ea
Lh
Begun in 1895
NUMBER 361 NOVEMBER 28, 2001
Paleoecological Studies
of South Forida
edited by
i Bruce R. Wardlaw
Published in cooperation with
The United States Geological Survey
ZUSGS
science for a changing world
Paleontological Research Institution
1259 Trumansburg Road
Ithaca, New York, 14850 U.S.A.
PALEONTOLOGICAL RESEARCH INSTITUTION
Officers
PRESTON sree areal ean ea aI ACTED ne a Male REA ee Tad CHRISTOPHER G. MAPLES
FURST WAGEAR RESTO EIN oy ce ieieyet HRL eed Mise ay Bie ett MalISPOA undone TEA Oy Deh enh JOHN PoseTA, JR.
SECONDAVIGE
' a wa
in - yin - Gd AOS Ae Caen Couly Ah al
ony : (\) om? 1.0 axe? Giagewh aoe? :
= om = Cone! it Ut Ge wt We Reet
come ‘2 oe
. l@ 7
4 4 We
CHAPTER
INTRODUCTION TO PALEOECOLOGICAL STUDIES OF SOUTH FLORIDA AND THE IMPLICATIONS
FOR LAND MANAGEMENT DECISIONS
BRUCE R. WARDLAW
U.S. Geological Survey, 926A National Center, Reston, Virginia 20192
ABSTRACT
The Florida Everglades developed from the interplay of sea level and climate. The subtle balance of these two factors over
the last two millennia is important to understanding restoration strategies. During the Medieval Warm Period and the Little Ice
Age, the Everglades showed a general drying trend. The Medieval Warm Period trend was due to a significant decrease in annual
precipitation while sea level was rising. The Little Ice Age trend was due to the combined effect of reduced precipitation and
the slowing in the relative rate of sea level rise. This failure to keep base level with Everglades sediment and peat accumulation,
yielded shorter hydroperiods and dryer conditions. Sea level has been steadily rising over the last century and should be considered
in water flow modeling. Land use and water management practices over the last century have greatly partitioned the Everglades,
compounding its ability to respond ecosystem-wide in predictable ways to climate and sea level change.
Salinity in Florida Bay is affected by marine circulation, climate (rainfall) and runoff. Marine circulation was reduced in the early
20" century by intense railroad, highway, and bridge construction connecting the Keys. Rainfall and runoff appeared coupled early
in the 1900’s and has slowly been decoupled throughout the last century, especially after 1970. A restoration goal should be to
couple rainfall and runoff again. Salinity in central, western and southern (Atlantic transitional) zones is influenced by direct rainfall;
salinity in the northern (northern transitional) and eastern zones is influenced to a large extent by runoff. These latter zones need to
be monitored to ascertain that rainfall and runoff are indeed coupled to the degree that they have been in the past.
Salinity can be predicted for specific sites in Florida Bay following known rainfall months.
INTRODUCTION
The South Florida ecosystem has undergone dynam-
ic changes over the last century but most of these are
known only by analogy. Can paleoecological studies
fill this important information gap? Yes, detailed pa-
leoecological studies provide the retrospective infor-
mation to understand the course and cause of these
changes. Can this information influence and be utilized
by decision makers? Yes, this information enables a
more coherent scientific foundation on which to base
policies and decisions.
This volume offers a wide variety of paleoecologi-
cal studies in many different areas of South Florida.
The area of study and the techniques utilized affect the
conclusions which vary from local and qualitative to
widespread and quantitative. The underlying data for
interpretation are invaluable.
Swart et al. (Chapter 2) show by the isotopic sig-
nature of the water that salinity in central, western and
southern (Atlantic transitional) zones of Florida Bay is
influenced by direct rainfall; salinity in the northern
(northern transitional) and eastern zones of Florida Bay
is influenced to a large extent by runoff (Text-fig. 1).
Holmes et al. (Chapter 3) discuss the utility of 7!°Pb
in dating cores from the mud banks of Florida Bay.
This technique provides age models for the last 150
years and is critical for unraveling the sequence of
variability and change in Florida Bay.
In the peats of the terrestrial ecosystem and the peats
of the initial transgression in the bays, radiocarbon
dates provide the age framework and are reliable up
until near modern time where the margin of error ex-
ceeds the age value. *!°Pb has provided reasonable age
models in the uppermost part of the peats for the last
century of deposition.
Willard et al. (Chapter 4) utilize modern census data
from the ridge and slough and southeastern Everglades
National Park along with that data collected by Riegel
(1965) to develop a robust modern analogue for inter-
preting down-core terrestrial samples. Core samples
are from the northern ridge and slough, Shark River
Slough, and Taylor Slough. The core data indicate that
the Everglades were generally wetter (experienced
deeper water and longer hydroperiods) than today, but
experienced three significant drying spells: one that
correlates with the Medieval Warm Period (800—1400
AD); one that correlates with the Little Ice Age (1550—
1850 AD); and one that is the present century. The
present dry spell shows a progressive deterioration
with shallower water and shorter hydroperiods by
1930, significant localization of the marsh in the
1950’s and 1960’s, and expansion of cattail marshes at
nutrient-enriched sites after 1960.
°
BULLETIN 361
Only Precipitatiamsand
sm
CL,
ATLANTIC OCEAN
Text-figure 1.—Zones of Florida Bay defining eastern, northern (Northern Transitional), central, western, and southern (Atlantic Transitional)
Florida Bay and isotopic signature of the waters (derived from Swart er al., this volume).
Winkler et al. (Chapter 5) concentrate on several
cores from the brackish marshes and wet prairies of
eastern Everglades National Park where marls and cal-
careous soils currently dominate. They provide a
wealth of data from a variety of fossil groups. An in-
teresting insight, based on the distribution of chitinous
exoskeletons of cladoceran crustaceans from two cores
shows a dramatic recent increase; it is commonly as-
sociated with periphyton; and implies that periphyton
may also be a recent phenomenom, corresponding to
nutrient and trace element increases from land use
changes during the last 50 years. This suggestion
needs careful documentation of the modern distribu-
tion and the distribution in many more cores before it
can be substantiated. Winkler er al. (Chapter 5) envi-
sion Everglades peat deposition initiating about 5,000
INTRODUCTION: WARDLAW 7
Lake
Okeechobee
Radiocarbon Ages
of Initial Sedimentation
in years before present
Elevation of Bedrock
Surface Showing Everglades
Trough (Shaded)
Contours in feet above mean
sea level
Text-figure 2——Basal ages of initial Everglades transgressive sed-
imentation modified from data presented in Winkler et al. (this vol-
ume) and elevation of bedrock surface modified from Brooks (1984)
outlining Everglades trough (shaded, elevations generally 6 ft or less
above mean sea level).
ame] Pinelands
x Wet Prairies
> al Cypress Forests
= Brackish Marshes
Sa] Mangroves
‘| Developed Area
= Rockland Pine Forests
ey] Sloughs
iii) Sawgrass Marshes
years ago by a widespread climate change; however,
based on their data, it is more easily explained as part
of the Holocene transgression (see below). They cau-
tion that dating at close intervals is critical to under-
standing the ecosystem history, yet fall prey to lack of
precision when their correlation of marls as contem-
poraneous is not supported by their sparse age data.
However, based on diatom abundances, they propose
a wet period between 2,000 and 1,600 years B. P,
which agrees with the wet period observed by Willard
et al. (Chapter 4) based on pollen at the base of their
sections (0 to 800 AD).
O’Neal et al. (Chapter 6) utilize census data from
Riegel (1965) which concentrated on the western por-
tion of Everglades National Park. Numerical analysis
(cluster and multivariate analysis) characterize spatial
pollen zones of the western Everglades and examine
the historical record in two cores from the mangrove
swamps of the same region. They document the over-
all late Holocene transgression and differentiate man-
grove swamp communities as coastlines that first trans-
gressed and then were stabilized.
Zarikian et al. (Chapter 7) show that rainfall is the
dominant driving force behind long-term variability in
microfaunal assemblages from one core from Oyster
Bay within the mangrove forests. This environment
(interior mangrove forest) generally is most suscepti-
bile to hurricane devastation and yet the core provides
Developed areas
Peat and muck soils -
sawgrass peat
Peat and muck soils -
loose peat
Sandy soils - light colored,
agi poor natural drainage
and low fertility
es) Rockdale - rockland pine
Wet rock land, marshes,
swamp and made land
= Marls and calcareous soils
eal Marls - saline
A Vegetation Type, B Soil Type
Text-figure 3.— Vegetation types (A) of South Florida (modified from Willard et al., this volume) and soil types (B) of parts of South Florida
(modified from Willard et al., 2001).
8 BULLETIN 361
Cattails
2000
ar
(ec)
lo)
oO
1600
Little Ice Age
1400
1200
1000
800
Medieval Warm Period
600
400
Weedy Sawgrass Marsh
; qf z | |
; ae z
Wet Prairie
Site 6
—M
Mixed |
Florida
B ay
Text-figure 4.—Pollen wetland subenvironments indicated for eight peat cores from South Florida over the past 3 millennia. Locality shown
in inset (modified from data in Willard et a/., this volume).
little correlation to hurricane events, though the au-
thors maintain they remain an important factor. Stable
isotopes from ostracode and foram tests suggest a de-
coupling of rainfall and freshwater flow during the
1970’s for this particular location.
Huvane and Cooper (Chapter 8) show that diatoms
are common to Florida Bay cores and useful in sea-
grass, salinity, and productivity studies.
Cronin et al. (chapter 9) utilizing ostracode-epiphyte
distributions show that seagrass and macro-benthic al-
gae fluctuate in abundance and coverage. They were
sparse in the late 19" and early 20" centuries, dense
in the 1950’s and 1960’s and declined in the 1970’s
and 1980’s; but they are still more common than at
the turn of the 20" century.
Brewster-Wingard et al. (Chapter 10) show that the
molluscan faunas that occur throughout the cores ba-
sically reflect the same fauna that is present at those
sites today. However, subtle fluctuations in patterns of
dominance and diversity track the changes in local
habitat, salinity, and water quality through time.
Ishman (Chapter 11) recognizes four distinct fora-
miniferal assemblages in Biscayne Bay: an open-bay
seagrass, an open-bay substrate, a restricted and a high
productivity assemblage. The restricted assemblage
represents reduced salinity, restricted circulation, and
point-source fresh water input. The high productivity
assemblage represents high nutrient input and is very
local in the northern part of the Bay.
Dwyer and Cronin (Chapter 12) show that Mg/Ca
ratios from adult ostracode shells reflect the salinity and
provide a reasonable value function; they conclude that
there is a strong climatic control on Florida Bay salinity.
THE EVERGLADES
Data presented in this volume suggest that the Ev-
erglades are the result of a delicate balance of sea level
rise and climate (precipitation). Comparison of the age
INTRODUCTION: WARDLAW 9
Salinity (ppt) and Rainfall (cm)
Text-figure 5.—Average annual rainfall for South Florida, in cm,
compared to salinity derived from shell elemental chemistry of os-
tracodes from the Russell Bank Core (for location see Fig. 6, mod-
ified from Dwyer and Cronin, this volume). Notice the strong match
of low rainfall years and high salinity up to 1967 and inconsistent
matches following that year.
of initial sedimentation of Everglade deposits to the
elevation of bedrock (Text-fig. 2) shows a direct cor-
relation of lower elevations and earlier deposition.
Clearly, the Everglades represent a growing terrestrial
marsh ecosystem with rising base level driven by sea
level rise. This is exactly what O’ Neal et al. (Chapter
6) could discern from their work, documenting a trans-
gressive sequence in the mangrove environments.
Vegetation and soil type (Text-fig. 3) distributions
agree well with each other and show a direct correla-
tion to the trough shown by lower elevation (Text-fig.
2) with the area that comprises the ridge and slough
vegetation of the Everglades today.
Willard et al. (Chapter 4) indicate three general dry-
ing phases in the history of the Everglades summa-
rized in Text-figure 4. These correspond to the Medi-
eval Warm Period (800-1400 AD), the Little Ice Age
(1550-1850 AD), and the present (post-1930 AD).
Much evidence suggests that the Medieval Warm Pe-
riod represented a period of decreased rainfall in the
subtropics (see Willard er al., Chapter 4); hence the
reason for the drying of the Everglades. However, the
explanation of the Little Ice Age is less clear. Fair-
bridge (1984) presents a lot of anecdotal evidence for
a decrease in the rate of sea level rise or even a slight
EVERGLADES
NATIONAL
PARK Decadal Salinity Averages
Florida Bay 4
®@ Lignumvitae
SALINITY (PPT)
Text-figure 6.—Decadal salinity averages across Florida Bay.
fall, during the Little Ice Age. It appears that a drop
in base level along with a slight decrease in the pre-
cipitation probably caused the drying conditions ob-
served for the Little Ice Age period in the Everglades.
This emphasizes the subtle interplay of base level and
precipitation in maintaining deep water levels and long
hydroperiods for the Everglades. Long term sea-level
curves (Robbin, 1984; Toscano and Lundberg, 1998)
are smoothed given the coarseness of the data com-
pared to the fluctuating curve of Fairbridge (1984). All
consistently show a decrease in the rate of sea-level
rise approaching the present, which suggests a natural
drying trend for the Everglades. However, the present
shows a dramatic drying trend, starting around 1930
(Willard er al., Chapter 4), and fragmentation after
1960 that appear to be beyond the interplay of precip-
itation and base level.
SALINITY IN FLORIDA BAY
Data presented in this volume and in an earlier sum-
mary by Brewster-Wingard ef al. (1998) indicate that
fossil and geochemical proxies yield fairly reliable sa-
linity information from Florida Bay. The distribution
10 BULLETIN 361
Bob Allen 6A e
High
Amplitude
Shifts
Ee
2
2
[o)
oO
=
=
a
oO
a
Russell Bank 19B o
Pass Key 37 @ Taylor T24 «
0)
10
20
30
40
50
60
70
~1900
ontoODmonTO
EVERGLADES
NATIONAL
PARK
Distribution of Benthic Salinity Indicators in Florida Bay Cores
Text-figure 7.—Distribution of benthic faunal indicators of euhaline and near euhaline conditions (> 25 ppt) in four cores from Florida Bay
(modified from Brewster-Wingard et al., 1998).
of benthic faunal indicators (foraminifer, ostracodes,
and mollusks) from four cores and the shell geochem-
istry of the ostracode shells from three of these cores
have been analyzed. We are continuing the shell geo-
chemistry analysis of additional cores, looking at the
results in different growth stages, and looking at the
elemental chemistry and carbon and oxygen isotopes
of mollusk shells in an effort to evaluate the seasonal
variability in salinity over the last century.
The salinity values derived from shell geochemistry
are compared to the average annual rainfall for South
Florida (Text-fig. 5). The salinity record is taken from
2 cm samples that average approximately 2 years/sam-
ple based on the age model which yields a sedimen-
tation rate of 1.28 cm/year (Russell Bank Core, see
Holmes et al., Chapter 2). The average rainfall curve
is a compilation of the available data from NOAA
website (http://www.ncde.noaa.gov/onlineprod.
drought/xmgr.html) for the three reporting districts that
makeup South Florida (the SW, the SE coast, and the
Bay and Keys), averaged to one composite and coy-
ering the years 1895 to the present. Our core was taken
in 1995. The rainfall pattern beginning the 19" century
shows high frequency, low amplitude shifts that
change to high frequency, high amplitude shifts in the
late 1930s, and change again to moderate frequency,
moderate amplitude shifts for the latter part of the 20"
century. The derived salinity curve matches the rain-
fall curve very well for the first and middle parts of
the century, with high salinity peaks occurring during
low rainfall years and low salinity peaks occurring in
high rainfall years. This is quite remarkable, given the
difference in sampling intervals. However, following
1967, a decoupling occurs, with only the lowest rain-
fall years matching salinity highs and moderate and
high rainfall years not corresponding consistently with
moderate or low salinities. Coincidently, 1967 is the
year that South Florida came under complete water
control management. Recent isotopic data from Swart
(Chapter 2) suggest that the salinity values in the east-
ern part of Florida Bay are affected directly by surface
waters as those waters have an isotopic composition
that reflects strong mixing with Everglades water.
Therefore, rainfall is coupled to decreases in salinity
directly by increased surface water flow and that flow
was disrupted in the late 1960's.
Swart et al. (1996b) presented salinity data based
on C and O isotopic studies of the growth bands in a
rare large coral head in Florida Bay from Lignumvitae
basin (inset, Text-fig. 6). The data were summarized
as decadal salinity averages. I compare this to the sa-
linity values derived from our combined fossil salinity
INTRODUCTION: WARDLAW 11
‘Best Fit' Age
140
120
100 Russell Bank Age Model
€
oO}
o 80
° *
: 4
£ LY
= 60 Sa
= yy
6 _
40 we LE
a ee
ge om
AE - a
20 JP _AA
1985S O75 N 19655 1955 S45 935
Year
Range in Age for a
given 2 cm Sample
Maximum, Median, and
Minimum Age based on
Sedimentation Rate
Park Key Age Model
1925) SiS) 1905) e895
Text-figure 8.—Best Fit Age model for Russell Bank and Park Key.
proxies (foraminifers, ostracodes, and mollusks). De-
cadal averages are plotted at the midpoint of each de-
cade (Text-fig. 6). The data for both Lignumvitae and
Bob Allen Key extend back to before 1850. The Rus-
sell Bank data extends back to 1870. Prior to 1910,
the three localities show a consistent gradient from
fresher to more saline values from north to south in
the bay. However, following 1910, Lignumvitae shows
less decadal fluctuation and slightly elevated average
salinities (above mean sea average) indicating more
restrictive conditions. Bob Allen and Russell Bank
switch; that is, though the decadal fluctuations still
track each other, Bob Allen in the center of the bay
becomes consistently fresher relative to Russell Bank,
disrupting the gradient noted prior to 1910. Large-
scale construction connecting the Keys and filling in
marine passages (building the Flagler Railroad to Key
West) and initial canal construction in the Everglades
occurred in the first two decades of the 20" century.
These activities may have directly affected the subtle
changes seen in the salinity patterns of the Bay.
The salinity values derived from faunal (foraminifer,
mollusk, and ostracode) proxies for our first four Flor-
ida Bay cores are shown in Text-figure 7. These are
plotted against depth in core with the age model high-
lighted in bold. The values for the longer cores (Bob
Allen and Russell Bank) show low amplitude shifts in
salinity prior to 1910 and high amplitude shifts in sa-
linity after about 1940, agreeing with the curve de-
veloped from shell geochemistry for Russell Bank.
These faunal derivations are too coarse to comment on
the frequency of salinity shifts. The overall trend in-
dicated by these benthic faunal indicators is that most
of the bay seems to have experienced a slight increase
in salinity values over the last century and specifically
in last 30 years (post-1970).
IMPLICATIONS FOR RESTORATION
Comparison of the average annual rainfall of South
Florida (Text-fig. 5) to the pollen assemblage data
which is summarized in Text-figure 4 shows that
broad-scale vegetational changes in the Everglades
wetlands began by 1930, indicating shallower water
depths and shorter hydroperiods, even though regional
precipitation increased concomitantly. Further, more
localized changes occurred after 1960. Thus, restora-
tion goals of achieving pre-1960 hydrologic regimes
are aimed at an already disrupted system; a more “‘nat-
ural” restoration target would be the 19" century Ev-
erglades, consisting of slough subenvironments which
had been stable for the past few centuries. Recent land-
use changes have resulted in localized rather than sys-
tem-wide ecosystem responses, at least in part because
of the fragmentation of the wetland. This artificially
induced ecological heterogeneity makes prediction of
future wetland responses to climatic changes problem-
atic.
Faunally and chemically derived salinity curves for
Florida Bay show that salinity values are directly re-
lated to rainfall and surface water flow for the eastern
half of the bay. However, following 1967, there is a
decoupling of salinity response to rainfall amount and
surface water flow. More subtle trends show that the
annual average salinity of the bay has increased slight-
ly during the 20" century and that the general pattern
of fresher to more saline water across the eastern bay
has been disrupted after the turn of the century. These
trends in the salinity values of the bay are related to
the disruption of both fresh-water and marine-water
delivery to the system that began shortly after the turn
of the century and culminated in total water-control
management in 1967. Recent water practices of re-
leasing fresh water into Taylor Slough appear to have
reestablished at least a normal gradient in salinity over
the past 5 years, based on our modern monitoring data
(post-coring).
Long-term data provided by faunal, floral and bio-
geochemical analysis provide essential information to
establish restoration targets for the South Florida eco-
system. These data clearly indicate that a more open
water delivery system is necessary to provide eastern
Florida Bay with more fresh water and the terrestrial
Everglades with less restricted flow and deeper water
during wet cycles, thus reestablishing more natural cy-
cles of change.
PREDICTING SALINITY IN FLORIDA BAY
The chemical proxy data from Florida Bay cores
allow us to develop predictive formulas for establish-
ing an anticipated salinity for a given rainfall record
that should help us test success in recoupling rainfall
and surface water flow. It also allows us to “‘retune”
the age model provided by Holmes et al. (Chapter 3).
Biotic and chemical proxies for salinity in Florida
Bay show that salinity values for the last century are
strongly correlated with climate, specifically, rainfall
(Brewster-Wingard and others, 1998; Cronin and oth-
ers, 1998). Elemental chemistry of ostracode shells
provides a methodology to discern salinity values of
the past (Dwyer and Cronin, 1999). Adult tests of the
ostracode Loxoconcha matagordensis are grown es-
sentially instantaneously in the late Spring and early
Summer. Salinity values derived from adult Loxocon-
cha matagordensis, generally grown between May and
July, each year serve as proxies for June salinities.
Because Loxoconcha matagordensis growth shows a
very tight normal distribution, population studies from
one sample should provide us with data on salinity
2» BULLETIN 361
Table 1.—Best Fit Model data for year, salinity (in parts per thou-
sand, ppt), and cumulative January—May rainfall (in cm).
Russell Bank Park Key
Year Salinity Rainfall Year Salinity Rainfall
1900 21 45.30 1895 31 33.46
1903 27 39.88 1897 30 30.80
1906 22. 42.71 1900 27 45.30
1912 20 41.50 1906 20 42.71
1913 19 35.54 1910 58 23.60
1915 21 42.59 1917 60 17.56
1916 30 26.93 1922 38 29.52
1917 34 17.56 1927 38 21.74
1919 21 43.67 1928 29 37.05
1922 30 29.52 1931 26 50.62
1923 20 35.61 1935 47 21.07
1927 37 21.74 1936 27 45.99
1928 27 37.05 1938 40 19.03
1929 42 25.05 1942 31 45.26
1931 20 50.62 1945 53 13.68
1932 28 41.44 1947 34 43.36
1933 40 29.49 1949 56 21.67
1934 28 45.75 1952 BS 28.06
1935 43 21.07 1955 51 18.84
1936 24 45.99 1957 38 51.61
1938 44 19.03 1958 33 64.36
1941 20 53.94 1959 27 41.65
1943 38 25.82 1962 60 20.67
1947 27 43.36 1966 35 42.40
1949 54 21.67 1971 49 16.33
1950 53 21.08 1973 36 27.60
1953 35 31.50 1978 25 41.01
1954 22 47.26 1981 51 20.05
1955 5y7/ 18.84 1982 34 52.45
1956 57 18.83 1983 32 67.63
1958 27 64.36 1985 45 28.03
1962 51 20.67 1987 33 38.32
1963 51 2el> 1991 28 48.61
1964 60 29.87 1993 26 50.50
1965 52 21.99
1967 43 19.43
1968 29 46.33
1969 35 43.01
1970 18 51.28
1974 31 20.40
1976 29 37.90
1977 36 36.60
1979 31 40.62
1981 45 20.05
1982 25 52.45
1984 25 45.80
1987 19 38.32
1988 29 32.83
1989 43 PYG)
1990, 42 31.73
1992 37 25°71
1994 35 42.59
changes occurring at the beginning of the rainy season
(usually June). Russell Bank has a long record and
relatively abundant Loxoconcha matagordensis. Park
Key has a relatively short record but also has relatively
INTRODUCTION: WARDLAW |
50
Salinity (ppt)
i}
1
20-4 1900-1928 1974-1994
0 10 20 30 40 50 60 70
Russell Bank
Salinity (ppt)
10 20 30 40 50
Cumulative Rainfall January-May) in cm
Park Key
Text-figure 9.—Regression trend lines for data points listed in
Table 1.
abundant Loxoconcha matagordensis. Biotic indicators
were analyzed from every other 2 cm sample and el-
emental chemistry of the ostracodes was performed
from every sample. The calculated sedimentation rate
of the Russell Bank core is 1.28 cm + 0.07 (Holmes
et al., Chapter 3), so our sampling represents approx-
imately every 1.5 to 2 years. The calculated sedimen-
tation rate of the Park Key core is 0.78 cm + 0.5
(Holmes et al., Chapter 3), so our sampling represents
approximately every 2.5 years.
ASSUMPTIONS AND VALUES FOR THE MODEL
January—May rainfall.——Monthly rainfall informa-
tion is available from NOAA website (www.ncdc.
Ww
Table 2.—Linear equations and R® values for regression trend
lines for all points, natural clusters, and model linear equation for
site at present for ‘natural’ restoration.
Russell Bank Park Key
All Points y = —0.78x + 60.08 y = —0.57x + 57.86
R? = 0.5508 R? = 0.5278
1900-1922 y = —0.62x + 46.60
R? = 0.6541
1895-1936 y = —0.97x + 68.25
R? = 0.7027
1929-1943 y = —0.73x + 58.83
R? = 0.9683
1895-1955 y = —0.86x + 65.24
R? = 0.6882
1946-1970 y = —0.85x + 69.94
R? = 0.7414
1938-1955 y = —0.66x + 60.75
R? = 0.6466
1974-1994 y = —0.50x + 49.98
R? = 0.4253
1957-1993 y = —0.45x + 55.27
R? = 0.4907
Model y = —0.79x + 64.38 y = —0.92x + 66.74
noaa.gov/onlineprod.drought.xmgr.html) dating back
to 1895. The eastern portion of Florida Bay is influ-
enced by rainfall and surface run-off. To compare rain-
fall to salinity values on a monthly basis, averages of
the three southern districts (southwest (Everglades),
southeast coast, and Bay and Keys) were taken be-
cause rainfall to all three may influence salinity in
Florida Bay. Because the focus is June salinity, rainfall
from the five preceding months was examined to es-
tablish a relationship. During El Nino years, high win-
ter rainfall is usually recorded in January; therefore,
January rainfall is included in the data set to indicate
the occurrence of these events. This methodology ap-
pears to combine the affects of both direct rainfall and
surface water runoff. The salinities derived from the
cores do not match well with direct rainfall for the
months of May, June, and July, and imply that surface
water flow has a strong influence on the salinity at
these sites.
June salinities.—Meg/Ca derived salinities from the
ostracode Loxoconcha matagordensis from Russell
Bank Core and Park Key Core as discussed above and
derived from the analyses listed in Dwyer and Cronin
(this volume). Wardlaw and Brewster-Wingard (2000)
have shown significant correlation between the Janu-
ary—May rainfall and derived salinities for the eastern
portion of the Bay. Here, I have taken the age model
of Holmes er al. (Chapter 3, this volume) and fine
tuned it (Text-fig. 8) matching better rainfall to salinity
values within or close to within the allowable ages
given the age model and sampling interval (see Table
1 for all data).
14 BULLETIN 361
Results.—The relationship between January—May
average rainfall for South Florida and June “‘‘ostra-
code”’ salinities can be expressed as a series of simple
linear equations of negative correlation (Text-fig. 9)
and provides us with a predictive value. The relation-
ships for all derived values and relationships that clus-
ter in age groups are shown in Table 2.
The lessening of slope appears to indicate disruption
or decoupling of the natural system. At Park Key, this
appears as early as the 1930’s. At Russell Bank, it
appears in the seventies. The correlation value (R?)
also decreases greatly at major disruption. Though
Park Key appears disrupted early in the century, its
impact appears minor in that the trend line for all data
1895-1955 is still very close to that before disruption.
A model for prediction of salinity can be derived be-
tween these two trend lines (Text-fig. 9).
Russell Bank shows a progressive change in slight
increase in slope and rising in overall salinity through
most of the century. Rainfall patterns are changing
from high amplitude, high frequency fluctuations to
more moderate amplitudes and frequencies and a mod-
el for prediction can be derived between trends of
1929-1943 and 1946-1970, suggesting a slight less-
ening of overall salinity for this site.
The slope of the trend line decreases as a function
of distance from surface water source and with a few
additional cores could be mathematically modeled.
In addition, the interplay of rainfall versus surface
water flow on salinity can be addressed by looking at
the biology of the ostracodes more closely and devel-
oping chemical proxies from longer-lived taxa such as
molluscs that can yield multi-year records of nearly
continuous carbonate secretion.
REFERENCES CITED
Alvarez Zarikian, C.A., Swart, P.K., Hood, T., Blackwelder,
P.L., Nelsen, T.A., Featherstone, C., Wanless, H.R.,
Trefrey, J., and Kang, W.-J.
2001. Chapter 7. A century of environmental variability in Oys-
ter Bay using ostracode ecological and isotopic data as
paleoenvironmental tools. Bulletins of American Paleon-
tology, no. 361, pp. 133-143.
Brewster-Wingard, G.L., Ishman, S.E., and Holmes, C.W.
1998. Environmental Impacts on southern Florida Coastal Wa-
ters: A history of change in Florida Bay. Journal of Coast-
al Research, Special Issue 26, pp. 162-172.
Brewster-Wingard, G.L., Stone, J.R., and Holmes, C.W.
2001. Chapter 4. Molluscan faunal distribution in Florida Bay,
past and present: An integrated of down-core and modern
data. Bulletins of American Paleontology, no. 361, pp.
199-231.
Brooks, H. K.
1984. Lake Okeechobee. in Gleason, P.J., ed., Environments of
South Florida, Present and Past II. Coral Gables, FL, Mi-
ami Geological Society, pp. 38-68.
Cronin, T.M., Holmes, C., Wingard, L., Ishman, S., Dowsett, H.,
and Waibel, N.
2001. Chapter 9. Historical trends in epiphytal ostracodes from
Florida Bay: implications for seagrass and macro-benthic
algal variability. Bulletins of American Paleontology, no.
361, pp. 159-197.
Dwyer, G.S., and Cronin, T.M.
2001. Chapter 12. Ostracode shell chemistry as a paleosalinity
proxy in Florida Bay. Bulletins of American Paleontolo-
gy, no. 361, pp. 249-276.
Fairbridge, R.W.
1984. The Holocene sea level record in South Florida. in Glea-
son, P.J., ed., Environments of South Florida, Present and
Past II. Coral Gables, FL, Miami Geological Society, pp.
427-436.
Holmes, C.W., Robbins, J., Halley, R., Bothner, M., Ten Brink,
M., and Marot, M.
2001. Chapter 3. Sediment dynamics of Florida Bay mud banks
on a decadal time scale. Bulletins of American Paleon-
tology, no. 361, pp. 31—40.
Huvane, J.K., and Cooper, S.R.
2001. Chapter 8. Diatoms as indicators of environmental change
in sediment cores from northeastern Florida Bay. Bulle-
tins of American Paleontology, no. 361, pp. 145-158.
Ishman, S.E.
2001. Chapter 11. Ecological controls on benthic foraminifer
distributions in Biscayne Bay, Florida. Bulletins of Amer-
ican Paleontology, no. 361, pp. 233-247.
O’Neal, M.A., Tedesco, L.P., Souch, C., and Pachut, J.F.
2001. Chapter 6. A pollen zonation of southwestern Florida us-
ing multivariate statistical methods and its application to
two vertical sedimentary sequences. Bulletins of Ameri-
can Paleontology, no. 361, pp. 101-132.
Riegel, W.L.
1965. Palynology of environments of peat formation in south-
western Florida: Unpublished PhD thesis, The Pennsyl-
vania State University, 189 pp.
Robbin, D.M.
1984. A new Holocene see level curve for the Upper Keys and
Florida reef tract. in Gleason, PJ., ed., Environments of
South Florida, Present and Past II. Coral Gables, FL, Mi-
ami Geological Society, pp. 437—458.
Swart, P.K., Healy, G.F., Dodge, R.E., Kramer, P., Hudson, J.H.,
Halley, R.B., and Robblee, M.B.
1996. The stable oxygen and carbon isotopic record from a coral
growing in Florida Bay, A 160 year record of climatic
and anthropogenic influence. Palaeogeography, Palaeocli-
matology, Palaeoecology, v. 123, p. 219-237.
Swart, P.K., Price, R.M., Greer, L.
2001. Chapter 2. The relationship between stable isotopic vari-
ations (O, H and C) and salinity in waters and corals from
environments in South Florida: implications for reading
the paleoenvironmental record. Bulletins of American Pa-
leontology, no. 361, pp. 17-29.
Toscano, M.A., and Lundberg, J.
1998. Early Holocene sea-level record from submerged fossil
reefs on the southeast Florida margin. Geology, vol. 26,
no. 3, pp. 255-258.
Wardlaw, B.R., and Brewster-Wingard, G.L.
2000. Predicting salinity in Florida Bay. Greater Everglades
INTRODUCTION:
Ecosystem Restoration Science Conference: Defining
Success, South Florida Ecosystem Restoration Task Force
and Working Group, Naples, Florida, 12/1 1—15/2000, pp.
177-179.
Willard, D.A., Holmes, C.W., and Weimer, L.M.
2001. Chapter 4. The Florida Everglades ecosystem: Climatic
and anthropogenic impacts over the last two millennia.
Bulletins of American Paleontology, no. 361, pp. 41-55.
W ARDLAW 15
Willard, D.A., Weimer, L.M., and Riegel, W.L.
2001. Pollen assemblages as paleoenvironmental proxies in the
Florida Everglades. Review of Palaeobotany and Paly-
nology, vol. 113, pp. 213-235.
Winkler, M.G., Sanford, P.R., and Kaplan, S.W.
2001. Chapter 5. Hydrology, vegetation, and climate change in
the southern Everglades during the Holocene. Bulletins of
American Paleontology, no. 361, pp. 57-99.
CHAPTER 2
THE RELATIONSHIP BETWEEN STABLE ISOTOPIC VARIATIONS (O, H, and C) AND SALINITY IN
WATERS AND CORALS FROM ENVIRONMENTS IN SOUTH FLORIDA: IMPLICATIONS FOR READING
THE PALEOENVIRONMENTAL RECORD
PETER K. SWART, RENE M. PRICE, AND LISA GREER
Stable Isotope Laboratory
Rosensteil School of Marine and Atmospheric Science
University of Miami
4600 Rickenbacker Causeway
Miami, FL 33149
ABSTRACT
The oxygen isotopic composition of calcareous material secreted by organisms is often used for the reconstruction of salinity
values in past environments. Implicit in this reconstruction is the assumption of a constant relationship between salinity and the
5'8O value of water. In this paper, we demonstrate through the analysis of water samples collected from several estuaries situated
adjacent to the Florida mainland that varying relationships exist between these two parameters depending upon the extent of
evaporation experienced by the freshwater which influences the salinity. We further show that these relationships are translated
into the 6'°C and 6'8O values of corals growing in these environments. Water samples collected from Florida Bay, Whitewater
Bay, and the Ten Thousand Island area between October 1993 and December 1998 show a range of correlations between the
5'8O of the water and salinity. These varying correlations arise because while water from the Everglades possesses a low salinity
it has a relatively positive 6'°O value. In contrast, precipitation has a more negative 65'*O value. Consequently, areas in which
the salinity is significantly influenced by freshwater derived from the Everglades tend to have different 6!*O to salinity relation-
ships compared to regions where salinity variations are caused solely by inputs of precipitation. These variations must be
recognized if one is to use proxy indicators of salinity such as the 6!*O value of skeletal carbonates to accurately reconstruct
past salinity records.
The 6'°C of estuarine waters is related to both the input of isotopically depleted runoff from the Everglades and the local
remineralization of organic material. Areas far removed from Everglades runoff, where variations in salinity resulted from input
of precipitation, show little relationships between salinity and 5'°C because 6'°C variations arise from the local remineralization
of organic material. In contrast, areas closer to the Everglades show stronger relationships between salinity and 6'°C. These
relationships are manifested in the isotopic composition of the coral skeleton as an absence of correlation between 6'C and 6!*O
in more marine areas compared to areas more directly influenced by runoff.
INTRODUCTION
The 6'°C and 6!8O values of calcareous organisms
(ostracods, foraminifera, molluscs, and corals) have
been used in the South Florida environment to ascer-
tain changes in past environmental conditions, in par-
ticular salinity (Swart et al., 1996; Swart et al., 1999;
Halley and Roulier, 1999; Nelsen et al., 2000). The
results of these studies indicate a change in salinity
patterns commencing in Florida Bay as early as 1900.
A concurrent shift in the 6'°C in both corals and mol-
luscs suggest that a reduction in the hydrodynamic cir-
culation of Florida Bay may be responsible for the
observed changes in salinity. Construction of the
Overseas Railway from Miami to Key West from
1906-1914, followed by water management practices
which diverted the natural flow of water from the Ev-
erglades into Florida Bay have been suggested as pos-
sible causes for the reduced circulation within Florida
Bay (Smith er al., 1989; Swart et al., 1996, 1999).
Anthropogenic changes in South Florida occurred
synchronously with climatic variability. Regional
droughts have been recorded in every decade from
1950 to 1990 resulting in salinity exceeding 50 in the
central portions of Florida Bay (Fourqurean and Rob-
blee, 1999; Fourqurean er al. 1992). In contrast, the
1997 El Nino, termed the “El Nino of the Century”
produced above average rainfall during the winter dry
season in South Florida. Decoupling these natural cli-
matic conditions from anthropogenic conditions is im-
portant in deciphering the paleoenvironmental record.
This paper examines the associations between salin-
ity, precipitation, and the 6'°O, 5D, and 6'°C of surface
waters from ‘fresh’, estuarine, and marine environ-
ments of South Florida and relates these data to the
interpretation of the 5'°C and 6'%O record of two corals
growing in Florida Bay. In order to study these rela-
tionships, we have conducted a temporal and spatial
investigation of the 6!°O, 6D, and 6'°C composition of
18 BULLETIN 361
waters from the Everglades, Florida Bay, Whitewater
Bay, and the Ten Thousand Islands areas. These data
were then correlated with salinity measurements made
on the same samples (Boyer ef al., 1999).
ACKNOWLEDGMENTS
The authors would like to thank Dave Rudnick for
having the insight to support this project, and the
SFWMD for providing funding. Help with sample col-
lection was provided by Mike Lutz. Amel Saied and
Vivian Gonzalez are acknowledged for assistance with
analyses. Collection of the water samples would not
have been possible without the assistance of Florida
International University/Southeastern Environmental
Research Program (FIU/SERP). In this regard Ron
Jones and Joe Boyer are especially acknowledged.
George Shardt of Everglades National Park is ac-
knowledged for collection of the Everglades samples.
BACKGROUND
HYDROLOGICAL SYSTEM IN SOUTH FLORIDA
South Florida experiences a subtropical climate with
a hot, wet season during the summer (mid-May—Oc-
tober), and a mild, dry season during the winter (No-
vember—mid-May). Annual precipitation is approxi-
mately 130 to 150 cm across the region, with approx-
imately 70% received during the summer wet season
(Leach et al., 1972). The seasonality of precipitation
and evaporation causes a seasonal fluctuation in sur-
face water levels in South Florida. Water levels rise
during the summer rainy season, with highest levels
commonly observed in September and October. Dur-
ing the winter dry season, water levels decline slowly.
Surface water can be absent from many areas of the
Everglades at the end of the dry season (April and
May) as evapotranspiration exceeds rainfall (Kushlan,
1990).
The Everglades depression extends southward from
Lake Okeechobee to Florida Bay and is bounded to
the east by Pleistocene limestones and quartz sands of
the Atlantic Coastal Ridge and to the west by exposed
Pliocene limestones of the Big Cypress Ridge. Holo-
cene freshwater peats and marls partially fill the Ev-
erglades trough, decreasing in thickness southward
from 4 m at the southern end of Lake Okeechobee to
0.7 m along the coastline of Florida Bay (Kushlan,
1990). Under natural, pre-development conditions,
most surface water flowed southward in the Ever-
glades from Lake Okeechobee to the Gulf of Mexico
via the Shark River Slough (Fennema ef al., 1994). A
smaller drainage area, Taylor Slough, discharged lo-
cally derived freshwater into Florida Bay. Beginning
in the late 1800s and throughout the 1900s, a complex
network of canals, levees, pump stations, and im-
poundment areas were constructed throughout South
Florida in an effort to manage water levels and flow
throughout the region. Because of the resultant im-
poundment of water in what are known as conserva-
tion areas, evaporation is high, with estimates of rain-
fall lost to evapotranspiration being as high as 70% to
95% (Leach et al., 1972). This decreased surface water
flow from Lake Okeechobee, through the Everglades
to the marine systems of the Gulf of Mexico and Flor-
ida Bay, and increased drainage eastward toward Bis-
cayne Bay (Fennema ef al., 1994).
Marine environments that bound the southern pen-
insula of Florida include the Gulf of Mexico to the
west, Biscayne Bay to the east, and Florida Bay to the
south. The west coast from Cape Romano in the north
to Cape Sable in the south, is characterized by man-
grove swamps, tidal channels and small bays. The
northern section contains numerous mangrove islands
separated by tidal channels with strong currents and is
referred to as the Ten Thousand Islands (TTI) area.
South of the TTI, tidal currents are more sluggish and
freshwater runoff from the Big Cypress Swamp and
the Everglades is transported to the Gulf of Mexico by
numerous streams (Davis, 1997).
Florida Bay is a large (2,000 km7?) triangular shaped
estuary lying between the Southern Peninsula of Flor-
ida and the Florida Keys. Florida Bay is relatively
shallow and underlain by a complex system of Holo-
cene carbonate mud-banks that subdivide Florida Bay
into smaller basins. Water in Florida Bay is a mixture
of precipitation, freshwater runoff from the Ever-
glades, and seawater from the Gulf of Mexico and the
Florida Straits. These sources combined with evapo-
ration results in salinity variations from 4 to 50 (Nuttle
et al., 2000).
STABLE ISOTOPES
The major influence on variations in the ratio of '*O
to '°O and D to H in the water cycle occurs during the
processes of evaporation and precipitation (Craig and
Gordan, 1965). During evaporation, the concentrations
of 'SO and D are reduced in the water vapor while
these same species are enriched in the condensate. As
a consequence, restricted bodies of water become en-
riched in the heavier isotopes during evaporation while
precipitation is typically enriched in '°O and H. Pre-
cipitation throughout the world has been measured by
the International Atomic Energy Authority (IAEA) and
this data set shows that precipitation values become
increasingly depleted towards the north and south
poles (Rozanski et al., 1993). In South Florida the
IAEA data estimates that rainfall should have a 65'8O
value between —2 and —4%c. The behavior of the iso-
topic composition of water during evaporation and
STABLE ISOTOPIC VARIATION: SWART ET AL. 19
Cape Romano < “BG
® » a
Outer Florida Bay
°
Bs A nee Miami
ox “ Biscayne Bay
x °
i °
ouned ‘ Precipitation
“ Precipitation “ _@ /
it cae) NX Manates Bay Coral
* ” . y Cora
F, ;
° e
De e Inner Florida Bay
ONS
Eastern-middle Florida Bay
Western-middle Florida Bay
e\Lignumvitae Basin Coral
Text-figure 1_—Location map of water sampling stations within each geographic region.
precipitation can be modeled using the Rayleigh Dis-
tillation Equation. However, the maximum amount of
enrichment in the heavier isotope that can be experi-
enced by an evaporating water body is limited by sub-
sequent exchange between the atmospheric moisture
and the water body. The resulting isotopic composition
that can be attained during evaporation is a function
of the relative humidity, temperature of evaporation,
and the isotopic composition of the atmosphere. The
principals which govern the isotopic composition of
such water bodies have been well studied and for fur-
ther discussion the reader is referred to Gat (1981) and
Gonfiantini (1986). For an environment such as South
Florida where the mean relative humidity is 75%, the
maximum 6'*O and 6D that can be attained by the
water body undergoing evaporation are approximately
+4%o and +42%o, respectively. Although the 6D ys.
6'8O of most precipitation throughout the world falls
along a line known as the meteoric water line (MWL)
(Craig and Gordan, 1965), data from evaporating wa-
ter basins, when plotted on a similar diagram, fall on
lines which deviate from the MWL depending upon
the relative humidity. For example, environments with
high relative humidity fall close to the MWL with a
slope of 8, while waters evaporated into an environ-
ment with a lower relative humidity will plot on a line
with a slope significantly less than 8 (Gonfiantini,
1986).
The dissolved inorganic carbon (DIC) pool is com-
posed of CO,, HCO, , and CO,’ . In waters of pH
between 7 and 9, the DIC is dominated by HCO, -.
Although there is a large fractionation factor between
CO, and HCO, (e = 8%c), the measured isotopic
composition of the DIC reflects that of the HCO,
fraction. Hence when CO, is added to water the iso-
topic composition of the HCO, will be approximately
8%c heavier than that of the CO,.
The 65°C of the DIC is governed by the input of
respiratory carbon delivered to the system through the
respiration of organic material and the consumption of
CO, through the process of photosynthesis. As organic
material contains less '*C than DIC which is in equi-
librium with the atmosphere, addition of oxidized ma-
terial will cause the 6'°C of the DIC to become en-
riched in the lighter isotope of C ('?C). Terrestrial or-
ganic materials tend to have more negative 6'°C values
than marine material, and therefore waters derived
from terrestrial sources associated with the oxidation
of organic material often have more negative isotopic
signatures than waters derived from marine sources. In
a relatively closed system, the utilization of CO, in the
process of photosynthesis can cause the residual pool
of carbon in an environment to become enriched in
5C. Hence algal blooms for example could lead to
more positive DIC 6'°C values.
METHODS
WATER SAMPLE COLLECTION
Surface water samples were collected monthly since
October 1993 in Florida Bay (Text-fig. 1). In October
20 BULLETIN 361
1996 and February 1997, additional stations were add-
ed first in Whitewater Bay and subsequently in the Ten
Thousand Islands area. Sampling for 6'8O at nine sites
in the Everglades was initiated in August 1996. All
water samples were collected in the field, filtered
through a 0.45 jm filter to remove carbonate and other
particles, and treated with a small amount of HgCl, to
suppress biological activity. In the laboratory, the sam-
ples were transferred to serum bottles and capped to
exclude as much of the air space in the bottle as pos-
sible. Repeated analyses of samples stored in this man-
ner have not detected appreciable change in the iso-
topic composition over periods of up to five years.
Salinity was measured at each surface water sam-
pling location (except for the Everglades locations) by
Florida International University/Southeastern Environ-
mental Research Program (FIU/SERP). A summary of
these data has been published by Boyer et al. (1999).
Automated water salinity monitors were deployed at
three locations in Florida Bay: Lignumvitae Basin,
East Key and Manatee Bay. Hourly salinity data were
obtained using a Hydrolab Datasonde 3 between 1995
and 1998 in Lingnumvitae Basin. Similar data from
East Key and Manatee Bay were collected over shorter
time periods. Each instrument was serviced at 6 to 8
week intervals. Servicing consisted of cleaning organ-
isms which encrusted the housing and sensors, replac-
ing batteries, and calibrating the instrument for salin-
ity. Drift in the instrument within a given sampling
period was corrected by measuring the salinity and
temperature with the instrument at the beginning and
end of each deployment period and assuming a linear
instrument drift during the deployment period.
Precipitation samples were collected between Au-
gust 1997 and December 1998 from two locations on
the South Florida mainland. At the Redlands site, lo-
cated southwest of Miami, precipitation was collected
in a bottle equipped with funnel and tube. On a daily
basis, the contents of the bottle were transferred to
another bottle that was kept indoors and capped to
form a bi-weekly to monthly composited sample. Dai-
ly amounts of precipitation were recorded with a stan-
dard wedge gage. Precipitation was also collected be-
hind the Iori Building within Everglades National Park
from December 1997 through December 1998, using
an Aerochemetrics Wet/Dry collector. This collector
consisted of two buckets, one to collect dry deposition
and the other to collect wet deposition. The instrument
was equipped with a heated sensor that when wet ac-
tivated a mechanical arm that moved a cover from the
wet collection bucket to the dry collection bucket,
thereby exposing the wet collection bucket to rainfall.
The heated sensor allowed for evaporation of rainfall
that collected on it, and once dry, the mechanical arm
was moved again to cover the wet collection bucket
to prevent evaporation of the rainfall sample. Amount
of precipitation was measured directly from the wet
collection bucket using a ruler.
CARBON ISOTOPIC COMPOSITION OF DISSOLVED
INORGANIC CARBON (DIC)
The 5'°C of waters was measured using two meth-
ods.
Method 1: The 6'°C of the DIC for samples col-
lected early in the project were determined by acidi-
fication under vacuum. Prior to analysis a sample (5
cc) was withdrawn from the serum bottle through a
septum. The displaced volume of water was replaced
by nitrogen gas. To acidify the sample it was injected
into a sealed vessel through a septum which contained
0.5 cc of H,PO,. The sample and acid were then re-
moved from the vacuum line and thoroughly shaken.
The vessel was then reattached to the vacuum line and
after evacuation the CO, produced was allowed to ex-
pand through a water trap and then frozen with liquid
nitrogen into a vacuum vessel. The vessel was subse-
quently attached to a stable isotope mass spectrometer
(Finnigan-MAT 251) and the ratios of C/'*C and !8O/
'©Q determined in a similar manner to the CO, pro-
duced from the acidification of carbonate. Standards
were made by dissolving NaHCO, of known 8C in
18.3 Mohm water to make a 4 mM solution. Stan-
dardization is relative to PDB and data are reported in
conventional 6 units. Reproducibility was determined
by replicate analysis of the same sample and has been
measured at + 0.1%0.
Method 2: The 5%C of the DIC for samples col-
lected during the latter stages of the project was de-
termined using a continuous flow mass spectrometer.
In this technique a vial with a septum containing 0.5
cc of ortho-phosphoric acid was evacuated. Two-cc of
the sample was injected into the vial and thoroughly
shaken. The sample was then placed along with other
samples and standards in an autosampler (Gilssen) at-
tached to a continuous flow isotope ratio mass spec-
trometer (Europa 20—20). Each sample in turn was
purged by helium and the water removed using mag-
nesium perchlorate. The CO, produced by the reaction
of the acid with the dissolved inorganic carbon (DIC)
was passed through a packed GC column (75°C) and
then analyzed using the mass spectrometer. In this
method up to 200 samples can be processed in one
batch. The precision of this method is approximately
the same as in the acidification under vacuum method
(+ 0.1%).
STABLE ISOTOPIC VARIATION: SWART ET AL. 21
OXYGEN ISOTOPIC COMPOSITION OF WATERS
The 5!8O of water was determined using the Epstein
and Mayeda (1953) method of equilibration. Two ad-
aptations of this method were used.
Method 1: In this technique, | cm? of sample was
placed in a vacuum vessel and all air removed through
a process of freezing, pumping and thawing. After all
air was removed, each flask was flooded in turn by |
atm of CO, and the flask sealed. The water and the
carbon dioxide were then allowed to equilibrate at
40°C overnight. In this process the oxygen in the water
isotopically exchanged with the CO,. The CO, then
attained an isotopic composition representative of the
water, although offset through a fractionation factor. In
each run 14 samples and 2 standards were equilibrated
with CO,. Standards are related to V-SMOW through
simultaneous equilibration with samples of V-SMOW
distributed by the National Bureau of Standards
(NBS). Reproducibility was determined by replicate
analysis of the same sample and has been measured at
+ 0.1%.
Method 2: \n this method the entire process includ-
ing the flushing of the samples with CO,, equilibration,
and extraction was handled automatically. The process
was controlled by a computer interfaced to a mass
spectrometer (Europa GEO) and an equilibration unit
(Europa WES). In each experiment, 0.5 cc of sample
was placed in a small serum vial with a septum top.
These samples were flushed with CO, at atmospheric
pressure and allowed to equilibrate for 12 hours at
35°C. The CO, was then extracted through a cryogenic
water trap and the isotopic composition determined us-
ing a mass spectrometer. A series of 59 specimens (in-
cluding samples and standards) was analyzed in a 12
hour period. The reproducibility was approximately
0.08%c.
HYDROGEN ISOTOPIC COMPOSITION OF WATERS
The 5D of the water was determined using two
methods. Initial samples were analyzed using the first
method described for each of the elements. Later sam-
ples, 1995 to the present, were analyzed using the sec-
ond method. In both methods, data are reported rela-
tive to V-SMOW.
Method 1: The first method utilized the reduction of
H,O to H, gas using uranium at 800°C (Friedman,
1953). The hydrogen gas was then analyzed using a
stable isotope mass spectrometer (Finnigan MAT 251).
Method 2: The second method used an equilibration
technique between H, gas and H,O in the presence of
a platinum catalyst (Hokko beads) (Coplen er al.,
1991). This method was used in conjunction with the
second method of determining the 6'°O of water using
the GEO and WES. In this method the same samples
which were used for the determination of the 6!°O
were flushed with H,. The equilibration between H,
and H,O occurred in a period of 30 minutes, allowing
for analysis of the first equilibrated samples immedi-
ately following flushing of the last sample H,. Preci-
sion of this method determined by replicate analyses
of the standards within each run is approximately
1.5%o.
CORALS
A specimen of Solenastrea bournoni was cored in
Lignumvitae Basin of Florida Bay in 1986 and 1993.
Isotopic analyses from this specimen were reported by
Swart et al. (1996, 1999). In 1996, a specimen of So-
lenastrea hyades was collected from Barnes Sound,
near the mouth of Manatee Bay in northeast Florida
Bay.
Coral cores were sliced into sections parallel to the
axis of coral growth. Core sections were then subjected
to X-radiography to facilitate the construction of an-
nual coral chronologies. Skeletal coral carbonate was
sampled for 6'°C and 6'8O using methods described
previously (Swart ef al., 1999). A powdered calcium
carbonate was collected in transects from coral slabs
using a computer-controlled drilling apparatus. All car-
bonate samples were processed on a Finnigan MAT
251 Mass spectrometer connected to an automated ex-
traction devise at the University of Miami. Data are
presented in standard delta notation relative to PDB.
RESULTS
IsoTOPIC COMPOSITION OF WATERS
The 6'8O of precipitation reported in this study in
the South Florida area is highly variable. Despite this
variability, the data fall near the MWL and have vol-
ume weighted mean values for 6'°O and 6D of —2.9%c
and —9.9%c, respectively (Table 1). The 6!*O data
agree well with precipitation data collected between
1983 and 1988 (Swart ef al., 1989) that have mean
5'8O and 6D values of —2.7%c and —11.4%c, respec-
tively. Actual 6'8O values of the precipitation ranged
from values as negative as —6%c during the wet season
to values close to 0%c during the drier portions of the
year.
Water from the Everglades is significantly enriched
in 'SO and D with respect to precipitation (Table 1).
Mean 56!8O for the Everglades samples is +0.55%c,
compared to —2.9%c for precipitation. The highest val-
ues occur during the end of the dry season (March—
May) with isotopically more negative values in the late
wet season (Text-fig. 2). The 6!°O and 6D data fall on
a line with a slope of 5.8 (R* = 0.75), less than the
slope of 8 for the MWL (Text-fig. 3). Although 6°C
22 BULLETIN 361
Table |.—Statistical summary of stable isotope and salinity data for surface waters by area and precipitation in South Florida.
Number of Mean S.D. (a) Mean S.D. Mean S.D. Mean S.D. (a)
Area samples ur) oe) 6D beC 6C Salinity Salinity
Everglades 221 0.55 1.59 6.75 10.83 =e). Ox (c) (c)
Florida Bay (total) 1318 1.62 1.00 ililesis) 6.66 ENT) 2.32 27.9 9.6
Inner 427 1.76 1.06 11.83 7.30 3-59 2.16 21.6 9.2
Eastern Middle 240 1.80 0.96 11.64 7.39 2,95 2.45 31.5 6.8
Western Middle 194 1.55 0.67 10.99 5.63 =) 7/7) 2.35 3319) 3.6
Outer 15] 1.41 0.60 9.50 4.99 2,23) 2.39 33.8 2.9
Whitewater Bay 247 1.58 1.19 10.07 8.07 133 1.85 13.5 10.1
Ten Thousand Islands 259 1.06 1.31 TEMP 7.89 =o) 2.34 21.6 11.6
Precipitation (b) 23 = 2.9 1.48 = 9195) 11 (c) (c) (c) (c)
(a) Standard deviation.
(b) Isotopic means for precipitation are weighted by the amount of precipitation.
(c) Not measured.
* Meyers (1990).
of Everglades waters were not measured in this study,
they were reported in Meyers (1990) and included in
Table | for comparative purposes. In general, the iso-
topic composition of DIC in Everglades waters were
negative with a mean 6'°C of —6.9%c (Meyers 1990).
The 6'8O of the water in all of the Florida Bay sam-
ples ranges from —2.9 to +5.7%c with a mean of
+1.62 (Table 1). The 5'%C for all of the Florida Bay
samples (—2.87%c) is higher than that reported for the
Everglades (—6%c; Table 1). A time series of the mean
d!°C values for all stations in South Florida indicate a
minimum in 1996 (Text-fig. 4).
The mean 46'°O for the Whitewater Bay and Ten
Thousand Islands areas are positive and +1.58%ec and
+1.06%c, respectively (Table 1). In contrast, the mean
56°C for both Whitewater Bay and the Ten Thousand
Islands area is significantly negative with values of
—7.33%e and —5.76%oc, respectively (Table 1). These
values are similar to those reported for the Everglades
(Table 1) by Meyers (1990).
COMPARISON OF OXYGEN AND CARBON ISOTOPIC
COMPOSITION WITH SALINITY
In order to examine the variation in the relationships
between 6!°O, 61°C, and salinity further, we have sep-
arated Florida Bay into four regions: Inner Florida
Bay, Eastern-middle, Western-middle, and Outer Flor-
ida Bay (Text-fig. 1). Salinity within these regions in-
creases from east to west across Florida Bay (Table 1),
typical of an estuary. The Inner Florida Bay region
has the lowest mean salinity of 21.6, while the West-
ern-Middle and Outer Florida Bay regions have a sim-
ilar mean salinities of 31—33. In general, the mean 6'°O
decreases from east to west across Florida Bay from
+1.8 to +1.4%c (Table 1).
The relationships between 6!°O and 6%C with salin-
ity of South Florida estuarine waters are investigated
by using 5!8O vs salinity and 6'3C vs salinity plots and
then fitting a linear regression through the data (Table
2). When the 6!°O for each site in Florida Bay is avy-
eraged on a monthly basis and compared to the cor-
responding mean monthly salinity for each station,
there is a positive linear correlation (R* = +0.36; p <
0.01) (Text-fig. 5). There is a positive linear relation-
ship between 6!°O and salinity for each of the four
regions of Florida Bay (Table 2; Text-fig. 6). Although
the linear regression can explain only 30 percent or
less of the variation between 5'8O and salinity in the
Florida Bay regions (Table 2), each of the linear re-
gressions are significant at p < 0.01. The linear rela-
tionship between 6!*O and salinity was strongest for
the two middle regions of Florida Bay (R* = 0.3), and
weakest for the Outer region of Florida Bay (R? =
0.09).
There is a variation in the intercepts of the linear
regressions for each of the regions in Florida Bay. The
linear regressions for the Inner and Eastern-middle ar-
eas of Florida Bay predict a 6!°O at zero salinity of
+0.33%c and —0.83%c, respectively. The predicted in-
tercept of the linear relationship at zero salinity for the
Western-middle and Outer areas of Florida Bay are
6.00
38'°0 ( Too)
-6.00
Apr-95 Oct-95 May-96 Dec-96 Jun-97 Jan-98
Shark S aylor
Text-figure 2.—Time series of mean 6'SO for Everglades. Error
bars represent + | standard deviation.
STABLE ISOTOPIC VARIATION: SWART ET AL. 23;
6D
o
-6 -4 -2 0 2 4 6
d°O1
Text-figure 3—Oxygen and hydrogen isotopic composition of
Everglades water. Black circle represents volume weighted average
isotopic composition of rainfall collected in this study. Open triangle
represents isotopic composition of rainfall reported by Swart er al.,
1989.
significantly more negative than zero at —2.39%c and
—1.07%c, respectively.
A positive linear relationship exists between 6!*O
and salinity for the Whitewater Bay and Ten Thousand
Islands areas (Table 2). The linear correlation coeffi-
cient (R*) for the Whitewater Bay and Ten Thousand
Islands is 0.34 and 0.28, respectively (p < 0.01). The
predicted intercept of the linear relationship at zero
salinity is close to zero for both the Whitewater Bay
(+0.35%c) and Ten Thousand Islands areas (—0.2%c,
Table 2).
There is no strong linear correlation between 61°C
and salinity throughout all of Florida Bay (Text-fig. 7).
Similarly there is no strong linear correlation between
56°C and salinity within the four regions of Florida Bay
(Table 2; Text-fig. 8), except for the Inner Florida Bay
area. The Inner Florida Bay area, Whitewater Bay area
and the Ten Thousand Islands areas have similar cor-
relations (R* = 0.2; p < 0.01) between 61°C and salin-
ity.
CORALS
The 65°C and 8'8O data for the skeleton of the coral
collected from Lignumvitae Basin have been reported
previously (Swart er al., 1996; 1999). The average an-
Table 2.—Results of linear regression analyses.
5'8O vs. salinity
Area Intercept Slope R? Number
Florida Bay
Inner +0.33 0.06 0.26 S3}7/
Eastern Middle —0.83 0.08 0.31 293
Western Middle —2.39 0.11 0.30 242
Outer == 07) 0.09 0.09 194
Whitewater Bay +0.35 0.07 0.34 395
Ten Thousand Islands —0.2 0.06 0.28 395
0.077
-2.0
-6.07]
-8.0 T l T
Oct-93 Jun-95 Feb-97 Oct-98 Jun-00
Text-figure 4+—Five month running mean of the 6'°C composition
of the DIC in all regions studied. Error bars represent + one standard
error.
nual values for both corals are shown in Text-figure 9.
The Manatee Bay coral (—2.58 + 0.3%c) is slightly
isotopically enriched in 6!°O relative to the Lignum-
vitae coral (—2.04 + 0.5%c), and has more negative
63C values (—6.45 + 0.43%o vs. —3.25 + 0.58%o for
the Manatee Bay coral).
CONTINUOUS SALINITY MONITORING DATA
Continuous salinity data collected from Lignumvi-
tae Basin in Florida Bay are shown in Text-figure 10.
The continuous data from Lignumvitae Basin have
been averaged into monthly values and are plotted
with the FIU/SERP data. The FIU /SERP data repre-
sent water samples collected on a single day within
the month. Generally the agreement between the two
datasets is excellent considering the different temporal
resolution between them.
DISCUSSION
SOURCES OF SALINITY VARIATION
Relatively few studies have examined the 6'*O and
6D of precipitation in South Florida. The data pre-
dC vs. salinity
P Intercept Slope R? Number 1p
<0.01 =3);7/5) 0.11 0.20 516 <0.01
<0.01 —6.54 0.12 0.11 296 <0.01
<0.01 —7.99 0.16 0.05 244 <0.01
<0.01 — 10.90 0.26 0.10 184 <0.01
<0.01 =8.33 0.10 0.20 364 <0.01
<0.01 STO 0.09 0.20 336 <0.01
24 BULLETIN 361
6.00 4
4.00 —
4
—~ |
~— 200
we 7
0.00 +
-2.00
T T T i T T T ] T ]
0.00 10.00 20.00 30,00 40.00 50.00
Salinity
Text-figure 5—Mean monthly 6'*O values versus salinity for all
sites in Florida Bay. Error bars represent +1 one standard deviation.
sented in this study show that the isotopic composition
of South Florida precipitation falls near the MWL
(Text-fig. 3). It is relatively depleted with volume
weighted means of 5D and 8'8O —9.95%ec and
—2.90%oc, respectively (Table 1).
Based on a correlation between 6D and 6!8O, the
data from the Everglades plot below the MWL (Text-
fig. 3). Deviation away from the MWL is a well-
known phenomenon and arises through the differential
influence of evaporation of the oxygen and hydrogen
isotopes. In the case of the Everglades data, the close-
ness of the slope of the best fit line through the data
Salinity
NR
uo
8'°C (°/o0)
Text-figure 7—Mean monthly 5'°C versus salinity for all sites in
Florida Bay.
(5.8) to the slope of the MWL, suggests evaporation
at a high relative humidity (Gonfiantini, 1986). Meyers
et al. (1993) predicted a mean relative humidity of
80% for the Everglades based upon a stable isotope
evaporation model. The time series trend of the Ev-
erglades data (Text-fig. 2) shows some seasonality with
isotopically more positive values toward the end of the
dry season when evaporation typically exceeds precip-
itation.
The intercept of the best fit line through the Ever-
glades data with the MWL is similar to the isotopic
composition determined for South Florida precipita-
tion (Text-fig. 3). These results indicate that precipi-
tation is the ultimate source of the evaporated water in
the Everglades, Everglades water can be distinguished
isotopically from its parent precipitation. The mean
5'8O and 5D of the Everglades water is +0.55%ce and
Outer FB
White Water Bay
SALINITY
30 40 50 0 10 20 30 40 50
Ten Thousand Islands
Text-figure 6.—Comparison of the linear relationship between 8'8O and salinity for the six major areas designated in Text-figure 1.
J 4) g g
STABLE ISOTOPIC VARIATION: SWART ET AL. 25)
a) “ee
Tra) a
a 10 20 30 40 50
3 Eastern Middle FB Western Middle FB
oe 2
°
oO
0 0
Qye2
-4
-6
-8 WA
A
-19 14 *
0 10 20 30 40
Outer FB
White Water Bay
SALINITY
Ten Thousand Islands
Text-figure 8.—Comparison of the linear relationship between carbon isotopic composition (DIC) and salinity for the six major areas
designated in Text-figure 1.
+6.75%c, respectively, as compared to —2.90%c and
—9.95%c for the precipitation.
Florida Bay exhibits a wide range of 5!°O values
related to evaporation, precipitation, and mixing with
water derived from the Everglades, the Gulf of Mex-
ico, and the Florida Reef Tract. The wide range of 6!°O
values was noted by Lloyd (1964) who was unable to
exhibit any relationship between salinity and 5!8O. In
contrast to the study of Lloyd (1964), the data in this
study were collected within a temporal and spatial
framework and consequently the apparent poor corre-
lation between salinity and 6!°O noted by Lloyd can
be explained by mixing between different end-mem-
— -1.00 + Bay
~~
Zo) -5.00
0.00 2
>
ss |
<< a) =
Oo
© 1000
Manatee Bay
Text-figure 9 —Oxygen and carbon isotopic data from corals col-
lected from Lignumvitae Bay and Manatee Bay.
bers with similar salinity, but differing 6'°O values.
These relationships can be visualized on a 6!*O vs.
salinity plot (Text-fig. 11) comparing the 6'°O com-
position and salinity of the relevant end-members wa-
ters: Everglades, precipitation, and seawater. Data for
the seawater end-member are derived from previous
studies (Ortner et al., 1995; Leder et al., 1996; Swart,
40.00
This Paper
36.00
Salinity
32.00
28.00
Jun-95 Apr-96 Feb-97 Dec-97 Oct-98
Date
Text-figure 10.—Comparison of data collected from the continu-
ous salinity monitoring stations and data collected by FIU on a
monthly basis. Salinity data have been averaged from hourly data
into monthly values. Error bars represent + one standard deviation.
26 BULLETIN 361
Oxygen
oOo
|
1
'
1
!
t
1
I
1
I
I
1
!
'
!
1
!
Nal
‘1
aE
>
a Gulf of Mexico/Atlantic Water
0 10 20 30 40 50
Salinity
Text-figure 11.—Model for the evolution of waters in the coastal
region of Florida Bay showing the salinities and 6'O of three end
members. Marine water from the Gulf of Mexico and the Atlantic
(A) has a salinity of approximately 35.5 and a 6'*O of +0.5 to
+1.5%c. This mixes with (B) rainwater (6'8O = 0 to —6%c; mean =
—2.9%c) or water from the (C) Everglades (6'9O = —4.5 to +
+4.5%e; mean = +0.55%c). Evaporation can take place at any time
from any mixture. Evaporation of the marine water will produce a
trend along a line (D).
unpublished data), and have a 6!°O of between +0.5
and +1.5%c and a salinity between 35 and 36. The
model proposed herein suggests that South Florida es-
tuarine waters plotting along a linear correlation be-
tween 6'8O and salinity with an intercept of 0 to + 1%c
reflect mixing of seawater with Everglades runoff. Es-
tuarine waters plotting along a linear correlation with
an intercept near —2%c or lower indicate mixing of
seawater with precipitation. Variation in the intercept
between these two extremes reflects varying contri-
butions from both runoff and precipitation and scatter
in these relationships reflect the presence of multiple
sources and processes such as evaporation (Text-fig.
11).
The intercept of the linear correlation between 6'*O
and salinity changes from being more negative in the
western areas (Outer and Western-middle) of Florida
Bay to being more positive in the eastern areas (East-
ern-middle and Inner) of Florida Bay (Table 2; Text-
fig. 11). These results indicate that inputs of precipi-
tation is most responsible for varying salinity in the
western areas of Florida Bay. Conversely, runoff from
the Everglades is more important for the variation in
salinity in the eastern portion of Florida Bay. For
Whitewater Bay and the Ten Thousand Islands areas,
the intercept of the correlation between 5'*O and sa-
linity is close to zero (Table 2) suggesting that the
salinity variation in these regions is principally a result
of runoff from the Everglades, and in particular from
Shark Slough.
SOURCES OF CARBON ISOTOPIC VARIATION
The 65°C of the DIC in all the surface waters from
the estuarine environments was highly variable and
showed a limited relationship with salinity (Table 2
and Text-fig. 7 & 8). Stronger correlations between
5'C and salinity were observed in the areas most in-
fluenced by runoff from the Everglades. These areas
include the Inner area of Florida Bay, White Water
Bay and the Ten Thousand Islands. Although the 6'°C
of the Everglades was not measured routinely through-
out this study, measurements that were made on se-
lected samples from the Everglades reveal that these
waters had negative 6'°C values (Meyers, 1990).
Hence when these waters flowed into the marine en-
vironment, the result was a strong correlation between
dC and salinity. In contrast, the absence of a strong
relationship between 6'°C and salinity in the more sa-
line portions of Florida Bay supports the relative lack
of influence of Everglades waters on these portions of
Florida Bay.
Despite the absence of a correlation between 64°C
and salinity in the central and western areas of Florida
Bay sites, the waters at these sites show a considerable
amount of variability in 6'°C which is probably related
to local remineralization of organic material. The most
negative 61°C values that are attained in Florida Bay
(mean = ~—3%c) is a reflection of the more marine
nature of this organic material compared to the White
Water Bay and Ten Thousand Island area (6'"C be-
tween —S5%o and —7%c) where most of the organic
material is derived from terrestrial sources.
CORALS
Considering the relatively poor correlation between
salinity and 6'°C of DIC, there appears to be little util-
ity in relating the 6'°C of calcareous skeletons to input
of freshwater except in the environments which are the
closest to the Everglades. In the central areas of Flor-
ida Bay, where it has been established that there is a
relative absence of freshwater input using the 6!°O
data, the changes in the 5'°C of calcareous organisms
can be interpreted as reflecting increased oxidation of
organic material, perhaps leading to an increase abun-
dance of nutrients.
The correlation between the 6'C and 5!*O of coral
skeletons can be used as an indicator of the relative
influence of freshwater. Corals from areas where there
is relatively little influence of freshwater will have rel-
atively little correlation between the two isotopes in
STABLE ISOTOPIC VARIATION: SWART ET AL. Di
-2.00 +
-4.00 4
4
Gam
wy | ~)
O -6.00
a | .
Ze) 4 ;
. Manatee Bay
192=0.18
Bape
8.00 oa
-10.00 ; 1 ] T ]
-6.00 -4.00 -2.00 0.00 2.00
8 OC»)
Text-figure 12.—Relationships between 6'°C and 6'%O in the coral
skeletons from Lignumvitae Basin and from Manatee Bay. Note that
in the Lignumvitae data there is no correlation between the two
isotopes (R* = 0.04)as predicted from an absence of a correlation
in the two isotopes in the water at that site. In Manatee Bay however
there is a correlation between the two isotopes which is statistically
significant at the 99% level (R* = 0.14).
their skeletons, while corals influenced by runoff
should have a better correlation. These correlations can
be seen in data from the coral collected from Manatee
Bay which shows a strong positive correlation between
skeletal 5'°C and 8!8O (R = .42, n = 362, p > .0001).
In contrast, corals from Lignumvitae Basin show no
correlation between the two isotopes (R = 0.003, n =
1894, p = not significant) (Text-fig. 12). These cor-
relations are similar to the relationships seen between
the 6'°C and 6!*O of waters. For example, results from
a linear regression model between 5'°C of DIC and
salinity has an R* of 0.2 (Table 2). In western-middle
Florida Bay, where the Lignumvitae Basin is located,
an R? of 0.05 is obtained (Table 2).
APPLICATION TO EXISTING PALEOENVIRONMENTAL DATA
Combining the salinity relationship beteen Lignum-
vitae Basin along with 6'8O record of the coral skel-
eton collected from Lignumvitae Basin, attempts have
been made at estimating salinity variations in other
regions of Florida Bay (Swart ef al., 1996; 1999).
Based on data presented in this paper, however, the
equation linking the skeletal 6'°O and salinity in Lig-
numvitae Basin (Swart ef al., 1999) is not valid for
other areas of Florida Bay. Consequently different
equations for estimating salinity from 6!*O need to be
derived for every location. While this is impractical in
real terms, One can obtain an idea of the influence of
the variations by calculating the relationships between
salinity and 6!°O for the different geographic sub-re-
gions designated for Florida Bay (Text-figs. 1, 6). As
an example of the influence that these different rela-
tionships have on the calculation of the salinity from
a 6'8O record, we examined the 6!8O record of a coral
from Lignumvitae Basin and a coral of a similar genus
growing in Barnes Sound near the mouth of Manatee
Bay. In order to compare the salinity estimates, we
used three methods to determine the salinity at the
mouth of Manatee Bay. These are (1) the transfer func-
tion approach presented by Swart er al. (1999) using
the 6'8O of coral skeletons growing in Lignumvitae
Basin and the modern salinity correlations between the
two basins, (2) using a coral actually growing in a site
and the correlation between salinity and 6!*O for Lig-
numvitae Basin, and (3) using the Manatee Bay coral
and the salinity vs. 6!°O relationship presented in this
paper. We then compared these salinity estimates di-
rectly with measurements by FIU/SERP and the Uni-
versity of Miami in Barnes Sound. In the first com-
parison, the data from the coral growing in Lignum-
vitae Basin was converted to salinity using the equa-
tion presented by Swart et al. (1999) which related
salinity to the temperature (T) and the 5'*O of the coral
skeleton (equation 1). This provided an estimate of
salinity with a slightly greater amplitude than the Lig-
numvitae coral, but significantly higher than the range
in salinity measured in Manatee Bay (Text-fig. 13).
Salinity = 4.9885!°O iP dlaiwe se IS oy? (1)
coral
In the second approach equation | was also used, but
in this case the data from the coral actually growing
near the mouth of Manatee Bay was used (Text-fig.
13). This gave an estimate similar in pattern to the
method used previously, but with much higher salinity
estimates than the salinities actually measured (Text-
fig. 13). The third method uses the new relationship
between salinity and 6!*O,,,,.. shown in Table 4. In or-
der to convert the 6!°O of the coral skeleton to salinity
we used the relationship established by Leder ef al.
(1996) between temperature and skeletal 5'°O for the
species Monastrea annularis and assumed an annual
mean temperature of 25°C. Combining these two equa-
tions a new formula can be derived which is specific
for corals growing in the inner portion of Florida Bay.
This relationship is shown in equation 2.
Salinity = 4.148'8O.,,4) + 0.978T + 9.33 (2)
cora
The results of these three calculations together with
the range of salinities actually measured at the site in
Manatee Bay are shown in Text-figure 13. This visual
comparison shows that the best estimate of salinities
28 BULLETIN 361
40
30
Salinity
1965 1975 1985 1995
Date
Text-figure 13.—Line (1) represents the salinity in Barnes Sound
estimated using the 6'SO data from the coral in Lignumvitae Basin
(Swart er al., 1996, 1999). Line (2) represents the salinity estimate
calculated from the 6'°O of the Manatee Bay coral using the rela-
tionship between salinity and 6'*O,,,., established for Lignumvitae
Basin. Line (3) is the new relationship between salinity and 6!°O for
the inner region of Florida Bay established in this paper (Table 4).
The ranges in salinity measured by FIU and presented in this paper
are indicated by the arrow.
measured at the site are provided by the method in
which we used the correlation between salinity and
5!°O, er for inner Florida Bay. Previously estimates for
the salinity in this portion of Florida Bay are therefore
likely to have been too high (Swart et al., 1999). How-
ever, in spite of the distance between the two locations
(Lignumvitae Basin and Manatee Bay), the coral in
Lignumvitae Basin shows a similar temporal pattern
in 6'8O to the Manatee Bay Coral and with the appro-
priate calibration the Lignumvitae coral record (Swart
et al., 1996) can be used to estimate salinities in other
portions of Florida Bay.
CONCLUSIONS
1) Relationships between salinity and 6!8O vary ac-
cording to relative distance from the Everglades. This
changing relationship is caused by an increasing influ-
ence of rainfall compared to runoff in controlling sa-
linity variations. For example, the western portion of
Florida Bay appears not to be significantly influenced
by runoff from the Everglades and salinity variations
there are related to changes in the amount of precipi-
tation. Application of a constant salinity vs. 5!°O re-
lationship in the interpretation of 6'°O values of cal-
careous organisms from South Florida estuaries is not
appropriate.
2) Carbon isotopic variations in Florida Bay are re-
lated to the oxidation of organic material. Large vari-
ations in the 6'°C of the DIC occur both within saline
and freshwater environments. Within the saline envi-
ronments, there is little relationship between salinity
and 6'°C while in the freshwater environment, there is
a strong correlation between salinity and 54C.
REFERENCES CITED
Boyer, J.N, Fourqurean, J.W., and Jones, R.D.
1999. Seasonal and long-term trends in the water quality of
Florida Bay (1989-1997). Estuaries, vol. 22, pp. 417—
430.
Coplen, T.B., Woldman, J.D., and Chen, J.
1991. Improvements in the gaseous hydrogen-water equilibra-
tion technique for hydrogen isotope ratio analysis. Ana-
lytical Chemistry, vol. 63, pp. 910-912.
Craig, H., and Gordan, L.I.
1965. Deuterium and oxygen-18 variations in the oceans and the
marine atmosphere. in Stable Isotopes in Oceanographic
Studies and Paleotemperatures, E. Tongiorgi, ed., Consig-
lio Nazionale delle Ricerche, Laboratori Geologia Nucle-
are, Pisa, pp. 9-130.
Davis, R.A., Jr.
1997. Geology of the Florida Coast. in The Geology of Florida,
Randazzo, A.F and D.S. Jones, eds., University Press of
Florida, Gainesville, pp. 155—168.
Epstein, S., and Mayeda, T.
1953. Variations in O'* content of waters from natural sources.
Geochimica et Cosmochimica, vol. 27, pp. 213-224.
Fennema, R.J., Neidrauer, C.J., Johnson, R.A., MacVicar, T.K.,
and Perkins, W.A.
1994. Understanding the Ecosystem. in The Everglades Hand-
book, Thomas E. Lodge, ed., St. Lucie Press, Delray
Beach, FL, pp. 249-289.
Fourqurean, J.W., and Robblee, M.B.
1999. Florida Bay: A history of recent ecological changes. Es-
tuaries, vol. 22, pp. 345-357.
Fourqurean, J.W., Zieman, J.C., and Powell, G.V.N.
1992. Phosphorus limitation of primary production in Florida
Bay: Evidence from the C:N:P ratios of the dominant sea-
grass Thalassia testudinum. Limnology and Oceanogra-
phy, vol. 37, pp. 162-171.
Friedman, I.
1953. Deuterium content of natural waters and other substances.
Geochimica et Cosmochimica Acta, vol. 4, pp. 89-103.
Gat, J.
1981. Isotopic fractionation. in Isotope Hydrology: Deuterium
and Oxygen-18 in the Water Cycle, J. Gat, and R. Gon-
fiantini, eds., Vienna, IAEA, pp. 21-31.
STABLE ISOTOPIC VARIATION: SWART ET AL. 29
Gonfiantini, R.
1986. Environmental isotopes in lake studies. in Handbook of
Environmental Isotope Geochemistry A., Volume 1, P.
Fritz, and J.C. Fontes, eds., Amsterdam, Elsevier, pp.
113-168.
Halley, R.B., and Roulier, L.M.
1999. Reconstructing the history of eastern and central Florida
Bay using mollusk-shell isotope records. Estuaries, vol.
22, pp. 358-368.
Kushlan, J.A.
1990. Freshwater Marshes. in Ecosystems of Florida, Myers,
R.L. and J.J. Ewel, eds., University of Central Florida
Press, Orlando, FL., pp. 324-363.
Leach, S.D., Klein, H., and Hampton, E.R.
1972. Hydrologic effects of water control and management of
oS —)
southeastern Florida, Florida Bureau of Geology Report
of Investigations, vol. 60, 115 pp.
Leder, J.J., Swart, P.K., Szmant, A, and Dodge, R.E.
1996. The origin of variations in the isotopic record of Scler-
actinian corals: | Oxygen, Geochimica et Cosmochimica
Acta, vol. 60, pp. 2857-2870.
Lloyd, M.R.
1964. Variations in the oxygen and carbon isotope ratios of Flor-
ida Bay mollusks and their environmental significance.
Journal of Geology, vol. 72, pp. 84-111.
Meyers, J.B.
1990. Stable isotope hydrology and diagenesis in the surficial
aquifer system, southern Florida everglades. Unpublished
M.S. thesis, University of Miami. 92 pp.
Meyers, J. B., Swart,P. K., and Meyers, J.L.
1993. Geochemical evidence for groundwater behavior in an un-
confined aquifer, south Florida. Journal of Hydrology, vol.
148, pp. 249-272.
Nelsen, T.A., Wanless, H.R., Trefry J.H., Alvarez-Zarikian, C.,
Hood, T., Blackwelder, P. Swart, P., Tedesco, L.,
Kang, W-J., Metz, S., Garte, G., Fetherstone, C.,
Souch, C., Pachut, J.F., O’Neal, M., and Ellis, G.
2001. Linkages between the South Florida peninsula and coastal
zone: A sediment-based history of natural and anthropo-
genic influences. in Linkages Between Ecosystems in the
South Florida Hydroscape: The River of Grass Continues.
K. Porter and J. Porter, eds., CRC Press, Boca Raton,
Florida (in press)
Nuttle, W.K., Fourqurean, J.W., Cosby, B.J., Zieman, J.C., and
Robblee, M.B.
2000. Influence of net freshwater supply on salinity in Florida
Bay. Water Resources Research, vol. 36, no. 7, pp. 1805—
1822.
Ortner, P.B., Lee, T.N., Milne P.J., Zidka, R.G., Clarke, M.E.,
Podesta, G.P., Swart, P.K., Testa, P.A., Atkinson, L.P.,
and Johnson, W.R.
1995. Mississippi river flood waters reached the Gulf Stream,
Journal of Geophysical Research, vol. 100, pp.13,595—
13,601.
Rozanski, K., Aragus-Aragus, L., and Gonfianitini, R.
1993. Isotopic patterns in modern global precipitation. in Cli-
mate Changes in Continental Isotopic Records, P. Swart,
K. Lohmann, J. McKenzie, and S. Savin, eds., Volume
78: Geophysical Monograph, Washington, DC, American
Geophysical Union, pp. 1-36.
Smith, T.J. II, Hudson, J.H., Robblee, M.B., Powell, G.V.N.,
and Isdale, P.J.
1989. Freshwater flow from the the Everglades to Florida Bay:
A historical reconstruction based on flourescent banding
in the coral Solenastrea bournoni. Bulletin of Marine Sci-
ence, vol. 44, pp. 274-282.
Swart, P., Sternberg, D., Steinen, R., and Harrison, S.
1989. Controls on the oxygen and hydrogen isotopic composi-
tion of waters from Florida Bay, USA. Chemical Geology,
vol. 79, pp. 113-123.
Swart, P.K., Healy, G., Dodge, R., Kramer, P., Hudson, H., Hal-
ley R., and Robblee, M.
1996. The stable oxygen and carbon isotopic record from a coral
growing in florida bay: A 160 year record of climatic and
anthropogenic influence. Palaeogeography, Palaeoclima-
tology, Palaeoecology, vol. 123, pp. 219-238.
Swart, P.K., Healy, G., Greer, L., Lutz, M., Saied, A., Anderegg,
D., Dodge, R.E., and Rudnick, D.
1999. The use of proxy chemical records in coral skeletons to
ascertain past environmental conditions in Florida Bay.
Estuaries, vol. 22, pp. 384-397.
bh aac wee (4) CGS ea
i : ; ; eee Ome et —s UA slp at -"
a
reat tee
CHARTERS
SEDIMENT DYNAMICS OF FLORIDA BAY MUD BANKS ON A DECADAL TIME SCALE
CHARLES W. HoLMEs!, JOHN ROBBINS?, ROBERT HALLEY!, MICHAEL BOTHNER®,
MARILYN TEN BRINK*, AND MArRcI Marot!
'U.S. Geological Survey, 600 4" St. South, St. Petersburg, FL 34701
°Great Lakes Environmental Research Laboratory, (NOAA), Ann Arbor, MI 48105
5U.S. Geological Survey. Woods Hole, MA 02543
ABSTRACT
Ecosystem management requires knowledge of environmental dynamics. If historical environmental records do not exist, other
methods must be employed to obtain this information. A well-known geochemical procedure that supplies this type of data is
the use of natural radioactive nuclides to “date” the timing of events. Of the many naturally occurring nuclides, *!°Pb is the best
suited for gauging the timing of episodes in Florida Bay. The age-depth relationships were calculated using the *!°Pb method for
thirty-five sites within Florida Bay. The ages were independently confirmed by comparing the distribution of known concentrations
of atmospherically anthropogenic lead recorded in dated cores to similar data in an annually banded coral. Sediments in the
western and northern fringe of Florida Bay are accumulating at ~0.3 cm/yr, a rate similar to that of sea level rise. In the north-
central part of the bay, sediments are accumulating at a faster rate of ~1.0 cm/yr. The highest rate, =2.0 cm/yr was measured
in the northeastern part of the bay on the bank between Pass and Lake Keys. The rapid rate of accumulation in the northeastern
part of the bay permits the deciphering of biological and geochemical changes with an accuracy of about two years. In contrast,
the intermediate sediment rate in the central part of the bay provides adequate age-depth for relationships deciphering the
environmental record of the past 100 years.
INTRODUCTION
Any retrospective analysis of natural systems re-
quires a temporal frame of reference against which dy-
namic processes can be measured. The lack of ade-
quate historical environmental records within the Flor-
ida Bay system necessitated the development of a
method to determine the timing of environmental al-
terations. Conventional wisdom implies that the natu-
ral hydrology was significantly altered over the past
100 years. However, prior to any “‘restoration”’ effort,
it is prudent to determine what changes are the result
of natural processes and which are not. The keys to
this information lie within the sedimentary record. To
read this record properly, the first requirement is to
establish accurately a temporal frame of reference. A
program was initiated in 1994 to define the environ-
mental record with the sediments of Florida Bay. The
project was a cooperative between the South Florida
Water Management District, the U.S. Geological Sur-
vey, and the Great Lakes Environmental Research
Laboratory (N.O.A.A.).
Sediments have been accumulating in Florida Bay
for approximately 4000 years (Scholl, 1964). The bas-
al sediments are brackish and freshwater peats and
marls, similar to those presently accumulating in the
Everglades (Davies and Cohen, 1989). Overlying these
sediments are marine sediments that form a network
of mud banks and islands that partition the bay into
more than 30 small basins. The basins range from one
to four meters in depth and are up to a few kilometers
wide (Text-fig. 1). The bulk of the sediments is in the
banks and islands and is predominantly silt- and clay-
sized carbonate particles. Most of the mud is derived
from the physical breakdown of larger skeletal frag-
ments, with a contribution from calcareous algae that
produce micron-sized carbonate needles (Boscence,
1989a). The minor amount of sand-sized and coarser
material is the result of the debris from a great variety
of calcareous invertebrates living in the bay (Bosc-
ence, 1989a).
Although much work has been devoted to the origin,
composition, and distribution of sediments in Florida
Bay (Boscence, 1989b), the processes that have cre-
ated the mudbanks and their rate of formation are
poorly understood. It is commonly held that the south
or west sides of the mud banks are accreting while the
northern and eastern sides are eroding (Wanless and
Taggert, 1989). '*C dating of peats and carbonate sed-
iments provide a general picture of the time of sedi-
ment formation. However, these data do not address
the dynamics of sediment transport—particularly in re-
gard to the mobile, fine-grained particles. This infor-
mation is important in defining stratigraphic relation-
32 BULLETIN 361
a
* Coral Site
25°00 N
Key
Contours in feet
Core Sites
——~— + A 7-30 —
a) Mud Banks
a Core Locations
Nautical Miles
After USC &G SCharts 1249,1250
1. Buchanan Bank N. 10. Russell Bank S. 19. Russell Bank S.
2. Buchanan Bank S. 11. Johnson Key 20. Bob Allen Bank
3. Rabbit Key 12. Porjoe Bank 21. W. Pass Key
4. Cluett Key 13. Trout Creek Spit 22. Russell Bank
5. Whipray Basin 14. Little Madeira Bay 23. Park Key Bank
6. Bob Allen S. 15. E. Pass Key 24. Bob Allen Bank
7. Rankin Bight 16. Crocodile Point 25. Whipray Basin
8. Lake Ingraham 17. W. Pass Key
9. Russell Bank N. 18. Russell Bank S.
81°00 W 80:00 W
Text-figure 1—Location of coring sites in Florida Bay.
ships within the deposits. To obtain this information,
chronological methods using short-lived isotopes were
employed.
The requirements for an isotope to be a candidate
for dating sediments are that: (1) the chemistry of the
isotope (element) is known; (2) the half-life is known;
(3) the initial concentrations of the isotopes are known
or can be accurately estimated; (4) the only change in
concentration is due to radioactive decay; and (5) in
order to be useful, it must be relatively easy to mea-
sure. If all these conditions are met, the effective range
for each isotope is about eight half-lives. Four natu-
rally occurring isotopes (’Be, tC, '*’Cs, and 7!°Pb) sat-
isfy these criteria and have been found useful for mea-
suring sedimentary dynamics in Florida Bay over the
past 100 years.
*10Pb is a common radioisotope that has become a
mainstay of radiometric dating of sediment deposited
during the last century (Goldberg, 1963). This method
is based on a radioactive disequilibrium between 7!°Pb
(t;. = 22.3 yr) and its parent, *?°Ra (t,. = 1600 yr).
The radioactive disequilibrium is the result of the sep-
aration in the decay chain caused by the diffusion of
the intermediate daughter, *’Rn. The physical and
chemical process that lead to this disequilibrium have
been described and critiqued in reviews by Robbins
(1978) and Appleby and Oldfield (1992).
ACKNOWLEDGMENTS
We wish to acknowledge the efforts of Peter Swar-
zenski and James Flocks who made a significant con-
tribution to this report in discussions with the senior
author. We would also like to acknowledge Fran Glick
of Mariner Hospital al in Tavernier and Larry Wise of
Fisherman’s Hospital in Marathon for providing the x-
radiographs, which played a vital role in the selection
of cores. The South Florida Water Management Dis-
trict provided much of the funding for this work. Da-
vid Rudnick was the contracting official and his work
(and worry) is greatly appreciated. There are many
others, too numerous to mention here, who labored
long and hard under the hot sun and late into the even-
SEDIMENT DYNAMICS: HOLMES ET AL. 333}
ing to collect and section cores; to those people we
are extremely grateful.
PROCEDURES
SAMPLING PROCEDURES
Twenty-five locations were selected for coring in
Florida Bay to obtain a comprehensive overview of
the different environments (Text-fig. 1). In keeping
with the goal of the program—to develop a method of
establishing a sediment chronology—most coring sites
were located on the accretionary side of the mudbanks.
It is there that the potentially longest sediment record
is likely. Additional cores were collected in other en-
vironments, such as on the “erosional” side of a mud-
bank (Site 18—Russell Bank and Site 12—Porjoe
Bank), within the central part of selected basins (Site
5—Whipray Basin; Site 7—Rankin Bight; Site 8—
Lake Ingraham) and on the fringe of the Everglades
(Site 13—Trout Creek Spit, Site 14—Little Madeira
Bay and Site 16—Crocodile Point). Duplicate cores
were taken at selected sites. In all, 27 cores were an-
alyzed. The cores (10 cm i.d.) were collected with a
conventional piston coring device.
All cores were driven to the limestone surface and
averaged ~1.5 meters in length. Extreme care was ex-
ercised to preserve the sediment-water interface and to
minimize disturbance. Prior to sampling, each core
was x-rayed to determine the presence of laminations
and to evaluate the extent of mixing. If there was no
observable disturbance during sampling, handling, or
transport to the laboratory, the cores were sectioned
into 2 cm intervals for analysis. Cores located on
Buchanan Bank (North and South), on Cluett Key,
Russell Bank north, and on the eastern part of Pass
Key were disturbed and therefore were not analyzed.
LABORATORY ANALYSIS
Wet and dry bulk density, acid insoluble residues,
loss on ignition, and grain size analysis were deter-
mined for each interval (Holmes er al., 1998). Since
2!°Pb is adsorbed on the clay-sized particles, analyses
were performed on the <0.062 mm fraction. The total
71°Pb content of the sediment samples was measured
on freeze-dried subsamples by both alpha spectrome-
try, with selected samples also analyzed by gamma
techniques. The samples, which were >99% carbon-
ate, were “ashed” at 450°C for 6 hrs to destroy any
organic material, and dissolved in 8N HCl. The alpha
spectrometry procedure closely followed the method
detailed by Flynn (1968). Overall counting uncertain-
ties were +2%. Replicate samples were analyzed by
gamma methods at the Woods Hole USGS Laboratory
(M. Bothner and M. Ten Brink) as well as at the Great
Lakes Environmental Research Laboratory (J. Rob-
bins) and the results were within counting error of
measurements made by alpha methods. The **°Ra val-
ues were measured on selected samples by gamma
spectrometry. Further detail of the analytical proce-
dures and results from the replicate analyses are pre-
sented in Holmes er al. (1998).
RESULTS AND DISCUSSION
SITE CHARACTERIZATION
In this study, attention was given to the selection of
coring sites that were thought to contain well-pre-
served sediments and high accumulation rates. Coring
focused on the recovery of finely laminated sediments,
which generally indicate a lack of physical or biolog-
ical mixing (Ball et al., 1967; Shinn, 1968; Wanless
and Taggert, 1989). Three factors are thought to con-
tribute to the formation of the laminated sediments in
Florida Bay:
1. Rapidly deposited mud becomes quickly de-wa-
tered by as much as 50% so that they become suffi-
ciently cohesive to exhibit pseudo-conchoidal fracture
and can resist erosion.
2. The interior of the Bay is characterized by ex-
treme variations in temperature and salinity relative to
the open Gulf of Mexico or the Atlantic Ocean. These
fluctuations limit the diversity of burrowing organisms
that can live in the Bay.
3. Florida Bay sediments are rich in sulfide (SO0—
5000 4M), which inhibits biological activity (William
Orem, 1998, personal communication).
The combination of these factors creates an envi-
ronment wherein short-lived radionuclides can be used
successfully to measure sedimentation rates.
Geochemical Models
The calculation of “‘ages” by the *!°Pb method is
usually based on one of two models: the constant rate
of supply model or the constant initial concentration
model. The constant rate of supply (CRS) model as-
sumes that the flux of 7!°Pb is constant over time re-
gardless of how the sediment flux might vary. The
constant initial concentration model is based on the
assumption that the concentration of *!°Pb at time zero
(when the sediment was deposited) has been constant
over time. Robbins er a/. (2000) examined several var-
iations of these models. As a result, a secular equilib-
rium model, which assumes that a fraction of the *!°Pb
present in the sample must be in equilibrium with
>°Ra, and that there is a constant flux of 7!°Pb with
the attendant adsorption, seems to be the most appro-
priate model for these sediments.
Individual cores were analyzed by applying a best-
fit curve to the data using an unweighted Marquardt-
Levenberg parameter optimization method (Press ef
34 BULLETIN 361
Table 1.—Concentration of 210Pb in Penicillus sp from site in northeastern Florida Bay.
Total Pb-210 activity
Sample no. Location (dpm/g) Latitude Longitude
June 1994
1 Peterson Keys 2.32 24.7330 80.7500
2 Russell Key 4.02 25.0660 80.6330
3 Rabbit Keys 1.22 24.9900 80.8330
4 Bob Allen Key 1.93 25.0300 80.6730
5 Johnson Basin 1.70 25.1330 80.9250
August 1994
1 Buttonwood Sound 3.64 25.0853 80.4542
2 Buttonwood Sound 2.79 25.0933 80.4480
3 Buttonwood Sound 3.92 25.1008 80.4425
4 Buttonwood Sound 3.49 25.1072 80.4367
4U Buttonwood Sound 2.43 25.1072 80.4367
5 Buttonwood Sound D728} 25.1128 80.4438
SA Buttonwood Sound 2.56 25.1128 80.4438
6 Buttonwood Sound 2.24 25.1063 80.4492
8 Buttonwood Sound 4.33 25.0908 80.4613
9 Buttonwood Sound 2.81 25.0963 80.4675
10 Buttonwood Sound 3.14 25.1033 80.4620
11 Buttonwood Sound 2.22 PPS IS) 80.4553
13 Buttonwood Sound gl 25.1242 80.4567
17 Buttonwood Sound 8555 25.1147 80.4755
18 Buttonwood Sound 3.58 25.1228 80.4688
19 Buttonwood Sound 4.16 25.1295 80.4642
19D Buttonwood Sound 1.90 25.1295 80.4142
25 Little Maderia Basin 8.66 25.1600 80.6233
26 Little Maderia Basin 6.10 25.1342 80.5833
October 1994
94-glw-201HAL Lignum Vitae Basin 0.72 24.9370 80.6957
94-glw-201PEN Lignum Vitae Basin Hater 24.9370 80.6957
94-glw-202HAL Twinkeys 0.44 24.9617 80.6669
94-glw-202PEN Twinkeys 1.35 24.9617 80.6669
94-g]w-203PEN Spy Keys 1.96 24.0123 80.7644
94-glw-204HAL Twin Key Basin 0.59 24.9898 80.7839
94-glw-204PEN Twin Key Basin 1.44 24.9898 80.7839
94-glw-205SPEN N. Rabbit Key 1.86 24.9816 80.8242
94-glw-207HAL Peterson Bank 0.85 24.9189 90.7472
94-glw-207PEN Peterson Bank 1.97 24.9189 90.7472
HAL = Halameda, U = Udotea, PEN = Penicillis, D = Dasycladaceae, A = Duplicate.
al., 1989). A distinct advantage of this procedure is
that all data points, including negative values (which
are the result of counting errors, etc.), are used in the
sediment accumulation rate calculations. This method
has the further advantage of correcting for the ‘old
age’ error, which is inherent in many other best-fit in-
terpolation routines.
Special Geochemical Considerations (Florida Bay)
*10Pb dating of marine sediments in the coastal en-
vironment is generally complicated by the multiplicity
and variability of sources and by physical perturba-
tions of sediment mixing. Thus, the assumptions used
in correctly applying *!°Pb as a geochronological tool
can be seriously violated. However, in northeastern
Florida Bay there are a variety of conditions, in ad-
ditional to the physio-chemical conditions previously
enumerated that suggest that the *!°Pb distribution can
provide a sound basis for dating the sediments.
First, a major source of excess 7!°Pb is the atmo-
spheric flux of 7!°Pb. In Florida Bay, input from other
sources such as groundwater and in situ production
(via *?Rn) must also be considered. Recent data by
Chanton and Burnett (1995) have demonstrated high
levels of *??Rn in Florida Bay waters. The likely source
is ground water exchange with the elevated levels of
parent radionuclides in the underlying Pleistocene
rocks. A measure of this reactivity is the measurement
of *!°Pb activities in the carbonate secreted by the
short-lived algae Penicillus sp. This alga, which has a
life span of ~45 days, is a good monitor (Stockton et
al., 1967). The concentration of *!°Pb in the algal car-
SEDIMENT DYNAMICS: HOLMES ET AL. 35
Radium (dpm/g) versus Lead-210 (dpm/g)
Lead -210 (dpm/g)
i)
o | 1 1 l 1 1
n
LE 4 1 ! si n 1
0.00 0.50 1.00
1.50 2.00 2.50
Radium (dpm/g)
Text-figure 2.—Plot of the *°Ra activity versus the total *!°Pb activity showing a positive relationship with the sediments of Florida Bay (r
= 0.68).
bonates is highest in areas where there is very little
sediment covering the “bedrock” (Table 1). This is
consistent with a ground water or bedrock source and
would provide a constant flux of *!°Pb to the waters
of Florida Bay.
Second, even though Rude and Aller (1991) mea-
sured the maximum excess *!°Pb activity (1.1 dpm/g)
to be only about half the background (noise) at the top
of the core on Bob Allen Bank, there are a number of
characteristics of the 7!°Pb and **°Ra profiles that ap-
pear to be favorable for geochronological purposes:
1. The radium values vary within a limited range
(<20%) over the one-meter long cores.
2. Sediment mixing, as evidenced from a 20-cm
zone of nearly constant excess *!°Pb activities, was
confined to the surface of the core. Conventional
steady-state sediment mixing and accumulation mod-
els can be accurately used to determine a mixing depth
and/or mixing rate as well as a sediment accumulation
rate (Robbins ef al., 2000).
3. Rude and Aller (1991) calculated a mean linear
accumulation rate of 1 cm/yr at the Bob Allen Key
site which favors the use of reasonably short sediment
cores to reconstruct historical records of system chang-
es over the past 100 years.
210Pbh and ?°°Ra Distribution
Calculating *‘?!°Pb ages” according to the model of
Robbins er al. (2000) requires the determination of a
portion of the *!°Pb that is “supported” by *°Ra. A
number of investigators assume that the radium distri-
bution is constant and equal to the value where the
total *!°Pb activity is constant with depth. They use
this value to subtract from the total 7!°Pb, eliminating
the need for independent radium measurements. In or-
der to validate this assumption for the Florida Bay
sediments, a large number of radium measurements
were made on samples from eight cores. These cores
were chosen to represent various environments of the
Bay. The Bob Allen Bank cores (6A and 6C), Russell
Bank cores (19B and 19C), and Pass Key Bank cores
(7D and 17G) were taken from the accreting sides of
the banks. The Whipray Basin core (5G), Trout Creek
Spit core (13), and Crocodile Point core (16B) repre-
sent the non-accreting fringing environments.
In each core from an accreting environment, radium
systematically decreased with depth. A linear best-fit
of the radium versus depth for individual cores had r-
squared values (goodness of fit) which ranged from
0.6-0.9. The slopes and the surface intercept of the
individual curves varied within narrow limits, between
—0.002——0.003 for the slope and 1.9—2.2 for the sur-
face intercept. The coherence of the radium distribu-
tions between these cores suggested that a singular ex-
pression, *?°Ra.,, = —0.0025(depth) + 2.0 (where cal
= calculated value) could be used to describe the ra-
dium distribution on accreting banks (Text-fig. 2).
36 BULLETIN 361
Russell Bank (Core 19C)
0
| ry |
. 226 }
Pee kee Ra
wea
40
| 210
—@
L
“4°
=
& 30 -
a
120 |—
|
39 | ee ee le
l 2 3 4 5 6
Total Activity - dpm/g (Radium-226 and Lead-2 10)
Text-figure 3.—Distribution of **°Ra and total *!°Pb in the Russell
Bank (Core 19C).
However, for non-accreting banks, where *!°Pb activity
approaches a minimum in the upper part of the cores,
the similar, more generic relationship was determined
for *!°Pb and radium *°Rag,,y. = 0.16 Total 7!°PBgpnjc
+ 1.37 (where dpm/g + total activity). Comparison of
ages using calculated versus measured radium values
yielded insignificant differences.
The total *!°Pb activities for all samples ranged from
1.5—6.0 dpm/g. Subtracting a background value of 2.2
dpm/g for the radium content of the most recently de-
posited sediments; the maximum excess 7!°Pb signal
was 3.8 dpm/g. A typical plot of excess 7!°Pb vs. depth
from Russell Bank (Core 19C) demonstrates that Flor-
ida Bay sediment are in sufficient disequilibrium for
calculating accurate geochronology (Text-fig. 3). Text-
figure 3 clearly shows that a discernable difference can
be observed in *!°Pb and *°Ra down to a depth of 100
cm. This diagram also shows the consistent decrease
in radium with depth.
The core in Whipray Basin (5G) has high ?!°Pb ac-
tivity at the top and, interestingly, also elevated activ-
Whipray Basin (Core 5G)
0
| eee
r =
ee
20 — oe
+
210
aT Pb (Gamma)
8 60 |—
<<
eh Tk
fa)
80
7%
SA \\210
LO is a \ Pb (Alpha)
r a
120 | 2 SA)
| oe ee
140 i it it | it it it | i it i | it 1 it | i it it
1 2 3 4 5 6
Total Activity (dpm/g)
Text-figure 4.—Distribution of **°Ra and total 7!°Pb in the Whi-
pray Basin Core 5G. Also plotted is the total *!"Pb as determined by
gamma spectroscopy that demonstrates very good agreement be-
tween the two methods.
ity at the bottom (Text-fig. 4). Gamma spectrometry
measurements of these cores suggest that the increased
activities observed in the bottom sediments are likely
the result of elevated radium and uranium activities.
This phenomenon has been detected in the cores taken
within the Everglades; in cores from the northern por-
tion of Florida Bay: Rankin Bight (Site 7), Johnson
Key (Site 11), Trout Creek (Site 13), and Little Ma-
deira Bay (Site 14); and in cores near the islands in
the southern Bahamas (Tedesco and Aller, 1997). The
remaining cores show a conventional exponential de-
crease in excess *!°Pb with depth.
The most obvious difference between these cores
appears to be the depth at which apparent equilibrium
(7.e., total activity equals supported activity) is
reached. This is clearly seen by comparison of the S.
Russell Bank (19C) and the Whipray Basin (5G) cores.
Age calculations for the S. Russell Bank cores indicate
that the bottom sediments are at least 150 years old,
whereas the Whipray Basin cores reach “‘equilibrium”
in the upper 20 to 40 cm and the bottom sediment has
been '*C dated as 3300 years BP. The Pass Key cores
SEDIMENT DYNAMICS: HOLMES ET AL. 37
Table 2.—Core location, average rate of accumulation for the last 100 years, and total excess *!°Pb inventory of cores used in this chapter.
Location Tait
Core Rate tory
Site No. N W cm/yr dpm/cm? Comment
Rabbit Key Basin SBP 24959!06) 9 80)50:25) 10!35) ==0!05, 3.3. Northern edge of basin in a grass bed.
Whipray Basin 5G 2504.26 804431 0.37 + 0.06 8 Eastern side of the basin in a barren patch.
Bob Allen Bank 6A 2501.39 8039.41 0.75 + 0.08 46 South side of the bank in a grass bed.
Bob Allen Bank 6@ > 25/0139) 18039741 1.08 + 0.10 79 10 m from 6A in a barren spot.
Rankin Bight [Bae 251092555 80i47262-—~ 0:338== 10105 28 Northeastern edge of the bight about 100 m from shore.
Lake Ingraham 8B 25 08.85 8105.88 3.57 + 0.50 32 West side of central channel in the middle of the lake.
Russell Bank 10B 25 03.84 80 37.52 1.00 + 0.08 141 Top of the bank'.
Johnson Key Basin 11A 2503.10 8054.39 0.37 = 0.04 5.6 Eastern side of basin in a grass bed.
Porjoe Bank 12B 25 08.04 80 28.42 0.50 + 0.08 35 Eastern side of the bank in a grass bed.
Porjoe Bank 12D 25 08.04 80 28.42 0.50 + 0.08 36 Eastern side of the bank in the same grass bed as 12B.
Trout Creek Spit ISD 25250 SOS 99 0411005 PE Spit on the eastern side of the entrance to Trout Creek.
Little Madeira Bay 14B 2510.65 8037.44 0.85 + 0.09 22 Mouth on the eastern side of Little Madeira Bay.
Crocodile Point 16B 2508.32 8043.68 0.33 + 0.06 85 Transverse north-south bank in grass bed’.
Pass Key Bank 17D 2508.86 8034.49 3.50 + 0.12 165* Top of bank in the middle of the old channel.
Pass Key Bank 17G 25 08.86 80 34.49 5.80 + 0.60 230* Top of bank, 5 m from 17G in a barren site.
Russell Bank I8B 25.03.98 8037.46 0.75 + 0.06 48 Northern side of the bank’.
Russell Bank 19B! 2503583) 80137748) 1-27 210108 54 South side of bank in deeper water than 10B (grassy).
Russell Bank 19C 25 03.83 80 37.48 0.97 + 0.08 71 South side of bank 10 m from 19B (barren).
Bob Allen Bank 20D 2501.39 80 39.41 1.56 + 0.20 182 South side of bank in a barren “hole” deeper than 6A.
Pass Key Bank 21B 25 08.86 80 34.49 3.00 + 0.08 462* Top of bank in the middle of the old channel.
Pass Key Bank ZE 25109106) | 803457, 78050) 19 Western side of bank, south of old channel.
Pass Key Bank 21F 25 08.83 = 80 34.21 1.91 + 0.06 114* Western side of bank, south of the old channel.
Russell Bank 22D 2503.88 8037.55 0.88 = 0.08 55) Top of bank to the north and west of core site 19.
Park Key 23A 2506.27 80 34.47 0.78 + 0.05 25 Northern side of the bank, sparse grass.
Bob Allen Bank 24A 2501.42 80 33.84 0.78 + 0.09 25 South side of bank, 1 km southeast of 6.
Bob Allen Bank 24C 2501.42 8033.84 0.73 + 0.08 20 South side of bank, 1 km southeast of 6.
Whipray Basin 25B 25 04.27 8044.31 0.43 + 0.07 23, Eastern side of the basin in deeper water than site 5.
' The core “Russell Bank 10C” appears to be in three segments: a top part that has either been rapidly deposited or is mixed, a middle part
that has the character of a grass bed (a highly variable *'°Pb signal and a lot of grass fragments), and a bottom part.
> The top 20 cm are disturbed. The rate of accumulation was calculated for the section of the core below the mixed layer.
3 The top of the core appears to represent material that has “slumped” into place. The rate was calculated using part of this zone.
do not attain equilibrium. Age calculations and pho-
tographic evidence indicate that the core bottom sed-
iments (~1 m) are less than thirty years old A sum-
mary of the rates of accumulation is given in Table 2.
Confirmation and Replication of *!°Pb Ages
Shen and Boyle (1987) measured stable lead con-
centrations in the annual bands of a Monastrea an-
nularis coral collected on the ocean side of the Florida
Keys, directly east of this study area. The total stable
lead concentration in this coral reached a maximum
around 1977, which corresponds to the peak use of
leaded gasoline. In three cores, 6A and 6C (Bob Allen
Bank) and 19C (S. Russell Bank), the down core pro-
file of stable lead was determined by ICP-MS. These
core and coral sites are located less than five miles
apart. The stable lead concentrations for both the cores
and coral are plotted against time (Text-fig. 5). The
ages of the cores were calculated from the *!°Pb data,
while the age of the coral was determined by counting
the annual bands. The clear concordance of these in-
dependently dated profiles confirms the validity and
accuracy of our sediment geochronology.
