakes and islands. Introduced
species, such as rats and cats, are cited as a
cause of extinction in nearly 40 percent of the
approximately 200 species where cause could
be attributed; the majority were island
forms". Seven endemic snails in French
Polynesia have been extirpated following the
late 1970s introduction of a carnivorous snail
species lEuglandina rosea] intended to
control another introduced species Ithe giant
African snail Achatina fulica], itself an
agricultural pest. Accidental introduction of
the brown tree snake Boiga irregularis to
Guam in 1968 led to decline of the entire

Humans, food and biodiversity 65

m>^''^^^

.r*7 A.

avifauna, of wfiicfi one species is now tfiougfit
extinct and one is extinct in the wild'".
Introduction of tfie Nite perch [Latss
nitoticus] to Lal<e Victoria has contributed to
the decline or extinction of nearly 200 native
and endemic cichlid fishes. Overall, Intro-
duction events probably number in the
thousands at the global level"and, although
cases involving animals have been cited to
illustrate the scope of the problem, intro-
duced plants are in many places as pervasive
and damaging.

High trade demand for certain species and
products, whether for international markets
consuming hardwoods, sea fish, live animals,
and plants and derivatives, or local markets
consuming commodities such as bushmeat
or turtle eggs, can readily push exploitation
beyond the production capacity of the

resource. The direct impact of hunting,
collecting and trade Is the second most
important threat category among globally
threatened mammals and birds, with around
35 percent In each group affected".

Rapid environmental change, such as that
associated with El Nino Southern Oscillation
(ENSOl events, can have significant impacts
on natural habitats. For example, in 1997-98
climate fluctuation associated with El Nlfio
was implicated In the persistence and spread
of fires in Brazil, Indonesia and elsewhere:
an estimated 1 million hectares of savannah
woodland burned in Brazil and a similar
area of forests in Indonesia were affected
by fire. The effect of events of this type will
be multiplied many times wherever habitats
are already fragmented and species are
depleted.

66 WORLD ATLAS OF BIODIVERSITY

REFERENCES

1 Coope, G.R. 1995, Insect faunas in ice age environments: Why so little extinction? In:
Lawton, J.H. and May, P.M. (edsl. Extinction rates. Oxford University Press, Oxford.

2 Adams, J. 1999. Sudden climate transitions during the Quarternary. Progress in Physical
Geography 23111: 1-36.

3 Van Couvering, J.A. led. I 1997. The Pleistocene boundary and the beginning of the
Quarternary, Cambridge University Press, Cambridge.

^ Petit et at. 1 999. Climate and atmospheric history of the past 420,000 years from the
Vostok ice core, Antarctica. Nature 399: 429-A36.

5 Larick, R. and Ciochon, R.L. 1996. The African emergence and early Asian dispersal of the
genus Homo. American Scientist November-December.

6 Haile-Selassie, Y. 2001 . Late Miocene hominids from the Middle Awash, Ethiopia. Nature
412: 178-181.

7 Wood, B. and Brooks, A. 1 999. We are what we ate. Nature 400: 21 9-220.

8 Gabunia, L. ef al. 2000. Earliest Pleistocene hominid cranial remains from Dmanisi,
Republic of Georgia: Taxonomy, geological setting, and age. Science 288: 1019-1025.

9 Zhu, R.X. eta/. 2001. Earliest presence of humans in northeast Asia. Nafure 413: 413-417.

10 Bermiidez de Castro, J.M. et al. 1997. A hominid from the Lower Pleistocene of
Atapuerca, Spain: Possible ancestor to Neandertals and modern humans. Science 276:
1392-1395.

11 Dennell, R. 1997. The worlds oldest spears. Wa(ure385: 767-768.

12 Cann, R.L. 2001. Genetic clues to dispersal in human populations: Retracing the past
from the present. Sc/ence 291: 1742-1748.

13 Rogers, A.R. 2001. Order emerging from chaos in human evolutionary genetics.
Proceedings of the National Academy of Sciences 98: 779-780.

14 Barbujani, G. and Bertorelle, G. 2001. Genetics and the population history of Europe.
Proceedings of the National Academy of Sciences 98: 22-25.

15 Thorn, A. et al. 1999. Australia's oldest human remains: Age of the Lake Mungo skeleton.
Journal of Human Evolution 36: 591-612.

16 Roberts, R.G. ef al. 2001. New ages for the last Australian megafauna: Continent-wide
extinction about 46,000 years ago. Science 292: 1888-1892.

17 DiUehay, T.D. 1997. Monte Verde: A Late Pleistocene settlement in Chile, Volume 1: The
archaeological context and interpretation. Smithsonian Institution Press, Washington DC.

18 Guidon, N. and Delibrias, G. 1986. Carbon-14 dates point to man in the Americas 32,000
years ago. Nature 321 : 769-771.

19 Haynes, C.V. 1984. Stratigraphy and Late Pleistocene extinction in the United States. In:
Martin, P.S. and Klein, R.G. (eds). Quaternary extinctions, pp. 345-353. University of
Arizona Press, Tucson.

20 Harcourt, C.S. and Sayer, J.A. 1995. The conservation atlas of tropical forests: The Americas.
lUCN-the World Conservation Union, Cambridge and Gland.

21 Pimm, L.S., Moulton, M.P and Justice, L.J. 1995. Bird extinctions in the central Pacific. In:
Lawton, J.H. and May, R.M. ledsl. Extinction rates, pp. 75-87. Oxford University Press,
Oxford.

22 Dewar, R.E. 1984. Mammalian extinctions and stone age people in Africa. In: Partin, P.S.
and Klein, R.G. ledsl. Quaternary extinctions, pp. 553-573. University of Arizona Press,
Tucson.

23 Semaw, S. et al. 1997. 2.5-million-year-old stone tools from Gona, Ethiopia. Nature 385:
333-336.

24 Thieme, H. 1997. Lower Palaeolithic hunting spears from Germany Nature 385: 807-810.

25 Morwood, M.J. et al. 1998. Fission-track ages of stone tools and fossils on the east
Indonesian island of Flores. Nature 392: 173-176.

Humans, food and biodiversity 67

26 Bird, M.I. and Call, J. A. 1998. A nnillion-year record of tire in sub-Satiaran Africa. Nature
394: 767-769.

27 Gowlett. J.A.J, ef al. 1981 . Early archaeological sites, hominid remains and traces of fire
from Ctiesowanja, Kenya. Naturel^U: 125-129.

28 Isaac, G. 1981. Early tiommids and tire at Ctiesowanja, Kenya. Nature 296: 870.

29 de Heinzelin, J. et at. 1999. Environment and betiavior of 2.5 million year old Bourl
fiominids. Science 28A: 625-629.

30 James, S.R. 1989. Hominid use of fire in the Lower and Middle Pleistocene; A review of
the evidence. Current Anthropology 3Q: 1-26.

31 Hoopes, J.W. 1996. In search of nature. Imagining the precolomblan landscapes of ancient
Central America. Working paper for the Nature and Culture Colloquium, University of
Kansas, Lawrence.

32 Teaford, M.F. and Ungar, P.S. 2000. Diet and the evolution of the earliest human
ancestors. Proceedings of the National Academy of Sciences 97: 13506-1351 1.

33 Ambrose, S.H. 2001. Paleolithic technology and human evolution. Science 29^: 1748-1753.

34 Richards, M.P. et a/. 2000. Neanderthal diet at VIndija and Neanderthal predatlon: The
evidence from stable Isotopes. Proceedings of the National Academy of Sciences 97:
7663-7666.

35 Richards, M.P. et al. 2001 . Stable isotope evidence for Increasing dietary breadth m the
European mid-Upper Paleolithic. Proceedings of the National Academy of Sciences 98:
6528-6532.

36 Stiner, M.C. 2001 . Thirty years on the 'Broad Spectrum Revolution' and paleolithic
demography. Proceedings of the National Academy of Sciences 98: 6993-6996.

37 Clutton-Brock, J. 1995. Origin of the dog: Domestication and early history In: Serpell, J.
led.) The domestic dog: Its evolution, behaviour and interactions with people, pp. 8-20.
Cambridge University Press, Cambridge.

38 Vila, C. etal. 1997. Multiple and ancient origins of the domestic dog. Science 27b: 1687-1689.

39 Smith, B.D. 1998. The emergence of agriculture. Scientific American Library.

40 Smith, B.D. 2001 . Documenting plant domestication: The consilience of biological and
archaeological approaches. Proceedings of the National Academy of Sciences 98: 1324-
1326.

41 Lulkart, G. et al. 2001. Multiple maternal origins and weak phylogeoqraphic structure In
domestic goats. Proceedings of the National Academy of Sciences K: 5927-5932.

42 FAO 1998. The state of the world's plant genetic resources for food and agriculture. Food
and Agriculture Organization of the United Nations, Rome.

43 World Conservation Monitoring Centre 1992. Global biodiversity: Status of the Earth's
living resources. Groombrldge, B. (ed.l Chapman and Hall, London.

44 Prescott-Allen, R. and Prescott-Allen, C. 1990. How many plants feed the world?
Conservation Biology M^]: 365-374.

45 Palsson, G. 1988. Hunters and gatherers of the sea. In: Ingold, T., Riches, D., Woodburn, J.
Hunters and gatherers 1: History, evolution and social change, pp. 189-204. BERG, Oxford.

46 FAOSTAT database. Food and Agriculture Organization of the United Nations. Available
online at http://apps.fao.org/ (accessed February 20021.

47 FAO 1996. Food requirements and population growth. World Food Summit technical
background document TBD VI E4. Food and Agriculture Organization of the United Nations.
Available online at http://www.fao.org/wfs/index_en.htm (accessed February 2002).

48 Uvin, P. The state of world hunger. Hunger report. Gordon and Breach. Available online at
http://www.brown.edu/Departments/World_Hunger_Program/hungerweb/intro/
StateofWorldHunger.pdf (accessed April 2002).

49 Committee on World Food Security 2001. Mobilizing resources to fight hunger 2002 World
Food Summit background document CFS:2001/lnf.7.

68 WORLD ATLAS OF BIODIVERSITY

50 de Ganne, I. and Harrison, G.A. ledsl 1988. Coping with uncertainty in food supply.
Clarendon Press, Oxford-

51 Fernandas, E.C.M., Okingati, A. and Maghembe, J. 1984. The Chagga homegardens: A
muUi-storied agroforestry cropping system on Mount Kilimanjaro [northern Tanzania].
Agroforestry Systems 2: 73-86

52 Juma, C. 1989. The gene hunters: Biotechnology and the scramble for seeds. Zed Books,
London.

53 Marsh, R.R. and Talukeder, A. 1994. Effects of the integration of home gardening on the
production and consumption of target interaction and control groups: A case study from
Bangladesh. Unpublished paper.

54 McEvedy, C. and Jones, R. 1978. Atlas of world population history. Penguin Books,
Harmondsworth.

55 Roberts, N. 1998. The Holocene. An environmental history. 2nd edition. Blackwell
Publishers, Oxford.

56 UN Population Division 2001. World population prospects. The 2000 revision. Draft.
ESA/P/WP.165. Available online at httpVMvwi^.un.org/esa/population (accessed April 20021.

57 Smil, V. 1999. How many billions to go? Nature AO]-. 429.

58 Yalden, D.W. 1996. Historical dichotomies In the exploitation of mammals. In: Taylor, V.J.
and Dunstone, N. The exploitation of mammal populations, pp. 16-27. Chapman and Hall,
London.

59 Wright, D.H. 1990. Human Impacts on energy ilovj through natural ecosystems, and
implications for species endangerment. Ambio 19|4|: 189-194.

60 Haberl, H. 1997. Human appropriation of net primary production as an environmental
indicator: Implications for sustainable development. Ambio 26131: 143-146.

61 Holdaway, R.N. and Jacomb, C. 2000. Rapid extinction of the moas (Aves:
Dlnornithiformesl: Model, test, and implications. Science 287: 2250-2254.

62 Alroy, J. 2001 . A multlspecles overkill simulation of the end-Pleistocene megafaunal mass
extinction. Sc/ence 292: 1893-1896.

63 Miller, G.H. et al. 1999. Pleistocene extinction of Genyornis newtoni: Human Impact on
Australian megafauna. Sr/ence283: 205-208.

64 UNEP-WCMC species conservation database Iplantsl. Available online at
http://wvi/w.unep-wcmc.org/specles.plants/plants-by-taxon.htm#search

65 Reid, W.V. 1992. How many species will there be"!" In: Whitmore, T.C. and Sayer, J. A.
ledsl. Tropical deforestation and species extinction, pp. 55-73. Chapman and Hall,
London.

66 lUCN 1994. lUCN Red List categories. lUCN-the World Conservation Union, Gland.

67 Hilton-Taylor, C. (compiler) 2000. 2000 lUCN Red List of threatened species. lUCN-the
World Conservation Union, Gland and Cambridge. Also available online at
http://www.redlist.org/ (accessed April 2002).

68 The Species Survival Commission of lUCN-the World Conservation Union. Red List
material. Available online at http://iucn.org/themes/ssc/red-llsts.htm (accessed
February 20021.

69 Master, L.L., Flack, S.R. and Stein, B.A. (eds) 1998. Rivers of life: Critical watersheds for
protecting freshwater biodiversity. Nature Conservancy, Arlington.

70 BIrdLlfe International 2000. Threatened birds of the world. Lynx Edicions and BIrdLlfe
International, Barcelona and Cambridge.

71 Richards, J.F. 1990. Land transformation. ln:Turner II, B.L. ef al. (eds). The Earth as
transformed by human action, pp. 163-178. Cambridge University Press, with
Clark University.

72 United Nations Environment Programme 1995. Heywood, V. (ed.) Global biodiversity
assessment. Cambridge University Press, Cambridge.

Humans, food and biodiversity 69

73 Martin, RS. 198i. Prehistoric overkill: The global model. In: Martin, P. and Klein, R.G.

ledsl. Quaternary extinctions, pp. 35i-403. University of Arizona Press, Tucson.
7A Juma, C. 1989. Biological diversity and innovation: Conserving and utilizing genetic

resources in Kenya. African Centre for Technology Studies, Nairobi.

75 Alvarez-Buylla Roces, M.A., Lazos Chavero, E. and Garcia-Burrios, J.R. 1989.
Homegardens of a humid tropical region in South East Mexico: An example of an
agroforestry cropping system in a recently established community Agroforestry Systems
8: 133-156.

76 Alcorn, J.B. 1984. Development policy, forests and peasant farms: Reflections on Haustec-
managed forest contribution to commercial and resource conservation. Economic Botany
38[4|: 389-406.

77 Davies, A.G. and Richards, P. 1991. Ram forest in Mende life: Resources and subsistence
strategies in rural communities around the Gola North Forest Reserve (Sierra Leonel. A
report to the Economic and Social Committee on Overseas Research (ESCORl, UK
Overseas Development Administration, London.

78 Michon, G., Mary, F. and Bompard, J. 1986. Multistoned agroforestry garden system in
West Sumatra, Indonesia. /Igroforesfry Sysfems 4[4|: 315-338.

79 Kunstadter, P., Chapman, E.C. and Sabhasri, S. 1978. Farmers in the forest: Economic
development and marginal agriculture in Northern Thailand. An East-West Center Book,
University Press of Hawaii, Honolulu.

80 Padoch, C. and de Jong, W. 1991. The house gardens of Santa Rosa: Diversity and
variability in an Amazonian agricultural system. Economic Sofany 45121: 166-175.

81 Denevan, W.M, and Treacy, J.M. 1987. Young managed fallows at Brillo Neuvo. In:
Denevan, W.M. and Padoch, C. (edsl. Swidden-fallow agroforestry in the Peruvian Amazon.
Advances in Economic Botany 5, pp. 8-46. New York Botanical Garden, New York.

82 Ohtsuka, R. 1993. Changing food and nutrition of the Gidra in lowland Papua New Guinea.
In: Hladik, CM. et al. ledsl. Tropical forests, people and food. Biocultural interactions and
applications to development. Man and the Biosphere Series. Volume 13. UNESCO and
Parthenon Publishing, Paris and Carnforth.

83 Conklin, H.C. 1954. An ethnoecological approach to shifting agriculture. Transactions of
the New York Academy of Sciences 17: 133-142.

84 Posey, D.A. 1 984. A preliminary report on diversified management of tropical forest by the
Kayapo Indians of the Brazilian Amazon. Advances in Economic Botany 1: 1 12-126.

85 Vickers, W.T. 1993. Changing tropical forest resource management strategies among the
Siona and Secoya Indians. In; Hladik, CM. et al. ledsl. Tropical forests, people and food.
Biocultural interactions and applications to development. Man and the Biosphere Series.
Volume 13. UNESCO and Parthenon Publishing, Paris and Carnforth.

86 Michon, G. 1983. Village forest gardens in West Java. In: Huxley, P. led.l Plant research
and agroforestry, pp. 13-24. International Centre for Research in Agroforestry, Nairobi.

87 Whitman, W.B., Coleman, D.C. and Wiebe, W.J. 1998. Prokaryotes: The unseen majority
Proceedings of the National Academy of Sciences 95: 6578-6583.

88 Rusek, J. 1998. Biodiversity of Collembola and their functional role in the ecosystem.
Biodiversity and Conservation 7: 1207-1219.

89 Zimmerman, P.R. et al. 1982. Termites: A potentially large source of atmospheric
methane, carbon dioxide, and molecular hydrogen. Science 218: 563-565.

90 Nicol, S. Time to krill? Antarctica Online: The website of the Australian Antarctic Program:
http://wv\nA/.antdiv.gov,au/science/bio/issues_krill/krilLfeatures.html

91 Gaston, K.J. and Blackburn, TM. 1997. How many birds are there'' Biodiversity and
Conservation 6: 615-625.

92 Said, M.Y. et al. 1995. African elephant database 1995. lUCN-the World Conservation Union,
Gland.

70 WORLD ATLAS OF BIODIVERSITY

93 SantiapiUai, C. 1997. The Asian elephant conservation: A global strategy. Gajah. Journal of

the Asian Elephant Specialist Group 18: 21-39.
9A Nowak, R.M. 1991. Walker's mammals of the world. 5th edition. Volunne II. Johns Hopkins

University Press, Baltimore.

95 Jefferson, T.A., Leatherw/ood, S. and Webber, M.A. 1993. t^anne mammals of the world.
FAQ Species Identification Guide. United Nations Environment Programme, Rome.

96 Kemf, E. and Philips, C. 1995. Whales in the wild. WWF Species Status Report, WWF -
Worldwide Fund for Nature, Gland.

97 Oldfield, S., Lusty, C. and MacKiven, A. 1998. The world list of threatened trees. World
Conservation Press, Cambridge.

98 Harrison, I.J. and Stiassny, M.L.J. 1999. The quiet crisis: A preliminary listing of the
freshwater fishes of the world that are extinct or 'missing in action'. In: (vlacPhee, R.D.E.
(ed.l Extinctions in near time: Causes, contexts, and consequences, pp.271 -332. Kluwer
Academic/Plenum Publishers, New York.

Terrestrial biodiversity 71

Terrestrial biodiversity

TERRESTRIAL ECOSYSTEMS EXTEND OVER LITTLE MORE than one quarter of the
Earth's surface but they are more accessible than aquatic habitats and so are far
better known. The land supports fewer phyla than the oceans but most of the global
diversity of species, and is characterized above all by an extensive cover of vascular plants,
with associated animals and other groups of organisms.

Forest and woodland ecosystems form the predominant natural landcover over most of
the Earth's surface. These systems generate around half the terrestrial net primary
production, and forests in the tropics are believed to hold most of the world's species.
Approximately half the area of forest developed in post-glacial times has since been cleared
or degraded by humans, and the amount of old-growth forest continues to decline.

Grassland, shrubland and deserts collectively cover most of the unwooded land surface,
with tundra on frozen subsoil at high northern latitudes. These areas tend to have lower
species diversity than most forests, with the notable exception of Mediterranean-type
shrublands, which support some of the most diverse floras on Earth.

Humans have extensively altered most grassland and shrubland areas, usually through
conversion to agriculture, burning and introduction of domestic livestock. They have had less
Immediate impact on tundra and true desert regions, although these remain vulnerable to
global climate change.

THE TERRESTRIAL BIOSPHERE
Land and water on the Earth's surface
Land extends over nearly 150 million square
kilometers (km*) or about 29 percent of the
total surface of the planet, the remainder
being covered by oceans. With the continents
in their present position, more than two thirds
of the land surface is in the northern
hemisphere, and the area of land situated in
the northern hemisphere above the Tropic of
Cancer slightly exceeds that in the rest of
the world put together (Table 5.1 1. A small
proportion of the total land area is occupied by
inland water ecosystems: takes and rivers
cover around 2 percent and swamps and
marshland a similar amount. About half of
the land surface, approximately 52 percent,
is below 500 meters (ml in elevation, and
the mean elevation is 840 m. A minor but
significant proportion of the land surface is
mountainous in nature, with alpine land-

scapes tending to replace trees above
1 000 m at higher latitudes and above 3 500 m
in the tropics.

The land as an environment for living
organisms

The environmental conditions prevailing in
any given place determine what kinds of
organism can live there. Relevant environ-
mental conditions include various aspects
of the physical environment, and the other

Table 5.1

Global distribution of land

area, by latitude bands

Region

Land area '^^H
(million km'l |

Norttiern tiemisphere

North of Tropic of Cancer

Equator nortti to Tropic of Cancer

Souttiern hemisptiere

Equator south to Tropic of Capricorn

23.5

Soutti of Tropic of Capricorn

23.5

(including Antarctica!

72 WORLD ATLAS OF BIODIVERSITY

Most terrestrial organisms
occupy an environment
where water loss is a
constant threat.

species that any given species directly or
indirectly interacts with. These interactions
may occur through a range of mechanisms,
including competition, predation, symbiosis,
mutualism and parasitism. For many living
species, interactions with humans are now a
significant feature of their environment.

The most fundamental distinction, at least
for macroscopic organisms, is between ter-
restrial and aquatic environments. All living

organisms require water because the basic
processes of life take place in the aqueous
medium found in cells. However, most
terrestrial organisms, in contrast to aquatic
ones, occupy an environment where water
loss is a constant threat. The anatomical and
physiological solutions to this problem are
many and varied, including epidermal and/or
cuticular layers with reduced permeability to
water, water storage tissues, and metabolic
or behavioral processes that conserve water
However, the greater the risk of dehydration,
the more carbon and energy are needed for
water conservation mechanisms, leaving less
available for other adaptations. Thus, life
tends to occur at low density and at low
diversity in hyperarid environments, whether
exceptionally hot or exceptionally cold.

There are a number of other extremely
important differences between air and water
as a medium for living things. It takes far less
energy (about 500 times lessl to raise the
temperature of a given mass of air by 1 degree
Celsius than to heat the same mass of water by
the same amount, and water conducts heat far

more rapidly than air As a result, while aquatic
organisms are buffered against rapid fluc-
tuation in their surroundings, terrestrial
organisms can be subjected to wide extremes
in temperature, corresponding to daily and
seasonal variation in insolation.

The air surrounding terrestrial organisms is
much less dense (about 800 times lessl than
water, and so land organisms must support
themselves against the full effects of gravity,
but are not subject to the large forces exerted
on aquatic organisms by moving water

Oxygen is freely and uniformly available in
the atmosphere, but in water is far less
concentrated and much more variable in time
and space. On the other hand, mineral
nutrients dissolve readily in water and it is
possible for aquatic organisms to extract
nutrients directly from their immediate
surroundings (although in many aquatic
habitats such nutrients are only present in
low concentrations!. On land, in contrast,
mineral nutrients occur in soil where their
distribution and concentration are spatially
variable, and they are directly accessible
primarily to microorganisms, plants, fungi
and algae.

Global variations in terrestrial habitats

The wide variations in water availability and
temperature regimes prevailing on the land
surface interact in often complex ways with a
number of other factors, including geology,
soil, terrain, wind, fire regimes and human
activities, to generate the immense range of
environments apparent on the Earth's land
surface. Classification of these into a manage-
able system is a major problem in biology,
one not merely of theoretical interest but of
considerable importance in the management
and conservation of the biosphere. The
problem arises largely from a need to divide
the natural environment into a series of
discrete bounded units for the purposes of
mapping, measuring and monitoring habitats,
whereas the world in reality appears to form a
highly variable continuum.

Where gradients exist between different
physical regimes (particularly of water and
temperature], habitat types tend to intergrade
imperceptibly, and it is impossible without

Terrestrial biodiversity 73

^^j

arbitrary definitions to distinguish, for example,
grassland with a few trees from open woodland
with grass ground cover. Even broad categor-
ies, such as 'forest' or 'wetland', inevitably
require arbitrary limits to be set, e.g. for the
density of tree cover necessary before an area
can be called a forest, or for the duration of
flooding necessary before an area can be
classified as a wetland rather than a terrestrial
system. In such circumstances, it is important
to keep in mind the inherent variability of
ecosystems, rather than attach undue signifi-
cance to their labels.

Vegetation and chlorophyll

The global distribution of actively growing
vegetation can be visualized without classi-
fication or criteria relating to structure, to
physiognomy or to species composition.
Advanced very high resolution radiometer
lAVHRRI satellite sensors measure the
reflectance of vegetation, primarily of the
green photosynthetic pigment chlorophyll, in
the visible and the near infrared part of the
spectrum. On land this can be interpreted as
broadly equivalent to the density and vigor
of green plant growth, represented as the
normalized difference vegetation index
(NDVII. This is plotted, aggregated over one
year, in Map 5.1. This reveals a clear
distinction between areas rich in standing
growth of plants, whether cropland or natural
vegetation, and areas where standing plant
growth is sparse or absent - these are
essentially the drylands and rangelands of
conventional landuse classifications.

Spatial variation in chlorophyll density is
only indirectly related to variation in primary
production levels; net primary production
depends further on soil and climate conditions
and community dynamics within ecosystems
(see Chapter II. So, while Map 5.1 shows high
chlorophyll in both tropical and high latitudes,
net primary production is far higher in the
former than the latter, where it is restricted
mainly by seasonally unfavorable climatic
conditions (compare with Map 1 .21.

Landcover and ecosystems

Landcover classification is more directly con-
cerned with differences in the physical aspects

of ground cover, mainly for landuse planning
and management, than with biodiversity or the
community aspects of vegetation cover Many
current landcover maps (Map 5.21 are based
on interpretation of remote-sensing data that
have the virtue of being quantitative in nature
and available in time series suitable for
monitoring applications. Because the source
data are typically interpreted in terms of
a classification system developed to take
account of conditions on the ground, landcover
maps are subject to some of the definition
problems mentioned above, no matter how
generalized the categories are at the highest
levels (e.g. 'forest', 'cropland', 'urban'].

One of the most useful ecological
distinctions to be made is between areas with
extensive or significant tree cover, and areas
with few or no trees. Terrestrial plant growth
IS favored by high soil water availability and
relatively elevated temperature during all or a
major part of the year Trees tend to be the
main plant growth form in such conditions,
and forest or woodland the main vegetation.
Conversely, primary production is strongly
limited by a shortage of soil water. Grasses
and low shrubs tend to be the main plant
growth forms in such dryland regions and,
where vegetation exists, it consists mainly of
grassland, savannah or shrubland.

Tree growth remains insignificant in those
parts of the world where water is present in
some form, but temperatures are too low
for growth during all or part of the year

Grasses and low shrubs are
the mam plant growth
forms in dryland regions.

74 WORLD ATLAS OF BIODIVERSITY

Map 5.1

Photosynthetic activity
on land

A map illustrating the
normalized difference
vegetation index INDVI)
calculated from AVHRR
satellite data and
aggregated over the year
1998. NDVI values vary with
absorption of red light by
plant chlorophyll and the
reflection of infrared
radiation by water-filled leaf
cells, and provide an
indication of photosynthetic
activity. The map reveals a
clear distinction between
areas rich in standing
growth of plants, whether
cropland or natural
vegetation, and areas where
standing plant growth is
sparse or absent.

Source. Adapted from image provided
by the SeaWiFS Project, NASA/Goddard
Space Fllgtil Center and GRBIfvlAGE.

Box 5.1 Defining ecosystems

Photosynthetic activity

High

Low

For example, in polar regions there is an
abundance of water, but because of permanent
low temperature it is mainly in solid form and

The word 'ecosystem' was introduced by the plant ecologisl Tansley in 1936, to refer to the
communities of plants, animals, other organisms and the physical environment of any given
place. The concept reflected a new and visionary approach to biological research which
focused on the system-level flow of materials and energy between components at any given
study site. The spatial boundaries of the system were originally of little or no significance.
The term is now widely used, particularly in the context of environmental planning, to refer
to broad biological communities of similar appearance, usually defined by physical, climatic,
structural or phenological features. Ecosystem diversity is generally understood to refer to
the range of different kinds of ecosystem, in this sense, within some defined area. In this
usage, the 'ecosystem' is treated as a map unit and the spatial boundaries of the system
assume major significance.

so unavailable to living organisms. In tundra
areas at high latitudes the subsurface is
permanently frozen and plant growth is
restricted to the few summer months when
thawing of the superficial layers makes liquid
water available. In upper alpine regions, liquid
water is often present in seasonal abundance,
but temperatures are too low over the year as
a whole to allow tree growth.

Many areas theoretically have sufficient
rainfall and a temperature regime suitable
for supporting significant tree cover but do
not do so. The most important reason for
this is undoubtedly human intervention but
in some areas this may also be a result of
natural causes. Nutrient availability may be
limiting. Slopes may be too steep to allow
formation and retention of soil, so that trees
cannot anchor themselves. Minerals may be

Terrestrial biodiversity 75

-^F—

,^<f-^ ?aira

present in toxic concentrations, or tlie fre-
quency and intensity of natural fires or floods
may be too higti. Large herbivores may also
prevent establisfiment of significant tree
cover, although as noted in Chapter 4 many of
those large herbivores in the Americas and
Australasia that might have exerted such an
influence in the past are now extinct.

Landcover classifications can be used as
the basis for habitat or ecosystem classi-
fications Isee Box 5.1], by incorporating
elements of species composition and com-
munity structure. Ecosystem maps should
have more direct application to biodiversity
conservation and management than maps of
landcover; hovi/ever the development of such
maps is constrained by the same problems of
defining boundaries and developing a consis-
tent approach to classification. Map 5.7 is a

representation of global forest cover, derived
by applying a simple classification of five
forest and woodland habitat types to the
landcover data plotted in Map 5.2. Forest
classification is discussed further below.

GLOBAL VARIATIONS IN TERRESTRIAL
SPECIES DIVERSITY

Just as habitat types show great variation
across the world's land surface, so does the
number of species that may be found in any
given place. The spatial heterogeneity, wide
range of present physical conditions and
complex history of the worlds land surface all
contribute to extreme variation in terrestrial
biological d'versity. Variation in species
number is not strictly related to variation in
habitat type because two areas that are
structurally similar may have very different

74 WORLD ATLAS OF BIODIVERSITY

Map 5.2

Global land cover

Map adapted from the
global landcover
classification developed by
ttie University of Maryland.
Tfie Maryland classification
includes 13 classes and
w^as based on AVHRR
remote-sensing data with a
spatial resolution of 1 km.
For presentation purposes
the data have here been
generalized to a 4-km grid.
Also for clarity, the two
needleleaf forest classes in
the original classification
have been combined, as
well as the two broadleaf
forest classes, and the
urban and built-up category
has been omitted.

Source: Data from University of
Maryland Global Land Cover Facility. For
full ijescription see Hansen .

€

numbers of species present. The most impor-
tant components of this variation can be
expressed in three different, though not com-
pletely independent, ways:

• the kinds of organisms, particularly pri-
mary producers, that live in any one place -
this influences the kinds of habitats found
there and is a reflection of ecosystem
diversity;

• the numbers of different kinds of organisms
that live in any one place - this is usually
assessed by measures of species diversity,
although other classification systems le.g.
guilds, functional groups, higher taxal can
also be used;

• the individual taxonomic identity of the
organisms that live in any one place -
this is determined by biogeography and

influences, among other things, how a
given area contributes to global biodiversity.

Measuring species diversity

Comparing the diversity of different parts of
the world is complex because of the way
diversity changes with scale. A vvfide range of
observations has demonstrated that, as a
general rule, the number of species recorded
in an area increases with the size of the area,
and that this increase tends to follow a
predictable pattern known as the Arrhenius
relationship, whereby: logS = c + zlogA where
S = number of species, A = area and c and z are
constants (see Figure 5.11. The slope of the
relationship (z in the equation) varies
considerably between surveys, although it is
generally between 0.15 and 0.^0, but some
survey data do not fit the relationship at all. A

Terrestrial biodiversity 77

•^^^

common generalization from tliis finding Is
tfiat a tenfold reduction in an area II. e. loss of
90 percent of habitat) will result in the loss of
between 30 percent (with z = 0.151 and 60
percent (with z = 0.401 of the species originally
present, or approximately half the species.

The relative diversity of different sites will
often partly depend on the scale at which
diversity is measured. Thus 1 m- of semi-
natural European chalk grassland may
contain many more plant species than 1 m' of
lowland Amazonian rainforest, whereas for
any area larger than a few square meters this
could be reversed. In other words, when an
area is sampled the number of species
recorded Increases with the size of the area,
but this rate of Increase varies from area
to area, I.e. the slope of the Arrhenlus
relationship is not constant everywhere.

The reason for this Increase may be
quite straightforward. When smalt areas are
sampled they are likely to be relatively
homogeneous in terms of habitat type.

Log area

Figure 5.1

A typical species-area

plot

Note; The data, consisting of
species counts in a series of
areas of different size, are
plotted on logarithmic axes
resulting in a stralght-tlne
graph, the slope of which Izl
indicates the rate at which
species number changes
with changing area.

78 WORLD ATLAS OF BIODIVERSITY

At a small scale, as sample area Increases, so
an increasing proportion of the species
present in that habitat is lil<ely to be included
in the sample. Beyond a certain point,
however, as larger areas are sampled so an
increasing number of different habitats will
be included in the sample area, each with
new species that are likely to be included
in the sample. The species/area relationship
therefore increasingly reflects habitat hetero-
geneity at larger scales. Ecologists attempt to
tal<e account of this by recognizing different
kinds of biological diversity The diversity
within a site or habitat is often referred to as
alpha (al diversity while the differences
between habitats are referred to as beta Ipl
diversity. Thus an area with a wide range
of dissimilar habitats will have a high f,-
diversity, even if each of its constituent
habitats may have low a-diversity Differences
in site diversity over large areas, such as
continents, are sometimes referred to as
gamma (yl diversity.

Measures of diversity can refer simply to
species richness but can also be more
sophisticated statistical measures that take
into account the relative abundance of different
species in a given place. A variety of different
measures of this kind has been developed (of

Box 5.2 Species and energy

Latitudinal variation in diversity on land is strongly correlated v^ith, and may be largely
explained by, vanation in incident energy over the Earth's surface. The relationship between
diversity and productivity, and related measures, has been the subject of long-standing
debate in ecology, but recent studies have shov\/n that at global or continental scale,
organismal diversity, particularly as measured at higher taxonomic levels, is strongly
correlated with available energy". This l<ind of relationship has been demonstrated, for
example, for flowering plants, for trees, lepidoptera, land birds and land mammals in a
range of countries and continents", and tor fishes in river basins at a global level'. One
simplistic explanation for this may be that higher energy availability leads to increased net
pnmary production (NPPl, and this broader resource base allows more species to coexist.
While the general relationship appears robust, the details are complex. Energy availability
can be measured in several ways: as heat energy, as potential (PET) or actual
evapotranspiration (AETl, or as NPP, and which is the best predictor of diversity has yet to be
determined. Some measure of the simultaneous availability of water and radiant energy may
provide the best general predictor of potential macro-scale species richness". More
complete explanation for richness variation would need also to consider the roles of
topography history and edaphic factors.

which H', the Shannon-Wiener function, is a
commonly used example). With many of these
measures, an area in which all species are of
similar abundance would generally be given a
higher diversity measure than an area with the
same number of species, a few of which were
very abundant and the remainder rare.
Deriving these statistical measures requires
intensive sampling- for this reason, simpler
measures of species richness tend to be more
useful at larger scales.

Major patterns of variation in global
species diversity

Despite the difficulties of establishing strictly
comparable quantifiable measures, a wealth of
empirical observations indicates that species
richness in eukaryotes tends to vary geo-
graphically according to a series of fairly well-
defined rules. In terrestrial environments:

• warmer areas hold more species than
colder ones;

• wetter areas hold more species than drier
ones;

• areas with varied topography and climate
hold more species than uniform ones;

• less seasonal areas hold more species than
highly seasonal ones;

• areas at lower elevation hold more species
than areas at high elevation.

The single most obvious pattern in the global
distribution of species is that overall species
richness increases as latitude decreases
toward the equator (see Box 5.2). At its
simplest this means that there are more
species in total and per unit area in temperate
regions than in polar regions, and far more
again in the tropics than in temperate regions.
This applies as an overall general rule, and
within most individual higher taxa (at order
level or higher), and within most equivalent
habitats. The pattern can be seen in Maps 5.3,
illustrating vascular plant species diversity,
and 5. A, which represents country-level
diversity in terrestrial vertebrates and
vascular plants.

There is good evidence that moist tropical
forests are the most species-rich environ-
ments on Earth. If current estimates of the

Terrestrial biodiversity 79

number of as yet unknown species Isee
Chapter 21 in the tropical forest microfauna
are accepted, these regions, extending over
perhaps 7 percent of the world's surface, may
conceivably hold more than 90 percent of the
world's species. If tropical forest small
Insects are discounted, then coral reefs and.
for flowering plants at least, areas of
Mediterranean climate in South Africa and
Western Australia may be similarly rich
in species.

Topographic heterogeneity may be expec-
ted to play a significant part in determining
species number for two reasons. First, such
heterogeneity will increase habitat variability.
thereby increasing the range of niches that
can be occupied by different organisms.
Second, depending on the size and mobility of
organisms, the chances of geographic Isolation
and speciatlon increase in topographically
diverse landscapes. The role of topography
has been demonstrated statistically at
continent scale for North American mammals',
and at both landscape and patch scale for
vascular plants' '".

The available Information on distribution
of species Is geographically very Incomplete,
and relates to only a small fraction of the
1.7 million known species. Geographically,
western Europe has been more thoroughly
sampled than elsewhere, while large areas
In the tropics, particularly of South America
and central Africa are very poorly known.
Taxonomlcally, the larger mammals, birds,
vascular plants and a few invertebrate
groups, such as Odonata. are better known
than other groups of species. Because of the
uneven availability of information there has
been considerable interest In Identifying
groups of Individual species that may serve
as surrogates for biodiversity more gener-
ally, or higher taxa. such as families, that
might predict patterns of richness In their
included species".

Maps 5.5 and 5.6 represent, respectively,
the distributions of flowering plant families
(phylum Anthophytal and of terrestrial
(non-aquaticl vertebrate families Iphylum
Chordatal. plotted as a global family diversity
surface for each group. The resulting maps
share key features, notably a marked latitud-

inal gradient in family richness, but also have
striking differences, particularly In areas of
high diversity. For example. Africa appears
to be very rich in vertebrate families, most
notably In moist forest areas around the
Gulf of Guinea and In the east, including less
humid woodland and savannah habitats, but
Is relatively poor In plant families compared
with other continents in the tropics. For
flowering plants, this mirrors the sequence
evident at species level, where there may be
around 90 000 species In the neotropics.
40 000 in tropical Asia and 35 000 in tropical
Africa'-.

Blogeography and endemism

While ecological factors Influence which kinds
of species, and how many of them, can persist
In a given area, history has already deter-
mined which actual lineages are present. A
complete explanation for global variation In
biodiversity must therefore involve both
historical events and current ecological pro-
cesses. The former are implicit in any explan-
ation of the origin of diversity, the latter In
explanations of Its maintenance; these being
two separate, although Intimately linked,
problems. On land, continental drift resulting
from plate tectonics, climate change, moun-
tain building or sea-level change, and prob-
ably the evolutionary lability of different
lineages, are among the Important historical
factors. Geographic features commonly re-
strict or prevent the further dispersal of
species: for example, a large river can pre-
sent a barrier to a terrestrial species, the sea
Is a barrier to non-flying Island forms and
land is a barrier to freshwater species.

Barriers to dispersal explain why the fauna
and flora of ecologically similar areas In
geographically separated parts of the world
tend to be composed largely of different
individual species. They also underlie the
phenomenon of endemism. An endemic
species Is one restricted to some given area,
which may be a continent or country, or more
significantly a relatively small area, such as a
mountain block. Island or lake. Discrete areas
of complex topography, particularly In the
tropics, often have high endemism In a
range of taxonomic groups, possibly because

WORLD ATLAS OF BIODIVERSITY

climate change has encouraged speciation by
isolating different lineages at different times.

Communities and ecoregions
The most comprehensive attempts to
describe and classify habitats try to combine
elements of all three sources of variation
outlined above. At fine scales these are
generally based on community ecology. On
land this is exemplified by the phyto-
sociological approach developed principally in
continental Europe during the 20th century

UNESCO"

Closed forest

Trees > 5 m tall with crowns interlocking

Woodland

Trees > 5 m tall with crowns not usually
touching but with canopy cover > 40%

US classification

Closed tree canopy

Trees with crowns interlocking, with

standards"

crowns forming 60-100% cover

Open tree canopy

Trees with crowns not usually touching
forming 10-60% or 25-60% cover

FAO"

Forest

Land with tree canopy cover > 10% and > 0,5 ha
in area; trees should be able to reach a
minimum height of 5 m

Other wooded land

Land with either a crown cover of 5-10% of trees
able to reach a height of 5 m at maturity; or
crown cover of more than 10% of trees not able
to reach a height of 5 m at maturity; or with
shrub or bush cover > 1 0%

Table 5.2

Different definitions of

forest cover

This was intended to describe and classify
plant communities on the basis of dominant
and other associated species, by inspection or
quadrat analysis of vegetation patches, taking
into account species identity, growth form and
abundance. One major problem with this is
that the more precisely a community is
defined, the more site-specific it becomes,
and hence the more limited its use in higher-
level analysis and planning. A more recent
related approach" attempts to delimit eco-
regions', these being defined as relatively
large units of land each of which contains a
distinct assemblage of natural communities
and species, with boundaries similar to the
original extent prior to anthropogenic change.
These are nested within a hierarchy having

traditional biogeographic realms and biome
systems [ecosystem types! as the first and
second levels.

FORESTS

Forests and woodlands probably once covered
about half of global land area and now cover
about one quarter They provide habitat for
half or more of the worlds species. They are
responsible for just under half of the global
terrestrial annual net primary production
[Table l.ll, and they and their soils house
about 50 percent of the world's terrestrial
carbon stocks. In addition to carbon storage,
forests perform many other important
ecosystem services, such as regulating local
hydrological and nutrient cycles, and stabiliz-
ing soils and watersheds.

Forests also provide a wide variety of
products, including food and fuel, medicines,
construction materials and paper, which are
important both for human subsistence and for
economic activity. Wood products are one of
the most economically important natural
resources. In the region of 3.3 billion cubic
meters (m-1 of wood is extracted from forests
and other habitats annually, the equivalent of
several hundred million trees. Just over half
of this volume is used as fuelwood and
charcoal, of which developing countries
consume 90 percent". The remainder is
industrial roundwood, which is processed into
various wood products. Forests are frequently
important culturally and play a significant role
in the spiritual life of communities worldwide.

What Is a forest?

Despite their importance in a number of
different human contexts and the large
amount of research that has focused on forest
ecosystems, a precise definition of 'forest'
remains elusive. Although it is generally
accepted that the term indicates an eco-
system in which trees are the predominant
life form, the problem arises because of the
broad range of systems in which trees occur
and the difficulty, even, in deciding what
constitutes a tree. For example, tree species
may dominate at high altitude, but be barely
recognizable as trees because of their
spreading prostrate forms. Savannahs may

Terrestrial biodiversity 81

"««/,(,

possess large numbers of trees, but as they
can occur at low density in association with
other life forms it may be difficult to define
precisely in which areas they are dominant. A
variety of different definitions of forest have
been proposed by organizations that evaluate
and monitor natural resources (Table 5.21.
Estimates of forest area may vary widely
depending on the definition adopted. Of sing-
ular importance is the degree of canopy cover
used as the threshold for dividing forests from
non-forests (Table 5.21. As a consequence, the
precise definitions employed should be borne
in mind when comparing forest cover data
provided by different institutions.

Forest types

There is great variation in the forms and types
of forest distributed throughout the world.
Information about this variation and the
distribution of forest vegetation types is
crucial to understanding the different roles of
forests in supporting biodiversity, in carbon
and hydrological cycles and other ecosystem
processes, and in supplying wood and non-
wood forest products. However, if deriving a
satisfactory definition of forest is problematic,
arriving at consensus on how to classify
forests is even more difficult.

A number of global classification systems
have been proposed, but as yet none has
gained universal acceptance. The UNESCO
(United Nations Educational, Scientific and
Cultural Organization) system proposed by
Ellenberg and Mueller-Dombois'"' is one such
system. It includes nearly 100 forest and
woodland subformations' and allows for yet
finer subdivisions, but many of the charac-
teristics that separate categories can only be
determined in the field. Other classifications,
such as the EROS Data Center seasonal
landcover regions, with nearly a thousand
classes, reflect more strongly the nature of
landcover data obtained from Earth-orbiting
satellites and the methods used in analyzing
and classifying them'. This complex system
has been translated into a much less complex
one in the International Geosphere-Biosphere
Programme (IGBP) classification, which
includes seven forest and woodland types that
reflect phenology and canopy closure world-

Country

Forest definition

Forest area Ikm'l

Australia

Tree canopy cover > 20%

384 115

Tree canopy cover > 70%

30 729

Senegal

Tree canopy cover > 10%

(includes dry woodland]

78 689

'Closed' forest (canopy cover

> iO%]

3 934

wide, but provides little other information on
forest physiognomy, composition or environ-
ment within the class names'. Map 5.7
represents the global distribution of a range
of forest types, based on forest physiognomy
and phenology, here aggregated into five
broad categories, discussed further below.

Temperate and boreal needleleaf forests
Distribution, types and characteristic taxa

Temperate and boreal needleleaf forests cover
a larger area of the world than other forest
types. They mostly occupy the higher latitude
regions of the northern hemisphere, as well as
high-altitude zones and some warm temper-
ate areas, especially on nutrient-poor or
otherwise unfavorable soils. These forests are
composed entirely, or nearly so, of coniferous
species IConiferophyta). In the northern
hemisphere, pines Pinus, spruces Picea,
larches Larix, silver firs Abies, Douglas firs
Pseudotsuga and hemlocl<s Tsuga dominate
the canopy, but other taxa are also important.
In the southern hemisphere coniferous trees,
including members of the Araucariaceae,
Cupressaceae and Podocarpaceae, often
occur in mixtures with broadleaf species in
systems that are classed as broadleaf and
mixed forests.

Forest type Area ImiUlon kmM

Temperate and boreal needleleaf

13.1

Temperate broadleaf and mixed

7.5

Tropical moist

11.7

Tropical dry

2.5

Sparse trees and parkland

6.9

Total

41.7

Table 5.3

Sample effects on forest
area estimates of different
forest definitions

Table 5.4

Global area of five main

forest types

Note; Based on forest
cover as shown in Map 5.7.
Estimates of this kind
vary significantly with
different source data and
classifications.

82 WORLD ATLAS OF BIODIVERSITY

Structure and ecology

The structure of temperate and boreal
needleleaf forests is often comparatively
simple, as conifer canopies are efficient
[igfit absorbers, reducing the possibilities for
development of lov^er strata in the canopy.
The tallest of these forests, the giant redwood
forests of the west coast of the United States,
may reach 100 m in height, but most are
much shorter, and indeed some pine forests
at high altitude or in arid environments are
quite stunted.

The distribution of temperate and boreal
needleleaf forest is limited at high altitudes
and latitudes by lack of enough days with
temperatures suitable for growth, and at
lower altitudes and latitudes by competition
with broadleaf species. In about a quarter of
the area of these forests, deciduous conifers
of the genus Larix replace the evergreen
species. This is especially true In far northern
continental areas with extremely low winter
temperatures.

In many areas, wildfire is an important
factor affecting the dynamics and main-
tenance of the forest ecosystem. Many coni-
ferous species produce resins that Increase
flammability, and many are characterized by
thick bark that increases the resistance of
adult trees to fire-induced mortality. A
number of tree species, such as the jack pine

Box 5.3 Fire in temperate and boreal forest

In general, as the frequency of fire increases, the intensity of individual fires tends to
decrease, because of reductions in the standing amount of fuel. Increased fire frequency
also tends to Increase the diversity of the herb layer by severely affecting the shrub layer
Fires are caused by natural events such as lightning strikes and by human activities.
Changing forest management has significantly altered fire regimes In coniferous forests.
The 1988 fires that affected over 5 000 km' In and around Yellowstone National Park,
United States, were attributed to the accumulation of fuel in the forests resulting from a
long-term policy of fire suppression.

Climatic variation plays an important part in determining fire occurrence and severity. The
Yellowstone fires, Canadian fires in 1989, and the 1987 Black Dragon fire In the boreal
forest region of China and Siberia In 1987, were all ascribed to unusual drought conditions.
The Black Dragon fire burned over 70 000 km^ and qualified as the largest forest fire in
recorded history'. There is concern that global climate change may increase the frequency
and impact of fires in boreal coniferous forests.

Pinus banksiana. have serotinous cones,
which depend on the high temperatures of
forest fires to open and release their seeds.
Non-coniferous species are generally less
resistant to fire than conifers, so periodic fires
are an Important factor In maintaining the
composition and extent of these forests.

Biodiversity

Although tree species richness is low in most
temperate and boreal needleleaf forests,
many conifer species are of great con-
servation concern. A well-known example
occurs In the giant redwood forests of
northern California, where the redwood
Sequoiadendron giganteum is considered
vulnerable to extinction'. About 22 percent
(140) of the world's 630 conifer species have
been assessed as globally threatened". Most
of the threatened taxa are characteristic of
mixed forests, particularly in the southern
hemisphere. Old growth conifer stands, which
may be many centuries in age, represent an
irreplaceable gene pool and are an important
habitat for many other organisms.

Species richness in these forests is
commonly Increased by a relatively high
diversity of mosses and lichens, which grow
both on the ground and on tree trunks and
branches. For example, there are at least 100
species of moss growing in the coniferous
forests between 1 300 and 2 000 m altitude on
Baektu Mountain, on the Chinese-Korean
border^ Mosses and lichens are Important
sources of food for many animals of coni-
ferous forest.

Vertebrate richness is generally lower in
boreal needleleaf forests than In broadleaf
temperate and tropical forests. Many species
are wide-ranging generallsts, often with a
holarctic distribution, e.g. wolf Canis lupus,
brown bear Ursus arctos.

There are a number of animals of con-
servation concern that are dependent on
temperate needleleaf forests. The northern
spotted owl Strix occidentalis caurina requires
large expanses of old-growth coniferous forest
in the northwest United States to provide
nesting habitat and adequate food resources;
Kirtland's warbler Dendroica kirtlandii needs
young regrowing jack pine as a nesting habitat.

Terrestrial biodiversity 83

"^'^h

Fire suppression programs have reduced the
available habitat for this species to critical
levels. While there is relatively little infor-
mation available on the conservation status of
Invertebrates, many common old-growth
species are known to become much rarer in
modern managed forests, often through the
loss of essential microhabltats'.

Rote in carbon cycle

Temperate and boreal needleleaf forests make
a significant contribution to the global carbon
balance, accounting for more than a third of
the carbon stored In forest ecosystems (Table
5.61 and about 8 percent of global annual net
primary production (Table 1.1|. Furthermore,
the soils under these forests store large
amounts of carbon (up to 250 metric tons per
hectare), some of which may be liberated by
Increasing decomposition rates related to
climate change. Of particular note in this
context are the giant conifer forests of the
Pacific northwest of the United States. These
forests may store more than twice as much
carbon per hectare as tropical rainforests.

Use by humans

Global Industrial roundwood production is
dominated by coniferous species. Pines Pinus.
spruces Picea, larches Larix, silver firs Abies.
Douglas firs Pseudotsuga and hemlocks Tsuga
from the needleleaf forests of the northern
hemisphere are the major sources of softwood.
Some conifer species from the southern
hemisphere and the tropics also provide
excellent timbers. Large-scale exploitation of
natural coniferous and mixed forests is taking
place around the Pacific Rim, notably In North
America, Russia and Chile. Temperate and
boreal needleleaf forests are also a principal
source of pulpwood for paper production.

Other ecosystem services

These forests, like others, stabilize soils on
sloping topography, especially in mountainous
regions. Recognition of this function in early
20th century Switzerland was the basis for a
new program of forest planting to control
avalanches in the Alps. Coniferous forests
have high recreational and cultural values,
especially in regions such as northern Europe.

Temperate broadleat and mixed forests
Distribution, types and characteristic taxa

Temperate broadleaf and mixed forests cover
some 7.5 million km- of the Earth's surface
(Table 5.4). They Include such forest types as
the mixed deciduous forests of the United
States and their counterparts in China and

Table 5.5

Important families and
genera, and numbers of
species, in four areas of
temperate broadleat
deciduous forest

Source: Afler Rohng .

■Family

Genus

Common Northeast
name America

Europe

East Asia

South V
America^

Fagaceae

Quercus Oak
Lithocarpus
Castanopsis

Cyclobalanopsis Asian oak
Castanea Chestnut
Fagus Beech
Nothofagus Southern beech

37
1

4
1

1
2

66
47
45
30
7
7

Aceraceae

Acer
Dipteronia

Maple

Betualceae

Betulus
Alnus

Birch
Alder

4
4

36
14

Salicaceae

Salix

Willow

Juglandaceae

Car/a

Juglans

Platycarya

Hickory
Walnut

11
5

4
4
2

Legumlnosae

Cercis

Cleditsia

Gymnodadus

Maackia

Robinia

Acacia

Albizia

Redbud
Honey locust

Locust
Acacia

1
2
1

2
7
3
3
1

1
1

Magnoliaceae

Liriodendron
Magnolia

Tulip tree
Magnolia

1
8

1
50

Oleaceae

Fraxinus
Osmanthus

Ash

4
2

20
10

Rosaceae

Malus

Prunus

Pyrus

Sorbus

Kageneckia

Quillaia

Apple

Cherry
Pear
Mountain ash

1
3
1
3

4
1

3
59

5
18

Tllllaceae

Tilia

Basswcod,
lime, linden

Lauraceae

Phoebe

Sassafras

Beilschmidia

Cryptocarya

Persea

Sassafrass

16
2

("lyrtaceae

Amomyrtus
Myrceungenelli
Myrceugenia
Nothomyrcia

8
1

Ulnnaceae

Celtis
Ulmus
Zelkova

Hackberry
Elm

2
4

1
3

Carpinaceae

Carpinus
Ostrya

Hornbeam

2
1

25
3

84 WORLD ATLAS OF BIODIVERSITY

Table 5.6

Biomass and carbon
storage in the world's
major forest types

Note: The carbon storage
figures are over- rather tfian
underestimates as they
incorporate no weighting tor
anthropogenic disturbance or
for the variation m biomass
and area among different
forest classes within the
broad types-

Source: After Adams and Huston ,

Japan, the broadleaf evergreen rainforests of
Japan, Chile, New Zealand and Tasmania, and
the sclerophyllous forests of Australia, the
Mediterranean and California. These last are
characterized by a predominance of often
evergreen trees with small, hard, leathery
leaves.

Trees belonging to the Anthophyta and the
Coniferophyta grow in mixtures in many of
these forests, especially in the southern
hemisphere. For example, the Valdivian and
Magellanic rainforests of Chile include
mixtures of Nothofagus Ian anthophytel
with Podocarpus and members of the
Cupressaceae Iboth conifers)". Much of
the forest of New Zealand was originally
a mixture of the conifers Podocarpus,
Phyllocladus, Dacryocarpus and Dacrydium.
with such Anthophyta as Metrosideros,
Elaeocarpus and Weinmannia. In North
American mixed forests, pines Pinus, hem-
locks Tsuga and cypress Taxodium, among
other conifers, are mixed in various propor-
tions with oaks Quercus. maples Acer, ashes
Fraxinus, hickories Carya, beeches Fagusand
other hardwoods.

The beech family IFagaceael is generally
important in temperate broadleaf forests
(Table 5.51, with such genera as Castanopsis
and Cyclobalanopsis playing an important
rolein Japan and China, and many different
Quercus species being important elements
of hardwood forests in North America,
Asia and Europe and of sclerophyllous
forests in California and the Mediterranean,
Sclerophyllous forests in Australia are largely
made up of Eucalyptus species, as are the
wet forests of Tasmania, which may also
include Nothofagus'.

Forest type

Above-ground

Estimated total

Estimated global

biomass per area

carbon storage

total carbon

(metric tons/lial

including soil and

storage

roots (metric tons/hal

(petagrams = 10" gl

Temperate needleleaf

200-1 500

300 1700 giant conifer]

394

Temperate broadleaf

and mixed forest

150-300

350

262

Tropical moist forest

195-500

300

350

Tropical dry forest

98-320

250

Structure and ecology

Depending on the precise forest type, these
forests tend to be structurally more complex
than pure coniferous forests, having more
layers in the canopy. It is not uncommon for an
upper canopy layer to have as many as six
distinct subcanopy and understorey layers
below it. The tallest of these forests, the mixed
forests of southern Chile and some Eucalyptus
forests of Australia, can reach over 50 m in
height. On the other hand, some sclero-
phyllous forests barely reach 5 m. Deciduous
forests may support rich herb layers, which
depend on the increased penetration of
sunlight early in the growing season.

As in needleleaf forests, fire plays an
important role in many types of temperate
broadleaf and mixed forest. Both natural and
anthropogenic fires are important ecological
factors affecting the maintenance of forest
structure and composition, especially in the
Eucalyptus forests of Australia and the
sclerophyllous forests of the Mediterranean
and California. Spatial variation in forest
structure and composition may be influenced
by fire and other kinds of natural and
anthropogenic disturbance. When canopy
trees die, the resulting gaps in the canopy
increase light availability locally, and such
areas may be colonized by a different subset
of the forest flora. These gap dynamic pro-
cesses are important in maintaining stand
diversity. Relatively few old-growth broadleaf
and mixed forests remain in the temperate
zones because of the historical exploitation of
these forests by human populations.

Biodiversity

As might be expected from their structural
diversity, temperate broadleaf and mixed
forests tend to be richer in species than
coniferous forests. Southern mixed hardwood
forests in the United States are commonly
composed of as many as 20 canopy and
subcanopy tree species and may include as
many as 30 overstorey species'". In comparison,
European forests tend to be less species rich,
while the deciduous forests of East Asia may be
the richest of all" (Table 5.51.

In the late 1990s, some 370 temperate
dicotyledonous tree taxa worldwide were

Terrestrial biodiversity 85

"^K

considered to be of conservation concern".
Japan alone has 43 threatened endemic tree
species, which are mostly characteristic of
its temperate broadleaf forests'". Fitzroya
cupressoides, an endangered large conifer
once important for the local and international
timber trade, is characteristic of mixed
forests in southern Chile. It has been over-
exploited and proves to be highly dependent
for its regeneration on large-scale natural
disturbances, such as landslides and light-
ning strikes. F. cupressoides is nov/ listed in
CITES (Convention on International Trade in
Endangered Species of Wild Flora and Fauna!
Appendix I, and commercial international
trade is accordingly prohibited.

While temperate broadleaf forests gen-
erally support moderate animal diversity,
species richness is often lower than in
comparable tropical habitats. However, there
is considerable geographical variation in
richness in the northern hemisphere; the
forests of East Asia are generally the most
species rich". As in the boreal needleleaf
forests, many of the mammal and bird
families of northern broadleaf forests are
holarctic in distribution and can be found in
other habitat types.

In contrast to most northern hemisphere
forests, the temperate forests of southern
South America, Australia and New Zealand
contain several restricted-range mammals
and birds. Analysis" of habitat requirements
of Australian mammals indicates that the
forests of southeast Australia and Tasmania
are of particular importance for wildlife
conservation.

Examples of temperate broadleaf forest
species of special conservation concern
include the huemul deer Hippocamelus
bisulcus of the southern Andes, threatened by
both habitat loss and hunting; a number of
New Zealand forest birds including the
kakapo Strigops habroptilus and some kiwi
species Apteryx spp., which are threatened
mainly by introduced predators; Leadbeater's
possum Gymnobelideus leadbeateri, which is
threatened by the loss of specific habitat
within the montane ash forests of Victoria
lAustralial; the Amami rabbit Pentatagus
furnessi of Amam\ Island iJapan], threatened

by habitat loss and introduced predators; and
the European bison Bison bonasus of central
and eastern Europe, with low genetic diversity
and at risk from disease.

Ttie figure shows tfie living planet index for a sample of 170 forest-occurring birds in Nortli
America and Europe. The prevailing trend during 1970-99 viias for a small net increase over
the period. This could be interpreted as reflecting a phase of relative stability in these forests
during recent decades, following centuries of decline in area, but may also be correlated
with local increase in forest area through the spread of plantations, and management in
some forests may have exerted a positive effect. The separate North America and Europe
samples |123 and 47 species] show extremely similar trends.

1970 1975 1980 1985

Source: UNEP-WCMC. from dala collaled for Lofi".

Role in carbon cycle

Above-ground biomass of temperate decid-
uous forests in Europe, the United States and
the former Soviet Union ranges from 140 to 500
metric tons per hectare depending on stand
age and altitude, among other factors". Soil
carbon storage in these systems is lower than
in the needleleaf forests, owing both to climat-
ically favorable conditions for decomposition
and to the inherent greater decomposability of
leaf litter from broadleaf trees.

One survey" suggests that average soil
carbon storage in these forests is some-
where between 135 and 160 metric tons per
hectare. Thus, global carbon storage across
the forest types included in this broad cate-
gory may total as much as 231 petagrams
(Table 5.61.

1990

1995

/—

86 WORLD ATLAS OF BIODIVERSITY

Tropical moist forests
are often particulary ricfi
in palms.

Use by humans

Temperate broadleaf and mixed forests have
provided large amounts of timber over tfie
centuries, but ttiey have largely been replaced
by tropical forests as the primary source of
hardwood timber in global trade. Hardwood
production continues in the temperate zones,
for products such as wood chips, furniture
and finishing wood.

Some of the non-timber products of
temperate mixed forests include camphor
from Cinnamomum camphorum in Japan

(though this has now largely been replaced
by synthetics], and sweet chestnuts
Castanea sativa from southern Europe.
Mushroom production is also a major
income source in some parts of Europe,
North America and Asia.

Other ecosystem services

Services provided by temperate broadleaf and
mixed forests Include soil and watershed
protection. This is especially important in the
southern Andes" and other areas where steep
topography is responsible for a high incidence
of landslides, but it is also a benefit in
Mediterranean regions where soils are prone
to degradation. Natural temperate forests
are important reservoirs of genetic material
of trees such as eucalypts that are now
commonly grown as plantation species. These
forests, perhaps more than any other type,
experience significant recreational use in
many areas.

Tropical moist forests

Distribution, types and characteristic taxa

Tropical moist forests cover more than
11-5 million km' of the humid tropics (Table
5. A and Map 5.7) and include many different
forest types. The best known and most
extensive are the lowland evergreen broadleaf
rainforests, which make up over half the total
area. These include, for example: the season-
ally inundated varzea and igapo forests and
the terra firme forests of the Amazon basin;
the peat forests and moist dipterocarp forests
of Southeast Asia; and the high forests of the
Congo basin.

Mountain forests are generally divided
into upper and lower montane formations on
the basis of physiognomy. These include
cloud forest - the middle- to high-altitude
forests that derive a significant part of their
water supply from cloud, and support a rich
abundance of epiphytes. Mangrove forests
(see Chapter 6] also fall within this broad
category.

The high diversity of many tropical forests
(see below] makes it difficult to characterize
them taxonomically. However, some plant
families are more prevalent than others. In
neotropical moist forests, the legumes
Leguminosae are particularly abundant and
are often the most species-rich family.
Other families that are generally among the
richest in tree species in lowland neotropical
moist forests are Moraceae, Lauraceae,
Annonaceae, Sapotaceae, Myristicaceae,
Meliaceae, Euphorbiaceae and Palmae". In
Southeast Asian lowland moist forests the
dominant family is the Dipterocarpaceae; the
Myrtaceae is also very speciose". In African
rainforests legumes are again important, and
the ten richest families usually include the
Olacaceae, Sterculiaceae, Dichapetalaceae,
Apocynaceae, Sapindaceae and Ebenaceae.
The other most abundant families in both
Africa and Asia are often the same as those in
the Americas''\

Structure and ecology

Many tropical moist forests have canopies
iO to 50 m tall, and some have emergent trees
that rise above the main canopy to heights of
60 m or more. Such large-stature forests are

Terrestrial biodiversity 87

characteristic of lowland forests and some
lower montane forests on relatively nutrient-
rich soils. Another characteristic of these
forests IS a relatively high frequency of woody
lianas and, especially in the nootropics,
palms". Moist tropical forests are also known
for a high abundance and diversity of vascular
epiphytes, which take advantage of the higher
light availability found in the canopy and can
survive because of abundant rainfall and high
atmospheric moisture. On more nutrient-
poor soils and at higher altitudes forest
stature decreases substantially; communities
such as those on white sands (bana and
campinal and in upper montane environments
lelfin forests] may be no more than a few
meters tall. With increasing altitude, decreas-
ing forest stature is accompanied by a reduc-
tion in the frequency of lianas and palms, and
an increase in tree ferns ICyatheaceae,
Blechnaceael and non-vascular as well as
vascular epiphytes".

Unlike the other forest types discusaed
here, tropical moist forests have relatively
little seasonal limitation to their growth,
though seasonal drought may be a limiting
factor, particularly in the semi-evergreen
formations. However, the tropical moist forest
environment is an intensely competitive one.
Though solar energy inputs are high, canopy
closure and complexity are also substantial,
resulting in efficient capture of incident
radiation and understorey light availability
frequently much less than 2 percent of that
above the canopy. This in turn limits the
growth of understorey species and regener-
ating trees. Some species can tolerate low
light availability, while others grow or
regenerate only in gaps in the canopy. Such
gaps are formed by the death of one or more
canopy trees and represent a significant
contribution to overall environmental hetero-
geneity, which is an important contributor
to high diversity within tropical forests'".
Infrequently, catastrophic disturbances such
as blowdowns caused by hurricanes or
convective storms may create large areas of
regenerating forest" ™ and perhaps alter the
long-term forest composition, as can logging
and other forms of forest disturbance caused
by human activities (see belowl.

Soil nutrients are another limiting re-
source in many forests. Soils in the humid
regions of the tropics are notoriously poor in
nutrients owing, among other factors, to loss
through leaching by the high annual rainfall
and to retention in the high-standing biomass.
The formation of gaps in the canopy allowing
regeneration of tree species is important for
increasing the heterogeneity of available
nutrients as well as the availability of solar
energy. Forests such as the Amazonian
varzea, which are seasonally inundated by
sediment-bearing rivers, are an exception to
this nutrient limitation.

Biodiversity

In numerical terms, global terrestrial species
diversity is concentrated in tropical rain-
forests. Many theories have been proposed to
explain this phenomenon*". Generally speak-
ing, the wet tropical forests of Africa have a
lower tree species richness than those of Asia
and America (Table 5.7). However, there is
great local variation in species richness.
Within the Amazon basin, tree species rich-
ness ranges from 87 species per hectare in
the east" to 285 species in central Amazonia"
and nearly 300 species in the west'l

The high diversity of tree species in
lowland evergreen rainforests is mirrored in
the diversity of epiphytes and lianas, which is
also much higher in neotropical forests than
in other regions^'. Fifty-three families in the
Anthophyta and at least nine pteridophyte
(Filicinophyta and allies) families include
epiphytes. Of nearly 25 000 species of
vascular epiphytes, around 15 000 belong
to the Orchidaceae. Nearly a thousand
others are members of the pineapple family
Bromeliaceae, which is primarily neotropical.
Other groups having a high diversity of

Table 5.7

Tree species richness in

tropical moist forests

Sources Afler Phillips etat."'.

Region

Africa

Southeast Asia
Americas

No. of tree species

l> 10 cm diameter!

per hectare

56-92
108-2i0
56-285

WORLD ATLAS OF BIODIVERSITY

::?=^

Map 5.3

Diversity of vascular plant

species

This map shows the
species richness of
vascular plants, plotted as
a world density surface. It
is based on some 1 iOO
literature records from
different geographic units,
with richness values as
mapped calculated on a
standard area of 10 000 km'
using a single species-area
curve. Value categories
range between extremes of
more than 5 000 species
and fewer than 100 species
per 10 000 km\

Source: Data and analysis © Wilhelm
BarlhloU IBotanic Institute and Botanic
Gardens. University of Bonn).
Reproduced by permission, with
modification to colors. For furtfier
details see Bartfilott and website
hltp://www.botanik,uni-bonn,de/system/
biomaps,htm#worldmap [accessed
f^arch 20021.

Diversity

High

Low

epiphytic species are the cactus family
Cactaceae, the aroids Araceae, the pepper
family Piperaceae and the African violet family
Gesneriaceae.

Not all tropical moist forests have high
species richness. Mangrove ecosystems have
a low diversity of tree species despite their
sometimes high productivity and high animal
diversity (see Chapter 6). Extremely nutrient-
poor soils, such as white sands, lead to the
development of low-diversity forests including
bana and campina''. As climate becomes
more seasonal, tree species richness tends to
decline [see dry forests, below). Increasing
altitude also tends to reduce species rich-
ness, with montane forests typically having
fewer tree species than lowland ones".

Regionally, the forests of Asia and South
America are rich in animal species, generally

more so than those of Africa"'. Many forest
animals are largely confined to moist forests,
e.g. the okapi Okapia johnstoni, but some
are widespread outside, such as leopard
Panthera parclus'\ In Africa, the Guineo-
Congolean forest block contains more than 80
percent of African primate species, and nearly
70 percent of African passerine birds and
butterflies". About half of the 1 100 South
American reptile species are found in moist
forests, with around 300 of these endemic to
the habitat". Amphibian species are partic-
ularly diverse in tropical moist forests"; 90
percent of 225 species identified in the
Amazon basin forests are endemic.

The importance and diversity of the fish
communities in forest streams and rivers is
often overlooked; the Amazon basin has the
richest freshwater fish fauna known, with at

Terrestrial biodiversity 89

■^^^

[east 2 500 species, many Important as major
seed predators and dispersal agents'". Local
species richness of insect and other arthro-
pod groups in tropical forest canopies is much
higher than in temperate forests'". Around
one third of the animal biomass of the
Amazon terra firme rainforesi consists of ants
and termites, and each hectare of soil is
estimated to contain more than 8 million ants
and 1 million termites".

Numerous tropical moist forest species are
of conservation concern. Notable animals
include the Sumatran rhino Dicerorhinus
sumatrensis of Southeast Asia, endangered by
habitat fragmentation and hunting; the bonobo
Pan paniscus of the Democratic Republic of
the Congo, threatened by habitat destruction
and hunting for food; the Philippine eagle
Pithecophaga jefferyi of the Philippines,

reduced to small fragmented populations
through habitat loss and hunting; and the
indri Indri indri of eastern moist forest on
Madagascar, threatened by habitat destruction
as is the recently rediscovered Edward's
pheasant Lophura ec/wards/ of Viet Nam.

Role in carbon cycle

Lowland evergreen broadleaf rainforests can
have high above-ground biomass (Table 5.61,
though not as high as some giant conifer
forests. Soil carbon, however, is relatively low
in most tropical moist forests, with the
exception of the peat forests of Southeast Asia
and some swamp forests. On this basis it has
been estimated that the remaining tropical
moist forests store over 300 petagrams of
organic carbon, or about one fifth of global
terrestrial organic carbon. They account for

90 WORLD ATLAS OF BIODIVERSITY

Map 5.4

Biodiversity at country

level

Country-level biodiversity.
represented by an index
based on species diversity
in the four terrestrial
vertebrate classes and
vascular plants, adjusted
according to country area.
Countries at the high end of
the scale have a higher
value of the index than
w/ould be expected on area
alone. The index is
unreliable for smaller
countries (such as Togo and
Luxembourg in this plot].

Note: To reduce ambiguity
Alaska (United States! has for
the purposes of this map
been assigned to the same
class as adjacent Canada
rather than the conterminous
United States,

Source: Based on national biodiversity
indices developed by UNEP-WCMC; see
Appendix 5.

Diversity

High

Low

nearly a third of global terrestrial annual
net primary production (Chapter II, and are
therefore key to the global carbon cycle and in
regulating global climate.

Use by humans

Tropical hardwood species contribute almost
one fifth of world industrial roundwood
production^ Of the several thousands of
species [in more than 200 families in the
phylum Anthophytal that show commercial
potential, a few hundred may be found in
international trade. Important families
include Dipterocarpaceae, with species of
meranti and balau Shorea. and keruing
Dipterocarpus, Meliaceae, with mahogany
Swietenia and Khaya, and cedar Cedrela and
Toona; and Leguminosae with rosewood
Dalbergia and Pterocarpus.

The exploitation of tropical hardwood from
moist forests has been the subject of much
publicity in the past two decades, coinciding
as it has done with increased rates of
deforestation and forest degradation. Timber
supplies from some countries are now widely
exhausted, generating openings for other
countries to take over as suppliers. A few
major producers, however, continue to
dominate supply. Indonesia, Malaysia, Brazil
and India accounted for 80 percent of tropical
log production in International Tropical
Timber Organization (ITTOI countries in 1999-
2000-.

Important non-timber products from tropi-
cal moist forests include rattans, which are the
second most important source of export earn-
ings from tropical forests. A few other craft
products and some medicinal products, such as

Terrestrial biodiversity 91

the bark of Prunus africana. are significant in
international trade. Brazil nuts Berthoiettia
excelsa and native rubber Hevea brasiliensis
are other extractive products from natural
tropical moist forests that are important in
international marl<ets, and many tropical moist
forests provide fruit, bushmeat and other
products for local markets.

Other ecosystem services

Like other forest types, tropical moist forests
often play an important role in soil and water-
shed protection. The high rainfall regimes of
the humid tropics mean that exposed soil is
particularly liable to leaching of mineral
nutrients and to erosion. Montane forests, and
especially cloud forests, serve to intercept
and store water, thus regulating local and
regional hydrological cycles". Lowland forests

are also important in hydrological cycles; it
has been estimated that about half the rainfall
in the Amazon Basin is derived from water
recycled by forest transpiration"'.

Regenerating forest, or forest fallow, is an
important part of the cycle of shifting culti-
vation, which is vital for restoring fertility to
areas that have been previously cultivated.
This regeneration can only take place if
nearby forest cover is adequate to provide a
source of propagules.

Tropical dry forests

Distribution, types and characteristic taxa

Tropical dry forests are characteristic of areas
in the tropics affected by seasonal drought.
Such seasonal climates characterize much of
the tropics, but less than A million km' of
tropical dry forests remain. The seasonality of

92 WORLD ATLAS OF BIODIVERSITY

rainfall is usually reflected in the deciduous
habit of the canopy trees, with most being
leafless for several nnonths of the year
However, under some conditions, e.g. less
fertile soils or less predictable drought
regimes, the proportion of evergreen species
with leaves highly resistant to water loss
increases I'sclerophyllous' forest). Thorn
forest, a dense forest of low stature with a
high frequency of thorny or spiny species, is
found where drought is prolonged, and
especially where grazing animals are plenti-
ful. On very poor soils, and especially where
fire is a recurrent phenomenon, woody
savannahs develop" " (see sparse trees and
parkland' below).

Perhaps the best-known tropical dry
forest tree species is teak Tectona grandis
IVerbenaceae). a deciduous hardwood charac-
teristic of the seasonal forests of south
and Southeast Asia, widely exploited for
furniture and other uses, and now an important
plantation species. In Southeast Asia the
dipterocarps Shorea and Dipterocarpus, and
Lagerostroemia iLythraceae). and a number of
species of legumes ILeguminosae) are also
important components of seasonally dry
forests'. In Africa, dry forests occur both north
and south of the equatorial rainforests. In the
north they are characterized by Afraegete
IRutaceae). Diospyros (Ebenaceae), Kigelia
IBignoniaceae) and Monodora lAnnonaceae).
among other taxa, while in the south the
characteristic genera are Entandophragma
iMeliaceae). Brachystegia ILeguminosae),
Diospyros, Parinari IChrysobalanaceae).
Syzigium (Myrtaceae) and Cryptosepatum
ILeguminosae)"".

In the neotropics. tropical dry forests
occur in Yucatan and Pacific slopes of Central
America, on the leeward sides of Caribbean
islands, in Venezuela. Caribbean Colombia, in
northeast Brazil and in the Chaco region of
Bolivia and Paraguay (see Map 5.7). The neo-
tropical dry forests are quite rich in species
and include Leguminosae, Bignoniaceae,
Rubiaceae. Sapindaceae. Euphorbiaceae.
Falcourtiaceae. and Capparidaceae as the
families with the largest numbers of
species. Important genera include Tabebuia
(Bignoniaceae). Trichitia (Meliaceae). Eryth-

roxytum (Erythroxylaceae). Randia (Rubi-
aceae). Capparis (Capparidaceae), Bursera
(Burseraceae). Acacia (Leguminosae) and
Coccoloba (Polygonaceae)".

Structure and ecology

Tropical dry forests are generally of lower
stature than moist forests, with canopy
heights ranging from only a few meters to
30 m or occasionally AO m". The taller forests
have multilayered canopies. Dry forests tend
to have more small trees than moist forests
and a lower above-ground biomass. The trees
have a greater proportion of their total
biomass below ground, as more extensive
root systems help the trees to obtain water
from the soil and avoid drought. Dry forests
have a much lower incidence of epiphytes
than wet forest, and tend to have both higher
frequencies and higher diversity of vines
and lianas".

Plants with specialized mechanisms for
avoiding drought or conserving water are an
important feature of these forests. The
proportion of deciduous tree species is
thought to increase steadily with decreasing
annual rainfall, but factors such as the sub-
strate and the between-year variation of
seasonal rainfall patterns are also important.
Many species have water storage tissues such
as succulent stems or tubers, and specialized
photosynthetic mechanisms that conserve
water are especially common among the
epiphytes. Most tropical dry forest trees tend
to flower and sometimes to re-leaf before the
end of the dry season, and stored water within
the plant is essential to this pattern.

As in tropical moist forests, environ-
mental heterogeneity is linked with in-
creased species diversity. Gallery areas
along water courses are one source of such
variation, and they serve as a refuge for
animals during the dry season". Termite
mounds provide an important source of
environmental variation in African dry
forests, adding to local topography and
supplying high-nutrient microenvironments
to the system, increasing tree species
richness by ^0-100 percent". Fire is also a
major factor determining the dynamics and
extent of tropical dry forests.

Terrestrial biodiversity 93

•^Iri.

Biodiversity

Though of lower species richness than
tropical moist forests, dry forests still have
appreciably more tree species than most
temperate forests. The richest neotropical dry
forests, which are not the wettest ones but
those in western Mexico and in the Chaco of
southeast Bolivia, have around 90 woody
species per 0.1 hectare sample".

Although a comprehensive assessment
has yet to be made, dry forests are thought to
have high rates of plant species endemism
relative to wet forests in the tropics". Sixteen
percent of the plant species of the Chamela
dry forest in western Mexico are local
endemics, and 20 percent of the flora of
Capeira, Ecuador, are endemic to western
Ecuador". Many of the dipterocarps in
Thailand's seasonal forests are national
endemics and distinct from the species in the
country's moist forests'.

Vertebrate species diversity is lower in dry
forests than in moist forests, but many dry
forests have high rates of endemism among
mammals, especially among groups such as
insectivores and rodents, characterized by low
body weights, low mobility and short gener-
ation times". Among neotropical dry forests,
those of Mexico and the Chaco have the
highest numbers of mammal endemics (26
and 22 respectively"). Remaining areas of dry
forest are often important refuges for once
widespread species. The Gir forest of Gujarat
[India] contains the only population of Asiatic
lion Panthera leo persica, once found
throughout much of southwest Asia; the dry
forests of western Madagascar are inhabited
by around 40 percent of the island-endemic
lemurs; some, such as red-tailed sportive
lemur Lepitemur ruficaudatus, are almost
entirely confined to this habitat. Invertebrate
species richness tends to be poorly known, but
in groups such as lepidoptera and hymenop-
tera richness in some dry forest areas may be
comparable to adjacent wet forest*'.

Because of their high degree of endem-
ism and because degradation and conversion
have progressed further than in wet forests,
the biota of tropical dry forests are often
highly threatened. Hunting, especially for the
wildlife trade, and habitat conversion are

important pressures on dry forest animal
species.

Threatened dry forest species include Spix's
macaw Cyanospitta spixii, which is nearly
extinct globally as a result of trapping and
habitat loss in Brazil's northeastern caatinga
region; the Chacoan peccary Catagonus
wagneri, which was rediscovered in the Gran
Chaco of central South America during the
1970s and is threatened by overhunting,
habitat loss and disease; Verreaux's sifaka
Propithecus verreauxi of western Madagascar.
which is at risk from loss of spiny and gallery
forest habitat; and the Madagascar flat-tailed
tortoise Pyxis planicauda, which is restricted
to the western Andranomena forest of
Madagascar and is believed to be declining as
a result of habitat destruction.

Many of ihe endemic
lemurs of Madagascar are
found in dry forest.

M WORLD ATLAS OF BIODIVERSITY

.^5^=3' -^"a^s^

Map 5.5

Flowering plant family

density

Global diversity of flowering
plants represented as a
density surface derived
from distribution maps for
284 non-aquatic plant
families. Scale indicates
number of families present,
up to a maximum of 182,
divided into 15 classes.

Source: Compiled primarily frorr
Heywood'°° by UNEP-WCMC

Density

High

Low

Role in carbon cycle

Their lower biomass means that tropical dry
forests represent a smaller reservoir of stored
carbon per unit area than the other forest types
discussed so far With a total biomass ranging
from 98 to 320 metric tons per hectare" and soil
carbon storage in the region of 100 metric tons
per hectare''. It is unlikely that relatively undis-
turbed tropical dry forests store more than 250
metric tons of carbon per hectare (Table 5.61,
This, combined with the fact that little Intact
tropical dry forest remains worldwide, suggests
that the total contribution of seasonally dry
tropical forests to global carbon storage is far
less than that of other forest types.

Use by humans

Notable among the economically important
species of seasonally dry tropical forests is teak

Tectona grandis, which accounts for about 1
percent of reported global tropical timber
exports. More than ten species of Thailand's
seasonal forests are significant for the timber
trade', with timber species such as mahogany
Swietenia and several species of Tabebuia
(Bignoniaceae) characteristic of neotropical
dry forests. The southern dry forests of Africa
also contain useful timber species such as
Entandrophragma spp. Tropical dry forests also
provide large quantities of fuelwood for local
populations. A number of food plants are native
to tropical dry forests and medicinal uses have
been reported for many dry forest plant
species*'. Craft products are also Important.

Other ecosystem services

Protection of relatively fragile soils is a vital
ecosystem service provided by tropical dry

Terrestrial biodiversity 95

€^

forests. Rains may be intense during the wet
season, and erosion can be a severe problem in
tropical dry forest areas, where soils are often
thin and soil formation processes slow".
Tropical dry forests may also be important
resources for native pollinators as well as
nectar sources for domestic bees. Many dry
forest trees produce conspicuous flowers with
specialist pollination mechanisms. Their mass
flowering provides a major nectar resource for
pollinating insects at the end of the dry season
when other such resources may well be
limited" '^ Honey production is one of the liveli-
hoods being promoted for local communities in
dry forest areas in Mexico and elsewhere.

Sparse trees and parkland

Sparse trees and parl<land are forests with
open canopies of more than 10 percent crown

cover They occur principally in areas of
transition from forested to non-forested
landscapes. The two major zones in which
these ecosystems occur are the boreal region
and the seasonally dry tropics.

At high latitudes, north of the main zone of
boreal forest or taiga, growing conditions are
not adequate to maintain a continuous closed
forest cover, so tree cover is both sparse and
discontinuous. This vegetation is variously
called open taiga, open lichen woodland or
forest tundra". It is species poor, has high
bryophyte cover, and is frequently affected by
fire. It is important for the livelihoods of a
number of groups of indigenous people, inclu-
ding the Saami and some groups of Inuit.

In the seasonally dry tropics, decreasing
soil fertility and increasing fire frequency are
related to the transition from closed dry forest

96 WORLD ATLAS OF BIODIVERSITY

The large savannah
vertebrates of Africa are
largely absent from other
continents.

through open woodland to savannah. The
open woodland ecosystems include the more
open Srachysteg/a and /sofaer//n/a woodlands
of dry tropical Africa and parts of both the
caatinga and cerrado vegetations of Brazil''.
Open woodlands in Africa are more species
rich than either closed dry forest or savannah.
The cerrado supports a high diversity of
woody plants, though many of them are of
shrubby habit.

Animal diversity is generally low in forest
tundra; few species are restricted to this
habitat, many also occurring in boreal forest
or tundra proper The sparsely wooded tropi-
cal savannahs are generally more species
rich than temperate forests or grasslands".
Wooded savannahs vary greatly. Those of
America are relatively species poor, while
African savannah sometimes attains a rich-
ness not far below rainforest in the same
continent. Sparsely wooded areas in Australia
are amongst the richest wildlife habitats on
the continent, sometimes more so than
adjacent wet forests''''"'. The large savannah
vertebrates present in such high diversity
in Africa are largely absent from other

wmsii^^

Air^itlsiJfiilf s'WxHft'

continents''' (see Chapter 41. The density and
biomass of tropical savannah soil inver-
tebrates (mostly earthworms, ants and
termitesl is generally lower than temperate
grasslands, but greater (at least in biomassl
than tropical rainforests'".

Species of conservation concern include
the black rhinoceros Diceros bicornis, threat-
ened primarily by hunting for their horn, and

the golden-shouldered parrot Psephotus
chrysopterygius of northern Queensland
(Australia!, threatened by the burning of
seeding grasses during the breeding season
and predation by feral cats.

Forest plantations

Forest plantations, generally intended for the
production of timber and pulpwood, increase
the total area of forest worldwide. FAQ"
estimates that forest plantations covered 187
million hectares in 2000, of which Asia
accounted for 62 percent. This represents a
significant increase from the 1995 estimate of
124 million hectares. Commonly mono-
specific and/or composed of introduced tree
species, plantation forests tend to be less
valuable as a habitat for native biodiversity
than are natural forests. However, they can be
managed in ways that enhance their habitat
value. Plantations are also important pro-
viders of ecosystem services, such as main-
taining nutrient capital and protecting water-
sheds and soil structure as well as storing
carbon. They may also alleviate pressure on
natural forests for timber and fuelwood
production.

In some countries, wood from plantation
sources makes up a significant portion of the
industrial wood supply. For example. New
Zealand is more than self-sufficient in wood
production based on plantations". New forest
plantation areas are increasing globally at a
rate of 4.5 million hectares per year, with
particularly high rates of increase in Asia and
South America'''. In future, forest plantations
are likely to play an increasingly important
role in mitigation of greenhouse gases, as
encouraged by the Kyoto Protocol.

Changes in forest cover
About half of the forest that was present
under modern (i.e. post-Pleistocene) climatic
conditions, and before the spread of human
influence, has disappeared, largely through
the impact of human activities. The spread of
agriculture and animal husbandry, the har-
vesting of forests for timber and fuel, and the
expansion of populated areas have all re-
duced forest cover. The causes and timing of
forest loss differ among regions and forest

Terrestrial biodiversity 97

types, as do the current trends in ctiange in
forest cover

The temperate forests of western Europe
have diminished by far more than the
50 percent estimated for forests globally.
Much of this deforestation occurred between
7 000 and 5 000 years ago as Neolithic
agriculture expanded'-. The expansion of
human populations and increasing demand
for fuel during classical times and the Middle
Ages put further pressure on European
forests. Between Neolithic times and the late
11th century, forest cover In what is now the
United Kingdom decreased by 80 percent.
As European forests dwindled they became an
Increasingly valuable resource that was more
carefully managed. Forest cover stabilized
during the 19th century in much of western
Europe In response to both improved man-
agement and reduced demand for forest
products lowing to the increasing use of fossil
fuels and changes in construction materials).
Since the early 20th century, forest cover In
Europe has expanded, often through the
establishment of conifer plantations.

In eastern and central Europe and in
Russia, forest clearance accelerated during
the 16th and 17th centuries as sedentary
agriculture expanded. One estimate" sug-
gests that around 1 million km' of forest had
been cleared In the former Soviet Union up to
1980. Timber exploitation continues to drive
forest clearance In the coniferous forests of
Siberia and parts of eastern Europe.

In North America, Indigenous groups had
impacts on the forests from at least 12 000
years ago, but most forest clearance took
place after European settlement. Forest cover
In eastern North America reached Its
minimum around 1860, but then increased
with the westward movement of the agri-
cultural frontier and subsequent urbanization
and Industrialization. Forests west of the
Appalachians suffered the most severe
impacts In the late 19th and early 20th
centuries, but are still under pressure from
demand for timber and pulp.

In Oceania, just as in North America,
Indigenous groups had significant Impacts on
the forests before the arrival of Europeans.
This was especially true of the aboriginal

Region

Total forest area

Forest cover

change 1990-2000

In 2000 ImlUlon ha|

Annual change

Annual rate of

[thousand ha|

change l%|

Africa

650

-5 262

-0.78

Asia

548

-364

-0.07

Europe

1039

881

0.08

North and Central America 549

-570

-0.10

Oceania

198

-365

-0.18

South America

886

-3 711

-0.41

World total

38&9

-9391

-0.22

use of fire In Australia, but European colon-
ization greatly increased the rate of forest
conversion. Over 230 000 km' of forest and
120 000 km- of woodland in Oceania are
estimated to have been converted to cropland
between 1860 and 1980".

In tropical Asia, Africa and Latin America,
large-scale deforestation was precipitated
by European colonial activities. Including
agriculture and timber exploitation. It Is
estimated that more than 1 million km-' of
forest and a similar amount of woodland in
tropical developing countries were converted
to cropland between 1860 and 1980'''. The bulk
of this conversion was in south and Southeast
Asia, where forest area declined by 39 percent
from 1880 to 1980.

Globally, tropical dry forest has lost the
greatest proportion of Its original area of
the four major types of closed forest, nearly
70 percent. About 60 percent of the original
area of temperate broadleaf and mixed
forests has disappeared, and tropical moist
and temperate needleleaf forests have lost
about 45 percent and 30 percent of their
original area, respectively.

Current trends in change in forest cover,
which are shown in Table 5.8, reveal that the
rates of deforestation continue to be high In

Table 5.8

Estimated annual change

in forest cover 1990-2000

Note: Figures refer to natural
forests and plantation forests
combined.

Table 5.9

Global protection of

forests within protected

areas in lUCN categories

l-VI

Note: Forest areas and
protection as assessed in
1999.

Source: UNEP-WCMc".

Forest type

Global protection

Ikm^l

i%i ^^

Temperate and boreal needleleaf

675 470

5.4

Temperate broadleaf and mixed

457 535

7.0

Tropical moist

1 382 004

12.2

Tropical dry

413 524

11.2

Sparse trees and parldand

274 401

5.8

Total

3 202 934

8.3

98 WORLD ATLAS OF BIODIVERSITY

Fuelwood and charcoal
consumption more than
doubled between 1961
and 1991.

the developing countries of the tropics, in both
absolute and proportional terms. In contrast,
temperate countries are losing forests at
lower rates, or indeed showing an increase in
forest area".

Pressures on forest biodiversity

The principal pressures on forests and their
biodiversity are conversion to other landuses,
principally forms of agriculture, and logging.
Conversion of forest to agriculture is the main
cause of tropical moist forest loss. This is
largely due to expanding populations and the
use of shifting cultivation at an intensity that
does not permit adequate fallow periods.
Government resettlement programs that have
moved large numbers of poor farmers have
increased the rate of land colonization and
clearance In parts of Southeast Asia and Latin
America. In some areas, land has been

converted to ranching principally as a means
of gaining title In order to permit speculation
in land values. Thus, population growth,
poverty and inequitable land tenure are
among the causes underlying deforestation by
conversion to agriculture.

Timber extraction puts great pressure on
biodiversity in both tropical and temperate
forests. Global consumption of industrial
roundwood was more than 1 521 million m' in
1998" and, on current trends, is projected to
continue to rise. Although some timber
species are naturally abundant, a factor that
can help ensure their survival from com-
mercial exploitation, many have suffered

extensive and irreversible population and
genetic losses. Furthermore, rare species
that are indistinguishable In the field from
their commercially important relatives are in
danger of extinction through overexploitation.
This particular problem exists, for example,
among the dipterocarp groups meranti, balau
and l^eruing. A tew hundred species may be
traded under these names, and a significant
proportion of these are geographically and
ecologically restricted and so at high risl<
of extinction.

Furthermore, logging operations create
access to forest areas that may otherwise
have remained Isolated. This improved access
facilitates hunting and other activities that
exert pressure on forest biodiversity, and may
ultimately lead to colonization and conversion
of the land to agricultural use. There is also
strong evidence that logging can increase the
probability of wildfire In temperate forests
and even in tropical moist forests not usually
subject to burning".

Particularly in the drier areas of the
tropics, fuelwood extraction can have serious
impacts on forests and open woodlands.
Fuelwood and charcoal consumption more
than doubled between 1961 and 1991, and is
projected to rise by another 30 percent to
2 395 million m= by 2010". About 90 percent of
the consumption is In developing countries,
but wood fuel may play an Increasing role In
some developed countries. Increasing demand
for wood still further".

In addition to loss of area, forest conver-
sion and logging lead to changes In the
condition or quality of the remaining forest.
These can include fragmentation of large
areas of continuous forest. Tropical forest
fragments are distinct from continuous for-
ests in both ecology and composition". There
are physical and biotic gradients associated
with fragment edges, and forest structure
undergoes radical change near the edges as a
result of the impacts of wind and increased
tree mortality. Some animal species are
'edge-avolders' and decline in abundance In
forest fragments, while others become more
abundant. Some non-forest and even non-
native species of plants and animals success-
fully invade forest fragments but not con-

Terrestrial biodiversity 99

"^'ik

tinuous forest. In addition to directly affecting
canopy composition, removal of large timber
trees may also affect the availability of seed
for regeneration and may affect animal
species that depend on the timber species.

Other factors that affect forests and their
biodiversity include acid rain and global
climate change. So far. most of the effects of
acid precipitation, which Is caused by indus-
trial air pollutants, have been documented in
temperate needleleaf forests and associated
vi/aterways of Europe and North America.
The likely impacts of global climate change
on forests are still being debated, but there
seems to be general consensus that the
boreal coniferous forests are particularly
vulnerable to both range restrictions and
Increasing fire frequency".

Another forest type that has been shown to
be vulnerable to climate change is tropical
montane cloud forest, which depends upon
clouds to supply It with atmospheric moisture.
Research has shown that the mean cloud
base is moving upwards on tropical moun-
tains as a result of climatic shifts. The forest
species are not able to migrate at a
comparable rate and. In any case, range shifts
will be limited by the land area existing at
higher elevations. Local extinctions in cloud
forest amphibians, including the golden toad
Bufo pengtenes assessed as critically
endangered, have been attributed to climatic
fluctuations that may be linked to long-term
global climate change".

NON-FOREST ECOSYSTEMS

The parts of the Earth that are too cold, too
dry or too severely affected by fire and/or
grazing do not support forest or woodland
ecosystems. However, as can be seen from
Map 5.1, many of them do support active plant
growth. Natural non-forest ecosystems in-
clude tundra (both arctic and montanel,
grasslands and savannahs, and shrublands.
Less productive, but with unique elements of
biodiversity, are the deserts and semi-deserts
IMap 5.4).

Tundra

Tundra Is the vegetation found at high
latitudes beyond the limits of forest growth;

the same term is sometimes used for similar
vegetation at high elevation at lower latitudes,
but these may be distinguished as polar
tundra' and 'alpine tundra', respectively. In
the Arctic, polar tundra occurs north of the
northern tree line, which Is determined by a
number of climatic factors Including the
summer position of arctic air masses" and
the depth of permafrost (permanently frozen

In polar tundra systems,
temperatures are low for
much of the year

subsurface soil]. Similarly, alpine tundra
occurs above the climatic tree line on
mountains, and Its elevation varies in a
complex fashion with latitude, continental or
oceanic climate, and the maximum elevation
and overall size of the massif™.

The characteristics of polar and alpine
tundra environments differ In many respects.
In high-latitude polar tundra systems,
temperatures are low for much of the year,
while permafrost limits both drainage and
root extension, and the growing season may
last for as little as six to ten weeks. Rainfall is
low. usually less than 200 mm per year, and
at extreme latitudes may be so low that the
environment Is described as polar desert.
At high elevation In temperate regions
temperatures may be similarly low. although
permafrost Is rare. However, at high altitudes
in the tropics, although low temperatures
occur every night, high Insolation causes
warming during the day so that adequate
temperatures for active plant growth occur

100 WORLD ATLAS OF BIODIVERSITY

<~ ^3,-^^^^

Map 5.6

Terrestrial vertebrate

family density

Global diversity of
terrestrial vertebrates
represented as a density
surface derived from
distribution maps for all
350 non-aquatic families of
mammals, birds, reptiles
and amphibians. Scale
indicates number of
families present, up to a
maximum of 126, divided
into 15 classes.

Source: Compiled from multiple sources
byUNEP-WCMC.

Density

Higti

Lovw

tfiroughout the year and the diurnal temper-
ature range is large.

Despite these environmental differences,
polar and alpine tundra vegetation have some
features in common. Both lack trees, but
contain woody species growing in dwarf or
prostrate forms, especially in locations with
less extreme climates. As latitude or altitude
increases, grasses, sedges, bryophytes and
lichens increase in importance while shrubs
decrease. Many plants have tussock or cush-
ion growth forms. At extremes of latitude or
elevation a high proportion of bare ground is
characteristic.

In the arctic tundra, plants cover 80-100
percent of the ground", and cover decreases
along the climatic gradient to polar desert.
The important woody plants are birches
Betula, willows Salix and alders Atnus. These

#S7^

are all genera that occur as trees in
temperate regions but as spreading, prostrate
or dwarf forms, sometimes less than 20 cm
tall, in tundra. Other shrub species are also
important, including Dryas, Vaccinium and
Empetrum. Interspersed with the shrubs and
of increasing importance in more extreme
sites are the sedges Carex and the cotton
grasses Eriophorum, among other gramin-
oids, and a high diversity of mosses and
lichens. Biomass is often much lower above
than below ground, and annual production is
low. This combination means that the poten-
tial of these ecosystems for recovery following
disturbance is limited".

Temperate alpine systems are charac-
terized by many of the same taxa as the arctic
tundra, but at lower latitudes other groups
become important. At high altitude in the

Terrestrial biodiversity 101

tropics, giant rosette plants are a notable
feature of alpine communities. These dis-
tinctive plants, whicfi have a number of
morphological and physiological adaptations
to the high insolation, large temperature
fluctuations and desiccating conditions on
tropical mountains, include Espeletia and
Puya in the paramos of the Andes, and
Senecio and Lobelia in the high mountains
of Africa".

In comparison with forested ecosystems,
both polar and alpine tundra are relatively
species poor It is estimated that species
richness declines by a factor of between
three and four between the boreal and arctic
zones", and species richness in the polar
desert is one fifth that of the tundra. The
entire North American Arctic has a vascular
flora of about 600 species". Bryophytes and

lichens may add more than 300 additional
species to the circumpolar flora. The tropical
alpine systems are richer; Venezuelan
paramos include more than iOO angiosperm
species'', but this is still much lower than in
surrounding forests.

Because most arctic plant species have
wide geographic distributions, few are of
significant conservation concern, but many
alpine areas are isolated by lowlands with
contrasting climates so increasing endemism
in many alpine floras. The floras of isolated
mountains in North America have significant
rates of endemism"; about 15 percent of the
flora above the timber line in the Alps are
locally endemic species" and about 80 percent
of the flora at high altitude in East Africa and
Ethiopia are endemic". Restricted distrib-
utions tend to make species more vulnerable

102 WORLD ATLAS OF BIODIVERSITY

to extinction, while the harsh environment of
alpine regions adds to the risl<.

Animal species richness tends to be low;
groups represented by several species in
boreal forest are often reduced in diversity by
up to one third in tundra habitat". In contrast,
a few groups, particularly water birds and
waders, are able to exploit the large numbers
of Invertebrates found in tundra soil and can
be both diverse and abundant". Although the
species richness of the most common in-
vertebrate groups (Collembola and oribatid
mitesi decreases with increasing latitude,
their total abundance may increase.

There are relatively few globally threaten-
ed species that are completely dependent on
tundra. An exception Is one of the worlds
most severely endangered species, the once
abundant Eskimo curlew Numenius borealis
of the Americas, which nests almost ex-
clusively within this habitat. Two globally
threatened bird species, Stellers [Polysticta
stelleri] and spectacled elder [Somaleria
fischeri] remain within the Arctic throughout
the year Although low In number of species,
the Arctic is home to most of the world's
geese and calidrid sandpipers".

Although the tundra accounts for less than
2 percent of global annual net primary
production (Table 1.11, the high below-ground
biomass and soil carbon in arctic tundra
means that it makes an important contribution
to global carbon stocks. Total biomass in
tundra communities of the Russian Arctic falls
in the range 7-30 metric tons per hectare, of
which some 60-70 percent is below ground".
Sedge-moss communities in the North
American Arctic may have 15-30 times as
much biomass below ground as above
ground", and three to eight times the amount
of dead as live material may accumulate.
Tundra soils may store around 200 metric tons
of carbon per hectare^'.

Because of Its Inhospitable climate, tundra
is not subject to severe pressure for
conversion for other landuses. However, Its
lack of ecological resilience means that
disturbances, e.g. those associated with
settlements or long-distance pipelines, tend
to have long-lasting effects. It is anticipated
that the effects of global warming on the

arctic tundra will be significant, as relatively
large temperature increases are predicted
for this zone". These will cause changes in
the permafrost regime and decomposition
of accumulated soil organic matter, which in
turn will release additional carbon dioxide
into the atmosphere. Evidence suggests that
the period of active plant growth has recently
lengthened in parts of North America. There
is also evidence that species are already
migrating northwards in response to climate
change, so that arctic tundra Is likely to be
compressed Into a much smaller area of
remaining appropriate climate.

Grasslands and savannahs
Grassland ecosystems may be loosely defined
as areas dominated by grasses (members of
the family Gramlneae excluding bamboos] or
other herbaceous plants, with few woody
plants'". Grassland ecosystems are typically
maintained by drought, fire, grazing and/or
freezing temperatures™. In addition, they are
often associated with soils of low fertility'".
Savannahs are tropical ecosystems charac-
terized by dominance at the ground layer of
grasses and grass-like plants. They form a
continuum from treeless plains through open
woodlands to closed-canopy woodland with a
grassy understorey. Some savannah areas
therefore meet general definitions of grass-
land (fewer than 10-15 woody plants per
hectare"! while others meet the definition of
woodland. Some polar and alpine tundra
communities also meet the definition of
grassland.

Around 20 percent of the Earth's land
surface (excluding Antarctica) supports
grassland, with these regions differing greatly
In naturalness'". Temperate grasslands make
up approximately one quarter of this area, and
savannahs the remainder (tvlap 5.81. The most
extensive areas of temperate grasslands are
the prairies of North America, the pampas
and campos of southern South America and
the steppes of central Europe, southwest and
central Asia and Russia. Temperate grass-
lands are sometimes divided Into formations
of tall grass, mixed grass and short grass,
which differ both floristically and ecologically;
short-grass communities are usually associ-

Terrestrial biodiversity 103

•^%

ated with drier climatic regimes™. Tropical
grasslands and savannahs include the llanos
of the Orinoco basin in Venezuela and
Colombia, the cerrado of central Brazil, and
the savannahs of tropical and subtropical
Africa, in addition to many species of grasses,
the sedges ICyperaceae) and many different
groups of dicotyledonous herbs are also
important.

Key ecological factors in grasslands are
grazing pressure and the effects of fire. Most
natural grasslands had, at one time, large
populations of native grazing mammals.
These have been replaced to a great extent by

environments, are often rich in organic matter
and are therefore particularly vulnerable to
conversion to cropland, v\/ith replacement of
native grasses by their domestic derivatives
Icerealsl and other plants".

At very tine spatial scales, natural grass-
lands can be extremely species rich. For
example, a square meter of 'meadow steppe'
in the former Soviet Union may have 40-50
plant species"', a tall grass prairie remnant of
less than 2 hectares may contain 100 species,
and 250 hectares may contain 250-300 plant
species". However, grassland communities
tend to be similar over large areas, and

The world's grasslands and
savannahs support
distinctive plant and animal
communities.

domesticated ungulates, which also exert a
significant degree of grazing pressure [the
magnitude depending on stocking densities).
Grazing tends to increase abundance of less
palatable species and to increase species
richness in productive areas, or decrease it in
less productive areas. At intermediate fre-
quencies, fire tends to increase diversity and
suppress invasion by woody species. Frequent
fires favor grasses, which usually recover
easily, whereas low fire frequency may allow
the density of woody species to increase.

These factors have important conse-
quences for vegetation structure in grassland
and savannah systems. A high proportion of
plant biomass, in the form of roots and
rhizomes, is located underground; there is a
high turnover of those parts of the plant above
ground". One important consequence of this is
that grassland soils, especially in more humid

structurally simple, so that at the landscape
scale diversity is relatively low compared with
tropical moist forest or t^^editerranean-type
ecosystems. Grasslands tend to have low
rates of endemism; however, the climatic
and soil gradients within them have led
to substantial ecotypic variation and high
genetic diversity".

The world's grasslands and savannahs
support distinctive plant and animal com-
munities. Although species diversity tends to
increase towards the tropics, it tends to be
moderate or low at the landscape scale and
above. Little more than 5 percent of the
world's mammal and bird species are pri-
marily dependent on grasslands habitats'".
All these ecosystems hold, or formerly held,
an array of native herbivores, and these in turn
support a number of high-profile mammalian
and avian predators.

104 WORLD ATLAS OF BIODIVERSITY

, Box 5.5 Grassland bird trends
k

The figure illustrates the living planet index in a sample of 29 grassland birds from North
Annerica 125 species! and Europe |4|. The clear trend over four decades is dc»/nward, and as
with forest birds the separate continent samples show a similar overall pattern.

UOr

120

100

1970 1975 1980

Source: UNEP-VtfCMC. from data collated lor Loh".

1985

1990

Ttie savannafi communities of East Africa
are typified by large herds of ungulate
herbivores, including a remarkable diversity -
more than 70 species - of antelope and other
medium- to large-sized bovids. The biomass of
ungulate herbivores here may rise to 30 metric
tons per km*, which is the highest recorded in
any terrestrial environment"'. Most grassland
invertebrate biomass is found within the soil
(most commonly nematodes, enchytraeid
worms and mitesi and may be in the order of
100 to 1 000 times as great as vertebrate
biomass; soil invertebrate biomass above the
soil is often dominated by Orthoptera"'.

Ivlany grassland birds require targe areas
of habitat to take full advantage of sparsely
distributed food resources, and fragmentation
of natural grasslands has made conservation
of these wide-ranging species difficult'".
An analysis of the location of bird species in
the neotropics indicated that nearly 12
percent of threatened birds 138 species! are
confined to grasslands. In this region, the
grasslands of southern Brazil and northern
Argentina are particularly threatened as a
result of agricultural improvement". In North
America, grassland species have experienced
the most consistent declines of any group of

birds monitored in a national survey of
breeding birds; available data suggest that
there has been a constant decline in these
species over the past 30 years'".

Among notable threatened mammal
species in grasslands are the greater one-
horned rhinoceros Rhinoceros unicornis.
associated with tall riverine grassland in
northern India and southern Nepal, where it
is threatened by poaching and further loss of
its restricted habitat. Other targe mammals
are the saiga antelope Saiga tatarica of
central Asian grasslands, and the vicufia
Vicugna vicugna in arid grasslands and plains
in the Andes; both of these are threatened in
parts of their range. Among birds, the short
grassland habitat of the plains wanderer
Pedionomus torquatus of southern Australia
continues to be lost to cultivation; Rudd's lark
Heterornirafra ruddi inhabits the montane
grassland plateaus of eastern South Africa
and is threatened by habitat degradation;
white the lesser ftorican Syptieotides indica,
of western India and Nepal, is critically
endangered through toss of suitable grass-
land habitat.

Grassland biomass ranges from around
2 to over 80 metric tons per hectare". As a
rule, more than 50 percent of this is below the
soil surface, and the ratio of root to shoot
biomass ranges from betow five in warm
humid grasslands to as much as 30 in the
desert grassland of Ivlongotia". Usually more
than half of the below-ground biomass is in
the upper 10 cm of the soil, and soil carbon
stocks may be as high as 250 metric tons per
hectare-'. Annual production reaches 30-50
metric tons per hectare per year in some
warm humid grasslands. It is estimated that
grasslands and savannahs together account
for about a third of global terrestrial net
primary production (Table 1.1], so despite their
relatively low biomass grasslands play an
important rote in the global carbon balance.

Because the high below-ground biomass
tends to increase the fertility of their soils,
grasslands have been subject to high rates of
conversion to agriculture. Less than half of
North American grasslands remain in natural
or semi-natural states"', the steppes of the
former Soviet Union have been extensively

Terrestrial biodiversity 105

•^%

irrigated and converted to agriculture and
much of the pampas has been converted to
agriculture or grazing land. Some anthropo-
genic grassland consists of short-term
monospecific sown pasture, VKhile some
areas support species-rich semi-natural
grassland created over centuries by pastoral-
ists in conjunction with livestock grazing.

Domestic livestock grazing is the most
extensive human use of unconverted or
anthropogenic grassland ecosystems, and
of most arid or semi-arid ecosystems.
Livestock have an impact on ecosystems
through trampling, removal of plant biomass,
alteration of plant species composition
through selective grazing, and competition
with native species. The impact of this on the
biological diversity of these ecosystems has
been variable. In some areas where the
native vegetation is well adapted, the impact
on plant species diversity has been relatively
small. In other areas, where vegetation has
not evolved in the presence of hooved
herbivores, the changes have been great.
Sometimes, particularly in tropical and semi-
tropical grasslands, the dominant com-
ponent of vegetation has shifted from grass
to woody plants.

Overgrazing can lead to reduction in plant
cover, loss and degradation of soil, and
invasion by non-native plant species. In
almost all cases, wild animal diversity has
been greatly affected (mostly through compe-
tition and hunting, but also through spread of
pathogens), so that the biomass of domestic
livestock greatly exceeds that of native wild
herbivores. In some areas, feral species (e.g.
rabbits, camels, donkeys, horses, goatsi may
also have a marked impact on natural or
semi-natural ecosystems.

Shrublands

Shrub communities, where woody plants,
usually adapted to fire, form a continuous
cover, occur in all parts of the world with 200-
1 000 mm of rainfall". In more arid areas
including some semi-desert systems, shrubs
are the dominant life form, but cover is
discontinuous. The most distinctive and best-
known shrublands are those of Mediterranean
climate regions.

Mediterranean climates are typified by
cool, wet winters and warm, or hot, dry sum-
mers. However, no single climatic or bio-
climatic definition of a Mediterranean eco-
system has yet been established, so that
these areas remain rather loosely defined.
Mediterranean ecosystems encompass a
wide range of habitat types including forest,
woodland and grassland, but are typified by
a low, woody, fire-adapted sclerophyllous
shrubland (maquis, chaparral, fynbos, malleel
on relatively nutrient-poor soils. These sys-
tems occur in five distinct parts of the world:
the Mediterranean basin; California (United
States); central Chile; Cape Province (South
Africa): and southwestern and south Australia.
Each of these regions occurs on the west
side of a continent and to the east of a cold
ocean current that generates winter rainfall.
They cover around 2.5 million km' in total,
or between 1 percent and 2 percent of the
Earth's surface (according to definition). More
than two thirds of the total Mediterranean-
type ecosystem area is found within the
Mediterranean basin.

'^^^^^r

Approximate area

Total no. of '

[million km'l

olant soecies |

Cape Province, South Africa

0.09

8 550

Southwestern Australia

0.31

ca 8 000

California

0.32

5 050

Chile

O.U

ca2100

Mediterranean basin

1.87

25 000

Differences in vegetation structure between
regions are in part a consequence of differ-
ences in the annual distribution of rainfall.
In South Africa the sclerophyllous fynbos
community contains an abundance of erica-
ceous species as an understorey to low
broader-leaved shrubs including members of
the Proteaceae and Myrtaceae".

The Australian heaths are structurally
similar, with Epacridaceae replacing Ericaceae.
Californian shrublands, known as chaparral.
are characterized by Adenostoma (Rosaceae)
and a high richness of Arctostaphylos species
and other members of the Ericaceae. The
shrublands, or matorral, of Chile include many

Table 5.10

Estimated plant species
richness in the five
regions of Mediterranean-
type climate

Source; UNEp".

106 WORLD ATLAS OF BIODIVERSITY

Map 5.7

Current forest distribution

Map adapted from tfie
global landcover
classification developed by
the University of Maryland.
The Maryland classification
includes 13 classes and
was based on AVHRR
remote-sensing data with a
spatial resolution of 1 km.
The map shown was
derived by reclassifying the
Maryland landcover data to
accord with an ecologically
based classification of five
major forest types. For
presentation purposes the
data have here been
generalized to a i-km grid.

Source; Data from University of
Maryland Global Land Cover Facility For
full description see Hansen .

Forest type

Tropical moist

Tropical dry

Temperate broadleaf and mixed

Temperate and boreal needleleaf

Sparse trees and parkland

of the same genera as those of California, while
in the Mediterranean basin itself Ericaceae,
Cistaceae, Leguminosae and Oleaceae are
all important.

Species richness in Mediterranean-type
ecosystems, particularly among plants, is
generally high - approaching values for
moist tropical forest areas - and levels of
endemism are also very high. Among the five
Mediterranean-type ecosystems, species
richness appears highest on the poorer soils
of South Africa and southwest Australia
(Table 5.101, and lower on the richer soils of
California, Chile and the Mediterranean
basin^'. Countries around the Mediterranean
Sea hold some 25 000 vascular species
(about 10 percent of all vascular plants! of
which around 60 percent are endemic to the
Mediterranean region.

The remaining four Mediterranean-type
ecosystem regions are all considered to hold
a disproportionately high floristic diversity in
relation to their area".

At fine scale, mean plant richness in the
fynbos of South Africa is moderate, i.e. around
16 species per m*. but many species have
small ranges, and there is a uniquely high
turnover in the species composition of plant
communities along ecological and geogra-
phical gradients. At landscape scale, richness
accordingly rises to very high values, for 2 256
species occur in 471 km' on the Cape
peninsula and the entire Cape floristic region
(including some non-fynbos vegetation) holds
some 8 550 species, about 70 percent of which
are endemic.

The Mediterranean-type ecosystems in
general have a relatively high proportion of

Terrestrial biodiversity 107

their species categorized as ttireatened.
The Cape flora, largely within a Mediterranean-
type ecosystem, occupies only h percent of the
land area of southern Africa, but accounts for
nearly 70 percent of the region's threatened
plant species. About one third of the natural
vegetation has been transformed by human
activity; the remaining natural vegetation is at
risl< from a number of invasive introduced
woody plants, and the effects of an introduced
ant (that suppresses native seed-storing ants
and thus renders seed liable to destruction
by rodents or firel. Around 10 percent of the
Californian flora is considered threatened
(equivalent to approximately one quarter of
all threatened plants in the United States).
In Australia, heath habitats, primarily in the
southwest Mediterranean-type ecosystem
region, rank third after 'woodland' and 'scrub'

in numbers of 'endangered' category plants.
Given their much smaller extent, this indi-
cates that a far higher proportion of their flora
is threatened than in either woodland or
scrub habitats.

Vertebrate diversity tends to be lower than
in plants. To take an example, in the Cape
Mediterranean-type ecosystem, reptile diver-
sity IS only moderate while bird and mammal
diversity is relatively low. The absence of large
mammals in California and the Mediterranean
basin may be linked to overhunting by
humans during the late Pleistocene".

Several threatened animal species rely on
shrubby or scrub habitat. The Iberian lynx
Lynx pardinus, found in the light woodland
and maquls of Spain and Portugal, where
habitat loss and hunting have led to decline, Is
possibly the most threatened cat species.

108 WORLD ATLAS OF BIODIVERSITY

Map 5.8

Non-forest terrestrial

ecosystems

Map adapted from tfie
global landcover
classification developed by
the University of Maryland.
The Maryland classification
includes 13 classes and
was based on AVHRR
remote-sensing data with a
spatial resolution of 1 km.
The map shown was
derived by reclassifying the
Maryland landcover data to
accord with a highly
generalized classification of
non-forest terrestrial
ecosystem types. For
presentation purposes the
data have here been
generalized to a 4-km grid.

Source; Data from University of
Maryland Global Land Cover Facility. For
full description see Hansen .

The riverine rabbit Bunotagus monticutaris of
South Africa is restricted to a small area of
riverine bush in the central Karoo, where it is
threatened by further loss of this habitat to
agriculture. Among birds, the island cisticola
Cisticola haesitatus is endemic to the island
of Socotra in the western Indian Ocean, where
it is threatened by loss of light scrub and
grassland habitat, possibly through over-
grazing by goats.

Mediterranean-type shrublands are not
notably high in either biomass or net primary
production, Biomass at mature fynbos sites is
typically 15-16 metric tons per hectare, and in
chaparral may be twice that". Combined with
their relatively low rates of primary production,
related to both climate and soil fertility factors,
the incidence of fire tends to reduce the
accumulation of carbon in these ecosystems

and their soils. Total carbon storage is probably
between 100 and 150 metric tons per hectare
in Mediterranean-type shrublands''.

The Mediterranean basin itself has for
many centuries been subject to intense
human activities, including forest clearance
and grazing, such that little genuinely natural
vegetation remains. It has been suggested
that the plant diversity is locally high because
of the number of species that have evolved
as components of successional vegetation in
response to frequent disturbance, in other
Mediterranean-type shrublands. expanding
human populations and conversion of land to
agricultural or residential use are important
pressures. These changes are often accom-
panied by changing fire and grazing regimes,
and both these changes tend to facilitate
invasion by non-native plant and animal

Terrestrial biodiversity 109

'■'^ i.

species, which threaten native species pop-
ulations, especially in California and South
Africa™.

Deserts and semi-deserts

Nearly 10 million km' of the Earths land area
is hyperarid, or true desert, where rainfall is
extremely low and unpredictable in space and
time, so that in some years none falls at all.
These areas have a ratio of rainfall to
potential evapotranspiration (P/PETI of less
than 0.05". The Sahara desert alone makes
up nearly 70 percent of the world hyperarid
zone. Other extensive areas are found in the
Arabian Peninsula and central Asia, with
smaller areas in southwest Africa, the Horn of
Africa, western South America and western
North America. In semi-deserts, areas with
less arid climates, the vegetation is usually

more substantial than in deserts, but covers
no more than 80 percent of the ground.
Temperate deserts and semi-deserts cover
nearly 6 million km' in Eurasia and North
and South America". Polar regions and some
high mountain areas with a permanently cold,
dry climate also meet the definition of desert,
but have completely different ecological
characteristics from true drylands and are not
usually considered with them.

Plant cover in desert ecosystems ranges
from areas without vegetation to areas with
low densities of small shrubs and perennial
grasses, often with populations of annuals
that vary in density depending on seasonal
precipitation. Deserts are often character-
ized by short periods of relatively high pro-
ductivity during rainy periods, interspersed
with long periods of very low productivity. As

110 WORLD ATLAS OF BIODIVERSITY

Desert species show a
wide range of adaptations
to extreme environments.

Dating back some
15 million years,
ttie Namib appears to be
Earth's oldest desert.

a result, most herbivorous insects develop
only one generation per season, leading to a
time lag in the interaction between herbi-
vorous insects and their food resources".
Consequently, most of the primary produc-
tivity in deserts ends up as dry plant material
or seeds, which accumulates because micro-
bial decomposition is limited by the low
moisture availability. These resources form
the basis for populations of detritivores such
as termites, darkling beetles and isopods,
and seed predators such as ants. These
groups can support a rich fauna of predatory
arthropods and reptiles as well as omni-
vorous birds and mammals. It is often
assumed that all deserts have low species
diversity because of the harsh environmental
conditions, but among animals almost all

—

^^B^^

■KTl

■

^^!^5|^B

?^^^SI

terrestrial higher taxa are represented, and
their species diversity may in some situ-
ations be comparable to that of more mesic
habitats".

True desert species show a wide range of
adaptations to the conditions of an extreme
environment. Characteristic plants Include
the Cactaceae In the Americas and the
succulent Euphorbiaceae in Africa. Semi-
desert species include salt bush Atnplex, and
creosote bush Larrea. Amongst animals,
groups that are Intrinsically adapted to very
low moisture environments Include reptiles
and many arthropods, although species in a
wide range of other groups have also evolved
to cope with these conditions. Strategies for
survival amongst both plants and animals
often include long periods of dormancy (as
seeds, in the case of many plants! between
rainfall events.

In the particular conditions prevailing In
the so-called fog deserts (notably the
Atacama desert and the Skeleton Coast
desert], different strategies have evolved.
Here plants and animals make use of the
regular moisture-laden fogs, which roll in
from the cold offshore currents, to obtain a
low but predictable supply of water

The often overlooked inland water habitats
of deserts may contain a particularly high
proportion of locally endemic species; the
pools of Cuatro Clenegas, tvlexico, for example,
contain numerous mollusk and fish species
found nowhere else.

Many of the larger desert and sub-desert
vertebrates are threatened; the openness of
these arid areas means that species such as
antelopes and other bovids are more con-
spicuous than forest species and thus more
vulnerable to overhuntlng. Threatened verte-
brates include the wild bactrian camel
Camelus bactrianus with a few remnant
populations in the Gobi desert of Mongolia
and China, and Przewalskl's gazelle Procapra
przewatskii of China's subdesert steppes, now
restricted by overhuntlng and habitat loss to a
few small areas surrounding Lake Quinghai.
The Mexican prairie dog Cynomys mexicanus
is confined to prairies and intermontane
basins with herbs and grasses where It is
threatened by persecution and continuing

Terrestrial biodiversity in

habitat loss. The Addax antelope Addax
nasomaculatus was originally widespread
from the western Sahara to Egypt and Sudan,
but as a result of uncontrolled hunting it
persists only in small and scattered local
populations and is extinct through much of its
former range.

Principal threats to desert environments
include the activities of domestic livestock
such as cattle, which cause soil compaction
through trampling, and can damage vege-
tation and waterholes. Introduced species,
such as rabbits in Australia, have been highly
damaging to some desert environments.
Other human activities that have affected
desert habitats include use of off-road
vehicles, irrigation and afforestation schemes,
and housing projects. In drylands, most

REFERENCES

adverse impacts that lead to some form of
land degradation can be categorized as
'desertification'. Under the UN Convention to
Combat Desertification, the latter term is
defined explicitly as land degradation in arid,
semi-arid and dry sub-humid areas resulting
from various factors, including climatic
variations and human activities'. According to
the above definition, hyperarid lands [true
deserts! are not susceptible to desertification,
because their productivity is already so low
that it cannot be seriously decreased by
human action. The effects of desertification
on arid and semi-arid areas promote poverty
among rural people, and by placing greater
pressure on natural resources, poverty tends
to reinforce any existing trend toward
desertification.

1 Loveland, T.R. et al. 1999. An analysis of the IGBP global land-cover characterization
process. Photogrammetric Engineering and Remote Sensing 65: 1021-1032.

2 ITTO 2001. Annual review and assessment of ttie world timber situation 2000.
International Tropical Timber Organization.

3 Salisbury, H.E. 1989. The great black dragon fire: A Chinese inferno. Little, Brown and
Co., Boston.

A Farjon, A. and Page, C.N. 1999. Status survey and conservation action plan: Conifers.
lUCN-the World Conservation Union, Gland.

5 Hoang Ho-dzung 1987. The moss flora of the Baektu mountain area. In: Yang, H. et al.
(edsl. The temperate forest ecosystem, pp. 29-31. Institute of Terrestrial Ecology
Symposium 20, National Environment Research Council, UK.

6 Vaisanen, R., Bistrom, 0. and Heliovaara, K. 1993. Sub-cortical Coleoptera in dead pines
and spruces: Is primeval species composition maintained in managed forests?
Biodiversity and Conservation 2121: 95-1 13.

7 Rundel, P.W. and Boonpragob, K. 1995. Dry forest ecosystems of Thailand. In: Bullock,
S.H., Mooney, H.A. and Medina, E. (eds). Seasonally dry tropical forests, pp. 93-123.
Cambridge University Press, Cambridge.

8 Rohrig, E. 1991. Floral composition and its evolutionary development. In: Rohrig, E. and
Ulrich, B. [edsl. Ecosystems of the world 7: Temperate deciduous forests, pp. 17-2^.
Elsevier, Amsterdam.

9 Ovington, J.D. and Pryor, L.D. 1983. Temperate broad-leaved evergreen forests of
Australia. In: Ovington, J.D. [ed.l Ecosystems of the world 10: Temperate broad-leaved
evergreen forests, pp. 72-102. Elsevier, Amsterdam.

10 Barnes, B.V. 1991. Deciduous forests of North America. In: Rohrig, E. and Ulrich, B. (edsl.
Ecosystems of the world 7: Temperate deciduous forests, pp. 219-344. Elsevier, Amsterdam.

11 Williams, P.H., Gaston, K.J. and Humphries, C.J. 1997. Mapping biodiversity value
worldwide: Combining higher-taxon richness for different groups. Proceedings of the
Royal Society, Biological Sciences 26A: 141-U8.

12 Thomas, W.W. 1999. Conservation and monographic research on the flora of tropical
America. Biodiversity and Conservation 8: 1007-1015.

112 WORLD ATLAS OF BIODIVERSITY

13 Olson, D.M. e( at. 2001 , Terrestrial ecoregions of the world: A new map of life on Eartfi.

BioScienceb]-. 933-938. Also see website tittpV/www.panda.org/resources/programmes/

global200/pages/li5t.titm (accessed March 20021.
U FAO 2001. Sfate of the world's forests 2001. Food and Agriculture Organization of the

United Nations, Rome.

15 UNESCO 1973. International classification and mapping of vegetation. United Nations
Educational, Scientific and Cultural Organization, Pans.

16 Hilton-Taylor, C. (compiler) 2000. 2000 lUCN Red List of tfireatened species. lUCN-the
World Conservation Union, Gland and Cambridge. Available online at
http://www.redlist.org/ (accessed March 20021.

17 Veblen, T.T, Schlegel, FM. and Oltremari, J.V. 1983. Temperate broad-leaved evergreen
forests of South America. In: Ovington, J.D. [ed.l Ecosystems of thie world 10: Temperate
broad-leaved evergreen forests, pp. 5-32. Elsevier, Amsterdam.

18 Ching, K.K. 1991. Temperate deciduous forests in East Asia. In: Rdhrig, E. and Ulrich, B.
(edsl. Ecosystems of ttie world 7: Temperate deciduous forests, pp. 539-556. Elsevier,
Amsterdam.

19 Lear, M. and Hunt, D. 1996. Updating the threatened temperate tree list. In: Hunt,

D. (ed.l Temperate trees under ttireat, pp. 161-171. International Dendrology Society, UK.

20 Ohba, H. 1996. A brief overview of the woody vegetation of Japan and its conservation
status. In: Hunt, D. (ed.l Temperate trees under tfireat, pp. 81-88. International Dendrology
Society, UK.

21 Schaefer, M. 1991. The animal community: Diversity and resources. In: Rohrig, E. and
Ulrich, B. (edsl. Ecosystems of the world 7: Temperate deciduous forests, pp. 51-120.
Elsevier, Amsterdam.

22 Ovington, J.D. and Pryor, L.D. 1981. Temperate broad-leaved evergreen forests of
Australia. In: Ovington, J.D. [ed.l fcosysfems of the world 10: Temperate broad-leaved
evergreen forests, pp. 73-99. Elsevier, Amsterdam.

23 Rohrig, E. 1991. Biomass and productivity. In: Rohrig, E. and Ulrich, B. ledsl.
Ecosystems of the world 7: Temperate deciduous forests, pp. 165-17A. Elsevier,
Amsterdam.

2A Zinke, P.J. et at. 198^. Worldwide organic soil carbon and nitrogen data. Environmental
Sciences Division, Publication No. 2212. Oak Ridge National Laboratory, US Department
of Energy.

25 Gentry, A.H. 1988. Changes in plant community diversity and floristic composition on
environmental and geographical gradients. Annals of the Missouri Botanical Garden
75: 1-34.

26 Whitmore, TC. 1984. Tropical rain forests of the Far East. Clarendon Press. Oxford.

27 Grubb, P.J. 1977. Control of forest growth on wet tropical mountains. Annual Review of
Ecology and Systematics 8: 83-107.

28 Huston, M.A. 1994. Biological diversity: The coexistence of species on changing
landscapes. Cambridge University Press, Cambridge.

29 Nelson, B. et al. 1994. Forest disturbance by large blowdowns in the Brazilian Amazon.
Eco/ogy 75: 853-858.

30 Tanner, E.V.J. , Kapos, V. and Healey, J.R. 1991. Hurricane effects on forest ecosystems in
the Caribbean. Biotropica 23: 513-521.

31 Pires, J.M. 1957. Nocoes sobre ecologia e fitogeografia da Amazonia. Norte
AgronomicoS: 37-53.

32 de Oliveira, A. A. and Mori, S.A. 1999. A central Amazonian terra firme forest. I. High tree
species richness on poor soils. Biodiversity and Conservation 8: 1219-1244.

33 Gentry, A.H. 1 988. Tree species richness of upper Amazonian forests. Proceedings of the
National Academy of Sciences 85: 156-159.

Terrestrial biodiversity 113

34 Benzing, D.H. 1989. Vascular epiphytism in America. In: Lieth. H. and Werger. M.J. A. leds).
Ecosystems of the world HB: Tropical rain forest ecosystems: Biogeograpfiical and
ecological studies, pp. 133-154. Elsevier, Amsterdam.

35 Prance, G.T. 1989. American tropical forests. In: Lieth, H. and Werger, M.J.A. leds).
Ecosystems of tfie world HB: Tropical rain forest ecosystems: Biogeograpl^ical and
ecological studies, pp. 99-132. Elsevier, Amsterdam.

36 Jenl<ins, M. 1992. Biological diversity. In: Sayer, J.A.. Harcourt, C.S. and Collins, N.M.H. Ttie
conservation atlas of tropical forests: Africa, pp. 26-32. Macmillan Publishers, London.

37 Harcourt, C.S. and Sayer, J. A. ledsl 1996. Tfie conservation atlas of tropical forests: The
Americas. Simon and Schuster, New York.

38 Lynch, J.D. 1979. The amphibians of the lowland tropical forests. In: Duellman, WE. led. I
The South American herpetofauna: Its origin, evolution, and dispersal, pp. 189-215.
Monograph No. 7 of the Museum of Natural History Kansas, Lawrence.

39 Goulding, M., Leal Carvalho, M. and Ferreira, E.G. 1988. Rio Negro: Rich life in poor
waters: Amazonian diversity and foodchain ecology as seen through fish communities.
SPB Academic Publishing, The Hague.

40 World Conservation Monitoring Centre. 1992. Groombridge, B. led.) Global biodiversity:
Status of the Earth's living resources. Chapman and Hall, London.

41 HoUdobler, B. and Wilson, E.G. 1990. The ants. Harvard University Press, Cambridge MA.

42 Hamilton, L.S., Juvil<, J.G. and Scatena, F.N. 1993. Tropical montane cloud forests. East-
West Center, Honolulu.

43 Salati, E. and Vose, P.B. 1984. Amazon basin: A system in equilibrium. Science 225:
129-138.

44 Mooney, H.A., Bullock, S.H. and Medina, E. 1995. Introduction. In: Bullock, S.H., Mooney,
H.A. and Medina, E. leds). Seasonally dry tropical forests, pp. 1 -8. Cambridge University
Press, Cambridge.

45 Sarmiento, G. 1992. A conceptual model relating environmental factors and vegetation
formations in the lowlands of tropical South America. In: Furley, P.A., Proctor, J. and
Ratter, J. A. ledsl. Nature and dynamics of forest-savanna boundaries, pp. 583-601.
Chapman and Hall, London.

46 Menaut, J.C, Lepage, M. and Abbadie, L. 1995. Savannas, woodlands and dry forests in
Africa. In: Bullock, S.H., Mooney, H.A. and Medina, E. leds). Seasonally dry tropical forests,
pp. 64-92. Cambridge University Press, Cambridge.

47 Gentry, A.M. 1995. Diversity and floristic composition of neotropical dry forests. In: Bullock,
S.H., Mooney, H.A. and Medina, E. ledsl. Seasona//y dry tropical forests, pp. 146-194.
Cambridge University Press, Cambridge.

48 Murphy, P.G. and Lugo, A.E. 1995. Dry forests of Central America and the Caribbean. In:
Bullock, S.H., Mooney, H.A. and Medina, E. leds). Seasonally dry tropical forests, pp. 9-28.
Cambridge University Press, Cambridge.

49 Ceballos, G. 1995. Vertebrate diversity, ecology and conservation in neotropical dry forests.
In: Bullock, S.H., Mooney, H.A. and Medina, E. leds). Seasonally dry tropical forests, pp.
195-220. Cambridge University Press, Cambridge.

50 Dodson, C.H. and Gentry, A.H. 1991. Biological extinction in western Ecuador Annals of the
Missouri Botanical Garden 78: 273-295.

51 Janzen, D.H. 1988. Tropical dry forests, the most endangered tropical ecosystem. In:
Wilson, E.G. 1988 led.) Biodiversity, pp. 13-137. National Academy Press, Washington DC.

52 Murphy, P.G. and Lugo, A.E. 1986. Ecology of tropical dry forest. 4nnL/a/ ffewew of fco/ogy
and Systematics 17: 67-88.

53 Bye, R. 1995. Ethnobotany of the Mexican tropical dry forests. In: Bullock, S.H., Mooney,
H.A. and Medina, E. ledsl. Seasonally dry tropical forests, pp. 423-438. Cambridge
University Press, Cambridge.

1U WORLD ATLAS OF BIODIVERSITY

5A Maass, J.M. 1995. Conversion of tropical dry forest to pasture and agriculture. In: Bullock,
S.H.. tv/looney, H.A. and Medina, E. ledsl. Seasonally dry tropical forests, pp. 399-422.
Cambridge University Press, Cambridge.

55 Bullock, S.H. 1995. Plant reproduction in neotropical dry forests. In: Bullock, S.H., Mooney,
H.A. and Medina, E. (edsl. Seasonally dry tropical forests, pp. 277-30A. Cambridge
University Press, Cambridge.

56 Tukhanen, S. 1999. Tfie northern timberline in relation to climate. In: Kankaapaa, Tasanen,
S.T. and Sutinen, M.-L. ledsl. Sustainable development in northern timberline forests, pp.
29-62. Finnisfi Forest Research Institute Research Papers 734.

57 Solbrig, O.T., Medina, E. and Silva, J.F. 1996. Determinants of tropical savannas. In:
Solbrig, O.T., Medina, E. and Silva, J.F. ledsl. Biodiversity and savanna ecosystem
processes: A global perspective, pp. 31-41. Springer, Berlin.

58 Newsome, A.E. 1983. The grazing Australian marsupials. In: Bourliere, F. led.] Ecosystems
of the world 13: Tropical savannas, pp 441-459. Elsevier, Amsterdam.

59 Solbrig, O.T. 1996. The diversity of the savanna ecosystem. In: Solbrig, O.T, Medina, E. and
Silva, J.F. ledsl. Biodiversity and savanna ecosystem processes: A global perspective, pp.
1-27. Springer, Berlin.

60 Lavelle, P. 1983. The soil fauna of tropical savannas. I. The community structure. In:
Bourliere, F. led. I Ecosystems of the world 13: Tropical savannas, pp. 477-484. Elsevier,
Amsterdam.

61 FAO 2001. Global forest resources assessment 2000. Food and Agriculture Organization of
the United Nations, Rome.

62 Williams, M. 1989. Deforestation: Past and present. Progress in Human Geography
13: 176-208.

63 Williams, M. 1991. Forests. In: Turner II, B.L. et al. ledsl. The Earth as transformed by
human action, pp. 179-201. Cambridge University Press vs/ith Clark University,
Cambridge.

64 FAO 1999. State of the world's forests 1999. Food and Agriculture Organization of the
United Nations, Rome.

65 Holdsworth, A.R. and Uhl, C. 1997. Fire in Amazonian selectively logged ram forest and the
potential for tire reduction. Ecological Applications?: 713-725.

66 Laurance, W.F and Bierregaard, R.O. ledsl 1997. Tropical forest remnants: Ecology,
management and conservation of fragmented communities. University of Chicago Press,
Chicago.

67 Smith, T.M., Leemans, R. and Shugart, H.H. 1992. Sensitivity of terrestrial carbon storage
to COj-induced climate change: Comparison of four scenarios based on general
circulation models. Climatic Change 21 : 367-384.

68 Pounds, J. A., Fogden, M.P.L. and Campbell, J.H. 1999. Biological response to climate
change on a tropical mountain. Nature 398: 611-615.

69 Bliss, L.C. 1997. Arctic ecosystems of North America. In: Wielgolaski, FE. led.l
Ecosystems of the world 3: Polar and alpine tundra, pp. 551-683. Elsevier, Amsterdam.

70 Wielgolaski, FE. 1997. Introduction. In: Wielgolaski, FE. [ed.l Ecosystems of the world 3:
Polar and alpine tundra, pp. 1-6. Elsevier, Amsterdam.

71 Chernov. Yu I. and Matveyeva, N.V. 1997. Arctic ecosystems in Russia. In: Wielgolaski, FE.
led.l Ecosysfems of the world 3: Polar and alpine tundra, pp. 361-507. Elsevier,
Amsterdam.

72 Diaz, A., Pefaur, J.E. and Durant, P 1997. Ecology of South American paramos with
emphasis on the fauna of the Venezuelan paramos. In: Wielgolaski, FE. [ed.l fcosysfems
of the world 3: Polar and alpine tundra, pp. 263-310. Elsevier, Amsterdam.

73 Campbell, J.S. 1997. North American alpine ecosystems. In: Wielgolaski, FE. led.l
Ecosysfems of the world 3: Polar and alpine tundra, pp. 211-261. Elsevier, Amsterdam.

Terrestrial biodiversity 115

•^iA

lU Grabherr, G. 1997. The high-mountain ecosystems of the Alps. In: Wielgolaski, F.E. led.)
Ecosystems of the world 3: Polar and alpine tundra, pp. 97-121. Elsevier, Amsterdam.

75 Hedberg, 0. 1997. High-mountain areas of tropical Africa. In: Wielgolaski, F.E. led.)
Ecosystems of the world 3: Polar and alpine tundra, pp. 185-197. Elsevier, Amsterdam.

76 Zockler, C. 1998. Patterns of biodiversity in arctic birds. WCMC Biodiversity Bulletin 3: 1-15.

77 Bazilevich, N.I. and Tishkov, A. A. 1997. Live and dead reserves and primary production in
polar desert, tundra and forest tundra of the former Soviet Union. In: Wielgolaski, F.E. led.l
fcosysfems of the world 3: Polar and alpine tundra, pp. 509-539. Elsevier, Amsterdam.

78 United Nations Environment Programme 1995. Heywood, V. (ed.l Global biodiversity
assessment. Cambridge University Press, Cambridge.

79 Risser, P.G. 1988. Diversity in and among grasslands. In: Wilson, E.G. and Peter, FM, ledsl.
Biodiversity, pp. 176-180. National Academy Press, Washington DC.

80 Coupland, R.T. 1992. Overview of the grasslands of North America. In: Coupland, R.T led.l
Ecosystems of the world 8A: Natural grasslands: Introduction and western hemisphere,
pp. 147-150. Elsevier, Amsterdam.

81 Coupland. R.T 1993. Review. In: Coupland, R.T (ed.l Ecosystems of the world 8B: Natural
grasslands: Eastern hemisphere and resume, pp. A7] -482. Elsevier, Amsterdam.

82 Lavrenko, E.M. and Karamysheva, Z.V. 1993. Steppes of the former Soviet Union and
Mongolia. In: Coupland, R.T. led.) Ecosystems of the world 8B: Natural grasslands:
Eastern hemisphere and resume, pp. 3-60. Elsevier, Amsterdam.

83 Wege, D.C. and Long, A.J. 1995. Key areas for threatened birds in the neotropics. Birdlife
International, Cambridge,

84 Sauer, J.R. e( al. 1997. The North American breeding bird survey results and analysis
Version 96.3. Patuxent Wildlife Research Center, Laurel. Available online at:
VTOnw.mbrpwrc.usgs.gov/bbs/bbstext.html laccessed April 20021.

85 Adams, J. 1997. Estimates of preanthropogenic carbon storage in global ecosystem
types. Available online at http://www.esd.ornl.gov/projects/qen/carbon3.html [accessed
April 20021.

86 Cowling, R.M. and Holmes, P.M. 1992. Flora and vegetation. In: Cowling, R.M. led! The
ecology of fynbos: Nutrients, fire and diversity, pp. 23-61. Oxford University Press, Cape
Town.

87 Myers, N. 1990. The biodiversity challenge: Expanded hot-spots analysis. Environmentalist
10:243-256.

88 Davis, G.W., Richardson, DM., Keeley, J.E. and Hobbs, R.J. 1996. Mediterranean-type
ecosystems: The influence of biodiversity on their functioning. In: Mooney, H.A. ef al. ledsl.
Functional roles of biodiversity: A global perspective, pp. 151-183. John Wiley and Sons,
Chichester

89 Keeley, J.F 1992. A Californian's view of fynbos. In: Cowling, R.M. led.l The ecology of
fynbos: Nutrients, fire and diversity, pp. 372-388. Oxford University Press, Cape Town.

90 Richardson, D.M., Macdonald, I.A.W., Holmes, PM. and Cowling, R.M. 1992. Plant and
animal invasions. In: Cowling, R.M. led.) The ecology of fynbos: Nutrients, fire and
diversity, pp. 271-308. Oxford University Press, Cape Town.

91 Middleton, N. and Thomas, D. (edsj 1997. World atlas of desertification. Second edition.
United Nations Environment Programme. Arnold, London.

92 West, N.E. 1983. Approach. In: West, N.E. led.) Ecosystems of the world 5: Temperate
deserts and semi-deserts, pp. 1-2. Elsevier, Amsterdam.

93 Polls, G.A. 1991. Desert communities: An overview of patterns and processes. In: Polls,
G.A. (ed.l The ecology of desert communities, pp. 1-26. University of Arizona Press,
Tucson.

94 Phillips, O.L. etai 1994. Dynamics and species richness of tropical rain forests.
Proceedings of the National Academy of Sciences 91 : 2805-2809.

114 WORLD ATLAS OF BIODIVERSITY

100

FGDC 1995. FGDC Vegetation Classification Standards. Federal Geographic Data
Committee, Reston. Unpublished.

UNEP-WCMC 1999, Contribution of protected areas to global forest conservation. In:
International forest conservation: Protected areas and beyond. A discussion paper for
the intergovernmental forum on forests. Commonwealth of Australia, Canberra.
Loh. J. led.) 2000. Living planet report 2000. WWF - World Wide Fund for Nature, Gland.
Hansen, M. et al. 2000. Global land cover classification at 1 km resolution using a
decision tree classifier. InternationalJournat of Remote Sensing 21: 1331-1365.
Barthlott, W. et al. 1999. Terminological and methodological aspects of the mapping and
analysis of global biodiversity. Acta Botanica Fennica 162: 103-1 10.
Heyvifood, V.H. led. I 1993. Flowering plants of ttie world. B.T Batsford Ltd, London.

Marine biodiversity 117

•^H

O Marin

e biodiversity

MOST OF THE PLANET IS COVERED BY OCEAN waters whose average depth is four
times the average elevation of the land, making the open sea by far the largest
ecosystem on Earth. Despite this volume, marine net primary production remains
similar to or less than that on land because photosynthesis in the sea Is carried out by micro-
scopic bacteria and algae restricted to the sunlit surface layers (plants are virtually absent).
The diversity of major lineages (phyla and classes! is much greater in the sea than on land
or in freshwaters, and many phyla of invertebrate animals occur only in marine waters.
Species diversity appears to be far lower, perhaps because marine waters are physically
much less variable in space and time than the terrestrial environment.

Marine fisheries are the largest source of wild protein, derived from fishes, moltusl<s and
crustaceans. The world catch from capture fisheries has grown fivefold over the past five
decades, but appears to have declined during the 1990s despite increased fishing effort.
More than half of the world's major fishery resources are now in need of remedial
management, mainly because of excess exploitation.

THE SEAS

Oceans cover 71 percent of the world's
surface. They are on average around 3.8 kilo-
meters Ikml deep and have an overall volume
of some 1 370 million kml The whole of the
world ocean (all contiguous seasi is theor-
etically capable of supporting life, so that the
marine part of the biosphere is far larger
than the terrestrial part. However, as on land,
life in the oceans is very unevenly distributed
- some parts are astonishingly productive
and diverse while others are virtually barren.

With the present configuration of land
masses, a major part 137 percent] of the world
ocean is within the tropics, and about 75
percent lies between the 45° latitudes. The
largest continental shelf areas are in high
northern latitudes (Table 6.21, but about 30
percent of the total shelf area is in the tropics.
Within the tropics, the shelf is most extensive
in the western Pacific (China Seas south to
north Australia).

Although knowledge of the functioning
of the marine biosphere has increased enor-
mously in the past few decades, overall it

remains far less well known and understood
than the terrestrial part of the globe. The
main reason for this is, quite simply, that

Table 6.1

Area and maximum depth

of the world's oceans and

seas

Source: Couper .

1 Ocean/Se^^^^^^H

■ Area (km'l

Depth (meters) J

Pacific Ocean

165 384 000

11524

Atlantic Ocean

82 217 000

9 560

Indian Ocean

73 481000

9 000

Arctic Ocean

14 056 000

5 450

Mediterranean Sea

2 505 000

4 846

South China Sea

2 318 000

5 514

Bering Sea

2 269 000

5 121

Caribbean Sea

1 943 000

7 100

Gulf of Mexico

1 544 000

4 377

Sea of Okhotsk

1 528 000

3 475

East China Sea

1 248 000

2 999

Yellow Sea

1 243 000

Hudson Bay

1 233 000

259

Sea of Japan

1 008 000

3 743

North Sea

575 000

661

Black Sea

461 000

2 245

Red Sea

438 000

2 246

Baltic Sea _. , ,.,

,, 422 000 ,__

— „ -'i^O .. _u..

118 WORLD ATLAS OF BIODIVERSITY

S-

^^^B^HH

^^_ Open ocean ^^M

^HrContinentalsh|iy|

Total area Imillion kml

Latitude bands 1% of total!
Polar and boreal 145-90°!
Temperate [20-45°!
Tropical [0-20°l

360.3

26,6
36.8
36.6

26.7

40.9
28.8
30.3

Table 6.2
Relative areas of
continental shelves and
open ocean

Source: Adapted from Longhurst and
Pauty , after t^oiseev.

much of it is inaccessible to humans. Study of
any part below the top few meters requires
specialized equipment and is expensive and
time consuming. Knowledge of most of the
sea IS thus based largely on a range of
remote-sensing and sampling techniques and
often remains sketchy. As these techniques
become more sophisticated, so our under-
standing of marine ecosystems, particularly
those away from the coastal zone, is under-
going constant revision.

Sea water and ocean currents
Sea water is a complex but relatively uniform
mixture of chemicals. Most of the 92 naturally
occurring elements can be detected dissolved
in it, but most only in trace concentrations;
the most abundant are sodium las Na+! and
chlorine las Cl-1, which occur at a concen-
tration some ten times higher than the next
most abundant element, magnesium. The
term used to quantify the total amount of
dissolved salts in sea water is salinity, a
dimensionless ratio, which generally ranges
between 33 and 37 and averages 35. Most of
the substances dissolved in sea water are
unreactive and remain at relatively stable
concentrations; however those that play a
part in biological systems can be highly
variable in time and space.

The world's sea waters are constantly in
motion, at all scales from the molecular to
the oceanic. Large-scale ocean circulation
plays a vital role in mediating global climate
as well as influencing the functioning of
marine ecosystems. It is driven by complex
interactions between a number of physical
variables, notably latitudinal variations in
solar radiation land consequent heating and
cooling!, precipitation and evaporation, trans-
fer of frictional energy across the ocean

surface from winds, and forces resulting
from the rotation of the planet.

Surface currents are largely driven by pre-
vailing winds. The most important features are
vast, anticyclonic gyres in the subtropical
regions of the world's oceans. These are pri-
marily driven by the westerly trade winds in
the Roaring Forties and circulate clockwise in
the northern hemisphere and counter-
clockwise in the southern hemisphere. Also
important are the eastward-flowing Antarctic
circum-polar current and equatorial current.
At smaller scales, eddies and rings are ubiqui-
tous and are analogous to weather systems in
the atmosphere; typical oceanic eddies may be
1 00 km across and persist tor a year or more.

The density of sea water increases with
increasing salinity and with decreasing temp-
erature until it freezes at around -1.9°C. Sea
water of different density does not mix readily,
and so the oceans tend to be well stratified
vertically, with bodies of less dense sea water
[warmer and less saline! sitting on top of
cooler, more saline and denser bodies. The
stratification is rarely stable over the long
term, however, as the influence of climate
changes the properties of surface waters and
causes various forms of vertical mixing. The
single most important factor controlling large-
scale deep-water circulation appears to be the
generation of cold, high-salinity water near the
surface in the Weddell Sea and off Greenland in
the North Atlantic. Here, during winter, the sea
freezes and the floating ice sheet is virtually
free of salt, leaving the underlying water more
saline, cold and dense. This sinks to the bottom
and moves south along the Atlantic floor, some
passing south of the equator and circulating
throughout the world ocean, producing what is
known as the Great Conveyor - the interlocking
system of major circulation currents in the
deep sea. Because it originates near the
surface it is well oxygenated as well as cold,
and is the major reason why aerobic organisms
can thrive in the deep sea.

Major zones of upwelling occur along the
western boundaries of continents where trade
winds blow towards the equator, causing
surface waters to be pushed offshore to be
replaced by cooler deep waters. There are five
major upwelling areas: the Humboldt current

Marine biodiversity 119

region off the Cfiilean and soutfiern Peruvian
coast of South America; the California current
region off western North Annerica; the Canary
current off the coast of Mauritania in north-
w/est Africa; the Benguela current region off
southern Africa; and the northwest Arabian
Sea. Marine production is much enhanced in
these regions which typically support major
pelagic fisheries. Outside these upwelling
zones, stratification tends to persist through-
out the year in the tropics and subtropics.

MAJOR MARINE ZONES
The continental shelf

Marine waters around major land masses
are typically shallow, lying over a continental
shelf which may be anything from a few
kilometers to several hundred kilometers
wide. The most landward part Is the littoral or
intertidal zone where the bottom is subject to
periodic exposure to the air Water depth here
varies from zero to several meters. Seaward
of this the shelf slopes gently from shore to
depths of one to several hundred meters,
forming the sublittoral or shelf zone. Waters
below low-tide mark in the continental shelf
region are referred to as neritic.

The extent, gradient and superficial geo-
logy of continental shelf areas are determined
by many factors. Including levels of tectonic
activity in the Earth's crust. More than 80
percent of the global volume of river-borne
sediment Is deposited In the tropics (and an
estimated 40 percent of it by just two river
systems: the Huang He or Yellow River and
the Ganges-Brahmaputra)', and this Is ref-
lected in the extent of shelf areas In parts
of the tropics, and in the high turbidity of
coastal waters In monsoon regions. Most
shelf areas in the tropics are overlain by
sands or muds composed of sediment of
terrestrial origin (terrigenous deposits).

Shelf regions support the marine com-
munities most familiar to humans, and many
of the marine resources of particular value to
them. Although mangrove and coral reefs are
two of the best known tropical coastal
ecosystems, they dominate only a minor part
of the world coastline: the former mainly In
deltaic or other low-lying coastal plains, and
the latter only in shallow waters where

terrestrial sedimentation is very low. Soft-
bottom habitats with sparse vegetation are
probably the most widespread coastal marine
ecosystem type, and virtually the entire
seabed away from the coastline Is covered in
marine sediments (see Box 6.1).

The deep sea

At the outer edge of the shelf there is an
abrupt steepening of the sea bottom, forming
the continental slope which descends to
depths of 3-5 km. The sea bottom along the
slope IS referred to as the bathyal zone. At
the base of the continental slope are huge
abyssal plains which form the floor of much of
the world ocean. Open waters above these
plains make up the oceanic pelagic zone.
Given that the oceans cover some 71 percent
of the globe, and that the shelf area is
relatively narrow, the oceanic pelagic zone is
by far the most extensive ecosystem on Earth.
The plains are punctuated by numerous sub-
marine ridges and sea mounts which may
break the surface to form Islands. There are
also a number of narrow trenches which have
depths of from 7 000 to 11 000 meters (m).
These constitute the hadal zone. Ocean

^Box 6.1 Life in sediments

Spaces between tfie particles of marine sediments contain water, air, detritus and
organisms, tfie last divided by size Into two main categories. The microfauna includes
archeans, bacteria and protoctists. These include a number of pnmary producers at the base
of the food web in shallow waters, and have a role in interstitial sulfur chemistry and
oxygenation of sediment. The melofauna includes sediment-living forms between 0,1 and
1 mm in size, and Is a major component of seabed ecosystems, particularly In the deep sea.
Four phyla contain only marine melofauna: the Gastrotricha, Gnathostomullda, KInorhyncha
and Loriclfera. Nematodes are typically the most numerous component, with harpacticoid
copepods and foraminiferans also important. High concentrations occur around the burrows
of deposit-feeding mollusks and polychaetes, with high bacterial numbers and elevated
nutrient flux. They play a key role In the flow of nutrients from the microfauna to larger
organisms, of which several distinctive species exist In and on the seabed. Sea cucumbers,
crinoids, polychaete worms, sea spiders, Isopods and amphipods are most abundant. Some
taxa are only found here. Pogonophorans occur mostly below 3 000 m and many of the
known species are restricted to the sediments of deep ocean trenches (more than 6 000 m).
Small-scale variation In food supply, such as the fortuitous appearance of a large animal
carcass, may enhance spatial structure and provide opportunity for other species to colonize.
A micro-landscape of hills and valleys is created by the burrowing and fecal mounds of
echiurid and polychaete worms.

jea

120 WORLD ATLAS OF BIODIVERSITY

Coral reefs develop within
the photic zone because of
the symbiotic relationship
between some coral
species and zooxanthellae.

trenches are formed as a consequence of
plate tectonic processes where sectors of ex-
panding ocean floor are compressed against
an unyielding continental mass or island arc,
resulting in the crust buckling downwards
IsubductingI and being destroyed within the
hot interior of the Earth.

CLASSIFYING THE MARINE BIOSPHERE

Much less effort has been devoted to catego-
rizing and mapping units within the marine
biosphere than to terrestrial environments.
As on land, basic marine habitat types are
commonly defined by some combination of
structural, climatic and community features
le.g. 'tropical coral reef'l but, probably in part
reflecting issues of scale and resolution, few
global maps of marine ecosystem types are
available.

marine ecosystem ILMEI units have been
defined' ', with a view to improving planning
and management. These are large regions of
marine space, some 200 000 km' or larger,
that extend from river basins and estuaries
out to the seaward boundary of the conti-
nental shelf, with distinct combinations of
bathymetry, hydrography and production.
With similar aims, the conservation organi-
zation WWF has distinguished 43 large
marine ecoregions, based broadly on oceano-
graphic and community features, and also
intended to be globally representative'.

THE BASIS OF LIFE IN THE SEAS

In the sea, as on the land, photosynthesis is
the driving force behind maintenance of life.
Because photosynthesis in nature depends on
sunlight, with few exceptions primary pro-

The principal global scheme intended to
classify the entire world ocean within an
objective oceanographic system is based on
long-term data on sea surface color
(obtained by the CZCS radiometer carried by
the Nimbus orbiting satellite during 1978-
861' \ These data reflect chlorophyll concen-
trations and provide a basis for estimating
primary production rates, and changes over
time, on a 1-degree grid. These values,
together with numerous other data sets, have
been used as the basis of a classification of
the world ocean into four ecological domains
and 56 biogeochemical provinces. Taking a
less quantitative approach, some 64 large

ductivity is confined to those parts of the
ocean that are sunlit. Water absorbs sunlight
strongly so that light intensity decreases
rapidly with increasing depth. Red wave-
lengths are most rapidly absorbed, except
where turbidity is high, while blue-green
wavelengths penetrate the deepest. Even in
the clearest waters the latter are completely
absorbed by around 1 km depth, this marking
the extreme limit of the so-called photic zone.
Photosynthesis is thus limited to the contin-
ental shelf area and the first few hundred
meters of surface waters (and often much
less] of the open ocean which together make
up a very small proportion of the total volume

Marine biodiversity 121

of the oceans. The portion of the photic zone
where sunlight is strong enough to support
appreciable amounts of photosynthesis is the
euphotic zone-
Virtually all other marine organisms.
including those of the unlit middle depths and
the deep sea. are dependent ultimately on
growth of primary producers in areas that
may be widely distant from them in time and
space. The most important exceptions to this
are the archeans and bacteria living around
hydrothermal vents associated with rift zones
in the ocean floor. The water here can be 10°C
warmer than in adjacent areas and these
microorganisms are able to grow using hydro-
gen sulfide gas emitted at the vents as an
energy source, and they in turn are used by
other organisms.

With some exceptions primary production
per unit area tends to be lower in marine
environments than in terrestrial ones, espec-
ially if highly managed terrestrial agricultural
systems are considered. This is because over
the vast majority of the ocean the euphotic
zone is far distant from the lithosphere. The
latter provides essential nutrients for life, and
these therefore have to be transported to the
euphotic zone to allow life processes to
continue. There is also usually a steady loss of
nutrients and organic compounds from the
euphotic zone, owing to the sinking of particles
and bodies into the dark regions of the sea
where no photosynthesis can take place.

Continued productivity in the open sea is
contingent on the replacement of the lost
nutrients. The latter may originate either on
land, in the form of river outflow, or from mar-
ine sediments. Their replacement in the pelagic
euphotic zone is dependent on the mixing or
vertical movement of the water column. At
latitudes higher than A0°, winter mixing allows
replenishment of the euphotic zone, partic-
ularly in continental shelf areas. However, in
permanently stratified subtropical and tropical
oceanic waters there is little vertical mixing,
and therefore little influx of new nutrients.
Productivity in these areas is correspondingly
low. In zones of upwelling. surface waters are
regularly replaced by nutrient-rich bottom
waters and very high levels of productivity can
be achieved, at least seasonally

Until the 1980s it had been believed that
photosynthesis in the pelagic ocean was car-
ried out only by single-celled phytoplankton.
between 1 and 100 microns in diameter
II micron = 0.001 mm|, and also that vast
expanses of open ocean where phytoplankton
could not be detected were, in terms of
productivity, the marine equivalent of deserts.
New observational techniques have since
revealed the presence in great abundance of
exceptionally small and previously unknown
organisms, collectively termed picoplankton.
These appear to be predominantly photo-
synthesizing unicellular cyanobacteria 0.6-1
micron in size, such as Prochlorococcus.
Because of their extraordinary abundance
(some 100 million cells may be present in
1 liter), and despite their minute size, these
organisms play a crucial role in the produc-
tivity of open ocean waters' and have led to
marked upward revisions in estimates of
overall marine productivity

The major role played by microscopic
organisms in marine productivity has a
number of important implications for marine
ecology. Although much remains unknown
about the structure and dynamics of pelagic
food webs, it seems that a high proportion
of marine primary production is used directly
by microscopic organisms Iboth autotrophic
and heterotrophic! and cycled back into non-
living forms (dissolved carbon dioxide and
organic carbon! rather than supporting
populations of larger organisms.

Oceanic primary producers divert a high
proportion of their energy into reproduction
rather than accumulating biomass. in con-
trast to terrestrial primary producers (plants!.
As a result of this, average standing biomass
per unit area in the oceans has been esti-
mated at around one-thousandth that on land.
Population turnover of oceanic primary pro-
ducers is also several orders of magnitude
higher than turnover of major terrestrial pri-
mary producers [plants!.

The small size and rapid turnover of
oceanic primary producers m.ean that there
are no organisms directly analogous to the
woody plants that so enrich terrestrial
environments by providing structurally com-
plex habitats for other organisms. The nearest

122 WORLD ATLAS OF BIODIVERSITY

Table 6.3

Marine diversity by

phylum

Notesr Strictly marine groups
shown in bold. Estimates are
from a variety of sources and
in some cases, e.g. Mollusca,
differ in detail from ttiose in
Table 2.1.

equivalents are the large brown algae known
as kelp Iphylum Phaeophyta], whose structure
is less complex and which are much more
narrowly distributed. Structurally complex
habitats may in contrast be created by
animals, particularly corals Iphylum Cnidarial
and, to a lesser extent, sponges (phylum
Ponferal, mollusks and serpulid worms
Iphylum Annelida].

BIOLOGICAL DIVERSITY IN THE SEAS

It is well known that diversity at higher
taxonomic levels (phyla and classes! is much
greater in the sea than on land or in
freshwater Of the 82 or so eukaryote phyla
currently recognized (see Chapter 2), around

60 have marine representatives compared
with around 40 found in freshwater and 40 on
land. Amongst animals the preponderance is
even higher, with 36 out of 37 phyla having
marine representatives (Table 6.31.

Some 23 eukaryote phyla, of which 18
are animal phyla, are confined to marine
environments. Most of these are relatively
obscure and comprise few species. The major
exceptions are the Echinodermata, of which
some 7 000 species are known, and the
Foraminifera, with around 4 000 known, ex-
tant species. A number of other important
phyla including the coelenterates (Cnidarial,
sponges (Poriferal and brown and red algae
(Phaeophyta and Rhodophyta, respectively!

amain Kingdom

Arcfiaea
Bacteria

2 ptiyla
12 pfiyla

■H

inidieii nn.

of described
> marine species

Eukaryota Protoctista 27 phyla, including:

Chloroptiyta Green algae

Pfiaeoptiyta Brown algae

Rfiodophyta Red algae

others incl.
Foraminifera

Animalia

7 000

1500

4 000

23 000

Placozoa

Porifera

Sponges

10 000

Cnidaria

Coelenterates

10 000

Ctenophora

Comb jellies

Platyhelminthes

Flatworms

15 000

Nemertina

Nemertines

750

Gnathostomulida

Gnathostomulids

Rhombozoa

Rhombozoans

Orthonectida

Orthonectids

Gastrotncha

Gastrotnchs

400

Rotifera

Rotifers

Kinorhyncha

Kinorhynchs

100

Lorlcifera

Lorlclferans

Acanthocephala

Spiny-headed worms 600

Entoprocta

Entoprocts

170

Nematoda

Nematodes

12 000

Nematomorpha

Horsehair worms

<240

pnyia

tsiimaieo no.

of described

marine specie

Ectoprocta

Ectoprocts

5 000

Phoronida

Phoronid worms

Brachiopoda

Lamp shells

350

Mollusca

Mollusks

?75 000

Priapulida

Priapullds

Sipuncula

SIpunculans

150

Echlura

Echiurids

Annelida

Annelid worms

12 000

Tardigrada

Water bears

few

Chelicerata

Chelicerates

1000

Mandibulata

Mandibulale

arthropods

few

Crustacea

Crustaceans

38 000

Pogonophora

Beard worms

120

Bryozoa

Bryozoans

4 000

Echinodermata

Echinoderms

7 000

Chaetognatha

Arrow worms

Hemichordata

Hemichordates

100

Urochordata

Tunicatesand

ascidians

2 000

Cephalochordata

Lancelots

Craniata

Craniates

15 000

Fungi 3 phyla

Plantae Anthophyta

500

Total

ca 250 000

Marine biodiversity 123

•^i<

''til,

are very largely marine, each with only a
small number of non-marine (usually fresh-
water] species.

The reason for this predominance of
marine higher taxa (particularly amongst
animals) is believed to be because most of
the fundamental patterns of organization
and body plan. i.e. the different basic kinds
of organism that are distinguished as phyla,
originated in the sea and remain there, but
only a subset of them has spread to the land
and into freshwaters. It is noteworthy that
only a third or so of marine phyla are found
in the pelagic realm, the remainder being
confined to sea bottom (benthicl areas - the
habitat where eukaryotic organisms are
believed to have evolved.

In contrast, known species diversity in
the sea is much lower than on land - some
250 000 species of marine organisms are
currently known, compared with more than
1.5 million terrestrial ones. Much of this
difference is because of the large numbei of
described terrestrial arthropods, for which
there is no marine equivalent. Amongst
fishes, almost as many freshwater species as
marine are known, despite the fact that
freshwater habitats account for only around
one ten-thousandth of the volume of marine
ones. Similarly, the most diverse known
marine habitats (coral reefsl are far less
diverse in terms of species number than the
moist tropical forests that are often taken as
their terrestrial counterparts.

The apparent lower total species diversity
of the marine biosphere is likely in part at least
to be a result of the physical characteristics of
water, particularly its high heat capacity and
its ability to mix. Because of these, marine
environments (particularly deep-water ones)
tend to show much less variation in time and
space in their physical characteristics than
terrestrial ones. This lack of physical variation
seems to result in a similar lack of ecological
variation over wide areas.

In contrast to terrestrial faunas, where
one phylum - the Mandibulata (insects and
relatives) - vastly outnumbers all others in
terms of known species, marine species are
much more evenly distributed across higher
taxa. The largest marine phyla - Mollusca and

1 Class J^l

r No, of marine species

Myxini

Hagfishes

Cephalaspidomorphi

Lampreys

Elasmobranctiii

Sfiarks, skates and rays

900

Holocephali

Chimaeras

Actinopterygii

Ray-fin fishes

13 8C0

Sarcopterygii

Lobe-fin fisties

Reptilia

Reptiles

Aves

Birds

307

Mammalia

Mammals

110

Crustacea - each comprise far fewer than
100 000 known marine species, in contrast
with the Mandibulata, of which around
1 million terrestrial species have been
identified to date. The only major eukaryote
phyla (i.e. those with 10 000 or more de-
scribed species! that are believed to have
comparable levels of diversity both on land
and in the sea are the Platyhelminthes,
Nematoda, Mollusca and Craniata. As on
land, vertebrates are by far the best-known
group of marine organisms. Of the 50 000 or
so described extant species, more than
15 000 may be considered marine (Table 6.41,
the overwhelming majority of which are fishes
(Table 6.51 and a few tetrapods (Table 6.6).

Marked latitudinal gradients in species
richness have been described in a number of
groups, for example in benthic isopods and
mollusks", and it is generally true that coastal
waters within the tropics tend, in parallel with
terrestrial environments, to be richer in
number of species than those at higher
latitudes (although many exceptions to this are
known, such as penguins, pinnipeds and auksl.
Although marine biogeography had advanced
relatively little since the early 20th century,
improved distribution data and more robust
methods have recently been developed. These
have been applied to definition of coastal
biogeographic regions', and important areas
for marine biodiversity analogous to those
delimited on land. For example, 18 centers of
endemism within the world tropical reef zone
have been identified using data on reef fish,
coral, snails and lobsters (Map 6.1); ten
centers were regarded as hotspots' because of
their higher threat score'".

Table 6A

Diversity of craniates in

the sea by class

Note: Because of changing
taxonomy, incomplete
information and occupation of
multiple habitat types, these
estimates are indicative only.

124 WORLD ATLAS OF BIODIVERSITY

Common fishes No. of

No. of

JH^

—Common fislies No. of

No. of

No. off

families

genera

species

B families

genera

species

occurring

in marine

■I^H

^^^^^HH

tiabitats

J^B

habita^

Myxiniformes

Hagfishes

Atheriniformes

Silversides

139

Petromyzontiformes Lampreys

Beloniformes

Needlefishes, sauries,

Chimaeriformes

Chimaeras

flyingfishes, halfbeaks

140*

Heterodontiformes

Bullhead sharks and
horn sharks

Cyprinodontiformes

Rivulines, killifishes,
pupfishes, four-eyed fishes.

Orectolobiformes

Carpet sharks

poeciliids, goodeids

Carcharhiniformes

Ground sharks

207

Stephanoberyclformes Gibberfishes, pricklefishes.

Lamniformes

Mackerel sharks

whalefishes, hairyfish.

Hexanchiformes

Cow sharks

tapetails

Squaliformes

Dogfishes and

Beryciformes

Fangtooths, spinyfins.

sleeper sharks

lanterneyefishes, roughies.

Squatiniformes

Angel sharks

pinecone fishes, squirrelfishes 7

123

Pristiophoriformes

Saw sharks

Zeiformes

Dories, boarfishes.

Rajiformes

Rays

432*

oreos, parazen

Coelacanthiformes

Coelacanths

Gasterosteiformes

Pipefishes, seahorses.

Acipenseriformes

Sturgeons

sticklebacks, sandeels.

Salmoniformes

Salmonlds

seamoths, snipefishes.

Stomiiformes

LIghtfishes, hatchetfishes,

shnmpfishes. trumpetfishes

238'

barbeled dragonfishes

321

Synbranchiformes

Swamp-eels

Ateleopodiformes

Jellynose fishes

Scorpaeniformes

Gurnards, scorpionfishes.

Aulopiformes

Greeneyes, pearleyes,
waryfishes,
lizard fishes,

velvetfishes, flatheads.
sablefishes, greenlings,
sculpins, oilfishes, poachers.

barracudinas, lancetfishes

219

snailfishes, lumpflshes

266

1219*

Myctophiformes

Lanternfishes

241

Perciformes

Perches, basses, sunfishes,

Lampridiformes

Garfishes, ribbonfishes.

whitings, remoras, jacks.

crestfishes, opahs

dolphinfishes, snappers.

Polymixiiformes

Beardfishes

grunts, damselfishes.

Ophidiiformes

Pearlfishes, cusk-eels.

dragonfishes, wrasses,

brotulas

350*

butterflyfishes, etc.

148

1496

7 371*

Gadiformes

Cods, hakes, rattails

481

Pleuronectifornnes

Plaice, flounders, soles

123

564*

Batrachoidiformes

Toadfishes

64*

Tetraodontiformes

Triggerfishes,

Lophiiformes

Anglerfishes. goosefishes,

puffers, boxfishes.

frogfishes. batfishes,

filefishes, molas

100

327*

seadevils

297

Mugilitormes

Mullets

65*

Total Iconservative working estimatel

uooo

Table 6.5

Diversity of fisties in the

seas by order

Note: Strictly marine orders in
bold; ottier orders that are
mainly marine (more than
50% of species! marked with
an asterisk'.

Source: Afler Nelson*'.

Algae

The macro-algae are superficially plant-like
protoctists that lack the vascular tissue used
by higher plants to transport w/ater and
nutrients. They are almost exclusively aquatic;
three of the four principal groups consisting
of large-sized species are mainly marine in
occurrence. These three, the green, brown
and red algae (seaweeds'!, are all cosmo-
politan in distribution and occur in a range of

environments, although some constituent
families have somewhat restricted ranges.
There are more marine species of red algae
IRhodophyta! - around 4 000 - than the
greens (Chlorophyta, ca 1 0001 and browns
(Phaeophyta, ca 1 500! combined.

As with pinnipeds and seabirds, the cold
and cool temperate regions of the world
appear to be surprisingly rich in species. On
present incomplete information, the region

Marine biodiversity 125

around Japan (norttiwest Paciticl. the North
Atlantic, and the tropical and subtropical
western Atlantic hold the most species of
marine algae. Southern Australia is not so
species rich but appears to have the highest
proportion of endemics. There are few
species of larger algae in regions of cold-
water upwelling; small isolated islands and
polar regions also have few species. In
contrast, coral reefs support a unique and
generally diverse algal flora that includes
many crustose coralline algae (more species
of which are likely to be discovered).
Mangrove areas also support a well-defined
algal vegetation. Sandy coastlines hold few
species of large algae and often form barriers
to seaweed dispersal.

Marine fishes

Fishes are considered a paraphyletic group
(see Chapter 21: that is, all living species are
thought to share a common ancestor but to
share this ancestor with another group (in this
case, the tetrapodsl not categorized as fishes.
Apart from some 50 or so species of generally
parasitic lampreys and hagfishes in the
superclass Agnatha, fishes are divided into
two unequal-sized groups, the cartilaginous
fishes (Chondrichthyesj including chimaeras
(class Holocephalil and sharl<s and rays
(class Elasmobranchiil, and the bony fishes
(OsteichthyesI, including the 'typical' ray-
finned fishes (class Actinopterygiil and the
lobe-finned coelacanths and lungfishes (class
Sarcopterygiil. Some 60 percent of all l<nown
living fish species (i.e. around 14 000 species!
occur in marine habitats. They range in size
from an 8 mm-long goby Trimmatom nanus in
the Indian Ocean to the 15-m whale shark
Rhincodon typus, respectively the smallest
and largest of all fish species, and occur in
virtually all habitats, from shallow inshore
waters to the abyssal depths.

The elasmobranchs (sharks, skates and
raysl are an overwhelmingly marine group
with around 850 living species in ten orders.
Although far less diverse than the bony fishes,
the cartilaginous fishes include many of the
largest fish species, a number of which are top
predators in marine ecosystems. There tend to
be more shark species at lower latitudes, but

at family level richness tends to be higher on
the edge of the tropics (Map 6.2). The bony
fishes are a remarkably diverse group, with an
enormous range of morphological, physio-
logical and behavioral adaptations. Of the 26
orders of bony fishes with marine represen-
tatives (Table 6.5], by far the largest and most
diversified is the Perciformes. This is the
largest of all vertebrate orders and dominates
vertebrate life in the ocean, as well as being
the dominant fish group in many tropical and
subtropical freshwaters.