The within-station variability in sedimentation rates
was estimated using two cores collected 10 meters
apart on Bob Allen Bank (Text-fig. 6). Core 6A was
collected in a grass bed, while core 6C was taken in
a barren area. Core 6A shows a fairly constant sedi-
ment flux prior to 1965 with the minimum accumu-
lation occurring during the mid-1930s. Since 1965, the
flux rate has rapidly decreased. In comparison, at the
non-vegetated site, the sediment flux fluctuated widely
before the 1920’s. A distinct decrease in sedimentation
occurred in the 1930’s, followed by a sharp increase
up to 1950. After 1950, the sediment flux has been
decreasing steadily. The patterns between the two
cores are remarkably similar, suggesting that the same
forces control sediment accumulation in both areas,
though to slightly different degrees. There is also a
difference in bulk density between the two cores. The
average density of sediment within the grass bed is 0.7
g/cm’, whereas the average density of the sediment at
38 BULLETIN 361
Bob Allen Core
4.5
a "
mt
3.5/_ ot
oll Coral Data _, 3
2.5 as
2.0 | - AC ‘Core Data
i) 1D ie - @e -. a
x 4 es 2s ma ee (le Eee Sr Se]
e 1860 1880 1900 1920 1940 1960 1980 2000
— Year
eo}
a
Sie South Russell Bank Core
Ol eS [e
= ane .
ae rh a
oe Core a?
2.5 Data ~ wich
2.0 — e e° ta L
45 a0 °° 8°°a" "\Coral Data
0 e@
(ye ee er ee
1860 1880 1900 1920 1940 1960 1980 2000
Year
Text-figure 5.—Comparison of the distribution of stable lead with
dated cores (6A and 19C) and a dated coral.
the barren site is 0.9 g/cm?. The core taken in the grass
bed contains a high percentage of plant material, which
decreases the overall density of the sediment. A similar
comparison can be seen in the Pass Keys Cores—17D
and 17G (Text-fig. 7).
Sedimentary Dynamics
As mentioned above, the source of excess ?!°Pb is a
result of the severed decay chain by the gaseous ***Rn.
As a result *!°Pb can reach a site of deposition via
atmospheric fallout or a combination of atmospheric
fallout and particle transport in which lead is added
along the route of transport. The atmospheric flux
along the northeastern coast of the United States has
been measured as 1.0 + 0.2 dpm/cm®? (Turekian er ai.,
1983). This would produce a steady state inventory of
excess *!°Pb of 32 + 6 dpm/cm?. Similar inventories
determined in lowland soil profiles in the eastern Unit-
ed States average 26 + 4 dpm/cm,. Baskaran ef al.
(1993) measured the atmospheric flux at Galveston
and College Station, Texas, as varying between 0.67
and 1.71 dpm/ cm?/yr, with a mean of 1.03 dpm/cm?/
yr. Their data also demonstrated the flux differences in
between continental and marine sources, with the ma-
rine sources producing a smaller flux than the conti-
nental source. In south Florida a flux of 0.8 dpm/cm?/
Bob Allen Bank
1.40 — — — —_—___—
r ’
1.20 F I Core 6C - Barren Site
° le ||
L mm) fs e fait ii Te
fo Ct hdl |
g Wor gs LS lel Nhe
ES rt oe es? one) : ®
= 0.80 Hike y
2) \~ 8 \
Sura Mu J 5 wa ll |
os | \
5 0.60 ¢ & Ay % |
[ Phe “sl laa Core 6A-Grass Site
%
0.40 4
a
0.20 4 |
a
0.00 & He | Si | 1 Se | L
0 50 100. 150 200 250
Years (BP)
Text-figure 6—Comparison of sedimentation rates between core
6A (Grass) and 6C (Barren) from Bob Allen Bank. The high vari-
ation in sedimentation rates in the barren core (6C) may be related
to climate fluctuations (see Cronin er al., this volume).
yr was measured at the University of Miami for the
year 1997 by Dr. Frank Millero (written communica-
tion, 1998). This would yield an inventory of ~25
dpm/cm?’. Dalphino ef al. (1993) estimated an inven-
tory of 19 dpm/cm? which would suggest an average
flux for south Florida of ~0.6 dpm/cm?/yr. These low-
er values are consistent with the proposed marine
source of air masses that move across south Florida.
Inventories of excess 7!°Pb have been calculated for
each core from:
Q, = > (:hA;)
where Q, = inventory of excess 7!°Pb, p,; = dry bulk
density of the ith depth interval (gy,, /em*), h, = thick-
ness of the ith interval (cm), and A; = excess 7!°Pb of
the ith interval (dpm/g). The summation of the excess
Pass Key Bank Cores 17D and 17G
2.50 +
F Core 17D (West of Bank) Is o
| a 2 1 WA AK
2.00 | a | te M Ye M\ 7 |
3 nit one Pe ary AR t |
2 , Mk \
Rae | a mx :
§ Nee oe
B 1.50 + A
E | n » A,
5b we ON \ of Core 17G (Top of Bank)
a
| mo 4
1.00
aba
Fei ete Seal ree es TS eR Ln,
0 10 20 30 40 50 60
Years (BP)
Text-figure 7.—Comparison of sedimentation rates between core
17D and 17G from Pass Key. The difference in sedimentation rates
can be attributed to their relative locations to the central channel.
SEDIMENT DYNAMICS: HOLMES ET AL. 39
values was carried out from the surface to that level
in which the difference between background and ex-
cess 7!°Pb activity was indistinguishable. The inven-
tories record in the cores ranges from 3 dpm/cm, to
>400 dpm/cm?. Assuming that the excess *!°Pb inven-
tory should be in the 20 to 30 range if all of the lead
was a result of atmospheric flux, these extreme values
lead to some indications of the sediment dynamics.
The inventories are presented in Table 2. Those sites
with inventories significantly lower than can be ac-
counted by the atmospheric flux indicate that some
sediment has been removed. The three such sites are
on the southern and western sides of the Bay. In Whi-
pray basin, the core with the lowest inventory (5G) is
much shallower than core 25. The *'°Pb distribution
curve in this core has an apparent discontinuity at
about 16 cm hinting at an interval of missing sediment.
Along the northern fringe of the bay, the inventory in
cores 7, 13, 14 and 16 imply that the source of lead
was wholly atmospheric. At all the rest of the sites,
the high inventories are indicative of sediment *!°Pb
“enriched” by focusing and/or an external non-atmo-
spheric source, possibly ground-water fluxes. In gen-
eral, the highest inventories were measured on Pass
Key bank, where the highest rate of sediment accu-
mulation occurs forming the bank within the last 30
years. The high rate on Bob Allen Bank (core 20) was
measured at a site in deeper water further south than
the other cores, consistent with the idea of southward
accretion of the banks.
SUMMARY
The determination of sediment geochronologies us-
ing *!°Pb at various sites within the sedimentary banks
of Florida Bay has yielded a valid time scale on the
order of decades. These calculated ages have been con-
firmed by independent means. At sites in the western
portion and along the northern fringe of the Bay, sed-
iment accumulation is ~0.3 cm/yr. At sites in the west-
ern part of the Bay, the inventory of *!°Pb is much less
than atmospheric flux, indicating that sediment has
been removed, probably by storms. In the northeastern
section of the Bay, which is most impacted by chang-
ing salinity regimes, the sedimentation rates vary from
0.7 cm/yr to greater than 2.0 cm/yr. In the central part
of the study area on the leeward side of Russell Bank
and Bob Allen Bank, the sediment accumulation rate
is moderate (~1.0 cm/yr). It is from these sites that
the ecological history of Florida Bay during the last
century could best be developed. Additional ongoing
studies are addressing other geochemical and paleon-
tological issues (Cronin ef al., 1998). This investiga-
tion on the feasibility of dating the sediments in Flor-
ida Bay has proven to be quite successful; the infor-
mation is being used to reconstruct the paleoecological
history of the bay and to assess the transport and de-
position of Florida Bay sediments.
REFERENCES CITED
Appleby, P.G., and Oldfield, F.
1992. Application of lead-210 to sedimentation studies. in Ura-
nium-series Disequilibrium. Applications to Earth, Marine
and Environmental Studies. M. Ivanovich, and R.S. Har-
mon, eds., Oxford Science Publishers, Clarendon Press,
Oxford, Second Edition, pp. 731-778.
Baskaran, M., Colman, C.H., and Santschi, P.H.
1993. Atmospheric depositional fluxes of 7Be and *!°Pb at Gal-
veston and College Station, Texas. Journal of Geophysical
Research, vol. 98, pp. 20,555—20,571.
Boscence, D.
1989a. Biogenic carbonate production in Florida Bay. Bulletin of
Marine Science, vol. 44, pp. 419—433
1989b. Surface sublittoral sediments of Florida Bay. Bulletin of
Marine Science, vol. 44, pp. 434—453.
Chanton, J., and Burnett, W.
1995. Delivery of nutrients to Florida Bay by submarine
groundwater. FSU Project report no. 1368-689-37, 15 p.
Cronin, T. M., Holmes, C.W., Wingard, G. L., Ishman, S., Dow-
sett, H., and Waibel, N.
2001. Historical trends in epiphytal ostracodes from Florida
Bay: Implications for seagrass and macro-benthic algal
variability. Bulletins of American Paleontology, no. 361,
pp. 159-197.
Dalphino, J.J., Crisman, T.L., Gottgens, J.F., Rood, B.E. and
Earle, C.D.A.
1993. Spatial and temporal distribution of mercury in Ever-
glades and Okefenokee wetland sediments, Final report
SEWMD, Contract C91-2237, 138 pp.
Davies, T. D., and Cohen, A. D.
1989. Composition and significance of the peat deposits of Flor-
ida Bay. Bulletin of Marine Science, vol. 44, pp. 387—
398.
Flynn, W.W.
1968. The determination of polonium-210 in environmental ma-
terials. Analytica Chimica Acta, vol. 43, pp. 121-131.
Goldberg, E. D.
1963. Geochronology with Lead-210. in Radioactive Dating,
[.A.E.A., Vienna, pp. 121-122.
Press, W. H., Flannary, B. P., Teukolsky, S.A., and Vettering,
V.T.
1989. Numerical Recipes: The Art of Scientific Computing.
Cambridge University Press, New York, NY, 702 pp.
Robbins, J. A.
1978. Geochemical and geophysical applications of radioactive
lead. in Biogeochemistry of Lead in the Environment,
Part A. J. O. Nriagu, ed., Elsevier Scientific Publishers,
Amsterdam, Netherlands, vol. 1A, pp. 285-393.
Robbins, J.A., Holmes, C.W., Halley, R. B., Bothner, M., Shinn,
E. A., Graney, J., Keeler, G., Ten Brink, M., and Rud-
nick, D.
2000. Time constants characterizing pre-depositional integration
40 BULLETIN 361
of '*’Cs and Pb fluxes to sediments in Florida Bay. Journal
of Geophysical Research, vol. 105, pp. C28,805—28,822.
Rude, P. D., and Aller, R. C.
1991. Fluorine mobility during early diagenesis of carbonate
sediment: An indicator of mineral transformation. Geo-
chimica et Cosmochimica Acta, vol. 55, pp. 2491—2509.
Scholl, D. W.
1964. Recent sedimentary record in mangrove swamps and rise
in sea level over the southwestern coast of Florida, parts
I and I. Journal of Marine Geology, vol. 1, pp. 344-366.
Shen, G. T., and Boyle, E. A.
1987. Lead in corals: Reconstruction of historical industrial
fluxes to the surface ocean. Earth and Planetary Science
Letters, vol. 82, pp. 289-304.
Stockman, K. W., Ginsburg, R. N., and Shinn, E. A.
1967. The production of lime mud by algae in south Florida.
Journal of Sedimentary Petrology, vol. 37, pp. 633-648.
Tedesco, L.P., and Aller, R.C.
1997. ?'° Pb chronology of sequence affected by burrow exca-
vation and infilling; examples from shallow marine car-
bonate sediment sequences, Holocene South Florida and
Caicos Platform, British West Indies, Journal of Sedimen-
tary Research, Section A: Sedimentary Petrology and Pro-
cesses, vol. 67, no. 1, pp. 36-46.
Turekian, K.K., Benninger, L.K., and Dion, E.P.
1983. /’Be and ?!°Pb total depositional fluxes at New Haven,
Connecticut and at Bermuda. Journal of Geophysical Re-
search, vol. 88, pp. 5411—S415.
Wannless, H., and Taggert, M. G.
1989. Origin, growth and evolution of carbonate mudbanks in
Florida Bay. Bulletin of Marine Science, vol. 44, pp. 454—
495.
CHAPTER 4
THE FLORIDA EVERGLADES ECOSYSTEM: CLIMATIC AND ANTHROPOGENIC IMPACTS OVER THE
LAST TWO MILLENNIA
DEBRA A. WILLARD!, CHARLES W. HOLMES?2, AND LISA M. WEIMER!
'U.S. Geological Survey, Eastern Surface Processes Team, 926A National Center, Reston, VA 20192
°U.S. Geological Survey, Center for Coastal and Marine Geology, 400 6th Street South
St. Petersburg, FL 33701
ABSTRACT
We document the response of the Everglades ecosystem to climatic and environmental changes over the last two millennia
using pollen records. Centennial-scale intervals of vegetational stasis characterize the time between 0 A.D. and 1900 A.D. From
0-800 A.D., marsh and slough vegetation characteristic of deeper water and longer hydroperiods (annual duration of inundation)
than today dominated the study region. Drier conditions between about 800 A.D. and 1200 A.D. resulted in shallower water
depths and shorter, fluctuating hydroperiods in Everglades marshes as well as salinity increases near Florida Bay. After a recovery
to deep-water conditions in the 14" century, somewhat drier conditions are suggested between the 17" and 19" centuries. These
climatically-induced periods of relative dryness are correlated with regional droughts during the intervals known as the Medieval
Warm Period (9"—-14" centuries) and Little Ice Age (15""—19" centuries).
Although regional precipitation was greater than average after 1920 A.D., vegetational changes throughout the area indicate
reduced water depths and hydroperiods after construction of water-contro] structures in the early 20" century. Further, more
localized changes occurred after 1960, when the Central & South Florida (C&SF) Project was completed. Thus, restoration goals
of achieving pre-C&SF Project hydrologic regimes are aimed at an already disrupted system; a more “natural” restoration target
would be the 19" century Everglades, which had been stable for the past few centuries. Recent land-use changes have resulted
in localized rather than system-wide ecosystem responses, at least in part because of the fragmentation of the wetland. This
artificially induced ecological heterogeneity makes prediction of future wetland responses to climatic changes problematic.
INTRODUCTION
In wetland ecosystems, climatic warming is expect-
ed to cause lowered water tables, increased rates of
evapotranspiration, and, in the Florida Everglades,
greater incidence of saltwater intrusion from both sea-
level rise and hydrologic changes (Watson ef al.,
1996). Additionally, anthropogenic changes in hydrol-
ogy and nutrient availability due to land-use alterations
affect wetland ecosystems on both local and regional
scales. The accurate prediction of wetland response to
future climatic and environmental changes is ham-
pered by a lack of broad-scale, long-term studies of
past changes in wetland ecosystems. Such studies are
feasible in the Florida Everglades, where we have col-
lected a series of cores that preserve records of the last
2,000 to 3,000 years of Everglades vegetation (Text-
fig. 1). These cores provide data on past responses of
wetland vegetation to late Holocene climatic events as
well as hydrologic alteration of the system over the
last century. A possible analog for Everglades re-
sponse to a predicted warming of 1°C to 3.5°C by 2100
A.D. (Watson ef al., 1996) is preserved in sediments
deposited during the Medieval Warm Period (MWP)
‘ Deceased.
(800-1300 A.D.) (Hughes and Diaz, 1994) when sea-
surface temperatures at least 1°C warmer than today
have been documented in the Sargasso Sea (Keigwin,
1996). The impact of a cooling event, the Little Ice
Age (LIA) (A.D. 1550-1850) (Bradley and Jones,
1993) on Everglades vegetation also is preserved in
many cores throughout the region. Of particular inter-
est are the timing and extent of vegetational changes
in response to hydrologic changes and changing land-
use practices of the last century. This suite of cores
provides the temporal and spatial coverage necessary
to address these questions on a regional basis.
Vegetational distribution in the Everglades wetland
is controlled primarily by water depth, hydroperiod
(average annual length of inundation), substrate, fire
regime, and water chemistry (Kushlan, 1990). Al-
though plant community composition can vary locally,
several broad vegetational categories have been iden-
tified (Davis et al., 1994; McVoy et al., 2000; Willard
et al., 2001a). These include: the Ridge and Slough
region consisting of a mosaic of sawgrass-dominated
marshes, sloughs, and tree islands in the north and
central Everglades and ranging southwest through the
Shark River Slough; sparse sawgrass marshes and wet
prairies in the southern Everglades; cypress swamps
42 BULLETIN 361
PI
LAKE
OKEECHOBEE
Hillsboro
Canal
| WCA3A\
it 26°
‘Shark River Tamiami Trail
ae BISCAYNE
BAY
ATLANTIC
OCEAN
10 20
7 We ==
F Z MILES
LORIDA BAY UY 950
fo} fe}
{2} (=)
eo) oO
foo)
Text-figure 1.—Map of core locations discussed in this study. De-
tailed locality information is provided in Table 1.
and domes and pinelands in the Everglades and Big
Cypress National Preserve; and mangrove forests
along the Florida Bay and Whitewater Bay coasts
(Text-fig. 2). This report presents reconstructions of
the vegetational history of the Everglades ecosystem
at several sites over the last two millennia and its re-
sponse to both natural climatic fluctuations and altered
land-use practices of the last century.
ACKNOWLEDGMENTS
The research was supported by the U.S. Geological
Survey Integrated Natural Resource Science Program
(INATURES) in cooperation with the South Florida
Water Management District. For field and technical as-
sistance, we thank C. Bates, P. Buchanan, A. Fagen-
holz, M. Cangelosi, C. Fellman, J. Herring, H. Lerch,
M. Marot, J. Murray, W. Orem, T. Sheehan, and N.
Waibel. L. Brewster-Wingard and S. Ishman provided
data on molluscan and foraminifer assemblages for site
8. For their helpful input during the project, we thank
T. Cronin, S. Davis, W. DiMichele, T. Fontaine, A.
Higer, C. Labandeira, C. McVoy, W. Park, P. Rawlik,
K. Rutchey, FE Sklar, D. Verardo, S. Verardo, B. War-
dlaw, and Y. Wu. Valuable comments on earlier ver-
sions of the manuscript were provided by E. Grimm,
O. Davis, D. Peteet, and P. Swart. We thank the staffs
of Big Cypress National Preserve, Everglades National
Park, Loxahatchee National Wildlife Refuge, and Ro-
tenberger and Holey Land Wildlife Management Areas
for assistance gaining access to their lands.
STUDY SITES AND METHODS
Peats preserved in the historic Everglades provide a
detailed archive of biotic patterns over the last few
millennia. Cores discussed herein were collected in
Water Conservation Areas 1, 2, and 3, in Taylor
Slough, and in the dwarf mangrove fringe adjacent to
Florida Bay (Text-fig. 1, Table 1). These sites were
selected to evaluate past vegetational patterns in the
Ridge and Slough area, where relatively minor chang-
es in hydroperiod and water depth may alter the spatial
distribution of vegetation types, and in the saline-in-
fluenced part of the Everglades, where hydrologic
changes may result in salt-water intrusion.
Pollen assemblages and geochronology were ana-
lyzed from samples collected at I—2 cm intervals
throughout the cores. Most cores are about 0.7 meters
long, cover the last two millennia (Tables 1, 2), and
reach the underlying Pleistocene limestone. Age mod-
els for the last century of deposition are based on 7!°Pb,
'7Cs, and 'C from atomic testing (Robbins er al.,
2000) and, where applicable, first occurrences of pol-
len of the exotic plant Casuarina which was intro-
duced to south Florida in the late 19" century (Lande-
land, 1990). In older sediments, models are derived
from linear interpolation between radiocarbon data
points (Table 2), which were calibrated to calendar
years using the Pretoria Calibration Procedure (Stuiver
et al., 1993; Talma and Vogel, 1993; Vogel et al.,
1993). In cores where both bulk radiocarbon dates and
AMS dates on plant fragments were obtained, age
models are based on the AMS dates, which minimize
the possibility of contamination by sediments outside
the sample interval. Pollen was isolated from samples
using standard palynological techniques (Traverse,
1988), including carbonate and silicate removal with
HCl and HF when necessary, acetolysis to reduce the
amount of phytodebris, sieving through 8 4m mesh to
remove clay-size particles, heavy liquid treatment
when needed, and staining with Bismarck Brown be-
fore mounting on microscope slides with glycerin jel-
ly. Tablets with known numbers of Lycopodium spores
also were added to the dry samples for determination
of pollen/gram dry sediment (see (Maher, 1981). To
calculate percent abundance in each sample, at least
300 pollen grains were identified and counted. In some
cases, pollen preservation was poor, and counts of at
least 100 grains were included in the analyses. Pollen
data and closest analogs are available via the internet
CLIMATIC AND ANTHROPOGENIC IMPACTS: WILLARD ET AL. 43
Gulf of
Mexico
Florida Bay
81°
Pinelands
Wet Prairies
Cypress forests
ELE
Southern marl-
forming marshes
Mangroves
Developed areas
Rockland Pine Forests
Sloughs
Sawgrass marshes
Ey |
Biscayne
Bay
/
80.5°
Text-figure 2—Generalized map of vegetation types, developed areas (urban and agricultural), and boundaries of Water Conservation Areas,
Big Cypress National Preserve, Everglades National Park, and Loxahatchee National Wildlife Refuge. Compiled from Davis (1943) and Davis
et al. (1994).
from the North American Pollen Database (NAPD) at
the World Data Center for Paleoclimatology in Boul-
der, CO (http://www.ngdc.noaa.gov/paleo/pollen.htm1)
and at the US Geological Survey South Florida Eco-
system History Database at http://flaecohist.erusgs.
gov/database.
We interpreted vegetational changes in cores from
patterns in both percent abundance and concentration
(pollen/gram dry sediment) of pollen. The timing of
major vegetational changes is clear from examination
of pollen diagrams. However, to improve the accuracy
of vegetational interpretations based on pollen data, we
statistically compared assemblages from cores to those
from modern samples using both cluster analysis and
Table 1.—Site locations, core length, and date of core collection for sediment cores, South Florida.
Latitude Longitude Core length
Site (N) (W) General location (cm) Date collected
l 26°28.873' 80°24.931' Loxahatchee NWR 72 4/21/1995
2 26°21.580' 80°22.230' Water Conservation Area 2A 72 4/26/1995
3 26°17.254’ 80°24.660' Water Conservation Area 2A 71 4/25/1995
4 25°58.455’ 80°40.129' Water Conservation Area 3A 70 4/27/1995
5) 25°17.231' 80°38.780' Taylor Slough 76 5/20/1996
6 DS mle ale 80°41.227' Taylor Slough 78 5/24/1996
7 25712339! 80°38.64 1 Everglades National Park 86 3/29/1995
8 25°13.190' 80°36.279' Everglades National Park 82 3/15/1996
44 BULLETIN 361
Table 2.—Radiocarbon dates from peat cores, South Florida.
Site Depth (cm) Beta no. Sample type 3C/"C ratio Corrected ''C age (BP) Calendar age
1 30-32 110295 Bulk 26:2, 480 + 40 AD 1410-1470
| 50-52 110296 Bulk —27.4 960 + 40 AD 1005-1180
1 70-72 110297 Bulk —26.2 1590 + 40 AD 405-575
2 14-16 104814 AMS—leaf —P)''6 175.2% + 0.4 post 1950
2 28-30 104815 AMS—leaf lox, 1040 + 50 AD 860-960
2 32-36 90032 Bulk —26.3 1090 + 60 AD 800-920
2 40-42 104816 AMS—leat —16.8 19110) = 50 10 BC-AD 90
2 52-56 90033 Bulk —26.3 1990 + 60 100 BC-AD 20
2 60-64 90034 Bulk —26.1 2270 + 60 380-260 BC
3 26-28 129850 Bulk = 2)7fa I 400 + 50 AD 1425-1635
3 38—40 129849 Bulk —27.4 930 + 50 AD 1005-1220
3 50-52 129848 Bulk —27.4 W/OOKERS 0) AD 235-435
3} 64-68 90034 Bulk —26.6 2090 + 60
4 30-32 110292 Bulk —?)5)(6) 610 + 40 AD 1295-1420
4 50-52 110293 Bulk =2519) 1880 + 40 AD 65-240
4 64-68 110294 Bulk — Psy 2470 + 40 BC 780—405
5 30-32 110298 Bulk —P5)-5) 1500 + 40 AD 465-645
5 50-52 110299 Bulk = 72333) 2460 + 40 BC 780—405
5 72-74 110300 Bulk 23:8 3560 + 40 BC 1975-1760
6 30-32 115197 AMS—leaf —27.0 310 + 40 AD 1600-1680
6 44-46 114967 Bulk —26.8 320 + 50 AD 1580-1680
6 60-62 115644 AMS—leat —30.6 530 + 60 AD 1305-1460
7 0-4 104507 AMS—twig PHS) 109.8% + 0.6 post 1950
7 4-8 104508 AMS—twig = 26:5) 96.0% + 0.4 post 1950
7 13-14 86194 Bulk ye: 370) 60 AD 1520-1640
7 20-24 104510 AMS—twig —26.8 230 + 40 AD 1680-1760
a 24-26 96195 Bulk — TS} 700 + 70 AD 1180-1320
a 75-76 86785 Bulk —26.5 1610 + 60 AD 280—400
8 30-32 96610 Bulk Uf 300 + 80 AD 1570-1730
8 58-60 93311 Bulk =i 750 + 80 AD 1120-1280
8 78-84 93312 Bulk ihe 1950 + 70 70 BC-AD 70
Note: Site numbers correspond to sites in Figure 1. Radiocarbon dating was performed by Beta Analytic, Miami, Florida, dating both bulk
sediment and selected plant fragments with AMS. When aerial material could not be separated confidently from the peat, homogenized bulk
samples were used. Due to the possibility of contamination by younger roots, dates on bulk samples represent minimum ages for the intervals.
Corrected dates are based on fractionation calculated by 6'°C. Calendar ages have been calibrated according to Stuiver et al., 1993, Talma and
Vogel, 1993, and Vogel et al., 1993, + 2 o, and 95% probability.
* These C13/C12 ratios were not determined but are estimates based on the peat composition.
the modern analog technique. Cluster analysis and oth-
er hierarchical classification techniques are used to
show relationships among different samples based on
species abundance in those samples. These relation-
ships are determined by first calculating a sample-by-
sample distance matrix and then linking samples into
clusters based on their dissimilarity/similarity to each
other (Gauch, 1982). We used average-linkage clus-
tering (UPGMA) with the Pearson product moment
correlation coefficient and the computer program
MVSP 3.1 (Kovach, 1999) to cluster our samples.
The modern analog technique compares fossil pol-
len assemblages with modern assemblages from a
number of vegetation types to identify the closest mod-
ern analog (Overpeck et al., 1985). If close analogs
exist for the fossil assemblage, it is assumed to have
a similar vegetational composition and, by extension,
similar environmental parameters, such as hydroperiod
or water depth. Modern analogs are identified by cal-
culating a dissimilarity coefficient between a fossil and
modern assemblage; samples with dissimilarity coef-
ficients less than a critical value are considered to rep-
resent the same vegetation type. To determine the ap-
propriate critical value for these samples, we calculat-
ed dissimilarity coefficients (using squared chord dis-
tance, or SCD) among all modern samples using the
ANALOG program of Schweitzer (1994) (see Willard
et al., 2001a). Samples from the same vegetation type
typically had SCD values <0.15. Recognizable wet-
land subenvironments include sloughs, sawgrass
marshes, sawgrass marshes near tree islands, southern
Everglades sawgrass marshes/wet prairies, tree islands,
southwestern mangrove forests, and cattail marshes.
This modern dataset incorporates both samples col-
lected between 1995 and 1998 and samples collected
in the early 1960s. The latter group includes a number
of sites that have since been altered by construction of
the Central & South Florida (C&SF) Project since
CLIMATIC AND ANTHROPOGENIC IMPACTS: WILLARD ET AL. 45
Table 3.—Clusters from Q-mode cluster analysis of modern and fossil pollen assemblages, modern analogs, and pollen signatures for each
cluster.
Cluster Modern analog Pollen signature
I TREE ISLANDS Fern Spores > 25%
1a LOXAHATCHEE SAWGRASS MARSHES Myrica > 30%
I SOUTHWESTERN MANGROVE FORESTS AND Rhizophora > 15%; Avicennia and Conocarpus present
BRACKISH MARSHES
IV SOUTHERN EVERGLADES MARSHES AND SLOUGHS
IV a No modern analogs, but apparently fresh-water marshes Cyperaceae 22-32%; Myrica 18-30%
IV b Wet Prairies (no fossils) Myrica 5—20%; Asteraceae > 15%: Sum of Cyperaceae and
Sagittaria > 5%
IV c Dwarf Mangrove Stands Rhizophora 5—25%; Myrica 2—12%; Cyperaceae 1—5%
IV dl Wet Prairies, Cypress Stands, Pinelands Pinus > 40%; Myrica 5—12%:; Asteraceae > 5%; Cyperaceae
> 2%
IV d2 Sawgrass marshes Pinus 20-40%; Myrica > 15%; Asteraceae 3—10%; Cypera-
ceae >3%
Vv NORTHERN EVERGLADES MARSHES AND SLOUGHS
Va Dense Sawgrass Marshes
V bl Sloughs (primarily fossils)
V b2 Sloughs, Dense Sawgrass Marsh
Vc Sawgrass Marshes near Tree Islands or Disturbed Sites
Cyperaceae > 10%; Asteraceae >10%; Sagittaria > 5%
Nymphaea 5—20%; sum of Cyperaceae and Sagittaria > 5%
Cheno/Ams 30-40%; sum of Nymphaea, Sagittaria, and Cy-
peraceae > 10%
Cheno/Ams > 40%; sum of Typha, Cyperaceae, and Sagit-
taria > 10%
1960, and they include deeper water sloughs than oc-
cur today.
ECOLOGY AND POLLEN SIGNATURES OF
EVERGLADES VEGETATION TYPES
Eight vegetation types are clearly distinguishable
from modern pollen assemblages collected throughout
the Everglades (Willard ef al., 2001a). Slough vege-
tation occurs in the wettest sites and typically is dom-
inated by floating aquatics such as Nymphaea (water
lily) and Utricularia (bladderwort). This community
type grows in relatively deep water over thick peat
substrates and requires long hydroperiods (>9 months
annual inundation) (Kushlan, 1990). Pollen assemblag-
es from sloughs are distinguished by >3% Nymphaea
pollen (Willard ef al., 2001a) and include samples in
cluster Vb (Table 3, Text-fig. 3).
Marshes requiring moderate water depths (<1 me-
ter) and hydroperiods (6—9 months annual inundation)
include those dominated by Cladium (sawgrass), Sag-
ittaria (arrowhead), Eleocharis (spikerush), and Typha
(cattail). Substrate type, fire frequency, nutrient con-
tent of surface waters, and disturbance determine the
dominant component (Kushlan, 1990). Sawgrass
marshes consist primarily of Cladium, grading into
flag marshes with more Sagittaria and Eleocharis
where surface sediments change from deeply organic
peats to shallow peats and marls (Kushlan, 1990). Pol-
len assemblages from typical sawgrass marshes are
dominated by Cladium and Sagittaria pollen and are
represented in clusters Va and Vb2 (Table 3). Cattail
marshes, dominated by Typha, may replace sawgrass
marshes after disturbances, such as severe fires or nu-
trient enrichment (Davis, 1994). Sawgrass marshes
near tree islands, near levees and canals, and with fluc-
tuating water levels are dominated by pollen of weedy
annuals, particularly Amaranthus (water hemp) or the
Asteraceae (aster family) (Willard er al., 2001a). These
plants are known to become particularly abundant after
periods of rapid drying over a few (<5) years or pro-
longed droughts (Loveless, 1950; Mc Voy et al., 2000).
Pollen assemblages from both cattail marshes and saw-
grass marshes associated with tree islands or distur-
bance are included in cluster Ve (Table 3).
Short hydroperiod marshes such as wet prairies re-
quire shallow water depths, seasonal drying (O—3
months annual inundation), and either marl or shallow
peat substrates (Kushlan, 1990). Sparse vegetation in
these marshes consists primarily of taxa in the Poaceae
(grass family) and Cyperaceae (sedge family) that tol-
erate both dry and flooded conditions. These marshes,
most common in the southern Everglades, have the
greatest abundance of Myrica (wax myrtle) pollen, as
well as Poaceae and Cyperaceae (Willard et al., 2001
a); wet prairies are represented in clusters [Vb and [Vd
(Text-fig. 3, Table 3). Pinus pollen also is abundant in
these assemblages; however, it represents extra-local
deposition and reflects low pollen productivity of
plants in this community type.
Tree islands are interspersed throughout the marshes
and sloughs of the Everglades, and two types exist.
“Pop-up” or battery islands, dominated by bays (Per-
46 BULLETIN 361
|
!
aval Il
[
ae ae
-0.2 0 0.2 0.4 0.6 0.8 1
Pearson Coefficient
Text-figure 3.—Dendrogram from Q-mode cluster analysis (using UPGMA and Pearson correlation coefficient) of pollen assemblages from
surface samples and down-core samples from sites 1-8, Florida Everglades. Cluster I represents tree-island samples. Cluster II represents
samples from sawgrass marshes in Loxahatchee NWR. Cluster III includes samples from southwestern mangrove forests and brackish marshes.
Cluster IV includes samples from southern Everglades marshes and sloughs. Cluster V includes samples from northern Everglades marshes
and sloughs. Finer details on clusters are provided in Table 3.
CLIMATIC AND ANTHROPOGENIC IMPACTS: WILLARD ET AL. 47
sea) and holly (lex), are the most common type in
Loxahatchee National Wildlife Refuge; these sites are
characterized by abundant //ex and Myrica pollen and
are included in cluster II (Table 3). Fixed, tear-drop
shaped tree islands, which are elongated parallel to
flow, occur throughout the Everglades; vegetation on
tree island heads consists of trees, shrubs, and ferns
and the tapering tails are vegetated by shrubs, ferns
and thick Cladium or Typha. Pollen assemblages from
fixed tree islands are characterized by high abundances
of fern spores and are represented in cluster I (Table
3). The high abundance of ferns is a local signature of
the tree islands themselves, with fern spores rare in
marshes and sloughs surrounding the islands (Willard,
et al., 2001b).
Two types of mangrove assemblages occur in South
Florida: dwarf mangroves consisting primarily of short
(<3 meters) Rhizophora near Florida Bay, and the tall,
diverse forests near Whitewater Bay in Shark River
Slough. Both types are distinguished in the pollen re-
cord by abundant mangrove pollen, particularly that of
Rhizophora. Near Florida Bay, mangrove pollen com-
prises only about 20% of assemblages and consists
primarily of Rhizophora with minor amounts of Con-
ocarpus and rare Avicennia and Laguncularia. These
are included in cluster [Vc (Table 3). In southwestern
Florida, however, Rhizophora is the dominant element,
and the four mangrove genera listed above combine to
make up at least 40% of the assemblages (Riegel,
1965; Willard et al., 2001a). These samples comprise
most of cluster III (Table 3).
VEGETATIONAL HISTORY: RIDGE AND
SLOUGH SITES
PRE-20T4 CENTURY VEGETATION
Four cores (sites 1—4: Text-fig. 1) in the Ridge and
Slough region (Text-fig. 2) provide records of the last
two millennia. From about 0-800 A.D., pollen evi-
dence from these sites indicate that they typically were
dominated by slough vegetation, indicating long hy-
droperiods and relatively deep water (Text-fig. 4). Be-
tween 800 A.D. and 1200 A.D., all sites show evi-
dence for shallower water and shorter hydroperiods,
although temporal resolution varies among the cores.
At sites | and 4, comparison with modern analogs in-
dicates that sawgrass marshes replaced sloughs for
varying intervals between 800 and 1200 A.D. The
great abundance of weedy annuals such as Amaranthus
at site 4 and, to a lesser extent site 2, indicate that
droughts or intervals of rapid drying (over <5 years)
(Loveless, 1950; McVoy et al., 2001) occurred during
that time, allowing weedy species to flourish. At site
2, Cladium pollen abundance increased and samples
are grouped in cluster Vb2, indicating shorter hydro-
periods and/or shallower water depths than before.
During this interval at site 3, there are no close ana-
logs, but samples cluster with marsh and slough sam-
ples from the southern Everglades (Text-fig. 4), where
water depths are shallower and peat substrates thinner.
Although this is suggestive of drier conditions, the
high percentages of Nymphaea make confident inter-
pretation of hydrology problematic at this site. Be-
tween 1200 A.D. and 1600 A.D., deeper water con-
ditions associated with slough vegetation returned at
all four sites, although site 3 remained similar to south-
ern Everglades sloughs. Early in the 17" century, Myr-
ica pollen doubled in abundance at all four sites, and
Nymphaea pollen decreased in abundance at most
sites. Although the assemblages remained analogous
to sloughs, the increased abundance of tree pollen is
suggestive of drier conditions beginning in the early
17" century. Vegetational composition remained stable
from then until the early 20" century.
20™ CENTURY VEGETATION
From about 1200 A.D. to 1930 A.D., vegetation at
all four sites was characteristic of sloughs. After 1930,
however, shallower water conditions are indicated at
sites 2, 3, and 4 by the change to sawgrass marshes
with abundant weedy annuals. Although shallower wa-
ter depths are suggestive of drier conditions, precipi-
tation levels were higher than the 20" century average
during this time (Text-fig. 5), indicating that reduced
water levels and drier marsh conditions at these sites
resulted from hydrologic alteration of the system rather
than climatic change. Hydrologic changes that were
completed by 1930 include construction of the Hoover
Dike on the south side of Lake Okeechobee and dredg-
ing of major muck canals linking the lake to the At-
lantic Ocean (Light and Dineen, 1994). Construction
of the Hoover Dike prevented seasonal influx of water
to the Everglades from overflow of the southern lake
bank, a major source of water for sheet flow across
the Everglades. Additionally, substantial areas of the
Everglades were drained as the muck canals diverted
much of the remaining sheet flow from the Everglades
to the ocean.
Additional changes occurred between 1950 and
1960. At site 1 in WCA 1, vegetation changed from
sloughs to sawgrass marshes; at site 2 near the Hills-
boro Canal in WCA 2A, cattail marshes began to dom-
inate the vegetation after 1960. Both changes occurred
after completion of the Water Conservation Areas and
Everglades Agricultural Area (EAA), part of the
C&SF Project. Construction of a levee on the western
border of WCA | between 1954 and 1959 effectively
stopped the remaining sheet flow across WCA | (Light
BULLETIN 361
SITE 1
Depth
cm
1950AD =
~1900AD =
1410-1470AD fj 30
40
1005 - 1180 AD I 50
60
405-575AD ff 70
0 10 0 0 10 20 30 0 0 10 20 30 40 50 60
PERCENT ABUNDANCE
SITE 2
Post 1950
20
1900AD
seo-960aD ff =
800 - 920 AD |
40
1oBc-90A0 ff
100 BC |
50
20 AD i
60
380 - 260 BC |
0 10 20 30 40 50 60 70 80 90
0 10 0 10 20 30 40 50 60 70 80 & 40 20 0
PERCENT ABUNDANCE
Sawgrass Marsh
Vbi
V b2
0 01
SQUARED CHORD
DISTANCE
Slough
Sawgrass Marsh - Slough
Mosaic
Slough
Cattail Marsh
Sawgrass Marsh
Near Tree Island
V b2
Vbi
Slough
oe
00102
Squared Chord Distance
Sawgrass Marsh
Near Tree Island
Myrica
Typha
Asteraceae
Cladium
OE O
Te 10 > Ierce © © 2 Fi > , Yl PD « t j 1 7
Text-figure 4.—Percent abundance of pollen of major plant groups at sites 1—4 in northern and central Everglades. SCD values show lowest
squared chord distance values for modern analogs; SCD <0.15 are considered to be close analogs for fossil assemblages. Cluster designations
refer to clusters designated in Text-figure 3 and Table 3.
CLIMATIC AND ANTHROPOGENIC IMPACTS: WILLARD ET AL.
SITE 3
e
Re Soo xe
S oe w
RS S & se
Radiometric Dat P ge oe
adiometric Dates ¢ eS we s
~1960AD « \
~1900AD =”
1425 - 1635 AD §
30 }
1005 - 1220AD E,, =
50 +
235-483 AD
60
0 1020 0 10 20 30 0 10 20
PERCENT ABUNDANCE
SITE 4
~1960AD =
~1900AD s
1295-1420AD J ~ =
40
65-240AD § ~
780-405 BC |
0 10 20 0 10
PERCENT ABUNDANCE
0 10 20 30
f) 0 10 20 30 40 50 60 70
SQUARED CHORD
DISTANCE
|
0 QO 10 20 30 40 50 60 70 80 0 010
SQUARED CHORD
DISTANCE
Text-figure 4.—Continued.
Va
Vb1
V b2
Vb1
EeVic2me
Vb1
ING
Vb1
IV d1
Vb1
V b2
Ve
V b2
Vb1
V b2
Vb1
Vb2
Vb1
Sawgrass Marsh
Slough
Sawgrass Marsh Near Tree Island
Slough
Sawgrass marsh
Slough/Sawgrass
marsh transition
Slough
Sawgrass Marsh
Near Tree Island
Slough
Sawgrass Marsh
Near Tree Island
Slough
E34 Myrica
= Asteraceae
Cladium
49
50 BULLETIN 361
1998
1995
1992
1989
1986
1983
1980
1977
1974
1971
1968
1965
1962
1959
1956
1953
1950
1947
1944
1941
1938
1935
1932
1929
1926
1923
1920
1917
1914
Wy
1908
1905
1901
1898
1895
120) 25 Isa Wee W4'o) 4s
Annual Precipitation (cm)
Year (A.D.)
Text-figure 5—Annual rainfall (9 point smoothing) for Ever-
glades, 1895-1998. Vertical line indicates 1895-1998 average.
Based on precipitation data for Florida region 5 obtained from
NOAA at http://www.ncdc.noaa.gov/onlineprod/drought/xmgr.html.
and Dineen, 1994), and the reduced water depths re-
corded in sediments from site 1 after 1960 illustrates
the impact of that hydrologic change. The increased
abundance of cattail pollen at site 2 after 1960 is cor-
related with increased levels of phosphorus in both
sediments and pore water from cores (Orem ef al.,
1999). After 1960, agricultural activities in the EAA
expanded rapidly: the subsequent increased phospho-
rus concentrations and cattail abundance recorded in
sediments at site 2 appears to be tied to greater nutrient
runoff from the EAA after that time.
VEGETATIONAL HISTORY: SOUTHERN
EVERGLADES SITES
PRE-20'! CENTURY VEGETATION
The impact of climatic changes on hydrology and
salinity in the southern Everglades is illustrated in four
cores collected in Taylor Slough and near Florida Bay
(Sites 5—8, Text-fig. 1). From about 0 A.D. to 1200
A.D., fresh-water marshes, analogous to sawgrass
marshes and wet prairies found in the southern Ever-
glades today, dominated the four sites (Text-figs. 6—9).
Even at the southernmost site near Florida Bay (site
8), mollusc assemblages of this interval consist exclu-
sively of terrestrial and fresh-water taxa (Text-fig. 9).
The first occurrence of brackish mollusc and benthic
foraminifer taxa there in the late 13" century coincides
with a decreased abundance of pollen typical of fresh-
water marshes. By the 16" century, vegetation at that
site fluctuated between wet prairies and brackish
marshes, indicating periodically shallower water and,
possibly, salt-water intrusion. The last occurrence of
terrestrial and fresh-water molluscs and algae was in
the 17"-18" century, indicating minimal fresh-water
influence after that time. Marsh vegetation persisted at
this site until late 19" or early 20" century, when man-
groves became the dominant element of the vegetation.
Pollen assemblages from site 5, located in very thick
peats in the central part of Taylor Slough, indicate
changes from sloughs and deep-water conditions to
wet prairies/sawgrass marshes with moderate water
depths at about 2,800 yrBP (Text-fig. 6). Although
sedimentation rates were very low at this site, little
vegetational variation is recorded at this site from
2,800 yrBP until the 20" century. Sites 6 and 7 provide
little data on vegetational trends prior to about 1300
A.D. because the basal age of the core at site 6 is about
1200 A.D. and because 5—10 cm thick intervals were
analyzed for the lower 60 cm of the core collected at
site 7. At site 6, pollen assemblages indicate fluctuat-
ing hydrologic conditions from about 1300 A.D. to
1900 A.D., with vegetation ranging from wet prairies
to brackish marshes to sawgrass marshes, sloughs, and
mangroves (Text-fig. 7). Conditions appear to have
been less variable at site 7 during that time (Text-fig.
8), although slower sedimentation rates decrease the
time resolution possible in that interval. As seen at
northern Everglades sites, Myrica abundance doubled
in the 17" century, possibly indicating regionally drier
conditions. From the 17" through 19" centuries, veg-
etation at all four sites varied little.
CLIMATIC AND ANTHROPOGENIC IMPACTS: WILLARD ET AL. 51
SITES
Radiometric Dates of
; ~ 1960 AD s
(4) ~1900AD 8
<< 10
20
465-645AD g ~
40 }
780-405BC §
60 |
70 |
1975-1760BC §
0 10 20 0 10 20 30 40 50 60 70 80 0 10
PERCENT ABUNDANCE
Rou
3 we
x &
Cua ee
con w
Ny s >
a es &
o” o
IV d1
IV d2
Wet Prairies/
Ivd1 Sparse Sawgrass
Marshes
| vil Sloughs
|
|
j !
10 20 30 0 10 20 0 0 0.10 — Myrica
SQUARED CHORD
DISTANCE Asteraceae
[7] Sagittaria
Text-figure 6.—Percent abundance of pollen of major plant groups at site 5, central Taylor Slough. SCD values show lowest squared chord
distance values for modern analogs; SCD <0.15 are considered to be close analogs for fossil assemblages. Cluster designations refer to clusters
designated in Text-figure 3 and Table 3.
20TH CENTURY VEGETATION
Vegetation at the southernmost site (site 8) changed
little throughout the 20" century, because present sa-
linity levels and vegetational composition apparently
were established by late in the 18" century. Site 7,
collected farther north in dwarf mangrove scrub,
changed from fresh-water marshes to mangrove as-
semblages in the early to mid-20" century. This
change probably reflects the combined effects of nat-
ural sea-level rise and reduced fresh-water flow from
the Everglades. Additional evidence for hydrologic
changes along Taylor Creek is provided by examina-
tion of aerial photographs taken between 1940 and
1994. These photographs show infilling of a lake along
Taylor Creek, reducing its area by about half and in-
filling parts of the coastal zone west of the Taylor
Creek mouth (Schaffranek er al., 2001).
In Taylor Slough, vegetation at both sites 5 and 6
changed after 1930. At site 5, pollen of asters and
sedges increased greatly, which is suggestive of greater
disturbance, fluctuating water levels, and/or drier con-
ditions. At site 6, mangrove abundance decreased by
1960, and marsh pollen (Cladium, other Cyperaceae,
Poaceae, and Typha) increased to abundances unprec-
edented in the past 2,000 years. This shift to sawgrass
marshes may reflect the influence of construction of
the C-111 canal and South Dade Conveyance System
in the early 1960s; management of these control struc-
tures ensures minimum monthly deliveries of water to
Taylor Slough (Light and Dineen, 1994), possibly pro-
viding more annually constant hydrology, deeper wa-
ter, and longer hydroperiods than occurred naturally at
this site.
DISCUSSION
Over most of the last two millennia, water depths
in much of the Everglades appear to have been deeper
than today with longer hydroperiods. In the Ridge and
Slough, the presence of slough vegetation between 0
A.D. and 800 A.D. and between 1200 A.D. and 1600
A.D. indicates the long-term existence of deep water
and long hydroperiods. In Taylor Slough and sites near
Florida Bay, the presence of wet prairie and sparse
sawgrass marsh vegetation prior to 1200 A.D. indi-
Nn
bo
SITE 6
Radiometric PS Ay
Dates &*
~1960 AD # 10-
20 £
~1900AD #8
1475 - 1665 AD §
1455 - 1665 AD 8
50
1305 - 1460 AD § © =
0 20 40 60 80
PERCENT ABUNDANCE
BULLETIN 361
2
Re)
Rs
&
we
we
se S
” oe
IVdt
IV d2 Wet Prairies
Tb
WV d2
Wet Prairies with
Dwarf Mangroves
IV d2
Brackish Marshes
Ww
V b2
Sawgrass Marshes
Near Tree Islands
Ve
Cattail Marshes
Brackish Marshes
m Southwest Florida
Mangrove Forests
wee Dwarf Mangroves
Vb1
IV dt Wet Prairies
SQUARED CHORD :
DISTANCE ES) myrica
EE] Asteraceae
Eq Wha
Text-figure 7—Percent abundance of pollen of major plant groups at site 6, southern Taylor Slough. SCD values show lowest squared chord
distance values for modern analogs: SCD <0.15 are considered to be close analogs for fossil assemblages. Cluster designations refer to clusters
designated in Text-figure 3 and Table 3.
cates fresh-water conditions and short to moderate hy-
droperiods. These long-term differences in hydrologic
regimes between northern and southern sites is reflect-
ed in the present vegetational distribution and substrate
type and probably reflects underlying geologic con-
trols, including differences in the underlying lime-
stone. Northern sites are underlain by the Fort Thomp-
son Formation, which consists of a series of alternating
freshwater and marine limestones. Southern sites are
underlain by the Miami Limestone, consisting of an
upper, oolitic limestone overlying a bryozoan facies
(Gleason and Stone, 1994).
The impact of climate variability on Everglades
vegetation and hydrology is indicated in sediments de-
posited between the 9" to 13" centuries. During this
time, water was shallower at most Ridge and Slough
sites, with the greater abundance of weedy annuals
suggesting rapid drying or prolonged droughts. Near
Florida Bay, salinity increased at this time. Also during
this interval, known as the Medieval Warm Period
(MWP) (9" to 14" centuries) (Hughes and Diaz, 1994),
prolonged droughts have been documented in Haiti,
Mexico, and North Carolina, as have lower lake levels
in Mexico, and increased fire frequency in Costa Rica
(Hodell et al., 1995; Horn and Sanford, 1992; Metcalfe
and Hales, 1990; Stahle er al., 1988). Although slow
sedimentation rates at Everglades sites preclude de-
tailed temporal analysis of vegetational patterns
throughout this interval, the evidence for drier condi-
tions is consistent with a broader regional pattern of
warmer, drier conditions during the MWP.
Our data also indicate slightly drier terrestrial con-
ditions in the 16" through 19" centuries. Although sed-
imentation rates in these cores do not allow detailed
analysis of this interval, the broad patterns shown dur-
ing the interval known as the Little Ice Age (LIA)
(1550-1850 A.D.) (Bradley and Jones, 1993) are con-
sistent with records of 65!°O from coral skeletons in
Biscayne Bay indicating drier conditions in the 18"
and 19" centuries (Swart et al., 1996). Sea-surface
temperatures 1°C cooler than today also are indicated
by 6'8O data from corals and foraminifers in the Flor-
ida Straits and Sargasso Sea for the late part of the
LIA (Druffel, 1982; Keigwin, 1996). Cooling of Gulf
surface temperatures in climate models is correlated
with lower precipitation levels (Overpeck ef al., 1989),
which is consistent with drying trends shown by LIA
pollen records from the Everglades.
CLIMATIC AND ANTHROPOGENIC IMPACTS: WILLARD ET AL. 5/3)
SIME?
SF s
ss
2) 2)
Radiometric Dates O “
~1960AD =
~1900 AD =
1680-1760 AD |
30
40
60
70
280-400 AD
so =
90 £
0 10 20 30 40 50 60 70 80 90 0 10
0 10 20 30 40
PERCENT ABUNDANCE
2
&
os
ys
s
e
AS *
3@ Pa
&” &
Vc
Wd2
Ve Dwart Mangroves
IV d2
Va Brackish and
i Freshwater Marshes
Va
IV d1
IV d2
Va
Wet Prairies/
Sparse Sawgrass
Marshes
IV di
0 10 20 30 40 50 0 0 10 20°
SQUARED CHORD ES Myrica
DISTANCE
E-] Asteraceae
Text-figure 8.—Percent abundance of pollen of major plant groups at site 7, Taylor Creek. SCD values show lowest squared chord distance
values for modern analogs; SCD -
2 2
designated in Text-figure 3 and Table 3.
The advent and evolution of water-control practices
in south Florida has resulted in unprecedented changes
in vegetational distribution. The greatest and most re-
gional impacts were seen by 1930, when shallower
water and shorter hydroperiods are documented in
most Everglades cores. These drier conditions oc-
curred at a time of higher than average precipitation,
indicating that vegetational and hydrologic changes re-
sulted from hydrologic changes to the system. The
combined effects of reduced sheet flow due to water
diversion through canals and disrupted seasonal water
supply due to construction of the Hoover Dike were
sufficient to shift the Ridge and Slough ecosystem
from a deep-water slough to a sawgrass marsh with
abundant weedy annuals. At both sites in Taylor
Slough, weedy species also increased in abundance,
although the impact on water depths is less clear. In
contrast, alterations associated with construction of the
Central & South Florida (C & SF) Project in the 1950s
and 1960s appear to have had more localized impacts
on marsh vegetation. Some sites underwent striking
changes after 1960, such as expansion of cattail marsh-
0.15 are considered to be close analogs for fossil assemblages. Cluster designations refer to clusters
es at nutrient-enriched sites, whereas other sites saw
minimal change.
These observations are particularly relevant in light
of suggestions that restoration goals for the Everglades
be aimed at rainfall and runoff conditions that existed
before the C&SF Project was completed in the early
1960s (Light and Dineen, 1994). Our data clearly in-
dicate that the major alterations of the system occurred
in the early 20" century, when construction of muck
canals linking Lake Okeechobee to the Atlantic Ocean,
the Hoover Dike along the southern shore of Lake
Okeechobee. Thus, the early- to mid-20" century
should not be construed as the “natural” state of the
Everglades ecosystem. Rather, reconstructions of the
early 19"-century Everglades provide a more realistic
estimate of the “natural” state that our data indicate
had persisted for the previous two to three centuries.
These data also provide a model for how an undis-
turbed wetland ecosystem responds to changing cli-
matic conditions; data indicating drier conditions dur-
ing the Medieval Warm Period are consistent with the
predicted response of the Everglades wetland to pos-
Nn
L
SITE 8
Radiometric Dates
~1960AD * "|
-1900AD = 2 |
1570-1730AD g |
50
1120-1280AD F
70BC-70AD § |
10 0 10 20 30 a)
0 10 20 30 40 50 60 70 80 90
PERCENT ABUNDANCE
BULLETIN 361
@
F
S
Ks &
¥
s
& &
eo S
\V dt
WV d2
[an Dwart Mangroves
IV d2
)
ty
Y) ——
Wat Brackish Marshes
H Vide and Wet Prairies
°
)
e
e dt
°
e
J
e
e
We Wet Prairies/
) Sparse Sawgrass
Marshes
dt
E43 Myrica
% 40 0 0 02 04
SQUARED CHORD F=) Asteraceae
DISTANCE
Text-figure 9.—Percent abundance of pollen of major plant groups at site 8, Mud Creek. SCD values show lowest squared chord distance
values for modern analogs; SCD <0.15 are considered to be close analogs for fossil assemblages. Cluster designations refer to clusters
designated in Text-figure 3 and Table 3. Ovals indicate presence of molluscs or foraminifers picked from sediments. Foraminifer presence also
was noted when foraminifer linings (““microforams”’) were observed in palynological preparations. Fresh-water algal presence is based on
presence of Ovoidites in palynological preparations.
sible global warming over the next century (Watson et
al., 1996). Additionally, the drier conditions indicated
during parts of the Little Ice Age are consistent with
climate model predictions of lower precipitation due
to cooling of Gulf surface temperatures (Overpeck et
al., 1989). A more difficult question to address confi-
dently is the impact of future warming on an already
disturbed ecosystem. Prior to compartmentalization of
the marshes by canals and levees in the 20" century,
it appears that broad-scale subenvironments of the Ev-
erglades responded similarly to climatic fluctuations.
Changes over the last century have altered water levels
and the natural rhythm of seasonal hydroperiod fluc-
tuations in the Everglades wetland, resulting in a land-
scape characterized by localized fluctuations in hydro-
period, disturbance, nutrient status, and salinity. The
resulting ecological heterogeneity makes modeling
predictions of system-wide changes due either to fu-
ture climatic change or alterations in water-control
practices much more problematic.
REFERENCES CITED
Bradley, R.S., and Jones, P.D.
1993. ‘Little Ice Age’ summer temperature variations: Their na-
ture and relevance to Recent global warming trends. The
Holocene, vol. 3, pp. 367-376.
Davis, J.H., Jr.
1943. The Natural Features of Southern Florida. Geological Bul-
letin No.25. Florida Dept. of Conservation, Tallahassee,
FL, pp. 130-215
Davis, S.M.
1994. Phosphorus inputs and vegetation sensitivity in the Ev-
erglades. in S.M. Davis, and J.C. Ogden, eds., Everglades:
The Ecosystem and its Restoration. Delray Beach, FL, St.
Lucie Press, pp. 357-278.
Davis, S.M., Gunderson, L.H., Park, W.A., Richardson, J.R., and
Mattson, J.E.
1994. Landscape dimension, composition, and function in a
changing Everglades ecosystem. in S.M. Davis and J.C.
Ogden, eds., Everglades: The Ecosystem and Its Resto-
ration. St. Lucie Press, Delray Beach, FL, pp. 419-444.
Druffel, E.M.
1982. Banded corals: changes in oceanic carbon-14 during the
Little Ice Age. Science, vol. 218, pp. 13-19.
CLIMATIC AND ANTHROPOGENIC IMPACTS: WILLARD ET AL.
Gauch, H.G., Jr.
1982. Multivariate Analysis in Community Ecology. Cambridge
University Press, Cambridge, 298 pp.
Gleason, P.J., and Stone, P.
1994. Age, origin, and landscape evolution of the Everglades
peatland. in S.M. Davis and J.C. Ogden, eds., Everglades:
The Ecosystem and its Restoration. St. Lucie Press, Del-
ray Beach, FL, pp. 149-197.
Hodell, D.A., Curtis, J.H., and Brenner, M.
1995. Possible role of climate in the collapse of Classic Maya
civilization. Nature, vol. 375, pp. 391-394.
Horn, S.P., and Sanford, J.R.L.
1992. Holocene fires in Costa Rica. Biotropica, vol. 24, pp.
354-361.
Hughes, M.K., and Diaz, H.F.
1994. Was there a “Medieval Warm Period’, and if so, where
and when? Climatic Change, vol. 26, pp. 109-142.
Keigwin, L.D.
1996. The Little Ice Age and Medieval Warm Period in the Sar-
gasso Sea. Science, vol. 274, pp. 1504—1508.
Kovach, W.L.
1999. MVSP—A MultiVariate Statistical Package for Windows.
Kovach Computing Services, Pentraeth, Wales, UK.
Kushlan, J.A.
1990. Freshwater marshes. in R.L. Myers and J.J. Ewel, eds.,
Ecosystems of Florida. University of Central Florida
Press, Orlando, pp. 324—363.
Langeland, K.
1990. Exotic woody plant control. Florida Cooperative Exten-
sion Service Circular, vol. 868: pp. 1-16.
Light, S.S., and Dineen, J.W.
1994. Water control in the Everglades: a historical perspective.
in S.M. Davis and J.C. Ogden, eds., Everglades: The Eco-
system and its Restoration. St. Lucie Press, Delray Beach,
FL, pp. 47-84.
Loveless, C.M.
1950. A study of the vegetation in the Florida Everglades. Ecol-
ogy, vol. 40, pp. 1-9.
Maher, L.J., Jr.
1981. Statistics for microfossil concentration measurements em-
ploying samples spiked with marker grains. Review of
Palaeobotany and Palynology, vol. 32, pp. 153-191.
MeVoy, C.W., Park, W.A., and Obeysekera, J.
2001. Landscapes and hydrology of the Everglades, circa 1850:
U.S. Geological Survey Open-file Report (in press).
Metcalfe, S., and Hales, P.
1990. Holocene diatoms from a Mexican crater lake—La Pisci-
na de Yuriria. Memoirs, California Academy of Science,
vol. 17, pp. 501-515.
NOAA CLIMVIS internet climate data site.
2000. — http://www.ncdc.noaa.gov/onlineprod/drought/xmgr.html
Orem, W.H., Lerch, H.E., Bates, A.L., Boylan, A., and Corum, M.
1999. Nutrient geochemistry of the South Florida wetlands eco-
system: sources, accumulation, and biogeochemical cy-
cling. in S. Gerould and A. Higer, eds., U.S. Geological
Survey Program on the South Florida Ecosystem—Pro-
ceedings of South Florida Restoration Science Forum,
Nn
Nn
May 17-19, 1999, Boca Raton, FL. U.S. Geological Sur-
vey Open-file Report, pp. 80-81.
Overpeck, J.T., Peterson, L.C., Kipp, N., Imbrie, J., and Rind, D.
1989. Climate change in the circum-North Atlantic region dur-
ing the last deglaciation. Nature, vol. 338, pp. 553-557.
Overpeck, J.T., Webb, T., III, and Prentice, I.C.
1985. Quantitative interpretation of fossil pollen spectra: dissim-
ilarity coefficients and the method of modern analogs.
Quaternary Research, vol. 23, pp. 87-108.
Riegel, W.L.
1965. Palynology of environments of peat formation in south-
western Florida. Unpublished PhD thesis, The Pennsyl-
vania State University, 189 pp.
Robbins, J.A., Holmes, C.W., Halley, R., Bothner, M.A., Shinn,
E., Graney, J., Keeler, G., Tenbrink, M., Orlandini,
K.A., and Rudnick, D.
2000. Time-averaged fluxes of lead and fallout radionuclides to
sediments in Florida Bay. Journal of Geophysical Re-
search, Oceans, vol. 105, pp. 28,805—28,821.
Schaffranek, R.W., Smith, T.J., I[1, and Holmes, C.W.
2001. An investigation of the interrelation of Everglades hy-
drology and Florida Bay dynamics to ecosystem processes
in south Florida. U.S. Geological Survey Open-file Re-
port, FS-049-01 (in press).
Schweitzer, P.N.
1994. ANALOG: A program for estimating paleoclimate param-
eters using the method of modern analogs. U.S. Geolog-
ical Survey, Reston, VA.
Stahle, D.W., Cleaveland, M.K., and Hehr, J.G.
1988. North Carolina climate changes reconstructed from tree
rings: A.D. 372-1985. Science, vol. 240, pp. 1517-1519.
Stuiver, M., Long, A., Kra, R.S., and Devine, J.M.
1993. Calibration—1993. Radiocarbon, vol. 35, pp. 244.
Swart, P.K., Dodge, R.E., and Hudson, H.J.
1996. A 240-year stable oxygen and carbon isotopic record in
a coral from south Florida: implications for the prediction
of precipitation in southern Florida. Palaios, vol. 11, pp.
362-375.
Talma, A.S., and Vogel, J.C.
1993. A simplified approach to calibrating C14 dates. Radio-
carbon, vol. 35, pp. 317-322.
Traverse, A.
1988. Paleopalynology. Unwin Hyman, Boston, 600 pp.
Vogel, J.C., Fuls, A., Visser, E., and Becker, B.
1993. Pretoria calibration curve for short lived samples. Radio-
carbon, vol. 35, pp. 73-86.
Watson, R.T., Zinyowera, M.C., Moss, R.H., and Dokken, D.J.
1996. Impacts, Adaptations and Mitigation of Climate Change:
Scientific-Technical Analyses. Cambridge University
Press, New York. 878 pp.
Willard, D.A., Weimer, L.M., and Riegel, W.L.
2001a. Pollen assemblages as paleoenvironmental proxies in the
Florida Everglades. Review of Palaeobotany and Paly-
nology, vol. 113, pp. 213-235.
Willard, D.A., Holmes, C.W., Korvela, M., Mason, D., Murray,
J.B., Orem, W.H., and Towles, T.
2001b. Paleoecological insights on fixed tree-island development
in the Florida Everglades: I. Environmental controls. in
F Sklar and A. van der Valk, eds., Tree Islands of the
Everglades. Kluwer Publishers, Amsterdam (in press).
L& ie Vokhio? lia
aw ~ “wt
7 ein 2s
_ - i coll eee {
i ~ot
iva _ J eh
rid . ‘ se a6 _ ¢ #8 honey AOR
= sl = = =~
; x on
ry A ne j , 4 a
7 Ti) : - ll v Lhd Pande _
: 7 ‘ , “4 a pie Mahaney
e iré 7 nd %
: a a”
= load ¥ 7 i y 33% Jona) UPD
k ee 2 Act oy Atte aml
4 . wv end
” » ig eal he —ith aime = ile ie
>a.
‘ x . . i a lye | SVE
= ‘ ’ he iL
& a : ns
wen » ee Vie sy 16 €& <4) seek wheat ~
ie & ,pe - tae
riot ae | ee ee
Sate a iad. Sh ae)
be AP
; 2 Qete= 81) adi 64 00
16 ~ ld pat oo) le 4 ree a iw a
Lee
o! « avs,
a4 = ie
Sty Z te
CHAPTER 5
HYDROLOGY, VEGETATION, AND CLIMATE CHANGE IN THE SOUTHERN EVERGLADES DURING
THE HOLOCENE
MARJORIE GREEN WINKLER!, PATRICIA R. SANFORD!, AND SAMANTHA W. KAPLAN?
‘Center for Climatic Research, Institute for Environmental Studies, University of Wisconsin—Madison,
1225 West Dayton Street, Madison, Wisconsin 53706
*Department of Geography and Center for Climatic Research, University of Wisconsin—Madison,
1225 West Dayton Street, Madison, Wisconsin 53706
ABSTRACT
Paleoecological study of 18 AMS-dated sediment cores from the southern Everglades provides evidence of a shifting mosaic
of biotic communities in the past similar to those on the Everglades landscape today. Our results also indicate initiation of
Everglades peatlands at 5000 yr B.P., the importance of fire (from charcoal analysis) as a structuring agent in the tropical
Everglades, evidence of past plant communities (/soetes marshes) not present in the modern landscape, and the introduction of
exotic plants (Schinus, Casuarina, etc.) today. Past vegetation changes are documented by pollen and sclereid changes. Past water
level changes are documented by plant community changes, diatom species and habitat changes, sponge spicule changes, and by
intervals of peat (wet) or marl (dry) deposition in the sediment cores. Marl deposition dominates today at these sites in the
southern Everglades, a long-term trend exacerbated by human impacts. The Everglades may become less complex in the near
future as introduced plants outcompete native vegetation and decreased water levels result in decreased peat production. A
rewatering plan must include both wet and dry seasonal cycles in order to preserve the shifting mosaic nature of the landscape
and to maintain the Everglades as a functional habitat for both plants and animals.
INTRODUCTION
The Everglades is an unique subtropical and tropical
hydroecologic region found at the southern end of the
Florida peninsula below latitude 28°N. Sheetflow from
the Kissimmee-Okeechobee drainage maintains the di-
verse biotic communities in the Everglades as it flows
through marshes, hammocks, sloughs, and swamps to
the ocean. Human activities in the past 100 years have
decreased flow of water critical to the health of the
natural Everglades. As South Florida population 1n-
creases in coming years, pressure to divert more water
and drain more land for development will increase.
Recently Congress increased federal and local funding
for a large-scale Everglades restoration effort. Con-
flicting demands from developers, landowners, and
preservationists make restoration decisions by federal
and state agencies charged with water and environment
management extremely difficult.
To make informed choices managers need to under-
stand the evolution of the Everglades landscape. Only
paleoecologic analyses can provide an understanding
of the processes which made the Everglades. Previous
paleoecologic studies of the region (Riegel, 1965;
Nichols, 1974; Gleason and Spackman, 1974; Gleason
and Stone, 1994) were concentrated in the coastal
southwestern or northern (the Water Conservation Ar-
eas, WCA) sections of the historical Everglades. We
undertook paleoecological analysis of 38 sediment
cores (Text-fig. 1) from 17 sites in the southern section
of the Everglades concentrating on Everglades Nation-
al Park (ENP) and the southern parts of WCA3 to
Ty © 4800 ]
Everglades hs /\\ Basal Ages Nan 4590 | |
Core Sites = 235 Poi. au |
Miles K— Dass ~Y
Q 1 20 \
6 = 2% 32 |3055\\ ©2000;
Kilometers oor 061445 "4
3460
| |
ll - | A
lente mae a ————
lf Le7Ag/
t \f 3500 @/ (\ sd
{ Vf, \ \(e
ims a Y/,
TN] 4 lease i) y, |
=Sra es y tN 4350 04990 | asi
es F ons ine SSS
| | ) @ 23) \ ‘f | { #2840 | (| y
epee? \ \ \/ // Biscayne eae i \ / -/ Biscayne
| i | 7 f Bay 1 f | ve \ Bay
al Nc ae | (Gwar
| Rocky Glade | |eCastellow Hamnmock | |e 4000 {
\ }
ee OES
Kyeress Dome eGumbs Limbo Trail
@cii
e Ever 6 /
|9Mile Pond * Craighead Pi Pond’ N
Try) as
Paurotis ards
e
yer
Florida Bay Pe 25°
v4
LY
ie 2290
@Lignum Vitae Key ee
A
Text-figure |.—a and b. Map of coring sites for this study in the
southern Everglades (a). Distribution of basal organic sediment ra-
diocarbon ages for cores in the southern Everglades (b). Filled cir-
cles denote sites in this study and are all AMS dates. Open circles
show basal ages from studies cited in text and are all conventional
bulk radiocarbon dates. All ages are expressed in radiocarbon years
before present (1950 A. D.) (yr B. P.).
L67A L67C
Large diam
0=60 cm
Depth Ocm
5
d modern /-31 40
10
15 | modern /-31 40
fa 1790+ 30 /-26 67
20
500+ 40 /-25.50
25
58 BULLETIN 361
C111 Canal C111 Canal
P-core 1 Hiller 4
0=40 cm
Ocm g
20
40
710 +55 /-25.23
100
550+40/-26 18 120
140
160
H3010+50/-25 71
180
Ey limestone
peat
algal floc
gyttja
silt
clay marl
Daee
peaty marl ®@@ shells
organic marl i] unsampled
marly peat sandy marl
charcoal-stained marl
Text-figure 2.—Sediment stratigraphies of cores obtained from Everglades sites. Radiocarbon dates and stable carbon isotope values (6'C)
are to the right of each sediment core. Water depth is represented by 0 = # at the top of each stratigraphic column.
provide a chronology for the development of the Ev-
erglades and to compare the past vegetation and hy-
drology with that of the modern landscape.
ACKNOWLEDGMENTS
We thank the many people at both Everglades Na-
tional Park South Florida Natural Resources Center
and at the Center for Climatic Research (CCR) who
helped facilitate this study. The field work was aided
by D. McJunkin, R. S. Webb, P. Killoran, J. Stott, S.
S. Winkler, W. Loftus, W. Robertson, R. Rehrer, and
M. Nesbitt. AMS-radiocarbon dates were run at the
AMS-facilities at Woods Hole Oceanographic Institute
and Lawrence Livermore Laboratory. I. Treichel, Uni-
versity of Wisconsin Chemistry Department ran the
samples for mass spectrometry. The Fairchild Tropical
Garden, Miami, Florida, allowed access to their her-
barium. This study was funded by NPS grant agree-
ment #CA 5280—2—9010 from Everglades National
Park to M. G. Winkler and by NSF grants ATM-
9101919 (COHMAP), ATM-9510668 (TEMPO),
ATM-9318973 (Interdisciplinary Research Program in
Climatology) to J. E. Kutzbach at CCR, University of
Wisconsin—Madison. The stable carbon isotope mea-
surements were partially supported by NIH Grant
GM18939 to W. W. Cleland, Dept. of Biochemistry,
UW-Madison.
AMS Radiocarbon dates from the National Ocean
Sciences Accelerator Mass Spectrometry (NOSAMS)
facility at Woods Hole Oceanographic Institute were
partially funded by NSF cooperative agreement
OCE801015. This is CCR Contribution No. 757.
METHODS
Cores were taken with the Hiller corer, hand-driven
polycarbonate tubes, freeze corer, or modified Living-
ston piston-corer as field conditions, transport consid-
erations, and/or sediment type dictated. All sites were
cored to bedrock except for Gator Lake, a deep sink-
hole lake, although at some sites (Rocky Glades and
C111) coring was discontinuous (Text-fig. 2). Sedi-
ments from all sites were analyzed by loss-of-weight-
HYDROLOGY, VEGETATION, AND CLIMATE: WINKLER ET AL. 59
Ever 6 Craighead Pond Craighead Pond Castellow Lignum Vitae
Small diam Hiller Hammock Hiller 2
0=35 cm 0=35 cm 0=7.5 cm 0=10 cm
Depth Ocm
20
30
40
50
60
70
80
90
110
algal floc ==s| limestone
gyttja peat
silt marly peat
clay marl
905+30 /-27 47
250+30/-25 00
algal floc
peaty marl shells
[al unsampled
organic marl
sandy marl
charcoal-stained marl
Text-figure 2.— Continued.
on-ignition (LOI) (Berglund and Ralska-Jasiewiczowa,
1986) and high temperature LOI (HLOI) (Dean, 1974)
to determine sediment composition (marl or peat). Fif-
ty samples of basal and other organic sediments from
the cores were radiocarbon-dated by Accelerator Mass
Spectrometry (AMS) techniques. Sediment chemistry
was obtained by Inductively Coupled Plasma-Optical
Emission Spectrometry (ICP-OES) analysis run at the
University of Wisconsin—Madison Extension Soil and
Plant Analysis Laboratory (Schulte er al., 1987) to de-
termine changes in the concentrations of atmospheri-
cally- and locally-deposited elements and nutrients.
Selected cores were analyzed for charcoal following
the method of Winkler (1985); diatoms and other algae
(Winkler, 1994a); pollen (Gray, 1965; Faegri and Iver-
sen, 1975); and cladocera (Frey 1986). Sponge spic-
ules and Nymphaea (white water lily) and Cladium
(sawgrass) sclereids were counted during cladoceran
and/or pollen counts (Winkler et al., 1996a).
RESULTS AND DISCUSSION
STRATIGRAPHY AND DATING
The unique landscape that is the Everglades in
southern Florida is one that presents dating and inter-
pretive difficulties for the paleoecologist. The land-
scape evolved from and is maintained by seasonal wa-
ter-level extremes and frequent peat-consuming fires
Charcoal is found at many levels in sediment cores
(Cohen and Spackman, 1974; Duever ef al., 1985).
Low water levels during drought, exacerbated by mod-
ern drainage and canalization, expose ancient peats
which are then destroyed by oxidation and decompo-
sition (Shih, 1980; Duever er al., 1985; Snyder and
Davidson, 1994). Dewatered peat is also loose and fri-
able and can be scattered by winds or moved down-
stream when water levels increase. In addition, gases
from the decomposition of peat during low water lev-
els weaken tree islands and permit storm-force winds
to cause tree islands and chunks of hammock peat to
break off and be transported downstream where they
60 BULLETIN 361
Rocky Glades Rocky Glades Paurotis H. Paurotis H. WCA 3B
Hiller Frozen F Hiller Small diam Hiller
0=24 cm 4cm 0=0 cm 0=0 cm 0=37 cm
Depth ocm Ocm Ocm
20
iG 10 20
310+ 40/-26 57
60
80 20 ay
100 30:/-25.a¢
30 /-21.4¢
120 30 60
140
160.) | AG 80
180.
134
B00 = 100
220 23.45
240 29 tao
260
140
280 70
300
+40/-18 09
160
320 a
180
90
ORES UG26-59) 200 4990+ 40 /-21 67
100
C 90 513
algal floc Erg limestone peaty marl ®@ shells
id gyttja peat organic marl =| unsampled
silt marly peat ee sandy marl
= clay marl charcoal-stained marl
Text-figure 2.— Continued.
(Robertson, 1953; Gunderson and Snyder, 1994). become the locus of further peat accumulation thus
confounding interpretation of the Everglades environ-
ments. When there is marl in the sequence there is the
possibility of hiatus in the record because transition to
marl-forming conditions may involve sediment de-
composition. In addition, when peat is subject to low
water levels and/or fire there is also the possibility of
hiatus in the record. Obtaining a continuous sedimen-
tary record is not possible at many sites in the Ever-
glades. In order to interpret this volatile region, many
radiocarbon dates are required to determine the chro-
nostratigraphic context at each site. Until a regional
pattern is clear, intervals of interest in Everglades cores
should be dated individually. Our goal was to date all
HYDROLOGY, VEGETATION,
AND CLIMATE: WINKLER ET AL.
61
A23 A06 Gator Lake Gator Lake
Hiller Hiller Livingstone P-core 1
0=35 cm 0=30 cm 0=259m 0=25.9m
Depth Ocr Ocmg : Ocm Ocm
5 00 20 100
40
10 : 40 200
x 900+60 80
0
1110+40 /-27 69
15 60 nee SO 26-307300
100 +30/-25.56
20 80 400
0
25 60 500
30 600 1950+45 50
2040 +80
35
40
45
50 “i }2840 +70
algal floc a limestone peaty marl shells
= gyttja peat organic marl =a unsampled
silt marly peat sandy marl
=| clay marl charcoal-stained marl
Text-figure 2.—Continued.
basal sediments and bracket-date as many of the ob-
vious sediment transitions in our cores as funds al-
lowed. Additional dates will be obtained as studies on
these cores continue.
Text-figure 2 presents the sediment stratigraphy, ra-
diocarbon dates, and stable carbon isotope values
(64C) available for the sites. Sediment core depths
ranged from 7 m of sediment at Gator Lake to 21 cm
from a cypress dome. Alternations between peat and
marl deposition are documented at several sites by sed-
iment changes, LOI and carbonate analysis, and stable
carbon isotope values. Peat deposition represents deep
water-long hydroperiod, while marl deposition repre-
sents low water level-short hydroperiod. We hypothe-
size that temporal concurrence in hydrologic indicators
suggests climatic forcing while local hydrological var-
iability may be a result of topographical differences.
Rocky Glades, L67C, Castellow Hammock and
WCA 3B have basal ages between 5000 and 4000
years Before Present (yr B. P.) (Table 1). These early
dates occur at the base of peat overlying bedrock and
correspond to basal ages obtained in earlier studies
(Gleason and Spackman, 1974; Gleason and Stone,
1994). Cores taken at sites L67A, C111 Canal, and
A23 show organic sediments or peat overlying lime-
stone bedrock with basal ages ranging from 3500 to
2500 yr B. P. A23 radiocarbon dates indicate that the
bottom 30 cm of sediment accumulated rapidly. Sites
BULLETIN 361
Table 1.—Radiocarbon dates (accelerator mass spectrometric [AMS] and conventional) and stable carbon isotope values (5'°C) for this study.
CAMS dates are from Center for Accelerator Mass Spectrometry, Lawrence Livermore National Laboratory. OS dates are from Woods Hole
National Ocean Sciences AMS facility. WIS dates are from University of Wisconsin—Madison Radiocarbon Dating Laboratory. 6'°C values of
—25 A are assumed rather than measured values.
Lab. ID Site Depth (cm) Material Date Le:
CAMS-7483 A23 2.03.0 org. marl 1380 + 70 —25 A
CAMS-7484 A23 5.09.0 charcoal-stained marl 1660 + 100 —25 A
CAMS-7485 A23 9.0-12.0 marl 1890 + 70 — Psy IN
CAMS-7486 A23 12.0-21.0 org. marl 2600 + 70 —25 A
CAMS-7487 A23 21.0—25.0 org. marl 2840 + 80 —25 A
CAMS-7529 A23 25.0-33.0 marl 2830 + 60 —25 A
CAMS-7499 A23 41.0-S0.0 marl 2840 + 70 —25 A
OS-3547 Rocky Glades, FF 101.6—-106.7 peat 1040 + 30 —=25:3
OS-3554 Rocky Glades, FF 116.8-121.9 marl 2370 + 30 —21.4
OS-3549 Rocky Glades, H2 195-200 org. carbon 1770) = 30 —21.34
OS-3753 Rocky Glades, H2 195-200 inorg. carbon 5270 + 45 —0.23
OS-3550 Rocky Glades, H3 232-242 wood 2900 + 50 —23.45
OS-3548 Rocky Glades, H1 295-300 gyttja 4880 + 40 —18.09
OS-3556 Lignumvitae 9.0-14.0 peat DOS) —25.44
OS-3557 Lignumvitae 20.0—25.0 peat 2502230) —26.85
OS-3558 Lignumvitae 50.0—60.0 peat 600 + 25 —24.72
OS-3559 Lignumvitae 75.0-79.5 org. carbon 1040 + 35 =P Py
OS-3752 Lignumvitae 75.0-79.5 inorg. carbon 2610 + 40 +0.62
OS-3560 Lignumvitae 96-100 peat 2290) 250 =25 250)
OS-3551 Castellow Hammock 61.0—62.3 wood 4000 + 30 = 7ze ils)
OS-3561 A06, H3 57.0—60.5 peat 1960 + 30 —26.31
OS-3562 A06, H3 70.0-75.0 peat 2100 + 30 =25:56
OS-8203 A06, H3 20.0—25.0 peat modern —26:53
OS-3731 L67A, F 135) peat modern —31.4
OS-3732 L67A, L 22.5 marl 3500 + 40 =75)-5)
OS-2746 L67C, P-core 2 5.5-6.5 peaty marl modern —25 A
OS-3733 L67C, P-core 2 14.3-15.5 peat 1790 + 30 —26.97
OS-3734 L67C, P-core 2 25.3-26.5 wood 4350 + 35 —25.86
OS-3735 Paurotis Hammock, P-core 2 10.2-10.9 wood 310 + 40 —26.57
OS-3736 Paurotis Hammock, Hiller 75.0-87.0 peat 200 + 45 =26.59
OS-2742 Paurotis Hammock, Hiller 100-103 marl i170) == ©I0) =). 118)
OS-3737 GU P=core! 1 10.6—11.7 peat TNO) == 35) = 25},313)
OS-3738 C111, P-core 1 17.6-18.6 (top) peat 550 + 40 —26.18
OS-2739 C111, Hiller 4H 169-174 peat 3010 + 50 = 27/1
OS-3744 Gator Lake 93.0-95.6 gyttja 1110 + 40 —2 769)
OS-3745 Gator Lake 190-191 gyttja 1010 + 50 PNT
OS-3746 Gator Lake 660-68 | gyttja 1950 + 45 = ies)
WIS-2338 Gator Lake 694-700 gyttja 2040 + 80 —27.67
OS-2745 Craighead Pond, P-core 22.8-23.8 peat 1500 + 30 20:59)
OS-2744 Craighead Pond, P-core 36.0-37.5 peat w/ clay 2120 + 40 —24.92
OS-2743 Craighead Pond, Hiller | 74.0-79.0 peat 905 + 30 —27.47
OS-2741 Ever6, Hiller | 68.0—73.0 marl w/ peat 2560) 2335 —24.24
OS-2740 WCA 3B, Hiller 187-198 peat 4990 + 40 —21.67
at AO6, Craighead Pond, and Lignumvitae have basal
ages between 2300 and 2100 yr B. P., although only
Craighead Pond has peat at the base. Basal ages on
the remaining cores fall between 1170 and 530 yr B. P.
Paurotis Hammock core shows a transition from marl
to peat at 310 yr B. P.,, possibly indicating the date of
formation for this hammock. Sedimentation probably
began at all of these sites in response to a rising water
table. As noted below, our sequence of sediment cores
would suggest at least four periods of long hydroper-
iod.
Marl deposition is observed at Rocky Glades and
A23 between 2900 and 2800 yr B. P. and at Ever6 at
2560 yr B. P. Radiocarbon ages on basal portions of
other marl deposits tend to cluster at about 1770 yr
B. P., 1040 yr B. P,, and between 210 yr B. P. and the
present.
Data from these cores indicate that water levels rose
high enough to allow sediment accumulation begin-
ning about 5000 yr B.P. At all of these sites, except
WCA 3B, basal sediments consisted of peat overlying
bedrock. While WCA 3B has one of the earliest basal
HYDROLOGY, VEGETATION, AND CLIMATE: WINKLER ET AL. 63
dates on peat (4990 + 40 yr B. P.), it overlies a marl
deposit. The presence of marl under peat at WCA 3B
suggests a chronological sequence similar to that of
calcitic mud underlying peat in southern Lake Okee-
chobee (Brooks 1974; Gleason and Stone, 1994) and
the Corkscrew Swamp (Kropp, 1976)—deposits which
have all been radiocarbon dated at ca. 6500 yr B. P.
After the earliest peat deposition (SOOO—4000 yr
B. P.) water levels continued to rise until c. 2900 yr
B. P. when a number of sites show a transition to marl.
The period from 2900—2300 yr B. P. is characterized
by marl deposition. Gleason and Stone (1994) also re-
port calcitic mud deposition at 2432 yr B. P. from Core
25 in south Water Conservation Area 3A west of Le-
vee 67 and interpret hydroperiods to be short. How-
ever, an increase in water levels or hydroperiod may
have occurred locally between 2600 and 2500 yr B. P.
as indicated by organic sediments at Ever6 and A23.
Deposition of organics here when widespread marl de-
position occurred over the rest of the Everglades in-
dicates the mosaic landscape form persisting even dur-
ing times of short hydroperiod.
Craighead Pond, A06, and Lignumvitae show sed-
iment beginning to accumulate over bedrock between
2290 and 2100 yr B. P. indicating a rise in water level
or increase in hydroperiod. Gleason and Stone (1994)
provide a radiocarbon age of 2036/2082 yr B.P. on
peat in Core 25 and hypothesize wetter conditions dur-
ing this period. A transition to peat deposition at AO6
about 1960 yr B.P. and to marl mixed with organic
material (charcoal-stained marl) at A23 at 1890 yr
B. P., as well as a peat layer deposited prior to 1790
yr B.P. at L67C, suggest water levels remained high
for several hundred years.
Marl deposition resumed at Rocky Glades at 1770
yr B. P. and at A23 at 1890 yr B. P. This drier period
may have lasted until c. 1400 yr B. P. when organic
deposition resumed at A23. Peat is also present prior
to 1500 yr B. P. at Craighead Pond.
Conditions were probably dry or variable between
1400 and 1000 yr B. P. as some sites show marl and
others peat. Again the mosaic of contemporaneous
wetter and drier habitats is demonstrated in the sedi-
ment record. Between 1040 and 900 yr B. P. longer
hydroperiods resumed and peat formation commenced
at Craighead Pond and Rocky Glades. Hydroperiods
remained long until about 550 yr B. P. as indicated by
peat and organic sediment deposition at C111 and Lig-
numvitae.
Hydroperiods were again variable between 550 and
210 yr B. P. Paurotis Hammock and Lignumvitae re-
veal peat formation, while the C111 canal site exhibits
a change to marl. The uppermost portions of the ma-
jority of cores have marl or algal deposits indicating
shortened hydroperiods. These sediments are dated as
modern at L67C, L67A, and AO06 or younger than 210
yr B. P. at Lignumvitae Key.
SEDIMENT CHEMISTRY
LOI and HLOI Analysis
Text-figure 3 presents results from the LOI (% or-
ganic matter) and HLOI (% carbonate) analyses for
selected sites. The values given in the graphs are the
direct measurements due to the LOI and HLOI ana-
lytical procedures (Dean, 1974). The amount of or-
ganic matter in the Everglades sediments varies from
a mean close to 85% in Gator Lake (a value that re-
mains quite stable throughout the 7 meter sediment
column) to less than 4% in Ever6. The percent car-
bonate also varies between 40% (at some time in the
development of most sites) and less than 1% (at Gator
Lake). At most sites the percent carbonate is greatest
(approaching 40%) near the basal section of the cores,
nearest to the limestone bedrock. 6'°C values of the
carbonate (inorganic carbon) component of the sedi-
ment (see Table 1, Rocky Glades, 195-200 cm and
Lignum Vitae 2, 75—79.5 cm) are —0.23 and +0.62,
respectively, close to 0 (marine HCO, ), demonstrat-
ing the importance of the marine Miami and Key Lar-
go limestones as the carbon source for much of the
sediment deposition at these sites. Everglades peat,
marly peat, and gyttja (lake sediments) are defined in
our study terminology as sediments with more than
30% organic matter by LOI, while the Everglades
marls in our cores contain a maximum of 30 to 40%
carbonate by HLOI. Alternations between peat and
marl are evident at 17 of the radiocarbon-dated cores
including those from C-111, A06, A23, LV-2, Rocky
Glades sinkhole site, L67C, Craighead Pond, and Pau-
rotis Hammock. Ever6 and 9-Mile pond sites (Text-
fig. 3) have elevated organic matter in the upper 15
cm, while WCA-3B (Text-fig. 3) has close to 90% or-
ganic matter throughout most of the core. Higher %
HLOI values in the top meter of the Gator Lake core
(Text-fig. 3), suggest increased atmospheric deposition
of dust and minerals from 20th Century land-use
changes in the region.
Charcoal Analysis
Fire burns off peat deposits locally and lowers the
ground surface which allows ponds to form (Loveless,
1959a; Gunderson and Loftus, 1993) thus contributing
to the mosaic nature of the landscape and causing hi-
atuses in the sediment record. A dramatic recent ex-
ample of this destructive process occurred in spring of
1999 when wild and human-set accidental fires burned
more than 20,000 ha of Everglades (Bragg, 1999). It
may not be possible to find a complete paleohistorical
64 BULLETIN 361
Craighead Pond Craighead Pond
Hiller P-Core
% Loss of Weight % Loss of Weight
40 60 20 40 60
—— +
7.5-8.5
17.8-18.8
Depth
(cm.)
22.8-23.8
28.9-29.9
33.9-35
Paurotis Hammock Paurotis Hammock
Hiller P-Core
% Loss of Weight % Loss of Weight
20 40 60 20 40 60 80 100
+ + + +— f____+- _____}
8.2-9.2
10.9-11.9
13.9-14.1
87-89
17.1-18.4
89-100
100-103 20.8-22.0
Everglade 6 Castellow Hammock
% Loss of Weight % Loss of Weight
40 60 80 40 60
3-4
15.9-16.9
24.5-25.5
35.4-36.3
44.3-45.4
53.4-54.4
57.7-58.8
Carbonate
Text-figure 3.—Percent loss-on-ignition (LOI) and percent high temperature loss-on-ignition (HLOI) analyses for selected Everglades sedi-
ment cores. LOI is a measure of organic carbon and HLOI measures sediment carbonate.
HYDROLOGY, VEGETATION, AND CLIMATE: WINKLER ET AL.
Gator Lake Gator L. P-Core 1
% Loss of Weight % Loss of Weight
20 40 20 40 60 80
51.2-52.2
61.2-62.2
73.5-74.7
86.5-87.6
A23 Hiller AO6 Hiller
% Loss of Weight
% Loss of Weight
20 40 60 80 100
40 60 80
+ +
C111 Canal C111 Canal
Hiller-4 P-Core 1
% Loss of Weight % Loss of Weight
40 60 40 60 80
0-5 + + +
15-19 2.0-3.0
20-25 4.0-5.0
6.2-7.5
149-154
85-95
154-159
10.6-11.7
ISAS 13.2-14.4
164-169 15.4-16.4
169-174 17.6-18.6
Organic Carbon Carbonate
Text-figure 3.—Continued.
66
Rocky Glades Hiller
175-180
195-200
232-237
247-252
270-275
280-285
290-295
300-310
0-8
25-33
50-58
75-83
100-108
116-125
150-156
167-171
175-187
198-200
0
% Loss of Weight
20 40 60 80
BULLETIN 361
Drives 2-4
L67A
% Loss of Weight
20 40 60 80
+
WCA 3B
% Loss of Weight
40 60 80
Organic Carbon
10.2-15.2
30.5-35.6
50.8-55.9
71.1-76.2
91.4-96.5
111.8-116.8
0-1.5
3.9-4.4
5.5-6.5
88-98
12.0-13.3
14.3-15.5
16.8-18.0
20.8-21.8
22.8-23.8
25.3-26.5
Text-figure 3.—Continued.
Rocky Glades FF
% Loss of Weight
20 40 60
80
L67C P-Core 2
% Loss of Weight
20 40 60
80
+
+
Lignum Vitae Key
% Loss of Weight
20 40 60
Carbonate
80
Depth (cm)
12-25
11.7-12.7
20.8-22.0
31.3-32.3
410-42 0
51.2-52.2
61.2-62 2
71.3-72.3
167-168
267-268
367-368
467-468
567-568
667-668
Text-figure 5.—Percent charcoal in Gator Lake sediments.
°
HYDROLOGY, VEGETATION, AND CLIMATE: WINKLER ET AL. 67
Charcoal as % organic weight
0 10
50
| 1
0-7
7-145
14-207
20-257
25-327
32-397
39-4575
45-507
50-577
57-60.5
c0 5-044
64-70
70-75
Depth cm
Charcoal as % dry weight
Depth cm
A23
Charcoal as % organic weight
0 20 40 60 80 100
> 1380 +/- 70 YBP
1660 +/-100 YBP
1890 +/- 70 YBP
2600 +/- 70 YBP
2840 +/- 70 YBP
> 2830 +/- 60 YBP
Ld 2840 +/- 70 YBP
Charcoal as % dry weight
—O— Charcoal as % dry weight
Charcoal as % organic weight
Text-figure 4—Charcoal analyses of AO6 and A23 presented as percent dry weight and as percent organic weight.
Gator Lake Charcoal
Percent Charcoal
100
record in any single Everglades peat deposit. Much of
the dark organic matter in some cores is charcoal
(A23, Text-fig. 4) and even in peat sections which ap-
pear to be continuous, hiatuses occur. These findings
dramatically underscore one of our conclusions: dating
at close intervals is critical to understanding Ever-
glades development.
Charcoal analysis of contiguous samples from cores
at AO6 and A23 (Text-fig. 4) showed that wildfires
have been an agent of change on the Everglades land-
scape for at least the last few millennia. High charcoal
abundances are noted after about 1600 yr B. P. at both
sites when drier plant communities are evident from
the pollen assemblages. In diatom slides abundant
charcoal was noted throughout 70—45 cm at the A06
site and from 33—12 cm and again in the 9—5 cm sam-
ple interval at A23.
Charcoal analysis of the Gator Lake sediments
shows higher charcoal abundances in the upper 20 cm
(Text-fig. 5). This may indicate an increase in regional
land-clearing fires and/or atmospheric deposition of
particles of soot from industrial growth in the region
since the 1940’s. Fires used to burn off sugar cane
growing areas in the Everglades Agricultural Area to
the north of Gator Lake, may be one of the sources of
increased charcoal in the recent sediments as well as
a source of increased mercury (see Sediment Chem-
68 BULLETIN 361
istry below). Fires due to drought, however, have been
documented in WCA-3, the region surrounding Gator
Lake, as recently as 1971 (Goodrick, 1984) and prob-
ably again in 1999 (Bragg, 1999). The charcoal evi-
dence helps explain the presence of willow scrub sur-
rounding Gator Lake as willow is known to colonize
burned-over peat (Hofstetter, 1974; Olmstead and Loo-
pe, 1984). The pollen analysis of Gator Lake sedi-
ments (see below) suggests that willow was a recent
addition to the vegetation near to Gator Lake.
Mineral and Heavy Metal Analysis
Total minerals and heavy metals in subsamples from
Rocky Glades, C111, A06, A23, LV2, and Gator Lake
were analyzed by ICP-OES (Table 2; Text-figs. 6 and
7). The results of these studies will be considered in-
dividually below. When contrasting prehistoric and re-
cent concentrations of most elements (Table 2b) we
found that lead, zinc, copper, and arsenic increase in
the surface sediments (upper I1—2 cm) at most sites.
Even at the Lignumvitae LV2 site where past and pre-
sent saltwater intrusion has confounded element
changes, surface sediments are highly enriched with
lead. Most lead is deposited to these remote places
through the atmosphere or, in the Everglades, through
waterflow as roads and other concrete highway struc-
tures are washed by heavy summer rains and pollutants
drain into surface waters. Lead in airboat fuels could
also contribute to the metals pollution.
Mercury is elevated in both fish scales and sedi-
ments from recent levels of the Gator Lake core (Text-
fig. 8). Mercury is used in the growing and processing
of sugar cane in the Everglades Agricultural Area in
the northern Everglades. Mercury has become a health
problem for wildlife and humans in ENP (Loftus and
Bass, 1992). The mineral and metals analyses show
that even in the same drainage, sites such as AO6 and
A23 (Table 2), can have different allochthonous inputs.
Rocky Glades
Total mineral (Text-fig. 6a) and heavy metal (Text-
fig. 6b) analyses of top 128 cm of sediments from
Rocky Glades provides evidence of elevated heavy
metal concentrations above 45 cm. Samples analyzed
below 110 cm (dated at about 1000 yr B. P.) have low-
est heavy metals concentrations. Additional dates are
needed to identify the cultural horizon in this core. The
total mineral analysis (Text-fig. 6a) indicates peak ley-
els of all minerals except calcium at 95 cm depth in
the core, almost directly overlying the 1000 B. P. date.
High sodium at this point suggests that saltwater in-
trusion may have been the cause of elevated mineral
concentrations. A similar, but smaller, peak occurs at
45 cm in the core, but is accompanied at this higher
interval by elevated heavy metals. The surface sedi-
ment has high concentrations of copper, zinc, manga-
nese, arsenic, lead, and selenium (Text-fig. 6b). Cal-
cium (Text-fig. 6a) is elevated in the bottom marl sed-
iments of the frozen core (Text-fig. 2) and in the top
surface sediment where LOI and HLOI analyses (Text-
fig. 3) document marl deposition (decreased organic
carbon and increased carbonate).
Clll-core 4H
ICP analyses of the top and bottom sediments of a
179 cm core from C111 (core 4H) (Text-figs. 7a and
7b) suggest that saltwater intrusion and agricultural
runoff both affect the sedimentation in this region of
the Everglades. A near basal date for C111 core 4H at
170 cm is about 3000 yr B. P. However, in these bot-
tom sediments there are elevated metal and mineral
levels which suggest that there may have been a salt-
water intrusion, an hiatus in the peat and concentration
of elements, or diagenesis and mixing at this site. Ca-
nal construction may have also affected the stratigra-
phy at this site although we moved away from the
canal road and berm before coring. Calcitum and man-
ganese peak in the marl sediments at 15—19 cm. depth
and lead has near peak values in the top sediments
(Text-fig. 7b). Boron, phosphorus, and copper, miner-
als used in fertilizers (Hem, 1992) have high concen-
trations in the top sediments. Sodium is elevated in the
top sediments also suggesting possible saltwater intru-
sion in this area.
Gator Lake
ICP total minerals were analyzed for two sediment
samples, 3.6 and 681 cm, from Gator Lake (Table 2a).
The 3.6 cm sample is from the modern sediments and
the 680—681 cm sample has been AMS radiocarbon-
dated at 1950 + 45 yr B. P. Similar to the results from
C111 core 4H, boron and copper are elevated in the
surface sediments of Gator Lake. These values suggest
enrichment from agricultural practices in the region.
A06, A23, and LV2
Surface and core bottom sediments were analyzed
at AO6 and A23 sites for both total minerals and trace
metals (Tables 2a and 2b). Percent N (% total nitrogen)
was also analyzed for the top and bottom samples of
the AO6 and A23 cores (Table 2a). Only trace metals
were analyzed for LV2 (Table 2b). As at some of the
other Everglades sites, phosphorus, boron, and copper,
as well as nitrogen, are elevated in modern sediments
at AO6 suggesting agricultural enrichment, while cal-
cium and manganese are high in the modern marls of
69
HYDROLOGY, VEGETATION, AND CLIMATE: WINKLER ET AL.
rol THT Pr co6 cle c6'b rO'SL 687 [98S 97'S Bes rS'l
CSI 67 EO Lb 606 66'T Sb €08 66¢ 7869 9801 Sis orl
CSI 67 LO'EP 60'6 66°C ais 9S TL re S6TS LL LL orl
9'LT Lr Lr Vol zeg pr OLLE ZG raids) 60'°9 Ip 97
OSI €t €0S 60'6 10°€ Ub PILI SCAG 66LL 716 90'€ Lr'l
TST TLE OE bE Lv1 98'P LLY GOS +8 +06L LET cL9 SET
Sl Ltt SO'9S 86'8 (bn GAG 80'0L SE 6LSL 16L L8'6 €
C8 SILC 8'8L 801 LOE £89 LO IL GEG Ssh LL9 GUE SLI
ssl €€T IP LZ Sv6 Soe tre 68S 8c LOS 19b crt 86r' I
VLI kSG 10°67 86'6 6c ere 16°61 vT'T €90b S91 86r'l 19°
r6S1 SL 07 LIS ces 769 9¢°9 L88b 98'S C8bT IL9€ 68LI or
786 = LOE OLte 86'E1 te) rs'S 6TLS 669 OISZ EL'LI 90'r1 GHG
ETI sl 69'9 6'L PTE t0'r IL8E 88 GLSL b6'S ef Ol oll
6LI € 97 61ST 6E 17 pre $9'T GSH VAS prSr 6£ 9¢'1 69'1
676 ~—sSPL'SI LUS Lys SLY arr £069 €LTI 6£69I ZO" 7 SI IS]
IL6 ~=r681 £901 19°€] 6c br 6'| Ga ler €18L 6T€ SO'r IW
2S sv uz qd IN ow uN a od nD ID oD
el 97E 6969 b6'b SPT I ST€8 as) 6$9 OLLI res SOLSOI £981
I TOEE €60C 6€'b 9669 Corl LOOT SOT = POI Os6l Proce! 7 COI
61 €9Ib 18S SOL ITE ST IE aa 8E'8 Z6IE ITEl TELE cial
Ve T EBL ISLE ~~ LOSE Ip8r $S'9T Or 69ST 9099 96£1 60007 7691
— € 167 Lr9E —-€9°8 psor 6895 F8'tl ¢St LEIP cEeeo £1691 8887
= Itl OL8I 79'8I 800r L061 TIE 90E O£6S L908 OLIFI 9881
N UN IV nD ad uN d uz Ss 3IN Le) M
wm Ap uidd
980
c80
c8'0
Saale
c8'0
cell
c8'0
86°0
£80
60
6c €
61
ro 0
99°T
189 18-08 “L “"PoOwD CT
cor T8-O08 “¢ "POD +1
C8e 16-06 “‘F “POD cI
067 06-S8 “€ “POD Tl
O81 18-08 CG “PHOrD II
QOL | a100-d = quoiey OI
less | at0o-d = quojyey 6
Taec | at0o-d yioyweyH g
9¢ ] o100-d qaowey 1
96 GeO tz CAT 9
6 6-0 ‘I TAT S
os ¢¢-0¢ ‘tH 90V
L L-0 ‘IH 90V £
IZ 1@-TI ‘TH €tW T
@ Z as09-d €TtV I
siskjeuy qOl—SIPIW 21,
IZ IZ-TI ‘1H ETV ST
I | a109-d €TW 9T
os $7-0T ‘TH 90V FT
L L-0 ‘IH 90V €T
189 18-08 “ZL “HOD gI
9¢ | aoo-d qiowey LI
yidaq aA aus
(wo)
siskjeuy qO[—S]|P1ouly [PIOL
qm Aap wid ut uaatd aie suonesUadUOD “Wd [89g 0} WHO g'¢ WO’ Saydues aye] 10IVH J0J puv ‘TAT ‘9OV “CTY Wor sajdues Woo pur do} J0OF stIsAjeUY s[PIe 9PIL SHO
-dOl QM Aap (uoypru tad syed) widd ur uaars are suoneussuo0d “¢7Ty Pure ‘QOY ‘Aye Joy wos saydwes woyog pue do} 10} sisAjeuy s[esoulP [PIOL SHO-dOl “RP —T AIGRL
70
Y GLADES, Everglades National Park
ICP Total Mineral Analyses, 1993 core
& cs
(e
Ss s
Z S
& Se
BULLETIN 361
SS,
ty
S
Depth (cm)
2370 + 30] 2
—SeasS
——
1040 + 30] 4 |
1
}
TVTTTT [TT |
7
yj
1
r = =
500 1000 1500 100 200
4000 6000 500 1000 1500’ 200 400
w
So
S
nm
2000
Depth (cm)
754
=|
804
854
90-4
954
100 4
1054
1040 + 30]
105
1154
TITTLE
T
+ 1 T + 1
100 200 300 50 100 150 200
400 600 200 400 600 800° 500 1000 ~=1500
ppm
Text-figure 6.—a and b. ICP-OES Total Mineral Analyses of the top 130 cm of Rocky Glades FF core (a). ICP-OES Heavy Metal Analyses
of the top 130 cm of Rocky Glades (ENP) sediments from same core (b). Minerals and metals concentrations are given in ppm.
A23 p-core 1. Lead, zinc, and arsenic are elevated in
the surface sediments of A23 p-core 2 (Table 2b),
while copper, lead, zinc, arsenic, and selenium are high
in the surface sediments of A06. LV2 has elevated
lead, lithium, and manganese in the surface sediments.
Abundant marine diatoms in the LV2 surface sedi-
ments indicate frequent saltwater inundation of this
mangrove-fringed coastal pond.
71
AL.
’, VEGETATION, AND CLIMATE: WINKLER ET
i
HyYDROLOC
National Park
Fyerglades
vergiad
f
TPT
rz
r
pil l = oe |
pea eal |
Pa |
ai | eee
ET es | od pop eae
ss es | ey ee ee I) ere
at at Sel
moonmnownwnoaonc;9qonononnonrwnaone79eru owownmnnsao
-NN MOM AD TST MNONMWOWOMm~Mm™~ DW DWADWOCC — NN 32)
&
i
|
}
a ra lela L i
| | ae mE Eta
(ise | in E
=,
1000 1500 2000' 50 100 1
Hi,
500
|
|
= rosa US) a ea Fe es Fo fd a
MOMONONOHNONONHNON OH OHNOHOHNONS
SFRPANA*TTDN DORK DAARADAGO-—--—-ANM
Text-figure 6.—Continued.
72 BULLETIN 361
——- = — — —
L =
m)
Depth («
—— |
ak _———_
= ; = : ; ——— ; :
00 1000 1500 100 200 300° 1000 2000 3000 100 200 300° 20
& d ppm
és os <
Ae) y
S 7 LO): R & oe
i= — eS
= IK ————d |
== => , |
| |
|
e
5
2s
a
S
(=)
SSS oa
3010 + 508
= __ :
1 tenaq ht T T Te 1 a “al Tose "ont T 7 T T Tr il
40 60 100 200 300 400 20° 100 200 300 400 500 100 200 300 400
ppm
Text figure 7—a and b. ICP-OES Total Mineral (a) and Heavy Metal Analyses (b) of the top 25 cm and the bottom 25 cm (155-180 cm)
from sediment cores from C-111, ENP. Minerals and metals concentrations are given in ppm.
3010 + 509
3010 + 50§
C11
ICP
Depth (cm)
Depth (cm)
Heavy Metals Analyses, 1993 core 4H
ee)
So
nn & OO
BDNONGHGS
an
HYDROLOGY, VEGETATION, AND CLIMATE: WINKLER ET AL.
Everglades National Park
Text-figure 7.—Continued.
73
ppm
74 BULLETIN 361
Percent ug/g Hg dry weight
io} 0.S
0 100} 1
Age
Model
Stratigraphy
Core Description
4 ‘t =
nN
o
+- before European settlement
E
Ss
° +
230 30 |
8
c
£404, 40+
a
fa
sot so +
60 | 60
70 44 70
8044 * 80 +
Hy
g x
so} 3 8 90+
om BS
2
100-4 +99+—
Homogeneous soupy
$4 and fibrous olive-
@ brown peat.
Flocculent sediment
0-6.6 cm becoming
slightly more
fibrous and
consolidated
Es
He
Be
g
2
#
4g with depth.
Text-figure 8.—Mercury concentrations in Gator Lake sediments (p-core 1) and in fish scales found at the same depth in the sediments
(modified from Jensen, 1994).
Sediment Stable Carbon Isotope Analysis
Charcoal isolated from the sediments of the A23H
core was used for stable carbon isotope analyses (Table
3). The charcoal is considered to represent air- or wa-
ter-borne upland carbon at most sites (Winkler, 1994b),
but in the Everglades, charcoal may also come from
the direct burning of peat at the site during drought.
Stable carbon isotope values (6'C) higher than —23
signal a proportional contribution of carbon from C-4
plants (such as Muhlenbergia and other drought-tol-
erant plants) in the region. The highest values in the
A23H sequence (6'°C —21.04) are found from 9-12
cm when marl was being deposited at the site (Text-
fig. 2). This result suggests an expansion of arid C-4
vegetation in the region, perhaps Muhlenbergia grass
prairies, before about 1890 + 70 yr B. P. and indicates
that this vegetation association (which contains com-
paratively little biomass) burned frequently and con-
tributed windborne charcoal to the sediment at A23H.
Other 85'°C results accompany the radiocarbon-dated
samples. These values (Table 1; Text-fig. 2) are from
analysis of whole sediment organic carbon which in-
cludes carbon from both upland and wetland sources.
Table 3.—6'°C-Stable carbon isotope values for six A23H sedi-
ment samples. Analysis was performed on charcoal fraction isolated
from the sediments at each depth.
Depth (cm) GAS
5-9 —23.48
9-12 —21.04
12-21 —24.01
21-25 Dep,
25-33 252i
41-50 —23.00
Highest values on organic carbon (values above 6°C
—21) are found at the bottom of the WCA 3B core
(4990 yr B. P.) influenced by the underlying marls, or
in the Rocky Glades sediments before 4800 yr B. P.
and from about 2500 to 1900 yr B.P., times when
other sediment analyses also indicate drought.
In the two cases where separate samples of inorgan-
ic carbon and organic carbon were analyzed (Rocky
Glades 195—200 cm and Lignumvitae LV-2, Dr. 4, 75—
79.5 cm), the 6'°C values of inorganic carbon docu-
ment dissolution of underlying marine coral (64°C
—0.23 and +0.62, respectively, close to marine lime-
stone 85°C of 0.0), rather than modern atmospheric
CO, (68°C —7.0), as the carbon source for marl-form-
ing organisms. This is an interesting result in light of
recent increases in atmospheric CO, and the need to
identify carbon sources and sinks in the global carbon
cycle. Additional stable carbon isotope sequences
would aid interpretation of vegetation community
change data from pollen.
BIOLOGICAL INDICATORS
Diatom Analysis
Diatom associations integrate environmental factors
such as water chemistry, water temperature, water
depth, and aquatic macrophyte presence or absence by
their speciation, distribution, and abundance. Accu-
mulation of diatom frustules in lake or peat sediments
represents the response of the biota at various time
scales. Seasonal and microhabitat variations are
smoothed at medium frequency sampling intervals.
Diatoms respond relatively rapidly to environmental
changes such as pH and salinity gradients, and to
changes in nutrient availability, water levels, and tem-
HYDROLOGY, VEGETATION, AND CLIMATE: WINKLER ET AL. 15
perature (Hustedt, 1939; Patrick and Reimer, 1966;
Bradbury, 1975; Patrick, 1977; Gasse, 1980; Diger-
feldt et al., 1980; Rasanen and Tolonen, 1983; Charles,
1985; Charles et al., 1986; Smol et al., 1986; Winkler,
1988; Whitmore, 1989; Sweets et al., 1989; Fritz,
1990; Sweets er al., 1990; Blinn, 1993; Dixit er al.,
1993). Therefore, diatoms can be used to reconstruct
the environmental history of lakes and wetlands and
to delineate the biotic response to modern trophic and
chemical conditions caused by land use and other
changes. Correlation of chemical variables, morpho-
metric factors, disturbance indicators, and diatom per-
centages, enables separation of some of the factors af-
fecting the diatom distribution within lakes (Winkler,
1988).
Diatoms were identified using floras compiled by
Hustedt (1930, 1939, 1955), Huber-Pestalozzi (1942),
Patrick and Reimer (1966, 1975), Pergallo and Per-
gallo (1965), Koppen (1975, 1978), Florin (1980), Fo-
ged (1980, 1984), Gasse (1980), Germain (1981), Hak-
ansson and Locker (1981), Gasse et al., (1983), Gasse
and Tekaia (1983); Haworth (1983), Bateman and
Rushforth (1984), Camburn et al. (1984—1986), Hak-
ansson and Stoermer (1984), Stoermer and Hakansson
(1984), Krammer and Lange-Bertalot (1985, 1986,
1988, 1991a,b), Camburn and Kingston (1986), Hak-
ansson (1986a,b), Dodd (1987), Klee and Schmidt
(1987), Kling and Hakansson (1988), and Scherer
(1988).