As with other groups of organisms, the
majority of fishes in the sea are strictly
marine, occurring only in salt water. A

Table 6.6

Marine tetrapod diversity

Notesr Birds follow the list of
seabirds recognized in Croxall
et al." with the additional
inclusion of four eider ducks
and three steamer ducks in
the family Anatidae and the
red phalarope Phalaropus
fulicana (Scolopacidael.
Taxonomy follows Sibley and
Monroe". Figures in
parentheses indicate species
that breed largely or entirely
inland. Strictly marine
families shown in bold.

Class

Order

Family S

No. of marine
species^^

Reptilia

Chelonia

Dermochelyidae

Leathery turtle

Cheloniidae

Sea turtles

Squamata

Elapidae

Sea snakes and sea kraits 55

Acrochordidae

File snakes

Iguanidae

Iguanas

Aves

Ansenformes
Ciconiiformes

Anatidae

Scolopacidae

Laridae

Gulls, terns, skuas,
auks, skimmers

7
1

120(131

Phaethontidae

Tropicbirds

Sulldae

Gannetsand boobies

Phalacrocoracidae

Cormorants and shags

36(21

Pelecanidae

Pelicans

Fregatidae

Frigatebirds

Spfieniscidae

Penguins

Procellariidae

Petrels, albatrosses,
shearwaters

115

Mammalia

Cetacea

Balaenidae

Right whales

Balaenopterldae

Rorquals

Esclirlclitiidae

Gray whale

Neobalaenidae

Pygmy right whale

Delpliinidae

Dolphins

Monodontidae

Beluga and nanvhal

Pliocoenidae

Porpoises

Phystereidae

Sperm whales

Platanistidae

River dolphins

ZIphiidae

Beaked whales

Sirenla

Trichechidae
Dugongldae

Manatees
Dugong

1
1

Carnivora

Mustelidae
Odobenidae

Otters and weasels
Walrus

Otarildae

Eared seals

i^JbSiA

Phocidae

Earless seals _^^

126 WORLD ATLAS OF BIODIVERSITY

•<*-

Map 6.1

Coral reef hotspots

The location of 18 areas
defined by fiigfi endemisrm
in reef fisfies, corals, snails
and lobsters, including the
ten areas identified as
'hotspots' on the basis of
high threat score.

Source: Adapted from Roberts .

Endemic-rich areas

Endemic-rich area at higher risk
Endemic-rich area

proportion, however, may also occur in
inland waters, often passing a particular part
of their life cycle there. Species that spend
most of their life in marine waters but ascend
rivers to breed, such as many salmonids
Ifamlly Salmonidae, order SalmoniformesI
and sturgeons (family Acipenseridae, order
Acipenseriformesl, are referred to as anad-
romous. Those that breed at sea but spend
their lives otherwise In freshwater, such as
most eels In the family Anguillldae (order
Angulliformesl, are referred to as catadro-
mous. Species with a wide salinity tolerance
that may occur In marine, brackish and
fresh waters (e.g. some sawfishes, family
Pristidae, order Rajlformesl are referred to
as euryhaline while those with narrow
tolerances, be they to marine, brackish or
fresh water, are referred to as stenohallne.

Reptiles

Present-day diversity of reptiles in the seas
Is low. One Important reason for this appears
to be that modern reptilian kidneys cannot
tolerate high salinities and thus the only
reptiles that have adapted to marine environ-
ments are those which have developed
specialized salt-excreting glands. The most
thoroughly marine reptiles are undoubtedly the
sea snakes In the subfamily Hydrophlinae
(family Elapidael. These spend their entire
lives In the sea, giving birth to live young there.
Although largely air-breathing like other
reptiles, they can also absorb some oxygen
directly from sea water and are thus able to
remain submerged for long periods. Around
50 species are known, widely distributed
In tropical parts of the Indo-Paclfic region. In
addition the little file snake Acrochordus

Marine biodiversity 127

granulatus (family Acrochordidael, from
northern Australia and Southeast Asia is also
entirely aquatic, but occurs in brackish estu-
aries as welt as sea water.

Five species of sea krait in the subfamily
Laticaudinae are also largely marine, feeding
mainly on eels. However they return to land
to breed, generally on small tropical islands.
They, too, are confined to the Indo-Pacific
region. One species of lizard, the Galapagos
marine iguana Amblyrhynchus cristatus
(family Iguanidael, feeds underwater on mar-
ine algae but spends a considerable proportion
of time on land. Several other reptile species
regularly enter sea water, most notably a
number of homalopsine mangrove snakes
(family Colubridae) from the Indo-Pacific and
the estuarlne crocodile Crocodyius porosus
(family Crocodylidael from the same region.

Undoubtedly the most prominent group of
marine reptiles is the sea turtles, comprising
the leathery turtle Dermocheiys coriacea in
the family Dermochelyidae and six members
of the family Cheloniidae. Alt species are
large (ranging from 70-centimeter (cm| adult
carapace length in Lepidochelys kempii
to, exceptionally, 250 cm in Dermocheiys
coriacea] and most are widely distributed in
tropical and subtropical waters (Map 6.31. Sea
turtles are almost completely marine; only the
females emerge to nest on land, mostly within
the tropics. One species, the loggerhead
Caretta caretta, nests largely In temperate
areas of the northern hemisphere. Sea turtles
typically have a long period to maturity (often
up to 25 years In the case of the green turtle
Chetonia mydas] and a long lifespan. Females
often nest only every two or three years. They

128 WORLD ATLAS OF BIODIVERSITY

Map 6.2

Shark family diversity

Diversity in sinarks, based
on the distribution of all 30
families plotted as a density
surface. Most sharks are
coastal in occurrence and
for illustration purposes the
family density is shown
within a band extending out
400 km from the coastline
(slightly further than the
200 nautical-mile EEZ limit).

Source^ Prepared by UNEP-WCMC:
family distributions aggregated from the
species range maps published by
Compagno

Level of diversity

High

Low

habitually return to the same nesting beaches,
sometimes undergoing protracted migrations
from feeding grounds. They may lay two or
three clutches in a season, sometimes consis-
ting of more than 100 eggs each, depending on
the species. Nest, hatchling and juvenile
mortality are often high.

Birds

Defining marine birds, or seabirds, is
somewhat more problematic than defining
marine species in other groups. All birds
breed in terrestrial habitats, but a large
number (almost all of them non-passerinesi
obtain all or much of their food from aguatic
or littoral habitats. Some of these, including
all frigatebirds [Fregatidael, tropicbirds
(Phaethontidael, gannets and boobies (Sulidael,
penguins (Spheniscidael and petrels, alba-

trosses and shearwaters (Procellanidael are
indisputably marine, in that they obtain all
their food from marine habitats, almost
invariably breeding along coastlines and
spending most or all of their time when not
breeding out at sea. Many others, however,
have less clear-cut habits. Some, such
as a number of cormorants and shags
(Phalacrocoracidael have both resident in-
land and marine populations. Others, such as
a number of gulls and terns (Laridae) and
some ducks and geese (Anatidael. may breed
inland but spend the rest of the year living in
coastal areas or out at sea. Yet others, such
as sandpipers (Scolopacidael and other
waders, typically feed in littoral or intertidal
habitats rather than in the sea itself; many of
these species also occur inland.

Adopting a somewhat arbitrary division,

Marine biodiversity 129

and excluding all wading birds with the
exception of the red phatarope Phataropus
fulicaria la truly pelagic species outside the
breeding season), over 300 species of birds
can be considered wholly or largely marine.

In common with pinnipeds, seabirds show
a latitudinal distribution in which diversity is
much higher at higher latitudes [temperate
and polar regions! than it is in the tropics. Two
thirds of all seabirds are confined as breeding
species to these latitudes, compared with only
7 percent that are exclusively tropical. This
stands in sharp contrast to the pattern found
in most major terrestrial groups Isee Chapter
51 and many marine groups such as sea
turtles, mangroves and reef-building corals
Isee, respectively. Maps 6.3, bA, 6.61 in which
species diversity increases dramatically with
decreasing latitude. Diversity is also markedly

higher in the southern than in the northern
hemisphere, with over half of all seabird
species breeding in southern temperate and
polar latitudes. Dominance of this region is
even more marked in the Procellariidae, the
family with the greatest number of truly
marine species, in which over 60 percent of
species breed at these latitudes and half are
confined to it (Table 6.7).

Table 6.7

Regional distribution of

breeding In seabirds

Notes: Several species breed
in more than one latitudinal
band so that overall totals
exceed actual number of
seabirds. Numbers in
parentheses indicate
approximate number
confined to the region.

Breeding ^^^^1

■

Ocean area

Procellariidae

Laridae

Others

Tote^

I distribution

million km'l

^Hl

Northern temperate and pola

79.5

24117)

50 135)

50 1361

124138)

Northern tropical

75.2

1515)

2710)

19101

6115)

Southern tiopical

78.1

2518)

3014)

26151

81117)

Southern temperate

and pola

130.1

70 158)

341151

56141)

1601114)

130 WORLD ATLAS OF BIODIVERSITY

Map 6.3

Marine turtle diversity

An overview of marine
turtle diversity, represented
by the number of turtle
species nesting in any one
area. Each symbol shows
the location of a nesting
site or area, colored
according to the number of
species present, up to a
maximum of five species in
some parts of the tropics.

Source: Prepared using spatial data
from a GIS database of marine turtle
nesting beaches maintained at
UNEP-WCI^C

Number of species
nesting in area

Mammals

Wholly aquatic mammals (those that never
normally emerge on to land! are confined
to two orders, the Cetacea and the Sirenia.
The Cetacea comprises some 78 species,
all except five marine, distributed throughout
the worlds seas. They include the largest
living animals - the rorquals in the family
Balaenopteridae. All cetaceans are carniv-
orous; the baleen or whalebone whales
(families Balaenidae, Balaenopteridae, Esch-
richtiidae and Neobalaenidael are filter
feeders, feeding on organisms several orders
of magnitude smaller than they are.

Of four living members of the order Sirenia,
only one - the dugong Dugong dugon - is
exclusively marine, occurring widely in coastal
waters of the Indo-Pacific. One other, the
Caribbean manatee Trichechus manatus, is

0-"^

• t

'"Vfl^,,'

found in both marine and inland waters
while the other two Ithe Amazonian manatee
Trichechus inungulsand West African manatee
Trichechus senegalensis] enter coastal waters
marginally if at all. One other species, the very
large Steller's sea cow Hydrodamalis gigas,
survived in waters around Bering and Copper
Islands in the North Pacific until the early 18th
century. All sirenians are herbivores; marine
populations feed mainly on seagrasses.

The remaining marine mammals are all
included in the order Carnivora. Two New
World otters in the family Mustelidae, the sea
otter Enhydra lutris from the north temper-
ate Pacific coast and the marine otter Lutra
felina from the south temperate Pacific coast,
feed very largely or exclusively in marine
waters; other otter species may frequent
coastal areas but are predominantly inland

Marine biodiversity 131

r^— C
/,

•1 ._v^

•\

water animals. Members of the tliree pinni-
ped families Odobenidae (the walrus],
Otariidae leared seals! and Phocidae (earless
seals] are all largely aquatic, emerging on
land to breed and rest, particularly when
molting; all are marine with the exception of
one or two species of Phocidae (the Baikal
seal Phoca sibirica and, if the Caspian is
regarded as a lake rather than a sea, the
Caspian seal Phoca caspica]. One member of
the family Phocidae, the Caribbean monk
seal Monachus tropicalis, has become extinct
this century. All species are carnivorous.

In contrast to most terrestrial mammal
families, pinnipeds are considerably more
diverse and more abundant at higher rather
than lower latitudes. Of the 32 extant or
recently extant species, only five occur within
the tropics (two marginally]. Part of the

explanation for this undoubtedly lies in the
greater availability of suitable habitat at
higher latitudes; as noted above, 70 percent
of continental shelf waters and just over 60
percent of the world's marine area are found
outside the tropics. However, this in itself is
unlikely to account for the entire difference.
It is probable that the greater productivity
of shelf waters at high latitudes, discussed
above, and of upwellmg areas at mid-
latitudes (e.g. the Benguela current off the
western coast of South Africa and the
Humboldt current off Chile and Peru] plays a
major part.

The isolated character of many island
breeding sites, such as the Galapagos group,
in temperate and sub-polar parts of the
southern hemisphere may also have en-
couraged speciation of pinnipeds here.

132 WORLD ATLAS OF BIODIVERSITY

Table 6.8

Diversity of mangroves

Note: Where two figures are
given, second figure indicates
number of fiybrids. Families
composed solely of mangrove
species are stiown in bold.

Source: Adapted from Duke and
Spalding e( 3l. .

COASTAL AND SHALLOW WATER COMMUNITIES
Mangroves

Mangrove woodland is indeed a truly hybrid
terrestrial/marine ecosystem, unique in that
terrestrial organisms can occur in the canopy
and marine species at the base". Mangroves,
or mangals, are a diverse collection of shrubs
and trees lincluding ferns and palms) which
live in or adjacent to the intertidal zone and
are thus unusual amongst vascular plants in
that they are adapted to having their roots at
least periodically submerged in sea water A
wide variety of organisms is associated with
mangroves including a number of epiphytic,
parasitic and climbing plants, and large
numbers of crustaceans, mollusks, fishes
and birds'l

Mangrove species are generally divided
into those found only in mangrove habitats
and those that may also be found elsewhere
but which are nevertheless an important
component of mangrove habitats. Both
groups come from a wide range of families.
The former includes around 62 species and
seven hybrids in some 22 genera (Table 6.81.
The appearance of mangroves is far from
uniform: they vary from closed forests

^ ^^1

^^^^■(o. of

Filicopsida

Adiantaceae

Plumbaginales

Plumbaginaceae

Theales

Pelliciceraceae

Malvales

Bombacaceae

Sterculiaceae

Ebenales

Ebenaceae

Primulales

Myrsinaceae

Fabales

Leguminosae

Myrtales

Combretaceae

Lythraceae
Myrtaceae

4+1
1
1

Sonneratiaceae

6+3

Rhizophorales

Rhizophoraceae

17+2

Euphorbiales

Euphorbiaceae

Sapindales

Meliaceae

2+1

Lamiales

Avicennlaceae

Scrophuliariales

Acanthaceae
Bignoniaceae

2
1

Rubiales

Rubiaceae

Arecales

Palmae

40-50 m high to widely separated clumps of
stunted shrubs less than 1 m high". They are
only able to grow on shores that are shelter-
ed from wave action, and are particularly
well developed in estuarine and deltaic
areas; they may also extend some distance
upstream along the banks of rivers, e.g.
some 300 km up the Fly River in Papua New
Guinea.

Mangrove communities are largely re-
stricted to the tropics between 30°N and 30°S,
with extensions beyond this to the north in
Bermuda and Japan, and to the south in
Australia and New Zealand" (Map 6.4). They
occur over a larger geographical area than
coral reefs (see below) and, unlike reefs, are
well developed along the western coasts of
the Americas and Africa. They have a more
restricted distribution than coral reefs in the
South Pacific.

There are two main centers of diversity.
The eastern group occurs in the Indo-Pacific
(the Indian Ocean and western part of the
Pacific Ocean) and is the most species rich'^".
The western group is centered around the
Caribbean and includes mangrove com-
munities along the west coast of the Americas
and Africa.

Global mangrove area is believed to slightly
exceed 180 000 km', divided regionally as
shown in Table 6.9. Mangroves occur in over
100 countries (including dependent terri-
tories) but exist in very small areas in many of
these. Four countries (Indonesia, Brazil,
Australia and Nigeria) between them account
for over 40 percent of the world's mangrove
area, and Indonesia alone possesses nearly
one quarter of the global mangrove area".

Although it is known that mangrove eco-
systems in most parts of the world have been
extensively degraded and cleared, it is difficult
to obtain reliable data on the global extent
of mangrove loss over time. Mangroves by
their very nature occupy highly dynamic and
unstable environments so that even without
human action the location and extent of
mangrove cover would be constantly chang-
ing. One assessment" suggested that more
than 50 percent of the worlds mangrove
forest cover had been destroyed.

Mangroves stabilize shorelines and de-

Marine biodiversity 133

crease coastal erosion by reducing the energy
of waves and currents and by holding the
bottom sediment in place with their roots.
They also act as windbreaks and provide
protection from coastal storms. They are
generally highly productive ecosystems and
are important habitats for crustaceans,
shellfish and finfishes. Most of the larger
commercial penaeid shrimps are mangrove
dependent; these and other species are
harvested both on a subsistence basis and
commercially, and may provide a major source
of income in some countries.

As well as providing habitat for adults of
many species of finfish and invertebrates,
mangroves serve as spawning and nursery
areas for many others, often of major
economic importance. The wood provides
building material, used locally in houses, as
fence poles and to build fish traps, and is also
harvested on a large scale for production of
pulp and particle board. In many areas
mangroves are also an important source of
fuel, as firewood and charcoal. Mangrove
foliage may provide an important source of
fodder for domestic livestock in some coun-
tries, particularly during dry seasons when
other sources of greenery are in short supply.

Salt marshes

Salt marshes are coastal communities of
rooted salt-tolerant Ihalophyticl plants of
terrestrial origin, dominated by grasses,
herbs and dwarf shrubs. They share many
characteristics with mangroves but replace
them geographically in higher latitudes,
except for some overlap at the extremes of
mangrove distribution (in the Gulf of Mexico,
Japan, southern Australia and northern
New Zealand). Globally, salt marshes are
estimated to cover around 350 000 km' in
total". They tend to develop in sheltered areas
of mud and silty sand that are flat and slow
draining; as plant growth and sedimentation
elevates the marshland, so the period of tidal
submergence decreases and a network of
creeks develops. The two genera which are
most prominent as pioneer saltmarsh plants
are Salicornia, or samphire, and Spartina, or
cord grass. Species of Puccinetla, Sc/rpus and
Juncus are also common.

Salt marshes are highly productive, around
2 500 g C per m' per year Occurring in highly
seasonal latitudes, salt marshes rapidly take
up and accumulate nutrients during the
growing season. In the autumn, when plants
die or become dormant, the uptake of nutri-
ents is greatly diminished and dead organic
matter may be transported out of the marsh.
The physical features of salt marshes mean
that pollutants are not easily or rapidly
flushed out. Other threats include infilling,
especially around the North Sea", and in-
creased erosion through channel dredging.

Rocky shores

Rocky shorelines are generally exposed to
oceanic swells and extreme wave action
(except for more sheltered fjordlandsl but are
topographically variable, occurring as wide
platforms, steep cliffs or other formations,
depending on local geology™. They provide a
unique habitat for plant growth, with a stable
substrate for attachment, and shallow well-
lit water that tends to be turbulent and
rich in nutrients. Consequently macro-algae
communities, dominated by kelps [Laminaria,
Ecktonia, Macrocystis] and fuccoids [Fucus,
Ascophyllum] flourish, mainly in temperate
regions but also in areas of the tropics where
seasonal upwellings of cold water occur, such
as Chile and the southern Arabian coast-'. The
net primary productivity of kelp forests
is comparable to tropical rainforests, and
Macrocystis pyrifera, the giant kelp, can attain
growth of up to 45 cm a day

Two physical factors - water movement
and desiccation - have influenced the diversity
of intertidal rocky shore species. Strong wave
action has favored use of crevices, dense
aggregations of individuals, and the evolution
of strong attachment devices (algal holdfasts.

Table 6.9

Current mangrove cover

Source: Spalding e( a/. .

Lg^^^^pP

^^f Mangrove area ^^H
(to nearest thousand km')

■|^Uftotal|

■

South and Southeast Asia
Americas

75 000
49 000

42
27

West Africa

28 000

Australasia

19 000

East Africa and Middle East

10 000

Total

181 000

13« WORLD ATLAS OF BIODIVERSITY

Map 6.4
Mangrove diversity

The location of current
mangrove forest, togetfier
witfi contours representing
gradients of mangrove
species ricfiness. Note
tfiat grapfiic presentation
at tliis scale enormously
exaggerates actual forest
area.

Source; Reproduced with modification,
from Spalding

Mangrove locality and
species richness

• Mangrove forest

High diversity

Low diversity

cementation of barnacles, the byssus threads
of mussels and the adhesive feet of gastro-
pods and echinodermsl. The need to avoid
being washed off the rocks means that most
organisms on rocky shores are sessile or have
limited motile ability. Competition for space is
therefore intense and organisms inhabiting
rocky shores occupy well-defined zones. It
also means that they can be particularly
vulnerable to disturbances such as oil spills,
especially if these occur during calm periods
which extend the residence time of pollutants.
However, because they are generally exposed
and high-energy environments, rocky shores
tend to be less vulnerable than most to
pollution. These features also limit some
destructive human activities, such as
construction, responsible for degrading other
marine habitats.

r-qr

Seagrasses

Seagrasses are a mixed group of flowering
plants Inot true grasses) that are adapted to
live submerged in shallow marine and estuar-
ine environments at a wide range of latitudes.
About 58 species are recognized by many
authorities, in four families, all within the
monocotyledons. They occur from the littoral
region to depths of 50 or 60 m but are most
abundant in the immediate sublittoral area.
There are more species in the tropics than in
the temperate zones, and of the 12 seagrass
genera seven are confined to tropical seas
and five to temperate seas". Most seagrass
species are similar in external morphology,
with long thin leaves and an extensive
rhizome root system which enables them to
fasten to the substrate. A variety of substrates
are occupied from sand and mud to granite

Marine biodiversity 135

fr''

rock, but the most extensive beds occur on
soft substrates".

While species diversity (see Map 6.51 is
highest in Southeast Asia las with mangroves
and corals) there are two important centers in
temperate regions: in mainland Japan and in
southwest Australia. Secondary centers of
diversity include East Africa, the Red Sea
and the Mediterranean. Although seagrasses
themselves are not a diverse group, they
support considerable diversity in some assoc-
iated species, including an estimated 450
species of epiphytic algae''.

In many areas seagrasses form extensive
but simple communities, referred to as sea-
grass beds or seagrass meadows. Seagrass
beds have high productivity and contribute
significantly to the total primary production of
inshore waters. Seagrasses are particularly

important in nutrient-poor (oligotrophicl
waters because they can extract some nutri-
ents from sediments, unlike other marine
primary producers. Many commercially impor-
tant species are dependent on seagrass beds,
often as nursery habitat, providing shelter
from predators and adverse sea conditions.
These include mollusks (such as the queen
conch Strombus gigas], shrimp, lobster, holo-
thurians and many finfish (such as grunts,
Haemulidae; rabbitfish, Siganidae; emperors,
Lethrinidae; and snappers, Lutjanidael. A
small but important number of threatened
species depend on seagrasses, including
sirenians, the green turtle Chelonia mydas
and many species of seahorse (Syngnathidae).
In addition to such direct values, seagrass
beds also play an important role in binding
sediments, providing some protection from

136 WORLD ATLAS OF BIODIVERSITY

Map 6.5

Seagrass species diversity

This first global map of
seagrass diversity indicates
tfie extent of seagrass
fiabitat (inventory sites) and
stiovtfs contours of species
ncfiness. Diversity contours
not sfiown for parts of West
Africa because species
inventory incomplete.

Source: Prelimtnarv plot, compiled using
multiple sources, from a seagrass atlas
in preparation at UNEP-WC(»1C.

Seagrass locality and
species richness

• Seagrass

High diversity

Low diversity

coastal erosion, and may help remove excess
nutrients and toxins from coastal waters.

A preliminary estimate'' suggested that
seagrass beds extend over some 600 000 km'
globally, but a precise figure is not yet
available. Although seagrasses can be highly
dynamic ecosystems, with individual seagrass
beds undergoing significant shifts in distri-
bution over relatively short periods, there is
considerable evidence that there have been
major net losses over the past century.
Extensive losses along the Atlantic coasts of
Europe and North America in the 1930s
followed disease caused by a marine slime
mold; some evidence suggests that other
environmental impacts (possibly increased
turbidity in coastal waters) may have increas-
ed susceptibility to disease. More recently,
nutrient loading has continued to increase in

'^M'^

•^^' '•?->

v^, -

coastal waters worldwide, often leading to
enhanced growth of phytoplankton, epiphytes
and macroalgae, all of which can out-
compete seagrasses for available sunlight.
Sedimentation also has a major impact,
reducing the passage of light through the
water column and physically smothering
seagrass plants. To a lesser degree, toxic
pollutants and physical disturbance from
activities such as trawling and dredging have
also played a role in seagrass losses".

Shallow tropical coral reefs

The term coral reef applies to a variety of
calcium carbonate structures developed by
stony corals. They are tropical shallow-water
ecosystems, typically with high biodiversity
and largely restricted to coastal seas between
the latitudes of 30°N and 30°S", They are

Marine biodiversity 137

V^gi-

"V,

^k:

-r^-.-: ii.'-^

> \

Vj:

>^-. ^

most abundant in shallow, well-flushed
marine environments characterized by clear,
warm, low-nutrient waters that are of oceanic
salinity". There are two basic categories:
shelf reefs, which form on the continental
shelves of large land masses, and oceanic
reefs, which are surrounded by deeper waters
and are often associated with oceanic islands.
Within these two categories there are a num-
ber of reef types: fringing reefs, which grow
close to shore; barrier reefs, which develop
along the edge of a continental shelf or
through land subsidence in deeper waters,
and are separated from the mainland or
island by a relatively deep, wide lagoon; and
atolls, which are roughly circular reefs around
a central lagoon and are typically found in
oceanic waters, probably originating from the
fringing reefs of long-submerged islands.

Two other less clearly defined categories are
patch reefs, which form on irregularities on
shallow parts of the seabed, and bank reefs,
which occur in deeper waters, both on
continental shelf and in oceanic waters".

The global extent of coral reefs is not
l<nown with certainty. A recent estimate,
derived by measuring the total reef extent in
a comprehensive set of national maps, sug-
gests a world total of 28^ 300 km-. This is
equivalent to a little more than 1 percent of
the total world ocean shelf area. New
information and mapping techniques may
increase the accuracy of such estimates, but
the total is unlikely to exceed 300 000 km'.
A regional breakdown is provided in Table
6.11. Although coral reefs occur in around
110 countries and territories, just five
countries - Indonesia, Australia, Philippines,

138 WORLD ATLAS OF BIODIVERSITY

Table 6.10

Diversity of stony corals in

the order Scleractinia

Note: Includes only the reef-
forming scleractinians with
zooxanthellae.

r- w 2'

Source: Veron .

France (overseas departments and territories
in the Indian Ocean and Pacific) and Papua
New Guinea - account for over half of the
global total.

Five different orders within the phylum
Cnidaria (coelenteratesl include species with
calcified skeletons, or stony corals'. Many of
these are reef building Ihermatypicl and some
are solitary lahermatypicl. The great majority
of hermatypic corals belong to the order
Scleractinia. the true stony corals, although
not all scleractinians are reef-builders.
Together with sea anemones and sea fans, the
Scleractinia mal<e up the class Anthozoa.
Most scleractinian coral polyps have symbiotic
algae Izooxanthellael within their tissues;
these use the nitrates, phosphates and carbon
dioxide produced by the coral, and through
photosynthesis generate oxygen and organic
compounds that provide much of the polyps'
nutrition. The zooxanthellae give corals their
color and, because they photosynthesize,
restrict the corals that contain them to the
photic zone-'. Corals without zooxanthellae
typically do not form reefs and can exist in
deeper colder waters (see deep-water reefs,
below]. Because of difficulties with synonymy
and in defining species, it is difficult to
estimate precisely how many extant species

^l^y^^^^HPi

Acroporidae

262

Agaricidae

Astrocoenidae

Caryopiiylliidae

Dendrophyllidae

EupiiylUdae

Faviidae

125

Fungiidae

Meandriniidae

Merulinidae

Mussidae

Oculinidae

Pectiniidae

Pocilloporidae

Poritidae

Rhizangiidae

Siderastreidae

Trachypfiyllidae

of reef-building coral there are. A recent
synthesis-' deals with around 800 zooxan-
thellate scleractinians (see Table 6.101;
another" lists more than 1 300 scleractinians
in all, and 260 calcified hydrozoans.

Not all reefs are constructed primarily
by corals. Within the red algae (phylum
Rhodophytal and the green algae (phylum
Chlorophytal in particular several genera
grow heavily calcified encrustations which
bind the reef framework and in places are the
main contributors to shallow reef growth.

Coral reefs are among the most productive
and diverse of all natural ecosystems. The
mam center of diversity for reet-building
corals is Southeast Asia, with an estimated
minimum of 450 species found associated
with reefs around the Philippines, Borneo,
Sulawesi and associated islands. This area
IS part of a single, vast, Indo-West Pacific
biogeographic province that extends from the
Red Sea in the west to the Pitcairn Islands in
the east. Many coral genera and a significant
number of species are found throughout the
region, although overall diversity in the pro-
vince decreases on leaving this center. In the
east of this region, the central and eastern
Pacific forms a series of somewhat distinct
subregions, characterized by a number of
genera and species (particularly in Hawaii]
not found further west. This area also shares
many species with the Indo-West Pacific
province but overall has much lower diversity
than most of the latter The Atlantic, including
the Caribbean and the Gulf of Mexico, forms a
distinct province with few species in common
with the Indo-West Pacific. It is also very
depauperate compared with most of the latter
(Map 6.6].

Deep-water reefs

Beyond the coral reefs of shallow tropical
waters, a few species of coral that lack
zooxanthellae form deep-water reefs on hard
substrates in high-current areas associated
with topographic rises, such as ridges and
pinnacles. The best known is Lopheiia
pertusa, a colonial coral that forms structures
ranging from patches a few meters in width
to reefs many hundreds of meters in size, at
depths of 100-3 000 m and temperatures of

Marine biodiversity 139

i-SX. Although best known through trawling
and oil exploration activities in the northeast
Atlantic, these deep-water reets are global in
occurrence.

The complex matrix of living and dead
branches of Lophelia increases spatial
heterogeneity above that of the surrounding
seabed and provides a habitat for many
species. Boring sponges, anemones, bryo-
zoans, gorgonians, polychaetes, barnacles and
bivalves occur in large numbers" and their
diversity is comparable to some shallow-
water tropical systems'". Although targe
aggregations of fish are associated with
Lophelia reefs, and reef areas support higher
catches than adjacent seabed", the species
diversity of fish and coral is much lower than
in tropical coral reefs (some 23 species of fish
have been recorded on Lophelia reefs in the
northeast Atlantic!.

The reefs are delicate structures easily
destroyed by demersal fishing gear which
routinely operate to depths of 2 000 m. The
total destruction of some Norwegian reets
has already been documented and an esti-
mated 30-50 percent of others have been
damaged. Slow growth, in the region of
4-25 mm per year, severely limits the ability
of Lophelia to recover The trawl fishery for
orange roughy Hoplostethus atlanticus and
oreos Allocyttus niger on sea mounts south
of Tasmania has been responsible for sub-
stantial destruction of Solenosmilia variabilis
reefs, with some reduced to more than 90
percent bare rocks". If recovery ever occurs it
will take hundreds of years.

OCEANIC PELAGIC COMMUNITIES

There is a fundamental distinction between
the processes and patterns observed in open
oceans, dominated by global winds and large-
scale vertical and horizontal movement of
water masses, and those observed nearer to
coasts, where shelf bathymetry, coastal winds
and local input of nutrients, pollutants and
sediments generate a diversity of smaller-
scale phenomena.

The oceanic pelagic zone is dominated by
the activity of plankton in the euphotic surface
waters. Plankton are by definition drifting or
weakly swimming organisms, comprising a

1 Region ^^^^^^^^^^H

Shallow reef area
Ito nearest thousand km'l

% of world total jH

Indo-Pacific

261 200

91.9

Red Sea, Gulf of Aden

17 400

6.1

Arabian Sea, Persian Gulf

4 200

1.5

Indian Ocean

32 000

11.3

Southeast Asia

91700

32.3

Pacific

115 900

40.8

Wider Atlantic

2U00

7,6 -

Caribbean

20 000

7.0

Atlantic (excl. Caribbean]

1600

0.6

Eastern Pacific

1600

0.6

Total

284 300

wide range of small to microscopic animals,
protoctists and bacteria. Free-swimming
pelagic organisms, predominantly fishes but
also cetaceans and cephalopod mollusks
(squid), are collectively termed nekton. These
organisms, when adult, are predators of
plankton or smaller nekton. They in turn - as
vertically migrating fishes or larvae, and as
dead organic material - provide food for deep-
sea and benthic organisms. With few excep-
tions, the only other food source for creatures
in the aphotic zone is the 'rain' of organic
matter, such as feces, molted crustacean
exoskeletons, and a variety of other organic
material derived from plankton in the surface
waters of the ocean. Plankton and larger free-
swimming organisms tend strongly to con-
centrate along major circulation currents
(gyresl, contact zones and upwellings, and
this can give rise to significant local variation
in diversity.

The marked vertical gradients within the
pelagic zone - of light, temperature, pressure,
nutrient availability and salinity - lead to
vertical structuring of pelagic species assem-
blages. Several zones based on changes in
species composition with depth have been
recognized, including epipelagic (usually
taken as from the surface to a depth of 200-
250 m and including the euphotic zone);
mesopelagic, which underlies the epipelagic
zone to a depth of 1 000 m or so; and below
this the bathypelagic which changes in a
somewhat less well-defined fashion to
abyssopelagic at around 2 500-2 700 m depth.

Table 6.11
Coral reef area

Source: Spalding ef at.

uo WORLD ATLAS OF BIODIVERSITY

Map 6.6
Coral diversity

The location of coral reefs,
togettier with contours
representing gradients of
species richness annong
reef-building scleractmian
coral species. Note that
graphic presentation at
this scale enormously
exaggerates actual reef
area.

Source: Reproduced by permission, wilt
modification, from Veron . revised by
Veron, pers. comm. November 2001.

Coral locality and
species richness

Coral

High diversity

Low diversity

These zones, however, tend to fluctuate in
time and space. As well as seasonal changes
in water characteristics, many components of
the epipelagic and mesopelagic nel<ton under-
go marked diet migrations (i.e. on a 24-hour
cycle], ascending to surface waters at night to
feed and descending, sometimes over 1 l<m,
during the day. Many species of nel<ton,
particularly cetaceans and larger fishes, are
also highly migratory, ranging over enormous
expanses of ocean in more or less regular and
predictable patterns.

It has generally been assumed that bio-
mass in the pelagic zone everywhere below
the euphotic zone is low. However, recent
studies have indicated that biomass of tropical
mesopelagic animals may be surprisingly
high. Study of the mesopelagic fauna has been
limited to date, as it requires the use of

expensive high-seas research vessels; i<now-
ledge of taxonomy, distribution and biology of
most of the species concerned remains very
incomplete'. One study" recognized around
160 fish genera in 30 families as important
components of the fauna. Most species are
small (less than 10 cm in length) and often
bizarrely shaped. Estimates based on a variety
of surveys carried out indicate that global
biomass of this stocl< may be large: a figure of
650 million tons (some six to seven times total
current marine fisheries landings! has been
suggested, although this should be regarded
with extreme circumspection'.

From available data, it would appear that
the mesopelagic biomass is greatest in the
northern Indian Ocean, and particularly in the
northern Arabian Sea, one of the five major
upwelling zones. Surveys here indicated

Marine biodiversity 141

extremely high blomass 125-250 g per m') in
the Gulf of Aden and Gulf of Oman as well as
off the western coastline of Pakistan. These
figures are around an order of magnitude
higher than those recorded elsewhere in the
tropics, indicating either a great overestimate
for the northern Indian Ocean or an under-
estimate elsewhere. Alternatively this region
genuinely is ten times as productive as the rest
of the tropical ocean system. Although this
appears as yet unresolved, it is nevertheless
apparent that there is substantial global
mesopelagic fish biomass ar.d that the Arabian
Sea is particularly rich in these species.

DEEP-SEA COMMUNITIES

Approximately 51 percent of the Earth's
surface is covered by ocean over 3 000 m in
depth, so deep-sea communities are preva-

lent over a significant proportion of the planet.
All deep-sea habitat is in the aphotic zone,
well below the distance sunlight can pene-
trate. As deeper and deeper levels are
reached biomass falls exponentially^'.

Despite their enormous volume, the deep
oceans appeared to be relatively simple
ecosystems and to make little contribution to
global species diversity, but discoveries
during the past decade or so have shown that
in some regions species diversity in the
benthic community increases with increasing
depth. This was revealed by novel sampling
techniques, principally the epibenthic sled^^
The rate of discovery of new species and the
proportion of species currently known from
only one sample both suggest that a great
number remain to be discovered"". As
with arthropods in tropical moist forests.

142 WORLD ATLAS OF BIODIVERSITY

estimates for the number of unknown species
vary widely, witfi some suggestions tfiat there
may be as many as 10 million undescribed
species in the deep sea". Others consider that
the true figure is more likely to be around
500 000".

Ocean trenches

Ocean trenches are typically close to land
masses and tend to have high rates of
sedimentation, a significant amount of which
is of organic origin and an important available
food source for trench communities. Several
trenches also underlie highly productive cold-
water upwelling zones, the organic fallout
from which contributes greatly to productivity
there. The water within trenches generally
originates from the surrounding bottom
water, which is derived from cold surface
water at high polar latitudes and is relatively
well oxygenated"'.

Trenches tend to be isolated linear systems
that because of their seismic activity form a
habitat that is unstable and unpredictable
compared with the relative environmental
stability of the adjacent abyssal plains"'.
Trench faunas are not rich in species but are
often high in numbers of endemic species.
There are some 25 genera restricted to the
ultra-abyssal [hadall zone, representing some
10-25 percent of the total number of genera
present, and two known endemic hadal
families: the Galatheanthemidae (Cnidarial
and Gigantapseudidae (Crustacea). The latter
family contains a single species: Giganta-
pseudes adactylus. The greatest number of
endemic species known from a single trench
is a sample of 200 from the Kurile-Kamchatka
trench; this maybe compared with 10 endemic
species known from the Ryukyu and Marianas
trenches.

Hydrothermal vents

Hydrothermal vent communities were first
discovered in 1977, at a depth of 2 500 m on
the Galapagos Rift, but contribute to what
might be one of Earths most archaic eco-
systems. They are now known to be assoc-
iated with almost all known areas of tectonic
activity at various depths. These tectonic
regions include ocean-floor spreading centers,

subduction and fracture zones and back-arc
basins". Cold bottom water permeates
through fissures in the ocean floor close to
ocean floor spreading centers, becomes
heated at great depths in the Earth's crust
and finds its way back to the surface through
hydrothermal vents. The temperature of vent
water varies greatly, from around 23°C in the
Galapagos vents to around 350°C in the vents
of the East Pacific Rise, and they may be rich
in metalliferous brines and sulfide ions'".
Most species live out of the main flow at
temperatures of around 2°C, the ambient
temperature of deep-sea water The biomass
of vent communities is usually high compared
with other areas of similar depth, and dense
colonies of tube worms, clams, mussels
and limpets typically constitute the major
components.

Hydrothermal vent communities are of
particular interest because they flourish
in the dark at high pressures and low
temperatures", and unique because they are
supported by chemolithoautotrophic archeans
and bacteria, notably Thiomicrospira species
(phylum Proteobacteria), which form dense
microbial carpets in the rich hydrothermal
fluid and derive their energy chiefly from
oxidizing hydrogen sulfide"". Many of the
eukaryote vent species filter-teed on these
microorganisms, whilst others rely on sym-
biotic sulfur bacteria for energy".

The overall species diversity at vents is low
compared with other deep-sea soft sediment
areas", but endemism is high. More than 20
new families or subfamilies, 50 new genera
and nearly 160 new species have been
recorded from vent environments, including
brine and cold seep communities"". Vent
communities are separated by gaps of
between 1 and 100 km, and although they may
persist only for several years or decades, sites
of vent activity move relatively slowly allowing
dispersal of vent organisms".

Two further seep patterns are known. Cold
sulfide and methane-enriched groundwater
seeps occur near the base of the porous
limestone of the Florida escarpment, as well
as in the Gulf of Mexico and elsewhere. These
support a community similar in taxonomic
composition to the hydrothermal vents of the

Marine biodiversity 143

''%

east Pacific. Tectonic subduction zone seeps
are more diffuse and lower in temperature
than hydrothermal vent seeps and are rich in
dissolved methane. They are known to occur
off Oregon and in the Guaymas basin in the
Gulf of California, and cold seeps occur at a
depth of 1 000 m in Sagami Bay near Tokyo
and in the subduction zones of the trenches
off the east coast of Japan.

HUMAN USE OF AND IMPACT ON THE OCEANS

The seas provide many biological resources
used by humans. A wide range of animal
species, notably fishes, mollusks and crus-
taceans, contribute to marine fisheries and
these provide by far the most Important
source of wild protein, of particular Impor-
tance to many subsistence communities
around the world. Marine algae are also an
Increasingly important foodstuff, notably in
the Far East, with current annual world
production of around 2 million metric tons.
Marine organisms are also proving extremely
fruitful sources of pharmaceuticals and other
materials used In medicines. Relatively minor
although locally Important uses include
exploitation of coastal resources for building
materials (e.g. coral limestone, mangrove
poles] and other industrial products (e.g.
tannins from mangroves).

Access to marine resources is not equitably
distributed amongst the world's nations. Most
obviously, some 39 states are landlocked, i.e.
have no seaboard (five of these border the
Inland Caspian Seal. Those that do have sea-
boards show great variation in length of
coastline, and area of territorial waters and
exclusive economic zones lEEZsl, both abso-
lutely and relative to their land areas. They also
show great variation In their capacities to
exploit marine resources, both on the high seas
and within their territorial waters and EEZs.

Human activities, directly and indirectly.
are now the primary cause of changes to
marine biodiversity. Approximately one third
of the world's human population lives in the
coastal zone Iwithin 60 km of the seal and
indications are that this proportion will rise
during the 21st century". Pressures exerted
by the human population on the marine
biosphere are substantial and Increasing.

-:^# tr^^

-^ ^s«.

Most identified threats relate to coastal
and Inshore (continental shelf] areas. How-
ever, threats to the oceanic realm are
undoubtedly Increasing: fisheries and their
attendant physical effects, such as habitat
alteration owing to dredging and trawling,
have entered deeper continental slope waters
having previously been largely confined to the
epipelagic zone, and deep-water oil and gas
mining is planned. Even abyssal and hadal
areas are susceptible to human Impact. A
small, steady increase In abyssal temperature
of 0.32°C In 35 years has been attributed to
global climate change brought about by
human activities. Ocean waste dumping and
the potential for deep-water mining and
mineral extraction are also causes for

Many species of fishes,
mollusks and crustaceans
provide humans with
their largest source of
wild protein.

u« WORLD ATLAS OF BIODIVERSITY

Just five species of finfish.
among them tfie Atlantic
herring, account for a
quarter of the global catch.

concern, as are the changes in biomass and
species composition in the waters above
these regions".

The following five activities have been
identified as the most important agents of
present and potential change to marine bio-
diversity at genetic, species and ecosystem
levels":

• fisheries operations;

• chemical pollution and eutrophication;

• alteration of physical habitat;

• invasions of exotic species;

• global climate change.

WORLD MARINE CAPTURE FISHERIES

World marine capture fisheries have grown
fivefold in the past half-century, with annual
landings increasing from nearly 18 million
metric tons between 1948 and 1952 to around
87 million metric tons during the period 199A-
97, with a fall caused mainly by El Nino in 1998
and a subsequent recovery in 1999. Marine
capture fisheries made up just over 70 percent
of current recorded global production of
aquatic resources in the late 1990s, the

l^".'M

remainder being accounted for by inland
capture fisheries Isee Chapter 7| and
aquacutture"". With capture fisheries appar-
ently remaining more or less stable, the
increase in total marine production during the
1990s was due to a continuing increase in
aquaculture production.

Composition of marine fisheries
Marine fisheries encompass a wide range of
organisms, including algae, invertebrate ani-
mals in various phyla and vertebrates including
fishes (often termed finfishes in fisheries analy-
sis], reptiles, mammals and birds (although by
convention the last of these groups is not
normally considered in fisheries analysis).

The Food and Agriculture Organization of
the United Nations IFAOI recognizes in total
just under 1 000 species items' (species,
genera or families! that feature at least
periodically in national catch statistics.
However, globally important marine fisheries
are confined to relatively few groups, with
over 80 percent of landings by weight being
finfishes and virtually all the remainder
mollusks and crustaceans.

In terms of major species groups, by far the
most important are the herrings and anchovies
in the order Clupeiformes, which in 1998
accounted for over 22 million metric tons, or
around 25 percent of marine landings. These
are followed by cod, hake and haddock
(GadiformesI, and jacks and mullets (some Per-
ciformes and MugiliformesI, with production of
more than 9 million and nearly 8 million metric
tons respectively The most important inver-
tebrate group overall is cephalopod mollusks
(squid, cuttlefish and octopus! of which some
3.A million metric tons were reported landed.

In terms of individual species, for several
years during the 1990s just five species of
finfish, anchoveta Engraulis ringens, Alaska
pollock Theragra chalcogramma, Chilean jack
mackerel Trachurus murphyi, Atlantic herring
Clupea harengusar\6 chub mackerel Scomber
japonicus, together made up around one
quarter of marine landings. Each accounted for
over 2 million metric tons annually, and among
several others around this level, Japanese
anchovy Engraulis japonicus and Skipjack
tuna Katsuwonus pelamis were increasingly
important. In most years by far the most im-
portant single species is the anchoveta
Engraulis ringens, whose fishery (off the west
coast of South America) was nearly 13 million
metric tons in 1994, constituting by far the
largest single-species fishery the world has
ever seen; but just under 8 million in 1 997 and
less than 2 million in 1998.

Marine biodiversity 145

Distribution of marine fisheries

The geographical distribution of marine
fisheries is determined both by the distri-
bution of harvestable fish stocl<s and by a
range of complex socioeconomic factors. The
former is largely determined by variations in
productivity, which, as noted above, are them-
selves largely determined by nutrient availa-
bility, so that overall the most productive
fisheries areas are on continental shelves at
higher latitudes and in upwelling zones at
lower latitudes. As a generalization, the latter
are associated with pelagic fish stocks and
the former more with demersal or semi-
demersal (deep-water or bottom-dwelling)
stocks, although pelagic stocks play an in-
creasingly important role even here.

As might be expected purely on the basis
of its size, the Pacific Ocean is by far the most
important major fisheries area, accounting
for over 60 percent of marine landings. The
northwest Pacific alone - an area with
extensive continental shelf development -
accounts for nearly half this total.

The various upwelling zones are not all of
equal importance in fisheries. That associated
with the Humboldt current off Peru and Chile
is the single most productive, while those
associated with the California, Benguela and
Canary currents are of somewhat lesser
importance, although each is still a major
fisheries area. The Arabian Sea upwelling
appears to be anomalous, in that it evidently
supports major populations of mesopelagic
li.e. middle-depth) rather than epipelagic
species. Not only are the former generally
considered of low value, with an identified
market only as animal feed, but capture and
processing requires expensive, advanced
technology They thus remain virtually un-
exploited at present and are considered along
with the Antarctic krill stocks to be the major
unexploited fisheries resource left.

Trends in marine fisheries

National fishery statistics are collated by the
FAG. These data are the principal source of
information on global fishery trends, although
it is widely acknowledged that they are
variable in quality During the 1950s and
1960s, total landings increased steadily as

■ 1996 ^Sl 1998

Million metric tons
2 4 6 8 10

i ' \ I 1 1

Alaska pollock (Theragra chalcogrammal
4.5
I 4.0

, Atlantic herring IClupea harengusi
2.3

2.4

I Japanese anchovy lEngraulisjaponicusI
1.3
I 2.1

Cack mackerel ITrachurus murphyil
4.4
2.0

Skipjack tuna IKatsuwonus pelamisi
1.6
1.9

I Chub mackerel IScomber japonicusi
2.2
1.9

Largehead hairtail ITrichiurus lepturusl

1.3

HH 1.4

Atlantic cod ICadus morhuaj

1.3

^1 1.2

Yellowfin tuna IThunnus albacaresl

1.1

^1 1.2

Blue whiting IMicromesistius poutassoul

D_6
HI 1.2

new stocks were discovered, while improved
fishing technology and an expansion of fishing
effort enabled fuller exploitation of existing
stocks of both pelagic and demersal species.
Long-range fleets increased in size during
this period and, as traditional fishing grounds
in the North Atlantic and North Pacific
became fully exploited, moved into new fish-
ing grounds closer to the tropics and in the

Figure 6.1

Species contributing most

to global marine fisheries

U6 WORLD ATLAS OF BIODIVERSITY

Figure 6.2

Marine fisheries landings

by major group

Note: 1984-99 data are
for capture fisheries only:
pre-1984 data include
aquacuUure.

S30

= 20

southern hemisphere. By concentrating their
efforts in the richest ocean areas, these fleets
were largely responsible for the rapid
increase in world catches.

At the beginning of the 1 970s, the Peruvian
anchoveta fishery alone contributed some 20
percent of marine fisheries production. These
stocks collapsed around 1972, at the same
time as the important South African pilchard
fishery in the Atlantic, seemingly in asso-
ciation with an ENSO lEl Nino Southern
OscillationI event. There was a sharp drop in
overall marine fisheries production, after
which the global catch increased more slowly
than before, reaching the early 1970s level
by the end of the decade. Landings of most
demersal fish stocks remained relatively
constant, however, implying that they were
close to full exploitation. Long-range fleets
continued to expand in importance.

The 1980s once again saw a period of
continuous growth (averaging 3.8 percent a
year! in reported world landings. As in the
1970s landings of demersal stocks were
generally static or declining so that shoaling
pelagic species provided most of the increase
in fish production. In fact, just three pelagic
species [Peruvian anchoveta. South American
sardine Sardinops sagax, and Japanese

Pelagic

Demersal

Species not elsewhere included

Crustaceans

Mollusks lexcl. cephalopodsl

Cephalopods

1950

1999

sardine Sardinops melanostictus] and one
semi-demersal species (Alaska pollock)
accounted for 50 percent of the increase in
world landings during the 1980s". Most of
this increase appears to have been because
of favorable climatic effects on stock sizes
rather than new fishery developments or im-
proved management practices".

Following a sharp decline at the end of the
1980s, FAQ data indicate slow net growth in
marine capture fisheries through the 1990s.
Four of the five most important fishes in
fisheries in the late 1990s are pelagic, the
exception being the Alaska pollock. This
dominance of pelagic over demersal species
is reflected in overall fisheries figures, with
pelagic landings well over twice demersal
landings globally. This contrasts sharply with
the situation in the early 1950s when pelagic
landings were only some 30 percent greater in
volume than demersal landings (Figure 6.21.

This increasing dependency on pelagic fish
stocks is symptomatic of a major crisis in
global marine fisheries. In general demersal
fishes are more valuable per unit weight than
pelagic species so that all else being equal
the former are preferentially harvested.
The increased importance of the latter in the
past AO years is indicative of the growing
overexploitation of fisheries stocks worldwide
- as valuable demersal stocks have been
depleted so attention has turned to the
intrinsically less valuable pelagic stocks.

Of some 441 fishery stocks for which status
data were available in 1999, FAO considered
that only 4 percent were underexploited, and
with a further 21 percent assessed as
moderately exploited, around 25 percent
of stocks analyzed were above the level of
abundance thought to correspond to maxi-
mum sustainable yield (ivlSYI level (or have a
fishing capacity below this level). The remain-
ing 75 percent of stocks were considered to
require strict control of fishing capacity and
fishing effort in order for them to recover to
MSY biomass". The proportion of stocks in
this condition has increased between 1974,
when first systematically reviewed, and 1999
(Figure 6.3). In terms of ocean regions, the
situation worsened steadily in the North
Atlantic and North Pacific until the early

Marine biodiversity 147

•^*l*

1990s, when there were signs of possible
stabilization, mainly in the former. Stocks
appear still to be in decline in the tropical and
southern parts of these oceans, with some
possible stabilization in the tropical Atlantic,

Recent analysis of fishery data suggests that
the widespread overexploitation of marine
fisheries is probably more serious than the
global catch statistics indicate because mis-
reporting by countries with particularly large
fisheries, coupled with wide fluctuation in
Peruvian anchoveta stocks (linked with El Nino
events], can produce spurious trends at global
level. When more realistic estimates of the
catch in China, for example, are substituted for
reported catch figures believed to be incorrect,
the global catch appears to have declined by
0.36 million metric tons annually since 1988,
rather than increased by 0.33 million metric
tons (Figure 6.A1. The declining trend is much
steeper if the pelagic Peruvian anchoveta are
excluded^".

There are three major reasons for the
declining state of many marine fisheries. First,
and most fundamental, most fisheries have
traditionally been regarded as an 'open access'
resource, so that, in effect, it pays any one
fisher to harvest as much as possible at any
given time because, if they do not, somebody
else will. Secondly, technological innovations
have made fishing much more efficient.
Thirdly, there has been high investment in the
world's commercial fishing fleet (partly a
consequence of the nature of fisheries as an
open access resource but also for complex
socioeconomic and political reasons!.

Bycatches and discards

The effects of overfishing are compounded by
the wastefulness of many marine capture
fisheries. FAO estimated in 1994 that global
marine fisheries bycatch and discards
amounted to 18-40 million metric tons (mean
27 miUionI (Map 6.7|. This represented just
over 25 percent of the annual estimated total
catch (i.e. landings represent around 75
percent of actual catchl. Although figures are
not available, it is generally assumed that the
great majority of discards die. Further losses
are caused by the mortality of animals which
escape from fishing gear during fishery

Overexptoited ^
depleted +
recovering

1970

1975

1980

1985

1990

1995

2000

2005

operations, but it is impossible at present to
estimate the importance of this. Shrimp fishing
produces the largest volume of discards
(around 9 million metric tons annually).

Bycatches include non-target, often low-
value or 'trash' species, as well as undersized
fish of target species. Non-target species
may include marine mammals, reptiles (sea
turtles! and seabirds, as well as finfishes
and invertebrates. Of particular concern in
recent years has been mortality of marine
mammals, especially dolphins, in pelagic drift
nets, of sea turtles in shrimp trawls and
more recently of diving seabirds, especially
albatrosses, in long-line fisheries. Discarding

Figure 6.3

Global trends in the state

of world stocks since 1974

Source: Adapted (rom Figure iO in
FAo".

Figure 6.4

Trends in global fisheries

catch since 1970

Source: Adapted from Watson and
Pauly=°.

1970

1975

1980

1985

1990

1995

2000

us WORLD ATLAS OF BIODIVERSITY

Map 6.7

Marine fisheries catch and

discards

The location of the fishery
areas recognized by FAO for
statistical purposes is
shown on this map, with
symbols representing the
approximate late 1990s
yield from capture fisheries
and the volume of
discarded catch, most of
which IS presumed not to
survive. Each symbol
represents approximately
1 million metric tons.

Source: Data from FAO ,

Fisheries catch

<C^^^^!^ 1 million metric tons landed

^ \ {{{h" i 1 million metric tons discarded

may be a side-effect of management systems
intended to regulate fisheries (e.g. non-
transferable quotas may cause discarding of
over-quota catch; species-specific licensing
may cause discard of non-licensed but still
commerciaUy valuable speciesl.

Solutions to bycatches and discards will
be found essentially through improvement in
the selectivity of fishing gear and fishing
methods. Much of the research in this has
been carried out in higher latitudes and is
not readily transferable to multispecies
tropical fisheries, where the tropical shrimp
trawls still produce high rates of bycatch.
Improved use of bycatch either as fishmeal
or human food is also a possibility; however,
this does not address the problem of
mortality of potentially threatened species
(sea turtles, seabirds, cetaceans), nor the

wasteful capture of immature specimens of
harvestable species.

A further problem in the efficient use of
marine resources is post-harvest loss. It is
almost impossible to estimate this accurately,
but FAO believes it to exceed 5 million metric
tons per year li.e. around 5 percent of harvest!.
Most significant are physical losses of dried
fish to insect infestations and loss of fresh fish
through spoilage. These problems are partic-
ularly significant in developing countries.

AQUACULTURE

One major response to the growing crisis in
marine capture fisheries has been the rapid
rise in various forms of aquaculture (Figure
6.5). The latter may be defined as the rearing
in water of organisms (animals, plants and
algae) in a process in which at least one

Marine biodiversity 149

phase of growth is controUed or enhanced
by human action. The animals used are gen-
erally finfishes, mollusks and crustaceans,
although a number of other groups such as
sea squirts ITunicatal, sponges (Poriferal and
sea turtles are cultured in small quantities.
Seaweeds of various kinds are also cultured,
some in large amounts. Most of the species
grown in any quantity are low in the food
chain, being either primary producers, filter-
feeders or finfishes that in their adult stages
are either herbivores or omnivores.

FAO notes that aquaculture is the worlds
fastest growing food production sector,
annual output having increased at an average
rate of some 10 percent in the period 1984-98
(compared with less than 2 percent for
capture fisheries) (see Figure 6.5). In 1999
aquaculture provided around one quarter of

recorded global fisheries production. Of the
total 32.9 million metric tons recorded in
1999, almost 20 million originated inland, and
nearly 13 million were produced in marine
and brackish environments". In 1996. some
7.7 million metric tons of algae and plants
were produced, almost all of this seaweed,
chiefly Japanese kelp Laminaria japonica,
nori Porphyra tenera and wakame Undaria
pinnatifida. The first of these was, in terms of
volume, the most important of all aquaculture
species, with around i.4 million metric tons
produced.

In marine and brackish (usually estuarine)
environments, by far the m.ost important
animal group in terms of volume is the moll-
usks, whose 1997 recorded production of
some 8.6 million metric tons made up more
than 75 percent of all animal production in

150 WORLD ATLAS OF BIODIVERSITY

■^i.

Seaweeds
MoUusks

Diadromous fishes
Marine fishes
Crustaceans
Miscellaneous
aquatic animaLs

1984

Figure 6.5

Marine aquacuUure

production

1998

these environments. Around 50 moUusk
species are produced in significant quantity,
almost all bivalves. As with most culture
systems, production is heavily skewed to a
small number of species, with 65 percent of
production composed of just three: the Pacific
cupped oyster Crassostrea gigas, Japanese
carpet shell Ruditapes philippinarus and
Yesso scallop Pectenyessoensis. The Far East
dominates production, with around 75 percent
of that recorded taking place in China and
most of the remainder in Japan.

Although production of marine crusta-
ceans accounts for only 10 percent or so by
volume of marine and brackish water animal
aquacuUure, it has disproportionately high
economic importance, and is also the sector
that has given rise to most environmental
concerns. Between 1984 and 1998 annual
production grew nearly sixfold, from less than
200 000 to over 1 million metric tons. The
great majority of production takes place in
tropical and subtropical Asia and is dominated
by Penaeus species; globally this genus pro-
duces over 90 percent of aquacuUure crusta-
cean supply by weight. Three species of
Penaeus account for around three quarters of
crustacean production. The giant tiger prawn
P. monodon is the most widely cultivated and
accounts for nearly half; the whiteleg shrimp
P. vannamei is cuUured in the Americas and
accounts for around 15 percent of estimated
global supply (around 70 percent of this

originating in Ecuador); and the fleshy prawn
P. chinensis is cultured in China and currently
accounts for around 10 percent of production,
having declined considerably since the early
1990s when around 200 000 metric tons were
produced annually. Other marine crustaceans
cultivated include other Penaeus species,
some Metapenaeus, and spiny lobsters
Panulirus. These groups, however, make an
insignificant contribution to global supply.

Growth in crustacean aquacuUure has been
fuelled by the high value of the product: the
market in 1996 was estimated to be worth
nearly US$7.5 billion, or around one quarter of
the total value of marine and brackish water
aquacuUure". The great majority of production
takes place in low-income countries - the five
countries producing over 100 000 metric tons
annually being China, Thailand, Indonesia,
Ecuador and Bangladesh - and is aimed mainly
at the export market (primarily to Europe, the
United States and Japan) and to a lesser extent
at the domestic luxury market. Pressure is
high to produce maximum returns on invest-
ment so that increasingly intensive farming
methods are used. These are widely acknow-
ledged to be having adverse social and
environmental impacts in the countries of
production, as well as leading to increasing
difficuUies in maintaining supply, owing to the
spread of major diseases. The last of these
accounts for the major decline in the Chinese

leshy prawn industry during the 1990s.

mpacts include:

loss of mangrove habitat;
abstraction of freshwater;
introduction of pathogens and other
damaging non-native species;
escape of cultured non-native species;
pollution;

diversion of low-quality or cheap fish food
resources (may lead to more efficient use
of bycatches and trash fish but also to
more indiscriminate catch fisheries);
• diversion of effort from other forms of aqua-
culture (notably milkfish Chanos chanos].

The aquarium trade

Up to 2 million people worldwide (about half in
the United States and a quarter in Europe! are

Marine biodiversity 151

■"^'*

thought to keep marine aquanunns, most of
which are stocl<ed with wild-caught species. In
1997 a total of 1 200 metric tons of coral was
traded internationally, with 56 percent impor-
ted by the United States and 15 percent by the
European Union. Approximately half of this was
live coral for aquariums, a tenfold increase on
the amount of live coral traded in the late
1980s^'. Qualitative estimates of trade suggest
that 14-30 million fish may be traded per year,
representing some 1 200 species, about two

thirds of which are from coral reefs. Aquarium
species are typically gathered by local fishers
using live capture techniques or chemical
stupefactants (such as sodium cyanidel which
are non-selective and adversely affect the
health of specimens as well as killing non-
target organisms. Inappropriate shipping
methods and poor husbandry along the supply
chain often cause high mortality among the
fish and invertebrates collected.

While the current impacts of the aquarium
trade remain poorly known, the industry has
considerable potential to contribute to sus-
tainable development. It is relatively low in
volume but very high in value - a kilo of
aquarium fish from one island country was
valued at almost US$500 in 2000, whereas
reef fish harvested for food were worth only
US$6*'. Aquarium species are a high-value
source of income in many coastal com-
munities with limited resources, with the
actual value to the fishers determined largely
by market access. In Fiji many collectors pay

an access fee to the villages to collect on their
reefs, but by selling directly to exporters they
can have incomes many times the national
average. By contrast, in the Philippines there
are many middlemen, and collectors them-
selves typically earn only around US$50 per
month. Targeting mostly non-food species,
aquarium fisheries could in principle provide
an alternative economic activity for low-
income coastal populations and an important
source of foreign exchange for national
economies, as well as an economic incentive
for the sustainable management of reefs.
The application of international certification
schemes may provide an important tool for
achieving this.

OTHER (vlAJOR IMPACTS ON THE MARINE
BIOSPHERE

Alteration of physical habitat
Physical alteration of habitats through human
action chiefly affects coastal and inshore
areas. Impacts here can be severe, although
few attempts have been made to quantify
them on a global basis. Major causes include
coastal development, particularly landfilling
and construction of groynes and jetties, aqua-
culture, dredging of channels for navigational
purposes, extraction of materials such as
sand and coral, stabilization of shorelines,
and destructive fishing methods such as
beam-trawling, use of explosives and muro-
anni (using rocks on ropes to drive fishes into
netsl. Upstream activities, such as defores-
tation and dam construction can greatly alter
sediment loads in rivers, affecting patterns of
sediment deposition in estuarine areas.

Chemical pollution and eutrophication

Human activities have increased inputs of a
huge range of organic and inorganic chem-
icals into marine ecosystems. Such inputs
may enter by direct discharge (e.g. in sewage
outflow pipes], via river and stream outflow,
as land runoff, through the atmosphere or
from seagoing vessels. Because virtually all
such input originates on land, as with most
other human Impacts on marine ecosystems,
areas most affected are coastal and inshore
regions, particularly enclosed or semi-
enclosed water bodies. In oceanic regions.

The aquarium trade targets
mostly non-food species
and many invertebrates
such as crabs, anemones
and shrimp.

152 WORLD ATLAS OF BIODIVERSITY

Worldwide, human
activities have increased
inputs of nitrogen and
phosphorus in rivers and
coastal waters fourfold

mixing of the enormous volume of sea water
generally ensures that inputs become rapidly
diluted.

Major categories of input include nutrients
of various l<inds (e.g. nitrates and nitrites,
phosphates, dissolved organic matter], per-
sistent organic pollutants IPOPsl, including a
range of chlorinated hydrocarbons, and heavy
metals such as cadmium (Cd), copper (Cu|,
mercury IHgl, lead (Pbl, nickel (Nil and zinc
(Zn|. Quantifying these inputs and assessing
their impact is problematic, particularly
because many occur naturally in sea water

Many POPs and heavy metals can act as
toxins above certain concentrations, inducing
mortality or morbidity or impairing repro-
ductive success, particularly in cases where
they become increasingly concentrated to-
wards the top of food chains. Their overall
impact on marine ecosystems remains un-
certain. More easily observable is the impact

of eutrophication resulting from the increased
input of organic and inorganic nutrients
(particularly nitrogen and phosphorus) into
coastal waters, mainly through fertilizer
runoff and sewage disposal. It is believed that
human intervention has increased river inputs
of nitrogen and phosphorus worldwide into
coastal areas by more than fourfold over
background levels. These inputs lead to
increases in productivity in coastal waters,
often in the form of algal blooms. These
blooms may themselves be noxious; they also
typically cause the euphotic zone to reduce
in vertical extent and are implicated in the

development of hypoxic (low dissolved oxygen
concentration] and anoxic (zero dissolved
oxygen] zones. A shallowing of the euphotic
zone may cause die-off of photosynthesizing
benthic algae in shallow-water areas. This
has occurred, for example, in the Black Sea
where the euphotic zone had decreased from
50-60 m vertical extent in the early 1960s to
around 35 m by 1990, leading to a decrease of
up to 95 percent in living biomass of benthic
macrophytic algae such as Phyllophora,
formerly an important harvested resource.

Hypoxia and anoxia result from the
activities of oxygen-respiring bacteria below
the euphotic zone feeding on accumulated
dead algae and other organisms and waste
matter raining down from above. Hypoxia re-
sults in the emigration of mobile aerobic
species and mortality of sedentary ones. This
may have catastrophic impact on local
fisheries. Most hypoxic zones vary in extent
through the year and from year to year and
some are only seasonal, disappearing when
winter mixing causes re-oxygenation of
bottom waters. They may be very extensive -
the hypoxic zone to the west of the Mississippi
delta covered some 16 000 km' in 1997, having
covered some 9 000 km' in 1989. Over 50 such
zones have been identified worldwide to date;
some appear to be at least in part induced by
natural phenomena while others are believed
entirely anthropogenic.

Invasions of exotic species

As on land, the breakdown of biogeographic
barriers in the sea appears to be having a
major, and increasing, impact on marine
ecosystems. The chief source is the deliberate
or accidental translocation of organisms, but
the construction of marine corridors between
previously isolated areas has had a major
impact on geographically restricted areas.
These factors have resulted in a rapid spread
of alien species in all the worlds oceans.

Translocation can result from deliberate
introduction of harvestable species, acciden-
tal escapes from aquacuUure and aquarium
operations, transport in ballast water of ships
and release of fouling organisms that adhere
to the hulls of ships and boats. The extent of
introductions of this kind has yet to be fully

Marine biodiversity 153

assessed but is certainly large. In many cases
(see Box 6.21 the introduced species appear to
be having a major impact on native biota,
although In general it is difficult to separate
the effects of a particular species from
general ecosystem deterioration.

Deliberate introductions include the plant-
ing of mangroves and Nips palms along
coastlines, and the cultivation of fish, crus-
taceans, mollusks and algae In many coastal
regions. Atlantic salmon that have escaped
from aquaculture are reportedly affecting vs/ild
stocks In the northeast Pacific, and their
pathogens have themselves moved into the
wild populations of closely related species.
Ballast water is commonly pumped Into the
hold of ships as a means of controlling balance
and position In the water, and is liable to be
flushed out far distant from where It was taken
on. It has been estimated that on any one day
the ballast water of the world's ocean fleets
contains around 10 000 different species".

The principal artificial corridors are [he
Suez Canal lopened In 1869) and the Panama
Canal lopened In 19U1. The 165-km long Suez
Canal Is a continuous seawater channel with
a water level at the Red Sea end some 1.2 m
higher than at the Mediterranean end, leading
to a constant northward flow of water It Is
estimated that to date some AOO-500 marine
species have migrated through the canal (so-
called Lessepsian migrants'", after Ferdinand
de Lesseps who planned the canall and
established themselves In the Mediterranean,
while a far smaller number have moved in the
other direction. New species are believed to
arrive in the Mediterranean at the rate of four
to five annually. Because the Panama Canal
has a separate freshwater section, some 25 m
above sea level, migration through It has been
limited to date.

Global climate change

Human-induced climate change is liable to
Impact directly on marine and coastal areas
by warming (particularly of the surface
layers], by sea-level rise (associated both with
thermal expansion and the melting of terres-
trial Ice caps and glaciers!, and through
change In the gases dissolved in surface
waters. These impacts are well understood.

and measurable changes are already appar-
ent. A more complex array of secondary
changes may also occur, including changes
to ocean stratification and surface mixing,
changes to patterns of surface current, and
perhaps to global systems such as the El Nino
Southern Oscillation.

Tropical coral reefs appear particularly
sensitive to temperature change. Reef-
building corals are adapted to stable thermal
conditions and In most areas appear to be
growing close to their upper temperature
limit. Temperatures tittle more than 1°C
above the normal maximum for a period of
a few weeks are sufficient to drive a stress
response known as coral bleaching. During
a particularly strong El Nino event in 1998,
warmer waters around the Seychelles and the
Maldives Induced a bleaching event In which
60-90 percent of all corals in the area died,
equivalent to 5 percent of the world coral reef
area. Although this event was linked to an
extreme climatic perturbation, it occurred
at a time of rising global temperatures and
provides an indication of the impact of
potential future climate change'". More subtle
changes associated with the gradually
changing background conditions, particularly
temperatures, have been recorded in the

Box 6.2 Marine introductions

Leidy's comb jelly Menopsis leidyi was introduced from the American Atlantic into the waters
of the Black Sea in 19B2, presumably In ballast water Unchallenged by natural predators,
this species proliferated to a 1988 peak estimated at around 1 000 million metric tons wet
weight (about 95 percent of the entire wet weight biomass in the Black Seal. The species
depleted the natural zooplankton stocks, with subsequent algal blooms and decline of the
fishing industry in the Black Sea".

. The Asian clam Potamocorbula amurensis spread through the northern San Francisco
' estuary (United StatesI following its introduction, possibly in ballast water, in 1986. The
species reaches high density, up to 2 000 Individuals per m', and has caused sharp declines
In the abundance and extent of several plankton species; its impact on fisheries is not yet
clear With more than 200 Introduced species, this bay may be the most Invaded aquatic
habitat In North America.

The green alga Cauterpa taxifolia is thought to have escaped from an aquanum In the
western Mediterranean and is spreading rapidly In the coastal waters of Spain, France and
Italy, with severe Impact on the native seagrass beds and on coastal fisheries". It has
recently been reported In the coastal waters of California^'.

154 WORLD ATLAS OF BIODIVERSITY

*K<

Much less readily observed,
marine species are in
general more difficult to
monitor and assess tfian
terrestrial ones.

distribution of pelagic seabirds along the
Californian coast"; the faunal composition of
intertidal communities'"; and penguin distri-
bution in the Antarctic".

The mean sea level has risen 18 cm during
the past 100 years and further increases
could have a severe impact on coastal
communities. Rising sea levels will lead to
the inundation of some coastal lands, whilst
in many other areas they will alter patterns
of coastal erosion, and they may increase
groundwater salinization. While many inter-
tidal habitats are highly adaptable, the
growing human presence in most coastal
areas will prevent the natural migration of
these habitats, leading to overall losses of
saltmarsh, mangrove, or even beach and
rocky shore habitats.

The biomass of the world's oceans is very
low compared with terrestrial environments,
but because of the rapid turnover in oceanic
carbon cycles, marine phytoplankton Icyano-
bacteria and algael play an important role
in removing dissolved carbon dioxide from
solution, and are intermediates in the trans-
port of organic carbon to the deep ocean.
Once in the deep ocean, this carbon is effec-
tively removed from exchange with the
atmosphere tor millennia. In this way, marine
photosynthesizers help to buffer the rising
concentration of carbon dioxide in the
atmosphere, but this service, and marine
primary productivity, may be affected if ocean

temperatures rise (warmer waters hold
less carbon dioxide in solution] and if the
broad patterns of ocean circulation change
significantly". It is thought that increasing
atmospheric temperatures may affect the
generation of cold, oxygen-nch bottom waters
beneath Arctic and Antarctic ice sheets, with
major implications for deep-sea biota and
for global patterns of seawater circulation, in
particular the Great Conveyor, driven by
bottom water generated in the North Atlantic.

THE CURRENT STATUS OF MARINE
BIODIVERSITY

Because they are usually much less readily
observed, marine species are in general
much more difficult to monitor and assess
than terrestrial ones. Assessment is based
on sampling and, in the case of harvested
species, often on the basis of catch rates,
although the latter may vary in response to
a wide range of factors in addition to changes
in the population of the species concerned.
An exception lies with those groups such as
pinnipeds, sea turtles and seabirds that nest or
breed on land. Because many of these tend to be
colonial species and because they tend to breed
in open habitats (beaches, cliff tops, ice sheets),
they may be easier to monitor than many other
species, either terrestrial or aquatic. In the case
of large, commercially valuable fish stocks,
monitoring at large scale has in some cases
been carried out tor many years, so that
estimates of the stock level are obtainable.

Threatened and extinct species

The only major marine species groups
(classes or above) that have been com-
prehensively assessed in terms of threatened
species status to date are mammals and
birds. In addition, sea turtles and a number of
fish families and genera (e.g. the sturgeons
in the order Acipenseritormes and the sea-
horses Hippocampus in the order Syngnathi-
formes) have also been assessed. Other
threatened marine species have been identi-
fied on more of an ad hoc basis. Data are
summarized in Table 6.12.

Relatively speaking, far fewer marine
species are known to have become extinct
since 1600 than either terrestrial or fresh-

Marine biodiversity 155

^'h

water ones. Cataloged extinctions comprise
two marine mammals (the Caribbean monk
seal Monachus tropicaiis and Steller's sea
cow Hydrodamalis gigas] and five seabirds
(three island petrels, Pallas's cormorant
Phalacrocorax perspicillatus and the great
auk Alca impennis]. In addition five coastal
or island duck species have disappeared at
various times from the late 17th century
onwards; however there is In most cases
insufficient information to determine whether
these species were predominantly marine
or terrestrial.

As a gross generalization, marine species
appear to be somewhat less extinction prone
as a result of mankind's activities than fresh-
water or terrestrial ones. There are arguably
two main reasons for this. First, because of
the size of the world ocean and the fact that
people do not actually live In it, the marine
biosphere remains as a whole considerably
more buffered from human Intervention than
terrestrial and inland water areas. Second,
marine species on the whole appear to be
more widespread than terrestrial or inland
water ones. In the open ocean, there are vast
areas with apparently similar habitat con-
ditions and there are few barriers to dispersal
so that many species have circumglobal
distributions. In addition, many forms that as
adults are sessile (e.g. sponges and coralsl or
sedentary (many mollusks and crustaceans!
have planktonic larvae that are often widely
dispersed In water currents. For this reason,
many coral reef species, for example, are
found in suitable habitat throughout the Indo-
Paclfic region. In addition, many of the most
heavily exploited fish species have high
fecundity (In the case of some tunas amoun-
ting to several million eggs In a single
spawning), so that they have at least poten-
tially high population growth rates unparallel-
ed in terrestrial vertebrates.

There are of course significant exceptions to
all these. Coastal regions In many parts of the
world, and enclosed or semi-enclosed marine
areas such as the Baltic, Black and Yellow
Seas, are often under intense pressure from a
range of human activities. A number of marine
species do appear to have restricted ranges
(e.g. the Hawaiian coral reefs are relatively rich

in species found nowhere else while many
southern hemisphere seabirds are apparently
confined to a small number of breeding sites)
and significant numbers have low or very low
reproductive rates (many chondrichthyme
fishes, marine mammals and seabirds).

Marine species which are
easily exploitable or have
high economic value may
suffer catastrophic declines
if exploitation is not strictly
controlled.

Until recently by far the most important
human activity affecting marine species was
uncontrolled exploitation. Where species are
either easily exploitable or are highly sought-
after [i.e. have high unit value], or both, they
may suffer catastrophic declines. This Is the
case with sea turtles and a number of marine
mammals and birds that are or have been
harvested principally at their terrestrial
breeding sites (which are often colonial), as
well as with the great whales and the dugong,
which although strictly marine are air-
breathing and therefore spend some time at
the sea surface (when they may be spotted).
Most of these species have relatively low
reproductive rates, so that even if they are
ultimately afforded protection population
recovery rates may be very slow.

A recent synthesis of a wide range of
information, including paleoecologlcal and
archeologlcal data relating to early human

156 WORLD ATLAS OF BIODIVERSITY

Table 6.12

Taxonomic distribution
and status of threatened
marine animals

Note: Only the birds and
mammals have been
comprehensively assessed for
species at risk; numbers refer
to species (units such as
subspecies and geographic
populations appear in the Red
List database but these are
not tabulated here).

Source: Status categories from 2000
Red List database, www.redlist.org
(accessed February 20021.

communities, suggests that overfishmg of
larger vertebrates and mollusks is charac-
teristic of indigenous and colonial human
use of coastal ecosystems, and the first of
what is typically a series of impacts'-,
tvlassive losses in biomass and abundance
appear to have occurred, on a scale largely
unsuspected, and seemingly amounting to
the loss of entire trophic levels of consumer
organisms, with radical consequences for
ecosystem status. Overfishing is likely to be
followed by impacts of pollution, mechanical
habitat loss, introduced species and climate

change. Loss of filter-feeding organisms
that maintain water quality is liable to be
followed by eutrophication, hypoxia and
disease, as exemplified by conditions in
Chesapeake Bay following the collapse of
the oyster fishery in the early 20th century.
Synergistic interactions of this kind are
making the effective, long-term manage-
ment of marine resources one of the major -
and most intractable - problems currently
facing humankind.

Land-breeding species may also be sus-
ceptible to other threats, such as predation,

Phylum ^^^^^^^^1

F ' Common name 'Critically

'Endangered'

'Vulnerable'

_ class or order ^^^^^^H

|^^_ endangered'

Cnidaria

Anthozoa

Stony corals

Mollusca

Bivalvia

Bivalves

Gastropoda

Gastropods 1

Craniata- fishes

Carcharhiniformes

Ground sharl<s 1

Hexanchiformes

Cow/ sharks

Lamniformes

Mackerel sharks

Orectolobiformes

Carpet sharks

Pristiformes

Saw/fishes 2

Rajiformes

Rays

Rhinobatiformes

Giiitar fishes 1

Squaliformes

Dogfishes and
sleeper sharks

Squatiniformes

Angel sharks

Coelacanthiformes

Coelacanths 1

Acipensenformes

Sturgeons 2

Batrachoidiformes

Toadfishes

Clupeiformes

Herrings and anchovies

Gadiformes

Cods, hakes, rattails 1

Gasterosteiformes

Sticklebacks

Lophiiformes

Anglerfishes, etc. 1

Ophidiiformes

Pearlfishes, cusk-eels, bnatulas

Perciformes

Perches, etc. 7

Pleuronectiformes

Plaice, flounders, soles

Salmoniformes

Salmonids 1

Scorpaeniformes

Gurnards, scorpionfishes, etc. 1

Siluriformes

Catfishes

Syngnathiformes

Pipefishes, seahorses, etc. 1

Tetraodontiformes

Triggerfishes, etc.

Marine biodiversity 157

coastal development and pollution. It is note-
worthy in this context that the family
Procellariidae contains nearly three times as
many threatened species (36 out of 115
species, or 28 percent! as the average bird
family, in Vi^hich 11 percent of species are
threatened, and nearly six times as many
critically endangered species as would be
expected at random. It is almost certainly the
tendency of these birds to nest on islands,
whose biotas have in general suffered enor-
mously more from mankind's influence in the
past few centuries (see Chapter A], rather

than their seagoing habits that has led
to this.

For truly marine species (chiefly finfishes
and invertebrate animals! the situation appears
somewhat different. Even when these have
been exploited to the point of stock collapse,
as has occurred for example with the cod
Gadus morhua stocks off Newfoundland in the
North Atlantic, the species concerned do not
appear to have become imminently threatened
with biological extinction. This is in part
because once stocks are reduced below a
certain level it is often no longer economically

phylum and

Family

Common name

'Critically

'Endangered'

'Vulnerable'

■ class or order ^^^

endangered'

Cranlata - Reptilla

Squamata

Iguanidae

Iguanas

Chelcnia

Dermochelyidae

Leathery turtle

Cheloniidae

Sea turtles

Craniata - Aves

Anseriformes

Anatldae

Ducks

Cfiaradriiformes

Alcldae

Auks, puffins

Charadriidae
Laridae

Plovers

Gulls, terns, skuas,
auks, skimmers

2
1

Ciconiiformes

Ardeidae

Egrets, herons

Pelecaniformes

Fregatidae
Pelecanidae

Frigateblrds
Pelicans

1
1

Phalacrocoracldae

Cormorants and shags

Sulidae

Gannetsand boobies

Procellarilformes

DIomedeldae
Hydrobatldae

Albatross
Strom petrels

12
1

•

Pelecanoidldae

Diving petrel

Procellariidae

Petrels, shearwaters

Sptienisciformes

Spheniscidae

Penguins

Craniata - Mammalia

Carnivora

Muslelidae

Otters, etc.

Otarlidae

Eared seals

Cetacea

Phocidae
Balaenidae

Earless seals
Right whales

1
1

■§

Balaenopteridae

Rorquals

Delphlnidae

Dolphins

(i^onodontidae

Beluga

■

Phocoenidae
Physeteridae

Porpoises
Sperm whales

1
1

Sirenia

Dugongidae

Dugong

Trichechidae

Manatees

158 WORLD ATLAS OF BIODIVERSITY

*i<m

120

100

1970

1975

viable to continue harvesting them. Generally,
the residual population at this stage is still
large enough to allow recovery if harvesting
ceases, particularly in the case of species with
high fecundity and therefore high potential
intrinsic rates of increase. Exceptions to this
are species that have low fecundity, partic-
ularly if they also have a long period to
maturity, with limited ranges, and which may
either have high unit value or be caught as
bycatches.

In the case of bycatches, because the
fishery is not directed at the species
concerned, its intensity will not decrease as
population levels decrease so that it may
theoretically be possible at least locally to
extirpate species, particularly if they are
habitually caught before they reach maturity.
Examples include several sawfish species

1980

1985

1990

1995

1999

Figure 6.6

Marine population trends

Note: A simplified
representation of the average
population change in a
sample of 217 marine
species, see text.

(family PrIstidaeL These are large, slow-
growing, predominantly inshore species that
give birth to relatively small numbers of live
young. Population densities appear to be
naturally low and animals are widely caught
as bycatch in inshore fisheries before they
are large enough to reproduce. As a result
five species are classified as 'endangered' and
two as 'critically endangered'. In addition, it is
possible that trophic shifts may occur when
populations of some species are severely
reduced, inhibiting recovery of these popu-
lations when exploitation ceases. This has
been suggested in the case of some great

whale populations that have not apparently
recovered as rapidly as projected following the
cessation of their harvest.

The marine living planet Index

An impression of the overall trend in a large
sample of species for which population
indicators are available can be derived from
the WWF living planet index"'. This approach
IS designed to represent the change in the
average species' in the sample from one five-
year interval to the next, starting in 1970.
The marine sample represents 217 aquatic
and coastal species of mammals, birds,
reptiles and fishes, and the overall trend is for
a significant decline In population levels over
the last three decades of the 20th century
(Figure 6.61. The sample is dominated by the
stocks and species that humans have an
interest in monitoring, most of the fishes
among these being of commercial importance
as a fisheries resource. These should also
be stocks that humans have an Interest In
managing as well as possible. That the index
has declined In every five-year interval since
1970 is evidence that such management Is
failing, as confirmed by the picture painted
above of global marine capture fisheries.

Assessing the status of marine and coastal
ecosystems

Threatened species inventories and the marine
living planet index can give a very general
overall Impression of the status of marine
biodiversity. Assessing marine ecosystem
health' is much more problematic. However,
snapshots can be obtained from examining
particular ecosystems, such as mangroves
and coral reefs". In the former, an overall
assessment can be made on the basis of the
area destroyed or severely degraded. In the
latter areal measures are more problematic. In
part because reef extent is much more difficult
to measure than mangrove extent and, of
greater importance, because the vast majority
of a reef Is composed of non-living calcareous
deposits. Measures of the change in extent of
these give little insight Into the state of the
living component of the reef. For this reason
other measures, such as estimates of inci-
dence of coral disease, may be feasible.

Marine biodiversity 159

REFERENCES

1 Longhurst, A.R, and Pauly, D. 1987. Ecology of tropical oceans. ICLARM (International
Center for Living Aquatic Resources Management! Contribution No. 389. Academic Press
Inc.. California.

2 Longhurst. A.R. 1995. Seasonal cycles of pelagic production and consumption. Progress
in Oceanography 36: 77 -Ml.

3 Longhurst, A.R. 1998. Ecological geography of the sea. Academic Press, San Diego.
A Sherman. K. and Busch. DA. 1995. Assessment and monitoring of large marine

ecosystems. In: Rapport, D.J., Guadet, C.L. and Calow, P. (edsl. Evaluating and
monitoring the health of large-scale ecosystems. Springer-Verlag. Berlin, published in
cooperation with NATO Scientific Affairs Division. NATO Advanced Science Institutes
Series I: Global Environmental Change. 28: 385-A30.

5 Available online at http://www.edc.uri.edu/lme/default.htm (accessed February 20021.

6 Olson. D.M. and Dinerstein, E. 1998. The Global 200: A representation approach to
conserving the earth's distinctive ecoregions. Conservation Biology }2: 502-515.

7 Vaulot. D. 1 995. Marine biodiversity at the micron scale. International Marine Science
News/efter 75/76. United Nations Educational. Scientific and Cultural Organization.
Paris.

8 Rex. M.A., Stuart, C.T. and Coyne, G. 2000. Latitudinal gradients of species richness in
the deep-sea benthos of the North Atlantic. Proceedings of the National Academy of
Sciences 97: A082-i08b.

9 Adey, W.H. and Steneck, R.S. 2001. Thermogeography over time creates biogeographic
regions: A temperature/space/time-integrated model and an abundance-weighted test
for benthic marine algae. Journal of Phycology 37 : 677-698.

10 Roberts, CM. et ai 2002. Marine biodiversity hotspots and conservation priorities for
tropical reefs. Sdence 295: 1280-1284.

11 Nybakken, J. 1993. Marine biology: An ecological approach. 3rd edition. HarperCollins
College Publishers, New York.

12 World Conservation Monitoring Centre 1992. Groombridge, B. (ed.l Global biodiversity:
Status of the Earth's living resources. Chapman and Hall, London.

13 Finlayson. M. and Moser, M. leds) 1991. Wetlands. Facts on File Ltd, Oxford.

U Woodroffe, CD. and Grindrod, J. 1991. Mangrove biogeography: The role of Quarternary
environmental sea-level change. Journal of Biogeography 18: 479-492.

15 Tomtinson, P.B. 1986. The botany of mangroves. Cambridge University Press, Cambridge.

16 Spalding. M.D., Blasco, R and Field, CD. (edsl 1997. World mangrove atlas. International
Society for Mangrove Ecosystems. Okinawa.