Diatoms have been analyzed at four sites, Gator
Lake, A23H, AOQ6, and LV2. Each site will be dis-
cussed individually below, while the Diatom Discus-
sion section will include consideration of results from
all of the sites.
Gator Lake
Diatom analysis of the Gator Lake sediments (Text-
fig. 9) reveals two zones devoid of diatoms: before
1900 yr B. P. (before 650 cm) and after about 1600 yr
B. P. (above 450 cm). The interval between 1900 and
1600 yr B.P. has abundant acid to circumneutral
planktonic diatoms dominated by Aulacoseira ambi-
gua and other Aulacoseira species, including a large
number of diatoms which may be a new species of
Aulacoseira (Plate 1, figs. 1 and 2). This zone of deep-
er water, as indicated by the planktonic diatoms, is
concurrent with one described below at sites in the
Northeast Shark River Slough and may indicate a
change in the regional hydrology at that time. A di-
verse assemblage of littoral diatoms including Eunotia,
Navicula, and Neidium, and other algae (Pediastrum
and desmids) are also present during this wet time be-
tween 650 and 450 cm.
At Gator Lake, the intervals with sparse diatoms are
the only ones with abundant sponge spicules (Pl. 1,3
and 4), as well as abundant rhizopod amoeba tests,
chrysophyte cysts, Nymphaceae sclereids (Pl. 1,6),
fungal spores, and pyrite framboids. The few diatoms
in the surface sediment sample suggest an increase in
taxa such as Aulacoseira granulata and varieties, and
Rhizosolenia resting spores. These species dwell in
more alkaline and more nutrient-rich water than dia-
tom species found in the lower levels of the core.
A23H
The diatom assemblage at the A23H site (Text-fig.
10) in Northeast Shark River Slough indicates that al-
kaline littoral (shallow water), epiphytic (plant cling-
ing) species such as Mastogloia smithii v lacustris, An-
omoneis vitrea, Cymbella microcephala have domi-
nated since 2840 + 70 yr B. P. However, for the period
after 2600 (?2000) to 1660 yr B.P, circumneutral
planktonic diatoms (Cyclotella, Aulacoseira ambigua,
Tabellaria flocculosa v linearis) appeared and made
up more than 10% of the diatom sum. These plank-
tonic species indicate deeper water at that time. The
chronology of this deepwater event is concurrent with
the interval of abundant planktonic diatoms at Gator
Lake dated from 1900 to 1600 yr B. P., suggesting hy-
drologic changes across the southern Everglades. The
inverse relationship of sponge spicule/diatom presence
or absence found in Gator Lake sediments, does not
hold true at A23H site. Sponge spicules at this site are
present only sporadically in intervals when diatoms are
abundant. Diatoms are very sparse in the marly peat
below 2600 yr B. P.
A06
Diatoms are relatively sparse throughout the A06
core and, although entire slides were examined, the
highest diatom sum for any level at this site was 156
frustules. The diatom assemblage at the A06 site (Text-
fig. 11) in Northeast Shark Slough is dominated main-
ly by ephiphytic alkaliphilic/alkalibiont diatoms from
the genera Mastogloia, Amphora, Cymbella, and Na-
vicula. However, the basal levels of the core contain
epiphytic Gomphonema spp. and evidence of alkali-
philic planktonic species including Asterionella, Ste-
pPhanodiscus, Synedra, and benthic Surirella spp. sug-
gesting slightly deeper water after about 2100 yr B. P.
Planktonic/benthic diatoms (small Fragilaria spp.) are
present above 55 cm in the core, but are the dominant
taxa only when other diatoms are scarce. At A06
sponge spicules and chrysophyte cysts are most abun-
dant in the top 30 cm of the sediment, the time when
diatoms are also most numerous. The number of lit-
toral diatoms increases in the AO06 top sediments with
much of this increase due to circumneutral Eunotia
76 BULLETIN 361
Ls n
-—f + ie 9 a
| | Joell Lae
| i} | |
|
| | | par
= | | |
Ey |
|
|
ae : = 2000 3 at
TOM PERCENTAGE
~ \
\
\ |
| 440 | 44 j
| | 7 |
‘Aaa
/ | Abundant
=n 40
A
| /
| / | |
/ Test |
| 13 5 + t4 | Sparse di l= 1950 + 45
2 = 2040 +8
DIATOM PERCENTAGE
Text-figure 9.—Gator Lake Diatom Percentage Diagram. Solid curves represent actual percentages. Stippled marking represents 10 ex-
aggeration. Radiocarbon dates are placed to the right of the diagram.
HYDROLOGY, VEGETATION, AND CLIMATE: WINKLER ET AL. WI
site { Je
vi f
4
t
|
| |-16
10 |
4
|
Sao cellll
rs Jy]
354
i}
|
455 |
zalll| ya 840 +
t 70 (
TOM PERCENTAGE
17 5 36 A
mall | = ;
| 97 | 24 1-1660 + 100
104 | |
| Nar = |
| 4 | | 30 +
| | |
1
a | A
ee is Lyell A
= | }
S | | 4
= \ | Ve g
= | f B40 +
A (ie ep
a | | |
304 | 4
| 4
| ) , 4
| i 4
| t 4
404 | | | 4
| | y Y |
| q |
Wl | | m7 |
45-4 | Y Z
| | | | | L
| | | | Yy YY
50J } =I pamel man [ [> Fo T - 1 LOTUS:
20 eC 20 50 100 = 15
DIATOM PERCENTAGES
Text-figure 10.—A23H Diatom Percentage Diagram. Solid curves represent actual percentages. Radiocarbon dates are placed to the right
of the diagram. Oldest sediment is at the bottom of the column.
78 BULLETIN 361
oy
Perot \
I
| a a
maf
0) 20 19
DIATOM PERCENTAGES
Text-figure 11.—A06 Diatom Percentage Diagram. Solid curves represent actual percentages. Radiocarbon dates are placed to the left of
the diagram. Oldest sediment is at the bottom of the column.
HYDROLOGY, VEGETATION, AND CLIMATE: WINKLER ET AL. 79
LIGNUM VITAE KEY, Florida
LV-2 Hiller Core
Diatom Percentages
M. Winkler, Analyst
s oe.
© 2 e~
s RS Ss ee : Ss ey RS RS
& Ny S < XO
SSS sor” Sor Fe 1s SF Se
SE SHS LS wf Qo TRS PV Mick of
10 on Pasties é [a fae eerie 143 18 34 0
210k | Sa SSeS 10 6 19 4
154 i
20
250 + 30] |
25
34
354
404
454
€ 50
600 + | = 55
3s a anP a ee ee) SCAR AA Oh 3 |e
65
ul
75
1040 + ssf
a
85
904
I 95
2290 + 50
100 0 i 0;9
Uns 20 20 40 In ae 20 20 40 eyiceti= 20 40 20 20 40 60 80 100!"
DIATOM PERCENTAGES
Text figure 12.—Lignum Vitae Key, Florida, LV2 Diatom Percentage Diagram. Radiocarbon dates are placed to the left of the diagram.
Oldest sediment is at the bottom of the column.
and Pinnularia species, as well as more alkaline Am-
phora, Anomoneis, Nitzschia, Cocconeis, and Diplo-
neis spp. A study by Wood and Maynard (1974) may
illuminate reasons for the unexpected presence of Eu-
notia and Pinnularia diatoms at this site. These inves-
tigators analyzed algae trapped in air nets placed on
the 7-mile fire tower in ENP. They found that aside
from marine algae which had traveled more than 64
km from the sea, there were several species of acid
and circumneutral Eunotia and Pinnularia diatoms in
the traps which had probably been airborne from
northern and central Florida sites. Eunotia and Pin-
nularia are frequently found growing in lakes under-
lain by acidic sands in northcentral Florida (Sweets er
al., 1989; Whitmore, 1989; Sweets et al., 1990). Other
aerophilic diatoms found in the air nets included Mas-
togloia, Amphora, Nitzschia, and Hantzschia species,
and Asterionella formosa (Wood and Maynard, 1974).
They also analyzed ENP surface water foams which
contained diatoms and theorized that these foams were
easily picked up, shredded, and transported by the
wind to colonize or recolonize distant sites. The pres-
ence of Hurricane Andrew debris at several Everglades
sites during our fieldwork in February 1993 (Winkler,
personal observation) illustrates the importance of
wind as a factor for distribution of diatoms and other
organisms in storm-swept southern Florida.
Lignumvitae Key (LV2)
At Lignumvitae Key site LV2 (Text-fig. 12), dia-
toms are relatively abundant only in the surface sedi-
ments. Marine diatoms, such as marine Amphora spp.
and Navicula spp., Pleurosigma, and other saline di-
atoms (Triceratium, Thalassiosira, and Paralia sulcata
genuina f. radiata) comprise the majority of species
above 15 cm in the core. Many of the freshwater spe-
cies, such as Amphora cf acutiuscula, Cyclotella kuet-
zingiana, and Navicula cf salinarum are also found in
brackish water. Sponge spicules and chrysophyte cyts
are abundant only in the top 70 cm of the core. Pyrite
framboids, indicative of reducing conditions, are fre-
quent in the basal sediments of LV2.
Diatom Discussion
The presence of diatoms in the Everglades and the
Keys sediments has been either feast or famine. In
much of the sediment, diatoms are absent or very
sparse. However at times, and at several different sites
80 BULLETIN 361
(some more than 40 km apart), there are abundant
planktonic diatoms indicating high water levels in the
southern Everglades. These findings are synchronous
and indicate high water between about 2000 and 1600
yr B.P. at site A23 (Text-fig. 10), around 1900 and
possibly 1600 yr B. P. at site AO6 (Text-fig. 11), and
between 1900 and possibly 1600 yr B. P. at Gator Lake
(Text-fig. 9). The high water levels in this time period
may be concurrent with a local sea-level highstand in
Florida Bay (H. Wanless, personal communication,
1998). High water in ENP during this time interval
also correlates well with interpretations of high water
in northern Guatemala from similar findings of abun-
dant planktonic diatoms in otherwise diatom-barren
sediment cores from Laguna de Petenxil in northern
Guatemala (Patrick, 1966). This extra-regional hydro-
logic similarity is not surprising. In an Americas lake-
level database which includes late-glacial and Holo-
cene sites, there are similarities in water level changes
in lakes within the Caribbean latitudes throughout their
history (Winkler er al., 1996b). The Everglades and
northern Guatemala are within this latitudinal range.
Aside from the diatom-rich period between 1900 and
1600 yr B.P., there are very sparse diatoms before
about 1900 yr B. P. at all Everglades sites examined.
After 1600 yr B. P. diatoms are either rare at some sites
such as Gator Lake, or are abundant shallow water
taxa that grow in high nutrient conditions, among
heavy macrophyte cover, and/or with heavy algal
blooms.
Reasons for the scarcity of diatoms in some wetland
sediments might include competition for a limited sil-
ica resource. This is unlikely in the Everglades. At
only one site (Gator Lake) do diatoms and other sili-
ceous objects (chrysophyte cysts and sponge spicules)
have an inverse relationship. At all other sites exam-
ined, these siliceous remains are abundant at the same
time and disappear together at other times. Dissolution
of diatoms and other silica objects may be due to
changes in water chemistry. Strong brines, especially
Na,CO, and NaCl, and salts of other single valent cat-
ions, dissolve silica rapidly. In an experiment by Bark-
er et al. (1994), diatoms dissolved in these brines in a
matter of days. High pH accelerates diatom dissolu-
tion. In fens that receive very alkaline (pH >9) ground
or spring water, diatoms are present in the top sedi-
ments, but only highly silicified remains are found a
few centimeters downcore (Winkler, unpublished
data). Dissolution conditions can be very complex.
Some elements, such as iron and aluminum, make
complexes with brines and retard dissolution. Seawater
incursions at some sites in ENP may destroy fresh-
water diatoms but may not provide high enough salin-
ities for a long enough duration for growth of saltwater
taxa. In combination with drought which concentrates
salts found in freshwaters, water level and water com-
position changes in the Everglades environment may
be too harsh for sustained growth of many organisms.
Increased UV-B penetration of surface waters in recent
decades, as documented at other sites, may be causing
changes in Everglades biota. Diatom growth is sup-
pressed by 40% by UV-B in alpine lakes (Leavitt and
Vinebrooke, 1999).
The morphometry of the diatom itself can prevent
or delay dissolution—(Barker, 1992; Barker et
al.,1994) those diatoms with low surface area to vol-
ume ratios remain, while other diatoms dissolve. Tur-
bidity can also cause decreased diatom growth as can
excessive nutrient or metal loading. If increased silts
and clays and/or macrophyte growth such as water lil-
ies shut out light, diatom growth would be reduced.
Dense blue-green algae blooms also shut out light and
may interfere with diatom growth. Wood and Maynard
(1974) and Gleason and Spackman (1974) have mea-
sured blue-green algae mats in excess of 10 cm depth
in surface waters. Very warm temperatures (as high as
36°C, Browder ef al., 1981) may also favor blue-green
algae growth and discourage (or prevent) diatom
growth. In temperate lakes diatom growth is dominant
during cool spring and fall months (Hutchinson, 1975),
giving way to blooms of green and then blue-green
algae during hot summer months. Blue-green algae
dominance in the Everglades surface waters may be
related to their tolerance of high temperatures and high
carbonate environments, and to their more efficient uti-
lization of other abundant nutrients. Indeed, some
blue-green algae precipitate carbonate (Gleason and
Spackman, 1974) and contribute to the formation of
marl which covers large areas of the shallow-water
Everglades today. The periphyton are the basis of the
aquatic food chain upon which most other Everglades
organisms are maintained. Some blue-green algae also
fix nitrates from atmospheric nitrogen, adding to the
already abundant nutrients in Everglades waters which
stimulate excess periphyton growth and, at times, lead
to anoxic conditions at some sites (especially under
low flow conditions). Some blue-greens (e.g., Micro-
cystis) secrete phytotoxins which may exacerbate poor
growing conditions for other forms of plankton.
Agricultural enrichment by nitrogen and phosphorus
in the northern Everglades has been studied by Rader
and Richardson (1992). They found that annual pri-
mary productivity and algal biomass increased at least
3-fold from nutrient enrichment, but algal diversity did
not change (although there were significant species
changes and a surprising shift from cyanobacterial
mats to dominance by filamentous green algae). How-
ever, there was increased diversity of macro-inverte-
HYDROLOGY, VEGETATION, AND CLIMATE: WINKLER ET AL. 81
brates. The increased minerals identified in our surface
samples and interpreted as resulting from agricultural
enrichment may explain the abundance of remains of
biota in the top sediments at some sites and the dearth
of microfossils at depth in some of the cores.
Comparison of Paleoecological Results with Studies
of Modern Periphyton Ecology and Distribution
Everglades periphyton taxonomy, chemistry, and
distribution have been studied by Van Meter (1965),
Gleason and Spackman (1974), Wood and Maynard
(1974), Browder et al. (1981), Swift (1984), and Swift
and Nicholas (1987). Much of their findings are sum-
marized in Browder et al. (1994). The dominant algal
taxa of the important periphyton communities are list-
ed in Table 16.1 in Browder er al. (1994). The periph-
yton communities have been divided by their relation-
ship to water chemistry into |). high mineral (carbon-
ate), high nutrient communities, 2). low carbonate,
high nutrient communities, and 3). low carbonate, low
nutrient communities. The high mineral, high nutrient
communities are very alkaline with pH values in ex-
cess of 9 and are dominated by carbonate-precipitating
filamentous blue-green algae. The low mineral, high
nutrient communities are dominated by filamentous
green algae in early summer, filamentous blue-greens
by late summer and early fall, and in winter by diatoms
characteristic of high nutrient waters. The low mineral,
low nutrient waters had pH values below 6 and con-
tained a diverse association of desmids and filamen-
tous green algae, and diverse (but not dominant) dia-
toms characteristic of oligotrophic water and wetlands.
Looking at the spatial variations of the periphyton
(Wood and Maynard, 1974; Browder er al., 1994) and
specifically, of the diatoms (Whitmore, 1989), it is
clear from our core sediment studies that some of the
same relationships occur on a temporal basis. For ex-
ample, Mastogloia smithii v lacustris, an alkaliphilic
diatom found in mesotrophic to eutrophic surface wa-
ters, is the dominant diatom in the marls found in the
sediment cores (see especially AO6 and A23 diatom
stratigraphies, Text-figs. 11 and 10). These marls are
high carbonate and either high (A06) or low nutrient
(possibly A23) environments. pH changes not directly
obvious from sediment changes identified in the field
or the laboratory may be interpreted by the presence/
absence of Mastogloia species at specific sites. At oth-
er times, however, diatom assemblages unlike those
found in the modern sediments were dominant at the
sites analyzed and indicate deeper water and different
water chemistry from today.
Pollen and Sclereid Analysis
Pollen analytical data from 5 sites, AO6 and A23 in
Shark River Slough, L67A and L67C in WCA3B, and
Lignumvitae Key illustrate the development of Ever-
glades vegetation. In this section pollen taxa will be
divided into 6 groups: Chenopodiaceae-Amarantha-
ceae (cheno-am), pine, slough and marsh taxa, fern
taxa, upland vegetation taxa, and mangrove taxa.
Cheno-am is a pollen morphological group comprised
of two related families with genera that have similar
pollen that is virtually indistinguishable using ordinary
light microscopy. Plants of these families are annual
weeds of disturbed soil or perennial forbs or shrubs of
desert, alkaline, or coastal areas. Based on pore num-
ber and grain size more than one species may be rep-
resented in our samples. Most grains are between 16—
22 wm in diameter and have more than 100 pores,
typical of the annual weed genera. Cheno-am pollen
found in Everglades sediment has not been assigned
to a particular Everglades landscape type because it
may be derived via wind transport from habitats far
removed from the Everglades, although Loveless
(1959b) found that Acnida cuspidata ( = Amaranthus
cannabinus), pigweed, invaded sawgrass marsh during
drought periods. Cheno-am is often the dominant pol-
len type found in Everglades sediments. Pine pollen is
often co- or sub-dominant with cheno-am pollen. It too
is wind transported and may be of extra-local origin,
although the pine forests of the Atlantic Coastal Ridge
of South Florida may be the most likely source. Slough
and marsh taxa include Nymphaea, Sagittaria, Utri-
cularia, Cyperaceae, Cladium, Typha, Potamogeton,
Justicia, Umbelliferae, and Gramineae. These species
represent more local vegetation although grains can be
transported by flowing water from their parent plants’
growth positions. Cyperaceae including Cladium may
be underrepresented in Everglades pollen spectra be-
cause these species propagate vegetatively as well as
by seed and in the case of Cladium only older culms
produce flowering stalks (Steward and Ornes 1975).
Ferns are monolete and trilete spores including Poly-
podiaceae, Anemia sp., and Osmunda sp. Ferns may
be emergent plants of marshes or understory plants on
hammocks. Upland taxa are mainly tree, shrub, vine,
or perennial herb species of hammocks, tree islands,
and willow heads. These include Myrica (wax myrtle),
Quercus (oak), Myrtaceae (stoppers), and Cephalan-
thus (buttonbush). Mangrove taxa are Rhizophora (red
mangrove), Conocarpus (buttonwood), Laguncularia
(white mangrove), and Avicennia (black mangrove).
Sclereids are structures that provide hardness or ri-
gidity to plant tissues. These structures may remain in
sediments after degradation of other plant tissues and
some sclereid forms are identifiable to genus. Both
Cladium and Nymphaceae have recognizable sclereids
which are diagnostic features of sawgrass and waterlily
peats (Cohen and Spackman, 1977). Cladium sclereids
82 BULLETIN 361
A06
Percent
€ au0Z
modern
Depth cm.
2
1960 +/- 30 yrs B. P.
N
2100 +/- 30 yrs B. PS
o
=
L67A
Percent
=~" © © 2 Wan @ © o
OCcenroaon eos 6
0-5.7
5.7-8.2
8.2-10.1
10.1-10.6~
10.6-12.7
12,7-14,2
Depth cm.
14.2-15.4
15.4-15.9
N
°
3
oO
=
A23
Percent
a N
= % © BHwanowwvos °
o o o o o °o o o o o =] Ss
oO
0-2 g >
2-3 Fj 1380 +/- 70 yrs B. P.
3-5 N
°
5-9 1660 +/- 100 yrs B.P. 3
9-12 [1890 +/- 70 yrs B.P. wo
12-21 F] 2600 +/- yrs B. P. RG
21-25 ee] Ml 28404/- 80 yrs BP. &
SS SS i
oO
25-33
np
33-41 N
41-50 P| 2840 +/- 70 yrs B. P. 2
=
b
L67C Core 2
Percent
% Pine
% Fern
HOg@a @
% Upland
% Mangrove
% Chenopodiaceae-Amaranthaceae
% Slough + Marsh
Text-figure 13.—a-d. Pollen diagrams of A06, A23, L67A, and L67C. Pollen is presented as percentages of total pollen divided into
Everglades community groups plus Chenopodiaceae-Amaranthaceae and Pinus. See discussion in text.
are flat net-like masses with distinctive triangular
openings (Plate 1, fig. 5); Nymphaceae sclereids are
irregular stellate structures (Plate 1,fig. 6).
A06
The AO6 pollen sequence is divided into three pol-
len zones (Text-fig. 13a). Zone 1, 75-64 cm, is stati-
graphically marl, more organic at the base and becom-
ing less so at the top of the pollen zone. It has a basal
date of ca. 2100 yrs B. P. Cheno-am dominates at 45—
50%, but pine is a secondary dominant. The slough-
marsh component and the upland component each
comprise ca. 12—15%. Sclereids of both Cladium and
Nymphaceae occurred at very low concentrations
(Text-fig. 14a). Zone 1 represents a low hydroperiod
with a shallow mixed sawgrass-waterlily marsh and a
drier upland area nearby.
Zone 2, 64—25 cm, transgresses stratigraphic zones.
HYDROLOGY, VEGETATION, AND CLIMATE: WINKLER ET AL. 83
A06 A06
Cladium sclereids Nymphaceae sclereids per
per g dry wt 0.6m!i sediment
fF 15000
20000
f 5000
10000
15000
20000
LF 10000
0-7
7-14
14-20
= 20-25 —)
© 25-32 —
= 32-39
a :
ee 39-45
Q 45-50
50-57
57-60.5
60.5-64
64-70
70-75
a.
A23 A23
Cladium sclereids Nymphaeacae sclereids per
per g dry wt 0.6m!i sediment
° ° ° ° °
S S S S
° ° ° ° ° 3 3 3 S
3 S 3 S a S x S
ca & x s
oy rn L 1
f = ; = 0-2
0-2 [scan
2-3 ps: a 2-3
3-54 355
z E 5-9
°o 5-9 rs) =
= 9-124 +fH4 € 9-12
Q 12-21 @ 12-21
21-25 21-25
25-33 25-33
33-41 33-41
41-50 41-50
Text-figure 14.—A06 Cladiwm and Nymphaceae sclereids (a);
A23 Cladium and Nymphaceae sclereids (b).
64—57 cm is marl; 57-25 cm is fibrous peat. Marl at
60.5—57 cm yielded a radiocarbon date of c. 1960 yr
B.P. The upper part of Zone 2 has no radiocarbon
dates, but the change in pollen spectra between Zone
2 and | suggests that there is a hiatus; however, the
charcoal profile does not show a sharp increase at the
top of Zone 2. This zone is characterized by cheno-
am dominance (Text-fig. 13a), ca. 60% of the pollen
spectrum throughout the zone. At the beginning of
Zone 2 slough-marsh and upland pollen groups occur
at lower percentages than in preceding Zone 1, but by
the end of Zone 2 the slough-marsh component in-
creases to ca. 20% of the total pollen, while the upland
component remains at less than 10%. Ferns occur at
low frequency and the mangrove component makes a
small contribution in the middle of Zone 2. Cladium
sclereid concentrations rise gradually to a peak at 50—
45 cm (Text-fig. 14a), decline and then rise to a higher
peak at 32-25 cm. Nymphaceae sclereid concentra-
tions rise faster than Cladium sclereids and reach a
peak at 57-50 cm, decline slightly, rise to a second
higher peak at 45-39 cm, and then decline again. The
high percentage of cheno-am pollen may be an indi-
cator of rapidly fluctuating water levels and/or variable
hydroperiod. Cheno-ams are early successional weeds
which can take advantage of briefly available habitats
such as areas alternating between inundation and des-
sication over short time intervals. In late Zone 2 at
50-25 cm the slough and marsh pollen group per-
centages rise and both Nymphaea pollen and Nym-
phaceae sclereid concentrations increase. This suggests
a longer hydroperiod during this interval.
Zone 3, 25—0 cm is stratigraphically fibrous peat. A
radiocarbon date on sediment from 25—20 cm yielded
a modern (<~100 years old) date. Zone 3 cheno-am
pollen spectra show a sharp decrease, falling from ca.
60% to ca. 30% and then to less than 20% (Text-fig.
13a). The slough-marsh component increases to almost
30% of the pollen total. Cladium sclereid concentra-
tions (Text-fig. 14a) peak early in Zone 3 (continuing
the trend seen in Zone 2), decline in the middle of
Zone 3, and increase again at the top of Zone 3 (the
surface sample). Nymphaceae sclereid concentrations
continue the decline begun in late Zone 2. Upland pol-
len increases from ca. 15% at the base of Zone 3 to
almost 40% at the top of Zone 3 (Text-fig. 13a). The
most important species in the upland component is
wax myrtle, but buttonbush and Vitis (wild grape) also
contribute to the increase. The increase in the upland
and fern groups may indicate development of a ham-
mock nearby, or may be the result of upland species
colonizing the newly available habitat of levee and
highway embankments (Leyden in Delfino er al.
1994). Mangrove pollen (Rhizophora and Conocar-
pus) occurred sporadically in Zones | and 2 and was
very scarce (mean = 0.28%). In Zone 3 it occurred in
every sample and accounted for almost 1.5% (mean =
1.42%) of the total pollen sum. In all zones mangrove
pollen is probably derived from distant sources, prob-
ably the southwest end of the Shark River Slough. Its
increase in Zone 3 may indicate recent deeper incur-
sions of salt water up the Harney and Shark rivers into
the Slough accelerating the expansion of mangroves
up-slough so that the pollen source area is now a little
closer to AO6 than in the past.
A23H
The A23H pollen sequence is divided into four
zones (Text-fig. 13b). Pollen Zone 1, covers the basal
sandy marl deposits (50-33 cm) and is characterized
by a high slough and marsh percentage, accounted for
almost entirely by /soetes microspores. Nymphaceae
and Cladium sclereids occur at very low concentra-
tions (Text-fig. 14b); Cladium sclereids were not found
84 BULLETIN 361
in the upper sample of Zone 1. Pollen Zone 2 covers
two stratigraphic units, a marl layer from 33-25 cm
and the lower sample from an organic marl layer, 25—
21 cm. It is characterized by cheno-am dominance,
60-70% of the pollen total, and a very low slough-
marsh component percentage (Text-fig. 13b). Both
Cladium and Nymphaceae sclereid concentration in-
crease slightly over levels in Zone | (Text-fig. 14b).
Pollen Zone 3, 21—2 cm covers five stratigraphic units:
the upper sample of the first organic marl layer (21—
12 cm), a second marl unit (21—9 cm), a charcoal
stained marl unit (S—9 cm), a third marl unit (S—3 cm),
and the lower sample of the second organic marl unit
(3-2 cm). Zone 3 is characterized by cheno-am and
pine co-dominance (Text-fig. 13b). The slough-marsh
component is variable, but increases to almost 20% by
the end of Zone 3. Cladium and Nymphaceae sclereid
concentrations increase sharply in Zone 3 (Text-fig.
14b). Zone 4, the surface sample (upper sample of the
second organic marl unit, 2—O cm) is once again dom-
inated by cheno-ams (Text-fig. 13b). The pine and
slough-marsh components percentages are very low.
The upland component percentage which increased
throughout Zone 3 holds steady. Cladium sclereid con-
centrations reach an all time high, while Nymphaceae
sclereid concentrations decline precipitously (Text-fig.
14b).
Zone | is dominated by /soetes microspores (Text-
fig. 13b). Isoetes flaccida, the only species mentioned
in Long and Lakela (1976) for central and southern
Florida today, inhabits sluggish waters. According to
Tryon and Tryon (1982) ‘in the American tropics [/s-
oetes| species grow in lakes, usually in rather shallow
water, in pools or in streams, in grassy turf at the bor-
der of lakes, in wet open sandy soil...” (Tryon and
Tryon, 1982: 829). The presence of /soetes does not
necessarily imply deep or continuously present water.
Pollen and sclereid data indicate low frequencies of
Nymphaea and Cladium/Cyperaceae and a relatively
greater frequency of upland taxa. LOI and HLOI evi-
dence of low percent organic and low percent carbon-
ate, microscopic observation of abundant sand grains,
Isoetes dominance and low pollen concentration, and
low occurrences of Nymphaea and Cladium suggest a
seasonally inundated, wet open sandy habitat, with ar-
eas of clear standing water, possibly an /soetes marsh.
Zone 2 is dominated by cheno-ams both relatively and
absolutely. This zone may represent a time of drought
or highly fluctuating water levels and hydroperiod.
Zone 3 is characterized by Pinus and cheno-am co-
dominance with these two taxa switching back and
forth as dominant. The slough and marsh group in-
creases fairly steadily early in the zone, declines in the
charcoal stained marl unit at 9—5 cm, but recovers in
the upper two samples. Presence of Nymphaea sug-
gests at least pockets of deeper water in the area of
A23 and a longer hydroperiod than that of Zones |
and 2. However the deposition of marl layers during
the time represented by Pollen Zone 3 indicates radical
changes in hydroperiod. Cladium/Cyperaceae pollen
increases during these marl-short hydroperiod epi-
sodes. Zone 3 appears to represent a time when hy-
droperiods were very variable and the landscape was
very patchy with small areas of deeper water sloughs
surrounded by marsh and drier areas.
Zone 4, a single sample, represents modern pollen
deposition (Casuarina is found in this sample). The
pollen spectrum is dominated by cheno-ams. Pollen
from slough-marsh and upland groups decrease in per-
cent and pollen concentrations of these groups fall pre-
cipitously. The Zone 4 pollen spectrum appears to rep-
resent a depauperate wet prairie community.
L67A
The L67A pollen sequence is divided into three
zones (Text-fig. 13c). Zone 1 is older than 3500 yr
B. P. and is characterized by cheno-am dominance less
than 50%, slough-marsh group slightly greater than
pine at ca. 25%, and a small incidence of upland pol-
len. Zone | is interpreted to represent relatively long
hydroperiods. Zone 2 has a basal date of 3500 yr B. P.
The top of the zone is undated. Zone 2 pollen spectra
are dominated by cheno-am pollen at 60-70%.
Slough-marsh and pine groups decrease from Zone |
percentages and pine is always greater than slough-
marsh. By the end of Zone 2 slough-marsh is at its
lowest percentage. Upland group pollen percentages
increase throughout Zone 2. Fern and mangrove
groups appear in upper Zone 2. Zone 2 is interpreted
to represent shorter hydroperiods than Zone 1 with
drought increasing through time. Zone 3 has a basal
modern radiocarbon date. The abrupt change in pollen
spectra signals a probable hiatus. Cheno-am percent-
ages decline abruptly to ca. 25% and continue to de-
cline upcore. Pine increases initially, but then also de-
clines upcore. The slough-marsh group percentage tri-
ples over Zone 2 values and then increases upcore to
dominate the surface sample at ca. 80%. Fern and up-
land groups increase in Zone 1, but decline in the sur-
face sample. Pollen from mangrove taxa is present
only sporadically. Zone 3 is interpreted to represent
recently artificially ponded conditions resulting from
construction of the Tamiami Trail and L67A and C
canals. The L67A pollen diagram resembles those of
AO6 and L67C in the decline of cheno-ams and the
increase in slough-marsh and upland groups pollen in
recent samples.
HYDROLOGY, VEGETATION, AND CLIMATE: WINKLER ET AL. 85
L67C
Only a few samples from L67C have been examined
(Text-fig. 13d). These samples resemble those from the
A06 core. Cheno-am dominates the pollen spectrum in
the oldest sediment; pine is variable, as is the slough-
marsh component. The upland component increases
over time and is a large proportion of the surface sam-
ple. The mangrove component is also most evident in
the surface sample.
Gator Lake
Subsequent to our coring effort at Gator Lake, we
learned that the lake had been cored by a group from
the Environmental Engineering Dept. of the University
of Florida in Gainesville, Florida (Delfino et al., 1994).
This group obtained a 40 cm core from Gator Lake
which was 7'°Pb-dated and found to cover 90 years.
The core was used for Hg analysis of the sediment and
for pollen analysis (carried out by B. Leyden [Delfino
et al., 1994]). The pollen indicated that willow trees
expanded near the lake in the last 50 years becoming
most abundant since the 1980’s. Cheno-am_ pollen
dominated throughout the 90 years covered by the sed-
iment core. Pine pollen decreased in the last 35 years,
while pollen from Compositae increased in the last 50
years. The aquatic environment was dominated by Nu-
phar (yellow pond lily) and Sagittaria (arrowhead).
Pollen analysis of our longer sediment cores from
Gator Lake is in progress. One interesting finding is
that Nuphar pollen is abundant in Gator Lake sedi-
ments from 400 cm depth (about 1500 yr B. P.) to the
present. At other sites in the Everglades, Nymphaea
(white water lily) is the abundant water lily pollen.
Pollen analysis also indicates that Salix (willow) pollen
is only present in the top 40 cm representing the recent
several decades at Gator Lake. The lake is presently
surrounded by shrub willows. These may have spread
from stands that grew on spoil piles from construction
in the 1930’s and 1940’s of the nearby Miami Canal
or may have become established after local fires
burned the surrounding peat during drought (see char-
coal stratigraphy, Text-fig. 5). Willows are known to
colonize burned-over peats in the Everglades (Hof-
stetter, 1974) and the high charcoal in the recent sed-
iments of Gator Lake (Text-fig. 5) suggests that willow
growth near the lake may have been initiated after fire
(probably during drought or drawdown due to canali-
zation) burned into the deep sawgrass peat around the
lake.
Lignumvitae
Lignumvitae Key was chosen as a coring site be-
cause it 1s a protected state botanical site, has a diverse
hardwood hammock vegetation, and has an elevation
Lignum Vitae, LV 2
Percent Percent
Depth cm
Wo (n = 883) 1 (n = 481)
26-100 So YL ALAR)
2290 +/- SO vey BP
Rhizophora
% Cheno-am
|
Conocarpus
[a
e
Oo Laguncularia
lm aviewen
% Other
a b
Text-figure 15.—a and b. Lignum Vitae (LV2) pollen diagrams
showing percent Chenopodiaceae-Amaranthaceae versus all other
pollen (a) and mangrove pollen divided into constituent species (b).
In (b)n = the sum of mangrove pollen.
of close to 5 m, one of the highest keys in Florida
Bay. The LV-2 coring site was a relatively shallow
(about 10 cm), saline, mangrove-fringed (Conocarpus,
Avicennia) pond (pH 8) separated by an old limestone
ridge from a tidal flat. Saltmarsh plants such as Sali-
cornia (saltwort) and Limonium sp. (sea lavender)
were growing on the tidal flat, while the cactus Opun-
tia cereus was found near the pond. Two samples have
been analyzed for pollen: 96—100 cm and 20—25 cm.
(Text-fig. 15). The 96-100 cm sample is the basal or-
ganic sediment which was AMS radiocarbon dated at
2290 + 50 yr B. P. The 20—25 cm sediment is undated
but is within the historic horizon.
The pollen in the younger sample is quite different
from that in the older. The younger sample is domi-
nated by Chenopodiaceae/Amaranthaceae (pigweed
family) pollen, which constituted only 14% of the old-
er sample (Text-fig. 15a). Most of the pollen in the
older sample is Conocarpus, a mangrove plant. Con-
sidering mangrove pollen only (Text-fig.15b) Rhizo-
phora and Avicennia are present in only trace amounts
and Laguncularia is absent in the older sample; while
in the younger sample Rhizophora and Conocarpus are
co-dominant with only minor representation of other
86 BULLETIN 361
mangroves, Laguncularia and Avicennia. Pollen of Al-
nus, Salix, Acacia, Guaiacum officinale, and G. sanc-
tum is present in the older sample, but not the younger
sample. Pollen of Pinus, Taxus-Cupressaceae-Taxo-
dium, Juniperus, Myrica, Schinus, Anacardiaceae, Pal-
mae, Solanaceae, Tubuliflora Compositae, and Legu-
minosae are present in the younger sample, but not the
older one. Taxa found in both samples include: Bur-
sera, Myrtaceae, Alternanthera/Gomphrena, and
Rhamnaceae.
Of the mangrove taxa Conocarpus is the least de-
pendent on immersion in fresh or saline water. It pre-
fers drier ground than the other three taxa and is also
a component of the tropical hardwood forest. Its dom-
inance of the older sample suggests drier conditions at
2290 years ago and lower sea level. It is, however,
growing abundantly near the coring site now, co-dom-
inant with Avicennia (black mangrove). Rhizophora is
also present today and is replacing Conocarpus as sa-
linity of the site increases due to a rising sea level.
The sparseness of Avicennia pollen in the two samples
examined and abundance of Avicennia at the site today
suggests that this mangrove species has expanded in
recent decades at the LV2 site. However, Riegel (1965)
states that Avicennia produces little pollen which has
a restricted local distribution. He considers Avicennia
pollen percents of 1—2% to indicate presence of the
plants at the site and pollen percents more than 5% to
indicate dominance at a site.
Chenopodiaceae dominance of the younger sample
and the presence of Schinus pollen may indicate clear-
ance of some of the island for habitation thus opening
up the tropical forest habitat to invasion by weeds. The
pollen spectra document the declining presence of Lig-
num Vitae (Guaiacum spp.) on the island. While ham-
mock flora is still well-represented, anthropogenic
pressures have greatly changed the forest composition
on Lignumvitae Key.
Cladocera Analysis
Cladocera (Arthropoda: Crustaceae, Branchiopoda)
are small aquatic invertebrates (0.2—2.0 mm) found
primarily in fresh water. A few taxa are marine and
some freshwater taxa can tolerate brackish water. All
of these creatures can swim, but while some forms are
primarily free-swimming in open water others are bot-
tom-dwelling or epiphytic. Cladocera are indicators of
water chemistry and nutrient status. Loftus ef al.
(1986) found cladocera in the Shark River Slough.
Chitinous exoskeletons of cladocera are common com-
ponents of lake sediment and should therefore be pre-
sent in marsh-slough deposits in the Everglades.
A06
Chitinous exoskeletal remains of 28 cladoceran taxa
were identified from AO6 sediments (Table 4). Of
these, Euryalona and Oxyurella have not been found
in previous neolimnological surveys of ENP (Loftus,
unpub. data, 1993), however they have been found in
southern Florida (Crisman 1980, Frey 1982) and it is
reasonable to expect to find them in ENP.
Cladoceran remains occurred throughout the core
(Text-fig. 16a). They were very rare in the organic
marl at the base of the core and only two taxa were
represented, Alona spp. and Chydorus brevilabris. The
number of cladoceran parts and taxa increased slightly
in the lower part of the peat, 57-20 cm. Cladoceran
parts and taxa increased again beginning at 20 cm and
continued at an exponential rate up to the surface, 7—
0 cm. One possible explanation for this distribution is
post-depositional degradation of cladoceran remains
over time although chitinous exoskeletons are known
to resist decay in lake and bog deposits over even lon-
ger spans of time than are represented by this core
(Frey 1964). Oxidation during dry phases would not
affect cladoceran remains more than the organic matrix
surrounding them, therefore the number of cladoceran
remains in a given sample of Everglades peat or marl
probably fairly represents the number of cladocera liv-
ing in the area at the time the deposit formed. Hunt
(1952) found that cladocera were never very abundant
in the open water of the Tamiami Canal, but that lit-
toral cladocera were found in periphyton growing on
emergent and submerged plants. Most of the species
listed in Table 4 are littoral substrate dwellers which
live on plants rather than in open-water. Thus another
explanation for the distribution of cladoceran remains
in AOQ6 deposits is that cladocera are tracking periph-
yton production, which is in turn responding to water
level and nutrient changes. The cladoceran data would
suggest that periphyton production increased dramati-
cally at this site in the last 50 years. This kind of
limnological change would be expected in light of ev-
idence of increased allochthonous nutrients to the site
in recent decades.
A23H
Chitinous exoskeleton remains of at least 17 cla-
doceran taxa were recovered from A23H deposits (Ta-
ble 5, Text-fig. 16b). Cladoceran remains are so scarce
in the deposits from 50—21 cm that little can be in-
ferred from them except that conditions at the A23 site
during this time never favored development of a cla-
doceran assemblage. Rapidly fluctuating water levels
and low nutrients could be contributing factors. Pe-
riphyton production may have been insufficient in this
HYDROLOGY, VEGETATION, AND CLIMATE: WINKLER ET AL. 87
Table 4.—A06 cladocera taxonomy and counts. Minimum number (MNI) molts/deaths on counts from four slides per sample.
Depth (cm)
Taxon 0-7 7-14 14-20 20-25 25-32 32-39 39-45 45-50 50-57 57—60.5 60.5—64 64—70 70-75
Alona affinis 2 1 2 — —_ 1 -
A. verrucosa 4 | 1 — 1 — —_— 0.5 1
A. setulosa 2 1 —_ — —- 1
A. rustica 4 1
A. guttata type 3 1 1 2
A. karua type 0.5 0.5 1
Alona spp. | | | — — — 05° —
Euryalona = — 1 — —_— 1
Chydorus brevilabris 1 — 1.5 — 0.5
Chydorus spp. — 2 — 1.5 — 1 1.5 — — — 1 —_- —
Chydorus nitidulus/Ephemeroporus ES 0.5 — 0.5
Ephemeroporus hybridus — 0.5
E. cf. archboldii 0.5
Dunhevedia 1
Oxyurella l
Leydigia acanthocercoides 1 1S = 1
L. cf. parva |
Leydigia spp. ] 1 1
Alonella excisa 2
Alonella spp. A) 1 — 1 _ — — 0.5
Pleuroxus/Alonella — — 0.5 0.5
Camptocercus l 1
Ilyocryptus 2 0.5 165) 0.1 0.5 —- —- —
Grimaldina 1
cf. Eubosmina tubicen IGS) 1
Ceriodaphnia 0.5 — — —
Daphnidae ies) — 0.5 — 1 — — 155) — 0.5 — —_-_ —
Scaphoberis 0.5
Total MNI 34.5 9'5 10 Sal DS Tes) 3 6.5 3.5 1 1 05° —
Total No. Taxa 21 10 9 5 3 il 3 8 4 2 1 1 —
Total Cladoceran Pieces 193 133 106 29 30 34 32 37 13 10 5 5 3
A06 A23
Cladoceran pieces per g dry wt Cladoceran pieces per g dry wt
° © 2 iS iS S = ° ° ° ° ° ° °
L 1 1 1 4 1 ai]
0-7 0-25 —
foie 2:34) Ab 1380 +/- 70 YBP
14-20
20-25 S25
— 25-32 — 5-9 1660 +/-100 YBP
a) a)
2 a e912 1890 +/- 70 YBP
B 45-50 Q 12-21 2600 +/- 70 YBP
Oevsp-57, OW ai-25 2840 +/- 70 YBP
57-60.5 1960 +/- 30 YBP 25-33 4 2830 +/- 60 YBP
60.5-64
64-70 33-41 |
70-75 2100 +/- 30 YBP 41-50 2840 +/- 70 YBP
a. b.
Text-figure 16.—Cladoceran pieces per gram dry weight for AO6 and A23. Error bars = | standard deviation based on 4 replicates per
sample.
88 BULLETIN 361
Table 5.—A23H cladocera taxonomy and counts. Minimum number (MNI) molts/deaths on counts from four slides per sample.
Depth (cm)
Taxon 0-2 2-3 3-5 5-9 9-12 12-21 21-25 25-33 33-41 41-40
Alona affinis I 1
A. circumfimbriata 2
A. costata 1
A. guttata 1
A. intermedia 1
A. rustica l
A. setulosa 2 1
A. verrucosa 3
Alona spp. — — 3 2.5 — — 1
Oxyurella 1
Camptocercus |
Leydigia parva/acanthocercoides 1 i
Alonella cf. dadayi 2
Alonella spp. 0.5
Chydorus brevilabris 2 _ 0.5
Chydorus/Pleuroxus 1]
cf. Ceriodaphnia 1
Daphnia spp. 1 1
Ilyocryptus ct. gouldeni 0.5
Ilyocryptus spp. —— — 1
Macrothrix paulensis 1.5
Macrothricidae 0.5
Total MNI 19 4 3)5) 3 2 10) 3.5 0 0 (0)
Total No. Taxa 12 4 4 2 2 (0) 4 0 (0) 0
Total Cladoceran Pieces 189 38 45 8 5 3 5 0 (0) 10)
area from 2800-1650 years ago to support many cla-
docera. Cladoceran remains increase in samples from
12-3 cm suggesting more permanent water and/or in-
creased nutrients during this time interval. Cladoceran
remains increase exponentially in the uppermost two
samples. Because other data suggest lower water levels
or shorter hydroperiods during this time, increased nu-
trients may be a factor in the increase of cladocera at
this site.
Sponge Spicule Analysis
Spicules are the siliceous remains of sponges (phy-
lum Porifera). Pennak states that ““Sponges are most
common in waters less than 2 m deep ... and are
seldom found in extremely rapid waters’ (Pennak
1978). They require hard substrates and rarely grow
on mud or silt. Aquatic macrophytes can, however,
provide a substrate for sponges to encrust. They are
quite sensitive to environmental conditions such as
light intensity, pH, conductivity, bicarbonate ion con-
centrations, and calcium concentrations, and are very
sensitive to eutrophication and pollution (Harrison
1974, 1988; Poirrier, 1974; Pennak 1978; Frost 1991).
The Everglades would seem to provide ideal habitat
for freshwater sponges, but they have rarely been stud-
ied (Poirrier, 1976; Harrison et al., 1979) and their role
in the Everglades ecosystem is not known; however,
G
sponges can have a “substantial influence on total nu-
trient cycling and primary production ... [and may
operate] primarily to limit the availability of materials
to other organisms within the food web” (Frost 1991).
Because sponges are potentially competitive with pe-
riphyton for substrate, their role in the development of
the Everglades may be of interest.
A06
Sponge spicules from A06 deposits appeared to be
primarily smooth, cylindrical fusiform, amphioxic
megascleres when viewed with ordinary light micros-
copy (Text-fig. 17a, Plate 1, fig. 3). All megascleres
were fragmentary and some showed varying degrees
of axial canal widening and degradation. This situation
is similar to that of spicular remains studied by Racek
(1966) in deposits from Laguna de Petenxil, El Petén,
Guatemala. He attributed axial canal widening and
degradation to bacterial reduction of gypsum in the
sediment releasing Ca(OH), which dissolved hydrated
silica. The degraded condition of AO6 megascleres
suggests that this process may have occurred in the
Everglades in the past. Increased amounts of Ca(OH),
in the sediment from bacterial action may also be op-
erative in the dissolution of diatoms.
Some A06 megascleres were virtually complete,
missing only their tips. Measurement of 50 virtually
HYDROLOGY, VEGETATION, AND CLIMATE: WINKLER ET AL. 89
A06 A23
Sponge spicule pieces per g dry wt Sponge spicule pieces per g dry wt
2 8 S S S S S S o J iJ =] J i=) i=J i=
o o o i=) o o Oo o o o o o o o
i=) i=] o o i=} i=} o o o i=} i=] o i=} o
N v o o o N Tv N tc o o Oo N vt
SSS ES 4
0-7 0-2
7-144
2-3 3 a Rp
aera 4 1380 +/- 70 YBI
» 20-2544 SAS
E 25-324 E 5-9 1660 +/-100 YBP
me SG) = 9-12 1890 +/- 70 YBP
B 39-454 a 3
& 45-504 Bie 2600 +/- 70 YBP
50-57- 21-25 2840 +/- 70 YBP
57-60.5 4 1960 +/- 30 YBP 25-33 2830 +/- 60 YBP
60.5-64 coe
64-70
70-75 2100 +/- 30 YBP 41-50 2840 +/- 70 YBP
a. b.
Text-figure 17—Sponge spicule pieces per gram dry weight for AO6 and A23. Error bars = | standard deviation based on 4 replicates per
sample.
complete spicules from 20—14 cm depth indicates that
megascleres average at least 396 4m long (range 300—
476 pm, S.D. 43 pm) by 30 pm w (range, 16.5—43
wm, S.D. 6.4 4m). Several taxa have smooth mega-
scleres in this size range: Ephydatia fluviatilis, Spon-
gilla alba, and S. cenota (Frost 1991). Unfortunately,
identification is based on microscleres and gem-
moscleres, which are scarce in A06 deposits. The few
gemmoscleres seen resemble those of Spongilla lacus-
tris, but the spines are not very well defined. These
spicules are slightly bent cylindrical fusiform am-
phioxes from 60—70 pm | with sparsely, but + uni-
formly distributed, short, broad-based spines (Plate 1,
fig. 4). With light microscopy the axial canal is visible
running lengthwise within this spicule type. The ab-
sence of definite microscleres and the large size of the
megascleres suggest that E. fluviatilis may also be pre-
sent, but its birotulate gemmoscleres were not seen.
Sponge spicule pieces are present in every sampling
increment of the AO6 core, but they are rare in the
dark organic marl (75-57 cm, Text-fig. 17a). Two
peaks in sponge spicule pieces are, however, apparent
in the peat: one at 45—39 cm and another at 20-14 cm.
when diatom frustules are also most numerous.
A23H
Sponge spicules were found throughout the A23 de-
posits (Text-fig. 17b). A23 spicules were similar to
those already described for AO6. They occurred in low
numbers in deposits from 50—21 cm with a small peak
at 41-33 cm. A very large peak occurred in the organic
marl unit from 21—12 cm at the same time that plank-
tonic diatoms occur. From 12—5 cm, spicules were
very rare. Then from 5 cm up to the surface sponge
spicules increased exponentially as did Cladium scler-
eids and cladocera pieces. There was no numerical re-
lationship among these three biotic indicators in earlier
levels, therefore the increase in all three indicators may
reflect some structural or functional change in the
marsh ecosystem, possibly increased nutrients.
HOLOCENE CLIMATE ON THE FLORIDA
PENINSULA
Our study of sites in the southern Everglades indi-
cates a shift from a dry to wet climate in the middle
Holocene at about 5000 yr B. P. which initiated for-
mation of the peat-forming Everglades. Similar chang-
es were taking place circum-Caribbean (Text-fig. 18).
We looked for continental- and regional-scale factors
which would effect this dramatic change. Climatic
changes in North America during the Holocene were
determined by collapse of the Laurentide ice sheet, the
position of sea level due to ice retreat, the relative
amounts of land and sea in the hemisphere, changes
in the orbital parameters of the Earth which determine
insolation variations and seasonality, volcanic erup-
tions, and other boundary conditions (Text-fig. 19).
Changes in the North Atlantic thermohaline circulation
pattern would also affect what happens on land in
southern Florida. Sea surface temperatures adjacent to
the Florida peninsula are important in determining the
amount of annual and seasonal precipitation with
warmer water causing greater rainfall. In our investi-
gation of the cause of the change from dry to wet
90 BULLETIN 361
| | | 5 ©}
| | | ice) =I
77) (re om
| o = = =o gr a
| E 2 fe) =| eS as 2
g 2 s | O28 0 w| |
BS w ao oe oS Oc} —|
© = pO eo =
i=) ne) ne] x 5 a0 0 or ®—
a= ) oc | cS Dow iL co
ao ® 2 ® >| Osc 2Q- =-c% ao] Bw
ow gs go Els o— oO C SE 3 0
So ahs ior ax Oc o Fe
2 DE 2a ayes a2 024 £2 oe
S| 28 Go| 995) 25 =8S/ 58| s=
>| wo ws} Wor sa 5300 zs JO
3
}
Qa
@
3
~)
So
oO
We
Dry
NoData[
lext-figure 18.—Time and space comparison of qualitative hydrologic changes (Wet/Dry) from regional paleoecological data. Sites range
from thern to southern (ENP sites) Florida and from Haiti in the Caribbean to northern Guatemala. Timing of wet/dry transitions may not
be a e as the line separating them suggests at some sites.
HYDROLOGY, VEGETATION, AND CLIMATE: WINKLER ET AL. 9]
f T T Ti T T 1
July + % solar radiation |
Eslean Peg aurentics
Sg d ONES Seasonality 4
fee Envelope
Cee | TOM. Yinceh. a} 30
5 L : Everglades 2
re Hydrology @
o Warm n Ory =
a x a 1 es
=
o ®
* Modern 25 3
2 3
z H
s Cool+ x
a z
l :
fe} 20 G4
Vv SST Au
Coldkaa m 9
‘ 4& SST Feb
January - % solar radiation
| | | | | |
18 15 12 ) 6 3 te)
yr. B-P. (x 10%)
Text-figure 19.—Late-glacial and Holocene climate change dia-
gram. The seasonality envelope illustrates changes in % solar radi-
ation in July (+) and January (-) throughout the last 18,000 radio-
carbon years (possibly 21,000 years if compared to the Uranium-
Thorium coral chronology) determined by changes in the orbital
parameters of the Earth (the Milankovitch cycle). It demonstrates
that at about 12,000 through 8,000 yr B. P. seasonality was different
from today because the Earth was closest to the sun in July (today
Earth is closest to the sun in January) and summers were warmer
and winters colder than modern in middle latitudes. The Laurentide
ice sheet melted from the continent by about 7000 yr B. P. (Dyke
and Prest, 1987). Sea Surface Temperature scale (SST, closed tri-
angles: August; open triangles: February) is on right and indicates
from foraminifera changes in ocean cores that SST (southeast of
South Florida) was slightly warmer in winter than modern at about
6000 yr B. P. (Ruddiman and Mix, 1993). Sea Level Rise, the lower
dashed-line curve, reaches to 4.5 m by about 5000 yr B. P. (Dyke
and Prest, 1987; Wanless er al., 1994) but the rate of rise varied
after about 3000 yr B. P. The Everglades peat was initiated at about
5000 yr B. P. possibly due to increased winter precipitation because
of the warmer SST’s and the rising sea level. The changes in Ey-
erglades hydrology (Wet or Dry) interpreted from our results is
shown from 5000 yr B.P. to the present. Diagram modified from
Kutzbach and Street-Perrott (1985).
climate in Florida and the Caribbean region at about
5000 yr B. P., we consider two mechanisms: sea level
rise and changes in ocean temperatures after collapse
of the Laurentide ice sheet. Calculations of orbital con-
ditions at 5000 yr B. PR. shows maximum insolation in
early fall. Sea surface temperatures would therefore be
higher at that time and the incidence of tropical cy-
clones would increase. Increased storminess would
bring more moisture to the Florida peninsula. Fossil
data from foraminifera in ocean cores from the region
(Ruddiman and Mix, 1993) indicate that sea surface
temperatures southeast of the Florida peninsula did
change at about that time. Sea surface temperatures
were about 0.3°C warmer than modern in winter 6000
yr B.P., but about the same as modern (~28.5°C) in
summer. In addition, because the S. Florida peninsula
is relatively narrow with low relief and has a porous
limestone aquifer, sea level rise during the Holocene
was important to initiation of Everglades peat. Sea lev-
el was 120 m lower than the present mean sea level
at the last glacial maximum (18,000—21,000 yr B. P.),
had returned to about 6.2 m below modern levels by
5500 yr B. P. (Fairbanks, 1989; Wanless ef al., 1994),
to about 4.5 m below modern levels by 5000 yr B. P.
(Dyke and Prest, 1987) and to about 3 m below mod-
ern levels by 4000 yr B. P. (Colquhoun and Brooks,
1986). Rising sea level would raise the freshwater ta-
ble in the Everglades wetlands (the freshwater aquifer
overlies saltwater) and would slow sheetflow through
the Everglades from the north.
Most sites throughout Florida indicate drought be-
fore 5000 yr B. P. and wet conditions after that time
(Text-fig. 18), regardless of their proximity to the sea,
indicating that changing climatic conditions such as
increased precipitation due to rising sea surface tem-
peratures in the middle Holocene (Ruddiman and Mix,
1993) especially in winter, may have been equally im-
portant as sea level rise in initiation of the Everglades
landscape. Furthermore, increased precipitation in
winter in the middle Holocene (about 6000-5000 yr
B. P.)—winter today being the driest time of the Ev-
erglades year—may have provided the needed increased
effective moisture for peat to accumulate at many sites.
Other global-scale climatic events such as more nu-
merous El Nino years since 5000 yr B. P. (Sandweiss
et al. 1996) would also increase precipitation to the
southern peninsula of Florida as it does today (Winkler
et al., 1999).
The interaction of rising sea level and increased pre-
cipitation resulted in long hydroperiod in low-lying ar-
eas and the accumulation of peat, especially in the Ev-
erglades Depression south of Lake Okeechobee. High-
er elevation areas adjacent to the Everglades Depres-
sion did not accumulate peat until sea level rise raised
the water table to intersect local low points. Once peat
is present on the landscape it retards waterflow, further
increasing the hydroperiod (water residence time) at a
site. The presence of peat would also raise the water
table at a site, separating the water in the peat from
the carbonate-rich Everglades aquifer and thereby
changing the chemistry of the site. The peat would
then be watered by a larger proportion of water from
rainfall than by groundwater or surface flow. The for-
mation of barrier beaches surrounding southern Florida
and the topography within the old limestone bedrock,
also affect the rate of waterflow through the Ever-
glades drainage and the accumulation of freshwater
peats. Some of the higher elevation areas may never
have accumulated peat. Instead seasonally wet prairies
developed on short hydroperiod, higher elevation sites
and marl deposition began.
92 BULLETIN 361
Our evidence for an increase in effective moisture
in winter at about 6000—SO000 yr B. P. and initiation of
peat deposition in the southern Everglades supports the
Genesis! climate model simulations for the Caribbean
region which show that in winter (December, January,
February) at 6 ka, surface temperatures increased by
0.74°C, precipitation increased by 0.79 mm/day, and P
— E increased by 1.2 mm/day, compared to modern
values (Thompson and Pollard, 1995).
What we probably see in the Everglades today is a
landscape that has become increasingly fragmented af-
ter 2900 yr B. P. by shrinkage of the peat-based plant
communities and expansion of the drier plant com-
munities. The modern dry conditions are further ex-
acerbated by anthropogenic drainage, canalization, and
drawdown of freshwater.
IMPACTS FOR LAND MANAGEMENT AND
DECISION-MAKING
Today the Everglades hydrologic environment is in-
creasingly uncoupled from climatic changes as water
is artificially sequestered in some parts of the historic
Everglades and diverted from other parts. Recent an-
thropogenic nutrient and hydrologic changes seem to
favor periphyton production leading to widespread
marl deposition in contrast with climate- and sea-level-
induced changes over the past 5 millennia when stand-
ing deepwater peats formed. However, some of our
cores (e.g. A23, Ever6, and 9-Mile) show that some
areas have experienced marl production only and have,
therefore, always had shorter hydroperiods. These marl
prairies are important habitats for native Everglades
fauna, notably the Cape Sable Sparrow. Therefore,
simply increasing the amount of water standing over
the whole Everglades by artificially maintained long
hydroperiod/high water levels will not preserve the
natural Everglades. Neither will suppression of wild-
fires, which are necessary to the maintenance of the
mosaic landscape by burning off peat and naturally
lowering the topography so that ponds and sloughs can
develop and provide sites for peat to accumulate again.
A rewatering plan must include both wet and dry sea-
sonal cycles in order to preserve the natural landscape
of the southern Everglades—a shifting mosaic of wet
and dry communities in space and time—as a func-
tional habitat for both plants and animals. The mosaic
landscape of the Everglades may become less complex
if long-term water management does not take into ac-
count the natural dynamics of the hydroscape and the
topographically variable landscape which offers di-
verse habitat for flora and fauna.
Sea-level rise is another confounding factor that
must be taken into account in an Everglades rewater-
ing plan. There is pollen (mangrove taxa replacement)
and diatom evidence of sea-level changes that correlate
with findings in cores from Florida Bay. Chemical
analyses of C111 sediments suggest a salt water in-
cursion several millennia ago at sites east of Taylor
Slough followed by a subsequent change to fresher
water as channels were closed off. Therefore, changes
in rates of sea-level rise and in the topography of Flor-
ida Bay, both volatile processes, should be considered
during water delivery to the lower Everglades.
CONCLUSIONS
Paleoecological study tells us that the Everglades
landscape was (and is) a shifting mosaic of biotic com-
munities initiated at about 5000 yr B.P. when the
southeastern U. S. and a broader Caribbean region un-
derwent transition from dry to wet climate. Our results
indicate that most sites alternated between deposition
of wet peat or drier marl at times throughout their ex-
istence. These facts document that the modern mosaic
landscape existed in time as well as space for at least
five millennia. Fire and low, drought-induced, water
levels are critical natural structuring agents in the de-
velopment of the Everglades landscape but these
events cause severe sediment decomposition. Due to
fire and drought there are probably no marsh or prairie
deposits with complete paleostratigraphic records and
it is therefore necessary to piece together the history
of the Everglades from many cores. These cores
should be well dated because hiatuses in the record are
not always stratigraphically obvious. A wide variety
of physical and biotic parameters should be examined
as well because chemical and hydrologic extremes also
result in gaps in the biologic record.
Most plant communities in the past were similar (in-
ferred from pollen and sclereid analyses) to those pre-
sent in today’s landscape, although in the course of
our research we found at least one habitat type, an
Isoetes marsh, which is not recognized in the modern
Everglades. /soetes marsh is found today on the Yu-
catan Peninsula, but is not listed as a biotic community
in any of the vegetation surveys of the historic Ever-
glades. Other vegetation changes, including the peri-
phyton composition, have been caused by nutrient and
trace element increases from land use changes in the
last 50 or so years and the introduction of aggressive
exotic plant species. However, dominant diatoms
found in surface marls were also found to be the same
taxa that dominated in ancient marl sediments. Cla-
doceran assemblage changes appear to correlate with
nutrient-driven periphyton changes in top sediments.
Pollen also provided evidence of changes in man-
grove species in Lignumvitae Key in Florida Bay.
There Rhizophora is replacing Conocarpus as the sa-
linity of the site increases due to sea level rise. At that
HYDROLOGY, VEGETATION, AND CLIMATE: WINKLER ET AL. 93
site pollen also documents the declining presence of
Lignum Vitae trees on the island and other hammock
forest changes due to anthropogenic pressures includ-
ing the introduction of exotic plant species.
Planktonic diatom increases documented a wet pe-
riod between 2000 and 1600 yr B. P. at several sites
over a broad region of the Everglades. This wet period
appears to correlate with an increase in the rate of sea-
level rise suggesting that sea-level changes affect sites
far inland in the southern Everglades as well as sites
in and near Florida Bay. However, although sea-level
rise continued, freshwater at inland sites lowered after
1600 yr B. P., as indicated by subsequent marl sedi-
ments and increases in epiphytic diatoms.
The recent hydrologic environment of the Ever-
glades and ENP appears to be one of marl deposition.
Relatively old dates on sediments near the upper part
of cores suggest that in the last centuries there were
large water-level changes leading to decomposition
and loss of recent sediments during times of drought
and also possibly flushing of sediments out of the sys-
tem during times of high flow. The lack of accumu-
lation of modern peat sediments except in places
where water pools due to human structures, the con-
centration of charcoal particles in top sediments, the
algal floc and marl found in recent sediments, and the
old dates of near-top sediments, suggest this change in
hydrologic regime.
REFERENCES CITED
Barker, P.
1992. Differential diatom dissolution in Late Quaternary sedi-
ments from Lake Manyara, Tanzania: an experimental ap-
proach. Journal of Paleolimnology, vol. 7, pp. 235-251.
Barker, P., Fontes, J.-C., Gasse, F., and Druart, J.-C.
1994. Experimental dissolution of diatom silica in concentrated
salt solutions and implications for paleoenvironmental re-
construction. Limnology and Oceanography, vol. 39, pp.
99-110.
Bateman, L. and Rushforth, S.R.
1984. Diatom floras of selected Uinta Mountain Lakes Utah,
U.S.A. Bibliotheca Diatomologica, vol. 4, pp. 1-72.
Berglund, B.E. and Ralska-Jasiewiczowa, M., editors
1986. Handbook of Holocene Paleoecology and Paleohydrolo-
gy. J. Wiley, New York. 869 pp.
Blinn, D.W.
1993. Diatom community structure along physicochemical gra-
dients in saline lakes. Ecology, vol. 74, pp. 1246-1263.
Bradbury, J.P.
1975. Diatom stratigraphy and human settlement in Minnesota.
Geological Society of America Special Paper 171, 74 pp.
Bragg, Rick
1999. Florida rain fails to remove threat of fires. New York
Times, April 19, 1999.
Brooks, H.K.
1974. Lake Okeechobee. in Environments of South Florida: Pre-
sent and Past. PJ. Gleason, ed., Miami Geological Society
Memoir 2, Miami, Florida. pp 256-286.
Browder, J.A., Black, S., Schroeder, P., Brown, M., Newman,
M., Cottrell, D., Black, D., Pope, R., and Pope, P.
1981. Perspective on the ecological causes and effects of the
variable algal composition of Southern Everglades pe-
riphyton. Report T-643. South Florida Research Center,
Homestead, Florida. 110 pp.
Browder, J.A., Gleason, P.J., and Swift, D.R.
1994. Periphyton in the Everglades: spatial variation, environ-
mental correlates, and ecological implications. in Ever-
glades: The Ecosystem and its Restoration. S.M. Davis
and J.C. Ogden, eds., St. Lucie Press, Delray Beach, Flor-
ida, pp. 379-418.
Brown, J.G. and Cohen, A.D.
1985. Palynologic and petrographic analyses of peat deposits,
Little Salt Spring. National Geographic Research, vol. 1,
pp. 21-31.
Camburn, K.E., Kingston, J.C., and Charles, D.F., editors
1984-1986. Diatom Iconograph. Paleoecological Investigation
of Recent Lake Acidification (PIRLA). unpublished re-
port.
Camburn, K.E. and Kingston, J.C.
1986. The genus Melosira from soft-water lakes with special
reference to northern Michigan, Wisconsin, and Minne-
sota. in Diatoms and Lake Acidity. J.P. Smol et al., eds.,
Junk, Dordrecht, pp. 17—34.
Charles, D.F.
1985. Relationships between surface sediment diatom assem-
blages and lakewater characteristics in Adirondack lakes.
Ecology, vol. 66, pp. 994-1011.
Charles, D.F., Whitehead, D.R., Anderson, D.S., Bienert, R.,
Camburn, K.E., Cook, R.B., Crisman, T.L., Davis,
R.B., Ford, J., Fry, B.D., Hites, R.A., Kahl, J.S., King-
ston, J.C., Kreis, R.G., Jr., Mitchell, M.J., Norton,
S.A., Roll, L.A., Smol, J.P., Sweets, P.R., Uutala, A.J.,
White, J.R., Whiting, M.C., and Wise, R.J.
1986. The PIRLA project (Paleoecological Investigation of Re-
cent Lake Acidification): preliminary results for the Ad-
irondacks, New England, N. Great Lakes States, and N.
Florida. Water, Air and Soil Pollution, vol. 30, pp. 355—
365.
Clausen, C.J., Cohen, A.D., Emiliani, C., Holman, J.A., and
Stipp, J.J.
1979. Little Salt Spring, Florida: a unique underwater site. Sci-
ence, vol. 203, pp. 609-614.
Cohen, A.D. and Spackman, W.
1974. The petrology of peats from the Everglades and coastal
swamps of southern Florida, in Environments of South
Florida: Present and Past. J.P. Gleason, ed., Miami Geo-
logical Society Memoir 2, pp. 233-255.
Cohen, A.D. and Spackman, W.
1977. Phytogenic organic sediments and sedimentary environ-
ments in the Everglades mangrove complex. Part II. The
origin, description and classification of the peats of south-
ern Florida. Palaeontographica Abt. B, Band 162, pp.71—
114.
Colquhoun, D.J. and Brooks, M.J.
1986. New evidence from the southeastern U.S. for eustatic
components in the late Holocene sea levels. Geoarchaeol-
ogy, vol.1, pp. 275-291.
94 BULLETIN 361
Crisman, T.L.
1980. Chydorid cladoceran assemblages from subtropical Flor-
ida. in The evolution and ecology of zooplankton com-
munities. W.C. Kerfoot, ed., University Press of New
England, Hanover, New Hampshire, pp. 657—668.
Dean, W.E., Jr.
1974. Determination of carbonate and organic matter in calcar-
eous sediments and sedimentary rocks by loss on ignition:
Comparison with other methods. Journal of Sedimentary
Petrology, vol. 44, pp. 242-248.
Delfino, J.J., Rood, B.E., Andres, J.M., Jr., and Earle C.D.A.
1994. Mercury spatial heterogeneity in Everglades soil cores
and a comparison of mercury accumulation in wetlands
and asssociated lakes. DEP Final Project Report Volume
1. 75 pp.
Digerfeldt, G., Hakansson, H., and Persson, Th.
1980. Paleoecological studies of the recent development of the
Stockholm lakes Langsjon, Lillsjon, and Laduviken. Uni-
versity of Lund, Sweden, Dept. of Quaternary Geology.
Report, 20. 66 pp.
Dixit, S.S., Cumming, B.F., Birks, H.J.B., Smol, J.P., Kingston,
J.C., Uutala, A.J., Charles, D.F., and Camburn, K.E.
1993. Diatom assemblages from Adirondack lakes (New York,
USA) and the development of inference models for ret-
rospective environmental assessment. Journal of Paleo-
limnology, vol. 8, pp. 27—47.
Dodd, J.J.
1987. Diatoms. The Illustrated Flora of Illinois. Southern Illinois
University Press, Carbondale, 478 pp.
Duever, M.J., Carlson, J.E., Meeder, J.F., Duever, L.C., Gun-
derson, L.H., Riopelle, L.A., Alexander, T.R., Myers,
R.L., and Spangler, D.P.
1985. The Big Cypress National Preserve. Research Report No.
8, National Audubon Society, New York, New York.
Dyke, A.S. and Prest, V.K.
1987. Late Wisconsin and Holocene history of the Laurentide
Ice Sheet. Geographie physique et Quaternaire, vol. 41,
pp. 237-263, and Geological Survey of Canada Maps
1702A and 1703A.
Faegri, K. and Iversen, J.
1975. Textbook of pollen analysis. Hafner Press, New York. 295
PPp-
Fairbanks, R.G.
1989. A 17,000-year glacio-eustatic sea level record: Influence
of glacial melting rates on the Younger Dryas event and
deep-ocean circulation. Nature, vol. 342, pp. 637-642.
Florin, M.-B.
1980. The taxonomy of some Melosira species: A comparative
morphological study. H. in 6th Diatom Symposium. R.
Ross, ed., pp. 43-73.
Foged, N.
1980. Diatoms in Oland, Sweden. Bibliotheca Phycologica, vol.
49. 193 pp.
1984. The diatom flora in springs in Jutland, Denmark (Springs
If). Bibliotheca Diatomologica, vol. 4, pp. 1-119.
Frey, D.G.
1964. Remains of animals in Quaternary lake and bog sediments
and their interpretation. Archiv fur Hydrobiologie Beih-
efte Ergebnisse der Limnologie, vol. 2, pp.1—114.
1982. Contrasting strategies of gamogenesis in northern and
southern populations of Cladocera. Ecology, vol. 63, pp.
223-241.
1986. Cladocera analysis. in Handbook of Holocene palaeoe-
cology and palaeohydrology. B.E. Berglund, and M. Ral-
ska-Jasiewiczowa, eds., John Wiley & Sons, New York,
pp. 667-692
Fritz, S.C.
1990. Twentieth-century salinity and water-level fluctuations in
Devils Lake: test of a diatom-based transfer function.
Limnology and Oceanography, vol. 35, pp. 1771-1781.
Frost, T.M.
1991. Porifera. in Ecology and classification of North American
freshwater Invertebrates. J.H. Thorp and A.P. Covich,
eds., Academic Press, New York, pp. 95-124.
Gasse, F.
1980. Les Diatomees lacustres Plio-pleistocenes du Gadeb
(Ethiopie). Revue Algologie Memoire hors serie 3, 249
Pp-
Gasse, F. Talling, J.F., and Kilham, P.
1983. Diatom assemblages in East Africa: classification, distri-
bution, and ecology. Revue Hydrobiologie Tropique, vol.
16, pp. 3-34.
Gasse, F. and Tekaia, F.
1983. Transfer functions for estimating paleoecological condi-
tions (pH) from East African diatoms. Hydrobiologia, vol.
103, pp. 85—90.
Germain, H.
1981. Flore des diatomees eaux douces et saumatres. Societe
Nouvelle des Editions Boubee, Paris. 441 pp.
Gleason, P.J. and Spackman, W.
1974, Calcareous periphyton and water chemistry in the Ever-
glades. in Environments of South Florida: Present and
Past. P.J. Gleason, ed., Miami Geological Society, Memoir
2, Miami, Florida, pp 146-181.
Gleason, P.J. and Stone, P.
1994. Age, origin, and landscape evolution of the Everglades
peatland. in Everglades: The Ecosystem and its Restora-
tion. S.M. Davis and J.C. Ogden, eds., St. Lucie Press,
Delray Beach, Florida, pp. 149-197.
Goodrick, R.L.
1984. The wet prairies of the northern Everglades. in Environ-
ments of South Florida: Present and Past II. PJ. Gleason,
ed., Miami Geological Society, Coral Gables, Florida, pp.
185-190.
Gray, J.
1965. Extraction techniques. in Handbook of Paleontological
Techniques. B. Kummel and D. Raup, eds., Freeman and
Co., New York, pp. 530-587.
Grimm, E.C., Jacobson, G.L., Watts, W.A., Hansen, B.C.S.,
Maasch, K.A.
1993. A 50,000-year record of climate oscillations from Florida
and its temporal correlation with the Heinrich events. Sci-
ence, vol. 261, pp. 198—200.
Gunderson, L.H. and Loftus, W.F.
1993. The Everglades. in Biodiversity of the Southeastern Unit-
ed States, Lowland Terrestrial Communities. W.H. Martin,
S.G. Boyce, and A.C. Echternacht eds., John Wiley &
Sons, Inc., New York, pp. 199-255.
Gunderson, L.H. and Snyder, J.R.
1994. Fire patterns in the southern Everglades. in Everglades:
The Ecosystem and its Restoration. $.M. Davis and J.C.
Ogden eds., St. Lucie Press, Delray Beach, Florida, pp.
291-305.
Hakansson, H.
1986a._ A study of species belonging to the Cyclotella bodanica/
comta complex (Bacillariophyceae). 9th Diatom Sympo-
sium, pp. 329-354.
HYDROLOGY, VEGETATION, AND CLIMATE: WINKLER ET AL. 95
1986b. A taxonomic reappraisal of some Stephanodiscus species
(Bacillariophyta). British Phycological Journal, vol. 21,
pp. 25-37.
Hakansson, H. and Locker, S.
1981. Stephanodiscus Ehrenberg 1846, a revision of the species
described by Ehrenberg. Nova Hedwigia, vol. 35, pp.
117-150.
Hakansson, H. and Stoermer, E.F.
1984. Observations on the type material of Stephanodiscus
hantzschii Grunow in Cleve & Grunow. Nova Hedwigia,
vol. 39, pp. 477-495.
Harrison, F.W.
1974. Sponges (Porifera: Spongillidae). in Pollution ecology of
freshwater invertebrates. C.W. Hart, Jr. and S.L.H. Fuller,
eds., Academic Press, New York, pp. 29—66.
1988. Utilization of freshwater sponges in paleolimnological
studies. Palaeogeography, Palaeoclimatology, Palaeoecol-
ogy, vol. 62, pp. 387-397.
Harrison, F.W., Gleason, P.J., and Stone, P.A.
1979. Paleolimnology of Lake Okeechobee, Florida: An analy-
sis utilizing spicular components of freshwater sponges
(Porifera: Spongillidae). Notulae Naturae, vol. 454, pp.
1-6.
Haworth, E.Y.
1983. A scanning electron microscope study of Cyclotella stel-
ligera and related taxa. in Rapport fra Diatomesymposium
i Bergen mai 1983. Botanical Museum Rapport, vol. 33,
pp. 15-25.
Hem, J.D.
1992. Study and interpretation of the chemical characteristics of
natural water. Third edition. U. S. Geological Survey Wa-
ter-Supply Paper 2254.
Hodell, D.A., Curtis, J-H., Jones, G.A., Higuera-Gundy, A.,
Brenner, M., Binford, M.W., Dorsey, K.T.
1991. Reconstruction of Caribbean climate change over the past
10,500 years. Nature, vol. 352, pp. 790-793.
Hofstetter, R.H.
1974. The effect of fire on the pineland and sawgrass commu-
nities of southern Florida. in Environments of South Flor-
ida: Present and Past. J.P. Gleason ed., Miami Geological
Society, Memoir 2, Coral Gables, Florida, pp. 201—212.
Huber-Pestalozzi, G.
1942. Das Phytoplankton des Susswassers. Die Binnengewasser,
vol. XVI, pp. 367-549.
Hunt, B.P.
1952. Food relationships between Florida spotted gar and other
organisms in the Tamiami Canal, Dade County, Florida.
Transactions of the American Fisheries Society, vol. 82,
pp. 13-33.
Hustedt, F.
1930. Die Kieslalgen, von Deutshlands, Osterreichs, und der
Schweiz. In L. Rabenhorst Kryptogamenflora, vol. 7, no.
1. Geest und Portig, Leipzig, Germany, pp. 1-920.
1939. Systematische und okologische Untersuchungen uber die
Diatomeen-Flora von Java, Bali, und Sumatra nach dem
Material der Deutschen Limnologischen Sunda-Expedi-
tion III. Die okologischen Factorin und ihr Einfluss auf
die Diatomeenflora. Archiv fur Hydrobiologie Supple-
menten, vol. 16, pp. 274-394.
1955. Marine littoral diatoms of Beaufort, North Carolina. Duke
University Press, Durham, North Carolina. 67 pp.
Hutchinson, G.E.
1975. Limnological botany. A treatise on limnology. Vol. III.
John Wiley & Sons, New York, 660 pp.
Jensen, J.E.
1994. Mercury Concentrations in Sediment and Fish Scales of
Gator Lake in the Everglades. Unpublished B.S. Senior
Thesis for the Center for Environmental Studies, Brown
University, Providence, R.I., 75 pp.
Klee, R. and Schmidt, R.
1987. Eutrophication of Mondsee (Upper Austria) as indicated
by the diatom stratigraphy of a sediment core. Diatom
Research, vol. 2, pp. 55—76.
Kling, H. and Hakansson, H.
1988. A light and electron microscope study of Cyclotella spe-
cies (Bacillariophyceae) from central and northern Cana-
dian Lakes. Diatom Research, vol. 3, pp. 55-82.
Koppen, J.D.
1975. A morphological and taxonomic consideration of Tabel-
laria (Bacillariophyceae) from the northcentral U. S. Jour-
nal of Phycology, vol. 11, pp. 236-244.
1978. Ecology of the diatom Tabellaria. The American Midland
Naturalist, vol. 99, pp. 383-397.
Krammer, K. and Lange-Bertalot, H.
1985. Naviculaceae. Bibliotheca Diatomologica, vol. 9, pp.
1-230.
1986. Bacillariophyceae: Band 2, Teil 1. Naviculaceae. in Sus-
swasserflora von Mitteleuropa. H. Ettl, et al., eds. Gustav
Fischer Verlag, Stuttgart, 876 pp.
1988. Bacillariophyceae: Band 2, Teil 2. Bacillariophyceae, Ep-
ithemiaceae, Surirellaceae. in Susswasserflora von Mitte-
leuropa. H. Ettl, et al., eds. Gustay Fischer Verlag, Stutt-
gart, 596 pp.
1991a. Bacillariophyceae: Band 2, Teil 3. Centrales, Fragilari-
aceae, Eunotiaceae. in Susswasserflora von Mitteleuropa.
H. Ettl, er al., eds. Gustav Fischer Verlag, Stuttgart, 576
Pp-
1991b. Bacillariophyceae: Band 2, Teil 4. Achnanthaceae, Kritis-
che Erganzungen zu Navicula (Lineolatae) und Gompho-
nema. in Susswasserflora von Mitteleuropa. H. Ettl, er al.,
eds. Gustav Fischer Verlag, Stuttgart, 437 pp.
Kropp, W.
1976. Geochronology of Corkscrew Swamp Sanctuary. /n Cy-
press Wetlands for Water Management, Recycling and
Conservation. 3rd Annual Report to NSF and Rockefeller
Foundation. H.T. Odum, K.C. Ewel, J.W. Ordway, and
M.K. Johnston, eds., Center for Wetlands, University of
Florida, Gainesville, pp. 772—785.
Kutzbach, J.E. and Street-Perrott, F.A.
1985. Milankovitch forcing of fluctuations in the level of trop-
ical lakes from 18 to 0 kyr BP. Nature, vol. 317, pp.130—
134.
Loftus, W. and Bass, O., Jr.
1992. Mercury threatens wildlife resources and human health in
Everglades National Park. Science, vol. 12, pp.18—20.
Loftus, W.F., Chapman, J.D., and Conrow, R.
1986. Hydroperiod effects on Everglades marsh food webs, with
relation to marsh restoration efforts. in Fisheries and
Coastal Wetlands Research. Vol. 6. G. Larson and M.
Soukup, eds., Proceedings of the Conference on Science
in the National Parks. Colorado State University, Fort
Collins, Colorado, pp. 1—22.
Long, R.W. and Lakela, O.
1976. A flora of tropical Florida. Banyan Books, Miami, Flori-
da, 962 pp.
Loveless, C.M.
1959a. The Everglades deer herd life history and management.
96 BULLETIN 361
Technical Bulletin No. 6. Florida Game and Fresh Water
Fish Commission.
1959b. A study of the vegetation in the Florida Everglades.
Ecology, vol. 40, pp. 1-9.
Nichols, D.J.
1974. Palynological analysis of Core 7 (Everglades peat). Re-
port to the Central and Southern Florida Flood Control
District, 29 pp.
Olmstead, I. and Loope, L.L.
1984. Plant communities of Everglades National Park. in En-
vironments of South Florida: Present and Past II. PJ.
Gleason, ed., Miami Geological Society, Coral Gables,
Florida, pp. 167—184.
Patrick, R.
1966. The diatoms. in The History of Laguna de Petenxil, A
Small Lake in Northern Guatemala. U.M. Cowgill, et al.,
eds., Memoirs of the Connecticut Academy of Arts &
Sciences, vol. XVII, pp. 74-77.
1977. Ecology of freshwater diatoms-diatom communities. in
The Biology of Diatoms. D. Werner, ed., Botanical Mono-
graphs, Vol. 13, Blackwell Scientific Publications, Ox-
ford, England, pp. 284—332.
Patrick, R. and C.W. Reimer
1966. The diatoms of the U. S. Academy of Natural Sciences of
Philadelphia Monograph 13, vol. I, 688 pp.
1975. The diatoms of the U.S. Academy of Natural Sciences of
Philadelphia Monograph 13, vol. IH, no.1., 213 pp.
Pennak, R.W.
1978. Freshwater invertebrates of the United States. 2nd Edi-
tion. Wiley-Interscience, New York, 803 pp.
Pergallo, H. and Pergallo, M.
1965. Diatomees Marines de France. Atlas. A. Asher & Co.,
Amsterdam.
Poirrier, M.A.
1974. Ecomorphic variation in gemmoscleres of Ephydatia flu-
viatilis Linneaus (Porifera: Spongillidae) with comments
upon its systematics and ecology. Hydrobiologia, vol. 44,
pp. 337-347.
1976. A taxonomic study of the Spongilla alba, S. cenota, S.
wagneri species group (Porifera: Spongillidae) with eco-
logical observations of S. alba. in Aspects of Sponge Bi-
ology. EW. Harrison and R.R. Cowden, eds., Academic
Press, New York, pp. 203-214.
Racek, A.A.
1966. Spicular remains of freshwater sponges. in The History
of Laguna de Petenxil, A Small Lake in Northern Gua-
temala. U.M. Cowgill et al., eds., Memoirs of the Con-
necticut Academy of Arts & Sciences, vol. XVII, pp. 78—
83.
Rader, R.B. and Richardson, C.J.
1992. The effects of nutrient enrichment on algae and macroin-
vertebrates in the Everglades. Wetlands, vol. 12, pp. 121—
135.
Rasanen, M. and Tolonen, K.
1983. Changes in diatom flora deposited annually during the
freshening of the impounded, sea-bay of Gennarbyviken
in southern Finland. Hydrobiologia, vol. 103, pp. 147—
52)
Riegel, W.L.
1965. Palynology of environments of peat formation in South-
western Florida. Unpublished Ph.D. thesis, The Pennsyl-
vania State University. 190 pp.
Robertson, W.B.
1953. A survey of the effects of fire in Everglades National
Park. U.S. National Park Service Report. 169 pp.
Ruddiman, W.F. and Mix, A.C.
1993. The North and Equatorial Atlantic at 9000 and 6000 yr B. P.
in Global Climates since the Last Glacial Maximum. H.E.
Wright, Jr., J.E. Kutzbach, T. Webb II, W.E Ruddiman, FA.
Street-Perrott, and PJ. Bartlein, eds., University of Minne-
sota Press, Minneapolis, Minnesota, pp. 94—124.
Sandweiss, D.H., Richardson, J.B. III, Reitz, E.J., Rollins, H.B.,
Maasch, K.A.
1996. Geoarchaelogical evidence from Peru for a 5000 years
B. P. onset of El Nino. Science, vol. 273, pp. 1531-1533.
Scherer, R.P.
1988. Freshwater diatom assemblages and ecology/paleoecolo-
gy of the Okefenokee swamp/marsh complex, southern
Georgia, USA. Diatom Research, vol. 3, pp. 129-157.
Schulte, E.E., Peters, J.B., and Hodgson, P.R., editors
1987. Wisconsin Procedures for Soil Testing, Plant Analysis and
Feed and Forage Analysis. #6 Soil Fertility Series. Dept.
of Soil Science Publication. UW—Madison, 65 pp.
Shih, S.F.
1980. Impact of subsidence on water management in Everglades
Agricultural Area. in Proceedings of the 6th International
Peat Congress. International Peat Society Publication, pp.
473-479.
Smol, J.P., Battarbee, R.W., Davis, R.B., and Merilainen, J., ed-
itors
1986. Diatoms and lake acidity. Dr. W. Junk, Publisher, Dor-
drecht, The Netherlands.
Snyder, G.H. and Davidson, J.M.
1994. Everglades agriculture: past, present, and future. in Ey-
erglades: The Ecosystem and its Restoration. S.M. Davis
and J.C. Ogden, eds., St. Lucie Press, Delray Beach, Flor-
ida, pp 85-115.
Steward, K.K. and Ornes, W.H.
1975. The autecology of sawgrass in the Florida Everglades.
Ecology, vol. 56, pp. 162-171.
Stoermer, E.F. and Hakansson, H.
1984. Stephanodiscus parvus: validation of an enigmatic and
widely misconstrued taxon. Nova Hedwigia, vol. 39, pp.
497-511.
Sweets, P.R., Garren, R.A., and R.G. Kreis, Jr.
1989. The relationship between surface sediment diatoms and
pH in Northern Florida. in Paleoecological investigation
of recent lake acidification (PIRLA): 1983-1985. D.E
Charles and D.R. Whitehead, eds. EPRI Technical Report,
Project RP2174—10, Indiana University Foundation, pp.
13-1 to 13-17.
Sweets, P.R., Bienert, R.W., Crisman, T.L., and Binford, M.W.
1990. Paleoecological investigations of recent lake acidification
in Northern Florida. Journal of Paleolimnology, vol. 4,
pp. 103-133.
Swift, D.R.
1984. Periphyton and water quality relationships in the Ever-
glades Water Conservation Areas. in Environments of
South Florida: Present and Past I. PJ. Gleason, ed., Mi-
ami Geological Society, Coral Gables, Florida, pp. 97—
117.
Swift, D.R. and Nicholas, R.B.
1987. Periphyton and water quality relationships in the Ever-
glades Water Conservation Areas, 1978-1982, Tech. Publ.
87-2, South Florida Water Management District, West
Palm Beach, Florida. 44 pp.
Thompson, S.L. and Pollard, D.
1995. A Global Climate Model (GENESIS) with a Land-Surface
HYDROLOGY, VEGETATION,
Transfer Scheme (LSX). Part I: Present Climate Simula-
tion. Journal of Climate, vol. 8, pp. 732-761.
Tryon, R.M. and Tryon, A.F.
1982. Ferns and allied plants with special reference to tropical
America. Springer-Verlag, New York.
Van Meter, N.
1965. Some quantitative and qualitative aspects of periphyton
in the Everglades. Unpublished M.S. thesis, University of
Miami, Florida.
Vinebrooke, R.D. and Leavitt, P.K.
1999. Differential responses of littoral communities to ultravi-
olet radiation in an alpine lake. Ecology, vol. 1, pp. 223—
237.
Wanless, H.R., Parkinson, R.W., and Tedesco, L.P.
1994. Sea level control on stability of Everglades wetlands. in
Everglades: The Ecosystem and its Restoration. S.M. Da-
vis and J.C. Ogden, eds., St. Lucie Press, Delray Beach,
Florida, pp. 199-223.
Watts, W.A. and Hansen, B.C.S.
1988. Environments of Florida in the Late Wisconsin and Ho-
locene. in Wet Site Archaeology, B. Purdy, ed., The Tel-
ford Press, Caldwell, N.J., pp. 307-323.
Whitmore, T.J.
1989. Florida diatom assemblages as indicators of trophic state
and pH. Limnology and Oceanography, vol. 34, pp. 882—
895.
Winkler, M.G.
1985. Charcoal analysis for paleoenvironmental interpretation:
A chemical assay. Quaternary Research, vol. 23, pp. 313—
326.
AND CLIMATE: WINKLER ET AL. 97
1988. Paleolimnology of a Cape Cod kettle pond: diatoms and
reconstructed pH. Ecological Monographs, vol. 58,
pp. 197-214.
1994a. Development of the Gull Pond Chain of Lakes and the
Herring River Basin, Cape Cod National Seashore. Tech-
nical Report NPS/NAROSS/NRTR/95-27. National Park
Service, North Atlantic Region, Natural Resources Man-
agement and Research, Boston, MA. 239 pp.
1994b. Sensing plant community and climate change by charcoal-
carbon. Ecoscience, vol.1, pp. 340-345.
Winkler, M.G., Sanford, P.R. and Kaplan, S.
1996a. Paleoecology of the Everglades National Park. Final Re-
port to the National Park Service, Everglades National
Park, Homestead, Florida, to be published as South Flor-
ida Natural Resources Center (SFNRC) Technical Report
97-006, 249 pp.
Winkler, M.G., Kaplan, S.W., Coe, M. and Americas Lake-Level
Database Working Group Members.
1996b. Americas lake-level database. Program and Abstracts of
the 14th Biennial Meeting of the American Quaternary
Association. Flagstaff, Arizona, 191 pp.
Winkler, M.G., Sanford, P.R.,and Kaplan, S.
1999. Southern Everglades and the Holocene Climatic History
of Florida and the Caribbean. (in preparation).
Wood, F. and N.G. Maynard.
1974. Ecology of the micro-algae of the Florida Everglades. in
Environments of South Florida: Present and Past. PJ.
Gleason, ed., Miami Geological Society, Memoir 2. Mi-
ami, Florida, pp. 123-145.
98 BULLETIN 361
EXPLANATION OF PLATE |
Figure
Page
1, 2. Aulacoseira, possible new diatom species, Gator Lake, 471 and 571 cm, 13 * 5—6 pm, with parallel rows of very fine pores (almost
unresolvabletat000> mapnification) snoteralsoylongyspinesiesn-iiensnemeiei- n-ne eee nee nent ee ieee 77
35 Unidentified/sponge megasclere;, A065 14—20)icem> 266) imi longs =o erencneye) teil) ecient pee seems ere citsncnen antennae ee 90
4) Unidentified!sponseigemmoscleres-A06;)20—25icm> 7 Ohpumlon ge) senate one teenenene ieee cuen stanton cliente eee 91
5. Cladium sclereid piece, A06, 0-7 cm, longest dimension 333 jm, triangular openings ca. 23 wm. ...........--...----. 84
6
* INymphaceae!sclereid,A23>33—4 lem 600—G67.O)uml long. iss cia = estes eucre cieneae cetonsteaeeits etree ariel eer eae 84
99
HYDROLOGY, VEGETATION, AND CLIMATE: WINKLER ET AL.
GHAPTER 6
A POLLEN ZONATION OF SOUTHWESTERN FLORIDA USING MULTIVARIATE STATISTICAL METH-
ODS AND ITS APPLICATION TO TWO VERTICAL SEDIMENTARY SEQUENCES
M. A. O’NEAL!, L. P. TEDESco!, C. SOUCH?, AND J. EF PAcHutT!
Departments of Geology! and Geography’,
Center for Earth and Environmental Science, Indiana University-Purdue University,
Indianapolis, IN 46202
ABSTRACT
Extensive and reliable modern analogs are the key to palynological assessments of habitat change. This study uses multivariate
statistical techniques to reevaluate surface pollen distributions within Everglades National Park, South Florida. Results indicate
that multivariate techniques provide a sensitive and informative measure of variations in the distribution of pollen and spores.
The distribution of pollen and spores in surface sediments of southwest Florida were grouped into five distinct clusters, or pollen
zones, containing a total of 10 sub-clusters that help delineate subtle differences within each zone. These pollen zones are: 1)
mangrove swamp, 2) brackish marsh, 3) freshwater marsh, 4) headwater marsh, and 5) upland complex. The spatial distribution
of these pollen zones strongly corresponds to the major physiographic provinces of the area. Cluster analysis indicates that
individual taxa typically are not environmentally restricted and, in most environments, there is no single indicator species because
of overlapping ranges. The results generated from the analysis of surface assemblages are applied to two vertical sedimentary
sequences to interpret past environmental changes. Results show that these methods can be used to identify broad environmental
changes (i.e., the overall late-Holocene transgression of South Florida) and can differentiate among different mangrove swamps
where time integration of in situ samples has occurred. The method is limited in its ability to differentiate short-lived changes
in coastal configuration (i.e., sea-level oscillations) as long as the overall environment remains in close proximity to the coastal
zone. In these cases, detailed sedimentologic work, combined with palynological analysis is required, especially where sediment
reworking has occurred.
INTRODUCTION
Pollen and spores incorporated into the sediment re-
cord of the coastal systems of South Florida can pro-
vide detailed information about environments of de-
position. However, the variety of pollen and spore
types present in most samples makes it difficult to
compare and group assemblages by visual inspection
of frequency histograms. Multivariate statistical meth-
ods provide more robust techniques for comparing and
grouping surface pollen sites, and also permit the si-
multaneous investigation of multiple sample sites con-
taining several taxa. Such techniques have been used
successfully for ecological interpretations at the re-
gional scale elsewhere: see, for example, Birks et al.
(1975). To date, however, such methods have not been
applied in South Florida, even though the vegetation
communities have a distinct spatial distribution that is
strongly related to salinity, hydroperiod, elevation, and
substrate, and are likely to be amenable to such anal-
yses. Moreover, given the considerable attention that
has been, and is now being, directed towards the use
of palynological data in interpreting the paleoecology
of vertical sediment sequences in the area (Riegel,
1965; Spackman et al., 1966; Kuehn, 1980; Bartow et
al., 1996; Meeder et al., 1996; Brewster-Wingard et
al., 1997; Willard and Holmes, 1997; Winkler et al.,
1998), such methods have great potential to associate
vertical sequences of assemblages with surface pollen
and spore groups in order to interpret paleoenviron-
mental changes through time.
The purpose of this study is two-fold. First, to de-
lineate surface pollen zones in Everglades National
Park, South Florida, using multivariate statistical
methods. The objectives here are to (1) provide an
objective model for interpreting the distribution of pol-
len and spores in surface sediments, (2) demonstrate
the sensitivities of cluster and discriminant analysis in
delineating surface pollen clusters, and (3) provide a
framework for paleoecological reconstructions in
South Florida. Then second, to use the objective pollen
zonation to analyze and interpret two vertical sedi-
mentary sequences, one from the Harney River, the
other from Gopher Key, in terms of late Holocene en-
vironmental conditions. With these objectives, the pa-
per is structured in two parts: pollen data for contem-
porary sediments and their analysis are presented first;
details of the vertical sequences, pollen counts, and
their analysis, second.
102 BULLETIN 361
ore 59-
(©) Core 59-T1 —
© Core 9411-7
~
Harney Rive O7 = ‘
GULF OF
MEXICO
Rod MANGROVE SWAMP
Es BRACKISH TO
TRANSITIONAL MARSH
FRESHWATER MARSH
WITH HAMMOCKS
FRESHWATER MARSH - SLOUGH
[4 umestone upLAND COMPLEX
SHALLOW MARINE MUDBANK
KILOMETERS
Text-figure 1—Major physiographic provinces of the South Florida peninsula (modified from Wanless er al., 1989). Cores discussed in Part
B are located.
ACKNOWLEDGMENTS
This research was supported by a grant from
NOAA’s Coastal Ocean Program and the South Florida
Ecosystem Restoration and Monitoring Program. Nu-
merous people have improved this manuscript through
thought-provoking discussions and comments. In par-
ticular we would like to thank H. R. Wanless (Uni-
versity of Miami), T. A. Nelsen (NOAA-AOML), A.
D. Cohen, and FE J. Rich. D. W. Kuehn (University of
Western Kentucky), P. G. Mueller and the Illinois State
Museum generously provided reference sets of tropical
pollen. Thanks to Everglades National Park, T. Ar-
mentano, and W. Meshaka for permission to sample
the vouchered herbarium at ENP and help in doing so.
A. Vlad volunteered her time to help sample the her-
barium. C. J. Ross helped computerize the reference
set and the original data set. M. A. Capps helped es-
tablish laboratory procedures for pollen preparation.
STUDY AREA
South Florida is part of the broad, Florida platform
(Parker and Cooke, 1944). Although presently attached
to the North American continent, the South Florida
peninsula is a partly inundated carbonate platform re-
moved from most continental influences (Wanless et
al., 1989). Sea level has flooded the platform resulting
in an expansive shelf margin to the west and a narrow,
steep-sided margin to the south and east. The upper-
most rock units in the study area are of Tertiary and
Quaternary age (Puri and Vernon, 1959). The north-
west portion of the study area is underlain by the Pli-
ocene Tamiami Formation, a calcareous sandstone to
sandy limestone (Parker and Cooke, 1944). The Big
Cypress Ridge area is underlain by shallow reefal
limestones of Pliocene age (Meeder, 1987). The re-
maining portions of the study area are underlain by the
Pleistocene Miami Oolite (Parker et al., 1955; Hoff-
meister ef al., 1967). Unconsolidated Holocene sedi-
ments blanket much of the bedrock surface. Holocene
carbonate, organic and siliceous sediments of biogenic
origin are forming and accumulating, and quartzose
sands are being supplied by littoral transport and by
dissolution of limestones (Wanless ef al., 1989). The
gentle southerly slope of the bedrock surface results in
an increase in the thickness of unconsolidated Holo-
cene sediments seaward. Local differences in the
thickness and distribution of Holocene sediments can
be attributed, in part, to hydrologic and topographic
variability in which the preexisting limestone surface
plays a fundamental role (Wanless et al., 1995).
PHYSIOGRAPHY
The distribution of pre-Holocene bedrock highs and
thick accumulations of Holocene sediments define the
major physiographic provinces of the South Florida
peninsula (Text-fig. 1). The Everglades depression ex-
tends southward through the center of South Florida
and is bounded by exposed Pliocene limestones of the
Big Cypress Ridge to the west and the Pleistocene
quartz sand and oolitic limestone of the Atlantic Coast-
POLLEN ZONATION: O’ NEAL ET AL. 103
al Ridge to the east (Text-fig. 1). The broad Everglades
depression carries fresh water flow southward from
Lake Okeechobee to the sea via the Shark River
Slough complex (Text-fig. 1). The Everglades depres-
sion is underlain by Pleistocene lagoonal deposits
termed the bryozoan facies of the Miami Limestone
by Hoffmeister et al. (1967). The depression is choked
with Holocene freshwater peat and calcitic mud de-
posits (Wanless et al., 1989). A natural coastal dam of
mangrove peat and storm-levee marl inhibits saline in-
trusion into the depression. The oolitic limestone ridge
of the Miami Limestone crests 2—6 m above sea level,
extends north-south from north Miami to the south,
where it then curves westward. The Atlantic Coastal
Ridge is cross-cut by numerous fossil tidal channels.
These channels form swales that are filled with quartz
sand and peat and served as conduits for fresh water
flow through the ridge prior to drawdown of Ever-
glades water levels in the early 1900’s (Wanless et al.,
1989).
In addition to the physiographic units defined by
Pliocene and Pleistocene ridges, Holocene sediment
buildups generate coastlines and marine buildups
which define the coastal complex. Along the northern
shore of Florida Bay, transgressive to regressive coast-
al storm levees of carbonate mud form the shoreline.
Broad progradational peat and coastal sand and mud
accumulations form a coastal mangrove forest and
transitional marsh complex along the western shoreline
(Wanless et al., 1995). The study area is characterized
by elevations that are generally less than 0.3 m above
mean sea level (MSL) (Spackman et al., 1966). In the
area south of Lake Okeechobee, elevations are as
much as 10 m above MSL and gently slope to the
south (Spackman et al., 1966). Changes in topography,
although subtle, are significant in that natural highs
and lows strongly influence the depth of the water ta-
ble, thickness of Holocene sediment, and the surface
flow of water in a given area (Riegel, 1965). The pri-
mary inputs of surface water to the study area are
overflow from Lake Okeechobee and local rainfall. In
the approximately 4000 sq. miles (6,400 km7?) of the
Everglades depression, water moves as sheet flow
(Spackman et al., 1966), although historically, human
channelization of flow has disrupted the natural flow
system to varying degrees (Mclvor et al., 1994; Smith
et al., 1989).
VEGETATION COMMUNITIES
South Florida has a very distinct zonation of vege-
tation that is strongly related to salinity, hydroperiod,
elevation, and substrate (Riegel, 1965; Text-fig. 2).
The boundaries between zones are usually abrupt (Co-
hen, 1968) except in estuaries. Here, transitional com-
» 84-€7 (SO+E): /)
» 64-10
"s 54-16 (SG+E) AR HOR
Al wh
; wy
# 64:1680(SG¥E), 1151/11
v\ ay uJ
20 kilometers
Mangrove Swamp
(J Brackish Marsh
(°] Freshwater Marsh with Cypress Hammocks
EJ Freshwater Marsh (Slough)
|} Freshwater Marsh with Hardwood Hammocks
E] Beach Community -
(i Flamingo Marl Prairie oe
Bi Pine - Upland eee an
(©) Hardwood Forest - Upland AREA | \,
| Big Cypress Swamp - Upland
d
Text-figure 2.—Generalized vegetation map of southwest Florida
showing the location of samples. SR samples are along the Shark
River Slough complex. The 86 sample sites used for statistical anal-
ysis are shown (modified from Riegel, 1965).
munities occur where tidal channels and river systems
compete for dominance of influence in a given area.
Fires, hurricanes, droughts, floods, and freezing tem-
peratures also control the distribution and extent of
vegetation communities; however, the long-term ef-
fects of such forces are not well understood (Olmstead
and Loope, 1984; Gunderson and Snyder, 1992). Veg-
etation communities in southwestern Florida, particu-
larly in Everglades National Park, vary in complexity.
Some communities are dominated by a few species,
while others are composed of more complex mixtures
with no truly dominant species. The vegetation com-
munities used in this study were described by Riegel
(1965) based on the work of Davis (1943) and Love-
less (1959). Gunderson (1994) provided a detailed de-
scription of freshwater marsh and tree island commu-
104 BULLETIN 361
nities, as well as upland vegetation. The spatial distri-
bution of communities is shown in Text-figure 2.
Coastal Mangrove Fringe (Mangrove Swamp)
Mangrove forests of varying heights dominate the
coastal margin of the study area. These forests are es-
sentially a continuous cover of woody vegetation that
is tolerant of high salinity and changes in hydroperiod
related to inundation by tides. The forests produce root
peat deposits that may develop into peat islands or
shore-parallel wedges that are dissected by river chan-
nels draining the Everglades Basin. The interior of
these islands is commonly higher in elevation and de-
velops into a saltwater marsh or coastal prairie sup-
porting less water-tolerant woody species (Bancroft et
al., 1994). Rhizophora (red mangrove), Avicennia
(black mangrove) and Laguncularia (white mangrove)
dominate the mangrove forest. Light gaps permit the
growth of Batis (saltwort). The coastal levees of the
mangrove swamp are capable of supporting a variety
of upland vegetation. Prior to devastation by Hurricane
Donna (1960), Avicennia forests were prevalent in the
areas around Flamingo and the lower Shark River
Slough (Craighead and Gilbert, 1962). These regrowth
forests are now primarily composed of Rhizophora.
Brackish and Headwater Marsh
A brackish zone develops parallel to the coast where
fresh water runoff from the mainland mixes with ma-
rine waters of the Gulf and Florida Bay. The location
and extent of this brackish transitional zone varies with
tidal influences and seasonal changes in the amount of
freshwater input. Physiographically, the vegetation
communities of southwestern Florida are parallel to the
coast consistent with the salinity gradient (Text-fig. 2).
Two brackish transitional zones can be distinguished.
One occurs parallel to the coast and away from the
major freshwater outflow of the Shark River Slough.
This brackish, transitional marsh is dominated by Rhi-
zophora, Avicennia, Laguncularia, Typha (cattail),
Spartina (eelgrass), and Cladium (sawgrass). A second
transitional zone, the headwater marsh, occurs adjacent
to the outflow of the Shark River Slough in the zone
of fluctuating freshwater and marine influences. This
area is dominated by a network of river channels of
the Shark River Slough/Little Shark River complex.
Vegetation in the area is a mixture of Rhizophora, Av-
icennia, Laguncularia, Conocarpus (buttonwood),
Spartina, Scirpus (bulrush), Typha, Cladium, and Myr-
ica (wax myrtle) (Bancroft et al., 1994).
Freshwater Marsh and Slough Environments
Inland from the coastal bays, waters become fresher
and mangrove communities diminish. These more in-
land vegetation communities can once again be divid-
ed based on species tolerance to fluctuations in salinity
and hydroperiod. The freshwater areas can be divided
into freshwater marsh—slough (the primary vegeta-
tion-choked waterways developed in the Everglades
Basin) and freshwater marsh with hammocks (tree is-
lands). Sloughs and marshes are prairies of grasses and
sedges that comprise the largest percentage of the
freshwater area. Cladium, Eleocharis (spikerush), Pan-
icum (maidencane), Muhlenbergia (Muhly grass) and
Rhynchospora (beakrush) are the primary sedges and
grasses. In the wetter slough areas, emergent aquatics
also are abundant including Sagittaria (arrowhead),
Nymphaea (water lily) and Utricularia (bladderwort)
(Gunderson, 1994). In the freshwater marsh with ham-
mocks, tree islands or hammocks commonly develop.
Communities can be divided into Bayheads, Willow
Heads, and Cypress Forests (Gunderson, 1994) and in-
clude a diverse mixture of Persea (red bay), [lex (da-
hoon holly), Anona (pond apple), Myrica (wax myr-
tle), Salix (willow), Chrysobalanus (cocoplum), and
Taxodium (bald cypress), in addition to numerous oth-
er species that occur in low abundances.
Upland Complex
Toward the rim of the Everglades depression are
bedrock highs on which pinelands and hardwood for-
ests dominate. Pine uplands are found along the east-
ern margin of the Everglades on the Atlantic Coastal
Ridge and extend into Everglades National Park along
limestone ridges (Gunderson, 1994). Rockland pine
forests are dominated by monotypic stands of Pinus
elliotti (slash pine). Additionally, an assortment of
palms, including Sabal (cabbage palm) and Serenoa
(saw palmetto) are common.
Tropical hardwood hammocks occur on rock sub-
strates most commonly in the southern Everglades.
Quercus (live oak), Lysiloma (wild tamarind), Bursera
(gumbo-limbo), and Swietenia (mahogany) are com-
mon hammock species. In many areas, dense ham-
mocks primarily comprised of the bald cypress Taxo-
dium distichum have developed. Orchids, bromeliads,
and ferns are also present, but are not abundant.
PART A: CHARACTERIZATION OF MODERN
POLLEN ZONES IN SOUTH FLORIDA FROM
SURFACE POLLEN SAMPLES
Data for the first part of this study, the delineation
of contemporary pollen zones in South Florida, was
obtained from Appendix II of Riegel (1965) who com-
pleted the first extensive palynological analysis of Ey-
erglades National Park. His objectives were effectively
the same as ours: to determine the relationship be-
tween pollen from surface sediments and the local
POLLEN ZONATION: O’ NEAL ET AL. 105
plant communities of the same area, and to character-
ize late Holocene ecological changes by comparing
pollen preserved in sedimentary sequences with those
found in the surface sediments. However, Riegel’s
(1965) methods differ from ours. His data were pre-
sented in a series of tables and abundance histograms
designed to display the distribution of pollen and
spores both at the surface and in sediment cores. His
analysis focused on the means and variances of indi-
cator species and their abundance. Results indicated,
as expected, that the content of pollen assemblages
was not directly proportional to the species composi-
tion of the parent plant communities; however, depo-
sitional environments could be recognized by pollen
types that were transport-limited, and from vegetation
communities restricted by salinity, hydroperiod, and
elevation.
The original data set is a matrix consisting of the
percent frequency of 92 taxonomic and morphologic
pollen and spore categories collected from 102 sample
sites, representing six of the ten major vegetation com-
munities presented in that report (Text-fig. 2). Riegel
(1965) followed standard preparation procedures and
counting was conducted on at least three slides per
sample. For each site, Riegel (1965) attempted to
count a minimum of 150 grains. Pollen was not abun-
dant in some samples and, on occasion, counting was
stopped after reaching a minimum of 100 grains. Sev-
eral of Riegel’s (1965) samples taken from detrital sed-
iments were removed from the data set for this study
in order to restrict the analysis to samples from in situ
depositional localities. The 92 taxonomic and morpho-
logic categories in the original data set were reduced
to 47 categories that could be associated with known
parent plants present in surface vegetation communi-
ties. Twelve samples were collected from the Joe River
area along a transect less than 300 m long extending
from the mangrove fringe into the transitional marsh
of northwest Cape Sable. This represents a much great-
er density of sites than for any of the other environ-
ments sampled by Riegel (1965) in South Florida. In
order to avoid biasing the present data set, only one
sample, JR-1, was selected from the Joe River transect.
The final data set consisted of 47 taxonomic and mor-
phologic pollen and spore categories and 86 surface
sampling sites (Text-fig. 2).
Statistical Methods
Two multivariate statistical methods, cluster analysis
and discriminant analysis, were applied to the data.
Cluster analysis is a technique that classifies samples
or taxa into relatively homogenous groups based on
inter-object similarities (Davis, 1986; Kachigan, 1991).
Two separate cluster analyses were used; one for as-
sociations of pollen and spores (taxonomic) and the
other for surface sample associations (environmental).
A hierarchical clustering technique, using Ward’s sum
of squares linkage method (Ward, 1963) and Chi-
squared distances, were used in each analysis. Ward’s
method, which sums the squared Euclidean distances
between each sample and the cluster means, is the pre-
ferred linkage technique because at each stage in the
cluster analysis two samples or clusters are merged in
a manner that arrives at the lowest within-group sum
of square totals (Davis, 1986; Kachigan, 1991). All
statistical calculations presented in this study were per-
formed using SYSTAT version 6.0 (SPSS Inc., 1996).
Cluster analysis produces a dendrogram with objects
(i.e., samples or taxa) on the vertical axis and a dis-
tance measurement that defines where clusters join on
the horizontal axis. Any vertical slice through the den-
drogram should produce distinct groups that have
maximized inter-object similarities. The statistical va-
lidity of the groups obtained from cluster analysis of
surface samples was tested using discriminant analysis.