17 Lasserre. P. 1995. Coastal and marine biodiversity International Marine Science
Newsletter 7b/76: 13-14.

18 Risser, P.O. 1986. Spatial and temporal variability of biospheric and geospheric
processes: Research needed to determine interactions with global environmental
change. Report of a workshop. International Council of Scientific Unions Press. Paris.

19 Ducrotoy. J.R. Elliott. M. and de Jonge, V.N. 2000. The North Sea. In: Sheppard, C (ed.l Seas
at the millennium: An environmental evaluation. Vol. I. pp. 43-64. Elsevier Science.
Amsterdam.

20 Lewis. J.R. 1977. Rocky foreshores. In: Barnes. R.S.K. (ed.l The coastline, pp. 147-158.
Wiley. London.

21 Wilson. S.C 2000. Northwest Arabian Sea and Gulf of Oman. In: Sheppard. C (ed.l Seas
at the millennium: An environmental evaluation. Vol. II. pp. 17-34. Elsevier Science.
Amsterdam.

22 Phillips, R.C. and Mehez, E.G. 1988. Seagrasses. Smithsonian Contributions to the
Marine Sciences 3i: 104.

160 WORLD ATLAS OF BIODIVERSITY

23 Hemminga. MA. and Duarte. CM. 2000. Seagrass ecology. Cambridge University
Press, Cambridge.

24 Short, F.T. and Wyllie-Echeverria, S. 2000. Global seagrass declines and effects of
climate change. In: Sheppard, C. led.l Seas at the millennium: An environmental
evaluation, Vol. Ill: Global issues and processes. Elsevier Science, Amsterdam.

25 Wilkinson, C.R. and Buddemeier, R.W. 1994. Global climate change and coral reefs:
Implications for people and reefs. Report of the UNEP-IOC-ASPEI-IUCN Global Task
Team on the Implications of Climate Change on Coral Reefs.

26 Hulm, P. and Pernetta, J.C. 1993. Reefs at risk: Coral reefs, human use and global
climate change. A programme of action. lUCN-the World Conservation Union, Gland.

27 Veron, J. 2000. Corals of the world. 3 vols. Australian Institute of Marine Science,
Queensland, Australia.

28 Cairns, S.D., Hoeksema, S.B. and van der Land, J. 1998. List of extant stony corals.
Available online at http://www.nodc.noaa.gov/col/projects/coral/hardcoral/
Hardcoralmain.html (accessed February 20021.

29 Mortensen, P.B. et al. 1995. Deep water bioherms of the scleractlnian coral Lophelia
pertusa at 64°N on the Norwegian shelf: Structure and associated megafauna. Sarsia
80: U5-158.

30 Rogers, A.D. 1 999. The biology of Lophelia pertusa and other deep-water reef-forming
corals and impacts from human activities. International Review of Hydrobiology8i[A]:
315-406.

31 Fossa, J.H. et al. 1999. Effects of bottom trawling on Lophelia deep water corals In
Norway. Poster presented to the ICES (International Council for Exploration of the SeasI
workshop on the ecosystem effects of fishing, February 1999.

32 Koslow, J. A. and Gowlett-Holmes, K. 1998. The seamount fauna of southern Australia:
Benthic communities, their conservation and impacts of trawling. Report to
Environment Australia and the Fisheries Research Development Corporation, FRDC
Project 95/058.

33 Gjosaeter, J. and Kawaguchi, K. 1 980. A review of the world resources of mesopelagic
fish. FAQ Fisheries Technical Paper 193: 1-151.

34 Rowe, G.T 1 983. Blomass and production of the deep-sea macrobenthos. In: Rowe, G.T
(ed.l Deep-sea biology, Vol. 8 The sea. pp. 453-472. John Wiley and Sons Inc., New York.

35 Hessler, R.R. and Sanders, H.L. 1967. Faunal diversity in the deep-sea. Deep-Sea
Research 14: 65-78.

36 Angel, M.V. 1991. Biodiversity In the deep ocean. A working document for the UK
Overseas Development Administration. Unpublished ms.

37 Grassle, J.F 1991. Deep-sea benthic biodiversity. Bioscience 41(71.

38 Grassle, J.F and Maciolek, N.J. 1992. Deep-sea species richness: Regional and local
estimates from quantitative bottom samples. American Naturalist 139: 313-341.

39 May, R.M. 1992. Bottoms up for the oceans. Nature 357: 278-279.

40 Angel, M.V. 1982. Ocean trench conservation. Commission on Ecology Papers No. 1.
lUCN-the World Conservation Union, Gland.

41 Gage, J.G. and Tyler, PA. 1991. Deep-sea biology: A natural history of organisms at the
deep-sea floor. Cambridge University Press, Cambridge.

42 Grassle, J.F 1986. The ecology of deep-sea hydrothermal vent communities. Advances in
Marine BiologyMol 23. Academic Press.

43 Hecker, B. 1985. Fauna from a cold sulphur-seep In the Gulf of Mexico: Comparison with
hydrothermal vent communities and evolutionary Implications. Bulletin of the Biological
Society of Washington 6: 465-473.

44 Grassle, J.F 1989. Species diversity in deep-sea communities. TREE [Trends in Ecology
and EvolutionI iiM.

Marine biodiversity 161

^5 UNCED 1992. The global partnership for environment and development A guide to
Agenda 21 . United Nations Conference on Environment and Development, Geneva.

Ab Committee on Biological Diversity in Marine Systems 1995. Understanding marine

biodiversity: A research agenda for the nation. National Academy Press, Washington DC.

^7 FAQ 1990. Review of the state of world fishery resources. Marine Resources Service, Food
and Agriculture Organization of the United Nations, Rome.

48 FAQ 1999. The state of world fisheries and aquacutture /998. Food and Agriculture
Organization of the United Nations, Rome.

49 FAQ 2000. The state of world fisheries and aquacutture 2000. Fisheries Department, Food
and Agriculture Organization of the United Nations, Rome. Available online at
http://www.fao.org/DOCREP/003/X8002E/X8002EOG.htm (accessed April 20021.

50 Watson, R. and Pauly, D. 2001. Systematic distortions in world fisheries catch trends.
Nature iU: 534-536.

51 Green, E. and Shirley, F 1999. The global trade in coral. World Conservation Press, Cambridge.

52 Edwards, A.J. and Shepherd, A.D. 1992. Environmental implications of aquarium fish collection
in the Maldives, with proposals for regulation. Environmental Conservation 19: 61-72.

53 Carlton, J.T. 1999. Invasions in the sea: Six centuries of re-organising the Earth's marine
life. In: Sandlund, O.T., Schei, P.J. and Viken, A. ledsl. Invasive species and biodiversity
management, pp. 195-212. Kluwer Academic Publishers, Netherlands.

54 Goldschmid, A. 1999. Essay about the phenomenon of Lessepsian migration. Colloquial
Meeting of Marine Biology I, Salzburg 1999. Available online at
http://www.sbg.ac.at/ipk/avstudio/pierofun/lm/Lesseps.htm (accessed April 2002).

55 GESAMP 1997. Opportunistic settlers and the problem of the ctenophore Mnem/ops/s
leidyi invasion in the Black Sea. IMO/FAO/UNESCO-IOC/WMO/WHO/IAEA/UN/UNEP Joint
Group of Experts on the Scientific Aspects of Marine Environmental Protection. Rep. Stud.
GESAMP. 58: 84.

56 Anon 1997. t^edwaves2A: 11-13. News Bulletin of the Mediterranean Action Plan, Athens.

57 Dalton, R. 2000. Researchers criticize response to killer algae. Nature 406: 447.

58 Spalding, M., Teleki, K. and Spencer, T. 2001. Climate change and coral bleaching, (n:
Green, E. ef al. (eds). Impacts of climate change on wildlife, pp. 40-41 .

59 Veil, R.R. ef al. 1997. Apex marine predator declines 90% in association with changing
oceanic climate. Global Change Biology 3: 23-28.

Sagarin, R.D. ef al. 1999. Climate-related changes in an mtertidal community over short
and long time scales. Ecological Monographs 69: 465-490.

Ainley, D., Wilson, P. and Eraser, W.R. 2001. Effects of climate change on Antarctic sea ice
and penguins. In: Green, E. et al. ledsl. Impacts of climate change on wildlife, pp. 26-27.

62 Balino, B.M., Fasham, M.J.R. and Bowles, M.C. 2001. Ocean biogeochemistry and global
change: JGOFS research highlights 1988-2000. International Geosphere-Biosphere
Programme Science No. 2. International Council of Scientific Unions, Stockholm.

63 Jackson, J.B. et al. 2001 . Historical overfishing and the recent collapse of coastal
ecosystems. Science 293: 629-638.

64 Loh, J. led.) 2000. Living planet report 2000. WWF - World Wide Fund for Nature, Gland.

65 Couper, A.D. 1993. /"/mes af/as 0^ fhe oceans. Times Books.

66 Nelson, J.S. 1994. Fishes of the world. John Wiley and Sons inc.. New York.

67 Croxall, J.P, Evans, RG.H. and Schreiber, R.W. 1994. Sfafus and conservation of the world's
seabirds. ICBP Technical Publication No. 2. ICBP (now BirdLife International], Cambridge.

68 Sibley, C.G. and Monroe, Jr, B.L. 1990. Distribution and taxonomy of birds of the world.
Yale University Press, New Haven and London.

69 Duke, N.C. 1992. Mangrove floristics and biogeography In: Robertson, A.I. and Alongi,
D.M. (edsl. Tropical mangrove ecosystems. Coastal and estuarine Studies 41, p. 329.
American Geophysical Union, Washington DC.

162 WORLD ATLAS OF BIODIVERSITY

70 Spalding, M.D.. Ravilious, C, and Green, E. 200L World atlas of coral reefs. University of
California Press, Berkeley.

71 lUCN 2000. 2000 lUCN Red List database available online at v^ww/.redlist.org (accessed
February 2002).

72 Compagno, L.J.V. 198^. Sfiarks of the world. An annotated and illustrated catalogue of
sfiark species known to date. FAQ Fisheries Synopsis No. 125 Vol. A (Parts 1-21.

73 Spalding, M.D. 1998. Biodiversity patterns in coral reefs and mangrove forests: Global and
local scales. Unpublisfied PhD thesis, University of Cambridge.

74 Veron, J.E.N. 1995. Corals in space and time: Tf)e biogeography and evolution of tlie
Scleractinia. University of New South Wales Press.

75 FAQ 2002. http://www.fao.Org/DOCREP/003/X8002E/x8002e04.htm#Pl_6 (accessed April
2002).

Inland water biodiversity 163

/ Inland water bi

odiversity

INLAND WATERS h4AKE UP A MINUTE PROPORTION Imuch less than a hundredth of
1 percent! of the world's water resource. Despite this, they encompass a wide range
of habitat types and contain a disproportionately high fraction of the world's biodiversity.
Freshwater is also a vital resource for human survival. In consequence, Inland water
ecosystems are placed under many, often conflicting, pressures, with Increasingly adverse
consequences for their biodiversity. There are indications that, overall, a higher proportion
of inland water species are In decline than marine or terrestrial forms.

INLAND WATERS

The hydrosphere is estimated to contain about
1 386 million cubic kilometers Ikml of water,
almost all of which (97.5 percent) is saline
water making up the world ocean, leaving
some 35 million km' 12.5 percent) of fresh-
water. A major proportion (about 69 percent]
of this freshwater is locked up in the form of
ice and permanent snow, where it is unavail-
able to living organisms. The Earths liquid
freshwater is mostly in subterranean ground-
waters, with a small proportion in soils and
wetlands, and the smallest proportion of all -
about 0.01 million km^ or 0.3 percent of all
freshwater - makes up the world's lakes and
rivers on which inland water biodiversity
depends (Table 7.1).

There are very large regional differences in
the concentrations of water in all its forms
(e.g. about twice as much atmospheric water
in equatorial as in temperate latitudes], and in
the occurrence of different types of inland
waters (e.g. South America has an enormous
concentration of river water but few large
lakes, while the converse is true for Africa).
Over the oceans, evaporation exceeds input
from rivers and rainfall, and over land precipi-
tation exceeds evaporation. This excess on
land amounts to about ^3 000 km^ annually
and this represents the global runoff that
replenishes the world's rivers, lakes and
marine waters, and which humans draw on.

together with groundwater, to meet their
domestic, agricultural and industrial needs.

Chapter 5, on terrestrial ecosystems,
outlines some of the principal differences
between terrestrial and aquatic environments
as they determine the living conditions of
organisms in them, (n general, freshwater sys-
tems and the organisms within them are far
more strongly affected by daily and seasonal
changes as a result of weather patterns and
climate conditions than are marine aquatic
environments, in parallel with change in the
broader terrestrial environment. For example,
upland streams may receive an influx of cold
meltwater in spring, and be subject to strong
insolation during midsummer; the volume,
speed and transparency of river water is
liable to change radically following rainfall in

Table 7.1

Components of the
hydrosphere

Source; Anon . Shiklomanov

Area

% total

Volume

% total

1
%

m^l^

Imilllon km')

area

(million km^l

water

freshwater

Earth's surface

510

Land

World ocean

361

1351

97.5

Freshwater

34.65

2.5

100

Ice and permanent

snow 16

23.8

1.76

68.9

Groundwater

10.4

0.77

29.9

Wetlands, soil viiater

permafrost 2.6

0.3

0.02

0,9

Lakes and rivers

1.5

0,01

0.0007

0.3

164 WORLD ATLAS OF BIODIVERSITY

Box 7.1 Saline and soda lakes

Some coastal or high altitude lakes and lagoons combine high concentrations of dissolved
chemicals with high water temperatures, sometimes as high as 70°C. Few kinds of species
thrive in these harsh alkaline or saline conditions. Communities typically include
cyanobacteria, diatoms, a few small invertebrate animals such as brine shrimps [Artemia],
with flamingos often being the only large organisms present. Flamingos, of which five living
species are recognized, are filter-feeders, with the bill and mouthparts specialized to extract
small particles from water All share the same basic morphology, but tend to extract prey of
different sizes: e.g. in sub-Saharan Africa the lesser flamingo feeds almost exclusively on
cyanobacteria and other microorganisms, while the greater flamingo (often found in the
same lakes), feeds mainly on small macroscopic invertebrates, such as brine shrimps and
other crustaceans. Fishes are often absent from lakes where flamingos are abundant, and
flamingos are rarely abundant where fishes are common. This may be because the two
compete with each other for their food supply but, in the inhospitable lakes where they
abound, flamingos have no real competitors. Their early adaptation to an extreme
environment that no other large animal was capable of exploiting may explain why they have
survived relatively unchanged in morphology for so long (fossils are known from the Eocene,
about 50 million years before the present).

ttie catctiment basin. The soil and surface
geology of watershed areas have a strong
influence on the chemical composition of both
river and lake waters and, for example, buffer
or reinforce the effect of acid rain.

Inland water habitats

Although the terms 'inland water' and
'freshwater' are often used more or less
interchangeably they are not equivalent.
A considerable number of inland waters are
saline, some much more so than sea water.
Conversely, waters of the deltaic regions of
some major river systems (most notably the
Amazon] may be fresh a considerable dis-
tance out to sea.

Despite their vastly smaller extent, inland
aquatic habitats show far more variety in their
physical and chemical characteristics than
marine habitats. They encompass systems as
varied as the world's great lakes and rivers,
small streams and ponds, temporary puddles,
thermal springs and even the minute pools
of water that collect in the leaf axils of certain
plants, such as bromeliads. Chemically they
range from almost pure water to highly
concentrated solutions of mineral salts,
toxic to all but a few specialized organisms
(see Box 7.1).

Inland water habitats can be divided into
running or lotic and standing or lentic sys-
tems. They may also be divided into per-
manent water bodies, periodically (usually
seasonally) inundated, and ephemeral or
transient. Each of these has its own distinct
set of ecological characteristics.

There is not necessarily a rigid dividing line
between an inland aquatic habitat on the one
hand and a terrestrial or marine habitat on
the other. Any temporarily inundated area,
such as a river floodplain, is effectively a
hybrid or transitional system, being at some
times essentially aquatic, at other times
terrestrial. Similarly there are many areas
that consist of shifting mosaics of land and
shallow water, or areas of saturated vege-
tation, such as sphagnum moss bogs that are
strictly neither land nor water. These trans-
itional areas are often collectively termed
'wetlands'. Similarly, estuarine areas are
transitional areas between inland and marine
systems.

Lotic systems: rivers and catchment basins
A river system is a complex but essentially
linear body of water draining under the
influence of gravity from elevated areas of
land toward sea level. The typical drainage
system consists of a large number of smaller
channels (streams, rills, etc.) at higher ele-
vation merging as altitude falls into pro-
gressively fewer but larger channels, which in
simplest form discharge by a single large
watercourse. Most such systems discharge
into the coastal marine environment. Some
discharge into lakes within enclosed inland
basins; a few watercourses in arid regions
enter inland basins where no permanent
lake exists.

The source area of all the water passing
through any given point in the drainage
system is the catchment area for that part of
the system. In parallel with the hierarchical
aggregation of tributaries of the major river
system, sub-catchments aggregate into a
single major catchment basin; this is the
entire area from which all water at the final
discharge point of the system - i.e. usually
the sea - is derived. Strictly, the watershed
is the line of higher elevation dividing one

Inland water biodiversity 165

catchment basin from another, but this
term is increasingly used as a synonym
of catchment.

The speed and internal motion of river
water depends largely on water volume and
the shape of its channel. These factors typi-
cally differ greatly through the river system,
from narrow, steep and fast upland feeder
streams to broad, level and slow downstream
reaches. In virtually all river systems water
volume also varies seasonally. Some rivers in
arid or semi-arid catchments flow for only
part of the year, or in extreme cases only once
every several years.

Large rivers may span many degrees of
latitude and pass through a wide range of
climatic conditions within their catchments.
Variations in water flow and underlying geo-
logy also create a wide range of habitats
within any river and often within a short
distance. Different organisms are typically
adapted to different parts of any given
river system.

River systems can change course radically
as a result of deposition and erosion of their
channel, and the uplift and erosion of
watershed uplands. Despite the dynamic
physical state of these systems, large rivers
rarely disappear, and although direct evi-
dence is scarce, indications are that some
have been in continuous existence for tens of
millions of years. This is consistent with the
fact that running waters include represen-
tatives of almost all taxonomic groups found
in freshwaters, and that several invertebrate

taxa occur only in running waters or attain
greatest diversity there. By far the largest
river catchment in the world is the Amazon
(with the Ucayalil in South America which
covers just under 6 million km' and is nearly
60 percent larger than the next largest, the
Congo in central Africa. Unsurprisingly, the
former is the major repository of the world's
freshwater biodiversity. Between them the
20 largest river catchments cover around
A5 million km', or about one third of the
world's ice-free land surface'.

Lentic systems

Lakes and ponds

The great majority of existing lakes, of which
around 10 000 exceed 1 km' in extent, were
formed as a result of glacial activity, with most
of the rest a result of tectonic activity'. Tectonic
lakes are formed either as a result of faults
caused by deep crustal movements or by
volcanism. In the case of the former, a lake
may form in a depression caused by a single
fault, or in a depressed area between two or
more faults - these being graben lakes - or in
a rift valley Most volcanic lakes form in craters
or calderas of volcanoes while a few (usually
short-livedl may form behind dams caused by
lava flows. Glacial lakes occupy basins caused
by the scouring action of ice masses.

Most of the world's existing lakes are
glacial and geologically very young, dating
from the retreat of continental ice sheets at the
start of the Holocene. around 11 500 years
before present. All such lakes are expected to

The majority of lakes
were formed as a result
of glacial activity.

u« WORLD ATLAS OF BIODIVERSITY

Table 7.2

Physical and biodiversity
features of major long-
lived lakes

Notes: Lakes ordered by
volume. A few other lakes
have notable endemism
among fishes, mollusks,
crustaceans or other groups.
Among these are lakes Inle
IMyanmarl. Lanao
IPhilippmesI, Malili
(Indonesia] and the Cuatro
Cienegas basin (Mexico] - but
their ages are not yet firmly
established. Qualitative
remarks (e.g. very high',
Tow'] in the Biodiversity
column are related to long-
lived lakes, not to lake
systems in general. All
biodiversity data are
approximate and subject to
change with new survey data
or different taxonomic
opinion,

i Evidence indicates that the
lake dried out completely, or
nearly so. around the late
Pleistocene, 10-12 000 years
ago".

Source: Collated from data in Martens
et at. . Fish estimates for East African
Lakes from Snoeks°^.

Lake

Country Age

Max.

Vol.

^^^^^^^^^M

(million

deptit

(km'l

^^^^^^^^M

years)

(ml

'^V^^H

Baikal

Russia 25-30

1637

23 000

Very high sp, richness; exceptional

Largest, deepest.

endemism in fishes and several

oldest extant freshwater

invertebrate groups

lake (20% of all liquid fresh

Total animal spp.: 1 825, endemic: 982

surface water on Earth]

Fishes: 56 spp., 27 endemic

Tanganyika

Burundi 20
Tanzania
Zambia
DR Congo
(former Zaire]

1470

18 880

Very high sp. richness; high
endemism, especially high among
cichlid fishes

Total animal spp.: 1 470. endemic: 632
Fishes: 325 spp., including 250 cichlids
of which 98% endemic

Malawi

Ivlalawi >2

Iwlozambique

Tanzania

780

8 400

Very high sp. richness; high
endemism, especially high among
cichlid fishes

Fishes: 845 spp., including 800 cichlids
of which 99% endemic

Victoria

Kenya >4?'

2 7i0

High sp, richness, especially of

World's second largest

Tanzania

fishes; exceptional endemism

freshwater lake (area]

Uganda

among cichlid fishes -many fish
endemics depleted or extirpated
following introduction of Nile perch
Fishes: 545 spp,, including 500 cichlids
of which 99% endemic

Titicaca

One of world's highest

Bolivia 3

280

890

Moderate sp. richness and endemism

altitude lakes

Peru

(highest among fishes]

Total animal spp,: 533, endemic: 61

Fishes: 29 spp., 23 endemic

Biwa

Japan 4

104

674

Moderate sp. richness and
endemism (highest in gastropod
mollusks and fishes]

Total animal spp.: 595, endemic: 54 !
Fishes: 57 spp., 4 endemic i

Olirid

Albania 3

295

Moderate sp, richness; 5

Fed mainly by

Macedonia

exceptional endemism in several '

subterranean karst

IFYR]

groups (planarians, oligochaetes, '■

waters

gastropod mollusks, ostracod

crustaceans]

Fishes: 17 spp., up to 10 endemic

Inland water biodiversity 167

fill slowly with sediment and plant biomass.
and to disappear within perhaps the next
100 000 years, along with any isolated biota.
Lakes may also be caused by the dissolution of
soluble rocks, most notably limestone in karst
regions which is gradually dissolved by dilute
acids in water running through it, and by
changes in the course of rivers in floodplain
regions, which result in ox-bow and scroll
lakes.

Only about ten existing lakes are known
with certainty to have origins much before the
Holocene (Table 7.2P, and most of these
occupy basins formed by large-scale sub-
sidence of the Earths crust, dating back to
at most 20 million (Lake Tanganyikal or 30
million (Lake Baikal) years before the present.

There is good evidence that some extinct
lake systems in the geological past were very
large and very long-lived under different
climatic and tectonic conditions. In general,
the long-lived lakes are of particular interest
in terms of biodiversity because they tend to
be rich in species of several major groups of
animals and many of these species are
restricted to a single lake basin.

Wetlands

As indicated above, the distinction between a
wetland, an aquatic system and a terrestrial
system may be essentially arbitrary. However,
a number of mixed shallow-water and terres-
trial habitat types share several characteristics
and are habitually grouped as wetlands.
Wetlands in this sense are typically hetero-
geneous habitats of permanent or seasonal
shallow water dominated by large aquatic
plants and broken into diverse microhabitats\
The four major broad habitat types are:

Bogs

Bogs are peat-producing wetlands in moist
climates where organic matter has accum-
ulated over long periods. Water and nutrient
input is entirely through precipitation. Bogs
are typically acid and deficient in nutrients
and are often dominated by sphagnum moss.

Fens

Fens are peat-producing wetlands that are

influenced by soil nutrients flowing through

the system and that are typically supplied
by mineral-rich groundwater Grasses and
sedges, with mosses, are the dominant vege-
tation. Fens are typically more productive and
less acidic than bogs.

Marshes

Marshes are inundated areas with herb-
aceous emergent vegetation, commonly dom-
inated by grasses, sedges or reeds. They may
be either permanent or seasonal and are fed
by ground or river water, or both.

Swamps

Swamps are forested freshwater wetlands on
waterlogged or inundated soils where little or
no peat accumulation occurs. As with marshes,
they may be either permanent or seasonal.

Biogeography and important areas
Freshwater lineages that originated in contin-
ental water systems may show general
patterns of distribution similar to terrestrial
groups, corresponding more or less to broad
biogeographic realms. Lineages of marine
origin may remain restricted to peripheral
systems corresponding to the area where the
ancestral forms moved into freshwater

Unlike many terrestrial species, which can
disperse widely in suitable habitat, the spatial
extent of the range of strictly freshwater
species tends to correspond to present or
formerly continuous river basins or lakes.
These species include fishes and most moll-
usks and crustaceans. Watersheds between
river basins are the principal barriers to their
dispersal between systems, and their ranges
are extended mainly by physical changes to
the drainage pattern (e.g. river capture
following erosion or uplift can allow species
formerly restricted to one system to move into
another!, or by accidental transport of eggs by
waterbirds, or by flooding.

In many instances, the range within a
system will also be restricted by particular
habitat requirements (variations in water
turbulence or speed, shelter, substrate, etc.].
These frequently differ at different stages in
the life cycle (for example in fishes the
conditions and sites required for egg depos-
ition and development, for early growth of fry,

168 WORLD ATLAS OF BIODIVERSITY

Continent

Area name ^^^^H

Taxonomic group

^^1

Africa

L. Malawi

Fishes

Mollusks

Africa

L Tanganyika

Fishes

Mollusks

Crabs

Africa

L.Victoria

Fishes

Mollusks

Africa

Madagascar

Fishes

Mollusks

Crabs

Africa

Niger-Gabon

Fishes

Crabs

Africa

Upper Guinea

Fishes

Mollusks

Crabs

Africa

Lower Congo

Fishes

Mollusks

Crabs

Australia

SE Australia & Tasmania

Fishes

Mollusks

Crayfish

Australia

SW Australia

Fishes

Fairy shrimp

Eurasia

SE Asia and lower

Ivlekong River

Fishes

Mollusks

Crabs

Eurasia

Balkans (southwest 1

Fishes

Mollusks

Eurasia

L. Baikal

Fishes

Mollusks

Eurasia

L. Biwa

Fishes

Mollusks

Eurasia

L Inle

Fishes

Mollusks

Eurasia

L Poso

Fishes

Mollusks

Eurasia

Malili Lakes

Fishes

Mollusks

Eurasia

Sri Lanka

Fishes

Crabs

Eurasia

Western Gtiats

Fishes

Mollusks

Crabs

Nortfi America

East Mississippi drainage
(Ohio, Cumberland,

Tennessee rivers)

Fishes

Mollusks

Crayfish

North America

Mobile Bay drainage

Fishes

Mollusks

Crayfish

North America

Western USA

Fishes

Mollusks

Fairy shrimp

South America

L. Titicaca

Fishes

Mollusks

South America

La Plata drainage

Fishes

Mollusks

South America

Amazon basin

Fishes

Crabs

Table 7.3

Partial list of global

hotspots of frestiwater

biodiversity

Notes: This table lists areas
of special importance for
diversity in fishes and either
mollusks or crustaceans or
both. See text and Appendix
6. Six of the seven long-lived
lakes in Table 7.2 also
appear here

Source: See sources cited in
Appendix 6,

and for feeding and breeding of adults are
often different).

Many cave or subterranean freshwater
aquatic species (e.g. of fishes, amphibians
and crustaceans) have restricted ranges,
perhaps consisting of a single cave or aquifer,
and limited opportunities for dispersal,
depending on the surrounding geology and
the consequent morphology of the water
system occupied.

Analysis of data from some 151 river
basins indicates that there is a strong cor-
relation between the spatial extent of a river
catchment and the number of fish species
therein. The size' of a river can be represen-
ted by the area of the basin or by the volume
of water flowing through the river system in
any given period; the latter is a better
predictor of fish species richness than is
basin area. When area is taken into account.

there is also a strong relationship between
species richness and the latitude of the basin.
Recent analysis suggests that latitude may be
a surrogate measure for energy availability
and productivity within the basin' ^ factors
known to be well correlated with variation in
terrestrial diversity (Chapter 5). No taxonomic
class restricted to inland waters has yet been
mapped globally at species level but, at a
higher taxonomic level, a density surface of
freshwater fish families has been developed
(see Map 7.1] with a view to providing an
indication of global variation in inland water
diversity analogous to those available for
terrestrial groups (Chapter 5).

A recent analysis of areas important for the
maintenance of global freshwater biodiversity'
was based on the expert view of a number of
regional and taxonomic specialists. The analy-
sis was designed to make effective use of
readily available information and. although
preliminary, yielded the first global overview of
freshwater biodiversity hotspots'. fvtaps 7.2, 7.3
and 7.A show, respectively, important areas for
freshwater fishes, mollusks and selected crus-
tacean groups. Further details of all these
areas can be found in Appendix 6. Table 7.3
lists the sites and areas that have been
identified as of special importance for more
than one of the above groups. It is not a
comprehensive global listing because it omits
several large but imprecisely defined areas of
known high diversity, and it omits diverse taxa
not covered in the assessment (e.g. amphi-
pods, copepodsl; nor does it mention sites of
key importance mainly for one group of
animals. Although the Amazon basin is a vast
region rather than an identifiable site, it has
such an exceptional diversity of fishes that it
could not reasonably be excluded from a list of
globally important areas.

More detailed continental reviews are now
also available for Asia, including discussion of
taxonomy, hotspots and policy', tor Latin
America' and North America'. In order to help
prioritize investment, the conservation organi-
zation WWF has selected 53 freshwater eco-
regions, based on a combination of biogeo-
graphic region, water body type, biodiversity and
representativeness' ". Perhaps more so than in
other biomes, these freshwater ecoregions

Inland water biodiversity 169

■'^^

have a firm objective basis because they
correspond broadly with catchment basins.

BIOLOGICAL DIVERSITY IN INLAND WATERS

At high taxonomic levels the diversity of
freshwater organisms is considerably lower
than on land or in the sea. Only one extant
eukaryote phylum IGamophyta - green conju-
gating algael is apparently confined to fresh-
water habitats. The number of species overall
is low in absolute terms in comparison with
marine and terrestrial groups, but species
richness in relation to habitat extent is
relatively high. For example, about 10 000 (40
percent! of the 25 000 known fish species are
freshwater forms'. Given the distribution of
water on the Earth's surface this is equivalent
to one fish species for every 15 km^ in fresh-
waters compared with one for every 100 000
km-' of sea water. This high diversity of
freshwater fishes relative to habitat extent is
undoubtedly promoted by the extent of isol-
ation between freshwater systems. Many
lineages of fishes and invertebrates have
evolved high diversity in certain water sys-
tems, and in some cases, species richness
and endemism tend to be positively correlated
between different taxonomic groups".

As is the case with terrestrial habitats,
species richness increases strongly toward
the equator, so that in most groups of
organisms, there are many more species in
the tropics than in temperate regions,
although in a tew specific cases (e.g. fresh-
water crayfish! this appears to be reversed.

Protoctlsts

The larger algae comprise some 5 000 species
in three major groups (the green, brown and
red algae!, the great majority of which are
marine or brackish water forms ('seaweeds'].
The green algae Chlorophyta includes one
order of around 80 species (Ulotrichales) that
is mainly freshwater However, one major
group usually associated with the green algae
- the stoneworts (charophytes) - is almost
entirely freshwater The stoneworts include
some 440 species, most of which are endemic
at continent level or below; they tend to be very
sensitive to nutrient enrichment and have
declined in many areas'^

Fungi

There are more than 600 species of fresh-
water fungi known, currently more from
temperate regions than from the tropics,
although probably only a small fraction of
existing species have been described, and the
tropics have been little sampled''. Virtually all
described freshwater fungi are ascomycotes
with few basidiomycotes and zygomycetes
having been identified. They occur wherever
vascular plant material is available as a
substrate. They appear to be important as
parasites, endotrophs and saprotrophs of
emergent aquatic macrophytes, as decom-
posers of submerged allochthonous woody
debris, and as a food resource for inver-
tebrates' '-. Most are very small, with
sporomes (fruiting bodies! less than 0.5 mm
in diameter.

Plants

Wetland or aquatic species occur with some
frequency in the non-vascular plant phyla,
which generally prefer moist habitats, and
among the ferns and allies. Mosses in the
order Sphagnales (a single family Sphag-
naceae and genus Sphagnum] often grow
submerged, and are key components of
peat bogs. Many groups of damp-loving
(hygrophilousj terrestrial mosses (e.g.
Thamnium, Bryum, Mnium] have aquatic
forms. Several genera of Bryales are aquatic
or have aquatic species. A number of

Species richness in
inland waters, in relation
to habitat extent, is
relatively high.

170 WORLD ATLAS OF BIODIVERSITY

Map 7.1

Freshwater fish family

diversity

Family richness of typical
bony fishes (Actinopterygii)
in inland waters, plotted as
a world density surface. It is
based on generalized range
maps of 157 families. Color
depth represents the
number of families, up to a
maximum of kit, potentially
present at any point. Two
families of cartilaginous
fishes lElasmobranchiil that
together have a very few
inland water species are
omitted. About ten or so
families of bony fish that
occur in coastal and
estuarine waters, but do
not extend significantly into
inland freshwaters, are
omitted. Several families
range more or less widely
in inland waters and also
occur in coastal and
estuarine waters around
the continents, but this
peripheral part of the range
is in most cases not
represented.

Source: Produced by UNEP-WCMC
using range maps prepared from
information in Berra .

Density

High

Low

liverwort species growing otherwise in wet
terrestrial situations may also live sub-
merged, sometimes at considerable depths.
Truly aquatic liverworts include several
species of Riccia and Ricciocarpus natans
[Ricciaceae; MarchantialesI that live free-
floating on the surface of eutrophic lakes. At
least 16 species of Riella in the Riellaceae
(JungermannialesI are aquatic forms charac-
teristic of temporary waters in semiarid
regions, reaching highest diversity in north-
ern and southern Africa. Among the lyco-
phytes, most of the 60 or so species of
Isoetes (family Isoeteacael are aquatic, some
of great limnological importance, and Stytites
is an endemic member of the littoral
community of Andean lakes. Sphenophytes
Ihorsetailsl often occur in moist situations,
including around water margins. Equisetum

icr-'v-

Jlfi>c

\ 1

fluviatite, for example, is a notable emergent
littoral form of north temperate lakes.

The Filicinophyta include several aquatic
forms. The genus Ceratopteris (four species,
family Pteridaceael has the only truly aquatic
(floating! homosporous ferns; some are culti-
vated ornamentals, others are edible. A
few other species, e.g. Microsorum pteropus
(Polypodiaceael and Microlepia speluncae
(Dennstaedtiaceae) can grow in water. Among
heterosporous ferns, the family Marseliaceae
comprises three genera and 55-75 species
which are either amphibious or fully aquatic.
All members of the families Salviniaceae (one
genus and about ten species) and Azollaceae
(one genus and around six species] are
floating aquatic ferns. The latter supports the
symbiotic nitrogen-fixing Anabaena azollae
(phylum Cyanobacteria).

Inland water biodiversity 171

ry-

■V

Vascular plants are essentially terrestrial
forms, and existing aquatic species are
derived from terrestrial ancestors; several
different lineages include aquatic species and
this transition has therefore occurred several
times. It has been estimated that at most
1 percent of angiosperms, i.e. up to 2 700
species, are aquatic". Around U angiosperm
families consist largely or exclusively of in-
land water forms (Ceratophyllaceae, Hippuri-
daceae, Hydrostachydaceae, Nymphaeaceae,
Podostemaceae, Trapaceae, Butomaceae,
Hydrocharitaceae, Lemnaceae, Limnochari-
taceae, Najadaceae, Pontederiaceae, Potamo-
getonaceae, Zannichelliaceael.

Most inland water plant species are
relatively widespread, ranging over more than
one continental land mass; many are cosmo-
politan, occurring around the world and on

, r^ k

remote islands. Of the widespread forms,
some are essentially northern temperate
species extending to a great or lesser extent
into the tropics; some are mainly tropical".
The Podostemaceae is particularly note-
worthy for its many monotypic genera, and a
large number of narrowly endemic species, in
at least one instance with several forms
restricted to different stretches of a single
river Tropical South America, Madagascar,
Sri Lanka, India, Myanmarand Indonesia hold
such localized species".

Inland water animals

Animal species are considerably more diverse
and numerous in inland waters than plants.
Most of the major groups include terrestrial
or marine species as well as freshwater
forms. Apart from fishes, important groups

172 WORLD ATLAS OF BIODIVERSITY

Table 7.4

Insects of inland waters

Notes: Data refer to number
of species; all estimates
approximate.

i Partially aquatic as adult

and sometimes as nympfi.

ii Number m parentfieses

refers to fully aquatic

Nepomorpha.

ill All tfiese species are

parasitoids as larvae.

Source: Collated from data in
Hutctiinson .

With inland water species include crus-
taceans (crabs, crayfishes, shrimps, as well
as planktonic forms such as filter-feeding
Cladocera and filter-feeding or predatory
Copepodal, moUusks (including mussels
Bivatvia, and snails Gastropoda], insects
(including stoneflies Plecoptera, caddisflies
Trichoptera, mayflies Ephemopteral, sponges,
flatworms, polychaete worms, oligochaete
worms, numerous parasitic species in various
groups, and numerous microscopic forms.
Information is incomplete for many groups,
but crustaceans and mollusks have speciated
profusely In certain freshwater systems, with
a tendency to form local endemic species.
Because of the feeding mode - attached
bottom-living filter-feeders - bivalves can
help maintain water quality but tend to be
susceptible to pollution (their larvae are
parasitic on fishesl. The diversity and eco-
logical role of microorganisms and micro-
Invertebrates In freshwater sediments have
been reviewed".

Insects with an aquatic larval phase but a
winged adult phase are often restricted to
particular river basins (even If adults disperse
widely, they may not find suitable habitat], but
in general are much less restricted in this way
than entirely aquatic species. A relatively
large number of species, particularly of
crustaceans, occupy temporary pools and
have a stage that is desiccation-resistant and
can undergo long-range passive dispersal

Ephemeroptera

614

224

2 250

Odonata

302

415

127

4 875

Plecoptera

196

578

387

2140

Orthoptera'

ca20

Blattodea'

calO

Hemlptera

236

404

129181]"

3 200

Megaloptera

300

Neuroptera

calOO

Coleoptera

730

1655

1072

5 000

Hymenoptera"

calOO

Diptera

1300

5 547

4 050

>20 000

Trichoptera

478

1340

895

7 000

Lepldoptera

calOO

between drainage basins; some such species
are thus widely distributed.

Insects

As on land. Insects (phylum Mandibulata] are
as far as is known by far the most diverse
group of organisms In Inland waters. The
true number of aquatic Insects remains
unknown; data for three relatively well
known areas (Europe, Australia and North
America] and extrapolations for possible
global totals are Included in Table 7.A. In
contrast to terrestrial faunas, where beetles
(order Coleoptera] are the most diverse, flies
and their relatives (order Diptera] appear to
be by far the most abundant group In inland
water habitats, although also one of the less
fully known.

In terms of life histories, there are two
main groups of aquatic insects: those In which
the adult stage and the active Immature
stages are passed in water (in some cases
with a terrestrial pupal stage]; and those in
which, afte'- a nymphal or larval stage in
water, the adult stage is spent on land or In
the air The great majority of Diptera, and
therefore most aquatic insects, form part of
the latter group. Included amongst their
number are several of enormous economic
significance to humans, of which the most
significant are almost certainly mosquitoes of
the genus Anopheles, Intermediary hosts of
the malaria parasite.

Most aquatic insects are benthic, living In
or on the bottom; a small number are
planktonic and live suspended In the water
column. Around half of the aquatic Hemlptera
and a few other insects and non-Insect
invertebrates live on the water surface
(epipleuston].

Fishes

Around 40 percent of known fish species
occur In freshwater: almost exactly 10 000
species are confined to freshwater, and a
further 1 100 or so occur in freshwater but are
not confined to It (Table 7.5]. These last
Include species, such as many salmonlds,
that grow in the seas but ascend rivers to
spawn (anadromous], and others, such as
eels, that grow in inland waters but spawn at

Inland water biodiversity 173

sea (catadromousl. Freshwater fishes are
taxonomically diverse, although not as diverse
as marine ones. Thirty-tour of the 57 or so
extant orders of fishes have at least one
strictly freshwater species, while a further
two, the sawfishes (Pristiophoriformesl and

tarpons (Elopiformes) have species that occur
in freshwater but are not confined to it".
This compares with 38 orders that have at
least one marine species.

Of the orders of fishes with freshwater
species, ten are entirely freshwater and

Table 7.5

Fish diversity In Inland

waters, by order

Source: Nelson (differs in detail from
tfie later taxonomy of Eschmeyer 1.

■flll!^""""

ToT^^

■^JffiF

^^0^^

^To^t

^^oTt

Approximate Approximate

[strictly freshwater

families

genera

species

strictly

species

% strictly

% using

orders in boldl

freshwater
species

using
freshwater

freshwater

Petromyzontiformes

Lampreys

100

Carcfiarhiniformes

Ground stiarl<s

208

Pristiophoriformes

Sawfishes

Rajiformes

Rays

456

Ceratodontiformes

Australian lungfish

100

Lepidoslreniformes

Lungfishes

100

Acipensiformes

Sturgeons

100

Amiiformes

Bowfin

100

Anguiliformes

Eels

141

738

Atheriniformes

Silversides

285

146

171

Batrachoidiformes

Toadfishes

Beloniformes

Needlefisfies, sauries.

flyingfishes, halfbeaks

191

Characlformes

Characins

237

1343

100

Clupeitormes

Herrings and anchovies

357

Cypriniformes

Carp, minnows, loaches

279

2 662

100

Cyprinodontiformes

Rivulines, killifishes, pupfishes,

poeclliids, goodeids

807

794

805

100

Elopiformes

Ladyfishesand tarpons

Esociformes

Pikes and mudminnows

100

Gadiformes

Cods, hakes, rattails

482

Gasterosteiformes

Pipefishes, sticklebacks, sandeels, etc.

257

Gonortiynchiformes

Milkfish and beaked sandfishes

Gymnotlformes

Knifefishes

100

Muglliformes

Mullets

Ophidiiformes

Pearlflshes, cusk-eels, brotulas

355

Osmeriformes

Smelts

236

Osteoglossiformes

Bonytongues

217

100

Perciformes

Perches, basses, sunfishes, whitings, etc.

1496

9 293

1922

2815

Percopslformes

Trout-perches, pirate perch, cavefishes

100

Pleuronectiformes

Plaice, flounders, soles

123

570

Polypteriformes

Bichirs

100

Salmoniformes

Salmonids

100

Scorpaeniformes

Gurnards, scorpionfishes, velvetfishes, etc.

266

1271

Semionotiformes

Gars

100

Siluriformes

Catfishes

412

2 405

2 280

2 287

Synbranchiformes

Swamp-eels

100

Tetraodontiformes

Triggerfishes, puffers, boxfishes.

fllefishes, molas

100

339

■1

174 WORLD ATLAS OF BIODIVERSITY

Virtually all reptiles of
inland waters return to
land, at least to nest.

anottier five are very largely so (witfi more than
80 percent of their known species in fresh-
watersl. A further 13 are very largely marine
with a small proportion of freshwater species
l<10 percent] while the remainder have
significant numbers of both marine and
freshwater species. Over 80 percent of fresh-
water species are confined to just four orders:
the carps and their relatives ICypriniformesI;
the characins iCharaciniformesl; the catfishes
(Siluriformesl; and the perches and their
relatives (Perciformesl. The first three of these
are wholly or almost entirely freshwater, while
the last, the largest order of fishes with nearly
AO percent of known species, is unusual in
having significant numbers of both marine and
freshwater species.

Amphibians

The great majority of the 5 000 or so living
amphibian species have aquatic larval stages
and, as none is known to occur in sea water,
all these are dependent on inland waters of
various kinds for continued survival of
populations. In some cases such water bodies
may be temporary pools or puddles, or water
in the leaf axils of plants. Relatively few
species are fully aquatic (Table 7.6). Although
the number of fully aquatic species in each of
the three extant orders is roughly similar
lea 20-30 in each], these represent very
different proportions of each order, being less
than 1 percent of anurans, around 5 percent
of caudate amphibians and more than
10 percent of caecilians.

Aquatic caudate amphibians are neotenic,
that IS retain features of the larval stage, most
notably external gills. In addition to the fully
aquatic amphibians (several of which can
survive for short periods in damp conditions
out of water], many other species may lead
largely aquatic lives or may, as in the case of
the Mexican axolotl Ambyostoma mexicanum,
have completely aquatic neotenic populations.

Reptiles

Very few completely aquatic inland water
reptiles are known. The three file-snakes in
the family Acrochordidae are live-bearing and
may pass their entire lives in water, often in
coastal and estuanne areas as well as

freshwaters. Virtually all other reptiles of
inland waters are egg-laying and return to
land at least to nest; most also spend a
proportion of their time on land, often basking
on banks or logs. The two most aquatic orders
are the Crocodilia and the Chelonia. All the
22 or so extant species of the former are

predominantly aquatic and occur in fresh-
waters, although one or two may also be found
in marine areas. Of the latter, some two thirds
of the 250 or so extant species are largely or
predominantly aquatic, and a further 30 or so
species may be considered amphibious.

Aside from the file-snakes and the homal-
opsine snakes, a number of other snake
species are at least semi-aquatic. These in-
clude several genera of natricine snakes,
including the North American Nerodia and the
Asiatic Sinonatrix, as well as the anacondas
Eunectes [family Boidae] and the water cobra
Boulengeria annulata (family Elapidae].
Amongst lizards, no wholly aquatic species is
known. However, many can swim proficiently,
often using water as a means of escape from
predators, and a number are semi-aquatic.
These last include several Australian and
Old World monitors Varanus spp. [family
Varanidae], water dragons Hydrosaurus and
Physignathus [Agamidae], the crocodile lizard
Shinisaurus from south China [Xenosauridael,
the Bornean earless monitor Lanthanotus
[Lanthanotidae], and a number of New World
teiids [Teiidael.

Inland water biodiversity 175

^*K:

^ Class and order

Family

No. of freshwater species

AMPHIBIA

Caudata

Amphiumidae

Congo eels

Cryptobranchldae

Giant salamanders and hellbenders

Plellnodontidae

Lungless salamanders

Proteidae

Mudpuppies and olm

Sirenidae

Sirens

^nura

Pseudidae

Paradox frogs

PIpidae

Clawed frogs and pipid toads

Gymnophiona

Typhlonectidae

Typhlonectid caecilians

19"

REPTILIA

Chelonia

Carettochelidae

Pig-nosed soft-shelled turtle

Trionychidae

Soft-shelled turtles

Platysternidae

Big-headed turtles

•',

Chelydrldae

Snapping turtles

Dermatemydidae

Central American river turtle

■

Chelldae

Austro-american side-necl<ed turtles

■

Kinosternidae

Mud and musl< turtles

■:

Pelomedusidae

Side-necl<ed turtles

Emydidae

Pond and river turtles

Crocodilia

Alllgatoridae

Caimans and alligators

Crocodylidae

Crocodiles

Gavialidae

Gharialand false gharial

Squamata

Acrochordidae

File-snal(es

Colubridae

Colubrid snal<e5

iO''

AVES

Anseriformes

Anseranatidae

Magpie goose

Dendrocygnidae

Whistling-ducks

Anatidae

Ducl(s, geese and swans

Gruiformes

Heliornitiiidae

Limpkin andsungrebes

Rallidae

Rails, gallinules and coots

Ciconiiformes

Jacanidae

Jacanas

Laridae

Gulls, terns, skuas, auks, skimmers

22 1131'

Podicipedidae

Grebes

Anhingidae

Anhingas

Phalacrocoracidae

Cormorants and shags

5121'

Phoenicopteridae

Flamingos

Pelecanidae

Pelicans and shoebill

Gaviidae

Divers

Passeriformes

Cinclidae

Dippers

MAMMALIA

Monolremala

Ornithorhynchidae

Platypus

Didelphiomorpha

Didelpliidae

Opossums

Insectivora

Tenrecidae

Tenrecs and otter shrews

Soricidae

Shrews

Talpidae

Moles and desmans

Rodentia

Castoridae

Beavers

Muridae

Voles and mice

Hydrochaeridae

Capybara

[ Cetacea

Platanlstidae

River dolphins

Sirenia

TrIchechidae

Manatees

Carnivora

Mustelidae

Mustelids, otters

Viverridae

Viverrids

Phocidae

Earless seals

Artiodactyla

Hippopotamidae

Hippopotamus

Table 7.6

Tetrapod diversity in

inland waters

Notes: Entirely freshwater
families in bold. Taxonomy
based on the same vertebrate
sources cited in Table 2.1.
Bird groups here differ to
some extent from Table 7.9
which uses a more traditional
arrangement of bird higher
taxa.

i Genera Leurognathus,

Haideotnton. Typhlomolge

only

11 Sometimes included in

the Caecilidae.

ill Subfamily Homalopsinae.

iv Genus Fulicula Icoots]

only

V Figures in parentheses

indicate those species of

the total that breed largely

or entirely inland also

included as seabirds in

Table 6.6.

176 WORLD ATLAS OF BIODIVERSITY

Map 7.2

Major areas of diversity

of inland water fish

This map represents an
informal synthesis of
documented expert opinion
on globally important areas
for freshwater fish diversity,
taking into account species
richness and endemism.
Two categories are shown:
discrete areas and systems
known to be of high
diversity, and areas where
diversity is globally
important but less
concentrated.

Note: For numbered
locations see Appendix 6.

Source: Comprled with the help of
members of lUCN/SSC specialist groups
and other ichthyologists; fjrsl published
inWCMc'.

Fish diversity

^^^H Key areas

Other important areas

Birds

Unlike mammals, there are no wholly aquatic
birds, because all species lay eggs that
cannot survive prolonged immersion in water;
however, a much higher proportion of bird
than mammal species is associated with
inland water ecosystems. As with other
tetrapod groups, it is impossible to separate
rigidly inland water species from primarily
terrestrial forms or from seabirds. Table 7.6
includes bird species that are highly adapted
to aquatic ecosystems and largely or ex-
clusively inhabit inland waters, rather than
marine or coastal areas. Nearly 60 percent of
the roughly 250 species belong to one family -
the Anatidae - all of whose members are
more or less associated with aquatic habitats,
although some [such as most goose species!
feed very largely on land. It is noteworthy

?■!'
W

103 9^

O 101 96

99
97

100~-:98<

104 t ,

123-1

124

N^131;

116

that, among the Passeriformes or perching-
birds - by far the most species-rich order
(accounting for over half of all bird species) -
only one small family, the dippers (Cinclidael.
can be considered truly aquatic in habits.

In addition to these, there are a large
number of wading bird species associated
with inland or coastal wetland and littoral
habitats. These include all or most members
of the following families: Anhimidae (scream-
ers - three species); Eurypygidae (sunbittern -
one species); Gruidae (cranes - 15 species);
Rallidae (rails, gallinules and coots - 142
species); Scolopacidae (sandpipers and their
relatives - 88 species); Rostratulidae (painted-
snipe - two species); Charadriidae (plovers
and their relatives - 89 species); Ardeidae
(herons, bitterns and egrets - 65 species);
Scopidae (hammerkop - one species);

Inland water biodiversity 177

"OX

65 ,73 39 ?'^'

41/

^42

Ttiresl<iornitlnidae (ibises and spoonbills - 3A
species!; and Ciconiidae (storks - 26 species).
There are also a considerably smaller number
of non-wading birds that feed largely on fishes
and other aquatic animals and are adapted to
diving or surface-snatching. Among these are
many kingfishers (families Alcedinidae,
Dacelonidae and Cerylidael, the fish-owls
[Ketupa and Scotopelia spp., family Strigidael,
fish-eagles iHaiiaeetus and Ichthyophagus
spp., family Accipitridae) and a few other
raptors.

Mammals

Wholly aquatic inland water mammals are
confined to two orders, Cetacea and Sirenia.
In the former, four of the five species of the
family Platanistidae (the river dolphins! are
confined to river systems and the fifth occurs

in estuarine and coastal waters. A number of
other cetaceans may enter the lower reaches
of river systems but all are predominantly
marine. Two of the four living species of
Sirenia - the Amazonian manatee Trichechus
inunguis and the West African manatee
T. senegalensis - are wholly or very largely
confined to freshwaters, and a third, the
Caribbean manatee T. manatus, is found in
both inland and marine waters.

Amongst other groups, a number of
species may lead more or less aquatic lives
but all these are effectively amphibious, in that
at the very least they produce young on land.
For most of these, a predominantly aquatic life
is evident from direct observations, but for
some (chiefly the Muridae! it is inferred from
morphological adaptations. In direct contrast
to terrestrial systems, where the majority of

^«-

178 WORLD ATLAS OF BIODIVERSITY

mammal species are herbivores, a high
proportion of these amphibious species
laround 50 percent! are carnivores, with only
eight true herbivores and most of the aquatic
murids believed to be omnivores (though
predominantly insectivorous or piscivorousl. It
is also notew/orthy that at least four of these
species - the two beavers Castor spp., the
hippopotamus Hippopotamus amphibius and
the capybara Hydrochaeris hydrochaeris -
feed very largely or entirely on land plants. In
addition to the species included in Table 7.6, a
number of other mammals are largely or
wholly confined to wetland habitats (marshes,
floodplains and swampsl. These include three
African antelopes - the Nile lechwe Kobus
megaceros. red lechwe Kobus teche and
sitatunga Tragelaphus spekei - and the
South American marsh deer Blastocerus
dichotomus.

Freshwater is essential to
human surA/ival.

HUMAN USE OF AND IMPACT ON INLAND
WATERS

Freshwater - as precipitation, groundwater or
in inland water ecosystems - is essential for
human survival, chiefly because humans
must drink and also because it is needed, in
far greater quantity, to produce food. It also
has a wide range of subsidiary uses - for
transport, industrial production, cleaning,
waste disposal, generation of hydroelectric
power, recreation, esthetic purposes and in
the form of inland water ecosystems as sites
for the production of food.

Many of these demands conflict with each
other, so that for example the use of water for

disposal of noxious wastes is incompatible with
the provision of safe drinking water Moreover,
while the amount of freshwater available is
limited, demands on it continue to grow
steadily as the global human population con-
tinues to expand. This problem is exacerbated
by the fact that freshwater Is unevenly
distributed around the world, so that it is often
not available where and when needed, nor in
the appropriate amounts, nor with the
necessary quality The two last are particularly
important for the maintenance of freshwater
biodiversity. Freshwater systems are therefore
under growing pressure, as flow patterns are
disrupted and the load of waste substances
increases. Inevitably, per capita shares of
water for human use are decreasing and water
stress is becoming more widespread".

Agriculture consumes around 70 percent
of all water withdrawn from the world's riv-
ers, lakes and groundwater^". In places, more
than half the water diverted or pumped for
irrigation does not actually reach the crop,
and problems of waterlogging and salln-
ization (deposition in soil of salts left by
evaporation of pumped groundwater! are
increasing. However, irrigated agriculture
produces nearly AO percent of world food and
other agricultural commodities on only 17
percent of the total agricultural land area, and
is thus disproportionately important to global
food security-".

The principal use of freshwater species,
not considering the properties of aquatic sys-
tems themselves. Is as food. Subsidiary uses
Include the aquarium trade, materials for
medicinal or ornamental use, and as fertilizer
Inland water fishery production has two com-
ponents: capture fisheries and aquaculture,
although as discussed below the distinction
between the two is becoming increasingly
blurred. For many human communities, par-
ticularly in countries less developed indus-
trially, capture fisheries provide a major
portion of the diet. Although it appears to be
under-reported, inland water production has
usually been regarded as far less important
than marine fisheries and. with few excep-
tions where countries have access to both
marine and inland aquatic resources, repor-
ted yield from inland waters is a small

Inland water biodiversity 179

^*t^

fraction of marine yield. Even m landlocl<ed
countries, the recorded inland harvest is often
[o\N both in absolute size and in relation to
consumption of meat and other agricultural
produce.

INLAND WATER FISHERIES
Capture fisheries and aquaculture

Globally, the reported inland w/ater capture
fishery for 1999 amounted to 8.2 million
metric tons, w/ith 19.8 million metric tons of
aquaculture production recorded''; over 85
percent of the former and about 98 percent of
the latter comprised finfishes, with virtually
all the remainder being freshwater crus-
taceans and mollusks"". The crustaceans are
mainly crayfishes and freshwater shrimps,
both exploited for food, and most of the
mollusks are bivalve, taken tor pearls and for
food. These reported totals compare with
reported marine capture fisheries of some 84
million metric tons, and marine and brackish
water aquaculture animal production of
around 13 million metric tons (see Chapter 6|.

The reported global inland water capture
fishery has increased slowly in the period
1984-99, by nearly 2 percent per year, al-
though this masks considerable regional
variation, with declines in some areas (e.g.
Europe and North America) and more marked
increases elsewhere (notably Asia]". Reported
inland aquaculture production has been rising
at a higher rate, and was well over twice the
reported production of inland capture fisheries
in 1999 (Figure 7.11. A major proportion of
global inland aquaculture is produced by
countries in Asia. China alone reportedly gen-
erates more than one quarter of the global
total (Table 7.71, and has been responsible for
most of the recent increase in this sector In
this particular instance the dividing line
between aquaculture and capture fisheries is
indistinct; no husbandry is involved beyond
release of hatchery stock, and the fishery
operates as a capture fishery''.

However, national statistics do not ad-
equately reflect the actual magnitude, location
or importance of inland fisheries. The repor-
ted inland capture production is certainly a
gross underestimate because much of the
catch is made far from recognized landing

c20

Aquaculture

1994

1995

1996

1997

1998

1999

places where catches are monitored, and is
consumed directly by fishers or marketed
locally without ever being reported. The evi-
dence suggests that actual capture fisheries
catch may be twice or conceivably even three
times the reported total, i.e. around 15-23
million metric tons per year". Because a far
higher proportion of inland fisheries than
marine fisheries harvest is apparently used
directly for human consumption (rather than
production of oils and meals, often used for
livestock feed), and because discards are
believed to be negligible, it has been argued by
some that the provision of foodfish from inland
waters is not that much less than that from
recorded marine catch".

Inland water capture fisheries, particularly
in countries less developed industrially,
certainly provide a staple part of the diet for
many human communities. This is the case in
West Africa generally, locally in East Africa,

Figure 7.1

Reported global inland

fisheries production

Table 7.7

Major inland fishery

countries

Note. The top ten producing
countries in 1998.

Production (metric tons)

% of world total

China

2 280 000

28.5

India

650 000

8.1

Bangladesh

538 000

6.7

Indonesia

315 000

3.9

Tanzania

300 000

3.7

Russia

271 000

3.4

Egypt

253 000

3.2

Uganda

220 000

2.8

Thailand

191 000

2.4

Brazil

180 000

2.3

180 WORLD ATLAS OF BIODIVERSITY

Map 7.3

Major areas of diversity of

Inland water moUusks

This map illustrates the
location of areas regarded
as globally important for
diversity in the bivalve and
gastropod moUusks of
inland waters, taking into
account species richness
and local endemism.

Note: For numbered
locations see Appendix 6.

Source: Compiled using information and
expertise provided by ttie lUCN/SSC
Mollusc Specialist Group; tirst published
inWCMC'.

MoUusk diversity

^^^H Innportant areas

and in parts of Asia and Amazonia. In some
landlocked countries inland fisheries are of
crucial importance, providing more than
50 percent of animal protein consumed by
humans In Zambia" and nearly 75 percent
In Malawi''. In the low-income food-deficit
countries fish protein may be particularly
important In times of food scarcity.

It is impossible at the global level to carry
out any meaningful analysis of the relative
contribution of different species or species
groups to inland capture fisheries because of
the inadequacy of reporting. In FAO IFood and
Agriculture Organization of the United Nations!
statistics, by far the largest group recorded Is
'freshwater fishes not elsewhere included',
that is those that are completely unclassified
other than being identified as finfishes. These
make up just under half of all reported landing

il07

;^H'

■f

134

by weight with a further 15 percent consisting
of moUusks and crustaceans similarly classi-
fied. The majority of the remaining catch Is
classified Into broad species groups (e.g.
cyprlnlds, characlns, siluroldsl, with only three
individual fish species having annual reported
global landings of more than 100 000 metric
tons. These are the Nile perch Lates niloticus
lea 330 000 metric tons reported in 1997), Nile
tllapia Oreochromis niloticus (226 000 metric
tons in 19971 and the common carp Cypnnus
carpio (100 000 metric tons in 19971.

It Is, however, evident that the importance
of different species of freshwater finfishes
varies considerably between different areas.
In terms of food security for local subsistence
or mixed market/subsistence communities,
particularly In the tropics, there is an increas-
ing amount of evidence that the diversity of

Inland water biodiversity 181

yfP^^^^

v£)

83 -''1^ ;!|^,89 y,

-.r

.'iS

species harvested is in itself a major factor in
ensuring a continuous food supply. Many of
the species that contribute to these fisheries
are often small and would be considered
'trash' fishes in orthodox fisheries, Vi^ith much
lower market value than larger (often non-
nativel species that might therefore be
considered for introduction. However, these
small species are easy to preserve and keep
under local conditions and moreover are
eaten whole, providing a valuable source of
calcium and other minerals. Larger species,
such as the introduced Ni'ie perch in Lake
Victoria, cannot be easily preserved locally
and are in any case not eaten whole, leading
to a danger of calcium deficiency. Fisheries
for such species tend to become indus-
trialized or semi-industrialized, producing
fish products for commercial high-value

markets, often for export. While these may
improve the balance of payments for the
countries concerned, they may ultimately
worsen the nutritional status of local people.
Additionally, there are some indications that
fish populations in mixed species fisheries are
more stable over time, less susceptible to
boom and bust' than those based on a small
number of often introduced species.

It is difficult rigorously to assess the
condition of inland fish stocks because they
appear able to respond rapidly to changing
environmental conditions. However, there is a
consensus that, regionally, most stocks are
fully exploited and in some cases over-
exploited. Exploitation has become more
efficient because of new technologies, and
developing infrastructure has allowed easier
access to freshwater resources. Some stocks.

182 WORLD ATLAS OF BIODIVERSITY

Map 7.-4

Major areas of diversity of
selected inland water
crustacean groups

This map represents a
preliminary assessment of
areas believed to support
high diversity among three
of the several major
crustacean groups that
occur in inland waters,
taking into account species
richness and local
endemism

Note: For numbered
locations see Appendix 6.

Source: Compiled using information and
expertise provided by the lUCN/SSC
Inland Water Crustacean Specialist
Group; first published in WCt-tcV

especially in nver fisheries, appear to be in
decline, but this is seemingly a result mainly
of anthropogenic changes to the freshwater
environment.

In many parts of the world fishing has high
recreational value, as well as being a means
of food gathering. Locally, notably in the
Amazon basin and in parts of Southeast Asia,
capture for the ornamental fish trade may be
an important source of income with potential
impact on wild populations. Increasingly it is
becoming difficult to distinguish between
truly wild fish stocks and those that are
artificially managed or enhanced in some way.

Other harvested species

Other exploited animal groups in inland
waters are far less important globally than
finfishes, but may still be highly significant.

•N).

112

Apart from crustaceans and mollusks, men-
tioned above, these include: frogs (chiefly
family Ranidael, exploited for food; crocodil-
ians, hunted mainly for leather; freshwater
chelonians, taken for food and to a lesser
extent for medicinal purposes, particularly in
eastern Asia; waterfowl which are hunted for
recreation and for food; fur-bearing mam-
mals, such as beavers Castor spp., otters
(subfamily Lutrinael and muskrats {Ondatra
zibetfiicusand Neofiber atleni], taken for their
skins; and manatees (family Trichechidael,
taken mostly for food although also used non-
consumptively on a small scale for biological
control of weeds.

Rice is the principal cultivated wetland
plant of global importance to food security.
Most of the relatively few plants associated
with inland waters that are heavily exploited in

Inland water biodiversity 183

.^'

the wild state are also marginal or wetland
species. Some species le.g. Aponogeton, in
Madagascar) are collected for use as orna-
mentals; reeds are used as building materials
le.g. thatch); and some are collected for
food or as medicines le.g. Spirulina algael.
Rhizomes, tubers and seeds [rarely leaves!
of aquatic and wetland plants are used as a
food source, mainly in less-developed regions
where they can be important to food security
in times of shortage, but globally they make
a relatively minor contribution to human
nutrition, fvlost important are some forms of
edible aroid lAraceae), notably some cultivars
of Colocasia Itarol and the giant swamp taro
Cyrtosperma chamissonis that grow in flood-
ed conditions and are important food crops in
the Caribbean. West Africa and the Pacific
islands. Conservation and collection of wild

forms of these is considered a high priority.
Sago palms Metroxylon spp. in Southeast
Asia and the Pacific and watercress Rorippa
nasturtium-aquaticum in Europe are other
examples of cultivated aquatic plants, the wild
relatives of which merit conservation. Aquatic
plants have been widely used for medicinal
purposes, documented for at least two
millennia, but such use appears at present to
be minor and probably of real significance in
few areas. However, interest in ornamental or
aquarium water plants is widespread and of
some economic importance.

OTHER MAJOR IMPACTS

Physical alteration and destruction of habitat

Destruction of inland water ecosystems is
most simply effected by the removal of water
Although humans have always made use of

184 WORLD ATLAS OF BIODIVERSITY

freshwater systems, the last 200 years
(spanning the Industrial Revolution, the growth
of cities, the spread of high-input agriculture!
have brought about transformations on an
unprecedented scale. The global rate of water
withdrawal rose steeply at the start of the 20th
century, and further after mid-century. Major
changes in the distribution of water have
resulted mainly from withdrawals for Irrigation
and secondarily from domestic and industrial
use. It has been estimated that humans use 26
percent of the total evapotranspiratlon from

Aral Sea"

Until the mld-20th century the Aral Sea was the world's fourth largest inland water body
(after the Caspian Sea, Lake Superior and Lal<e Vlctorial. Located within a catchment area of
some 1 .9 million l<m' extending over six countries, the Aral is fed by two major rivers, the
Amu Darya, rising in the Pamir, and the Syr Darya, rising in the Tien Shan. Starting in the
19605, excess water withdrawal from these rivers, primarily for cotton irrigation, has
severely affected the Aral Sea. Its area had reduced from more than 65 000 km' to about
28 500 km' In 1998, with volume falling by 75 percent, water level falling by around 20
meters (ml, and salinity greatly Increasing. In consequence, problems with drinking water
quality and availability and with dustborne pollutants, have severely affected the health
status of the human population; the commercial fishery has collapsed; waterlogging and
salmlzation have degraded agricultural lands; and the deltaic marshlands of the two feeder
rivers have largely been replaced by sandy drylands.

Mesopotamia""

Serial satellite images confirm a loss of around 90 percent of the lakes and marshlands in
the lower fvlesopotamlan wetlands during the last three decades. The only significant
permanent marshland remaining is in the Al-Hawlzeh region. The large number of dams
now present on upstream parts of the Tigris-Euphrates system may have contributed to this
loss, but it appears to be pnmarlly the result of major hydrologlcal engineering works In
southern Iraq, notably the completion of the major outfall drain (or 'third river'] which diverts
water to the head of the Gulf This loss has placed further pressure on the t^la'dan (l»1arsh
Arabs!, now largely displaced within Iraq or in refugee camps In Iran. Recent Information Is
scarce, but biodiversity In the region will inevitably have been affected, probably including the
endemic form of smooth-coated otter iLutrogale perspiciUata, an otherwise oriental species,
assessed as globally threatened!.

Azraq oasis''

Groundwater extraction for urban needs in Jordan rose from about 2 million m' in 1979 to
about 25 million m' in 1993, with an additional 25 million m' per year used for agricultural
Irrigation. The important Azraq wetlands natural reserve, formerly extending over some
12 000 hectares, and a vital staging site for bird migrants, now supplies around one quarter
of Amman's water needs, and as a consequence has lost most of its marshland and migrant
bird populations.

land surfaces and 54 percent of the accessible
runoff". Unregulated withdrawal can lead to the
wholesale destruction of inland water eco-
systems, as has occurred with the Aral Sea In
central Asia (Box 7.2!. Similarly, many wetlands
have been completely destroyed by drainage,
often for conversion to agriculture. Other fac-
tors can modify or destroy particular habitats
within Inland water ecosystems. For example,
canalization, usually undertaken to improve
navigability, generally destroys riparian (shore-
line! habitats while flood-control systems
drastically alter regimes In floodplalns".

Dams and reservoirs

Dams, particularly large dams, have a major
impact on the rivers on which they are
built. They affect flow regimes, often dram-
atically, destroy large areas of existing habitat
(while at the same time creating new ones!
and can catastrophlcally disrupt the life cycles
of species that migrate up and down rivers".
Large dams are unevenly distributed across
the world's major catchments, with a partic-
ularly high concentration in North America,
especially within the contiguous states of
the United States, where at least eight catch-
ments have In them more than 100 large
dams each. In contrast, small and medium-
sized dams, which may cumulatively have as
substantial an Impact, are concentrated In
eastern Asia, particularly China"'. Dams may
be primarily for the generation of hydro-
electric power, or to create reservoirs for the
storage of water, or both. The size of a dam is
not necessarily directly related to the area or
volume of the Impoundment created or to its
downstream impact.

Pollution and water quality

Assessment of anthropogenic changes in
water quality Is not always easy as such
changes are Invariably superimposed on
natural background variations. Historically, a
similar sequence of water quality issues has
became apparent In both Europe and North
America during rapid socioeconomic develop-
ment over the past 150 years. Problems of
fecal and other organic pollution were evi-
dent In the mld-19th century, followed by
salinization, metal pollution and eutrophi-

Inland water biodiversity 185

^^\

cation in tlie first lialf of tfie 20tfi century, with
radioactive waste, nitrates and other organic
micropoUutants, and acid rain most promin-
ent in recent decades. Newly industrializing
countries are likely to face these problems
over a much more compressed period, and
typically without the capacity to monitor and
analyze water quality, or manage water use
appropriately^l

Different kinds of pollutants appear to
affect different classes of water system to
differing extent^'. With regard to quality for
human use, contamination by pathogens of
fecal origin is the major problem in river
systems, and eutrophication probably the
most widespread problem affecting lake and
reservoir waters" "".

Acid deposition through precipitation has
been recognized as a regional transboundary
phenomenon since the 1960s. Industrial
emissions of sulfur and nitrogen oxides ISO2,
NOJ, mainly a result of fossil fuel combustion,
are the principal source of acid rain. Most
evidence of acid rain and its effects relates to
North America and Europe, but emission rates
are rising steeply in rapidly industrializing
countries elsewhere. Acid rain in one country
may be a consequence of compounds released
into the atmosphere by industry in another
country hundreds of kilometers distant. The
geology, soil and vegetation of drainage basins
strongly influence the acidification process.
Acid rain has been shown to decrease species
diversity in lakes and streams but has not
been implicated in any recorded species
extinction or any major species decline. It has
not yet been shown to be a significant issue in
tropical freshwaters, where global freshwater
diversity is concentrated'.

Sedimentation

Removal or extension of forest cover, or any
anthropogenic interference with soils and
landcover (e.g. agriculture, urbanization, road
construction, mining], modifies the rate of
runoff from catchment slopes and also the
density of particles carried in the drainage
system. All moving waters carry some mass
of suspended material, and there is consider-
able natural variation in this in space and
time, but logging can increase sediment load

by up to 100 percent for a short period, and
20-50 percent over the longer term. Sediment
reaching lakes will be deposited and in effect
enter long-term storage; depending on water
velocity, sediment in rivers will settle out on
floodplains or other parts of the course, or be
carried into the coastal marine environment.
Increased sedimentation can have several
effects on aquatic biodiversity: deposition can
radically change the physical environment of
species restricted to particular conditions of
depth, light penetration and velocity; it is a
major carrier of heavy metals, organic pollu-
tants, pathogens and nutrients; and it can
interfere mechanically with respiration
In gill-breathing organisms'. The endemic

cichlid fishes of the African Great Lakes rely
on complex visual signals in breeding, and
reduced water clarity because of sedimen-
tation is suspected to be affecting their
breeding success.

Introduced species

Unplanned or poorly planned introduction of
non-native species and genetic stocks is a
major threat to freshwater biodiversity. Such
introductions can have negative or positive
effects on fishery production; it is a reasonable
assumption that all successful introductions
will have an impact on existing population size
and community structure, and many changes
are likely to be undesirable''. The incomplete

Although some factories
have introduced cleaner
production procedures,
industrial emissions are
still the principal source
of acid ram.

18« WORLD ATLAS OF BIODIVERSITY

1 000

Decade total

■ ■

•^^

■^

Cfl

-o

CT-

CNJ

-<f

tr^

cr-

Cr-

c>-

Figure 7.2
Inland water fish
introductions

Source: Welcomme .

Figure 7.3

Fresliwater population
trends

Note: A simplified
representation of tfie average
population change in a
sample of 194 inland water
species, see text.

information available suggests tfiat althougfi a
significant number of fish introductions took
place during the 19th century, the three dec-
ades starting with the 1950s were particularly
important (Figure 7.21. A classic example of
the effect of introduced species is the impact
of the Nile perch Lafes niloticus on the
haplochromine cichlids of Lake Victoria
discussed further below.

Several species of aquatic plant, in
particular free-floating species that are able
to spread rapidly by vegetative growth (most
notoriously the South American water
hyacinth Eichhornia crassipes], but also other
forms, have dispersed widely over the globe
and become major pest species. They block

1970

1975

1980

1985

1990

1995

1999

drainage channels, sluices and hydroelectric
installations, impede boat traffic and hinder
fishing. In recent decades the question of how
best to control or eradicate pest species has
been the foremost issue in conservation and
management of aquatic plants'.

THE CURRENT STATUS OF INLAND WATER
BIODIVERSITY

As with marine species, assessment of the
status of wholly aquatic inland water species
is hampered by difficulties of direct obser-
vation. However, because these species also
in general have far more restricted ranges
than marine species, it is easier to infer their
status from assessment of habitat condition
and from sampling efforts. Amphibious or
surface-dwelling species may be relatively
easier to monitor. Where such species are of
economic importance - as for example with
those European and North American water-
fowl that play a role in the recreational
hunting industry - they may be among the
best monitored of all wild species.

Threatened and extinct species

In the few cases where elements of inland
water faunas - usually fishes - have been
studied in any detail, it has generally been
found that more species than suspected are
threatened or cannot be re-recorded'°"".
Among the 20 or so countries where the entire
inland water fish fauna has been evaluated,
an average of 17 percent of the species are
regarded as globally threatened (categories
critically endangered', endangered' or
vulnerable' in the lUCN (World Conservation
Union) threat categorization system"! (Table
7. 81 A far larger proportion is likely to be in
local decline, although not in danger of global
extinction. The proportion of inland water
chelonians that is believed threatened is even
higher: 99 such species were categorized as
threatened in 2000, equivalent to about 60
percent of the number of inland water chelon-
ians listed in Table 7.6.

Amongst mammals and birds the pro-
portions are considerably lower, probably
because many semi-aquatic species are able
to disperse from one inland water body to
another relatively easily. Nevertheless, some-

Inland water biodiversity 187

wtiat more species ttian average for the
groups as a wtiole are regarded as threatened.
Table 7.9 shoves the category and taxonomic
distribution of threatened inland water
vertebrates.

Two groups of species, the Lake Victoria
cichlid fishes and the Mobile Bay drainage
gastropod moUusks, serve as exemplary case
studies illustrating the major threats faced by
inland water biota worldwide.

Lake Victoria, the largest tropical lake in
the world, provides a classic example of the
potential negative impacts of species intro-
ductions. Until some 30 years ago, when the
large top predator, the Nile perch La(es
niioticus, was introduced, the lake supported
an exceptional 'species flock' of more than 300
species of haplochromine cichlid fishes, as
well as smaller numbers from other families.
The cichlids are of enormous interest in the
study of evolutionary biology. Not all the
species have yet been formally described;
many of these are known among aquarists and
others only by informal common names. At
least half of the native species are believed
to be extinct or so severely depleted that too
few individuals exist for the species to be
harvested or recorded by scientists.

Predation by the Nile perch is believed to
be the major cause of this decline, but
important additional factors include increas-
ing pollution and sediment load, excess fishing
pressure, and possible competition from intro-
duced tilapiine cichlids. The lake itself has
now become depleted of oxygen, and a shrimp
tolerant of oxygen-poor waters provides a
major food source for the Nile perch. In recent
years the Nile perch and one of the introduced
tilapiines have formed the basis of a high-
yielding fishery and an important national and
export trade. However, it is thought unlikely
that such high yields will be maintained
because of continued overfishing and the
suspected instability of the already highly
disturbed lake ecosystem'.

Dam construction is the prime cause of
extinction in the gastropod fauna of the Mobile
Bay drainage in Alabama, United States.
Historically, the freshwater snail fauna of
Mobile Bay basin was probably the most
diverse in the world, followed by that of the

Mekong River Nine families and about 118
species were known at the turn of the century
to occur in the Mobile Bay drainage. Several
genera and many species were endemic,
particularly in the Pleuroceridae. Recent
surveys suggest at least 38 species are extinct
(32 percenti; decline in species richness
ranges between 33 percent and 84 percent in
the main river systems. The richest fauna was
in the Coosa River and this system has
undergone the greatest decline (from 82 to 30
species]. Almost all the snail species
presumed extinct were members of the
Pleuroceridae and grazed on plants growing
on rocks in shallow oxygen-rich riffle and
shoal zones. The system has 33 major hydro-
electric dams and many smaller impound-
ments, as well as locks and flood control
structures. A combination of siUation behind
dams and submergence of shallow water
shoals has removed the snails' former
habitat. Where habitat remains it has dimin-
ished in area and become fragmented".

The inland water living planet index

An impression of the overall trend in a large
sample of species for which indicators of

Table 7.8

Numbers of threatened
freshwater fishes in
selected countries

Notes: These are the 20
countries whose fish faunas
have been evaluated
completely, or nearly so, and
which have the greatest
number of globally
threatened freshwater fish
species. The estimates of
total fish species present are
all approximations.

Source: Total species estimates from
UNEP-WCMC database; threatenetj
species (Jata from online Red List
fitfpV/www redlist.org [accessed
Marcfi 20021

I^B

Total species

Threatened species

% threatened

USA

822

120

Mexico

384

Australia

216

South Africa

Croatia

Turkey
Greece

174
98

22
19

13
19

Madagascar
Canada

41
177

13
12

Papua New Guinea
Romania

195
87

11
11

6
13

Italy
Bulgaria

45
72

11
11

24
15

Hungary

Spain

Moldova

79
50
82

10
10
9

13
20
11

Portugal
Sri Lanka

28
90

9
9

32
10

Slovakia

Japan

150

188 WORLD ATLAS OF BIODIVERSITY

Table 7.9

Taxonomic distribution
and status of threatened
inland water vertebrates

Notes: Family selection
based on Table 7 6.
Numbers refer to species
(not subspecies and
geographic populations]
recorded in the Red List
database as occurring in
freshwater; only the birds
and mammals have been
comprehensively assessed
lor species at risk.

i Emydidae here includes
batagurme turtles
sometimes treated as a
separate family (several of
the species are primarily
terrestrial not aquaticl.

Source: Status categories from 2000
Red List database, www.redlistorg
iaccessed February 2002!

Phylum and

Family

Common name

Critically

Endangered'

'Vulnerable- 1

class or order

endangered'

Craniata - fishes

Petromyzontiformes

Lampreys

Carchariniformes

Ground stiarks

Prisliformes

Sawfish

Myliobatiformes

Rays

Acipenseriformes

Sturgeons

Alheriniformes

Silversides

Beloniformes

Needlefishes, sauries, etc.

Characiformes

Characins

Clupeiformes

Herrings and anchovies

Cypriniformes

Carp, minnow, loaches

117

Cyprinodontiformes

Rivulines, ((illifish, etc.

Gasterostiformes

Sticl<lebacks

Ophidiiformes

Pearlfishes, etc.

Osteoglossiformes

Bonylongues

Perciformes

Perches, etc.

104

Percopsiformes

Trout-perches, cavefishes

Salmoniformes

Salmonids

Scorpaeniformes

Gurnards, scorpionfishes, etc.

Silunformes

Catfishes

Synbranchiformes

Swamp eels

j^_

Syngnathiformes

Pipefishes, seahorses, etc.

Craniata - Amphibia

Caudate

Cryptobranchidae

Giant salamanders and hellbenders

1^^

Proteidae

Ivtudpuppies and olm

Anura

Pipidae

Clawed frogs and pipid toads

Craniata - Reptilia

Chelonia

Carettochelidae

Pig-nosed soft-shelled turtle

Chelidae

Austro-american side-necked turtles 3

- 7

Chelydridae

Snapping turtles

Dermatemydidae

Central American river turtle

Emydidae'

Pond and river turtles

population change are available can be
derived from the WWF living planet index.
This method is designed to represent the
change in the average species' in the sample
from one five-year interval to the next,
starting in 1970. The inland waters sample
represents 194 species of mammals, birds,
reptiles and fishes, and the index suggests a
significant declining trend over the last three
decades of the 20th century (Figure 7.31".

The sample includes a large number of
wetland and water margin species in addition
to truly aquatic forms. The declining trend in
the inland water sample is a little steeper
than the equivalent marine index, and sub-
stantially steeper than the terrestrial index.

Assessment of the status of inland water
ecosystems

Indicators of habitat condition in river
catchments

While the living planet index methodology
provides an indication of global trends in inland
water biodiversity, an alternative approach
aims to assess the overall condition of inland
water ecosystems. In one approach to a global
assessment' two high-order indicators of likely
habitat condition in different river catchments
were combined. First, a wilderness measure
for each river catchment was calculated using
the wilderness index methodology developed
by the Australian Heritage Commission""
(see Chapter 4 and particularly Ivlap 4.51. This

Inland water biodiversity 189

■\V

E Phylum and

Family

Common name

'Critically

'Endangered'

'Vulnerable'

■ class or order

endangered'

Kinosternidae

Mud and musk turtles

Pelomedusidae

Side-necked turtles

Trionychidae

Soft-shelled turtles

Crocodilia

Alligatondae

Caimans and alligators

Crocodylidae

Crocodiles

Gavialidae

Ghanal and false gharial

Craniata - Aves

Anseriformes

Dendrocygnidae

Whistling-ducks

Anatidae

Ducks, swans and geese

Charadriiformes

Charadriidae
Laridae
Rhynchopldae
Scolopacidae

Plovers, etc.

Gulls, terns, skua, auks

Skimmers

Curlews, etc.

1
1

2
1
1

Ciconiiformes

Ardeidae

Herons, egrets

Ciconiidae

Storks

Threskiornithidae

Ibis, spoonbill

Phoenicopteridae

Flamingos

Gruiformes

Heliornithidae

Limpkin and sungrebes

Rallidae

Rails, gallinules and coots

Pelicaniformes

Pelecanidae

Pelicans and shoebill

Passeriformes

Cinciidae

Dippers

Podicipediformes

Podlcipedidae

Grebes

Craniata - Mammalia

Artiodactyla

HIppopotamldae

Hippopotamus

Carnivora

Mustelidae
Phocidae

Mustellds, otters
Earless seals

Cetacea

Platanistidae

River dolphins

Insectivora

Soricidae

Shrews

Talpidae

Moles and desmans

Tenrecidae

Tenrecs and otter shrews

Rodentia

Murldae

Mice, voles, etc.

Sirenia

Trichechldae

Manatees

Figure 7.4

River basin richness and

vulnerability

Notes: Eacti symbol
represents a river basin
scored on fish family
richness and basin
vulnerability: the darker
symbols score high on both
counts and may be regarded
as high priority. See Map 7,5
and Table 7,10.

Source: WCMC'

provides a measure of the spatial extent of
human impacts in the land area of the
catchment as a whole and does not directly
reflect the condition of riverine ecosystems.
Secondly, a national water resource vulnera-
bility index IWRVIl" developed on the basis of
three water resource stress indices (reliability,
use-to-resource and coping capacity) was
resolved at catchment level Iby measuring that
proportion of each catchment that lies within
any given country and weighting this proportion
by the national WRVIl. This provides an indirect
measure of vulnerability tor each catchment.
By normalizing and combining these two
measures, a single value representing vulner-
ability or stress level for each major river

S 10

-10

-20
-0

t • •

••

•

rf>

r •

.8 -0

6 -0

4 -0

2 (

)

6 0.8

Overall vulnerability

190 WORLD ATLAS OF BIODIVERSITY

Map 7.5

Priority river basins

A possible scheme for
prioritizing river basins at
global level Major
catchment basins are
assessed for fish diversity
lat family level) and for
vulnerability (a combined
indication of disturbance
and potential water stress].
Systems with both high
diversity and vulnerability
are proposed as high
priority for investment and
management action; those
with low diversity and little
disturbance are lower
priority. See text and source.

Source: WCMC'.

Table 7.10

Thirty high-priority river

basins

Notes: These are 30 river
basins that support high
biodiversity (assessed as fish
family richness! and are most
vulnerable to future
pressures (have a low
wilderness score, and are
high on the water resource
vulnerability index). See text
for further explanation and
Figure 7.4. Basins are listed
in alphabetical sequence.

Priority

High

tuledium

Low

catchment can be calculated. Overall, the
pattern mapped agrees well with what might
intuitively be expected, with few evident
anomalies. Globally, the most stressed catch-
ments are to be found in South Asia (the Indian
subcontinent), the Middle East and western
and northcentral Europe. The least stressed

^^^^^^^^^^I^^I^^^^^^H

Irrawaddy

Nile

Senegal

Cauvery

Krishna

Pahang

Sittang

Ctiao Phraya

Parana

Song Hong (Red)

Gambia

Magdalena

Parnaiba

Tapti

Ganges-

Mahanadi

Penner

Tembesi-Hari

Brahmaputra

Mekong

Peral<

Uruguay

Godavari

Narmada

Salween

Volta

Indus

Niger

Sao Francisco

are those in the northwestern part of North
America. Further refinements of this analy-
sis would involve applying water resource
vulnerability measures for individual catch-
ments" rather than to countries and incor-
porating measures of direct impacts on inland
waters, in particular water quality and the
number and kind of dams.

The number of freshwater fish families
present in each basin can be calculated from
the family density surface (Map 7.1) in order to
provide an indication of biodiversity value.
Plotting family number against vulnerability
allows the basins with high diversity and high
vulnerability to be identified, and these can
reasonably be regarded as global priorities
for management intervention designed to
minimize biodiversity loss (Figure 1 A, Table
7.10, Map 7.5).