Discriminant analysis evaluates the distinctiveness of
predefined groups by calculating discriminant func-
tions representing a linear combination of measured
variables (Davis, 1986). Likewise, discriminant func-
tion scores are calculated for each sample by trans-
forming the original set of measurements into a single
value on each function. The mean, variance, and con-
fidence intervals for function scores were calculated
for each group and used to evaluate the distinctiveness
of cluster-generated groupings. Function coefficients
are calculated for each variable in the data set (i.e.,
taxa) permitting the identification of variables that
most strongly contribute to group separation along
each function.
Results for Surface Samples
Cluster Analysis of Taxa
The dendrogram from cluster analysis of the 47 pol-
len and spore categories can be divided into four clus-
ter groups (Text-fig. 3). Cluster A is further divided
into three sub-clusters. Cluster Al contains Rhus,
Sphagnum, Caryophyllaceae, Fraxinus, Eugenia, Car-
ya, Rhabdadenia, Ericaceae, Hamelia, Ulmus, Pteris,
Rapenea, cf. Thomsonopollis, Nyssa, Myrtaceae, Nu-
phar, Ficus, Polygona, Morus, and Alnus. Cluster A2
contains Sapotaceae, Vitis, Ilex, Anacardium, and Os-
munda. Cluster A3 contains Compositae, Quercus, Ty-
pha, Gramineae, Salix, Cephalanthus, Polypodium,
and Taxodium. Cluster B contains chenopods. Cluster
C contains triporates, Cyperaceae, Sagittaria, Utricu-
laria, Umbelliferae, cf. Ovoidites, Nymphaea, and Pi-
106 BULLETIN 361
Table 1.—Southwestern Florida surface pollen and spore data arranged by environmental cluster groupings. Data from Riegel (1965) are
percentage abundance of the original sample. All samples do not total 100% because of the exclusion of taxonomic/morphologic categories
that could not be associated with known parent vegetation. Average (Avg.), standard deviation (st. dev.), and 95% confidence intervals (CI)
are presented for each cluster group.
Lagun- Cheno-
Site Cluster = Rhizophora Avicennia cularia Conocarpus _ Batis pods Gramineae Cyperaceae
Mangrove Swamp (N = 10)
62-01 1 58.7 7.3 0.7 3, 0.0 7.4 0.0 0.7
62-06 1 Si/ 5.4 0.0 4.2 1.8 7.8 ile? 0.6
62-08 1 64.0 3.1 1.2 Bh 3.1 6.1 0.0 0.0
62-11 1 63.5 0.0 0.6 1.8 2.4 6.0 0.0 0.0
62-12 1 66.1 0.0 DS 0.0 0.6 4.3 0.6 0.6
62-15 1 58.7 0.6 1.2 4.3 0.6 11.8 0.6 1.2
62-17 1 61.0 1.8 1e2 0.6 0.6 7.9 0.0 0.6
62-20 1 56.0 33 DD, 2.2 1.6 4.9 0.0 eit
62-21 1 43.7 7.6 PPS) 6.3 P25) 8.3 0.0 1.3
62-22 1 46.9 Ue 1.7 0.6 ed 2.8 0.6 67/
Avg. 57.4 3.0 1. 3.0 5 6.7 0.3 0.8
Stdev. 73 2.9 0.8 2.5 1.0 2.5 0.4 0.5
Cl 4.5 1.8 0.5 1.5 0.6 1.6 0.3 0.3
Mangrove Swamp (N = 15)
62-02 2 16.6 9.6 0.0 Se) 29.6 25.8 1.0 0.0
62-03 2 29.4 2) 0.5 S5f/ 10.8 28.4 2.1 0.0
62-05 2 Dilee 9.5 1.2 4.7 si 21.9 1.2 0.0
62-07 2 34.3 1.2 1.2 5.4 4.2 18.1 2.4 0.0
62-09 2 38.9 Spl 0.6 Su7/ 3.8 10.2 3} 1153}
62-10 2 43.5 2.0 1.3 2.0 0.0 11.7 1.3 1.3
62-13 2 37/3) 33 1.3 3:9 1.3 hil 0.7 0.0
62-14 2 39.6 45 1.1 2.3 177 10.8 ileal 1
62-18 2 49.0 1:9 3.9 32 2.6 8.5 0.0 0.7
62-19 2 25.1 6.7 1.7 1.7 0.0 10.7 1.1 tl
64-11 2 11.7 0.6 SE 1.8 0.0 12.2 19) 4.3
64-21 2 19.8 IEG 16.4 11.8 0.0 519) 0.9 0.0
SR-O1 2 37.5 ils) 1.3 3.3 319 8.5 0.0 0.7
SR-02 2 32.3 6.2 12. 2.3 3.5 IS\-2? 0.0 0.0
59-T1 2 30.7 6.1 8.9 2.4 0.0 9.2 1.2 0.0
Avg 31.5 4.3 3.0 4.1 4.6 14.0 1.0 0.7
Stdev 10.2 2.9 4.3 2.6 7.6 6.7 0.7 1.1
CI 5:2 1.5 2.2 1.3 3.8 3.4 0.4 0.6
Mangrove Swamp (N = 11)
SR-03 3 44.4 Dei, 0.9 14.2 1.3 ile 0.0 0.0
SR-04 3 40.9 3i2 1:9 11.1 19 6.7 0.0 0.9
SR-05 3} 32.6 0.0 1.1 4.9 0.0 23.9 0.0 0.5
SR-06 3 33.5 ier 3 6.5 0.0 79) 0.3 1.4
SR-07 3} 3257 0.3 0.6 8.0 0.3 95 0.0 0.6
SR-09 3 52.4 2.9 1.3 6.8 0.3 6.5 0.0 0.7
SR-10 3 36.2 0.9 1.9 S57] 0.0 18.1 0.0 0.0
SR-13 3 13.6 0.0 1.1 8.2 1.1 315)-3) 0.0 0.5
SR-14 3 13.5 2.4 0.6 1229 0.0 21.2 0.6 2.4
SR-15 3 36.9 0.0 1.9 25 0.0 18.1 0.6 0.6
SR-16 3 29.1 0.0 0.7 iTh,3) 0.0 8.6 0.7 2.0
Avg 33.3 1.2 1.1 8.4 0.4 16.1 0.2 0.9
Stdev 11.7 1.3 0.5 3.6 0.7 8.8 0.3 0.8
CI 6.9 0.8 0.3 2.1 0.4 $.2 0.2 0.5
Freshwater Marsh (N = 9)
62-34 4 1.2 0.0 0.6 12 0.6 2.9 75 6.4
64-04 4 0.5 0.0 0.0 0.5 0.0 2.8 4.2 2.8
64-07 4 0.0 0.0 0.0 0.4 0.0 4.3 22, 3.0
64-E7SG 4 0.0 0.0 0.0 0.0 0.0 13:9 12 5.2
64-09 4 1.2 0.0 0.0 0.0 0.0 8.8 Ps) 73)
64-15 4 0.0 0.0 0.0 0.5 0.0 16.9 1.4 5.5
64-16SG 4 0.0 0.0 0.0 0.5 0.0 7.4 2.1 6.4
SLTR-03 4 0.0 0.0 0.0 0.0 0.0 eS) 1.4 Tes
SLTR-04 4 0.0 0.0 0.0 1.0 0.0 13.1 0.0 4.7
Avg. 0.3 0.0 0.1 0.5 0.1 Chi7/ 2.5 6.0
Stdey. 0.5 0.0 0.2 0.4 0.2 5.8 2.2 2.8
io)
L ol
—)
i2*)
i—)
-
=
w
—
-_
i)
ie)
_
=
_—
oo
Table |.—Extended.
Typha
Nym-
phaea
Sagit-
taria
POLLEN ZONATION: O'NEAL ET AL. 107
Caryo-
Umbel- Utricu- Com- Poly- phylla- Trip- Eu- Cephal- Frax- Sapot-
liferae laria positae gona ceae_ Ilex Salix Morus Ficus orates genia anthus inus — aceae
0.0 0.0 2A 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 1.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 1.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 2.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 1.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.6 0.0 S7/ 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 2.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 3.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 3.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 6.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.1 0.0 2.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.2 0.0 1.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.1 0.8
0.0 0.0 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.0 0.0 0.0 0.0
0.0 0.0 0.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.6 0.0 12. 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 2a 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.7
0.0 0.0 3:3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 No 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.6 0.0
0.0 0.0 0.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 9.6 0.0 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.6 0.0
0.6 0.0 2 0.0 0.0 0.0 0.0 0.0 0.0 Sal 0.0 0.5 0.5 0.0
0.4 0.0 1.8 0.0 0.0 0.0 0.0 0.0 0.0 ile7/ 0.4 0.0 0.0 0.0
0.0 0.0 0.6 0.0 0.7 0.0 0.7 0.0 0.0 2.0 0.0 0.0 0.0 0.0
0.9 0.0 2.4 0.0 0.4 0.0 0.0 0.0 0.0 3.9 0.0 0.0 0.0 0.0
0.6 0.0 3.0 0.0 0.0 0.6 0.0 0.6 0.6 1.9 0.0 0.0 0.0 0.6
0.2 0.0 1.9 0.0 0.1 0.0 0.0 0.0 0.0 0.9 0.0 0.0 0.1 0.1
0.3 0.0 2.3 0.0 0.2 0.2 0.2 0.2 0.2 1.3 0.1 0.1 0.2 0.2
0.2 1.2 0.1 0.1 0.1 0.1 0.1 0.7 0.1 0.1 0.1 0.1
0.0 0.0 0.8 0.0 0.0 0.0 0.0 0.0 0.0 4.8 0.0 0.0 0.0 0.0
0.3 0.0 1.8 0.0 0.0 0.3 0.0 0.0 0.0 4.4 0.0 0.0 0.0 0.3
0.0 0.0 4.2 0.0 0.0 0.0 0.0 0.0 0.0 3.8 0.0 0.0 0.0 1.1
0.3 0.3 3.0 0.0 0.0 0.0 0.0 0.0 0.3 5.4 0.0 0.0 0.0 0.0
0.3 0.0 0.6 0.0 0.0 0.0 0.3 0.0 0.3 5.4 0.0 0.0 0.0 0.0
0.3 0.0 33) 0.0 0.0 0.0 0.0 0.0 0.0 SiS) 0.0 0.0 0.0 0.0
0.0 0.0 3.8 0.0 0.0 0.0 0.0 0.0 0.0 2.9 0.0 0.0 0.0 0.0
0.0 0.5 2.0 0.0 0.0 0.0 0.5 0.0 0.0 4.3 0.0 0.0 0.0 0.0
0.0 0.0 3.0 0.6 0.0 0.6 0.0 0.0 0.0 S)) 0.0 0.0 0.0 0.0
0.0 0.0 0.6 0.0 0.0 0.0 0.0 0.0 0.6 8.1 0.0 0.0 0.0 0.0
0.0 0.0 1.4 0.0 0.0 0.7 0.0 0.0 0.0 IS) 0.0 0.0 0.0 0.0
0.1 0.1 2.2 0.1 0.0 0.1 0.1 0.0 0.1 Seo) 0.0 0.0 0.0 0.1
0.2 0.2 1.3 0.2 0.0 0.3 0.2 0.0 0.2 3.0 0.0 0.0 0.0 0.3
0.1 0.1 0.8 0.1 0.2 0.1 0.1 1.8 0.2
0.6 0.0 4.0 0.6 0.0 0.6 0.0 Pee) 0.0 17.9 0.0 0.0 0.0 le7/
0.5 0.0 See 0.0 0.0 0.0 0.9 0.0 0.0 9.9 0.0 0.0 0.5 0.0
1.3 0.9 2.5 0.0 0.0 0.0 0.4 0.0 0.0 8.7 0.4 0.4 0.0 0.0
od N72 0.6 0.0 0.0 0.0 0.0 0.0 0.0 12a 0.0 (leg 0.0 0.0
1.2 1.8 2.9 0.0 0.0 1.2 1.2 0.0 0.0 9.4 0.0 12 0.0 0.0
0.9 0.9 2.3 0.0 0.0 0.5 0.5 0.0 0.0 6.9 0.0 0.9 0.0 0.0
2.1 0.5 Dr 0.0 0.0 0.5 0.0 0.0 0.5 2Ae2 0.0 0.0 0.0 0.0
0.0 0.0 0.9 0.0 0.0 0.0 0.0 0.5 0.0 Tell 0.0 0.5 0.0 0.0
0.0 1.0 0.5 0.0 0.0 0.0 1.0 0.0 0.0 11.5 0.0 0.5 0.0 0.0
0.9 0.7 2.2 0.1 0.0 0.3 0.4 0.3 0.1 11.7 0.0 0.6 0.1 0.2
0.7 0.6 1.2 0.2 0.0 0.4 0.5 0.8 0.2 4.8 0.1 0.6 0.2 0.6
0.5 0.4 0.8 0.1 0.3 0.3 0.5 0.1 3.1 0.1 0.4 0.1 0.4
108 BULLETIN 361
Table 1.—Continued.
Anacar- cf. Ovoidi-
Site Cluster Quercus dium Alnus Carya Nyssa Taxodium tes Vitis Rapenea
Mangrove Swamp (N = 10)
62-01 1 0.0 ils) 0.0 0.0 0.0 0.7 0.0 0.0 0.0
62-06 1 1:2 5.4 0.0 0.0 0.0 1.2 0.0 0.0 0.0
62-08 1 1.2 0.0 0.0 0.0 0.0 2.4 0.0 0.0 0.0
62-11 1 0.0 2.4 0.0 0.0 0.0 3.0 0.0 0.6 0.0
62-12 1 0.0 1.2 0.0 0.0 0.0 2.5 0.0 0.0 0.0
62-15 1 1.2 0.0 0.0 0.0 0.0 0.6 0.0 0.0 0.0
62-17 ! 0.0 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0
62-20 1 0.0 1.2 0.0 0.0 0.0 1.6 0.0 0.0 0.0
62-21 1 1.9 0.0 0.0 1.3 0.0 3.2 0.0 0.6 0.0
62-22 1 0.0 0.0 0.0 0.0 0.0 1.1 0.0 0.0 0.0
Avg. 0.6 1.2 0.0 0.1 0.0 1.6 0.0 0.1 0.0
Stdev. 0.7 1.7 0.0 0.4 0.0 1.1 0.0 0.3 0.0
cI 0.5 1.0 0.3 0.7 0.2
Mangrove Swamp (N = 15)
62-02 2 0.5 1.2 0.0 0.0 0.0 0.2 0.0 0.0 0.0
62-03 2 0.0 0.0 0.0 0.0 0.0 2.1 0.0 0.0 0.0
62-05 2 3.0 1.8 0.0 0.0 0.0 0.5 0.0 0.0 0.0
62-07 2 0.0 0.0 0.0 0.0 0.0 2.4 0.0 0.0 0.0
62-09 2 13 1.3 0.6 0.0 0.0 1.3 0.0 0.0 0.0
62-10 2 1.4 0.7 0.7 0.1 0.0 1.3 0.0 0.0 0.0
62-13 2 4.0 2.6 0.7 0.0 0.0 0.0 0.0 0.7 0.0
62-14 2 2.3 1.1 0.0 0.0 0.0 1.7 0.0 0.0 0.0
62-18 2 3.9 0.0 0.0 0.0 0.0 4.5 0.0 0.0 0.0
62-19 2 4.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
64-11 2 DES 0.6 0.0 0.0 0.0 2.5 0.6 0.0 0.0
64-21 2 Del 0.0 0.0 0.0 0.0 0.0 0.9 0.0 0.0
SR-O1 2 1.3 0.0 0.0 0.0 0.0 3:3 0.0 0.0 0.0
SR-02 2 4.3 0.4 0.8 0.0 0.0 0.4 0.4 0.0 0.0
59-T1 2 3:5 1.2 0.0 0.0 0.0 2.5 0.0 0.0 0.0
Avg. 2.2 0.7 0.2 0.0 0.0 cs 0.1 0.0 0.0
Stdev 1.7 0.8 0.3 0.0 0.0 1.4 0.3 0.2 0.0
CI 0.7 0.4 0.2 0.0 0.7 0.1 0.1
Mangrove Swamp (N = 11)
SR-03 3 0.9 0.0 0.0 0.0 0.0 0.0 0.9 0.0 0.0
SR-04 3 322 0.0 0.0 0.0 0.0 0.3 0.6 0.0 0.0
SR-05 3 1.0 0.5 0.5 0.0 0.0 0.6 0.0 0.0 0.0
SR-06 3 2 0.3 0.0 0.0 0.0 0.6 1.7 0.0 0.0
SR-07 3 2.1 0.0 0.0 0.0 0.0 0.9 1.2 0.0 0.0
SR-09 3 0.3 0.0 0.0 0.0 0.0 0.0 1.3 0.0 0.0
SR-10 3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
SR-13 3 Del 0.0 0.0 0.0 0.0 1.1 0.0 0.5 0.0
SR-14 3 1.2 0.0 0.0 0.0 0.0 1.2 0.0 0.0 0.0
SR-15 3 0.6 0.0 0.0 0.0 0.0 1.3 0.0 0.0 0.0
SR-16 3 0.7 0.0 0.7 0.0 0.0 1.3 0.7 0.0 0.0
Avg 1.4 0.1 0.1 0.0 0.0 0.7 0.6 0.0 0.0
Stdev 1.1 0.2 0.2 0.0 0.0 0.5 0.6 0.2 0.0
CI 0.6 0.1 0.1 0.3 0.4 0.1
Freshwater Marsh (N = 9)
62-34 4 il-72 0.6 0.6 0.0 0.0 il -72 2.9 1.7 0.0
64-04 4 Shr 0.0 0.0 0.0 0.0 3.8 15.0 0.0 0.0
64-07 4 13 0.4 0.4 0.0 0.0 EY 10.3 0.0 0.0
64-E7SG 4 1.2 0.0 12 0.0 0.0 2.3 8.7 0.0 0.0
64-09 4 2.4 0.6 0.0 0.0 0.0 53) 15.8 1.2 1.2
64-15 4 1.0 0.0 0.0 0.0 0.0 1.4 12.8 0.0 0.0
64-16SG 4 0.5 0.0 0.5 0.0 0.0 0.5 iL? 0.0 0.0
SLTR-03 4 0.9 0.9 0.0 0.0 0.0 1.8 8.2 0.0 0.0
SLTR-04 4 1.0 0.0 0.0 0.0 0.0 Sell 5.8 0.0 0.0
Avg 1.4 0.3 0.3 0.0 0.0 2.3 10.1 0.3 0.1
Stdev. 0.8 0.4 0.4 0.0 0.0 1.5 4.2 0.7 0.4
o)
L al
=
a
—
iv
=
ta
cme
—)
iS)
Q
S
os
o
w
Hame-
lia
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Rhus
0.7
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
0.2
0.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
cil
0.0
0.0
0.0
0.0
0.0
0.1
0.3
0.1
0.0
0.3
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
0.1
0.0
0.0
0.0
0.0
0.0
POLLEN ZONATION: O’NEAL ET AL. 109
Table 1.—Continued Extended.
cf. Grains
Thom- Rhab- Grains used
Myrt- Poly- Eric- Sphag- sono- daden- Liquid- count- for
aceae Osmunda podium — aceae num Ulmus pollis Nuphar ia Pteris amber Pinus Other ed Stats
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.0 10.1 150 133
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 6.6 De) 167 156
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 4.3 8.7 164 148
0.0 0.0 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Te? 8.5 167 151
0.0 0.0 ibs) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 10.5 Tell 162 149
0.0 0.0 ed, 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 8.0 4.7 162 153
0.0 0.0 2, 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Ue) 12.6 164 142
0.0 0.0 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 15.8 4.6 184 174
0.0 0.6 1.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3.8 10.9 158 139
0.0 0.0 igi 0.0 0.6 0.0 0.0 0.0 0.0 0.0 0.0 20.7 ily 179 156
0.0 0.1 0.7 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 8.7 9.4
0.0 0.2 0.6 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.0 5.7 69.0
0.1 0.3 0.1 3.5 80.8
0.0 0.5 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3yS) 0.5 591 576
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 8.8 4.5 194 181
0.0 0.6 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 8.9 OD n/a n/a
0.0 2 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 9.0 16.2 166 136
0.0 0.0 1.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 18.9 15.2 157 130
0.0 0.7 2.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 16.2 7.6 154 139
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 14.4 10.7 153 134
0.0 1.7 2.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 15.3 8.5 177 158
0.0 0.0 0.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 13.6 3.4 155 147
0.0 0.0 0.0 0.0 0.6 0.0 0.0 0.0 0.0 0.0 0.0 27.4 6.6 179 164
0.0 Sul 3.1 0.0 0.0 0.0 0.0 0.6 0.0 0.0 0.0 6.8 36.1 n/a n/a
0.0 18.1 0.9 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 9.4 4.1 23 223
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 17.8 15.1 52 26
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 15.2 7.4 25 233
0.0 1.8 1.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 S33 5.1 163 SI!
0.0 1.8 0.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 12.8 12.1
0.0 4.6 1.0 0.1 0.2 0.0 0.0 0.2 0.0 0.0 0.0 5.6 41.0
2.3 0.5 0.1 0.1 0.1 2.8 70.1
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 8.0 7.0 225 203
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 11.4 6.9 315 284
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 12.5 8.8 184 162
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 12.5 6.4 352 319
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 8.6 25.3 336, 241
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 10.4 4.3 307 285
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 7.6 199) 105 81
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 17.3 We? 184 165
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 8.8 OKT, 170 131
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 6.3 18.3 160 126
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 O33 14.5 151 125
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 10.2
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3.1
1.8
0.0 0.0 L7/ 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 13.9 25:9) 17/3) 121
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 25.4 0.9 213 139
0.0 0.0 3.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 47.8 49.2 232 109
0.0 Shs) 35 0.0 0.0 0.6 0.0 0.0 0.0 0.0 0.0 33.5 35.0 173 106
0.0 0.0 1.8 0.0 0.0 0.0 0.6 0.0 0.0 0.0 0.0 17.0 13.1 171 142
0.0 0.9 1.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 32.4 38.2 219 127
0.0 1.0 2.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 22.9 25.2 218 154
0.0 6.3 Drill 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 28.2 30. 220 145
0.0 2.6 3.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 33.5 36.7 213 126
0.0 1.6 2.3 0.0 0.0 0.1 0.1 0.0 0.0 0.0 0.0 28.3
0.0 2.2 1.2 0.0 0.0 0.2 0.2 0.0 0.0 0.0 0.0 10.1
1.4 0.8 0.1 0.1 6.6
110 BULLETIN 361
Table 1.—Continued.
Lagun- Cheno-
Site Cluster = Rhizophora Avicennia cularia Conocarpus _ Batis pods Gramineae Cyperaceae
Freshwater Marsh (N = 8)
62-23 5 0.6 0.0 0.0 0.0 0.0 34.8 0.0 5.6
64-E7E 5) 0.0 0.0 0.0 0.0 0.0 Bye. 1.0 1S
64-10 5 0.0 0.0 0.0 0.0 0.0 33.8 0.0 3.6
64-16E 5 0.0 0.0 0.0 0.6 0.0 eT 0.0 By)
64-16ASG 5) 0.0 0.0 0.0 0.0 0.0 47.2 0.5 1.8
64-16AE 5 il 0.0 0.0 0.0 0.0 Syileil 1.2 122)
SLTR-O1 5) 0.4 0.0 0.0 0.0 0.0 22.9 3a 1.9
SLTR-02 5) 0.5 0.0 0.0 0.0 0.0 Dial 0.5 31)
Avg. 0.3 0.0 0.0 0.1 0.0 32.7 0.8 Del
Stdev. 0.4 0.0 0.0 0.2 0.0 7.5 1.0 1.5
cI 0.3 0.1 S52 0.7 1.0
Headwater Marsh (N = 5)
64-33 6 0.0 3.4 0.0 0.6 16.2 62.1 0.0 0.6
SR-25 6 0.5 0.0 0.3 0.0 0.0 80.5 0.3 2.1
SR-26 6 0.9 0.0 0.0 0.0 0.0 78.4 0.0 3
SR-28 6 0.4 0.0 0.4 0.0 0.0 70.6 0.0 0.4
SR-29 6 1.3 0.3 0.0 0.0 0.0 89.9 0.0 0.3
Avg 0.6 0.7 0.1 0.1 3.2 76.3 0.1 0.9
Stdev 0.5 1.5 0.2 0.3 72 10.5 0.1 0.8
CI 0.4 1.3 0.2 0.2 6.4 9.2 0.1 0.7
Headwater Marsh (N = 10)
62-24 7 2.3 0.0 0.0 0.0 0.0 37.3 2.3 1.7
SR-17 7 18.2 iNet 0.0 1.7 0.0 22K 0.6 Ne7/
SR-18 7 24.4 0.0 0.0 3.6 0.6 31.5 0.0 4.2
SR-19 7 11.0 0.5 1.0 4.0 0.0 33.0 0.0 3.0
SR-20 7 20.1 0.7 0.0 2.0 0.0 22.8 0.0 0.7
SR-21 7 28.7 0.5 0.5 1.1 0.0 23.2 0.0 1.1
SR-22 7 8.9 0.0 0.0 0.0 0.0 47.3 0.0 0.8
SR-23 7 15.4 1.1 0.0 0.0 0.5 44.0 0.0 22.
SR-24 7 6.8 0.9 0.0 0.9 0.0 47.7 0.0 1.4
SR30 Ti 25.6 1.8 0.6 1.2 0.0 14.9 0.0 1.8
Avg 16.1 0.7 0.2 1.5 0.1 32.4 0.3 1.9
Stdev 8.8 0.6 0.4 1.4 0.2 11.5 0.7 1.1
CI 5.4 0.4 0.2 0.9 0.1 Aoi 0.5 0.7
Brackish Marsh (N = 9)
62-25 8 7.9 0.0 0.0 3.4 0.0 11.9 0.0 4.5
62-30 8 7.0 0.0 0.0 10.2 0.6 8.3 1.3 8.9
64-18 8 9.5 0.0 0.0 1.1 1.6 10.6 2.0 16.4
64-22 8 2.3 0.0 0.6 1.1 il57/ 5.6 0.5 17.0
HR-O1 8 12.2 0.0 0.0 19.1 0.9 73 1.9 0.9
HR-02 8 16.0 0.6 1.9 13.5 0.6 7.0 1.9 1.9
JR-O1 8 3.7 0.0 0.7 4.1 0.0 4.8 0.0 3.0
TB-O1 8 16.9 0.0 0.0 4.5 atl 19.1 0.6 11.2
TB-02 8 16.4 0.0 0.0 3.0 0.0 9.7 7 8.5
Avg 10.2 0.1 0.4 6.7 0.7 9.4 4.2 8.0
Stdev fs) 0.2 0.6 6.2 0.7 4.3 9.5 6.0
CI 3.6 0.1 0.4 4.1 0.4 2.8 6.2 3.9
Brackish Marsh (N = 5)
62-26 9 1.3 0.0 0.0 1133 0.0 6.3 1.3 2S
62-33b 9 0.0 0.0 0.0 0.9 0.9 0.9 0.0 2.8
62-35 9 0.0 0.0 0.0 0.5 0.0 0.5 0.0 1.6
64-09H 9 0.0 0.0 0.0 0.0 0.0 15.1 0.6 35
64-14 9 2, 0.0 0.0 0.0 1.2 3.6 2.4 3.6
Avg. 0.5 0.0 0.0 0.5 0.4 55) 0.9 2.7
Stdev 0.7 0.0 0.0 0.6 0.6 6.0 1.0 0.7
CI 0.6 0.5 0.5 5.2) 0.9 0.7
Upland (N = 4)
62-27 10 0.0 0.0 0.0 0.0 0.0 ite 0.6 isi
62-28 10 0.6 0.0 0.0 0.6 0.0 12 3.8 3.8
62-29 10 0.6 0.0 0.0 0.0 0.0 lez 0.0 0.0
EE-11 10 0.0 0.0 0.0 0.0 0.0 0.6 0.6 4.1
Avg. 0.3 0.0 0.0 0.2 0.0 1.2 1.3 2.3
Stdev. 0.3 0.0 0.0 0.3 0.0 0.5 1.7 2.0
POLLEN ZONATION: O’ NEAL ET AL. 111
Table 1.—Continued Extended.
Caryo-
Nym- Sagit- Umbel-Utricu- Com- Poly- phylla- Trip- Eu- Cephal- Frax- Sapot-
Typha phaea taria liferae laria positae gona ceae_ Ilex Salix Morus Ficus orates genia anthus inus — aceae
0.6 ls7 Ded, 0.0 0.0 0.6 0.0 0.0 0.0 0.0 0.0 0.0 9.0 0.6 0.0 0.0 0.0
0.0 S)5) 3.0 0.3 1.3 0.6 0.3 0.0 0.0 0.3 0.0 0.0 3.0 0.0 0.0 0.0 0.0
0.5 4.1 0.5 1.0 0.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 6.7 0.0 0.0 0.0 0.0
0.0 9.0 12 0.6 1.9 0.6 0.0 0.0 0.0 0.0 0.6 0.0 135 0.0 0.0 0.0 0.0
0.0 5.0 2.3 0.0 0.5 0.0 0.0 0.0 0.0 0.0 0.0 0.5 2.8 0.0 0.5 0.0 0.5
0.8 12.0 5.0 0.4 0.4 2.0 0.0 0.0 0.0 0.4 0.0 0.0 2 0.0 0.0 0.0 0.0
OOF 253 1.0 0.4 0.4 il) 0.0 0.0 0.0 0.0 0.0 0.0 0.8 0.0 0.0 0.0 2.8
0.0 14.7 5.0 0.5 0.5 2.7 0.0 0.0 0.0 0.0 0.0 0.0 6.9 0.0 0.5 0.0 0.0
0.2 9.6 2.5 0.4 0.7 1.0 0.0 0.0 0.0 0.1 0.1 0.1 5.5 0.1 0.1 0.0 0.4
0.3 Tet 1.7 0.3 0.6 1.0 0.1 0.0 0.0 0.2 0.2 0.2 4.4 0.2 0.2 0.0 1.0
0.2 5.3 1.2 0.2 0.4 0.7 0.1 0.1 0.1 0.1 3.0 0.1 0.2 0.7
0.6 0.0 0.0 0.0 0.0 4.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.3 0.8 5.9 0.0 0.0 0.0 0.5 0.0 0.0 0.0 0.0 0.0 Dell 0.0 0.0 0.0 0.0
0.4 0.4 3.0 0.4 0.0 0.8 0.0 0.0 0.0 0.0 0.0 0.4 319 0.0 0.0 0.0 0.0
0.0 0.8 225 0.0 0.0 1.6 0.0 0.0 0.0 0.0 0.0 0.0 ie? 0.0 0.4 0.0 0.0
0.0 0.0 2.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.6 0.0 0.3 0.0 0.0
0.3 0.4 2.9 0.1 0.0 1.3 0.1 0.0 0.0 0.0 0.0 0.1 2.0 0.0 0.1 0.0 0.0
0.3 0.4 2.1 0.2 0.0 1.7 0.2 0.0 0.0 0.0 0.0 0.2 1.4 0.0 0.2 0.0 0.0
0.2 0.4 1.8 0.2 1.5 0.2 0.2 1.3 0.2
23 lev 1.0 1.3 0.0 10.3 0.0 0.0 0.3 ile4/ 0.0 0.0 lea 0.0 1.0 0.0 0.0
0.0 23 1.7 0.0 0.0 2.4 0.0 0.0 1.1 0.6 0.0 0.0 20.4 0.0 0.0 0.0 0.0
0.0 0.0 i) 0.0 0.0 3.6 0.0 0.0 0.0 0.0 0.0 0.0 Hell 0.0 0.0 0.0 0.0
0.0 1.0 2:5 0.0 0.5 2.0 0.0 0.0 0.5 0.5 0.5 0.0 ae) 0.0 0.5 0.0 0.0
0.0 1.3 13, 0.0 0.0 4.2 0.0 0.7 0.0 0.0 0.0 0.0 11.4 0.0 0.0 0.7 0.0
0.5 1.1 1.1 0.0 0.0 5.4 0.0 0.0 0.6 0.0 0.6 0.0 10.3 0.0 0.0 0.6 0.0
0.0 Med 1.1 0.0 0.0 5.6 0.0 0.0 1.1 0.0 0.0 0.0 12.4 0.0 7 0.0 0.0
0.5 0.0 1.6 0.5 0.0 1.0 0.0 0.0 0.5 1.6 0.0 0.0 9.3 0.5 0.0 0.5 0.0
0.9 0.5 3.6 0.9 0.5 2.8 0.5 0.0 1.4 0.0 0.0 0.0 8.2 0.5 0.5 0.5 0.0
0.6 0.6 0.6 0.0 0.0 1.2 0.0 0.0 0.6 0.0 0.0 0.0 12.5 0.0 125) 0.0 0.0
0.5 1.0 1.6 0.3 0.1 3.9 0.1 0.1 0.6 0.4 0.1 0.0 10.9 0.1 1.6 0.2 0.0
0.7 0.8 0.9 0.5 0.2 2.8 0.2 0.2 0.5 0.7 0.2 0.0 4.1 0.2 3.9 0.3 0.0
0.4 0.5 0.5 0.3 0.1 1.7 0.1 0.1 0.3 0.4 0.1 2.5 0.1 2.4 0.2
0.0 0.0 0.0 0.0 1.8 0.0 0.0 2.8 0.0 0.0 ei 17.6 0.0 ite 0.0 0.6
0.0 0.0 0.0 0.0 6.4 0.0 0.0 0.0 0.6 0.0 0.6 14.6 0.0 0.0 0.0 0.0
1.6 0.0 0.0 0.0 0.0 4.2 0.0 0.0 0.5 0.5 1.1 0.5 ules 0.0 0.0 0.0 0.0
0.0 0.6 0.6 0.6 0.0 2.8 0.0 0.0 1.1 0.6 0.0 0.0 20.9 0.0 0.0 0.4 0.0
0.0 0.0 0.9 0.0 0.0 4.4 0.0 0.0 0.9 0.9 0.0 0.9 19.1 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0 0.0 7.0 0.0 0.0 1.3 0.6 0.0 0.0 22.4 0.0 0.0 0.0 0.0
0.0 0.0 0.4 0.0 0.0 7.4 0.0 0.0 4.1 0.0 0.0 0.0 23.3 0.7 0.0 0.0 0.0
0.0 0.6 0.0 0.0 3.4 0.0 0.6 0.6 0.0 0.0 0.6 17.4 0.0 0.6 0.0 0.0
a3) 0.0 0.6 0.0 0.0 1.8 0.0 0.0 2.4 0.6 0.0 0.0 20.6 1.3 0.0 12; 0.0
0.0 0.0 1.9 19) 0.0 12.6 0.6 0.0 0.0 1.3 0.0 0.0 5.0 0.0 8.2 0.0 0.0
0.0 0.0 0.0 eS) 0.0 4.7 0.9 0.0 0.0 1.9 0.0 1.9 3.8 0.0 19) 0.0 18.8
1.6 0.0 1.1 0.0 0.0 12.7 0.0 0.0 0.0 9.5 0.5 0.0 0.5 0.0 0.5 0.0 0.0
0.6 0.0 1S) 0.6 0.0 6.9 0.6 0.0 0.0 26.4 0.0 0.6 14.5 0.0 2.5 0.0 0.0
0.0 0.6 8.3 7A 0.6 3.6 0.0 0.0 0.0 3.0 0.0 0.0 4.2 0.0 27.4 0.0 0.0
0.4 0.1 2.5 2.2 0.1 8.1 0.4 0.0 0.0 8.4 0.1 0.5 5.6 0.0 8.1 0.0 3.8
0.7 0.3 3.3 2.8 0.3 4.3 0.4 0.0 0.0 10.6 0.2 0.8 5.3 0.0 11.2 0.0 8.4
0.6 0.2 2.9 2.5 0.2 3.8 0.4 9.3 0.2 0.7 4.6 9.8 7.4
0.0 0.0 0.0 0.0 0.0 Prp3) 0.0 0.0 0.6 0.6 0.0 0.0 0.6 0.0 2.8 0.0 0.0
0.6 0.0 0.6 0.0 0.0 4.6 0.6 0.0 0.0 Sil 0.0 0.0 1.3 0.0 0.0 0.0 0.0
0.6 0.0 0.0 0.0 0.0 Nod 0.0 0.0 0.0 1.1 0.0 0.0 1.1 0.6 0.0 0.0 0.0
0.6 0.0 te, 0.0 0.0 1.2 0.0 0.0 0.0 2.3 0.6 0.0 2.3 0.0 4.1 0.0 0.0
0.5 0.0 0.5 0.0 0.0 2.5 0.2 0.0 0.2 2.4 0.2 0.0 1.3 0.2 1.7 0.0 0.0
0.3 0.0 0.6 0.0 0.0 1.5 0.3 0.0 0.3 2.3 0.3 0.0 0.7 0.3 2.1 0.0 0.0
0.3 0.6 1.5 0.3 0.3 2.3 0.3 0.7 0.3 2.0
1 BULLETIN 361
Table 1.—Continued.
Anacar- cf. Ovoidi-
Site Cluster Quercus dium Alnus Carya Nyssa Taxodium tes Vitis Rapenea
Freshwater Marsh (N = 8)
62-23 5 el 0.6 0.0 0.0 0.0 3.9 9:5 0.0 0.0
64-E7E 5 0.7 0.0 0.0 0.0 0.0 0.0 7.0 0.0 0.0
64-10 5 1.0 0.0 0.0 0.0 0.0 il5) 12.8 0.0 0.0
64-16E 5 0.0 0.6 0.0 0.0 0.0 3.9 8.4 0.0 0.0
64-16ASG 5) 1.4 0.0 0.0 0.5 0.0 0.9 6.0 0.0 0.0
64-16AE 5 0.4 0.0 0.0 0.0 0.0 1.2 1.4 0.0 0.0
SLTR-O1 5 0.0 0.0 0.0 0.0 0.0 P)s8) 9.6 0.0 0.0
SLTR-02 5) 0.5 0.5 0.0 0.0 0.0 1.8 8.3 0.0 0.0
Avg. 0.6 0.2 0.0 0.1 0.0 1.9 7.9 0.0 0.0
Stdev 0.5 0.3 0.0 0.2 0.0 1.4 3.3 0.0 0.0
cI 0.4 0.2 0.1 1.0 2.3
Headwater Marsh (N = 5)
64-33 6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
SR-25 6 0.0 0.0 0.0 0.0 0.0 0.5 0.8 0.0 0.0
SR-26 6 0.0 0.0 0.0 0.0 0.0 DD, 0.0 0.4 0.0
SR-28 6 0.0 0.0 0.0 0.0 0.0 ile7/ 0.0 0.0 0.0
SR-29 6 0.3 0.0 0.0 0.0 0.0 0.0 0.3 0.0 0.0
Avg. 0.1 0.0 0.0 0.0 0.0 0.9 0.2 0.1 0.0
Stdev 0.1 0.0 0.0 0.0 0.0 1.0 0.3 0.2 0.0
cI 0.1 0.9 0.3 0.2
Headwater Marsh (N = 10)
62-24 7 1.0 0.0 0.0 0.0 0.0 0.7 De, 0.0 0.0
SR-17 q/ 0.0 0.0 0.0 0.0 0.0 4.6 0.0 0.6 0.0
SR-18 iT, 0.0 0.0 0.0 0.0 0.0 2.9 0.0 0.0 0.0
SR-19 7 13) 1.5 0.0 0.0 0.0 2.4 0.5 0.0 0.0
SR-20 7 2.0 0.7 0.0 0.0 0.0 47 0.0 0.0 0.0
SR-21 W 0.0 ey 0.0 0.0 0.0 1.6 0.6 0.0 0.0
SR-22 vii ED 0.0 0.0 0.0 0.0 2) 1.1 0.0 0.6
SR-23 7 1.0 0.0 0.0 0.0 0.5 1.1 0.5 0.5 0.0
SR-24 7 0.9 0.0 0.0 0.0 0.0 1.4 1.4 0.5 0.0
SR-30 7 1.2 0.0 0.0 0.0 0.6 2.4 0.0 0.0 0.6
Avg 0.9 0.4 0.0 0.0 0.1 2.4 0.7 0.2 0.1
Stdev 0.7 0.7 0.0 0.0 0.2 1.4 0.9 0.3 0.3
cI 0.4 0.4 0.1 0.8 0.5 0.2 0.2
Brackish Marsh (N = 9)
62-25 8 0.6 0.0 0.6 0.0 0.0 61 0.0 0.6 0.6
62-30 8 1.9 0.0 0.6 0.0 0.0 8.3 0.6 0.6 0.6
64-18 8 Del 0.0 0.0 0.0 0.0 3}37/ 0.5 0.0 0.0
64-22 8 5 0.0 0.0 0.0 0.0 2.8 4.5 0.0 0.0
HR-O1 8 Net 0.9 0.0 0.0 0.0 4.3 0.0 0.0 0.0
HR-02 8 0.6 0.0 0.0 0.0 0.0 1.9 0.0 0.6 0.0
JR-O1 8 0.0 0.4 0.0 0.0 0.0 0.0 0.0 8.5 0.0
TB-O1 8 ED 0.0 0.0 0.0 0.0 el 0.0 0.0 0.0
TB-02 8 3.0 0.0 0.0 0.0 0.0 2.4 1.2 0.0 0.0
Avg. 1.3 0.1 0.1 0.0 0.0 2.8 0.8 1.1 0.1
Stdev 1.0 0.3 0.3 0.0 0.0 2.4 1.5 2.8 0.3
CI 0.6 0.2 0.2 1.6 1.0 1.8 0.2
Brackish Marsh (N = 5)
62-26 9 1.3 0.0 0.0 0.0 0.0 32.1 1.9 0.0 0.0
62-33b 9 0.9 0.0 0.0 0.0 0.0 2.8 0.0 2.8 0.0
62-35 9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.6 0.0
64-09H 9 0.6 0.0 0.6 0.0 0.0 6.9 0.6 0.0 0.0
64-14 9 0.6 0.6 0.0 0.0 0.0 1.2 1.8 is? 0.0
Avg. 0.7 0.1 0.1 0.0 0.0 8.6 0.9 1.1 0.0
Stdev 0.5 0.3 0.3 0.0 0.0 13.4 0.9 1.2 0.0
CI 0.4 0.2 0.2 11.7 0.8 1.0
Upland (N = 4)
62-27 10 el 0.0 0.0 0.0 0.6 72.5 0.0 D3, 0.0
62-28 10 0.6 0.0 0.6 0.6 0.0 55.4 i133} 0.0 0.0
62-29 10 ile 7/ 0.0 0.0 0.0 0.0 87.4 0.6 0.0 0.0
EE-11 10 0.0 0.0 0.6 0.0 0.0 73.4 0.0 0.6 0.0
Avg. 0.9 0.0 0.3 0.2 0.2 1p 0.5 0.7 0.0
Stdev. 0.7 0.0 0.3 0.3 0.3 13.1 0.6 1.1 0.0
POLLEN ZONATION: O’ NEAL ET AL. 113
Table 1.—Continued Extended.
ch Grains
Thom- Rhab- Grains used
Hame- Myrt- Poly- Eric- Sphag- sono- daden- Liquid- count- for
lia Rhus aceae Osmunda podium — aceae num Ulmus pollis Nuphar ia Pteris amber Pinus Other ed Stats
0.0 0.0 0.0 23 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 20.2 Pailes) 178 13
0.0 0.0 0.0 0.3 3.3 _ 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 31.6 30.1 301 195
0.0 0.0 0.0 0.5 4.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 24.6 24.4 195 138
0.0 0.0 0.0 2.6 1.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 16.8 18.7 155 118
0.0 0.0 0.0 1.8 1.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 TL) 21.0 218 161
0.0 0.0 0.0 0.0 2.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 28.2 32.2 241 151
0.0 0.0 0.0 0.0 0.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 23.0 21.8 261 191
0.0 0.0 0.0 0.5 0.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 21.6 20.4 218 163
0.0 0.0 0.0 1.0 2.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 23.7
0.0 0.0 0.0 1.1 1.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 4.6
0.7 0.9 3:2
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 8.2 12S) 179 157
0.0 0.0 0.0 0.8 0.0 0.0 0.0 0.0 0.0 0.3 0.0 0.0 0.0 1 3h] 375 361
0.0 0.0 0.0 0.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3.9 6.7 231 216
0.0 0.0 0.0 11.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 ALS) 8.4 242 222
0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.6 Ds) 307 299)
0.1 0.0 0.0 2.5 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 3.7
0.1 0.0 0.0 4.8 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 2.7
0.1 4.2 0.1 2.3
0.0 0.0 0.0 2.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 93 le7/ 300-244
0.0 0.0 0.0 1.8 0.0 0.0 0.0 0.6 0.0 0.0 0.0 0.0 0.0 10.2 10.9 176 144
0.0 0.0 0.0 6.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 9.5 tT) 168 143
0.0 ile) 0.0 6.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 6.0 13.3 200 159
0.0 0.0 0.0 10.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 7.4 9.6 149 124
0.0 1.1 0.6 7.0 0.6 0.0 0.6 0.0 0.0 0.0 0.0 0.0 0.0 7.0 3.9 185 165
0.0 0.0 0.6 2.9 0.0 0.0 0.0 0.0 0.0 0.6 0.0 iI 0.0 4.5 2.1 178 162
0.0 0.0 0.5 4.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 4.4 3.5} 182 160
0.0 0.0 0.0 6.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 4.6 3.5 220 197
0.0 0.0 0.6 4.8 0.0 0.0 0.0 0.0 0.0 1.2 0.0 0.0 0.0 3.6 Tel 168 144
0.0 0.3 0.2 5.2 0.1 0.0 0.1 0.0 0.1 0.2 0.0 0.1 0.0 6.7
0.0 0.6 0.3 2.6 0.2 0.0 0.2 0.0 0.2 0.4 0.0 0.3 0.0 2.4
0.3 0.2 1.6 0.1 0.1 0.1 0.3 0.2 1.5
0.0 0.6 0.0 1e, 19.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 9.0 13.2 n/a n/a
0.0 0.0 0.0 13 7.6 0.0 0.0 0.0 0.0 0.0 0.6 0.0 0.0 7.6 6.3 n/a n/a
0.0 0.0 0.0 10.1 6.4 0.0 0.0 0.0 0.5 0.0 0.0 0.0 0.0 IE 8.0 n/a n/a
0.0 0.0 0.0 2.3 6.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 IG)? 18.6 n/a n/a
0.0 0.0 0.0 4.4 7.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 6.1 4.3 n/a n/a
0.0 0.0 0.0 3.8 82, 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 5.8 7.2 n/a n/a
0.0 0.0 0.0 8.2 19.2 0.0 0.0 0.0 0.4 0.0 0.0 0.0 0.0 4.1 351 271 241
0.0 0.0 0.0 1.2 il 72 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 6.7 5.0 n/a n/a
0.0 0.6 0.0 0.6 4.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 11.5 4.9 n/a n/a
0.0 0.1 0.0 Abi 8.3 0.0 0.0 0.0 0.1 0.0 0.1 0.0 0.0 9.0
0.0 0.3 0.0 3.3 6.5 0.0 0.0 0.0 0.2 0.0 0.2 0.0 0.0 4.5
0.2 2.2 4.3 0.1 0.1 3.0
0.0 0.0 0.0 0.0 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 8.2 10.9 n/a n/a
0.0 0.0 0.0 2.8 6.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 7.6 34.2 n/a n/a
0.0 0.0 0.0 0.0 54.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.5 38) n/a n/a
0.0 0.0 0.0 0.6 6.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 LES) 2.6 n/a n/a
0.0 0.0 0.0 0.6 6.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Tell 11.7 168 133
0.0 0.0 0.0 0.8 14.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 5.2
0.0 0.0 0.0 1.2 22.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3.7
1.0 19.5 ah
0.0 0.0 0.0 0.0 ite 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.8 27 n/a n/a
0.0 0.0 0.0 0.6 il.) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 7.0 15.6 n/a n/a
0.0 0.0 0.0 0.0 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.6 0.0 0.6 Dal n/a n/a
0.0 0.0 0.0 0.0 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.3 ee. n/a n/a
0.0 0.0 0.0 0.2 1.1 0.0 0.0 0.0 0.0 0.0 0.0 0.2 0.0 3.2
0.0 0.0 0.0 0.3 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.3 0.0 2.7
114 BULLETIN 361
Nuphar
Ficus
Polygona
Morus
Alnus
Sapotaceae
Vitis
Ilex
Anacardium
Osmunda
Composits
Quercus
Typha
3 Gramineae
Salix
Cephalanthus
Polypodium
Taxodium
Chenopods
Triporates
Cyperaceae
Sagittaria
Utricularia
E Umbelliferae
cf. Ovoidites
Nymphaea
Pinus
(a Rhizophora
Conocarpus
D Laguncularia
Avicennia
Batis
Rhus
Sphagnum
Carophyllaceae
Fraxinus
Eugenia
Carya
Rhabdadenia
Ericaceae
it Hamelia
Ulmus
Pteris
A Rapenea
cf. Thomsonopollis
Nyssa
Myrtaceae
tO
:
Distance
Text-figure 3.—Dendrogram illustrating associations of surface
pollen and spore categories. Four cluster groupings are based on a
Chi-squared distance of ~2.8.
nus. Cluster D contains Rhizophora, Conocarpus, La-
guncularia, Avicennia, and Batis (Text-fig. 3).
Cluster A contains taxa that are generally found in
lower frequencies than those of other clusters (Table
1). Sub-cluster Al includes taxa that have very low
pollen frequencies and commonly appear in <2% of
the sample sites. In addition, there are no clear patterns
of spatial distribution for these taxa. The taxa in sub-
cluster A2 appear in slightly higher frequencies than
those of sub-cluster Al, but are still generally rare with
no clear spatial distribution. The taxa in sub-cluster A3
are more common than the taxa in the other two sub-
clusters, appearing in 30—50% of the sites. In addition,
these taxa tend to occur in the freshwater marsh
(slough), freshwater marsh with hammocks, and brack-
ish marsh vegetation communities (Text-fig. 2). Clus-
ter B is comprised of only chenopods. Chenopods are
correlated most strongly with sites in the freshwater
marsh—slough vegetation community (Text-fig. 2).
However, they occur at all sites and tend to have an
inverse relationship with abundances of Rhizophora.
Riegel (1965) also noted this relationship. Cluster C
consists of taxa that are generally more abundant than
those of cluster A and are consistently found at ap-
proximately 50% of the sample sites. In addition, taxa
from cluster C show a strong spatial correlation with
the freshwater marsh—slough and freshwater marsh
with hammocks vegetation communities (Text-fig. 2).
Finally, cluster D is comprised of taxa that are domi-
nant in mangrove swamp and brackish marsh vegeta-
tion communities (Text-fig. 2).
Cluster Analysis of Sample Sites
The dendrogram from cluster analysis of the 86 sur-
face samples can be divided into ten distinct cluster
groups (Text-fig. 4). Samples in clusters 1, 2 and 3 are
from locations in the mangrove swamp vegetation
community and the margin of the brackish marsh veg-
etation community near the mangrove swamp. Cluster
| is comprised of samples 62—01, 62—06, 62—08, 62
1 62-12. 62-15; 62-175 62-20562—21FvandsG2—22"
These samples are from locations that are | to 2 ki-
lometers from the coast (Text-fig. 2). Cluster 2 is com-
prised of samples 62—2, 62-3, 62—5, 62—7, 62-9, 62
10, 62-13, 62-14, 62-18, 62-19, 64-11, 64-21, SR-
Ol, SR-02, and 59-T1. These samples are generally
from sites along the coastal margin of the mangrove
swamp (Text-fig. 2). Samples in cluster 2 have 35 taxa
present, the widest variety of any cluster. Cluster 3 is
comprised of samples SR-03 to SR-07, SR-09, SR-10,
and SR-13 to SR-16. These samples are from sites
along the mouth of the Shark River where there is a
strong marine influence. The three clusters are domi-
nated by mangroves (Rhizophora, Avicennia, and La-
guncularia), Batis, chenopods (pigweed family), Con-
ocarpus, and Compositae (aster family) pollen types.
The frequency of Rhizophora in cluster | is 20% high-
er on average than in clusters 2 and 3. Cluster 2 sam-
ples differ primarily from cluster | by having higher
frequencies of Avicennia and chenopods. Cluster 3 dif-
fers from clusters | and 2 by possessing more tripor-
ates, chenopods, and Conocarpus (Table 1).
Clusters 4 and 5 contain samples distributed
throughout the freshwater marsh-slough, freshwater
marsh with hardwood hammocks, and freshwater
marsh with cypress hammocks vegetation communi-
ties (Text-fig. 2). Cluster 4 is comprised of samples
62-34, 64-04, 64-07, 64-09, 64-E7SG, 64-15, 64—
16-SG, SLTR-03, and SLTR-04. Cluster 5 is com-
prised of samples 62—23, 64-E7E, 64—10, 64—16AE,
64—-16ASG, 64—16E, SLTR-01, and SLTR-02. AI-
though neither cluster occurs exclusively in a single
vegetation community, samples in cluster 4 are pri-
marily from the freshwater marsh with hardwood ham-
POLLEN ZONATION: O’ NEAL ET AL. 115
Environment Cluster
l
Mangrove 2
Swamp
3
4
Freshwater
Marsh
5
| See
6 SR2¢
Headwater SR-22 —
Marsh sR23 J
7 SRS j
8 1Bai
Brackish e423
Marsh +
9 414 Sli
H
62. =]
Upland 10 ;
[ T T =a 1
0 5 10 15 20
Distance
Text-figure 4.—Dendrogram illustrating association of surface
sample cluster groups to environments. Ten clusters are designated
based on a Chi-squared distance of ~2 (dashed line).
mocks and freshwater marsh with cypress hammocks
vegetation communities, while those in cluster 5 are
mainly from the freshwater marsh slough vegetation
community. The dominant pollen types associated with
freshwater marshes are chenopods, triporates, Nym-
phaea, Sagittaria, cf. Ovoidites, Compositae, and Cy-
peraceae (sedge family). Both of these clusters contain
a wide variety of freshwater-dependent species, but
chenopod abundances are approximately 33% higher
in cluster 5 (Table 1).
Samples from clusters 6 and 7 are from sites located
near the transition between the Everglades Depression
and the mangrove swamp (Text-fig. 1). Although this
area is designated as part of the brackish marsh veg-
etation community by Riegel (1965) (Text-fig. 2), Rie-
gel (1965) also describes this area as a unique zone
called the headwater marsh. The assignment of head-
water marsh will be used to designate this area in this
study as well. Cluster 6 is comprised of samples 64—
33, SR-25, SR-26, SR-28, and SR-29, whereas cluster
7 is comprised of samples 62—24, SR-17 to SR-24, and
SR-30. Samples in cluster 6 were obtained from lo-
cations along the margin between the freshwater
marshes and brackish marsh, while samples in cluster
7 were selected from a more seaward position within
the brackish marsh vegetation community (Text-fig. 2).
The dominant pollen types associated with the head-
water marsh are Rhizophora, chenopods, triporates,
Conocarpus, and Osmunda (ferns). Cluster 6 samples
have more than twice the average frequency of che-
nopods of cluster 7, whereas cluster 7 samples display
a marked increase in Rhizophora and triporates (Table
i):
Cluster 8 groups samples from within the brackish
marsh vegetation communities. This cluster includes
samples HR-O1 and HR-02, TB-01 and TB-02, JR-O1,
62-25, 62—30, 64—18, and 64—22. Taxa common to the
brackish marsh (1.e., Rhizophora, triporates, cheno-
pods, Polypodium (ferns), Conocarpus, and Cypera-
ceae) dominate these samples. However, there is a
wide diversity of taxa present in this cluster (Table 1).
Cluster 9 contains samples that are from the fresh-
water marsh with hardwood hammocks and freshwater
marsh with cypress hammocks vegetation communi-
ties, but are grouped with the brackish marsh cluster.
This cluster contains samples 62—26, 62—33b, 62-35,
64—09H, and 64—14. These samples include a variety
of common freshwater species such as Sagittaria, Um-
belliferae, and Salix (willow family), but they have
more Compositae and less Pinus pollen than is found
in the freshwater marshes. Although these samples
contain a variety of taxa, each has an individual taxon
that comprises at least 25% or more of the sample and
does not conform to any common dominant species
distribution found in other cluster groups (Table 1).
For example, 27.4% of sample 64—14 is Cephalanthus
(Buttonbush) and 26.4% of sample 64—09H is Salix.
Cluster 10 contains samples from sites along the At-
lantic Coastal Ridge within the freshwater marsh with
cypress hammocks vegetation community. This cluster
contains samples 62-27 to 62—29, and EE-11. The
dominant taxon associated with these sites is Taxo-
dium, although there are also a variety of freshwater
components present in these samples (i.e., Gramineae,
Cyperaceae, and Compositae) (Table 1).
Discriminant Analysis
Discriminant analysis correctly assigned all 86 sam-
ples to the environmental groups delineated by cluster
116 BULLETIN 361
Table 2.—Pooled within-groups correlations (loadings) between taxa and discriminant functions. The 18 key contributing taxa are highlighted.
Taxa Function | Function 2 Function 3 Function 4
cf. Ovoidites —0.05813 0.22115* —0.21270 —0.06030
Triporates —0.00777 0.21434* 0.21033 0.04846
Tlex 0.00486 0.07691 0.15489* —0.01152
Utricularia —0.02764 0.09736 —0.10737* —0.00624
Rhabdadenia 0.00280 0.02188 0.05313* —0.01507
Chenopods —0.08009 —0.01600 —0.11739 0.46002*
Taxodium 0.21718 0.02709 —0.22441 —0.24851
Umbelliferae 0.03061 0.03766 —0.03234 0.5863
Pinus —0.06834 0.12545 —0.13453 —0.12608
Salix 0.06142 0.01412 —0.00748 0.06407
Cephalanthus 0.04336 0.01126 —0.00109 0.08223
Composits 0.05375 0.01563 0.07966 0.06185
Sapotaceae 0.02655 0.00773 —0.00527 0.04206
Rhizophora —0.05571 — 0.30883 0.14459 —0.24059
Ficus 0.02610 0.03491 0.06934 0.02143
Sphagnum —0.00375 —0.02183 0.00589 0.00232
Quercus —0.00703 —0.00955 0.02081 —0.09885
Myrta —0.00405 —0.00153 0.02151 0.10310
Osmunda —0.00653 0.02657 0.05460 0.07390
Fraxinus —0.00419 0.01061 0.04842 0.01722
Nuphar —0.00428 —0.00930 0.00679 0.05765
Nyssa 0.02168 0.00030 —0.01830 0.01646
Caryophyllaceae —0.00337 —0.01131 0.02565 —0.00836
Laguncularia —0.01176 —0.05997 0.01209 —0.06271
Carya 0.00979 —0.01318 —0.02525 —0.03736
Rhus —0.00284 —0.00467 0.03766 0.02378
Pteris 0.01606 0.00022 —0.01355 0.01219
Rapenea —0.00428 0.03605 0.02974 0.00614
Ericaceae —0.00287 —0.01593 —0.00225 —0.01830
cf. Thomsonopollis —0.00258 0.03484 0.03885 0.00052
Conocarpus —0.01460 —0.04109 0.14725 — 0.09632
Avicennia —0.02068 —0.11662 0.00641 — 0.08752
Anacardium —0.01116 —0.04637 0.00337 —0.04570
Batis —0.00755 —0.04279 0.00075 —0.01688
Polypodium 0.04074 0.04546 0.05585 0.02431
Cyperaceae 0.00343 0.14085 0.09104 —0.04395
Typha 0.00558 0.05675 0.07508 —0.01937
Gramineae 0.00381 0.04199 0.03805 —0.03976
Vitis 0.02567 0.03047 0.04343 —0.00396
Nymphaea —0.03497 0.08198 —0.12149 0.01812
Sagittaria —0.01212 0.08797 — 0.08782 0.07566
Hamelia —0.00513 —0.00912 —0.01654 0.06530
Ulmus —0.00701 0.02828 —0.02115 —0.01904
Alnus 0.00442 0.02143 —0.00730 —0.05957
Eugenia 0.00908 0.03237 0.03283 —0.01658
Polygona 0.04516 0.01269 —0.03082 0.06516
Morus —0.00386 0.03707 —0.00408 —0.00298
analysis. The first four discriminant functions account
for 92.5% of the variance in the data set and display
high canonical correlations of 0.996, 0.989, 0.981, and
0.971 respectively, underscoring the effectiveness of
the functions in discriminating between cluster group-
ings. The function loadings from the first four func-
tions indicate that the surface cluster groupings are
strongly influenced by 18 key taxa (Table 2). Taxo-
dium, the dominant genus of the freshwater marsh with
cypress hammocks physiographic zone, and chenopods
are the statistically significant taxa on function 1. Sig-
nificant taxa on function 2 include cf. Ovoidites, tri-
porates, Ilex, Utricularia, Pinus, Rhizophora, Avicen-
nia, Cyperaceae, Nymphaea, and Sagittaria. Important
taxa on function 3 are cf. Ovoidites, triporates, Ilex,
Utricularia, chenopods, Taxodium, Pinus, Composi-
tae, Rhizophora, Conocarpus, Cyperaceae, Typha,
Nymphaea, and Sagittaria. On function 4, chenopods,
Taxodium, Pinus, Cephalanthus, Rhizophora, Quer-
cus, Osmunda, Conocarpus, Avicennia, and Sagittaria
POLLEN ZONATION: O’ NEAL ET AL. WaL7/
Function 2
\o
te
0 10
Function 1
71K 9
Function 3
rl
tb
Function 2
Text-figure 5.—(A) Plot of means and 95% confidence intervals
for environmental cluster groups on discriminant functions one and
two. (B) Plot of means and 95% confidence intervals for environ-
mental cluster groups on discriminant functions two and three. Num-
bers refer to cluster groups in Text-figure 4. The symbol used to
identify cluster group 7 (Figure A) encompasses the 95% confidence
interval bars for that group.
contribute most to group separation. Means and 95%
confidence intervals for each group are plotted in Text-
figure 5. A plot of discriminant score group means in
three-dimensional discriminant function space displays
broad spatial separations that underscore the statistical
distinctiveness of the environmental groupings ob-
tained from cluster analysis (Text-fig. 6).
DISCUSSION OF SURFACE POLLEN ZONATION
Visual inspection of the pollen data matrix sorted
by site and taxonomic/morphologic groupings derived
from cluster analysis (Table 1) indicates that there is
a large amount of overlap in the species composition
of the environmental cluster groups. Individual pollen
and spore taxa typically are not environmentally re-
stricted so that no single pollen or spore type can be
used to characterize a particular environment. Rhizo-
phora pollen, for instance, is most abundant in man-
grove swamp clusters where it comprises from 16 to
66% of samples. However, it contributes as much as
17% to samples from brackish marsh cluster 8 and as
much as 29% to samples from the headwater marsh
cluster 7. Chenopods are most common in samples
from the headwater marsh clusters with abundances
ranging between 15 and 90%. However, they also
comprise 7 to 47% of samples from the freshwater
marsh clusters. Similarly, triporates are most abundant
in samples from the brackish marsh clusters, where
they comprise 11 to 23% of the samples, but they com-
prise as much as 21% of samples from freshwater
marsh cluster 4. Because of these occurrence patterns,
visually determined associations of taxa may not be
reliable indicators of environments. Riegel (1965) and
Kuehn (1980) both relied on the appearance of Typha
(cattail) and Conocarpus in vertical sequences to in-
dicate an environmental shift to a brackish marsh. Dis-
criminant analysis indicates that Conocarpus is an im-
portant component of the brackish marsh. However,
our results do not indicate that the abundances of these
species co-vary and suggest that the mere presence of
these species is not a reliable environmental indicator.
In contrast, Rich et al. (1982) interpret cf. Ovoidites
to be an indicator of freshwater environments. Statis-
tical analysis confirms that cf. Ovoidites is an impor-
tant species of the freshwater marsh, but this interpre-
tation is strengthened by the co-occurrence of cf. Ovo-
idites with Sagittaria, Nymphaea, Utricularia and Um-
belliferae (hornwort), all of which are aquatic species
(Text-fig. 3; Table 1). Pollen from the genus Pinus has
a ubiquitous distribution because it is transported read-
ily throughout the system by both wind and water (Ha-
bib et al, 1994). Although it is generally most abun-
dant near its origin in the Pine-Upland vegetation com-
munity, the presence of Pinus in all other vegetation
communities, and its tendency to concentrate along the
coast because of fluvial and marine transport, limits its
usefulness. However, a study in Taylor Slough (which
drains a small portion of the southern Everglades into
Florida Bay) indicates that Pinus pollen comprises a
significant proportion of samples and may be a useful
taxon for paleoenvironmental interpretation there (Wil-
lard and Holmes, 1997). Similarly, trees of the genus
Quercus are physically restricted to freshwater up-
lands. However, Quercus pollen tends to be wide-
spread and present in low abundances, which may also
limit its use for environmental interpretation.
As discussed above, numerically dominant pollen
118
Cluster Group Membership
e Cluster 1
cheno 1) Cluster 2 Mangrove Swamp
Cephalanthus
Sagittaria, \qQ Cluster 3
Osmunda e Cluster 4
8 Freshwater Marsh
© Cluster 5
6 a Cluster 6
les} 1 Headwater Marsh
| 4 o Cluster 7
Qo 4\n Aa Cluster 8 Brackish and
O° & Aa Cluster 9 Transitional Marsh
=
iS g U Cluster 10 | Upland
S
Taxodium S pet
Rhizophora 4 >
e cf, Ovoidites
Pinus, Quercus, S,
Conocarpus, 15 triporates, z
Aiea s6 3 erneracesey eins Percent of Variance
Utricularia i
wa 5 Sagittaria, Function | = 56.76
triporates, //ex
Conocarpus 5
Rhizophora. On
Cyperaceae, S} -10 -10
composits, aw
Taxodium,
. cf Ovoidites, Pinus
Nymphaea,
chenopods,
Utricularia, Sagittaria
BULLETIN 361
vl Nymphaea, Ilex Function 2 = 18.59
oe Function 3 = 10.39
as Function 4 = 6.82
Rhizophora
Avicennia
Text-figure 6.—Three-dimensional plot of environmental cluster-group discriminant-score means with salinity regimes and key taxa displayed
on each axis. Key taxa were identified by running stepwise iterations of discriminant analyses to identify the threshold where 100% discri-
minantion was no longer achieved.
taxa are not restricted to a given cluster. Of note is
also the fact that several species are not present in
certain environments (Table 1). For example, Avicen-
nia never occurs in the freshwater marshes. Lagun-
cularia and Vitis (wild grape) pollen are extremely rare
in the freshwater marshes, appearing in low numbers
in only one sample. Cephalanthus and Salix are not
present in the mangrove swamp or headwater marsh.
However, other pollen and spore types that are restrict-
ed to one, or even two cluster groups, for example
Ericaceae, are uncommon and comprise less than 1%
of any given sample. Thus, their presence appears to
be very sensitive to sampling and preparation proce-
dures and very large sample volumes and counts
would be required to be certain that their absence re-
flects environmental or depositional conditions. Addi-
tional problems may occur in that many rare pollen
types generally comprise less than 2% of a sample and
may not be present in statistically significant numbers
in small samples. In most cases, the identification of a
given environment cannot be based on the presence or
absence of any single taxon and multivariate statistical
analysis of pollen assemblages is necessary.
The surface sample groupings obtained from cluster
analysis do, however, display a spatial distribution that
is strongly correlated with the physiographic provinces
of the study area (Text-fig. 7). This is evident when
the sample sites (from Text-fig. 2) are categorized by
clusters (from Text-fig. 4) and overlaid on the phys-
iographic map of the study area (Text-fig. 1).
In samples from the Mangrove Swamp _ physio-
graphic province, there is a strong correlation between
the frequency of Rhizophora, Avicennia, chenopods,
triporates and Conocarpus in a given sample and the
elevation, salinity, and proximity to major rivers from
which samples were taken. Cluster | contains samples
that are located approximately 1 km from the current
shoreline along tidal channel margins where Rhizo-
phora is dominant. These samples contain an average
of 20% more Rhizophora than samples in the other
mangrove swamp clusters. Cluster 2 samples are from
areas along the coastal margin where there is more
variability in substrate type, tidal influence, and ele-
vation than is seen in cluster | areas. This variability
is reflected in the increased frequency of mangrove
forest taxa (Avicennia, Laguncularia, and Conocar-
pus) in cluster 2 samples. Cluster 3 contains samples
with more triporate and Conocarpus pollen than sam-
ples in cluster 1 and 2. The samples in cluster 3 are
from locations along the mouth of the Shark River, a
primary route of freshwater delivery to the Gulf of
Mexico. Therefore, it is reasonable to expect taxa com-
mon in brackish and freshwater marsh communities
upstream to be present in cluster 3 samples.
Differences between clusters 4 and 5, from the
freshwater marshes, are less evident than for other
POLLEN ZONATION: O’NEAL ET AL. 119
GULF OF
MEXICO
Cluster Group Membership Physiographic Provinces
© Cluster 1
© Cluster 2 Mangrove Swamp ba) Mangrove Swamp
© Cluster 3
@ Cluster 4 ; as) Brackish & Transitional Marsh
E Freshwater Marsh
© Cluster 5 -
mw Cluster 6 y Freshwater Marsh (Slough)
Headwater Marsh
O Cluster 7 (Ezal
A Cluster 8 Bracken and Freshwater Marsh with Hammocks
4 Cluster 9 Transitional Marsh
UT Cluster 10 Upland Ba Limestone Upland Complex
Text-figure 7—Overlay of sample site cluster groupings on phys-
iographic map of southwest Florida showing strong correlation be-
tween environmental pollen clusters and physiographic provinces.
Clusters 1, 2, and 3 are the mangrove swamp; clusters 4 and 5 are
the freshwater marsh; clusters 6 and 7 are the headwater marsh;
clusters 8 and 9 are the brackish marsh; and cluster 10 is the upland
complex.
clusters. The abundance of a variety of freshwater-de-
pendent taxa such as Nymphaea, cf. Ovoidites, and
Utricularia assist in delineating this pollen zone, but
compositional differences between the slough and ad-
jacent marshes with hammocks result primarily from
a large increase in the frequency of chenopods in the
slough. Riegel (1965) attributed the dominance of che-
nopods and Nymphaea in the slough to their need for
a higher water table in this low-lying area.
Differences in the headwater marsh clusters are
based on changes in the frequency of Rhizophora and
chenopods. Chenopods comprise more than 60% of
samples from the headwater marsh (cluster 6) and are
less dominant in cluster 7. Rhizophora, which com-
prises less than 2% of any of the samples in cluster 6,
increases to an average of 16% in cluster 7. The in-
crease in Rhizophora occurs in samples nearer to shore
where the slough shifts to channels and Rhizophora is
localized along the margins.
Samples from the brackish marsh, cluster 8, contain
a wide variety of taxa. Tidal variability as well as fluc-
tuations in freshwater and marine input strongly influ-
ence the taxa that can survive in the brackish marsh
zone. The pollen spectrum resembles the mixture of
taxa that result from in situ deposition in the brackish
marsh (i.e. triporates, Conocarpus, Cyperaceae, Os-
munda, and Polypodium) plus components from the
adjacent mangrove swamp and freshwater marsh en-
vironments. Many samples are dominated by individ-
ual taxa that do not occur in high frequencies else-
where. These samples are grouped primarily in cluster
9 where individual taxa comprise 25% or more of the
pollen totals. Each of these species is associated with
the hardwood hammocks of the freshwater marsh (i.e.
Cephalanthus, Taxodium, and Salix). It is likely that
the broad distribution of taxa in conjunction with the
low frequency of mangrove swamp taxa, resulted in
the strong association of these samples with the brack-
ish marsh rather than with freshwater marshes.
The samples in cluster 10 are from the freshwater
marsh with cypress hammocks vegetation community.
Each sample contains at least 50% Taxodium pollen,
confirming Riegel’s (1965) conclusion that an abun-
dance of this species is indicative of the parent vege-
tation and the environment in which it occurs.
The loadings from the first four discriminant func-
tions indicate that the surface pollen cluster groupings,
or zones, are strongly influenced by 18 key taxa (Table
2). When the taxa are plotted along the functions
(Text-fig. 6) it is evident that they are organized by
salinity regimes. Although, hydroperiod, elevation,
and salinity control the distribution of taxa and limit
the pollen spectra of a sample, the functions only
clearly display the salinity regimes when groups of
taxa are arrayed on each axis towards the regime
where they provide their strongest contribution. Text-
figure 6 displays functions 2, 3, and 4 with the key
taxa that drive each function towards the different re-
gimes. Function 1, which is driven by Taxodium and,
to a lesser degree by chenopods, was not plotted in
the figure because it only excludes cluster 10 from the
other groups (i.e., it serves to separate the Taxodium
dominated hammocks of the freshwater marsh from
other environments). Function 2 is driven by a suite
of taxa from freshwater regimes at one end and dom-
inant marine taxa on the other. Function 3 represents
fresh to brackish regimes and helps separate the fresh-
water clusters from the others. Function 4 separates
clusters located in and near the brackish areas from
those with more freshwater and marine influences. The
taxa that are listed at the end of each axis represent
the key taxa for each regime that drive a cluster in
either direction. Thus, a cluster’s position on each axis
is related directly to the key taxa that are present for
each function. When the clusters are viewed on all
three functions, domains representing different salinity
120 BULLETIN 361
Table 3.—Range and average frequency of the 18 key taxa derived from discriminant analysis of the 86 surface sample sites.
Cluster Rhizophora Avicennia Conocarpus | Chenopod Cyperaceae Nymphaea Utricularia
1 Mangrove Swamp Range 43.7-66.1 0-7.6 0-7.3 2.8—11.8 O-1.7 0-0.6 0-0
Average 57.4 3 3 6.7 0.8 0.1 10)
Std. Dev cs) 2.9 2.5 Poe) 0.5 0.3 (0)
2 Mangrove Swamp Range 11.7-49 0.6-9.6 1.7-11.8 5.9-28.4 0-4.3 0-1.3 0-0
Average BirS 4.3 4.1 14 0.7 0.3 0
Std. Dev 10.2 29) 2.6 6.7 Hil 0.5 0)
3 Mangrove Swamp Range 13.5-52.4 0-3.2 2.5—14.2 6.5—35.3 0-2.4 O-1.1 0-0.5
Average 33.3 12 8.4 16.1 0.9 0.4 0.1
Std. Dev Wile 13) 3.6 8.8 0.8 0.4 0.2
4 Freshwater Marsh Range O-1.2 0-0 O-1.2 2.8-17.3 2.8-12.3 0-8.9 0-1.8
Average 0.3 0 0.5 9.7 6 BoD, 0.7
Std. Dev 0.5 0 0.4 5.8 2.8 SY 0.6
5 Freshwater Marsh Range O-1.2 0-0 0-0.6 22.9—47.2 1.2-5.6 1.7-25.3 0-1.9
Average 0.3 0) 0.1 B27 Qui 9.6 0.7
Std. Dev 0.4 (0) 0.2 tS 1.5 Well 0.6
6 Headwater Marsh Range O-1.3 0-3.4 0-0.6 62.1-89.9 0.3-2.1 0-0.8 0-0
Average 0.6 0.7 0.1 76.3 0.9 0.4 0)
Std. Dev 0.5 IES 0.3 10.5 0.8 0.4 (0)
7 Headwater Marsh Range 2.3-28.7 O-1.8 0-4 14.9-47.7 0.7—4.2 0-2.3 0-0.5
Average 16.1 0.7 1.5 32.4 IES) 1 0.18
Std. Dev 8.8 0.6 1.4 11.5 1.1 0.8 0.2
8 Brackish Marsh Range 2.3-16.9 0-0.6 1.1-19.1 4.8-19.1 0.9-17 0-0.6 0.0
Average 10.2 0.1 6.7 9.4 8 0.1 0
Std. Dev V5) 0.2 6.2 4.3 6 0.2 0)
9 Brackish Marsh Range 0-1.3 0-0 O-1.3 0.5—15.1 1.6-3.6 0-0.6 0-0.6
Average 0.5 (0) 0.5 5.3 Del 0.1 0.1
Std. Dev 0.7 0 0.6 6 0.7 0.3 0.3
10 Upland Range 0-0.6 0-0 0-0.6 0.6-17 0-4.1 0-0 0-0
Avg 0.3 (0) 0.2 12 Py.) 0 0
Std. Dev 0.3 0) 0.3 0.5 2 0 10)
regimes are apparent (Text-fig. 6). Although hydro-
period, salinity and elevation are strongly correlated in
the study area, hydroperiod and elevation differences
do not appear to be controlling distributions that are
identifiable in this form of multivariate statistical anal-
ysIs.
A discriminant analysis of the surface sample data
set based solely on the 18 key taxa produced 98.84%
correct discrimination, with only one sample from
cluster 2 misclassified as a member of cluster 3. This
misclassification is a shift within the mangrove swamp
environment and is not a complete reassignment to a
different environment. These results indicate that anal-
yses like this could potentially be completed using
only the 18 key taxa in Table 3, although none of these
taxa are restricted to a single cluster. Analysis of all
of the taxa in a sample may be required to determine
relationships between samples. Text-figure 8 is a gen-
eralized schematic of the Everglades physiographic
provinces with the key pollen types that help delineate
each area. This conceptual diagram can be used to de-
termine the distribution and relative importance of the
18 key taxa in the different physiographic zones. A
primary reason for this is that a variety of taxa are
present in relatively low abundances in some clusters
such as the freshwater marsh clusters 4 and 5. This
problem reiterates the need for high pollen counts. The
suggested minimum count of 150 grains per sample
used by Riegel (1965) is still very low.
PART B: APPLICATION TO VERTICAL
SEDIMENTARY SEQUENCES
In order to assess the value of the approach for pa-
leoecological interpretations, pollen data were collect-
ed for two cores. The first, from the coastal margin of
the Harney River (referred to hereafter as 59-T1) was
collected by Reigel (1965). The second, core 9411-7,
was collected from the Gopher Key region for a sed-
imentologic study by Gelsanliter (1996) (Text-fig. 1).
Both are long cores, taken to the Pleistocene bedrock
surface and potentially contain sediments that record
much of the late Holocene depositional history of
southwest Florida.
Core Descriptions
The sedimentological description for core 59-T1 is
presented in Riegel (1965). The core is 3.97 m in
POLLEN ZONATION: O’ NEAL ET AL. 121
Table 3.—Extended.
Cephalan-
Ilex Triporates Taxodium Ovoidites Pinus Sagittaria Composits thus Quercus Typha Osmunda
0-0 0-0 0-3.2 0-0 2-20.7 0-0 1.8-6.1 0-0 0-1.9 0-0.6 0-0.6
(0) 0) 1.6 0 8.7 2.9 0) 0.6 0.1 0.1
0 10) 1.1 0 S)si/ il} 0) 0.7 0.3 0.2
0-0.6 0-3.9 0-4.5 0-0.9 5.3-27.4 O0-0.7 0-9.6 0-0.5 0-4.3 O-0.7 O-18.1
0 0.9 NES) 0.1 12.8 0.1 i) 0 2.2 0.2 1.8
0.2 1.3 1.4 0.3 5.6 0.2 2.3 0.1 1.4 0.3 4.6
0-0.7 2.9-13.9 O0-1.3 O-1.7 6.3-17.3 O-1.8 0.6—4.2 0-0 0-3.2 0-0.7 0-0
0.1 Se) 0.7 0.6 10.2 0.5 2.2 0 1.4 0.1 0
0.3 3 0.5 0.6 Bul 0.6 13 0 Vil 0.2 0
0-1.2 6.9-21.2 0.5-5.3 2.9-15.8 13.9—47.8 0.9-11.5 0.5—4 0-1.7 0.5-3.2 0-6.1 0-6.3
0.3 11.7 2.3 10.1 28.3 4.8 2:2, 0.6 1.4 0.9 1.6
0.4 4.8 1.5 4.2 10.1 4.4 ile? 0.6 0.8 2 22
0-0 0.8-13.5 0-3.9 1.4-12.8 16.8-31.6 0.5—5 0-2.7 0-0.5 O-1.4 0-0.8 0-2.6
0 5.5 1.9 VS) 235) DES I 0.1 0.6 0.2 1
10) 4.4 1.4 33 4.6 1.7 1 0.2 0.5 0.3 1.1
0-0 0-3.9 0-2.2 0-0.8 1.6-8.2 0-S.9 0-4 0-0.4 0-0.3 0-0.6 O-11.1
0 2 0.9 0.2 Su 2.9 IES 0.1 0.1 0.3 25)
0 1.4 1 0.3 2.7 2.1 1.7 0.2 0.1 0.3 4.8
O-1.4 5.5—20.4 0.7—4.7 0-2.7 3.6—10.2 0.6-3.6 1-10.3 0-12.5 0-2 0-2.3 1.8-10.1
0.6 10.9 2.4 0.7 6.7 1.6 3.9 1.6 0.9 0.5 5.2
0.5 4.1 1.4 0.9 2.4 0.9 2.8 3.9 0.7 0.7 2.6
0-4.1 11.1-23.3 0-8.3 0-4.5 4.1-19.2 0-0.9 1.8-7.4 O-1.1 0-3 0-7.3 0.6-10.1
1.5 18.6 2.8 0.8 9 0.3 4.4 0.2 3 DED, 3.7
1.3 3.9 2.4 1S 4.5 0.4 2.1 0.4 1 2.8 333
0-0 0.5—14.5 0-32.1 O0-1.9 0.5—-8.2 0-8.3 3.6-12.7 0.5-27.4 O0-1.3 O-1.6 0-2.8
0 5.6 8.6 0.9 5.2 2S) 8.1 8.1 0.7 0.4 0.8
0 2)53) 13.4 0.9 Shi 3.3 4.3 11.2 0.5 0.7 1.2
0-0:6 0.6—2.3 55.4-87.4 O-1.3 0.6-7 O-1.2 1.2-4.6 0-4.1 0-1-7 0-0.6 0-0.6
0.2 1.3 72.2 0.5 3.2 0.5 PES) 1.7 0.9 0.5 0:2
0.3 0.7 1341 0.6 Del] 0.6 Iles) Dal 0.7 0.3 0.3
length (Text-fig. 9). Its base contains a mixed marl and
peat (3.97 m to 3.73 m) overlain by a pure peat that
grades into a muddy mangrove peat (3.73 m to 0.76
m), and is capped by organic-rich carbonate muds
(0.76 m to the surface in 1965).
The sedimentological description of core 9411-7 is
presented in Gelsanliter (1996). This core is 2.63 m in
length (Text-fig. 10). The base of the core (between
2.63 m and 2.10 m) consists of quartz sands overlain
by a red mangrove peat (between 2.10 m and 1.74 m).
A gray mudstone to wackestone is present between
1.73 m and 1.50 m. Between 1.50 m and 1.37 m is a
red mangrove peat overlain by a gray mudstone to
wackestone (1.37 m and 0.97 m) that grades into a tan
mudstone to wackestone (0.97 and 0.75 m). Capping
the sequence, between 75 cm and the depositional sur-
face in 1994, is a red mangrove peat.
Laboratory Methods for Pollen Analysis of
Core Samples
Identification of pollen and spore types in the cores
followed standard procedures (Moore et al., 1991;
Wood et al., 1996). Core sub-samples of a known vol-
ume were taken with a 10-cm? syringe, water content
was determined, and bulk density calculated for sub-
sequent count normalization. Organic content of pre-
viously dried sediment was estimated from weight loss
on ignition at 450°C for 6 hours (Dean, 1974). Initially,
1 g sub-samples were processed for pollen concentra-
tion. However, the large quantity of organic material
present in all samples yielded low pollen concentra-
tions using standard processing techniques. Using
standard methods on larger quantities of material (7.e.,
>1 g) often resulted in incomplete digestion of organic
debris that masked the presence of pollen. Therefore,
multiple, smaller sub-samples of approximately 0.2 g
to 0.25 g were dried, disaggregated and spiked with a
Lycopodium yield tracer tablet to facilitate the calcu-
lation of absolute pollen concentrations (Stockmarr,
1971). Lycopodium tablets were dissolved and calcium
carbonate was removed from the sample by the addi-
tion of 10% HCl. All samples were treated with HF
in a hot water bath to remove silica. After neutraliza-
tion of the acid, samples were dried and treated in a
hot water bath with 5% KOH for 3 minutes. Samples
were sieved with a 200 pm screen to remove large
122, BULLETIN 361
Upland Freshwater
P Marsh Slough
with Hammocks
Freshwater Marsh ° Headwater ' Brackish and
Marsh Transitional
Mangrove Swamp
High Tide
Marine | Pollen Type
Brackish
| Ric ophora
Avicennia
a | Conocarpus
Osmunda
composits
triporates
Ilex
Cyperaceae
Typha
Cephalanthus
chenopods
Sagittaria
Nymphaea
Utricularia
cf. Ovoidites
| Taxodium |
Pinus
| Quercus
Text-figure 8.—Conceptual diagram displaying the distribution of the 18 key taxa across the major physiographic zones of the study area.
Shading on bars represents relative within-taxa frequency in a given environment (darker = more abundant).
organic detritus. After sieving, the samples were ace-
tolyzed in a hot water bath for 3 minutes. After neu-
tralization, pollen concentrates were dehydrated with
alcohol, suspended in silicon oil, stained with safranin
and mounted on microscope slides for examination.
Sample Counts
Pollen identification for core 941 1—7 was completed
at Indiana University/Purdue University at Indianapo-
lis. Pollen grains were identified under incandescent
and fluorescence light microscopy by comparing them
to a reference set of tropical pollen consisting of her-
barium samples from Everglades National Park, a set
of modern pollen taxa provided by D. Kuehn (Western
Kentucky University), and a computerized database.
Under fluorescence, details of exine microstructure and
ornamentation are enhanced thereby facilitating grain
identification.
Core 59-Tl was processed and counted by Riegel
(1965). Riegel (1965) indicates that standard prepara-
tion procedures were followed and counting was con-
ducted on at least three slides per sample. For each
site, Riegel (1965) attempted to count a minimum of
150 grains. Due to the scarcity of grains in some sam-
ples, on occasion, counting was stopped after a mini-
mum of 100 grains. Riegel (1965) did not use marker
grains; thus absolute pollen concentrations are not pre-
sented. In total, Riegel counted 28 samples for core
59-T1 (Table 4).
For core 9411—7 every attempt was made to count
300 grains for each sample interval. In total seventeen
samples from this core were counted (Table 5). For
most samples however, the scarcity of pollen necessi-
tated lower counts (range 102 to 306). Most palynol-
ogists working in South Florida attempt to identify a
minimum of 300 grains, although many data sets are
characterized by low pollen sums of commonly around
150 grains (Riegel, 1965; Kuehn, 1980; and Willard
and Holmes, 1997).
Statistical Analysis of the Core Data
Discriminant analysis was used to assign core sam-
ples to surface pollen zones, identified earlier in this
paper, in order to identify environmental changes in
the vertical sequences over time. The input data were
the frequency counts. The data sets for the cores were
reduced to the same 48 taxonomic and morphologic
pollen and spore categories presented in the first sec-
tion of this study. Discriminant function scores were
calculated for each sample by transforming the original
set of measurements into a single value on each func-
tion. Using these scores, each sample was assigned to
the surface cluster group with which it was most sim-
POLLEN ZONATION: O’ NEAL ET AL. |
i)
>)
-
Core 4 . Depth A 5 y
Derrintion Cluster Physiographic Zone eS Sediment Type Age Cluster} Physiographic Zone
2 Mangrove Swamp (Coastal)
2 Mangrove Swamp (Coastal) 10. =| present
4 Freshwater Marsh with Hammocks 5 pe) 6 Headwater Marsh (Upper)
Organic-rich “ : 20. 3 Sa! "4,
Gashounte 2 Mangrove Swamp (Coastal) 30 Mangrove Swamp (Channel)
Muds Z Mangroves wauin) constal) 40 =| Red Mangrove Peat 2 Mangrove Swamp (Coastal)
: Mangrove Swamp (Coastal) 50am 2 Mangrove Swamp (Coastal)
Mangrove S' Inner) =
: bah eer Naas 60. | 2 Mangrove Swamp (Coastal)
2 Mangrove Swamp (Coastal) ==]
70S 1 Mangrove Swamp (Inner)
2 Mangrove Swamp (Coastal)
80 Excluded / Low Pollen Sum
2 Mangrove Swamp (Coastal) ~_| Tan Mudstone to
2 Mangrove Swamp (Coastal) pOgeres| Wackestone Excluded / Low Pollen Sum
Muddy Peat with 2 Mangrove Swamp (Coastal) 100 __| =
Mangrove Roots aa) 2 Mangrove Swamp (Coastal)
2 Mangrove Swamp (Coastal) LLU
_| Gray Mudstone to | 2,390+60
1 Mangrove Swamp (Inner) 120 peewee] Se aa YBP 2 Mangrove Swamp (Coastal)
2 Mangrove S Coastal =
angrove Swamp) (Cosetal) 130, 2: Mangrove Swamp (Coastal)
= 1 Mangrove Swamp (Inner) 140 _]
€ 1 Mangrove Swamp (Inner) 1500 —) Red Mangrove Peat 2 Mangrove Swamp (Coastal)
3 =
a=] Mang: S 4 . 2,880+60 2, Mangrove Swamp (Coastal)
= 1 Aangrove Swamp (Inner) 160+ Gray Mudstone to ans
cs) 2 Mangrove Swamp (Coastal) 4 Jacke 2 Mangrove Swamp (Coastal)
170 Wackestone g Pp
3 Mangrove Swamp (Channel) 180 4
Red Mangrove 1) aneroNe Sere (nes) 190° 4 sone Excluded / Low Pollen Sum
Peat 3 Mangrove Swamp (Channel) _| Red Mangrove Peat Sam
2 Mangrove Swamp (Coastal) 200 __] YBP
8 Brackish 210° 2 Mangrove Swamp (Coastal)
8 Brackish 220) aaa
4 Freshwater Marsh with Hammocks 23 0sums|
Mixed Marl 8 Brackish 240 "| Quartz Sand
and Peat 8 Brackish 950 7 8 Brackish
Text-figure 9.—Stratigraphic column of core 59-T1l (modified 260... Excluded / Low Pollen Sum
from Riegel, 1965) and surface cluster group associations for sam- 7 Bedrock
ples. Sample depths are indicated in gray.
ilar. All multivariate statistical analyses were complet-
ed using SYSTAT version 6.0 (SPSS Inc. 1996).
Results for Vertical Sedimentary Sequences
The results of discriminant analysis of core 59-T1
with the surface pollen zone data set assigned samples
84-92 cm, 190-198 cm, 221—229 cm, 236-242 cm,
251-259 cm, and 297-305 cm to the mangrove swamp
surface cluster group | (Text-fig. 9). Samples O—8 cm,
8-15 cm, 38—45 cm, 53-61 cm, 69-77 cm, 99-107
cm, 114—122 cm, 130-138 cm, 145—153 cm, 160—168
em, 175-183 cm, 205-213 cm, 266—272 cm, and 320—
328 cm were assigned to the mangrove swamp surface
cluster group 2. Samples 282-290 cm and 312-320
cm were assigned to the mangrove swamp surface
cluster group 3. Samples 22—30 cm and 366-374 cm
were assigned to the freshwater marsh surface cluster
group 4. Samples 335-343 cm, 351—359 cm, 381—389
cm, and 389-397 cm were assigned to the brackish
and transitional marsh cluster group 8 (Text-fig. 9).
The discriminant analysis of core 9411—7 with the
surface data set grouped sample 69 cm with the man-
grove swamp surface cluster group | (Text-fig. 10).
Samples 35 cm, 46 cm, 58 cm, 104 cm, 120 cm, 130
Text-figure 10.—Stratigraphic column of core 9411-7 (modified
from Gelsanliter, 1996) and surface cluster group associations. Sam-
ple depths are indicated in gray.
em, 145 cm, 155 cm, 165 cm, and 205 cm were
grouped with the mangrove swamp surface cluster
group 2. Sample 21 cm was grouped with the man-
grove swamp surface cluster group 3. Sample 10 cm
was grouped with the headwater marsh surface cluster
group 6. Sample 248 cm was grouped with the brack-
ish and transitional marsh surface cluster group 8.
Samples 79 cm, 90 cm, and 188 cm were excluded
from discriminant analysis because of low pollen
counts (Text-fig. 10).
DISCUSSION
Discriminant analysis of samples from core 59-T1
with the surface data set indicates that the overall se-
quence went from basal brackish and freshwater marsh
environments to dominantly mangrove swamp envi-
ronments towards the present surface. These mangrove
environments shifted from mixed mangrove environ-
ments to predominantly coastal environments upwards
in the core. This sequence suggests an overall trans-
gression of southwest Florida, an interpretation sup-
ported by Riegel (1965) and Spackman er al. (1966)
BULLETIN 361
124
90 VC 6s 00 Pl 81 (ETE (G)! 00 00 VC (Gi)! Wd L6E—-86€
Se Site IS 90 OTI eal: (ew) 00 00 90 00 $6 W9 68E-18e
SO EXE 86 00) £6 Cy! rl 00 a0) 00 00 LG wd pLe—-99e
FO ST V8 L0 (46 Jer! ool 00 00 L0 LO NS wd 6SE-1SE
00 00 s0 (ONE CLI 00 CLI 00 OT s0 sO cel wid EpE-SEE
00 90 00 00 90 00 96 OG cal all 90 ses Wd 8Ct—-OCE
00 00 00 90 Oe Gl OTT 00 00 00 (Gli 88P wo Q@E-CIE
00 00 00 90 00 cl OCI 00 90 00 90 pss wd COE—-L6~
00 00 00 00 00 Cl Ol TI 90 90 90 £9 W9 06T-T8T
00 00 00 00 Vl Dal ost ry! [E\! (ae Ei VEP Wd CLT-99C
00 90 00 00 00 90 OCI 90 (e! 00 vY 8°TS wd 6S7-IST
00 90 00 00 90 90 L6 00 Ul 90 ve OL9 Wd CHT-9ET
00 00 00 00 00 90 6Cl 00 Dae ei! ve 109 WIS 6CC~I1CC
00 00 00 00 Ve 00 O9T 00 90 cl Up 6LY wd €1c-SO0C
00 SO OT 00 Ol ol (Si! 00 OT ol (SAL eLy WW 861-061
00 00 SO 00 SO ima 6s 00 SO a EG L8e wd €8I-SLI
00 90 00 00 90 90 Gi Vl ST LE £8 9 eV wd 891—O9I
00 00 00 00 cl 61 90 L8 61 cl 89 CLV wd €CI—-SPrl
00 00 00 00 61 00 SYP 60 se ete 66 OEP wo §el—-Oel
00 SO a0) 00 en 00) ES ima HEX (NS Lol Cir ws @cI-vII
SO 00 00 00 00 Ol OV 00 Ore ST 9TI £0S wu 101-66
00 00 00 00 00 SO ts al OT oO SE! [ESS WS 6-8
00 L0 L0 ry L0 00 Or 81 Ul LO | Lt w9 11-69
00 00 00 00 FO vO 08 el Ge Te Cl VB8e wo 19-€¢
00 00 00 00 Vl 00 9S 00 OPI! GG [Ee Or Wd Cp-8e
00 00 00 00 Gi Gl ONS 90 eV Gn os Tle wd 0€—-CC
90 00 00 00 90 00 oH, 00 ONE 90 oP VCP wd CI-8
90 00 Gil 00 00 Gil T6 00 VT 68 19 L0¢ wd 8-0
Arey peoymouy ATI] Jey pened Ajwuey aspag A[turey sseig spodousyD HOMeS poomuoyNg c2oIsURLY SAOISURIY dAois UR ajdures
Aasied DUDINISDS vanydwcy pyda I arooriedAD = aROUTLURID suvg sndip20u0) aU, yori pow
aviajfy[oqui~) D1 piuuaaAy vsoydozyy
-{D]NIUNSDT
“T.L-6§ 2109 107 wep Aouanboiy us]][od—t 29PL
WY
N
O’NEAL ET AL.
POLLEN ZONATION
00 00 00 90
00 00 00 Sal
00 00 00 00
00 00 00 vl
00 00 00 OT
00 00 00 (elt
00 90 00 1
00 00 00 90
00 00 00 cl
00 00 00 Vl
00 00 00 90
00 00 00 90
00 00 00 00
00 00 00 00
00 00 00 0)
00 00 00 00
00 00 00 00
00 00 00 90
00 00 00 a0)
00 00 00 a0)
00 00 00 00
00 00 00 a0)
00 00 00 00
00 FO 00 80
00 00 00 ol
00 00 90 VC
00 00 00 6€
00 00 00 61
SNUIXDAY snylupjpyday viuasny soqeiodiy,
00
00
00
r0
00
00
90
00
00
90
00
00
00
00
00
00
00
00
sO
00
00
SO
r0
00
SO
90
00
90
SNL]
00
00
00
00
a0)
00
00
00
90
00
00
90
00
00
00
a0)
90
90
00
00
00
00
r0
00
00
00
00
90
sniopw
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
i
00
00
50
00
00
50
r0
r0
50
90
00
i)
MOTTIAN
xyes
90
ox)
ox)
00
ox)
00
00
00
00
00
00
00
00
00
00
ox)
00
00
00
00
00
00
+0
ox)
00
00
00
90
ATIOH
xal]
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00 DE 00 Wd /6¢-86£
00 8s 00 Wd 68E-I8E
00 or 0) Wd plLe—-99E
00 ev a0) wd 6CE-ICE
00 CT 00 wd €pE-CEE
00 VT 00 wid 87E-OTE
00 ST 00 wo OCE-TIE
00 00 00 wd COE—-L6T
00 90 00 Wd ()6T-T8T
00 90 00 wd TL7—-99T
00 El 00 Wd 6CT-IST
00 00 00 Wd ThT7—-9ET
00 Gi! 00 Wd 677-177
00 90 00 Wd €1Z—-SOT
00 OT 00 WO 861-061
00 00 00 wo €8I-CLI
00 00 00 Wd 891-091
00 00 00 wo €CI—Crl
00 OT 00 wo §¢I—-O€l
00 OT 00 wo CCI-¢I11
00 00 00 wd 101-66
00 a0) 00 Wd 76-8
00 00 00 wid //—-69
00 80 00 wd 19-¢€¢
00 0) 00 Wd Ch-8e
00 ST 00 Wd O€-CT
00 (Gl 00 wo CI-8
00 Oe 00 wd 8-0
sysoduioj = woMIappel_ ajduies
araory|AydoAiea) =paam yRuUIS
arooruosA|og
DIADINII)
‘papuarxy—p a1qeL
BULLETIN 361
126
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
snyy
00 00 00 LAV,
00 00 00 90
00 00 00 ONE
00 00 00 vt
00 00 a0) s0
00 00 00 00
00 00 00 00
00 00 00 00
00 00 00 00
00 00 00 00
00 00 00 00
00 00 00 00
00 00 00 90
00 00 00 00
00 00 00 00
00 00 00 00
00 00 00 00
00 00 00 90
00 00 00 00
00 00 00 Ul
00 00 00 00
00 00 00 00
00 00 00 00
00 00 00 00
00 00 00 a0)
90 00 00 00
00 00 00 el
00 00 00 00
pIawuvoy pauadpy adrin SINPIOAQC, “JO
SHIA
(G! 00 00 00
00 00 00 00
00 00 00 00
vO 00 00 vO
00 00 00 00
90 00 00 90
00 00 00 90
00 00 90 00
90 00 00 00
90 00 00 00
00 00 00 00
00 00 00 90
90 00 00 90
00 00 00 00
SO 00 00 00
0) 00 00 00
ny 00 00 90
cl 00 00 00
vl 00 00 00
SO 00 00 00
00 00 00 00
ol 00 00 00
LO 00 00 00
(GE 00 00 00
0) 00 00 00
(Ql 00 00 00
Sal 00 00 00
SC 00 00 00
ssaidAp DSSANT DMD IOplv
UNIPOXD | snuyy
00
00
00
FO
00
00
90
00
cl
90
00
00
90
00
iS"
00
ea
00
val
EG
00
Ol
81
(|
91
81
90
(|!
unipibovUuy
Cts
ST
oO
00
VC
61
ONS
00
VT
90
OG
90
00
SI
ia0)
at
6T
SiG
61
00
0:0
a0)
Tl
60
OT
VC
DME
cae
APO
SNIAANE)
00
00
00
00
00
00
90
00
00
Lae
00
00
00
00
0
Ul
00
00
00
00
00
0
00
FO
sO
00
90
90
avaonjodes
WS /6£-86e
W9 68E-18e
wd pLE—99E
W9 6¢e-1Se
WOlEveSSce
WS 8CE-0CE
wd OCE-CLE
WS COt-L6C
WS ()6T-C8T
WS CL7-997
wd 6SC-ISZ
WIS CVC—-9ET
WS 6¢CC-1C7
ws €1c-S07
WW 861-061
WS €8I-SLI
wW9 891-091
wo €CI—SrI
m9 8el—Oel
wd CCI-FIT
WS 101-66
wd ¢6-8
wo //—69
wd T9-€S
wd Ch-8e
wld 0E—CC
uid ¢[-8
wuld g-()
ajduirs
‘popuarxq— hp alge
127
POLLEN ZONATION: O’NEAL ET AL.
OSI 691 Vil 0'6C 00 00 00 00 00 00 00 00 00 00 00 WS L6E-86E
trl SSI S6 OBE 00 00 00 00 00 00 00 00 00 00 00 WS 68¢-18e
Ss tol 6 Me) 00 00 00 00 00 00 00 00 a0) 00 00 WO PEESOOS
O8T S8C [Es V8c 00 00 00 00 00 00 00 00 00 00 00 WS 6S ESSE
86l £0C VC O9¢ 00 00 00 00 00 00 00 00 00 00 00 WE) SASL
L¥rl 9S HENS Pst 00 00 00 00 00 00 00 00 90 90 00 Wo"8CE—0CE
9S I F9l TS) £81 00 00 00 00 00 00 00 00 00 00 00 WOO CtROhs
OSI 991 L6 L8i 00 00 00 00 00 00 00 00 00 00 00 WS SOESLOC
Srl P9l Sl SOI 00 00 00 00 00 00 00 00 90 00 00 WS ()67-C8C
SSI SLI Fil (ENS | 00 00 00 00 00 00 00 00 90 00 00 KCL GRI9G
trl 6SI Vol 9TI 00 00 00 00 00 00 00 00 90 00 00 WS 6ST-1ST
col OLI GE Ls 00 00 00 00 00 00 00 00 Ul 00 00 WOTCVG-9GG
8S 8ZI Cal LEAS 00 00 00 00 00 00 00 00 00 00 00 WO 6CCaCS
Srl 691 cri Galil 00 00 00 00 00 00 00 00 00 00 00 Wo €TC-S07
esl SOT 601 LOl 00 00 00 00 00 00 00 00 00 SO 00 W9 §61-061
891 981 66 6Cl 00 00 00 00 00 00 00 00 a0) a0) 00 0) SS S/O!
991 18] £8 ULI 00 00 00 00 00 00 00 00 90 90 00 Wo 891—-09T
srl 19] v8 orl 00 00 00 00 00 00 00 00 om 00 00 WOES TSSyI
981 LIT oll ST 00 00 00 00 00 00 00 00 60 00 00 wo SE I—-OET
O91 L8I LYl LOI 00 00 00 00 00 00 00 00 a0) a0) 00 Wo ColSV PT
CLI 661 Cel oT 00 00 00 00 00 00 00 00 a0) 0) 00 W195" 201566
Os rol OL Vel 00 00 00 00 00 00 00 00 a0) 00 00 WS" C6578
CET PLE Psi 601 00 00 00 00 00 00 00 00 00 r0 00 COT 69)
961 PCC Ll 6Cl 00 00 00 00 00 00 00 00 00 00 00 ws [9-E¢
orl c8I Vs Srl 00 00 00 00 00 00 00 00 ol 00 00 Wd Cr-8E
trl rol STI OST 00 00 00 00 00 00 00 00 90 (Gi) 00 MSIOESCC
cri cs] 18 DOT 00 00 00 00 00 00 00 00 90 00 00 ws CI=8
Isl £9] UL SS 00 00 00 00 00 00 00 00 cl 81 00 eee) t=(0)
sonsneis 1of payunod aI, SNUI aq SAL piuap — Avydnny syjjodouoswoyy, wy wnuspydg — Ayturey wnip — ppuntusEQ. aeaoeyK|y ajdurg
posn suieig = suring -wppinbiT -Dpqvyuy ‘yo smu) ueoH -od\jog
avaorol”
‘papuaxq—"p 2qRL
128 BULLETIN 361
Table 5.—Pollen frequency data for core 9411-7.
Rhizophora Avicennia Laguncularia
Gramineae Cyperaceae
Sample Red Black White Conocarpus — Batis Cheno- Grass Sawgrass Typha Nymphaea Sagittaria
depth Mangrove Mangrove Mangrove Buttonwood Saltwort pods family family Cattail Water Lilly Arrowhead
10 cm 24.3 1.9 1.3 4.2 0.3 51.8 0.6 129) 1.3 0.0 1-3
21 cm 24.4 1.7 0.4 1.2 0.0 31.8 0.4 12 0.0 0.8 0.4
35 cm 15.1 5.8 1.2 5.8 0.6 33.1 3i5 2.9 3:5 0.0 1e7/
46 cm Syle3} TES 0.0 1.5 0.0 29:9 37 0.0 0.0 0.0 0.7
58 cm 37.9 Del 0.7 2.8 0.0 32.4 2.1 0.7 0.0 0.0 0.0
69 cm 41.1 3.1 1.2 1.2 0.0 25.8 DES Pies) 0.0 0.0 0.0
79 cm 34.8 6.8 7.6 DS 0.0 12.9 eS) 0.8 0.0 0.0 0.0
90 cm 9.3 10.1 0.8 1.6 0.8 25.6 1.6 0.8 0.0 0.0 0.8
104 cm 17.4 SS) 2.2 0.0 0.7 15.9 0.0 4.3 0.0 0.0 0.0
120 cm 19 16.3 0.0 1e3} 1.3 11.3 6.3 6.3 0.0 0.0 0.0
130 cm 2235 10.3 1.1 ile 7/ 1.1 8.0 Pes} 1.1 0.0 0.0 0.0
145 cm 28.0 8.4 2.8 4.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0
155 cm 24.6 8.4 NeF/ 1.1 0.0 5.6 1.1 1.1 0.0 0.0 0.0
165 cm 26.9 5.4 sf 1.1 0.5 O77 0.5 1.1 0.0 0.0 0.0
188 cm 29.3 0.0 0.7 4.1 0.0 0.0 0.0 0.0 2.0 0.0 0.0
205 cm 19.7 0.0 0.0 0.0 Le 22.5 45 Pep2 i697 0.6 1.7
248 cm Sul 0.0 1.3 0.0 0.0 3.8 3.8 11.5 0.0 4.5 3.8
Sample Umbelliferae Utricularia Polygonaceae Ilex Salix
depth Celery family Bladderwort Composits Smart weed Holly Willow Morus Ficus Triporates
10 cm 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.0
21 cm 0.4 0.0 1.2 0.0 0.0 0.0 0.8 0.0 2.1
35 cm 0.0 0.0 0.0 0.6 0.0 0.6 0.0 0.0 122:
46 cm 0.7 0.0 0.7 0.0 0.0 0.0 0.7 0.0 0.7
58 cm 0.0 0.7 0.0 0.7 0.0 0.0 0.0 0.0 0.7
69 cm 0.0 0.0 1.2 0.0 0.0 0.0 0.0 0.0 le
79 cm 0.0 0.0 0.8 0.0 0.0 0.0 0.0 0.0 0.0
90 cm 0.0 0.0 23 0.0 23 0.8 0.0 0.0 2.3
104 cm 0.0 0.0 D2, 0.0 0.0 1.4 1.4 0.0 2.9
120 cm 0.0 0.0 6.3 0.6 0.0 4.4 0.0 0.0 0.6
130 cm 0.0 0.0 1.1 0.0 0.6 1.1 0.0 0.0 4.0
145 cm 0.0 0.0 0.0 1.4 0.7 0.0 0.0 0.0 0.0
155 cm 0.0 0.0 0.6 0.0 0.0 0.0 0.6 0.0 0.0
165 cm 0.0 0.0 1.1 0.0 0.0 0.0 0.0 0.0 0.0
188 cm 0.0 1.4 2 0.0 0.0 0.0 0.0 0.0 0.7
205 cm 0.0 0.0 NZ 0.6 0.0 0.6 0.0 0.0 Dee
248 cm 0.0 0.0 12.7 0.0 0.6 0.0 0.0 0.0 i)
for the same core. Discriminant analysis using the sur-
face pollen zonation is a sensitive enough measure to
differentiate generally between the mangrove-domi-
nated assemblages that are prevalent in the core. The
assignment of core intervals between 328 cm and 190
cm to a mixture of mangrove swamp surface clusters
(Text-fig. 9) is expected during the early stages of
coastal transgression when variability in substrate type,
hydroperiod, elevation, and varying proximity to
brackish and freshwater marsh communities would
contribute a variety of taxa to the sediment record. The
upper 190 cm of core (essentially the last 2,800 years;
Text-fig. 9) is assigned almost exclusively to mangrove
swamp surface cluster 2, or that cluster represented by
the coastal mangrove physiographic province, the de-
position site of the core. This suggests that essentially
modern conditions existed in the vicinity of the Harney
River coastline for the last 2,800 years, a time surface
that occurs within the muddy mangrove peat unit and
is otherwise indistinguishable.
Discriminant analysis of core 9411—7 with the sur-
face pollen data set indicates a predominantly man-
grove coastal environment of deposition for the core.
However, core 9411-7 has been interpreted by Gel-
sanliter (1996) to contain sedimentologic evidence that
suggests a short-lived, high-frequency sea-level oscil-
lation occurred in the overall transgressive sequence
at approximately 2,400 ybp. Sedimentologic evidence
for this sea-level oscillation first occurs in the form of
Rhizophora peats prograding over subtidal marine car-
POLLEN ZONATION: O'NEAL ET AL. 129
Table 5—Continued.
Sample Quercus Alnus Taxodium
depth Cephalanthus Fraxinus Sapotaceae Oak Anacardium Tricolpates Alder Carya Pinus Cypress
10 cm 0.0 0.0 0.0 0.3 0.0 1.0 0.0 0.0 2.9 1.0
21 cm 0.0 0.0 0.4 0.0 0.0 0.8 0.0 0.4 AUS) 33
35 cm 0.6 0.0 0.0 0.0 0.0 1.7 0.0 0.0 4.7 5.8
46 cm 0.0 0.0 3S 0.0 0.0 0.0 0.7 0.0 2.2 1S
58 cm 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.8 Ql
69 cm 0.0 0.0 0.0 0.0 0.0 0.6 0.0 0.0 Sh7/ 0.6
79 cm 0.0 0.0 0.8 3.0 0.0 1.5 0.0 0.0 1.5 0.0
90 cm 0.0 0.0 0.0 0.0 0.8 1.6 0.0 0.0 6.2 0.0
104 cm 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 11.6 0.0
120 cm 0.6 0.0 0.0 0.0 0.6 0.6 0.0 0.0 6.3 0.0
130 cm 1.1 0.0 0.0 0.0 0.0 2.9 0.0 0.0 15.4 0.0
145 cm 0.7 0.7 0.0 0.0 0.0 0.0 0.0 0.0 24.5 0.0
155 cm 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 28.5 1.1
165 cm 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 25.3 0.0
188 cm 0.0 1.4 0.0 0.0 0.0 0.0 0.0 0.0 15.6 2.0
205 cm 0.0 0.0 0.0 0.0 0.0 0.0 0.6 0.0 5.6 0.0
248 cm 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 12.1 0.0
Sample Vitis cf. Polygala Others and Grains used
depth Monocolpates Grape Inaperturate Ovoidites Milkwort Osmunda Polypodium unidentified Grains counted for statistics
10 cm 0.0 0.0 0.6 0.6 0.0 0.0 0.3 4.9 306 294
21 cm 0.0 0.4 2.1 0.0 0.0 0.4 1.2 20.7 242 183
35 cm 0.6 0.0 0.0 0.6 0.0 0.0 1.2 18.0 163 152
46 cm 0.0 0.0 2.2 0.0 0.7 0.0 0.0 11.2 122 113
58 cm 0.0 0.0 0.7 0.0 0.0 0.0 0.0 14.5 130 124
69 cm 0.0 0.0 0.0 0.0 0.0 0.0 0.0 16.0 145 137
79 cm 0.0 0.0 3.0 1.5 0.0 0.0 0.0 25.8 111 98
90 cm 1.6 0.0 2/3 0.0 0.0 0.0 1.6 27.1 105 87
104 cm 0.0 0.0 0.0 0.0 1.4 0.0 0.0 225 111 106
120 cm 0.0 0.0 2.5 0.0 0.0 1.9 4.4 16.3 130 112
130 cm 1.1 0.0 1.7 0.0 0.0 0.6 0.0 3357) 142 126
145 cm 0.0 0.0 0.0 0.0 0.0 0.0 0.0 52.4 107 102
155 cm 0.0 0.0 iT 0.0 0.0 0.0 0.0 32:9 140 133
165 cm 0.0 0.0 0.0 0.0 0.0 0.0 0.0 43.5 144 138
188 cm 0.0 0.0 3.4 1.4 0.0 0.0 0.0 48.3 102 89
205 cm 0.0 0.0 Ded: L7/ 0.0 leith 1.1 28.1 130 123
248 cm 0.0 0.0 0.0 25) 0.0 0.0 0.0 19.1 106 100
bonates (regression) at about 2,800 years before pre-
sent (Text-fig. 10). This may be the event that stabi-
lized the coastline in core 59-T1. Regressional man-
grove peats are overlain by marine carbonate mud as
the coast is transgressed. From 98 cm to 75 cm is a
tan, supratidal carbonate mudstone. This represents a
period of rapid sediment reworking following the sea-
level oscillation centered around 2,400 ybp (Gelsan-
liter, 1996). Rhizophora peats overlie the sequence to
the depositional surface and indicate continual trans-
gression. Discriminant analysis of core samples with
the surface pollen data set yields variable results that
attest to both the sensitivities and limitations of this
method. Sample 248 cm is correctly attributed to the
brackish marsh, as the sample is from reworked basal
Holocene quartz sands. Samples 205 cm to 104 cm are
all assigned to mangrove swamp cluster 2, represent-
ing the coastal margin physiographic province (Text-
figs. 7 and 10). Cluster 2 is represented by the widest
variety of taxa in the mangrove swamp clusters and is
characterized by higher percentages of chenopods and
Avicennia pollen. These characteristics reflect the fact
that the coastal margin mangrove swamp has the most
variability in substrate type, hydroperiod, and eleva-
tion. It is therefore reasonable that discriminant anal-
ysis assigned all of these intervals to cluster 2 regard-
less of the fact that the host sediments varied from
Rhizophora peats (both transgressive and regressive)
to marine carbonate muds. All these samples shared
high overall percentages of Rhizophora pollen with
130 BULLETIN 361
complex mixtures of Avicennia, chenopods, Pinus, and
other taxa, yet all were deposited in close proximity
to the coastal margin.
Samples 90 cm and 79 cm were both excluded from
discriminant analysis because of low overall pollen
sums. This is significant in that both samples come
from the tan, supratidal carbonate mudstone that rep-
resents a period of rapid sediment reworking and
build-up of a supratidal coastal barrier (Gelsanliter,
1996). These intervals undoubtedly have low pollen
sums because the pollen signal was inundated by re-
worked marine sediment. The upper 58 cm of the core
sequence are discriminated to cluster 2 (coastal man-
grove swamp, 3 samples), cluster 3 (mangrove
swamps along the Shark River Slough), and cluster 6
(upper headwater marsh). This sequence represents an
overall change upwards to reflect the influence of
proximity to brackish and freshwater taxa. Cluster 3
incorporates higher percentages of triporates, cheno-
pods, and Conocarpus, while cluster 6 is dominated
by chenopods. These discriminant assignments reflect
the present proximity of the depositional site to the
inner bays and its inclusion in the brackish to transi-
tional marsh physiographic province (Text-fig. 7).
While discriminant analysis of core 941 1—7 with the
surface pollen data set initially suggests that this tech-
nique is not sensitive enough to differentiate signifi-
cant sedimentologic variability in a core sequence, it
does yield predictable results. Discriminant analysis
indicates an overall transgression in the core sequence
that is present and correctly positions the contempo-
rary depositional environment. The method, however,
fails to recognize a sea-level oscillation because it as-
signs all samples throughout the interval to the coastal
mangrove swamp. Undoubtedly, the pollen signature
of sediments deposited during this oscillation were, in
fact, representative of the coastal mangrove swamp.
CONCLUSIONS
In South Florida, the boundaries between vegetation
communities are commonly well defined, but bound-
aries between pollen zones are blurred because of mix-
ing of pollen by fluvial, tidal, and atmospheric pro-
cesses. Multivariate statistical techniques provide a
sensitive and objective means of delineating the dis-
tribution of pollen and spores in the study area.
Cluster analysis of pollen and spore types from sur-
ficial deposits in the study area indicates that individ-
ual taxa typically are not environmentally restricted
and, in most environments, there are no single indi-
cator species because of broadly overlapping geo-
graphic ranges. Multivariate analysis of pollen assem-
blages is required to separate surface sample pollen
assemblages into distinct pollen zones. Cluster analysis
of pollen and spores from surface sample sites pro-
duced five clusters, or pollen zones, containing a total
of 10 sub-clusters that help delineate subtle differences
within each zone. The spatial distribution of these pol-
len zones strongly corresponds to the major physio-
graphic provinces of the area. These pollen zones are:
1) mangrove swamp, 2) brackish marsh, 3) freshwater
marsh, 4) headwater marsh, and 5) upland complex.
The correlation between the spatial distribution of
clusters and physiographic provinces in the study area
strongly suggests that the surface pollen cluster group-
ings have environmental meaning and that pollen can
provide valuable information for paleoenvironmental
reconstructions. Discriminant analysis underscores the
distinctiveness of the environmental groupings ob-
tained from cluster analysis. It also supplies the objec-
tive criteria and a methodological protocol for accu-
rately assigning group membership to pollen and spore
assemblages from vertical stratigraphic sequences and
ultimately facilitates paleoecological interpretations. A
visual comparison of unknown samples with the 18
key taxa identified herein indicates that the reduced set
is also useful in making a first approximation in as-
signing an unknown sample to a surface pollen zone.
Results from two cores, 59-T1 and 9411-7, indicate
the methods used in the study can identify the overall
late Holocene transgression of South Florida from pol-
len assemblages in vertical sequences and can differ-
entiate among different mangrove swamp communities
as coastlines are first transgressed and then stabilized.
The method is limited in its ability to differentiate
short-lived changes in coastal configuration (7.e., sea-
level oscillations) as long as the overall environment
remains in close proximity to the coastal zone. De-
tailed sedimentologic work, combined with palynolog-
ical analysis is required, especially where sediment re-
working has occurred.
REFERENCES CITED
Bancroft, G.T., Strong, A.M., Sawicki, R.J., Hoffman, W., and
Jewell, S.D.
1994. Relationship among wading bird foraging patterns, colony
locations, and hydrology of the Everglades. in Everglades:
The Ecosystem and its Restoration. S.M. Davis and J.C.
Ogden, eds., St. Lucie Press, Delray Beach, Florida, pp.
615-659.
Bartow, S.M., Craft, C.B., Richardson, C.J.
1996. Reconstructing historical changes in Everglades plant
community composition using pollen distributions in peat.
POLLEN ZONATION: O’ NEAL ET AL. 131
Journal of Lake and Reservoir Management, vol. 12, pp.
313-322.
Birks, H.J., Webb, B. T., and Berti, A.A.
1975. Numerical analysis of surface pollen samples from central
Canada: a comparison of methods. Review of Paleobot-
any and Palynology, vol. 20, pp. 133-169.
Brewster-Wingard, G.L., Ishman, S.E., Willard, D.A., Edwards,
L.E., and Holmes, C.W.
1997. Preliminary paleontologic report on cores 19A and 19B,
from Russell Bank, Everglades National Park, Florida
Bay. U.S. Geological Survey Open-file Report 97—460,
29 pp.
Cohen, A.D.
1968. The petrology of some peats of southern Florida (with
special reference to the origin of coal). Unpublished Ph.D.
thesis, Pennsylvania State University, 352 pp.
Craighead, F.C., and Gilbert, V.C.
1962. The effects of Hurricane Donna on the vegetation of
southern Florida. Quarterly Journal of the Florida Acad-
emy of Science, vol. 25, pp. 1-28.
Davis, J.H., Jr.
1943. The natural features of southern Florida, especially the
vegetation and the Everglades. Bulletin 25, Florida Geo-
logical Survey, Tallahassee, Florida. 311 pp.
Davis, J.C.
1986. Statistical and data analysis in geology. 2 edition. John
Wiley & Sons, New York, 646 pp.
Dean, W.E., Jr.
1974. Determination of carbonate and organic matter in calcar-
eous sediments and sedimentary rocks by loss on ignition:
comparison with other methods. Journal of Sedimentary
Petrology, vol. 44, p. 242-248.
Gelsanliter, S.G.
1996. Holocene stratigraphy of the Chatam River Region,
Southwestern Florida; With a reevaluation of the Late Ho-
locene sea-level curve. Unpublished M.S. thesis, The Uni-
versity of Miami, Miami, Florida. 182 pp.
Gunderson, L.H.
1994. Vegetation in the Everglades: Determinants of community
composition. in Everglades: The Ecosystem and its Res-
toration. S.M. Davis and J.C. Ogden, eds., St. Lucie Press,
Delray Beach, Florida. pp. 323-340.
Gunderson, L.H., and Snyder, J.R.
1992. Fire patterns in the southern Everglades. in Everglades:
The Ecosystem and its Restoration. S.M. Davis and J.C.
Ogden, eds., St. Lucie Press, Delray Beach, Florida, pp.
291-306.
Habib, D., Eshet, Y., and Van Pelt, R.
1994. Palynology of sedimentary cycles. in Sedimentation of
Organic Particles. A. Traverse, ed., Cambridge University
Press, pp. 311-335.
Hoffmeister, J.E., Stockman, K.W., and Multer, H.G.
1967. Miami Limestone of Florida and its recent Bahamian
counterpart. American Association of Petroleum Geolo-
gists Bulletin, vol. 78, pp. 175-190.
Kachigan, S.K.
1991. Multivariate statistical analysis: A conceptual introduc-
tion. Radius Press, New York, 303 pp.
Kuehn, D.W.
1980. Offshore transgressive peat deposits of southwest Florida:
Evidence for a late Holocene rise of sea level. Unpub-
lished MLS. thesis, Pennsylvania State University, 104 pp.
Loveless, C.M.
1959. A study of the vegetation of the Florida Everglades. Ecol-
ogy, vol. 40, pp. 1-9.
Mclvor, C.C., Ley, J.A., and Bjork, R.D.
1994. Changes in freshwater inflow from the Everglades to Flor-
ida Bay including effects on biota and biotic processes:
A review. in Everglades: The Ecosystem and its Resto-
ration. S.M. Davis and J.C. Ogden, eds., St. Lucie Press,
Delray Beach, Florida, pp. 117—148.
Meeder, J.F.
1987. The paleoecology, petrology, and depositional model of
the Pliocene Tamiami Formation, southwest Florida (with
special reference to coral and reef development). Unpub-
lished Ph.D. dissertation, University of Miami, 249 pp.
Meeder, J.F., Ross, R.S, Telesnicki, G., Ruiz, P.L., and Sah, J.P.
1996. Vegetation analysis in the C-111/Taylor Slough Basin. Fi-
nal Report Contract C-4244, Southeast Environmental Re-
search Program, Florida International University, Miami,
Florida, 90 pp.
Moore, P.D., Webb, J.A., and Collinson, M.E.
1991. Pollen Analysis. Blackwell Scientific Publishing, Oxford,
216 pp.
Olmstead, I.C., and Loope, L.L.
1984. Plant communities of Everglades National Park. in En-
vironments of South Florida: Present and Past II. PJ.
Gleason, ed., Miami Geological Society, Coral Gables,
Florida, pp. 167—184.
Parker, G.G., and Cooke, C.W.
1944. Late Cenozoic geology of southern Florida, with a dis-
cussion of ground water. Florida Geological Survey Bul-
letin 27, 119 pp.
Parker, G.G., Hoy, N.D., and Schroeder, M.C.
1955. Geology of water resources of southeastern Florida with
special reference to the geology and groundwater of the
Miami area. U.S. Geological Survey Water Resources Pa-
per 1255.
Puri, H.S. and Vernon, R.O.
1959. Summary of the geology of Florida and a guidebook to
classic exposures. Florida Geologic Survey, Special Pub-
lication 5, Tallahassee, Florida, 255 pp.
Rich, F.J., Kuehn, D., and Davies, T.D.
1982. The paleoecological significance of Ovoidites. Palynolo-
gy, vol. 6, pp. 19-28.
Riegel, W.L.
1965. Palynology of Environments of Peat Formation in South-
western Florida. Unpublished Ph.D. thesis, Pennsylvania
State University, 189 pp.
Smith, T.J., III, Hudson, J.H., Robblee, M.B., Powell, G.V.N.,
and Isdale, P.J.
1989. Freshwater flow from the Everglades to Florida Bay: A
historical reconstruction based on fluorescent banding in
the coral Solenastrea bournoni. Bulletin Marine Science,
vol. 44, pp. 274-282.
Spackman, W., Dolsen, C.P., and Riegel, W.
1966. Phytogenic organic sediments and sedimentary environ-
ments in the Everglades—Mangrove Complex, Part I: Ey-
idence of a transgressing sea and its effects on environ-
ments of the Shark River area of southwestern Florida.
Palaeontographica, vol. 117, pp. 135-152.
Stockmarr, J.
1971. Tablets with spores used in absolute pollen analysis. Pol-
len et Spores, vol. 13, 614-621.
Wanless, H.R., Tedesco, L.P., Rossinsky, V., and Dravis, J.J.
1989. Carbonate Environments and Sequences of Caicos Plat-
132 BULLETIN 361
form with an Introductory Evaluation of South Florida. in
28" International Geologic Congress Field Trip Guide-
book T1374, American Geophysical Union, 75 pp.
Wanless, H.R., Tedesco, L.P., Risi, J.A., Bischof, B.G., and Gel-
sanliter, S.
1995. The role of storm processes on the growth and evolution
of coastal and shallow marine sedimentary environments
in South Florida. 1*‘ SEPM Congress on Sedimentary Ge-
ology, St. Petersburg, Florida, 137 pp.
Ward, J.H.
1963. Hierarchical grouping to optimize an objective function.
Journal of the American Statistical Association, vol. 58,
pas
Willard, D.A. and Holmes, C.W.
1997. Pollen and geochronological data from South Florida:
Taylor Creek Site 2. U.S. Geological Survey Open-file
Report 97-35, 14 pp.
Winkler, M.J., Sanford, P.R., and Kaplan, S.W.
1998. Holocene hydrologic and vegetation history of the south-
ern Everglades. Geological Society of America, Abstracts
with Programs, vol. 30, p. A309.
Wood, G.D., Gabriel, A.M., and Lawson, J.C.
1996. Palynological techniques—processing and microscopy, in
J. Jansonius, and D.C. McGregor, eds., Palynology: Prin-
ciples and Applications. American Association of Strati-
graphic Palynologists. Salt Lake City, Utah, pp. 29-50.
CHAPTER 7
A CENTURY OF ENVIRONMENTAL VARIABILITY IN OYSTER BAY USING OSTRACODE
ECOLOGICAL AND ISOTOPIC DATA AS PALEOENVIRONMENTAL TOOLS
Cartos A. ALVAREZ ZARIKIAN', PETER K. SWART!, TERRI Hoop!, Pat L. BLACKWELDER!2,
TERRY A. NELSEN*, AND CHARLES FEATHERSTONE:
'Marine Geology and Geophysics, University of Miami-RSMAS, 4600 Rickenbacker Causeway,
Virginia Key, FL 33149
°Nova Southeastern University, Oceanographic Center, 8000 N. Ocean Dr., Dania, FL 33004
SNOAA-AOML-OCD, 4301 Rickenbacker Causeway, Virginia Key, FL 33149
ABSTRACT
Stable isotopic analysis (6'SO and 6'C) and characterization of the ostracode community structure were carried out from a
high-resolution sediment-core recovered from Oyster Bay in the west of the Everglades National Park. Because of its location,
between the Shark River Slough (SRS) and the Gulf of Mexico, the Oyster Bay core locality experiences extreme salinity
fluctuations due to the interaction of freshwater run-off, precipitation, and marine water inputs. Ostracode population dynamics
and isotopic variability over the 20" century are linked to natural and anthropogenic forces that affect the South Florida coastal
ecosystem on interannual to decadal time scales. Three ostracode assemblages can be recognized within the 100-year sediment-
core record: the first extending from the turn of the century to about 1950; the second, from the early 1950s to the late 1970s;
and the third to core recovery in 1995.
An abrupt decrease in ostracode abundance, species diversity, and shifts in species dominance occur in the mid-1980s and reflect
episodes of environmental stress. Markedly enriched 6!SO values from the ostracode Peratocytheridea setipunctata and the benthic
foraminifer Ammonia parkinsoniana typica at this time are concurrent with a major regional drought in South Florida, as well as
with documented algal blooms and major die-off of sea grasses in Florida Bay. In addition, the timing of these events is contem-
poraneous to the onset of the South Florida Water Management District (SFWMD) “Rainfall Plan” and the closing of the Buttonwood
canal. Higher ostracode abundance and species richness occurs between the late-1950s and late-1970s. Stable isotopic data and
ostracode assemblage characteristics suggest a period of relative environmental stability and possibly improved water circulation in
Whitewater Bay and Oyster Bay. Fluctuations in community structure during this time are more systematic and appear to be
temporally correlated to rainfall variability patterns. Water management policies at this time are also discernable from the microfaunal
and isotopic record, particularly the Congressionally mandated Monthly Minimum Allocation Plan of water supply to SRS. Before
1950, hurricane events and their effects are the major cause for immediate modifications within the ostracode community, though
our data show that ostracode populations are capable of rapid recovery. Over the complete record of the last century, the effects of
water management practices can be assessed from information embedded in the ostracode record. Nevertheless, the effects of natural
climatic variability in Oyster Bay appear to outweigh the impact of anthropogenic forces.
INTRODUCTION
Over the past decade, scientists have used proxy re-
cords of environmental conditions in Florida Bay and
its environs to examine the paleoecological history of
this region (Brewster- Wingard et al., 2000; Halley and
Roulier, 1999; Swart et al., 1996, 1999; Nelsen et al.,
in press). Benthic microfaunal communities preserved
in undisturbed sediment records have shown great
promise for the reconstruction of paleoenvironmental
fluctuations (Alvarez Zarikian et al., 1999a, 1999b;
Brewster-Wingard and Ishman, 1999; Cronin ef al.,
2000; Dwyer and Cronin, 2000; Hood et al., 1998:
Ishman, 2000). Ostracodes have proven particularly
valuable for such reconstructions because they are
abundant in the marine and brackish water environ-
ments of South Florida (Puri and Hulings, 1957; Puri,
1960; Benda and Puri, 1962; Keyser, 1975, 1977). In
addition, they have specific ecological requirements
and well defined salinity tolerances (Neal, 1988).
Moreover, ostracodes rapidly secrete a calcitic bivalve
shell (Turpen and Angell, 1971), which with genetic
influences, records the isotopic composition of the
aquatic environment, thus providing a geochemical
snapshot of the chemical conditions of the water in
which they lived.
The lower Everglades National Park is comprised of
a north-south ecosystem succession ranging from large
flat wetlands in the north to shallow bays, lakes and
channels in the south, which are mostly filled with
mangrove swamp deposits and bordered by mangrove
forests. The environmental transition into the coastal
zone is gradual and it extends into the shallow mud-
banks and semi-restricted basins of Florida Bay. To the
west and north of Cape Sable, the 1—2 meter-deep
Whitewater Bay extends to the northwest through the
mangrove forest and Oyster Bay until opening up to
134 BULLETIN 361
# Oyster Bay core
;Whitewater Bay
: d Sees
Florida Bay* *~_
v 2 oF 5
Text-figure 1—Location Map of Oyster and Florida Bays. Sediment core location in Oyster Bay represented by black cylinder.
the Gulf of Mexico. Direct freshwater runoff from the
Everglades into Florida Bay and adjacent bays is pri-
marily through the natural drainage basins of Shark
River and Taylor Sloughs, and artificially from the C-
111 Canal. To the west, the park is heavily influenced
by the marine waters of the Gulf. Freshwater input
from canals and watersheds that flow into Florida Bay
is strongly influenced by rainfall and freshwater man-
agement policies of the South Florida Water Manage-
ment District (Nelsen ef al., in press; Rudnick er al.,
1999).
Numerous hydrological modifications have impact-
ed the Everglades natural ecosystem during the 20th
century. Among these are the fragmentation of the
wetlands due to construction of a massive network of
canals and levees for water supply and flood control,
and the loss of extensive wetland area to agriculture
and residential development (Davis and Ogden, 1994).
Prior to human impact, natural forces and seasonal cli-
matic variability were the main factors controlling hy-
drological conditions in the area. Increased salinity and
nutrient concentration in Florida Bay (Mclvor et al.,
1994), as well as algal blooms and major seagrass die-
offs (Fourqurean and Robblee, 1999; Robblee er al.,
1991), have raised growing concerns as to whether the
origin of these changes are anthropogenic or are part
of a natural cycle of climatic variability.
Micropaleontological and stable isotopic data ob-
tained from a high deposition rate sediment-core col-
lected in Oyster Bay are presented here. These data
have allowed interpretation of interannual to decadal
changes in hydrological conditions from this area
within the Everglades National Park. Because of its
location, adjacent to Shark River and the Gulf of Mex-
ico, the Oyster Bay core locality experiences extreme
salinity fluctuations (4 to >40 parts of salinity, Boyer
et al., 1997) due to the interaction of freshwater run-
off, precipitation and marine water inputs. Because of
the wide range of environmental conditions, this site
is ideally located to examine the response and sensi-
tivity of ostracode populations and their isotopic com-
position to environmental change over the last century.
CORE SELECTION AND GEOCHRONOLOGY
An undisturbed sediment record is a pre-requisite
for the meaningful interpretation and reconstruction of
paleoenvironmental changes. The Oyster Bay core
(Text-fig. 1) used in this study is one of many collected
in Florida and adjacent bays during the last four years.
It was recovered from seagrass-free soft bottom sedi-
ments in shallow water (1 m). This high sedimentation
rate core has been the focus of an interdisciplinary
study to reconstruct Oyster Bay’s paleoenvironmental
record. Analyses that have been completed include mi-
crofaunal and pollen assemblage, stable isotope, or-
ganic and inorganic carbon, mineralogy and heavy
metals (Nelsen ef al., in press).
The location of the sediment core was carefully tar-
OSTRACODE PALEOENVIRONMENTAL RECONSTRUCTION: ALVAREZ ZARIKIAN ET AL. 135
geted from review of sequential aerial photographs
(1927—present) and published data on the dynamics of
banks and deltas in Florida Bay (Wanless and Tagett,
1989). Aerial photographs helped identify sites that
had been free of macro vegetation since the time of
the earliest aerial photograph. Subsequent reconnais-
sance coring sought to verify that the site was an area
of active sediment accumulation and that it had well-
preserved stratification (minimal bioturbation activity)
(Nelsen et al., in press).
The |.5-m long Oyster Bay core was examined for
stratification/disruption utilizing X-radiography to as-
sure that sediment geochronology would yield inter-
pretable sequences. Stratification in the sediment core
was clear, with laminations consisting of alternations
of mangrove organics, carbonate silt, and fine quartz.
The core was age-dated using the *!°Pb-dating tech-
nique (Nelsen ef al., in press), described, and photo-
graphed for evaluation of sediment type and structures.
The sediment accumulation rates derived from the pro-
file of excess 7!°Pb were approximately 1.1 cm/yr (0.6
g/cm?/yr). The geochronology for the Oyster Bay core
was extended back to the turn of the century by ex-
trapolating the activity of excess *!°Pb downcore. In
addition to providing a time-line and accumulation rate
information, fine-structural features were observed in
the excess 7!°Pb profile including sediment erosion/de-
position bands that are tied by year to hurricane events.
These discontinuities appear as lower levels of excess
*!°Pb than would be expected from the overall trend.
These lower levels are most likely a result of the com-
bined effects of losing sediment from the core site dur-
ing a storm and input of sediments with lower levels
of excess 7!°Pb, i.e., older sediments or organic-rich
debris, from upland or other sources (Nelsen ef al., in
press). In the Oyster Bay core site, sudden shifts in
the *!°Pb activity correspond (within +3 years) to the
years of 1960, 1948, and 1935 (Text-fig. 2) and are
likely the result of sediment movement due to Hurri-
cane Donna, the hurricane of 1948, and the Great La-
bor Day Hurricane in 1935, respectively.
Given the established age model, the Oyster Bay
core was subsampled at l-cm intervals for geochemi-
cal (organic carbon, heavy metals, stable isotopes), mi-
cropaleontological (ostracodes and benthic foramini-
fers) and palynological characterization. These data
were complemented by an analysis of historical rain-
fall, gauged freshwater flow, and limited near shore
salinity data (see Nelsen ef al., in press).
It is important to note that geochronology-based ages
have associated error estimates that increase downcore.
As a result, 7!°Pb-derived core ages in the 1940s and
1920s, for example, have a range of +3 and +4 years
at the core location in Oyster Bay. For more recent
Excess 210Pb activity (Ln(dpm/g))
-1 0 1 2
1 o oO —
Hurricane Andrew
11
721\ 5
a skis
SI
5 4
= 414 © Hurricane
€ 4 Donna
E51
rs (e) Hurricane
— of 1948
1
Ee
o
1}
n
un Hurricane of 1935
81
91
101
Text-figure 2—Excess *!°Pb profile used to establish the geochro-
nology for the Oyster Bay core (from Nelsen er al., in press).
years, between 1980 and 1995, sediment geochronology
varies by only +1 year (Nelsen ef al., in press).
ACKNOWLEDGMENTS
The authors would like to thank Amel Saied and
Vivian Gonzalez, from the University of Miami-
RSMAS Stable Isotope Laboratory, for running the
samples for isotopic analysis. Dr. Harold R. Wanless,
chairman of the University of Miami’s Dept. of Ge-
ology, for fieldwork logistics and expertise about sed-
imentological processes in Florida Bay. Dr. John Tre-
fry and W-J. Kang, from the Florida Institute of Tech-
nology’s Div. of Marine and Environmental Systems,
for performing the core’s geochronology. Water qual-
ity data was provided by Drs. Ron Jones and Joe Boy-
er of the Southeast Environmental Research Center at
Florida International University with support by the
South Florida Water Management District (SFWMD/
SERC #C-10244) through Everglades National Park
(NPS/SERC Cooperative Agreement #5280-2-9017).
The National Climate Data Center for the rainfall data
and the National Oceanic and Atmospheric Adminis-
tration for their financial assistance. In addition, we
136 BULLETIN 361
would like to thank Dr. Larry Peterson for reviewing
the manuscript and for his helpful suggestions.
METHODS
MICROFAUNAL ANALYSIS
Core sediment intervals were sub-sampled, wet-
sieved using a 63 fm mesh, and the coarse fraction
retained for microfaunal characterization. High organic
content in the Oyster Bay core required the use of
ashing and 30% hydrogen peroxide treatments. Sub-
sample weights were used to determine total popula-
tion abundances expressed as tests or valves per gram
dry sediment weight. A minimum of three hundred
specimens of both ostracodes and foraminifers were
counted for each sample unless limited by sample size.
Temporal fluctuations in microfaunal community
structure were evaluated using relative and absolute
abundances, and the Shannon-Weiner diversity index
(SWDI). The SWDI is defined as: H = — XN, p(i)
In p(i), where H is the measure of diversity, N is the
number of species counted and p(i) is the proportion
of the total number of individuals which belong to the
i species (MacArthur, 1983). This index incorporates
a measure of species evenness as well as number
(communities with many species with equal-size pop-
ulations have the highest index) (Patrick, 1983), mak-
ing it sensitive to the presence of rare species.
STABLE ISOTOPE ANALYSIS
Stable isotopic analyses (6!°O and 6°C) were car-
ried out on valves of the ostracode Peratocytheridea
setipunctata. This species was chosen for isotopic
analysis due to its ubiquitous occurrence throughout
the Oyster Bay core as well as its high abundance in
surface sediment samples collected from Oyster and
Whitewater bays and Florida Bay. Well-preserved
adult valves of P. setipunctata were isolated from the
>63 wm sediment fraction prior to the ashing/H,O,
treatments. Specimens were cleaned with a small
paintbrush in a distilled water bath and then in an eth-
anol solution to remove adhering debris. Between 6 to
8 valves of P. setipunctata were then immersed in
methanol-filled “‘copper boats” and crushed to a pow-
der. The carbonate samples were processed by an au-
tomated device attached to a Finnigan-MAT 251 mass
spectrometer. The external precision (0.02%o for 6'°C
and 0.08%c for 6'8O) was calculated from replicate
analyses of the RSMAS Stable Isotope Laboratory cal-
cite standard. All isotopic data are reported in per mil
(%o) units relative to the Pee Dee Belemnite (PDB)
standard and are corrected for the usual isobaric inter-
ferences. The stable isotopic composition of the sedi-
ment’s organic C fraction was analyzed as well (Nel-
sen et al., in press).
RAINFALL AND FRESH WATER FLOW DATA
Continuous monthly freshwater flow data for Shark
River Slough extending from 1939 to the present were
obtained from the United States Geological Survey
(http://fl.water.usgs.gov/). Rainfall data were obtained
from NOAA's National Climate Data Center. Contin-
uous monthly data from the Everglades, Flamingo,
Homestead and Tavernier stations were obtained, ex-
tending from present back to 1926, 1951, 1911, and
1936, respectively (Nelsen ef a/., in press).
RESULTS
The three major sediment components in the Oyster
Bay core are carbonates, alumino-silicates and organic
matter. The average CaCO, content is 75 + 5% and is
comprised mostly of benthic foraminifers, ostracodes,
small mollusks and calcareous fragments. The organic
C content averages 7.1 + 1.3%, while the alumino-
silicate contribution is minor, averaging only 1.6 +
0.3% (Nelsen et al., in press).
OSTRACODE ASSEMBLAGE
The Oyster Bay core contained 43 ostracode spe-
cies. The number of species in any one sample ranged
from 2 to 27, with an average of 11 species per 1-cm
sample. Ostracode diversity, total abundance and spe-
cies evenness show large temporal fluctuations (Text-
fig. 3). Significant variability in species dominance and
temporal loss of numerous species towards the present
time also occurred. In addition, microfaunal preser-
vation is lower towards the top of the core.
The Shannon-Weiner diversity index (SWDI), used
here to expressed ostracode diversity, is sensitive to
the presence and absence of minor species, such that
larger positive values reflect greater diversity while
lower numbers indicate a community dominated by
fewer species. The SWDI ranged from 0.22 at a depth
of 18.5 cm (~1977) to 2.51 at a depth of 59.5 cm
(~1941). Ostracode abundance in the sediment core
ranged from 0 to 397 valves per gram. Highest ostra-
code abundance and diversity occurred in sediments
deposited in the mid-1970s, and in the late-1960s and
1950s. In contrast, the bottom half of the core, record-
ing deposition between the turn of the century and the
late- 1940s, is characterized by a fairly diverse and sta-
ble community and a relatively low ostracode abun-
dance (Text-fig. 3). The lowest abundances were found
in sample intervals at 90, 74 and 57 centimeters. These
samples contained less than 10 ostracode valves each,
mainly of Perissocytheridea brachyforma, and repre-
sent the years of 1915 + 4, 1929 + 4 and 1941 + 3,
according to the extended 7!°Pb-derived geochronolo-
gy. The timing of two of these events temporally cor-
relates with hurricanes that occurred over areas of
OSTRACODE PALEOENVIRONMENTAL RECONSTRUCTION: ALVAREZ ZARIKIAN ET AL.
Phase III Phase II
Phase |
< r¢ 4
0.5 (1995) 7
40.5 (1985)
20.5 (1975)
30.5 (1966)
40.5 (1958)
50.5 (1948)
Wag Ostracodes per gram
—@— Ostracode diversity (SWDI)
300
c. Ye
: , © ee
: , e | 200
100
lh IB wt,
Se
+
a
=
in
S
rr)
80.5 (1923)
90.5 (1915)
100.5 (1906)
Pb Geochronology
Text-figure 3.—Ostracode diversity (Shannon-Weiner diversity index; left axis) and abundance (valves/g: right axis) in the Oyster Bay core.
Geochronology based on 7!°Pb and '*’Cs dating techniques (cm (year)).
—@— Mailzella floridana
—a&— Peratocythendea setipunctata
50 4
25
*
—.
40.5 vos
0.5 (1995) +
40.5 (1985)
20.5 (1975)
30.5 (1966)
50.5 (1948)
60.5 (1940)
70.5 (1932) 8
80.5 (1923)
90.5 (1915)
100.5 (1906)
—m- P. brachyforma
Text-figure 4.—Relative abundances of: a) Malzella floridana; b)
Peratocytheridea setipunctata, and c) Perissocytheridea brachyfor-
ma in the Oyster Bay core. X-axis: *!°Pb geochronology (cm (year)).
Florida Bay: the Hurricane of 1929, and the Great La-
bor Day Hurricane of 1935.
From 1956 + 2 to 1977 + 1, the ostracode assem-
blage at this location exhibited high abundance and
species diversity indicating a stable and robust ostra-
code community. Population evenness was relatively
high and constant at this time. Following this long,
stable interval, in about 1978 + 1, a dramatic decline
in assemblage diversity occurred (Text-fig. 3). The
SWDI was reduced by half (from ~2 to ~1) and os-
tracode abundance dramatically declined from more
than 300 ostracodes per gram to less than 50. This
marked the second long-term decline in ostracode
community structure during the 20th century. The
number of species dropped from an average of 18 spe-
cies per sample to less than five and the ostracode
assemblage became dominated by Peratocytheridea
setipunctata (Text-fig. 4b). However, despite its high
relative abundance, P. setipunctata displays low ab-
solute abundance which suggests survivor type domi-
nance rather than opportunistic behavior. SWDI and
evenness increase slightly at the top of the core, al-
though, ostracode abundance remains low.
The most common ostracode species in the core
were Peratocytheridea setipunctata, Malzella flori-
dana, Neocaudites triplistriatus and Perissocytheridea
brachyforma with maximum absolute abundances of
113, 80, 71 and 22 valves per gram, respectively. Ac-
tinocythereis subquadrata, Loxoconcha matagordensis
and Xestoleberis mixohalina were also common
throughout the core, but their relative abundances nev-
er exceed 15% of the total assemblage at any particular
138 BULLETIN 361
x Neocaudites tnplistratus
18 @ Actinocythereis subquadrata
0.5 (1995)
10.5 (1985)
20.5 (1975)
30.5 (1966)
40.5 (1958)
50.5 (1948)
60.5 (1940)
70.5 (1932)
80.5 (1923)
90.5 (1915)
o
3
o)
o
r=)
5S
@ Loxoconcha matagordensis
Text-figure 5—Relative abundances of a) Neocaudites triplistria-
tus, b) Actinocythereis subquadrata; and c) epiphytic ostracode spe-
cies Loxoconcha matagordensis in the Oyster Bay core. X-axis:
!0Pb geochronology (cm (year))
sediment interval. Inter-specific dominance among P.
setipunctata, M. floridana, N. triplistriatus and P. bra-
chyforma has changed through time (Text-fig. 4a,b,c;
and Text-fig. 5a). The ostracode Peratocytheridea se-
tipunctata showed a marked increase in relative abun-
dance in about the late 1970s and remained the dom-
inant species (>50%) well into the mid-1990s. In con-
trast, the relative abundance of Malzella floridana re-
mains rather constant throughout the core, exhibiting
a peak in abundance in the mid-1940s, and a decline
during the 1980s. Although P. setipunctata and M.
floridana are the two most common species in the
Oyster Bay core, their relative abundance appears to
be inversely correlated (r: —0.72).
In the other hand, the mesohaline species Perisso-
cytheridea brachyforma is found sporadically in the
lower portion of the Oyster Bay core in abundances
of up to 50%. This corresponds to the time before
1950. Since then, abundance of P. brachyforma re-
—@— Oligo-to-mesohaline ostracodes
aioe
‘ rr nee aoe
*!°b Geochronology
10.5 (1985)
20.5 (1975)
30.5 (1966)
40.5 (1958)
50.5 (1948)
60.5 (1940)
70.5 (1932)
80.5 (1923)
2
x
D
2)
wo
(=)
90.5 (1915)
100.5 (1906)
—#— Bairdia sp. —g-— Cytherella sp.
Text-figure 6—Relative abundance of: a) Grouped oligohaline
and B-mesohaline species: Candona sp., Cypridopsis okeechobei,
Cytheridella illosvayi, Darwinula stephensoni, Limnocythere flori-
densis and Phisocypria sp. A; and b) Euhaline species: Bairdia sp.
and Cytherella sp. in the Oyster Bay core. X-axis: ?!°Pb geochro-
nology (cm (year))
mained generally low until the early 1990s when its
relative abundance increased again reaching 25% of
the total ostracod assemblage. On the contrary, Neo-
caudites triplistriatus exhibited its greatest abundance
just before P. brachyforma in the early 1910s (>40%),
and at about 1960 (~50%). Although present to some
extent throughout the core, N. triplistriatus has been
essentially absent from the Oyster Bay core since early
in the 1980s (Text-fig. 5a).
A group of oligo-mesohaline species, such as, Can-
dona sp, Cypridopsis ockeechobei, Cytheridella illos-
vayl, Darwinula stephensoni, Dolerocypria fastigata,
Limnocythere floridensis and Physocypria sp. occurred
at sporadic intervals in the sediment core correspond-
ing to the mid-1910s, early 1930s, early 1960s, early
1980s, and in the early-1990s (Text-fig. 6a). Cypri-
dopsis ockeechobei, Cytheridella illosvayi and Dar-
winula stephensoni are the most common species of
this group, reaching in some intervals over 7% of the
total ostracode assemblage.
In contrast, euhaline species such as Bairdia sp. and
Cytherella sp. are only present in the Oyster Bay core
at lower percentages (<4%). Although less significant
in number, their distribution is clearly limited to the
period of highest diversity and population abundance
that occurred between the early 1950s and late 1970s
OSTRACODE PALEOENVIRONMENTAL RECONSTRUCTION: ALVAREZ ZARIKIAN ET AL. 139
4 9 7 5
Ol(1995) cia > 7
9.5 (1986) 4 %e
19.5 (1976) 4 4° NY 2?
28.5(1968)} © e
eee A
48.5(1950)} 4% ,
58.5 (1942) 4
68.5 (1934) °®
78.5 (1925 ¢
(1925)7 6% 4
88.5 (1916) 7 @ e
98.5 (1908) %
Bjejoundijas eapuayjAIO}e18q JO D.,2 }
ejejOUuNdHas eapLEyjAIOJe/Ad JOO, 2 @
Ado[OUoIYyIOAH qdO [TZ
wo
7 5 3 4
Text-figure 7.—Staple isotopic composition (8'°O and 6'°C) of the
ostracode species Peratocytheridea setipunctata in the Oyster Bay
core.
(Text-fig. 6b). A gap in the occurrence of euhaline spe-
cies 1s visible around 1960.
The 6!°O isotopic composition of the ostracode Per-
atocytheridea setipunctata ranged from —2.80 to
+2.37%c. Strong negative excursions are found prior
to 1925 + 4, while the most positive values occurred
during the 1980s (Text-fig. 7). Between the early
1950s and 1970s, the 'SO values do not show large
deviations. During this time, values oscillate between
—1 and +1%c on time scales of | to 3 years.
The 5'°C of Peratocytheridea setipunctata at Oyster
Bay exhibits very negative values, ranging from —7.59
to —10.72. Relatively constant and lighter values (ap-
proximately —10%c) occur between early 1900s and
late 1950s. Isotopic trends during this time are gradual,
trending to more positive or negative values in roughly
decadal timescales. In contrast, abrupt positive shifts
in the 5%C record take place at about 1970 + 1| and
1983 + 1. The 56°C of P. setipunctata at these times
is enriched by more than 2%c. An increase in the mag-
nitude of fluctuations of the 6'°C is evident upcore.
DISCUSSION
MICROFAUNAL RESPONSE
Large decreases in ostracode abundance and diversity
and shifts in species dominance that occurred in the Oys-
ter Bay core represent episodes of environmental stress
due to natural and/or anthropogenic forces. Distinctive
characteristics of the ostracode assemblages from the
core locality were used to divide the 100-year ostracode
record into three zones. The first zone extends from the
turn of the century to about 1950; the second, from the
early 1950s to the late 1970s; and the third to the time
of core recovery in 1995. Before 1950, the significant
declines in ostracode community structure that occurred
in 1915 + 4, 1929 + 4, 1941 + 3 and 1948 + 2 are
temporally coincident with severe tropical storms or hur-
ricanes: the tropical storm of 1916, the hurricane of
1929, the Great Labor Day hurricane in 1935, and the
hurricane of 1948. Hurricanes are capable of causing im-
mediate or long-term environmental changes in coastal
areas due to sediment erosion and deposition, and storm
induced bi-directional sediment transport (Tabb and
Jones, 1962; Smith eft al., 1994; Wanless er al., 1994).
Decline in the ostracode populations and the input of
reworked freshwater species as seen in the Oyster Bay
core are evidence of hurricane disturbance. However, ac-
cording to our data, reasonably rapid population recov-
eries from these events indicate a relatively resilient com-
munity. The first zone culminated with a decade-long
diversity low, and an ostracode assemblage dominated
initially by Malzella floridana and later by Peratocy-
theridea setipunctata. Rainfall data from the Homestead
station (Nelsen ef al., in press) showed lower than av-
erage rainfall during the early 1940s, with a later increase
in rainfall from the mid to late 1940s (Text-fig. 8). In
contrast, freshwater flow data for Shark River Slough
during that time shows a diminished flow during most
of the 1940s (Nelsen ef al, in press). The Homestead
rainfall data are used as a comparison in this case be-
cause the record is longer (back to 1911) than that of
the Flamingo meteorological station, whose data only
extend from 1951 to the present.
The second zone, between the late 1950s and late
1970s, exhibited the richest ostracode fauna (average:
>200 valves/g) and the highest diversity (SWDI >2).
Community fluctuations during this interval are more
gradual and can be temporally correlated to SFWMD
policies. The second zone generally represents a
“bloom” period interrupted only by a brief hiatus in
140
250
200 -
rainfall (cm/yr)
100 -
50
1997
1992
1987
1982
1977
1972
1967
1962
1957
BULLETIN 361
1952
1947
1942
1937
1932
1927
1922
1917
1912
== Hmstd (trend) -e- Homestead —«— Flamingo
Text-figure 8.—Rainfall record during the 20th century from the Flamingo and Homestead meteorological stations (NOAA's National Climate
Data Center). Gray area: trend analysis for Homestead rainfall station (right axis).
the ostracode assemblage continuity about 1960, co-
incident with Hurricane Donna. During this period ma-
jor construction also occurred (1960—63) with vast ar-
eas of the Everglades north of Tamiami Trail delimited
with new levees and canals. Water flow into the Shark
River Slough (SRS) at this time, began to be controlled
by a series of gate spillways in order to create the
current system of Water Conservation areas (Light and
Dinnen, 1994). In addition, higher concentrations of
organic carbon in the sediment intervals of this period,
and at about 1935 (Nelsen er al., in press), support the
hypothesis of hurricane effects. Large amounts of or-
ganics derived from sediment transport and subsequent
settling of suspended material occurred at this time.
Oxidation of high concentrations of organic material
at the sediment/water interface may have reduced ox-
gen levels, temporarily inhibiting occupation by most
of the benthic microfaunal assemblage.
During the late 1950s through late 1970s, small num-
bers of the euhaline ostracodes Bairdia sp. and Cyther-
ella sp. also appeared (up to 4%, see Text-fig. 6b). Typ-
ical marine species were not present in relative abun-
dances greater than 0.3% anywhere in the core until
this time. Two reasons potentially explain the appear-
ance of euhaline taxa at the core locality. First, hurri-
cane-related onshore sediment transport due to the Hur-
ricane of 1948 and Hurricane Donna (1960). Second,
less than average rainfall in the area during the 1950s
may have favored marine water input from the Gulf of
Mexico, allowing shelf taxa to adapt to the otherwise
brackish waters of Oyster and Whitewater bays. Be-
tween the 1960s and early 80s, the SFWMD imple-
mented a series of short-term water management plans
(1963-70), and more importantly the congressionally
mandated Monthly Minimum Allocation Plan of water
supply to SRS (Light and Dineen, 1994). The latter
practice resulted in essentially, no correlation between
rainfall and flow (r = —0.06) (Nelsen ef al., in press).
The time interval encompassed in zone 2 was also tem-
porally coincident with the opening and closing of the
Buttonwood Canal (late-1950s and late-1970s). The
Buttonwood Canal may have led to improved water ex-
change between Florida Bay, Whitewater Bay and the
Gulf of Mexico by means of tidal water movements,
ultimately improving water quality in Whitewater Bay.
The second ostracode zone ended circa 1979, when an
abrupt decline in population diversity and abundance
occurred, from which the ostracode assemblage had not
yet fully recovered as of 1995.
The third zone extends from about 1979 until 1995
when the sediment core was taken. It is highlighted by
a large decline in ostracode diversity and abundance
at about 1980 and the survivor-mode dominance ex-
hibited by Peratocytheridea setipunctata, one of the
two most common species in the core, known to be
tolerant to extreme salinity fluctuations (<5 to 42)
(Keyser, 1977; pers. comm., 1999). The early 1980s
marked the end of the SFWMD’s Monthly Minimum
Water Delivery Plan, which was replaced by a regu-
lated flow plan believed to be more natural, known as
the Rainfall Plan. In addition, a great part of the nat-
ural flow of freshwater running through Shark River
Slough was deviated into the drainage basin of Taylor
Slough. This cutback in freshwater renovation is likely
to have induced strong evaporative conditions in Oys-
ter Bay at this time. Natural climatic variability (the
mid-1980s drought in South Florida) appears to have
further impacted environmental conditions and prob-
OSTRACODE PALEOENVIRONMENTAL RECONSTRUCTION: ALVAREZ ZARIKIAN ET AL. 141
ably outweigh the impact of anthropogenic forces in
this time period. Extracting information embedded in
the ostracode record to assess and isolate the effects
of water management practices may then, not be easy
in this case. The modest presence of the oligohaline
species Cypridopsis vidua, Cytheridella illosvayi, Dar-
winula stephensoni (7, 3 and 7% respectively) in the
early 1980s is considered the result of sediment transport.
INTERPRETATION OF STABLE ISOTOPIC COMPOSITION
Stable isotopic analysis of coastal ostracodes is not
routine and it is fraught with many complexities as
such studies are still in an early stage. Freshwater and
deep sea ostracodes are comparatively simpler than in
coastal taxa because variations in salinity and temper-
ature in those environments are relatively small and
require the understanding of control by only a few
parameters (7.e., ice volume). Coastal species are con-
trolled by many other environmental parameters (i.e.,
rainfall, evaporation) that have the potential to change
dramatically in a seasonal basis. Therefore, the under-
standing of the interaction between these processes and
the isotopic composition of ostracode valves is not
straightforward, but we will attempt to demonstrate the
feasibility of such studies in paleoenvironmental re-
construction of coastal ecosystems.
The ubiquitous occurrence of Peratocytheridea se-
tipunctata made this species ideal for an evaluation of
the utility of stable isotopic analysis as a paleoenvi-
ronmental tool. The stable isotopic composition (6!°O
and 6"%C) of ostracodes principally responds to two
factors: water chemistry and temperature (Chivas er
al., 1986). Differences in the physiology of organisms
give rise to the so-called vital effects (von Grafenstein
et al., 1999). This means that different species inhab-
iting the same environment have similar responses to
environmental factors, but can be offset from one an-
other. Vital effects can explain when an organism ap-
pears to be out of equilibrium from the predicted values.
Important to this study is the understanding of the
environmental factors that control 6!*O and 6'C in-
corporation in ostracode valves. These are primarily
water chemistry and in the case of 5!8O, temperature,
rainfall and evaporation (Swart er al., 1996). Strong
excursions to more negative values of 5'8O of Pera-
tocytheridea setipunctata that take place prior to 1925
+ 4 correlate to high-rainfall years, and are consistent
with relatively fresher conditions or rather constant
brackish water conditions in a typical estuary. In con-
trast, strong positive excursions to heavier 5!*O values
occurred during the mid-1980s. As evaporation pro-
ceeds, it preferentially removes '°O and leaves the wa-
ter behind enriched in '%O; thus, the 6!°O value of the
water becomes more positive. This may have been the
case for the mid-1980s when heavier oxygen isotopic
values were concurrent with a major regional drought
in South Florida, as well as with algal blooms and a
major die-off of sea grasses (Swart er al., 1999). Evap-
oration of marine or brackish waters will result in a
correlation between salinity and 6!°O, whereas evap-
oration of freshwater produces no such correlation.
Some of the intricacies this difference causes in inter-
pretation of 'SO results in the Everglades/Florida Bay
system are as follows.
Commonly marine waters have 6'8O values between
O and +1%o SMOW, while freshwater values are more
negative. In South Florida, the 6'%O of rainfall is ap-
proximately —3%c (Swart ef al., 1989). During drier
years, when precipitation is appreciably lower than the
annual mean of 150 cm, the intense evaporation in the
Everglades can cause freshwater to have a 5!*O value
as high as +3%o (in contrast to the marine signal of 0
to + 1%c). Reduction in freshwater flow may also allow
a greater intrusion of marine water resulting in an in-
crease of isotopic composition as well. However, in
this instance the 6'°O will increase because of the
movement of the salt water/freshwater transition in-
land and not because of the enhanced evaporation.
Both cases will result in a negative correlation between
salinity and 6'8O. In wet years, the Everglades water
is closer in isotopic composition to the original rainfall
value (~—3%c). Finally, during very wet periods, the
freshwater will retain its original negative 6'8O value
and when mixing with marine water it will produce a
positive covariance between oxygen and salinity (Nel-
sen et al., in press).
In coastal waters, the 61°C is also highly variable
and the dissolved inorganic carbon (DIC) will be high-
ly depleted as a result. The carbon isotopic composi-
tion of calcareous organisms can be principally related
to the 8'°C of the DIC pool as well as the physiological
state of the organism. Under normal marine condi-
tions, physiological processes are probably most im-
portant in controlling interannual variations in the 6%C
of the calcitic skeleton. In situations such as Oyster
Bay, variations in the DIC probably dominate. Varia-
tions in the 6%C of the DIC are caused by a combi-
nation of the oxidation of organic material, which pro-
duces water highly depleted in 5'°C, and the introduc-
tion of marine waters, which are relatively enriched in
54C (Nelsen et al., in press). During the wet season,
Oyster Bay will be inundated by freshwater that will
be isotopically depleted in 6'°O and 6/3C. In contrast,
during the dry season, the area will be influenced more
by marine water from the Gulf, which will be heavier
in 6!8O and 6'°C. Generally, the ostracode isotopic data
from Oyster Bay shows positive correlations between
5C and '8O suggesting a normal interaction between
142 BULLETIN 361
freshwater and marine waters (Text-fig. 7). Positively
correlated 5'°O and 6'°C deviations suggest an asso-
ciation between terrestrially derived, isotopically light,
C and O.
Exceptions to the positive correlation between the
6'8O and 6'C in the ostracode data occur in the 1970s,
and the 1930s and 1940s. The offset in the 1970s is
significant because it occurs immediately prior to a
substantial shift towards more positive oxygen and
carbon isotopic compositions that are evident in the
ostracode isotopic record. At this time, flow in the
Shark River Slough and the rainfall record at Home-
stead and Flamingo are decoupled (Nelsen ef al., in
press), suggesting a major modification in water flow
prior to this time period. The other periods (1930s and
1940s), which exhibit a negative correlation between
dC and 6'8O, are more difficult to understand because
of an absence of hydrological data and the lower abun-
dance of ostracodes.
CONCLUSIONS
The following conclusions can be derived from the
ostracode fauna and isotopic data. Natural variability
in the ostracode communities over time is the result of
natural seasonal fluctuations in climate and hydrolog-
ical conditions. Hurricane events and their effects are
the major cause for immediate modifications within
ostracode populations. However, these populations are
capable of a rapid recovery. Rainfall is the dominant
natural driving force behind long-term variability in
microfaunal assemblages.
The Oyster Bay core showed a large decline in os-
tracode diversity and abundance at about 1980. The
timing of this event is best correlated to the onset of
the SFWMD “Rainfall Plan” and the modification of
the natural flow of freshwater running through Shark
River Slough to the drainage basin of Taylor Slough.
This cutback in freshwater renovation to Oyster Bay
is likely to have induced strong evaporative conditions.
In addition, the closing of the Buttonwood canal, also
at this time, may have reduced water exchange be-
tween Whitewater Bay and Florida Bay, and further
stressing Oyster Bay’s microfaunal communities.
Strong positive deviations in 6'C and 6!*O for both
species, Peratocytheridea setipunctata and Ammonia
parkinsoniana in the mid 1980s were concurrent with
a documented major regional drought in South Florida,
as well as with algal blooms and a major die-off of
sea grasses in Florida Bay (Fourqurean and Robblee,
1999; Robblee er al., 1991).
In general, the isotopic data from Oyster Bay shows
positive correlations between '%C and '8O for both P.
setipunctata and A. parkinsoniana suggesting a normal
interaction between freshwater and marine waters. An
exception that occurred in the 1970s is probably the
result of decoupling of rainfall and freshwater flow
down Shark River Slough during this time period due
to changes in water management.
REFERENCES CITED
Alvarez Zarikian, C.A., Hood, T., Blackwelder, P.L., Swart,
P.K., Nelsen, T.A., Wanless, H.R., Trefry, H.J., and
Kang, W-J.
1999a. Paleoecological significance of temporal variations in as-
semblages and isotopic composition of Ostracoda and Fo-
raminifera from the lower Everglades/Florida Bay eco-
system. Abstracts, 1999 AGU Spring Meeting, Boston,
MA, OS31A-17, p. S178.
Alvarez Zarikian, C.A., Blackwelder, P.L., Hood, T., Swart,
P.K., Nelsen, T.A., and Featherstone, C.
1999b. A century of hydrological variability in the lower Ever-
glades National Park as interpreted from stable isotopes
on ostracodes and foraminifers. Program and Abstracts,
1999 Florida Bay and Adjacent Marine Systems Science
Conference, Key Largo, Florida, p. 198.
Benda, W., and Puri, H.
1962. The distribution of Foraminifera and Ostracoda off the
Gulf Coast of the Cape Romano area, Florida. Transac-
tions, Gulf Coast Association of Geological Societies, vol.
12, pp. 303-342.
Boyer, J.N., Fourqurean, J.W., and Jones, R.D.
1997. Spatial characterization of water quality in Florida Bay
and Whitewater Bay by multivariate analyses: Zones of
similar influence. Estuaries, vol. 20, pp. 743-758.
Brewster-Wingard, G.L., and Ishman, S.E.
1999. Historical trends in salinity and substrate in central Flor-
ida Bay: A paleoecological reconstruction using modern
analog data. Estuaries, vol. 22, no. 2B, pp. 369-383.
Brewster-Wingard, G.L., Stone, J.R., and Holmes, C.W.
2001. Molluscan faunal distribution in Florida Bay, past and
present: An integration of down-core and modern data.
Bulletins of American Paleontology, no. 361, pp. 199—
231:
Chivas, A.R., DeDeckker, P., and Shelley, J.M.G.
1986. Magnesium content of non-marine ostracod shells: a new
palaeosalinometer and paleothermometer. Palaeogeogra-
phy, Palaeoclimatology, Palaeoecology, vol. 54, pp. 43—
6l.
Cronin, T.M., Holmes, C., Wingard, L., Ishman, S., Dowsett, H.,
and Waibel, N.
2001. Historical trends in epiphytal ostracodes from Florida
Bay: implications for seagrass and macro-benthic algal
variability. Bulletins of American Paleontology, no. 361,
pp. 159-197.
Davis, S.M., and Ogden J.C., editors.
1994. Everglades: The Ecosystem and its Restoration. St. Lucie
Press, Delray Beach, Florida.
Dwyer, G.S., and Cronin, T.M.
1999. Reconstructing the salinity history of Florida Bay using
ostracode shell chemistry. Program and Abstracts, 1999
Florida Bay and Adjacent Marine Systems Science Con-
ference, Key Largo, Florida, p. 158.
OSTRACODE PALEOENVIRONMENTAL RECONSTRUCTION: ALVAREZ ZARIKIAN ET AL. 143
2001. Ostracode shell chemistry as a paleosalinity proxy in Flor-
ida Bay. Bulletins of American Paleontology, no. 361, pp.
249-276.
Fourqurean, J.W., and Robblee, M.B.
1999. Florida Bay: A history of recent ecological changes. Es-
tuaries, vol. 22, no. 2B, pp. 345-357.
Halley, R.B., and Roulier, L.M.
1999. Reconstructing the history of eastern and central Florida
Bay using mollusk-shell isotope records. Estuaries, vol.
22, no. 2B, pp. 358-368.
Hood, T.A., Alvarez Zarikian, C. A., Blackwelder, P.L., Swart,
P.K., Nelsen, T.A., Wanless, H.R., Trefry, J.H., Kang,
W-J., Tedesco, L., and O’Neal, M.
1998. The sediment record as a monitor of natural and anthro-
pogenic changes in the lower Everglades/Florida Bay eco-
system. Proceedings, 1998 Florida Bay Science Confer
ence, p. 2.
Ishman, S.E.
2001. Ecological controls on benthic foraminifer distributions in
Biscayne Bay, Florida. Bulletins of American Paleontol-
ogy, no. 361, pp. 233-247.
Keyser, D.
1975. Ostracodes of the mangroves of south Florida, their ecol-
ogy and biology. Bulletins of American Paleontology, vol.
65, no. 282, pp. 489-499.
1977. Ecology and zoogeography of recent brackish water Os-
tracoda (Crustacea) from southwest Florida, in Loetfler,
H., and Danielopol, D.L., eds., Aspects of Ecology and
Zoogeography of Recent and Fossil Ostracoda. Proceed-
ings, 6th International Symposium on Ostracodes, Salz-
burg, 1976, pp. 207—222.
Light, S.S., and Dinnen, J.W.
1994. Water control in the Everglades: a historical perspective,
pp. 47-84, in Davis, S.M., and Ogden, J.C., eds., Ever-
glades: The Ecosystem and its Restoration. St. Lucie
Press, Delray Beach, Florida.
MacArthur, R.H.
1965. Patterns of species diversity. Cambridge Philosophical So-
ciety, vol. 40, pp. 510-533, in Patrick, R.P., editor, 1983.
Diversity, Benchmark Papers in Ecology, vol. 13, Hutch-
inson Ross Publishing Company, Stroudsburg, PA, pp.
14-37.
Mcelvor, C.C., Ley, J. A., and Bjork, R.D.
1994. Changes in freshwater inflow from the Everglades to Flor-
ida Bay including effects on biota and biotic processes:
A review, in Davis, S.M., and Ogden, J.C., eds., Ever-
glades: The Ecosystem and its Restoration. St. Lucie
Press, Delray Beach, Florida, pp. 117—146.
Neal, J.W.
1988. Ostracodes and paleosalinity reconstruction, in De-
Deckker, P., Colin, J-P., and Peypouquet, J-P., eds., Ostra-
coda in the Earth Sciences, Elsevier, p. 302.
Nelsen, T.A., Wanless, H.R., Trefry, J.-H., Alvarez-Zarikian,
C.A., Hood, T., Blackwelder, P.L., Swart, P.K., Tedes-
co, L., Kang, W-J., Metz, S., Garte, J., Featherstone,
C., Souch, C., Pachut, J.F., O’Neil, M., and Ellis, G.
2001. Linkages between the South Florida peninsula and coastal
zone: a sediment-based history of natural and anthropo-
genic influences in Linkages Between Ecosystems in the
South Florida Hydroscape: The River of Grass Continues.
K. Porter and J. Porter, eds., CRC Press, Boca Raton,
Florida (in press).
Patrick, R.P., editor.
1983. Diversity. in Benchmark Papers in Ecology, vol. 13,
Hutchinson Ross Publishing Company, Stroudsburg, PA,
413 pp.
Puri, H.
1960. Recent Ostracoda from the west coast of Florida. Trans-
actions, Gulf Coast Association of Geological Societies,
vol. 10, pp. 107-149.
Puri, H., and Hulings, N.
1957. Recent ostracode facies from Panama City to Florida Bay
area. Transactions, Gulf Coast Association of Geological
Societies, vol. 7, pp. 167—190.
Robblee, M.B., Barber, T.R., Carlson, P.R., Durako, M.J., Four-
qurean, J.W., Muehlstein, L.K., Porter, D., Yarbro,
L.A., Zieman, R.T., and Zieman, J.C.
1991. Mass mortality of the tropical sea-grass Thalassia testu-
dinum in Florida Bay (USA). Marine Ecology Progress,
Series, vol. 71, pp. 297-299.
Rudnick, D.T., Chen, Z., Childers, D.L., Boyer, J.N., and Fon-
taine, III, T.D.
1999. Phosphorous and nitrogen inputs to Florida Bay: The im-
portance of the Everglades watershed. Estuaries, vol. 22,
no. 2B, pp. 398-416.
Smith, T.J., Robblee, M.B., Wanless, H.R., and Doyle, T.W.
1994. Mangroves, hurricanes, and lightning strikes. Biosciences,
vol. 44, p. 256.
Swart, P.K., Healy, G., Dodge, R.E., Kramer, P., Hudson, H.,
Halley, R., and Robblee, M.
1996. The stable oxygen and carbon isotopic record from a coral
growing in Florida Bay: A 160-year record of climate and
anthropogenic influence. Palaeogeography, Palaeoclima-
tology, Palaeoecology, vol. 123, pp. 219-237.
Swart, P.K., Healy, G., Greer, L., Lutz, M., Saied, A., Anderegg,
D., Dodge, R.E., and Rudnick, D.
1999. The use of proxy chemical records in corals skeletons to
ascertain past environmental conditions in Florida Bay.
Estuaries, vol. 22, no. 2B, pp. 384-397.
Swart, P.K., Sternberg, L., Steinen, R., and Harrison, S.A..
1989. Controls on the oxygen and hydrogen isotopic composi-
tion of waters from Florida Bay. Chemica et Geochemica
(Isotopic. Geoscience Section), vol. 79, pp. 113-123.
Tabb, D.C., and Jones, A.C.
1962. Effect of Hurricane Donna on the aquatic fauna of North
Florida Bay. Transactions of the American Fisheries So-
ciety, vol. 9, p. 375.
Turpen, J. B., and Angell, R.W.
1971. Aspects of molting and calcification in the ostracod Het-
erocypris. Biological Bulletin, vol. 140, pp. 331—338.
Von Grafenstein, U., Erlernkeuser, H., and Trimborn, P.
1999. Oxygen and carbon isotopes in modern fresh-water ostra-
code valves: assessing vital effects and autecological ef-
fects of interest for paleoclimate studies. Palaeogeogra-
phy, Palaeoclimatology, Palaeoecology, vol. 148, pp.
133-152.
Wanless, H.R., and Tagett, M.G.
1989. Origin, growth and evolution of carbonate mudbanks in
Florida Bay. Bulletin of Marine Science, vol. 44, p. 454.
Wanless, H.R., Parkinson, R.W., and Tedesco, L.P.
1994. Sea level control on stability of Everglades wetlands, in
Davis, S.M., and Ogden, J.C. eds., Everglades: The Eco-
system and its Restoration. St. Lucie Press, Delray Beach,
Florida, pp. 199-223.
we
ie it Ww
af?
i ms af é wt yoy
7 yy »
a) & i hee ‘
° a 13 ini tel a in)
7 5 7 “ U
. ha" oh nah eh ae ae
s\ Se. ot co ay one ; — \b uh wey - 198 7 awe a i, d
ie Saas | nae: 2 Va Wi Wiel”?
is n qe é Pcotm 7 ” ~ oe | | mee ;
a any S
Co Der: » 7 _ i be : — Od, sy te : ~
ete eres | ree ae) ms
_ ap ta Peel Ti mh : ‘a) previ 7 py wh
l= j vad tee
wr: 7 sie sv : i ea Ti) Y, si)
oe en re oat Oona = Ohi tnt a
Set aa<- fiat @ 5 Psy =e “aut Acer —_ :
Aw = 4 =i) » aw
'\ | ae? AR swe | Py Cee este 7
9a = ae, ) re py Se =~ ke ee a
f a "ii
ican -_— reat a tera '
ies 0 oer
2 y
a - ~~
CHAPTER 8
DIATOMS AS INDICATORS OF ENVIRONMENTAL CHANGE IN SEDIMENT CORES
FROM NORTHEASTERN FLORIDA BAY
J. K. HUVANE AND S. R. COOPER
Duke University Wetland Center
Nicholas School of the Environment
Durham, NC 27708
ABSTRACT
Diatom assemblages found in two sediment cores from northeastern Florida Bay indicate fluctuations in both salinity and
extent of sub-aquatic vegetation cover during the past 100 years. Diatoms from Russell Bank core 19A indicate that salinity was
high prior to the 20th century. Between 1890 and 1920, assemblages dominated by Mastogloia species prevail. After 1920, there
is a dramatic shift to assemblages dominated by an epipelic taxon, Nitzschia granulata. Around 1950, this taxon declines in
abundance and epiphytic species increase. Epiphytic species remain common until about 1972. After 1972, epiphytic diatoms
decrease in abundance, and diatoms indicative of higher salinity increase. Similar trends are seen in the Pass Key core 37 (which
only spans about 35 years, between 1960-1996). Epiphytic diatoms are common between about 1967 and 1977, and there is an
overall trend towards increasing salinity upcore. These findings are in general agreement with other studies of fossil indicators
related to salinity and seagrass cover found in Florida Bay sediment cores. These fluctuations may be related to anthropogenic
activities, such as the reduction of freshwater flow to Florida Bay from the Everglades as a result of water management practices.
INTRODUCTION
In the past decade, the ecological health of South
Florida has been questioned, particularly that of the
Everglades and Florida Bay (e.g., Davis and Ogden,
1994; McPherson and Halley, 1996). For instance, re-
cent seagrass die-offs, algal blooms and periods of hy-
persalinity have generated interest in the extent of an-
thropogenic impacts on Florida Bay. The advent of
ecological studies and monitoring programs in Nation-
al Parks has allowed managers to move from a belief-
based management system to science-based ecosystem
management (Davis and Halvorsen, 1996). For ex-
ample, the Everglades Forever Act was passed in
1994, which mandates that the ecosystem be returned
to its “natural state”. One of the pivotal issues that
arise from this mandate is how this “natural state”
should be defined. Commonly, the goal is to restore
the environment to conditions that existed prior to ma-
jor European anthropogenic disturbance, such as ur-
banization (e.g., Holling et al., 1994; MacMahon,
1998). In the absence of historical monitoring data that
defines the natural variation of the ecosystem, resto-
ration management decisions are exceedingly difficult.
Paleoecological applications, which allow for the re-
construction of past environments, provide proxy data
to managers interested in restoration (e.g., Brewster-
Wingard et al., 1998). Paleoecological techniques have
been used successfully to help solve environmental
problems such as lake acidification (e.g., Battarbee et
al., 1999) and eutrophication (e.g., Hall and Smol,
1999). More recently, paleoecological techniques have
been applied to estuarine systems (e.g., Juggins, 1992;
Cooper, 1995a; Brewster-Wingard et al., 1998a; Ish-
man et al., 1998; Cooper, 1999).
This study deals with the use of diatoms as indica-
tors of environmental change in Florida Bay sediment
cores. Diatoms are microscopic algae that commonly
occur in aquatic environments. They have a siliceous
shell, or frustule, that is generally well preserved in
aquatic sedimentary environments. Species designa-
tions are based on the morphology of their silica shell,
making them particularly attractive as fossil indicators.
In addition, species distributions are closely linked to
water quality parameters such as pH, salinity, and nu-
trient availability (e.g., Sweets et al., 1990; Fritz et al.,
1991; Dixit et al., 1993; Fritz et al., 1993; Wilson et
al., 1994; Cooper et al., 1999).
We present diatom assemblage data from sediment
cores from two sites in northeastern Florida Bay. The
main objective of the study is to provide a diatom-
based reconstruction of environmental conditions in
this sub-region of Florida Bay during the last century.
We examine past changes in salinity and sub-aquatic
vegetation cover.
ACKNOWLEDGMENTS
Funding for this study was provided by the U.S.
Geological Survey South Florida Ecosystem Program.
The Duke University Wetland Center, Nicholas School
146 BULLETIN 361
|
ADES NATIONAL’ | J
PARK— —/
Text-figure 1—A map of Florida Bay with the 26 USGS monitoring sites labeled. The Russell Bank and the Pass Key sites are indicated
by the “xX.”
of the Environment, provided additional support. We
thank Laura Pyle for performing the diatom counts for
Pass Key core 37. We would also like to thank G.
Lynn Brewster-Wingard of the U. S. Geological Sur-
vey, Reston, VA, who provided sediment samples for
diatom analyses. The Keys Marine Lab, Florida Insti-
tute of Oceanography, Long Key, Florida provided
field assistance.
METHODS
CORE COLLECTION AND CHRONOLOGY
Sediment cores were collected from two sites: Rus-
sell Bank and Pass Key (Text-fig. 1). Russell Bank
sediment cores were collected in February 1995 by the
U.S. Geological Survey (St. Petersburg, FL) in co-
operation with the South Florida Water Management
District and the Everglades National Park (Brewster-
Wingard ef al., 1997). Cores were collected from the
south side of Russell Bank (N 25° 03’ 50.04”, W 80°
37’ 29.28") from a seagrass bed, at a water depth of
one half meter. Russell Bank core 19A (RB 19A) was
taken alongside Russell Bank core 19B (RB 19B), and
the two are considered replicate cores. A third core,
Russell Bank 19C (RB 19C) was taken 54 m north of
cores 19A and 19B. Cores were dated using 7!°Pb ra-
diometric-dating techniques. At present, only cores RB
19B and RB 19C have been analyzed for 7!°Pb activity
(Brewster- Wingard et al., 1997).
Samples from core RB 19A were used for diatom
analyses. Core RB 19A was 140 cm in length and
subsamples were taken at | cm intervals between O—
30 cm and at 2 cm intervals between 30—140 cm. Se-
lect samples were analyzed for diatoms.
Pass Key core 37 (PK 37) was collected by the U. S.
Geological Survey (St. Petersburg, FL) in May 1996.
The core was taken at N 25° 08’52.08” W 80°
34'28.20". This core was 76 cm in length and was
subsampled at 2 cm intervals. Core chronology was
established by *!°Pb radiometric-dating techniques. De-
tails of the dating methods can be found in Robbins
et al. (in press).
DIATOM EXTRACTIONS
A modified version of the procedure described by
Funkhauser and Evitt (1959) was used to extract dia-
toms from sediment core samples. Approximately 0.5
mg dry weight sediment was digested for each sample
from core RB 19A. For core PK 37, the amount of
sediment digested varied between approximately 0.7
and 2.4 mg dry weight per sample. Samples were treat-
ed with H,O,, followed by HCl, and finally an HNO,
and K,Cr,O, digestion. Sand particles were removed
by a coarse fractionation procedure.
Measured aliquots of the final diatom slurries were
permanently mounted in Naphrax® (Northern Biolog-
ical Supply, Ipswich, UK) (Refractive Index = 1.7) on
microscope slides. Slides were examined with a light
microscope at a magnification of 1000X. Light micro-
scope digital images of most diatom taxa were cap-
tured via CCD video camera and frame grabber. These
images are on file at the Duke University Wetland
Center.
DIATOMS AND ENVIRONMENTAL CHANGE: HUVANE AND COOPER 147
DIATOM IDENTIFICATIONS
Taxonomic identifications were aided by a number
of resources including Peragallo and Peragallo (1897—
1908); Hustedt (1927-1964, 1955); Patrick and Rei-
mer (1966); DeFelice (1976); Stephens and Gibson
(1979, 1980a, 1980b); Yohn and Gibson (1982a,
1982b); Foged (1986); Navarro (1982a, 1982b); Kram-
mer and Lange-Bertalot (1986, 1988, 1991a,b); and
Cooper (1995b).
However, there are diatom taxa that have yet to be
identified to species level. Some of these taxa may not
be described in the literature. These taxa were identi-
fied to the generic level and given a species letter des-
ignation, e.g., sp. A.
For core PK 37, 15 samples were analyzed and be-
tween 300-500 diatom valves were enumerated and
identified from each sample. For core RB 19A, 31
samples were analyzed and approximately 400 diatom
valves were enumerated and identified from each sam-
ple.
DATA ANALYSIS
Down-core changes in diatom assemblages were ini-
tially analyzed by examining changes in the relative
abundance of dominant taxa. Shannon-Wiener diver-
sity, (~~ pn p;, where p; = the proportion of the total
number of valves belonging to the ith taxa in the as-
semblage) was calculated for each sample following
Wetzel, 1983. Centric: Pennate (C:P) ratios were cal-
culated for each sample. Centric and pennate are tax-
onomic terms related to diatom valve (shell) symme-
try. In these particular Florida Bay samples, the centric
diatoms are primarily planktonic and the pennate dia-
toms are benthic and epiphytic. The C:P ratio provides
a simple index of the relative abundance of planktonic
versus benthic and epiphytic diatoms in these samples.
A multivariate ordination technique, detrended cor-
respondence analysis (DCA), was used to analyze the
species assemblage data. Detrended correspondence
analysis is an indirect ordination analysis that arranges
sites along axes based on species composition (ter
Braak, 1995). Ordination analyses indicate which sites
are similar in species composition. The ordination
analyses were based on the percent abundances of the
most common diatom taxa in each core. These anal-
yses were done separately for each core. For core PK
37, a total of 43 taxa were included in the analysis.
Each of these 43 taxa comprised at least 1% of the
total diatom count in at least two samples. Similarly,
for core RB 19A, 36 taxa that comprised at least 1%
of the diatom assemblage in two or more samples were
used. The gradient length for the Pass Key dataset was
relatively short (<2.0 standard deviation units). There-
fore, a linear ordination technique, principal compo-
nents analysis (PCA) was used instead of the unimodal
DCA in the final analysis of the Pass Key data (ter
Braak, 1995).
A stratigraphically constrained cluster analysis,
CONISS (Grimm, 1987), was performed on the same
data sets for both cores (PK 37 and RB 19A). This is
a program that performs cluster analysis by incremen-
tal sums of squares, with the constraint that only ad-
jacent core samples can be merged.
Diatom species distributions are displayed as a ratio
to the individual species maximum in the core sam-
ples. For example, if the maximum abundance of a
given species were 30%, then a particular sample with
a percent abundance (for that species) close to 30%
would have a ratio close to 1. This form of display
allows one to easily examine relative changes in the
abundance of the less common species. Some of these
less common species are indicative of certain environ-
ments, and thus changes in their abundance provide
information concerning past environmental conditions.
RESULTS
CORE CHRONOLOGIES
The *!°Pb activity curves for cores RB 19A and RB
19C were very similar. Sedimentation rates for these
two cores were estimated at 1.22 cm/yr + 0.05 cm
(Brewster-Wingard et al., 1997). Given this sedimen-
tation rate, core RB 19A is approximately 115 years
old, spanning the period from about A.D. 1880 to
1995;
The sedimentation rate for core PK 37 was esti-
mated to be on the order of 2.06 cm/yr + 0.2. Evi-
dence from other sources supports this estimation. Ae-
rial photo evidence indicates that the opening south of
Pass Key has been filling in since the 1950’s (Brew-
ster-Wingard ef al., 1998b). In addition, Casuarina
pollen is present throughout the core indicating rela-
tively recent deposition (less than 100 years) because
this species was not introduced to the region until the
late 1800’s or early 1900’s and would therefore not be
present at an earlier date. (Brewster-Wingard ef al.,
1998b).
Russell Bank Core 19A
Diatom Flora
A total of 157 diatom taxa have been identified from
core RB 19A. The five most abundant taxa in the core
samples examined are Cyclotella cf. litoralis Lange
and Syversten, Nitzschia granulata Grunow, Cyclotel-
la cf. striata (Kutz.) Grunow, Grammatophora cf.
oceanica var. macilenta (Wm. Smith) Grunow, and Sy-
nedra sp. A. Representative images of these taxa are
BULLETIN 361
148
“
DIATOMS AND ENVIRONMENTAL CHANGE: HUVANE AND COOPER 149
max. = 71.8%
Nitzchia granulata
max. = 12.8%
Depth (cm)
Cocconeis cf. placentula
var. euglypta
max. = 7.3%
100
125
150
Mastogloia crucicula
max. = 61.2%
max. = 10.3%
Synedra sp. A
max. = 2.9%
Mastogloia corsicana
1995
1970
1945
1920
max. = 15.5%
Grammatophora cf.
oceanica var. macilenta
1895
1870
1995
1970
max. = 11.9% | 1945
Date
1920
aes 1895
Mastogloia sp. A
1870
1995
1970
1945
max. = 2.8% 1920
1895
Mastogloia ovalis
1870
OOM O:2 10406) 0:8) 1101010" 0:2" 0:4 O16 0/8" 1.0. 010 0:2 0:4 016 0:8 1.0
Proportion
Text-figure 3.—Relative abundances of select taxa from the Russell Bank core 19A. Relative abundances are standardized to the maximum
abundance for each species. Values of 1 indicate a maximum abundance for that species. Samples are plotted by sample depth in centimeters.
Dates (based on the average sedimentation rate derived from Pb-210 analyses) are shown on the right.
shown in Text-figure 2. The percent abundance of
these species standardized to the maximum abundance
of each species are shown in Text-figure 3. The two
species of Cyclotella are grouped together in the Cy-
clotella graph (Text-fig. 3) because these species were
sometimes difficult to distinguish. The maximum
—
abundance (71.8%) of Cyclotella species occurs in the
lower part of the core at the 88—90 cm sample interval.
A second peak occurs at the top of the core, where the
percent abundance reaches 60% of the total diatom
count. Mastogloia species are most common in the
bottom third of the core, comprising an average abun-
Text-figure 2.—Light microscopic photographic images of representative diatoms from Russell Bank core 19A. a) Mastogloia corsicana b)
Mastogloia crucicula c) Mastogloia sp. R d) Cocconeis placentula var. euglypta e) Mastogloia ovalis f) Mastogloia sp. A g) Nitzschia granulata
h) Cyclotella cf. striata i) Synedra sp. A j) Grammatophora cf. oceanica var. macilenta k) Cyclotella cf. littoralis.
150 BULLETIN 361
dance of 24% between the 124-126 and 90-92 cm
samples. The most common Mastogloia species in this
portion of the core are Mastogloia sp. A and Masto-
gloia sp. R. Mastogloia species decline in abundance
between 80 and 56 cm, with an average abundance of
only 6% over this interval (Text-fig. 3). Above the 56—
58 cm sample, Mastogloia species once again increase
with an average abundance of 15%. However, the spe-
cies differ from those present in the bottom third of
the core. In the upper portion of the core, Mastogloia
crucicula Grunow, Mastogloia corsicana Grunow, and
Mastogloia ovalis A. Schmidt are among the species
that increase in abundance (Text-fig. 3).
Nitzschia granulata is an important component of
the diatom assemblages in the middle of the core be-
tween 80 and 28 cm (Text-fig. 3), with an average
relative abundance of 26% over this interval, and a
maximum abundance of 61% at the 70—72 cm sample.
Nitzschia granulata declines above 60 cm and decreas-
es to less than 2% of the assemblage in samples above
28 cm. Grammatophora ct. oceanica var. macilenta
becomes more common above 80 cm, with an average
abundance increasing from 1.5% (below 80 cm) to 9%
(above 80 cm), although abundances decline near the
top of the core. Cocconeis placentula var. euglypta
(Ehr.) Cleve is most abundant in the 50—52 cm sample,
with a maximum abundance of over 12% (Text-fig. 3).
The relative abundance of this species is fairly high
until about 28 cm. Synedra sp. A occurs throughout
the core, but percentages of this taxon increase above
30 cm, with a maximum abundance of 10.3% in the
20—21 cm sample (Text-fig. 3).
Detrended Correspondence and Cluster Analyses
In the DCA diagram, samples with similar diatom
assemblages have similar axis | and 2 scores and are,
therefore, closer together on the plot (Text-fig. 4). Ar-
rows drawn between samples (from the bottom to the
top of the core) show how the diatom assemblages
have changed over the time period represented by the
core. Several groups of samples emerge from this anal-
ysis. These are the bottom-most group, between 136
and 130 cm, a second group between 130 and 90 cm,
a transition group between 90 and 76 cm, a third group
between 76 and 56 cm, another transition group be-
tween 28 and 56 cm, and the uppermost samples be-
tween 28 and 4 cm.
The cluster analysis shows that the greatest change
in diatom assemblages occurs between the 80-82 cm
and 90—92 cm samples (Text-fig. 5). Other sub-clusters
contain samples between 130 and 136 cm, 130 and 90
cm, 90 and 56 cm, 56 and 28 cm, and samples between
28 and 4 cm.
These groupings indicate that the diatom flora can
Russell Bank Core 19A
Axis 2A =0.11
Axis 2A = 0.36
Text-figure 4.—Detrended correspondence analysis ordination
plot based on the diatom assemblages in Russell Bank core 19A.
The numbers on the figure indicate sample numbers (the top depth).
Circles are drawn around the samples from each zone. Arrows are
drawn sequentially from zone I to zone V and indicate how diatom
assemblages have changed during the last century. Diatom zones are
indicated by the following symbols: Zone I (136-130 cm; 1882—
1886 A.D.): filled circles. Zone I (130-90 cm; 1882-1921 A.D.)
filled squares. Zone HI (90—56 cm; 1921—1949 A.D.) filled diamond.
Zone IV (56—28 cm, 1949-1972 A.D.) open squares. Zone V (28—
4 cm; 1972-1992 A.D.) stars.
be separated into five time zones represented by depth
intervals in the core. The average sedimentation rate
of 1.22 cm/yr was used to calculate approximate time
intervals: Zone I ca. A.D. 1882—1886 (136-130 cm),
zone II ca. A.D. 1886-1921 (130—90 cm), zone III ca.
A.D. 1921-1949 (90-56 cm), zone IV ca. A.D. 1949—
1972 (56-28 cm), and zone V ca. A.D. 1972-1992
(28—4 cm).
Zone I is characterized by higher percentages of Cy-
clotella, and generally poorer valve preservation than
the rest of the core. Zone II is best characterized by
high percentages of Mastogloia species. Zone III
shows a major shift to diatom assemblages dominated
by Nitzschia granulata. Grammatophora cf. oceanica
var. macilenta also increases during this zone. Zone IV
has high percentages of N. granulata, as well as higher
relative abundances of Mastogloia crucicula, M. cor-
sicana, M. ovalis, and Cocconeis placentula var. eu-
glypta. A decrease in N. granulata and increases in
Synedra sp. A and Cyclotella species characterize zone
V. According to the DCA, zone V and zone I are sim-
ilar. This reflects the high abundance of Cyclotella in
these samples and the low abundance of N. granulata.
Due to the poorer preservation of valves in zone I, it
is not clear if the two zones represent similar diatom
communities in other respects.
DIATOMS AND ENVIRONMENTAL CHANGE: HUVANE AND COOPER 151
o
ee Russell Bank Core 19A
FS
0 ¢ CONISS
1975 25
50
1950
i
© &
8 ee
(7)
19257 O
1004
1900
1254
1875
150°
Total sum of squares
Text-figure 5.—A stratigraphically constrained cluster analysis of the Russell Bank core 19A diatom assemblages. The Y axis shows sediment
depth with the top of the core at the top of the diagram. The second Y axis shows dates based on Pb-210 analyses. The samples counted
along with the diatom zones are indicated on the left.
Diatom Species Diversity
The Shannon-Weiner index of species diversity
takes into account both the number of species (species
richness) in an assemblage and the number of diatom
valves represented by each species. For these samples,
species richness mirrors the total diversity. For 1in-
stance, when most of the valves belong to one or a
few species, both species diversity and species rich-
ness are low. There are several shifts in diatom diver-
sity in core RB 19A (Text-fig. 6). Diversity is low in
the bottom-most samples, partly due to a larger num-
ber of poorly preserved valves that were difficult to
identify, even to the generic level. However, the Pear-
son correlation coefficient for diversity and the percent
1995
1970
1945
Date
1920
Depth (cm)
1895
1870
Difference
Difference
Text-figure 6.—Diversity and C:P profiles for Russell Bank core
19A. Both diversity and the C:P ratio are shown as a difference
from the core average. The core mean diversity = 2.30. The core
mean C:P = 0.56.
unidentifiable valves (typically broken, poorly pre-
served valves) is very weak (1? = 0.004), suggesting
that low diversity of samples is not related to poor
preservation of diatom valves. Diversity is highest be-
tween 90-110 cm, where diatom assemblages are
characterized by an increase in Mastogloia species, yet
preservation quality of the valves is similar to that of
the samples in the bottom of the core. There is a very
sharp decrease in diversity at 88-90 cm, a sample
dominated by Cyclotella species. Diversity values re-
cover and then decrease between 56 and 70 cm. This
decrease reflects the dominance of Nitzschia granulata
valves in these samples. Above 50 cm in the core there
is an increase in diatom diversity. In general, trends in
diversity correspond to the five zones determined by
diatom species composition. Diversity values are be-
low average during zone I (A.D. 1882-1886). During
zone II (A.D. 1882-1921), values increase and then
drop to below average again during zone III (A.D.
1921-1949). Values are above average during zone IV
(A.D. 1949-1972) and then decline slightly during
zone V (A.D. 1972-1995), but do not fall much below
average until ca. A.D. 1995.
Centric : Pennate Ratio
The C:P ratios in Russell Bank core 19A generally
reflect the abundance of planktonic Cyclotella species.
For example, there is a high C:P ratio in the bottom
of the core and a high abundance of Cyclotella species
152 BULLETIN 361
max. = 45.3%
Cyclotella spp.
Nitzschia granulata
max. = 14.2%
Depth (cm)
Mastogloia crucicula Mastogloia erythrea
Fragilaria cf. tabulata
Mastogloia ovalis
00 02 04 06 08 1000 02 04 06 08 1.0 0.0
Proportion
Nitzschia frustulum
Mastogloia elegans
1995
1985
max. = 12.2% max. = 9.4%
Cocconeis placentula var. euglypta
cconeis pi ul iglyp 1956
1985
1976
Date
1966
Mastogloi icana
lastogloia corsical 1956
0.0 02 04 06 O08 1.0
1995
1985
1976
1966
rammatophora cf. peaaticay
var. macilenta} 4956
02 04 06 O08 1.0
Text-figure 7.—Relative abundances of select taxa from the Pass Key core 37. Relative abundances are standardized to the maximum
abundance for each species. Values of 1 indicate a maximum abundance for that species. Samples are plotted by sample depth in centimeters.
Dates (based on the average sedimentation rate derived from Pb-210 analyses) are shown on the right.
(Text-fig 6). The sharp increase in C:P at the 88—90
cm sample interval also reflects the large amount of
Cyclotella valves in this sample. Above 88 cm, the
ratio declines. There is a general increase in the C:P
ratio between 70 and 40 cm. The ratio declines slightly
between 40 and 30 cm, and above 30 cm there is a
gradual increase.
Pass Key Core 37
Diatom Flora
A total of 95 diatom species and varieties were iden-
tified from Pass Key core 37. The five species with
the highest mean percent abundances are Cyclotella cf.
litoralis, Nitzschia granulata, Mastogloia crucicula,
Cocconeis placentula cf. var. euglypta, and Fragilaria
cf. tabulata var. tabulata (Ag.) Lange-Bertalot. Select
images of diatom taxa from PK 37 can be found in
Pyle et al. (1998).
The percent abundances of many of the more com-
mon species show changes throughout the time period
represented by the Pass Key core (Text-fig. 7). Nitz-
schia granulata shows some of the largest changes in
abundance (Text-fig. 7). This species is much more
common deeper in the core, with percent abundances
as high as 45% and 36% in the 64—66 cm and 50-52
cm samples, respectively. Above the 44—46 cm sam-
ple, N. granulata is much less common, with percent
abundances of 10% or less. Cyclotella cf. litoralis also
shows changes in abundance, ranging from 12—18% in
the deeper samples, and increasing to over 20% above
the 44—46 cm sample (Text-fig. 7). The percent abun-
dance of C. cf. litoralis reaches a high of 33% in the
DIATOMS AND ENVIRONMENTAL CHANGE: HUVANE AND COOPER 15
Pass Key Core 37 PCA
0.13
Axis 2A
Axis 1A = 0.69
Text-figure 8.—Principal components analysis ordination plot
based on the diatom assemblages in Pass Key core 37. The numbers
on the figure indicate sample numbers (the top depth). Circles are
drawn around the samples from each zone. Arrows are drawn se-
quentially from zone I to zone V and indicate how diatom assem-
blages have changed during the last century. Zones are indicated by
the following symbols next to the sample number: Zone I (70 cm;
1962 A.D.) filled circle. Zone II (64—50 cm; 1965-1972 A.D.) open
square. Zone III (44—24 cm; 1974-1984 A.D.) diamond. Zone IV
(20—4 cm; 1987-1995 A.D.) plus sign. Zone V (0 cm; 1996) open
circle.
24—26 cm sample, decreases to 17% in the 20—22 cm
sample, and resumes an increasing trend towards the
surface of the core. Nitzschia cf. frustulum (Kutz)
Grun. appears to have become less common through
time (Text-fig. 7). The percent abundance of N. cf.
frustulum decreases from a high of 12% in the 70—72
cm sample to <1% in the 64—66 cm sample. Relative
abundance of this species remains low until the 10—
12 cm sample, where it increases to 5%.
Principal Component and Cluster Analyses
Several groups of samples can be identified from
the PCA ordination diagram of the diatom assemblag-
es from core PK 37 (Text-fig. 8). There is a shift in
species composition from the bottom-most sample at
70-72 cm to the next sample at 64—66 cm. Samples
between 64 and 54 cm group together on the ordina-
tion diagram. Between 50 cm and 44 cm there is a
transition group. The samples between 40 and 24 cm
form a third group. Then there is a shift to a fourth
group consisting of the 20—4 cm samples. The top
oS)
most sample (O—2 cm) contains an assemblage that is
more like the third (40—24 cm) group.
The cluster analysis clearly divides the samples into
three main groups: 70-50 cm, 40—24 cm, and 20—0
cm (Text-fig. 9).
Based on the cluster analysis and the PCA, five di-
atom zones, spanning between 7 and 9 years each, are
evident. The average sedimentation rate of 2.06 cm/yr
was used to calculate the dates. Zone I ca. A.D. 1962
consists of only the one sample at 70—72 cm. Zone II
occurs between ca. A.D. 1965 and 1972 (64—50 cm).
The third zone occurs between ca. A.D. 1974 and 1984
(44—24 cm) and zone IV occurs between ca. A.D. 1987
and 1995 (20—4 cm). The last zone V occurs in A.D.
1996 and contains only the uppermost sample at 0—2
cm.
Zone I is characterized by a peak (12%) in the rel-
ative abundance of Nitzschia cf. frustulum (Text-fig.
7). Nitzschia granulata (Text-fig. 7) has an abundance
of 9% in zone I, and during zone II, the average abun-
dance increases to 34% with a maximum of 45% and
a minimum of 25%. Epiphytic taxa, including Coc-
coneis placentula var. euglypta, Mastogloia crucicula,
and M. ovalis increase in abundance between ca. A.D.
1967-1969 (56—60 cm). These taxa show another in-
crease at the lower boundary of zone III, between ca.
A.D. 1963 and 1975 (44—40 cm). Zone III is also
marked by a sharp decline in N. granulata to a zone
average of 6%. Cyclotella cf. litoralis increases in
zone III with a peak relative abundance of 33% (Text-
fig. 7). In zone IV, there are increases in Amphora
coffeaeformis (Ag.) Kutzing, which has an average of
6.3% during this zone compared to an average of 2%
for the previous three zones combined. There is also
an increase in Mastogloia elegans Lewis to a zone
average of 2% (Text-fig. 7). The average abundance
of this species in the previous three zones is only
0.6%. Zone V is marked by an increase in N. granu-
lata to 7.5%, along with a slight increase in Cyclotella
cf. litoralis.
Diatom Species Diversity
As in core RB 19A, species diversity in core PK 37
mirrors species richness (Pyle et al., 1998). Species
diversity in this core is negatively correlated (1 =
0.75) with the abundance of Nitzschia granulata, the
species that dominates the assemblages of zone II. In
general, diversity is lowest in zone II and highest in
zone IV when there are the lowest percentages of N.
granulata (Text-fig. 10).
Centric : Pennate Ratio
C:P ratios increase in the Pass Key core from be-
tween 0.12 and 0.22 in the deepest samples to over
154 BULLETIN 361
=)
e@
ee Pass Key Core 37
ce as
e CONISS
yp ya
1990
1985
25
1980
E
o S
& r=
19754 o&
fa)
1970 oo
1965
1960 0.2 0.4 0.6
75 Total sum of squares
Text-figure 9.—A stratigraphically constrained cluster analysis of the Pass Key core 37 diatom assemblages. The Y axis shows sediment
depth with the top of the core at the top of the diagram. The second Y axis shows dates based on Pb-210 analyses. The samples counted and
diatom zones are indicated on the left.
0.35 in the 24—26 cm sample, decreasing to 0.18 in
the 20—22 cm sample and then resuming a gradually
increasing trend toward the top of the core. The high-
est ratios occur between 36 and 24 cm (A.D. 1979—
1985).
DISCUSSION
RUSSELL BANK 1882-1949
The bottom zone I (A.D. 1882-1888; 130-138 cm)
in Russell Bank core 19A is difficult to interpret due
to the poor diatom valve preservation. Many of the
1995
Diversity
mean = 3.00
1985
1976
Depth (cm)
Date
cP 1966
mean = 0.25
1956
Difference Difference
Text-figure 10.—Diversity and C:P profiles for Pass Key core 37.
Both diversity and the C:P ratio are shown as a difference from the
core average. The core mean diversity = 3.0. The core mean C:P =
0.25.
broken valve pieces were unidentifiable to species or
even genus level. However, it was possible to deter-
mine whether the valves were pennate or centric. Thus
C:P values are probably not affected by preservation
quality. However, the low species diversity may have
been affected by preservation quality. Salinity may
have been relatively high during this time, as Cyclo-
tella litoralis is a marine species (Lange and Syver-
ston, 1989).
The prevalence of two Mastogloia species (sp. A
and sp. R) in RB 19A zone II (A.D. 1888-1921; 130—
90 cm) is also difficult to interpret at the present time.
Diversity was generally high during this zone. De-
Felice (1976) found that epipelic (bottom-dwelling) di-
atom communities were more diverse as compared to
epiphytic communities in his study of Florida Bay di-
atoms. Therefore, these taxa may be part of an epipelic
assemblage. Further study of the literature is necessary
in order to determine whether these particular Masto-
gloia taxa have been described, and if any autecology
is known. These species were not common in an anal-
ysis of diatoms from modern surface sediment samples
in Florida Bay (unpublished data). This suggests that
there have been significant changes in the diatom as-
semblages at the Russell Bank site in Florida Bay since
around A.D. 1920.
DIATOMS AND ENVIRONMENTAL CHANGE: HUVANE AND COOPER 15
During zone III (ca. A.D. 1921-1949; 90-56 cm),
the salinity around Russell Bank was likely relatively
low based on the abundance of Nitzschia granulata, a
species found in modern assemblages from sites with
salinities between 8 and 24 ppt (Huvane, in press).
However, the presence of Grammatophora cf. ocean-
ica var. macilenta, a taxon associated with more saline
conditions in Florida Bay (Huvane, in press) increases
during this period. This conflicting evidence may be
explained by salinity fluctuations in the bay. The pres-
ence of taxa indicative of less saline conditions may
reflect average or minimum salinities during that time
period. It is possible that intermittent periods of higher
salinity allowed for the increase in the Grammatopho-
ra species.
RUSSELL BANK AND PASS KEy 1949-1996
The increase in Cocconeis placentula var. euglypta
frustules during zone IV (A.D. 1949-1972; 56-28 cm)
suggests that the water was at least periodically brack-
ish at this time. This taxon is characteristic of fresh-
water (e.g., Patrick and Reimer, 1966), but also occurs
in brackish water (e.g., Cooper, 1995b). In a study of
Florida Bay surface sediment samples, this taxon was
most abundant at the least saline site that had an av-
erage salinity of 7.7 ppt (Huvane, in press).
Mastogloia corsicana, a taxon associated with sites
in Florida Bay that have salinities between 20 and 33
ppt (Huvane, in press), increases after A.D. 1953
(zones IV and V), although there is also a peak around
A.D. 1932 (76-78 cm) during zone III. Synedra sp. A,
another taxon associated with higher salinity sites in
Florida Bay (between 19 and 32 ppt: Huvane, in press)
starts to increase at the upper boundary of zone IV.
The increase in these taxa along with the decrease in
Cocconeis placentula var. euglypta suggests that salin-
ity was increasing during zone V (A.D. 1972-1992;
28—4 cm).
The diatom assemblages in the Pass Key core 37
suggest that salinity increased during the same general
time period, between A.D. 1974 and 1995. This is ev-
idenced by a decrease in Nitzschia granulata, and in-
creases in Mastogloia corsicana, and M. elegans that
have been described in the literature as marine (Hu-
vane, in press; Foged, 1984; Yohn and Gibson, 1983b).
Mastogloia corsicana, M. crucicula, M. ovalis and
Cocconeis placentula var. euglypta were found in sea-
grass samples collected in February 1999 from Florida
Bay (Huvane, unpublished data). These taxa have also
been noted elsewhere in the literature as epiphytic
(e.g., Patrick and Reimer, 1966; Hustedt, 1927-1964;
Navarro, 1982; DeFelice, 1976; Stephens and Gibson,
1979). Higher percentages of these epiphytes in core
RB 19A starting around A.D. 1953, indicate that dur-
Nn
ing zone IV (A.D. 1949-1972; 56-28 cm,) there was
more sub-aquatic vegetation available for colonization
by epiphytic diatoms. In the Pass Key core 37, a sim-
ilar diatom-inferred increase in sub-aquatic vegetation
occurs later, between A.D. 1967 and 1977.
In the Russell Bank core 19A, increases in C:P ra-
tios occur between A.D. 1972-1992 (zone V). In the
Pass Key core, increases occur between A.D. 1979 and
1984 (zone III). These increases can be interpreted in
several ways. The increase in the centric diatom Cy-
clotella cf. litoralis could be an indication of increases
in salinity as other changes in the stratigraphy of both
the Pass Key and the Russell Bank cores suggest. This
species has been described as a marine planktonic
form (Lange and Syversten, 1989). Another explana-
tion is that there are more planktonic algal blooms,
possibly related to increases in nutrients. Such blooms
could increase the number of centric planktonic dia-
toms as well as decrease the number of pennate ben-
thic diatoms by reducing light availability to the bot-
tom. Increased levels of sediment suspension could
also serve to inhibit the growth of benthic diatoms.
Population declines of seagrass in Florida Bay could
also increase this ratio, as many of the pennate diatoms
are epiphytic.
COMPARISONS TO OTHER INDICATORS
The diatom inferences concerning salinity and sea-
grass cover generally coincide with the inferences
made from benthic faunal indicators. In the Russell
Bank core 19B, benthic foraminifera show significant
assemblage changes around 1940, indicating an in-
crease in salinity (Brewster-Wingard ef al., 1997). Di-
atom assemblages also show shifts during this interval,
but the salinity inference is ambiguous. Foraminifera
assemblages indicative of lower salinities are common
between about A.D. 1960—1980 (42-18 cm) (Brew-
ster-Wingard et al., 1997). The increase in the diatom
Cocconeis placentula var. euglypta between A.D. 1953
and 1972 (28-50 cm) also suggests lower salinities.
Foraminifera assemblages indicate a shift to higher sa-
linities between A.D. 1975 and 1980 (24-18 cm)
(Brewster- Wingard et al., 1997). This up-core increase
in salinity is also indicated by changes in diatom as-
semblages starting around A.D. 1972.
In the Pass Key core 37, foraminifera remains in-
dicate increases in salinity between A.D. 1965 and
1978 (64 and 38 cm), and after about A.D. 1989 (14
cm) (Brewster-Wingard ef al., 1998a). Diatom indi-
cators suggest an increase in salinity starting around
A.D. 1975 (44 cm). However, one of the salinity in-
dicators, Mastogloia elegans, does not increase until
after A.D. 1986 (20 cm), suggesting that there may
156 BULLETIN 361
have been shifts in salinity between A.D. 1975 and
1986.
Mollusc data suggests that seagrass is the dominant
substrate in the time period between A.D. 1921—1976
(88—22 cm) in core RB 19B (Brewster-Wingard et al.,
1998b). However, the highest percentages of mollusc
species indicative of seagrass occur after A.D. 1948.
The relative abundances of epiphytic diatoms in core
RB 19A indicate that seagrass is dominant between ca.
A.D. 1953 and 1972 (50-28 cm). Between about A.D.
1921-1948 (88-56 cm), Nitzschia granulata abun-
dances increase, and this species has not been noted
in grass samples from Russell Bank, but it has been
noted in surface sediments from both Russell Bank and
Pass Key (unpublished data). Thus, the diatom data
suggest that there was some bare sediment available
for the colonization of this diatom species during this
interval.
In the Pass Key core, mollusc data suggest that sea-
grass cover increases between A.D. 1965 and 1978.
This agrees with the diatom assemblage data, which
shows high abundances of epiphytic diatoms during
this period.
Minor discrepancies between diatom indicators and
faunal indicators (foraminifera and molluscs) may be
explained by a variety of factors. For example, the
spatial scale that is reflected by each group may differ.
Microscopic diatoms may be deposited in sediments
further away from their original habitat than the larger
mollusc shells.
Other factors to consider are the relative sensitivity
of the indicators to environmental change and the re-
productive rates of the organisms. Diatoms are known
to respond fairly quickly to changes in water quality
(Dixit et al., 1992) and therefore may show an earlier
response to changes in salinity. This appears to occur
at the Russell Bank site where diatom indicators of
increasing salinity appear around A.D. 1972, while
faunal indicators do not appear until A.D. 1975. How-
ever, this pattern was not recognized in the Pass Key
core, and may only reflect variations in the sedimen-
tation rates between cores RB 19A and RB 19B.
IMPACTS ON DECISION MAKING
These data have several implications for manage-
ment decisions and research goals. Fluctuations in both
seagrass cover and salinity suggest that management
plans should be designed to incorporate some degree
of natural variability. Future research should address
questions concerning specific limits of natural vari-
ability such as the natural extremes of salinity and the
periodicity of such extremes. It would also be useful
to know the extent of seagrass cover prior to the major
anthropogenic disturbances in South Florida. The fos-
sil data suggest that seagrass cover has fluctuated in
the past, and these fluctuations are important in our
understanding of seagrass population dynamics. In ad-
dition, the mechanisms that initiate both natural and
anthropogenic-induced variability need further explo-
ration in order to implement remediation.
SUMMARY
The data from Russell Bank core 19A suggest that
there have been fluctuations in salinity and seagrass
cover during the last hundred years. This is consistent
with what has been found in other studies of sediment
cores from Florida Bay. Diatom indicators of salinity
and seagrass coverage show the same general patterns
as other fossil indicators from these cores. In the Rus-
sell Bank core 19A, the earliest diatom assemblages
indicate a period of higher salinity. Between A.D.
1890 and 1920, epiphytic diatoms were not common,
supporting mollusc data that indicates bare sediment
as the dominant substrate during this period. Epiphytic
diatoms increase in abundance after about A.D. 1950
and remain common until about A.D. 1972, which is
consistent with the mollusc data that indicates a de-
cline in epiphytes around A.D. 1970. In the Pass Key
core 37, diatom epiphytes increase at the same time as
mollusc epiphytes (between A.D. 1967-1977).
There are indications of salinity fluctuations after ca.
A.D. 1950 in both cores. The presence of Cocconeis
placentula var. euglypta between ca. A.D. 1953 and
1970 in core RB 19A suggests that salinities were low-
er during this time. Diatom indicators suggest increas-
es in salinity in both cores after ca. A.D. 1970. This
is also consistent with faunal indicators of salinity.
REFERENCES CITED
Battarbee, R.W., Charles, D.F., Dixit, S.S., and Renberg, I.
1999. Diatoms as indicators of surface water acidity. in The Di-
atoms: Applications for the environmental and earth sci-
ences. E.R Stoermer and J.P. Smol, eds., Cambridge Uni-
versity Press, pp. 85-127.
Brewster-Wingard, G.L., Ishman, S.E., Willard, D.A., Edwards,
L.E., and Holmes, C.W.
1997. Preliminary paleontologic report on cores 19A and 19B,
from Russell Bank, Everglades National Park, Florida Bay:
U.S. Geological Survey Open-file Report 97-460, 29 pp.
Brewster-Wingard, G.L., Ishman, S.E., and Holmes, C.W.
1998a. Environmental impacts on the Southern Florida coastal
waters: A history of change in Florida Bay. Journal of
Coastal Research, vol. 26, pp. 162-172.
DIATOMS AND ENVIRONMENTAL CHANGE: HUVANE AND COOPER 157
Brewster-Wingard, G.L., Ishman, S.E., Waibel, N.J., Willard,
D.A., Edwards, L.E., and Holmes, C.W.
1998b. Preliminary paleontologic report on core 37 from Pass
Key, Everglades National Park, Florida Bay. U.S. Geo-
logical Survey Open-file Report 98-122, pp. 162-172.
Cooper S.R.
1995a. Chesapeake Bay watershed historical land use: impact on
water quality and diatom communities. Ecological Appli-
cations vol. 5 (3), pp. 703-723.
1995b. Diatoms in sediment cores from the mesohaline Chesa-
peake Bay, U. S. A. Diatom Research, vol. 10 (1), pp. 39—
89.
1999. Estuarine paleoenvironmental reconstructions using dia-
toms. in The Diatoms: Applications for the environmental
and earth sciences. E.E Stoermer and J.P. Smol, eds.,
Cambridge University Press, pp. 352-373.
Cooper, S.R., Huvane, J.K., Vaithiyanathan, P., and Richardson,
C.J.
1999. Calibration of diatoms along a nutrient gradient in Florida
Everglades Water Conservation Area-2A, U.S. A. Journal
of Paleolimnology, vol. 22, pp. 413-437.
Davis, G.E., and Halvorson, W.L.
1996. Long-term research in National Parks: From beliefs to
knowledge. in Science and ecosystem management in the
national parks. G.E. Davis and W.L. Halvorson, eds., Uni-
versity of Arizona Press, Tucson. 362 pp.
Davis, S. M., and Ogden, J.C.
1994. Everglades: the ecosystem and its restoration. Delray
Beach, FL: St. Lucie Press, 826 pp.
DeFelice, D.
1976. Model studies of epiphytic and epipelic diatoms of upper
Florida Bay and associated sounds. Unpublished M.S.
thesis, Duke University, Durham, N.C., 193 pp.
Dixit, S.S., Cumming, B.F., Kingston, J.C., Smol, J.P., Birks,
H.J.B., Uutala, A.J., Charles, D.F., and Camburn, K.E.
1993. Diatom assemblages from Adirondack lakes (N.Y. USA)
and the development of inference models for retrospective
environmental assessment. Journal of Paleolimnology,
vol. 8, pp. 27—47.
Dixit, S.S., Smol, J.P., Kingston, J.C., and Charles, D.F.
1992. Diatoms: powerful indicators of environmental change.
Environmental Science and Technology, vol. 26 (1), pp.
22-33.
Foged, N.
1984. Freshwater and littoral diatoms from Cuba. Vol. 5, Bib-
hotheca Diatomologica, J. Cramer, Berlin, 243 pp.
1986. Diatoms in the Volo Bay, Greece. Vol. 12, Bibliotheca
Diatomologica, J. Cramer, Berlin, 153 pp.
Fritz, S.C., Juggins, S., Battarbee, R.W., and Engstrom, D.R.
1991. Reconstruction of past changes in salinity and climate us-
ing a diatom-based transfer function. Nature, vol. 352, pp.
706-708.
Fritz, S.C., Kingston, J.C., and Engstrom, D.R.
1993. Quantitative trophic reconstructions from sedimentary di-
atom assemblages: A cautionary tale. Freshwater Biology,
vol. 30, pp. 1—23.
Funkhauser, J.W., and Evitt, W.R.
1959. Preparation techniques for acid insoluble microfossils.
Micropaleontology, vol. 5 (3), pp. 369-375.
Grimm, E.C.
1987. CONISS: A FORTRAN 77 program for stratigrapically
constrained cluster analysis by the method of incremental
sum of squares. Computers and Geoscience, vol. 13, pp.
13-35.
Hall, R.I., and Smol, J.P.
1999. Diatoms as indicators of lake eutrophication. in The Di-
atoms: Applications for the environmental and earth sci-
ences. E.E Stoermer and J.P. Smol, eds., Cambridge Uni-
versity Press, pp. 128-168.
Holling, C.S., Gunderson, L.H., and Walters, C.J.
1994. The structure and dynamics of the Everglades System:
Guidelines for ecosystem restoration. in Everglades: the
ecosystem and its restoration. S.M. Davis and J.C. Ogden,
eds., Delray Beach, FL: St. Lucie Press, pp. 741-756.
Hustedt, F.
1927-1964. Die Kieselalgen Deutschlands, Osterreichs und der
Schweiz unter Beriicksichtigung der iibrigen Landen Eu-
ropas sowie der angrenzenden Meeresgebiete. 3 vols, Dr.
L. Rabenhorst’s Kryptogamen-flora von Deutschland, Os-
terreich und der Schweiz. Leipzig: Akademische Verlags-
gesellschaft, 920 pp.
1955. Marine littoral diatoms of Beaufort, North Carolina. Dur-
ham, N.C., Duke University Press, 67 pp.
Huvane, J.K.
2002. Modern diatom distributions in Florida Bay: A prelimi-
nary analysis. in Everglades, Florida Bay and coral reefs
of the Florida Keys: an ecosystem source book. J.W. Por-
ter, and K.G. Porter, eds., CRC Press, Boca Raton, Flor
ida, pp. 471—487 (in press).
Ishman, S.E., Cronin, T.M., Brewster-Wingard, G.L., Willard,
D.A., and Verado, D.J.
1998. A record of ecosystem change, Manatee Bay, Barnes
Sound, Florida. Journal of Coastal Research, vol. 26, pp.
125-138.
Juggins, S.
1992. Diatoms in the Thames estuary, England: ecology, pa-
laeoecology, and salinity transfer function. H. Lange-Ber-
talot, ed., Vol. 25, Bibliotheca Diatomologica. Berlin:
Gerbruder Borntraeger, 316 pp.
Krammer, K., and Lange-Bertalot, H.
1986. Bacillariophyceae 2/1. Naviculaceae. in StiBwasserflora
von Mitteleuropa, H. Ettl, J. Gerloff, H. Heynig, D. Mol-
lenhauer, eds., G. Fischer, Stuttgart, 876 pp.
1988. Bacillariophyceae 2/2. Bacillariaceae, Epithemiaceae, Su-
rirellaceae. in StBwasserflora von Mitteleuropa, H. Ettl,
J. Gerloff, H. Heynig, D. Mollenhauer eds., G. Fischer,
Stuttgart, 596 pp.
1991a. Bacillariophyceae 2/3. Centrales, Fragilariaceae, Eunoti-
aceae. in SiiSwasserflora von Mitteleuropa. H. Ettl, J. Ger-
loff, H. Heynig, D. Mollenhauer eds., G. Fischer, Stutt-
gart, 393 pp.
1991b. Bacillariophyceae 2/4. Achnanthaceae. in SiiBwasserflora
von Mitteleuropa. H. Ettl, J. Gerloff, H. Heynig, D. Mol-
lenhauer eds., G. Fischer, Stuttgart.
Lange, C.B., and Syvertsen, E.E.