Inland water biodiversity 191

Habitat condition in lakes
One semi-quantitative study attempted to
evaluate the ctianging condition of freshwater
lakes during the last three decades of the 20th
century^'. A baseline was provided by Project
Aqua, a project initiated by the Societas
Internationalls Limnologlae In 1959, which
collated and later published" Information
provided by national and regional specialists in
relation to more than 600 water bodies. Many
of these lakes were treated in later Infor-
mation sources relating to the 1980s and
1990s", and In some 93 cases it was possible
to make an assessment that, although
Imprecise, is likely to be indicative of changing
conditions. Each lake was scored according to
whether Its condition appeared to have
deteriorated (or Impacts Increased], or to have
Improved, or whether no change (or no new

information) was reported. Improvement was
reported In a very small number of lakes, but
the overwhelming trend was for a deterior-
ation In conditions (see Figure 7.5|.

Figure 7.5

Changes in condition of a
sample of freshwater lakes
between 1950s and 1980s

Note: Number following area
name is number of lakes In
sample.

Source: various sources

Better

Not known

Worse

100
%
80

Africa
(201

Asia
(241

Central & South Australia
America (91 (3|

Europe
(371

TOTAL
(931

192 WORLD ATLAS OF BIODIVERSITY

REFERENCES

1 WCMC 1998. Freshwater biodiversity: A preliminary global assessment. By Groombridge,

B. and Jenkins, M. WCMC-World Conservation Press, Cambridge.

2 Gorthner, A. 1994. What is an ancient lake? Speciation in ancient lakes. In: Martens, K.,

Goddeens, B. and Coulter, G. (eds). Archiv fijr l-lydrobiologie. Ergebnisse der Limnologie
UU: 97-100.

3 Home, J. A. and Goldman, C.R. 1994. Limnology. McGraw-Hill Inc., New York.

A Guegan, J. -P., Lek, S. and Oberdorft, T. 1998. Energy availability and habitat heterogeneity
predict global riverine fish diversity. Nature 391 : 382-384.

5 Oberdorff, T., Guegan, J.-F. and Hugueny, B. 1995. Global scale patterns of fish species

richness in rivers. Ecograp/iy 18: 345-352.

6 Kottelat, M. and Whitten, T. 1 996. Freshwater biodiversity in Asia with special reference to

fish. World Bank Technical Paper No. 343.

7 Olson, D.M. et al. (edsl 1 997. Freshwater biodiversity of Latin America and the Caribbean: A

conservation assessment. Proceedings of a workshop. World Wildlife Fund, Washington DC.

8 Abell, R. et al. 1 998. A conservation assessment of the freshwater ecoregions of North

America. Final report submitted to the US EPA, April. World Wildlife Fund-US.

9 Olson, D.M., and Dinerstein, E. 1998. The Global 200: A representation approach to

conserving the earth's distinctive ecoregions. Conservation Biology M: 502-515.

10 WWF. Online at http://www.panda.org/resources/programmes/global200/pages/list.htm
(accessed March 20021.

11 Watter, G.T. 1992. Unionids, fishes, and the species-area curve. Journal of Biogeography
19:481-490.

12 Tittley, I. 1992. Contribution to chapter 7, Lower plant diversity. In: World Conservation
Monitoring Centre. Global biodiversity. Groombridge, B. led. I Chapman and Hall, London.

13 Goh, T.K. and Hyde, K.D. 1996. Biodiversity of freshwater fungi. Journal of Industrial
ivlicrobiologyM: 97-100.

14 Shearer, C.A. 1993. The freshwater ascomycetes. Nova Hedwigia 56: 1-33.

15 Shearer, C.A. Freshwater ascomycetes and their anamorphs. Available online at
http://fm5web.life.uiuc.edu:23523/ascomycete/ (accessed March 20021.

16 Sculthorpe, CD. 1967. The biology of aquatic vascular plants. Edward Arnold, London.

17 Palmer, M. et al. 1997. Biodiversity and ecosystem processes in freshwater sediments,
/^mfa/o 26(81: 571-577.

18 Nelson, J.S. 1994. Fishes of the world. 3rd edition. John Wiley and Sons, Inc., New York.

19 UN (CSDI 1997. Comprehensive assessment of the freshwater resources of the world.
Economic and Social Council. Fifth Session, 5-25 April. E/CN. 17/1997/9. Available online at
http://www.un.org/esa/sustdev/csd.htm (accessed April 20021.

20 FAO 1996. Food production: The critical role of water. Technical background document.
World Food Summit. Available online in pdf at http://www.fao.org (accessed April 20021.

21 FAO 2000. The state of world fisheries and aquaculture 2000. Fisheries Department, Food
and Agriculture Organization of the United Nations, Rome. Available online at
http://www.fao.org/DOCREP/003/X8002E/X8002E00.htm (accessed April 2002).

22 FAO 1999. Review of the state of world fishery resources: Inland fisheries. FAO Fisheries
Circular No. 942. FIRI/C942.

23 Coates, D. 1995. Inland capture fisheries and enhancement: Status, constraints and
prospects for food security. Paper presented at international conference on sustainable
contribution of fisheries to food security. Kyoto, Japan, 4-9 December 1995. Food and
Agriculture Organization of the United Nations, Rome. KC/FI/95/Tech/3.

24 Borgstrbm, R. 1994. Freshwater ecology and fisheries. In: Balakrishnam, M., Borgstrom, R.
and Bie, S.W. Tropical ecosystems, pp. 41-69. International Science Publishers. Oxford and
IBH, NewDehti.

Inland water biodiversity 193

25 Scudder, T. and Conelty, T. 1985. Management systems for riverine fisheries. FAO Fisheries
Technical Paper No. 263. Food and Agriculture Organization of the United Nations. Rome.

26 Munthali, S.M. 1997. Dwindling food-fish species and fishers' preference:
Problems of conserving Lal<e Malawi's biodiversity. Biodiversity and Conservation 6:
253-261.

27 Postel, S.L., Daily, G.C. and Ehrlich, PR. 1995. Human appropriation of renewable fresh
water Science 271; 785-788.

28 L'Vovich, M.I. et al. 1990. Use and transformation of terrestrial water systems. In: Turner
II, B.L. Ttie eartti as transformed by tiuman action. Global and regional ctianges in the
biosphere over the past 300 years, pp. 235-252. Cambridge University Press with Clark
University.

29 FAO AQUASTAT. Online at http://www.fao.org/landandwater/aglw/aquastat/
reglons/fussr/mainS.htm (accessed March 20021.

30 Evans, M.I. 2001. The ecosystem. Draft paper for conference, Iraqi Marshlands: Prospects,
London, 21 May 2001. Available online at http://www.amarappeal.com/documents/
Draft_Report.pdf (accessed April 20021.

31 UNEP 2001. UN study sounds alarm about the disappearance of the Mesopotamian
marshlands. 18 May Available online at http://www.grida.no/inf/news/news01/
news42.htm. and also see http://www.grid. unep.ch/./activities/sustainable/
tigris/marshlands/marshlands.pdf (accessed April 20021.

32 Fariz, G.H. and Hatough-Bouran, A. 1998. Population dynamics in arid regions: The
experience of the Azraq Oasis Conservation Project (Case study: Jordan). In: de Sherbmin,
A. and Dompka, V. (eds). Water and population dynamics: Case studies and policy
implications. Available online at http://www.aaas.org/international/ehn/waterpop/
front.htm (accessed April 20021.

33 Dynesius, M. and Nilsson, C. 1994. Fragmentation and flow regulation of river systems in
the northern third of the world. Science 266: 735-762.

34 Avakyan, A.B. and lakovleva, V.B. 1998. Status of global reservoirs: The position in the late
twentieth century Lakes and Reservoirs: Research and /Management 3: 45-52.

35 Meybeck, M. and Helmer, R. 1989. The quality of rivers: From pristine stage to global
pollution. Palaeogeography, palaeoclimatology. palaeoecology 75: 283-309.

36 Chapman, D.V. 1992. Water quality assessment. Chapman and Hall, London.

37 UNEP 1991. United Nations Environment Programme. Freshwater pollution. UNEP/GEMS
Environment Library No. 6.

38 UNEP 1995. United Nations Environment Programme. Water quality of world river basins.
UNEP Environment Library No. 14.

39 Moyle, P.B. 1996. Effects of invading species on freshwater and estuarine ecosystems. In;
Sandlund, O.T. ef al. (edsl. Proceedings of the Norway/UN Conference on Alien Species,
Trondheim, 1-5 July 1996. pp. 86-92. Directorate for Nature Management and Norwegian
Institute for Nature Research. Trondheim.

40 Moyle, P.B. and Leidy, R.A. 1992. Loss of biodiversity in aquatic ecosystems: Evidence from
fish faunas. In; Fielder, PL. (eds). Conservation biology, the theory and practice of nature
conservation, preservation and management, pp. 129-169. Chapman and Hall, New York
and London.

41 Stiassny, M. 1996. An overview of freshwater biodiversity with some lessons learned from
African fishes. Fisheries 2'\{9]: 7-13.

42 Reinthal, P.N. and Stiassny, M.L.J. 1991. The freshwater fishes of Madagascar: A study of
an endangered fauna with recommendations for a conservation strategy. Conservation
Biology 5[2]: 231-243.

43 Kirchhofer, A. and Hefti, D. (eds) 1996. Conservation of endangered freshwater fish in
Europe. Advances in Life Sciences. Birkhauser Verlag, Basel.

194 WORLD ATLAS OF BIODIVERSITY

44 HiUon-Taylor. C. Icompiler] 2000. 2000 lUCN Red List of threatened species. lUCN-the
World Conservation Union, Gland and Cambridge. Also available online at
http://www.redlist.org/ (accessed April 20021.

45 Bogan, A.E., Pierson, J.M. and Hartfield, P. 1995. Decline in the freshwater gastropod
fauna in the Mobile Bay basin. In: LaRoe, E.T. et al. (edsl. Our living resources: A report to
the nation on the distribution, abundance, and health of U.S. plants, animals, and
ecosystems, pp. 240-252. US Department of the Interior, National Biological Service,
Washington DC.

46 Loh, J. led.l 2000. Living planet report 2000. WWF - World Wide Fund for Nature, Gland.

47 Lesslie, R., in litt., 30 May 1998.

48 Lesslie, R. and Maslen, M., 1995. National wilderness inventory, handbool< and
procedures, content and usage. 2nd edition. Commonwealth Government Printer,
Canberra.

49 Raskin, P. et al. 1997. Water futures: Assessment of long-range patterns and problems.
Background Report No. 3 of Comprehensive Assessment of the Freshwater Resources of
the World. Stockholm Environment Institute.

50 Alcamo, J., Henrichs, T. and Rosch, T. 2000. World water in 2025. Global modeling and
scenario analysis for the World Commission on Water for the 21st Century. Kassel World
Water Series. Report No. 2. University of Kassel, Centre for Environmental Systems
Research.

51 Loh, J. etal. 1998. Living planet report 1998. WWF-World Wide Fund for Nature, Gland.

52 Luther, H. and Rzoska, J. 1971. Project Aqua: A source book of inland waters proposed for
conservation. IBP Handbook No. 21. lUCN Occasional Paper No. 2. International Biological
Programme. Blackwell Scientific Publications, Oxford and Edinburgh.

53 For example, Scott, D.A. led.) 1989. A directory of Asian wetlands. lUCN, Gland and
Cambridge; Scott, D.A. and Carbonell, M. 1986. A directory of neotropical wetlands. lUCN,
Gland and Cambridge; International Lake Environment Committee (ILEC), World lakes
database, available online at http://www.ilec.or.jp/database/database.html

(accessed March 2GQ2I.

54 Anon 1978. Centre for Natural Resources, Energy and Transport of the Department of
Economic and Social Affairs, United Nations. Register of International Rivers. Water
Supply and Management!: 1-58. (Special issue). Pergamon Press.

55 Shiklomanov, I. A. 1998. World water resources at the beginning of the XXlst century.
Report to UNESCO (Division of Water Sciences). Available online in summary form at
http://webworld.unesco.org/water/ihp/db/shiklomanov/summary/html/summary.html
(accessed March 2002).

56 Martens, K., Goddeeris, B. and Coulter, G. (eds) 1994. Speciation in ancient lakes. Archiv
fur l-lydrobiologie. Ergebnisse der Limnologie 44.

57 Hutchinson, G.E. 1993. A treatise on limnology. Vol. IV. The zoobenthos. John Wiley and
Sons, Inc., New York.

58 Eschmeyer, W.N. ef al. 1 998. A catalog of the species of fishes. Vols 1 -3. California
Academy of Sciences, San Francisco. Available online in searchable format at
http://www.calacademy.org/research/ichthyology/catalog/fishcatsearch.html [accessed
January 20011.

59 Welcomme, R.L. (compiler) 1988. International introductions of inland aquatic species. FAO
Fisheries Technical Paper No. 294. Food and Agriculture Organization, Rome.

60 Berra, TM. 2001. Freshwater fish distribution. Academic Press, San Diego and London.

61 Johnson, T.C. et al. 1996. Late Pleistocene desiccation of Lake Victoria and rapid evolution
of cichlid fishes. Soence 273: 1091-1093.

62 Snoeks, J. 2000. How well known is the ichthyofauna of the large East African Lakes?
Advances in Ecological Research 31 : 17-38.

Global biodiversity: responding to change

Global biodiversity:
responding to change

CONCERN OVER THE RATE OF CHANGE IN THE BIOSPHERE, particularly In populations
of larger animals and In the naturalness of landscapes, dates primarily from the turn
of the 20th century. Although some early measures were taken, prompted by
pioneering conservation organizations, concerted International effort did not develop until
mid-century. Since this period, actions have tended to focus on conservation of Individual
species or of large areas of habitat, as national parks and other protected areas.

During the 1970s several International agreements aimed at conserving v/ild species and
habitats were agreed and entered Into force. During the 1980s the word 'biodiversity' was
coined, and a new paradigm formulated, aiming to Integrate biodiversity conservation with
sustainable human development. In 1992 the pivotal Earth Summit was held In Rio,
culminating In agreement on the Convention on Biological Diversity, and Agenda 21 - a plan
of action for sustainable human development.

After a decade of planning and Implementation, the status of some species and the
condition of some ecosystems have Improved. In some areas, this has been achieved by
restoration of degraded ecosystems, an approach likely to be of Increasing Importance In the
future. However, given future trends In human population growth and development. It seems
probable that pressures on biodiversity will continue to Intensify In coming decades. In many
parts of the world, the most significant challenge to conservation will be to minimize losses
of biodiversity while improving the livelihood of human populations, particularly those
experiencing severe poverty.

BIODIVERSITY CHANGE

Change is a donninant theme in the biosphere;
species have diversified and become extinct
throughout the history of life, and habitats too
have expanded and declined, along w/ith
changes in community composition, climate
and landforms. However, the recent rate of
change in global biodiversity appears higher
than that prevailing over most periods of
geological time. Earlier chapters have indi-
cated how in well-assessed groups of
organisms, such as birds and mammals, a
significant proportion of species now appears
to be threatened with global extinction.
Countless other species exist in reduced
numbers and as fragmented populations,
many of which are threatened with extinction
at national or more local scale. The rate of
extinction over the past few centuries, so far

as this can be reliably estimated, appears to
be much higher than the average background
rate estimated from the fossil record.
Whatever proximate causes may be impli-
cated, the increasing numbers and material
aspirations of the human species, the in-
creasing burden of waste and continuing
ineguitles in the distribution of wealth and
resources, together appear to drive most con-
temporary biodiversity loss.

Despite abundant evidence of change in
biodiversity, often Involving radical modifi-
cation of landcover or water bodies, much
of the relevant information on the status of
species or populations is qualitative or anec-
dotal in nature. At the level of individual
species, a qualitative assessment of trends In
numbers or range may be adequate evidence
of the need for management intervention, or a

196 WORLD ATLAS OF BIODIVERSITY

Concern about the decline
of species became the
focus of sustained
international attention only
in the last 30 years.

guide to its effectiveness, but it remains diffi-
cult to develop a comprehensive and quantit-
ative view of global species trends. The
relatively few large-scale population monitor-
ing programs that exist have tended to
concentrate on marine fishery stocks, on par-
ticular groups le.g. farmland birdsl or the
larger threatened species of animals. The res-
ulting series of data are not always directly
comparable, and often not amenable to pres-
entation in a manner appropriate for assisting
global and regional planning. Much recent
discussion' has focused on the design of
biodiversity indicators that could serve this
purpose, but few operational systems yet exist.
This chapter first provides an overview of
current approaches to conservation, focusing

on species, areas and ecosystems, and the
particular role of protected areas, as well as
the potential for ecological restoration. The
international dimension is then considered,
with specific reference to multilateral environ-
mental agreements and conventions relating
to biodiversity. Finally, changes in global
biodiversity that might be anticipated in future
are explored through description of recently
developed scenarios. Such changes highlight
the challenges facing current and future
efforts aimed at the conservation and sus-
tainable use of global biodiversity.

RESPONSES TO BIODIVERSITY CHANGE

Although concern about the decline of species
and loss of undisturbed habitats arose within
developed industrialized societies around the
start of the 20th century, they became subject
to sustained international attention only during
the 1970s. In general, major sectors that used
biodiversity, such as forestry and fisheries,
failed to incorporate consideration of bio-
diversity into their planning and regulations.
The issue was widely perceived as marginal to
other concerns, and conservation of nature
was commonly seen as an impediment to
human development.

The past three decades have been marked
by the emergence of concerted responses to
the crisis in biodiversity at all levels from the
local to the global. Civil society, largely in the
form of a hugely diverse and increasingly
sophisticated network of non-governmental
organizations INGOsI, has undoubtedly been
the driving force for most of this change (see
Box 8.11. Response to pressure from civil
society can be seen at both governmental and
intergovernmental levels.

Much of the progress that has been made in
recent decades can be attributed to the vision
articulated in the World Conservation Strategy
by the International Union for Conservation of
Nature and Natural Resources, the United
Nations Environment Programme and the
World Wildlife Fund In 19801 This document
helped set the conservation agenda during the
period leading up to the UN Conference on
Environment and Development (the Earth
Summit] in Rio de Janeiro in 1992. It em-
phasized conservation for development, and
embraced the notion of the sustainable use of
natural resources as a means of achieving
this. Maintenance of biological diversity as part
of a functioning biosphere was presented as a
fundamental prerequisite of sustainable hu-
man development, not an impediment to It. On
the ground, this was manifested in increasing
numbers of integrated conservation and de-
velopment projects In developing countries.
The Earth Summit was undoubtedly the major
environmental milestone of the 1990s, and
from it emerged the Convention on Biological
Diversity, which entered Into force in
December 1993.

Global biodiversity: responding to change

197

At the start of the 21st century concerns
about loss of species and habitats have
intensified. Additional issues have emerged,
such as climate change, with its inevitable
effects on coastal, montane and other eco-
systems, and the ability of humans directly to
modify gene structure and expression in living
organisms, with debate over the risks involved
when genetically modified organisms are re-
leased into the biosphere.

Protecting species

Maintaining biodiversity requires that viable
populations of diverse organisms are main-
tained in viable ecosystems. This may involve
activities carried out on site or off site. While
the latter lex situ] has an important role, as
with the seed banks and germplasm collec-
tions for agricultural plants, it can only have
very limited application. The former {in situ] is
essential for the vast majority of organisms,
and ecosystem conservation selt-evidently
depends entirely on maintaining environ-
ments in which communities of organisms
can interact and evolve.

Because the vast majority of the worlds
biodiversity exists within the territorial boun-
daries established by nations, most conser-
vation action is carried out within the policy
and legal systems established by national
governments lor in a few instances, by
regional or provincial governments). A wide
range of national measures exists, varying to
some extent from country to country depen-
ding on the social, political and economic
environment. Despite this variety, most such
measures involve regulation of the taking,
possession and trade in a set of species,
typically named and listed in legislation. The
protection of wild fauna has generally been
given much more attention than the pro-
tection of wild flora, but most modern
examples of species-specific legislation cover
both flora and fauna, as exemplified by the US
Endangered Species Act, passed in 1973.

The species approach has been supported
at the international level by the lUCN (World
Conservation Union] Red Data Book program
of activities and information tools. The focus is
on documenting and disseminating infor-
mation on species at risk of extinction, and on

Box 8.1 Pioneering NGOs in biodiversity conservation

Some of tfie earliest establisfied non-governmental organizations remain influential today.
The Sierra Club was founded in the United States in 1392, with a focus on maintaining
wilderness areas in North America. The Society for the Preservation of the Wild Fauna of the
Empire - now Fauna and Flora International IFFI) - was founded in 1903 in order to safeguard
southern Africa's declining large mammal populations and is now worldwide in scope.
Initiated in 1962, Operation Oryx was a landmark project of FFI, and averted extinction of the
Arabian oryx by means of a captive breeding and reintroduction program. The International
Council for Bird Preservation - now BirdLife International IBLII - was founded in 1922. It has
played a lead role in advancing standards for conservation information and has promoted
grassroots involvement in conservation through its worldwide membership structure.

The International Union for the Protection of Nature - now lUCN-The World Conservation
Union - was created in Switzerland in 1948. Sir Peter Scott became the chair of two of
lUCN's key commissions: on protected areas and on species survival (SSCI, and later of FFI,
and in 1966 he initiated the Red Data Book approach to documenting species at risk. The
World Wildlife Fund (WWFI - now known as WWF-World Wide Fund for Nature outside North
America - was designed originally to generate public contributions to support lUCN's work,
and was launched in 1961 with a campaign on black rhino. These two NGOs have been
driving forces in global policy development.

detailing the management steps - often in-
cluding preservation of important areas or
reduction in exploitation levels - needed to
reduce that risk. In some cases, targets and
indicators of progress may also be defined.

In practice, conservation attention has
tended to focus on species that are large,
charismatic and possibly also ecologically
important or highly threatened, or both.
The tiger Panthera tigris, Arabian oryx Oryx
[eucoryx, white rhinoceros Ceratotherium
simum and the blue whale Balaenoptera
musculus are familiar mammalian examples.
Considerable success has been achieved with
the species named above and the many others
(mainly animals) subject to similar conser-
vation efforts. However, the species approach
could never be extended to cover more than a
minute fraction of the approximately 300 000
larger organisms [plants and vertebrates) in
the biosphere, let alone the other several
million that probably exist. In fact, the primary
benefit may be that large organisms, and
terrestrial ve'^tebrates in particular, generally
require large areas of suitable habitat, and if
such areas can be managed to minimize risk,
other species may be safeguarded.

198 WORLD ATLAS OF BIODIVERSITY

Figure 8.1

Development of the global
network of protected
areas

Source: UNEP-WCMC database,
maintainted in collaboration with lUCN
World Commission on Protected Areas

2 500

2 000

1 500

1 000

500

Protecting areas

The regulation of access to and use of
particular areas has always been seen as
complementary to the focus on individual
species. Protected areas are in many ways the
most important form of legislative measure
for the conservation of biodiversity. Whereas
the initial purpose of many such areas was to
protect spectacular scenery and provide
recreational facilities, the concept evolved to
encompass habitats of threatened species
and ecosystems rich in biodiversity.

By the beginning of the 20th century many
countries had either already established
protected areas or were contemplating doing
so. The concept, however, was slow to be put
into practice and it was not until the 19^0s
that protected areas were being established
in any significant number The rate at which
land was being incorporated into the system
did not increase markedly until the early
1960s. The first World Parks Congress held in
Seattle, in the United States, in 1962 was an
important stimulus for the increase. This
meeting signified the emergence of the mod-
ern protected area network with over 80
percent of the worlds protected areas being
established since thenl The cumulative and
periodic growth of world protected areas are
plotted in Figure 8.1 (the creation of the
Greenland National Park in 1974, covering
some 97 million hectares, and the Great

BARS:
Number of sites

lover 1 000 ha. lUCN categories I to VII
gazetted over a five-year period
(left-hand scale!

LINE:

Cumulative protected area

[right-hand scale!

UOOO

12 000

- 10 000

-8 000

6 000

4 000

2 000

1970

1975

1980

1985

1990

1995

2000

Barrier Reef Marine Park in the 1980s,
extending over around 34 million hectares,
have had marked individual effects on the
global totals). The location of protected areas
in the lUCN/WCPA (World Commission on
Protected Areas! categories l-VI greater than
100 000 hectares in extent is shown in
Map 8.1.

Many protected areas, particularly those
where secure funding allows management to
be maintained, are effective in conserving
species, habitats and landscapes of value.
An extensive questionnaire study involving
93 protected areas in the tropics, in 22
countries, showed that areas where basic
management activities are in place tend to
avoid wholesale land clearance (but often still
suffer some degree of disturbance, from
hunting or logging, for example!'. An inte-
grated measure of their effectiveness is not
yet available, and it may be that firm manage-
ment within park boundaries can have the
effect of increasing pressure on outside land
that is less controlled; on the other hand,
many protected areas are badly under-
resourced, reducing their effectiveness.

Wise management of areas outside the
protected areas network, by means of
planning controls and voluntary agreements
and by incorporating conservation principles
in landuse planning, also plays an essential
role in conservation of biodiversity. Indeed, in
many countries, management of landuse
outside the national network of protected
areas - in agricultural landscapes, for
example - will play as important a role in the
maintenance of national biodiversity as will
the network itself.

Maintaining ecosystems

Increasingly in recent years a more holistic
approach has emerged in which ecosystems
themselves have become the focus of con-
servation efforts. Many would argue that the
primary goal of conservation action is the
long-term maintenance of ecosystem pro-
cesses at the global scale and over
foreseeable human generations. This is
essentially equivalent to achieving sustain-
ability in human development and use of
biodiversity. The Convention on Biological

Global biodiversity: responding to change

199

Diversity adopted 'the ecosystem approach'
as the guiding framework for actions taken in
pursuit of its goals and, in elaborating on the
meaning of this term, stressed the need to
conserve ecosystem structure and function
in order to maintain ecosystem services.
However, making an assessment of the
organization, vigor and resilience of eco-
systems - the key components of ecosystem
health' - is fraught with immense practical
difficulties. Much conservation activity there-
fore is guite properly based on a strong form
of the precautionary principle iBox 8.2), and
focuses on maintaining so far as possible
the prominent elements of ecosystems, i.e.
species and populations and their physical
environment, in anticipation that the system
will thus be perpetuated.

benefit will accrue from ensuring the integrity
of areas of higher biodiversity value than
areas of lesser value.

One early area-based approach identified
some 12 'megadiversity' countries, which
between them include a large proportion of
global biodiversity in selected major groups^

Box 8.2 The precautionary principle

A concise definition of the precautionary principle is provided by Principle 15 of the Rio
Declaration, made at the 1992 Earth Summit:

'In order to protect the environment, the precautionary approach shall be widely applied by
States according to their capabilities. Where there are threats of serious or irreversible
damage, lack of full scientific certainty shall not be used as a reason for postponing cost-
effective measures to prevent environmental degradation.'

Defining priority areas for protection

Management action aimed at biosphere con-
servation demands financial resources, but
these are limited, while human numbers and
impacts continue to increase. This implies
that choices must be made between possible
actions, whether by design or by default.

Rational and informed decision-making
should seek to increase the efficiency with
which conservation funds are used. Several-
studies have been undertaken that aim to
identify and sometimes to rank areas of high
biodiversity value. The approach is generally
based on the premise that more is better'.
That is, an area with greater biodiversity value
is more worth conserving than an area with
lower value. At its simplest biodiversity may
be eguated with species number overall or
per unit area, but other selection criteria
are possible. Higher priority may be accorded
to an area with populations of threatened
species, or one rich in endemic species
lespecially if there are endemics in several
different groups), or in species of commercial
or cultural importance; or one which is par-
ticularly representative of an ecosystem
(perhaps one that is widely degraded else-
where). While the results of such assess-
ments are usually of considerable biological
interest, they may also have direct application
as a basis for choice between different
courses of action, on the grounds that greater

Another early study delineated 18 endemic-
rich botanical 'hotspots', which between them
included around 20 percent of the world's
known plant species in less than 1 percent of
the land surface, and were also undergoing
rapid habitat conversion". A strength of the
former approach is that it is focused at country
level, and it is at this administrative level that
most conservation action is undertaken; a
weakness is that by evaluating species rich-
ness alone it was not able to address unigue-
ness, and adjacent countries rich in species
are likely to have many species in common. A
strength of the latter approach is that it focus-
ed on areas rich in restricted-range endemic
species and such areas by definition make a
large contribution to global biodiversity.

The country-based approach has been
extended in a study" using a database of
estimates of richness and endemism in land
vertebrates and vascular plants for all coun-
tries of the world (Appendix 5). Indices
of overall diversity (weighting richness and
endemism equally) and diversity adjusted
for country area have been produced. Within
the relatively wide margin of error associated
with species inventory (see Chapter 2), these
indices can yield a useful view of variation in
diversity in geopolitical terms or. If the data
are treated as geographic samples, a view of
general global variation In diversity (see Map

200 WORLD ATLAS OF BIODIVERSITY

Map 8.1

World protected areas

An overview of the world's
surface nominally subject
to protection and
appropriate management.
Tfie location of protected
areas in lUCN/WCPA
categories l-VI greater than
100 000 hectares in area is
shown, represented by a
point symbol. Protected
areas greater than 1 million
hectares in extent are
represented by polygons
instead of point symbols
whenever boundary data
are available.

Source; UNEP-WCMC database (data
extracted Marcti 20021, maintained in
collaboration witti lUCN World
Commission on Protected Areas.

Protected areas

^^^H > 1 million hectares
> 100 000 hectares

5. A). The hotspots approach has been further
developed by Conservation International in a
recent synthesis, and adopted as the basis for
its conservation planning and action'. Twenty-
five terrestrial hotspot regions were identi-
fied, combining high species endemism,
particularly among plants, and high rates of
habitat loss, with at least 70 percent of origi-
nal natural vegetation having been lost. Coral
reef areas of the world have now also been
evaluated in this way [see Chapter 6 and Map
6.1), and freshwater ecosystems are currently
being evaluated. The term 'hotspof is now
often applied loosely to any area that has a
concentration of diversity, whether by a
measure of simple richness, or endemism or
number of threatened species, and whether
undergoing habitat conversion or not.

Of studies based on biogeographic rather

^^hh'i^^'^

■'U^^i

than geopolitical areas, the Centres of Plant
Diversity project'" remains one of the largest.
It relied heavily on extensive consultation
among botanists to identity several hundred
important sites worldwide, defined semi-
quantitatively on the basis of a general
combination of richness and endemism (Map
8.2|. A broadly similar but less structured
approach, also based on expert knowledge,
has been taken to identify 43 areas of special
importance for amphibian diversity, again
using species richness and endemism as
criteria" (Map 8.31. A preliminary selection of
areas of importance for inland water bio-
diversity has also been made on the basis of
expert knowledge of richness and endemism
among fishes, mollusks and crustaceans [see
Chapter 7, and Maps 7.2-7.-41.

However, the most systematic and com-

Global biodiversity: responding to change

::-^:':c^

%€/^.

plete global level assessment to date has
involved bird species. Birds are by far the
best-known major group of organisms on the
planet, with a relative wealth of distribution
and population data available, and global
analyses by BirdLife International and its
partners have set a standard yet to be
matched for other taxonomic groups. The dis-
tributions of all restricted-range bird species
(defined as those in which the area encom-
passing all distribution records is less than
50 000 km1, amounting to 25 percent of all
birds, have been mapped in digital format.
The co-occurrence of restricted-range species
defines a set of 218 endemic bird areas
lEBAs)''; these are shown, ranked in three
categories according to biodiversity impor-
tance, in Map 8.4. Importance' here takes
account of the number of restricted-range

species in the EBA and the number of EBAs in
which they are present, taxonomic unique-
ness and EBA area.

There are a number of limitations to area-
based methods for establishing priorities
among possible conservation actions. First,
there is no single unequivocal way of com-
paring value in different categories: how
many vulnerable species is a single critically
endangered species worth? Or how many
endemic beetles is a single endemic bird
worth? Second, information is always incom-
plete, in that it never covers all taxa at all sites
of interest. Resolving the former requires
more or less arbitrary assigning of value.
Attempts to resolve the latter entail the
search for indicators, that is groups of species
or other variables that can act as surrogates
for wider measures of biodiversity.

202 WORLD ATLAS OF BIODIVERSITY

Map 8.2

Centers of plant diversity

This map stiows the
location of the sites and
areas identified as
important centers of plant
diversity at regional and
global levels, using expert
knovi/ledge. and mixed
criteria emphasizing
species richness and
endemism. See the three-
volume source cited below
tor further details and
extended documentation

Source; WWF and IUCn'°

Plant diversity

^^H • Areas and centers

The search for biological indicators of this
kind has generated a great deal of research
and discussion. Findings to date have gen-
erally been equivocal, and depend in part on
spatial scale. In general, though, it seems that
at coarse scales there may be quite close
agreement between different taxa so that, for
example, many EBAs (which may be up to
600 000 km- in extenti also hold significant
numbers of other restricted-range species".
At this scale, birds may serve as indicators of
high biodiversity value more generally In
other instances, and perhaps more generally
at finer resolution, the relationship appears to
break down. Studies in areas as disparate as
North America, South Africa", Cameroon"
and the British Isles"" indicate that areas
important for rare species in different groups
often do not coincide [and these may be

■ ■^.

negatively correlated with areas of high
species richness). Also, richness levels in any
one group do not necessarily provide an
indication of species richness in others.

A slightly different approach that avoids
some of the limitations faced by methods
based on species-specific information is to
identify areas on the basis of their biological
communities, and on biogeographic criteria,
with reduced emphasis on richness or
endemism in particular taxa, and none on
current habitat condition. The conservation
organization WWF's 'ecoregions' system,
mentioned in Chapters 5-7 above, is the
principal example of this approach, and
currently covers much of the world's marine
and inland waters, together with the entire
land surface. It has been adopted as the basis
for conservation planning and action by WWF.

Global biodiversity: responding to change

'^C^

Whilst a number of different groups and
organizations have explored the use of spatial
biodiversity information in suggesting con-
servation priorities, these generally remain
disconnected and to some extent duplicative".
Recent developments suggest that a more
integrated approach is feasible. Firstly, the
conceptual and methodological framework of
a more systematic approach to priority setting
has been elaborated" (see Box 8.3). Secondly,
sustained effort to map species distribution in
a globally consistent manner is beginning to
allovi/ robust analysis of large-scale multi-
taxon patterns of biodiversity" ". Combining a
systematic approach with a geographically
extensive set of spatial biodiversity data
has the potential to allow priorities to be
defined in a way that is flexible, transparent,
and robust.

Box 8.3 Systematic conservation planning

WL.

Key stages in systematic planning for conservation
by means of adding to or modifying protected area
systems:

1. Compile data on the biodiversity of the planning
region.

2. Identify conservation goals for the planning
region.

3. Review existing conservation areas.

4. Select additional conservation areas.

5. Implement consen/ation actions.

6. Maintain the required values of conservation
areas.

Source: Margules and Pressey .

204 WORLD ATLAS OF BIODIVERSITY

Map 8.3

Major areas of amphibian

diversity

The location of areas
identified on the basis of
expert knowledge as
globally important for
amphibian diversity, using
data on species richness
and endemism. Areas are
here shown classified in
five categories according to
species richness. See
source cited below for
further details.

Source: Adapted from Duellman

37- 76
21 -36
4-20

It should, however, be emphasized that
many priority-setting exercises are likely to
remain only theoretical. This is because the
opportunities to establish entirely new terres-
trial protected areas, or redesign existing
protected area networks in the light of priorities
identified, are in general extremely limited. The
designation and management of protected
areas, as so much other conservation action, is
generally driven and constrained far more by
socio-political and economic considerations
than it is by conservation biology.

Management of inland waters
To a greater extent than is typical for terrestrial
ecosystems, inland waters are often subject to
management and use decisions made within
many different sectors, including agriculture
and forestry, navigation, public utility supply

'^^- t

Ipower, water and waste disposal! and recre-
ation. Most inland waters or their catchments
also typically intersect several subnational
administration units Icounties, provinces, etc.)
Although it has been recognized for some time
that the catchment basin is the fundamental
unit within which management must be
formulated, it has proved difficult to reconcile
the many different interests concerned and
coordinate action. This is particularly problem-
atic because sources of adverse impacts on
such systems often originate far distant from
where they are felt: thus pollutants and other
inputs that enter the top reaches of a river
system may have an impact in all downstream
parts of the system as far as the river mouth
and beyond. Similarly, water abstraction may
have little effect in upper reaches, but be a
serious constraint downstream.

Global biodiversity: responding to change 205

-^,

X.-,

K>T\

>;iv#>

Biodiversity maintenance and conser-
vation of inland w/ater capture fisheries lother
than Sonne lucrative sports fisheries! have
not ranked highly among these competing
interests, so that it has been difficult to impose
catchment-vifide regulations or remedial
measures for their benefit. This problem is
exacerbated by the fact that it is often difficult
to pinpoint a distant source of a problem or to
unequivocally demonstrate that actions in one
place are having an adverse effect somewhere
else (e.g. to convince farmers that application
of large doses of nitrogenous fertilizer on up-
stream agricultural land is causing deleter-
ious eutrophication of estuarine wetlands!.

Although they cannot solve catchment-wide
problems, inland water protected areas may
play a valuable role in safeguarding particular
sites or populations of species from immediate

threats. Protection may be most effective
where sites are relatively small and thus
manageable, and have a relatively low level of
allochthonous inputs. Wetlands, with their
often abundant and highly conspicuous avi-
fauna, have in general received most attention
in this regard.

Notable wetland protected areas include
the Moremi game reserve in the Okavango
delta (Botswana!, Camargue national reserve
(France!, Keoladeo (Bharatpur! national park
(India), Dohana national park (Spain! and
Everglades national park (United States!.
Inland water ecosystems are unusual in that
an international convention is dedicated
specifically to them: the Convention on
Wetlands of International Importance espec-
ially as Waterfowl Habitat (the Ramsar
Convention, see belowl.

206 WORLD ATLAS OF BIODIVERSITY

Map 8.^

Endemic bird areas

More than one quarter
12 5611 of ttie world's bird
species, including more
than 70% of the threatened
birds, have a range
restricted to less than
50 000 km'. Virtually all
occur within the 218
endemic bird areas lEBAsI
defined by BirdLife
International'-.
The world's EBAs are
shown on this map
categorized 1. 2 or 3
according to increasing
biodiversity importance
(based on the number of
restricted range species,
whether shared between
EBAs, taxonomic
uniqueness and EBA sizel

Source: Data provided by BirdLife
International, and see Stattersfield .

Category

BSBJUS

Transboundary inland waters

Waters that delineate or cross international
boundaries present a special class of
management issue. Such waters and the
living resources they contain are shared by
one or more countries, and require positive
international collaboration for effective use
and management.

Available water in any given country within
an international basin lor other administrative
unit within a basin more generaltyl can be
divided into endogenous, i.e. locally generated
runoff available in national aquifers and
surface water systems, and exogenous, i.e.
remotely generated runoff imported in flow
from upstream. Some countries (e.g. Canada
and Norway! have an abundance of water
from endogenous sources; others (e.g. Egypt
and Iraqi have a small endogenous supply but

large exogenous volumes (others have small
supplies from both sources). Use of exogen-
ous water carries an increasing risk because
of dependence on sufficient supply from
upstream countries.

There are well over 200 major inter-
national rivers and a host of smaller ones''.
As demands on inland water resources
continue to grow through the 21st century, as
they undoubtedly will, management of these
and the biological resources they contain will
also grow ever more challenging.

Management of marine ecosystems
Management of the terrestrial environment is
typically carried out alongside more or less
severe anthropogenic disturbance. Although
the particular nature of the marine biosphere
has to some extent buffered it from the

Global biodiversity: responding to change

207

impacts of humans, it also imposes its own
set of difficulties and constraints for rational
management. Firstly, because almost all of it
is generally out of sight, impacts are not
immediately apparent, so that extreme deter-
ioration may take place before anyone is
aware of the fact. There may also be less
incentive to take action than with terrestrial
ecosystems where such deterioration has a
direct impact on people. Secondly, with some
exceptions (such as communal property
resources on reefs and other inshore areas in
parts of the South Pacific], living marine
resources have been widely considered open-
access resources, particularly those outside
territorial waters (usually up to 12 nautical
miles (nml from shore]. There is thus, quite
simply, an incentive for any given individual
to exploit a resource as fast and as intensively

as possible before someone else does.
Thirdly, the ability of water to transport large
amounts of dissolved and suspended mater-
ials, including living organisms, means that it
is extremely difficult to manage limited areas
of marine habitat in isolation.

With the introduction of the exclusive
economic zone (EEZ) under the United
Nations Convention on the Law of the Sea
lUNCLOS], which allows nations control over
resources in an area up to 200 nm offshore, a
far greater proportion of the world's seas now
come within the control of individual nations.
At present 99 percent of world fisheries catch
is taken within EEZs. Although this should
theoretically allow more rational manage-
ment of marine resources, and more effective
enforcement of management measures,
progress in both has in practice been limited

208 WORLD ATLAS OF BIODIVERSITY

Map 8.5

Marine protected areas

The map shows the location
of protected areas in lUCN
categories l-VI that are
entirely or in part marine,
with map symbols graded
according to protected area
size (including any land
present!.

Note; For presentation
purposes it has been
necessary to use symbols
that in most cases at this
map scale greatly exceed the
size of the protected area
represented, giving the
impression that much more
of the world's coastal waters
are protected than
is in fact the case.

Source: UNEP-WCMC database Idala
extracted March 20021. mamtained m
collaboration with lUCN World
Commission on Protected Areas.

to date, as evidenced by the increasing
proportion of world's fisheries that are over-
exploited. This is because fisheries manage-
ment regimes are frequently subject to
political pressure, so that in many countries
quotas are habitually set higher than those
recommended by fisheries biologists, and
also because the active enforcement of regu-
lations is difficult and expensive. Many
countries lack the resources or the political
will, or both, to enforce such regulations
adequately.

It is also increasingly apparent that individ-
ual marine resources cannot be effectively
managed in isolation from each other
Complex interactions between populations of
different organisms, when combined with
perturbations in the environment and vari-
ations in human impact le.g. changes in

fisheries technology or fishing effort! create
responses that may be far from intuitively
predictable. Recognizing that the under-
standing of such responses will require
modeling and management of large-scale
ecosystem processes, a number of large
marine ecosystem (LMEl units have been
identified-, based on the world's coastal and
continental shelf waters, which are regarded
as central to such analysis. Over 95 percent of
the usable annual global biomass yield of
fishes and other living marine resources is
produced within 64 identified LMEs, nearly all
of which lie within and immediately adjacent
to the boundaries of EEZs of coastal nations.
Many LMEs include the coastal waters of
more than one state. In these cases, it will be
effectively impossible for individual nations to
assess whether their use of marine resources

Global biodiversity: responding to change

209

t'i^^

¥'^-^_^-

-rJ

.vj;

A f^'

'-.-TC-v ;^

"'^^ XW^'

\K^--''

is sustainable in isolation from neighboring
nations. Coordination between states in mon-
itoring and resource management will thus
become increasingly necessary as the press-
ures placed on these areas increase.

A critical need in monitoring marine
ecosystems is the development of consistent
long-term databases for understanding
between-year changes and multi-year trends
in biomass yields. For example, marked
alterations in fish abundances were observed
during the late 19605 when there was intense
fishing within the northeast US continental
shelf LME. The biomass of economically
important finfish species le.g. cod, haddock,
flounders! declined by approximately 50
percent, and this was followed by increases
in the biomass of lower-valued small elasmo-
branchs Idogfish and skates). Management of

marine fisheries will need to take these kinds
of species dominance shifts into account in
the development of strategies for long-term,
economic sustainability of the fisheries".
Monitoring the changing states of LMEs has
received considerable attention, with several
now being assessed and managed from a
more holistic ecosystem perspective*- ■'.

Marine protected areas

The long-term management of LMEs is highly
complex and there is an urgent need for
smaller-scale and more immediate app-
roaches. As with terrestrial ecosystems, the
establishment of protected areas in marine
ecosystems has been viewed as a major
contribution to maintenance of biodiversity. A
1995 review' identified just over 1 300 marine
protected areas in existence at that time.

210 WORLD ATLAS OF BIODIVERSITY

Map 8.6

International protected

area agreements

The map shows the location
of protected areas managed
under the Ramsar
Convention on Wetlands,
under the World Heritage
Convention or as a
biosphere reserve within
the UNESCO Man and the
Biosphere Programme.

Source: UNEP-WCMC database Idata
extracted March 20021. maintained in
collaboration with lUCN World
Commission on Protected Areas,

^^'

World protected areas
Ramsar site
World Heritage site
■ MAB biosphere reserve

ranging in size from 1 hectare to 2UM million
hectares Ithe Great Barrier Reef Marine Park)
IMap 8.5). Effective management and control of
marine protected areas is problematic, partic-
ularly if, as is often the case, they are in areas of
intensive and potentially conflicting resource
use. As noted above, marine ecosystems are
also in general more difficult to protect than
terrestrial ones from allochthonous inputs (i.e.
those originating elsewhere). Although a no-
catch regime can be effective in small marine
reserves, in general it has been found that
large, carefully zoned, multiple-use areas are
more practical and effective than small re-
serves. Sanctuaries or strict reserves may still
be required for critical habitat areas such as
nutrient sources, areas of high biological diver-
sity, nesting sites of threatened species or to
protect breeding stocks of important fishes" ".

■^-.

Reversing change: restoration and
reintroduction

Increasing recognition of widespread environ-
mental degradation has led to a growth of
interest in both the science and practice
of ecological restoration. The main aim of
such restoration is to reestablish the key
characteristics of an ecosystem, such as
composition, structure and function, which
were present before degradation took place. It
has been suggested that ecological restor-
ation IS a crucial complement to the estab-
lishment of protected areas for safeguarding
biodiversity'', and it is widely expected that
restoration will become a central activity in
environmental management in the future.
Such efforts are being supported by develop-
ment of national and international policies.
For example, the UN Convention on Biological

Global biodiversity: responding to change

'Jp-

• w^f

Diversity, Article 8f, states ttiat parties should
'rehabilitate and restore degraded eco-
systems and promote the recovery of threat-
ened species, through the development and
implementation of plans or other manage-
ment strategies'.

A large number of restoration projects
have now been initiated in different parts of
the Vi^orld, focusing on a variety of different
ecosystem types, including grasslands, wet-
lands and forests. Although a number of
national governments are active in ecological
restoration, sometimes on a very large scale
(notably in North America], most projects are
being undertaken by NGOs, often as grass-
roots or community-based initiatives. For
example, together with a variety of local
partners, the Forests for Life program of
WWF/IUCN is implementing restoration

programs in areas such as the Lower Mekong,
New Caledonia, the Mediterranean, India and
the Carpathians. WWF/IUCN is increasingly
recognizing the critical importance of de-
veloping plans for restoration at the land-
scape scale, and the need to provide benefits
to local communities as well as to biodiversity
Experience of restoration projects to date
has highlighted how difficult such ecological
rehabilitation can be in practice. Although
many degraded ecosystems display an ability
to recover through natural processes if the
causes of degradation are removed, in many
areas the extent of degradation has been so
severe that greater management intervention
is required for restoration to be effective. For
example, severely deforested areas may
require large-scale tree planting in order for
forest ecosystems to reestablish on a partic-

212 WORLD ATLAS OF BIODIVERSITY

ular site. Restoration projects may also be
difficult to manage or monitor, as it is often
fiard to define with precision what the
structure, composition or function of a given
ecosystem was prior to degradation, partic-
ularly in areas where degradation occurred a
long time ago. Another key challenge to

Restoration projects have
been initiated in different
parts of the world, focusing
on a variety of ecosystem
types, including grasslands,
wetlands and forests.

restoration projects is the high cost involved:
for example, a plan to restore the Florida
Everglades has recently been launched, at a
total cost of US$7.8 billion^'.

Efforts at restoring degraded ecosystems
can be complemented by programs focusing
on the reintroduction or reestablishment of
species that have become extinct within a
particular area. For such reintroductions to be
successful, thorough knowledge of a species
and its habitat requirements are needed, in
addition to a clear understanding of the
original causes of extinction. In some cases,
such as large vertebrate predators, there may
be considerable public antipathy to reintro-
duction being attempted. However, there have
been some notable examples of successful
reintroductions, such as the Arabian oryx
[Oryx leucoryx] to Oman", the white-tailed
eagle [Haliaeetus albicilla] to Scotland, and
the tvlexican gray wolf ICan/s /upusfaa/(ey/l and
California condor [Gymnogyps californianus]
to parts of the United States. Such examples
provide important lessons for successful
reintroductions and, together with habitat

restoration projects, illustrate how positive
action can contribute to reversing the trends of
biodiversity loss.

THE INTERNATIONAL DIMENSION

National boundaries do not enclose all the
world's biological diversity: the high seas, the
deep seabed and Antarctica all contain natur-
al resources, some of great interest or
economic importance. (Management of bio-
diversity in such areas can, by definition, only
be achieved by means of international meas-
ures. Similarly, areas or communities of
particular interest may be crossed by national
boundaries, and in such cases international
cooperation is essential for conservation
measures to be planned and implemented
effectively.

More generally, it is important to develop
policy and planning at the global level to place
national efforts in a broader context. In a
hypothetical example, an individual country
might devote more effort to conserving
species that are rare or peripheral at national
level, but widespread elsewhere, than to
more common, nationally endemic species.
Although the former may be regarded as
national priorities, the latter may be more
important to global biodiversity. Regardless of
differences between national and global
priorities, an international forum is needed to
develop conservation science, to provide
exposure for diverse opinions and an oppor-
tunity for NGOs and other bodies to comment
on policy, and to formalize agreements that
can guide the way individual countries manage
their environments. Such opportunities are
offered by the international agreements that
have recently been developed.

International agreements
A multilateral treaty is an international
agreement concluded between three or more
states and governed by international law (see
Box 8.AI. Existing international treaties that
deal entirely or in part with biological diversity
have evolved in an uncoordinated manner.
Despite this, and the consequent gaps and
duplications in overall coverage, a handful
of such treaties whose text was agreed
during the 1970s have come to exert a

Global biodiversity: responding to change

powerful influence on the conservation and
management of elements of biodiversity.
Among the most notable are the 1971
Convention on Wetlands of International
Importance especially as Waterfowl Habitat
iRamsarl; the 1972 Convention Concerning
the Protection of the World Cultural and
Natural Heritage (World Heritage); the 1973
Convention on International Trade in
Endangered Species of Wild Fauna and Flora
(CITES); and the 1979 Convention on the
Conservation of Migratory Species of Wild
Animals ICMS). The 1982 UN Convention on
the Law of the Sea (UNCLOS). which entered
into force In 1994, has strong potential for
enhancing marine and coastal conservation.

The names of these major treaties indicate
their sectoral focus and, even if the many
Important regional and species-related trea-
ties are also considered, it is clear that the
total obligations explicit in existing treaties
fall short of the demands of an adequately
comprehensive system. The Convention on
Biological Diversity (CBD), agreed at the 1992
Earth Summit In Rio. attempted to meet many
of these demands. It was the first treaty
planned to concentrate specifically on the
conservation and use of global biodiversity. Its
text establishes the conservation and use of
biological resources as a matter of common
Interest to all. It has as Its objectives the con-
servation of biological diversity, the sustain-
able use of biological resources and the
equitable sharing of benefits arising from the
use of genetic resources, recognizing that a
careful balance must be maintained between
these If biological resources are to be used
wisely. It not only acknowledges the control
of individual states over their biological
resources, but it also states their respon-
sibility for protecting them and using them
sustalnably.

The United Nations Framework Conven-
tion on Climate Change lUNFCCCl was also
agreed at the Earth Summit and Is relevant to
biodiversity management. The CBD, the
UNFCCC and the United Nations Convention
to Combat Desertification (UNCCD) (which
arose from the Summit but was not agreed
until 199i) are sometimes termed the 'Rio
conventions'.

Those with a particular focus on bio-
diversity (CBD, CITES, CMS, Ramsar, World
Heritage) are informally termed 'the bio-
diverslty-related conventions'. They each
Impose more or less rigorous reporting
requirements on parties to them, and also
generate a significant demand for information
from their parties and others. Meeting these
demands can place a substantial burden on
governments, particularly those with limited
resources, and work is proceeding on
harmonizing information management among
the treaties.

In addition to the Rio conventions, the Rio
Declaration (a set of guiding principles), and a
comprehensive plan of action - Agenda 21 -
were also agreed and adopted by more than
178 governments at the Earth Summit.
Agenda 21 Is designed to support sustainable

Box 8.i Negotiating a multilateral treaty

Text

Negotiation of tfie text of tfie treaty can require many years and numerous meetings. It Is
concluded by the adoption of the text, typically when all the participating states reach
agreement over the wording. Adoption of a treaty does not by itself create any obligations.

Consent

A treaty does not come into force until two or more states consent to be bound by it.
The expression of such consent is usually by signature, notification, acceptance, approval or
accession or by other means where so agreed. Signature followed by ratification is the most
frequent means of expressing consent. Signature refers to the signature of the diplomats
negotiating the treaty and Is often synonymous with the adoption of the treaty. Ratification is
the need for approval of the treaty by the head of state or the legislature. Accession is the
normal way that states which did not participate In the negotiations become parties to the
treaty and has the same effect as signature and ratification combined.

Entry into force

This final stage usually occurs when all the negotiating states have expressed their consent
to be bound by the treaty; the date may be delayed to provide time for parties to adapt
themselves to its requirements. When a large number of states participate In the drafting of
a large multilateral treaty it often enters into force once a specified number have ratified.

human development and stewardship of the
environment, and to be implemented at a
range of scales - globally, nationally and
locally - by organizations within the United
Nations system, governments and other
major groups.

2U WORLD ATLAS OF BIODIVERSITY

Table 8.1

Major global conventions

relevant to biodiversity

maintenance

Notes: Conventions are
listed in order of entry into
force. Year' is date of
agreement, Entry' is year in
Vi/hicfi agreement entered
into force, 'Parties' is
number of party states as
indicated at eacti agreement
website in Marcti 2002.

^^^^^^^^^^^F

Scope ^^^^^^^^^^^^^^^^^H

Convention on Wetlands of International Importance

All aspects of wetland conservation and wise use.

especially as Waterfowl Habitat

Parties are required to list at least one w/etland of

international importance for special management and

(Convention on Wetlands or Ramsar Convention!

protection.

Year Entry Parties

1971 1975 131

Convention Concerning the Protection of the World

To define and conserve the world's fieritage, by dravKing

Cultural and Natural Heritage

up a list of sites whose outstanding values should be

preserved for all humanity, and to ensure their pro-

(World Heritage Convention!

tection through a closer cooperation among nations.

Sites may be of importance as cultural heritage or

Year Entry Parties

natural heritage or both.

1972 1975 167

Convention on International Trade in Endangered

Aims to prevent species being threatened with

Species of Wild Fauna and Flora

extinction because of international trade. Parties act by

banning commercial international trade in an agreed

[CITES!

list of endangered species (Appendix-l listed species!

and by regulating and monitoring trade in others that

Year Entry Parties

might become endangered or whose trade needs to be

1973 1975 154

regulated to ensure control over trade in Appendix-l

species |Appendix-ll listed species!.

Convention on the Conservation of Migratory Species

Aims to protect migratory species and their habitats.

of Wild Animals

Parties cooperate in research relating to migratory

species and provide immediate protection for species

ICMS or Bonn Convention!

listed in Appendix 1 of the convention. For those species

listed in Appendix 11, parties are required to endeavor to

Year Entry Parties

conclude range state' agreements on their con-
servation and management, a number of which have

1979 1983 79

been concluded.

Convention on Biological Diversity

The major international agreement on biodiversity, the

CBD sets out a framework within which parties

ICBD!

undertal<e to carry out national and international

measures aiming to: conserve biodiversity, make

Year Entry Parties

sustainable the use of its components, and share

1992 1993 183

equitably the benefits derived from the use of genetic

resources The Conference of the Parties has so far met

five times. The Cartagena Protocol on Biosafety was

adopted in 2000 but is not yet in force.

Global biodiversity: responding to change

Ronve

:onvention

United Nations Frameworl< Convention on Climate
Change

lUNFCCCI

Year

Entry

Parties

1992

1994

179

United Nations Convention on the Law of the Sea

lUNCLOSI

Year

Entry

Parties

1982

1994

138

United Nations Convention to Combat Desertification
in those countries experiencing serious drought
and/or desertification, particularly In Africa

lUNCCD - Desertification Convention]
Year Entry

Parties

1994

1996

179

Scope

Aims to stabilize greenliouse gas concentrations in tlie
atmosphere at safe levels. Parties are required to make
an inventory of their sources and sinks of greenhouse
gases and to formulate policies and measures to
mitigate and/or adapt to the effect of climate change.
Developed country parties were required to reduce their
emissions of greenhouse gases to their 1990 level by
the year 2000. The Kyoto Protocol establishes further
reduction commitments for developed country parties.

Contains a comprehensive codification of the principles
and rules relating to the seas. UNCLOS establishes
rights and obligations relating to navigation, the
conservation and use of marine resources, and the
protection of the marine environment

Aims to ensure improved management of dryland
ecosystems and use of development aid. National action
programs (NAPsI will address the underlying causes of
desertification and drought and seek to identify
preventative or remedial measures. Subreglonal and
regional action programs ISRAPs, RAPsI will be devel-
oped, particularly when transboundary resources such
as lakes and rivers are involved.

Agreement for the Implementation of the Provisions of The objective is to ensure the long-term conservation

the UN Convention on the Lavif of the Sea relating to
the Conservation and Management of Straddling Fish
Stocks and Highly Migratory Fish Stoclts

and sustainable use of straddling and highly migratory
fish stocks. Emphasizes the precautionary approach,
protection of marine biodiversity and the sustainable
use of fisheries resources.

(Straddling Fish Stocks Agreement]

Year

Entry

Parties

1995

2001

International protected area systems

Two international conventions and one inter-
national program include provision for desig-
nation of sites internationally important for
biodiversity conservation. These are the World
Heritage Convention, the Ramsar (Wetlands]
Convention, and the UNESCO Man and the
Biosphere (MABI Programme. The location of

protected areas managed under these
agreements is shown in Map 8.6.

Ramsar Convention

The Convention on Wetlands of International
Importance especially as Waterfowl Habitat
was signed in Ramsar (Iran] in 1971, and
provides a framework for international

214 WORLD ATLAS OF BIODIVERSITY

The Ramsar Convention
provides a framework for
International cooperation
for tfie conservation of
wetland fiabitats.

cooperation for tfie conservation of wetland
habitats. It places general obligations on con-
tracting party states relating to the
conservation of wetlands throughout their
territories, with special obligations pertaining
to those wetlands which have been added
to the List of Wetlands of International
Importance. Each state party is obliged to list
at least one site. Wetlands are defined by the
convention as: areas of marsh, fen, peatland
or water, whether natural or artificial, per-
manent or temporary, with water that is static
or flowing, fresh, brackish or salt, including
areas of marine waters, the depth of which at
low tide does not exceed 6 meters. There are
currently (March 20021 131 contracting
parties to the convention; 1 1^8 wetlands have
been designated for Inclusion in the List of
Wetlands of International Importance, cover-
ing more than 96 million hectares.

World Heritage Convention

The Convention Concerning the Protection of
the World Cultural and Natural Heritage was
adopted in Paris in 1972, and provides for the
designation of areas of 'outstanding universal
value' as World Heritage sites, with the prin-
cipal aim of fostering International co-
operation In safeguarding these important

areas. Sites must be nominated by the signa-
tory nation responsible and are evaluated for
their world heritage quality before being
listed by the international World Heritage
Committee. Of the 721 sites distributed
among 12^ countries that are currently listed,
144 cover natural heritage and 23 sites are
mixed cultural and natural.

Article 2 of the World Heritage Convention
considers as natural heritage: natural fea-
tures consisting of physical and biological
formations or groups of such formations
which are of outstanding universal value from
the esthetic or scientific point of view;
geological or physiographlcal formations and
precisely delineated areas which constitute
the habitat of threatened species of animals
and plants of outstanding universal value
from the point of view of science or con-
servation; and natural sites or precisely
delineated areas of outstanding universal
value from the point of view of science,
conservation or natural beauty. Criteria for
inclusion in the list are published by the
United Nations Educational, Scientific and
Cultural Organization lUNESCOl.

Biosptiere reserves

The establishment of biosphere reserves is
not covered by a specific convention, but is
part of an international scientific program,
the UNESCO Man and the Biosphere (MABI
Programme. The objectives of the network of
biosphere reserves, and the characteristics
which biosphere reserves might display, are
Identified in the Action Plan for Biosphere
Reserves. There are currently 411 biosphere
reserves, spread over 94 countries.

Biosphere reserves differ from the pre-
ceding types of site in that they are not
exclusively designated to protect unique
areas or important wetlands, but for a range
of objectives which include research, mon-
itoring, training and demonstration, as well
as conservation. In most cases the human
component Is vital to the functioning of the
biosphere reserve, which does not nec-
essarily hold for either World Heritage or
Ramsar sites. Some biosphere reserves
coincide spatially with Ramsar or World
Heritage sites.

Global biodiversity: responding to change

217

POSSIBLE FUTURES

Previous chapters have introduced some of
the impacts on the biosphere of human
expansion and development, and this chapter
has outlined some of the general ways in
which humans have attempted to manage
the extent or severity of these impacts on
biodiversity. These approaches reflect current
conditions of climate and human development
and the present policy environment, but is
there a way to incorporate possible future
conditions within the planning process?

Scenarios

Scenarios are increasingly being used to inform
policy development and implementation, by
illustrating the possible outcome of current
trends, and by highlighting the implications of
different policy decisions. The development of
scenarios has become an important tool by
which scientists communicate the results of
their research to decision-makers, as well as
constituting a significant research endeavor
in its own right. An example is provided by the
Intergovernmental Panel on Climate Change
[IPCCI, which provides scientific, technical
and socioeconomic advice to the world
community on the specific issue of climate
change. Climate change scenarios have been
developed by the IPCC through workshops
and meetings involving experts in modeling,
climate impact assessment and emissions
scenarios. These scenarios have formed the
basis of the decisions made by parties to
the UN Framework Convention on Climate
Change, which provides the overall policy
framework for addressing the climate
change issue.

Few attempts have been made to develop
scenarios of biodiversity change. One recent
approach^" involved consultation between
ecological specialists from a number of
different regions, who developed global scen-
arios for ten terrestrial biomes for the year
2100, based on global scenarios of changes in
environment and landuse. Five key deter-
minants of changes in biodiversity were
identified: landuse change, atmospheric car-
bon dioxide concentration, nitrogen depo-
sition and acid rain, climate and the introduc-
tion of exotic species into an ecosystem. The

expected changes in these drivers were then
considered for different ecosystem types
Ibiomesl, and the relative impact of the
different drivers on biodiversity was also
estimated. Three scenarios of future bio-
diversity were developed for each biome.
based on various assumptions about the
interactions between the drivers affecting
biodiversity change.

The results of this investigation suggest
that when all biomes are considered together,
landuse change is the driver expected to have
the largest impact on biodiversity at the global
scale, with climate change ranking second in
importance. Considering the impact of all
drivers together, Mediterranean ecosystems
appear to be most at risk, with grassland and
savannah ecosystems also being at relatively
high risk.

To further illustrate possible future trends
in global biodiversity, examples of two alter-
native approaches to scenario development
are described below. Both these approaches
were developed as contributions to the Global
Environment Outlook IGEOl produced by the
United Nations Environment Programme
lUNEPl, in collaboration with a wide range of
partners. The latest report of the GEO process
[GEO-31 provides further details of these
scenarios, and the methods employed in their
development".

RIVM IMAGE scenario evaluation

For GEO-3, UNEP developed scenarios through
a highly participative process involving GEO
collaborating centers and other partners
throughout the world, allowing particular
issues identified at regional scale to be incor-
porated. The scenarios involved development
of narratives or storylines, respectively termed
Markets First, Policy First, Security First and
Sustainabitity First. These each describe
possible futures based on different inter-
pretations of prevailing global driving forces.

A number of models and analytical tools
were used to develop the scenarios, to help
guantify the dynamics described qualitatively
in the narratives, and to compare impacts
across regions. These included PoleStar,
developed at SEI-Boston: IMAGE, developed
at RIVM in the Netherlands; WaterGAP,

218 WORLD ATLAS OF BIODIVERSITY

Figure 8.2

Possible future scenarios
from GEO 3, evaluated
witfi RIVM IMAGE:
pressures on remaining
natural areas in 2002 (topi
and changes in pressure
between 2002 and 2032
under the Markets First
scenario (bottom)

Notes: The pressures
selected are those believed to
have a major influence on
biodiversity and for which
data are available. These
include habitat loss,
population density, primary
energy use, temperature
change, and restoration time
for agricultural land and
deforested areas. The loviier
map shows the relative
increase or decrease in
pressure between 2002 and
2032 {difference divided by
the pressure in year 20021.
No change means less than
10% change over the
scenario period; small
increase or decrease means
between 10 and 50% change,
substantial increase or
decrease means 50 to 100%
change: and strong increase
means more than doubling.

Source; RIVM IMAGE; Bnnk"*; Bnnk";
Brink e( at.".

Low pressure
Very high pressure

.■it '-•''. ,. ■■>

■■ Domesticated area
I I Ice and polar area

^H Substantial decrease in pressure

Small decrease in pressure

No change in pressure

Small increase in pressure

Substantial increase in pressure
^H Strong increase in pressure

developed by CESR at Kassel in Germany;
and the AIM model developed at NIES and
Kyoto University in Japan. PoleStar offers a
simple modeling approach to explore how
assumptions affect social and environmental
performance, and incorporates scenarios
developed by the Global Scenario Group'^ For

From domesticated to natural area
From natural to domesticated area
Remains domesticated

I I Ice and polar area/no data

the GEO-3 exercise, the scenario assumptions
were applied to different regions individually,
and were then used to initiate and inform the
scenario design process. PoleStar was also
used to test the plausibility of storylines
identified through the participatory process in
most of the regions.

Global biodiversity: responding to change

219

The GEO-3 scenarios look 30 years ahead.
Markets First \s a world in which market-driven
developments converge on the values and
expectations that prevail in industrialized
countries. Policy First \s a world in which con-
certed action on environment and social issues
occurs through incremental policy adjust-
ments. Security First is a world of fragmen-
tation, where inequality and conflict prevail,
brought about by socioeconomic and environ-
mental stresses. Sustainability First is a world
in which a new development paradigm emerges
in response to the challenge of sustainability,
supported by new values and institutions.

In order to assess the impact on bio-
diversity of such diverging trends, IMAGE was
used to assess changes in the natural capital
index. This index is designed as a measure of
biodiversity in natural ecosystems and agri-
cultural land, and is calculated as the product
of remaining area and its quality. The current
remaining natural area is taken as a percent-
age of the total land area. Changes over time
are caused by the conversion of natural
ecosystems into agricultural and built-up
areas, and vice versa. Ideally, ecosystem
quality should incorporate measures of the
abundance of characteristic species relative to
a low-impacted baseline state, but such data
are not generally available at global scale, and
ecosystem quality is therefore approximated
by means of four pressure factors that are
assumed to have a major influence on bio-
diversity and for which data are available. For
each pressure factor, a preliminary range is
defined from a review of relevant literature,
from no effect to complete deterioration of the
habitat if the maximum value is exceeded. The
four selected pressure factors are population
density (min-max: 10-150 persons per km-);
primary energy use 10.05-100 petajoules per
kml; rate of temperature change 10.2-2.0°C in
a 20-year period); and a restoration time for
abandoned agricultural land and deforested
zones in reconversion towards natural, low-
impacted ecosystems Imin-max: 100-0 years
after conversion or clear-cutting). The proxy
for ecosystem quality is a reversed function of
these pressures, calculated as a percentage of
the low-impacted baseline state. The higher
the pressure, the lower the quality. Finally, the

percentages for habitat area and quality are
multiplied, resulting in a pressure-based
natural capital index. The calculations have
been carried out on a detailed latitude-
longitude grid, before aggregation to sub-
regions and regions.

The maps Isee Figure 8.2) illustrate the
combined effect of decreasing area and qual-
ity of habitat. The Markets F/'rsf scenario sug-
gests a strong decrease of habitat quality and
quantity in most regions, but in some regions
agricultural land is taken out of production and
may therefore revegetate naturally through the
process of succession. However, in biodiversity
terms this reconverted land must be con-
sidered of low quality during initial decades.

GLOBIO

GLOBIO (Global methodology for mapping
human impacts on the biosphere! was
launched by UNEP in 2001 as a simple global
model to help visualize the growing impacts
of infrastructural development on biodiversity
and ecosystem resources". GLOBIO provides
an assessment of the probability of such
impacts occurring by defining buffer zones
around infrastructure. The extent of impact
zones varies with the type of human activity
and density of infrastructure, region, vege-
tation type, climate and sensitivity of species
and ecosystems, and is based on extensive
literature surveys of scientific studies assess-
ing environmental impacts resulting from
human development.

By linking impacts in different ecosystems
and regions with satellite imagery lAVHRR
data from 1992-93 on 1 km- resolution in the
Global Land Cover Characterization database
version 2.0), available resources and infra-
structure, overviews of the cumulative im-
pacts of continuous development can be
derived. Scenarios for impacts on biodiversity
are based upon data describing existing
infrastructure, historic growth rates of infra-
structure, availability of petroleum and min-
eral reserves, vegetation cover, population
density, distance to coast and projected
development. The outcome is a simple over-
view of the current and possible future
cumulative impacts of infrastructural develop-
ment on biodiversity, assuming continued

220 WORLD ATLAS OF BIODIVERSITY

Figure 8.3

Possible future scenarios,

GLOBIO: the pressures on

the environnnent in 2002

(topi and in 2032 under

the Markets First scenario

(bottom)

Notes; GLOBIO provides
projections of human impacts
based on the development of
infrastructure. With an
economy dominated by
market forces, biodiversity
will be affected by
infrastructural development
on over 72% of the land area
by 2032, according to GLOBIO
projections.

Source: GLOBIO-'. GRID Arendal.
UNEP.

High impact
Medium-high impact
Lo»/-medium impact

Semi-deserts and deserts

Wetlands

Forests

Grasslands/savannahs

Croplands

Water

growth in demand for natural resources and
the associated infrastructure development.

GLOBIO defined four different degrees of
human impact:

• Human populations and urban areas (high
impact). This includes all areas that are
within 0.5 km of a road, or within a few
hundred meters of any development, inclu-
ding urban areas.

• Converted land (medium-high impact). This
includes areas converted to agricultural
lands, plantations, and with a high density of
infrastructure (i.e. generally within 0.3-2 km
of any infrastructure).

• Areas under conversion and fragmentation

Imedium-moderate impact). This includes
areas within 2-10 km of infrastructure.
• Relatively intact (low impact). These are
areas that still retain their natural state to a
large extent, i.e. more than 10 km from any
infrastructure.

GLOBIO scenarios are based on two assum-
ptions: that stable gross domestic product
(GDP) or an increase in GDP will require
further development in infrastructure, and
that impacts of such developments on bio-
diversity will decrease with distance from the
infrastructure. The scenario illustrated here
[Markets First, Figure 8.3) assumes a 3-^
percent annual increase in economic growth.

Global biodiversity: responding to c ti a n g e

Because of differences in tilstory, population
density, coastal distance, landcover, availa-
bility of mineral and petroleum resources
[included as part of the model) and current
economic capacity, ttie rate of change is not
the same in all parts of the world. The
implication of this scenario is that if further
expansion is not controlled, increasingly
severe losses of biodiversity will occur as a
result of continuing human development.
GLOBIO projections suggest that increased
industrial exploration for oil, gas and minerals
will accelerate road construction, draining of
rivers for hydro-power and irrigation, and
increased immigration, logging, and conver-
sion of land to plantations and farmland,
resulting in heavy losses of biodiversity in
many areas. During the last 150 years,
humans have directly impacted and altered
close to Ul percent of the global land area.
Under the scenario presented here, bio-
diversity will be threatened in almost 72
percent of the land area by 2032. Losses of
biodiversity are likely to be particularly severe
in Southeast Asia, the Congo basin and parts
of the Amazon. As much as 48 percent of these
areas will become converted to agricultural
land, plantations and urban areas, compared
with 22 percent today, suggesting wide
depletions of biodiversity.

Conclusions

In considering the results of such scenario-
building exercises, it is important to recognize

that the future is very uncertain. None of
the approaches described above attempts to
provide accurate predictions of what is likely
to happen in the future; rather, the aim is
to illustrate what could occur if current
trends continue. Although differing in the
methodologies adopted and the results
obtained, each of these approaches suggests
that human activities will continue to have
a major impact on biodiversity in coming
decades. It is clear that those areas charac-
terized by the highest numbers of species,
such as tropical forests and Mediterranean
ecosystems, are at particularly high risk of
suffering major losses of biodiversity. In
contrast, northern temperate areas appear
less likely to experience such severe impacts,
although significant losses of biodiversity may
still be anticipated. The challenge for future
efforts at scenario development will be to
identify those policy decisions that could make
the greatest contribution to the conservation
of biodiversity while securing the economic
benefits of sustainable development. With
widespread declines in biodiversity likely to
continue, given current trends in economic
development and population growth, it seems
probable that such decisions will at best
contribute to slowing the rate of biodiversity
loss. Significantly, none of the scenario
approaches suggests much scope for the
ecological recovery or rehabilitation of eco-
systems that have already been degraded,
except in localized areas.

REFERENCES

1 UNEP 1997. Recommendations for a core set of indicators of biological diversity.
Background paper prepared by the Liaison Group on Indicators of Biological Diversity
UNEP/CBD/SBSTTA/3/lnf.13. Available online from the CBD Secretariat website,
http://www.biodiv.org/ [accessed April 20021.

2 lUCN, UNEP, WWF 1980. V^odd conservation strategy: Living resource conservation for
sustainable development. lUCN-the World Conservation Union, Gland.

3 lUCN 1998. 1997 United Nations list of protected areas. Prepared by UNEP-WCMC and
WCPA. lUCN-the World Conservation Union, Gland and Cambridge.

It Bruner, A.G. et al. 2001. Effectiveness of parks in protecting tropical biodiversity Science
291: 125-128.

5 Mittermeier, R. 1988. Primate diversity and the tropical forest. Case studies from Brazil
and Madagascar and the importance of megadiversity countries. In: Wilson, E.O. [ed.l
Biodiversity, pp. U5-154. National Academy Press, Washington DC.

6 Myers, N. 1988. Threatened biotas: Hotspots' in tropical forests. Environmentalist B: 1-20.

222 WORLD ATLAS OF BIODIVERSITY

7 Myers, N. 1990. The biodiversity challenge: Expanded hot-spots analysis. Environmentalist
10; 2^3-256.

8 WCMC 1998. Development of a national biodiversity index. Version 2. Unpublished report
for the World Bank. World Conservation Monitoring Centre.

9 Mittermeier, R.A.. Myers, N. and Mittermeier, C.G. 1999. Hotspots: Earth's biologically
rictiest and most endangered terrestrial ecoregions. Conservation International,
Washington DC, distributed by University of Chicago Press. Available online at
http:/Mwvtf.conservation.org/xp/CIWEB/strategies/hotspots/hotspots.xml laccessed March
2002).

10 WWF and lUCN 199/i. Centres of plant diversity. A guide and strategy for their
conservation. 3 vols. lUCN Publications Unit, Cambridge.

11 Duellman, W.E. (ed.l 1999. Patterns of distribution of amphibians: A global perspective.
Johns Hopkins University Press, Baltimore.

12 Stattersfield, A.J. et al. 1998. Endemic bird areas of the world. BirdLife International,
Cambridge.

13 Van Jaarsveld, A.S. et al. 1997. Biodiversity assessment and conservation strategies.
Science 279: 2106-2108.

U Prendergast, J.R. et al. 1998. Biodiversity inventories, indicator taxa and effects of habitat
modification in tropical forest. Nature 391 : 12-1 b.

15 Prendergast, J.R. et al. 1993. Rare species, the coincidence of biodiversity hotspots and
conservation strategies. Nature 365: 335-337.

16 Prendergast, JR. and Eversham, B. 1997. Species richness covariance in higher taxa:
Empirical tests of the biodiversity indicator concept. Ecography 20: 210-216.

17 Mace, G. et al. 2000. it's time to work together and stop duplicating conservation
efforts. Nature A05: 393.

18 Margules. C.R. and Pressey, R.L. 2000. Systematic conservation planning. Nature U05:
2/13-253.

19 Da Fonseca, G.A.B. et al. 2000. Follovuing Africa's lead in setting priorities. Nature i05:
393-394.

20 Brooks, T. et al. 2001 . Towards a blueprint for conservation in Africa. BioScience 51 :
613-624.

21 Wolf, A.T. et al. 1 999. International river basins of the world. International Journal of Water
Resources Development 15: 387-427.

22 Large marine ecosystems of the world. Available online at
http://www.edc.uri.edu/lme/default.htm laccessed March 20021.

23 Murowski, S.A. 1996. Can we manage our multi species fisheries? In: The northeast shelf
ecosystem: Assessment, sustainability, and management, pp. 491-510. Blackwell
Science Ltd, Oxford.

24 Sherman, K. and Busch. D.A. 1995. Assessment and monitoring of large marine
ecosystems. In: Rapport D.J., Guadet, C.L. and Calow, P. ledsl. Evaluating and monitoring
the health of large-scale ecosystems. Springer- Verlag, Berlin, published in cooperation
with NATO Scientific Affairs Division. NATO Advanced Science Institutes Series 1: Global
Environmental Change 28: 385-430.

25 Salm, R.V. and Clark, J.R. 1984. tvlarine and coastal protected areas: A guide for planners
and managers. lUCN-the World Conservation Union, Gland.

26 Kelleher, G., Bleakley, C. and Wells, S. ledsl 1995. A global representative system of
marine protected areas, \/o[s 1-4. IBRD/World Bank, Washington DC.

27 Dobson. A.P., Bradshaw, A.D. and Baker, A.J. M. 1997. Hopes for the future: Restoration
ecology and conservation biology. Science 277: 51 5-522.

28 WRI 2000. World resources 2000-2001: People and ecosystems. The fraying web of life.
World Resources Institute, Washington DC.

Global biodiversity: responding to change

223

29 Spalton, A. 1990. Recent developments in the relntroduction of the Arabian oryx [Oryx
leucoryx] to Oman. Species 15: 27-29.

30 Sala, O.E. ef at. 2000. Global biodiversity scenarios for the year 2100. Science 287:
1770-177^.

31 UNEP 2002. Global environment outlook. Report of GEO-3. United Nations Environment
Programme, Nairobi.

32 Raskin, P. ef at. 1998. Bending the curve: Toward global sustainability. Stockholm
Environment Institute, Stockholm.

33 GLOBIO. Online at http://www.globio.info (accessed April 2002).

34 Brink, ten B. 2000. Biodiversity indicators for the OECD Environmental Outlook and
Strategy, a feasibility study RIVM report 4020010U, Bilthoven.

35 Brink, ten B. 2001. The state of agro-biodiversity in the Netherlands. Integrating habitat
and species indicators. Paper for the OECD workshop on agri-biodiversity indicators, 5-8
October, Zurich, Switzerland.

36 Brink, ten B. et at. 2000. Technical report on Biodiversity in Europe: An integrated
economic and environmental assessment. Prepared by RIVM, EFTEC, NTUA and IIASA in
association with TME and TNO under contract with the Environment Directorate-General
of the European Commission. RIVM report 481505019, Bilthoven.

APPENDIX 1 225

APPENDIX 1:

THE PHYLA OF LIVING ORGANISMS

The phyla are listed in alphabetical sequence
within each higher group, as are the 'kingdoms'
within the Eukarya. The symbols associated with
each phylum name indicate whether the species
occur in marine, inland water or terrestrial
habitats. Where more than one symbol is shown,
this does not mean that species are equally
distributed between them. In some cases, the text
notes the principal habitat. For parasitic forms
the symbol refers to the host habitat.

Where possible, an estimate has been given of
the number of described extant species in each
eukaryote phylum. Because of the vagaries of
taxonomy and lack of a consolidated catalogue of
such species Isee Chapter 21, even this number is
subject to considerable uncertainty in many
cases; '?' indicates no estimate available.

An attempt has also been made to gauge
whether the given number represents a low,
medium or high proportion of the possible total
number of species in each phylum (i.e. including
currently undescribed species). In rough terms, a
low proportion is taken to indicate that the total

diversity of the phylum may be an order of
magnitude lor morel higher than the number of
currently described species. A high proportion
indicates that well over half the total number of
species has almost certainly already been
described. Generally, these estimates reflect the
likelihood that microscopic, aquatic Especially
marine! and parasitic groups are less thoroughly
sampled than most macroscopic terrestrial forms.
Because applying the species concept to
prokaryotes is so problematic, no figures for
species diversity are given for Archaea or
Bacteria: it can be safely assumed that in almost
all these phyla the proportion of existing diversity
characterized to date is low.

Source: Margulis, L. and Schwartz. K V. 1V98. Five kingdoms. An
Illustrated guide to the ptiyla of lite on earth. 3rd edition. W.H.
Freeman and Company, New York.

T Marine

♦ Freshwater

• Terrestrial

Crenarchaeota T ♦

Bacteria adapted to hot, acidic sulfur-rich

Thermoacidophils

environments often found in hot springs and around

submarine vents. Pyrotobus grows at temperatures of

Size Microscopic

1 13°C. Sulfolobus tolerates temperatures up to 90°C

Nutrition Chemoautotrophs or heterotrophs

and may die if the temperature drops below 55°C; it

l^odeoflife Free-living

also tolerates highly acid conditions (pH of less than

1, or stronger than concentrated sulfuric acidl.

Euryarchaeota T ♦ •

swamps, bogs and estuary sediments; many others

Methanogensand halophils

live in the guts of herbivorous animals, from termites

to cows. They are cnemoautotrophs that obtain

Size lulicroscopic

energy by reducing COj (carbon dioxidel and oxidizing

Nutrition Methanogens are chemoautotrophs

H2 IhydrogenI to produce CH^ (methanel and H^O

halophils are photosynthetic

(waterl. They are responsible for production of most

Mode of life Free-living or symbiotic, inhabiting the

natural gas and for liberation of organic carbon from

intestines of animals

sediments into the atmosphere where it can be

Euryarchaeota share similanties m ribosomal RNA

reused, involving around 2 billion tons of methane

sequence but consist of two very different groups.

annually Halophils are aerobes that live in extremely

f/lethanogens cannot tolerate oxygen lare obligate

salty or highly alkaline environments, such as soda

anaerobesi and free-living forms tend to occur in

lakes, where tliey may be visible as a pink scum.

ARCHAEA

226 WORLD ATLAS OF BIODIVERSITY

BACTERIA

Actinobacteria ♦ •

Actinomycetes. actinomycota and their relatives

Size Microscopic

Nutrition Helerotroptiic

Mode of life Some free-living; some symbionts

A large and diverse group of heterotrophic
unicellular rod-shaped bacteria Icoryneforms), and
filamentous, multicelled bacteria lactinomycetesl
originally regarded as fungi. Some form pathogenic
lesions on skin, while others are found in leaf litter;
some of the latter can break down cellulose. Frankia
15 a nitrogen-fixmg symbiont in plants. Streptomyces
produces streptomycin and other antibiotics.

Aphragmabacteria

Size Microscopic

Nutrition Heterotrophic

Mode of life All symbionts, some parasitic

? ? • Very small bacteria, lacking a cellwall, widespread

in insect, plant and vertebrate tissues. Normally
benign, but pathogenic in some conditions, and
responsible for some forms of pneumonia and tick-
borne diseases le.g. Ehrlichia]. Eight named genera
to date.

Chlorobia T ♦ •

Anoxygenic green sulfur bacteria

Size Microscopic

Nutrition Photosynthetic
Mode of life Mainly free-living; some symbiotic with
other bacteria

Phototrophic obligate anaerobes, inhabiting sunlit
sulfide-nch habitats, particularly anaerobic muds.
Some are tolerant of extremely high or low
temperatures and salinities. Most use hydrogen
sulfide or sodium sulfide in photosynthesis, instead
of water, releasing sulfur instead of oxygen. Others
form symbiotic associations with oxygen-respiring
heterotrophic bacteria.

Chloroflexa

Green nonsulphur phototrophs

Size Microscopic

Nutrition Photosynthetic
Mode of life Free-living

♦ 9 Anaerobic filament-forming bacteria known from

sulfur-rich habitats such as hot springs. Whilst
these forms are typically photosynthetic,
Chlorollexus can also grow heterotrophically in
the dark. Three genera named to date.

Cyanobacteria ▼ ♦ •

Blue-green bacteria and chloroxybacteria

Size Microscopic but relatively large

Nutrition Photosynthetic
Mode of life Free-living

Photosynthesizmg bacteria, present in a great variety
of habitats Until recently called 'blue-green
algaeand considered to be plants. These bacteria
dominated the landscape in the Proterozoic eon
between 2 400 and 545 million years ago.
Prochlorococcus occurs at the base of the photic zone
throughout the world's oceans and may be one of the
commonest bacteria. Many fix atmospheric nitrogen.
Form reef-like stromatolites in some shallow-water
marine environments. Around 1 000 named genera.

Deinococci

Heat- or radiation-resistant bacteria

Size Microscopic

Nutrition Heterotrophic
Mode of life Free-living

Spherical, heterotrophic, obligate or facultative
aerobic bacteria highly resistant to heat [Thermus] or
radiation iDe/nococcusl. Most metabolize sugars.
Thermus aqualicus, isolated from hot springs in
Yellowstone National Park, United States, is the
source of Tag polymerase used in the polymerase
chain reaction technigue.

APPENDIX 1 227

Endospora T ♦ •

Endospore-forming and related bacteria

Size Microscopic

Nutrition Heterotrophic

Mode of life Many synnbionts and parasites

A very large, important and varied group of
tieterotrophic bacteria, some obligate anaerobes,
others facultative or obligate aerobes. Most form
endospores Ipropagules within the parent cell
resistant to heat and desiccation!. Some can break
down lignin and cellulose, others are fermenters,
breaking down sugars to produce compounds such
as lactic acid and ethanol. Some, such as
Streptococcus, are associated with infections.

Pirellulae ♦

Protemaceous-walled bacteria and their relatives

Size Microscopic

Nutrition Heterotrophic
Mode of life Mostly aquatic in freshwaters; some
symbionts. some parasitic

Diverse bacteria with proteinaceous cell walls;
mostly obligate aerobic heterotrophs living in
freshwaters. Chlamydia is parasitic, inhabiting
animal cells and with apparently no independent
means of producing energy C. psiHao causes
psittacosis; C. tracliomatis causes trachoma
blindness.

Proteobacteria

Purple bacteria

▼ ♦•

Size Microscopic

Nutrition Include virtually all nutritional modes

known
Mode of life Major parasites and symbionts. some

free-living aquatic forms
An enormous and extremely varied group of bacteria
including many disease-causing forms leg.
Salmonella, a cause of food poisoning, and
Neisseria, which causes gonorrheal and symbionts

such as Escherichia coli. Proteobacteria show a
great range of physical structure and metabolic
activity; the group includes heterotrophs,
chemotrophs. chemoheterotrophs,
chemolithoautotrophs, photoautotrophs,
photoheterotrophs, methylotrophs, hydrogen-
oxidizers and sulfide-oxidizers. Many are facultative
aerobes, respiring oxygen when this is available but
able to survive by respiring, for example, nitrogen
IN2I or sulfate ISO4' I when not. Responsible for a
significant proportion of atmospheric nitrogen
fixation.

Saprospirae
Fermenting gliders

T4«

Size Microscopic

Nutrition Heterotrophic

Mode of life Free-living, or symbiotic in animals

One group includes anaerobic fermenters restricted
to anoxic environments. Some free-living; some le.g.
Bacteroides] inhabit intestinal tract of vertebrates,
including humans, in enormous numbers. A second
group includes aerobic forms inhabiting decaying
vegetation and other organic-rich environments.

Spirochaetae

Spirochaetes

T*«

Size Microscopic

Nutrition Heterotrophic

Mode of life Free-living or symbiotic; some parasitic

Spiral-shaped bacteria occurring in marine and
freshwater habitats, including deep muddy
sediments, and in animals, where many are major
parasites or symbionts. Some respire gaseous
oxygen, others are poisoned by it. Treponema
pallidum causes syphilis and yaws, and Leptospira
causes leptospirosis Twelve genera named to date.

TJiermotogae

Thermophilic fermenters

Size Microscopic

Nutrition Heterotrophic
Mode of life Free-living

▼ ♦ • Recently discovered obligate anaerobic bacteria

known from submarine hot vents, terrestrial hot
springs and subterranean oil reservoirs. Highly heat
tolerant, living at temperatures of 50°C to 80°C.
Ferment sugar and other organic compounds.

BACTERIA

228 WORLD ATLAS OF BIODIVERSITY

EUKARYA: ANIMALIA

Acanthocephala T ♦ •

Adult individuals anchor themselves to the gut wall

Thorny-headed worms

of vertebrates. Infection generally occurs after an

No. species described Over 1 000

intermediate invertebrate host is ingested. Thorny-

Proportion of group known Low/moderate

headed worms appear to alter host behavior so as to

Size Between 1 mm and 1 m in length

increase probability of host being ingested by

Nutrition Heterotrophic

predator, and so transfer parasite to further host.

Mode of life Parasitic worms that lack a free-liuing stage

Humans are seldom parasitized.

Annelida ▼ ♦ •

A large phylum including polychaetes 19 000

Annelids

species], oligochaetes 16 0001 and leeches 15001.

No. species described ca 16 000

Most are active predators and scavengers.

Proportion of group known Low/moderate

Polychaetes include free-living and tube-dwelling

Size From 0,5 mm to 3 m

marine species, mainly benthic but some pelagic.

Nutrition Heterotrophic

Oligochaetes occur in freshwater, estuaries and

Mode of life Segmented worms, mostly free-living m

deep sea, but are most numerous on land where

soils and sediments; some parasitic

earthworms are very important to soil structure.

Leeches are mainly free-living predators of

vertebrates and invertebrates in freshwaters or

water film on land; formerly more widely used

for medicinal purposes IH/rudo medianalisl.

Brachiopoda ▼

Cosmopolitan. Present between the intertidal zone

Lampshells

and 4 000 m depth. Usually occur cemented to surface

No. species described ca 350

by stalk Ipediclel; some species free-living on or in

Proportion of group known Low/moderate

marine sediment. Unlike moUusks, brachiopods have

Size From 2 mm to 10 cm

a lophophore, a specialized surface for gas exchange

Nutrition Heterotrophic

and food collection, and have dorso-ventral symmetry

t^ode of life Benthic, mainly sessile, marine animals

instead of lateral symmetry. Previously very diverse.

especially during the Paleozoic era Isee Chapter 3),

Some 30 000 extinct species have been described, and

Lingula, with fossils from 400 million years ago, may

be the oldest genus with living species.

Bryozoa ▼ ♦

Marine bryozoans mainly intertidal, but also on

Ectoprocts

seafloor to considerable depths About 50 freshwater

No. species described ca k 000

species known, with jelly-like colonies on plant

Proportion of group known Low

surfaces in slow streams. Large colonies 10.5 m in

Size Individuals mainly microscopic;

diameter! derived from the asexual budding of zooids

colonies to 0,5 m diameter

Hess than 1 mm in length! may contain several

Nutrition Heterotrophic

million individuals. Marine forms contribute to reef

Mode of life Sessile, colony-forming filter-feeders.

diversity

mainly marine

Cephalochordata T

Lancelets occur in estuary sediments and shallow

Lancelets

sandy seafloors. and live with the head protruding

No. species described 23

in order to screen out small plankton and organic

Proportion of group known Moderate/high

materials. They make up a small phylum of

Size From 5 cm to 15 cm

chordates with a cartilaginous rod dorsal to the

Nutrition Heterotrophic

gut InotochordI, a dorsal hollow nerve cord, and

f«1ode of life Free-living, filter-feeding marine animals

persistent gill slits in the pharynx, but without an

internal bony skeleton or cerebral ganglion. They

are the closest living relatives of vertebrate

animals. Used as human food in some areas.

APPENDIX 1 229

Chaetognatha ▼

Common plankton in open seas, especially abundant

Arrow worms

in warm seas down to 200 m. Detect prey, mainly

No. species described ca 7C

copepods, by vibration sensors, and can inject

Proportion of group known Low

neurotoxins. Important to marine fisheries as a

Size From 5 to 15 cm

source of food for fishes.

Nutrition Heterotrophic

Mode of life Worm-lil<e, planktonic marine predators

Cfielicerata T ♦ •

segmented bodies, chitinous exoskeleton, jointed

Chelicerates

appendages! with insects and crustaceans. Most

No. species described ca 75 000

chelicerates are in Arachnida (more than 75 000

Proportion of group known Low

speciesi; others are horseshoe crabs IMerostomatal

Size Macroscopic: the largest species of

and sea spiders jPycnogonidaj, Arachnids are

Pycnogonida have a leg span of almost

ubiquitous on land, with a few freshwater species; the

80 cm

group includes ticks and mites, some of considerable

Nutrition Heterotrophic

importance as vectors of disease in humans and

Mode of life Generally free-living and in most habitats

livestock Merostomata include Limulus, superficially

A very large and diverse arthropod phylum

unchanged since the Silurian (see Chapter 31,

characterized by claws Ichelicerae) on the anterior

Pycnogonids range from the shallows to the deep

pair of appendages, and sharing other features le.g.

ocean 16 800 ml and from pole to pole.

Cnidaria T ♦

A diverse phylum of radially symmetrical animals.

Cnidarians, hydras

including sea anemones, jellyfishes and corals.

No. species described ca 9 000

Specialized stinging cells called cnidoblasts are

Proportion of group known Moderate

diagnostic. Largest individuals le.g. lion's mane

Size Mainly macroscopic

Cyanea] may have tentacles many meters long. Reef-

Nutrition Heterotrophic, mostly carnivorous; reef-

building corals are of major importance in clear-

building corals contain photosynthetic

water coastal shallows in tropics and subtropics.

symbionts

Coral reefs often highly diverse and of great economic

Mode of life Aquatic, almost all marine; colonial and

value, Photosynthetic symbionts IdinomastigotesI of

solitary, free-swimming and sedentary

reef corals occur in polyp tissue at density up to 5

forms known

million/cm'; these require sunlight and limit reef

growth to upper part of the photic zone. Non-reef

coral without symbionts range down to 3 000 m.

Craniata T ♦ •

Agnatha in contrast to other vertebrates, all of which

Craniates or vertebrates

possess jaws IGnathostomatal. About half of all

No. species described ca 52 500

described vertebrate species are fishes: the

Proportion of group known High

Chondrichthyes (sharks and raysl, Osteichthyes (bony

Size From about 1 cm to 35 m

fishesi, the lungfishes and coelacanths. These make

Nutrition Heterotrophic

up around 25 000 species in lotal. The tetrapods (the

Mode of life Free-living species present in most

four-limbed non-fish vertebratesi include amphibians,

habitat types

reptiles, birds and mammals. Although not so

This group contains the vertebrates; a very large and

versatile as bacteria, vertebrates between them

very diverse phylum of chordates. all of which, unlike

extend from the air above the highest mountain to

the acraniate chordates lurochordates.

abyssal ocean depths, from sand desert to tropical

cephalochordatesi, have a brain enclosed within a

forest, and from hot springs to polar ice and subzero

skull Icraniuml. The majority have a bony internal

waters. The vertebrates include the most familiar

skeleton. A small group of mainly marine species

animals, and, with mollusks and crustaceans, most of

(lampreys and hagfishl lack jaws and are grouped in

those of direct nutritional importance to humans.

EUKARYA: AN I MALI A

230 WORLD ATLAS OF BIODIVERSITY

EUKARYA: ANIMALIA

Crustacea T ♦ •

water fleas, woodlice, etc. Size ranges between 0.25

Crustaceans

mm [Alonella] and 2.8 m [Macrocheira claw span].

No. species described ca 40 000

They are the predominant arthropods in most

Proportion of group known Low/moderate

freshwaters and widespread in all marine habitats.

Size Mostly macroscopic

from pelagic waters to ocean depths at 5 000 m, and

Nutrition Heterotrophic

in moist terrestrial situations. Numerous parasitic

Mode of life Mostly free-living in aquatic and humid

and commensal forms exist; pentastomids

terrestrial habitats, some parasitic

sometimes parasitize humans. Crustaceans sucti as

A very large and very diverse arthropod phylum

krill \Euph3U5ia superba] form key components of

distinguished by having two pairs of antennae.

the marine food web. There are numerous important

Species occur in virtually all habitats. Includes crabs,

crustacean fisheries.

crayfish, prawns, barnacles, copepods. brine shrimp.

Ctenophora T

A small phylum of translucent, soft-bodied predators.

Comb jellies

Widespread in marine waters and possibly the most

No. species described ca 100

abundant planktonic animals between 400 and 700

Proportion of group known Low

m depth. Their fragility makes them difficult to

Size Typically around 1 cm; largest up to 2 m

collect and study.

in length

Nutrition Heterotrophic

Mode of life Free-swimnning marine organisms

Ectiinodermata ▼

Invertebrates with five-part radial symmetry, an

Echinoderms

internal calcium carbonate skeleton, and a water

No. species described ca 7 000

vascular system. Includes starfish, sea urchins, sea

Proportion of group known Moderate

cucumbers and others. Mostly benthic in intertidal or

Size A few centimeters to near 2 m

subtidal habitats. Sea lilies extend to 10 000 m, and

Nutrition Heterotrophic

sea cucumbers in places make up nearly the entire

Mode of life Mostly free-living, benthic marine species

animal biomass at these abyssal depths. Viviparous

forms exist. Some, such as dried sea cucumbers

ItrepangI, used as human food.

Echiura T

Echiurans live in U-shaped burrows in marine

Spoon worms

sediments, rock crevices and mangrove, with some

No. species described ca UO

forms extending to abyssal depths around 10 000 m.

Proportion of group known Low/moderate

Echiurans have a flexible proboscis which may

Size From a few millimeters to 40 cm

extend to 1.5 m from the bulbous, unsegmented

Nutrition Heterotrophic

body Cilia move food items down the proboscis to

Mode of life Exclusively free-living marine organisms

the mouth.

Entoprocta T

Vl/idely distributed in shallow coastal waters.

Entoprocts

One freshwater species known. Colonies

No. species described ca 150

permanently attached by stalks, horizontal stolons

Proportion of group known Low

and basal discs to solid substrate, algae or other

Size Very small macroscopic animals, up to 1 cm

animals. Often form conspicuous mat-like growth on

Nutrition Heterotrophic

seaweed and rocks. Filter-feeders, consuming

Mode of life Mostly sessile, colonial marine

diatoms, desmids, other plankton and detritus.

organisms

Loxosometla is free-living, and moves by

somersaulting basal disc over tentacles.

APPENDIX 1 231

— ^y

Gastrotricha

Gastrotrichs

No. species described ca 400

Proportion of group known Low
Size Average length 0.5 mm

Nutrition Heterotrophic
Mode of life Free-livmg, worm-lil<e animals,
mainly marine

▼ ♦ The ventral side is ciliated and often glued to

substrate; exposed surfaces bear bristles or scales.
Most occur in subtidal or intertidal sediments where
form part of the marine meiobenthos: freshwater
forms most abundant in small still waters. Important
scavengers of dead bacteria and plankton.

Gnathostomulids

Jaw worms

No. species described ca 80

Proportion of group known Low

Size Average length around 1.5 mm

Nutrition Heterotrophic

Mode of life Free-living marine worms

A small phylum of translucent, benthic. worms
capable of surviving in sediments very low in oxygen
and high m hydrogen sulfide. Graze on baclena,
proloctists and fungi in marine sediments. Have
been found at several hundred meters depth.
Population densities may exceed 6 000 per liter of
sediment, outnumbering nematodes.

Hemichordata

Acorn worms

No. species described ca 90

Proportion of group known Low/moderate

Size Adults between 2.5 and 250 cm

Nutrition Heterotrophic

Mode of life Sedentary, benthic marine species

A small phylum of soft-bodied, benthic marine
worm-like animals. Adults mostly sedentary and live
burrowed in soft sediment of shallow seas, or in
secreted tubes. Sexual and asexual reproduction
occurs; colonies may be formed by budding.
Hemichordates were previously classified as
chordates. and resemble them in having ciliated
gill slits in pharynx.

Kinorhyncha

Kinorhynchs

No. species described ca 150

Proportion of group known Low/moderate

Size Up to 1 mm in length

Nutrition Heterotrophic

Mode of life Free-living marine animals

Kinorhynchs are cosmopolitan in muddy bottom
habitats, including estuaries and the intertidal zone,
and to a depth of approximately 5 000 m. Some
species are commensal with hydrozoans, bryozoans
and sponges.

Loricifera

Loriciferans

No. species described ca 100

Proportion of group known Low

Size Microscopic

Nutrition Heterotrophic

Mode of life Benthic marine species

Widespread, probably cosmopolitan part of
mlerstitial fauna. Life history incompletely known.
Adults are sedentary on sand or graveL sometimes
ectoparasites; the larvae are believed to be free-
living and mobile. Protective plates cover the
abdomen, into which the neck and head with
mouth cone can be retracted.

Mandibulata T ♦ •

Mandibulates

No. species described ca 950 000

Proportion of group known Low

Size From near microscopic to many

centimeters
Nutrition Heterotrophic
Mode of life Most are free-living terrestrial species
An exceptionally large and diverse arthropod phylum
distinguished by a pair of crushing mandibles.
Jointed chilinous exoskeleton and a single pair of

antennae. Includes insects IHexapodal. centipedes
and millipedes iMyriapodal. symphyla and
pauropods. The largest group of animals, Hexapoda.
contains some 950 000 described species and may
number in millions. Insects range up to 30 cm length
(giant stick insect Pharnacia serratipes]. Social
insects such as termites, ants, some bees and
wasps can form large colonies. Many insects are
important crop pests or disease vectors; others are
beneficial becarse of crop pollination or pest control.

EUKARYA: ANIMALIA

232 WORLD ATLAS OF BIODIVERSITY

EUKARYA: ANIMALIA

Mollusca T ♦ •

deserts. Includes snails, slugs, mussels, chitons.

Mollusks

octopods, squid and others. Most species are free-

No. species described At least 70 000 and

living, although some are parasitic or commensals.

possibly more than

Many, e.g. bivalves, are sedentary as adults. Size

100 000

reaches maximum, approaching 20 m, in the giant

Proportion of group known Moderate

squid Architeuthis. Many important as food source.

Size Range from near microscopic to

Some forms act as intermediate hosts to parasites

several meters

(e.g. Sch/5(osom3| that cause serious human

Nutrition Heterotrophic

disease; other species can cause significant damage

Mode of life Mainly free-living species present in

to crops and constructions le.g. Dre/ssenal. Venom

most habitat types; some parasitic

of some marine gastropods is of medical interest.

A large and highly diverse phylum, with species

Monoplacophorans, the most primitive mollusks.

occurring in benthic and pelagic marine viiaters, in

were first seen alive in the 1970s but abundant in

freshvioters of all kinds, and on land, from forests to

Paleozoic Isee Chapter 31.

Nematoda T ♦ •

Possibly the most abundant animals living on

Nematodes

Earth, found in virtually all habitats and in many

No. species described ca 20 000

other organisms. Free-living forms are key to

Proportion of group known Low

decomposition and nutrient cycling. Many species

Size From 0.1 mm to 9 m in length

are important parasites of plants and animals.

Nutrition Heterotrophic

including humans le.g. filariasisl. Nematodes have

Mode of life Free-living or parasitic worm-like

provided important research animals in genetics and

animals

cell differentiation.

Nematomorpha T ♦ •

A small phylum of leathery, unsegmented, worm-like

Nematomorphs

animals. Occur widely in aquatic or moist terrestrial

No. species described ca 240

habitats. Eggs hatch into minute motile larvae which

Proportion of group known Low

enter host and metamorphose into immature worms.

Size Ranging from 10 to 70 cm in length

These burst out, killing the host, when near water or

Nutrition Heterotrophic

during ram. Hosts include annelids and arthropods.

Mode of life Adults are free-living and usually

Rarely found in humans, where appear non-

aquatic; all are endoparasitic at some

pathogenic.

stage

Nemertina T ♦ •

Characterized by the slender anterior proboscis,

Ribbon worms

used for predation. defense and locomotion.

No. species described ca 900

Abundant in the intertidal zone; some forms are

Proportion of group known Low/moderate

pelagic. Freshwater and terrestrial forms are known.

Size Macroscopic, from 0.5 mm to 30 meters

Some species are symbionts or parasites; recorded

in length, mostly small

from echmoderms, nemertmes, annelids, mollusks

Nutrition Heterotrophic

and flatworms. May affect host reproduction.

Mode of life Mostly free-living, predatory marine

Malacobdella is a filter-feeder, inhabiting the mantle

worms

cavity of clams.

Onychophora •

Require high humidity levels to counter water

Velvet worms

loss through thin chitinous cuticle. Many occur

No. species described ca 100

in forest; some in caves. Alt carnivorous. Walk

Proportion of group known Low

slowly on U-43 pairs of stumpy legs. Two geographic

Size From 1 to 20 cm

groups exist; one mainly warm northern hemisphere;

Nutrition Heterotrophic

the other southern hemisphere. Many tropical

Mode of life A small phylum of free-living,

species are viviparous.

terrestrial, worm-like animals

APPENDIX 1 233

Orthonectida T

Recorded from echinoderms, nemertines,

Orthonectida

annelids, mollusks and flatworms. Less benign

No. species described ca 2G

than rhombozoans; may affect host reproduction.

Proportion of group known Low

Size Microscopic

Nutrition Heterotrophic

Mode of life Worm-lil(e internal parasites or

symbionts of marine invertebrates

Ptioronida T

Most inhabit leathery chitinous tubes encrusted with

Phoronids

sand or shell fragments, some burrow in mollusk

No. species described U

shells or rock. From coastal shallows to 400 m

Proportion of group known Low/moderate

depth. Filter-feed on plankton and detritus.

Size From 1 mm to 50 cm

Cosmopolitan but not abundant; half the known

Nutrition Heterotrophic

species occur on Pacific coast of North America.

Mode of life Sedentary filter-feeding marine worms

Placozoa T

Discovered in a seawater aquarium in 1883, and

Trichoplax

since reported in shallow marine water and marine

No. species described 1

research stations, Tnchoptax adhaerens is the least

Proportion of group known High

complex of all living animals, consisting of a few

Size Up to 1 mm

thousand cells but no distinct tissues, little is known

Nutrition Heterotrophic

of its life history.

Mode of life Very small marine animal

Platytielmintties T ♦ •

Flatworms, flukes and tapeworms. Four classes.

Flatworms

Free-living soil flatworms most abundant in tropics;

No. species described ca 20 000

aquatic forms mainly temperate. Some can survive

Proportion of group known Low/moderate

in environments low in oxygen by oxidizing hydrogen

Size Often a few millimeters; tapeworms to

sulfide. Parasitic forms include flukes such as

30 m in length

Schistosoma, the cause of schistosomiasis, and

Nutrition Heterotrophic

tapeworms, obligate parasites of vertebrate gut-

Mode of life Free-living or symbionts, many

Many flatworm parasites have complex life cycle

parasitic; found in freshwater, marine

with infective larvae and intermediate hosts.

and terrestrial environments

Pogonophora ▼

shallow polar seas or Ithe vestimentiferansi around

Beard worms

hot submarine vents with a high hydrogen sulfide

No. species described More than 1 20

and methane content. Greatest diversity in the

Proportion of group known Low

western Pacific, Adult pogonophorans have no gut

Size From 10 cm to 2 m in length

and probably absorb nutrients directly from tentacles.

Nutrition Heterotrophic

Vent-living forms derive nutrients and energy from

Mode of life Sessile, benthic marine worms

the oxidation of hydrogen sulfide through symbiotic

Pogonophora live in fixed upright chitin tubes secreted

chemoautotrophic bacteria, which can occur at

in sediments, shell or decaying wood on the ocean

densities of 1 billion per gram of body tissue.

floor Most abundant in cold, deep waters.

Porifera T ♦

The vast ma]onty of sponges are marine and about

Sponges

100 species are freshwater Filter-feeders lone

No. species described 5 000-10 000

Mediterranean form passively captures crustaceans

Proportion of group known Moderate

and digests them externallyl. Many include

Size Macroscopic, some to 2 m in height

pholosynthetic symbionts, e.g. cyanobacteria. and

Nutrition Heterotrophic

brown, red or green algae. Sponges have simple

Mode of life Sedentary aquatic animals

structure with no tissues or organs and are generally

supported by calcareous or silicaceous spicules or

fibrous, proteinaceous matrix.

EUKARYA: ANIMALIA

234 WORLD ATLAS OF BIODIVERSITY

EUKARYA: AN I MALI A

Priapulida ▼

Found in sand or mud. from intertidal pools to

Pnapulids

abyssal depths, and from tropical waters to the

No. species described 17

Antarctic, Approximately half of the described

Proportion of group known Low

species are part of the marine meiobenthos li.e.

Size Range between 0.5 mm and 30 cm

small bottom- or sediment-living species between

Nutrition Heterotrophic

about 0.5 and 1 mm).

Mode of life Exclusively marine, free-living, worm

like animals

Rhombozoa ▼

Mostly found in the kidneys of squid and octopus m

Rhombozoans

temperate waters.

No. species described ca 70

Proportion of group known Low

Size Up lo 5 mm

Nutrition Heterotrophic

Mode of life Worm-like internal parasites or

symbionts of benthic cephalopod

mollusks

Rotifera ▼ ♦ •

Mainly freshwater, also in moist habitats on land;

Rotifers

about 50 species occur in benthic and pelagic

No. species described ca 2 000

marine habitats. Rotifers are the most abundant

Proportion of group known Low

and cosmopolitan of the freshwater zooplankton.

Size Mostly microscopic, some to 2 mm

Mostly free-living. Many live on other invertebrate

Nutrition Heterotrophic

organisms; many are endoparasites of invertebrates.

Mode of life Mostly free-swimming in freshwaters

Most free-living rotifers reproduce parthenogenetically.

Sipuncula ▼

Benthic species, mainly in shallow, warm marine

Peanut worms

habitats; most abundant on rocky shores, but also

No. species described ca150

present in polar regions and down to 7 000 m in the

Proportion of group known Low

abyssal ocean. Ingest diatoms and other protoctists

Size A few millimeters to 0.5 m in length

or organic debris. Used locally as human food in the

Nutrition Heterotrophic

Indo-Pacific and China.

Mode of life Exclusively marine, worm-like.

burrowing or crevice-dwelling

organisms

Tardigrada T ♦ •

Widely distributed from pole to pole. All are aquatic;

Water bears

land species live in the water film on mosses, forest

No. species described ca 750

litter and other habitats. A few marine species. Move

Proportion of group known Low

on four pairs of stumpy legs. Mainly ingest liquid

Size Ivlainly microscopic

food obtained by piercing protoctists, animals or

Nutrition Heterotrophic

plants The Mesotardigrada Igenus Thermozodiuml

Mode of life Free-living animals, mostly in moist

inhabit hot springs. Can survive extreme desiccation

terrestrial and freshwater habitats

with low metabolism or m encysted form, and when

dormant Icryptobioticl may be tolerant of

exceptionally high or low temperatures, approaching

absolute zero.

APPENDIX 1 235

Urochordata ▼

Sea squirts

No. species described ca 1 400

Proportion of group l<nown Low/moderate

Size From 1 mm to 2 cm

Nutrition Heterotroptiic

Mode of life Small, marine, filter-feeding animals

Adults may be either benthic and sedentary Iclass
Ascidiacae, tunicatesl or pelagic and free-swimmmg
Iclass Larvaceal; Thaliacea or salps also free-
swimming. All are ciliary filter-feeders, Urochordata
have a dorsal hollow nerve cord, a cartilaginous rod
dorsal to the gut InolochordI and gill slits in the
pharynx at some stage, these being features of
chordates.

Asconnycota T ♦ •

Ascomycotes

No. species described ca 30 000

Proportion of group l<nown Low

Size Mainly microscopic but reproductive

body up to several centimeters
Nutrition Heterotrophic
Mode of life Mainly terrestrial, in soil and leaf litter,

or on and in other organisms
A large diverse phylum distinguished from other
fungi by the microscopic, sac-like, spore-producing
reproductive structure lascusL Includes yeasts,
morels, truffles, blue-green molds and lichens.
Thread-like hyphae form network Imyceliuml

through substrate. Many are Iree-living; many
are parasitic. Several hundred forms occur in
freshwaters. More than 10 000 are the heterotrophic
components of lichens, which are joint organisms
formed by ascomycotes with either photosynthetic
green algae or cyanobacteria. As with other fungi,
ascomycotes fulfil a key ecological role in breaking
down organic material and transferring inorganic
nutrients and water from soil to plants. Some are
important in food preparation (e.g. baker's yeastl.
Many cause important diseases of plants and
animals, including humans, while others are
sources of key medicinal substances such as
penicillin and similar antibiotics.

Basidiomycota ? ♦ •

Basidomycotes

No. species described ca 22 250

Proportion of group known Low/moderate
Size Generally microscopic or somewhat

larger, but hyphae extend considerable

distance, and fruiting bodies to 10 cm

or more
Nutrition Heterotrophic
Mode of life Mainly terrestrial, in soil and leaf litter,

or on trees and other organisms
A large phylum including typical mushroonns,
puffballs, stinkhorns, and rusts and smuts (some of
which cause economically important plant diseases!.

All are characterized by microscopic, club-shaped,
spore-producing reproductive structures, typically
borne in great numbers on a basidiocarp - the
familiar mushroom. Largest known mushroom
specimen grew to U6 cm wide and 54 cm high.
Many basiomycotes form mycorrhizas with trees
and shrubs; as with zygomycotes. the fungi
move basic nutrients from soil to plant, and
plant carbohydrates move into the fungus. Many
ascomycotes fulfil a key ecological role in breaking
down organic material, and transferring inorganic
nutrients and water from soil to plants. Several
species are valued wild food items. A few freshwater
forms are known.

Zygomycota ? ♦ •

Zygomycotes

No. species described ca 1 100

Proportion of group known Low

Size Generally microscopic or a little larger,

but hyphae may extend considerable

distance through soil
Nutrition Heterotrophic
Mode of life Mainly terrestrial in soil and leaf litter

Most are saprobic Isaprophyticl, secreting digestive
enzymes into organic material and absorbing
nutrients released. Many are parasites on
protoctists, small animals, plants or other fungi.
Zygomycotes include about 100 species that form
mycorrhizal associations, in which fungal partner
contacts or enters plant roots and assists inflow of
nutrients and absorbs organic plant substances in
exchange. Most vascular plants probably have such
relationships with fungi and these may be critical in
nutrient-poor soils. A few occur in freshwaters.

EUKARYA: ANIMALIA

EUKARYA: FUNGI

236 WORLD ATLAS OF BIODIVERSITY

EUKARYA: PLANTAE

Anthocerophyta

Horned liverworts

No. species described ca 100

Proportion of group l<nown Moderate

Size Low-growing plants

Nutrition Ptiotosynttietic

Mode of life Terrestrial, in moist tiabitats

A small group of non-vascular plants of moist
tiabitats, typically on woodland floor or water
margins. Present worldwide in temperate and
tropical regions. Among first colonists of bare
substrates, including rocks. Some species have
associated nitrogen-fixing cyanobacteria.

Anthophyta ▼ ♦ •

Flowering plants, angiosperms

No. species described ca 270 ODD

Proportion of group known High

Size From less than 1 mm in length IWolffia

angusta] to more than 100 m

|fuca/yp(us regnansi
Nutrition Photosynthetic
Mode of life Flowering plants occurring in most

habitat types
An extremely diverse, geographically cosmopolitan,
phylum of vascular seed plants, distinguished by
flowers, and fruits that enclose the fertilized seeds.
The great majority of species are terrestrial, in
virtually all habitat types. Many occur in or around

lakes, rivers and wetlands, and seagrasses occur
subtidally in shallow marine waters. Two main
groups are distinguished, according to whether
the germinating seed has one or two seed leaves;
Monocotyledones and Dicotyledones. Monocots
include palms, lilies and the economically vital
grasses; most monocots are herbaceous and
woody forms lack special tissue that secondarily
adds width to the trunk. Dicots form the larger
group. Success of anthophytes appears linked
to coevolution with animals, m particular with
specialized modes of pollination and seed dispersal.
All major food and medicinal plants, and hardwood
timber trees, are found in this ohylum.

Bryophyta ♦ #

Mosses

No. species described ca 10 000

Proportion of group known Moderate/high

Size Low-growing plants

Nutrition Photosynthetic

Mode of life Terrestrial, mainly in moist habitats and

wetlands
A large phylum of non-vascular plants li.e. lacking
specialized xylem and phloem transport tissue!

consisting of mosses and the genus Takakia. Often
conspicuous in cold or cool temperate habitats,
particularly tundra, where mosses are the dominant
plants, and also in heathland, bogs, woodland,
waterlogged areas and freshwater margins. Most
diverse in moist tropical habitats. Many mosses well
adapted to withstand desiccation; some occur in
warm and regions. Peat moss [Sphagnum]
contributes to the development of new soils.

Coniferophyta #

Conifers

No. species described 630

Proportion of group known High

Size Shrubs or large trees, up to 100 m in

height
Nutrition Photosynthetic
Mode of life Terrestrial

Conifers are cone-bearing gymnospermous vascular
plants, with needle-shape leaves, and are mostly

evergreen trees. They form extensive forests at high
latitudes in the northern hemisphere, and also occur
more locally, often on and mountains; also common
in the tropics and in temperate southern forests,
where Araucana is widespread. In Sequoiadendron
and Sequoia, conifers include the largest living plants.
Many species in mountainous and northern areas
have characteristic symbiotic mycorrhizal fungi.
Conifers provide timber, paper pulp and ornamental
plants, and some have food or medicinal value

Cycadophyta t

Cycads

No. species described U5

Proportion of group known High

Size From shrubs to small trees of 18 m in

height
Nutrition Photosynthetic
Mode of life Terrestrial

A small phylum of seed-bearing, often palm-like,
vascular plants restricted to the tropics and

subtropics, where are present in a range of habitats,
from moist forest to deserts and coastal mangroves.
Diverse in the Cretaceous. Cycads are gymnosperms,
I.e. seeds do not become enclosed in a fruit. Many
are insect pollinated, often by beetles. All species
have symbiotic, nitrogen-fixing cyanobacteria.
Cycads provide a variety of materials, including
thatch, food, medicines and ornamental plants.
Cycad starch for bread requires special treatment to
destroy potentially fatal toxins.

APPENDIX 1 237

Fillcinophyta 4 •

Ferns

No. species described ca 12 000

Proportion of group known High

Size From a few centimeters to 25 m

Nutrition Ptiotosynttietic

Mode of life Terrestrial; a few in freshwater

A diverse phylum of vascular plants, the most

species-rich group of plants lacking seeds.

widespread from cold temperate areas to the tropics.

Mainly in moist areas, such as forest floor and
stream margins; species diversity highest In tropics,
where many forms are epiphytic and some species
grow as trees to 25 m in height. The aguatic Azolla. a
very small floating fern, has symbiotic, nitrogen-
fixing cyanobacterla. Several food, medicinal and
other products are derived from ferns. Ferns,
especially tree ferns, were diverse and very abundant
in Devonian and Carboniferous times.

Ginkgophyta

Ginkgo

No. species described 1

Proportion of group known High
Size To 30 m in height

Nutrition Photosynthetic
("lode of life Terrestrial

A vascular seed-bearing tree, characterized by a fan-
shaped leaf with bifurcating veins and a fleshy
exposed ovule, now restricted as a wild species to
steep forest in southern China. A wide diversity of
ginkgophytes, of which Ginkgo biloba is the only
survivor, existed during the Mesozoic, Now widely
planted for ornamental purposes. It is a gymnosperm.
I.e. seed not enclosed in a fruit. Leaf extract used as a
traditional food and medicine in East Asia.

Gnetophyta

Gnetophytes

No. species described ca 70

Proportion of group known High

Size Small trees, shrubs or vines

Nutrition Photosynthetic

Mode of life Terrestrial

A small phylum of vascular seed plants, distinguished
from other gymnosperms by having vessels for water
transport similar to those of flowering plants. The
three living genera differ greatly from each other
Some plants of the genus Gnetum in tropical moist
forest grow to 7 m in height; some Welwitschia, a
unique, low-growing, cone-bearing plant of southwest
African deserts, may be 2 000 years old.

EUKARYA: PLANTAE

Hepatophyta

Liverworts

No. species described ca 6 000

Proportion of group known Ivloderale

Size Low-growing plants

Nutrition Photosynthetic

Mode of life Terrestrial in moist habitats

♦ • Non-vascular plants typically found in moist habitats

growing on woodland floor, shaded stream banks,
waterfalls or rocks; often epiphytic and often occur
with mosses IBryophytal. Widespread in cold
temperate regions, present in Antarctica, but species
diversity highest in tropics Often among first plants
to colonize burned or newly exposed substrates.

Lycophyta ♦ #

Club mosses *

No. species described ca 1 000

Proportion of group known Moderate

Size Mainly low-growing herbaceous plants

Nutrition Photosynthetic

Mode of life Terrestrial, in moist and dry habitats

Small seedless evergreen vascular plants found in
temperate and tropical habitats, typically on forest
floor in temperate regions although most tropical
species are epiphytic. A few occur in and areas.
Lycophytes were prominent in Paleozoic plant
communities before evolution of flowering plants;
although all living species are small, trees up to AO
m in height were dominant in Carboniferous coal
forests. Some similarity to mosses and conifers but
unrelated to either

238 WORLD ATLAS OF BIODIVERSITY

EUKARYA: PLANTAE

EUKARYA:
PROTOCTISTA

Psilophyta 9

Whisk fern

No. species described 10

Proportion of group l(nown Moderate/fiigfi

Size Small herbaceous plants

Nutrition Photosynthetic and symbiotic with fungi

Mode of life Terrestrial

A very snnall group of vascular plants, the only
ones lacking both roots and leaves. Similar to
earliest simple, leafless land plants of late Silurian
and Devonian times, 400 million years ago. and
conceivably direct descendants of them. Present as
epiphytes or ground-living species with a restricted
range in subtropics and temperate areas. These
plants have a mycorrhizal association lalso seen
in earliest fossil forms! with fungal hyphae that
increase the flow of soil nutrients to the non-
photosynthetic plant cells.

Splienophyta

Horsetails

No. species described 15

Proportion of group knovKn High
Size Herbaceous plants

Nutrition Photosynthetic
Mode of life Terrestrial

A small phylum of seedless vascular plants,
with jointed ridged stems and tiny scale-like
leaves. Found in moist or disturbed areas, including
urban areas and roadsides; more typically in moist
woods and wetland margins, and also in salt flats.
Historically consumed as food in Europe and North
America, poisonous to livestock. As with lycophytes,
sphenophytes were diverse and abundant in
Devonian and Carboniferous forests, with tree-like
forms up to 15 m high.

Actinopoda T ♦ •

Radiolarians

No. species described ca 4 000

Proportion of group known Low

Size Ivlicroscopic

Nutrition Heterotrophic but most hold symbiotic

photosynthetic haptomonads
Mode of life Mostly marine, although the Heliozoa is

mainly freshwater

Relatively large, generally unicellular protoctists with
radial symmetry. Some form large colonies in which
many individuals are embedded in a jelly-like matrix.
Some occur in open ocean waters, some are benthic.
Many have silicaceous skeletons with spines or oars
used for swimming. Most acantharians include
photosynthetic grass-green haptomonad or yellow or
green algae symbionts.

Apicomplexa ? ? •

Sporozoa

No. species described ca 5 000

Proportion of group known Low

Size Single-celled

Nutrition Heterotrophic

Mode of life Symbiotic with or parasitic on animals

Many of these spore-forming protoctists are
bloodstream parasites with complex life cycles.
Coccidians are the best-kno«in group because
infection often causes serious or fatal intestinal
tract infection. Many e.g. Eimena. infect livestock;
kospora hominis is the only direct coccidian parasite
of humans. Plasmodium causes malaria, probably at
present the single most important infectious disease
affecting humans.

Archaeoprotista T ^ #

Amitochondriates
No. species described '

Proportion of group known Low''
Size Single-celled

Nutrition Mostly heterotrophic
Mode of life Free-living in aquatic habitats, and
symbionts, often parasitic, in animals

Anaerobic and lacking mitochondria. Many forms are
parasitic or symbiotic in the intestines of animals,
eg. wood-eating termites and cockroaches. Giardia
causes giardiasis in humans.

APPENDIX 1 239

Chlorophyta T ♦

Green algae

No. species described ca 16 000

Proportion of group known Low
Size Range from single-ceiled to

macroscopic green seaweeds
Nutrition Photosynthetic
Mode of life Diverse, mostly marine and freshwater

algae; a few symbiotic with other

organisms

Chlorophyles include unicellular and complex
multicellular species as well as forms with many
nuclei sharing the same cytoplasm, |v|a|or primary
producers, they are estimated to fix over 1 billion
tons of atmospheric carbon annually Symbiotic
forms include Ptatymonas in the flatworm Convolula
roscoffensis. Some forms are resistant to al least
periodic desiccation. Some early form of chlorophyta
almost certainly gave rise to plants.

▼ ♦

Chrysomonada

Chrysophyta

No. species described i

Proportion of group known Low'

Size fvlost single-celled; some form large

branching colonies
Nutrition Photosynthetic
Mode of life Free-living, mainly in freshwaters

A large and diverse group of algae with golden-
yellow pigments. The silicotlagellates are a
component of marine plankton and extract
silica from sea water to form shells.

Chytridiomycota

No. species described ca 1 000

Proportion of group known Low
Size Ivlicroscopic

Nutrition Heterotrophic
Mode of life Decomposers or parasites in
freshwater or moist soils

♦ • Feed by extending threadlike hyphae into living

hosts or dead material. Simplest forms grow entirely
within the cells of their hosts. Some are associated
with plant diseases, e.g. Physoderma zea-maydis
causes brown-spot in maize. Cell walls of chitin;
" some with cellulose also. Chytrids may be ancestral

to fungi.

EUKARYA:
PROTOCTISTA

Ciljopfiora T ♦

Ciliates

No. species described ca 10 000

Proportion of group known Low
Size Ivlostly microscopic and single-celled

Nutrition Mostly heterotrophic
Mode of life Mostly free-living; some symbionts or
parasites

Although most are unicells, a few multicellular
forms resembling slime molds exist. Ciliates feed
on bacteria or absorb nutrients from the surrounding
medium, Entodmiomorphs live as symbionts in the
stomachs of ruminants; the parasite Balantidium
sometimes causes disease in humans. The free-
living Paramecium and Slentor are well studied
and much used in research and education.

Cryptomonada

Cryptophyta

No. species described '

Proportion of group known Low''

Size Single-celled

Nutrition Some heterotrophic; others

photosynthetic
Mode of life Mostly free-living

T ♦ Cosmopolitan in moist areas. Most cryptomonads are

flattened, ellipticaL free-swimming cells in freshwater.
Marine species may form blooms on beaches; others
are intestinal parasites. Heterotrophs ingest bacteria
and protoctisls. Some of the photosynthetic forms
possess yellow and red pigments in addition to
chlorophyll, and some also contain blue-red
phycocyanin pigments. Some form colonies of
non-mobile cells embedded in a gel-like matrix.

Diatoms

T*«

Widely distributed in the photic zone of marine and
inland waters worldwide. Some occur in moist soils.

No. species described ca 1 000

Diatoms have distinctive paired tests or shells of

Proportion of group known Low

organic material impregnated with silica extracted

Size Single-celled; some colonial

from surrounding water Important basal

Nutrition Mostly photosynthetic;

components or marine and freshwater food webs.

some saprophytes

Mode of life Mostly free-livmg

2*0 WORLD ATLAS OF BIODIVERSITY

EUKARYA:
PROTOCTISTA

Dinomastigota ▼ ♦

Dinoflagellafes

No. species described ca 4 000

Proportion of group known Low
Size Single-celled, up to 2 mm.

occasionally colonial
Nutrition Some are heterotrophic;

others photosynthetic
Mode of life Mostly free-living marine plankton

Typically planktonic; some symbiotic with or live
on marine animals or seaweed, some occur in
freshwalers. Many adopt very different forms at
different life stages. Gymnodintum microadnaticum
IS the most common intracellular photosynthesizing
symbiont in corals. Some produce powerful toxins
and are an important cause of fish mortality le.g.
Pleistena piscicida] and may form toxic 'red tides'
le.g. Gonyaulax tamarensis]. Ciguatera poisoning in
humans is caused by accumulations of dinoftagellate
toxins in fishes and marine invertebrates. Many
species le.g. Noctiluca] are bioluminescent.

Oiscomitochondria ▼ ♦ •

Flagellates, zoomastigotes
No. species described ?

Proportion of group known Low?
Size Mostly unicellular

Nutrition Generally heterotrophic; most

euglenids are photosynthetic
Mode of life Mainly free-living in a wide range of

aquatic and terrestrial habitats

All formerly regarded as protozoan animals, and in
medical literature are commonly termed flagellates.
Most feed on bacteria or absorb nutrients directly
from surroundings; some, i.e. euglenids. are usually
photosynthetic. Some are symbiotic or parasitic, the
latter including organisms I rfypanosomal responsible
for sleeping sickness and Chagas disease.

Eustigmatophyta

Green eyespot algae

No. species described '

Proportion of group known Low?

Size Single-celled

Nutrition Photosynthetic

Mode of life Free-living algae, mostly freshwater

Planktonic algae with yellowish-green pigments,
typically at the base of freshwater food webs. A few
multicellular forms are known. Nine genera
described.

Gamophyta i

Conjugating green algae

No. species described Several thousand

Proportion of group known Low

Size Multicellular forms are macroscopic

Nutrition Photosynthetic

Mode of life Freshwater algae

Multicellular filament-forming or unicellular green
algae found in freshwaters. Many contribute to algal
blooms and pond scum. Filamentous forms include
Spirogyra. Desmids consist of paired cells joined
at a narrow bridge through which their cytoplasm
IS continuous.

Granuloreticulosa ▼ ♦

Foraminifera and reticulomyxids

No. species described ca 4 000

Proportion of group known Low

Size Mostly microscopic, but some several

centimeters in diameter
Nutrition Heterotrophic, some with

photosynthetic symbionts
Mode of life Mostly benthic, but some are free-

swimmmg planktonic organisms; nearly

all marine

Foraminifera have multipored shells ItestsI
composed of organic matter reinforced with
minerals Isand or calcium carbonatel. Important
in marine food webs. Many marine sediments are
composed largely of foraminifera, and fossil species,
about 40 000 of which are known, are important in
stratigraphy Some of latter, e.g. Nummulites. can
be up to 10 cm in diameter Reticutomyxids lack
shells and form soft reticulate masses.

APPENDIX 1 241

Haplospora

No. species described 33

Proportion of group known Low
Size Single-celled

Nutrition Heterotrophic
Mode of life Unicellular symbionts living in tiie
tissues of marine animals

Life history incompletely known, but characterized by
production o( spores into water or host tissue. Many
are benign symbionts, and exist in multinucleate
"Plasmodium form, but several are parasitic and
damage host tissues. Host animals include
mollusks, nematodes, trematodes and polychaetes.
Often found as parasites of parasites, e.g. within
Irematode parasites of oysters. Formerly regarded
as sporozoans.

Haptomonada ▼ ♦

Prymnesiophytes
No. species described
Proportion of group known Low
Size Mostly single-celled

Nutrition Photosynthetic
Mode of life Aquatic, with free-living and resting
stages

Most are marine; some occur in freshwaters. Two
distinct life stages: a motile, golden-colored alga
and a resting, coccolithophorid stage, covered
in distinctive calcareous plates Icoccolithsl.
Coccolithophorids are important in calcium
carbonate sediments and in stratigraphic studies.
Some are endosymbionfs of radiolaria lActinopodal.

Hyphochytriomycota ♦ #

Hyphochytrids

No. species described 23

Proportion of group known Low
Size Microscopic

Nutrition Heterotrophic
Mode of life Present in freshwaters and soil
moisture, saprophytic or parasitic

Feed by extending threadlike hyphae into host
tissue, typically algae or fungi, or into organic
remains, e.g. insect or plant debris, where digestive
enzymes are released and nutrients absorbed.
Formerly regarded as fungi.

Labyrinthulata ▼

Slime nets and thraustochytrids

No. species described '

Proportion of group known Low?

Size Colonies up to a few centimeters long

Nutrition Heterotrophic

Mode of life Colonial marine protoctists

Slime nets consist of a complex colonial network of
cells that move and grow within an extracellular
slime matrix of their own making. Labynnthula
grows on eel grass \Zostera\ where possibly
pathogenic. Eight genera described.

Microspora ? ? •

Microsporans

No. species described ca 800

Proportion of group known Low

Size Single-celled

Nutrition Heterotrophic

Mode of life Intracellular parasites of animals

Anaerobic and lacking mitochondria Frequently
orm large single-cell tumors in host animals, some
highly pathogenic, some harmless. Nosema causes
pebrine, a disease of silkworm larvae.

Myxomycota ? ? •

Plasmodial slime molds

No. species described ca 500

Proportion of group known Low

Size Microscopic cells but macroscopic, up

to several centimeters, in plasmodial
form

Nutrition Heterotrophic

Mode of life Free-living organisms in damp
terrestrial habitats

Similar in some respects to cellular slime molds
IRhizopodal. Myxomycotes have a sexual stage,
and the Plasmodium that develops from the zygote
is multinucleate. Fruiting stage develops in drier
conditions. Feed by enveloping bacteria and
protoctists growing on decaying vegetation.
Key organisms in studies of cell motility

EUKARYA:
PROTOCTISTA

242 WORLD ATLAS OF BIODIVERSITY

A'-

EUKARYA:
PROTOCTISTA

Myxospora T ♦

Myxospondians

No. species described ca 1 100

Proportion of group known Low

Size Infected tissue may tiave growths of

several centimeters in diameter

Nutrition Heterotrophic

Mode of life t»lulticellular symbionts, mostly

parasites of fishes but also of marine
and freshwater invertebrates

Myxospondians penetrate the host integument
and travel to the intestine where amoeboid forms
carried to target organs are released. Many form
large plasmodial masses attached to internal
organs. Hosts include sipunculans and freshwater
oligochaete worms. Most appear benign, including
the fish symbionts, but some are important
pathogens, e.g. Myxostoma cerebralis causes
twist disease of salmon. Formerly regarded
as sporozoans.

Oomycota ♦ •

Oomycetes

No. species described Several hundred

Proportion of group l<nown Low
Size Microscopic

Nutrition Heterotrophic symbionts
Mode of life Mostly in freshwaters or soil; some
parasitic on land plants

Feed by extending threadlike hyphae into host tissue
where release digestive enzymes and absorb
nutrients. Familiarly known as water molds, white
rusts and downy mildews. Many oomyctes are very
important crop pests, e.g. Phytophthora inlestans
causes potato blight; Saprolegnia parasitica attacks
freshwater and aquarium fishes. Formerly regarded
as fungi.

Paramyxa

Low

No. species described
Proportion of group known
Size Microscopic

Nutrition Heterotrophic
Mode of life Obligate symbionts living within the
cells of marine invertebrates

Characterized by production of multicelled spores
within host tissue Live within annelids, crustaceans,
mollusks and probably other groups of marine
invertebrates. Formerly regarded as sporozoans.

Phaeophyta T -^

Brown algae

No. species described ca 900

Proportion of group known Moderate/high
Size Macroscopic plant-like organisms,

mostly a few centimeters; sometimes

much larger
Nutrition Photosynthetic
Mode of life Most live anchored to the substrate on

rocky coasts

Most widespread in temperate regions, where they
usually dominate the mtertidal zone. Generally fixed
but some, e.g. Sargassum, form large floating mats
far out to sea. The largest proloctists: Pacific giant
kelp iMacrocysds pynfera] sometimes reach 65 m in
length. Brown algae are major primary producers in
inshore environments and also provide habitat for a
large number of macroscopic marine organisms.

Plasmodiomorpfia

? ?

•

Zoospores occur in soil and infect the host; a
Plasmodium with many cell nuclei but no dividing

No. species described 29

walls develops within the host cell. Most species do

Proportion of group known Low

not appear to harm their hosts but Plasmodiophora

Size Microscopic

brassicae causes clubroot disease of brassicas and

Nutrition Heterotrophic

Spongospora sublerranea powdery scab of potatoes.

Mode of life Obligate intracellular syml

lonts, mainly

of terrestrial plants; some

parasitic

APPENDIX 1 243

■#>

Rhizopoda ▼ ♦ •

Amastigote amoebas and cellular slime molds

No. species described ca 200

Proportion of group known Low

Size Single-celled or multicellular

Nutrition Heterotrophic

Mode of life Mainly benthic in aquatic habitats, or in

water film on land; some amoebas are

parasitic

Often abundant in soil, where cyst-forming types
highly resistant to desiccation. Entamoeba histolytica
IS responsible for some forms of amoebic dysentery
in humans. Some amoebas construct a coating ItestI
from detritus and these have a fossil record from
Paleozoic times, some fossil acntarchs Isee Chapter
31 may represent testate amoebas. Cellular slime
molds typically exist amid decaying vegetation, on
logs or bark, and feed by enveloping bacteria and
protoctists. The reproductive form of slime molds is
an aggregation of cells each formerly having
independent existence. Key experimental organisms
in studies of cell communication and differentiation.

Rtiodophyta T ♦ •

Red algae

No. species described ca i 000

Proportion of group known Moderate/high
Size tvlacroscopic plant-like organisms, up

to 1 meter in size
Nutrition Nearly all photosynthetic; a few are

symbionts on other red algae
Mode of life Virtually all are marine, a few species

are freshwater or terrestrial

Red algae occur attached to substrate on beaches
and rocky shores worldwide. Most abundant in
tropics. Many forms become encrusted with calcium
carbonate; calcified red algae have a fossil record
from the early Paleozoic. Agar jelly is extracted from
red algae, and other extracts are used in food
manufacture. Along with the Phaeophyta Ibrown
algael the largest and most complex protoctists.

Xanthophyta T ♦ •

Yellow-green algae

No. species described ca 600

Proportion of group known Low

Size Single-celled or colonial

Nutrition Photosynthetic

Mode of life Free-living, mostly freshwater algae

Free-swimming unicells, or highly structured
multicellular or multinucleated organisms, with
gold-yellow xanthin pigments. Often form scum in
pond water and margins. Typically form pectin-rich
cellulosic cell walls; cysts often rich in iron or silica.

Xenophyophora

Xenophyophores

No. species described i2

Proportion of group known Low

Size Sometimes several centimeters in

diameter
Nutrition Heterotrophic
Mode of life Benthic marine forms

Little-known bottom-living marine protoctists from
deep sea and abyssal regions. Make shells itestsl
from detritus leg. forammiferan shells, sponge
spicules). Xenophyophores are the most abundant
macroscopic organisms in some deep-sea
communities, with several individuals per square
meter Some acritarch fossils Isee Chapter 3| may
have been xenophyophores.

Zoomastigota ▼ ♦ •

Zoomastigotes

No. species described ^

Proportion of group known Low''

Size Single-celled, some colonial

Nutrition Heterotrophic

Mode of life Some are free-living in marine and
freshwater environments, others are
symbionts m the intestines of
vertebrates

Many feed by ingesting bacteria. Parasitic forms
occur in the intestine of aquatic vertebrates, e.g. the
opalinids. found in frogs and toads. One group of
colonial forms, the choanomastigotes, may be
ancestral to the sponges.

EUKARYA:
PROTOCTISTA

244 WORLD ATLAS OF BIODIVERSITY

CEREALS

APPENDIX 2:
IMPORTANT FOOD CROPS

FOOD CROPS OF MAJOR SIGNIFICANCE

These species and groups of species are those
that, according to national food supply data
maintained by the Food and Agriculture
Organization of the United Nations IFAOI', provide

5 percent or more of the per capita supply of
calories, protein or fat in at least ten countries.

Source- List ol crops from Prescott-AUen and Prescott-AUen ; other

J i 5

information from f^abberley , Smartt and Simmonds .Smith et at- ,
Vaughan and Geissler ; conservation status from Walter and Gillett .

Millets

Echinochloa frumentacea, Eieusine coracana, Panicum miliaceum, Pennisetum glaucum,
Setaria italica (Gramineael

Main uses: Ec frumentacea: Quicl<est growing of all
millets, available in six weeks. Used for human
consumption in India and East Asia and for animal
fodder in the United States. f( coracana: Important
staple in East and Central Afnca and India. Wild cereal is
harvested dunng times of famine. In Africa it is the
preferred cereal for brewing beer Seed heads may be
stored for ten years. Pa. miliaceum: Cereal cultivated for
human consumption, mainly in northern China, Russia,
Mongolia and Korea, and as animal feed elsewhere. In
Europe it is grown mainly as bird seed. Millets are
generally tolerant of poor soils, low rainfall and high
temperatures, and are quick matunng. Pe glaucum:
Most widely grown of the millets. The mam cereal of the
Sahel and northwest India. Heat and drought resistant.
May contribute to incidence of goitre. 5. italica: A once
important cereal that has declined in popularity, but still
grown on a relatively large scale in India and China,
mainly for home consu.mption Used also for animal
fodder and bird seed. Early maturing and stores well.

Origins: Echinochloa: Different strains are thought to
have partially different origins. Approximately 35 spp.
exist in the genus, distributed in warm areas. Eieusine:
Eastern and southern Afnca highlands. Nine spp. in the
genus in Africa and South America. Panicum: Unknown
in wild state. The closest relative P. miliaceum var.
ruderale is native to central China. At least 500 spp. in
the genus in tropical to warm temperate areas.
Pennisetum: Cultigen originated in West Africa from P.
violaceum. Total of 130 spp, in the genus, found in
tropical and warm areas. Setaria: Native to temperate
Eurasia. Approximately 1 50 spp. in the genus in tropical
and warm areas.

Related species: Echinochloa: E. pyramidalis (tropical
and southern Africa and Madagascar], used as fodder
and locally as flour; £ turnerana channel millet
lAustralial is a promising forage and gram crop.
Several other spp. are weeds, Panicum: P. hemiotum
[pifine grass. North America] and P. texanum (Colorado

grass. North Amencal used as fodder. P. maximum
(Guinea grass. Afnca, naturalized Amencal is cultivated
as a forage crop, P. sonorum (Mexico! a minor grain;
P. sumatrense (little millet, Malaysia! a minor gram.
Pennisetum: Used as fodder lawn grasses, some
grains. P. hohenacken (moya grass. East Africa to
Indial IS suggested lor papermaking; P. clandestinum
(Kikuyu grass, tropicat Afnca! pasture grass, erosion
control, lawns; P. purpureum (elephant or Napier
grass. Africa! fodder and paper P. violaceum (Africa!
harvested during famines. Setaria: S. glauca (yellow
foxtail! cattle fodder; S. paWdilusca is a cereal in
Burkina Faso; S patmifolia (India! shoots are eaten in
Java; S. pumita is cultivated as cereal; S. sphacelata
(South Atncal is an important silage crop.

Genetic base: Echinochloa: One sp. listed as
threatened in 1797. Eieusine: Five races of cultivated
finger millet recognized from Africa and India.
Excellent prospects for improvement. Significant
annual yield increases in India, mainly due to the
incorporation of Afncan germplasm. Panicum: 16 spp.
listed as threatened in 1997. Pennisetum: P. violaceum,
the wild progenitor, is an aggressive colonizer and may
be found in large populations around villages in West
Afnca. The cultivated crop is relatively undeveloped.
Open-poUinated cultivars are popular in Afnca and
India. Five spp. listed as threatened in 1997. Setaria:
Largely a crop of traditional agnculture systems. Two
spp. listed as threatened in 1997.

Breeding: El. coracana: Wild spp. in Africa cross with
domesticated finger millet to produce fertile hybnds
which can be obnoxious weeds. Pe. glaucum: Genetic
exchange with related wild forms in same geographical
area is possible, S, i'(a(/ca; Hybridizes easily with wild
relative S. italica var viridis to produce fertile offspring.

Germplasm collections: 90 500 general millet
accessions. 45-60 percent of landraces and 2-10
percent of wild spp. represented.

APPENDIX 2 245

Main uses; Early maturing gram with high yield
potential; can be grown where other crops fail, e.g.
above Arctic circle, at high altitude and in desert and
saline areas. Most important as animal feed, also for
brewing beer and human food. Mam producers in
Europe. North Africa. Near East. Russia, China, India,
Canada, United Slates.

Origins: Southwestern Asia. Approximately 20 spp. in
the genus, distributed m the north temperate region.

Related species: H. distichon |2-rowed barley] is
possibly H. vulgare x H. spontaneum

Genetic base: Landraces have been almost completely
replaced by pure line cultivars and the change in
genetic structure of barley populations has been
profound. Important contribution of Ethiopian barleys
highlights need to broaden genetic base. Two spp were
listed as threatened in 1997.

Breeding: Fertile hybrids between wild and cultivated
forms occur naturally where ranges overlap. Crosses
possible with other spp. in the genus but not utilized in
barley cultivars. Ethiopian barleys have been important
as genetic source of disease resistance and improved
nutritional value.

Main uses: Highest world production of all grains.
Primary source of calories and protein in humid and
subhumid tropics. The gram is relatively low in protein;
brown rice a source of some B vitamins Rice bran is
used m animal feeds and industrial processes. Can
grow in flood-prone areas Mam producers are China,
India, Indonesia, Bangladesh, Viet Nam. Only 4 percent
of world production is exported. Mam exporters
Thailand, Viet Nam, Pakistan, United States.

Origins: Two cultigens appear to have been
domesticated independently. The origin of 0. sativa
is uncertain, possibly derived in several centers from
0. rufipogon (selected weed in Colocasia fields).
Archeological evidence suggests origin m China or
Southeast Asia. Center of diversity of 0. glabernma is
the swampy area of the Upper Niger Approximately
18 spp. in the genus, distributed in the tropics.

Genetic base: Following agricultural intensification
many populations of wild relatives have disappeared or
intergraded with domesticated rice. Reduced genetic
base has also led to repeated pest epidemics. Great
genetic diversity exists in 0. sativa cultivars, much less
in 0. glabernma. Rapid spread of improved rice
varieties has displaced tens of thousands of landraces.
many now extinct. 0. glabernma is rapidly being
replaced by 0. sativa. Genetic erosion is reported in
China, Philippines. Malaysia, Thailand and Kenya.
Three spp. were listed as threatened in 1997.

Breeding: sativa has formed numerous hybrids with
wild spp. 0. nivara and 0. rufipogon. Genes improving
tolerance to diseases or adverse conditions have been
derived from African rices and wild relatives.

CEREALS

246 WORLD ATLAS OF BIODIVERSITY

CEREALS

Main uses: Cereal crop used as animal feed and for
human consumption. Eaten mainly as rye bread and
cnspbread Higher in minerals and fiber than wheat
bread- Previously more popular as a bread flour; now
largely replaced by wheat. Still important in cooler
parts of northern and central Europe and Russia,
cultivated up to the Arctic circle and up to k 000 m
altitude. Tolerates poor soils. Also used in brewing
industry and young plants produce animal fodder Ivlain
producers are Russia and Europe,

Origins; Probably originated from weedy Secale types
in eastern Turkey and Armenia. S, montanum is
possible ancestor Total of three spp in the genus in
Eurasia,

Genetic base; A number of weed ryes, found
associated with agriculture throughout the Near East.
are now considered to be subspecies of S, cereale. A
complex of subspecies of 5. montanum extends from
Ivlorocco east to Irag. Five spp, were listed as
threatened in 1997,

Sorghum

Sorghum b/co/or (Gramineae

Main uses: Staple cereal in semi-arid tropics. Mostly
grown in developing countries, especially for domestic
consumption by small farmers in Africa and India.
Used in brewing beer and as animal fodder Gram
stores well. Main producers are United States, India,
China, Nigeria, Sudan,

Origins; Developed primarily from the wild

S arundinaceum in Africa, Total of 24 spp, in the

genus in warm areas of the Old World and Mexico.

Related species; Backcrosses with 5. arundinaceum
gave S drummondii cultivated for forage; S, halepense
IMediterraneanI is a widely naturalized fodder plant,
often weedy.

Genetic base: Wide variation in landraces. Genetic
erosion reported in Sudan. Modern varieties have not
been widely popular for use as human food. Two spp.
listed as threatened in 1997.

Breeding: Wild relatives may be an important source
for disease resistance.

Germplasm collections: 168 500 accessions. 21
percent in the International Crops Research Institute
for the Semi-Arid Tropics. 80 percent of landraces, 1
percent wild spp. represented.

APPENDIX 2 247

Wheat

Triticum aestivum, T. turgidum iGramineael

Main uses: Most widely cultivated crop. Gram 15 gluten
ricti and tiigtily valued for bread making. 90 percent of
wheat grown 15 T. aestivum. Wheatgerm oil is highly
unsaturated and high in vitamin E. Durum wheat,
T. turgidum, has higher protein content. Used for
making pasta. High nutritive value; easy processing,
transport and storage. Ivlain producers are China, India,
United States, France and Russia.

Origins: tvlediterranean and Near East. Origin is
complex and not fully understood, probably involving
Aegitops spp. Total of four spp. in the genus,
distributed from the Ivlediterranean to Iran.

Genetic base: Large variation in the crop, around
25 000 different cultivars. However, large areas are
planted with genetically uniform crops and the inflow of
landrace material into breeding programs is low
18 percentl. Genetic erosion is reported in China,
Uruguay, Chile and Turkey

Germplasm collections: Approximately 850 000
accessions. Largest collection (13 percentl in Centre
for Maize and Wheat Improvement. 95 percent of
landraces and 60 percent of wild spp. collected.

Main uses: Mostly grown for human consumption in
parts of Africa and Latin America; elsewhere mainly
for animal fodder Starch may be extracted and used in
food processing. The germ oil is important. Also used
in the brewing industry. Mam producers are the United
States, China, Brazil, Mexico, France.

Origins: Probably derived from teosmte, Zea mays ssp.
mexicana. Total of four spp. in the genus, confined to
Central America,

Genetic base: Most of the world's maize crop is
derived from a few inbred lines. Landraces represent
40 percent of the crop grown in developing countries.
Genetic erosion is reported m Mexico, Costa Rica,

Chile, Malaysia, Philippines, Thailand. Z. perenn/s was
presumed extinct in the wild until its rediscovery m
1977. Z- diploperennis was recently discovered and is
now protected m the Sierra de Manantlan Biosphere
Reserve, Mexico. Three spp. were listed as threatened
in 1997.

Breeding; Teosinle crosses readily with maize to produce
fertile offspring, Tnpsacum crosses with less success.
Neither has been widely used in breeding programs.

Germplasm collections: 277 000 accessions, the
largest existing at Indian Agricultural Research
Institute. 95 percent of landraces and 1 5 percent
wild spp. are represented.

CEREALS

2A8 WORLD ATLAS OF BIODIVERSITY

TUBERS

Yams

Dioscorea spp. D. alata, D. bulbifera, D. cayenensis, D. dumetorum, D. esculenta,
D. rotundata, D. trifida (Dioscoreaceael

Main uses: Edible stem tuber, 28 percent starch and
limited vitamin C. Important staple in the humid and
subhumid tropics. Also major ingredient in oral
contraceptives. Religious and cultural role. Good
storage properties. West Africa produces 90 percent of
world production; Nigeria atone produces 70 percent.

Origins: Three mam independent centers of diversity or
domestication in Asia, Africa and America.
Approximately 850 spp. exist in the genus, distributed
in tropical and warm regions.

Genetic base: Predominantly a subsistence crop.
Apparently little genetic erosion. Large genetic
variability in wild edible forest yams. 68 spp. were listed
as threatened in 1997.

Breeding: New World and Old World spp. show strong
genetic isolating barriers and crosses between them
are not successful.

Cassava

Manihot esculenta (Dioscoreaceae!

Main uses: Edible tuber containing 35 percent starch
and vitamin C. Cultivated in almost all tropical and
subtropical countries, mainly by smallholders. One of
the most efficient crops for biomass production. A good
famine reserve, able to withstand harsh conditions.
Also animal feed. Mam producers are Brazil, Nigeria,
Dem. Rep. of Congo, Thailand, Indonesia.

Origins: Unknown in the wild state. Total of 98 spp. exist
in the genus, occurring between southwest United
States and Argentina. Most diversity occurs in northeast
Brazil and Paraguay and in west and south Mexico.

Related species: M. glaziovii is the source of Ceara or
Manicoba rubber and oilseeds.

Genetic base: Estimated 7 000 landraces. Local
preferences in flavor, root texture and growth habit vary

greatly, many farmers retain traditional cultivars
despite improvements in new cultivars. Genetic erosion
reportedly a risk in South and Central America,
Thailand and China. 65 spp. were listed as threatened
in 1997.

Breeding: Variability of cultivated forms has probably
increased through crosses with wild forms. M. glaziovii
and M. melanobasis have contributed to improvement
of cultivated form. High diversity in germplasm
provides good improvement potential. Interspecific
crossing with wild relatives may be employed further to
broaden tolerance of different conditions.

Gernnplasnn collections: 28 000 accessions, mostly in
international centers of research. 35 percent of
landraces and 5 percent wild spp. collected were listed
as threatened in 1997.

APPENDIX 2 249

Potato

Solanum tuberosum (Dioscoreaceael

Main uses: One of the most important world crops.
Cultivated in 150 countries, mainly for local
consumption. Little international trade. Tubers are
cooked or processed into a range of products. Starcfi,
alcohol, glucose and dextrin are also ma|or products.
Tubers also make animal feed. Potatoes are 80 percent
water, 18 percent carbohydrates, with range of
minerals, and a good source of vitamin C. I^ain
producers are Russia, China, Poland, Germany, India.

Origins: Maximum diversity in cultivated and wild spp.
on the high plateau of Bolivia and Peru A number of
ancestral spp involved. The gene pool consists of
S. tuberosum ssp. andigenum and tuberosum, S.
stenotomum, S. aj3nt)wn. S. goniocalyx, S. x cbauca, S.
X juzepczukii, 5. x curtilobum, 5 phure/a. Total of 1 700
spp in the genus, distributed worldwide.

Related species: S. mebngena llndial (eggplant). S.
centrale land Australia! and 5. muncatum Ipepinol
lAndesI have edible fruit; 5. quitoense Inaranjidol (AndesI

IS used for fruit juice; S. melanocerasum ('cultigenl
Icultivated in tropical West Africa! fruit; S hyporhodium
lupper Amazon!; 5. amencanum lyerba mora!.

Genetic base: Between 3 000 and 5 000 varieties of
potato are recognized by farmers in the Andes. Genetic
erosion is reported in centers of origin, including Chile
and Bolivia. In Peru, of the 90 wild potato spp. 35 are
now extinct in the wild. Wild spp. and ancient cultivars
largely replaced by modern varieties. Attempts to
broaden the narrow genetic base have been slow.
1 25 spp. were listed as threatened in 1 997.

Breeding: Much introgression from wild relatives has
been attempted, improving disease resistance and
other traits.

Germplasm collections: 31 000 accessions worldwide.

20 percent are held by the Centre Internacional de la
Papa, Lima. Peru. 95 percent of landraces and 40
percent of wild spp. are collected.

Main uses: Mainly temperate vegetable crop, but grown
worldwide. Large number of edible and ornamental
varieties, including cauliflower, calabrese and kohlrabi.
Important component of human nutrition throughout
the world: a good source of fiber, vitamins E, B and C,
and also Vitamin A in the greener parts. Main
production in Europe including Russia.

Origins: The wild cabbage is native to Europe;
development of cultivars took place in the
Mediterranean region. Total of 35 spp. exist in the
genus, distributed in Eurasia.

Related species: Wide range of crops [variously leaves,
buds, florets, stems and roots! eaten; also used for oil
production. S. campestns and 6. napus Irapeseedl.
B.cannata [Texsel greens! (northeast Africa!; 6. hirta

(white and yedow mustard! (Mediterranean!; 6. juncea
(Indian mustard! (Eurasia(; S.yuncea var cnspifolia
(Chinese mustard!.

Genetic base: Outbreeding nature. Large amounts of
genetic variation in most crops, where not highly
selected. Continuing emphasis on uniformity in recent
decade and controls on release of new cultivars have
led to significant reduction in genetic variation in
commercial cultivars Wild relatives in Mediterranean
are threatened; U spp. listed as threatened in 1997.

Germplasm collections: Efforts made to ensure
different crops are represented, including obsolete and
locally popular varieties. Cultivars from southern
Europe are less well collected.

TUBERS

LEAF VEGETABLES

250 WORLD ATLAS OF BIODIVERSITY

BEANS

Main uses: L purpureus. Young pods and young and
mature seeds of lablab are eaten; pulse contains 25
percent protein, little fat and 60 percent carbotiydrate.
Mam producers: India, Souttieast Asia, Egypt, Sudan.
P lunatus: Dried or immature seeds of lima bean
are used as pulses; seeds contain 20 percent protein,
1 .3 percent fat, 60 percent carbohydrate; flour also
obtained from seed. Mam producer is the United
States. P. vulgaris: Most widely cultivated of all beans;
in temperate areas grown mainly for the pod, which
contains 2 percent protem and 3 percent carbohydrate
with vitamins A, B, C and E; seeds have 22 percent
protein, 50 percent carbohydrate, 1.6 percent fat and
vitamins B and E. Mam producer: Brazil.
V. unguiculata: Cov^pea is a nutritionally important
minor crop in subsistence agriculture in Africa; dry and
green seeds, green pods and leaves are eaten; highly
resistant to drought.

Origins: Lablab: African or Asian origin; only one
species m the genus (previously Dolichos lablabl.
Phaseolus: It is thought that separate domestications
occurred in Central and South America from
conspecific races; total of 36 spp. in the genus, found
in tropical and warm America, Vigna: Center of
diversity of wild relatives in southern Africa; greatest
diversity of cultivated form exists m West Africa;
subspecies dekindtiana is probable progenitor; total of
1 50 spp in the genus, mainly in the Old World tropics.

Related species: Phaseolus: Five cultigens exist in the
genus: apart from P. lunatus and P. vulgaris, there are
P. aculitolius Itepary bean, America); P. coccineus
Iscarlet runner. Central America] and P. polyanthus
lyear bean. Central America!. Various other spp. are
important pulse crops, previously listed as Vigna spp.
Vigna: Other spp. are used for forage and green

manure, etc. Other pulses include: V. aconitilotia Imoth
bean. South Asia; V. angulans lAduki bean. Asia];
V. mungo lurad, tropical Asia); I', radiata (mung bean,
'Indonesia!; V. subterranea IBambara groundnut. West
Africa). V. umbellata Ince bean, southern Asia);
V. unguiculata )cowpea. Old World); V. vexillata (tropical
Old World) which has edible roots.

Genetic base: Lablab: Mainly grown in small plots and
home gardens; larger areas under cultivation in
Austraha. No threat of genetic erosion. Phaseolus:
Much dry bean cultivation in the United States depends
on very small germplasm base; improved varieties also
widely adopted by smallholder farmers. Relatively wide
genetic base provided by landrace groups, if conserved;
most wild relatives widespread but populations of
several taxa being lost to overgrazing in southwest
United Slates and northern Mexico. Two spp. listed as
threatened in 1997. Vigna: Breeding relies on narrow
genetic base and hybridization with other Vigna spp. is
important; more variability in wild relatives m the
primary gene pool than in cultivated cowpea. Four spp.
were listed as threatened in 1997.

Breeding: Phaseolus: Several wild relatives are fully or
partly compatible; populations of wild hma bean with
larger seeds recently discovered in northwest Peru and
Ecuador Vigna: cowpea crosses successfully with wild

subspecies of V. unguiculata.

Germplasm collections: Lablab: 1 1 500 accessions in
Africa and Caribbean. Phaseolus: 268 500 accessions
of Phaseolus spp. in total. 15 percent are held by
Centro Inlernacional de Agricultura Tropical. Call,
Colombia. On average 50 percent diversity m the genus
IS represented.

APPENDIX 2 251

Groundnut

Arachis hypogaea ILeguminosael

Main uses: The edible nut contains 50-55 percent oil, 30
percent protein, and is good source of essential minerals
and E and B vitamins. Cultivated for ttie nut or for oil in
many tropical and subtropical countries. Seed residue
useful as animal feed. Nutshells are used as fuel and in
industry. Stems and leaves used as forage. Mam
producers are India. China. United States, Argentina,
Brazil. Nigeria. Indonesia. Myanmar. Mexico. Australia.

Origins: Mato Grosso in Brazil is the primary center
of origin and diversity for the genus. The cultivated
groundnut is thought to have originated in southern
Bolivia and northv^est Argentina. Total of 22 spp. in the
genus, all from South America.

Coconut

Cocos nucifera IPalmae)

Main uses: The endosperm of the nut contains 65
percent saturated oil, used in manufacture of
margarine, soap, cosmetics and confectionery. Also
eaten fresh, desiccated or as a coconut milk. Residue
IS a high-protein animal feed. There are many more
uses: source of naturally sterile water, fiber, wood,
thatch. Mainly a smallholders' crop. Mam producers
are the Philippines, Indonesia, India. Sri Lanka,
Malaysia, Mexico. Pacific Islands.

Origins: Possibly originated in Melanesian area of
Pacific. Wild types predominate on the African and

Genetic base: Cultivated as a marginal crop with
relatively little selection pressure. Many varieties exist
worldwide with broad adaptability

Breeding: A. monlicola freely crosses with

A. hypogaea- Wild Arachis material confers resistance

on domestic form.

Germplasm collections: 13 000 accessions at the
International Crop Research Institute for the Semi-
Arid Tropics.

Indian coasts of the Indian Ocean, and scattered in
Southeast Asia and the Pacific. Single species in
the genus.

Genetic base: Tendency to plant uniform, improved
hybrids is reducing genetic variation particularly in
domesticated types.

Breeding: Wild and domestic coconuts are fully
compatible. Hybridization with wild types has increased
genetic diversity of cultivated crups.

Main uses: The mesocarp on the fruit yields oil for
human consumption. Unrefined oil is high in vitamin A.
Oil may also be extracted from the kernel. An export
crop and important for local consumption. Very high
yielding. Malaysia supplies 70 percent of world exports.

Origins: West Africa, originally a species of the
transition zone between savannah and rainforest.
Only 2 spp. exist in the genus.

Related species: £ oleifera (tropical Americal is less
important as an oil crop than £ guineensis.

Genetic base: Populations in Africa are semi-wild. They
are being thinned to make way for other crops.
Plantations in Malaysia were based on material from
only four specimens. New material is being introduced
to broaden the genetic base.

Breeding: Fertile offspring produced with £ oleilera.

OIL CROPS

252 WORLD ATLAS OF BIODIVERSITY

OIL CROPS

Soybean

Glycine max iLeguminosael

Main uses: The most important oil crop and gram
legume in terms of production and international trade.
An important basis of Asian cuisine, developed into
various forms of food from soy sauce to tofu. Immature
green beans and sprouts also eaten. Seeds contain 18-
23 percent oil and 39-45 percent protein. Oil is used in
various forms. Most of the meal is used as a high-
protein animal feed fvlain producers are the United
States, Brazil, China, Argentina, India.

Origins: A cultigen, not known m the wild. Soybean is
thought to have arisen as a domesticate in the eastern
half of northern China about 3 000 years ago probably
from S. soja. Total of 18 spp. exist in the genus,
distributed from Asia to Australia.

Genetic base: The genetic base of varieties is narrow
worldwide Conservation of traditional landraces is
urgently needed. Two spp. listed as threatened in 1997.

Breeding; Wild spp. are increasingly used for
improvement and there is good potential for further
valuable characteristics to be found in wild Glycine spp,
6. soj3 easily crosses with soybean.

Germplasm collections: 174 500 accessions, 9 percent
in Institute of Crop Germplasm Resources. Chinese
Academy of Agricultural Sciences, Beying, China.
60 percent of landraces and 30 percent wild spp.
are represented.

Cotton seed

Cossypium barbadense, G. hirsutum (Malvaceael

Main uses: Cotton is the second most valuable oil crop,
as well as being the most important textile fiber Crop
development is concentrated on fiber production
because value is three or four times greater New
World cottons took over from Old World forms after the
European exploration of the Americas. Ivlain producers
of G. barbadense are Russia, Egypt, Sudan, India,
United States, China.

Origins: Unique in that four spp. were domesticated
independently for the same use as a fiber and oil crop,
in Africa and India: G. arboreum and G. herbaceum; in
Central and South America: G. hirsutum and 6,
barbadense. Total of 39 spp, in the genus, found in
warm temperate to tropical zones.

Related species: 6. arboreum is still important in India
and Pakistan. G, herbaceum is grown only on a small
scale in Africa and Asia.

Genetic base: Modern cultivars of G. hirsutum are
responsible for over 90 percent of world production.
New Gossypium spp, possibly occur in Arabia and
Africa, Wild forms of G, herbaceum, G. hirsutum and 6.
barbadense are known. Past breeding involved much
introduction of genetic material from different
geographic regions, but a severe narrowing of the
genetic base has occurred in the production of modern
G, hirsutum varieties Large amounts of fertilizers and
pesticides required in modern cotton production. Eight
spp. listed as threatened in 1997.

Breeding: At least six related spp, have contributed
genes of importance to the cultivated crop Ivlaterial
from wild gene pool used in genetic engineering, G,
herbaceum and G, arboreum are able to interbreed,
although later generations have a high probability of
failing reproductively.

APPENDIX 2 253

Sunflower seed

Helianthus annuus (Compositae

Main uses; Seeds contain 27-iO percent
potyunsaturated oil and 13-20 percent protein. Oils and
margarines used for human consumption, and for
industrial uses, and waste products useful in animal
feed. Pollinating bees frequently used for fioney
production, (ulam producers are Russia, United States.
Argentina.

Origins: Probably originated in southwest North
America, Total of 50 spp. exist in the genus, distributed
in North America.

Related species: Also ornamental; H. tuberosus

{Jerusalem artichoke! is also eaten. H. petiolaris used
for hybridization.

Genetic base: Increased yields in hybrids led to
increased interest and production in 1960s. Large gene
pool exists in wild and weed sunflowers in North
America, although habitat loss is resulting in
population declines. 16 spp. listed as threatened in
1997.

Breeding: Resistance to several diseases was secured
through hybridization with H. tuberosus.

Main uses; Fruit with iO percent oil content. Highly
superior oil for cooking, margarines, dressing; also
used in cosmetics and pharmaceutical industry. Fruit
eaten pickled. Despite competition with more modern
oil-producing crops, olive oil still commands premium
price. Recent rise in popularity and recognition of
nutritional value. Mam producers are Spam, Italy.
Greece. Turkey, Tunisia.

Origins: Olive is a cultigen. evolved in eastern
Mediterranean, 0, europaea ssp. oleasfer recognized
as progenitor Total of 30 spp. in the genus, in tropical
and warm temperate parts of Old World.

Related species: Related species provide good timber

Genetic base: A long-lived tree. The turnover of clones
should be slow. Hundreds of distinct cultivars, found in
different geographic groups. Olive production still relies
on traditional cultivars. Few new varieties have been
released- Decline in area under cultivation. Marginal
groves have been abandoned with serious
consequences for Mediterranean wildlife. Wild
populations outside area of cultivation under pressure
from cutting and land clearance; two spp. were listed
as threatened in 1998.

Breeding; Closely related to wild subspecies in the
Mediterranean, Africa. Arabia, Iran and Afghanistan.

OIL CROPS

254 WORLD ATLAS OF BIODIVERSITY

SUGAR CROPS

FRUITS

Sugarcane

Saccarhum officinarum (Gramineae

Main uses: Ma|or source of calories worldwide.
Cultivated in about 70 countries, mainly in tropics.
Requires good rainfall and rich soil for successful
growth. Stems are easily transported. Mam producers
are Brazil, India, China, Thailand, Pakistan.

Origins; A cultigen with origin and center of diversity in
New Guinea. Between 35 and iO spp. in the genus,
distributed in tropical and warm zones.

Related species: Other cultivated sugar canes include
S barben, 5 edule and 5 sinense. S.robustum and S.
spontaneum are wild sugar canes.

Genetic base: Risk of genetic erosion reported in
Assam and suspected in Indonesia, Papua New Guinea
and Thailand, where monocrop plantations have taken

Banana, plantain

Musa spp. IMusaceael

Main uses: One of the most popular dessert fruits in
industrial nations, a major source of calories and export
earnings in developing countries. Bananas and plantains
are high in carbohydrates and potassium; bananas are a
good source of vitamins C and 06, and plantains contain
high levels of vitamin A. Numerous other uses. Mam
producers are India, Brazil, Ecuador. Philippines and
China for the banana; Uganda, Colombia. Rwanda,
Dem. Rep. of Congo and Nigeria for the plantain.

Origins: Bananas evolved in Southeast Asia from M.
acuminata or combinations of M. acuminata and M.
balbisiana. Plantains probably originated in southern
India. Primary areas of diversity exist in Southeast Asia.
Secondary areas also occur in tropical Africa, Indian
Ocean islands and the Pacific. Fe'i bananas I2nl,
thought to be derived from M. mactayi and possibly
other related spp. Greatest diversity of fe'i bananas is
on Tahiti. Total of 35 spp. in the genus, distributed
throughout the tropics.

Related species: Fe'i bananas are a significant source
of food in New Guinea and the Pacific. M. textilis recent
domesticate in Philippines used for Manila hemp.
Related Ensete ventncosum cultivated in Ethiopia for
starchy pseudostem. M. balbisiana produces edible
fruit and contributed to present-day cultivars.

Genetic base: About 500 genetically distinct cultivars.

over from indigenous spp. Modern hybrids have a
narrow genetic base Plantations are prone to severe
pest and disease epidemics. Attempts to incorporate
more genetic diversity is slowly having effect. Only 10
percent wild germplasm used in breeding. S. robustum
exhibits the most genetic diversity, but has had little
application in breeding

Breeding: Commercial varieties are derived from
interspecific crosses with other wild and cultivated
sugarcane spp.

Germplasm collections: 19 000 accessions in total,
nearly a quarter of them in Centra Nacional de
Pesquisa de Recursos Geneticos e Biofecnologia,
Brasilia, Brazil. 70 percent of landraces, 5 percent of
wild spp. represented.

90 percent of global banana production is from
smallholdings International trade in bananas relies on
very few cultivars, based on the Cavendish type.
Dangerously narrow genetic base and very susceptible
to diseases. Increased disease resistance is extremely
important given the economic importance Of the export
crop. The number of Fe'i banana cultivars has declined
severely as a result of human demographic changes in
the Pacific and the spread of pests. Banana is an
aggressive weed. Wild populations of Musa benefit
from forest clearance if succession is allowed to take
place. Three spp. listed as threatened in 1997.

Breeding: An extensive contact zone between cultivated
and weedy types exists in several areas, e.g. Sri Lanka.
Much mtrogression is believed to have enriched the
gene pool of cultivated types. M. balbisiana has
valuable traits. Several other wild relatives have useful
characteristics. Germplasm collections have been
poorly used; better selections could be made to suit
subsistence farmers.

Germplasm collections: Edible bananas, being
seedless, are not storable. Seeds from wild spp. may
be stored. Field gene banks hold collections. 10 500
accessions in lotal. The International Network for the
Improvement of Bananas and Plantains holds 10
percent. Most of the diversity of wild and cultivated
bananas is thought to be covered.

APPENDIX 2 255

Main uses: Seeds are fermented and roasted to
produce cocoa powder and chocolate Waste goes lo
produce animal feed, mutcfi or fertilizer Cocoa is a
nutritional beverage: tfie powder is 25 percent
saturated fat, 16 percent protein and 12 percent
carbohydrates, fvlain producers are West African
countries, Brazil, fvlalaysia.

Origins; Upper Amazon basin. Center of cultivation in
Central America. Total of 20 spp. in the genus, confined
to tropical America.

Related species: All the following are cultivated:
T. grandiflorum Icupuacu, Amazonia):
r. speciosum Icacaui, Central and South America!:
T. subincanum (South Americal: T. obovatum (AmazonI:
r. angus(/fo//um (Central Americal: T. b/co/or (Central
and South Americal: T. glaucum (Amazonial,

Genetic base: Undoubted genetic erosion has occurred
In recent years. Currently cacao plantations are

established by seed with varying degrees of genetic
heterogeneity Production in West Africa is based on a
particularly narrow gene pool. Originally three mam
cultivated types. Criollo yields the most superior
chocolate but has been largely replaced because of
low yields. Forastero dominates world production.
Wild cacao is highly variable, especially in its core
area. Dramatic increase in plantations of coca and
pulp-producing spp in various parts of the Amazon,
agricultural expansion and movement of human
populations have caused severe losses to the wild
gene pool.

Breeding: Little use of or research into wild genetic
reserves because they are relatively hard to cross.

Germplasm collections: Seeds do not remain viable for
long, i 000 to 5 000 accessions kept in field gene
banks. International Cocoa Genebank in Trinidad has
the most comprehensive collection. Close relatives are
poorly represented. Vegetative germplasm is collected.

BEVERAGE CROPS

256 WORLD ATLAS OF BIODIVERSITY

CEREALS AND
PSEUDO-CEREALS

FOOD CROPS OF SECONDARY OR LOCAL IMPORTANCE

These species and groups of species are those that,
according to national food supply data maintained
by the FAO' ■', provide a significant amount of the
per capita supply of calories, protein or fat, but on
the criteria followed here are not of equal
importance to the crops in the previous table

(i.e. they provide below/ 5 percent of the total
per capita supply and/or do so in fewer than
ten countries].

Source: List of crops from Prescotl-AUen and Prescott-Allen ;
ottier information from Mabberley . Smartt and Simmonds .
Smitti et 3l , Vauqtian and Geissler ; conservation status from
Vifalterand Gillett ,

Oats

■

Avena sativa IGraminael

Origin; West and north Europe from weed oat

A. byzantina also cultivated. Genetic erosion from

components of wheat and barley crops

intensive breeding has resulted in efforts to

conserve landraces and early varieties. Crosses

One of the fnajor temperate cereals, althoiigh currently

between A. sativa and A. byzantina have led to

declining in production and generally regarded as a

numerous cultivars. Fertile hybrids obtained from

secondary crop, f^ostly used for animal feed. Oat

crosses between cultivated oats and weed species.

kernel is higher in high-quality protein and fat than

Some success in incorporating desirable genes

any other cereal. Oat bran is a good dietary fiber

from more distant relatives.