1989. Cyclotella litoralis sp. nov. (Bacillariophyceae) and its re-
lationship to C. striata and C. stylorum. Nova Hedwigia,
vol. 48, pp. 341-356.
MacMahon, J.A.
1998. Empirical and theoretical ecology as a basis for restora-
tion: An ecological success story. in Success, limitations,
and frontiers in ecosystem science. M.L. Pace and P.M.
Groffman, eds., New York, Springer Verlag, pp. 220-246.
McPherson, B.F., and Halley, R.
1996. The South Florida environment—a region under stress.
U.S. Geological Survey Circular 1134, 61 pp.
Navarro, J. N.
1982a. A survey of the marine diatoms of Puerto Rico. V: Sub-
158 BULLETIN 361
order Raphidineae: Families Achnanthaceae and Navicu-
laceae (excluding Navicula and Mastogloia). Botanica
Marina, vol. 25, pp. 321-338.
1982b. Marine diatoms associated with mangrove prop roots in
the Indian River, Florida, U.S. A. Vol. 61, Bibliotheca
Phycologica. Hirschberg, J. Cramer, 154 pp.
Patrick, R., and Reimer, C.W.
1966. The diatoms of the United States exclusive of Alaska and
Hawaii. Vol. 1. Philadelphia: The Academy of the Natural
Sciences of Philadephia, 688 pp.
Peragallo, H., and Peragallo, M.
1897-1908. Diatomées marines de France et de districts mari-
times voisins. M. J. Tempeére, ed., Arez-Sur-Loing, 491
Pp-
Pyle, L., Cooper, S.R., and Huvane, J.K.
1998. Diatom paleoecology Pass Key core 37, Everglades Na-
tional Park, Florida Bay. U. S. Geological Survey Open-
file Report 98-522, 37 pp.
Robbins, J.A., Holmes, C.W., Halley, R.B., Bothner, M., Shinn,
E.A., Graney, J., Keller, G., Tenbrinck, M., and Rud-
nick, D.
2001. Time constraints characterizing predeposition integration
of 137Ce and Pb fluxes to sediments in Florida Bay. Geo-
chemica et Cosmochemica Acta (in press).
Stephens, F.C., and Gibson, R.A.
1979. Ultrastructural studies on some Mastogloia (Bacillario-
phyceae) species belonging to the group Ellipticae. Bo-
tanica Marina, vol. 22, pp. 499-509.
1980a. Ultrastructural studies of some Mastogloia (Bacillariophy-
ceae) species belonging to the groups Undulatae, Apicu-
latae, Lanceolatae, and Paradoxae. Phycologia, vol. 19
(2), pp. 143-152.
1980b. Ultrastructural studies on some Mastogloia species of the
group /naequales (Bacillariphyceae). Journal of Phycol-
ogy, vol. 16, pp. 354-363.
Sweets, P.R., Bienart, R.W., Crisman, T.L., and Binford, M.W.
1990. Paleoecological investigations of recent lake acidification
in northern Florida. Journal of Paleolimnology, vol. 4, pp.
103-137.
ter Braak, C.J.F.
1995. Ordination. in Data analysis in community and landscape
ecology. R.H.G. Jongman, C.J.E ter Braak and O.ER. Van
Tongeren, eds., Cambridge University Press, pp. 91-169.
Wetzel, R.G.
1983. Limnology. CBS College Publishing, 763 pp.
Wilson, S.E., Cumming, B.F., and Smol, J.P.
1994, Diatom-salinity relationships in 111 lakes from the Inte-
rior Plateau of British Columbia, Canada: the develop-
ment of diatom-based models for paleosalinity reconstruc-
tions. Journal of Paleolimnology, vol. 12, pp. 197-221.
Yohn, T.A., and Gibson, R.A.
1982a. Marine diatoms of the Bahamas: I. Mastogloia Thw. ex
Wm. Sm. Species of the groups Decussatae and Ellipticae.
Botanica Marina, vol. 25, pp. 41-53.
1982b. Marine diatoms of the Bahamas III: Mastogloia Thw. ex
Wm. Sm. species of the groups Inaequales, Lanceolatae,
Sulcatae, and Undulatae. Botanica Marina, vol. 25, pp.
277-288.
CHARTERS
HISTORICAL TRENDS IN EPIPHYTAL OSTRACODES FROM FLORIDA BAY: IMPLICATIONS FOR
SEAGRASS AND MACRO-BENTHIC ALGAL VARIABILITY
T. M. CRonin!, C. W. HoLMeEs?, G. L. BREWSTER-WINGARD!, S. E. ISHMAN?, H. J. DowsettT',
D. KEYSER*, AND N. WAIBEL!
'US Geological Survey, Reston, Virginia 20192
°US Geological Survey, St. Petersburg, Florida 33701
sDepartment of Geology, University of Southern Illinois, Carbondale, Illinois 62901
4Zoologisches Institut und Museum, Hamburg, Germany
ABSTRACT
Living and fossil epiphytal ostracodes from central and eastern Florida Bay were investigated to determine historical trends in
seagrass and macro-benthic algal habitats during the past century. Living assemblages collected in February and July, 1998 from
15 sites throughout Florida Bay revealed that Loxoconcha matagordensts and Malzella floridana are the dominant species living
on Thalassia and that Xestoleberis spp. is the most abundant ostracode group living on Syringodium and marine algae such as
Chondria. Peratocytheridea setipunctata is a species that is common on muddy substrates and on Halodule.
Temporal trends in epiphytal ostracode abundance were reconstructed from isotopically-dated sediment cores from Whipray
mudbank, Russell Bank, Bob Allen mudbank, Pass and Park Keys, Taylor Creek (near Littlke Madeira Bay) and Manatee Bay.
The results show that there have been frequent changes in the relative frequencies of L. matagordensis, M. floridana, and
Xestoleberis over the past century. Prior to the mid-20" century, seagrass- and algal-dwelling ostracode species were relatively
rare at our sites in central and eastern Florida Bay. Ostracode assemblages living between about 1900 and 1940 were characterized
by moderate to large proportions (10O—>60%) of Peratocytheridea setipunctata when Thalassia- and macro-benthic algal-dwelling
species were less common than during the later half of the 20" century. Beginning about 1930, and continuing until 1950, P.
setipunctata populations experienced significant declines while L. matagordensis and Xestoleberis increased progressively from
0-10% to >25—40%, depending on the site. This faunal shift suggests that there has been a greater abundance and/or density of
subaquatic vegetation in central Florida Bay over the past 50 years than in the prior half century. Since 1950, several sites in
central and eastern Florida Bay have experienced large swings in the proportion of seagrass and macro-benthic algal-dwelling
species suggesting subaquatic vegetation has been extremely dynamic spatially and temporally over decadal timescales. Some of
these oscillations, such as the decline in L. matagordensis, Xestoleberis, and M. floridana during the 1970s and early 80s may
represent large-scale dieoffs. The possible influence of climatological factors on salinity and epiphytal ostracodes is also discussed.
INTRODUCTION
Growing efforts to restore polluted and disturbed
coastal and estuarine ecosystems to a more natural
state rely increasingly on retrospective paleoecological
analyses using isotopically-dated sediment cores to es-
tablish ecosystem trends and to better understand eco-
system functioning (e.g., Brush 1986; Vallette-Silver
1993). Paleoecological studies are especially important
because few monitoring programs have been in place
for a long enough period to sufficiently establish the
natural range of habitat variability both prior to, and
subsequent to human influence. Moreover, in addition
to anthropogenic disturbance of ecosystems, extrinsic
factors such as climatic variability are also critical as-
pects to estuarine functioning (Peterson ef al., 1995;
Cronin et al., 2000).
Florida Bay, a large (2,200 km/?), shallow (<3 m),
tropical to subtropical lagoon is one such ecosystem
believed to have been affected by human activities over
the past century. For example, Smith et al. (1989) sug-
gested that canal construction from 1912 to 1931 de-
creased freshwater flow into Florida Bay, especially
through Shark River Slough, and led to hypersalinity.
Declines in seagrass populations in parts of Florida Bay
since 1987 have been attributed to chronic hypoxia re-
sulting from human alteration of freshwater inflow from
the Everglades, and to pathogenic slime molds (Zieman
et al., 1989; Robblee er al., 1991). McIvor er al. (1994)
pointed out that, in addition to its impact on seagrasses,
hypersalinity of Florida Bay also may have contributed
to recent increases in mangrove mortality. They com-
piled measurements of salinity obtained over the past
40 years showing that, since at least the late 1950’s,
seasonal hypersalinity has characterized some regions.
Salinity values reached as high as 50 to 70 parts per
thousand (ppt) in eastern and central parts of the bay.
Light and Dineen (1994, p. 48) expressed the widely
held view that human alteration of Everglades hydrol-
ogy is unprecedented for the late Holocene as follows:
160 BULLETIN 361
“The first reclamation efforts in the Everglades
began in 1881. Just over 100 years later, Governor
Graham formally embraced a system-wide restora-
tion initiative entitled “Save Our Everglades.’ More
has happened to the Everglades between these two
benchmarks than in the preceding 50 centuries.”
Given the temporally and spatially incomplete his-
torical monitoring record of Florida Bay, several pa-
leoecological studies have been conducted to recon-
struct various ecosystem parameters. Smith ef al.
(1989) concluded, on the basis of analyses of autoflu-
orescence of a Solenastrea coral skeleton from Peter-
son Key in the Atlantic transition zone of Florida Bay
that canal construction severely diminished fresh water
inflow into Florida Bay, especially through Shark Riv-
er Slough, reaching minimal flow around 1931. They
argued that the hydrological impact of canal construc-
tion was severe enough to disrupt the natural 4- to 6-
year periodicity of freshwater inflow resulting in un-
precedented episodes of hypersalinity. Swart ef al.
(1996a) attributed carbon and oxygen isotopic vari-
ability in a Solenastrea colony from the Lignumvitae
Basin, also in the Atlantic transition zone of Florida
Bay, to construction of the Florida East Coast Railway
between 1905 and 1912. They also suggested that fre-
quent hurricanes between 1912 and 1948 caused great-
er exchange of water between the bay and reef areas,
but that after 1948, decreased hurricane activity led to
the retention of organic material in the bay and the
initial stages of eutrophication. Brewster-Wingard ef
al. (1998) analyzed mollusks and foraminifers from
four of the sediment cores examined in this study (Bob
Allen, Russell, Taylor, and Pass) to determine the long-
term trends in salinity and benthic habitats. Their study
shows that subtle shifts in the faunal distributions oc-
cur around 1910, but after 1940, the pattern departs
substantially from the pre-1900 pattern. The amplitude
of shifts in salinity indicators increased from 15%—
20% about the pre-1900 mean to 40%—60% about the
post-1940 mean. In general, there was an increase in
epiphytal molluscan species throughout this century;
which Brewster-Wingard ef al. suggested might be due
to changes in water-management practices. Ishman er
al. (1998) found on the basis of paleoecological data
that environments of Manatee Bay had changed sig-
nificantly over the past century.
Despite the growing amount of paleoecological
data, there is still a large degree of uncertainty about
the ecosystem history of Florida Bay and the factors
that cause interannual and interdecadal variability in
Florida Bay environments. The present study is de-
signed to address the scale, timing and possible causes
of subaquatic vegetation (seagrasses and macro-ben-
thic algae) variability in Florida Bay using quantitative
analyses of ostracodes from sediment cores from cen-
tral and eastern Florida Bay, and from Manatee Bay
just outside the northeast margin of Florida Bay (Text-
fig. 1). This study has two parts. First, we conducted
a study of the dominant ostracode taxa living on mod-
ern seagrass (especially Thalassia) and marine algae
in Florida Bay. Second, we examined trends in sea-
grass and algal-dwelling species preserved in sediment
cores.
ACKNOWLEDGMENTS
We are especially grateful to Robert Halley and Eu-
gene Shinn for obtaining cores and helping with the
stratigraphy of mudbanks and to the South Florida Wa-
ter Management District (SFWMD) for support and
logistical aid throughout this study. We are grateful to
John Repetski and Sarah Gerould for critical reviews.
We also thank Dave Rudnick of SFWMD who has
championed paleoecological research in Florida Bay.
Mike Bothner and John Robbins kindly provided ac-
cess to their age model for the Bob Allen and Russell
cores. Jeffrey Stone and Kristi Alger, US Geological
Survey, and Thomas Scott and Harley Means, Florida
Geological Survey, assisted in collecting the modern
samples. Jeffrey Stone developed and maintains the
ecosystem history database. Sara Schwede, Marcy Ma-
rot, Jill D’Ambrosio, lan Graham, and Jean Self-Trail
helped with the laboratory processing of the ostra-
codes. The study is a contribution to the USGS Eco-
system Program.
MATERIAL AND METHODS
MODERN ECOLOGICAL SAMPLES
Epiphytal ostracodes were studied from samples of
living seagrasses and macro-benthic algae collected
from 15 stations in February and July 1998 in order
to obtain quantitative data on species that live on sub-
aquatic vegetation during winter and summer months
(Text-fig. 1). These samples were divided into three
substrate groupings, those with mainly Thalassia,
those with mainly macro-benthic algae, and those with
primarily Halodule grasses. The macro-benthic algae
included mostly Laurencia, Chondria, and Polysi-
phonia; the seagrasses were Thalassia, Halodule and
Syringodium. Two sites with coarse shelly substrate
were also examined.
After washing vegetation in a 63 4m sieve with tap
water, the first 100 individual ostracode specimens
were picked with fine brush under a binocular micro-
scope from the size fraction >150 to 850 pm. This
size fraction contains the adults and several juvenile
instars of all common podocopid ostracodes living in
Florida Bay.
SEAGRASS OSTRACODES: CRONIN ET AL. 161
m Vegetation Sites [[-—~juggg “a i
0 2 4 6
@ Sediment Cores
Text-figure 1.—Map showing location of modern ostracode samples (squares labeled FB-) and sediment cores at Whipray, Russell, Bob
Allen, Park, Pass, Manatee Bay (MB-1), and Taylor Slough.
SEDIMENT CORE SITES
Seven sediment cores were collected along a south-
west/northeast transect running from Whipray Key in
central Florida Bay to Manatee Bay on the margin of
Barnes Sound (Text-fig. 1, Table 1). Cores were taken
by hand using a large-diameter hand-driven piston cor-
er ranging in length from 74 cm (Pass Key) to 156 cm
(Bob Allen Key). Five cores came from mudbanks in
<1 m of water from the central part of Florida Bay
(Text-fig. 1): one each from Whipray mudbank (WR-
25B), Bob Allen mudbank (BA-6-A), Russell Bank
(RB-19-B). Pass Key (PA-37) and Park Key (PK 23-
A). One core came from the transition zone in Little
Madeira Bay near the mouth of Taylor Creek (T-24)
and another core was taken from Manatee Bay (MB-
1) located on the western edge of Barnes Sound, near
the northeast part of Florida Bay. Information on each
core site is given in Table 1, in preliminary reports
(Wingard et al., 1995; Brewster-Wingard et al., 1997;
Ishman, 1997), and at the website http://geolo-
gy.er.usgs/gmapeast/fla/home.html.)
Whipray Keys are located in central Florida Bay;
the core was taken on a mudflat on the eastern side of
Whipray basin. The core has a top mixed layer to 12
cm representing the past ~ 15—20 years; below this, the
710Pb-determined sediment accumulation rate is 0.43
cm/year until reaching equilibrium at 40 cm. Below
this level, the material is too old for *!°Pb dating. The
Table 1.—Core site information for South Florida paleoecology of ostracodes.*
Sediment
accumulation Sampling
Site Latitude Longitude Core length Water depth rate # interval
Russell Key (19-B) 25°03.83'N 80°37.49'W 140 cm 0.51 m 1.22 2 cm
Bob Allen Key (6-A) 25°01.39'N 80°39.41'W 156 cm 0.6 m 0.75-1.04** 2cm
Manatee Bay (MB-1) 25°15.69'N 80°24.06'W 120 cm ~1.0 m ~1.0 2 cm
Taylor Creek (T-24) 25°11.4'N 80°38.4'W. 86 cm ~0.5 m NA 2cm
Park Key (PKK-23A) 25°10.45'N 80°57.46'W 86 cm ~lm 0.78 2cm
Pass Key (PK-37) 25°14.78'N 80°40.82'W 74 cm ~lm 2.06 ~3 cm
Whipray (WR-25B) 25°07.12'N 80°73.85'W 80 cm ~1.25 m 0.43 5 cm
* Sources: Bob Allen and Russell cores: Wingard er al. (1995), Brewster-Wingard er al. (1997); Manatee Bay cores: Ishman et al. (1996, in
press), respectively. # centimeters per year. ** Higher rate was used in Figures 8—11.
BULLETIN 361
162
pnut yyos ur [jays
DIIUAANDT ‘a/NPO|VE] “‘DISSV]DY TL,
piuoydissjog
pisspjoyy ‘vioydojng ‘piupuoyy
DISSDIDYL, ‘VIUOYdIsSAJog ‘Dla4puoyy
anpo]|PH
piuoydisdjog ‘PIssv]OY
UMIPOSULIKS,
DIIUAANDT ‘ajNpo|PH] ‘vissvjpyy “viuoydissjog “viapuoyyD
DISSDIDY], ‘VI4puoyy
DIAPUOYD ‘DISSD]DY TL,
anpopHY ‘vissvjpyy ‘vioydoing ‘piuoydissjog
pisspjpyy, ‘vaoydoiwg ‘piauasnvy
psoydolvg ‘DIQUAINDT ‘DISSD]VYL,
DIIUIANDT ‘vISsDjDY], ‘ajnpojpH ‘piddny
dIQRURA ‘DISSDIDY |.
aynpo]DH
DISSDIDYL.
DISSDIDY], ‘DIQUaNDT ‘PiuoydisAjOg ‘pl4puoyy
UNSSDSADS
piuoydiskjog ‘Mi4puoyy ‘vIssv]vY
DISSDIDYL, ‘a/NPo]D
DISSDIDY], ‘DI4puoyD
piuoydisdjog ‘PIssp]DY],
DIIUAANDT ‘DIUOYAdISA]Og ‘DISSD]DY I.
DIIUIANDT ‘Aa]NPOIVE] ‘DISSV]DY TL.
DISSDIDY], ‘DIQUANDT
ajnpo]PH
DIQUIANDT ‘UMNIPOSULAAG ‘aj]npo
-[DE] ‘DISSDIDY] *{,DIQUaANDT/VLUpUuoYy) ‘j,viupuoyyD/{,,viuoydisAjog
A[NPO]VY ‘PIQUAANDT/VIAPUOY) ‘DIDUIANDT] “DISSD]DY TL,
pluoydisajJod ‘PISSPIDY
DIDSSDAIUL DpaMo]DP]T ‘WUiniposulsKg
anpo]MH ‘syplomagd ‘piuoydiséjOg ‘DISsv]VY I,
paoydolyg ‘vIIUAANDT] ‘VISSDIDY
a/npolpH
DISSD]DY [ ‘a/NPO|PH
{Dlupuoy)/{,puoydisKjog ‘VISsP]DY
uINSsDSADG ‘vAoydolwg ‘vISsv]DY], ‘DlADjNGvIaIy
‘{Dlupuoy,) ‘,,pluoydisdjog *{plouasnvT/piipuoyy
DISSDIDY I,
psoydolpg ‘,piuoydisasjog *{plouasnvy] *j,p14puoy)D
DIIUAANDT ‘a]NPOI|DE] ‘DISSVIDY T.
anpoppY ‘j,piuoydisjog “‘DIguasNVT ‘DISSD]DY [,
uonRj}asaAQ,
86
901
OOl
88
L6
8Ol
+8
[RI0OL
OMNON OANA
OnmrNDAONMl OC
Xe} Yemnae) ata
(oa)
6S
Iv
al
oI
S
s1laqa]
-O1SAX
Lv
n-
ce
v
I
DIDI
-ound
-1as
d
rat > (OV IQ (00 rt NON OS SO Saco
pupp
-MoUe
ppyjaz
TPW
rn
ononaa
SISUAPAOS ,.dV.IS
-DIDU “T
(O) GOOG) (OAS) <0) ee (OI (2) 1) (OO) eee (GI ee eae, (ONG) ee <6 (9) 17)
SE) ONSNONS) =e 1S) 1S) =e
OOtO<
-qng
ve 6 8C qeus Ica 86-994
VC 68C I 1cd4 86-994
661 LI g 0cdH 86-494
6 6l AT v Oca 86-994
6 ol | 7e)! € Ocdd 86-494
661 ULI c Ocdd 86-994
66! VL I Oca 86-994
61C L6T Cc 81d 86-494
6ST Lie c L194 86-994
6ST Lali c L194 86-994
6ST fO\Wte I L144 86-994
£97 1V0e € 91d 86-994
£97 0c C 91d 86-994
£97 T0e I 91d 86-994
8° EC 697 I rida 86-994
O1T LC S clad 86-994
OAS LOT v cid 86-994
DIT L0t (e cIdh 86-994
OIC LOT Cc clad 86-994
ONG L0c I cldd 86-994
rol £61 S 11d 86-994
rol ol € 11d 86-494
rol £6! c 1144 86-994
rol £61 I [144 86-994
It 8S S 1d 86-994
1c 8ST 4 1d4 86-994
Ic SSI I 144 86-994
ape ySe Cc 9cdH 86-1Nf
ole 8'9¢ v ec 86-10f
VOL 66 I ccd 86-1Of
(G3 T8T € voc 86-18f
£67 GLE (é 8144 86-18
P8T EADIE I L184 86-18
b'6T L'8¢e Cc 9144 86-1Nf
€€ Iv I bid 86-8
OVE Le Cc VCIdd 86-1
€€ Ss C VIIa 86-[8f
60£ 6 ET c V8sda4 86-1
I {ts OSE I V9dd 86-18
VEE Ole (é VSdi 86-[f
67CE (GAS I Vrdd 86-1Nf
SINE 69 € Vids 86-1
dwiay uyeg saquinu ql qy aduirs quo
‘sgourpunge apoor-nsO uoNnrjiasea Avg ePHOL{—'Z AqRL
seti-
Je
ella
Mal-
Table 2.—Continued.
SEAGRASS OSTRACODES: CRONIN ET AL. 163
slower sediment accumulation rate at Whipray Keys
means that a 2-cm sample represents on average about
j four years of sediment.
= Russell Bank is a few hundred meters across and
our uly & about | km long. Wanless and Tagett (1989) studied
a3 Se Wao = the stratigraphy of Russell Bank and concluded that
es SiS Gare during its construction, it had accreted in a southward
23 ey age apy = direction. The Russell Bank core (RB-19-B) was taken
(oye a a iy >
Ee et & Sis 853 w from the top of Russell Key mudbank. The Bob Allen
= eS Sys Sey = core (BA-6-A) was obtained from the flank of another
4 = 38 Tf 2a so 0 4 Sj :
30 se Zé Si §6&s8 8 accretional mudbank located about 4 km to the south-
A SG Soe ee. & west of Russell Key. Both the Russell and Bob Allen
> 3 Gee aa & y
Seon eS Sis ae & cores have been the subject of intensive isotopic dat-
3s 2s E= 5 8 Sic J Ip)
ESS SSeo5R SES 8 ing, and the age model based on 7!°Pb and '*’Cs of
Sa, Ss 8 5 83 S 0 Robbins er al. (2000) is used here.
Sh EPs SSH ena as Park Key core was taken on the western side of the
SseubvF Sessegssk 8 bank, which, based on aerial photographs, is the ac-
OBS 5 a, 8 Ss] vcs p grap
SMISTSSESSSSsus creting side of the bank. The core was taken in 10 cm
SES IS ISianse Wheelie of water on top of the bank.
The Whipray, Russell, Bob Allen, Park and Pass
a Keys sites are strategically situated in the central and
S12°S 2S & *A Ba eastern parts of Florida Bay where periods of sum-
Fe mertime hypersalinity have occurred, especially during
ne years with low freshwater discharge (Robblee ef al.,
2S ES OS eS 1991; Fourqurean et al., 1993).
aie The Manatee Bay core (MB-1) site is located to the
Los By Aico AONE rae, northeast of Florida Bay (the “inner zone” of Wanless
zs © and Tagett (1989) near the outflow of the C-111 canal.
Nn o
Manatee Bay is connected to the Biscayne Bay hydro-
EEloon CAD A OD OM logic system. The C-111 canal was completed in the
S815 t = =6©6 : ;
Su 1960s and its outflow has a large impact on Manatee
g P
42 Bay salinity (Light and Dineen, 1994; Ishman et al.,
S| eR ESO ays NAA ee 1998). Core MB-1 penetrates about 106 cm. The upper
~j S er 85-90 cm are carbonate muds; the lowermost 36 cm
* a are a peat and peaty-shelly marl containing limnic and
1 9 & 5 . .
2 FIOOO OOS = aa BO] oligohaline microfaunas (Ishman et al., 1998).
+ Z The Taylor Slough core (T-24) is located in Little
a : Madeira Bay, a region sensitive to changes in outflow
ret || G2 a SX eh 2 n : ;
S|Sah Raa A AA ARE from Taylor Slough, which drains the southeastern part
5 of the Everglades. Like the Manatee Bay core, T-24
3 does not have a isotopic age model, although the fau-
FEinnvo orm oO Ne} Nowe) fo) ikine | al ak
eiliorol Ue CIS NS Iie moat M nal patterns suggest that salinity-related changes oc-
i 0 curred in this region over what is believed to be the
5 5p past century or so of sedimentation based on the first
we) a 9 7 a
E I As stratigraphic appearence of Casuarina (Ishman et dl.,
=) || sat Sb At = nN (coma) Be Sofiilll
PB Be 1998).
5 a The seven cores cover a wide range of salinities that
Q se vary over monthly and interannual timescales. For ex-
Llano © 2 ample, Halley et al. (1995) showed that in June 1995,
araaag Qa Yel al no ‘o a é : i
SA eneeree ec pet bale pee Mine sed OPEN Coulee during a period of relatively low fresh water flow and
oi 2 rainfall, salinities were high ranging from ~25 ppt at
Ss 2 . .
sleexw eer 2 2m we : Taylor and Manatee Bay to >36 ppt at Whipray Basin.
E 4 aa 2 2 Sarita 2 Bale Conversely, in August 1995, during an apparently wet
21 ee Be ee ee lx period with higher fresh water flows, salinity ranged
164 BULLETIN 361
from 4—5 ppt at Taylor Creek to 30 ppt at Whipray.
Even larger historical swings in salinity described by
Mclvor et al. (1994) are likely to be a factor in inter-
preting paleoecological records.
Florida Bay sediments consist mostly of unconsol-
idated muds containing abundant macro- and micro-
fossils. Following lithological description and X-ray
radiography, sediments were sampled every 2 centi-
meters for *!°Pb dating and micropaleontological anal-
ysis. Approximately 20—40 g of dry sediment was
washed through 63 pm sieves and ostracodes in the
size fraction >150 and 850 4m were picked with a
fine brush. This size fraction yields adults and several
juvenile molt stages of essentially all major species
living in Florida Bay; our focus here is on trends in
epiphytal species.
OSTRACODE SPECIES CENSUS DATA
The counts of key ostracode species from the mod-
ern vegetation and sediment samples are given in Ta-
ble 2 with the major seagrasses and/or macro-benthic
algae genera found at each site. Appendix | contains
the species counts for the seven sediment cores.
Nore ABOUT OSTRACODE TAPHONOMY
There are four lines of taphonomic evidence sug-
gesting there is minimal transport of ostracode valves
at the studied sites. First, the size of adult ostracode
valves and carapaces (a carapace consists of two artic-
ulated valves held together after death by a tooth-and
socket hingement) are too large to have been trans-
ported large distances in Florida Bay. Halley er al.
(1997) showed that most mudbank sediment >~63
kLm is autochthonous within a mudbank. Depending on
the species, adult ostracode valves and carapaces are
>180 «m in length and thus, most adults are in the
size fraction of sediment that is not transported. Sec-
ond, juvenile valves representing | to 4 molt stages
prior to the adult molt are commonly associated with
conspecific adult valves and carapaces. Although time
constraints precluded detailed analyses of species’
population structure, these qualitative observations in-
dicate most species are represented by juvenile and
adult individuals. Third, articulated ostracodes cara-
paces are common in mudbank sediments and such
preservation would not be expected had the ostracodes
been transported by currents which tend to disarticu-
late the carapace. Fourth, fossil assemblages consist of
ecologically compatible species; there is no faunal ev-
idence for transport of allochthonous species from oth-
er parts of Florida Bay or adjacent habitats. In sum,
ostracode taphonomy indicates that the majority of os-
tracodes examined in this study from the >63 ym size
fraction represent individuals that lived, died and were
buried at the mudbank.
STATISTICAL METHODS
We used the changing proportions (also called rel-
ative abundances) of ecologically significant indicator
species to infer past environmental changes. In paleo-
ecological research, the number of individuals of a par-
ticular species n, is often expressed as
S
>a = 2
i=]
where S is all species in the assemblage and n is the
total number of individuals in the sample (see Buzas,
1990). The proportion of a species is expressed as p
= n/n, and p has certain confidence limits that depend
on both the value of p for that species and on the value
n. Because n, the abundance of ostracodes, varies with-
in each core and among cores, and because picking
microfossils is so time intensive, it is important to con-
sider what values of n ensure statistical reliability of
species’ proportions and still yield meaningful tem-
poral patterns. Ideally, the confidence limits for any
given p value should not exceed the observed down-
core temporal changes in p if trends and oscillations
in p are to be considered meaningful.
Confidence limits on species’ proportions are often
expressed in paleoecology using the binomial distri-
bution (e.g., Patterson and Fishbein, 1989; Buzas,
1990) as follows:
o = (pq/n)!”
where o = the binomial standard error, p = n/n, or
the proportion of the i species, q = 1 — p. The con-
fidence limits d are expressed as:
d = =+to
where t corresponds to the chosen level of confidence
(i.e., t = 1.96 for 0.95 confidence).
Table 3 illustrates the different confidence limits d,
for three values of n, 100, 200, 300 and 1000 individ-
uals. We see that for a species that comprises 10% of
an assemblage (p = 0.1), d decreases progressively
with increasing sample size (n) from 5.9% (100 indi-
viduals), 4.2% (200 individuals) and 3.4% (300 indi-
viduals). For a dominant species comprising 50% of
an assemblage, the comparable values are higher,
9.8%, 6.9% and 5.7% respectively. The values of d are
essentially meaningless for very rare species (p < 1%,
Buzas, 1990).
In general, the results presented below for Florida
Bay suggest that the long-term variability in indicator
species proportions (p) often exceeds the confidence
limits (d) imposed by total abundance (n). Thus, we
SEAGRASS OSTRACODES: CRONIN ET AL. 165
Table 3.—Confidence limits for 100, 200 and 300 individuals at
different relative frequencies.
Pp q n sigma t d
0.1 0.9 100 0.03 1.96 0.0588
0.2 0.8 100 0.04 1.96 0.0784
0.3 0.7 100 0.04582576 1.96 0.0898 1848
0.4 0.6 100 0.04898979 1.96 0.09602
0.5 0.5 100 0.05 1.96 0.098
0.6 0.4 100 0.04898979 1.96 0.09602
0.7 0.3 100 0.04582576 1.96 0.0898 1848
0.8 0.2 100 0.04 1.96 0.0784
0.9 0.1 100 0.03 1.96 0.0588
0.1 0.9 200 0.0212132 1.96 0.04157788
0.2 0.8 200 0.02828427 1.96 0.05543717
0.3 0.7 200 0.0324037 1.96 0.06351126
0.4 0.6 200 0.03464102 1.96 0.06789639
0.5 0.5 200 0.03535534 1.96 0.06929646
0.6 0.4 200 0.03464102 1.96 0.6789639
0.7 0.3 200, 0.0324037 1.96 0.06351126
0.8 0.2 200 0.02828427 1.96 0.05543717
0.9 0.1 200 0.0212132 1.96 0.04157788
0.1 0.9 300 0.01732051 1.96 0.0339482
0.2 0.8 300 0.02309401 1.96 0.04526426
0.3 0.7 300 0.02645751 1.96 0.05185673
0.4 0.6 300 0.02828427 1.96 0.05543717
0.5 0.5 300 0.02886751 1.96 0.05658033
0.6 0.4 300 0.02828427 1.96 0.05543717
0.7 0.3 300 0.02645751 1.96 0.05185673
0.8 0.2 300 0.02309401 1.96 0.04526426
0.9 0.1 300 0.01732051 1.96 0.0339482
0.1 0.9 1000 0.00948683 1.96 0.01859419
0.2 0.8 1000 0.01264911 1.96 0.02479226
0.3 0.7 1000 0.01449138 1.96 0.028403 1
0.4 0.6 1000 0.01549193 1.96 0.03036419
0.5 0.5 1000 0.01581139 1.96 0.03099032
0.6 0.4 1000 0.01549193 1.96 0.03036419
0.7 0.3 1000 0.01449138 1.96 0.028403 1
0.8 0.2 1000 0.01264911 1.96 0.02479226
0.9 0.1 1000 0.00948683 1.96 0.01859419
Pp = proportion of species; n = number of individuals, sigma =
standard error, t = level of confidence, d = c.
chose to obtain 100 individual valves and/or carapaces
from each 2-cm interval (except in the Park Key core,
where we picked 300 individuals) to determine decad-
al and sub-decadal faunal trends across a wide area of
Florida Bay. We were able to obtain about 100 indi-
viduals +20 from most samples, with the important
exception of the interval 72-134 cm in Bob Allen core
where there were usually less than 10 individuals.
Trends from this low abundance zone are interpreted
with caution.
TIME SERIES ANALYSES
We examined trends in L. matagordensis at Russell
Bank and compared them to trends in the Southern
Oscillation index (SOI, a measure of strength of El
Nino-Southern Oscillation) using time series analysis.
For time series analysis, the age of each data point
was estimated by linear interpolation between dated
depths. We then interpolated the Loxoconcha and SOI
records at constant |-year intervals. Both records were
separated into pre-1940 and post-1940 segments. All
segments were linearly detrended and the auto and
cross correlation functions for pre-1940 and post-1940
calculated. From these functions the coherency and
phase functions were calculated.
DATING AND CORRELATION METHODS
Isotopic dating by 7!°Pb and '*’Cs was carried out
by Robbins ef al. (2000) to obtain age models for the
Bob Allen and Russell Bank cores (see also Holmes
et al., 2000). These investigators found that Bob Allen
and Russell Bank cores BA-6-A and R-19-B have lin-
ear sediment accumulation rates of 0.75—1.04 and 1.22
cm yr! respectively, for the period covering approx-
imately the last 100-150 years. Park Key has an av-
erage accumulation rate of about 0.78 cm yr~! and the
86-cm core represents approximately the last 100
years. Sediment accumulation at Whipray was much
slower and there is an uncorrected radiocarbon age of
~3000 yr BP near the base.
The age model for the Manatee Bay core is not as
well established because 7!°Pb dating was unsuccessful
for unknown reasons (see Ishman ef al., 1998). How-
ever, a pollen age marker, the first appearance datum
(FAD) of the genus Casuarina between 60 and 70 cm,
suggests a maximum age for this interval at ~1890
and a probable age of 1920-1930, a time when this
genus became abundant in South Florida (D. A. Wil-
lard, personal communication). Similarly, the Casua-
rina FAD in the Russell Key core lies between 94—
104 cm, a level consistent with the isotopic age model.
TAXONOMIC AND ECOLOGICAL NOTES
There is considerable taxonomic and ecological data
on living ostracode populations collected from coastal
bays and estuaries along the Atlantic and Gulf coasts.
These include studies of Texas (Swain, 1955; King and
Kornicker, 1970; Garbett and Maddocks, 1978) and
Campeche, Mexico (Morales, 1966) lagoons; Baha-
man (Kornicker, 1961, 1963), Veracruz (Krutak,
1982), and Belize carbonate platforms (Teeter, 1975);
mangroves of southwest Florida (Keyser, 1975, 1976,
1977); the continental shelf off west Florida (Puri and
Hulings, 1957; Benson and Colman, 1963; Hulings
and Puri, 1964) and Florida Bay and other regions off
western Florida (Puri, 1960), the Cape Romano region,
northwest Florida (Benda and Puri, 1962); the Carib-
bean and Gulf of Mexico (Maddocks, 1975), and the
Atlantic coast (Cronin, 1979).
Species were identified following the taxonomy of
166 BULLETIN
Keyser (1976), Hazel (1983), Cronin (1979, 1990),
and Garbett and Maddocks (1978). In Florida Bay sed-
iment Malzella floridana (Benson and Coleman 1963)
is synonymous with Radimella floridana littorala
Grossman 1965 of Keyser (1976) and King and Kor-
nicker (1970). Loxoconcha matagordensis Swain 1955
is a widespread epiphytal species living along the At-
lantic and Gulf coasts. Xestoleberis spp. includes 3 to
4 species that were not distinguished from one another
because it is difficult to identify species of Xestoleberis
based solely on carapace features. Peratocytheridea
setipunctata is conspecific with Haplocytheridea seti-
punctata of some authors (see Hazel, 1983).
SUBSTRATE AND MUDBANK STRATIGRAPHY
Substrate is an important factor controlling the dis-
tribution and abundance of organisms throughout Flor-
ida Bay. Substrate types are generally divided into un-
consolidated carbonate muds, hard bottoms (mostly in
southern Florida Bay), mangrove islands (<2% of the
Bay), and seagrass “‘meadows.” Seagrasses, which
provide an important Florida Bay microhabitat for
many species (e.g., Sheridan 1997), constitute an in-
tegral component of mudbanks in the central part of
the bay in that they stabilize and help build up the
banks (Wanless and Tagett, 1989). The most important
types of seagrass species in Florida Bay are: Thalassia
testudinum (turtlegrass), Halodule (shoal grass,), Syr-
ingodium (manatee grass), and Ruppia (Widgeon
grass) (Robblee er al., 1991). There are also numerous
species of marine algae living in Florida Bay, which
provide habitat for ostracodes.
Seagrasses play an important role in sedimentation
on mudbanks in several ways. Wanless and Tagett
(1989) reported that broad-bladed Thalassia tend to
baffle wave and currents, an attribute that permits fine-
grained sediment to accumulate. Thalassia rhizome
mats also stabilize substrate. Swart and Kramer (1997)
reviewed evidence that many organisms, primarily red
algae and serpulid worms, and, to a lesser extent, mol-
lusks, foraminifers and other organisms, encrust on
Thalassia such that when seagrasses die, this carbonate
material makes a large contribution to the total car-
bonate production of the bay. In addition, Halley er al.
(1997) observed that, although seagrasses are rare-to-
absent in many mudbank sediment cores, seagrasses
may still have been present when “‘grassless” deposits
accumulated if those grasses had been living upstream
of the core site. While this remains a tenable hypoth-
esis at the local scale of mudbanks, our data on phytal
ostracode species presented below suggest that over
the last century, widespread changes in seagrass abun-
dance have occurred.
361
RESULTS
EPIPHYTAL OSTRACODE SPECIES IN
MODERN FLORIDA BAY
Vegetation and Substrate
Many ostracode species are adapted to inhabit spe-
cific substrates such as seagrasses and sandy bottoms.
Text-figure 2 presents data on the four ostracode taxa
from Florida Bay stations collected in February and
July 1998. For February and July samples, L. mata-
gordensis, M. floridana, and Xestoleberis spp. account
for a combined total of 74.5% and 70.9% of the total
assemblages, respectively. Two notable exceptions are
one sample from FB21 and one from FB6 from Feb-
ruary 1998, which are shelly substrate covering soft
mud with sparse Halodule. These two samples con-
tained abundant P. setipunctata. Another exception is
the abundance of P. setipunctata at sample FB-14. Of
the total 2,658 individuals collected from 36 marine
algae and seagrass samples during February 1998, 808
(30.4%) were L. matagordensis, 668 (25.1%) were
Xestoleberis spp., 614 (23.1%) were M. floridana, and
only 91 (3.4%) were P. setipunctata. Of 1,379 total
individuals obtained from 15 samples in July 1998,
248 (18%) were L. matagordensis, 274 (19.9%) were
M. floridana, 173 (12.5%) were P. setipunctata, and
442 (32.1%) were Xestoleberis spp.
Text-figures 3 and 4 compare the relative frequen-
cies of these four taxa by season and by vegetation
type. The relative frequencies of ostracode taxa living
on macro-benthic algae, and the seagrasses Thalassia
and Halodule are generally similar in both February
and July collections, except that L. matagordensis is
relatively more abundant on algae during February,
and both L. matagordensis and M. floridana are more
common on Thalassia in February. The species com-
position of ostracode assemblages on Halodule is very
similar during both months. Xestoleberis spp. predom-
inates on marine algae in both February and July sam-
ples. In general, these data show that L. matagordensis
and M. floridana are characteristic epiphytal species
living on Florida Bay seagrasses (Thalassia, Halodu-
le), and that Xestoleberis spp. prefer macro-benthic al-
gal habitats. P. setipunctata is most common in sam-
ples with Halodule, a pattern that may be an artifact
of the sampling of Halodule. Halodule is typically
found on very soft substrates, therefore root material
and adhering mud is often collected with the grass
sample.
These results confirm those of earlier studies that L.
matagordensis (Tressler and Smith, 1948; Swain,
1955; Morales, 1966) and related species (i.e., Loxo-
concha japonica, Kamiya 1988; L. fischeri, Teeter
1975) prefer mainly phytal habitats. Our evidence that
SEAGRASS OSTRACODES: CRONIN ET AL. 167
July 1998
Xestoleberis
P. setipunctata
M. floridana
L. matagordensis
25%) at winter temperatures of 19—
27°C. These may reflect different seasonal life histories
for these two species, allowing them to inhabit the
same epiphytal habitat on Thalassia. King and Kor-
nicker (1970) found a similar relationship as they doc-
umented seasonal changes in the relative abundance of
SEAGRASS OSTRACODES: CRONIN ET AL. 169
Florida Bay Ostracode Taxa: Salinity and
Temperature Relationships
salinity temperature
100
75
5
50: Loxoconcha
Malzella
percent total assemblage
P. setipunctata
Xestoleberis
15 20 25 30 35 40 45
ppt February: open diamonds
July: solid circles
Text-figure 5—Relationships between temperature and salinity and four ostracode taxa based on 1998 collections from Florida Bay.
170 BULLETIN 361
M. floridana and L. matagordensis in Texas Bays and
lagoons.
M. floridana also seems to be relatively common at
salinities between 15—20 ppt. However, King and Kor-
nicker (1970) studied the seasonal ecology of M. flor-
idana and demonstrated that this species dominated
assemblages from Texas lagoons during the summer
months when salinity often exceeded 35—40 ppt. Thus,
M. floridana is a euryhaline species tolerating a wide
range of salinities. The relationship of Florida Bay os-
tracodes to salinity and its application to reconstruction
of salinity history will be treated in a separate study.
In summary, it must be emphasized that multiple
environmental factors including, but not limited to,
substrate, salinity, and temperature, ultimately deter-
mine the relative proportion of various ostracode spe-
cies in any assemblage. There is nonetheless compel-
ling ecological evidence on Atlantic and Gulf Coast
ostracodes, accumulated over several decades of re-
search, in various marginal marine settings, and in the
new analyses presented above, that certain ostracode
species have preferences for phytal habitats. This hab-
itat preference makes M. floridana, L. matagordensis
and Xestoleberis spp. potentially valuable paleoecolog-
ical “proxy” indicators of changes in the relative
abundance of subaquatic vegetation in Florida Bay
when they are studied in well-dated sediment cores.
HISTORICAL TRENDS IN EPIPHYTAL OSTRACODES
Florida Bay Mud Island Stratigraphy and
Sedimentation
The stratigraphy, sedimentation, and microfaunal ta-
phonomy of Florida Bay mudbanks make them excel-
lent sedimentary environments in which to study the
paleoecology of Florida Bay. Before discussing his-
torical trends in ostracodes, we briefly describe some
important attributes of mudbank geology and paleon-
tology based on papers by Enos and Perkins (1977,
1979), Wanless and Tagett (1989) and Swart and Kra-
mer (1997) and references therein.
Florida Bay is divided into semi-isolated, shallow
(usually <1 to 2 m) sub-basins separated from each
other by small, partially emergent islands barely ex-
posed during low tide. These islands are called mud-
banks (Enos and Perkins, 1979; Wanless and Tagett,
1989). “Mud island” is the term given to the small
surfaces that sit on top of, or on the flanks of mud-
banks. There are more than 200 mud islands in Florida
Bay; they consist mostly of fine-grained carbonate
mud whose internal stratigraphy and abundant micro-
fossils make them amenable to isotopic dating and pa-
leoecological analysis (Wanless and Tagett, 1989;
Swart and Kramer, 1997; Robbins ef al., 2000).
Understanding sedimentary processes in Florida
Bay in general, and on mudbanks, in particular, is crit-
ical for establishing faunal and ecosystem trends from
sediment cores. Holocene sediment blankets the floor
of Florida Bay forming a thin veneer covering the
limestone “bedrock” consisting of the late Pleistocene
Miami Limestone. These sediments consist of about
95% unconsolidated carbonate derived mainly from
biogenic skeletal material from serpulids, calcareous
algae, foraminifers, coelenterates, ostracodes, mol-
lusks, and other organisms (Enos and Perkins, 1977;
Nelson and Ginsberg, 1986). Most sediments that form
the mudbanks are classified as bioturbated pelloidal
and molluscan wackestones and/or mudstones (Enos
and Perkins, 1979).
In their review of the origin and stratigraphy of
mudbanks, Wanless and Tagett (1989) classified Flor-
ida Bay mudbanks into four categories reflecting bank
morphology, stratigraphy, sedimentology and bank dy-
namics. In the “‘Central Migration Zone” of Florida
Bay, where the Bob Allen, Russell, and Park Keys are
located, sedimentary processes that form mudbanks
are generally dominated by erosion on the windward
side and deposition on the leeward side. These pro-
cesses produce features that have a history of migra-
tion that is dependent on the predominant winds, sed-
iment supply and wave energy (Enos and Perkins,
1979; Wanless and Tagett, 1989). Although mudbank
sediments are often bioturbated, in some cases they
consist of finely laminated muds suggesting little or no
bioturbation. Stratigraphic and isotopic studies also in-
dicate that some mudbanks have experienced relatively
high rates of fairly continuous sedimentation over the
past few centuries (Holmes ef al., 1997).
The provenance of mudbank sediments is also im-
portant in determining the significance of faunal and
paleoecological trends. Halley et al. (1997) concluded
on the basis of bomb-produced radiocarbon analyses
that mudbank sediment coarser than 250 «.m is not
transported, sediment between 63 and 250 ym is
sometimes transported, and sediment finer than 63 4m
is almost all transported.
In sum, the stratigraphy and sedimentation of Flor-
ida mudbanks is relatively well known, sediment ac-
cumulation rates are high and provide interdecadal
temporal resolution, and adult and most juvenile os-
tracodes studied from the >150 ym size fraction are
not transported (smaller juveniles are more likely to
have been transported), and abundant. These attributes
make paleoecological study of Florida Bay a viable
means to reconstruct the history of Florida Bay.
Trends in Epiphytal Species
Text-figures 6 and 7 show changes in relative fre-
quencies of the four indicator ostracode taxa in seven
SEAGRASS OSTRACODES: CRONIN ET AL.
a. Downcore Trends in Xestoleberis in Florida Bay
percent total assemblage
0 10 20 30 40 O 10 20 30 40 0 10 20 30 40
0 ) 0 0
10 20 20 10
6
20 40 40 20
, ; 30
30 60 60
a 40
40 80 80
low abundance 50
50 100 aaa 100 Be
Ee 60 120 120 an
gS =
<= 70 140 140 80
Qa
® 160 160 90
2 80
o ,
6 Whipray Bob Allen Russell
Oo
0 10 20 30 40 50 0 10 20 30 40 50 0 10 20 30 40 500
low abundance
S
0 0 0
10 20 20
20 40 40
30 60 60
40 80 80
50 100 5 100
60 120 120
70 140
80 160 160
171
0 10 20 30 4 Ol. 10 20 30 40
0 0
20
25
40
50 60
80
75
100
100 120
Park & Pass Taylor Manatee
10 20 30 40 S50 0 10 20 30 40 50 O 10 20 30 40 50
0 0
20
25
40
50 60
80
75
100
100 120
b. Downcore trends in Loxoconcha matagordensis in Florida Bay
Text-figure 6—Plot of relative frequencies of Xestoleberis (6a top) and L. matagordensis (6b, bottom) for seven cores. Park Key core is
solid line, Pass Key core is dotted line.
cores from Whipray, Bob Allen, Russell, Park, Pass,
Taylor, and Manatee Bay, plotted against core depth.
Several stratigraphic trends are evident from these re-
cords:
1. All cores except that from Park Key show increases
in the proportions of Xestoleberis from O to 10% in
the lower intervals of the cores to 20—40% in upper
intervals (Text-fig. 6a).
All cores except Pass Key show increases in the
proportions of L. matagordensis from O—10% to
> 10-30% (Text-fig. 6b).
3. Proportions of M. floridana oscillate widely in all
cores, generally with greater proportions in the up-
per intervals, except at Manatee Bay where this
species declines significantly (Text-fig. 7a).
There is a net decline in the proportions of P. se-
tipunctata in all cores (Text-fig. 7b).
i)
It is also noteworthy that at the Bob Allen and Rus-
sell Bank cores, two cores with the most detailed sam-
pling, the first order trends in all four taxa are gener-
ally similar, taking into account the artifacts produced
by low abundance between 70 and 120 cm at Bob
Allen.
The temporal significance of these trends is revealed
by plotting the relative frequencies of each species
against age derived from the lead-210 dating for five
cores from central and eastern Florida Bay (Text-figs.
8-11; Table 1; Holmes et al., 1997). It should be kept
in mind that the limitations of the *!°Pb dating preclude
precise inter-site comparison of interannual trends, es-
pecially during periods in the early 20" and late 19"
century when the precision of 7!°Pb dating diminishes.
Rather, our objective is to search for synchroneity or
asynchroneity in trends in indicator species over inter-
decadal timescales.
During the period 1910 through the 20’s, L. mata-
gordensis (Text-fig. 8) is relatively rare at Russell,
Park and Bob Allen, though the Bob Allen core has
sparse ostracodes in this interval. All three cores have
172
BULLETIN 361
a. Downcore Trends in Mailzella floridana in Florida Bay
10 20 30 40 50 60 0 102030405060 0 1020 30405060
0 0 0 0
10 20 20 10
, 20
20 40 40
30
30 60 60
40
40 80 80
a 50
50 100 100 a
60 120 120 70
70 140 140 80
80 160 160 90
Bob Allen
abundance
low
Whipray Russell
core depth (cm)
0 20 40 60 80 0 20 40 60 80 9 20 40 60 80
0 0 0 0
10 20 20 10
20 40 40 ay
3
30 60 60 a8
40
40 80 80
50
50 100 100 ,
60
60 120 120 ae
70 140 ] 140 80
80 160 160 90
a lise
0 10 20 30 40 50 60
0 10 20 30 40 50 60 0 20 40 60 80
Manatee
20 40 60 80 O 20) 740) 1607 80 0 20 40 60 80
0 0
20
25
40
50 60
80
75
100
100 120
b. Downcore Trends in Peratocytheridea setipunctata in Florida Bay
Text-figure 7.—Plot of relative frequencies of M. floridana (7a top) and P. setipunctata (7b, bottom) for seven cores. Park Key core is solid
line, Pass Key core is dotted line.
intervals where this species is relatively common
(>20—30%) during the late 1940’s through the 1950’s.
Another epiphytal ostracode species, Paracytheroma
stephensoni (not shown in figure) is also relatively
abundant in the Russell Bank core during this interval.
After the 1950’s, the abundance of L. matagordensis
fluctuates at Russell and Bob Allen Keys and shows a
gradual increase in abundance at Park Key. Three
cores exhibit significant declines in L. matagordensis
during the late 1970’s and early 1980’s: Bob Allen
(from 20 to 0%), Russell (from 28 to 8%) and Pass
Key (from 40 to <10%).
Malzella floridana appears to have been an impor-
tant member of the assemblage during the late 19" and
early 20" centuries at all sites (Text-fig. 9). At Russell
Bank, there was a progressive decline in M. floridana
between 1935 until the late 1950s when this species
reaches its lowest percentages. The Russell and Park
Key records suggest that following a period of mini-
mum abundance in the 1950’s, M. floridana became a
more abundant part of the ostracode assemblage
(~50%) during the 1960’s through 1980’s. However,
wide swings in its abundance (swings from less than
20% to almost 50%) occur over the past 30 years at
all the sites. The most noteworthy pattern is reminis-
cent of that observed for L. matagordensis. Two cores
(Russell and Bob Allen) show oscillations in M. flor-
idana during post-1960’s: Park Key shows a gradual
increase of M. floridana with minor declines during
the 1980’s and 1990’s, whereas Whipray shows a de-
cline from 35 to ~20% since the 1960’s. A decline in
M. floridana also occurred at Pass Key (from 40 to
10%) during the 1960s and 1970s. In general, these
data provide evidence for fluctuating abundances of M.
floridana including declines during the 1970s and ear-
ly 1980s at some sites.
An increase in the abundance of Xestoleberis (Text-
fig. 10) is dated at between the 1920’s and 1960's at
Whipray, Bob Allen and Russell Keys when Xestole-
beris rose from less than 10-15% to 30—40%. Xesto-
leberis abundance seems to have peaked during the
1950s and 1960s, declined during the 1970’s and the
early 1980s, then increased again at all sites except
Park Key during the late 1980’s and 1990’s. Except
SEAGRASS OSTRACODES: CRONIN ET AL. 173
Historical Trends in L. matagordensis in Florida Bay
2000
1990
1980
1970
1960
1950
1940
1930
1920
year
1910
1900
1890
1880
1870
®
c
°
N
)
°
=
©
v
=
=
a
©
S
°
1860
1850
1840
1830
1820
1810
0 10 20 30 40 50 0 10 20 30 40 S50 (0)
2000 2000
1990 1990
1980 1980
1970 1970
1960 1960
1950 1950
1940 1940
1930 1930
1920 1920
1910 1910
1900 1900
1890 1890
1880 1880
1870 1870
1860 1860
1850 1850
1840 1840
1830 1830
1820 1820
1810 1810
20 30 40 50 0 10 20 30 40 50 0 10 20 30 40 50
percent total ostracode assemblage
Winipray ey Bob Allen Key
Russell Key
Park Key Pass Key
Text-figure 8.—Trends in L. matagordensis in five cores from Florida Bay.
for the two samples dated at around 1890-1900, the
abundance of Xestoleberis at Park Key remained fairly
steady over the past century.
Significant declines in the relative frequencies of P.
setipunctata (Text-fig. 11) begin in the early 1930s at
Whipray, Bob Allen, Russell, and Park Keys and con-
tinued, with short reversals at Bob Allen and Park, up
through the 1990s. Reticulocythereis floridana (not
shown), another species that lives on sediment, also
declines in abundance from bottom to top of the Rus-
sell Bank core. All cores suggest that P. setipunctata
was a much more important part of the Florida Bay
ostracode fauna during the late 19" century and first
half of the 20" century that it was during the last half
century.
Because both L. matagordensis and M. floridana
have similar ecological requirements in terms of their
preference for Thalassia as a habitat, we lumped them
into a single “Thalassia” indicator index and com-
pared the temporal changes in their abundance in the
Russell and Park Key cores (Text-fig. 12). These sites
are located about eight kilometers apart, both have
abundant ostracodes throughout the core, excellent
*10Pb chronology and high sediment accumulation
rates, and thus should be comparable if the observed
trends are typical of the area. The results suggest that
prior to ~1920 oscillations in “Thalassia” species are
out of phase at the two sites, though we cannot exclude
the possibility that the age model for the lower parts
of these cores is incorrect. After a minimum abun-
dance at both cores around 1920, both cores show the
following similar fluctuations in the abundance of
these species. After ~ 1920, there is an increase to 40—
50%, then, a peak in abundance ranging from 50% to
65% occurred at Russell and Park respectively ~1936
and 1948, followed by a decline to <40% in the late
1950s, a steep rise in the 1960s, a decline in the 1970s
at Russell Key, and an increase to historically the high-
est levels at times during the late 1970s to the 1990s.
Thus, even though L. matagordensis and M. floridana
are known to have different patterns of relative abun-
dance during a single year due to seasonal effects
(King and Kornicker, 1970), the first-order decadal
patterns since ~1920 are similar at the two sites, sug-
174 BULLETIN 361
Historical Trends in M. floridana in Florida Bay
Whipray Key Bob Allen Key Russell Key Park Key Pass Key
2000 2000 2000 2000 2000 peest
1990 1990 1990 1990 1990
1980 1980 1980 1980 1980
1970 970 1970 1970 1970
1960 960 1960 1960 1960
1950 950 1950 1950 1950
1940 1940 1940 1940 1940
1930 930 1930 1930 1930
1920 920 1920 1920 1920
i 1910 1910 e 1910 1910 1910
S 1900 1900 2 1900 1900 1900
1890 1890 E 1890 1890 1890
1880 1880 FI 1880 1880 1880
1870 1870 E 1870 1870 1870
1860 1860 1860 1860 1860
1850 1850 1850 1850 1850
1840 1840 1840 1840 1840
1830 1830 1830 1830 1830
1820 1820 1820 1820 1820
1810 1810 1810 1810 1810
10 20 30 40 50 60 0 10 20 30 40 50 60 0 10 20 30 40 50 60 10 20 30 40 50 60 0 10 20 30 40 50 60
percent total ostracode assemblage
Text-figure 9.—Trends in M. floridana in five cores from Florida Bay.
gesting that similar changes in seagrass and algal hab-
itats have occurred.
POSSIBLE CLIMATIC EFFECTS ON FLORIDA BAY
EPIPHYTAL OSTRACODES
Environmental factors such as salinity variability
caused by climatological and hydrological processes
may also contribute to the observed decadal changes
in epiphytal ostracodes and their inferred seagrass hab-
itats. Dwyer and Cronin (this volume) demonstrated
that the influence of interannual and decadal variability
in rainfall, especially during winter season, has a large
impact on long-term salinity variability in Florida Bay.
Salinity in this portion of Florida Bay can range from
lower polyhaline values, <20 ppt, to hypersaline >45—
50 ppt. Thus, to examine the possible role of clima-
tological factors on Florida Bay epiphytes, we com-
pared trends in the relative frequency of L. matagor-
densis from the Russell Bank, Park Key and Bob Allen
cores with records of winter rainfall in south Florida
(Text-fig. 13). We chose this species because it its eco-
logical preference (salinity range is upper mesohaline-
polyhaline and epiphyte habitats of Thalassia and Zos-
tera beds) is well known from numerous studies along
the Gulf of Mexico and Atlantic coast of North Amer-
ica (see above).
The trends shown in Text-figure 13 suggest that for
most of the past century there is an inverse relationship
between winter rainfall and L. matagordensis abun-
dance. During wet periods (winter rainfall > 30 cm/
yr) such as the 1930s, 1950s and post-1980, L. mata-
gordensis is relatively abundant compared to drier pe-
riods (winter rainfall < 30 cm/yr) of the 1900-1910,
1915-1925, 1940s, 1960s and 70s. Two notable ex-
ceptions occur in the Russell core where during the
late 1940s/early 1950s and around 1970, L. matagor-
densis is abundant during periods of low rainfall. A
greater abundance of L. matagordensis, relative to oth-
er more marine species, during periods of wet climate
and lower Florida Bay salinity would be expected
based on this species’ preference for polyhaline con-
ditions.
SEAGRASS OSTRACODES: CRONIN ET AL. 175
Historical Trends in Xestoleberis in Florida Bay
2000 2000 2000 2000
1990 1990 1990 1990
1980 1980 1980 1980
1970 1970 1970 1970
1960 1960 1960 1960
1950 1950 1950 1950
1940 1940 1940 1940
1930 1930 1930 1930
= 1920 1920 1920
we oO
g 1910 5 1910 1910
1900 2 1900 1900
1890 : 1890 1890
1880 Fe 1880 1880
1870 é 1870 1870
1860 1860 1860
1850 1850 1850
1840 1840 1840
1830 1830 1830
1820 1820 1820
1810 1810 1810
0 10 20 30 40 0 10 20 30 40 0 10! 20) 30) 40 0 10 20 30 40
percent total ostracode assemblage
Whipray Key Bob Allen Key Russell Key Park Key Pass Key
Text-figure 10.—Trends in Xestoleberis in five cores from Florida Bay.
There are two important sources of uncertainty in
this analysis. First, it is not clear how rainfall-driven
salinity variations might affect the various species of
seagrass, especially Thalassia, which L. matagordensis
prefers as its habitat. Salinity is probably influencing
both the seagrass habitat of L. matagordensis as well
as this species ability to reproduce and survive in a
particular part of the bay (see Discussion). Second, we
assume that there is no “‘lag”’ time in the response of
populations to large-scale salinity oscillations such as
those observed by Robblee ef al. (1991) and Dwyer
and Cronin (this volume) for the past few decades.
Nonetheless, the broad patterns suggest that salinity,
as well as seagrass availability, is a factor in control-
ling the abundance of L. matagordensis over decadal
timescales.
DISCUSSION
The ecological data on ostracodes presented above
provide evidence that certain ostracode species are pre-
dominantly epiphytal in habitat in modern Florida Bay
and the fossil data show that there have been signifi-
cant interdecadal changes in their relative frequencies
at several localities in central and eastern Florida Bay.
The most important issue regarding the interpretation
of these interdecadal faunal trends is whether changes
at each site represent a local trend at a single mudbank
or whether they signify synchronous, regional events
that characterized the central and eastern parts of Flor-
ida Bay. If these do indeed signify regional changes
in Florida Bay benthic habitats, then what are their
causes?
It is unlikely that two of the major faunal trends
discussed above would occur by chance simultaneous-
ly at four or five mudbanks located kilometers apart.
The first is the long-term decline in the sediment- and
Halodule-dwelling P. setipunctata coincident with the
rise in species that inhabit Thalassia and macro-ben-
thic algae (L. matagordensis, M. floridana and Xesto-
leberis spp.) during the mid-20" century. An increase
in epiphytal species is not only evident in central and
eastern Florida Bay, but also near the northern tran-
176 BULLETIN 361
Trends in P. setipunctata in Florida Bay
2000 2000 2000 2000 2000
1990 1990 1990 1990 1990
1980 1980 1980 1980 1980
1970 1970 1970 1970 1970
1960 1960 1960 1960 1960
1950 1950 1950 1950 1950
1940 1940 1940 1940 1940
1930 1930 1930 1930 1930
1920 1920 = 1920 1920 1920
1910 1910 : 1910 1910 1910
.
S 1900 1900 FE 1900 1900 1900
> 1890 1890 5 1890 1890 1890
1880 ws === 2 1880 1880 1880
1870 1870 —1870 1870 1870
1860 1860 1860 1860 1860
1850 1850 1850 1850 1850
1840 1840 1840 1840 1840
1830 1830 1830 1830 1830
1820 1820 1820 1820 1820
1810 1810 1810 1810 1810
0 20 40 #460 0 20 40 60 0 20 40 #60 0 204 On GO 0 20 40 60
percent total ostracode assemblage
Whipray Key Bob Allen Key Russell Key Park Key Pass Key
Text-figure 11.—Trends in P. setipunctata in five cores from Florida Bay.
sition zone at the Taylor core site and in Manatee Bay,
though these sediments are not as well dated as the
other sites. Thus, central Florida Bay most likely ex-
perienced region-wide benthic habitat changes during
the mid-20" century. Moreover, similar synchronous
changes in mollusks and benthic foraminifers also oc-
curred in these areas (Brewster-Wingard et al., 1998;
Brewster-Wingard and Ishman, 1999).
The second event is the decline in seagrass and al-
gal-dwelling ostracodes species during the late 1970s
and 1980s. Although the strength of the decline varied
by species and by site, the stark contrast between the
abundance of epiphytal species during the 1950s and
60s and the very low abundance at times during the
following two decades suggests a broad regional shift
occurred at this time.
What were the causal factors behind these 20" cen-
tury changes? Many hypotheses have been proposed
to explain the causes of seagrass die-off in the late
1980s, but these theories do not address natural cycles
of change in seagrass distribution, and the possibility
that seagrass ““meadows”’ were not always as exten-
sive as they were in the early 1980’s. The most sig-
nificant and often cited factors proposed to explain the
die-off include changes in salinity (Boesch ef al.,
1993; Robblee et al., 1991; Eichinger et al., 1998;
Blakesley, 1998), light availability (Boesch ef al.,
1993; Stumpf and Frayer, 1998), water temperature
(Robblee et al., 1991; Zieman, et al., 1998), nutrient
availability (Lapointe ef al., 1994), disease (Blakesley
et al., 1998; Boesch et al., 1993; Durako, M. J. and
Kuss, 1994; Robblee et al., 1991), eutrophication (La-
pointe ef al., 1994), and hypoxia of roots (Carlson et
al., 1994; Robblee et al., 1991).
One approach to addressing the question of seagrass
dieoffs is to first examine what has caused both the
build-up of sub-aquatic vegetation beginning around
1950, and the subsequent decline in recent decades.
One hypothesis is that this mid-century shift was re-
lated to anthropogenic disruption of freshwater fiow
into Florida Bay, which has been described in papers
by Light and Dineen (1994) and Mclvor et al. (1994).
Beginning in 1910 and continuing for the next few
decades, canal and levee construction caused the shut-
SEAGRASS OSTRACODES: CRONIN ET AL. 177
Trends in seagrass ostracodes at
Russell and Park Keys
2000
ae
1990 :
Russell :
G aN.
Park
1970
oe
1960 6
oy
es Key CEC rer
Gitapncntaedsacassco °
= 1950 < oa
on
~
1940 &
1930 ~~
oO
ae
1920 Fe <
1910 ;
ee a
>
1900 Ei
nO
eo “oe
1890} “0
1880
- 7 T a 100
percent total
Text-figure 12.—Plot of summed proportion of L. matagordensis
and M. floridana in Russell Bank and Park Key cores.
off of freshwater flow southward from Lake Okeecho-
bee diverting it towards the Atlantic Ocean. Smith et
al. (1989) interpreted the changes in Florida Bay coral
fluorescence as a manifestation of these freshwater di-
versions. It is possible that as a result of these changes
to the natural hydrology, Florida Bay has, over the past
5 decades, become more sensitive to variability in
freshwater flow and/or to precipitation, at least com-
pared to the period prior to ~1940.
Another idea is that other anthropogenic events such
as the construction of the railway between Miami and
Key West disrupted the exchange of water between
Florida Bay and the Gulf of Mexico (Swart et al.,
1996a). This “railway” hypothesis holds that the
1907-1912 building of the railway restricted circula-
tion, which allowed the products of the oxidation of
organic carbon to build up in the bay. This event
might, in theory, have also disrupted Florida Bay cir-
culation and natural salinity variability. However, one
would have to explain the mid-century faunal shifts
preserved in central Florida Bay sediment cores in
terms of a lagged faunal response to the completion
of the railway.
The decreased frequency of hurricanes since 1948
was also noted by Swart er al. (1996a) as a possible
explanation for diminished exchange of Florida Bay
and Gulf of Mexico water and increased late 20" cen-
tury eutrophication that was evident from their coral
isotope record. It is interesting to speculate that the
carbon isotopic excursions recorded by Swart et al.
may be related to the appearance of seagrasses and
algae seen in our core records. It is nonetheless diffi-
cult to separate the hurricane influence from that of
ENSO (strong El Nino events are correlated with de-
creased hurricane landfall for the U.S. Atlantic coast,
Gray, 1984; O’Brien, 1996) and from the impact of
engineering projects in the 1950s and 60s (Swart et
al., 1996a).
An alternative hypothesis to explain changes in epi-
phytal ostracode species might be that, since the
1940’s, climatological factors have caused greater var-
iability in regional precipitation, at least relative to the
19" and early 20" centuries. This *‘climatic’’ hypoth-
esis would hold that post-1940 paleoecological vari-
ability is related to “natural’’ changes in the climate
system rather than human activities. There is, in fact,
considerable evidence that beginning in the mid-
1970’s, ENSO has behaved irregularly compared to its
behavior over the past few centuries. Trenberth and
Hurrell (1994), for example, showed that El Nino ep-
isodes became more frequent and La Nina episodes
were weaker and less common since 1976. In addition,
changes in longer-term climatic variability related to
the Pacific-North American (PNA) pattern (Wallace
and Gutzler, 1981) also occurred in the mid-1970s.
Henderson and Robinson (1994) showed that during
certain seasons changes in the PNA index coincided
with precipitation variability in the southeastern Unit-
ed States, though they cautioned that south Florida
precipitation exhibits patterns quite distinct from those
in states to the north.
178
Russell Bank
Park Key
BULLETIN 361
Bob Allen Key
% L. matagordensis
O10. 20,20. 20 5 10
2000
Sooo
1990 i
on
1980 ew
TSO.
ae
1970 ;
-——_
Fee
1960 =
er
_ 1950 saa
w AES
2 1940 Be
aaa
1930 3
Vee e
1920 []
f
a
1910 “4,
1900 Cie
1890
1880
1870
40
LSS 2025) S035
3570 25) 20 45,40 35) S025 ae20
Winter rainfall (Nov.-April) cm/yr
wetter
(7-point smooth)
ee ee Oe
Text-figure 13.—Plot of relative frequency of L. matagordensis at Russell Bank, Park Key and Bob Allen Key compared to winter rainfall
in South Florida (NOAA region #5) for the period 1880-1995. The rainfall record is a 7-point running mean. Note that during most dry
periods, when Florida Bay salinity was high, L. matagordensis abundance is low; the opposite situation is seen during wet periods. See text.
There is also a large body of literature on 20" cen-
tury trends in extreme weather events, including trop-
ical cyclones and extra-tropical storms (see Nicholls et
al., 1996). Most studies, however, have been carried
out on global or hemispheric scales and the impact of
long-term climate trends on Florida is uncertain. One
exception is the study of a 240-yr isotope record from
a Monastrea coral from Biscayne Bay, Florida by
Swart et al. (1996b). They discovered from oxygen
isotopic patterns that the 20" century may have been
significantly wetter that most of the 18" and 19" cen-
turies. If such a trend were characteristic of the latter
part of the 20" century in the nearby Florida Bay area,
one would expect the net effect to be a long-term de-
crease in salinity, rather than increasing periods of hy-
persalinity.
At this time, the climatological and canal hypotheses
seem to account for observed temporal trends in epi-
phytes and vegetation described above. Natural pro-
cesses have certainly exerted strong influence on Flor-
ida Bay habitats over the past century and continue to
do so. But the degree of impact may have been altered,
perhaps even exacerbated in some areas, by anthropo-
genic diversion of freshwater inflow. The shift in Flor-
ida Bay benthos during the mid-century, the relation-
ship between Loxoconcha and rainfall/salinity, and the
large amplitude of post-1950 faunal events, all suggest
that further investigation of the relationship between cli-
SEAGRASS OSTRACODES: CRONIN ET AL. 179
matological processes, such as ENSO and PNA, and
the timing of Florida Bay ecosystem changes are war-
ranted. Moreover, while climatological variability
would have affected the entire Florida Bay ecosystem,
information about the direct causes of seagrass die-offs,
such as disease or light attenuation, need further study
within the context of climate variability.
In regard to the topic of restoration of the “natural
state’’ of Florida Bay seagrass and algae, our data sup-
port historical observations that central Florida Bay
has experienced oscillations in the availability of sea-
grass and macro-benthic algal habitats that have been
especially severe during the past 30 years. Generally,
seagrass and algal habitats in the studied part of Flor-
ida Bay have been much more abundant and/or dense
during the last 50 years than during the first half of
the century. Thus, how one defines the “natural state”’
of Florida Bay then depends on the period one selects.
It is very likely that central and eastern Florida Bay
had extensive seagrass and macro-benthic algal
“meadows” during the mid-20" century and these
meadows diminished during the 1970’s and 1980’s.
However, seagrass and macro-benthic algal habitats
were probably sparse in this region during the 19" and
early 20" centuries, at least relative to the 1950s and
1960s.
REFERENCES CITED
Benda, W.S., and Puri, H.S.
1962. The distribution of foraminifera and Ostracoda off the
Gulf Coast of the Cape Romano area, Florida. Transac-
tions, Gulf Coast Association of Geological Societies, vol.
12, pp. 303-340.
Benson, R.H., and Coleman II, G.L.
1963. Recent marine ostracodes from the eastern Gulf of Mex-
ico. Kansas Univeristy Paleontological Contributions no.
31, Arthropoda, article 2, 52 pp.
Blakesley, B.A., Landsberg, J-H., Ackerman, B.B., Reese, R.O.
Styer, J.R., Obordo, C.O., and Lukas-Black, S.E.
1998. Seagrass disease in Florida Bay: pathogenicity of isolates
of the slime mold Labyrinthula in cultured Thalassia tes-
tudinum appears to parallel the distribution of Labyrin-
thula in the field. Proceedings, 1998 Florida Bay Science
Conference, May 12-14, 1998, pp. 84-87.
Boesch, D.F., Armstrong, N.E., D’Elia, C.F., Maynard, N.G.,
Paerl, H.W., and Williams, S.L.
1993. Deterioration of the Florida Bay Ecosystem: an evaluation
of the scientific evidence. Report to the Interagency Work-
ing Group on Florida Bay, 29 pp.
Brewster-Wingard, G.L., Ishman, S.E., Willard, D.A., Edwards,
L.E., and Holmes, C.W.
1997. Preliminary paleontologic report on cores 19A and 19B,
from Russell Bank, Everglades National Park, Florida
Bay. U. S. Geological Survey, Open-file Report 97-460,
29 pp.
Brewster-Wingard, G.L., Ishman, S.E., and Holmes, C.W.
1998. Environmental impacts on southern Florida coastal wa-
ters: a history of change in Florida Bay. Journal of Coastal
Research, special issue 26, pp. 162-172.
Brewster-Wingard, G.L., and Ishman, S.E.
1999. Historical trends in salinity and substrate at three sites in
central and northern Florida Bay: A paleoecological re-
construction using modern analog data. Estuaries, vol. 22,
no. 2, pp. 369-383.
Brush, G.S.
1986. Geology and paleoecology of Chesapeake Bay: A long-
term monitoring tool for management. Journal of Wash-
ington Academy Science, vol. 76, pp. 146-160.
Buzas, M.A.
1990. Another look at confidence limits for species proportions.
Journal of Paleontology, vol. 64, no. 5, pp. 842-843.
Carlson, P.R., Yarbro, L.A., and Barber, T.R.
1994. Relationship of sediment sulfide to mortality of Thalassia
testudinum in Florida Bay. Bulletin of Marine Science,
vol. 54, pp. 733-746.
Cronin, T.M.
1979. Late Pleistocene marginal marine ostracodes from the
southeastern Atlantic Coastal Plain and their paleoenvi-
ronmental interpretation. Geographie Physique et Quater-
naire, vol. 2, pp. 121-173.
Cronin, T.M.
1990. Evolution of Neogene and Quaternary marine Ostracoda:
United States Atlantic Coastal Plain: Evolution and spe-
ciation in Ostracoda IV. U.S. Geological Survey Profes-
sional Paper 1367-C, C1—C43.
Cronin, T.M., Willard, D., Kerhin, R., Karlsen, A., Holmes, C.,
Ishman, S., Verardo, S., McGeehin, J., Colman, S., and
Zimmerman, A.
2000. Climatic variability in the eastern United States over the
last millennium from Chesapeake Bay sediments. Geol-
ogy, vol. 28, no. 1, pp. 3-6.
Eichinger, C.D., Obordo, C.O., Blakesley, B.A., Carlson, P.R.,
Jr., Yarbro, L.A., Lukas-Black, S.E., and Reichert,
P.J.
1998. Turtle-grass (Thalassia testudinum) mortality on Florida
Bay mudbanks: Proceedings, 1998 Florida Bay Science
Conference, May 12-14, 1998, p. 90-93.
Enos, P., and Perkins, R.D.
1977. Quaternary sedimentation in South Florida. Geological
Society of America Memoir 147, 198 pp.
Enos, P., and Perkins, R.D.
1979. Evolution of Florida Bay from island stratigraphy. Geo-
logical Society of America Bulletin, vol. 90, pp. 59-83.
Fourqurean, J.W., Jones, R.D., and Zieman, J.C.
1993. Processes influencing water column nutrient characteris-
tics and phosphorous limitation of phytoplankton biomass
in Florida Bay, Fl. USA: Inferences from spatial distri-
butions. Estuarine Coastal Shelf Science, vol. 36, pp.
295-3 14.
Garbett, E.C., and Maddocks, R.F.
1978. Zoogeography of Holocene cytheracean ostracodes in the
bays of Texas. Journal of Paleontology, vol.53, pp. 841—
919.
Gray, W.M.
1984. Atlantic seasonal hurricane frequency. part 1. El Nino and
30 mb Quasibiennial oscillation. Monthly Weather Re-
view, vol. 112, pp. 1649-1668.
180 BULLETIN 361
Grossman, S.
1965. Ecology of Rhizopodea and Ostracoda of southern Pam-
lico Sound region, North Carolina, part 1, Living and sub-
fossil rhizopod and ostracode populations. Kansas Uni-
versity Paleontological Contributions no. 44, Ecology, ar-
ticle 1, pp. 1-82.
Halley, R.B., Smith, D., and Hanson, M.
1995. Florida Bay surface salinities. U.S. Geological Survey
Open-file Report 95-634.
Halley, R.B., Holmes, C.W., and Prager, E.J.
1997. Florida Bay mudbanks: relatively new piles of mostly old
sediment. U.S. Geological Survey Open-file Report 97-
385, pp. 26—27.
Hanson, K., and Maul, G.A.
1991. Florida precipitation and the Pacific El Nino, 1895-1989.
Florida Scientist, vol. 54, pp. 160-168.
Hartmann, G.
1981. Die ostracoden der ordnung podocopida G.W. Muller,
1984 der subtropisch-tropischen Ostkuste Australens
(zwischen Eden im Suden und Heron-Island im Norden).
Mitteilungen aus dem Hamburgischen Zoologischen Mu-
seum und Institut, vol. 78, pp. 97-149.
Hartmann, G.
1984. Zur kenntnis der ostracoden der polynesischen inseln hu-
ahine (gesellschaftsinseln) und rangiroa (taumotu-inseln).
Mitteilungen aus dem Hamburgischen Zoologischen Mu-
seum und Institut, vol. 81, pp. 117-169.
Hazel, J.E.
1983. Age and correlation of the Yorktown (Pliocene) and Croa-
tan (Pliocene and Pleistocene) formations at the Lee
Creek Mine. in C. E. Ray, ed., Smithsonian Contributions
to Paleobiology, no. 53, pp. 81-199.
Henderson, K.G., and Robinson, P.J.
1994. Relationships between the Pacific/North American tele-
connection patterns and precipitation events in the south-
eastern United States. International Journal of Climatol-
ogy, vol. 14, pp. 307-323.
Holmes, C.W., Marot, M., Willard, D., Weimer, L., and Brew-
ster-Wingard, L.
1997. Methods to establish the timing of ecological changes in
South Florida—good, better, best. U.S. Geological Survey
Open-file Report 97-385, pp. 33-34.
Holmes, C.W., Robbins, J.A., Halley, R.B., Bothner, M., Ten
Brink, M., and Marot, M.
2001. Sedimentary dynamics of Florida Bay Mud Banks on a
decadal time scale. Bulletins of American Paleontology,
no. 361, pp. 31—40.
Hudson, J.H., Powell, J.V.D., Robblee, M.B., and Smith, T.J.
1989. A 107-year-old coral from Florida Bay: a barometer of
natural and man-induced catastrophes? Bulletin of Marine
Science, vol. 44, pp. 283-291.
Hulings, N.C., and Puri, H.S.
1964. The ecology of shallow water ostracods from the west
coast of Florida. Pubblicazioni Stazione Zoologica di Na-
poli, no. 33, pp. 308-344.
Ishman, S.E.
1997. Ecosystem history of South Florida: Biscayne Bay sedi-
ment core descriptions. U.S. Geological Survey Open-file
Report 97-437, 13 pp.
Ishman, S. E., Cronin, T.M., Brewster-Wingard, L., Willard, D.,
and Verardo, D.
1998. A record of ecosytem change, Manatee Bay, Barnes
Sound, Florida. Journal of Coastal Research special issue
26, pp. 125-138.
Kamiya, T.
1988. Morphological and ethological adaptations of Ostracoda
to microhabitats in Zostera beds. in Hanai, T., Ikeya, N.,
Ishizaki, K., eds., Evolutionary Biology of Ostracoda. El-
sevier Developments in Palaeontology and Stratigraphy
11, pp. 303-318.
Keyser, D.
1975. Ostracoden aus den Mangrovegebieten von Sudwest-Flor-
ida. Abhandlungen und Verhandlungen des Naturwissen-
schaftlichen Vereins in Hamburg, vol. 18/19, pp. 255—290.
Keyser, D.
1976. Zur kenntnis der brackigen mangrovebewachsenen
Weichboden Sudwest-Floridas unter besonderer Beruck-
sichtigung ihrer Ostracodenfauna. Unpublished Ph.D. the-
sis, Hamburg University, 142 pp.
Keyser, D.
1977. Brackwasser-Cytheracea aus Sud-Florida. Abhandlungen
und verhandlungen des Naturwissen-schaftlichen Vereins
in Hamburg, vol. 20, pp. 43-85.
King, C.E., and Kornicker, L.S.
1970. Ostracoda in Texas bays and lagoons: an ecologic study.
Smithsonian Contributions to Zoology, no. 24, pp. 1-87.
Kornicker, L.S.
1961. Ecology and taxonomy of recent Bairdiinae (Ostracoda).
Micropaleontology, vol. 7, pp. 55-67.
Kornicker, L.S.
1963. Ecology and classification of Bahamian Cytherellidae
(Ostracoda). Micropaleontology, vol. 9, pp. 61—69.
Krutak, P.R.
1982. Modern ostracodes of the Veracruz-Anton Lizardo Reefs,
Mexico. Micropaleontology, vol. 28, pp. 257-288.
Lapointe, B.E., Tomasko, D.A., and Matzie, W.R.
1994. Eutrophication and trophic state classification of seagrass
communities in the Florida Keys. Bulletin of Marine Sci-
ence, vol. 54, pp. 696-717.
Light, S.S., and Dineen, J.W.
1994. Water control in the Everglades: a historical perspective.
in Davis, S.M., and Ogden, J.C., eds., Everglades: the
Ecosystem and its restoration, St. Lucie Press, Delray
Beach, FL, pp. 47-84.
Maddocks, R.F.
1975. Ostracodes. in Bright, T.J., and Pequegnat, L.H., eds., Bi-
ota of the West Flower Garden Bank. Flower Garden
Ocean Research Center, Gulf Publishing, Houston, Texas,
pp. 200-229.
Melvor, C.C., Ley, J.A., and Bjork, R.D.
1994. Changes in freshwater inflow from the Everglades to Flor-
ida Bay including effects on biota and biotic processes: a
review. in Davis, S.M., and Ogden, J.C., eds., Everglades:
the Ecosystem and its restoration, St. Lucie Press, Delray
Beach, FL, pp. 117-146.
Morales, G.A.
1966. Ecology, distribution, and taxonomy of recent Ostracoda
of the Laguna de Terminos, Campeche, Mexico. Bulletin,
Universidad Nacional Autonoma de Mexico Instituto de
Geologia, vol. 81, pp. 1-98.
Nelson, J.E., and Ginsberg, R.N.
1986. Calcium carbonate production of epibionts on Thalassia
in Florida Bay. Journal of Sedimentary Petrology, vol. 56,
pp. 622-628.
Nicholls, N., Gruza, G.V., Jouzel, J., Karl, T.R., Ogallo, L.A.,
and Parker, D.E.
1996. Observed climate variability and change. in Houghton,
J.T., Meira Filho, L.G., Callender, B.A., Harris N., Kat-
SEAGRASS OSTRACODES: CRONIN ET AL. 181
tenberg, A., and Maskell, K., eds., Climate Change 1995.
Cambridge University Press, pp. 133-192.
O’Brien, J.J.
1996. The effect of El Nino on U.S. landfalling hurricanes. Bulletin
American Meteorological Society, vol. 77, pp. 773-774.
Patterson, R. T., and Fishbein, E.
1989. Re-examination of the statistical methods used to deter-
mine the number of point counts needed for micropale-
ontological quantitative research. Journal of Paleontology,
vol. 63, no. 2, pp. 245-248.
Peterson, D., Cayan, D., DiLeo, J., Noble, M., and Dettinger, M.
1995. The role of climate in estuarine variability. American Sci-
entist, vol. 83, pp. 58-67.
Puri, H.S.
1960. Recent Ostracoda from the west coast of Florida. Trans-
actions, Gulf Coast Association of Geological Societies,
vol. 10, pp. 107-149.
Puri, H.S., and N.C. Hulings.
1957. Recent ostracode facies from Panama City to Florida Bay
area. Transactions of the Gulf Coast Geological Society,
vol. 7, pp. 167-190.
Robbins, J.A., Holmes, C.W., Halley, R.B., Bothner, M., Shinn,
E., Graney, J., Keeler, G., ten Brink, M., Orlandini,
K.A., and Rudnick, D.
2000. First order time-averaging of Cs-137 and Pb fluxes to Pb-
210-dated sediment in Florida Bay. Journal of Geophyscal
Research, Oceans, vol. 105, pp. 28,805—28,821.
Robblee, M.B., Barber, T.R., Carlson, Jr., P.R., Durako, M. J.,
Fourqurean, J.W., Muehlstein, L.K., Porter, D., Yar-
bro, L. A., Zieman, R.T., and Zieman, J.T.
1991. Mass mortality of the tropical seagrass Thalassia testu-
dinum in Florida Bay (USA). Marine Ecology Progress
Series, no. 71, pp. 297-299.
Sheridan, P.
1997. Benthos of adjacent mangrove, seagrass and non-vegetat-
ed habitats in Rookery Bay, Florida, U.S.A. Estuarine,
Coastal and Shelf Science, vol. 44, pp. 455-469.
Smith, T.J., Hudson, H., Robblee, M.B., Powell, G.V.N., and Is-
dale, P.J.
1989. Freshwater flow from the Everglades to Florida Bay: a
historical reconstruction based on fluorescent banding in
the coral Solenastrea bournoni. Bulletin of Marine Sci-
ence, vol. 44, pp. 274-282.
Stumpf, R.P., and Frayer, M.L.
1998. The distribution and areal extent of seagrass die-offs and
the light availability for seagrass in the Florida Bay region
from 1985 to 1997. Proceedings, 1998 Florida Bay Sci-
ence Conference, May 12-14, 1998, p. 80.
Swain, F.M.
1955. Ostracoda of San Antonia Bay, Texas. Journal of Pale-
ontology, vol. 29, pp. 561-646.
Swart, P.K., Healy, G.F., Dodge, R.E., Kramer, P., Hudson, J.H.,
Halley, R.B., and Robblee, M.B.
1996a. The stable oxygen and carbon isotopic record from a
coral growing in Florida Bay: a 160 year record of cli-
matic and anthropogenic influence. Palaeogeography, Pa-
laeoclimatology, Palaeoecology, vol. 123, pp. 219-237.
Swart, P.K., Dodge, R.E., and Hudson, H.J.
1996b. A 240-year stable oxygen and carbon isotopic record in
a coral from South Florida: Implications for the prediction
of precipitation in southern Florida. Palaios, vol. 11, pp.
362-375.
Swart, P.K., and Kramer, P.A.
1997. Geology of mud islands in Florida Bay. in Vacher, H.L.
and Quinn, T., eds., Geology and Hydrogeology of Car-
bonate Islands: Developments in Sedimentology, vol. 54,
pp. 249-274.
Teeter, J.W.
1975. Distribution of Holocene marine Ostracoda from Belize.
in Wantland, K.F, and Pusey III, C., eds., Belize Shelf-
Carbonate Sediments, Clastic Sediments, and Ecology.
American Association of Petroleum Geologists Studies in
Geology no. 2, pp. 400—498.
Trenberth, K.E., and Hurrell, J.W.
1994. Decadal atmosphere-ocean variations in the Pacific. Cli-
mate Dynamics, vol. 9, pp. 303-319.
Tressler, W.L., and Smith, E.M.
1948. An ecological study of seasonal distribution of Ostracoda,
Solomons Island, Maryland, region. Chesapeake Biolog-
ical Laboratory Publication no. 71, 61 pp.
Valette-Silver, N.J.
1993. The use of sediment cores to reconstruct historical trends
in contamination of estuarine and coastal sediments. Es-
tuaries, vol. 16, pp. 577-588.
Wallace, J.M., and Gutzler, D.S.
1981. Teleconnections in the geopotential height field during
Northern Hemisphere winter. Monthly Weather Review,
vol. 109, pp. 784-811.
Wanless, H.R., and Tagett, M.G.
1989. Onin, growth and evolution of carbonate mudbanks in Flor-
ida Bay. Bulletin of Marine Science, vol. 44, pp. 454—489.
Wingard, G.L., Ishman, S.E., Cronin, T.M., Edwards, L.E., Wil-
lard, D.A., and Halley, R.B.
1995. Preliminary analysis of down-core biotic assemblages:
Bob Allen Keys, Everglades National Park, Florida Bay.
U.S. Geological Survey Open-file Report 95-628, 35 pp.
Zieman, J.C., Fourqurean, J.W., and Iverson, R.L.
1989. Distribution, abundance and productivity of seagrasses
and macroalgae in Florida Bay. Bulletin of Marine Sci-
ence, vol. 44, pp. 292-311.
Zieman, J.C., Fourqurean, J., and Frankovich, T.
1998. Seagrass dieoff in Florida Bay (USA): Long term trends
in abundance and growth of Thalassia testudinum. Pro-
ceedings, 1998 Florida Bay Science Conference, May 12—
14, 1998, p. 101.
Appendix la.—Pass Key Ostracode census data.
Ie,
seti- Total
Depth punc- — Loxo- Mal- Xesto- Ostra-
(cm) Year tata concha zella leberis codes
1 1997 0 13 52 16 118
3 1996 (0) 1 43 14 67
5) 1995 0 0 34 7 Sp)
9 1993 0 14 35 14 116
13 1991 0 15 35 15 124
17 1989 i 6 36 22 94
21 1987 0 8 50 15 112
25 1985 0 17 45 4 114
29 1983 0 21 33 7 94
33 1981 0 15 21 7 94
Si/ 1979 0 3 26 17 86
41 1977 0 39 11 iil 1
53 1971 0 48 16 8 121
69 1964 5 18 38 7 98
73 1962 0 47 20 14 108
182 BULLETIN 361
Appendix 1b.—Ostracode species census data, Bob Allen Key core 6-A.
Actino- Acuticy- Cythero- Loxo-
cythereis thereis Aglaio- morpha_ Cyther- Cyther- Dolero- Lepto- _ concha Loxo- Malzella _ Neone-
Depth cf. sub- laevis- cypris Cytherel- paracas- ura ura cypriacythere matagor- concha flori- sidea
(cm) quadrata —_sima sp. loidea tanea_ elongata fiscina sp- sp- densis spp. dana sp-
0 5 1 20 17 3
2 8 3 3 3 6 53 2 54 21
6 10 1 7 5 45 3 32 7
8 1 1 9 1 5 2
10 16 2 i 6 5
12 1 4 | 4 3 3
14 1 5 1 1
16 1 1 2 1
18 1 2 3 7 1
2 1 1 4
22 1 1 3 1 1
24 1 2 1 6 1 10 1
26 5 1 6 18 7 20 5
28 1 1 8 9 1
30 1 2 1 4 5 1 8 6
32 ] 1 1 6 2 6 3}
34 1 5 1 10 4
36 2, 1 12 3 10 2
38 2 2 1 1 8 5 3
40 4 2 2 49 2 19 17
42 2 1 1 1 12 2 11 6
44 1 | 9 1 5 2
46 1 10 1 2 29 2 19 2
48 1 1 1 3 7 1 8 4
50 2 10 I 3 8 + 34 39 I
52 1 4 2 1 1 itil 19
54 1 2 14 13 1
56 2 2 1 2 16 7 5
58 3 1 1 1 10 6 3
60 1 2 1 1 9 7
62 3 1 1 1 3 1
64 2 1 15 7 1
66 2 4 1 1 11 11 1
68 1 3 2 2 11 5)
70 3 13 2, 2
TP) | 1 1
74 1 | 1 2
76
78 3
80 1 1
82 2, 1 1
84
86 1 1
88 1 1 2 2 3
90 1 1
92 1
94 1
96
98 2 1 3 i
100 2
102 1
104 1
106
108 1 3
110
112 2
114 1 1
116
118 1 1 1 3 1
SEAGRASS OSTRACODES: CRONIN ET AL. 183
Appendix 1b.—Extended.
Paracy- Peratocy-
Paracy-. theroma_ Peratocy-__ theridea Radimella Reticulo- Triangulo- — Xesto-
theroma stephen- theridea setipunc- Proponto- Proteo- | Puriana_ confra- cythereis cypris leberis | ?Micro- Total
repexa soni bradii tata cypris — concha Spp- gosa_ floridana __laeva spp. cythere Ostracodes
1 7 1 2 10 67
9 4 2 14 3 1 3 13 67 269
12 1 33 3 1 4 5 28 197
5 4 1 1 30
3 3} 2 43 4 1 92
4 1 10 52 1 2 ) 4 99
2 6 46 1 8 3 3 77
1 11 2 1 20
3 2 6 1 5 31
2 8 1 2 19
3 5 1 6 22)
1 3 H 1 3 10 41
10 21 1 1 7 il} 115
2 2 1 7 13 45
3 4 1 1 1 8 19 65
1 8 2 1 16 48
3 17 ] 11 53
2 1 12 12 11 68
3 1 1 3 8 38
12 2 2 18 17 43 189
4 1 17 ] 10 6 75
1 ®) 5 5 39
9 3 4 25 3 2; 1 13 25 151
4 1 1 13 2 1 6 8 62
4 6 22 40 4 3 1 2 26 30 235
4 3 9 Sy 3 1 6 21 126
4 3 16 1 5 5 7 72
2 17 Ss) 3 2 5) 10 129
1 3 8 3 1 4 5) 50
1 16 1 7 46
3 4 18 6 6 1 48
4 10 1 5 8 54
4 23 7 8 2)
2 4 13 1 1 1 5 1 52
2 13 1 2 38
1 2 6
13 4 4 26
2 7) 4
2 2 7
2 5) 9
1 6 11
6 6
1 1 18 3 2 2
1 1 1 1 1 14
8 2 1 13
2 1 3 7
2 3
1 1 2
1 2 3 5) 24
1 1 3 1 8
1 2
1 1 3
2 1 3
2 1 3 3 | 14
1 1
1 3 1 7
13 l 16
184 BULLETIN 361
Appendix 1b.—Continued.
Actino- Acuticy- Cythero- Loxo-
cythereis thereis — Aglaio- morpha_ Cyther- Cyther- Dolero- Lepto- — concha Loxo- Malzella __Neone-
Depth cf. sub- — laevis- cypris Cytherel- paracas- ura ura cypria cythere matagor- concha flori- sidea
(cm) quadrata — sima sp. loidea tanea_ elongata fiscina sp. sp. densis spp. dana sp.
120 2
122 1
124 1
126 1
128 1 1
130 1 1
132 1 l 2
134 1 3
136 1 1 1 i 4 7 1
138 | 1 1 1
140 1 2 9
142 2 1 3 1 16
144 4
146 1 2 3
148 1 2
150 2 8
152 6 1 3)
154 1 15 3 1 2 5 11
156 3 4
ee
SEAGRASS OSTRACODES: CRONIN ET AL. 185
Appendix 1b.—Extended Continued.
Paracy- Peratocy-
Paracy- theroma_ Peratocy- theridea Radimella Reticulo- Triangulo- Xesto-
theroma stephen- theridea_ setipunc- Proponto- Proteo- Puriana_ confra- cythereis — cypris leberis — ?Micro- Total
repexa soni bradti tata cypris — concha spp. gosa _floridana laeva spp. cythere Ostracodes
1 1 4
1 1 3
3 4
2) 6 1 10
2 4
2 6 1 2 13
2) 6 1 2 2 17
2 6
2 i 1 DZ 6 28
2 2 2 3 13
2 1 12 2 2 1 32
2 18 1 3 3 50
1 1 7 1 1 2 N7/
3 1 10
6 1 3} 13
4 2 4 20
8 1 1 3 25
2 2 18 2 1 7 8 78
1 3 1 5) 4 21
186 BULLETIN 361
Appendix 1c.—Ostracode species census data, Manatee Bay core MB-1.
Actino- Acuti- Cythero-
cythereis cythereis Cytheri- morpha
Depth sub- laevis- Candona Cyther- Cytherel- della Cytherura Cytherura — paracas-
(cm) — quadrata sima Bairdrids annae Cytherid ellid loidea alosa spp. forulata tanea
l 1 8 22
3 12 16
5 1 5) 16
7 l 1 15
9 8 16
11 e} 7] 3
13 2 4 15
15 2 25 1
Wii | 1 3 2 21
19 1 4 28
21 1 6 17
23 3 4
25 1 1 i 11
27 1 1 1 2 7 9 5
29 1 4 2 2 1 4 5 1
31 2 1 5 8 5 3 3 2
33 1 1 5) 1 5 5 3 4
35 5) 3 5) 1 5 4 4 5 5
3h 2 1 7 2 3 6 2 1 4
39 4 1 6 3 1 1 5)
41 7 4 5 l 2 4 6 1 3
43 2 6 2: 3 2 2 1 5
45 1 2 2 4 7 5
47 4 4 4 4 5 2 2
49 4 1 6 1 1 | 4 1 2
Sil 2 1 17 1 3 3 i 10
53 12 9 4 4
35 4 1 8 4 5
57 5 1 16 7 6 6
59 6 6 2 8 2 2
61 4 3 6 2
63 5 4 3 3
65 1 1 2 1 2 1 1
67 5 1 2 2
69 1
71 2 2 1 2 1
73 2 1 1
75 6 2 1 2 1
77 5 2 1 1
79 3 3
81 5 I |
83 3 ] 3
85 3 2
87 2) 1 7
89 3} 1
91 5) 1
93 2 1 8 1
95 2 4 1 1
97 1 12 11
99 2 2
101 1 5)
103 4 7 8 1
105 1
107 1
109
111
113 30 22
115 1 15 11
117 1 21 24
SEAGRASS OSTRACODES: CRONIN ET AL. 187
Appendix 1c.—Extended.
Perato-
Loxo- cythe-
Cythero- — concha ridea Reticulo- Triangulo- Xesto-
morpha matagor- — Malzella Mega- Paracy- setipunc- cythereis = Puriana Thalloso- cypris leberis Total
sp. densis floridana cythere theridea tata floridana convoluta cypris? sp. spp. Ostracodes
15S 31 ] 16 36 1 17 148
17 20 2 14 32 13 126
12 11 3 1 15 30 6 9 109
23 23 ] 7 26 6 17 120
19 21 1 8 9 2 22 106
19 21 1 2 7 15 6 1 20 124
11 Ail 1 1 9 10 1 4 3 18 100
19 17 5 ] 13 9 3 2 10 107
15 20 2 12 14 6 | 11 109
17 PL] 2 ] 14 10 1 5 1 10 121
19 28 4 10 1 1 6 6 8 107
18 12 1 2 13 27 80
14 30 5 6 9 2 4 2 IN7/ 109
15 22 3 10 3 l 5 13 98
30 28 8 1 3 5 12 i 26 134
1 19 20 2 4 5 8 17 3 28 131
1 19 Di, 1 1 1 1 3 15 9 Ds) 128
22 32 3 3 2 4 2 3 3 26 137
18 28 2 3 5 4 2 5 8 20 123
1 20 37 2. 4 2 3 10 6 11 28 145
2 23 26 1 6 4 3 4 4 6 23 135
3 26 28 1 3 p) 7 14 41 148
3 7 38 6 1 2 8 7 5 45 148
2 15 39 2 5 1 3 5 8 8 37 150
3 24 20 1 3 3 4 12 5 26 122
3 25 15 1 1 3 2) 3 4 21 116
22 56 10 20 2; 4 31 175
1 19 31 4 1 1 8 4 2 29 122
1 19 31 3 1 6 5 8 3 30 148
1 14 54 2 2 1 6 3 5) 3 25 142
2 16 52 1 5 10 1 4 3 23 132
10 42 5 5 2 2 22 103
7 18 1 3 2 5 38 84
14 44 2 1 6 1 2 7 19 106
3 15 1 3 1 12 36
i 69 Z 1 10 16 4 i 6 5 129
4 41 2 24 14 1 2 3 95
5) 93 2 10 11 2 1 18 154
3 101 3 B 19 1 13 152
5 81 2 8 19 2 12 135
4 71 2) 5 10 12 111
3 92 3 5 7 2 1 12 132
2 83 1 6 13 11 121
4 90 2 9 12 6 133
2 90 7 12 1 116
B 88 1 16 16 7 137
74 4 14 5 104
81 1 4 13 10 117
1 48 2 10 17 1 5 108
18 3 3 28
1 32 7 5 1 52
56 12 19 3 110
14 3 3 1 22
3 3 1 8
0
0)
17 5 2 3 79
2 51 1 18 11 1 111
2 46 1 19 10 3 127
188
Apppendix 1d.—Park Key core Ostracode census data.
BULLETIN 361
Depth Year
1 1995
3 1992
5 1990
7 1987
9 1984
11 1982
13 1979
15 1977
17 1974
19 1972
21 1969
23 1967
25 1964
27 1961
29 1959
31 1956
33 1954
35 1951
37 1949
39 1946
4] 1943
43 1941
45 1938
47 1936
49 1933
51 1931
53 1928
55 1925
57 1923
59 1920
61 1918
63 1915
65 1913
67 1910
69 1908
71 1905
73 1902
75 1900
77 1897
79 1895
81 1892
83 1890
85 1887
P. setipunctata
Loxoconcha
5
18
9
16
11
14
ms
Monn pf f
wo
85
84
83
70
79
72
65
76
69
Malzella
140
111
155
143
128
127
115
140
88
Sil
Xestoleberis
15
26
9
14
12
17
15S
13
Ww
SS SS Ss SSS)
NWN CO
ere fw f
Se NW eRe
NWre fAIYW Ore
lo}
Ww
Total ostracodes
Appendix le.—Ostracode species census data, Russell Bank core 19-B.
Depth
Year
1995
1993
1992
1990
1988
1987
1985
1984
1982
1980
1979
LOTT
1975
1974
1972
1970
1969
1967
1965
1964
1962
1961
1959
1957
1956
1954
1952
1951
1949
1947
1946
1944
1943
1941
1939
1938
1936
1934
1933
1931
1929
1928
1926
1924
1923
1921
1920
1918
1916
1915
1913
1911
1910
1908
1906
1905
1903
1902
Actino-
cyther- cyther-
Acuti-
eis cf. eis Aglaio- Aurila
subqua- laevis- cypris laevi-
drata — sima sp. cula
2 2
1
1
1
2
1
1 2
1 1
1 2 1
2 3} 2
1 1
1 1
1
1 1 1
2
2
1 3
2 1
3
3
2
1
1 1
1
3 2
3 5
6 2
2 1
3 2
3 1
2 7
6 3
10 5
9 5
7 2 1
2 9 1
8 6
4 q
3 4
5 1
2 5
2 8
6
3
2 2
2 3 1
2 4 2
3 3 1
3 a
2 1 1
1 4
5 3
SEAGRASS OSTRACODES: CRONIN ET AL. 189
Cythero- Cyther- Cyther- Loxo-
Bytho- Cythe- morpha ura’ Cyther- ura Hemi- Lepto- concha _ Loxo-
cy- rel- para- elon- ura sablen- cyther- cythere matagor- concha
therid loidea castanea_ gata __ fiscina Sis ura sp. densis spp-
2 2 2 3 14
1 2 6 3 18
1 2 1 5 1 15
4 2 3 10
a 2 1 1 4 16
1 1 1 1 1 12
5 7) 1 14
3 3 4 2 2 35
4 1 Z 1 22
3 1 19
1 2 9
1 2 1 1 16
4 I 19
1 3} 1 1 1 28
3 4 1 5 15
2 4 1 1 2 22
2 3} 1 30
1 4 2 3} 3 22
2 4 2 3 9
3 1 5 1 17
2 4 3 1 12
4 6 5 3 25
5 5 5 14
5 4 2 22
4 9 4 18
3 6 9 28 1
4 1 3 1 3 24
2 2 2 1 35
1 4 4 1 30
7 1 3 29
4 1 2 15
7 5 3 2 1 27
1 2 9 1 1 25
1 9 4 4 14
1 2 10 4 2 2 18
3 4 1 12
1 1 2 6 1 13 1
1 1 1 2 7
2 1 2 4 1
1 1 2 1 7 1
2 1 2 2 8
1 4 2} 9 1
1 5 3 3
7 2 1 12
1 2 2 1 1 5
1 1 3 3) 1 1 5
1 1 2 3 1 8
1 1 3 1 2 6 1
3 1 3 2 1 4
3 1 3} 2 11
5 1 3) 1 9
5 1 2 11
2 6 1 10
2 2 1 6 1
1 8 5) 1 13
1 Z 4 2 2 e)
3 4 4 1 1 2 5
1 1 4
190,
Appendix
le.—Continued.
Depth Year
100
102
104
106
108
110
112
114
1995
1993
1992
1990
1988
1987
1985
1984
1982
1980
1979
1977
1975
1974
1972
1970
1969
1967
1965
1964
1962
1961
1959
1957
1956
1954
1952
1951
1949
1947
1946
1944
1943
194]
1939
1938
1936
1934
1933
193]
1929
1928
1926
1924
1923
1921
1920
1918
1916
1915
1913
1911
1910
1908
1906
1905
1903
1902
BULLETIN 361
Perato-
Neo- Para- Para- cythe- Reticu-
Mai- caud- cyther- Para- cythe- Perato- ridea Radi- locyth- Tri-
zella ?Mi- ites idea cythe- roma cythe- _ seti- Puri- mella_ ereis angulo- Xesto-
flori- cro- triplis- Neone- tschop- roma steph- ridea_ punc- Proteo- ana con-_ flori- cypris leberis
dana cythere triatus sidea pi repexa ensoni bradyi tata concha spp. fragosa dana laeva _ spp.
25 1 3 1 2 40
32 1 2 3 1 1 4 1 23
39 1 3 3 35
36 2 1 1 1 1 39
39 4 1 1 1 2 32
37 1 3 1 2 1 41
35 6 l 2 3 1 32
17 3 4 1 1 1 36
45 1 1 2 2 2 2 1 1 28
39 1 1 1 1 41
52 2 2 1 1 2 1 1 26
46 3 2 3 2
23 1 1 1 5 1 1 2 2 2 3 30
21 3 4 1 2 2 28
39 1 3 1 1 2 1 21
35 3 R) 3 1 6 4 2 2 15
39 2 6 1 7 2 1 1 12
36 2 4 i 4 1 2 1 1 16
46 1 1 1 1 2 1 2, 22
57 4 3 5 1 4 1 1 17
38 2 1 2 3 1 8 1 1 2 27
16 5 5 5 1 36
19 3 4 1 5 1 1 1 42
16 1 2 2: 9 2 4 36
25 1 5) 2 4 7 4 1 2 23
11 1 2 6 2 2 1 47
14 2 2 15 1 1 26
13 8 4 9 3 2 32
29 3 2 12 4 1 1 27
11 6 10 5 2 1 7 14
33 3 2 13 6 2 1 1 10 il7/
23 1 2 2 14 7 ] 3 5 36
34 5 ] 1 11 v7 2 RS) 1 8 17
31 3 2 1 4 8 1 6 2 W 4 16
28 2 1 3 4 5) 2 4 2 4 13
46 2 2 3 9 4 11 4 4 4
3i7/ 1 2 3 5 2 9 3 10 1 5 10
42 2 5 ib) 2 8 1 3 5 5
51 1 2 1 3 23 2. 6 3 3) 3
42 1 23 3 8 1 3 6 9
40 2 1 4 2 2 12 3 10 5 2 4
44 1 3 2 13 3 6 1 3 4 9
43 4 2 6 2 5 2 2}
43 ] 1 3 2 1 15 4 3 4 5 9
35 8 2 15 2 10 1 1 3 10
32 1 5 i 19 3 1 5 8
30 1 2 1 4 11 2 6 4 4 14
47 1 5 3 6 6 3 9
3)5) 2 9 1 16 2 4 1 4 4 16
45 5 15 2 2 7
42 1 5 14 3 6 1 14
42 1 3 17 2 6 1 2 1 10
51 1 1 4 2 1 10 2: 6 5 1 21
42 1 1 2 1 10 5 4 3 1 14
31 6 1 1 5 3 10 1 2 17
31 2 4 2 3 14 1 6 1 3 9
33 1 3 1 18 3 3 4 6 9
28 3 1 21 1 1 4
Other Total
SS
98
106
104
109
104
103
114
116
110
100
104
100
103
99
1 11S
108
106
99
122
112
114
109
105
112
121
98
113
121
97
115
147
137
120
112
113
122
109
124
123
112
118
92
124
106
96
1 103
105
114
99
109
110
132
103
115
2 102
1 107
73
SEAGRASS OSTRACODES: CRONIN ET AL. 19]
Appendix le.—Continued.
Actino- Acuti-
cyther- cyther- Cythero- Cyther- Cyther- Loxo-
eis cf. eis Aglaio- Aurila Bytho- Cythe- morpha ura Cyther- ura — Hemi- Lepto- concha_ Loxo-
subqua- laevis- cypris laevi- cy- rel- para- elon- ura sablen- cyther- cythere matagor- concha
Depth Year drata — sima sp. cula_ therid loidea castanea_ gata __ fiscina sis ura sp- densis spp.
116 1900 2 1 ] |
118 1898 2 2 1 1 3 il 2 12 1
120 1897 3 1 2 1 3
122 1895 4 3} I 6
124 1893 4 6 1 1 7
126 1892 2 2 1 1 3 1 1 1
128 1890 7 4 1 1 1 4 1
130 1888 3 1 2 1 1 3 5
132 1887 3 3 | 2 3 ]
134 1885 1 2 1 1 2 1 1 4
136 1883 1 6 1 6 2 1 2 9 1
138 1882 6 2 1 4 2 1 13
140 1880 5 5 1 i 2 2 1
142 1879 2 14 1 1 2 2 1 13
Appendix 1f.—Ostracode species census data, Taylor Core T-24.
Acuti- Cypri- Cythero- Cytherura
cythereis dopsis | morpha cf. Dolero- Hetero- Lepto- — Limno-
Depth /aevis- Cypris Cyprideis okee-_ paracas- Cytherura’ reticu- — cypria cypris _cythere — cythere
(cm) sima Candona Cypretta sp. salebrosa chobei tanea — sandbergi lata sp- punctata sp. floridana
10) 1 a 16 1 3 11 1
2 1 14 Z 17
4 1 9 1 23 1
6 3 11 2 12 1 1
8 1 1 11 1 8 18 2 1
10 1 2 22 1 2 12 1 3
12 2 2 5) 251 1 4 18 1 1 1 1
14 2 15 1 8
16 2 122 1 9 i 1
18 2 8) 1 4 1
20 1 9 1 3 1
22 11 1 3 1 2 2
24 1 10 1 2
26 4 26 3 1
28 3 6 2 1
30 1 1S 4 2 1
32 1 11 2 W 2 1
34 1 1 15 3 1 1 1
36 3 11 1 3 1 1
38 1 12 2
40 2 13 1 1 2
42 2 10 2 1 5
44 2 14 2 1
46 8 1 1
48 1 5 1 1 1 2
50 2 7 1 1
ny) 1 1 9 3 6
54 1 9 2 4 2 2
56 1 7 2 8 1 3
58 3 11 3 5 2 1
192 BULLETIN 361
Appendix le.—Extended.
Neo- Para- Para- Perato- Reticu-
caud- cyther- Para- cythe- Perato- cythe- Radi- locyth- Triang-
Maizella ?Mi- _ ites idea cythe- roma cythe- ridea mella ereis ulo- Xesto-
flori- cro- triplis- Neone-tschop- roma steph- ridea _ seti- Proteo- Puriana con-_ flori- cypris leberis
dana_cythere triatus sidea pi repexa ensoni bradyi punctataconcha spp. fragosa dana _ laeva spp. Other Total
12 1 1 2 1 3 2 1 2 30
24 1 1 2 1 13 5 2 5 3 13 1 96
20 2 6 1 5 3 1 4 1 53
20 1 2 1 1 7 1 1 7 1 63
39 2 2, 4 8 2 3 95
39 1 1 1 3 YU 2 1 Z 1 10 86
34 1 1 3 4 10 3 11 ©) 2 7 104
48 1 1 4 1 1 15 4 6 1 1 8 107
46 1 1 11 1 8 2 14 97
49 1 1 8 1 6 1 8 88
35 1 1 2 1 13 4 2 6 14 108
35 4 2 d 1 6 2 2 6 94
Bi, 2 3 4 6 6 D 5 7 89
35 4 6 2 4 1 1 8 18 115
Appendix 1f.—Extended.
Para-
Loxo- Para- cythero- Peratocy- Perisso- Perisso-
concha Parapon- cythero- ma theridea cytheridea cytheridea Reticulo- Xesto-
matagor- Malzella _ toparta ma stephen- seti- brachy- cf. cythereis Thalasso- leberis Total
densis __floridana sp. repexa soni punctata forma cribrosa_ floridana_ cypria spp. Other Ostracodes
19 48 2 1 28 4 1 10 148
18 48 38 5 2 7 152
24 50 2 ] 5 47 3 1 1 9 3 181
24 42 3 50 7 2 7 165
27 59 3 1 68 7 2 3 13 226
23 1S 1 1 3 74 8 2 5) 236
24 103 2 3 1 1 135 10 3 13 358
16 41 1 54 16 1 4 1 160
10 65 1 1 1 Wil 4 1 6 2 188
10 21 1 44 10 1 104
11 Sil 1 53 8 119
10 43 3 713 5 1 3 1 159
4 28 2 43 2 2 95
5 13 45 11 1 1 110
5 21 59 4 101
9 21 2. 42 7 3 107
5 20 2 54 1 2 108
5 21 1 2 3h7/ 6 3 1 1 100
10 23) 1 2 33) 8 3 100
5 20 3 1 54 2 3 103
14 24 1 3 32 3 1 1 98
1] 22 1 1 39 3 1 4 102
9 Sif 33 5 1 5 103
12 31 2 2 36 2 2 2 5 1 105
17 Dill 4 4 1 35 2 5 106
6 38 3} 3 1 32 3 5 102
11 40 1 1 1 29 3 5 111
12 Sil 2 36 6 1 2 110
6 19 1 45 22 2 3 100
7 19 2 2 51 3 2D) 111
SEAGRASS OSTRACODES: CRONIN ET AL.
Appendix 1g.—Whipray Key Ostracode census data.
Xesto- Total
Depth P. seti-Loxo- leberis ostra-
(cm) Year = punctata concha Malzella _ spp. codes
1 1995 11 8 6 11 57
5 1985 9 8 24 37 107
®) 1976 0) 19 28 9 99
13 1967 2 10 31 23 95
17 1957 7 9 32 15 115
21 1948 6 10 27 14 109
25 1939 11 9 28 8 102
29 1930 29 9 22 5 101
33 1920 29 7 45 2 145
37 1911 PH] 5 32 3 100
41 1902 14 6 58 0) 97
45 1892 10 20 32 8 98
49 1883 12 3 17 8 52
53 1874 43 2 23 0) 81
S7/ 1864 34 3 6 2 56
61 1855 40 3 13 2 85
65 1846 Tih 10 21 1 127
69 1837 61 2 43 1 146
73 1827 40 2 31 1 108
77 1818 2 17 49 4 125
78) 1813 21 7 29 2 91
194
Appendix 2.—Ostracode species occurrence on Florida Bay seagrasses and marine algae.
BULLETIN 361
Acuti- Loxo-
cythe- Cythe- Cy- Hemi- concha
Actino- Aglaio- reis Aurila Cythe- Cythe- rura _ thero- cythe- mata- Loxo-
Sample ID Sub- cythe- cypris laevis- laevi- Bair- — rura rura — sand- morpha Cypri- Cythe- rura gor- concha
ID number strate _reis sp. simacula_ diids_ cybaea fiscina bergi sp. deis reliids sp. densis sp.
FB1 1 G 1 2 4 6 3 oy)
FBI 4 G 4 1 2 3 21
FB1 5 G 2 3 5 1 2 29
FB11 1 G 1 2 1 2 i 17 1
FBI 2 A 1 3} 1 1 18
PBS 3 G 2 1 1
ralesill 3) G 1 7) g 2. 10 2 21
FB12 1 A 1
BRI 2 A 1 1 28 3
FBI2, 3 G 2 4 6 3 2 8 1
FBI2 4 G 1 2
FBI2 5 G 1 4 17 1 11
FBI4 1 G 1 3 4 2 35
FBI6 1 G 100 1
FB16 2 A 1 57
FBI6 3 A 1 70
FB17 1 G 1 1 2 7
laesily 2 A 1 50 5 2 3 2 18
BB A 1 17 1 40
FBS, 2 A 11 2 1 1 20
FB20 1 G 2 1 3) 15
FB20 2 G 1 1 5) |
FB20 3 A 2
FB20 4 A | ]
FB20 5 A 5
BB2I G 2 2 1 42 2
FB21 shell S 2 2 a 4 22 12
FB22 1 G 2 1 1 1 1 26 2
BB23 G 2 1
FB26) 1 G 5 1 1 oy) 1
FB26 2 G 9
FB4 4 G 1 1 1 2 3} 2 16
FBS 1 A 1 1 2 9
FB5 2 A 1 l 3 3 2 4 24
FB5 3 A 4
FB5 5 A 1 6 2 1 | 29
FB6 shell S 3 1 4 29 4 1
FB8 1 G 1 l 18 3
Total for 36 veg
samples 15 17 1 12 85 35 29 16 56 7 30 5 808 14
% for 36 veg
samples 0.564 0.64 0.038 0.451 3.198 1.317 1.091 0.602 2.107 0.263 1.129 0.188 304 0.527
Total for 2 sand
samples 5 0 3 0 4 10) 7 0) 10) 33 0) 0 26 13
% for 2 sand
samples 2.092 0) 1-255 10) 1.674 10) 2.929 (0) O 13.81 0 0) 10.88 5.439
Total for 15 algal
samples 3 4 0) 12 74 14 9 0 6 0) 13 3 325) 3
% algal samples 0.281 0.375 0 1.126 6.942 1.313 0.844 0 0.563 0) 1.22 0.281 30.49 0.281
Total for 21
Grass samples 1 13 1 10) 11 21 20 16 50 W 17 2 483 11
% Grass samples 0.754 0.817 0.063 0) 0.691 1.319 1.256 1.005 3.141 0.44 1.068 0.126 30.34 0.691
* Perissocytheridea troglodyta and P. rugata.
** Some Triangulocypris, Thalassocypris, indeterminate taxa.
# Includes Puriana, Leptocythere, Caudites.
Mal-
zella
flori-
dana
23
92
24
38
— WwW pond
ONnNK DANK ODN WW
— -
aornnNneHe D~
i)
Appendix 2.—Extended.
SEAGRASS OSTRACODES: CRONIN ET AL.
Perato-
Para-_ cythe-
Megacy- cythe- ridea_ Perato-
there roma seti- cythe- Peris- Reticu-
John-— steph-— punc- _ ridea_ socythe- locythe-
soni ensoni tata bradyi ridea* reis
2 3 2
| 4 9 2
1 i 1
1 l 15 1
1 12 2
2
2 9 4
1 11 25 3 1
1 10 2
2
]
2
1 2
2 1
1 l
1
1
4 47 1
1 1 1
1
1
1
2,
2 1
3 3 3
2 2 2
69 1 2
1 9
21 21 91 8 Pei} 13
0.79 0.79 3.424 0.301 1.016 0.489
4 0 116 1 0 3
1.674 0) 48.54 0.418 (0) 1.255
8 2 9 5 2 1
0.75 0.188 0.844 0.469 0.188 0.094
13 19 82 3 25 12
0.817 1.193 S25 OMS8 1257 0.754
Thal-
asso.
&
Tri-
ang.**
1
1
1
i)
i)
14
0.879
Xesto-
leberis
Spp-
19
11
13
11
32
Other
#
BWNAD
Ne
i)
Total
118
151
92
112
65
31
2658
Vegetation
Laurencia, Thalassia
Thalassia, Halodule, Laurencia
Thalassia, Polysiphonia, Laurencia
Thalassia, Polysiphonia
Chondria, Thalassia
Halodule, Thalassia
Thalassia, Chondria, Polysiphonia
Sargassum
Chondria, Polysiphonia, Laurencia, Thalassia
Thalassia
Halodule
Thalassia
Ruppia, Halodule, Thalassia, Laurencia
Thalassia, Laurencia, Batophora
Laurencia, Batophora, Thalassia
Polysiphonia, Batophora, Thalassia, Halodule
Thalassia, Chondria
Chondria, Thalassia
Chondria, Polysiphonia, Thalassia, Halodule,
Laurencia
Syringodinium
Thalassia, Polysiphonia
Halodule
Chondria, Polysiphonia, Thalassia
Chondria, Batophora, Thalassia
Polysiphonia
Thalassia, Halodule, Laurencia
shell
Thalassia, Laurencia
Syringodium, Thalassia, Chondria,
feathery algae
Halodule, tulip-like algae, Thalassia,
Laurencia, Chondria
Halodule, Thalassia
Thalassia, Sargassum, Polysiphonia
Siphonia, Batophora, Chondria (2 species),
Thalassia, Laurencia, Halodule
Laurencia, Acetabularia, Enteromorpha,
Halodule, Batophora
Laurencia, Chondria, filamentous
green algae
Laurencia, green mossy filamentous algae,
Batophora, Chondria, Halodule, Thalassia
shell
Thalassia
196 BULLETIN 361
Appendix 3—Modern ostracode data for 1995.
Cy-
Actino- thero-
cythe- mor- Cy-
reis Proteo- pha _ ther- Cy-
sub- Acuti concha para- ura ther- Dolero-
quad- laevis- Aglaie- Caud- tuber- Cypri- Cythrel- cas- elon- Cyther- ura cypria
Slide Location rata sima cypris ites culata deis Cypris loidea tanea gata ura sp. fiscina _ sp.
95GLW 16 Bob Allen Keys 1 1 q 1
95GLW 17 Bob Allen Keys 3 2
95SGLW18A Shell Creek—mouth 1 3 ®)
95GLW18B Shell Creek—mouth 4 1 2 2 3 2
9S5SGLWI19A Trout Creek—mouth 9 4 0 Wf 1
95GLW 19B Trout Creek—mouth 2 5 4
95GLW20A Duck Key 1 1 6 1 5
95GLW20B Duck Key 1 3 ; 1 + 1 7 1
95GLW21 Nest Key 5 1 2 3 11 9 3 1
95GLW22A Porjoe Key 1 1 1 1 3 1
95GLW22B Porjoe Key 1 3 2 5 3
95SGLW22C Porjoe Key 1 4 8 4 1 4
9S5GLW22D Porjoe Key 1 4 2 i 2
9SGLW23 Butternut Key 2 1 11 6 1 il 0
Bob Allen Keys 0.95 0 (0) 10) 0.95 0 (0) 0.95 0 0) OPS 0) 10)
Bob Allen Keys 3516, 0 2-117 10 0 0 0 (0) 231 nO) 0 0 0
Shell Creek—mouth (0) (0) 0 (0) 10) 0 10) 1.01 0 3.03 0 9.09 0
Shell Creek—mouth (0) 10) 0 0 0 4.3 1.08 (0) 0 2.15 AMsy 3728} ZS
Trout Creek—mouth 10) 0) 0) 0 0) 9 (0) 0 4 10) 10) 7 1
Trout Creek—mouth 0 (0) 0 10) (0) (0) (0) 0) 6.67 0 16.7 10) ils\33)
Duck Key 1 0 0 1 (0) 6 10) 0 1 5 (0) 10) 0)
Duck Key 12037 33:09. 2:06; 0 1.03 O 0 avid 0} 7/2? 1.03 0 0)
Nest Key 4.5 0.9 1.8 0) Do 0 0) GO| fill We 0) 0 0.9
Porjoe Key 12 Dien De 10 0 1-22 10 0) 122 nO) 3.66 0 1.22 0
Porjoe Key 0 0) 0.96 0 0 0 (0) 2.88 1.92 4.81 10) 2.88 0
Porjoe Key 0.89 O (0) 0 (0) 0 (0) BS Gaia) BNe537/ 0.89 3.57 0
Porjoe Key 101 0 0 0) 0 0 0 4.04 2.02 7.07 (0) 0 2.02
Butternut Key 3.28 1.64 0O (0) 18 9.84 0 (0) 0 1.64 1.64 0 0
Bottle Key 122") 10:49" 0!5 0.07 1.71 2.08 0.08 1.98 2.8 2.92 167, E93 139)
SEAGRASS OSTRACODES: CRONIN ET AL. 197
Appendix 3—Extended.
Para- Perato- Peris-
cy- cy- SO-
Loxo- Para-_ thero- theri- Perisso- cythe-
Limno- concha Mal- cy- ma Para- Perato- dea cythe- ridea Reti- Trian-
cythere mata- zella Neo- thero- Ste- cy- cythe- seti- Perisso- ridea cf. cula gulo- Xesto-
flori-— gor- flori- nesidea ma_ phen- the- ridea punc- cythe- brachy- cri- Puri- flori- Thal. cypris leberis
dana densis dana sp. repexa soni ridea bradii tata ridea forma brosa ana dana vavrai laeva_ spp. Total
13 27 6 l 2 1 1 4 46 105
11 32 2 3 1 3 6 8 IF) 95
12 49 1 2 3 4 15 99
14 44 1 5 3 12 93
1 8 28 1 25 4 5 V 100
18 1 30
8 30 3 2 2 30 1 10 100
21 26 a) 1 4 4 5 1 1 12 97
22 20 3 2 5 1 | 19 111
36 16 1 2 2 1 16 82
22 32, 4 1 3 1 1 26 104
24 19 1 3 5 1 6 3 18 112
25 13 2 1 4 1 3 34 99
2 12 2 4 1 1 17 61
(0) 12742557 S271 0.95 129) 0:95 0 0 0 0 0 0.95 0 3.81 0 43.8 100
0 6S 3357, 10 2.11 10) 0 0) 3:16, 10 EOS 0) 3.16 632 842 0 17.9 100
(0) 12a 495 0 0) 1.01 0 0 (0) 10) 2102 3103; 10 0) 4.04 0O 15.2 100
(0) 15.1 473 O 108 0O (0) (0) (0) 0 5.38) 3123) 10 0) 0 0 12.9 100
1 8 2 (0) 1 10) 10) 10) 0) 10) 25 4 0 10) 5 (0) 76 ~=100
0 60 10) (0) 0) 10) 0 0) (0) (0) 0 (0) 0 3333770 (0) 0 100
(0) 8 30 3 0 2 0 0 2 30 10) 0 0 0) 1 0 10 100
0 DG 26:8 92:06) 50 N03) 4212) 451210 10) 10) Splisy 1.03 12035) 10 12.4 100
0 19.8 18 2.7 1.8 4.5 (0) (0) 0 0 10) 0 (0) 0.9 3.6 (0) 17.1 100
0 43.9 19.5 PP (0) 0) 0 (0) 244 O 10) (0) 244 O 12250) 19.5. 100
10) PIP) Shey Ifo) (0) 0.96 0 10) 2.88 0 10) 10) 0 0.96 0 0.96 25 100
10) 2A 2529" 10:89)" 2168" 4:46) 10. Ol89F 5:36) 70 0 10) 0 (0) 2.68 O 16.1 100
0 25.3 13.1 0 2.02 1.01 0 0 4.04 0O 10) 0 1.01 3:03' 0 0 34.3 100
10) 3.28 Ie @) 3.28 O 10) (0) 6.56 0 10) 10) 164 O 164 O 27.9 100
0.07 20135 26:3 1.39 1.07 1.21 0.36 0.36 1:89 2.14 2:39 1.1 0.66 1.11 2.32 0.07 18.5
CHAPTER 10
MOLLUSCAN FAUNAL DISTRIBUTION IN FLORIDA BAY,
PAST AND PRESENT: AN INTEGRATION OF DOWN-CORE AND MODERN DATA
G. LYNN BREWSTER-WINGARD!, JEFFERY R. STONE!, AND CHARLES W. HOLMES?
'U. S. Geological Survey, 926A National Center, Reston, Virginia 20192
2U. S. Geological Survey, 600 4" St., St. Petersburg, Florida 347011
ABSTRACT
Statistical comparison of modern molluscan fauna to down-core molluscan assemblages in four cores elucidates changes in
the Florida Bay ecosystem during the past 100 to 200 years. Fluctuations within molluscan faunal dominance and diversity
patterns suggest a response to changing environmental conditions. Faunal dominance patterns indicate an increase in salinity in
the northern transitional zone, and possibly the eastern portion of Florida Bay. Distinctive faunal shifts recorded at Russell Bank
occur approximately between 1913 and 1933 and at Bob Allen mudbank between approximately 1900 and 1910. The period
from approximately 1930 to 1980 within these cores shows rapid and dramatic fluctuations in species dominance and faunal
richness. Beginning around 1980, the mussel Brachidontes exustus, which can tolerate diminished water quality and a wide range
of salinities, increases in percent abundance in the upper portion of all four cores and becomes the dominant species at Russell
Bank and Bob Allen Mudbank.
While these fluctuations within assemblages are distinctive, they are not so profound that they represent a major shift in
estuarine zonations within northern, eastern, and central Florida Bay during the past 100 to 200 years. The majority of the
molluscan fauna that are present at the core sites today are generally present throughout the period of deposition. Fluctuations
in the molluscan faunal record down-core primarily express changes in dominance and diversity within assemblages and do not
reflect substantial changes in overall assemblages. It is these fluctuations in dominance and the appearance or disappearance of
critical indicator species that are indicative of salinity changes.
Understanding the dynamics of an ecosystem and the natural range of variation in the system over an extended period of time
is a critical component of effective restoration. Analysis of the modern environment provides a means to interpret biological data
preserved in cores, and to determine the physical and chemical variations in the environment indicated by the biota. Knowledge
of the past provides the best insight to predicting the impact of future change on the environment.
INTRODUCTION
Molluscs are an abundant and diverse component of
the shallow coastal ecosystems of Florida. Two hun-
dred and twenty-one families, 646 genera, and 1407
species of molluscs have been documented from the
shallow marine environments of the state (Camp, Ly-
ons, and Perkins, 1998). In Florida Bay alone, 235
species of molluscs were recorded by Lyons (1999)
over the course of a three-year sampling period (1994—
1996). This diversity can be attributed in part to south-
ern Florida’s position at the junction of two molluscan
faunal provinces: the Caribbean and the Carolinian.
Molluscs are an integral part of the Florida Bay eco-
system. Different groups of molluscs are components
of the micro-, meio-, and macro-fauna of the bay and
as some species pass through their life stages they
move from the plankton to the benthos. In addition,
molluscs employ a number of feeding strategies, in-
cluding filter feeding, grazing, scavenging, and carniv-
ory. Consequently, molluscs occupy a number of tro-
phic levels within the ecosystem and they fill many
important niches as both predators and prey.
Analyzing historical patterns of change in molluscan
assemblages is a very powerful tool in reconstructing
the history of an ecosystem. Estuarine molluscs tolerate
a wide range of fluctuating environmental conditions,
yet on a scale of seasons, their populations will migrate
to more ideal conditions (Dame, 1996). Lyons (1996,
1998) has demonstrated molluscan migration within
Florida Bay in response to seasonal and annual varia-
tions in salinity. Individual molluscs, however, are rel-
atively limited in their movement in post larval stages,
and typically provide information about a specific site
for the duration of their lifetime. In addition, most mol-
luscs are hard-shelled, and therefore well preserved in
cores. Turney and Perkins (1972) related the distribu-
tion of molluscan death assemblages to environmental
influences. On the basis of these previous studies, and
given the ecological preferences of molluscs, we be-
lieve changes in molluscan assemblages identify signif-
icant perturbations of an environment and filter out
“noise”? caused by day to day or short-term fluctua-
tions, such as those caused by tropical storms. By un-
derstanding the factors controlling modern distributions
of molluscs, we can interpret patterns of change in the
past, and in turn, predict future responses to human-
induced or natural environmental change.
200 BULLETIN
81/00i
PARK
BOUNDARY
25 J 00i
WATER DEPTH
10 Kilometers
81/00:
361
80) 30i
Sab
Allen 6A0c%> g 25 J 00i
a S
| + PISTON CORE SITES
_ © MODERN SAMPLING SITES
ATLANTIC OCEAN
80/30i
Text-figure 1—Map of Florida Bay showing location of 26 monitoring sites and four cores. Boundary of Everglades National Park is shown
by bold dashed line. Table 1 gives latitude and longitude.
ACKNOWLEDGMENTS
Our work has been done in collaboration with col-
leagues at a number of other agencies, including South
Florida Water Management District, National Oceanic
and Atmospheric Administration, and Everglades Na-
tional Park. We would like to thank them for their
cooperation and assistance in this investigation. We
would especially like to thank William Lyons, Florida
Marine Research Institute, for his invaluable and gen-
erous help with current molluscan taxonomy, and for
the many long discussions on the ecology and biology
of Florida Bay molluscs; we have benefited greatly
from his contributions. None of this work would have
been possible with out the assistance of William Gibbs
and the staff at the Keys Marine Laboratory, our base
of operations during fieldwork. We are indebted to our
colleagues Robert Halley and Eugene Shinn, U.S.
Geological Survey, St. Petersburg, FL, for collecting
the cores and making them available to us for analysis,
and to John Robbins (NOAA) for collaborating on the
development of the age model for the cores.
We would like to thank our reviewers Lucy Ed-
wards, U. S. Geological Survey, Reston, VA and John
Pojeta, U.S. Geological Survey and U.S. National
Museum of Natural History, Washington, DC for their
thoughtful reviews and suggestions to improve this
manuscript. Thomas Scott, and G. Harley Means, Flor-
ida Geological Survey, have provided assistance in the
field. The following U.S. Geological Survey person-
nel have assisted in field work and sample preparation:
Kristi Alger, Jill D’ Ambrosio, Alessandro Bagalia
(volunteer), Nancy Carlin, Carey Costello, Lauren
Hewitt, Sara Schwede, and Steve Wandrei. Sara
Schwede and Luke Blair, USGS, produced the GIS
salinity maps.
METHODS OF INVESTIGATION
FIELD METHODS
Twenty-six sites throughout Florida Bay (Text-fig.
1, Table 1), visited biannually between February 1995
and July 1999, constitute the source of modern obser-
vations. Biannual observation of sites always occurred
between February 12-23 and July 6—13 for seasonal
consistency. Sites 1-14, under regular observation
MOLLUSKS PAST AND PRESENT: BREWSTER-WINGARD ET AL. 201
Table 1.—Location of 26 modern sampling sites and 4 cores shown on Figure 1.
Site Number Site Name Latitude Longitude
1 Shell creek mouth N 25.20684 W 80.48701
2 Trout Creek water monitoring station N 25.21226 W 80.53345
3 Duck Key water monitoring station N 25.18012 W 80.48916
4 Northern Nest Key N 25.13104 W 80.50458
5) Porjoe Key S mud bank N 25.14121 W 80.47267
6 Butternut Key water monitoring station N 25.08677 W 80.51915
7 Bottle Key SE mud bank N 25.06092 W 80.55918
8 Little Madeira Bay mouth N 25.17580 W 80.63166
9 Mid basin S of Little Madeira Bay N 25.13433 W 80.62146
10 Mid basin (between Park & Russell Key) N 25.08494 W 80.58954
11 Park Key SW bank N 25.10628 W 80.56254
12 Russell Key SE bank N 25.07168 W 80.63814
13 Bob Allen Keys mud bank N 25.01892 W 80.66540
14 Bob Allen Key water monitoring station N 25.02647 W 80.68115
15 Gopher Pass N 24.98098 W 80.73686
16 Mud bank SE of Sid Key & SW of Corinne N 25.01227 W 80.76410
17 Little Rabbit Key water monitoring station N 24.98144 W 80.82515
18 Peterson Key bank water monitoring station N 24.91783 W 80.74662
19 Mid basin N of Barnes Key N 24.95348 W 80.78270
20 Pass Key S mud bank N 25.15314 W 80.57711
21 Buttonwood Keys water monitoring station N 25.07101 W 80.73588
22 Whipray basin mud bank N 25.08853 W 80.74182
23 Sprigger Bank N 24.9132] W 80.93768
24 Schooner Bank N 24.9597] W 80.97596
25 Cape Sable N 25.11344 W 81.07898
26 Johnson Key water monitoring station N 25.05240 W 80.90391
Core Number Date Collected
T24 Taylor Creek at mouth of Little Madeira Bay N 25.2000 W 80.6417
PK37 Pass Key - mud bank to south of Key N 25.1478 W 80.5745 5/26/96
RB19B Russell Bank to south of Russell Key N 25.0639 W 80.6248 2/24/95
BA6A Bob Allen mudbank, south east of Bob Allen Keys N 25.0232 W 80.6568 5/26/94
since February 1995, have been visited consistently
twice a year. Sites 15-26, added to the observations
after February 1995, were typically less accessible in
the ‘wet’ season (May—October) and as a result un-
derwent more regular observation during ‘dry’ seasons
(November—April) and less consistent observation dur-
ing ‘wet’ seasons. Observations were obtained by
wading or snorkeling at each site and recording the
substrate, fauna and flora present (living and dead),
and microhabitats. Observation procedures did not in-
clude digging within the sediment for infaunal organ-
isms or a systematic search of the site, but represent a
careful examination of the area. One hundred and sev-
enty-five detailed site observations from February
1996 to July 1999 are analyzed statistically in this pa-
per.
Each observation was coupled with a characteriza-
tion of the water at the site and a sampling of the
sediment. Water was characterized by salinity, tem-
perature, depth, and clarity. Salinity values were cal-
culated using a Hydrolab Surveyor 4, YSI Model 30/
10 FT, or a refractometer. Temperature values were
obtained using a Hydrolab Surveyor 4 or YSI Model
30/10 FT.
In February and July 1998, seventeen sub-aquatic
vegetation samples (macro-benthic algae and seagrass)
were collected at sites 8, 12, 13, and 20. Sub-aquatic
vegetation was sampled randomly, to characterize the
types of flora present, and the associated epifaunal or-
ganisms. Sub-aquatic vegetation was collected without
the roots or holdfasts (where present) to reduce the
chance of contamination by infaunal organisms. Veg-
etation samples were collected with a small amount of
concomitant water in an attempt to delay the effects
of biodegradation before laboratory examination of the
samples.
Sediment samples were collected using a 1.5 inch
diameter PVC tube (push core). Push cores were
pressed into the substrate, and a vacuum was main-
tained to extract the sediment. Push core tubes were
cut at the water-substrate interface and sealed to pre-
vent disruption of sediment during transport to the lab-
oratory. The twenty-eight push cores described in this
study were taken from four sites in northern and east-
202 BULLETIN 361
ern Florida Bay (Text-fig. 1) biannually between Feb-
ruary 1995 and July 1998 (collection at site 20 began
February 1997), and represent a subset of the total
number of push cores obtained. Push core samples
were chosen for this study based upon proximity of
modern sampling sites to the location of the four pis-
ton cores.
The four short piston cores (<2 m) examined were
taken from Florida Bay localities (Text-fig. 1) between
February 1994 and March 1996 by researchers from
the U. S. Geological Survey (St. Petersburg, Florida),
in cooperation with South Florida Water Management
District, Everglades National Park, and National Oce-
anic and Atmospheric Administration (NOAA). Mod-
ern sites 12, 13, and 20 occur at the same location as
piston cores RB19B, BA6A, and PK37, respectively.
Site 8, located at the mouth of Little Madeira Bay, is
the closest modern observation site to the site of piston
core T-24, located at the mouth of Taylor Creek in
Little Madeira Bay.
LABORATORY METHODS
Vegetation samples were characterized by type of
sub-aquatic vegetation present and condition of the
vegetation. All vegetation samples were then carefully
examined for extant epifaunal organisms before pro-
cessing. Samples were vigorously scrubbed to remove
epifaunal organisms, washed through 850 «m and 63
um sieves, and dried at 50°C. Dry vegetation material
from each sample was weighed, re-hydrated, and
washed through 850 jm and 63 jm sieves a second
time to help remove any remaining epifaunal organ-
isms.
All push cores were cut to 10 cm below the water-
substrate interface, and this upper 10 cm of sediment
was retained for processing. Twenty-eight push core
samples were analyzed for faunal content. Piston cores
collected for historical reconstruction were x-rayed
and visually examined for signs of sediment disrup-
tion, including bioturbation. Samples from the piston
cores were submitted for *!°Pb analysis (see Holmes,
et al., 2001, for method; Wingard eft al., 1995, and
Brewster-Wingard et al., 1997 for 7!°Pb results on
cores discussed here). All four piston cores were sam-
pled at 2-cm intervals. All push core and piston core
samples were washed through 850 wm and 63 pm
sieves to remove carbonate mud. Samples were dried
at 50°C.
All recognizable molluscs and molluscan fragments
were picked from the 850 m-size fraction from each
vegetation, push core, and piston core sample. Picked
specimens were sorted, identified, and characterized by
condition of the shell material present. Specimens
were characterized by stage (adult or juvenile) as well
as pristine (intact and retaining original shell material),
whole (intact but lacking original shell material or col-
or), broken (retaining original shell material and more
than 50% of shell), worn, or fragmented (retaining less
than 50% of the shell but still recognizable at the ge-
neric level).
Age models for the piston cores 6A and 19B are
based on ?!°Pb analysis of the <63 ,m-size fraction
of the samples. For details of the method, see Holmes,
et al. (this volume).
ANALYTICAL METHODS
Univariate and bivariate statistical analyses, includ-
ing minimum, maximum, mean, median, and standard
deviation, were conducted on environmental data (sa-
linity and temperature), and individual species distri-
butions using Microsoft Excel 97. For site observa-
tions, the presence-absence data were simplified for
statistical analysis by deleting samples with no mol-
luscs present and removing faunal groups that occur
less than two times throughout the dataset. Observa-
tion data were translated into a presence-absence data
matrix.'! For the piston core, push-core and vegetation
data, absolute species abundances were standardized
by calculating percent relative abundance of the total
molluscan component of the sample. Worn and frag-
mented molluscan specimens were removed from the
analysis of vegetation samples, and from one run of
push core samples, in order to isolate extant material
from older material.
All multivariate analyses were done using MVSP
(Kovach Computing Services, MVSP Plus, version
3.1) following the methods described by Kovach
(1989, 1995). A cluster analysis of the presence-ab-
sence data was conducted using unweighted paired
group method, average-linkage (UPGMA), with So-
renson’s Coefficient distance measurement, dual clus-
tering procedure, and random input order. Cluster anal-
yses of percent abundance data were conducted with
log-ratio transformed and centered data, then clustered
using UPGMA, with cosine theta distance measure-
ment, dual clustering procedure, and random input or-
der. Each cluster analysis was run several times using
random input order to determine robustness of clusters.
Constancy and fidelity measures were calculated for
each cluster in both the presence-absence data analysis
and the percent abundance data analysis. Method was
adapted from Hazel (1977). Constancy is a measure of
how frequently a species occurs within a given cluster,
expressed as a percentage. The formula is
' The full data matrix can be found at http://flaecohist/database/
Reference/synthesis.
MOLLUSKS PAST AND PRESENT: BREWSTER-WINGARD ET AL. 203
Total
Observations
20
10
60
35 Oe 040
e)
Oo
30 es ° ciet
e ry i> >
be Sali dues oe =
Y
g
B 5 A
oO
: oa
A
5 asia “aa
AA
20
: te
15 SSF =
0 5 10 15) 20 D> 30 35 40 45
Salinity (ppt)
Text-figure 2.—Plot of temperature (°C) versus salinity (ppt) recorded at sites from February 1996 through July 1999. Histogram on x axis
shows frequency of observations in 5 ppt increments. Histogram on y axis shows frequency of observations in 5°C increments. Bimodal nature
of temperature readings reflects seasonality.
Constancy = (Occurrence: 100)
+ Total number of samples in cluster.
A species with 100% constancy is present in every
sample in a given cluster. Fidelity is a measure of how
unique a species is to a given cluster (expressed as a
percentage). The formula is
Fidelity = [(Constancy in Cluster X)
+ (> constancy values for all
clusters) ]-100.
Thus, it is possible for a species to have a high con-
stancy value, if it is present in most samples in a clus-
ter, but a low fidelity value if it also is commonly
present in other clusters. Conversely, a species can be
very rare and have a low constancy for a cluster, but
a high fidelity because it occurs only in that cluster. A
species with 100% fidelity is unique to that cluster.
Several measures of diversity were calculated on all
piston core samples: 1) total number of individual mol-
luscan specimens; 2) faunal richness; 3) evenness; and
4) Shannon’s diversity index. Faunal richness in this
study is a measure of the number of faunal groups
present in a given sample (in the case of the presence-
absence data observed at a given site). In some cases,
the fauna are grouped into genera and occasionally
broader categories (e.g., marginellids), so this is not
“species”’ richness in the usual sense. Evenness and
Shannon’s diversity index were calculated using
MVSP. Evenness is a measure of how evenly dispersed
the total number of individuals are among the faunal
categories; the higher the value, the more evenly dis-
persed the individuals are. Shannon’s diversity index
(Shannon and Weaver, 1949; Ludwig and Reynolds,
1988) is a measure of the degree of uncertainty that a
randomly selected individual will belong to a certain
species. The formula is as follows:
204 BULLETIN 361
Table 2.—Univariate statistics for temperature (degrees C) at each site.
All years (February and July)
Mini- Maxi- Standard
Site 2-96 7-96 2-97 7-97 2-98 7-98 2-99 7-99 mum mum Mean Median deviation
1 19.00 30.40 23.00 30.50 21.00 31.30 20.62 29.16 19.00 31.30 25.62 26.08 Bysil@)
2 19.70 30.80 23.70 31.40 22.00 31.47 20.41 28.80 19.70 31.47 26.04 26.25 5.10
3 17.20 31.40 23.80 32.80 21.40 32.00 20.45 29.18 17.20 32.80 26.03 26.49 6.04
4 19.80 31.20 24.60 32.70 21.60 32.88 21.46 29.61 19.80 32.88 26.73 ALT 5.46
5 16.50 31.60 25.50 32.90 21.90 33.42 22.09 30.37 16.50 33.42 26.79 27.94 6.22
19.70 29.60 24.40 32.40 21.00 33.29 21.68 30.49 19.70 33.29 26.57 27.00 5.48
33.59 27.09 27.95 5.95
19.20 31.30 24.70 31.60 20.00 30.86 19.96 29.52 19.20 31.60 25.89 27 ull S05)
17.90 31.00 24.10 30.40 19.40 Siig? 19.96 28.31 17.90 31.32 25.30 26.21 5.65
10 17.50 31.40 24.20 31.30 18.40 31.94 19.96 28.39 17.50 31.94 25.39 26.30 6.15
11 19.80 30.50 23.60 33.30 19.40 32.99 20.19 30.53 19.40 33.30 26.29 27.05 6.14
12 18.50 31.50 24.50 30.10 21.60 34.58 21.60 28.95 18.50 34.58 26.42 26.73 5.67
13 19.80 31.30 24.60 30.80 22.20 32.10 22.37 30.90 19.80 32.10 26.76 27.70 5.01
14 19.40 31.30 26.00 31.70 23.80 33.01 22.92 34.10 19.40 34.10 27.78 28.65 5.45
6
7 19.40 30.80 25.10 32.70 20.70 33.59 21.89 32792 19.40
8
9
15 17.40 28.80 31.70 28.54 22.86 28.73 17.40 31.70 26.34 28.64 5.24
16 17.60 30.40 28.50 31.70 26.30 29.38 23.54 29.99 17.60 31.70 27.18 28.94 4.65
17 17.60 32.30 30.80 25.90 28.42 25.05 17.60 32.30 26.68 27.16 5.24
18 17.70 32.50 26.40 29.90 24.90 29.25 22.58 31.30 17.70 32.50 26.82 27.83 4.97
20 24.60 29.90 19.90 32.02 23.14 28.89 19.90 32.02 26.41 26.75 4.61
21 31.60 24.00 30.34 24.20 29.40 24.00 31.60 27.91 29.40 3.56
22, 32.60 24.90 30.42 23.90 29.72 23.90 32.60 28.31 29012) 3.74
23 31.00 23.50 31.94 21.47 29.14 21.47 31.94 27.41 29.14 4.66
24 31.20 23.90 32.08 21.54 29.28 21.54 32.08 27.60 29.28 4.64
25 32.70 24.90 32.42 30.70 24.90 32.70 30.18 31.56 3.63
26 34.20 27.60 33.26 23.82 29.28 23.82 34.20 29.63 29.28 4.24
Table 3.—Univariate statistics for salinity (ppt) at each site.
All years (February and July)
Mini- Maxi- Standard
Site 2-96 7-96 2-97 7-97 2-98 7-98 2-99 7-99 mum mum Mean Median deviation
1 18.30 14.10 18.90 15.40 15.80 26.90 28.16 35.48 14.10 35.48 21.63 18.60 7.66
2 13.00 2.70 22.60 1.40 13.00 24.61 25.24 12.70 1.40 25.24 14.41 13.00 Oy 7H/
3 18.50 19.60 25.10 18.80 17.10 29.10 29.27 37.12 17.10 SZ 24.32 22.35 7.08
4 17.50 17.40 24.50 21.40 16.70 32.15 27.81 36.47 16.70 36.47 24.24 PUPS) a-39,
5) 20.30 21.80 25.40 21.30 17.60 31.94 2.9)35) 36.84 17.60 36.84 AS\3)I 23.60 6.61
6 18.50 26.60 24.60 28.80 20.00 35.94 28.87 35.81 18.50 35.94 239) 27.70 6.44
7 22.10 26.40 24.60 30.50 20.60 37.43 30.47 36.92 20.60 37.43 28.63 28.44 6.34
8 15.80 10.20 19.90 10.90 15.20 23.88 19.38 31.31 10.20 sulesiil 18.32 1 7s)2) 6.97
9 18.10 21.10 21.50 19.20 18.40 28.80 24.06 33.58 18.10 33.58 23.09 21.30 5250)
10 18.10 24.60 21.70 22.30 20.50 35.47 26.86 34.70 18.10 35.47 2S):5)5) 23.45 6.45
11 18.70 24.00 21.50 22.80 19.30 S3ql3) 27.04 34.49 18.70 34.49 25.12 23.40 5.98
12 18.90 23.40 23.30 26.20 20.70 36.97 28.61 29.00 18.90 36.97 25.89 24.80 5.72
13 27.20 32.90 28.70 31.20 26.70 40.51 Ssehll7/ 29.80 26.70 40.51 SE2 7 30.50 4.44
14 28 33.4 28.9 29.2 41.03 32.94 30.1 28.00 41.03 31.94 30.10 4.50
15 31.30 31.30 33.30 26.90 36.75 34.62 40.00 26.90 40.00 33.45 33.30 4.23
16 31.40 34.50 29.60 33.20 30.10 38.73 35.34 36.85 29.60 38.73 8357/2, 33.85 3.25
17 30.70 34.40 33.30 31.70 36.25 32.54 30.70 36.25 33.15 32.92 1.98
18 31.30 35.00 35.10 35.30 29.70 37.15 35.76 29.00 29.00 37.15 33.54 35.05 3.07
20 22.60 19.80 17.10 28.15 24.97 33.39 17.10 33.39 24.34 23.79 5.88
21 25.60 28.90 39.34 34.72 39.71 25.60 SOM 33.65 34.72 6.28
22 24.40 30.20 39.93 35.11 41.30 24.40 41.30 34.19 35.11 7.00
23 34.30 33.00 36.76 34.04 38.39 33.00 38.39 35.30 34.30 2.21
24 34.30 31.90 35.93 35.13 39.00 31.90 39.00 35.25 35013 2.58
25 28.30 29.50 36.07 37.00 28.30 37.00 32.72 32.79 4.45
26 30.30 31.60 35.43 33°51 37.00 30.30 37.00 33.57 33.51 Pas
MOLLUSKS PAST AND PRESENT: BREWSTER-WINGARD ET AL. 205
Table 2.—Extended.
February (all years) July (all years) Number of
Mini- Maxi- Standard Mini- Maxi- Seam CESe Neen
mum mum Mean Median deviation mum mum Mean Median deviation Feb. July
19.00 23.00 20.91 20.81 1.64 29.16 31.30 30.34 30.45 0.88 4 4
19.70 23.70 21.45 21.21 1.78 28.80 31.47 30.62 31.10 1:25) 4 4
17.20 23.80 20.71 20.93 Af} 29.18 32.80 B1k35) 31.70 ESS 4 4
19.80 24.60 21.87 21.53 2.00 29.61 32.88 31.60 31.95 152) 4 4
16.50 25.50 21.50 22.00 3.72 30.37 33.42 32.07 32.25 1.37 4 4
19.70 24.40 21.70 21.34 1.98 29.60 33.29 31.45 31.45 1.70 4 4
19.40 25.10 21.77 21.30 2.44 30.80 83159) 32.40 32.61 ici / 4 4
19.20 24.70 20.97 19.98 2.52 29.52 31.60 30.82 31.08 0.92 4 4
17.90 24.10 20.34 19.68 2.65 28.31 Bir32 30.26 30.70 1.35 4 4
17.50 24.20 20.02 19.18 2.97 28.39 31.94 30.76 31.35 1.60 4 4
19.40 23.60 20.75 20.00 1.93 30.50 33.30 31.83 31.76 E52, 4 4
18.50 24.50 21.55 21.60 2.45 28.95 34.58 31.28 30.80 2.43 4 4
19.80 24.60 22.24 22.29 1.96 30.80 32.10 31.28 31.10 0.59 4 4
19.40 26.00 23.03 23.36 2.74 31.30 34.10 32.53 32.36 1.28 4 4
17.40 22.86 20.13 20.13 3.86 28.54 31.70 29.44 28.77 1.51 2 4
17.60 28.50 23.99 24.92 4.72 29.38 31.70 30.37 30.20 0.98 4 4
17.60 25.90 22.85 25.05 4.57 28.42 32.30 30.51 30.80 1.96 3} 3
17.70 26.40 22.90 23.74 3.80 29.25 32.50 30.74 30.60 1.45 4 4
19.90 24.60 225) 23.14 2.41 28.89 32.02 30.27 29.90 1.60 3 3
24.00 24.20 24.10 24.10 0.14 29.40 31.60 30.45 30.34 1.10 2 3
23.90 24.90 24.40 24.40 0.71 29.72 32.60 30.91 30.42 1.50 2 3
21.47 23.50 22.49 22.49 1.44 29.14 31.94 30.69 31.00 1.42 2 3
21.54 23.90 22.72 22.72 1.67 29.28 32.08 30.85 31.20 1.43 2 3
24.90 24.90 =. — — 30.70 32.70 31.94 32.42 1.08 1 3
23.82 27.60 25.71 25.71 2.67 29.28 34.20 82525) 33.26 2.61 2 3
Table 3.—Extended.
February (all years) July (all years) Number of
Mini- Maxi- Standard Mini- Maxi- Stamtierds me cesce waCHS
mum mum Mean Median deviation mum mum Mean Median deviation Feb. July
15.80 28.16 20.29 18.60 5.42 14.10 35.48 22.97 2S 10.13 4 4
13.00 25.24 18.46 17.80 6.40 1.40 24.61 10.35 7.70 10.76 4 4
17.10 29.27 22.49 21.80 Sy7/i 18.80 SNP 26.16 24.35 8.68 4 4
16.70 27.81 21.63 21.00 5.41 17.40 36.47 26.86 26.78 8.94 4 4
17.60 29.35 23.16 22.85 5.24 21.30 36.84 27.97 26.87 7.68 4 4
18.50 28.87 22.99 22.30 4.70 26.60 35.94 31.79 32.31 4.80 4 4
20.60 30.47 24.44 23.35 4.34 26.40 37.43 32.81 33.71 Soil 4 4
15.20 19.90 Si 17.59 2.41 10.20 31.31 19.07 17.39 10.30 4 4
18.10 24.06 20.52 19.95 2.82 19.20 33.58 25.67 24.95 6.71 4 4
18.10 26.86 21.79 21.10 3-710) 22.30 35.47 29.27 29.65 6.79 4 4
18.70 27.04 21.64 20.40 3.80 22.80 34.49 28.61 28.57 6.06 4 4
18.90 28.61 22.88 22.00 4.23 23.40 36.97 28.89 27.60 5.85 4 4
26.70 33.17 28.94 27.95 2.94 29.80 40.51 33.60 32.05 4.78 4 4
28.00 32.94 29.95 28.90 2.63 29.20 41.03 33.43 31.75 5.38 3 4
26.90 34.62 30.94 31.30 3.87 31.30 40.00 35.34 35.03 3.84 3 4
29.60 35.34 31.61 30.75 2.60 33.20 38.73 35.82 35.68 2.46 4 4
30.70 32.54 31.65 31.70 0.92 33.30 36.25 34.65 34.40 1.49 3 3
29.70 35.76 32.97 33.20 2.93 29.00 S77 34.11 35.15 3.54 4 4
17.10 24.97 21.56 22.60 4.04 19.80 33.39 2m! 28.15 6.85 3 3
28.90 34.72 31.81 31.81 4.12 25.60 39.71 34.88 39.34 8.04 2 3
30.20 35.11 32.66 32.66 3.47 24.40 41.30 35.21 39.93 9.39 2 3
33.00 34.04 33.52 S852 0.74 34.30 38.39 36.48 36.76 2.06 2 3
31.90 85513) 33.52 33.52 2.28 34.30 39.00 36.41 35.93 2.39 2 3
29.50 29.50 — — = 28.30 37.00 33.79 36.07 4.78 1 3
31.60 BStor 32.56 32.56 1.35 30.30 37.00 34.24 35.43 3.50 2D, 3
206 BULLETIN 361
‘Salinity
| Range (ppt) | Salinity
| i 10-15 Range (ppt)
15-20 i 15-20
GH 20-25 i 20-25
HB 25-30 GH 25-30
WB 30-35 | WH 30-35
February 1996 February 1997
Se aoa :
Salinity | | ‘ange (ppt)
| Range (ppt) | = os
Sos | 5-10
5-0 | @ 10-45
| (i 10-15 | 15-20
| GB 15-20 | 20-25
| [20-25 | HM 25-30
GH 25-30 Gi 30-35
| MB 30-35 | MM 35-40
July 1996 July 1997
Text-figure 3—Maps showing 5 ppt salinity gradients and the prominent cluster designations from the presence-absence Q-mode analysis
for the sampling periods from February 1996 through July 1999. Contours are primarily based on our own salinity measurements, however,
in areas of sparse sampling we utilized data from Halley (http://sofia.usgs.gov/projects/circulation) to fill in the gaps.
bay-wide salinity is less variable in the winter (dry)
season than during the summer (wet) season, but the
Opposite is true for water temperature. Recorded win-
H' = = p,log Pi
i=1
where s is the number of species of know proportions ter water temperatures have varied more than 10°C
(P|, Pr, Ps - +++ P,) (see Ludwig and Reynolds, 1988, from one year to the next, ranging from a low of
p. 92, for complete explanation). 16.5°C (site 5, 2/96) to a high of 28.5°C (site 16, 2/
RESULTS ON). Summer water temperatures have ranged from a
low of 28.3°C (site 9, 7/99) to a high of 34.6°C (site
MODERN ENVIRONMENTAL DATA 12, 7/98). Seasonal standard deviations for the period
Text-figure 2 shows seasonal and annual variations of record are relatively low for water temperatures at
recorded in temperature and salinity at our monitoring each site, and water temperatures do not display a
sites in Florida Bay. These measurements show that strong geographic gradient (Table 2).
MOLLUSKS PAST AND PRESENT: BREWSTER-WINGARD ET AL. 207
Salinity
Range (ppt)
February 19
FD tp,
4
Salinity
| Range (ppt)
20-25
GB 25-30
GM 30-35
MM 35-40
WB 40+
July 1998
Salinity
| Range (ppt)
Salinity
Range (ppt) |
July 1999 |
Text-figure 3.—Continued.
In contrast to temperature, salinity does display a
strong geographic gradient. Text-figure 3 illustrates
seasonal changes in salinity and Table 3 lists standard
deviations calculated for each site. In general, standard
deviations for salinity decrease as you move to the
south and west in the bay, indicating more stable sa-
linity on a seasonal and annual basis. Summer salinity
standard deviation values are significantly higher at
most sites than the winter values. Salinity at three sites
(1, 2, and 8) in the northern transition zone has varied
over 20 ppt during the July sampling period from
1996-1999, compared to winter variations at the same
sites of 5-12 ppt. Bay-wide salinity has ranged from
a low of 1.4 ppt (site 2, 7/97) to a high of 41.3 ppt
(site 22, 7/99).
MODERN MOLLUSCAN PRESENCE-ABSENCE DATA
An unweighted pair-group method (UPGM) dual-
cluster analysis, using Sorenson’s coefficient on the
modern presence-absence faunal dataset (rares delet-
ed), produced the clusters seen in Text-figures 4 and
5. The dual clustering method (Kovach, 1989, 1995)
allows direct comparison between the separate Q-
mode and R-mode analyses. Q-mode analysis pro-
duced seven primary clusters, grouping the modern
sample sites based on species found alive at those
208 BULLETIN 361
0 0.2 0.4
0.6 0.8 1
Sorenson's Coefficient Distance Measure
Text-figure 4.—Q-mode cluster of presence-absence data from modern monitoring sites. Cluster was generated through MVSP statistical
package (Kovach Computing Services, MVSP Plus, version 3.1) using unweighted paired group method, average-linkage (UPGMA), with
Sorenson’s Coefficient distance measurement, dual clustering procedure, and random input order. Full data matrix sorted in dendrogram order
and labeled clusters can be viewed at http://flaecohist/database/Reference/synthesis. Table 4 indicates which species are responsible for the
clusters seen.
sites. An examination of the clusters and the associated
data matrix, sorted in dendrogram order, indicated
which fauna are responsible for the clustering of sam-
ple sites seen. Text-figure 3 shows the distribution of
the clusters by sampling period.
Cluster A groups samples from sites 6, 7, 11, 12
and 14 in eastern and central Florida Bay. The occur-
rence of Batillaria minima and/or Melongena corona
defines this cluster; constancy (C) and fidelity (F) for
Batillaria minima equal 83.3 and 80.0 and for Melon-
gena corona 50.0 and 70.3 (Table 4). Fasciolaria spp.
and Truncatella spp. also have high fidelity values for
this cluster (Table 4). Sites in cluster A can be char-
acterized as having a subenvironment with shallow
MOoLLusksS PAST AND PRESENT: BREWSTER-WINGARD ET AL. 209
Il
0.68 0.84 1
0.04 0.2 0.36 0.52
Muricidae sp.
Arcopsis adamsi
Ostrea equestris
— Cerithid
Fasciolaria spp.
Batillaria minima
Chione cancellata
Diodora sp.
Cerithium spp.
Carditamera floridana
Argopecten irradians
Modulus modulus
Cerithium muscarum
- Prunum sp.
- Bittiolum varium
Crepidula spp.
Brachidontes exustus
- Pteria longisquamosa
Lucina spp.
Tellinid
Laevicardium mortoni
Marginellid
Bulla striata
Limaria sp. cf. L. pellucida
Truncatella spp.
Melongena corona
Acteocina canaliculata
Parastarte triquetra
Turridae sp. A
Caecum pulchellum
Transennella sp.
Anomalocardia auberiana
— Lithopoma americanum
= Columbella spp.
Melampus coffeus
= Busycotypus spp.
Pleuroploca gigantea
— Pinnidae
Sorenson's Coefficient Distance Measurement
Text-figure 5—R-mode cluster of presence-absence data from modern monitoring sites. Cluster analysis and full data matrix as in Text-
figure 4.
(<20 cm) water depth in relatively close proximity to
mangrove islands, sparse vegetation (<20% Halodule,
scattered Thalassia may be present), and soft, “gelat-
inous”’ calcareous mud is typical (799-6 is an excep-
tion). Average salinity for the cluster is 28.4 ppt, and
the range is 18.7—41.0 ppt. Average observed faunal
richness? at sample sites in cluster A is three, with a
maximum of six faunal groups seen alive during one
observation period.
? Faunal richness for the modern presence-absence database is a
measure of the number of faunal groups observed alive at a site
during one field season. The fauna are frequently grouped into gen-
era and occasionally into broader categories (e.g., Marginellids), so
Cluster B is defined by the presence of Ostrea
equestris (C = 76.9, F = 70.5) and/or Arcopsis adamsi
(C = 46.2, F = 87.6), with Brachidontes exustus pre-
sent in nine of the thirteen samples in this cluster (Ta-
ble 4). The sample sites (1, 2, 3, 8, 9, and 24) are
geographically widespread, encompassing the northern
transitional zone, eastern, and western portions of
this is not “species” richness in the usual sense. In addition, the
observation and recording process is definitely biased towards epi-
faunal molluscs, and so this is not meant to be a measure of the true
richness of the environment. However, because the same categories
were used at all sites and similar methods, we feel it is a basis for
comparing the faunal richness from one site and one observation
period to another.
BULLETIN 361
210
(snoru
=) = One I SI IS A OD VIE iS DOU VEL I fe) Me “Uul]) snjnpoi snjnpow
—- — — woe 9¢ I — 3869 6ZIl 6 —- —- — — — — &eg wnivosnu wuniyiuad 9
rt ae Ae Ea Ri ORS I = eat = — =] Wine OWS We ae AS te LEVIED nC ‘ds wnuntg
(Jay
= a Se) SOON OS: ae we Se ef AS I NOS wits (5 iwWilfe isi C —_- —- ~~ ~ sed) wniiva unjouyg DO
Lei Ol0 Caan. VET 68E L TCL O07 I GM Get FS WAG Wiis Ce (tell Iie 18 os €8 I ‘dds nynpidaid
(snaeuurq)
Se E18) OCS, MAY, VSt OOF G CS il iv SS (Ss I) te, reteh8) 0) ey OEE SnISNXA SowuOpIyIDAT
(4oyunqd)
Ee ORE OES I LOL 002 I Sy GNSS GUE MS NS iy EY, I (sie HEINE, DSOUPNDSISUO] Viald ey
O;001 OOF (é — ‘dds puionT qd
££6 OOF c meee ee II OTE c PruyleL d
(peiuo0))
css OOO! § en ee on /Le) MOLT} 9 oe ee Ge Dl. tee I IUOJLOW LUNIPADIIABVDT df
89 009 oye fSt3 IS OL iE ¢ ao) 9 ea SOE I pypeuisieyy
er ANGIE OMS I CLe O07 I Sibi AS I Oe OG © arginsnig VIDLUS DING OD
(suepy “€’D) »P
= os eee OSS Pan OES) I = —- eet HOS YEAS I OT 67T (6 -1onyjad “7 ‘jo-ds vpiuvuryT gd
EST Ore Cc Lop LL I [ty €8 I ‘dds vjjaipounty
(uly
2 aii es vie Ole = oa = €0L OOS 9 -aWD) pUOIOD DUasUuojJaW OD
(Aes)
OOOL 67 G DIDINIYoOUYI YUIIOAIIyY =D
(peiu0D)
OOOL Ls b DAJNDLY AADISDIVY
OOOL 67 c vy ds ovprun, 9
uosduins
ig Cl ONG t (1b, (eth I —- —> — wnjjayajnd unzvavy 9
a ae ema (Gil7é BMS I OVSIE Wye ¢ EES LU. I -- — a ‘ds pyjauuasupty
(AusiqiO.P)
oe c a OLS aC 6¢r O0T I = —— — Fra JES t (sonlnt fll I — — — pupiiaqnyd pipipsojouIoUy qi
(uljatu)
Ve ONO Sol My T= Sere ae) po Yee I SiG Val I unupoiaup puodoyyntyT
Ore O0T I 0199 68 LZ (u10g) MIMIOSP{ DjNBAaL OD
cOc =O00T I fos OOS 6 aa a SE ie Stall IS GC 61 C th = YE7E I oe a! ee Ul[EUID VeUuDjsDD Ogun],
ee ES Le L OS Se 0) ee ae Oe, Eva Cc Ocl 98 ©) ‘dds pyjaqunjoD 9
(snaeu
OOO! FL Tc -urq) snaffoo sndunjaw
Tvl O0¢ I 90t OS I es vl I ‘dds sndtjootsng 9
(s9uary)
OQOOL OOF TC paquns1s8 poojdoinajgd =D
TO08 O09 ¢€ (Sil WE ve ata: me) ONS LAS I sepiuuld d II
d 2) O fa 9) O al o) O A e) O a e) O a 7) O A o) O dnoipH yeune{ Sse J3sn[D
apoul-y
9) A el d o) d Vv
Jasn[Q e2pow-—
‘epodonsey = 5 ‘epodAoajag = d :UONRUSISEp sse[D Jepso
WieISOIpUap IpPpoul-yY UI PoLIOS ome eXEL, “SIsATeUR JO}SNID aduasqe-douasaid ay} WIZ Ja}SN{O YORA UT UOXE] YOR JOJ sanyRA (4) AWepy pue “(D) Aouvysuos (QO) 2:ouUeLNIIO— fF IIGBL
Table 4.—Continued.
Q-mode Cluster
G
R-mode
Faunal Group
Cluster Class
P Argopecten irradians
la
(cont'd)
(Lamarck)
P Carditamera floridana
33.3 66.9
6
14.9
7.4
a
Conrad
G_ Cerithium spp.
G
9
3
20.0 42.1
|
7.8
Diodora sp.
MOLLUSKS PAST AND PRESENT: BREWSTER-WINGARD ET AL.
3
P Chione cancellata (Lin-
naeus)
G_ Batillaria minima (Gme-
3.6
Bo)
83.3 80.0 —
10
lin)
G_ Fasciolaria spp.
G_ Cerithid
16.7
3
41.7 48.0
5
1 20.0
3.4
Bu),
10
Ostrea equestris Say
P
7.0
P Arcopsis adamsi
G_ Muricidae sp.
Outlier
18
70
Total number of samples/
cluster
Florida Bay. The common characteristic of these sites
is the presence of wood (mangrove roots, driftwood,
or pilings) or exposed limestone bedrock that provides
an attachment-point for the cemented Ostrea equestris,
or bysally attached Arcopsis adamsi. Salinity values
for the cluster range from 2.7—35.1 ppt, and the av-
erage is 18.6 ppt. A maximum of seven faunal groups
were found alive at sites in cluster B, and the average
observed faunal richness is three.
Cluster C comprises the largest group of samples
(70) from sites 1-17, and 20—22 in the northern tran-
sitional, eastern and central portions of Florida Bay.
The nearly ubiquitous occurrence of Brachidontes ex-
ustus (66 of our 70 samples, C = 94.3, F = 35.5)
distinguishes this cluster (Table 4). Within the larger
cluster C are six smaller groupings, characterized by
species that co-occur with Brachidontes exustus. Av-
erage salinity for cluster C is 28.1 ppt, and the range
is 10.9—41.3 ppt. The environments represented by the
20 sites in this cluster are varied, including mud banks,
mangrove islands, open basins, and channels, but all
have at least some subaquatic vegetation present. Sites
within cluster C attain the highest values for observed
faunal richness; the average is five and the maximum
value is 20.
Cluster D includes samples from eastern, central,
western, and Atlantic transition zone sites (3, 5—7, 10,
11, 13, 15-18, 20, 21, and 26). The occurrence of
Pteria longisquamosa |=Pinctada radiata Turney and
Perkins] (C = 85.2, F = 45.7) distinguishes cluster D;
and the occurrence of Modulus modulus (C = 51.9, F
= 71.8) is significant (Table 4). The environments rep-
resented by these samples can generally be character-
ized by the presence of healthy Thalassia beds (ex-
ceptions, 6, 10, and 26). Average salinity for the sam-
ples is 29.9 ppt and the range is 19.8—39.7 ppt. Ob-
served faunal richness at sites in cluster D averages
four, and the maximum value is eleven.
A small group of five samples from the northern
transitional zone, and eastern and central Florida Bay
(sites 2, 3, 9, and 21) constitute Cluster E. The pres-
ence of Laevicardium mortoni (C = 100, F = 85.5)
distinguishes this cluster (Table 4). All of the sites
have a subenvironment with exposed substrate and
sparse vegetation. The average salinity for the cluster
is 24.0 ppt and the range is 12.7—34.7 ppt. Observed
faunal richness ranges from 2—6, and the average is 4.
Cluster F includes two distinct sub-clusters, one dis-
tinguished by the co-occurrence of Turbo castanea (C
= 50, F = 50.5), Tegula fasciata (C = 38.9, F = 66)
and Columbella spp. (C = 55.6, F = 77.7), the other
by Columbella spp., Cerithium spp. (C 50, F =
64.4), Chione cancellata (C = 27.8, F 52.2) and
Carditamera floridana (C = 33.3, F = 66.9) (Table
PIMP BULLETIN 361
4). The cluster includes sites from the central, western,
Gulf transition, and Atlantic transition zones of Florida
Bay (13, 17, 18, 21, 23, 24, and 26). The sites can be
characterized as having lush subaquatic vegetation and
bare sediment areas available. Average salinity for
cluster F is 34.6 ppt, and the range is 29.0—39.3 ppt.
Average observed faunal richness is four and the max-
imum value is eight for the sites in cluster F
Cluster G comprises five samples from the western-
most sites (23 and 24) in the Gulf transition zone. Pin-
nidae (C = 60, F = 80.2), and Pleuroploca gigantea
(C = 40, F = 100) define this cluster (Table 4). Av-
erage salinity is 34.1 ppt, and the range is 31.9 to 36.8
ppt. These sites are predominantly calcareous sand
bars, with lush beds of mixed subaquatic vegetation
present, including Thalassia, Halodule, Syringodium,
and many species of calcareous algae, and macro-ben-
thic red, green and brown algae. The sites at cluster G
have the lowest observed faunal richness, with an av-
erage of two and a maximum of five faunal groups
found alive at the sites.
R-mode analysis produced two primary divisions of
the faunal categories, based on the samples in which
they are found (Text-fig. 5). In general, fauna in cluster
I tolerate a wide range of conditions and can be found
in the northern transitional, eastern, central and west-
ern portions of the bay. Fauna in cluster I were found
alive in salinities ranging from 2.7—41.3 ppt. The re-
mainder of the faunal groups fall into cluster I, with
the exception of Muricidae sp., a rare species that is
an outlier on the dendrogram. Cluster II groups fauna
that are typically found in the western portions of the
Bay, or at sites with near normal marine salinities. The
average salinity recorded at sites where fauna from
cluster II are found is 33.3 ppt and the range is 17.6—
40.0 ppt. The exception in cluster Il is Melampus cof-
feus. The presence of Melampus is indicative only of
the presence of a supratidal environment and the pres-
ence of mangroves.
MODERN VEGETATION SAMPLE DATA
Sixteen vegetation samples and one surficial mud
sample were analyzed from sites 8, 12, 13, and 20,
collected in February and July 1998.! Brachidontes ex-
ustus and Bittiolum varium account for 95% (10,981
individuals) of the 35 molluscan faunal groups found
alive in the samples from all four sites. These two
species were most frequently found in polytypic sub-
aquatic vegetation samples, including Thalassia and
macro-benthic algae. Crepidula spp., Cerithium mus-
carum, Pteria longisquamosa and Prunum spp. com-
bined constitute only 2.25% (260 individuals) of the
living fauna collected.
MODERN PUSH CoRE DATA
Twenty-eight push core samples were analyzed from
sites 8, 12, 13, and 20, collected in February and July
beginning in 1995.' The push core samples represent
death assemblages at each site, and thus the push core
data provide a link between modern living assemblage
data and down-core assemblages from the core data
set. Two cluster analyses were done (Text-fig. 6); one
included all specimens, regardless of preservation; the
other was limited to only pristine and broken shells.
The purpose was to contrast the death assemblage at
each site with the living or recently expired fauna. The
two clusters produced similar results. All samples from
site 8 formed a distinct cluster in both analyses. In
general, samples tended to cluster with other samples
from the same site in both analyses.
Brachidontes exustus and Bittiolum varium are
among the four most abundant species in the push
cores at all four sites (8, 12, 13, and 20), based on the
cumulative data. When the individual push core data
are examined for each site, Bittiolum varium is among
the four most abundant species and Brachidontes ex-
ustus 1S among the six most abundant species. Tran-
sennella sp. is among the four most abundant species
at sites 12, 13, and 20, and Crepidula spp. at sites 12
and 13. The discrete clustering of site 8 can be ex-
plained by the prominence of Hydrobiidae, Truncatel-
la bilabiata, and Anomalocardia auberiana.
HISTORICAL PISTON CORE DATA
Details of the individual core analyses have been
previously published (Brewster-Wingard ef al., 1997,
1998a, 1998b; Brewster-Wingard and Ishman, 1999;
Ishman ef al., 1996; Wingard et al., 1995).> Text-fig-
ures 7—10 illustrate the down-core distribution of key
species identified in the modern presence-absence data
analysis and in previous core analyses. Shannon’s di-
versity index, faunal evenness, the number of faunal
groups, and the number of specimens present are also
shown on the down-core plots (Text-figs. 7-10). The
age model for cores 6A and 19B 1s based on sedimen-
tation rates calculated from 2!°Pb analysis. The rate for
core 6A is 0.75 cm/yr + 0.08 and for core 19B is 1.27
cm/yr + 0.08 (Holmes, et al., this volume).
COMPILED DATA SET
Percent abundance data were compiled from the
modern vegetation samples and push cores from sites
8, 12, 13, and 20, and from the corresponding piston
cores. The compiled database was analyzed using the
unweighted pair-group method (UPGM) dual-cluster
‘ All open-file reports and data tables for each core are available
on line at http://sofia.usgs.gov/flaecohist.
MOLLUSKS PAST AND PRESENT: BREWSTER-WINGARD ET AL.
tO
—
Ww
ER 795-8A
798-8A
796-84 |
298-8A
297-8A
296-8A
797-8A |
295-8A
= — 795-128 |
797-20A
798-20A
298-20A
796-12B
798-12B
797-12B
298-12B
297-12B
296-12B
Push Cores
Pristine and Broken
Components Counted
12 1 0.8 0.6
Cosine Theta Distance Measurement
298-13C
297-20A
796-13C
797-13C
297-13C
296-13C
295-13C
798-13C
795-13B
295-12B
04 0.2 0
795-8A
798-8A
298-8A
796-8A
797-8A
297-8A
296-8A
295-8A
797-20A
298-13C
797-13C
297-13C
296-13C
796-13C
795-13B
Push Cores
All Identifiable
Components Counted
12 1 0.8 0.6
Cosine Theta Distance Measurement
295-13C
797-12B
mm) 798-128
my) 298-128
} 297-20A
297-12B
Se) 296-128
" 295-128
0.4 0.2 0
Text-figure 6—Q-mode cluster of push-core data from modern monitoring sites. Top cluster illustrates the results when only the pristine
and broken (retaining original material and more than 50% of shell) material are included. Bottom cluster includes all identifiable molluscan
components. Clusters were generated using MVSP statistical package (Kovach Computing Services, MVSP Plus, version 3.1) on log-ratio
transformed and centered percent abundance data, using unweighted paired group method, average-linkage (UPGMA), with cosine theta
distance measurement, dual clustering procedure, and random input order.
procedure, with cosine theta distance measures on a
log-transformed data matrix (Text-figs. 11 and 12). Q-
mode analysis produced three main clusters, grouping
the individual samples based on the faun they contain.
Faunal groups responsible for the Q-mode clusters
were determined by an examination of the associated
data matrix, sorted in dendrogram order.*
Cluster J is divided into two very distinct compo-
+See footnote 1, p. 202.
214 BULLETIN 361
Taylor Creek 24 Hydrobiidae Anomalocardia RXIc Group Transennella sp. Br Crepidula spp. Cerithium Bittiolumt variant Evenness No. of specimens
Q-Mode Cluster auberiana muscarum 0 09 coc ASS
; = £ =
]
|
Evenness
G
c 40 = Taal i onge
o
(
60 — L & =
—
z
E
80 = 2
g
2
Index B
86 ~
60 0 30 60 0 15 0 15 0 15 0 15 0 15 0 1 2 0 10 20
Percent abundance Shannon's Diversity Faunal Richness
Index
Text-figure 7—Down-core changes in core T24 from the mouth
mode clusters (see Text-fig. 11), significant molluscan fauna (show
of Taylor Creek in Little Madeira Bay. Columns illustrate changes in Q-
n as percent abundance), Shannon’s diversity index, evenness, number of
faunal groups, and number of individual specimens. RXIc group refers to the R-mode cluster shown in Text-figure 12; this group is composed
primarily of species that prefer salinities less than 18 ppt. Depth is
nents (Text-fig. 11). Cluster JI is primarily composed
of modern push core samples, the majority of these
from site 8. Cluster J2 is exclusively composed of
samples from core T24. A diverse faunal assemblage,
blending components of R-mode clusters XIa and XIc
(Text-fig. 12) defines cluster J. Anomalocardia auber-
iana (C = 93.4, F = 65.4), Acteocina canaliculata (C
shown in cm.
= 93.4, F = 69), and Hydrobiidae (C = 85.2, F =
92.9) are the dominant fauna in cluster J, and Poly-
mesoda maritima, and Cerithidea spp. are character-
istic (Table 5). Cluster J1 is further distinguished by a
group of species from R-mode cluster XII (Odostomia
spp.. Prunum sp., Schwartziella catesbyana, Bittiolum
varium juy., and Truncatella spp.). A number of faunal
No, of specimens
Anomalocardia
Pass Key 37 Pteria longisquamosa Cerith L di Evenness s
Brachidontes ransennella ol 70 « Modulus modulus auberiana & erenium Crepidul: aevicardium Ags ~
0 QMode Cluster piles Transennella sp. Bittiolum varium & Modulus modulus RX Group sisdaviinn Crepidula spp. Sa 5 ss oe 8 Sj
Zz
20 - at 6
=
3
3
& 3
ta} 8
§ | ®
cs
a. = aS
B40 KI = us L
a
|
|
| |
| |
60 = | > }———4
K2
A KI
: = rs = 0 50 0 50 0 20 0 20 0 20 0 20 0 1 2 0 10 20
= Shannon's Diversity Faunal Richness
Index
Percent Abundance
Text-figure 8—Down-core changes in core PK37 from south of Pass Key in eastern Florida Bay. Columns illustrate changes in Q-mode
clusters (see Text-fig. 11), significant molluscan fauna (shown as percent abundance), Shannon’s diversity index, evenness, number of faunal
groups, and number of individual specimens. RXIc group refers to the R-mode cluster shown in Text-figure 12; this group is composed
primarily of species that prefer salinities less than 18 ppt. Depth is shown in cm.
MOLLUSKS PAST AND PRESENT: BREWSTER-WINGARD ET AL.
Crepidula spp. Cerithiurt mu:
Russell Bank 19B
Q-Mode Clu
iS)
nr
Evenness No. of specimens
0 1 0 500 1000
7
KI
20
40
z
Depth in cm
Ki
100
120 {SST
K1
140
ai inne 1994
44
4
Evenness
1980
), of specimens.
1960
1ed4,
Index
1940
1920
Faunal 1900
Richness
= = — ora Talento T
0 50 100 0 50 100 0 50 0 50 0
Percent abundance
: ~ ——— tot “eon
SOO 30) Foe eee SOs 10-520 100 200
Shannon's
Diversity
Index
Faunal Richness Rainfall (cm),
Text-figure 9 —Down-core changes in core RB19B trom Russell Bank in eastern Florida Bay. Columns illustrate changes in Q-mode clusters
(see Text-fig. 11), significant molluscan fauna (shown as percent abundance), Shannon’s diversity index, evenness, number of faunal groups,
and number of individual specimens. Rainfall data is compiled from NOAA NCDC website http://www.ncde.noaa.gov/. RXIc group refers to
the R-mode cluster shown in Text-figure 12; this group is composed primarily of species that prefer salinities less than 18 ppt. Depth is shown
in cm. Age is based on *'°Pb analysis (Holmes ef al., this volume).
groups occur in cluster J that are absent in other clus-
ters (F = 100, Table 5), indicating cluster J represents
a relatively unique and diverse assemblage. These re-
sults are consistent with the analysis of the push cores
(Text-fig. 6); site 8 push cores formed a distinct cluster,
separate from the other sites. The average diversity
measures for cluster J are relatively high: Shannon’s
diversity index is 2.03; evenness is 0.74; and faunal
richness is 16.8.
Cluster K represents a mixture of samples from
cores 6A, 19B, and 37, with modern push core samples
from sites 12, 13, and 20, and modern vegetation sam-
ples from sites 8, 12,13, and 20. The common unifying
element for cluster K is the presence and relatively
high abundance of Brachidontes exustus in all but a
few samples (average abundance 32.13%, C = 98.1).
A number of species are present in a high percentage
of samples (Table 5), but at relatively low percent
abundances, forming a distinct and diverse assemblage
defined by R-mode clusters XIa and XIb. Cluster K
can be subdivided into four components (Text-fig. 11).
Cluster K1 is predominantly samples from cores 19B
and 37. In addition to Brachidontes exustus, cluster K 1
is defined by the presence of Bittiolum varium, Ceri-
thium muscarum juy., and Rissoidae. The core samples
in K1 can be further divided based on the abundance
of Pteria longisquamosa, Modulus modulus and mar-
ginellids in Kla and Transennella sp., Chione cancel-
lata, Olivella pusilla, and Vermicularia sp. in K1b.
Cluster K2 is faunally diverse and contains a mixture
of vegetation, push core and core samples primarily
defined by R-mode cluster Xla species, including
Laevicardium mortoni and Tellina spp., which distin-
guishes this cluster from K1. The occurrence of Cer-
ithidea spp. with abundant Brachidontes exustus iso-
lates the four samples that constitute cluster K3. Clus-
ter K4 is primarily vegetation samples and is lower in
diversity that the rest of cluster K. Bittiolum varium,
216
BULLETIN 361
Bob Allen 6A Transennella sp. Brachidontes exustus Crepidula spp. Cerithium muscarum Bittiolum varium Pteria longisquamosa Chione cancellata_ Anomalocardia = Mae:
Q-Mode Cluster & Modulus modulus & Perici auberiana a ashe : jen
\ 1994
\
~ Sy Zz
wns
& 1980
y a
3
< 3
20 | - — YS 2 3S fat
1960
6 : L = 1940
<> —
<=
<> => 1920
“<
E > >
o ¢
p> 1900
¢ )
o.
cy
Eo les a : ae |e )
80
) Evenness| jl
2 7, . 1880
> >- >
100 = ) = J } LL > 1860
>> ’
1840
120 } 4
=a 1820
Faunal 20
= Richnes§
Index A
F p> | > > mS Hx
Za 1800
= >
>
ES ) 37 9 8.4 63.1
G Crassispira spp. 9 8 82 1 1.9 18.0
12 Nucula proxima Say 7 6.5 77.9 1 1.9 22.1
G Rissoina cancellata Philippi 2 3 21 13 12 9
G Batillaria minima (Gmelin) 9 NS) 84 3 3 16
P Mysella planulata (Stimpson) 1 2 14 9 8.4 70.7 1 1.9 15.6
G Amaea retifera (Dall) | 2 10 16 15.0 90.1
G Muricidae sp. B 11 18 76 6 6 24
G Olivella ? sp. 5 4.7 71.6 1 1.9 28.4
G Muricidae sp. A 3 5) 32 9 8.4 55.4 1 1.9 12.2
G Finella sp. 6 6 75 1 1.9 24.8
G Conidae juv. 1 2, 23 6 5.6 77.4
G Pyramidellidae 6 10 47 8 a 36 2 3.7 17.6
1p Argopecten irradians (Lamarck) 1 2 14 7 6.5 55.0 2 aia Bi
12) Tagelus sp. 16 26 100
P Pitar simpsoni (Dall) 2 3 13 11 10.3 41.7 6 ates 45.0
G Vitrinellid 15 25 67 13 12 33
G Cyclostremiscus suppressus 5) 8 32 19 18 68
(Dall)
P Limaria sp. cf. L. pellucida 4 7] 18 3] 29.0 TES 1 1.9 5.0
(C.B. Adams)
XIIb Pp Arcopsis adamsi (Dall) 7 11 34 24 22.4 66.2
G Vermicularia sp. 2 3 16 18 17 84
G Caecum pulchellum Stimpson 4 7 20 26 24 74 1 1.9 5.7
G Rissoidae 2 3 8 38 3525 91.5
XIlc G Bittiolum varium (Pfeiffer) juv. 14 23} 39 38 36 61
G Odostomia spp. 10 16 39 27 25.2 60.6
G Prunum apicinum (Menke) 8 13) 69 220) 21 61
G Schwartziella catesbyana 13 21 46 23 DAS 46.2 2 Shi/ 8.0
(d’ Orbigny)
G Truncatella spp. 16 26 85 5 4.7 15.1
Total number of samples/cluster 61 107 54
(Halley, http://sofia.usgs.gov/projects/circulation;
Johns et al., 1999; NOAA NCDC website http://
www.ncde.noaa.gov/). In the northern and eastern
portions of Florida Bay, salinity varies more than
temperature at each site on both a seasonal and an
annual scale. Typically at each site, there is less fluc-
tuation in salinity from year to year in February (the
“dry” season), than during July (the ““wet™ season).
The exception is the 1998 pattern; during 1997-1998,
a very strong El Nino affected regional rainfall
throughout southern Florida, bringing heavy winter-
time rain. This was followed by a strong La Nina
causing a hot dry summer in southern Florida. Text-
figure 3 illustrates the effect of the anomalous re-
gional wintertime rainfall on the salinity of northern
and eastern Florida Bay. February 1998 resembles a
typical summer pattern, whereas July 1998 resembles
a winter pattern.
MODERN ANALOGUES
Five distinct molluscan associations were identified
from the analysis of the modern presence-absence da-
tabase. Assemblages identified by clusters B and E
(Text-fig. 4) are minor components of larger commu-
220
1.8
BULLETIN 361
Truncatella spp.
Schwartziella catesbyana
Prunum apicinum
Odostomia spp.
Bittiolum varium juv.
Rissoidae
Caecum pulchellum
Vermicularia sp.
Arcopsis adamsi
Limaria sp. cf. L. pellucida
Cyclostremiscus suppressus
Vitrinellid
Pitar simpsoni
Tagelus sp.
Argopecten irradians
Pyramidellidae
Conidae juv.
Finella sp.
Muricidae sp. A
Olivella ? sp.
Muricidae sp. B
Amaea retifera
Mysella planulata
Batillaria minima
Rissoina cancellata
Nucula proxima
Crassispira spp.
Caecum cornucopiae
Melongena corona
Turbonilla spp.
Lucina pectinata
Eulimidae
Lithopoma americanum
Anodontia alba ?
Codakia spp.
Hyalina sp.
Marshallora nigrocincta
Gastrocopta sp.
Eulithidium sp.
Ostrea equestris
Diodora sp.
Cyrenoida floridana
Lucina sp.
Semele bellastriata
Fasciolaria spp.
Lucinisca nassula
Parvilucina multilineata
Naticidae
Physidae
Polygyridae
Epitonium rupicola
Haminoea sp.
Turridae sp. A
Melampus coffeus
Busycotypus sp.
Assiminea succinea
Nassarius sp.
Lymnaea columella
Cerodrillia sp.
At sp
Rictaxis punctostriatus
Acteocina canaliculata
Anomalocardia auberiana
Cerithidea spp.
Hydrobiidae
Polymesoda maritima
Semele proficua
Mytilid
Mytilopsis leucophaeata
Parastarte triquetra
Cumingia tellinoides
Bulla striata
Marginellid
| Olivella pusilla
Chione cancellata
Cerithium muscarum
Transennella sp.
Tellina spp.
Laevicardium mortoni
Pteria longisquamosa
Modulus modulus
Crepidula spp.
Brachiodontes exustus
Cerithium muscarum juv.
J Bittiolum varium
1.5 iy 0.9 0.6 0.3 0
Cosine Theta Distance Measurement
MOLLUSKS PAST AND PRESENT: BREWSTER-WINGARD ET AL. 221
Table 6.—Salinity and temperature ranges for molluscan taxa observed at monitoring sites.
Salinity (ppt) Temperature (deg. C) Neue
Minimum Maximum Average Minimum Maximum Average Observations
Acteocina canaliculata 20.7 23.9 22.3 21.6 30.9 26.2 2
Anomalocardia auberiana 15.2 39.3 29.0 20.0 34.6 27.8 8
Arcopsis adamsi 15.8 35.1 24.8 19.2 32.0 24.8 9
Argopecten trradians 20.6 41.3 31.8 Wot 33.6 27.6 22
Batillaria minima 16.7 41.0 2TES) 19.4 34.6 26.6 24
Bittiolum varium 13.0 41.3 27.6 19.9 34.6 26.2 26
Brachiodontes exustus 10.2 41.3 27.4 1922. 34.6 27.4 85
Bulla striata 19.6 40.5 28.8 16.5 33.4 27.9 12
Busycotypus spp. Sil) 36.9 34.7 23.9 33.3 29m 3
Caecum pulchellum 15.2 28.2 22.4 20.0 32.0 27.6 3
Carditamera floridana 17.1 39.7 Sef 21.4 34.2 28.2 10
Cerithid 10.2 33.3 232 17.7 33.4 26.9 12
Cerithium muscarum 17.1 40.5 29.8 19.9 34.6 28.6 10
Cerithium spp. 15.8 38.4 30.9 21.0 34.6 28.8 21
Chione cancellata 24.4 41.3 34.1 25.5 34.2 30.4 18
Columbella spp. 17.6 40.0 34.3 DileS 34.2 29.0 18
Crepidula spp. 15.4 40.5 27] 19.4 34.6 26.3 43
Diodora sp. 20.6 37.0 30.0 20.7 34.6 28.5 9
Fasciolaria spp. 21.8 3957 31.4 17.4 34.2 All 19
Laevicardium mortoni 12.7 41.0 28.3 19.4 34.6 26.6 12
Limaria sp. cf. L. pellucida 27.0 36.9 30.9 20.2 32.0 27.0 4
Lithopoma americanum 29.0 36.8 33.4 PAGS) 32.3 28.1 5)
Lucina pectinata NPT 25.1 18.9 23.8 28.8 26.3 2
Marginellid 12.7 36.5 24.4 17.4 32.9 2553 15
Melampus coffeus PIES} 34.4 27.9 32.3 32.9 32.6 2
Melongena corona 14.1 41.0 24.9 19.8 33.0 25.8 16
Modulus modulus 19.6 3I9)7/ 30.7 17.6 34.6 28.4 21
Muricidae sp. 26.9 36.8 Sills) 20.0 30.4 259) 5)
Ostea equestris 2.7 39.0 20.5 19.2 31.3 24.8 1S
Parastarte triquetra 20.7 40.5 31.6 21.6 34.6 30.1 4
Pinnidae 27.0 38.4 33.6 20.2 8251 2729) 6
Pleuroploca gigantea Si-9) 34.3 33.1 23.9 31.0 DUS 2
Prunum sp. 17.1 41.0 30.4 16.5 34.6 28.3 25
Pteria longisquamosa 10.2 40.5 29'3 17.4 33.4 QED 62
Tegula fasciata 29.0 38.4 34.6 2IES 31.9 2/3) 7
Tellinid WAG 40.5 27.6 24.0 32.1 29.2 4
Transennella sp. SEZ: 39.3 28.4 20.0 34.6 28.3 q/
Truncatella spp. 15.2 30.5 22.2 20.0 30.9 23:2 4
Turbo castanea 213 38.4 33.2 17.6 BORO) 26.4 16
Turridae sp. A 20.7 28.2 24.4 21.6 32.0 26.8 2
nities within Florida Bay, and are not considered sig- ditions. This assemblage frequently co-occurs with the
nificant for this study. Text-figure 13 shows the geo-
graphic range of five communities in Florida Bay, and
Text-figure 3 shows the changing distribution from
season to season over the period of observation.
Batillaria/Melongena Assemblage
The Batillaria/Melongena assemblage defined by
cluster A represents a unique set of environmental con-
Brachidontes assemblage (cluster C), but occupies a
unique subenvironment at each site. Subenvironments
were not analyzed separately, therefore, the occurrence
of the Batillaria/Melongena assemblage is not always
reflected on the distribution maps (Text-fig. 3). This
assemblage appears to be restricted to eastern and cen-
tral Florida Bay, in very shallow areas with sparse sub-
aquatic vegetation and in close proximity to islands.
ae
Text-figure 12.—R-mode cluster of compiled data (cores, push-cores, and vegetation samples). Clusters were generated using MVSP statis-
tical package (Kovach Computing Services, MVSP Plus, version 3.1) on log-ratio transformed and centered percent abundance data, using
unweighted paired group method, average-linkage (UPGMA), with cosine theta distance measurement, dual clustering procedure, and random
input order. Full data matrix sorted in dendrogram order can be viewed at http://flaecohist/database/Reference/synthesis.
22) BULLETIN 361
é oes
D6 | ie Cluster G
.@, Extent of Cluster
ae, | Distribution |
i = Cluster A |
j~t— Cluster C ©
® tt Cluster D |
Cluster F
Text-figure 13.—Map showing range of distribution of presence-absence clusters over course of sampling period from February 1996 through
July 1999. This represents a compilation of data shown on individual seasonal plots in Text-figure 3. Clusters B and E are not shown; the
distributions are limited and non-contiguous.
Dense Batillaria minima consistently have been pre-
sent at site 14 since the beginning of our observation
period (Text-fig. 3). Salinity does not appear to be a
strong controlling factor; Batillaria minima and Me-
longena corona have been found from mesohaline to
hypersaline conditions (Table 6). These observations
are in general agreement with Turney and Perkins
(1972). They identify Melongena corona as a com-
mon species in the northern subenvironment and Ba-
tillaria minima as an intertidal species found in sedi-
ments near land; we have commonly found Melongena
corona in the northern transitional and eastern divi-
sions of Florida Bay.
° Turney and Perkins (1972) indicate it is unusual to find Batillaria
anywhere but in the immediate vicinity of dry land, and they place
a great deal of significance on finding it ““more than 1/4 mile from
present land” at site 7403. Site 7403 is in close proximity to our
site 11, on the southern end of the mud bank extending south from
Park Key. We have observed live Batillaria at site 11, and on other
mudbanks extending outward from land (sites 4, 7, 12). We believe
this is a typical habitat for Batillaria and disagree with Turney and
Perkin’s interpretation of its presence as being indicative of recent
migration of Park Key (Turney and Perkins, 1972, p. 29).
Brachidontes Assemblage
The Brachidontes assemblage defined by cluster C
is the most predominant molluscan community found
in eastern and central Florida Bay. Brachidontes ex-
ustus 18 commonly found alive at most sites in this
region of the bay, regardless of salinity, water depth,
substrate, or clarity. The low fidelity values for Brach-
idontes exustus in cluster C (Table 4) are due to the
ubiquitous nature of the species. Although Brachidon-
tes exustus is euryhaline, ranging from 10.2—41.3 ppt,
the average value is 27 ppt.° Brachidontes exustus
have been found in the deepest parts of the basins in
eastern Florida Bay, bysally attached to exposed lime-
stone, and on the tops of mud banks attached to sub-
aquatic vegetation. Brachidontes exustus are found at-
tached to almost all common types of vegetation, in-
cluding Thalassia, Halodule, Sargassum, Chondria,
° The mean and median salinity values for Brachidontes are 27.45
and 28.8 ppt, based on 85 observed occurrences of the species.
These values do not take into account the abundance of the species
at a site, and may in part reflect the frequency of sampling within
higher salinity ranges (see Text-fig. 2).
MOLLUSKS PAST AND PRESENT: BREWSTER-WINGARD ET AL.
and Laurencia, and to exposed limestone and shell de-
bris. When Brachidontes exustus are very abundant
they commonly form “nests” in clumps of Chondria,
Laurencia, and dead Thalassia, and they were ob-
served forming a dense “Brachidontes-mat” by at-
taching to each other across the top of an expansive
shoaling area (2/97, site 12). The most abundant
Brachidontes exustus concentrations we have observed
are typically found on the sides of the mud banks in
clumps of Chondria and/or Laurencia in relatively
murky water. In the process of collecting and process-
ing samples, we have determined that Brachidontes
exustus are very tolerant of low water quality; they
have been found alive several days after collection in
sealed plastic bags, filled with decaying organic ma-
terial. Given the wide range of environmental condi-
tions under which Brachidontes exustus have been
found, it is difficult to determine what factors control
their distribution. Given these tolerances it is note-
worthy that Brachidontes exustus are primarily re-
stricted to the northern transitional, eastern, and central
portions of Florida Bay. A possible explanation is that
their distribution is a function of decreased competi-
tion or fewer predators in this region of the bay.
An examination of the associated species that sep-
arate the sub-clusters within C indicates some of the
environmental variables that control the distribution of
the Brachidontes assemblages. Sub-cluster Cl is de-
fined by the co-occurrence of Brachidontes exustus
with Prunum apicinum; these species are most com-
monly found together on clumps of Laurencia or
Chondria. The co-occurrence of Brachidontes exustus
and Pteria longisquamosa with Argopecten irradians
or Chione cancellata delimits sub-cluster C2. This
group of species is indicative of relatively clear water
and abundant healthy Thalassia; macro-benthic algae
may be present. Sites with Brachidontes exustus alone
or with Batillaria minima form sub-cluster C3. Batil-
laria minima are present when the water is very shal-
low, the substrate is generally soft mud, and the veg-
etation is sparse. Due to Brachidontes exustus toler-
ance of varying salinity and substrate conditions and
relatively poor water quality, the presence of B. ex-
ustus alone may indicate the environment is not suit-
able for other epifaunal molluscs. Sub-clusters C4 and
C5 have relatively high values for faunal richness.
Pteria longisquamosa, Crepidula spp., Bittiolum var-
ium, and Prunum apicinum may be present with
Brachidontes exustus, and a number of other species
in R-mode cluster Ia. Species from R-mode cluster Ib
are included in sub-cluster C5. These diverse faunas
are typically present when the substrate consists of
mixtures of Thalassia and macro-benthic algae includ-
ing Polysiphonia, Chondria, and Laurencia. Sub-clus-
i)
i)
eS)
ter C6 is defined by the presence of Brachidontes ex-
ustus and Crepidula spp.; Crepidula are generally
found on Thalassia, and like Brachidontes exustus, ap-
pear to be relatively tolerant of reduced water quality.
Count-data from the modern vegetation samples and
the push cores are in agreement with presence-absence
observation data, and provide further evidence of the
predominance of Brachidontes exustus in the mollus-
can fauna of Florida Bay. Out of 11,597 total individ-
uals collected on vegetation samples in July 1998, we
found 5713 Brachidontes exustus (49.26%). These
numbers are comparable to the findings of Lyons
(1998). He found Brachidontes exustus comprised
78% of all the molluscs he collected in an extensive
sampling program throughout Florida Bay in the sum-
mer of 1994 that gathered nearly 14,000 living speci-
mens greater than 3 mm. In the pristine fraction of the
push core samples, Bittiolum varium comprises 30.3%
of the specimens, and Brachidontes exustus 13.0%,
followed closely by Transennella sp. at 11.43%. When
all the fragments and worn specimens are included
from the push core data, Brachidontes exustus com-
prises 20.64% of the total molluscan specimens, and
Bittiolum varium 19.74%. The lower values for Brach-
idontes in the pristine sub-set are probably due to the
fragility of the shell.
Pteria Assemblage
Pteria longisquamosa [=Pinctada radiata Turney
and Perkins] and its associated species Modulus mod-
ulus, and Argopecten irradians define cluster D. This
association of species is typically found on the sides
of mudbanks (40—150 cm of water) in dense Thalassia
beds, relatively clear water, and salinities between 20
and 40 ppt’ (Table 6), in eastern, central, western and
Atlantic transition zones of Florida Bay. Although
Pteria longisquamosa has been found at its densest
concentrations attached to Thalassia, it has also been
found on macro-benthic algae, particularly Chondria
and Laurencia, and it has the ability to camouflage
itself to match the color of the vegetation to which it
attaches. The occurrence of Preria longisquamosa in
relatively clear water raises an interesting question of
whether P. longisquamosa occurs there because the
water is clear, or whether the water is clear due to the
filtering activity of P. longisquamosa. The latter es-
pecially may be true where Pteria longisquamosa are
extremely abundant. Prteria longisquamosa, however,
seem unable to survive in water of diminished quality;
7’ The mean and median salinity values for Pteria longisquamosa
are 29.3 and 29.7 ppt, based on 62 observed occurrences of the
species at a site. These values do not take into account the abun-
dance of the species at a site, and may in part reflect the frequency
of sampling within higher salinity ranges (see Text-fig. 2).
224 BULLETIN 361
they generally die within hours of collection. Typical-
ly, the Preria assemblage occurs outside the 20-ppt
contour and to the east, south, or west of the Brachi-
dontes assemblage (Text-fig. 3). Based on our obser-
vations, we believe the distribution of the Preria as-
semblage is controlled by a combination of salinity,
substrate, water depth, and water clarity.* The domi-
nance of this combination of species in a death assem-
blage would be indicative of polyhaline to euhaline
relatively clear water, deposition on the side of a mud
bank in 40-150 cm of water, and the presence of a
relatively dense Thalassia bed.
Turney and Perkins (1972) found Brachidontes ex-
ustus, Pteria longisquamosa, Cerithium muscarum and
Bittiolum varium to be the characteristic species of
their interior subenvironment, which corresponds ap-
proximately to the eastern and central divisions of our
current usage. Their results agree with our analyses.
The distribution of the Brachidontes and Pteria assem-
blages overlap in central and eastern Florida Bay
(Text-fig. 13). Turney and Perkins (1972, p. 10) state
that the molluscs in the interior subenvironment are
able to “survive large salinity fluctuations and other
effects of poor circulation.” They point out the ability
of the characteristic species to survive in other sub-
environments, but suggest that one reason they flourish
in the interior may be a lack of competition. Our data
support the idea of wide salinity tolerances for these
assemblages, but our data indicate Pteria longisqua-
mosa is sensitive to water quality. This apparent dis-
crepancy can be explained by the fact that Turney and
Perkins were analyzing death assemblages.
The “Western” Assemblages
The Turbo, Tegula, and Columbella association that
forms cluster Fl has been found at three sites (17, 18,
23) in the western, Atlantic and Gulf transition zones;
these sites have a mixture of various types of sub-
aquatic vegetation including Thalassia, Syringodium
and Halodule and calcareous green algae. These sites
also have varying substrates and water depths, because
each site has a channel bordered by a mudbank or
sandbar. Currents flowing through the channels can be
strong at times, and the water is typically clear; salin-
ities have ranged from 25—38 ppt. Based on our ob-
servations, we consider this assemblage to be indica-
tive of near normal marine conditions (upper polyha-
line to euhaline), relatively clear water, and the pres-
ence of sub-aquatic vegetation. Currents and varying
substrates also may be determining factors in the dis-
* Water depth, water clarity, and vegetation are not independent
of each other. The tops of the mudbanks typically have less vege-
tation and more material in suspension than the sides of the banks.
tribution of the Fl assemblage, but we have not ob-
served this assemblage frequently enough to reach a
conclusion.
Cluster F2 is a weak association of species including
Cerithium spp., Chione cancellata, Carditimera flori-
dana, and Columbella spp. and other faunal groups
from R-mode cluster Ia. The cluster is formed of sam-
ples from sites 13, 17, 21, 24, and 26 in the central,
western and Gulf transition zones of Florida Bay. Sa-
linities for the sites in cluster F2 range from 30—39.
Our data suggest this assemblage represents euhaline
conditions, but the limited observations and weak as-
sociation of fauna make this a speculative conclusion.
Pinnidae and Pleuroploca gigantea are the critical
fauna in cluster G. These taxa are generally found in
the western most sites (23 and 24) in the Gulf transi-
tion zone and they are representative of open marine
euhaline conditions. Pinnidae are found in shallow
shoaling areas on top of banks, partially submerged in
the substrate. The substrates have varied from muddy
calcareous sand to firm calcareous mud. These fauna
have been rarely observed, but seem indicative of eu-
haline conditions.
Our observations of the more euhaline marine as-
semblages differ from Turney and Perkins (1972).
Lithopoma americanum |=Astraea americana Turney
and Perkins, 1972] and Tegula fasciata were listed as
characteristic species of Turney and Perkin’s (1972, p.
13) Atlantic subenvironment, and they stated that “the
Atlantic group is composed of species which are al-
most entirely restricted to that part of the bay which
is influenced by the Atlantic Ocean.” Although our
observations of living Lithopoma americanum and Te-
gula fasciata are limited, we have observed both spe-
cies in the central (site 17), Atlantic (site 18), and Gulf
transition (site 23) zones.
Comparison of Presence-Absence Data with
Other Data
Through quantitative and qualitative comparisons of
the death assemblages represented by the push core
data with other data sets (Lyons, 1995, 1998; Turney
and Perkins, 1972; and our modern vegetation and ob-
servation data discussed above) we have determined
that the push core data are a valid representation of
the fauna that have lived at each site during the 0-10
years preceding collection. The prominence of Brach-
idontes exustus and Bittiolum varium in the push cores
from sites 8, 12, 13, and 20 agrees with our modern
vegetation data from those sites. Transennella sp. and
other infaunal organisms (absent in the presence-ab-
sence observation data and the vegetation sample data)
are a significant component of the push-core assem-
blages. Their absence from the observation and veg-
MOLLUSKS PAST AND PRESENT: BREWSTER-WINGARD ET AL. aS)
etation data sets does not indicate a lack of agreement
with the push core data but is simply an artifact of our
method of observing and sampling (see discussion un-
der methods). Turney and Perkins (1972) did a detailed
analysis of the death assemblages at over 75 stations
in Florida Bay in the 1950’s and they concluded that
“transportation of shells larger than silt size is not sig-
nificant in Florida Bay” (1972, p. 32). Only local mi-
gration of shells takes place and “‘faunal dislocations”
are rare. Diagenetic and depositional processes have
little effect on the distribution of molluscan assem-
blages (Turney and Perkins, 1972, p. 35). Thus, their
work supports our conclusion that push core data pro-
vide a valuable addition for down-core comparison to
the historical data set.
HISTORICAL PISTON CORE DATA
Three primary clusters are formed by the analysis
of the compiled data set (modern vegetation samples,
push cores, and piston cores) (Text-fig. 11). Only the
Brachidontes assemblage that defines cluster C in the
modern analysis is represented in the compiled data
set as cluster K. The other modern assemblages iden-
tified in the cluster analysis of the presence-absence
data set are not represented in the piston core data.
The absence of the modern assemblages (with the
exception of the Brachidontes assemblage) from the
piston cores can be explained by a number of factors.
The Batillaria/Melongena assemblage typically is lim-
ited to shallow-shoaling areas in relatively close prox-
imity to land. The four piston cores were not taken
under these conditions in the modern environment, and
the nearly complete absence of Batillaria minima and
Melongena corona from the piston cores implies these
conditions did not exist in the past at the localities
where piston cores were taken. Clusters B and E, the
Ostrea/Arcopsis assemblage and the Laevicardium as-
semblage, are limited to only a few sites in the modern
sampling, are loosely associated groups, and are gen-
erally affiliated with Brachidontes. When these species
are present within the piston core, the samples join
with other clusters. Pteria longisquamosa are present
in the piston cores (primarily RB19B and PK37), but
do not form the discrete Pteria assemblage seen in
cluster D of the modern analysis. Instead, Pteria lon-
gisquamosa are a component of the Brachidontes as-
semblage in cluster Kla. The fragility and low pres-
ervation potential of Preria longisquamosa shells most
likely affect their relative abundance in the piston core
data, although they are easily recognizable from small
fragments. This observation agrees with the results of
Turney and Perkins (1972); they did not distinguish
separate Pteria longisquamosa = [Pinctada radiata
Turney and Perkins 1972] and Brachidontes exustus
groups in the death assemblages from the interior sub-
environment where they found both species to be
abundant.
The key species in the two western assemblages
(clusters F and G) are completely missing from the
piston cores. Turbo castanea, Tegula fasciata and Col-
umbella spp., which comprise cluster Fl, are not pre-
sent in any piston core samples. Cluster F2 is a more
loosely associated group, defined by the presence of
Columbella spp., Cerithium spp., Chione cancellata,
and Carditimera floridana; Carditimera and Colum-
bella are absent from all piston core samples. Ceri-
thium and Chione are present, but typically in associ-
ation with Brachidontes or Transennella dominated as-
semblages. Pinnidae and Pleuroploca gigantea (cluster
G) are absent from the piston cores, but Lithopoma
americanum is a rare component in the upper 50 cm
of piston core BA6A. The absence of the Turbo/Te-
gula/Columbella and Pinnidae/Pleuroploca assem-
blages from the piston cores indicates that during his-
torical times conditions in northern, eastern and central
Florida Bay did not reach the more open marine con-
ditions currently seen in the western, Gulf, and Atlan-
tic transition zones.
Compiled Data
The down-core distribution of the three primary
clusters formed by the compiled data analysis, and the
relationship of the modern vegetation and push core
data to the historical piston core data, are shown in
Text-figure 14. The unique character of the molluscan
assemblage in Little Madeira Bay (site 8 and piston
core T24 site) is apparent. Cluster J (Text-fig. 11) is
predominantly composed of samples from piston core
T24 and modern site 8. Abundant Hydrobiidae and
Anomalocardia auberiana, combined with the less
common Acteocina canaliculata, Polymesoda mariti-
ma, and Cerithidea spp. (R-mode XIc group, Text-fig.
12), separate these samples from the samples in eastern
and central Florida Bay. The species in R-mode cluster
XIc are relatively rare outside of Little Madeira Bay.
Only a few samples from other localities join cluster
J (piston core PK37 10 cm; push cores from site 20,
and a push core and vegetation sample from site 12);
these samples join the cluster due to the presence of
Anomalocardia and Acteocina. Turney and Perkins
(1972) identified Anomalocardia auberiana [=Anom-
alocardia cuniemeris Turney and Perkins, 1972] as the
characteristic species of the northern transitional zone,
and Acteocina canaliculata [=Retusa canaliculata
Turney and Perkins, 1972] as a common species. Ly-
ons (1996, 1999) has determined that zones of An-
omalocardia occurrence will shift with changes in sa-
linity. Thus, the higher abundance of Anomalocardia
@ L#8-867 ‘€#V8-864
=| ‘ZHVS-B6L ‘THV8-86L fet HTS
ae V8-967 ‘W8-L6L ‘V8-96L ‘W8-L6Z ‘W8-862 ‘V8-86L ‘V8-S6L ‘V8-S6Z ea on eee Seo ere eed
S > ae £86 ig °F Uke Bere
Sa mw ||O Ue 20s a Oe eOmemoES
iS) al
aa) gl S|
= Es
i?) +
2 ela | :
= 1S) g
GHOZ-B67 ‘ZHOZ-867 ‘FH0Z-867 Eee
‘EHV OC-R6L ‘THWOC-B6L C#07-86Z Se aie os a
omen
Sod
tp MM. 07-867 °9 VOZ-86Z ‘VOZ-L6L is >
is > bx
v2) Pet {=|
D Mi =: ™ ag
WD om
Ra a
Pu
ee)
foe)
Z
iF WM, THTI-862 ‘FHTT-R62 THZT-B6Z
a . PHATI-B6L sl AZI-L6Z ‘AZ1-962 ‘AZI-862 ‘AZI-L6L ‘ATI-S6Z ‘ATI-S6L ‘AZT-96L ra)
= lop)
>) 3)
ea faa S716 VN 121-862 y
Z
e & | M M4 M3 S on
3 SS =. —
Fs ‘ 3
w
ba
—— ia
ZHEL-86L MB <
cw ENEMA D€1-967 ‘DET-L6Z ‘DET-L6L ‘DET-B6Z ‘DET-S6Z ‘AET-S6L ‘DET-96L ‘DET-B6L e
mera
= = =; c
= oo — SRLS Se ee eee 8
< 2 Sk Mh , BS Sa Sh, Se as aS SS S SS ot
mo} i Bes ree =z
Qs s Sy ~ Ses
° S fee eleva OS Nae SS, so
pa pa
ve oS ° ro) =) =) = iS) eS
aovj1ayU] 18} -eoRJINS a Py ne Se = SS = vr
© (wo) yidaq
(co
|
Text-figure 14.—Relationship between Q-mode cluster assignment (compiled data set, Text-fig. 11) and sample position. Specific push-core
and vegetation samples are identified.
MOLLUSKS PAST AND PRESENT:
BREWSTER-WINGARD ET AL. D2,
Text-figure 15—*Blow-out” areas at Bob Allen mudbank, February 1999. Sharp transition between lush Thalassia beds and areas of
gelatinous calcareous mud are visible. Photo courtesy of Thomas Scott, Florida Geological Survey.
and Acteocina in some samples from Pass Key and
Russell Bank may be indicative of lowered salinities.
Samples from piston cores RB19B and PK37, and
the upper portion of BA6A, are primarily grouped in
cluster K (Text-figs. 11 and 14). This is equivalent to
the Brachidontes assemblage identified in the modern
presence-absence data analysis. Modern push cores
and vegetation samples from the corresponding sites
(12, 13, and 20) also fall into cluster K in the com-
bined analysis. These sites occur within eastern and
east-central Florida Bay, and they can be characterized
as shallow mud banks with sub-aquatic vegetation and
macro-benthic algae present. Salinities have ranged
from 17—33 ppt at site 20, 19-37 ppt at site 12 and
27-41 ppt at site 13 during our periods of observation.
Typically, either the Brachidontes or Pteria longisqua-
mosa assemblages have dominated at these sites in the
modern presence-absence data (Text-fig. 3). The prev-
alence of the Brachidontes assemblage down-core for
PK37 and RB19B, and down to approximately 54cm
in BA6A, indicates that the general environmental
conditions were the same in the past. The common
species in cluster K correspond to the characteristic
species identified by Turney and Perkins (1972) for
their interior subenvironment.
Samples from the lower portion of piston core
BA6A are dominated by the Transennella assemblage
that defines cluster L. This assemblage also is seen in
portions of RB19B (30 cm and 118 cm) and PK37 (6
cm). This is an infaunal assemblage and so it was not
identified in the modern presence-absence analysis.
Transennella sp. are present in the modern push core
samples, but they are components of diverse Brachi-
dontes assemblages, and thus the samples join cluster
K. Bob Allen mudbank today (site 13) is distinctive
among the sites discussed here. Lush Thalassia beds
adjoin “‘blow out” areas at site 13 (Text-fig. 15) and
the transition from grass to gelatinous carbonate mud
is very sharp. Low diversity Transennella assemblages
seen in the piston core may be the down-core repre-
sentation of these modern “‘blow-out” areas, but we
have not sampled the modern analogue for this cluster
and can only speculate on what it indicates about the
environmental conditions.
Piston Core T24
Historical changes in molluscan fauna at the mouth
of Taylor Creek (piston core T24) are shown in Text-
figure 7. The gradual decrease of the R-mode XIc and
the increase in Brachidontes exustus, Crepidula spp..
and Cerithium muscarum up-core illustrate the gradual
increase in salinity that has occurred at this site during
228 BULLETIN 361
the time of deposition. Brachidontes exustus increase
significantly in the upper 20 cm of the core. Modern
analogue data (discussed herein; Turney and Perkins,
1972; Lyons, 1996, 1998, 1999) have demonstrated
that Brachidontes exustus seem most abundant in areas
of fluctuating salinity and in areas where macro-ben-
thic algae are available for habitation. Fluctuations in
salinity at this site could be attributed to changes in
fresh-water flow due to climatic change, or changes in
water management practices, or it could be due to
long-term changes in sea level. Fluctuations in the rel-
ative abundance of Hydrobiidae may be directly in-
dicative of changes in fresh water flow. Hydrobiidae
are minute (generally less than 3 mm) fresh-water gas-
tropods that cannot survive in estuarine waters. They
arrive in Little Madeira Bay via terrestrial freshwater
flow through Taylor Creek, and perhaps their abun-
dance can be related to flow rates, but additional field-
testing is necessary. Despite the obvious changes in
the relative abundance of individual species, the entire
piston core falls within cluster J and Anomalocardia
auberiana, Turney and Perkins’ (1972) indicator of the
northern subenvironment, is present throughout the
piston core. In addition, diversity, evenness, faunal
richness, and number of individuals remain relatively
constant compared to the other piston cores.
Piston Core PK37
Deposition at Pass Key (piston core PK37) (Text-
fig. 8) has been relatively rapid, and the piston core
represents only the latter half of this century. Brachi-
dontes exustus has dominated the assemblage at this
site during the period of deposition, and below 26 cm,
the relative abundance of B. exustus is fairly constant.
Above 26 cm, however, a great deal of fluctuation oc-
curs. Seemingly dramatic changes in species domi-
nance near the top of piston core PK37 are due to
extremely low numbers of individual shells and low
faunal richness; which is probably a result of the rapid
sedimentation rate. The complete disappearance of all
species except Brachidontes exustus and Transennella
sp. in the upper 6 cm may be indicative of rapid sed-
imentation at the site and/or conditions unfavorable to
other species. The increases in Anomalocardia auberi-
ana and R-mode XIc group may denote periods of de-
creased salinity; this assemblage is characteristic of Lit-
tle Madeira Bay, and the 10 cm sample joins cluster J.
Piston Core RB19B
The piston core from Russell Bank (RB19B) (Text-
fig. 9) records approximately 110 years of change at
the site. The lower portion of the piston core (~1880—
1922) shows fluctuations in faunal richness and num-
bers of individuals, but the relative abundance of the
key species does not fluctuate dramatically. Transen-
nella sp. is the most abundant species in this portion
of the core, but the presence of Brachidontes exustus,
Cerithium muscarum, Bittiolum varium and other spe-
cies in most of the samples indicates the presence of
sub-aquatic vegetation. Significantly, Anomalocardia
auberiana and other species from the R-mode XIc as-
semblage typical of the northern transition zone, are
present before 1922. The presence of these species
may indicate lower salinities and possibly the influence
of a lower salinity plume. In the modern Florida Bay,
a less saline plume of water is sometimes present on
the western edge of the basin south of Little Madeira
Bay (Text-fig. 3, July 1996; Halley, http://
sofia.usgs.gov/projects/circulation). After 1922, signif-
icant changes occur in the relative abundance of the
key fauna. Transennella sp. decrease, and Brachidon-
tes exustus, Bittiolum varium and other components of
the Brachidontes assemblage increase somewhat. After
1942, fluctuations become even more pronounced, un-
til approximately 1980 when Brachidontes exustus in-
crease dramatically in relative abundance, and all other
key species, with the exception of Pteria longisqua-
mosa and Modulus modulus, decline. The increase in
Brachidontes exustus in the upper 20 cm of the piston
core corresponds to an increase in numbers of speci-
mens (this is consistent with the findings of Lyons
1998) and a decrease in faunal richness, evenness and
diversity.
Piston Core BA6A
Bob Allen mudbank (piston core BA6A) preserves
a record of deposition since approximately 1784. No
modern analogue has been identified for the Transen-
nella sp. assemblage (cluster L) that dominates the
lower portion of piston core BA6A, before 1922 (54
cm). The lowest portion of the piston core (140-158
cm) indicates either a rapidly fluctuating environment,
or deposition at the transition zone between a grass
bed (indicated by cluster K assemblage) and a low
diversity, sparse Transennella sp. habitat. Beginning
around 1890 to 1901, Brachidontes exustus becomes
a significant component of the molluscan assemblage
in piston core BA6A, and around 1901 the relative
abundance of Transennella sp. declines and never
again constitutes greater than 51% of the assemblage.
The Brachidontes assemblage (cluster K), indicative of
the presence of sub-aquatic vegetation) dominates the
upper portion of piston core BA6A. Beginning in the
late-1920’s, Lithopoma americanum, indicative of the
more euhaline western environments, is occasionally
present at Bob Allen mudbank. Around 1981, Brach-
idontes exustus becomes the most dominant species in
the piston core. In general, the measures of diversity
MOLLUSKS PAST AND PRESENT: BREWSTER-WINGARD ET AL. 229
(faunal richness, evenness, and the Shannon’s diversity
index) all fluctuate significantly throughout the depo-
sition of this core, and the values for numbers of in-
dividuals and faunal richness are lower in piston core
BA6A, in comparison to the other piston cores ex-
amined.