Quinoa

Chenopodium quinoa IChenopodiaceae)

Origin: High Andes

An important and sacred pseudo-cereal in Inca times.
Remains a staple in large parts of South America.
Nutrient composition is superior to other cereals, being
high in lysine and other essential amino acids, calcium,
phosphorus, iron and vitamin E. Can grow in marginal
conditions. Greatest diversity of genotypes in the

highlands of southern Peru and Bolivia. Cultivation
declined with Spanish conquest until 1970s when
grown as a sole crop only in parts of Peru and Bolivian-
Peruvian Altiplano. Agricultural and nutritional benefits
have now been recognized and acreage has increased
significantly Improvement so far has been based on
inbred populations and pure lines. Considerable
potential for improvement, both in the crop and its use.

Fonio

Digitaria exitis (Gramineae

Origin: West Africa; thought to be a cultigen

Popular cereal in parts of West Africa. Adapted to
marginal agricultural land. Several species are
harvested as cereals during times of famine.

APPENDIX 2 257

' Taro ^^^^^^

^^■^^^^■^H|

Colocasia escutenta lAraceae) "^

^m^^^^i

Origin: India

products. Young leaves eaten as spinach. Tolerates high

temperatures and poor soils. More than 1 000 cultivars

Edible tuber; 25 percent starch, low protein, good

have arisen through subsistence farming. Lack of

vitamin C source. Probably cultivated before rice. Widely

interest and germplasm exchange at a more

cultivated in China and staple in many Pacific islands.

commercial level. Serious danger of genetic erosion.

Also used by food and beverage industries, and in pasta

Carrot M

^^^^^^^^^^^H

Caucus carota (Umbelliferael '%

^^^mmiiii

Origin: Afghanistan

replaced by hybrids in United States, Japan and

Europe. Environmental health concerns over level of

Root crop, grown worldwide, and eaten raw, cooked or

pesticide has led to interest in genetic source of pest

processed. The best plant source of provitamin A; low

resistance. D. capiUifolius has passed some pest

in other nutrients. Numerous wild and cultivated

resistance to cultivated crop.

subspecies. Open pollinated crops almost entirely

Sweet potato

Ipomoea batatas (Convolvulaceae

Origin: Not known in the wild. Greatest species
diversity occurs between Yucatan and the mouth of the
Orinoco. Major variation is found in Guatemala.
Colombia, Ecuador and Peru.

The tuberous root is an important staple in the tropics.
Able to grow in high temperatures with low water and
fertilizer input. Good source of fiber, energy and
vitamins A and C. Also industrial source of starch and
ethanol. Although acreage has declined, increases are
likely as a crop able to respond to population growth in

Tannia

Xanthosoma sagittifoUum (Araceae

Origin: New World

marginal areas. China accounts for 80 percent of
production. Until recently, material used in breeding
programs represented a fraction of existing diversity.
Genetic base has now broadened but requires further
increase. Little work has been done on cultivar
improvement in areas of highest production li.e. where
sweet potato is a staple). Countries where breeding
programs exist have replaced native cultivars with
improved varieties. Sweet potato is thought to have
more potential for yield improvement than any other
major crop in Asia.

Similar use and nutritional composition as taro. but
starch is more difficult to digest. Used in preparation of
fufu in West Africa.

ROOTS AND
TUBERS

258 WORLD ATLAS OF BIODIVERSITY

BEANS AND OTHER
LEGUMES

Pigeonpea "^^^H

IH^HI^^Hl

Cajanus cajan (Leguminosae) ^^^|

^mufii^mii

Origin: Cultigen, India

species wiffi good potential in agroforestry systems and

on marginal lands, India contributes more than 90

One of the major pulse crops of tlie tropics, feature

percent of world production. Domestication has not

seeds contain 20 percent protein, 60 percent

altered the species as much as other crops.

carbofiydrate and little fat. Important in small-scale

C cajanifolius is closest relative;! 2 spp may be

farming in mainly semi-arid regions, A multipurpose

crossed with pigeon pea.

Chickpea

Cicer arietinum ILeguminosae

Origin: Southeast Turkey; C. reticulatum is probably
the progenitor

One of the most important pulse crops. The seeds
contain less protein 1 1 7 percent or morel but more fat
15 percent! than other pulses. Grown over large area
from Southeast Asia to Ivlediterranean, Only 2-i

percent of world production is exported. Recently
discovered C, reticulatum is confined to ten
populations in Turkey Two main cuUivars have
emerged. Traditional landraces have been selected to
suit local ecological conditions. Commercial breeding
is a recent phenomenon, C, reticulatum and C.
echinospermum are compatible with the chickpea.

Origin: Near East; wild progenitor L orientalis

Seeds contain 25 percent protein. 56 percent
carbohydrate and 1 percent fat. Young pods also eaten.
Seeds are commercial source of starch for textiles and

printing. Residues used as animal feed. Unique
assemblages of landraces in different geographic
regions. The crop has been altered little by modern
breeding, ti4uch variation in the crop unexploited.

Origin: Wild progenitor is unknown. Possible centers of
origin are Ethiopia, the Ivlediterranean and Central Asia.

The second most important pulse, 90 percent
production as dried peas. Seed coats are source of
protein, used in bread or health foods. Russia and

China produce 80 percent of the world production of
dried peas. United States and United Kingdom are
largest producers of green peas. Breeding relies on a
fairly narrow genetic resource base and efforts to
conserve genetic variability of the cultivated crop have
been fairly limited.

APPENDIX 2 259

Lupin

Lupinus mutabilis ILeguminosael

Origin: Andes; other lupins originated in two mam
centers of genetic diversity in Mediterranean and in
Americas

A relatively minor pulse crop, obtained from several
Lupinus spp. Seed contains U percent protein,
17 percent oil Seed is human food in subsistence
agriculture. Also used as coffee substitute and

high-protein animal feed. Seed flour used as soya.
Species also important in soil improvement Related
spp. have ornamental value, used as fodder, coffee
substitute, green manure or to stabilize sand dunes.
May act as substitute for soybean, where climate is
unsuitable for soybean growth. Other Lupinus spp.
potentially suitable for cultivation.

Broad bean

Vicia faba iLeguminosae'

Origin: Cultigen; wild ancestor unl^nown, possibly fronn
central Asia

A temperate pulse crop. Both immature seeds and dry
mature seeds are eaten. The latter are also used as

Mustard seed

Brassicajuncea (Cruciferae)

Origin: Central Asia: Himalaya; probably 6. n/gra x
6. campestns and other Brsssica spp.

The most important spice in the world in terms of
quantity. Four species contributing to mustard exist.

animal feed. Dried seed contains 25 percent protein,
1.5 percent fat, 47 percent carbohydrate; the immature
bean has much less of these nutrients but more
vitamin A and vitamin C Also used as green manure.

6. juncea took over from 6. nigra in 19505 as it allowed
completes mechanization of harvesting. Also valuable
as oil crop and salad crop, vegetable and fodder Long-
lived seed allows easy maintenance of large
collections. Wild material is widely distributed.

Origin: Probably a hybrid of B. oteracea x 6.
campestris

The seed is an important, relatively recent source of oil,
containing 40 percent unsaturated fat. with industrial
and culinary applications. The root crop provides animal
and human food Iswedel. Uncertain whether 6. napus

exists in wild form. Domestication was a relatively
recent event. The crop is tolerant of inbreeding, and
landraces have been replaced by improved cultivars
since the 19th century. Swedes, of which there are
only a few varieties, are the result of hybridization with
6- campestns- Various valuable contributions to oilseed
rape also from 6. campestns and B. oteracea.

BEANS AND OTHER
LEGUMES

OIL CROPS

260 WORLD ATLAS OF BIODIVERSITY

OIL CROPS

LEAF
AND FLOWER
VEGETABLES

Safflower seei

Carthamus tinctorius ICompositae

Origin: Turkestan, Turkey, Iran, Iraq, to Israel and
Jordan

Oilseed crop, produces two types of oil for margarine
and also cooking oils. Ingredient of animal feeds. Dried
flowers serve as substitute for saffron. Applications in

cosmetic industry and as medicine. Originally
domesticated for use as dye plant. Mucfi diversity
developed as the species was cultivated over a wide
area. Large-scale cultivation in few countries. No
reported genetic erosion. Related species cross easily
witli cultivated crop and form natural hybrids.

Sesame ^^^^^^^^^^^^^^^^^^^^^^^^^^^H

Origin: Origin and ancestors unknown, possibly
Ethiopia or peninsular India

An ancient oilseed crop. Seeds contain 50 percent
unsaturated oil and 20-25 percent protein, and are
used widely in bread and confectionery. Oil used in

cooking, margarine, soaps and other industries.
Residues are valuable animal feed. Interest in the crop
is in decline as difficulty mechanizing harvesting and
low seed yields compared with other oil crops. Good
genetic diversity in related species. 5. malabaricum
produces fertile offspring with S. indicum.

Shea nut

Viteilaria paradoxa ISapotaceael

Origin: Dem. Rep. of Congo, Sudan, Uganda

The roasted kernels are used to make purified shea
butter, rich in vitamin E, used in cooking and as an
alternative to cocoa butter for chocolate manufacture.
Also has commercial use in manufacture of soap,
cosmetics, candles. Various local uses. Fruit is eaten

Artichoke

Cynara scolymus ICompositae]

Origin: Mediterranean. Canary Islands

and is a good source of carbohydrates, iron and
B vitamins. A monospecific genus. Threatened by
overexploitation as a timber and source of fuel, and
also by land clearance. Stands may be conserved for
their valuable seed, but no official protection exists.
Mostly grown for local consumption.

Flowerheads and the receptacle are eaten. Small
amount of vitamin C.

APPENDIX 2 261

Origin: Probably evolved in Asia Ivlinor or Middle East
from L. sernota

Lettuce leaves are a useful source of fiber, minerals
lespecially potassium), vitamins A, E and C, Ivlay be

grown year round. Stem is boiled as a vegetable in China.
A flighty variable crop, resulting probably from long
history of selection. Increasing diversity of lettuce types
consumed. Wild species, including L sernola, L saligna
and L. wrosa have been used in breeding programs.

Origin: Southv^est Asia

Edible leaves contain range of minerals, vitamins A, E
and C and the B vitamin range.

Pineapple

Ananas comosus (Bromeliaceae

Origin: Obscure origin in South America, probably on
fringes of Amazon

Seedless fiber-rich fruit, source of vitamins C. A and B.
Highly suited for canning and as a juice. Unigue in that
timing of harvest can be controlled by externally applied
groviith hormone. Over 65 countries grow pineapple for
domestic consumption and export. No wild populations.

Genetically variable species, but genetic base of
commercial plantations very narrow. 70 percent of world
production and 96 percent of cannery industry comes
from one variety. Highest diversity of near relatives in
Paraguay and Brazil Poorly known, but A. ananassoides
has contributed several characteristics to cultivated
crop. A. erectilolius also considered for improvement
programs.

Origin; Obscure, probably hybrid of several Carica spp.,
arising in lowland tropical forest in eastern Andes or
Central America

Easily digested fruits produced all year round. Good
source of vitamin C; red-fleshed fruits also rich in
vitamin A. Papain extract is exported as a meal
tenderizer; also used medicinally, to tan leather and in
brewing beer May be produced by biotechnology in
future Commercially produced in over 30 countries.

mostly for domestic consumption. High diversity in
eastern Andes. At least six other spp. domesticated and
1 2 spp. are harvested for their fruit. Several commercial
cultivars come from highly inbred hermaphrodite lines.
Most production from backyard papaya trees, where
local variation is high. Many wild species have desirable
characteristics, useful in breeding. Hybridization already
carried out with five wild Canca spp. Highly susceptible
to viral and fungal diseases; some resistance detected in
wild relatives but conventional crossing impossible.

LEAF

AND FLOWER

VEGETABLES

FRUITS

262 WORLD ATLAS OF BIODIVERSITY

FRUITS

Lime, lemon, pomelo, tangerine, sweet orange

Citrus aurantiifolia, C. limon, C. maxima, C. x paradisi, C. reticulata, C. sinensis (Rutaceae)

Origin: Lime: cultivated tiybnd with obscure origins,
possibly a hybrid of C. medica with another sp- Lemon:
probably a hybrid of lime with C. medica. Pomelo:
probably a native of the tvlalay peninsula. Grapefruit:
probably hybrid between orange and pomelo, arising in
the Caribbean, Tangerine: possibly Indo-Chma. Orange:
probably introgressed hybrids of C. maxima and C.
reticulata, perhaps originating in China.

Fruits contain nearly 90 percent water, potassium,
vitamins A, B, E and high vitamin C. They are eaten

fresh, used as a flavoring and in marmalade. Orange
accounts for 70 percent of C/(rus production. Various
other spp. are cultivated. Wild populations located
in northern India. Wild species threatened by forest
clearance. In Southeast Asia wild groves are
being replaced by oil palm and cacao plantations.
C. taiwanica is critically endangered in Taiwan, mainly
because of extensive habitat loss but also because of
use as a rootstock for citrus plantations. Wide variation
within the genus. Can be crossed with several genera.
Economic Citrus spp. are highly interferlile.

Pumpkin, squash, gourd

Cucurbita moschata, C. maxima, C. argyrosperma, C. pepo, C. ficifolia ICucurbitae)

Origin: C. moschata is most like the wild species and
was domesticated independently in Central and South
America.

Fruits, containing 90 percent water, small amounts of
starch, sugars, protein, fat and vitamins A, B and C, are
used as vegetables and as animal feed. Leaves and
flowers may be cooked. Seeds eaten and sometimes
processed for oil. Grown worldwide in temperate and
tropical zones, commonly in home gardens and as

subsistence crops as well as commercially Long
shelf life. Broad gene pool because of wide use of
traditional or unimproved varieties in subsistence
farming and home gardens. Many Cucurbita spp. have
restricted geographical ranges. Disease resistance is
found in wild relatives, with some transfer to cultivated
species through interspecific crosses. Crosses between
crop species and wild or feral relatives have occurred
and genetic exchange takes place where their ranges
overlap.

Strawberry ^^^^^^^^^^^^^^^^^H

Origin: A hybrid between two American species, F.
chiloensis and F. virginiana

Soft fruit, 90 percent water, high vitamin C, eaten
fresh or in jams and confectionery. Grown in most
temperate and subtropical countries. All Fragaria spp.
produce palatable fruit. Hundreds of cultivars with

wide ecological adaptability. Considerable genetic
diversity lost in cultivated strawberries in the last
1 00 years. Attempts are being made to extend the
genetic base of the crop. Much unused genetic
variation in wild species.

APPENDIX 2 263

Origin: Eastern Mediterranean

Fruits contain 1 percent sugar when fresh and 50
percent when dried. Also substantial amounts of
potassium, especially in the dried fruit. Ivlost world
trade as dried figs. Figs are widely distributed in
tropical, subtropical and warm temperate areas
throughout the world. Ficus spp. are also source of
rubber, fibers, paper, medicines and ornamental plants.
Fig is largely grown for domestic consumption using

traditional cultivars, hundreds of which exist, with local
clones occurring in distinct geographical groups in the
Mediterranean basin. Closely related wild forms are
distributed throughout the Mediterranean basin. Fig
culture is in decline. Many old groves have been
abandoned or cleared. A number of wild relatives are
considered threatened. 27 Ficus spp. were listed as
threatened in 1998. Reproductive isolation is dependent
solely on the specificity of the wasp pollinator Artificial
crosses can be made between species.

Origin: An aggregate of over 1 I
and complex hybrid origin

) cultivars. of ancient

Apples, with pears, are the most important fruit crops of
cooler temperate regions. Fruit is eaten fresh or cool<ed,
as a |uice or brewed as cider Potassium is the mam
mineral with small amounts of vitamin C. Breeders in
the 19th century used wild species in breeding. Genetic

diversity accumulated in North America was greater

than in Europe because propagation was by seed
rather than by grafting. Current trend depending on
few varieties has caused rapid loss in genetic diversity
and potential breeding material. Widespread elimination
of wild stands is also talking place. Three spp. were
listed as threatened in 1998, Hybridization with many
wild species within the genus occurs readily.

Origin: Northeast India

Fleshy edible fruit; a good source of vitamins A and C.
Thrives on infertile marginal soils. Important tree in
Hindu mythology and religion. Kernel oil may be used
in chocolate manufacture. Demand for the fruit and its
juice IS increasing worldwide. India accounts for two
thirds of production. Fruits of more than 12 wild spp.
collected. Several are cultivated. The majority of fruit-
bearing trees are more or less wild. Genetically highly
heterogeneous. Over 1 000 cultivars exist, many in

Borneo and the Malay peninsula. Feral populations
are distributed throughout the tropics. Of the 40 to
60 spp. in the genus, many are poorly l<nown, severely
threatened or possibly extinct. 35 spp. were listed
as threatened in 1998. Logging, forest clearance and
replacement with commercial species in Southeast
Asia are largely responsible for population extinctions.
Various species are suitable for cultivation given
further selection. Many display valuable traits,
such as tolerance of waterlogged soils and more
regular fruiting.

FRUITS

264 WORLD ATLAS OF BIODIVERSITY

FRUITS

Origin: Mexico to northwest Colombia

A highty nutritious fruit, containing 15-25 percent
monounsaturated fat and vitamins C. B and E- Ttie
oil IS used in cosmetics. Trees fruit year round.
Importance tias increased over recent decades and tfie
crop IS now grown in most tropical and subtropical
countries. Ivlost production is for domestic
consumption. Other Persea spp. are used for timber
and fruit. There are three geographically distinct
varieties, which are able to interbreed. Commercially
important cultivars have arisen mostly in private

orchards by chance rather than as a result of
germptasm manipulation. Increasing use of grafting
and uniform varieties. Serious genetic erosion in
traditional varieties. Diversity appears greater in
traditional growing areas, where farmers still
propagate by seed. Genetic exchange occurs between
cultivated forms and wild populations. Wild populations
of the avocado and its close relatives are small and
becoming increasingly isolated. Deforestation poses a
severe threat to their survival. 15 spp. were listed as
threatened in 1 998. A number of wild relatives show
resistance or tolerance to disease, drought and frost.

Origin: Western India or Arabian Gulf

Edible fruit with sugar content of 30 to 80 percent,
corresponding to soft and dry dates. Vitamins are
relatively low in quantity. Eaten as an ingredient in a
variety of foods or as a juice. A staple where produced.

Good storage. One of the oldest cultivated tree crops.
Current cultivars resulted from thousands of years of
selection. Perhaps over 3 000 cultivars exist; only 60
grown widely All commercial cultivars are female. Wild
populations of some related spp. are highly restricted in
geographical range. All Phoenix spp. intercross freely.

Apricot, cherry, plum, almond _^^^^^^^

Prunus armenica, P. avium, P. domestica, P. du/c/slRosaceae

Origin: Apricot: west China. Cherry: west Asia. Plum:
an ancient 6n cultigen with complex origin, possibly in
southwest Asia; North American plums may be native
American spp. or hybrids with P. saticina. Almond:
central to west Asia. Peach: west China, possibly a
cultigen derived from P. davidiana.

Apricot, cherry, plum and peach are soft fruit with up to
10 percent sugar, good potassium and Vitamin A in the
case of apricots, but low vitamin C. Consumed fresh,
dried or as an ingredient in jams and confectionery.
Almond is the most important tree nut crop. The kernel
contains '10-60 percent unsaturated oil and 20 percent

protein. Eaten as a dessert nut and in confectionery and
marzipan. A major trading commodity. Many other
Prunus spp. have edible fruit. Many cultivars and much
genetic diversity. Plums are genelically central to the
genus and harbor the most useful genetic material
Narrow breeding has led cherries to be more isolated
from the rest of the genus. Increasing loss of genetic
diversity Developing countries are tending to replace
indigenous types and wild stands with western varieties,
e.g. the switch from seed to vegetatively propagated
almonds in Turkey A number of wild relatives are
confined to narrow ranges. 23 spp. were listed as
threatened in 1998.

APPENDIX 2 265

Origin: Asia minor, the Caucasus, central Asia and
China. Cultivars have come from P bretschneideni, P.
pynlolia, P. sinkiangensis and P. ussunensis- P. nivalis
for perry production.

With apples, pears are the most important fruit crops
of cooler temperate regions. The fruit is eaten fresh or
cooked, as a juice or brewed as perry. The fruit are a
good source of dietary fiber, potassium and reasonable
amounts of vitamin C. Currently about 20 spp. and

5 000 recorded cultivars. li/lajor loss of genetic diversity
through concentration on few varieties. Several wild
species in Turkey are under threat. Five spp. were
listed as threatened in 1998. Hybridization with high
proportion of wild species within the genus is possible,
providing useful rootstocks and possibly disease
resistance. Much use of wild species in breeding in the
past. Evolution of new varieties will be seriously limited
unless stands of wild species conserved.

Blackcurrants, redcurrants

Ribes nigrum, R. rubrum (Grossularlaceae

Origin: Europe and northern Asia, viiith the
blackcurrant extending to the Himalayas

Fruits with high vitamin C content. A luxury crop,
largely produced for processing into juice. Many spp.
with edible fruits, cultivated and wild. Wide use has
been made of wild or near-wild relatives.

Grape

^^■■j^^^^^H

Vitis vinifera (Vitaceael

^BHill

Origin: Eurasia

commercial products include grapeseed oil and vine

leaves. Various other spp. produce edible grape. One

The fruit has high sugar content. 68 percent of grape

estimate suggests there are 10 000 cultivars of grape.

production is for the manufacture of wine. 20 percent

Wild species still occurs in Middle Asia. Wild relatives

for dessert grapes, 1 1 percent for dried fruit - raisins.

are suffering genetic erosion m the United States. All

sultanas, currants - 1 percent for juice. Other

known Wdsspp. produce fertile offspring.

Melon seed/watermelon

CitruUus lanatus ICucurbitae)

Origin: Tropical and sub-tropical Africa, domestication
took place in Mediterranean

The flesh of the fruit is 90 percent water; also contains
vitamin C and A. The seeds contain 40 percent

unsaturated oil and 40 percent protein. Wild plants still
harvested in Kalahari. C. cotocynthis is fertile with the
watermelon. An African watermelon with extraordinarily
long storage life has been identified.

FRUITS

FRUIT
VEGETABLES

266 WORLD ATLAS OF BIODIVERSITY

FRUIT
VEGETABLES

BEVERAGE CROPS

Melon, cucumber

Cucumis melo, C. sativus (Cucurbitael

Origin: Wild melon populations appear to be distributed
south of the Sahara to Transvaal in South Africa.
Cucumber's wild or feral relative and possible
progenitor, var hardwickii, is native to the southern
Himalayan foothills.

Melon IS grown worldwide in temperate and tropical
countries, 90 percent water some sugar and vitamin C,
Pink or orange-colored fruit have a high percentage of
vitamin A, Also grown for their fragrance or ornamental
value. Cucumber produces edible fruits, containing 96

Tomato

Lycopersicon escuientum ISolanaceael

Origin: Cultigen. from Mexico

There are few growing areas, from the tropics to the
Arctic circle, where the tomato is absent. Fruit,
containing potassium, vitamins A, B, C and E, is eaten
fresh, dried or cooked as a vegetable, or processed in a

percent water, some vitamin C and reasonable amounts
of Vitamin A. Also used in production of fragrances,
cosmetics and medicines. Young leaves and shoots may
be cooked. Also cultivated C. angaria West Indian
gherkinl and C. metuliferus (African horned cucumber
or jelly melon). Wild and feral populations of melon
occur throughout Africa and southern Asia.
Cucumber produces fertile hybrids with its wild
counterpart C. saf/vusvar hardwickii. No interspecific
hybridizations have been used to improve crops.

wide range of food products. Disease is a common
threat. The wild relatives of the tomato have limited
ranges. Wild gene pools are prone to erosion by habitat
destruction. Tomato can be hybridized with all spp. in
the genus and wild relatives have been used as source
of numerous useful traits, including disease resistance.

Eggplant

Solarium melongena

Origin: India; wild progenitor,

S, incanum, occurs throughout Africa

and Asia.

Fruit IS eaten as a vegetable, contains over 90 percent
water, large amount of potassium, some vitamins A, E.
B. C. Highly productive and useful smallholders crop.
Various spp. cultivated and used as grafting stock.

Origin: Probably lower Tibetan mountains or central
Asia

Tender shoots are used to make tea. Important
plantation and smallholder crop throughout the tropics
and subtropics Planted commercially in at least 30
countries. Increasing consumption in developing

countries. High diversity of forms or species in East and
Southeast Asia. Ivlany distinct forms, hybrids and species
continue to be discovered. Recent trend to propagate
the plant vegetatively, which has led to large areas being
planted with one or few clones. No threat yet of genetic
erosion in the crop. 1 1 Camellia spp. in China and Viet
Nam were recorded as threatened in 1998.

APPENDIX 2 267

Origin: Arabica coffee originated in montane forest in
southwest Etfiiopia and the neighboring Boma Plateau
in Sudan; possibly also in Marsabit forest in Kenya,
Robusta grows wild in West and Central Africa.

Important sources of foreign currency in many
developing countries. Roasted seeds used for beverage
containing 1-2.5 percent caffeine, and niacin and
potassium. A number of other Collea spp. also
cultivated as coffee or as a source of edible berries.
Commercial arabica cultivars have very narrow genetic
base, especially in Latin America and the Caribbean.

Mate

//ex paraguariensls

Origin; South America ,

Tea IS made from leaves, contains 2 percent caffeine.
Little use in export market.

and are highly susceptible to disease. Robusta
outcrosses and has wider variation, t^luch production
by smallholders, but iO percent of coffee from
Americas and Caribbean from intensive monocrop
plantations. Recent recognition of importance of
conserving species-rich shade coffee systems.
Signihcant percentage of Ethiopian coffee from uniform
commercial cultivars; 400 000 ha remain of wild coffee,
accounting for half of Ethiopia's coffee production.
Several populations of wild species increasingly
restricted in distribution and fragmented. Nine spp.
from mainland Africa were listed as threatened in 1998.

^^■^—■■—■ffi II in H

Onion, garlic

Allium cepa, Allium sativum (Alliaceaei

Origin: Onion exists only in cultivation; may have come
from Afghanistan, Iran, and former USSR area.
Possible progenitors of garlic are A lonqicuspis or wild
A. ampeioprasum. The greatest number of AUium spp.
are in North Africa and Eurasia.

Underground bulbs are more important for their flavor
and antimicrobial properties than nutritional value.
Garlic contains large amounts of potassium and
significant vitamin C. Important component in the diet
of a wide range of cultures. Numerous medicinal

functions. The allins contained in Allium spp.,
especially garlic, may protect against cancer and
cardiovascular disease. Seven economically important
cultivated Ailium spp.; many other species consumed
on a lesser scale. Open pollinated populations still
represent most of the production in tropical and
subtropical countries. The habitat of some wild Album
spp. IS severely threatened. Poor results from
interspecific hybridizations The genetic variability
available in wild and cultivated relatives has not been
extensively used in crop improvement.

BEVERAGE CROPS

SPICES AND
FLAVORS

268 WORLD ATLAS OF BIODIVERSITY

SPICES AND
FLAVORS

Chili pepper, sweet pepper

Capsicum annuum (Solanaceael

Origin: Tropical America

Fruits of varying pungency are used either fresh as a
vegetable or dried or powdered as a spice. Fresh fruits
contain large quantities of vitamin A, plus vitamin C.

Cardamom

Elettaria cardamomum IZingiberaceael

Origin: India

Seeds used as a spice. Essential oil used in perfume
and as flavor for liqueurs. Wild populations exist in

Breeding has generally depended on pure lines. Wild
peppers are still collected and sold locally. Some
interfertility with other Capsicum spp. Wild spp. offer
valuable new traits.

monsoon forests in south India and Sri Lanka. There
are no essential differences between wild and cultivated
forms. Collection from the wild contributes to the
commercial trade. Wild populations are disappearing.

Pimento

Pimenta dioica IMyrtaceae

Origin: West Indies and Central America

Plants grow in semi-wild state. Populations of wild
relatives are confined to small areas of remaining dry

forest and coastal habitat in the Caribbean, especially
Cuba. They are poorly studied and are severely
threatened by habitat loss.

Origin: Western Ghats, India

The dried fruits, high in alkaloid content, represent one
of the oldest spice crops. Several other Piper spp.
important for local pepper production. Grown as a

smallholder crop in tropical countries. Large-scale
planting is based on one clone and is dangerously
vulnerable to disease. Wild pepper still grows in the
Western Ghats.

APPENDIX 2 269

Origin: Evolved from sea-beet (fi. vulgaris V3r.
maritima] in Europe and west Asia-
Swollen taproot provides nearly half tfie world
production of sucrose. Forms of ttie same species
include leaf beets and ctiards used as garden
vegetables, and other beets with swollen taproots, eg.

beetroot and mangold, for human consumption and
animal feed. All forms within the species may be
crossed. Wild relatives have already provided some
disease resistance. The only source of resistance
against the beet cyst nematode is detected in relatives
from a different section in the genus.

Origin: Europe and western Asia

Edible tree nut and ornamental. Kernels are used in
confectionery; 18 percent protein and 68 percent oil. All

Corytus spp. have edible nuts. C. colurna is also
cultivated for nuts. Populations of C. chinensis in China
have declined, largely because of overexploitation.

Walnut

Juglans regia Uuglandaceae)

Origin: Balkans to China

timber, ornamentals. No apparent threat of genetic

erosion in the crop. Major cultivation in the United

Edible nut. containing vitamins E. C and B. Use as

Stales but enormous unexplored potential elsewhere.

dessert and in confectionery; oil also extracted. The

Wild walnut forest has declined and become

l<ernel contains 15 percent protein and 70 percent

fragmented throughout its native range. Of the 21

unsaturated oil. Leaves matte good fodder The timber

species in the genus, seven were listed as threatened

IS highly valued. All Juglans spp. produce edible seeds.

in 1997.

Pistachio

Pistacia vera (Anacari

Origin: Near East and western Asia

Tree nut: low in sugar, more than 20 percent protein.
50 percent oil. Important food for nomads during
migration in Iran and Afghanistan. Highly drought
resislant. Trees used for ornamental and shade
purposes, also as a source of resin, dye. turpentine,
mastic and medicine. Cultivated in the Mediterranean
and western Asia for 3 000-4 000 years. None of the

related species has value as a nut crop, although seven
spp. are used as rootstocks and also for pollination.
Largely harvested from wild in Afghanistan and parts
of Pakistan. Iran has had commercial plantations for
hundreds of years. Wild species may have a role in
future improvements. Many wild populations have
been destroyed by forest clearance, over-cutting
for charcoal and grazing. Three spp. were listed as
threatened in 1998.

SUGAR CROPS

TREE NUTS

270 WORLD ATLAS OF BIODIVERSITY

TREE NUTS i

Brazil nut

Berthotletia excelsa ILecythidaceael

Origin: Tropical South America

Kernel contains 1 7 percent protein, 65-70 percent
monounsaturated oil. Largely an export crop. Also a
staple for indigenous people and important ecological
component of rainforest in the Amazon basin. Oil used
for cooking or as fuel or animal feed. Valuable timber
Attempts to establish plantations have generally failed
Well-managed plantations have the potential of
producing yields far exceeding natural groves. Almost
all nut production is from wild trees. Distribution and
density of groves may have been largely influenced by

indigenous groups in the past. Little information exists
on genetic variation. Populations appear to tolerate
different soil types. Sustainable system of harvesting in
extractive reserves, but considerable habitat loss and
illegal tree felling continues elsewhere. Development in
the Tocantins valley where there is high concentration
of brazil nut trees continues to cause population
decline. Developments elsewhere are also resulting in
serious genetic losses. The species was listed as
vulnerable in 1998. Protected populations are found in
biological reserves, Indian and extractive reserves and
corporate property

REFERENCES

1 Prescott-AUen, R. and Prescott-Allen, C. 1990. How many plants feed the world?
Conservation Biology A{A]: 365-37ii.

2 FAQ. FAOSTAT database. Available at Food and Agriculture Organization of ttie United
Nations website, http://apps.tao.org/ (accessed February 20021.

3 Mabberley, D.J. 1997. The plant-book. A portable dictionary of ttie vascular plants. 2nd
edition. Cambridge University Press, Cambridge.

4 Smartt, J. and Simmonds, N.W. 1995. Evolution of crop plants. 2nd edition. Longman
Scientific and Technical, Harlow.

5 Smith, N.J.H. et al. 1992. Tropical forests and their crops. Cornell University Press, Ithaca
and London.

6 Vaughan, J.G and Geissler, C.A. 1997. The new Oxford bool< of food plants. Oxford
University Press, Oxford.

7 Walter, K.S. and Gillett, H.J. (eds) 1998. 1997 lUCN Red List of threatened plants.
Compiled by the World Conservation Ivjonitoring Centre. lUCN-the World Conservation
Union, Gland and Cambridge.

APPENDIX 3 271

APPENDIX 3:
DOMESTIC LIVESTOCK

This table presents information on the major
domestic mammals used in agriculture and
related activities, such as hunting or transport,
and on the number and status of closely related
wild species.

At the local level, a great many wild animal
species are used primarily to meet subsistence
needs, the kind depending largely on availability,
convenience and tradition. Far fewer are used in
livestock production: breeds of goat, sheep, cattle
and pigs are cosmopolitan in distribution and,
along with domestic fowl, form the basis for most
of the world's agricultural animal food production
on land. These four principal mammalian
livestock species have diversified into more than
4 000 recognized breeds. While intensification of
production has typically gone hand in hand with
narrowing of the genetic base, such that semen
from individually documented and tested lines
commands a premium, there is increasing
recognition of the genetic potential resident in
less commercially developed breeds and blood
lines, and of the often neglected value of locally
adapted stock in comparison with commercial
stock from advanced industrial countries. The
pool of genetic resources represented by
domestic animal diversity is an essential basis for
efficient and sustainable food production, and is
likely to be of increasing importance in the more
demanding production environments.

There is no universally accepted system for
naming domestic stock. Some authorities
apply the earliest valid name to both the
wild species and to domestic stock derived from
it; others prefer to retain separate names for
domestic stock where such a name has been in
common use, and apply the next available valid
name to the wild species. In the first case, for
example, Capra hircus Linn. 1758 would be
applied to the wild goat and all domestic
derivatives; in the latter case, that name would be
restricted to domestic stock and Capra aegagrus
Erxleben 1 777 applied to the wild goat of Eurasia.
The second approach is adopted below.

In 'Number and status of breeds' the
first figure is the number of breeds given in the
Food and Agriculture Organization of the United
Nations database. Subsequent figures are, in
order: the number of threatened breeds in the
critical' category, the number endangered' and
the number 'extinct'. See Scherf for definitions.
For both threatened categories, the counts
include breeds maintained by active conservation
programs and by institutions.

Source: For general intormation. see Clutton-Brock' and Mason^;
for number and status of breeds see Scherf^; for number of
congeneric wild species see Wilson and Reeder*; and for status of
wild species see HiUon-Taylor^.

272 WORLD ATLAS OF BIODIVERSITY

Bos taurus Ihumpless, mainly European cattle,

taurine type]

Bos indicus Ihunnped, mainly Asian cattle, zebu type!

IBovidael

Meat, milk, transport, draught, dung. etc.

Domestic longhorn cattle from around 8 000 years
ago at several Middle East sites, later in Nile region,
and circum-Mediterranean by 3 000 years before the
present, European cattle probably of Middle East
origin. Humps, assumed result of artificial selection,
at base of neck or over shoulder in zebu type. Zebu
generally heat and parasite resistant, dominant in
Asia and Africa Isome longhorns persist, e.g.
trypanosome-resistant N'Dama in West Africal.
Cattle were first draught farm animals; in Europe

only specialized for meat or nnilk when replaced as
power source by horse. Very high breed diversity,
many now rare. British breeds to North America and
Australia in the IVth century; Iberian breeds earlier
to South America, Cattle certain to continue as major
farm animals for meat and milk. Much potential In
tropics for development of local stock, e.g. zebu dairy
breeds. Several feral herds.

Number and status of breeds: U79; C 106, E 193,
Ex 255

Derived from: 60s pnmigenius wild ox or aurochs
lextinctl

Congeneric wild species: i; all 4 listed as
threatened in 2000.

Bos frontalis iBovldae]

freely in forest during day or for months, restrained

at intervals, lack human control over breeding. May J

Ceremonial sacrifice, barter

breed with cattle and gaur j]

No firm evidence but probably of early origin.

Number and status of breeds: None formally \

Restricted to Bhutan, hills in northeast India

recognized '.t

bordering China and Myanmar, and Chittagong hills

of Bangladesh. Typically higher elevation than cattle

Derived from: Bos gaurus gaur. South and '

and lower than yak. Kept mainly by hill tribes.

Southeast Asia

usually by men of high status, for use in ceremonial

sacrifice, exchange and trophy display. Not much

Congeneric wild species: 4; all 4 listed as

used for draught or milk. Mithan generally forage

threatened in 2000 las for domestic cattle].'

Bos grunniens IBovidael

Milk, transport, meat, dam of 'dzo' Icattle x yak
hybrid draught animal]

Possibly domesticated at same time as cattle,
probably on Tibetan plateau or the Himalayas. Most
yak in west China, many in Mongolia, fewer in
Tajikistan. Kyrgyzstan, Nepal, Bhutan. Afghanistan,
India. Usually at 3 000-5 000 meters altitude-
Variable size and pelage, usually smaller than wild
yak. Yak tail in trade for centuries; white tips favored
for ease of dyeing. Yak can graze where other

livestock cannot. Much medical or religious use in
Tibet, where milk and butter most important; used
as meat source in Mongolia. Hair used for rope, felt;
skin for leather; dung for fuel; important pack
animal.

Number and status of breeds: 13; C 0, E 1. ExO

Derived from: Bos mulus yak, China: north of Tibet
plateau lAltun Shan. Qilian ShanI

Congeneric wild species: 4; all 4 listed as
threatened in 2000 las for domestic cattlel.

APPENDIX 3 273

Bos javanicus [Bowdae]

fertile, little fat, uses poor pasture in hot humid

conditions. Good draught animal for small fields and

Draught, meat

terraced slopes; much potential as meat or crossing

stock. Feral herd in Cobourg Peninsula lAustralial.

Domestic cattle present in Souttieast Asia ca 5 500

years before present, Banteng possibly domesticated

Number and status of breeds: No data

in prehistory in Southeast Asia or Java. Now in many

parts of Indonesia; small herds in Malaysia,

Derived from: Bosjavanicus banteng, Southeast Asia

Philippines, Australia. Uniform in type. Organized

selection in 20th century: no entire males exported.

Congeneric wild species: 4; all i listed as threatened

no crossing with other cattle. Small size, highly

in 2000 las for domestic cattlel.

Bubalus bubalis IBovidael

for draught, and river buffalo in South Asia,

mainly for milk. Do better than cattle on swamp and

Draught, milk

lloodplam grazing. Much potential for development

as meat producer. Milk rich in fat. Large feral herds

Probably domesticated earlier than 4 500 years ago

in Australia.

in Middle East. Wild ancestor occurred from

Mesopotamia east to Southeast Asia; by the 19th

Number and status ot breeds: 86; C 3, E 8, Ex

century restricted to India and adjacent areas, where

local. Domestic buffalo reached southeast Europe by

Derived from: Bubalus arnee wild water buffalo.

12th century where from Uth century much used in

Bhutan, India, Nepal. Thailand

Muslim communities; later taken to the Americas

and Australia, and Africa in 20th century. Breed

Congeneric wild species: 4; alU listed as

development centered in India and Pakistan. Broadly

threatened in 2000.

divided into swamp buffalo in Southeast Asia, mainly

Capra hircus IBovidael

where often adverse impact on native biota. Much

potential for further breed development, e.g. for

Meat, milk, hair

specialized tropical dairy animals. Ruminant

physiology allows efficient use of coarse vegetation

Goats and sheep next to be domesticated after dog.

in semi-arid and and regions.

Domesticated around 10 000 ago in Zagros

Mountains of western Iran; to Europe by mid-

Number and status of breeds: 587; C 31 , E 70. Ex 17

Neolithic. Worldwide distribution. Great variety in

form of horns and ears, hair color, etc. Highest

Derived from: Capra aegagruswild goat, southwest

numbers in South Asia. Milk breeds developed in

Asia: Turkey east to Pakistan

Switzerland have influenced many milk breeds

worldwide. The Boer ISouth Africal is major meat

Congeneric wild species: 9; 7 listed as threatened in

breed. Two fleece breeds: angora ITurkeyl and

2000.

cashmere Icentral Asial. Many feral populations.

274 WORLD ATLAS OF BIODIVERSITY

Ovis anes IBovidael

Meat, milk, wool

Sheep and goats next to be domesticated after dog.
Sheep in use in Mesolithic; evidence for
domestication around 1 1 000 years ago in Middle
East; to North Africa (where no wild sheepi by 6 000
years ago; to Annericas in 16th century. Worldwide
distribution; important in Europe, Middle East.
Central Asia. Coat of wild sheep has outer hairs over
woolly inner coat; hairs lost during domestication to
produce fine fleece breeds. Wool and mill< often
more important than meat. Wool trade basis of great
wealth in medieval and early modern Europe. Many
breeds: some multipurpose, others specialized for
milk, fleece or meat. Sheep numbers in decline in

some developed countries e.g. United States and
Australia, but elsewhere provide vital support to
human life in marginal and rangeland environments.
Ruminant physiology allows efficient use of coarse
vegetation in semi-arid and arid regions.

Number and status of breeds; 1 '195; C t
181

3, E 1V9, Ex

Derived from: Ovis onentslis mouflon, southwest
Asia: Turkey east to Iran; Mediterranean populations
(Corsica, Sardinia, Cyprus! possibly feral primitive
domestic stock

Congeneric wild species; 6; 4 listed as threatened in
2000.

Bus domesticus (Suidael

Production increasingly specialized, but still an

important role for local varieties in utilizing liouseliold

fvleat

waste and wild foods. Pigs have a major cultural

significance in parts of Southeast Asia and Melanesia.

First evidence of domestic pigs by 9 000 years ago in

Anatolia; widespread in Eurasia, incl. Egypt, by 5 000

Number and status of breeds: 6't9; C 58, E 1 06. Ex

years ago. Worldwide; nearly half the world's pigs

151

occur in non-Muslim Asia, mostly in China.

Management varied: may free-range in woodland or

Derived from; Sus scrofa, Eurasian wild pig, North

be sty-fed. Pigs introduced to the Americas from

Africa. Europe, Asia

Europe; few in Africa or Australia, New Zealand.

Several feral herds. Large number of breeds.

Congeneric v»ild species: 10; 6 taxa listed threatened

Commercial production now dominated by few lines.

in 2000

Came/us bactnanus (Camelidael

Draught, transport, meat, milk, wool, dung

Fossil camels known from North America (where no
extant camels! and Eurasia west to North Africa.
Rock drawing in Mongolia of two-humped camel may
be 10 000 years old. First evidence of domestication
in Iran and Turkmenistan about 5 000 years before
the present. Widespread in central Asia by ca 3 000
years ago. Mam transport on Silk Route between
Mesopotamia and China but replaced by dromedary
in west and south from ca 2 000 years ago.

Restricted to Central Asia, incl. Mongolia and China.

Numbers probably in decline.

Number and status of breeds: 1 1 ; C 1 , E 1 , Ex

Derived from: A domesticated form of wild Bactrian
camel, southwest Mongolia, northwest China

Congeneric virild species: Non-domesticated
populations in central Asia, listed threatened
(endangered) in 2000.

APPENDIX 3 275

Camelus dromedanus ICamelidael

Transport Idraught, meat, milk, wool, dung)

Remains of dromedary or similar species at
Paleolithic sites in North Africa about 80 000 years of
age. Wild camels apparently extinct in Africa by 5 000
years ago but persisted in Saudi Arabia Iwhere
perhaps first domesticated! until ca 2 000 years ago.
Domestic camel to Horn of Africa around 4 000 years
ago. Reached present importance with rise and
spread of Arab power from 7th century onwards.
Most camels in northeast Africa and Afghanistan/
Pakistan/India, where numbers rising; fewer and

Lama glama ICamelidael

Transport, wool Icoarsei. meat, dung

Domesticated by b 000 years before the present In
high-altitude Andean pastures, possibly centered
around Lake Titicaca basin of south Peru and west
Bolivia, Alpaca textiles known from 2 500 years ago.
Domestic camelids spread to lower altitudes and
along Andean chain by i 000 years ago and reached
greatest extent during Inca period; in decline since
Spanish conquest in early 16th century and
introduction of European stock. Remain important to
Andean culture and for superior adaptation to poor

Lama pacos ICamelidael

Wool Ifinel

See llama for background.

decreasing in Middle East, Primarily for transport;
specialized pack and riding breeds exist. Introduced
to Canaries and Australia (where feral herds]. Ability
to withstand long periods without drinking and use
thorny browse key to human use of hot deserts.

Number and status of breeds: 52; CI . E 2. Ex

Derived from; Ancestral form unknown, presumed
extinct Camelus species; Bactrian camel closest
living relative

Congeneric VKild species: See Bactrian camel.

high-altitude grazing. Pad feet may cause less
pasture damage than hoofs of sheep. Llamas and
most alpacas held by small-scale pastoralists on
communal grazing; some alpaca kept in large herds
by cooperatives In Peru, Not milked. Alpaca wool has
high commercial value. Llama flocks in the United
States and Europe,

Number and status of breeds: 8; C 0. E 0. Ex

Derived from: Probably Lama guanicoe guanaco,
south Peru, west Bolivia, northwest Argentina

Congeneric wild species: 1; not listed threatened.

Number and status of breeds: 6; C 0, E 1 . Ex

Derived from: Unknown in wild, presumed Lama sp.
or Lama x Vicugna hybrid

Congeneric wild species: See llama.

276 WORLD ATLAS OF BIODIVERSITY

Rangifer tarandus ICervidaej

potential tor better use; numbers have been

increasing but with local indications of overgrazing.

Meat, milk, transport

Lichens, the mam winter feed, are vulnerable to

atmospheric pollution.

Fossil evidence for use of reindeer from 80 000 years

ago, domesticated by 2 500 years before ttie present-

Number and status of breeds: No data

Management varies: riding or miilk animals may be

separated from herd and fed, or herds may roam

Derived from: Semi-domesticated forms of Eurasian

widely and be gathered annually for marking or

Rangifer tarandus. reindeer lEurasial, caribou INorlh

slaughter Reindeer industry important in north

America!

Scandinavia, northwest Russia and Siberian Russia,

less so in North America, Reindeer exploitation key

Congeneric wild species: See above; \ North

to settling the far north. Wild reindeer include four

American subspecies listed threatened in 2000.

major types, all used in husbandry systems. Some

.JVggg r donkey

Equusas/nuslEquidae)

Transport, draught, sire of mule lass x horse hybrid!

Probably domesticated in northeast Africa; records
from about 6 000 years ago in Egypt. The only
domestic animal certainly of African origin.
Widespread in Middle East by ca 2 000 years ago. To
Americas in 16th century. Much more important than
horse in Africa where present in north and west.
Common in south and central Asia; also present in
south Europe. Mostly lor transport; specialized riding
and pack breeds exist. Formerly milked, meat

sometimes used. Feral asses widespread incl.
Socotra, Galapagos, United States. Australia, Sahara,
etc. Numbers worldwide likely to decline, but
because of hardiness and low cost will retain
importance in less developed areas.

Number and status of breeds: 1 03; C 5, E 1 6, Ex 6

Derived from: Equus afncanus, African wild ass.
North Africa to Somalia

Congeneric wild species: 9; 8 taxa listed threatened
in 2000.

Equus caballus lEquidae!

Transport, draught, sport, dam of mule lass x horse
hybrid!

Some evidence of domestic horses 6 000 years ago in
central Eurasia lUkraine!. Spread through Eurasia
during Bronze and Iron Ages. Important early military
use. to draw chariots and lor riding, especially after
invention of stirrups before the start of the 6th
century. Wild horses present with Amerindians in
North America but extinct there by 12 000 years ago;
domestic horses introduced by European colonists.

Most horses occur in South America where numbers
also highest in relation to humans; numbers high in
North America and Asia. Specialized for draught or
riding, but both uses in decline. Feral horses on all
continents lexcept Antarctica!.

Number and status of breeds: 820; C 1 27. E 1 78.
Ex 94

Derived from: Equus ferus If. przewalskiil wild
horse, formerly Americas, Europe. Asia

Congeneric wild species: See ass.

APPENDIX 3 277

Cams lamiliaris ICanidael

Number and status of breeds: A few hundred, no

data on status

Companion, hunting, security, food

Derived from: Cams lupus wolf. North America,

Domestication may have begun long before that of
agricultural stock, possibly 100 000 years ago; first
direct evidence U 000 years ago in Middle East;
distinct kinds of dog evident by 7 000 years ago. The
dingo is a feral domestic dog taken to Australia

Europe. Asia

Congeneric wild species: 7; 3 taxa listed threatened
in 2000.

around 12 000 years ago.

Cavia porceltus ICaviidael

Number and status of breeds: No data

Meat, laboratory, companion

Derived from: Cavia aperea. widespread in

South America, or C. tschudil Peru, south

One of the few domestic animals of South American

Bolivia, northwest Argentina, north Chile

origin. Probably domesticated 3 000-6 000 years ago.

but in use long before. Taken to Caribbean and Europe

Congeneric wild species: i; 1 listed threatened

by mid-16th century. Some planned selective breeding

in 2000

during past 30 years. Potential for more development

as meat source, especially in original Andean range.

but broiler fowl increasingly used instead.

■BK

Number and status of breeds: fO; C 0, E 4, Ex

Derived from: Oryctoiagus cumculus European
rabbit, west and south Europe to northwest Africa

Congeneric wild species: None; a monospecific
genus (several related genera listed threatenedl.

Oryctoiagus cumculus iLeporidael

Meat, fur, laboratory, companion

Kept enclosed jin leporaria by Romans! for more
than 2 000 years. Kept by medieval monks; newborn
or unborn young were permissible food during Lent.
Distributed worldwide by manners; many feral
populations. Some development of meat breeds
since the second world war; much potential as low-
cost converter of surplus vegetation into meat.

REFERENCES

1 Clutton-Brock, J. 1999.4 natural history of domesticated mammals. 2nd edition.
Cambridge University Press, Cambridge, and the Natural History Ivluseum, London.

2 Mason, LL. led.) 1984. Evolution of domesticated animals. Longman, London and New York.

3 Scherf, B.D. led.) 2000. World Watch List for domestic animal diversity. 3rd edition.
Food and Agriculture Organization of tfie United Nations. Rome.

4 Wilson, D.E. and Reeder, D.M. (eds) 1993. Mammal species of the world: A taxonomic and
geographic reference. 2nd edition. Smithsonian Institution Press, Washington DC and London.

5 Hilton-Taylor, C. (compiler) 2000. 2000 lUCN Red List of threatened species. lUCN-the
World Conservation Union, Gland and Cambridge. Available online at
http://www.redlist.org/ (accessed April 20021.

278 WORLD ATLAS OF BIODIVERSITY

APPENDIX 4:

RECENT VERTEBRATE EXTINCTIONS

This list provides information on vertebrate
species thai have been listed as globally extinct. It
is based in part on lists produced using the lUCN
threat category system, but mammals, birds and
fishes have been revised following more rigorous
criteria developed by the Committee on Recently
Extinct Organisms |CREO|. The OREO system
relates to extinctions since AD1500. Where
information meets all OREO criteria, the
extinction event is considered fully resolved; these
species appear on a darker background. To date,
extinct and possibly extinct mammals and fishes
have been evaluated in this way, and data are
available at the CREO website. The bird list is
based on Threatened Birds of the World which
itself informally follows CREO guidelines. The
reptiles and amphibians listed have been regarded
as extinct, but in several cases data needed to
meet CREO criteria le.g. on taxonomy or survey
effort! appear to be unavailable. A number of

species have been removed, as new evidence
suggests that extant populations of that taxon
persist or because extinction took place before
AD1500. The Lake Victoria cichlid fishes in the
genus Haplochromis [sensu lata] at the end have
been widely regarded as extinct, but data available
leave some uncertainty whether these species are
extinct in the wild, or persist in small numbers
outside sampled areas. These fishes are listed as
questionably extinct and are distinguished by a
paler background. The Period column indicates,
very approximately, the probable period during
which extinction occurred. 'Early' refers to the
first four decades of a century ICl, 'mid' the next
three decades, 'late' the final three decades.
Where a more precise date is available, this is
given in parentheses.

Source: Data on mammals and fishes. CREO ; on birds, BirdLife
International ; on reptiles and amphibians, Hilton-Taylor .

Class MAMMALIA

Order DASYUROMORPHIA

Family Thylacinidae

Thylacinus cynocephalus

Thylacine

Tasmania

Early 2l)thC 119361

Order PERAMELEMORPHIA

Family Peramelidae

Chaeropus ecaudatus

Pig-footed bandicoot

Australia

Mid 20th C

Macrotis leucura

Lesser bilby

Australia

Mid20thCI1960sl

Perameles eremiana

Desert bandicoot

Australia

Mid20thCI19(i0sl

Order DIPROTODONTIA

Family Macropodidae

Cabprymnus campestris

Desert rat-kangaroo

Australia

Early 20th C 11932)

Lagorchestes asomatus

Central hare-wallaby

Australia

Mid 20th C1 19605)

Lagorchestes leporides

Eastern hare-wallaby

Australia

Late 19th C 118901

Macropus greyi

Toolache wallaby

Australia

Late 20th C 11972) _

Onychogalea tunala

Crescent nailtail wallaby

Australia

Mid 20th C 11956) ■

Polorous ptatyops

Broad-faced potoroo

Australia

Late 19th C 11875)^

Order INSECTIVORA
Family Nesophontidae
Nesophontes hypomicrus

Nesophontes longirostns

Atalaye island-shrew Hispaniola Early 16th C? "^

Long-nosed island-shrew Cuba Post-1500

APPENDIX U 279

I^^^H

^^B

■M^g

Nesophontes mapr

Cuba

Post- 1500

Nesophontes micrus

Western Cuban island-shrew

Cuba

Post- 1500

||[ Nesophontes paramicrus

St Michel island-shrew

Hispaniola

Early 16th C '

Nesophontes submicrus

Cuba

Post-1500

W Nesophontes superstes "^H

Cuba

Early 16th C? ;

t Nesophontes zamicrus

Haitian island-shrew

Hispaniola

Early 16th C? I

Nesophontes sp. 1

Grand Cayman island-shrew

Grand Cayman

Post- 1500

Nesophontes sp. 2

Cayman Brae island-shrew

Cayman Brae

Post- 1500

Family Solenodontidae

1 Solenodon marcanoi

Marcano's solenodon

Hispaniola

Early 16th C |

Order CHIROPTERA

Family Molossidae

E Mystacina robusta

New Zealand lesser
short-tailed bat

New Zealand

Mid 20th C (1965! ]

Family Pferopodidae

* Dobsonia chapmani

Dobson's fruit bat

Negros 1.

Late 20th C

(Philippines)

(1970s!

Nyctimene sanclacrucis

Nendo lube-nosed fruit bat

Santa Cruz
ISolomonsI

Late 19th C (1892! ,

Pteropus brunneus

Dusky flying fox

Percy 1.

Late 19lhC (1874!

i Pteropus pilosus

(Queensland. Australia!

Large Palau flying fox

Palau

Lalel9lhC(1874! ,

Pteropus subniger

Reunion flying fox

Mauritius,

Mid 19th C (1840s! '

I Pteropus tokudae

Reunion

Guam flying lox

Guam

Mid 20th C (1968!

Family Vespertilionidae

Kemoula slncana

Dobson's painted bat

Tanzania

Pre-1878

1 Pharolis imogene

Large-eared nyctophilus

Papua
New Guinea

Late 19th C (1890! ^

Order PRIMATES

Family Megaladapidae

f Megaladapis cf. M, edwardsi

Tretretretre

Madagascar

Early 16lhC?

Family Paleopropithectdae

|| Palaeopropithecus ingens

Large sloth lemur

Madagascar

Early 16th C? |

Family Xenotrichidae

J Xenothrix mcgregori

Jamaican monkey

Jamaica

Early 16th C |

Order CARNIVORA

Family Canidae

1 Dusicyon australis

Falkland island dog

Falkland Is

Late 19th C 11876| ^

'Sl

280 WORLD ATLAS OF BIODIVERSITY

^Kecres^^^^^^H

' UvmnW^^^^^^^^^^^^^^I

Perio^^^^B

Family Mustelidae

Mustela macrodon

Sea mink

USA

Late 19th C 118601

Family Phocidae

MonachuB Iropicalis

Caribbean monk seal

Caribbean Sea

Mid 20th C 119505]

Order SIRENIA
Family Dugongidae

Hydrodamalis gigas

Steller's sea cow

Russia. USA

Mid 18th C 117681 "■

Order ARTIODACTYLA

Family Bovidae

Gazella rufina
Hippotragus leucophaeus

Red gazelle
Bluebuck

Algeria
South Africa

Late 19th C 11894]
Early 19th C 11800]

Family Hippopotamidae

Hippopotamus lemedei

Malagasy dwarf hippo

Madagascar

Early 16th C

Hippopotamus
madagascanensis

Order RODENTIA
Family Capromyidae
Capromys sp. 1

Common Malagasy hippo
Cayman hutia

Madagascar
Cayman Is

Post- 1500 :m
Post- 1500 ^

Geocapromys colombianus
Geocapromys thoractus

Cuban coney
Swan Island hutia

Cuba Little
Swan 1.
IHondurasI

Early 16th C? W
Mid20thCll950sia

Geocapromys sp. 1
Geocapromys sp, 2

Great Cayman coney
Cayman Brae coney

Grand Cayman
Cayman Brae

Post- 1500
Post-1500

Hexolobodon phenax
Isolobodo!) montanus
Isolobodon portoricensis
Plagiodontia ipnaeum
Rtwzopiagiodontia lemkei

Imposter hutia
Montane hutia
Allen's hutia
Johnson's hutia
Lemke's hutia

Hispaniola
Hispaniola
Hispaniola
Hispaniola
Hispaniola

Early 16th C?
Early 16th C? _
Early 16th C? fl
Early 16th C ■
Early 16th C? J

Family Echiimyidae
Boromys offetla
Boromys torrei

Cuban esculent spiny rat

De la Torre's esculent spiny rat

Cuba

Post- 1500 1

Post-1500 '

Brotomys voratus

Muhoy

Hispaniola

Early 16th C? J

Family Heptaxodontidae
Quemisia gravis

Family Mundae

Quemi

Hispaniola

Post- 1500

Coniiurus albipes

White-footed rabbit rat

Australia

Late 19th C 11875]

Craleromys paulus
Leimacomys buetiner,

llin bushy-tailed cloud-ral
Groove-toothed forest mouse

Philippines
Ghana

1953

1890

Leporilius apicalis

Lesser stick-nest rat

Australia

Late 20lh 0(1970] ^

APPENDIX U 281

|Bli^"™

W^^

Malpaisomys insulans

Volcano mouse

Canary Is

Post- 1500

Megalomys audreyae

Barbuda giant rice rat

Barbuda
and Antigua

Pre- 1890

P Megalomys desmarestii

Martinique rice rat

Martinique

Early 20th C (1902)

Megalomys tuciae

St Lucia giant rice rat

Sainl Lucia

Late 19th C (18811

Megaoryzomys cunoi
L

Curio's giant rice rat

Santa Cruz.
Galapagos

Mid 18th C (1740) '

Megaoryzomys sp 1

Isabela giant rice rat

Isabela. Galapagos

Post- 1500

1 Nesoryzomys darwini

Darwin's rice rat

Santa Cruz,

Mid 20th C 11940s)

__J3alapagos

Nesoryzomys sp 2

Isabela Island nee rat B'

Isabela.
Galapagos

Post- 1500

Noronhomys vespuccii

Vespucci's rice rat

Fernando da
Noronha 1. (Brazil)

Early 16th C :

Notomys amplus

Short-tailed hopping-mouse

Australia

Late 19th C (1896)

^ Notomys bngicaudatus

Long-tailed hopping-mouse

Australia

Early20thC(19Gl)

Notomys macrotis

Big-eared hopping-mouse

Australia

1843

Notomys mordax

Darling Downs hopping-mouse

Australia

1846

Notomys sp 1

Great hopping-mouse

Australia

Pre- 1900

1 Oligoryzomys victus

St Vincent pygtITy rice rat

Saint Vincent

Late 19th C (1892)

1 Oryzomys antillarum

Jamaican rice rat

Jamaica

Late 19th 0(1877)

Oryzomys liypenemus

Barbuda rice rat

Barbuda
and Antigua

Post- 1500

f Oryzomys nelsoni

Nelson's rice rat

Maria Warire 1.
(Mexicol

Late l9thC (1897) ;

Oryzomys sp. 1

Barbados nee rat

Barbados

Pre- 1390

W Peromyscus pembertoni

Pemberton's deer mouse

San Pedro
Nolasco 1. [Mexicol

Early 20th C (1931)

Pseudomys gouldii

Gould's mouse

Australia

Early 20th C (1930)

Rattus madeari

Maclear's rat

Christmas 1.

Early 20th 01903)

■

lAustralial

' Rattus nativitatis

Bulldog rat

Christmas 1.
[Australial

Early20thC (19031

Uromys imperator

Giant naked-tailed rat

Guadalcanal

Mid 20th C (1960s)

(Solomons!

Uromys porculus

Little pig rat

Guadalcanal

Late 19th C (1887)

(Solomons)

Order LAGOMORPHA

Family Leporidae

Sylvilagus insonus

Omilteme cottontail

Mexico

Late 20th C (1991)

Family Ochotonidae

f Prolagus sardus

Sardinian pika

Corsica,
Sardinia

Late 18lhC (1777)

282 WORLD ATLAS OF BIODIVERSITY

||M|^|pHHHi^^pHI^^^^B

Class AVES

■■^H

Order CASUARIIFORMES

^^^^^^^H

Family Dromaiidae

-^^^^^^^H

Dromaius ater

King Island emu

King 1.
(Australia!

Early 19th C fl

Dromaius baudmianus

Kangaroo Island emu

Kangaroo 1.
(Australia]

Early 19th C

Order PODICIPEOIFORMES

Family Podicipedidae

Podiceps andinus

Colombian grebe

Colombia

Late 20th C (1977!

Podifymbus gigas

Atitan grebe

Guatemala

Late 20th CI1986!

Order PROCELLARIIFORMES

Family Procellariidae

Bulweria bifax

St Helena bulwer's petrel

St Helena

16th C M

Oceanites maorianus

New Zealand storm petrel

South 1.
(New Zealand!

^■fl

Pterodroma rupinarum

St Helena gadfly petrel

St Helena

16th c ^^m

Order PELECANIFORMES

Family Ardeidae

^^B

Ixobrychus novaezelandia

New Zealand little bittern

New Zealand

Late 19th C? ■

Nycticorax duboisi

Reunion night-heron

Reunion

Late 17th C?

Nycticorax mauritianus

Mauritius night-heron

Mauritius

Early 18th C?

Nycticorax megacephalus

Rodrigues night-heron

Rodrigues

Mid 18th C

Family Phalacrocoracidae

Phalacrocorax

Pallas's cormorant

Bering Straits

Mid 19th C(1850s|

perspiciUatus

(Russia!

Family Threskiornithidae

Threskiornis solitahus

Reunion fhghtless ibis

Reunion

Early 18th C ■

Order ANSERIFORMES

Family Anatidae

Alopocben mauribania

Mauritian shelduck

Mauritius

Late 17th C (16981 _
Early 19th C' 1
(1793!

Anas marecuta

Amsterdam Island duck

Amsterdam 1.

(France!

Anas Iheodon

Mauritian duck

Mauritius,

Early 18th C?

Reunion?

(I710I

Camptorhyncbus

Labrador duck

Canada, USA

Late 19th 0(18781

labradohus

Mergus australis

Auckland island
merganser

New Zealand

Early 20th C11902!

APPENDIX 4 283

Order FALCONIFORMES
Family Falconidae
Cracara lutosus

Order GALUFORMES
Family Phasianidae
Argusianus bipunctatus
Coturnix novaezelandiae

Order GRUIFORMES

Family Rallidae
Aphanapteryx bonasia
Aphanapteryx leguati
Aramides gutturalis
Atiantisia elpenor
Atlantisia podarces
Cabalus modestus
Fulica newloni

GaWnula nesiotis

GaWrallus dielfenbachii

GaWrallus pacificus

GaWrallus sharpei

Gallirallus wakensis
Nesodopeus poecilopterus
Porphyrio a/bus

Porphyria coerutescens
Porphyria hochsletten

Porphyria kukwiedei
Porzarta astnctocarpus
Parzana monasa

Porzana nigra

Parzana palmen
Porzana sandwichensis

Order CHARADRIIFORMES
Family Charadriidae
Haematopus meadewaldoi

Common name

Guadalupe caracara

Double-banded argus
New Zealand quail

Red rail
Rodrigues rail
Red-throated wood rail
Ascension flightless crake
St Helena crake
Chatham rail
Mascarene coot

Tristan moorhen

Oieffenbach's rail

Tahiti rail

Sharpe's rail

Wake Island rail
Bar-winged rail
Lord Howe swamphen

Reunion gallinule
North Island takahe

New Caledonia gallinule
St Helena rail
Kosrae crake

Miller's rail

Laysan crake
Hawaiian crake

Canary Islands oyslercatcher

Place

Guadalupe
IMexicol

Southeast Asia'
New Zealand

Early 20th C
11900)

Late 19th C?
Late 19th C 118751

Mauritius

Early 18th C 117001

Rodrigues

Mid 18th C 11761)

Peru

19th C

Ascension 1.

Early 19th C?

St Helena

Early 16th C

New Zealand

Late 19th C 11900')

Mauritius.

Early 18th C

Reunion

116931

Tristan da

Late! 9th C

Cunha

Chatham Is

Late 19th C

iNewZealandl

118721

Society Is

Early 20th C

IFrench Polynesial

11930sl

Indonesia?

Late 19th

or early 20th C?

Wake 1. lUSAI

Mid 20th C 119451

Fill

Late 20th C 119731

Lord Howe 1.

Early 19th C?

lAustralial

Reunion

Early 18th C 117301

North 1.,

Late 19th C

New Zealand

New Caledonia

17th C

St Helena

Early 16th C

Fed, States

Mid 19th C

Micronesia

Society Is

Late 18th C

IFrench Polynesial

Hawaii

Mid 20th C|)9W1

Hawaii

Late 19th C 118841

Canary Is

Mid or late 20th C

284 WORLD ATLAS OF BIODIVERSITY

Species I^^^^^H

{^Common name ^^^H

^^1

^^H

■^-TTr5'rs?57S»?r5«S5sa3Si

Family Laridae

Pinguinus impennis

Great auk

North Atlantic
coasts

Mid 19th C (18521

Family Scolopacidae

■

Prosobonia ellisl

White-winged sandpiper

Society Is
(French Polynesia]

Late 18th C

Prosobonia leucoptera

Tahitian sandpiper

Society Is
(French Polynesia!

Late 18th C?

Order COLUMBIFORMES

Family Columbidae

■It

Alectroenas nitidissima

Mauritius blue pigeon

Mauritius

Early 19th C (1830s!

Alectroenas rodericana

Rodrigues pigeon

Rodrigues
(Mauritius!

Early 18th C?

Columba duboisi

Reunion pigeon

Reunion

Early 18th C'

Columba jouyi

Ryukyu pigeon

Nansei-shoto
(Japan!

Early 20th C (19361

Columba versicolor

Bonin wood pigeon

Ogasawara-shoto
(Japan!

Late 19th C (1889! J

Dysmoropelia dekarchiskos

St Helena dove

St Helena

Early 1 6th C?

Ectopistes migratorius

Passenger pigeon

USA

Early 20th C (19001

Gallicolumba ferruginea

Tanna ground dove

Vanuatu

Late 18th C ■S

Gallicolumba norfolaensis

Norfolk Island ground dove

Norfolk 1.
(Australia!

Late 18th C? 1

Microgoura meeki

Choiseul pigeon

Choiseul
(Solomon Isl

Early 20th C (190^1!

Ptilinopus mercieni

Red-moustached fruit-dove

Marquesas Is

Early 20th C J

(French Polynesia!

119221 ^jl

Family Raphidae

Pezophaps solitana

Rodrigues solitaire

Rodrigues

Late 18lh C ^
(1760s!

Raphus cucutlatus

Dodo

Mauritius

Late 17th C (1665!

Order PSITTACIFORMES

Family Psittacidae

Amazona martinicana

Martinique parrot

Martinique

Late 18th C

Amazona violacea

Guadeloupe parrot

Guadeloupe

Late 18th C
(1779!

Ara atwoodi

Dominican green-and-yellow
macaw

Dominica

Early 19th C?

Ara erythrocephala

Jamaican green-and-yellow
macaw

Jamaica

Early 19th C

Ara gossei

Jamaican red macaw

Jamaica

Late 18th C?

Ara guadeloupensis

Lesser Antillean macaw

Guadeloupe,
Martinique

Late 18th C?

APPENDIX 4

B!? ^^^Hl

4ra Incotor

Cuban red macaw

Cuba,
Hispaniola

Late 19th C (1885!

Aratinga tabati

Guadeloupe parrot

Guadeloupe

Late 18th C

Conuropsis carolmensis

Carolina parakeet

USA

Early 20th C (1904!

Cyanoramphus ulietanus

Raiatea paral^eet

Society Is

Late 18th C?

(French Polynesia)

(1773!

Cyanoramphus zealandicus

Black-fronted parakeet

Society Is

(French Polynesial

Mid 19th C (1844!

Lophopsittacus bensont

Mauritius gray parrot

Mauritius

Late 18th C (1764!

Lophopsittacus mauritianus

Broad-billed parrot

Mauritius

Late 17th C (1675!

Mascahnus mascannus

Mascarene parrot

Reunion

Late 18th C (1770s!

Necropsitlacus rodencanus

Rodrigues parrot

Rodrigues
(Mauritius!

Late 18th C (17611

Neslor productus

Norfolk Island kaka

Phillip 1,

Early 19th C ,

(Australia!

Psephotus pulcherrimus

Paradise parrot

Australia

Mid 20th C (19271 '

Psitlacula exsul

Newton's parakeet

Rodrigues
(Mauritius!

Late 19lhC (1875!

Psittacula wardi

Seychelles parrot

Seychelles

Late 19th CdeSOs!

Order CUCUUFORMES

Family Cuculidae

Coua delalandei

Snail-eating coua

Madagascar

Mid 19th 0(18341

Nannococcyx psix

St Helena cuckoo

St Helena

18th C?

Order STRIGIFORMES

Family Strigidae

Mascarenotus gwcheti

Reunion owl

Reunion

Early 17th C

Mascarenotus munvows

Rodrigues owl

Rodrigues

Early 18th C (1726!

Mascarenotus sauzieri

Mauritius owl

Mauritius

Mid 19th C (18371

Sceloglaux albifacies

Laughing owl

New Zealand

Early 20thC (1914!

Order AP0DIF0RME5

Family Trochilidae

Chbrostilbon bracei

Brace's emerald

Bahamas

Late 19th C (18771

Chlorostilbon elegans

Gould's emerald

Bahamas?
Jamaica?

Late 19th C?

Order UPUPIFORMES

Family Upupidae

Upupa antaois

St Helena hoopoe

St Helena

Early 16th C?

Order PASSERIFORMES

Family Acanthisittidae

Traversia lyalli

Stephens Island wren

Stephens 1
(New Zealand!

Late 19th 0(18941

Xenicus bngipes

Bush wren

New Zealand Late 20th C (1972!

286 WORLD ATLAS OF BIODIVERSITY

^^^Place

Periot^^^^^l

'^^SiSS

. .-i-ijU

Family Cailaeidae

Heleralocha acutirostris

^^^^^^^M

1 New Zealand

Early 20th C 11907]

Family Drepanididae

Akialoa ellisiana

Oahu 'akialoa

Oahu. Hawaii

Early 20th C 7 118371

Akialoa lanaiensis

Maui Nui 'akialoa

Lanai. Hawaii

Early 20th C 7118921

Akialoa obscura

'Akialoa

Hawaii

Mid 20th C 7119401

Akiaba stejnegeri

Kaua'i akialoa

Kaua'i, Hawaii

Late 20th C 7 119691 ;

Ciridops anna

Ula-ai-hawane

Hawaii

Late 19th C (18921 ;

Drepanis funerea

Black mamo

Hawaii

Early 20th C (19071 .'

Drepanis pacifica

Hawaii mamo

Hawaii

Late 19lhC 118991 ]

Dysmorodrepanis munroi

Lana'i hookbill

Hawaii

Early 20th 0(19201 '

Paroreomyza flammea

Kakawihie

Hawaii

Mid 20th 0(19631

PsiWrostra kona

Kona grosbeak

Hawaii

Late 19th 0(18941 i

Rhodacanlhis flaviceps

Lesser koa-finch

Hawaii

Late 19th 0(18911 |

Rhodacanthis palmen

Greater koa-finch

Hawaii

Late 19th 0(18961

Viridonla sagittirostns

Greater 'amakihi

Hawaii

Early 20th 0(19011 j

Family Fringillidae

Chaunoprodus ferreorostns

Bonin grosbeak

Ogasawara-shoto
(Japan)

Late 19th 0(189071

Family Icteridae

Quiscalus palustris

Slender-billed grackle

Mexico

Early 20th 0(19101

Family Meliphagidae

Chaetoptila angustipluma

Kioea

Hawaii

Late 19th 0(18591

Moho apicalis

Oahu 00

Hawaii

Mid 19th G (18371

Moho bishopi

Bishop's 00

Hawaii

Early 20th 0(19041

Moho braccatus

Kaua'i 00

Kaua'i, Hawaii

Late 20th C (19871

Moho nobilis

Hawaii 00

Hawaii

Early 20th 07(19341

Family Muscicapidae

Anlhomis melanocephala

Chatham Island bellbird

New Zealand

Early 20th 0(19061

Bowdlena rufescens

Chatham Island fernbird

Chatham Is

Early 20th ■>

(New Zealand!

(19001

Gerygone insulans

Lord Howe gerygone

Lord Howe 1..
Australia

Early 20th

Myadestes oabensis

Amaui

Hawaii

Mid 19th C7 118251

Myiagra freycineti

Guam flycatcher

Guam

Late 20th C 119831

Nesillas aladabrana

Aldabra bush-warbler

Aldabra
ISeychellesI

Late 20th 0(19831

Pomarea pomarea

Maupiti monarch

Maupiti,
French Polynesia

Early 19th 07 (18231

Turdus ravidus

Grand Cayman thrush

Cayman Is

Early 20th 0(19381

Turnagra capensis

South Island piopio

New Zealand

Mid 20th 0(19631

Turnagra lurnagra

North Island piopio

New Zealand

Mid 20th 0(19551

APPENDIX 4 287

^^^^ffife^H^^^^H^^jyCAm m on^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^B

Zoothera terrestns Bonin thrush

Ogasawara-shoto

Mid 19th C 118281

IJapanj

Family Sturnidae

Apbnis corvina

Kosrae mountain starling

Kosrae

Mid)VlhCll828|

(Fed. States Micronesial

Aplonis fusca

Norfolk Island starling

Norfolk 1.
(Australia]

Early20thC 119231

Apbnis mavornata

Mysterious starling

Cook Is

Mid 19th C 11825]

Fregilupus vanus

Reunion starling

Reunion

Mid 19th C 118505]

Necrospar rodencanus

Rodrigues starling

Rodrigues

Early 18th C 117261

Family Zosteropidae

Zosterops strenuus

Robust white-eye

Lord Howe 1.
lAustralial

Early 20th C (1928)

Class REPTILIA

Order SAURIA

Family Anguidae

Cetestus occiduus

Jamaican giant galliwasp

Jamaica

Mid 19th 0(1840]

Family Gekkonidae

Hoplodadylus dekourti

New Zealand I'l

Mid 19th C

Phelsuma gigas

Giant day gecko

Rodrigues

Late 19th

Family Iguanidae

Leiocephalus eremitus

Navassa 1.
lUSAl

Early 20th C (19001

Leiocephalus herminieri

Martinique

Early 19th C (1830s]

Family Scmcidae

]

Leiolopisma mauntlana

Mauritius

Early 17th 0(1600]

Macwscincus coctei

Cape Verde giant skink

Cape Verde

Early 20th C

Tachygia microlepis

Tongan ground skink

Tonga

17th 0?

Tetradactylus eastwoodae

Eastwood's longtailed seps

South Africa

Early 20th C? ;

Family Teiidae

Ameiva cineracea

Guadeloupe

Early 20th J

Ameiva major

Martinique giant ameiva

Martinique

17th C? ^^

Order SERPENTES

Family Boidae

Bolyeria multocannata

Round 1.
iMauriliusI

Late 20th 0(1975]

Family Colubridae

Alsophis sanclicruas

St Croix racer

Virgin Is lUSAl

Mid 20th J

288 WORLD ATLAS OF BIODIVERSITY

OT^^H

^^^

mmmsm

m^^

Family Typhlopidae

TypMops canei

' Mauritius

17th C .

Order TESTUDINES

Family Testudinidae

Cylindraspis borbonica

Reunion

Early 19th C (1800!

Cylindraspis indica

Reunion

Early 19th C (1800)

Cylindraspis inepta

Mauritius

Early 18th C

Cylindraspis peltasles

Rodrigues

Early 19th C (18001

Cylindraspis tnserrata

Mauritius

Early 18th C

Cylindraspis vosmaen

Rodrigues

Early 19th C (1800!

Class AMPHIBIA

Order ANURA

Family Discoglossidae

Discoglossus nigmenter

Israel

jainted frog

Israel

Mid 20th C (1940!

Family Myobatrachidae

Uperoleia marmorata

Marbled toadlet

Australia

Late 19th C

Family Ranidae

Arthroleptides dutoiti

Kenya

Early 20th C

Rana fisheri

Relict

leopard frog

USA

Mid 20th C (1960!

Rana tlaloci

Mexico

Late 20th 0(1 990sl

Class ACTINOPTERYGII

Order CYPRINIFORMES

Family Catostomidae

' Xflas'mstes~n^u^iei

Snake River sucker

USA

Early 20th Cl1927r^

^ Moxostoma lacerum

Harelip sucker

USA

Late 19th C (1893! ]

Family Cyprinidae

Acanthobrama hulensis

Lake Huleh
(Israel)

Late 20th C ?

Anabanlius alburnops

Lake Dianchi
IChinal

20th C

Anabarilius polylepis

Lake Dianchi
(China!

20th C

Barbus microbarbis

Lake Luhondo
(Rwanda!

Mid 20th C

' Cephalakompsus pachychsilus

Lake Lanao
(Philippines!

Mid 20th C' (post 19211'

Chondrostoma scodrensis

Lake Skadar
(Albania. Yugoslavia

20th C (post 18811

Cyprinus yilongensis

Lake Yilong
(China!

Late 20th C (19771 j

APPENDIX i 289

P^gtnfilftBff'rianie

Pefiofd ^

Evarra buslamantei

Mexican dace

Mexico

Mid 20th C 11957!

Evarra eigenmanni

Plateau dace

Mexico

Mid 20th C11V54!

Evarra tlahuacensis

Endorheic dace

Mexico

Mid 20th C (19571 ,

Gita crassicauda

Thicktail chub

USA

Mid 20th C (19571

Hybopsis amecse

Ameca shiner

Mexico

Late 20th C 11969!

Hybopsis aubdion

Durango shiner

Mexico

Late 20th C

Lepidomeda altivelis

Pahranagat spinedace

USA

Mid 20th C 11933!

Mandibularca resinus

Bagangan

Lake Lanao
IPhilippinesI

Early 20th C 11922!

Notropis amecae

Anneca shiner

Mexico

Late 20th C (1970s?!

Notropis aubdion

Durango shiner

Mexico

Mid 20th C 11961! :

Notropis orca

Phantom shiner

Mexico. USA

Late 20th C '>

j Ospatutus pataemorphagus

Lake Lanao

Early 20th C 119241 ',

(Philippines)

Early 20th C 11921!

Ospatulus twnculatus

Bitungu

Lake Lanao

(Philippines!

Phoxinellus egndin

Yag baligi

Lake Egridir
ITurkeyl

Mid 20th C 119551

Phoxinellus handbrschi

Cicek

Lake Egridir
ITurkeyl

Mid 20th C (1955!

Pogonichthys ciscoides

Clear Lake splittail

USA

Late 20th C 11970! |

Puntius amarus

Pait

Lake Lanao
IPhilippinesI

Early20thC 11910!

Punbus baoulan

Baolan

Lake Lanao
(Philippines!

Early 20th C 119261

Punbus demensi

Bagangan

Lake Lanao
(Philippines!

Early 20th C 119211

Punbus disa

Disa

Lake Lanao
(Philippines!

Early 20th C 119321

Puntius flavifuscus

Katapa-tapa

Lake Lanao
(Philippines!

Early 20th C 119211

Puntius tierrei

Lake Lanao
(Philippines!

Early 20th C 119081

Punbus tanaoensis

Kandar

Lake Lanao
(Philippines!

Early 20th C 119221

Puntius man3lal(

Manalak

Lake Lanao
IPhilippinesI

Early 20th C 119241

Puntius Strang

Sirang

Lake Lanao
IPhilippinesI

Early 20th C 119321

Puntius tras

Tras

Lake Lanao

IPhilippinesI

Early 20th C 119251

Rhinichthys deaconi

Las Vegas dace

USA

Mid 20th C (19401

Sprattlicypris palata

Palata

Lake Lanao
(Philippines!

Early 20th C (19221

StvDodon sianifer

Slumptooth minnow

Mexico

Early 20th C (19031

290 WORLD ATLAS OF BIODIVERSITY

■fJ'^iJyl^i.-X'V.f-, /■ ,- ^,-.:^.^■..■..:»l^^.,

^^Hj^H

^rl^^^^H

Order CHARACIFORMES

Family Characidae

Brycyon acuminatus

Brazil

20th C

Order OSMERIFORMES

^^B

Family Retropmnidae

Prototrocles oxyrhyncbus

New Zealand grayling

New Zealand

Early 20th 0119231^

Order SILURIFORMES

Family Scliilbeidae

Plalytropius siamensis

Thailand

Late 20th C 119661

Family Tnchomycteridae

Rhizosomichthys totae

Lake Tola
IColombial

Mid 20th C 119571

Order SALMONIFORMES

Family Salmonidae

Coregonus alpenae

Longiaw cisco

Great Lakes
(Canada. USA]

Late 20th C 119751

Coregonus confusus

Parrig

Lake Moral
ISwitzerlandl

Mid 20th C ■:
'i

Coregonus fera

Fera

Lake Geneva
ISwitzerlandl

Mid 20th C

Coregonus gutturosus

Kilch

Lake Constance
ISwitzerlandl

Mid 20th C

Coregonus hiemalis

Gravenche

Lake Geneva
ISwitzerlandl

Mid 20th C i

Coregonus johannae

Deepwater cisco

Great Lakes

Mid 20th C

ICanada, USAI

Coregonus restnctus

Fent

Lake Moral
[Switzerland!

Mid 20th C

Salvelinus agassizi

Silver trout

Dublin Pond
lUSAl

Early 20th C 1193011

Salvetinus mlramundis

Orkney char

Hoyl.lUKI

Early 20th C 119091

Salvelinus scharlfi

Scharffs char

Lough Owel

llrelandl

Early 20th C 119081

Order ATHERINIFORMES

FanfiilyAtherinidae

Chirostoma compressum

Cuitzeo silverside

Lake Cuitzeo
IMexicol

Mid 20th C 7119401 •
!

Rheocles sikorae

Zona

Madagascar

Mid 20th C 119661

Order CYPRIN0D0NTIFORME5

Family Cyprinodontidae

"T3SR5FFitb de la Presa

Mexico

Late 20lh CI19881 ■
Late,25tJ)C119£l4)J

, Cyprinodon mmemomm

Cachorrito de la Trinidad

. Mexico

APPENDIX k 291

^^^^^^^^^^^Hr Common name ^^^^^^^^^^^^^^^^^^^^^^|

wr_

'''^^prinodon iatifasciatus

Pernio de Parras

Mexico

Early 20th C 119031

Cypnnodon sp. '

Monkey Spring puplish

USA

Late 20th C

Leptolebias marmoralus
Orestias cuvien

Ginger pearllish
Lake Tilicaca oreslias

Brazil

Lake Tilicaca
IBolivia. Peru]

Mid 20th C [mil]
Mid 20th C 119371

Family Fundulidae

Fundulus albolineatus

Whileline lopminnow

USA

Early 20th C ■> 1)899]

Family Goodeidae

Characodon garmani
Empetrichthys merriami

Parras characodon
Ash Meadows kiUilish

Mexico
USA

Late 19th C (18891 1
Mid 20th CI19A8I j

Family Poeciliidae

Gambusia amistadensis

Amislad gambusia

USA

Mid 20th C 119631

Order BELONIFORMES
Family Adrianichthyidae

Adriankhthys kruyti

Duck-billed buntingi

Lake Poso
llndonesial

Late 20lh C 119831 i

Order GASTEROSTEIFORMES
Family Gasterosteidae

' Pungitius kaibarae

Kyoto ninespine stickleback

Japan

Mid 20th C 119591 j

Order 5C0RPAENIF0RMES

Family Cottidae

Cottus echinatus

Utah Lakesculpin

Utah Lake
lUSAI

Early 20lh 0(19281

Order PERCIFORMES

Family Cichlidae

Tristramella magdalenae

Israel

Late 20th 0(19971

Family Gobiidae

■ Lake Poso.
1^ Sulawesi (Indonesia

Late 20th 0(19851

Hapbchromis attigenis

Lake Victoria

Mid/late 20th C ?

Hapbchromis apogonoides

Lake Victoria

Mid/late 20th C ?

Hapbchromis arcanus

Lake Victoria

Mid/late 20th C ?

Hapbchromis argenteus

Lake Victoria

Mid/late 20th C ?

Hapbchromis artaxerxes

Lake Victoria

Mid/late 20th C '

Hapbchromis barbarae

Lake Victoria

Mid/late 20th C ^

Hapbchromis bareli

Lake Victoria

Mid/late 20th ■>

Hapbchromis barbni

Lake Victoria

Mid/late 20th ?

Hapbchromis bayoni

Lake Victoria

Mid/late 20th ?

Hapbchromis boops

Lake Victoria

Mid/late 20th ?

292 WORLD ATLAS OF BIODIVERSITY

^^^^m^^^

^^^^^^ Place

^^^H

Haplochromis cassius

Lake Victoria

Mid/late 20th C ''

Haptochromis cinctus

Lake Victoria

Mid/late 20th C ?

Haplochromis cnester

Lake Victoria

Mid/late 20th C ?

Haplochromis decticostoma

Lake Victoria

Mid/late 20th C ?

Haplochromis dentex group

Lake Victoria

Mid/late 20th C ?

Haplochromis diplotaenia

Lake Victoria

Mid/late 20th C ">

Haplochromis estor

Lake Victoria

Mid/late 20th C

Haplochromis llavipinnis

Lake Victoria

Mid/late 20th C ?

Haplochromis gilberti

Lake Victoria

Mid/late 20th C '

Haplochromis gowersi

Lake Victoria

Mid/late 20th C ?

Haplochromis guiarti

Lake Victoria

Mid/late 20th C ■?

Haplochromis heusinkveldi

Lake Victoria

Mid/late 20th C '

Haplochromis hiatus

Lake Victoria

Mid/late 20th C ?

Haplochromis ins

Lake Victoria

Mid/late 20th C ?

Haplochromis longirostris

Lake Victoria

Mid/late 20th C ''

Haplochromis macrognathus

Lake Victoria

Mid/late 20th C ?

Haplochromis maculipinna

Lake Victoria

Mid/late 20th C '

Haplochromis mandibularis

Lake Victoria

Mid/late 20th C ?

Haplochromis martini

Lake Victoria

Mid/late 20th C ?

Haplochromis megalops

Lake Victoria

Mid/late 20th C '

Haplochromis michaeli

Lake Victoria

Mid/late 20th C ?

Haplochromis microdon

Lake Victoria

Mid/late 20th C ''

Haplochromis mylergates

Lake Victoria

Mid/late 20th C ?

Haplochromis nanoserranus

Lake Victoria

Mid/tale 20th C ?

Haplochromis nigrescens

Lake Victoria

Mid/late 20th C ?

Haplochromis nyanzae

Lake Victoria

Mid/late 20th C '

Haplochromis obtusidens

Lake Victoria

Mid/late 20th C ?

Haplochromis pachycephalus

Lake Victoria

Mid/late 20th C ''

Haplochromis paraguiarti

Lake Victoria

Mid/late 20th C ?

Haplochromis paraplagiostoma

Lake Victoria

Mid/late 20th C?

Haplochromis parorthostoma

Lake Victoria

Ivlid/late 20th C ''

Haplochromis percoides

Lake Victoria

Mid/late 20th C '

Haplochromis pharyngomylus

Lake Victoria

Mid/late 20th C ?

Haptochromis prognathus

Lake Victoria

Mid/late 20th C •>

Haplochromis pseudopellegrini

Lake Victoria

Mid/late 20th C ?

Haplochromis pyrrhopteryx

Lake Victoria

Mid/late 20th C ?

Haplochromis spekii

Lake Victoria

Mid/late 20th C ''

Haplochromis teegelaari

Lake Victoria

Mid/late 20th C '^

Haplochromis thuragnathus

Lake Victoria

Mid/late 20th C ?

Haplochromis tridens

Lake Victoria

Mid/late 20th C ?

Haplochromis victorianus

Lake Victoria

Mid/late 20th C ?

Haplochromis xenostoma

Lake Victoria

Mid/late 20th C '

Haplochromis 'bartoni-like'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'bicolor'

Lake Victoria

Mid/late 20th C ?

Haplochromis big teeth'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'black cryptodon'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'back pectoral'

Lake Victoria

Mid/late 20th C ?

APPENDIX 4 293

HHHI^^H^Hi^l^HH

Haplochromis 'chlorocephalus

Lake Victoria

Mid/late 20th C ?

Haplochromis 'citrus'

Lake Victoria

Mid/late 20th C ■>

Haplochromis 'coop'

Lake Victoria

Mid/late 20th C '

Haplochromis 'elongate rockpicker'

Lake Victoria

Mid/late 20th C '>

Haplochromis 'fiiamentus

Lake Victoria

Mid/late 20th C '

Haplochromis 'fleshy lips'

Lake Victoria

Mid/lale 20th C '

Haplochromis 'gray pseudo-nigricans'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'large eye guiarti'

Lake Victoria

Mid/late 20th C '>

Haplochromis 'lividus-frels'

Lake Victoria

Mid/late 20th C '

Haplochromis 'longurius'

Lake Victoria

Mid/late 20th C ''

Haplochromis 'macrops like'

Lake Victoria

Mid/late 20th C '

Haplochromis 'micro-obesus'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'morsei'

Lake Victoria

Mid/late 20th C ->

Haplochromis 'orange cinereus'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'orange macula'

Lake Victoria

Mid/late 20th C '

Haplochromis 'orange yellow big teeth'

Lake Victoria

Mid/late 20th C ■>

Haplochromis 'orange yellow small teeth'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'paropius-like'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'pink paedophage'

Lake Victoria

Mid/late 20th C '

Haplochromis 'pseudo-morsei' -

Lake Victoria

Mid/late 20th C '

Haplochromis 'purple head'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'purple miller'

Lake Victoria

Mid/late 20th C ?

Haplochromis 11] 'purple rocker'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'red empodisma'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'red eye scraper'

Lake Victoria

Mid/late 20th C ?

Haptochromis 'reginus'

Lake Victoria

Mid/late 20th C ^

Haplochromis 'regius

Lake Victoria

Mid/late 20th C ?

Haplochromis 'short supramacrops'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'small blue zebra'

Lake Victoria

Mid/lale 20th C ?

Haplochromis 'small empodisma'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'smoke'

Lake Victoria

Mid/late 20th C ?

Haplochromis "soft gray'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'stripmac'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'supramacrops'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'theliodon-like'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'tigrus'

Lake Victoria

Mid/late 20th C ■?

Haplochromis 'too small'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'twenty'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'two stripe white lip'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'wyber'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'xenognathus-like'

Lake Victoria

Mid/late 20th C ?

Haplochromis 'yellow'

Lake Victoria

Mid/late 20th C i

Haplochromis 'yellow-blue'

Lake Victoria

Mid/late 20th C ?

Hoplotilapia retrodens

Lake Victoria

Mid/late 20th C ?

Psammochromis cryplogramma group

Lake Victoria

Mid/late 20th C ?

294 WORLD ATLAS OF BIODIVERSITY

REFERENCES

1 BirdLlfe International 2000. Threatened birds of the world. Lynx Edicions and BirdLife
International, Barcelona and Cambridge.

2 CREO. Connmittee on Recently Extinct Organisms ICREO] website: http://creo.amnh.org/
(accessed January 20021.

3 Hilton-Taylor. C. (compiler! 2000. 2000 lUCN Red List of threatened species. lUCN, Gland and
Cambridge. Online at http://www.redlist.org/ (accessed April 2002).

APPENDIX 5 295

APPENDIX 5:

BIODIVERSITY AT COUNTRY LEVEL

This table includes estimates of the number of
mammals, breeding birds and vascular plants in
each country of the world, together with estimates
of the number of these endemic to each country.
The threat status of mammals and birds has been
comprehensively assessed, and for these classes
the number and percentage of globally threatened
species present in each country are given. In the
mammals, most estimates of the total number
present relate to native non-marine species only,
but the threatened species counts for many
countries include marine species (a few island
countries have a small number of non-marine
mammals but a higher number of threatened
species - the percentage figure is omitted in such
casesl. The richness and endemism data should

Methodology notes

Numerical indices representing national biodiversity
have been derived from available estimates of species
total and species endemism. Taxonomic groups covered
are: mammals, birds, reptiles, amphibians and vascular
plants Iferns to flowering plants!. For birds alone, the
database includes estimates of botfi the total number
recorded li.e. including non-breeding migrants and
accidental visitors! and the number of breeding bird
species; the latter has been used in the analysis. All
countries belovi/ 5 000 km^ in area have been excluded
from the analysis, leaving 169 countries.

Four arbitrary assumptions are made: the four
vertebrate classes included are of equal importance;
plants are of equal importance to the vertebrates
combined; richness and endemism are reasonably
vtfell correlated at country level across the taxonomic
groups covered; vertebrates plus plants provide
a valid surrogate for biodiversity in general.

Because interest lies in relative biodiversity rather
than absolute values, data in each column are first
normalized. Each estimate N/ for each country /is
divided by Nmax where Nmax is the highest value for

be taken as provisional; they are subject to
change with new taxonomic treatments and
new surveys. The columns headed Dl and Al,
respectively, contain the unweighted national
diversity indices, and the diversity indices adjusted
for area [see methodology notes below for
details!. The index is based on data for the groups
shown here, plus reptiles and amphibians, not
included here. The Al column is the basis for
Map 5.4.

Source: WCMC database; data derived from a large number of
published and unpublished sources, including country reports and
regional checklists Numbers of threatened species retrieved
from the online version of the 2000 tUCN Red List of threatened
species, at http;//www.redlist.org/ [accessed f^larch 20021.

that parameter This transforms the data to within the
range 0-1, with the most important country having the
value Nmax/Nmax- 1, and the least important having
the value closest to zero. Estimates for mean vertebrate
richness fV/?i and mean vertebrate endemism IVEIare
derived by averaging the figures for all classes, and
estimates for combined richness fWand combined
endemism /flare derived for each country by
averaging figures for vertebrates and plants. Inspection
of the data shows that PR and W correlate quite
closely, while Pf tends to be approximately half VE;
where estimates of PPand Pf are missing the
appropriate vertebrate-based value has been inserted
before calculating Rand £

An overall diversity index IDII is calculated for each
country as the mean of R and £ This treats richness
and endemism equally and so makes fewest
assumptions about their relative significance in terms of
overall biodiversity, but the 0/ could be weighted to give
greater importance to either. Because species richness
tends to increase with area, and with proximity to the
humid tropics, 0/ is strongly affected by country area

296 WORLD ATLAS OF BIODIVERSITY

and geographical position, but it also takes account of
levels of endemism, which are shaped by several
factors, including topography, geographic isolation
and tectonic history.

Because area is an important determinant of
species number, there is much interest in evaluating

relative levels of biodiversity per unit area, i.e. how
much more or less rich in species is any given area
lor country). This may be addressed by the Arrhenius
equation describing the species-area relationship
Hog S = g + z log A], where S = number of species,
A = area, z is the slope of the line, and g another

^flW^

^^^5!^^

^^!P^

^mSnfWS^

km'

total

endemic

no. theatened

Afghanistan

652 225

0.063

-0.296

119

Albania

28 750

0.035

-0.019

Algeria

2 381 7^5

0.045

- 1.003

American Samoa

197

Andorra

465

Angola

1 246 700

0.176

0.544

276

Anguilla

Antigua and Barbuda

Argentina

2 777 815

0.196

0.423

320

Armenia

29 800

0.042

0.153

Aruba

193

Australia

7 682 300

0.608

1.268

252

206

Austria

83 855

D.036

-0.293

Azerbaiian

86 600

0.05

0.027

Bahamas

13 865

0.017

- 0.503

5 ■

Bahrain

661

Bangladesh

laooo

0.059

0.058

125

Barbados

430

Belarus

207 600

0.029

-0.771

Belgium

30 520

0.023

- 0.441

11 :

Belize

22 965

0.056

0.526

125

Benin

112 620

0.08

0.437

188

Bermuda

Bhutan

46 620

0.058

0.366

160

Bolivia

1 098 575

0.239

0.882

316

23 1

Bosnia and Herzegovina

51 129

0.034

-0.200

10 1

Botswana

575 000

0.062

-0.287

164

' 1

Brazil

8 511 965

0.74

1.436

394

119

79 1

British Ind. Oc. Terr.

Brunei ^^|||||||^

5 765

0.071

1.145

157

' 1

Bulgaria ^^^^^

110910

0.044

-0.167

15 1

Burkina Faso

274 122

0.068

0.011

147

' i

Burundi

27 835

0.072

0.723

107

Cambodia

181 000

0.059

0.001

123

Cameroon

475 500

0.167

0.762

409

Canada

9 922 385

0.067

- 1.014

193

Cape Verde

4 035

Cayman Islands

259

Central African Republic

624 975

0.08

- 0.058

209

APPENDIX 5 297

constant. If it is assumed, given its area dependence,
that Dl IS scaled in the same way as 5, a regression
analysis of log Dl and log A allows the constants g and z
to be calculated. The regression line establishes the
expected biodiversity value of each country for its area,
and the distance of each country point from the line

gives a measure M//of how much more Itvel or less
l-vel diverse is a given country than expected. Al
attempts to assess diversity per unit area, rather than
overall biodiversity value per country, and it is noticeable
that several smaller mainland countries and island
states move up in rank order, as would be expected.

Hwa^R^^

^^ira^^

Hlant^^

^^linl^^

% threatened

breeding total

endemic

no. threatened

% threatened

total

endemic ,

235

4 000

800

230

3 031

192

3 164

250

100

471

113

1 350

765

5 185

1 260

321

1 158

897
242

38
4

4
2

9 372

1 100

3 553

460

649

350

15 638

14 074

213

3 100

248

4 300

240
118

1 111

195

295

5 000

572

221

2 100

180

1 550

356

2 894

150

307

2 500

167

448

5 468

17 367

4 000

218

386

2 151

1 492

185

113

56 215

101

359

6 000

240

3 572

320

335

1 100

451

2 500

307

19
15

690

8 260

156

426

3 270

147

774

539

537

3 602

100

= zero: - = no data.

298 WORLD ATLAS OF BIODIVERSITY

Area

°'jl

^^^^

J^mmals

Mammals

; km'

im^i

^nal

endemic

no. theatened

Chad

1 2U 000

0.049

- 0.739

134

Chile

751 625

0.112

0.229

China

9 597 000

0.392

0.767

394

Colombia

1 138 915

0.538

1.685

359

Comoros

1 860

Congo, Dem. Rep.

2 345i10

0.218

0.579

450

Congo. Republic

342 000

0.128

0.589

200

Cook Islands

233

Costa Rica

50 900

0.162

1.358

205

Cote d'lvoire

322 i65

0.116

0.507

230

Croatia

56 538

0.036

-0.169

Cuba

1U525

0.12

0.829

n J

Cyprus

9 250

0.017

- 0.429

3 1

Czech Republic

78 864

D.033

-0.356

8 1

Denmark

43 075

0.021

- 0.643

5 1

Djibouti

23 000

0.02

-0.528

^ J

Dominica

751

' 1

Dominican Republic

48 440

0.076

0.625

' 1

Ecuador

461 475

0.353

1.519

302

31 1

Egypt

1 000 250

0.038

- 0.936

12 1

El Salvador

21 395

0.048

0.393

135

Equatorial Guinea

28 050

0.084

0.869

184

Eritrea

117 600

0.057

0.088

112

Estonia

45 100

0.025

- 0.483

Ethiopia

1 104 300

0.145

0.383

277

3^ --

Falkland Islands

15 931

0.004

- 2.040

4 jl

Faroe Islands

3 jl

Fiji

18 330

0.028

-0.100

5 S

Finland

337 030

0.023

- 1.145

6 fl

France

543 965

0.051

- 0.473

18 H

French Guiana

91 000

0.079

0.483

150

9 M

French Polynesia

3 940

3 H

French S. and Antarctic Te

rr 7 241

0.001

-3.261

;fl

Gabon

267 665

0.116

0.56

190

15 fl

Gambia

10 690

0.036

0.308

117

3 S

Georgia

69 700

0.051

0.111

107

U H

Germany

356 840

0.033

-0.770

12 S

Ghana

238 305

0.114

0.571

222

13 9

Gibraltar

Greece

131 985

0.062

0.129

14 fl

Greenland

2 175 600

0.007

-2.821

7 ^1

Grenada

345

Guadeloupe

1 780

5 S

Guam

450

2 fl

Guatemala

108 890

0.142

1.014

250

6 S

Guinea

245 855

0.094

0.373

190

11 fl

Guinea-Bissau

36 125

0.05

0.289

108

2 B

APPENDIX 5 299

Birds Birds Birds Birds Plants Plants

breeding total endemic no. threatened % threatened total endemic

= zero; - = no data.

370

1 600

296

5 284

2 698

1 100

32 200

18 000

1 695

51 220

15 000

721

136

929

11 007

1 100

449

6 000

1 200

100

284

600

12 119

950

535

3 660

224

4 288

137

6 522

3 229

1 682

199

1 900

196

1
5

1450
826

1
6

126

1 228

136

5 657

1 800

1 388

19 362

4 000

153

2 076

251

2911

273

3 250

319

213

1 630

626
64

6 603

1 000

165

236

1 518

760

248

1 102

269

1
1

4 630

133

5 625

144

959

560

466

6 651

280

974

4 350

380

239

2 682

529

3 725

600

251
62

4 992

742

529

1 068

1 400

100

330

458
409

8 681

1 171

3 000

243

1 000

300 WORLD ATLAS OF BIODIVERSITY

■Countri^HH^^^I

■

■Mammals

' total

Mammals
endemic

Mammals
no. theatened

Guyana

2U970

0.133

0.758

193

Haiti

27 750

0.071

0.71

Honduras

112 085

0.094

0.597

173

Hungary

93 030

0.031

- 0.457

Iceland

102 820

0.006

- 2.080

6
86

India

3 166 830

0.326

0.896

390

Indonesia

1 919^5

0.731

1.844

515

222

140

Iran

1 648 000

0.091

-0.194

140

Iraq ^^^^^^^^

438 445

0.041

-0.629

^^l^^^^^l

68 895

0.013

- 1.248

5
14

^^^|H^H

• 20 770

0.043

0.285

116

Italy ^^^^^B

301 245

0.065

-0.056

Jamaica

11 425

0.051

0.619

Japan

369 700

0.124

0.536

188

Jordan

96 000

0.036

-0.310

Kazakhstan

2 717 300

0.071

-0.581

178

Kenya

582 645

0.145

0.56

359

Kiribati

684

Korea, DPR

122 310

0.025

-0.775

13 .

Korea, Republic

98 445

0.03

-0.518

13 1

Kuwait

24 280

0.007

- 1.564

Kyrgyzstan

198 500

0.036

-0.537

Lao PDR

236 725

0.081

0.229

172

Latvia

63 700

0.025

- 0.553

Lebanon

10 400

0.031

0.145

Lesotho

30 345

0.025

- 0.354

Liberia

111 370

0.059

0.132

193

Libya

1 759 540

0.029

-1.343

9 ■

Liechtenstein

160

3 ;

Lithuania

65 200

0.026

- 0.544

5
6

Luxembourg

2 585

Macedonia, FYR

25 713

0.037

0.077

11 1

Madagascar

594 180

0.298

1.277

141

50 1

Malawi

94 080

0.079

0,473

195

' 1

Malaysia

332 965

0.254

1.28

300

47 1
fl

Maldives

298

Mall

1 240 140

0.053

- 0.658

137

13 M

Malta

316

Marshall Islands

181

Martinique

1 079

■■

Mauritania

1 030 700

0.041

-0.856

10 ■

Mauritius

1 865

3 M

Mayotte

376

Mexico

1 972 545

0.589

1.621

491

140

69 ^

Micronesia, Fed. States

702

6 ■

Moldova

33 700

0.025

- 0.396

3 j

Monaco

APPENDIX 5 301

Birds
breeding total

Birds
endemic

Birds
no. threatened

Birds
% threatened

678

6 409

5 242

1 623

422

5 680

148

205

2214

88
923

377

18 664

5 000

1 519

408

113

29 375

17 500

323

8 000

172

1
12

1
7

950

180

2317

234

5 599

712

113

3 308

923

250

5 565

2 000

U1
396

6
4

2 100
6 000

844

6 506

265

115

2 898

107

112

2 898

224

234

4 500

487

8 286

217

1 153

154

3 000
1 591

372

2 200

103

1 825

134

124

1 410

202

1 796

126

1 246

210

3 500

202

105

9 505

6 500

521

3 765

501
23

37
1

7
4

15 500

3 600

583

397

1 741

914

100

1 287

273

1 100

750

325

500

769

26 071

12 500

100

1 194

293

177

1 752

= zero; - = no data.

302 WORLD ATLAS OF BIODIVERSITY

Countr^^^^^^^H

Arefll

^^■j

1 "

Mammals

km'^"

total

endemic

no. theatened

Mongolia

1 565 000

0.051

-0.767

133

Montserrat

104

Morocco

458 730

0.057

- 0.304

105

Mozambique

lU 755

0.09

0.005

179

Myanmar

678 030

0.141

0.493

300

Namibia

824 295

0.102

0.116

250

Nauru

Nepal

U1 415

0.096

0.549

181

Netherlands

41 160

0.022

-0.599

Netherlands Antilles

800

New Caledonia

19 105

0.078

0.904

New Zealand

265 150

0.065

-0.017

Nicaragua

148 000

0.098

0.555

200

Niger

1 186 410

0.061

-0.512

131

Nigeria

923 850

0.107

0.131

274

Niue

259

Northern Marianas

477

Norway

386 325

0.024

- 1.107

Oman

271 950

0.03

-0.812

Pakistan

803 940

0.08

-0.121

188

Patau

492

Panama

78 515

0.162

1.236

218

Papua New Guinea

462 840

0.271

1,254

214

58 J

Paraguay

406 750

0.115

0.429

305

' 1

Peru

1 285 215

0.396

1.344

460

47 1

Philippines

300 000

0.225

1.188

153

102

50 1

Pitcairn Islands

Poland

312 685

0.032

-0.761

Portugal

92 390

0.045

- 0.088

17
2 1

Puerto Rico

8 960

0.033

0.259

Qatar

11 435

0.005

- 1.770

,■■

Reunion

2510

3 H

Romania

237 500

0.039

- 0.490

17 .^1

Russia

17 075 400

0.179

-0.179

269

42 "9

Rwanda

26 328

0.087

0.925

151

B m

San Marino

1 ^

Sao Tome and Principe

964

3 '9

Saudi Arabia

2 400 900

0.04

- 1.129

7 S

Senegal

196 720

0.057

-0.065

192

11 ^1

Seychelles

404

4 ^1

Sierra Leone

72 325

0.083

0.588

147

11 ^1

Singapore

616

3 H

Slovakia

14 035

0.037

0.252

9 S

Slovenia

20 251

0.036

0.106

9 ^1

Solomon Islands

29 790

0.049

0.316

21 |H

Somalia

630 000

0.087

0.025

171

19 ^H

South Africa

1 184 825

0.252

0.915

247

41 H

APPENDIX 5 303

^^^nmals

Birds

Bird^^l

^^^M

Birds

Plants

^threatened

breeding total

endemic iliSiW^Shed

% threatened

total

endemic^H

426

2 823

229

671

210

3 675

625

498

5 692

219

867

35
9

7 000

1 071

469

3 174

687

611

6 973

315

191

1 221

107

3 250

3 200

150

2 382

1 942

482

7 590

299

1 460

681
15

9
1

4715

205

178

315

243

1 715

107

1 204

375

4 950

372

732

9915

1 222

644

11 544

556

7 851

1 538

112
186

17 144

5 356

196

8 931

3 500

227

2 450

207

5 050

150

105

8
6

2 493

235

355

546

165

247

3 400

628

11 400

513

2 288

895

134

155

2 028

384

2 086

250

182

466

2 090

118

2 282

209

3 124

207

3 200

163
422

43
11

3 172

3 028

500

596

23 420

= zero; - = no data.

304 WORLD ATLAS OF BIODIVERSITY

Country

Area

Mammals

km'

total

endemic

no. theatened

Spain

504 880

0.067

-0.172

Sri Lanka

65 610

0.082

0.606

St Helena and dep.

411

St Kittsand Nevis

261

St Lucia

619

St Vincent

389

Sudan

2 505 815

0.137

0.093

267

Suriname

163 820

0.092

0.471

180

Swaziland

17 365

0.044

0.353

Sweden

440 940

0.026

-1.067

Switzerland

41 285

0.033

-0.173

Syria

185 680

0.046

- 0.265

Taiwan

36 960

0.058

0.418

Taj il<i Stan

143 100

0.033

- 0.536

Tanzania

939 760

0.189

0.693

316

Thailand

514 000

0.162

0.709

265

Togo

56 785

0.094

0.781

196

Tokelau

Tonga

699

Trinidad and Tobago

5 130

0.045

0.729

100

Tunisia

164 150

0.033

- 0.572

Turkey

779 450

0.114

0.237

116

Turkmenistan

488 100

0.044

-0.572

103

Turks and Caicos Islands

430

Tuvalu

Uganda

236 580

0.12

0.624

345

Ukraine

603 700

0.05

- 0.509

108

United Arab Emirates

75 150

0.019

- 0.883

United Kingdom

244 880

0.024

-1.003

12
37

United States of America

9 372 614

0.342

0.638

428

105

Uruguay

186 925

0.05

-0.186

US Pacific Islands

658

Uzbekistan

447 400

0.051

-0.413

Vanuatu

14 765

0.014

-0.728

Venezuela

912 045

0.379

1.398

323

Viet Nam

329 565

0.147

0.737

213

37 J

Virgin Islands (British!

153

° 1

Virgin Islands (US)

352

' 1

Wallis and Futuna

255

° 1

Western Sahara

3 .1

Western Samoa

2 840

' i

Yemen

477 530

0.041

-0.654

' J

Yugoslavia

102 173

0.046

- 0.086

11 M

Zambia

752 615

0.096

0.074

233

12 m

Zimbabwe

390 310

0.099

0.298

270

APPENDIX 5 305

nmals

Birds

Plants

Plafl|

eatened

breeding total

endemic

no. threatened

% threatened

total

endei^

278

5 050

941

250

3314

890

165

659

1 028

108

1 166

680

3 137

603

5018

364

2715

249

2
2

1
1

1 750
3 030

193

204

3 000

160

3 568

5 000

822

24
2

33
37

4
6

10 008

11 625

1 122

616

391

3 085

463

260

2 259

236

173

2 196

302

8 650

2 675

448

11
2

4 900

830

263

5 100

230

1 623

650
237

19 473

4 036

11
-1

2 278

1
9

4 800

400

870

150

1 340

21 073

8 000

535

10 500

1 260

475

60
40

330

737

1 650

135

224

4 082

605

4 747

211

532

4 440

= zero; - = no data.

306 WORLD ATLAS OF BIODIVERSITY

APPENDIX 6:
IMPORTANT AREAS FOR
FRESHWATER BIODIVERSITY

This table presents information on areas
identified as of special importance for diversity
(species richness, and/or endemism) in the
inland water groups treated (fishes, mollusks
['MolL' in table], crabs, crayfish, fairy shrimps).
This is a preliminary synthesis, designed to
represent expert opinion on relative levels of

Source: This information was prepared for WCMC on
the basis of data kindly provided by a number of expert
ichthyologists and members of the lUCN/SSC Specialist
Groups for Inland Water Crustaceans and Molluscs. The
particular source of information is indicated by letters in
square brackets in the Remarks text. Please note that
numerical estimates and other information may have
been superceded by later survey and taxonomic work.

GA Gerald Allen, in litt. March 1998.

MK Maurice Kottelat, report compiled for

WCMC^'.
SK Sven Kullander. report compiled for

WCMC'^
M5G Adapted from a report by lUCN/SSC Mollusc

Specialist Group, primarily by Philippe

Bouchet and Olivier Gargominy, and also by

Arthur Bogan and Winston Ponder
KC Keith Crandall, information on

distribution of crayfish genera and species.
DB Denton Belk, summary of fairy shrimp

distribution patterns.
NC/RvS Neil Cumberlidge and R. von Sternberg.

report compiled for WCfvlC .

diversity for each taxon at continent level. In
the absence of global criteria for relative
importance, areas on different continents do not
represent strictly equivalent levels of diversity.

The rows of data in this list are sorted
first by continent, and secondly by the taxon
concerned.

For North American and African fishes, where source is
UNEP-WCMC, information has been extracted from
available literature, with additional data and advice for
Africa from Christian Leveque and Guy Teugels.

In most instances [MK, SK, MSG, NC/RvS, DB|
contributors indicated the approximate location of the
important areas concerned on a series of A3-5ized base
maps provided. These areas, and those identified from
literature by UNEP-WCfvlC. were digitized
for purposes of presentation [Maps 7.2. 7.3, 7.U\
and analysis.

APPENDIX 6 307

^rea name ^^^

^iUfe

Remarks ^^^^^^^^^^^^^B

^^BSRSSBST

■

L Tanganyika

Crabs

L, Tanganyika is Ihe only East African great lake where
endemic species of freshwater crabs occur: of the 9 species
and 2 genera present, 1 genus and 7 species are endemic.
iNC/RvSl

Lower Congo

Crabs

Diversity is marked in the Congo R. basin, but appears highest
in 2 areas, the lower parts oi the basin (including Congo,
Cabinda and DR Congo (former Zaire]] and the upper reaches
(including Rwanda/Burundi and parts of DR Congo]. [NC/RvS!

Madagascar

Crabs

4 genera and 10 species of freshwater crabs, all endemic,
occur in Ivladagascar [NC/RvSl

Niger-Gabon

Crabs

Southeast Nigeria, southern Cameroon and Gabon: 3 endemic
genera and more than 10 endemic species of freshwater
crabs'. [NC/RvS]

Upper Congo

Crabs

Diversity is marked in the Congo R. basin, but appears highest
in 2 areas, the lower parts of the basin (including Congo,
Cabinda and DR Congo (former Zaire]] and the upper reaches
(including Rwanda/Burundi and parts of DR Congo]. (NC/RvS]

Upper Guinea

Crabs

Upper Guinean rainforest, centered on Guinea, Sierra Leone,
Liberia, and western Cote d'ivoire (including Mount Nimbal: 2

endemic genera and 5 endemic species of gecarcinucids"'".

(NC/RvS]

' 7

Southern Africa

Fairy shrimp

2 endemic genera, 45 species, 38 endemic. South Africa
proper: 34 species, 22 endemic. [DB]

Cape rivers

Fishes

With 4 families and 33 species the fish fauna of southern Africa
IS rather poor in comparison with most other parts of the

continent; most species are cyprinids. However, there is
marked local endemism; most rivers in the southern Cape
region have 3 or 4 native endemics (several species are
threatened]. [UNEP-WCMC]

AFRICA

308 WORLD ATLAS OF BIODIVERSITY

AFRICA

9 Congo IZairel basin

10 Congo 'Cuvette
Centra le'

11 Congo rapids

12 Cross R.

13 L Barombi-Mbo

U L Bernnin

15 L Malawi

Fishes General region of very high richness; second only to the

Amazon basin in species richness. 25 families and 686 species
have been reliably reported from the Congo/Zaire basin, j

excluding L, Tanganyika and L. Moero*. Around 548 of the
species present lea 70%] are endemic to this basin. The basin
can be divided into i sections; Upper Lualaba, Cuvette
Centrale, Luapula-M»(eru and the rapids. [UNEP-WCMC]

Fishes High richness plus marked endemism. Around 690 species

occur in the Congo system; the Cuvette Centrale section
possibly has the highest species richness owing to the great
diversity of freshwater habitats available. lUNEP-WCMCl

Fishes High richness plus marked endemism. The rapids between

Kinshasa and the sea have a high concentration of fish species
(150 species!. 34 of which are endemic to this section. The
caves near Thysville are fed by the Congo system and support
one of Africa's few true hypogean fishes Caecobarbus geertsi.
Caecomastacembelus brichardi and Gymnanatlabes tihoni, not
strictly cave fishes, have been collected in the Stanley Pool in
riffles under flagstones or in crevices. [UNEP-WCMC!

Fishes Nigeria-Cameroon. 42 families, 166 species'. Very high species

diversity compared to the relatively modest catchment area,
and marked endemism. Transitional ichthyofauna between
the Nile-Sudan province and the Lower Guinea province.
(UNEP-WCMC!

Fishes This small (ca 4.5 km'! crater lake in Cameroon has 15 species

(plus another 2 present in the inflow stream, not the lake
proper!. At least 12 of the species are endemic, notably the 11
cichlids that form 1 of the 2 recorded 'species flocks' in West
Africa. 4 of the 5 cichlid genera are endemic: Konia, Myaka,
Pungu and Stomatepia. This very important site is at risk
from overfishing, the effects of introduced crustaceans and
fishes, siltation from local deforestation and water pollution.
(UNEP-WCMC!

Fishes A small (ca 0.5 km'! crater lake in southwest Cameroon with 2

non-endemic fishes and a remarkable species flock of 9
tilapiine cichlids. The cichlids are very small and not exploited;
they are at some risk because of the small distribution and
deforestation in the surrounding area. (UNEP-WCMC!

Fishes 30 800 km'. 1 2 families, more than 845 species, most of them

endemic to the lake. Rich species flocks among Cichlidae, and
a small species flock of Clariidae. (UNEP-WCMC!

APPENDIX 6 309

16 L. Tana

17 L.Tanganyika

18 L. Turkana

19 L.Victoria

20 Madagascar

21 Niger basin

22 NlemR.

Fishes The lish fauna olthis targe 13 150 kml lake includes 21

species in k families and is dominated by lake endemic
cyprinids. The large Sarbus cyprinids form 1 of 2 recorded
cypnnid species flocks Ithe other being that of L Lanao
in the Philippines, many species of which are severely
threatened!. (UNEP-WCMC]

Fishes 32 000 km'. In the lake itself, 16 families, around 250 cichlid

species and 72 non-cichlid species. Several species flocks
are present not only in Cichlidae, but also among Clariidae,
Bagridae, Mochokidae, Centropomidae. Ivlastacembelidae.

Fishes 6 750 km'. 51 species, 35 genera, 17 families. High family and

generic diversity, many of the species are lake endemic;
cyprinids form the most diverse family [UNEP-WCIvlCl

Fishes 68 800 km\ 12 families, around 545 species (many

undescribedl. High species diversity dominated by cichlids. The
majority of species are lake endemic. [UNEP-WCIvlCl

Fishes Around UO fish species have been recorded from the brackish

and freshwaters of Ivladagascar'; although species richness is
not remarkable, endemism is high. Two endemic families
IBedotiidae and Anchariidael have been recognized in
Madagascar, as well as 13 endemic genera and 43 endemic
species. Most endemic species are restricted to freshwater
habitats, mainly in eastern forested regions. About one quarter
of endemic species are known only from the type of locality.
Blind cave fishes have been described from Madagascar:
the gobiid Glossogobius ankaranensis and the elotrids
Typheleotns madagascarensis and T. pauliani. [UNEP-WCMCl

Fishes General region of high richness. 36 families, ca 243 species,

with 225 primary freshwater species'. Endemism moderate: 20
species endemic to Niger The basin includes 1 1 of the 13
primary freshwater families that are endemic to Africa. 164
primary freshwater fishes reported' from the Niger delta in
Nigeria, based on reference specimens for each species; the
high diversity 173% of the freshwater species in the entire
basinl in this area is seriously threatened by oil pollution.
[UNEP-WCMC)

Fishes Cameroon. High richness for area, plus marked endemism. 16

families. 94 species, 8 endemic. lUNEP-VCMCl

AFRICA

310 WORLD ATLAS OF BIODIVERSITY

AFRICA

23 Ogooue lOgowel R.

2i Sanaga R.

25 Upper Guinea rivers

26 Volta basin

27 L. Malawi

28 L. Tanganyika

29 L, Victoria

Fishes Gabon. High richness for area, plus marked endemism. 23

fan^ilies, 185 species, 48 species endemic (o Ogowe. A
relatively small drainage basin wiith a very high concentration
of species. Ivlany of the families represented are endemic to
Africa. Available data certainly underestimate actual diversity
(several new species are now being described, resulting from a
project of Tervuren Ivluseum, the American Museum of Natural
History and Cornell Universityl. (UNEP-WCMCl

Fishes Cameroon. 21 families; high concentration of species in a

small river basin; probably at least 135 land this figure is
believed to be a significant underestimate!'. Between 10 and 18
species endemic to the Sanaga. [UNEP-WCMC) 1

Fishes High richness for area, plus marked endemism. The Upper •

Guinea province includes coastal rivers from south of the ,'

Kogon R. in Guinea to Liberia, and has faunal affinities with the '
lower Guinea province and the Congo/Zaire. The fauna includes
many taxa endemic to the area"", fvlany small river basins,
many of Ihem still poorly investigated. Konkoure R. (Guinea!:
19 families, 85 species, at least 10 endemic. Kolente or Great
Scarcies R. (Gumea-Sierra Leone!: 19 families, 68 species.
Jong R. (Sierra Leonel: 20 families, 94 species. Saint-Paul R.
(Liberia!: 19 families, 76 species. Cess-Nipoue R. (Liberia-Cote
divoire!: 20 families, 61 species. (UNEP-WCMCl

Fishes General region of high richness. 27 families, about 139

species, 8 endemic to Volta basin. High species richness, with ,i
9 of the 13 African endemic primary freshwater fish families
represented". [UNEP-WCfvlC!

MolL Gastropods: 28 species, 16 endemic. Bivalves: 9 species, 1

endemic. (MSGI

fvlolL Gastropods: 68 species, 45 endemic. Bivalves: 1 5 species, 8

endemic. (MSGl

MolL Gastropods: 28 species. 13 endemic. Bivalves: 18 species, 9

endemic. (MSGl

APPENDIX 6 311

30 Lower Congo basin

Moll.

31 Madagascar

32 Western lowland forest
and VoKa basin

33 The Mollucas, New Guinea
and northerm Australia

34 Southeast Australia

35 Southwest Australia

36 Fly R., Papua New Guinea

37 Kikori R., L Kutubu,
Papua New Guinea

38 Kimberley District,
Western Australia

39 Aikwa llwakal R., Irian Jaya

40 Southeast Australia

41 Southwest

Western Australia

Moll

Moll.

Crabs

Crayfish

The region downstream of Kinshasa in Congo and OR Congo
(former Zairel. Gastropods: 96 species, 24 endemic. Endemic
gastropods are almost all prosobranchs; 5 endemic
'rheophilous' {specialized for life in the rapids] genera,
belonging to the Bithyniidae ICongodoma. Liminitesta] and
Assimineldae [Pseudogibbula. Septanellina, Valvatorbis].
Bivalves: no data. [MSGl

Gastropods: 30 species, 12 endemic. Genus Melanatha
endemic. Bivalves: no data. IMSGI

Upper Guinea region in Ghana, Cote d'lvoire. Sierra Leone,
Liberia, Guinea. Around 28 gastropod species of which 1 9
endemic (and 9 near-endemicl. Bivalves: no data. [MSGl

More than 30 species of freshwater crabs belonging to 5
genera, all in Parathelphusidae. [NC/RvS]

Large area of high richness and endemism, centered on
Victoria, 35 species, and Tasmania, 1 9 species. IKCl

Fairy shrimp 19 species, 12 endemic, [DB]
Fishes

Fishes

High species richness, 103 species in Fly proper, and high
local endemism, 12 endemics in system. [GAl

Headwaters of Kikori and Purari systems, with L. Kutubu.
High richness, 103 species and high endemism, 16 species in
Kikori: plus 14 species in L Kutubu. [GA]

14 endemic species (a density second only in Australia to
Tasmania and equal to southwest Western Australia], including
5 species within Prince Regent Reserve and 4 in the Drysdale
R. area; and 47 species in total. (GA] j

Near Timiki, Irian Jaya. High species richness: ca 78 species.
IGA]

I
1 1 endemic species occur in coastal southeast Australia, a

lower count per area than the other 3 areas cited here, and 42

species in total, [GA]

9 endemic species [i.e. density similar to the Kimberleysl, and
14 species intotaL [GA]

AFRICA

AUSTRALASIA

312 WORLD ATLAS OF BIODIVERSITY

AUSTRALASIA

EURASIA

kl Tasmania

43 Vogelkop. Irian Jaya

kk Great Artesian basin,
Australia

45 l^ew Caledonia

Fishes

Moll,

MolL

46 Western Tasmania, Australia MolL

47 Indonesia

Crabs

48 Myanmar-Malaysia

Crabs

49 South China

50 South India

Crabs

51 Sri Lanka

Crabs

12 endemic species, a greater number per area than anywhere
else in Australia, including 6 concentrated in the Central
Plateau area; and 24 species in total. [GA]

Moderate richness with high local endemism, ca 14 endemic
species, including Triton and Etna Bay lakes, IGAl

Springs and underground aquifers. Important area of
gastropod diversity. Bivalves: no data, [MSGl

Springs and underground aquifers. Gastropods: 81 species, 65
endemic. Bivalves: no data. [MSGl ,

Springs and underground aquifers. Important area of j

gastropod diversity Bivalves: no data. [MSGl

The area containing Sumatra. Java, Borneo, Sulawesi and the
southern Philippines has the greatest freshwater crab diversity
in Indo-Australia, with representatives of the Parathelphusidae
(10 genera and 71 speciesi and the Gecarcinucidae 15 genera
and 21 speciesi. [NC/RvSl

Northeast India lAssaml, Myanmar, Thailand, the Mekong
basin in southern Indochina, to the Malaysian peninsula and
Singapore. In this region there are an estimated 30 genera
and more than 100 species of freshwater crabs in 3 families,
the Potamidae, the Parathelphusidae and the
Gecarcinucidae'"'. [NC/RvSl

Only the Potamidae occur in China, but more than 160 species
and subspecies in 22 genera are present, most of which are
endemic. The southern provinces of China represent the
hotspot of biodiversity for this country""'". [NC/RvSl

The freshwater crabs of the Indian peninsula south of the
Ganges basin are all endemic to the subcontinent and belong
to 2 families, the Gecarcinucidae and the Parathelphusidae"".
The west coast of the peninsula and the south show most
diversity: an estimated 7 endemic genera and about 20 ■

endemic species in 2 families Ithe Parathelphusidae and
Gecarcinucidael. A third freshwater crab family, the Potamidae,
is found only in northern India but is not represented in the
Indian peninsula. [NC/RvS]

Sri Lanka has some 16 endemic species of freshwater crabs
belonging to 3 genera, 1 of which ISpirabthelphusa] is
endemic to the island'' ". iNC/RvS]

APPENDIX 6 313

52 Italy

53 Borneo highlands

54 Caspian Sea

55 Central Anatolia

56 Coastal peat swamps and
swamp forests of Malaysia.
Sumatra and Borneo

57 Coastal rainforest
of Southeast Asia

58 High Asia

Fairy shrimp 16 species, 7 endemic. IDBI

Fishes The fish fauna of the highlands of Borneo seems to be poor

in absolute number of species, but many of them have
developed specialization for hill-stream habitats and are
endemic to single basins. The area is still largely unsurveyed.
About 50 known endemic species, but actual figure might be
over 200" IMK]

Fishes Moderate species richness. Although many species are shared

with the Black Sea region, and/or the Aral basin, there is
marked endemism, including the monotypic lamprey
Caspiomyzon. ca 12 gobies, including monotypic genera Asra
and Analirostrum. also 3 Absa. [UNEP-WCMCl ,:

Fishes An and plateau with several endorheic lakes. About 20

endemic species, apparently underestimated by inadequate
taxonomy. Adjacent areas also have a number of endemics. In
urgent need of critical reassessment: probably one of the most
poorly known fish faunas in Eurasia. [MK]

Fishes Includes Bangka island. Extent along eastern coast of Borneo

not known. Probably formerly present on Java but apparently
cleared. About 100 endemic species in peat swamp forests, a
habitat type often restricted to a narrow fringe along the
coasts, still largely unsurveyed. Although peat swamps are
traditionally considered as a habitat with poor diversity, good
data for limited areas in Malay peninsula and Borneo indicate
that up to 50 species may be found within a small area (less
than 1 km'l, about half of them endemic and stenotypic. Most
species have small distribution ranges Isome possibly only a
fewkm'PMMK]

Fishes Thailand, Cambodia and southern Viet Nam. Southern extent

not known accurately This habitat is largely destroyed in
Thailand, and virtually unsurveyed in Cambodia and Viet Nam.
Endemic species expected in peat swamp forests"'". [MK]

Fishes Boundaries not known with accuracy; includes the Tibetan

plateau and probably parts of Chinese Turkestan. Distribution
and ecological data are sparse outside the Chinese literature.
About 150 known fish species, about half of Ihem endemic to
this area". Survey probably still superficial as a result of
difficulties of access. [MKl

EURASIA

3U WORLD ATLAS OF BIODIVERSITY

EURASIA

^^^^^^I^^^^^^^^^B

Karstic basins of Yunnan,

Fishes

Boundary not known with accuracy. About U known species of

Guizhou and Guangxi

cave fishes. Survey is still superficial and numerous additional
species are expected''. [MKl

L. Baikal, Siberia

Fishes

A species flock of 36 species of the family Cottidae Isculpins]
(including the endemic family Comephoridael, 4 'ecologically
differentiated stocks' Imany probably endemic species using
Western concepts! of Coregonus. 2 of Thymallus, 2 of Lota".
Endemic mollusks, gammarids, sponges and Baikal seal. [MKl

L. Biwa, Japan

Fishes

Reportedly 4 endemic species". [MKl ■1

L. El'gygytgyn. Siberia

Fishes

An old lake formed on the site of a meteorite crater 113 kml
Total fish diversity: 5 species, including an endemic genus and
species ISalvethymus svetovidovi], an endemic species
[Salvelinus elgyticus], and 1 species endemic to eastern
Siberia [Salvelinus boganidael. Endemic diatom species and
apparently endemic invertebrate[sl*''. IMKl

L Inle, Myanmar

Fishes

About 25 native fish species, ca 10 of them endemic, including
3 endemic genera"". [MK]

L Lindu, Sulawesi

Fishes

Very limited information. One native and endemic species; ,1
others might be expected". (MKl

L. Poso, Sulawesi

Fishes

1
10 native and endemic species, 2 endemic genera [both '

extinct?! and with L. Lindu comprises the entire known

distribution of the subfamily Adrianichthyinae"". Additional

species might still be expected, [MKl

L. Thingvalla,

Fishes

5 native hsh species, including 3 endemic Salvelinus [recent
summary in Kottelat"!. [MKl

Lakes of Isles

Fishes

A number of lakes host 1 or 2 species of British Salvelinus,
although information on individual lakes is usually inadequate.
At the beginning of the century up to U species were ^
recognized; although generally not accepted under later '
systematic concepts, recent work suggests that this figure may
be underestimated. Also at least 5 endemic Coregonus.
1 endemic Clupeidae and potential for endemic Salmo [recent
summary in Kottelat"!. [MK] .

APPENDIX 6 315

Lakes of Central Yunnan. Fishes

China

69 Lough Melvin, Ireland

70 Lower Danube

71 Mainland Southeast
Asian hills

Fishes

Lakes Dianchi. Fuxian, Er Hai, Yangling, Yangzong, Xingyun,
etc., have a distinctive fauna; despite the lakes being now in
different river basins IMekong. Yangtze. NanpangjiangI, they
have similar fauna, characterized by numerous endemic
species in the genera Cypnnus, Schizothorax, Anabantius and
Yunnanilus. Exact up-to-date figures of the number of species
are difficult to extract from the Chinese literature, but vne have
the following data: Oianchi: 25 native species. 1 1 endemic of
which apparently all but 2 are extinct. The lake basin has 2
other endemics'': Fuxian: 25 native species. 12 endemic plus 2
endemic shared only with Xingyun"; Er Hai: 17 native species,
9 endemic, several apparently extinct"; Yangzong has Ihad) at
least 2 endemics; Yangling and Xingyun at least 1 each. [MKl

Three endemic species of Salmo Irecent summary in
Kottelat'l [MK]

The lower Danube basin has a relatively richer fauna
(especially more diverse communities! than any other
European river. Endemics: about 6. possibly underestimated
(counted in Kottelat 1997-1, [MK(

Northern boundary not clear as published data on fish
distribution land ground surveys! in southern China are too
scanty Could be subdivided into a! upper Song Hong Imcludes
hills of Hainan and southern Nanpang Jiang!; bl Annamite
Cordillera; c! upper Mekong, ChaoPhraya and t^ae Khlong
basins; dl Salween, upper Irrawaddy and southeastern Assam
(including Tenassenm!. Recorded fish fauna estimated to be
over 1 000 species Iwith an estimated 200-500 species still
awaiting discovery!, 500 endemic to this area. Includes
ca 400 known species endemic to headwaters of individual
sub-basins. The fauna of the lower reaches of the main rivers
(excluded from this polygon! is richer (in terms of the number
of species that can be observed at a given locality! but most
have wide distributions crossing several river basins-'"'". IMK]

EURASIA

316 WORLD ATLAS OF BIODIVERSITY

EURASIA

72 Malili Lakes, Sulawesi Fishes

73 Maros karst, Sulawesi Fishes

74 Mindanao, Philippines Fishes

75 Northwest Mediterranean Fishes

drainage

76 Palawan, Philippines Fishes

Most important single site for aquatic biodiversity In Asia, A
complex of 5 lakes ITowuti, Matano, Mahalona, Wawontoa,
Masapil with endemic radiations of fishes of the families
Telmatherinidae 13 genera, 15 species, all but 1 endemici,
Hemiramphidae 13 endemic speciesi, Oryziidae 13 endemic
species!, Gobiidae lat least 8, all but 1 endemic], prawns lea 12
species?], crabs 14 species?], mollusks lea 60 endemic
species], etc. The distribution of the fishes is not uniform
within the lakes, all but 1 of the species of L. Matano are
endemic, while the others land 2 genera] are endemic to
Towuti. Mahalona, Wawontoa. Masapi has not yet been
surveyed. Only 2 species of the Telmatherinidae are known
outside this area. [MK]

One endemic genus ]possibly an artifact of limited collection;
more surveys might show it to be present outside this area]
and about 6 endemic species, including a cave species" [MKl

About 30 endemic species of cyprinid fishes, including about
18 endemic species of Puntius in L. Lanao lall but 2 or 3
reportedly extinct]"" Cyprinids are fishes which live only in
freshwater and cannot disperse in marine environments;
several other families also occur in the island's freshwaters,
but all are able to disperse through the seas. [MK]

Includes Spam, Portugal, southern France and northern Italy.
The total diversity in the whole area is quite low, the
communities are quite poor, but this area holds 55 endemics,
many with small distribution ranges. 3 of the Rhone endemics
extend almost to the northern extremity of the basin.
Endemics;! Petromyzonidae, 1 Acipenseridae, 1 Clupeidae.
34 Cyprinidae. 5 Cobitidae. 6 Salmonidae, 1 Valenciidae,
1 Cyprinodontidae, 2 Cottidae. 1 Percidae and 4 Gobiidae
(counted in Kottelafl. [MK]

About 10 recorded species of cyprinid fishes, actual figure
probably higher [MKl

APPENDIX 6 317

77 Southwest Balkans

Fishes

78 Southwest Sri Lanka

79 Subalpine lakes

Fishes

Sundaic foothills
and floodplains

81 Western Ghats, India

Fishes

82 Balkans region
{Former Yugoslavia,
Austria, Bulgaria, Greece!

83 ChilkaL

84 L. Baikal

Moll.

Moll

The total diversity in the whole area is quite low, the
communities are quite poor, but the area holds 84 endemics,
most of them with restricted or very restricted distribution
ranges: 1 Petromyzonidae. 2 Clupeidae, 48 Cyprinidae, 8
Cobitidae, 1 Balitondae, 1 Siluridae, ISSalmonidae, 1
Valenciidae, 1 Gasterosteidae, 1 Percidae and 7 Gobiidae.
The systematics of many groups is still very poorly known
and more species will be recognized or even discovered in
the future (possibly 10-201. Noteworthy are L. Ohrid with
apparently 4 endemic Salmo. L. Prespa with apparently
7 endennic species and the Vardar basin with at least 8
endemic species". [MK]

28 of the 91 native fish species of Sri Lanka are endemic to
this area. Several of the species traditionally given the same
name as Indian species are being revised and turn out to be
specifically distinct, so that the figure will rise'"". [MKl

Stretches from L. Bourget in the west to Traunsee in the east.
Numerous endemic Coregonus (possibly 27, several already
extinctl, at least 2 endemic Salvelinus and possibly some
endemic Salmo. Some lakes have more complex communities,
e.g. L. Konstanz with 4 Coregonus, 2 Salvelinus. 1 Salmo and
several other species (recent summary in Kottelaf. ||v1K]

About 400 known species. Most of the floodplain species are
widely distributed over the whole area, while those of foothill
streams have more localized distributions and are of greater
interest in terms of endemicity. Northern limit: Tapi basin in
peninsular Thailand '"'". (MK(

About 100 endemic fish species (estimated from Talwar and
Jhingran*'; Pethiyagoda and Kottelat'^l. Difficult to give
accurate figures. Many wide-ranging 'species' of fishes in
South Asia are in fact complexes of species, so that the actual
number of species is likely to increase significantly after
adequate systematic revision. [MK(

Springs and underground aquifers. Gastropods: ca 190 species,
some 180 endemic. Bivalves: no data. (MSGj

Brackish water Gastropods: 28 species, ca 1 1 endemic.
Bivalves: 43 species, 25 endemic. (MSGl

Gastropods: 147 species, 114 endemic. Bivalves: 13 species.
3 endemic. (MSGl

EURASIA

318 WORLD ATLAS OF BIODIVERSITY

EURASIA

NORTH AMERICA

85 L. Biwa

L, Inle

Moll,

Moll.

87 L, Ohrid and Ohnd basin Moll.

88 L Poso and Malili Moll.
Lakes system

89 Lower Mekong R., Thailand, Moll.
Laos, Cambodia

90 North Western Ghats

Moll.

91 Zrmanja R., Croatia

92 Southeast USA

93 Western USA

94 Bear L.

Moll.

Crayfish

Gastropods: 38 species, 19 endemic. Bivalves: 16 species,
9 endemic. [MSGl

Gastropods: 25 species, 9 endemic. Bivalves: 4 species,
2 endemic. IMSG]

Gastropods: 72 species, 55 endemic. Bivalves: no data. [MSG]

Sulawesi. Gastropods: ca 50 species, ca 40 endemic.
Bivalves: 5 species. 2 endemic. [MSG]

River habitat. Only ca 500 km of the lower

Mekong mam course (with the tributary Mun R.I

has been well studied. Gastropods: 121 species, 111 endemic.

Two rissoacean groups dominate this entirely prosobranch

assemblage of 120 plus species, the pomatiopsid Triculinae 192

endemic species, 11 endemic general and the Stenothyridae

119 endemic species). Bivalves: 39 species, 5 endemic. [MSG]

River habitat. Gastropods: ca 60 species, 10 endemic. 2
endemic genera Turbinicota, Cremnoconchus. The succineid
genus Lithotis is known from 2 species: L. tumida not collected
since its description in 1870. and L wpicola only known from a
single locaUty. The highly localized genus Cremnoconchus is
the only littorinid living in a freshwater/terrestrial environment.
Bivalves: 1 1 species, 5 endemic. [MSGl

Gastropods: all are hydrobioid snails, 11 species, 5 endemic.
Bivalves: no data. [MSG]

Large area of high richness and endemism at generic and
species level; including the eastern and southern Mississippi
drainage (Ohio R., Tennessee R., to Ozark and Ouachita
mountains!; 72 species in Alabama, 71 in Tennessee. [KC]

Fairy shrimp 26 species, 13 endemic. [DB]

I
Fishes This lake is part of the Bonneville R. basin and contains 1

local endemic \Prosopium gemmiferum] and 2 species that are
now restricted to this site [Prosopium spibnotus and
P. 3bys5icot3l-\ [UNEP-WCMC]

APPENDIX 6 319

95 Colorado basin

Cumberland Plateau
(Cumberland and
Tennessee rivers!

97 Death Valley region

Eastern USA

The largest basin of the western USA, this has high species
richness and endemism, including 5 endemic genera of which
only Plagopterus is monotypic''. About one third of the
ichthyofauna of the Colorado is threatened, endangered or
extinct due to dams and introduced species*'. lUNEP-WCt^Cl

Fishes This area has the highest species richness and local

endemism in North America. It is part of the highly diverse
Ivlississippi basin, vjith ca 240 species in total. 160 present in
both the Tennessee and Cumberland drainages, U endemic to
the 2 basins. Of these U. 10 are darters, 3 are minnowis and 1
IS a topminnow. The Tennessee has the greatest species
diversity with 224 species including 25 endemics las well as 64
not found in the Cumberland). The Cumberland has 176 native
species, including 9 endemics and 16 species not shared with
the Tennessee"'. [UNEP-WCMCl

Fishes There is a high level of local endemism associated with the

dispersed pattern of springs and marshes. 4 families are
present ICypnnidae. Catostomidae. Cyprinodontidae and
Goodeidael with 9 species including an endemic species of
Catostomidae''. Several are globally threatened, including
2 of the 5 Cypnnodon species ICypnnodon radiosus and j

C, disbolisi lUNEP-WCMC) |

Fishes This is a general area of high species richness and endemism

which, with the possible exception of the incompletely known
East Asian fish species, represents the most diverse of all the
freshwater faunas of the temperate zone". This includes a) the
Ozark Plateau, bl the Ouachita Mountains, cl the South Atlantic
Central Plain and dl the Tennessee-Cumberland Plateau.
[UNEP-WCMCl

Klamath-upper Sacramento Fishes

100 Ouachita Mountains

The Klamath R. basin contains 28 species in total with
relatively high endemism. The 6 endemic species include 2
Catostomus. 1 Chasmistes and 1 Gila^. The ichthyofauna of the
Sacramento differs from that of the Klamath and contains 4
genera that are confined to this river and a few neighboring
drainages'^IUNEP-WCMCl

Fishes This area includes parts of the lower Red and Ouachita rivers,

each containing 133 species. The Ouachita and the Red river
systems both contain 18 endemic species^. lUNEP-WCMC]

NORTH AMERICA

320 WORLD ATLAS OF BIODIVERSITY

NORTH AMERICA

^pm^^^m^mm^^^^^g^

101

Ozark Plateau

Fishes

The Ozark Plateau is an area of high species diversity and
particularly high local endemism in the southeast USA; it
represents a concentration of the species-rich southwestern
Mississippi drainage (more than 30 endemic fish speciesP.
[UNEP-WCMCl

102

Rio Grande-Pecos confluence

Fishes

The Rio Grande basin overall has more than 60 endemic
species" and the Pecos, a tributary, has 5 " Many of the '
endemics occur at the confluence of the 2 rivers, and many
are globally threatened. [UNEP-WCMCl

103

Southern Oregon-California
rivers

Fishes

These rivers share few family similarities with the eastern
USA and have about 25% of the number of species, but the
region is high in local endemism. [UNEP-WCMC]

Southern Atlantic
coastal plain

Fishes

This includes the Alabama-Tombigbee basin with a species-
rich fauna, including about iO endemic taxa" This region
also contains the Pearl R., with 106 species" and the species-
rich lower Mississippi. lUNEP-WCMC]

105

And/semi-arid western USA

Moll.

Springs and underground aquifers. Gastropods: all are
hydrobioid snails, ca 100 species, at least 58 endemic.
Great radiation in genus Pyrgubpsis. 3 extinct species, and
all others are candidates for listing by US Fish and Wildlife
Sereice. Bivalves: no data. [MSGl

106

Cuatro Cienegas basin,
Mexico

Moll.

Springs and underground aquifers. Gastropods: all are
hydrobiids; 12 species, more than 9 endemic. 5 genera
[Nymphophilus. Coabuilix, Paludiscala, Mexithauma,
Mexipyrgus] are endemic to this small area of 30 x 40 km.
Bivalves: no data. [d^SGl

107

Florida, USA

Moll.

Springs and underground aquifers. Gastropods: mostly
hydrobiid snails. 84 species, ca 43 endemic. No bivalves. [MSGl

108

Mobile Bay basin

Moll.

Tombigbee-Alabama rivers. River habitat. Gastropods: 118
species, 1 10 endemic; 6 endemic genera; greatest species
richness (76 speciesl in Pleurocercidae. 38 of the gastropod
species believed extinct. 70 candidates for listing by US Fish
and Wildlife Service. Bivalves: 74 species, 40 endemic.
25 extinct. [MSGl

109

Ohio-Tennessee rivers

Moll.

Eastern Mississippi drainage. River habitat. High species
richness and endemism. [MSG]

APPENDIX 6 321

10 Central America

111 South America

112 Southern South America

113 Altiplano of the Andes

lU Amazon R, basin

115 Aripuana R.. a tributary
of the t^adeira

Crabs The freshwater crabs of Central America belong to exclusively

neotropical families, Pseudothelphusidae and Trichodactylidae.
Central America from Ivlexico to Panama, including some of
the Caribbean islands, holds at least 22 genera and over
80 species of pseudothelphusid crabs, and 4 genera and
ca 10 species of trichodactylids. The Isthmus of Tehuantepec in
central Mexico is a hotspot of biodiversity for freshwater crabs
in Central America'"", and richness declines toward to the
south and north. The 7 species of freshwater crab belonging to
1 genus found in Cuba are all endemic to that island"
iNC/RvSl

Crabs 2 freshwater crab families iPseudothelphusidae,

Trichodactylidael endemic to the neotropics occur here.
Freshwater crabs do not extend to southern Chile or
southern Argentina, There are an estimated 17 genera and
over 90 species of pseudothelphusids, found mainly in the
highland regions of Peru, Ecuador. Colombia, Venezuela and
the Guianas, and on the islands of the southern Caribbean,
and 12 genera and over 40 species of trichodactylids in the
Amazon basin. The Cordilleras of Colombia"", coastal
Venezuela and the Guianas'"^ and the highland areas of
Ecuador and Peru" are all diversity hotspots for freshwater
crabs. The Amazon basin is rich in species*'", but most are
widespread in the basin, and it is not possible yet to delimit
special areas, [NC/RvSl

Fairy shrimp 18 species, 14 endemic. (DBl

Fishes Species flock of Orestias with 43 or more species, representing

an endemic subfamily. Orestiinae. of the Cyprinodontidae. [SKI

Fishes The Amazon Kvith adjacent TocantinsI basin probably has

about 3 000 species, and is one gigantic hotspot. The Amazon
fauna eguals or exceeds other continental faunas in species
richness. Endemism in tributaries and subtributaries makes
up most of the overall diversity, rather than the main Amazon
itself. Only a few of the constituent rivers have been studied in
any detail, [SK]

Fishes Known to have a highly endemic but still little-studied fauna

upstream of the lowermost falls, with at least 10 endemic
species, some restricted to rapids. [SKI

CENTRAL AND
SOUTH AMERICA

322 WORLD ATLAS OF BIODIVERSITY

CENTRAL AND
SOUTH AMERICA

116 Central America
between the Isttimus of
Tetiuantepec and

the Isthmus of Panama

117 Iguacu R.

Fishes

La Plata basin: Uruguay, Fishes

Paraguay and Parana rivers

119 L Titicaca and smaller lakes Fishes
of the Altiplano extending
from Chile to Peru

280 freshwater hsh species, all endemic. [SKI

On the border between Argentina and Brazil, tributary to the
Parana R. Its fish fauna is separated from the Parana by the
Iguacu falls, which do not permit any migration; highly
endemic, with ca 50 endemic species out of a total of 65
species lea 80%l. There are considerable difficulties with the
nomenclature and systematic status of the Iguacu fish species,
most belonging to groups that have never been revised.
Nonetheless, endemism will probably remain above 50%. The
endemic fauna, mainly a running water one, is highly
endangered by hydroelectric power projects, pollution and
Introduced species. The fauna is not protected. [SK]

fiiarked by numerous waterfalls providing isolation.
Mainly endemic species, including numerous local endemics.
Number of species unknown, but estimated fewer than 1 000,
possibly ca 600. The tributaries of the Parana down to about
Encarnacion have a very high number of local endemics,
often restricted to a single river, mostly separated from the
Parana by one or more waterfalls near the mouth. Many of
these have not been described or examined by a specialist,
but are known only from occasional collections made before
the Italpu. Acaray and Yacyreta dams were constructed.
Unfortunately, environmental impact assessment for those
dams did not result in any significant collections to show what
species were in the area before the dams were built. A lesser
collection of pre-dam fishes is available In Museo Naclonal de
HIstorIa Natural del Paraguay and the Museum d'histoire
naturelle de Geneve, ISKl

These lakes hold a large number of species of the genus
Orestias ICyprinodontidael, 23 are endemic to Titicaca.
The genus, with a total of 43 species, has a narrow range from
northern Chile to southern Peru. The lake species flocks may
not be monophyletic", but the group certainly attained its
present species richness in the area. The sister group is the
North American Cyprinodontidae. Other highland Andean fish
families include the Astroblepidae, ranging from Bolivia across
Peru and Ecuador into Colombia, and many trichomycterid
fishes ITrichomycterldae] occur L. Titicaca with its Orestias
fauna Is the only identifiable hotspot. ISKl

APPENDIX 6 323

^l^g^j^^m^^jj^^^^

120

Marowijne/Maroni R. drainage,

Fishes

Known to have many endemic species above the falls, with

Guyana and Suriname

the same genera as in the rest of the Guianas area. [SKI

121

Mata Atlantica

Fishes

Numerous endemic species in small mountain streams or in
the few major river systems, most incompletely known. The
Ribeira R, has 77 species, and similar numbers appear to be
in the other rivers. The Jequitinhonha is notable for several
endemic species, including 1 of Rbamdia, which is othenwise
represented by only a few widespread species in South
America. The Mata Atlantica fauna extends to eastern Uruguay
and southeastern Paraguay as numerous fragmented habitat
patches, and although not high in species richness Iperhaps
ca 1501. has a large number of locally restricted species, with
related species replacing each other from one river to another.

■

ISKl

122

Mazaruni and Potaro rivers,

Fishes

Separated from the rest of the Essequibo system by falls, with

Guyana highlands

several endemic species, but little explored. [SK]

123

Mesa Central, Mexico

Fishes

Endemic subfamily Goodeinae of family Goodeidae with about

36 species. [SK)

124

Mexican Plateau

Fishes

125

Negro R. and upper

Fishes

At least 700 species, probably nearer to 1 000. many of which

Orinoco R., Brazil, Colombia

are endemic to the clear and black waters distinguishing

and Venezuela

the basin. [SK]

126

Nicaraguan lakes

Fishes

The Nicaraguan great lakes [Nicaragua and Managua! in the
San Juan basin do not have great numbers of species (about
16 cichlidsl. but endemism is high [9 endemic species.
2 endemic general. [SK]

Orinoco R. basin

Fishes

More than 1 000 species, most of which may be endemic.
There is much local endemism as habitats vary considerably,
including lowland inundation savannahs, fast-flowing mountain
rivers, etc. Includes thus different biogeographic regions. [SK]

128

Oyapock R., Brazil and

Fishes

Known to have many endemic species, especially rheophilic.

French Guiana

from the lowermost falls upstream. Still little studied. [SK]

129

Pacific coast of Colombia

Fishes

Although a high-rainfall region there are few large rivers.

and Ecuador

The fauna is poor, but species are highly endemic to the region
and to particular rivers or portions of rivers. In particular, the
Baudo R. and San Juan R. in Colombia seem to have
numerous endemics. Possibly the Atrato R. should be
included. KK^^^^^^ ^^

CENTRAL AND
SOUTH AMERICA

324 WORLD ATLAS OF BIODIVERSITY

CENTRAL AND
SOUTH AMERICA

130 Patagonia lArgentma and
Chile, from around
the R. Negro southward -
except the most and areas]

Fishes

131 Panuco R. basin

132 Upper Uruguay R.

Fishes

133 Western Amazonia

Fishes

13^1 L, Titicaca

MolL

Low diversity but endemic relict fauna of more general
southern hemisphere type, with families such as Geotriidae
and Galaxiidae, and also the endemic catfish family
Diplomystidae with 6 species, the monotypic catfish family
Nematogenyidae. and k species of the percoid family
Percicihtyidae. This is not a hotspot of species richness,
but a region of considerable local endemism, and a fauna
completely different from that of the rest of South America.
The Nematogenyidae are related to the Loricariidae and
Trichomycteridae of northern South America (the Brazilian
faunal. but the Diplomystidae are the most primitive living
catfish family The scaleless characid Gymnocharacinus bergii
represents the Brazilian fauna, but lives isolated in one
Patagonian locality on the Sumuncura Mountain which
maintains about 22.5°C water temperature year-round. [SKI

A small drainage with ca 75 species, ca 30% endemic,
including several closely related species of Henchthys
ICichlidael, Eastern Mexico, ca 25 endemic of ca 75 known
species. ISKl

The river is relatively well known from collections made by
teams of the Museu de Zoologia of the Pontificia Universidade
Catolica do Rio Grande do Sul during environmental impact
assessment in the area, which is destined for numerous
hydroelectric power plants. The collections concern the
middle and upper portions, located in Brazil. More than 130
species of fish are recorded from the middle and upper
Uruguay, and the number is likely to rise to over 150 at least.
About half of those may be endemic. Lucena and Kullander"
described 1 1 species of Crenicichla from the Uruguay R.. and
noted that this is double the number of a similar Amazonian
river. 6 of the species form a species flock originating on site
and diversifying by trophic adaptation similar to cichlids of
East African lakes. The lower Uruguay R., along the
Argentinian-Uruguayan border, is very little studied and
may have fewer endemics, (SK]

Lowland Amazonian Peru. Ecuador and Colombia, and parts of
Brazil, representing a large expanse of lowland Amazonia, very
rich in species, but not well studied. Work in Peru and Ecuador
suggest that there may be at least 1 000 species in the area.
and at least half may be endemic. [SK]

Gastropods: 24 species. 15 endemic. Bivalves: no data. IMSGI

APPENDIX 6 325

HH^ipH^mi^^^m^^B

135 Lower Uruguay R and

Moll

Gastropods; 5i species, 26 endemic. Bivalves: 39 species,

Rio de la Plata, Argentina,

8 endemic. IMSGI

Uruguay. Brazil

136 Parana R.

Moll.

More than 7 species, 7 endemic, of vnhich 3 are extinct in the

3KS0II

wild. Bivalves: no data, IMSGI

REFERENCES

1 CumberUdqe, N. ^999. The freshwater crabs of West Africa. Family Potamonautidae. Faune
et Flore Tropicales 36. Institut de recherche pour le developpement IRD lex ORSTOM), Paris.

2 Cumberlidge, N. 1996. A taxonomic revision of freshwater crabs IBrachyura. Potamoidea.
Gecarcinucidaej from the Upper Guinea forest of West Africa. Crustaceana 69(61: 681-695.

3 Cunnberlidge, N. 1996. On the Globonautinae Bott, 1969, freshwater crabs from West Africa
IBrachyura: Potamoidea: Gecarcinucidael. Crustaceana 69171: 809-820.

A Tijrl<ay. M. and Cumberlidge, N. 1 998. Identification of freshwater crabs from Mount Nimba,
West Africa IBrachyura: Potamoidea: Potamonautidae). Senclfenbergiana biologica 77121.

5 Teugels, G.G. and Guegan, J.-F. 1994. Diversite biologique des poissons d'eaux douces de la
Basse Guinee et de I'Afrique Centrale. In: Paradi, Teugels, G.G. and Guegan, J.-F. ledsl
Biological diversity of African fresh-and brackish water. Geographical overviews. Symposium,
Ann. Mus. Roy. Afr. Centr., Zooi 275: 67-85.

6 Teugels, G.G., Reid, G. McG. and King, R.P. 1992. Fishes of the Cross river basin ICameroon-
Nigerial: Taxonomy, zoogeography, ecology and conservation. Ann. Mus. Roy. Afr. Centr.,
Zooi. 266.

7 Stiassny, M.L.J, and Raminosoa, N. 1994 The fishes of the inland waters of Madagascar. In:
Teugels, G.G., Guegan, J.-F. and Albaret, J. -J. ledsl Diversite biologique des poissons des
eaux douces et saumatres d'Afrique. Syntheses geographiques. Ann. Mus. Roy. Afr. Centr.,
Zooi 2Tb: 133-U9.

8 Teugels, G.G. and Powell, C.B. 1993. Biodiversity of fishes in the eastern Niger delta, Nigeria,
and species distribution in relation to white and blackwater rivers. Abstracts. Symposium
International sur la diversite biologique des poissons d'eaux douces et saumatres d'Afrique,
Dakar, 15-20 November 1993, p. 62. Plus unpublished data.

9 Leveque, C, Paugy, D. and Teugels, G.G. ledsl 1990. Faune des poissons d'eaux douces et
saumatres de lAfrique de I'Ouest. Tome I, pp. 1-384. Collection Faune Tropicale XXVIII,
ORSTOM, Paris/MRAC, Tervuren.

10 Leveque, C, Paugy, D. and Teugels, G.G. ledsl 1992. Faune des poissons d'eaux douces et
saumatres de I'Afrique de I'Ouest. Tome il, pp. 385-902. Collection Faune Tropicale XXVIII,
ORSTOM, Pans/MRAC, Tervuren.

11 Leveque, C. 1997. Biodiversity, dynamics and conservation. The freshwater fish of tropical
Africa. ORSTOM. Cambridge University Press. Cambridge.

12 Alcock, A. 1910. Catalogue of the Indian decapod Crustacea in the collection of the Indian
Museum. Part I. Brachyura. Fasciculus II. The Indian fresh-water crabs - Potamonidae.
Calcutta.

13 Bott, R. 1970. Betrachtungen uber die Entwicklungsgeschichte und Verbreitung der
SiJOwasser-Krabben nach der Sammlung des Naturhistorischen Museums in Genf/Schweiz.
Revue Suisse de Zoologie 77121: 327-344.

CENTRAL AND
SOUTH AMERICA

326 WORLD ATLAS OF BIODIVERSITY

U Ng, P.K.L. 1988. The freshwater crabs of Peninsular Malaysia and Singapore. Department of
Zoology, University of Singapore, Shing Lee Publishers Pte. Ltd, Singapore.

15 Ng, P.K.L. and Naiyanetr, P. 1993. New and recently described fresfiwater crabs (Crustacea:
Decapoda: Brachyura: Potamidae, Gecarcmucidae and Parathelphusidael from Thailand.
Zoologische VerhandeUngen 28^: 1-117.

16 Ng, RK.L. and Dudgeon, D. 1991. The Potamidae and Parathelphusidae (Crustacea:
Decapoda: Brachyura! of Hong Kong. Invertebrate Taxonomy 6: 741-768.

17 Dai, Zhou and Peng 1995. Eight new species of the genus Sinopotamon from Jiangxi
Province, China (Crustacea, Decapoda, Brachyura, Potamidae). 6eauforf/a 45(51: 61-76.

18 Turkay, M. and Dai, A.Y. 1997. Review of the Chinese freshwater crabs previously placed in
the genus Malayopotamon Bott, 1968 (Crustacea: Decapoda: Brachyura: Potamidael. Raffles
Bulletin of Zoology A5[2]: 189-207.

19 Dai, A.Y. 1997. A revision of freshwater crabs of the genus Nanhaipotamon Bott, 1968 from
China (Crustacea: Decapoda: Brachyura: Potamidael. Raffles Bulletin of Zoology A5[2\: 209-
235.

20 Dai, A.Y. and Turkay, M. 1997. Revision of the Chinese freshwater crabs previously placed in
the genus Isolapotamon Bott, 1968 (Crustacea: Decapoda: Brachyura: Potamidael. Raffles
Bulletin of Zoology A5[2]: 237-264.

21 Bott, R. 1970. Die SuOwasserkrabben von Ceylon. Arkiv fiir Zootogie 22: 627-640.

22 Ng, P.K.L. 1995. A revision of the Sri Lankan montane crabs of the genus Perbnnckia Bott,
1969 (Crustacea: Decapoda: Brachyura: Parathelphusidael. Journal of South Asian Natural
History M:]: 129-174.

23 Kottelat, M. et at. 1993. Freshwater fishes of western Indonesia and Sulawesi. Periplus
Editions, Hong Kong, with some updating.

24 Ng, P.K.L. 1994. Peat swamp fishes of Southeast Asia - Diversity under threat. Wallaceana
73: 1-5.

25 Ng, RK.L., Tay, J.B. and Lim, K.K.R 1994. Diversity and conservation of blackwater fishes in
Peninsular Malaysia, particularly in the North Selangor peat swamp forest. Hydrobiologia
285:203-218.

26 Kottelat, M. and Lim, K.K.P. 1 995. Freshwater fishes of Sarawak and Brunei Darussalam: A
preliminary annotated check-list. Sarawak Mus. J. 48(691: 227-256.

27 Kottelat, M. 1985. Fresh-water fishes of Kampuchea. Hydrobiologia 121: 249-279.

28 Kottelat, M. 1989. Zoogeography of the fishes from Indochinese inland waters with an
annotated check-list. Bull. Zool. Mus. Univ. Amsterdam 12(1): 1-54.

29 Wu, Y-R and Wu, C.-Z. 1992. [The fishes of the Qinghai-Xizang Plateau]. Sichuan Publishing
House of Science and Technology, Chengdu [Chinese, English summary).

30 Chen, Y-R. and Yang, J,-X. 1993. A synopsis of cavefishes from China. Proceedings of the XI
International Congress of Speleology, pp. 121-122, Beijing, updated.

31 Smith, G.R. and Todd, T.N. 1984. Evolution of species flocks of fishes in north temperate
lakes. In: Echelle, A.A. and Kornfield, i.L. (eds). Evolution offish species flocks, pp. 45-68.
University of Maine at Orono Press, Orono.

32 Masuda, H. et al. 1984. The fishes of the Japanese archipelago. 2 vols. Tokai University
Press, Tokyo,

33 Chereshnev, LA. 1992. Rare, endemic, and endangered freshwater fishes of Northeast Asia.
J. Ichthyol. 3218]: 110-124.

34 Chereshnev, LA. and Skopets, M.B. 1990. Salvethymus svetovidovi gen. et sp. nova - a new
endemic fish of the subfamily Salmoninae from Lake El'gygytgyn (Central Chukotkal. J.
/c/it/iyo/. 30(21:87-103.

35 Annandale, N. 1918. Fish and fisheries of the Inle Lake. Rec. Ind. Mus. 14: 33-64.

36 Kottelat, M. 1986. Die Fischfauna des Inle-Sees in Burma. Aquar Terrar. Ztschr 39(9): 403-
406, (101:452-453.

APPENDIX 6 327

37 Kottelat. M. 1990. Synopsis of the endangered Buntingi (Ostelchthyes: Adnanichthyidae and
Oryziidael of Lake Poso, Central Sulawesi, Indonesia, witfi a new reproductive guild and
description of three new species. Ichthyol. Exptor. Freshwat. 1111: '19-67.

38 Kottelat, M. 1990. The ricefishes lOryziidael of the Malili Lakes, Sulawesi, Indonesia, with
description of a new species. Ichthyol. Exptor. Freshwat. 1121: 151-166.

39 Kottelat, M. 1990. Sailfin silversides (Pisces: Telmatherinidael of Lakes Towuti, Mahalona
and Wawontoa (Sulawesi, Indonesia) with descriptions of two new genera and two new
species. Ichthyol. Explor. Freshwat. 1(3): 227-2^6.

40 Kottelat, M. 1991. Sailfin silversides (Pisces: Telmatherinidael of Lake Matano, Sulawesi,
Indonesia, with descriptions of six new species. Ichthyol. Explor. Freshwat. 1(4|: 321-344.

41 Kottelat, M. 1997. European freshwater fishes. An heuristic checklist of the freshwater fishes
of Europe (exclusive of fornner USSRl. with an introduction for non-systematists and
comments on nomenclature and conservation. Biologia, Bratislava, Sect. Zool. 52 (suppl. 51:
1-271.

42 Kottelat, M. and Chu, X.-L. 1988. Revision of Kunnan/7us with descriptions of a miniature
species flock and six new species from China (Cyprlnlformes: Homalopteridae]. Env. Biol.
F/s/i. 23(1-21: 65-93, updated.

43 Yang, J.-X. and Chen, Y.-R. 1995. [The biology and resource utilization of the fishes of Fuxian
Lake, Yunnan]. Yunnan Science and Technology Press, Kunming [Chinese, English
summary].

44 Li, S.-Z. 1982. [Fish fauna and its differentiation in the upland lakes of Yunnan). Acta Zool.
Sinica 28(21: 169-176, updated [Chinese, English abstract].

45 Kottelat, M. 1990. Indochinese nemacheilines. A revision of nemacheiline loaches IPisces:
CypriniformesI of Thailand, Burma, Laos, Cambodia and southern Viet Nam. Pfell, Munich.

46 Kottelat, M. 1998. Fishes of the Nam Theun and Xe Bangfai basins ILaosj, with diagnoses of
a new genus and twenty new species (Cyprinidae, Balltorldae, Cobltldae, Coiidae and
Eleotrldidae). Ichthyol. Explor Freshwat. 9(1]: 1-128.

47 Myers, G.S. 1960. The endemic fish fauna of Lake Lanao, and the evolution of higher
taxonomic categories. Evolution 14: 323-333.

48 Kornfleld, I.L. and Carpenter, K.E. 1984. The cyprinids of Lake Lanao, Philippines:
Taxonomic validity, evolutionary rates and speciatlon scenarios. In: Echelle, A. A. and
Kornfleld, I.L. (edsl. Evolution of fish species flocks, pp. 69-84. University of Maine at Orono
Press, Orono.

49 Pethiyagoda, R. 1991. Freshwater fishes of Sri Lanka. Wildlife Heritage Trust of Sri Lanka,
Colombo.

50 Pethiyagoda, R. 1994. Threats to the indigenous freshwater fishes of Sri Lanka and remarks
on their conservation. Hydrobiologia 28b: 189-201.

51 Kottelat, M. 1995. The fishes of the Mahakam River, East Borneo: An example of the
limitations of zoogeographic analyses and the need for extensive fish surveys In Indonesia.
Trop. Biodiv. 2(31:401-426.

52 Talwar, P.K. and Jhlngran, A.G. 1991. Inland fishes of India and adjacent countries. 2 vols,
Oxford and IBH Publishing Company, New Delhi, updated.

53 Pethiyagoda, R. and Kottelat, M. 1994. Three new species of fishes of the genera
Osteochilichthys (Cyprinldael, Travancoria (Balitoridael and Horabagrus (Bagrldae) from the
Chalakudy River, Kerala, India. J. South Asian Nat. Hist. 1(1]: 97-116.

54 Minckley et ai 1986. In: Hocutt, C.H. and Wiley, E.O. (eds]. The zoogeography of North
American freshwater fishes. John Wiley and Sons.

55 Banarescu. P. 1991. Zoogeography of freshwaters. 1. General distribution and dispersal of
freshwater animals. Aula Verlag, Wiesbaden.

56 Carlson, C.A. and Muth, R.T 1989. In: Dodge, D.P (ed.] Proceedings of the International
Large River Symposium. Can. Spec. Publ. Fish. Aquat. Sci. 106.

328 WORLD ATLAS OF BIODIVERSITY

57 Starnes and Etnier. 1986. In: Hocutt, C.H. and Wiley, E,0. ledsl. The zoogeography of North
American freshwater fishes. John Wiley and Sons.

58 Cross, F.B. et at. 1986. In: Hocutt, C.H. and Wiley, E.G. (edsl. The zoogeography of North
American freshwater fishes. John Wiley and Sons.

59 Smith, M.L. and Rush Miller, R. 1986. In: Hocutt, C.H. and Wiley, E.G. (eds). The
zoogeography of North American freshwater fishes. John Wiley and Sons.

60 Swift, CO. etal. 1986. In: Hocutt, C.H. and Wiley, E.G. ledsl. The zoogeography of North
American freshwater fishes. John Wiley and Sons.

61 Alvarez, F. and Villalobos, J.L. 1991. A new genus and two new species of freshwater crabs
from Mexico, Odontothelphusa toninae and Stygiothetphusa lopezformenti [Crustacea:
Brachyura: Pseudothelphusidael. Proceedings of the Biological Society of Washington
10^121:288-29^.

62 Rodriguez, G, 1986. Centers of radiation of fresh-water crabs in the neotropics. In: Gore, R.H.
and Heck, K.L. ledsl. Biogeography of the Crustacea. Crustacean Issues 4: 51-67.

63 Chace, F.A. and Hobbs, H.H. 1969. The freshwater and terrestrial decapod crustaceans of the
West Indies with special reference to Dominica. United States National Museum Bulletin
292: 1-258.

bU Rodriguez, G. and Campos, M.R. 1989. The cladistic relationships of the freshwater crabs of
the tribe Strengerianmi ICrustacea, Decapoda, Pseudothelphusidael from the northern
Andes, with comments on their biogeography and descriptions of new species. Journal of
Crustacean Biology9: Hl-156.

65 Rodriguez, G. and Pereira, G. 1992. New species, cladistic relationships and biogeography of
the genus Fredius (Decapoda: Brachyura: Pseudothelphusidael from South America.
Journal of Crustacean Biology 1 2: 298-31 1 .

66 Rodriguez, G. and von Sternberg, R. 1998. A revision of the freshwater crabs of the family
Pseudothelphusidae (Decapoda: Brachyura] from Ecuador Proceedings of the Biological
Society of Washington 111.

67 Rodriguez, G. 1982. Les crabes d'eau douces d'Amerique. Famille des pseudothelphusidae.
Faune Tropicale 22, ORSTOM, Paris.

68 Rodriguez, G. 1992. The freshwater crabs of America. Family Trichodactylidae and
supplement to the family Pseudothelphisidae. Faune Tropicale 31 , ORSTOM, Paris.

69 Magalhaes, C. and Tiirkay, M. 1996. Taxonomy of the neotropical freshwater crab family
Trichodactylidae I. The generic system with description of some new genera (Crustacea:
Decapoda: Brachyura). Senckenbergiana biologica 75(1-21: 63-95.

70 Magalhaes, C. and TiJrkay, M. 1996. Taxonomy of the neotropical freshwater crab family
Trichodactylidae II. The genera Forsteria, Melocarcinus, Sylviocarcinus and Zilchiopsis
(Crustacea: Decapoda: Brachyura). Senckenbergiana biologica 75(1-2): 97-130.

71 Magalhaes, C. and TiJrkay, M. 1996. Taxonomy of the neotropical freshwater crab family
Trichodactylidae III. The genera Fredilocarcinus and Goyazana (Crustacea: Decapoda:
Brachyura). Senckenbergiana biologica 75U-2\: 131-1/12.

72 Parenti, L.R. 1984. A taxonomic revision of the Andean killifish genus Orestias
ICypnnodontiformes, Cypridontidae). Bull. Amer. Mus. Nat. Hist. 178: 107-2U.

73 Lucena, C.A.S. de and Kullander, S.O. 1992. The Crenicichia species of the Uruguai River
drainage in Brazil. Ichthyol. Explor. Freshwat. 3: 97-160.

74 WCMC 1998. Freshwater biodiversity: A preliminary global assessment. Groombridge, B. and
Jenkins, M. World Conservation Press, Cambridge.

INDEX 329

Index

abyssal plains 119, 142
Aceraceae 83
Acheulian tools 37, 39
acid deposition 99, 185
acritarclis 25
Acrochordidae 126-7. 174
Acrochordus granulatus 126-7
Addax nasomaculatus 1 1 1
advanced very high resolution

radiometer lAVHRR]

satellite sensors 73, 74-5
Aepyornis maximus 53
Africa

dry forests 92. 94

Great Lal<es 166. 181, 185,

186, ]87. 307-10

human origins 36, 37

inland water biodiversity

307-7 7

mountain regions 101

rain forests 86, 87, 88

savannah 96, 104

see also named countries
Agenda 21 195
agriculture 48

crop biodiversity 40

land devoted to 48-9, 104-5

origins 33, 39-40

traditional systems 45-6

water consumption 178
air 72

Alaska pollock 145, 146
algae 7, 1 1

aquacuUure 149, 150

marine/coastal 122-3, 124-

5, 133, 138
alien species

deserts 1 1 1

inland waters 185-6

marine and coastal

ecosystems 152-3

Mediterranean-type

ecosystems 107

and recent extinctions 64-5
almond 264
alpaca 275
alpha-diversity 78
alpine tundra 99-102
Alps 83, 101
Amami rabbit 85
Amazon basin 87, 88-9, 165,

168, 182
Amazonian varzea forests 87
Amblyrhynchus cristatus 127
Ambyostoma mexicanum 174
Americas

first human settlement

36-7