SUMMARY OF SIGNIFICANT HISTORICAL FINDINGS
1) Molluscan assemblages that exist at each piston
core locality today, typically existed there throughout
the period of deposition represented by the piston
cores. Shifts have occurred in the patterns of domi-
nance and diversity within each piston core, but the
faunal composition has not changed significantly.
2) Piston core T24, at the mouth of Taylor Creek,
records gradually increasing salinity, an increase in eu-
ryhaline molluscan fauna, and a decline in mesohaline
molluscan fauna up core. Anomalocardia auberiana,
typical of the northern transition zone, is present
throughout the core.
3) Pass Key piston core (PK37), southeast of Little
Madeira Bay, shows significant fluctuations in Anom-
alocardia auberiana and the other species indicative
of the northern transition zone.
4) Between 1913 and 1933, distinctive changes oc-
cur in the molluscan assemblage at Russell Bank (pis-
ton core RB19B). Anomalocardia auberiana, consis-
tently present in low percentages prior to 1922, com-
pletely disappears from the site after that time.
5) At Bob Allen mudbank (piston core BA6A) a
dramatic change in faunal assemblages occurred be-
tween 1890 and 1901, from a Transennella sp. domi-
nated assemblage to a Brachidontes exustus dominated
assemblage.
6) From approximately 1930 to 1980, Lithopoma
americanum, typical of euhaline western and Gulf
transition environments, is occasionally present at Bob
Allen mudbank (piston core (BA6A).
7) From approximately 1930—1980 dramatic shifts
in percent abundance of different species within the
Brachidontes assemblage occur at both Bob Allen
mudbank and Russell Bank.
8) Beginning around 1980, Brachidontes exustus
becomes the dominant mollusc at Russell Bank, and
faunal diversity and evenness decline. Similar increas-
es in Brachidontes exustus are seen in the upper por-
tion of all four piston cores analyzed.
CONCLUSIONS
Changes in ecosystems take place on many scales,
from daily to decadal to millennial. Molluscs generally
are tolerant of short-term perturbations in their envi-
ronment, but they will respond to seasonal and annual
changes in salinity. Thus, changes recorded in mollus-
can populations are significant. A comparison of mol-
luscan faunal distributions in modern Florida Bay to
down-core faunal distributions has revealed important
details about the last 100 years of Florida Bay history.
All four cores examined show changes during the
period of time represented by deposition at the indi-
vidual sites, yet only the Bob Allen site recorded a
complete change in molluscan assemblages. The mol-
luscan fauna that occur throughout the cores are the
same fauna that are present at those sites today. Chang-
es that have occurred within the cores are fluctuations
in patterns of dominance and diversity within a single
assemblage, not complete replacement of the assem-
blage. For example, the Turbo/Tegula/Columbella as-
semblage indicative of more marine conditions has not
existed at Bob Allen mudbank or at Russell Bank dur-
ing the last 100-200 years. Nor has the northern tran-
sitional assemblage, typical of the Little Madeira Bay
sites, dominated at Bob Allen or Russell Bank. Even
the change that occurs around 1900 at Bob Allen mud-
bank does not represent complete replacement of a
fauna. We have not identified a modern analogue for
the Transennella assemblage that existed at Bob Allen
prior to the turn of the century, but we do know that
Transennella sp. occur at Bob Allen today.
The changes indicated by shifts in dominance pat-
terns, however, do suggest the influence of some en-
vironmental factors on molluscan fauna in northern,
eastern and central Florida Bay. The patterns in core
T24 clearly demonstrate increasing salinity at the
mouth of Taylor Creek. Freshwater outflow through
Taylor Creek has been controlled by water manage-
ment during the later half of this century. The change
in salinity at this site may be due to water management
practices, fluctuations in average annual precipitation,
changes in sea level, or a combination of all of these.
The record from Pass Key (core PK37) is not as clear,
due in part to the rapid sedimentation rates at that site,
but fluctuations in the abundance of Anomalocardia
auberiana and other typical northern transition species
at Pass Key probably reflect fluctuations in salinity.
This conclusion is consistent with Lyons’ findings
(1996, 1998).
The age models for Russell Bank (core RB19B) and
Bob Allen mudbank (core BA6A) allow us to compare
changes in the molluscan fauna to historical data on
human activities and rainfall (Text-figs. 9 and 10). The
changes that occur between 1900 and 1910 at Bob
Allen and between 1913 and 1933 at Russell indicate
a shift toward less stable conditions. The disappear-
ance of Anomalocardia auberiana from Russell Bank
after 1922 may indicate more saline conditions. It is
possible, given the increasing error in the age model
with depth, or the possibility of varying sedimentation
230 BULLETIN 361
rates with depth, that these events occurred simulta-
neously. During this time period, average rainfall did
not fluctuate significantly in southern Florida, but sub-
stantial human alteration of the environment occurred
with the construction of the Flagler Railroad in the
Keys. Given the current data, it is impossible to de-
termine if the construction contributed to changes in
the molluscan fauna, but evidence indicates that the
railroad restricted the natural exchange of water be-
tween the bay and the Atlantic (Swart et al., 1996).
Between approximately 1930 and 1980, the mollus-
can fauna at Bob Allen and Russell both went through
a period of fairly rapid and dramatic fluctuations.
These fluctuations correspond to a period of relatively
dramatic shifts in regional rainfall and to a period of
significant alteration of the terrestrial environment
with the construction of canals, the implementation of
water management practices, and a rapidly growing
population. Lithopoma americanum, a typically west-
ern Florida Bay euhaline species, is occasionally pre-
sent at Bob Allen from approximately 1930 to 1980.
The presence of Lithopoma may be indicative of pe-
riods of increased salinity or of fluctuating marine in-
fluence caused by the environmental perturbations.
Beginning around 1980, Brachidontes exustus be-
comes the most dominant species at Russell and Bob
Allen, and it increases in dominance in the upper por-
tion of the cores from Pass Key and Taylor. Brachi-
dontes exustus is a euryhaline species that can tolerate
diminished water quality, and it is nearly ubiquitous at
sites in central and eastern Florida Bay today. The con-
centration of this species in the upper portion of all
the cores implies that salinity fluctuations have in-
creased in the last 20 years, and/or water quality has
diminished.
Additional studies of the autoecology of molluscan
species will refine our knowledge of the environmental
parameters that control their distribution and will in-
crease their utility as bioindicators. We have estab-
lished that the following molluscan species can serve
as important biological indicators of conditions in
Florida Bay during restoration: Anomalocardia auber-
iana, Brachidontes exustus, Cerithidea spp., Pteria
longisquamosa, Polymesoda maritima, Turbo casta-
nea, Tegula fasciata, and Hydrobiidae. Additional
cores from eastern and central Florida Bay are being
examined to geographically extend our coverage, and
determine if the patterns seen here persist in other ar-
eas as well.
It is critical for effective restoration of the Florida
Bay ecosystem that we understand the dynamics of the
system and the degree to which physical, chemical and
biological components of the system have varied in
the past. Studies of modern fauna and flora provide a
means to interpret biological data preserved in cores,
and to determine the physical and chemical variations
in the environment indicated by the biota. Once the
patterns of the past are understood, we can better pre-
dict the impact of future change on the environment.
REFERENCES CITED
Brewster-Wingard, G.L., Ishman, S.E., Willard, D.A., Edwards,
L.E., and Holmes, C.W.
1997. Preliminary paleontologic report on cores 19A and 19B,
from Russell Bank, Everglades National Park, Florida
Bay. U.S. Geological Survey Open-file Report 97-460,
29 pp.
Brewster-Wingard, G.L., Ishman, S.E., Waibel, N.J., Willard,
D.A., Edwards, L.E., and Holmes, C.W.
1998a. Preliminary paleontologic report on Core 37, from Pass
Key, Everglades National Park, Florida Bay. U.S. Geo-
Brewster-Wingard, G.L., Ishman, S.E., and Holmes, C.W.
1998b. Environmental Impacts on the southern Florida Coastal
Waters: A history of change in Florida Bay. Journal of
Coastal Research, special issue 26, pp. 162-172.
Brewster-Wingard, G.L. and Ishman, S.E.
1999. Historical trends in salinity and substrate in central Flor-
ida Bay: A Paleoecological Reconstruction using modern
analogue data. Estuaries, vol. 22, no. 2b, pp. 369-383.
Camp, D.K., Lyons, W.G., and Perkins, T.H.
1998. Checklists of selected shallow-water marine invertebrates
of Florida. Florida Department of Environmental Protec-
tion, Florida Marine Research Institute Technical Report
TR-3, 238 pp.
Dame, R.F.
1996. Ecology of marine bivalves: an ecosystem approach. CRC
Press, Boca Raton, 254 pp.
Hazel, Joseph E.
1977. Use of certain multivariate and other techniques in assem-
blage zonal biostratigraphy: examples utilizing Cambrian,
Cretaceous, and Tertiary Benthic Invertebrates. in Con-
cepts and Methods in Biostratigraphy. E.G. Kauffman and
J.E. Hazel, eds., Dowden, Hutchinson, & Ross, Inc.,
Stroudsburg, PA, pp. 187—212.
Holmes, C.W., Robbins, J., Halley, R., Bothner, M., Ten Brink,
M., and Marot, M.
2001. Sediment Dynamics of Florida Bay Mud Banks on a De-
cadal Time Scale. Bulletins of American Paleontology,
no. 361, pp. 31—40.
Ishman, S.E., Brewster-Wingard, G.L., Willard, D.A., Cronin,
T.M., Edwards, L.E., and Holmes, C.W.
1996. Preliminary paleontologic report on Core T-24, Little Ma-
deira Bay, Florida. U.S. Geological Survey Open-File
Report 96-543, 27 pp.
Johns, E., Wilson, D.W., and Lee, T.N.
1999. Surface salinity variability of Florida Bay and southwest
Florida coastal waters. Program and Abstracts, 1999 Flor-
ida Bay and Adjacent Marine Systems Science Confer-
ence, pp. 169-171.
MOLLUSKS PAST AND PRESENT: BREWSTER-WINGARD ET AL. 231
Kovach, W.L.
1989. Comparisons of multivariate analytical techniques for use
in pre-quaternary plant paleoecology. Review of Palaeo-
botany and Palynology, vol. 60, pp. 255-282.
1995. Multivariate Data Analysis. in Statistical Modeling of
Quaternary Science Data, D. Maddy and J.S. Brew, eds.,
Quaternary Research Association, Technical Guide, no. 5,
pp. 1-38.
Ludwig, J.A. and Reynolds, J.F.
1988. Statistical ecology: a primer on methods of computing.
John Wiley & Sons, New York, 337 pp.
Lyons, W.G.
1995. Mapping Florida Bay benthic assemblages: using mol-
lusks to assess faunal change. Programs and Abstracts,
Florida Bay Science Conference, October 1995, Gaines-
ville, Florida, Florida Sea Grant, University of Florida,
pp. 167-169.
1996. An assessment of mollusks as indicators of environmental
change in Florida Bay. Programs and Abstracts, Florida
Bay Science Conference, December 1996, Key Largo,
Florida, Florida Sea Grant, University of Florida, pp. 52—
54.
1998. Florida Bay molluscan community dynamics: shifting
zones in a changing world. Proceedings, Florida Bay Sci-
ence Conference, May 1998, Miami, Florida, Florida Sea
Grant, University of Miami.
1999. Responses of benthic fauna to salinity shifts in Florida
Bay: evidence from a more robust sample of the mollus-
can community. Program and Abstracts, Florida Bay and
Adjacent Marine Systems Science Conference, November
1999, Key Largo, Florida, pp. 47—5S0.
Shannon, C.E., and Weaver, W.
1949. The mathematical theory of communication. University
Illinois Press, Urbana, IL.
Swart, P. K., Healy, G. F., Dodge, R. E., Kramer, P., Hudson,
J. H., Halley, R. B., and Robblee, M. B.
1996. The stable oxygen and carbon isotopic record from a coral
growing in Florida Bay: a 160 year record of climatic and
anthropogenic influence. Paleogeography, Palaeoclimatol-
ogy, Palaeoecology, vol. 123, pp. 219-237.
Turney, W.J. and Perkins, B.F.
1972. Molluscan distribution in Florida Bay. University of Mi-
ami, Miami, Florida, Sedimenta, vol. III, 37 pp.
Wingard, G.L., Ishman, S.E., Cronin, T.M., Edwards, L.E., Wil-
lard, D.A., and Halley, R.B.
1995. Preliminary analysis of down-core biotic assemblages:
Bob Allen Keys, Everglades National Park, Florida Bay.
U.S. Geological Survey Open-file Report 95-628, 35 pp.
x c
Sy »
a
ile :
t _ mi av “Sen
mil 12 nt ‘ ar A eB feet ila
Z ere 0u ae a
y i
ea ee ) sine . | ait 2
e : es
| r rs ath ae jo we) Ones fi
oe aod } a
a 2A . - ée tis o 7 - Tey o6f2
» « : , “ ; : i ikre® @ A
+~ nes : va
7 ¢ 4 il rea ieee
a
=,
F eabte 1) AA
: y % nei ‘jm Goren O86 .
ss a ie
e) a Le. jhe a 7
- a pou ‘as ian!’ @F ema
te PAE peter | Mie. Se
; : ; ‘ i , v= f > inn ee AR a) <, .
,
oa(tm ae —s |
¥ > ta » Se” ' - :
fh .6i'S,_ eae :
> he) mM 52215) 52-15) 0:00
Central Biscayne Bay
SPO1 25 39'28 80 16'06 30.25 29.50 4.18 393 SHO 10:23) 150.00 16.83 33.65 50.48 0.00
BB22 25 45'21 80 10'28 30.00 33.80 5.60 458 1.90 0.22 19 0.00 0.00 42.28 42.28 0.00
BB26 25 44'50 80 11'06 29'S 33:40) 555 473 200 O13 26096 8.04 28.30 36.33 0.00
BB34 25 39'03 80 15/33 30.25 32.00 4.34 419 2ALOR OR 200.00 0.00 0.00 0.00 0.00
BB35 25 38/37 80 11/30 30.01 37.00 5.71 449 10) 0:08) 25) 10:00 4.69 0.00 4.69 0.00
BB36 25 35'58 80 14’11 2990) 37-40) 5:82) 4395230) ORS 26 ~=—-0..00 0.00 0.00 0.00 0.00
BB37 25 34'13 80 11/31 29 S45 S/-20) SO 45 1.80 0.08 29 0.00 0.00 0.00 0.00 0.00
South Biscayne Bay
BB41 25 28'20 80 17'04 29716 S85 08 O04 SiO 2 aOR ON 2 ee 0.00 0.00 0.27 0.27 0.00
BB42 25 27/23 80 11'59 PENG) ifesK0) SHO} ISS) 1.30 0.24 9 0.00 0.00 0.00 0.00 0.00
BB44 25 24'02 80 15'20 29.67 37.60 5.53 448 ZO O85 5 267510200, 0.00 0.33 0.33 0.00
BB45 25 22'09 80 16/48 29.60 37.40 5.37 458 2.40 0.11 21 0.00 0.00 0.00 0.00 0.00
BB48 25 18'29 80 20'56 29373240 5:38) Als 2.60 0.14 24 0.00 0.32 0.63 0.95 0.00
BBS51 25 15'04 80 24'51 DIZ 2220) ee se 444 1.50 0.26 12 OL00) 0.00 25.00 25.00 0.00
fig. 2; SFWMD, 1994) occur in Central Biscayne Bay.
These are supported on substrates including calcareous
and quartz sands, calcareous mud, and organic-rich
muds (Wanless, 1976). Salinities within Central Bis-
cayne Bay range from 32.0 to 37.4 ppt in the open-
bay and 29.5 ppt at the discharge of Snapper Creek
Canal (Text-fig. 3; Table 1). Average August salinity
since 1988 in the open-bay is 35.90 ppt. The coastal
regions of Central Biscayne Bay vary significantly de-
pendent on point-source freshwater run-off. The north-
ern part of Central Biscayne Bay is strongly influenced
by the Miami River, which accounts for the high tur-
bidity, high nutrient, and high pollutant levels in this
region. Further south, Snapper Creek, Coral Gables
Waterway, and Cutler Drain have been identified as
pollutant point-sources (SFWMD, 1994). However,
flushing of these regions occurs on a regular basis due
to Government and Norris Cuts. The southern part of
Central Biscayne Bay becomes more pristine and in-
cludes Biscayne National Park. Significant effects on
the ecosystem in this region are localized, many relat-
ed to watercraft use such as sewage and solid waste
disposal, fuel leakage and _ spillage,
scouring of seagrass beds.
South Biscayne Bay includes the southern portion
of Biscayne National Park and the northern part of the
Florida Keys National Marine Sanctuary (Barnes and
Card Sounds; Text-fig. 1). Sediments in South Bis-
cayne Bay include non-tidal mud banks, calcareous
mud, and sands (Wanless, 1976), and they support sea-
grass and seagrass with hard-bottom matrix commu-
nities (Text-fig. 2; SFWMD, 1994). Salinities range
from between 38.5 and 37.3 ppt north of Card Sound,
to less than 30 ppt in Barnes Sound with average Au-
gust salinity since 1989 in Barnes Sound at 28.80 ppt
(Text-fig. 3; Table 1). Although not as severely af-
fected by urbanization as the North and Central Bay,
Southern Biscayne Bay is affected by channelized
freshwater input and nutrient enrichment from the ca-
nal systems. In addition, Card and Barnes Sounds are
very restricted, thus reducing their flushing cycles.
Other factors that influence the ecosystem of South
Biscayne Bay are pollutants from adjacent landfills,
and propeller
BENTHIC FORAMINIFERA: ISHMAN 237
Table 1.—Extended.
Boli-
Arti- Astro- Boli- __ Boli- vina Boliv- Buli- Buli-
Archaias culina Asti- nonion — vina vina pseudo- ina mina Buli- minella_ Cassi-
angula- mucro- gerina _ stella- low- pseudo- punc- Boliv- stria-_ margin- mina elegan- dulina Cibi- Clavu-
Sample tus nata sp. tum mani plicata tata ina sp. tula ata striata tissima sp. cides sp.lina sp.
North Biscayne Bay
BBO2 0.00 0.00 0.00 0.00 0.31 0.94 5.66 0.00 4.09 1.89 0.00 1.57 0.00 0.63 0.00
BBO3 0.00 0.00 0.00 0.00 0.63 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
BBO4 15292 0.00 2.42 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
BBOSA 0.00. 0.00 0.00 10.47 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00.
BBO9 0.00 0.00 0.00 0.00 0.29 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
BB1O 2.24 eT) 0.00 0.00. 0.00 0.45 0.90 0.00 0.45 0.00 0.00 0.45 1.35 0.00 0.00
BBI1 0.00 0.85 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
BB14 0.00 1.88 0.00 0.00 0.00 0.00 0.00 0.00 1.25 0.63 0.00 0.00 0.00 0.00 0.00
BBIS5 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
BB17 0.33 0.99 0.00 0.00 0.00. 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Central Biscayne Bay
SPO1 0.00 0.00 0.63 0.00 0.32 0.00 0.00 0.32 0.00 0.00 0.00 0.32 0.00 0.00 0.00
BB22 0.00 1.68 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
BB26 0.00 0.64 0.00 0.00 1.93 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
BB34 12.58 3.23 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.65
BB35 12.64 2.17 1.08 0.00 0.00 0.00 0.00 0.00 1.81 0.00 0.00 0.36 0.00 0.00 0.00
BB36 32.23 0.66 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.33 0.00 0.00 0.00 0.00
BB37 18.18 4.17 0.00 0.00 0.38 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.89
South Biscayne Bay
BB41 26.08 0.27 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
BB42 45.23 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
BB44 32.89 1.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.33
BB45 20.86 1.71 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2.00
BB48 1.59 4.76 0.00 0.00 0.32 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
BBS1 0.00 0.82 0.00. 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
the Turkey Point power facility, and propeller scour in METHODOLOGY
shallow regions.
An additional consideration with respect to the Bis-
cayne Bay ecosystem is the impact of ground water.
Ground water seepage at the coastal margins and from
subsurface springs has been noted historically (Kohout
and Kolipinski, 1967). This acts as an additional
source of fresh water, but also may provide an addi-
tional source of contaminants by pollutant-enriched
ground water. A significant source of groundwater dis-
charge is apparent along the coastal margin of Central
Biscayne Bay (V. Quinones, USGS, personal com-
munication).
The factors listed above play a primary role in the
distribution of aquatic communities within the Bis-
cayne Bay ecosystem (SFWMD, 1994). The identifi-
cation of the significant factors in controlling the dis-
tributions of the Bay’s biogenic components is critical
for our interpretation and understanding of the evolu-
tion of the Biscayne Bay ecosystem. This paper pre-
sents the modern distributions of benthic foraminifera
and their environmental controls within Biscayne Bay.
A systematic collection of surficial sediment sam-
ples from Biscayne Bay was conducted in collabora-
tion with Metropolitan Dade County Department of
Environmental Resources Management (DERM) dur-
ing the months of June and August, 1996. A total of
twenty-three localities (Text-fig. 1; Table 1) were sam-
pled for surficial sediments using an Eckman Grab
sampler. These sites have been sampled monthly start-
ing in 1988 (DERM Bay Run Program) for a variety
of water-quality variables of which the following are
reported herein: salinity, temperature, dissolved oxy-
gen, and oxygen-reduction potential (redox) (Table 1).
Approximately 100 cubic centimeters (cc) of the up-
per two centimeters of the grab samples were disag-
gregated and sieved through a 63 um sieve. The great-
er than 63 jm size fraction was recovered, dried at
50°C, and analyzed for foraminifera. More than 300
foraminifer specimens were picked randomly using a
45 compartment picking tray and random number ta-
ble. Samples not producing greater than 300 specimens
N
Ww
oo
Table 1.—Extended.
BULLETIN 361
Elphi-
Elphi- Elphi- dium Globo-
Cribro- Cyclo- dium — Elphi- dium galve- Elphi- Flori- Gaud- cassi- Milio- Milio-
stom- — gyra_ Discor- cf. dium galve- stonense dium lus Fursen- ry- dulina linella linella Milio-
oides — plan- bis adven- delica- sonense’ mexi-_ poey- auri-_—_koina ina subglo- circu- fich- __ linella
Sample sp. orbis — mira um tulum typicum) canum anum _— culus sp. exilis bosa laris teliana labiosa
North Biscayne Bay
BBO2 0.00 1.26 4.72 0.00 2.52 0.00 0.94 0.00 0.00 2.20 2.83 1.26 1.26 0.00 0.94
BBO3 0.63 0.00 0.00 3.80 0.00 7.28 5.70 0.32 0.95 0.00. 0.32 0.00 0.00 0.00 0.00
BBO4 0.00 0.00 1.38 0.35 0.00 0.00 14.88 0.35 0.35 0.00 0.69 0.00 0.00 0.00 2.08
BBOSA 0.00 0.00 0.00 0.00 0.00 0.00 21.80 0.00 0.87 0.00 0.00 0.00 0.00 0.00 0.00
BBO9 0.00 0.00 0.00 0.00 0.00 0.00 37.97 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
BB10 0.00 0.00 4.48 0.00 8.07 0.00 0.00 0.45 0.00 0.90 0.00 0.45 0.00 0.00 0.00
BBI1 0.00. 0.00 0.00 0.00 0.00 0.28 37.18 0.00 0.00 0.00 0.00 0.00 0.85 0.00 0.00
BB14 0.00 0.00 0.00 0.00 0.00 2.19 35.74 0.00 0.00 0.00 0.00 0.00 0.31 0.00 0.94
BBI5 0.00 0.00 0.00 0.00 1.81 0.00 25.68 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.81
BB17 0.00. 0.00 0.00 0.00. 0.00 0.00 24.75 0.00 0.00 0.00 0.00 0.00 0.00 0.00 10.56
Central Biscayne Bay
SPO1 0.00 0.00 0.00 0.00 0.00 25.71 1.27 0.00 0.00 0.00 0.00 0.00 0.32 0.00 0.00
BB22 0.00 0.67 0.00 0.00 0.00 6.04 16.11 0.00 0.67 0.00 0.00 0.00 1.34 0.00 0.00
BB26 0.00 0.00 0.32 0.00 1.61 9.65 13.18 0.00 2.89 0.32 0.00 0.00 1.61 0.00 0.64
BB34 0.00 0.00 0.00 2.26 0.00 0.00 0.00 0.65 0.00 0.00 0.00 0.00 0.65 0.65 1.61
BB35 0.00 0.00 0.00 0.00 0.00 10.83 1.81 0.00 1.81 0.00 0.00 0.36 0.00 0.00 0.00
BB36 0.00 0.00 0.66 0.33 0.00 0.00 2.99 0.00 0.00 0.00 0.00 0.00. 2.99 0.00 1.33
BB37 0.00 0.00 0.00 0.38 0.00 0.00 0.00 1.52 0.00 0.00 0.00 0.00 8.33 0.38 0.00
South Biscayne Bay
BB41 0.00 0.00 0.00 0.27 0.00 0.00 2.42 1.08 0.00 0.00 0.00 0.00 11.83 0.00 0.00
BB42 0.00 0.00. 0.00 0.00 1.41 0.00. 0.00 0.00 0.00 0.00 0.00 0.00 1.41 0.00 0.35
BB44 0.00 0.00 1.00 0.00 2.33 0.00 1.00 0.66 0.00 0.00 0.00 0.00 3.99 0.00 0.00
BB45 0.00 0.00 0.00 1.43 8.86 0.00 0.00 2.00 0.00 0.00 0.00 0.00 DEST 0.00 0.86
BB48 0.00 0.00 0.00 0.63 27.30 0.00 Bale), 0.00 0.00 0.00 0.00 0.00 13.02 0.00 0.00
BBS1 0.00. 0.00 0.00 0.00 0.00 0.00 41.80 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.23
were picked of all their specimens and used in this
study if they exceeded 50 counts (Table 1). The fora-
minifera were identified based primarily on Bock
(1971) and Poag (1981), counted, and species counts
were standardized to percent of total assemblage (Ta-
ble 1) and used as such in the quantitative analyses.
Because the results of the analyses presented here are
the basis on which future downcore paleoenvironmen-
tal analyses will be interpreted, total assemblage data
is presented. This is consistent with the findings of
Scott and Medioli (1980) who concluded that total as-
semblage data more accurately represent prevailing
modern conditions and are more useful for paleoen-
vironmental studies.
The foraminiferal data were analyzed using two sta-
tistical techniques, hierarchical cluster analysis and
principal components analysis (PCA). The hierarchical
clustering method, complete linkage of the Pearson
correlation similarity coefficient was applied to the fo-
raminiferal data to identify distinct sample groupings
(Q-mode) and foraminiferal assemblages (R-mode). Q-
mode varimax principal component analysis of the
Pearson correlation similarity matrix was used to de-
termine the robustness of the foraminiferal groupings
defined in the cluster analyses and to quantify the data
in multi-dimensional space, which places environmen-
tal constraints on the faunal distributions.
BENTHIC FORAMINIFERAL RESULTS
A total of 69 taxa of benthic foraminifers common
to the North American southeast coast and Gulf of
Mexico were identified from the Biscayne Bay modern
sediment samples (Table 1). Species diversity, as mea-
sured using Simpson’s index (Simpson, 1949), ranged
from 0.080 to 0.493. Simpson’s index is useful in that
it identifies dominance patterns within the data. Com-
bined with the number of species present the Simp-
son’s index should reflect low diversity, high domi-
nance assemblages associated with stress conditions
and low dominance, high diversity assemblages asso-
ciated with normal marine conditions.
Rotaliid forms dominate the foraminiferal assem-
blages with agglutinated taxa constituting a minor
component in most of the assemblages. Species dom-
inance varies considerably throughout the Bay with
Ammonia parkinsoniana and Elphidium galvestonense
BENTHIC FORAMINIFERA: ISHMAN 239
Table 1.—Extended.
Quin- — Quin- Quin-
Nodo- Nubecu- Patel- Pseudo- Pyrgo quelo- quelo- Quin- — quelo-
bacular- Nodo- Nonio- laria lina Penero- clave- Pyrgo sub- culina culina — quelo- — culina
iella sarl- nella luci- corru- plis Plank- lina ~~ denti- Pyrgo_ sphaer- agglu- bos- culina _ poly-
Sample sp- idae sp. fuga gata proteus tonic. gracilis culata serrata ica tinans ciana poeyana gona
North Biscayne Bay
BBO2 0.00 0.94 8.18 0.00 0.31 0.00 0.00 0.00 0.00 0.00 0.00 0.00 4.72 1.26 0.00
BBO3 0.00 0.00 0.00 0.00 0.00 0.00 0.63 0.00 0.00 0.00 0.32 0.00 0.63 0.63 0.63
BBO4 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 AG Sita 1.73 0.35 2.08
BBOSA 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2.03 0.29 1.45 0.87 0.00
BBO9 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.29 0.29 1.45 0.29
BB1O 0.00 0.00 0.00 0.00 0.00 0.45 0.00 0.00 0.00 0.00 0.00 1.35 5.38 2.24 0.90
BBI1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.28 0.85 0.00 0.00 0.00
BB14 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.94 0.00 0.00 0.31
BBI5 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.30 0.91 0.00 1.21 0.00
BB17 0.00 0.00 0.00 0.00 0.33 0.00 0.00 0.00 0.00 0.00 0.00 0.66 0.33 0.66 0.00
Central Biscayne Bay
SPO1 0.00. 0.00 0.00. 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.95
BB22 0.00 0.00 0.00 0.00 0.00 1.68 0.00 0.00 0.00 0.00 4.03 235 5.70 1.34 0.00
BB26 0.00 0.00 0.00 0.00 0.00 0.32 0.32 0.00 0.00 0.32 1.29 0.64 6.11 4.18 By Py)
BB34 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2.26 0.00 6.13 3.87 1.61 3.87 3.87
BB35 0.00 0.00 0.00 0.36 0.00 0.00 0.00 0.00 0.00 0.00 0.72 6.14 9.03 8.66 5.05
BB36 0.00 0.00 0.00 0.00 0.00 2.66 0.00 0.00 0.00 0.33 0.33 2.99 3.32 10.63 7.97
BB37 0.00 0.00 0.00 0.00 0.00 7.58 0.00 1.52 0.38 0.76 0.76 3.41 3.41 3.03 3.79
South Biscayne Bay
BB41 0.00 0.00 0.00 0.00 0.00 1.61 0.00 0.00 0.27 0.00 25 2.69 4.03 5.38 8.06
BB42 0.00 0.00 0.00 0.00 0.00 8.13 0.00 0.00 0.00 0.35 2.83 8.48 2.83 2.83 4.59
BB44 0.33 0.00 0.00 0.00 0.00 1.00 0.00 0.00 0.33 0.00 199 6.31 S52 6.31 3.65
BB45 0.00 0.00 0.00 0.00 0.00 2.00 0.00 0.00 0.00 0.00 3.43 2.00 7.43 6.86 6.86
BB48 0.00 0.00 0.32 0.32 0.00 0.32 0.00 0.00 0.63 0.00 0.32 0.63 18.41 8.89 2.54
BBS1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.23 4.51 0.00 TAS 2.46
constituting over 50% of the fauna in much of North-
ern Biscayne Bay and Barnes Sound where the Simp-
son’s index range from 0.114 to 0.493. Conversely,
Archaias angulatus is dominant (up to 45%) in sam-
ples collected from the open regions of Biscayne Bay
where Simpson’s index averages ~0.011. Other cal-
careous hyaline forms include Bolivina spp., Elphi-
dium delicatulum, Rosalina floridana, and R. globu-
laris. Common miliolids include Articulina mucrona-
ta, Miliolinella circularis, M. labiosa, Pyrgo sub-
sphaerica, Quinqueloculina agglutinans, Q. bosciana,
Q. poeyana, Q. polygona, Q. seminula, Q. tenagos,
and Triloculina spp. Soritids present are Archaias an-
gulatus, Peneroplis proteus, and Sorites marginalis.
Agglutinated taxa in Biscayne Bay include Ammoba-
culites sp., Clavulina tricarinata, Pseudoclavulina
gracilis, Textularia candeiana, T. conica, and Tro-
chammina spp.
Results of the quantitative analyses of the benthic
foraminifer data are shown in Text-figure 4 and Table
2. Q-mode cluster analysis defines two sample groups
or biotopes (Text-fig. 4a) with two sample outliers.
Biotope one consists of samples from Central and
northern South Biscayne Bay (BB34, BB35, BB36,
BB37, BB41, BB42, BB44 and BB45; Text-fig. 4a)
and represent open-marine conditions. The second bio-
tope consists of samples from restricted environments
or regions greatly effected by point-source freshwater
input (SPO1), North Biscayne Bay, and Barnes Sound
(BB3, BB4, BB5A, BB9, BB10, BB11, BB14, BB1I5,
BB17, BB22, BB26, BB51 and SPO1; Text-fig. 4a).
One outlier sample, BBO2, is from Dumbfoundling
Bay, the northernmost sample locality. Sample BB48
is the second outlier and is from Card Sound, Southern
Biscayne Bay.
R-mode cluster and principal components (PC) anal-
yses result in the definition of four major groups of
benthic foraminifers or assemblages (Text-fig. 4b). The
first group, Cluster 1, contains several significant spe-
cies including Peneroplis proteus, Triloculina spp.,
Miliolinella circularis, Articulina mucronata, Quin-
queloculina subpoeyana, and Q. bosciana (Text-fig.
4b) that have high positive PC1 and PC4 loadings (Ta-
ble 2). Cluster 2 contains Quinqueloculina polygona,
Q. agglutinans, Q. tenagos, Q. poeyana, Archaias an-
gulatus, and Rosalina floridana that have high positive
240
Table 1.—Extended.
BULLETIN 361
Quin- Spiro-
Quin- quelo- Quin- Rosa- —_Rosa- sigmot- Stain- Trilo- Trilo- Trilo-
quelo-culina’ quelo- Recto- Recur- lina lina lina Spirolo- Sorites forthia culina culina Trilo- culina
culina sub- culina bolivina voides flori-— globu- antil- culina margin- compla- line-__ plac-_culina _ tricar-
Sample seminula poeyana tenagos advena _ sp. dana laris larum sp- alis nata iana iana_ rotunda inata
North Biscayne Bay
BBO2 27.67 0.00 0.00 0.00 0.00 O75 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2.20
BBO3 4.11 0.00 6.96 0.32 0.32 222. 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2.22
BBO4 0.69 0.00 1°73 0.00 0.00 2.08 0.00 0.00 0.00 0.00 0.00 0.35 0.00 0.69 8.30
BBOSA 1.45 0.00 0.58 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
BBO9 2.32 0.00 1.45 0.00 0.00 0.29 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 5:22
BB1O 17.94 0.00 4.48 0.00 0.00 6.73 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.45 3.59
BBII1 0.85 0.00 3.94 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.41 3.94
BBI4 3.45 0.00 4.39 0.00 0.00 0.63 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.31 0.63
BBI15 1.81 0.00 0.30 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.30 0.00
BBI7 0.00 0.00 AES) | 0.00 0.00 0.00 132 0.33 0.00 0.00 0.00 0.00 0.00 0.00 4.62
Central Biscayne Bay
SPO1 13.02 0.00 0.95 0.00 0.00 2.54 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 ep)
BB22 7.38 0.00 1.34 0.00 0.00 2.35 0.00 0.00 0.00 0.34 0.00 0.67 0.00 1.34 2.68
BB26 6.43 0.00 0.32 0.00 0.00 1.61 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 4.50
BB34 0.00 0.00 8.71 0.00 0.00 20.97 0.00 0.32 0.00 0.00 0.00 0.65 0.00 0.00 16.13
BB35 7.58 0.00 0.36 0.00 0.00 6.50 0.00 0.00 0.00 0.36 0.72 1.08 1.44 1.08 12.64
BB36 13.62 0.00 1.00 0.00 0.00 1.99 0.00 0.00 0.66 0.33 0.00 0.33 0.00 0.66 8.64
BB37 0.76 1.89 9.09 0.00 0.00 9.47 0.00 1.89 1.14 0.76 0.00 7.20 1.89 1.14 1.14
South Biscayne Bay
BB41 8.87 0.00 1.88 0.00 0.00 4.57 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.27 12.90
BB42 8.13 0.00 2.12 0.00 0.00 2.12 0.00 0.00 0.00 0.35 0.00 0.00 0.35 0.00 1.06
BB44 4.65 0.00 2.33 0.00 0.00 2.33 0.00 0.00 0.00 0.00 0.00 0.66 0.00 2:33 12.96
BB45 5.71 0.00 2.29 0.00 0.00 3.71 0.00 0.00 0.00 0.29 0.00 0.00 0.00 0.29 17.71
BB48 2.54 0.00 0.32 0.00 0.00 5.71 0.00 0.95 0.00 0.00 0.00 0.00 0.00 2.86 3.49
BBS51 10.25 0.00 0.00 0.00 0.00 2.87 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.64
PC3 and PC4 loadings (Text-fig. 4b, Table 2). The
third cluster includes Bolivina spp., Bulimina margin-
ata, Buliminella elegantissima, and Fursenkoina sp.
that have high PC2 loadings (Text-fig. 4b, Table 2).
The fourth cluster is composed predominantly of Am-
monia parkinsoniana, Elphidium galvestonense, and
Ammobaculites exiguus that are represented by high
negative PC3 and PC4 loadings (Text-fig. 4b, Table 2).
In the R-mode principal components analysis the
first four principal components explain 94 percent of
the variance in the data set (Table 2). High positive
and negative (greater than 0.5 and —0O.5) principal
component one (PC1) and principal component four
(PC4) scores were obtained by foraminiferal species
common to normal marine salinities and found on the
continental shelf of the southeastern seaboard of the
United States (Schnitker, 1971). Several of these spe-
cies also are associated with epiphytal habitats. Spe-
cies with high principal component two (PC2) scores
are common to areas with high surface productivity
and organic-rich, fine-grained sediments. Principal
component three (PC3) contains species with high pos-
itive and negative scores that have specific salinity
preferences. High negative scores are associated with
the species Ammonia parkinsoniana and Elphidium
galvestonense, taxa associated with salinities ranging
from 5 to 25 parts per thousand (ppt) in the south
Florida region. Foraminifera associated with subtrop-
ical to tropical marine conditions (salinities ~ 35 ppt)
have high positive PC3 scores.
DISCUSSION
Few studies have considered the distribution of fo-
raminifera within Biscayne Bay. Bush (1958) present-
ed a descriptive account of benthic foraminifers from
Biscayne Bay by defining thirteen biotopes. Cole
(1972) observed benthic foraminifera from a specific
locality and attributed test deformation within benthic
foraminiferal populations to thermal stress associated
with power plant discharge. Benthic foraminiferal
studies from Florida Bay (Bock, 1971; Rose and Lidz,
1977: Lidz and Rose, 1989) showed foraminiferal dis-
tributions associated with physiographic features and
salinity conditions within the Bay. Results of the ben-
thic foraminiferal analyses presented herein indicate
BENTHIC FORAMINIFERA: ISHMAN 241
Table 1.—Extended.
Textu- Textu- Trocham- Trocham- Valvulin-
laria laria mina Trocham- mina Unident- Uvigerina Valvulina eria Wiesner-
Sample candeiana — conica inflata mina sp. squamata ified sp. sp. laevigata ella sp. Counts
North Biscayne Bay
BBO2 0.31 0.00 0.00 0.00 0.00 0.00 0.94 0.00 0.00 0.63 318
BBO3 0.00 0.00 6.96 0.00 0.00 0.00 0.00 0.00 0.00 0.00 316
BBO4 DTT 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 289
BBOSA 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 344
BBO9 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 345
BB1O 0.00 0.45 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 223
BBI1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 855
BB14 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 319
BBI5 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 331
BB17 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 303
Central Biscayne Bay
SPOL 0.00 0.00 0.00 0.00 0.32 0.00 0.00 0.00 1.59 0.00 315
BB22 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 298
BB26 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.64 311
BB34 0.00 0.00 0.00 0.00 0.00 0.00 0.00 9.35 0.00 0.00 310
BB35 0.00 0.72 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Die
BB36 0.00 0.66 0.00 0.00 0.00 0.00 0.00 1.00 0.00 133 301
BB37 0.00 3.79 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 264
South Biscayne Bay
BB41 0.00 0.00 0.00 0.27 0.00 0.00 0.00 4.84 0.00 0.00 372
BB42 0.00 0.00 0.00 0.00 0.00 0.00 0.00 7.42 0.00 0.00 283
BB44 0.00 0.00 0.00 0.00 0.00 0.33 0.00 ja 0.00 0.33 301
BB45 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.14 0.00 0.00 350
BB48 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 315
BBS1 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.41 0.00 0.00 244
distinct distribution patterns associated with salinity
and productivity.
The easiest explained grouping is Cluster 4, with
species having high negative PC3 and PC4 scores, Am-
monia parkinsoniana, Elphidium galvestonense, and
Ammobacculites sp. are interpreted as being controlled
by salinity. Studies from Florida Bay (Rose and Lidz,
1977; Lidz and Rose, 1988; Brewster-Wingard and
Ishman, 1999) and the Gulf Coast (Poag, 1981) indi-
cate that these taxa are dominant in low salinity (<25
ppt) environments. These taxa are dominant in the
samples that represent Biotope 2, samples from North
Biscayne Bay, the outflow from Snapper Creek
(SPO1), and Barnes Sound, areas of restricted environ-
ments or near point-source freshwater input (Text-fig.
5) where salinities less than 35 ppt persist throughout
the year. The distribution of these taxa in Biscayne Bay
is consistent with their distribution in Florida Bay and
other regions.
Conversely, Clusters | and 2 are composed of taxa
associated with more normal marine conditions (salin-
ities ~35 ppt), Articulina mucronata, Archaias angu-
latus, E. delicatulum, M. circularis, P. proteus, Q. po-
lygona, Q. agglutinans, Q. bosciana, Q. poeyana, Q.
subpoeyana, Q. tenagos, T. lineiana, T. rotunda, T.
tricarinata, and S. antillarum. These taxa comprise the
assemblages found in Biotope | from Central and
northern South Biscayne Bay (Text-fig. 5). Several of
these taxa, A. mucronata, T. lineiana, Q. bosciana, Q.
poeyana, and Q. polygona in Cluster | are commonly
associated with normal marine conditions on the
southeastern U.S. continental shelf (Schnitker, 1971;
Murray, 1991). In addition, at least two of these spe-
cies, M. circularis and P. proteus, have epiphytal hab-
itats in lagoonal environments (Brasier, 1975a). The
species representing these two clusters have high pos-
itive PC1, PC3, and PC4 scores. This indicates that the
distribution of these species is partially controlled by
normal marine salinity and the presence of seagrass.
However, it may also be that the distinction between
Clusters | and 2 represents a separation between a
Central Biscayne Bay assemblage (1), influenced by
an inflow of Atlantic shelf water, and a more southerly
assemblage (2) with greater affinities to higher salinity
assemblages found in Florida Bay. This pattern is con-
sistent with the molluscan distributions in Biscayne
242 BULLETIN 361
Dumbfoundling
Bay
Salinity (ppt)
REBE
1988 1989 1990 1991 1992 1993 1994 1995 1996
Yes North Bay
Port of Miami
Salinity (ppt)
24 _
1988 1989 1990 1991 1992 1993 1994 1995 1996 Central Bay
Year
South Bay
ca
=
=
2
=
Ss
a
1 E
1988 1989 1990 1991 1992 1993 1994 1995 1996 &
Year
ATLANTIC OCEAN
Text-figure 3—Map of the Biscayne Bay region of South Florida showing August 1988 through 1996 salinity distributions in parts-per-
thousand (ppt). The salinity contours represent August 1996.
BENTHIC FORAMINIFERA: ISHMAN 243
BB34 B. striata
BB3S Sea
BB45 @) See ae
= » graciis
1B
aon r T.conica Cluster 1
Bioto e 1 pirosigmoilina sp
BB36 P P ipretei
BB42 A, mur ata
% antillarum
M. Le
ae ected
BB48 Q. bosciana
BBO2 x 2 eae
ricarinata
SPO1 T tricarinata
Q. polygona
BBO3 a) poeyanaltendeos
/ ti
BB26 Q. agelutinads
BB22 La ese Cluster 2
3510 | hii
BBO4+ fa | Biotope 2 Hees a
BBSA | | Q. seminula inxs acta eaten :
BBIS iBone: Cluster 3
BB17 (alae ee Bicpeantisiime
BBO9 & | | | Ammobatulites sp. ~ ,
BB1I ® | een LES REA) ma Cluster 4
BBI4 & | sas sae au aio: |
| . globularis isso D)
S33 |) Aupartnconicna BANNER NESE as
SS aa SS Se. el E. galvestonenee
= a | nen me ie U
5 : 5)
ge ees FO : 0.0 0.5 1.0 LS 2.0
Distances * Distances
a
Text-figure 4.—Dendrograms showing the results of the (a) Q-mode and (b) R-mode cluster analyses.
Bay that show much stronger affinities to Atlantic mar-
gin faunas than Florida Bay faunas (Brewster-Win-
gard, USGS, personal communication).
Cluster 3 contains an assemblage that has high PC2
scores (Table 2), is restricted to one sample from Bis-
cayne Bay, BBO2, and represents one of the outliers
on the Q-mode cluster (Text-fig. 4a). It is distinguished
by the only occurrence of Bulimina marginata, Boli-
vina spp., and Fursenkoina sp. in Biscayne Bay. Mor-
phometric studies (Bernhard, 1986; Corliss and Chen,
1988) indicate that bolivinids are associated with or-
ganic rich, low oxygen environments. Their distribu-
tions are commonly associated with organic-rich sed-
iments, as is B. marginata in the Gulf of Mexico (Phle-
ger, 1951), South Atlantic (Mackensen ef al., 1993),
and eastern Pacific (Phleger and Soutar, 1973). Sample
BBO2 is from the northernmost part of Biscayne Bay,
Dumbfoundling Bay. This region is characterized by
high chlorophyll-a values, particularly at site BBO2.
Canal input to North Biscayne Bay contains the high-
est concentrations of dissolved nitrates and phosphates
in all of Biscayne Bay (Lietz, WRD, USGS personal
communication). Other data indicate that North Bis-
cayne Bay is a region of high organic sedimentation
(Wanless er al., 1984). Wanless et al. (1984) show the
highest concentration of diatoms in North Biscayne
Bay for a period from March 1982 through August
1983, with concentrations exceeding 600 10*/liter
during November 1982. These values are an order of
magnitude greater than those observed in South or
Central Biscayne Bay, and are somewhat reflected in
the high weight percent total organic content (up to 29
percent) reported for North Biscayne Bay sediments
(Wanless et al., 1984). However, the coastal regions in
Central and South Biscayne Bay are also rich in or-
ganic content (vegetative material) therefore making it
necessary to distinguish organics from phytoplankton
productivity versus macrofloral contribution. The dis-
tribution of the boliviniid taxa in Biscayne Bay indi-
cates a relationship to phytoplankton productivity in
Dumbfoundling Bay.
The benthic foraminiferal assemblages described in
this study of Biscayne Bay foraminifera are somewhat
consistent with previous results from Biscayne Bay,
Florida Bay, and similar environments in the western
Atlantic (Bush, 1958; Brasier, 1975a; Brasier, 1975b;
Rose and Lidz, 1977) that emphasize substrate as a con-
trolling factor. Brasier (1975a, 1975b) studied forami-
niferal ecology and distributions in the environments
surrounding Barbuda, West Indies, concluding that sub-
strate was a strong controlling factor in the distribution
of benthic foraminiferal assemblages. His three domi-
nant substrate conditions were stable back-reef facies,
consisting of fine-grained sediments that support a sea-
grass community in relatively clear water; mobile, bio-
clastic calcarenite substrate where diversity increases
with increased seagrass density; and a high-energy, mo-
bile, coarse bioclastic fore-reef substrate. Bush (1958)
described thirteen benthic foraminiferal biotopes from
Biscayne Bay: open-water east of the keys, tidally ef-
fected channel areas, turbulent shallow shoals, open-
water shallow shoals, grass covered shoals, well-sorted
244 BULLETIN 361
Table 2.—Result of the R-mode principal components analysis of the Biscayne Bay foraminiferal data showing the first 4 varimax rotated
principal components, which account for 59% of the variance in the data.
Principal component scores
Taxon PCl
Trochammina conica —0.964
Triloculina lineiana —0.938
Spiroloculina sp. —0.907
Sorites marginalis —0.874
Triloculina planciana —0.850
Spirosigmoilina antillarum —(0.745
Peneroplis proteus —0.741
Clavulina tricarinata —0.602
Fursenkoina sp. 0.082
Bolivina spp. 0.043
Buliminella elegantissima 0.051
Nonionella sp. 0.057
Bulimina marginata 0.065
Quinqueloculina seminula 0.072
Cyclogyra planorbis 0.052
Discorbis mira 0.090
Valvulina sp. O.111
Ammonia parkinsoniana 0.326
Triloculina tricarinata 0.135
Quinqueloculina polygona —0.261
Pyrgo subsphaerica 0.106
Archaias angulatus —0.318
Quinqueloculina poeyana/tenagos —0.385
Pyrgo denticulata 0.072
Rosalina floridana =OIS)5)
Quinqueloculina agglutinans —0.279
Elphidium galvestonense 0.240
Elphidium poeyanum —0.420
Miliolinella fichteliana (0S
Elphidium delicatulum 0.176
Articulina mucronata —0.377
Quinqueloculina bosciana 0.002
Miliolinella circularis —0.267
Triloculina rotunda —0.123
Textularia candeiana 0.053
Amphistegina lessonit 0.047
Ammobaculites sp. 0.104
Miliolinella labiosa 0.086
Elphidium advenum 0.112
Wiesnerella sp. —0.071
Rosalina globularis 0.044
PC2 PC3 PC4
—0.002 —0.003 0.153
0.044 0.056 0.147
0.004 0.041 0.068
0.038 0.194 0.017
0.020 0.052 0.108
0.097 0.013 0.509
0.053 0.295 —0.100
0.093 0.369 0.186
—0.965 —0.084 0.024
—0.963 SOS 0.024
—0.961 —0.092 0.016
—0.942 —0.058 0.012
—0.891 —0.124 —0.016
—0.848 0.011 —0.058
—0.841 —0.095 —0.002
—0.826 —0.007 —0.041
0.116 0.821 —0.095
0.237 —0.800 —0.234
0.106 0.787 0.027
0.102 0.743 —0.006
0.206 0.740 —0.141
0.082 0.718 —0.265
0.187 0.641 0.409
O.L1S 0.640 0.343
—0.318 0.629 0.317
0.148 0.617 —0.243
0.330 —0.615 —0.204
0.096 0.527 0.127
0.093 0.506 0.173
—0.053 0.078 0.787
0.132 0.248 0.771
—0.161 0.266 0.713
0.040 0.292 0.603
0.122 0.101 0.588
—0.053 0.089 —0.481
0.052 0.095 —0.479
—0.077 —0.030 —0.471
0.129 —0.182 —0.174
O.151 0.354 0.160
—0.425 0.124 —0.141
0.159 —0.265 —0.046
sediment substrate, near-shore surface runoff, Bay deep-
water axis, north bay, northeast bay, eastern margin,
western bay, and southwestern bay. Cluster | is consis-
tent with the composition of the back-reef biofacies of
Brasier (1975a) and the channel areas, shallow shoals,
vegetated shallow shoals, and well-sorted shoals bio-
topes of Bush (1958). These faunal associations are
dominated by miliolids, particularly Miliolinella circu-
laris that is associated with seagrass in this environment
(Brasier, 1975a). Cluster 2 contains taxa similar to the
mobile substrate biofacies of Brasier (1975a) and deep-
water bay axis biotope of Bush (1958), and is charac-
terized by A. angulatus, Q. agglutinans, and E. poey-
anwm. In addition to the similarity in species compo-
sition, the Simpson’s index decreases and species di-
versity of this assemblage increases as the density of
seagrass increases and sediments become finer grained.
Both Biscayne Bay assemblages contain species A. an-
gulatus and P. proteus, which are considered epiphytal
and commonly associated with Thalassia (Lee and
Zucker, 1969; Levy, 1991; Steinker and Steinker, 1976).
The spatial distribution of species such as A. angulatus
and E. poeyanum in the current study may differentiate
a seagrass-dominated substrate from a sediment domi-
nated substrate, but does not indicate an absence of sea-
grass.
to
iS
Nn
BENTHIC FORAMINIFERA: ISHMAN
ATLANTIC OCEAN
: ) = J ~ Biotope |
Florida y , Biotope 2
Bay Ys
Text-figure 5.—Map of the Biscayne Bay region of South Florida showing the distribution of the biotopes in Biscayne Bay: Biotope | (dark
shade) represents marine salinities; Biotope 2 (light shade) represents restricted circulation and oligohaline to polyhaline conditions.
246 BULLETIN 361
Ammonia parkinsoniana and Elphidium galvesto-
nense are dominant taxa (>25%) in samples from Bio-
tope 2 in Northern Biscayne Bay, and Barnes Sound,
Southern Biscayne Bay. The dominance of these taxa
indicates conditions of high variability and relatively
low salinity conditions. This assemblage is similar to
the Ammonia-Elphidium assemblage (A-E assemblage)
described from the Gulf of Mexico (Poag, 1981) and
northeastern Florida Bay (Brewster-Wingard et al.,
1996). The A-E assemblage is associated with oligo-
to mesohaline (O0.5—18 ppt) environments. In contrast,
benthic foraminiferal assemblages from the Gulf Coast
(Poag, 1981) and Florida Bay (Brewster-Wingard et
al., 1996) that are dominated by miliolid taxa are as-
sociated with polyhaline to euhaline (18 to 35 ppt)
conditions. Similarly, Brasier (1975a) found strong as-
sociations between foraminifera and salinity, where
low salinity habitats typically contained Ammonia bec-
carii, and hypersaline habitats were dominated by mil-
iolids Quinqueloculina spp. and Triloculina spp.
SUMMARY
Benthic foraminiferal analyses of 23 surficial sedi-
ment samples, collected from Biscayne Bay, indicate
four distinct foraminiferal assemblages identified using
cluster and principal components analyses. The Cluster
4 assemblage is dominated by Ammonia parkinsoni-
ana and Elphidium galvestonense, represents oligo-
haline to polyhaline conditions in areas where circu-
lation is restricted and point-source freshwater input
occurs. The Cluster 3 assemblage, characterized by the
presence of Bolivina spp. and Bulimina marginata,
represents high nutrient input and organic flux to the
substrate. The Cluster 2 assemblage is dominated by
miliolids with Archaias angulatus and Elphidium
poeyanum significant faunal components with affini-
ties to Florida Bay foraminiferal assemblages. The
Cluster | assemblage also is dominated by miliolids in
which Articulina mucronata and Peneroplis proteus
are significant components. This assemblage has affin-
ities with Atlantic shelf assemblages. The latter two
assemblages represent polyhaline to euhaline condi-
tions where seagrass is an important factor. These four
assemblages recognized from the modern Biscayne
Bay sediments will provide a basis on which the in-
terpretation of historical changes in environmental
conditions within Biscayne Bay can be made.
REFERENCES CITED
Bernhard, J.M.
1986. Characteristic assemblages and morphologies of benthic
foraminifera from anoxic, organic-rich deposits. Journal
of Foraminiferal Research, vol. 16, pp. 207—215.
Bock, W.D.
1971. A handbook of the benthonic foraminifera of Florida Bay
and adjacent waters. Miami Geological Society, Memoir
1, pp. 1—72.
Brasier, M.D.
1975a. Ecology of Recent sediment-dwelling and phytal fora-
minifera from the lagoons of Barbuda, West Indies. Jour-
nal of Foraminiferal Research, vol. 5, pp. 42—62.
Brasier, M.D.
1975b. The ecology and distribution of Recent foraminifera
from the reefs and shoals around Barbuda, West Indies.
Journal of Foraminiferal Research, vol. 5, pp. 193-210.
Brewster-Wingard, G.L., and Ishman, S.E.
1999. Historical trends in salinity and substrate in central Flor-
ida Bay: a paleoecological reconstruction using modern
analogue data. Estuaries, vol. 22, no. 2B, pp. 369-383.
Brewster-Wingard, G.L., Ishman, S.E., Edwards, L.E., and Wil-
lard, D.A.
1996. Preliminary report on the distribution of modern fauna
and flora at selected sites in North-central and North-east-
ern Florida Bay. U.S. Geological Survey Open-file Report
96-732, 34 pp.
Bush, J.
1958. The foraminifera and sediments of Biscayne Bay, Florida
and their ecology. Unpublished Ph.D. thesis, University
of Washington, Seattle, WA.
Cole, S.A.
1972. The effect of thermal stress conditions on benthic fora-
minifera in Biscayne Bay, Florida. Unpublished Masters
thesis, University of Illinois, Urbana-Champaign, IL, 106
Pp:
Corliss, B.H., and Chen, C.
1988. Morphotype patterns of Norwegian Sea deep-sea benthic
foraminifera and ecological implications. Geology, vol.
16, pp. 716-719.
Kohout, F.A., and Kolipinski, M.C.
1967. Biological zonation related to ground water discharge
along the shore of Biscayne Bay, Miami, Fla. in G.H.
Lauff, ed., Estuaries. American Association for the Ad-
vancement of Science, pp. 488—499.
Lee, J.J., and Zucker, W.
1969. Algal flagellate symbiosis in the foraminifera Archaias.
Protozoology, vol. 16, pp. 71-81.
Levy, A.
1991. Peuplements actuels et thanatocenses a Soritidae et Pe-
neroplidae des Keys de Floride (USA). Oceanologica
Acta, vol. 14, pp. 515-524.
Lidz, B., and Rose, P.R.
1989. Diagnostic foraminiferal assemblages of Florida Bay and
adjacent shallow water. Bulletin of Marine Science, vol.
44, pp. 399-418.
Mackensen, A., Futterer, D.K., Grobe, H., and Schmiedl, G.
1993. Benthic foraminiferal assemblages form the eastern South
Atlantic Polar Front region between 35° and 57°S: distri-
bution, ecology and fossilization potential. Marine Micro-
paleontology, vol. 22, pp. 33-69.
Mclvor, C.C., Ley, J.A., and Bjork, R.D.
1994. Changes in freahwater inflow from the Everglades to Flor-
ida Bay including effects on biota and biotic processes:
A review. in S.M. Davis and J.C. Ogden eds., Everglades:
BENTHIC FORAMINIFERA: ISHMAN 247
the ecosystem and its restoration. St. Lucie Press, Delray
Beach, FL, pp. 117-148.
Murray, J.W.
1991. Paleoecology of benthic foraminifera. John Wiley & Sons,
Inc., New York, NY, 397 pp.
Phleger, F.B.
1951. Ecology of foraminifera, northwest Gulf of Mexico. Part
| foraminifera distribution. Geological Society of Amer-
ica Memoir, vol. 46, pp. 1-88.
Phleger, F.B., and Soutar, A.
1973. Production of benthic foraminifera in three east Pacific
oxygen minima. Micropaleontology, vol. 19, pp. 110—
TIS:
Poag, C.W.
1981. Ecologic Atlas of Benthic Foraminifera of the Gulf of
Mexico. Hutchinson and Ross, Stroudsburg, PA, viii, 174
PP-
Rose, P.R., and Lidz, B.
1977. Diagnostic foraminiferal assemblages of shallow-water
modern environments: South Florida and the Bahamas.
Sedimenta, vol. VI, p. 55.
Schnitker, D.
1971. Distribution of foramininfera on the North Carolina con-
tinental shelf. Tulane Studies in Geology and Paleontol-
ogy, vol. 8, pp. 169-215.
Scott, D.B., and Medioli, F.S.
1980. Living vs. total foraminiferal populations: their relative
usefulness in paleoecology. Journal of Paleontology, vol.
54, pp. 814-831.
SFWMD
1994. Surface Water Improvement and Management Plan for
Biscayne Bay. Lower East Coast Planning Division South
Florida Water Management District, West Palm Beach,
FL, 139 pp.
Simpson, E.H.
1949. Measurement of diversity. Nature, vol. 163, p. 688.
Steinker, P.J., and Steinker, D.C.
1976. Shallow-water foraminifera, Jewfish Cay, Bahamas. Mar-
itime Sediments Special Publication, vol. 1, pp. 171-180.
Wanless, H.R.
1976. Geological setting and Recent sediments of the Biscayne
Bay region in Biscayne Bay: Past, Present and Future,
Thorhaug, L.P. and Volker, A., eds., University of Miami
Sea Grant, Miami, FL, pp. 1-32.
Wanless, H.R., Cotrell, D., Parkinson, R., and Burton, E.
1984. Sources and circulation of turbidity Biscayne Bay, Flori-
da. Unpublished report to Dade County and Florida Sea
Grant, 499 pp.
= ¢
ve
- ale
i
5 at 1h
Ap 5
'
&
: Ta ie
: . e 7 _ eine te a
e . a >
7 | % , =
pie AH sm Lhe
CHAPTER 12
OSTRACODE SHELL CHEMISTRY AS A PALEOSALINITY PROXY IN FLORIDA BAY
GARY S. DWYER
Division of Earth and Ocean Sciences
Duke University
Durham, North Carolina 27708
AND
THOMAS M. CRONIN
U. S. Geological Survey
926A National Center
Reston, Virginia 20192
ABSTRACT
We investigated the use of ostracode metal/calcium ratios (Mg/Ca, Sr/Ca, and Na/Ca) to reconstruct the salinity history of
central Florida Bay from its sedimentary record. We first developed provisional partition coefficients (Kp yj. and Ky.s,) from the
average me/Ca ratio for modern Florida Bay water and 171 modern shells of Loxoconcha matagordensis at a mean Florida Bay
water temperature of 26.5°C. Trends in Mg/Ca ratios for L. matagordensis and Peratocytheridea setipunctata from radiometrically
dated sediment cores from Russell Bank, Park Key, and Bob Allen Keys appear to provide a reasonable estimate of Florida Bay
salinity history for the past 130 years. The most complete Mg/Ca record from Russell Bank extends back to 1875 and shows a
good correspondence between estimated paleosalinity and measured annual rainfall in south Florida. Paleosalinity estimates for
the Russell Bank area range from ~13 to ~55 ppt. Estimates for the post-1950s period are comparable to instrumental mea-
surements of salinity near the Russell Bank site. The overall record indicates a decadal-scale variability in salinity in which large
amplitude shifts in ostracode Mg/Ca ratios with high Mg/Ca (higher salinity) values corresponding to periods of low rainfall
from ~1895 to ~1920 and from 1940 to present. The intervening 20 years (~1920—1940) are characterized by relatively low
amplitude shifts in Mg/Ca and rainfall. Since 1950, periods of high salinity (~45 to ~55 ppt) occur during each decade. Prior
to 1940, salinity never exceeded ~45 ppt and generally remained below 40 ppt, despite evidence in the rainfall record of dry
intervals in the early 1900s equivalent to those of the last 50 years. Salinity minima associated with the decadal-scale oscillations
range from ~15 to ~25 ppt. Four of the five extreme low values (~1915, ~1940,
times of strongly negative values of the Southern Oscillation Index, that is, El Nino conditions that bring anomalously high
winter rainfall in the southeastern U.S.
These results support the hypothesis that seasonal and decadal-scale salinity fluctuations are a natural part of the Florida Bay
ecosystem, and that these fluctuations are largely a function of natural variability of regional climate (rainfall). The relatively
low peak salinity values associated with apparent dry periods in the early 1900s suggests that, since ~ 1950, anthropogenic factors
~1983, ~1998) appear to correspond with
may have played a role in the magnitude of the salinity fluctuations in more recent times.
INTRODUCTION
Salinity variations play an important role in the
physical and biogeochemical processes of estuaries
and marine embayments. Within these marginal ma-
rine settings, either directly or indirectly, spatial and
temporal variations in salinity control turbidity and
sedimentation, the distribution of dissolved inorganic
and organic constituents, and the distribution and
abundance of flora and fauna (reviews in Chester,
1990; Berner and Berner, 1996).
Recently, Florida Bay—an estuarine-like, shallow
marine embayment that receives significant terrestrial
runoff from peninsular Florida—has experienced dra-
matic ecosystem changes including sea-grass die-offs,
decreases in populations of fishes, pink shrimp, and
water birds, and increased blooms of algae (Boesch er
al., 1993). At the same time, the frequency and inten-
sity of hypersaline events in Florida Bay have appar-
ently increased leading to the hypothesis that the re-
cent ecosystem changes in the bay have resulted from
changes in salinity (MclIvor ef al., 1994). Further, it
has been hypothesized that salinity variations within
the bay are the result of anthropogenic factors, such
as diversion of freshwater runoff from the Everglades,
freshwater that would otherwise have drained into
Florida Bay (Mclvor ef al., 1994). Hydrologic mod-
eling studies support this hypothesis, showing that
250 BULLETIN 361
large-scale decreases in freshwater influx to Florida
Bay from the Everglades will lead to extreme fluctu-
ations in salinity and an increased occurrence of hy-
persaline conditions, especially in near-shore sub-ba-
sins of central and northeastern Florida Bay (Fennema
et al., 1994).
While it is clear from historical accounts that the
bay has experienced hypersaline events in the last few
decades, very little is known about the long-term sa-
linity history of Florida Bay and whether or not the
recent events of hypersalinity are unprecedented prior
to human activities in South Florida. Unfortunately,
only short and discontinuous instrumental records of
Florida Bay salinity have been obtained over the last
50 years, and only anecdotal salinity data exists prior
to this time. In place of instrumental data, longer re-
cords of Bay salinity have been reconstructed from
fossil material deposited and preserved in the bay. For
example, on the basis of an oxygen isotopic (6'5O)
record from a Florida Bay coral, Swart et al. (1996)
postulated that there has been no major long-term in-
crease in salinity in Florida Bay, implying that the
events of hypersalinity are part of the natural variabil-
ity of Florida Bay. They concluded, however, that fol-
lowing construction of the Miami-to-Key West rail-
road (1905-1910) there was a decrease in the vari-
ability of the 6'SO signal, suggesting a stabilization of
salinity conditions in Florida Bay during the period
1905-present relative to 1825—1905. Brewster-Wingard
et al. (1998) and Brewster- Wingard and Ishman (1999)
interpreted molluscan and foraminiferal assemblages
from Florida Bay sediment cores to indicate progres-
sively increasing salinity during the later part of the
20th century and an increase in the amplitude of sa-
linity variability about 1940.
The coral record of Swart et al. (1996) is from Lig-
numvitae Basin near the southwestern bay-ocean
boundary where seasonal and inter-annual salinity var-
lations are substantially lower than in central Florida
Bay. Although they argue that salinity variations in
Lignumvitae Basin are representative of salinity
changes throughout most of Florida Bay, no compa-
rable studies from interior-basins (where corals are
rare) have confirmed this hypothesis. Further, Swart et
al. (1996) noted the complexity of the coral 5'°O sig-
nal, which is controlled by temperature, rainfall, rela-
tive fluxes to the bay of freshwater and seawater, and
evaporation, and, consequently, the difficulty in inter-
preting this record without ambiguity.
In contrast to the coral 6'8O-based salinity record,
paleoecological studies of salinity-dependent fauna
and flora from sediment cores suggest multi-decadal
scale fluctuations of Florida Bay salinity superimposed
on an overall increase in salinity during the last 200
years (Wingard er al., 1995). Like the 6!8O record,
interpreting paleoecological signals may also be com-
plicated by a number of factors, including changes
over time in bottom sediment type, temperature, sub-
aquatic vegetation, food availability, and predation,
leading to uncertainty in paleoecological-based salinity
reconstructions.
Here we develop a new geochemical proxy of Flor-
ida Bay salinity—the metal/Ca ratios preserved in fos-
sil ostracode shells—and use it to reconstruct the sa-
linity of Florida Bay for the last 180 years. We also
compare the paleosalinity record to instrumental rain-
fall records of South Florida to examine the effects of
regional climate variability on Florida Bay salinity.
ACKNOWLEDGEMENTS
We are grateful to L. Brewster-Wingard, S. Ishman,
R. Halley, C. Holmes, J. Stone, S. Schwede, and S.
Kim for their help in the field and in dating and pro-
cessing the core samples. Funded by USGS Place-
Based Studies Program.
REGIONAL SETTING
Florida Bay (Text-fig. |) is a triangular-shaped, shal-
low, restricted marine embayment bounded to the
north by the coastal wetlands of the Everglades and to
the south by the Florida Keys which separate the bay
from open ocean waters of the Straits of Florida. The
western boundary of the bay is less distinct, made up
of a system of shallow sediment banks that restrict
exchange between Florida Bay and the Gulf of Mex-
ico. Water depths range from less than | m to about 3
m generally deepening to the south and west (Text-fig.
1). Tidal effects are minimal throughout much of the
bay.
Within the bay are a series of sub-basins separated
by a network of small islands, called keys, and mud-
banks. The mudbanks are submerged, accretionary
features and are the primary sites in the bay for the
accumulation of sediment, most of which is fine-
grained calcium carbonate (mud) derived from calcar-
eous organisms living within the bay (Stockman ef al.,
1967; Enos and Perkins, 1979).
Waters of Florida Bay are composed primarily of a
mixture of two salinity end-members: fresh, terrestrial
runoff/discharge from the Everglades with a salinity of
less than | ppt, and saline, surface waters from the
Gulf of Mexico and Straits of Florida with salinity
around 36 ppt. Mixing of these two end members leads
to a general northeast-to-southwest salinity gradient.
The bay’s salinity can also greatly exceed 36 ppt as
a result of evaporation. Evaporative concentration of
Florida Bay waters generally occurs during the dry
season (~November to May) and salinity values ex-
OSTRACODE SHELI
~~ ¥
; OE,
ly
%
\ 270
BY
E
y \
—2 — Water Depth
(m)
h Mudbanks
cs (< 1m deep)
cs
y
/
os' 81° 55' 50' 45'
CHEMISTRY: DWYER AND CRONIN 251
§
A)
“ee
Atlantic
Ocean
o Modern Samples
* Cores
40' 35' 80°30' 25' 20'
Text-figure |—Map showing Florida Bay region and location of sample sites for this study. Also shown is the location of the coral studied
by Swart et al. (1996).
ceeding 50 ppt have been reported in interior portions
of the Bay (Ginsburg, 1972, and see reviews in Swart
et al., 1996 and Mclvor et al., 1994). Wet season
(~May to October) rains bring new runoff from the
Everglades and reverse the dry season trend of increas-
ing salinity.
The strong influence of rainfall/runoff on the salin-
ity of Florida Bay is evident from recent monitoring
studies and varies in magnitude and timing with dis-
tance from the mainland (Text-fig. 2, Text-fig. 3). Shal-
low groundwater levels in the Everglades, which are
largely controlled by rainfall, show a significant cor-
relation with Florida Bay salinity providing further ev-
idence of the impact rainfall/runoff on the salinity of
Florida Bay (Tabb, 1967; White, 1983; as reviewed in
Mclvor et al., 1994).
Unfortunately, these studies were conducted after
significant modifications were made to the natural run-
off system in South Florida, which previously was
characterized by sheet flow south from Lake Okeecho-
bee through the Everglades to Florida Bay (Miller,
1997). During this century, a complex network of ca-
nals and levees has been constructed in order to control
flooding and to drain wetlands for agriculture, ulti-
mately resulting in a lowering of the water table and
an alteration of the near-surface hydrogeology over
much of in southern Florida (RAIL, 1973: Miller,
1997). Just after the onset of canal construction around
the turn of the century, the Miami to Key West railroad
was completed, resulting in partial or complete closing
of many of the ocean inlets between the Florida Keys.
This has led to the hypothesis that the salinity of Flor-
ida Bay may have been markedly different prior to
large-scale human activities in south Florida, an idea
which we evaluate here by reconstructing the salinity
history of central Florida Bay using the chemistry of
fossil ostracode shells.
CONTROLS ON OSTRACODE SHELL
CHEMISTRY
PREVIOUS STUDIES
Ostracodes are microcrustaceans that inhabit fresh-
water, brackish, marine, and hypersaline environments.
They secrete a bivalved carapace, or shell, that is com-
monly preserved in sediments and whose chemistry is
increasingly being used for paleoenvironmental recon-
structions in various aquatic ecosystems (Chivas er al.,
1983; Holmes, 1996). The shell is composed mostly
of the mineral calcite (CaCO,), which co-precipitates
minor amounts of foreign metals such as magnesium
DOD BULLETIN 361
—2 — Water Depth
(m)
, Mudbanks
“-* (< 1m deep)
Salinity
October 1995
(ppt)
EVERGLADES
——
ms St
— 2 — Water Depth
(m)
>», Mudbanks
G@— (<1mdeep)
>
Salinity
June 1996
(ppt)
80°30" 25' 20'
Text-figure 2.—Examples of salinity distribution in Florida Bay near the end of the wet (October) and dry (June) seasons. Contoured from
Halley et al. data (1995).
and strontium, that substitute into the crystal lattice in
place of calcium. The uptake of Mg or Sr into the
calcitic ostracode shell can be described by the follow-
ing equation:
= (K )(me/Ca )
D-me water
(me/Ca),
»stracode calcite
where me/Ca represents the atomic ratio of Mg (or Sr)
to Ca, and Kj, 1s the element-specific partition co-
efficient. Thus, if K is constant or can be con-
D-me
strained, then the me/Ca ratio in ostracode calcite can
be used to determine the me/Ca ratio of the water in
which the shell was secreted (see Wansard, 1996).
In the waters of Florida Bay, me/Ca ratios appear
to vary with salinity as a result of mixing of freshwater
runoff and discharge from the Everglades, and sea-
water (Text-fig. 4). Thus, (me/Ca),,,,¢,.
from (Me/Ca) ciracode calcines COUld be used to estimate
Florida Bay salinity.
as calculated
OSTRACODE SHELL CHEMISTRY: DWYER AND CRONIN D253
5 e
g >
= <
oO =
x 5
012345 6 7 8 9 101112131415161718192021 22
Jan Jan
28 Month 9°
Rainfall ----O---- Mouth Little Madeira Bay
----2---- Russell Bank
---f--- Lignumvitae Basin
Text-figure 3.—Time series of monthly rainfall in South Florida
(Florida Area 5, NOAA Data Archive) and bi-monthly salinity (Hal-
ley et al., 1995) at four locations in Florida Bay for the period from
November 1994 to June 1996. Note apparent amplitude decrease
and delayed response of salinity change to rainfall with increased
distance from mainland.
In addition to (me/Ca),,,.,, other factors potentially
control the uptake of foreign ions into calcite. From
experimental and field studies with inorganic and bio-
genic calcite, Kj. has been shown to have a strong
thermodependence (Chave, 1954; Chivas ef al., 1983;
Cadot and Kaesler, 1977; Burton and Walter, 1991;
Dwyer et al., 1995). And K,,5, appears to be controlled
primarily by calcite precipitation rate (reviewed in
Morse and McKenzie, 1990).
Aside from temperature and Mg/Ca,,,,.,, salinity may
independently affect Kp .4.. While most workers have
concluded that Kj yj. 1s constant or increases with in-
creasing salinity, two survey studies of shells of ostra-
code genus Cyprideis trom shallow coastal ponds of
widely varying salinity (0 to >120 ppt) indicated a
negative relationship between Kp yj. and salinity (Tee-
ter and Quick, 1990; Bodergat, 1985). Similarly, Wan-
sard (1996) noted the apparent nonlinearity in Ky,
for Cyprideis when comparing Kp, calculated for
freshwater specimens from Lake Banyoles, Spain
(0.0169) to the Kj. calculated for marine specimens
by Chivas et al. (1986) (0.0045). Unfortunately, results
of empirical field studies from sites with significant
inter-annual environmental variability often are incon-
clusive because, even for live specimens, it is difficult
to verify that the ostracode shells grew under the en-
vironmental conditions measured at the time of collec-
tion (e.g., Xia et al., 1997). To avoid the inherent un-
Atomic
Ratio
OF 1020 S040 S50 60R 70
Salinity (ppt)
Text-figure 4—Model data (connected filled symbols) showing
the salinity dependence of Mg/Ca (mol/mol) and Sr/Ca (mmol/mol)
in waters of Florida Bay. The model assumes conservative mixing
of freshwater from the South Florida surficial aquifer system and
seawater with the following elemental concentrations (ppm):
Ca Mg Sr Na
Freshwater (Sonntag, 1987) 90 5.6 0.8 26
Seawater (salinity = 35 ppt, 412 1290 ES) 10,770
Chester, 1990)
Open symbols are ratios calculated from Florida Bay water anal-
yses reported by Berner (1966) and Burns and Swart (1992), and
they generally agree with the model data. At salinity above normal
seawater, me/Ca ratios of Florida Bay waters may behave in a num-
ber of ways. Under simple evaporative concentration, ratios will
remain unchanged. However, if evaporation concentration is accom-
panied by significant amounts of CaCO, precipitation (biogenic or
inorganic), me/Ca ratios of bay waters will continue to increase
along with salinity. Relative to Mg/Ca, this effect may be diminished
for St/Ca because of the relatively high Sr/Ca ratio of aragonite, a
mineral that makes up around 60% of the sediments deposited in
Florida Bay (Burns and Swart, 1992).
certainties associated with field studies, DeDeckker et
al. (1999) analyzed the chemistry of Cyprideis shells
raised in laboratory cultures under controlled temper-
ature, salinity, and Mg/Ca,,,,., conditions. Their results
confirm the strong effects of water temperature and
Meg/Ca,..-. on the uptake of Mg and the lack of any
salinity effect, positive or negative.
In addition to magnesium and strontium, ostracode
shell calcite coprecipitates minor amounts of sodium.
However, unlike Mg and Sr, Na is not incorporated in
the calcite crystal lattice, but instead resides in crystal
defect sites (Busenberg and Plummer, 1985). Evidently
sodium uptake is controlled by the sulfate (SO,*-) con-
centration of the mother solution (Busenberg and
254 BULLETIN 361
Plummer, 1985). As a conservative ion in estuarine
settings (Chester, 1990), SO,?- likely covaries with sa-
linity in Florida Bay much like Ca, Mg, Sr, and Na.
Therefore, the Na/Ca ratio of ostracode calcite may
also be indirectly controlled by salinity and be useful
for paleosalinity reconstructions in Florida Bay.
Phylogenetic and ontogenetic factors also influence
Ky me and Kj, of ostracode calcite (Chave, 1954; Ca-
dot and Kaesler, 1977; Chivas er al., 1983), requiring
the use of shells of same genus and of the same on-
togenetic stage for meaningful paleoenvironmental re-
constructions based on shell chemistry. Thus it is nec-
essary to calibrate K,.,,. for shells of the species se-
lected for paleoenvironmental reconsrtructions.
STRATEGY TO DEVELOP PALEOSALINITY EQUATIONS
Our strategy was to determine partition coefficients
(Kp.me) for foreign metals for adult shells of the com-
mon species Loxoconcha matagordensis on the basis
of average me/Ca ratios for Florida Bay water and for
average me/Ca ratios in modern specimens of Loxo-
concha matagordensis. Ky, Would then be used in
conjunction with fossil shell me/Ca ratios to calculate
past values of me/Ca,,,,.. Which in turn would be used
to estimate salinity based on the relationship displayed
in Text-figure 4.
This modern-average calibration method was pre-
ferred over a site-by-site direct calibration of measured
salinity to modern shell Mg/Ca at the site (7.e., Dwyer
et al., 1995) for three reasons. First, it is difficult to
separate the effects of water temperature, which an-
nually varies from about 15—33°C in Florida Bay, from
those of salinity. Previous studies, as well as our own
data shown below, suggest strong thermodependence
of ostracode Mg/Ca ratios. Separating salinity and
temperature effects in coastal environments is difficult
for most geochemical and faunal proxies of past en-
vironmental conditions. However, because temperature
and salinity are often positively correlated in Florida
Bay waters (Swart ef al., 1996; results below), it is
quite probable that downcore variability in ostracode
shell Mg/Ca is due to changes in both factors (see
discussion).
Second, although salinity and temperature were
measured at the time of collection of modern L. ma-
tagordensis, the exact time that each individual ostra-
code secreted its adult shells was not known. Because
adults of L. matagordensis live for several months and
this species secretes its adult shells year round (King
and Kornicker, 1970; Kamiya, 1988), then many
adults, in fact, did not secrete their shell at the time of
collection. Because salinity and temperature vary tem-
porally and spatially, it is possible only to obtain a
D-me
range of possible salinities and temperatures for the
site for the previous few months.
Third, during our period of sampling, the mean sa-
linity in the study area varied from ~23.9 ppt (Feb-
ruary 1998) to ~33.6 ppt (July 1998). With few ex-
ceptions, our material does not include individuals that
lived in extremely high (>40 ppt) or low (<18 ppt)
salinity environments such as those experienced by
Florida Bay historically (Robblee ef al., 1991). Con-
sequently, a revised calibration should be developed in
the future using individuals grown in cultures at
known temperatures, over the widest salinity range
possible, which for L. matagordensis is 10 to >50 ppt.
METHODOLOGY
ANALYSES OF WATER AND LIVING OSTRACODE SHELLS
Water and ostracode samples for this study were col-
lected from 25 U.S. Geological Survey sites across
most of Florida Bay (Text-fig. 1). As part of a larger
study to characterize the Florida Bay ecosystem, most
of these sites have been sampled seasonally (February
and July) since 1994 (Wingard ef al., this volume).
Water samples, filtered in the field using disposable
0.45 wm syringe filters, were collected from each of
the 25 sites in July 1998 and February 1999. In ad-
dition, water samples (unfiltered) were collected in
August 1997 from 51 sites in Biscayne Bay and the
small bays that lie between Biscayne and Florida Bays.
At each site, water temperature and salinity were mea-
sured in the field using routinely-calibrated YSI or Hy-
drolabs brand field instruments (Brewster-Wingard et
al., this volume).
Ca, Mg, Sr, and Na concentrations of the water sam-
ples were measured by direct current plasma atomic
emission spectrophotometry (DCP) in the Plasma Lab
at Duke University Earth and Ocean Sciences to de-
termine the relationship between water me/Ca ratios
and salinity of south Florida’s coastal waters. Final
data are reported as Mg/Ca, Sr/Ca, and Na/Ca atomic
ratios with analytical precision, based on replicate
analysis of [APSO standard seawater, of +1%, +2%,
and +3%, respectively.
Ninety-seven living (whole carapaces with visible
soft parts), adult specimens of Loxoconcha matagor-
densis, a cosmopolitan phytal ostracode in Florida Bay
(Keyser, 1975; Cronin et al., this volume), were col-
lected from sub-aquatic vegetation from 16 of the 25
USGS sites in Florida Bay in February 1998, July
1998, and February 1999. Shell Mg/Ca, Sr/Ca, and Na/
Ca ratios of these specimens were measured by DCP.
Procedures follow those in Dwyer er al. (1995). An
additional step included an overnight soaking of shells
in full-strength commercial bleach to remove organic
OSTRACODE SHELL CHEMISTRY: DWYER AND CRONIN
matter prior to triple rinsing in deionized water. Re-
sults are reported as atomic ratios of Mg/Ca, Sr/Ca,
and Na/Ca in mmol/mol. Analytical precision, based
on replicate analysis of internal-consistency standards
is less than +2% for Mg/Ca and Sr/Ca ratios, and
around +8% for Na/Ca.
From six of the sites we also analyzed the Mg/Ca,
Sr/Ca, and Na/Ca ratios of shells of dead adult speci-
mens of L. matagordensis collected from the upper 10
cm of sediment push cores. These core-top specimens
allowed a comparative evaluation of possible post-
mortem alteration effects on shell chemistry.
FossiL MATERIAL FROM SEDIMENT CORES
Mg/Ca, Sr/Ca, and Na/Ca ratios were measured on
fossil ostracode shells from five 1 to 2 m long sedi-
ment cores. Four cores lie along a roughly north-south
transect in central Florida Bay (Text-fig. 1). These in-
clude mudbank cores BA-6A, RB-19B, and PKK-23
collected in 1994 near Bob Allen, Russell, and Park
Keys, respectively, and core T-24 collected from with-
in Little Madeira Bay along the coast of the Everglades
near the freshwater outflow region of Taylor Slough.
The fifth core, MB-1, is from Manatee Bay in Barnes
Sound, an adjacent embayment northeast of Florida
Bay.
Chronstratigraphy for the cores was determined by
uranium disequilibrium series methods (Wingard ef al.,
1995: Brewster-Wingard et al., 1997; Robbins ef al.,
in press) and preliminarily corroborated with exotic
pollen horizon studies (Wingard ef al., 1995). Calcu-
lated linear sediment accumulation rates for BA-6A,
RB-19B, and PKK-23A are 1.04, 1.22, and 0.78 cm/
year, respectively, and were applied to the entire length
of the core. Chronstratigraphic information for core T-
24 and core MB-| are still under development.
Samples for ostracode analysis were collected at ap-
proximately 2-cm intervals. Most intervals in the 5
cores contained ostracodes and the majority of these
yielded adult valves for geochemical analysis. Shells
of Loxoconcha matagordensis were sufficiently abun-
dant to produce continuous records in the Russell and
Park Key cores and the upper half of MB-1. Kp inc
values for L. matagordensis calculated from modern
studies were then used to estimate salinity from fossil
me/Ca ratios. In cores or intervals with insufficient
numbers of L. matagordensis in (Bob Allen, T-24, and
the lower half of MB-1; Cronin er al., this volume) we
analyzed adult shells of second genus, Peratocytheri-
dea. However, no K,,,,. values are available for Per-
atocytheridea, so any down-core trends Peratocyther-
idea me/Ca ratios can only be interpreted qualitatively.
Shells were prepared and analyzed for Ca, Mg, Sr, and
Na by DCP as for modern specimens discussed above.
i)
Nn
Nn
RESULTS AND DISCUSSION
WATER CHEMISTRY
The results of water chemistry analysis are listed in
Table | and summarized in Text-figure 5. As shown in
Text-figure 5, Mg/Ca and Sr/Ca ratios of Florida and
Biscayne Bay decrease with decreasing salinity, shift-
ing gradually from 35 ppt and 15 ppt and more steeply
from 15 ppt to 0 ppt. Na/Ca ratios (not shown) behave
similarly.
The trends observed are similar to that predicted
from the preliminary water chemistry data presented
in Text-figure 4 and the assumption of conservative
mixing of Ca, Mg, Sr, and Na between seawater and
freshwater runoff from southern peninsular Florida.
Mg/Ca values overlap directly with Mg/Ca ratios cal-
culated from the data of Berner (1966) who measured
Mg and Ca concentrations at salinity ~5, 17, and 43
ppt in the 1960’s.
The steep decrease in me/Ca ratios with salinity is
not typical of estuaries of the eastern United States.
More typically, estuarine waters show little or no
change in Mg/Ca, Sr/Ca, or Na/Ca ratios, retaining
seawater-like values well below 15 ppt. This occurs
because, relative to South Florida runoff, terrestrial
runoff is generally far more depleted in Ca, Mg, Sr,
and Na and me/Ca ratios are thus overwhelmed by
seawater. The relatively high concentration of these el-
ements, especially Ca, in south Florida runoff is likely
the result of interaction between surface water and up-
per Tertiary carbonate rocks of southern Florida.
Most importantly for this study, the similarities in
water chemistry between Florida and Biscayne Bay
sampled from 1997 to 1999 and the comparable Mg/
Ca ratios obtained by Berner (1966) for Florida Bay
suggest that the relationships between me/Ca,,,,.. and
salinity are regionally and temporally persistent.
MODERN SHELL CHEMISTRY
Results of shell chemistry analyses of modern spec-
imens of adult Loxoconcha matagordensis from veg-
etation samples and sediment core tops are listed in
Table 2 and the results of vegetation specimens are
summarized in Text-figure 6. Text-figure 6 shows shell
Meg/Ca and Sr/Ca ratios versus measured ambient wa-
ter salinity and temperature from sampling trips in
February 1998, July 1998 and February 1999.
It is clear from Text-figure 6 that there are no ob-
vious trends between modern shell me/Ca ratios across
the entire range of measured water temperature and
salinity. Mg/Ca ratios range from 23 to 54 mmol/mol.
Shells from February 1999, covering the fewest num-
ber of sites in the three seasonal data sets, appear to
show a positive correlation with both salinity and tem-
256 BULLETIN 361
Table 1.—Chemistry of waters from Florida and Biscayne Bays. Calcium, magnesium, strontium, and sodium concentrations reported in
ppm; Mg/Ca and Na/Ca as mol/mol, and Sr/Ca as mmol/mol. * Salinity determined using sodium concentration. Florida Bay sites correspond
to map (Text-fig. 1). Biscayne Bay sample locations available from the authors.
Salinity
Sample site (ppt) Ca Mg Sr Na Mg/Ca Str/Ca Na/Ca
Florida Bay (July 1998)
1 26.9 332 1005 6 7792 4.99 8.39 40.86
2-Top 23.96 333 972 6 6098 4.81 8.59 31.89
3-Top 29.1 362 1105 7 8303 5.04 8.44 40.02
4 32.45 401 1272 8 9561 5.23 8.61 41.54
5 32.45 400 1249 7 9343 SAS 8.49 40.72
6 35.94 436 1379 8 10657 S22 8.58 42.63
6-Top 35.31 436 1370 8 10545 5.18 8.53 42.20
7 SMlaits) 457 1431 8 10745 5.17 8.50 41.01
8 23.51 319 914 6 5824 4.72 8.37 31.78
:) 28.24 412 1196 8 5831 4.79 8.52 24.67
10 34.63 435 1342 8 9860 5.08 8.48 39.47
11 30.78 380 1192 i 9161 Sel 8.56 41.98
12 36.85 432 1363 8 10523 5.20 8.60 42.42
13 40.12 503 1573 10 11372 5.16 8.67 39.45
14 40.1 498 1538 ) 11400 5.09 8.66 39.91
15 36.75 453 1386 8 10512 5.04 8.48 40.42
15-Top 36.18 450 1379 8 10684 5.05 8.61 41.41
16-Top 38.77 432 1327 8 10424 5.07 8.58 42.10
17-Top 34.61 422 1290 8 9573 5.04 8.46 39:52
18 36.74 438 1352 8 10454 5.09 8.56 41.63
20 27.82 349 1054 6 8023 4.97 8.29 40.03
21 39.34 469 1441 9 10946 5.06 8.67 40.66
21-Top 38.72 459 1405 9 10764 5.05 8.54 40.86
22-Top 38.56 469 1455 9 11104 5.11 8.69 41.24
23-Top 36.25 433 1368 8 10639 5.21 8.65 42.84
24 35.93 423 1307 8 10191 5.10 8.51 42.03
24-Top 35.65 431 1337 8 10326 Sez 8.50 41.80
25 35.57 427 1301 8 10126 5.02 8.46 41.31
26-Top 35.18 420 1304 8 10320 Spl 8.63 42.82
Florida Bay (February 1999)
1 28.16 298 824 6 5704 4.55 8.45 83332
1 28.16 299 814 5) 5384 4.48 8.24 31.34
2 25.24 359 1026 7 6205 4.71 8.35 30.14
3 29.27 385 1145 7 8520 4.90 8.30 38.54
4 27.81 366 1090 7 8281 4.91 8.33 39.43
5 2935 389 1145 7 8470 4.85 8.22 37.93
6 28.87 374 1111 7 8317 4.90 8.25 38.75
7 30.47 383 1160 7 9021 4.99 8.30 41.02
8 19.38 286 795 5) 5003 4.59 8.19 30.51
9 24.06 343 996 6 6141 4.79 8.35 31.23
10 26.86 367 1078 7 6357 4.84 8.38 30.19
11 27.04 358 1058 6 8047 4.88 8.24 89122
12 28.61 370 1106 7 8362 4.94 8.38 39.45
13 33a 418 1262 8 9454 4.97 8.21 39.41
14 32.94 422 1287 8 9735 5.03 8.37 40.24
15 34.62 443 1340 8 9880 4.99 8.34 38.87
16 35.34 447 1363 8 10012 5.03 8.50 39.06
17 32.54 427 1308 8 9953 5.05 8.65 40.63
17-banktop* 31.3 417 1262 8 9583 4.99 8.40 40.03
18 35.76 455 1401 9 10250 5.08 8.57 39.31
20 24.97 347 1023 6 6154 4.85 8.53 30.89
21 34.72 440 1313 8 9618 4.92 8.18 38.09
21 34.72 443 1333 8 9633 4.96 8.25 37.91
21-lake* 36.9 543 1646 11 11288 5.00 8.86 36.27
21-lake* 35.4 544 1638 10 10841 4.97 8.70 34.75
22 eppilil 427 1302 8 9753 5.03 8.56 39.81
23 34.04 428 1298 8 9707 5.00 8.49 39.54
24 35.13 416 1257 8 9426 4.98 8.37 39.48
26 33.51 409 1223 8 94.14 4.93 8.40 40.15
26 33.51 408 1215 V 9201 4.91 8.29 39.34
Table 1.—Continued.
OSTRACODE SHELL CHEMISTRY: DWYER AND CRONIN
Sample site
Biscayne Bay (August 1997)
MRO6
TMO2
MRO7
MROS5
MRO4
MRO3
WCO02
SPO1
SKOL
MRO2
LRO3
PROI
BLO2
AROI
BLO1
MWOl1
GLO02
MRO1
BBS5S1
BBO1
LROI
MIO1
CGO1
BBO3
CDOo1
BBS5O0
BB48
BB48
BBO9
BB39A
BB34
BB47
BB10
BB27
BBSA
BB23
BB29
BB31
BB28
BB44
BB41
BB45
BB45
BB32
NNH
BB25
BB46
BBO7
BB35
BB42
BB36
BB43
BB37
BB39*
Standards*
IAPSO
IAPSO
IAPSO
IAPSO
IAPSO
IAPSO
Salinity
(ppt) Ca Mg Sr Na
0.3 64 9 1 3
0.3 66 8 1 =
0.4 60 11 1 11
0.5 66 14 1 43
0.7 71 26 1 172
it 76 39 1 306
1.4 80 42 1 367
2.3 88 65 1 562
2.7 92 94 1 771
2.8 93 94 1 787
3.8 96 87 1 723
Spill 141 172 2 1380
9.4 174 334 3 2099
10.9 190 387 3 3020
12.4 203 469 3 3478
12.6 198 449 3} 3370
13.4 205 459 3 3408
16.4 142 279 2 2411
17.4 233 632 4 4418
18.2 267 697 5 4658
19.] 126 209 2 1835
20.8 269 708 5 5047
23.3 290, 755 5) 5253
23.6 328 918 6 5775
231) 319 874 6 5680
24.1 317 891 6 5798
26 320 945 6 7464
26 317 938 6 7537
26.4 314 919 6 7353
26.9 346 1029 6 7903
29 327 964 6 7562
30.2 366 1110 7 8686
Si 384 1166 7 8912
Shila7/ 386 1174 7 9009
Sl) 398 1220 7 9292
32 383 1165 7 9009
32.1 392 1191 7 9114
32.4 393 1206 1 9233
32.6 394 1200 7 9122
32.9 397 1219 7 9338
33 390, 1210 7 9274
$36). 400 1220 7 9240
B3el 393 1218 7 9362
33.8 372 1151 7 8917
33.9 412 1273 8 9538
8522 413 1283 8 10087
35.2 416 1305 8 10320
35.6 427 1330 8 10513
35.8 420 1322 8 10265
36.1 428 1353 8 10615
36.4 430 1335 8 10174
36.7 433 1354 8 10474
37.3 436 1374 8 10707
33.9 425 1329 8 10377
33.4 422 1314 8 10235
33.0 424 1310 8 10093
3257. 432 1328 8 10015
35.1 430 1334 8 10740
3355 440 1345 8 10247
32.9 436 1349 8 10077
251,
8.59
8.53
8.53
8.47
8.14
8.33
8.57
8.38
258 BULLETIN 361
Table 2—Chemistry of living and sediment coretop shells of Table 2.—Continued.
adult specimens of Loxoconcha matagordensis. Mg/Ca, St/Ca, and
Na/Ca ratios reported as mmol/mol. * Salinity determined using so- Sample Temp. Salinity
dium concentration. Florida Bay sites correspond to map (Text-fig. Site CC) (pp) VPI Mg/Ca_ Sr/Ca_Na/Ca
1). Sample descriptors (e.g. #1, A) refer to separate subsamples from July 1998
the sites. Biscayne Bay sample BBS51 is located near the MB1 core
site (Text-fig. 1). VPI (visual preservation index) is a measure of Lage BOSE 30.25 20s sO70 105 14.71
shell clarity ranging from | (clear, unaltered) to 7 (opaque, highly LA#3 31.3 26.9 2.5 39.71 4.31 17.09
Brereant 1A#3 31.3 26.9 25 35.88 3.96 13.99
23#1 31.94 36.76 2.5 4449 3.68 17.00
Sample Temp. Salinity 2241 30.42 39.93 2.5 36.11 420 14.34
Site (C) (ppt) VPI Mg/Ca Sr/Ca Na/Ca 2241 30142) 39193| 255) | 42.48) Ao 7 Sa
Florida Bay Living Specimens 8A#2 30.86 23.88 2.5 38°22 35977 16.45
Febiny 1908 8A#2 30.86 23.88 2.5 43.36 4.16 14.66
os 526 Aas a Pye eRe eee 16#2 29338) 3878) 255) S7:85i 4 Olea
7 AAO a7 A Nats 850 “GOaN 16#2 29138) 38078) 225) S271 Se
ie EO AI 4 a5 64 SMG. 1541 LLA#2 32:99 33:13 25 39195 3:83. mI400
7 75.9 317 4 3778 365 15.04 LLA#2 ewes) “eeulel Oey seis BOP 1238
17 25.9 317 4 5920 334 «13.77 1LA#2 Ewch) eis Os Ge B75 1G57
7 35.9 31.7 4 43.08 3.70 17.56 SA#2 33.42 31.94 2.5 39.79 3.55 15.46
l 21 15.8 4 3770). 3u19 15.70 12A#2 3458) 936197) 255 38.14 3.83 16.53
1 21 15.8 4 36.59 3.57 16.03 12A#2 34.58 36.97 DES) 35.41 4.02 14.87
| 21 15.8 3 38.64 3.86 15.85 18#1 ys 205 DS SER S89 14.65
1 21 15.8 D 46.84 3.29 17.03 18#1 29195, 3705-255) 44197 1 S4 misao
1#1 21 15.8 DSS HA StOS OWS 18#1 ps ey DS 309 BG less
1#1 21 15.8 25 27.01 418 14.53 20A#3 iy AIG AS Zyl 293} GO
#1 21 15.8 2) 28-118) 73:60 18.45 20A#3 B10 AIS SS ESO) LSS 18.97
44 21.6 16.7 2:5 29.85 3:81! 26:86 26#2 3326) 35:48) 255) 32189) 32a
444 216 167 25 35.58 4.10 16.43 26#2 83126) 35:43) 255) 34°51 SsSiemlarsy
SH 21.9 AG 2.5) 40:34 3:68" 14.57 26#2 33.26 35.43 25 35:40 341) ies?
Se) ge ioe Oa palm gl ae 6A 33.29 35.94 2.5 39.06 3.97 16.24
ie as ie ae ee ee oe 6A 33.29 35.94 25) |4154) 3193) Netstas
S#1 20 15.2 25 25.59 4.01 11.88 a 33:29 © 824s 28 NE ae
ae 56 ae eee Sas At a Cavan 4A#1 B03 Spy PS SBR SOO iias
11#5 19.4 19.3 2.5 37.53 3.69 15.44 see Sma OE Ra eee HOS
eS on foe 3 AAGR 580 AOE 4A#1 BD 88) 32051) 0 2654249) Omens
antes 194 163 DS Aaah 2a eee 14#1 33.01 41.03 2.5 30.04 3.69 17.62
“uD ME AO AS) AAG AGE) TAGE 14#1 33.01 41.03 2.5 29.70 3.90 18.85
12#2 16 0.7 25 2594 3.78 14.04 14#1 33101 41003) 2558 3429) 423 enIaSD
14#1 IBS 269) 25) 20:85 876, SIS6 22 SUS Bee gy eee ead) sos
14#1 BS SOD AS TT BSB GT 22 30.4 39.9 3 35.30 4.18 16.90
16#1 26.3 30.1 AS OT BOL eA 22 30.4 39.9 > 37.46 4.32 27.50
16#1 26.3 30.1 DES 2 Sel S OS mls e28: 8 30.9 23.9 4 44.42 3.86 13.72
16#1 26.3 30.1 MS DEeO” BV AA 8 30.9 Me) al 3705) 4:47 eelGes
17#3 25.9 Bile OS AGS GME 87 8 30.9 23.9 1 50.39 4.02 16.04
17#3 25.9 31.7 AS SOO B59 Gey 8 30.9 2319) 4 Mle Mi TAG
18#2 24.9 29.7 AS OOM Ae NGA 8 30.9 23.9 4 44.62 3.75 12.76
18#2 24.9 29.7 25 29.74 3.76 16.13
20#1 OO ig DS MSG AOL iAGI Hep umes
20#1 19.9 17.1 VS Bln Al Sil 18 22.6 35.8 3 44.99 3.32 15.30
20#1 19.9 17.1 D5 WOl ANG Wie 18 22.6 35.8 3 3353 3.45 15.60
21 Shell 24 28.9 2.55 29.53 4.02 15.79 18 22.6 35.8 3 46.06 3.75 17.42
21 Shell 24 28.9 DS WO Aq) 1S 18 22.6 35.8 3 38.74 3.74 16.33
21 Shell 24 28.9 2551929193) Sh7 0406 18 22.6 35.8 3 53.83 3.36 14.16
22#1 24.9 30.2 D5 Weil BIO las 8 20 19.4 3 SETS S Sines
22#1 24.9 30.2 AS Me AO WCB 20 19.4 3 37.92 4.03 18.78
22#1 24.9 30.2 25 3089 3.54 14.83 8 20 19.4 3 A277 AOS 15.60
23#1 DBS 313) OFS) 26.94 B35) 15.56 8 20 19.4 2 37.34 4.42 15.65
26# | 27.6 31.6 2D 36:37 3270 14.94 8 20 19.4 3 BB 00m AS 15.35
26#1 27.6 31.6 25 3411 3.45 14.50
Table 2.—Continued.
OSTRACODE SHELL CHEMISTRY: DWYER AND CRONIN 259
Table 2.—Continued.
Sample
Site
Temp.
(°C)
Coretop Specimens
Florida Bay (July 1998)
ws)
NNN NY
DODO Deo
ed
oS
>>
20A
30.42
30.42
33.26
33.26
33.26
33.26
30.34
30.34
30.34
30.34
31.94
31.94
sil
31
31
31
31
31
31
31
31
31
31
31
32.
Sle
Sik
ln bo ww Ww
WW WwW
WHoWWwWWowWwWWwWwwwbw
NNNNNN WY
OO 0 O
BE
Salinity
(ppv)
Sample Temp. Salinity
VPI Mg/Ca Sr/Ca_ Na/Ca Site CC) (ppt) VPI Mg/Ca Sr/Ca Na/Ca
9 3122 BBR Ae AROS) SH) BSE
9 Sl sy 28.8 DES 38.72 3.89 16.57
25 35.87 3.85 16.27 9 S132 28.8 DES 38.89 4.01 13.27
25 36.09 3.58 18.30 9 31.32 28.8 2.5 28.74 3.90 15.20
25 33.85 3.90 15.77 9 Silas 28.8 25) Siva 3153) 14.71
5 OW al) ivigg 15 28:54) 2675) 2 400083710 654
Pry esis) S/K2) 16.99 Biscayne Bay (August 1997)
= see ie 15.50 BBSI Slee ww 33.14 3.64 16.39
a ae ae bass BBS1 3035) 14 3 EN "SUE ily Ze)
ae Ae BGG eG BBS1 Bk35 17.4 3 37.59 4.11 SHG.
ae Fase 413 16.69 BBS1 Siil33)5) 17.4 4 42.50 3.91 14.98
A) & 34.42 3.05 15.97 BBS1 31.35 17.4 3 32.50 3.46 Sys
D5 37.62 3.84 16.73 BBS1 31.35 17.4 2 31.02 $}55) 16.47
25 40.72 3.85 16.13 BBS1 31.35 17.4 3 40.05 3.64 16.55
25 42.00 3.77 17.44 BBS1 S1k35 17.4 2 41.04 3.83 17.35
2.5 36.14 3.69 15.89
25 37.58 3.69 14.73
25 42.42 3.80 17.89 perature. There also seems to be broad positive cor-
oe aes aul sae relation between Mg/Ca and temperature. Further-
25 3818 367 15.50 more, shells collected from July 1998, when both sa-
5) 30127) 30211 15.15 linity and temperature were relatively high, all have
2:5) 49:81 4214! 13:99 Mg/Ca ratios > ~30 mmol/mol.
2 SOUS! oUt ISS Sr/Ca ratios range from 3.3 to 4.6 mmol/mol and
25 4146 3.71 14.84 : : i : :
25 3625 3.75 15.06 appear to display a broad negative correlation with
25 3735 346 16.03 both salinity and temperature, especially within the
D5 Sails Bs 1 11,0)5) February data. Na/Ca ratios (not shown) range from
2.5 48.07 4.02 15.59 12 to 19 mmol/mol and show no variation with salinity
25 39.17 3.52 IS}ES)7/ or temperature.
IF) 42.45 4.28 14.08 : Le <.
25 40.73 3.68 1437 Core-top specimens (Table 2), consisting of both ar-
25 4859 3.38 £14.35 ticulated and disarticulated valves, yield Mg/Ca, Sr/Ca,
2S 019 37/6) 14°20 and Na/Ca ratios that are indistinguishable from those
25 46.59 3.80 17.16 of vegetation samples. This implies that post-mortem
DiS5eees4'570 3:83) lial os Barres : i
Wee. 4645 AT 17.40 effects on shell me/Ca ratios in Florida Bay are neg-
Diy 36.07 B77) 16.75
DS 36.67 3.86 14.84 ;
25 51.86 2.90 12.46 Modern Water Chemistry
IY) 44.47 3.24 15.40
DS 34.52 3.39 15.53
2.5 32.05 3.74 14.71
2.5 43.11 3.59 15.46
2.5 43.41 3.71 16.24
2.5 36.39 3.26 16.21 °
MS “S95 S75 i138 g
DES 43.07 4.04 17.16 re)
25 39.89 3.75 15.58 E
2 39.55 3.98 17.76 x
Pies) 36.94 Bear 15.11
2S 44.04 4.06 17.16
ae ae Aor ea oO Sr/Ca (mmol/mol)
ae i e ates Fh J ° Mg/Ca (mol/mol)
DS 34.23 3.51 13.89
2 Ee a S18 0 5 10 15 20 25 30 35 40 45
PMS) 42.95 3.96 14.50
SN 36.02 9337 15.86 Salinity (ppt)
EES EN GAN: Joust) Text-figure 5—Mg/Ca and Sr/Ca ratios of waters from Florida
PN) 46.61 3.69 15.99
and Biscayne Bays collected in 1997-99.
260 BULLETIN 361
Table 3—Chemistry of fossil shells of Loxochoncha matagordensis and Peratocytheridea setipunctata trom downcore portions of sediment
cores from Florida Bay. Core numbers correspond to map (Text-fig. 1). Mg/Ca, Sr/Ca, and Na/Ca ratios reported as mmol/mol. Intra-sample
average ratios reported in right three columns. Most shells analyzed were adult stage, though some juvenile molts were also analyzed. VPI
(visual preservation index) is a measure of shell clarity ranging from | (clear, unaltered) to 7 (opaque, highly altered).
Depth Mg/Ca Sr/Ca Na/Ca
(cm) Year Genus Species Molt VPI Mg/Ca Sr/Ca Na/Ca (avg.) (avg.) (avg.)
Russell Bank Core 19B
Site 12* 1998.0 bs matagord. A D5, 29.20 3.93 16.55 SPRIIT/ 3.89 15.50
Sitemi24 1998.0 fs. matagord. A De) 25.94 3.78 14.04
Site 12* 1998.0 iby matagord. A DES 38.14 3.83 16.53
Site 12* 1998.0 he matagord. A py) 35.41 4.02 14.87
1 1994.2 ib matagord. A 1 40.57 4.07 16.60 37.84 4.03 16.25
| 1994.2 EE matagord. A 2 37.07 4.17 16.81
| 1994.2 Ibs matagord. A 1 35.89 3.85 15.34
3 1992.5 1b, matagord. A 3} 39.82 4.08 15.67 38.39 3.95 16.88
3 1992.5 16, matagord. A 3 36.97 3.81 18.10
5 1990.9 EE: matagord. A 3 40.19 3.44 16.04 39.46 3.78 15.64
5) 1990.9 Ih, matagord. A 2 36.97 3.98 16.23
5 1990.9 Ios matagord. A 1 41.22 3.93 14.66
7 1989.3 il, matagord. A 3 36.87 3.90 Sy 727/ 39.61 3.94 15.67
a 1989.3 1B matagord. A 3 42.35 3.98 16.07
9 1987.6 Ie, matagord. A 2 32.10 3.60 16.63 36.08 3.74 15.95
9 1987.6 ik matagord. A 4 36.94 3.56 15.48
9 1987.6 ie matagord. A 4 39.19 4.06 15.74
11 1986.0 IL; matagord. A 2D) 31.43 4.18 17.40 32.48 4.09 16.50
11 1986.0 Ji matagord. A 3 31.39 Sn? 16.68
11 1986.0 ibs matagord. A 4 34.61 4.37 15.42
13 1984.3 Ibs matagord. A 2 29.87 3.83 18.11 34.92 3.78 16.09
13 1984.3 iE matagord. A 4 36.75 3.68 15.44
13 1984.3 Ib matagord. A 4 38.14 3.84 14.71
NS) 1982.7 is matagord. A 4 41.81 3.97 i13}.113} 35.10 3.91 14.79
1S 1982.7 iL. matagord. A Dy 33.30 3)57/5) 15.58
15 1982.7 IL. matagord. A 3 30.17 4.01 15.65
17 1981.1 ee matagord. A 4 31.30 3t73 13.99 34.99 3.56 14.97
wi 1981.1 IL: matagord. A 4 39.24 3.18 15.58
17 1981.1 Ib matagord. A 3 34.42 3.76 15%33
19 1979.4 iL. matagord. A 2 35.40 3:79) 15373) 40.09 3.64 15.12
19 1979.4 ik matagord. A 4 44.78 3.50 14.50
21 1977.8 IE: matagord. A 3 36.75 4.27 18.29 36.75 4.27 18.29
23 1976.1 EE matagord. A 3 32.97 3.68 16.25 38.00 35)! 15.14
23 1976.1 BS matagord. A 3 43.43 3.36 15.78
23 1976.1 iE. matagord. A 3 37.60 Boi 13.40
25 1974.5 E, matagord. A 3 37.30 3.85 13.09 36.02 3.74 14.76
25 1974.5 Ib matagord. A 3 35.18 3.60 15.08
PPS) 1974.5 ib matagord. A 3 35.59 S75) 16.12
27 1972.9 ik, matagord. A 3 36.44 3.95 16.60 36.67 3.68 SSS)
27 1972.9 Jk. matagord. A 3 Sik25 3.60 16.36
27 1972.9 LE. matagord. A 3 42.33 3.50 12.42
29 1971.2 il, matagord. A 3 30.07 3555 16.42 SE95 3.78 16.92
29 1971.2 Ex matagord. A 3) 33.82 4.02 17.42
31 1969.6 1. matagord. A 3 28.06 3152 72) 37.85 3.78 15.53
31 1969.6 iL. matagord. A 4 42.47 4.13 14.05
31 1969.6 I: matagord. A 3} 43.01 3.68 15.36
33 1968.0 ibs matagord. A 4 33.07 3.88 15.93 36.21 4.09 15.39
33 1968.0 i matagord. A 3 39.89 4.24 15.71 39.59 3.78 15.11
33 1968.0 Ue: matagord. A D 35.68 4.13 14.53
35 1966.3 ib. matagord. A 3 41.36 4.06 15.71
35) 1966.3 ik: matagord. A 3 41.07 Si52 Sy 7/tl 41.37 4.03 15.63
35 1966.3 Ib, matagord. A 3} 36.34 ST 13.92
37 1964.7 ik. matagord. A 3 39.80 4.01 15.87 46.37 3.92 15.11
37/ 1964.7 JE; matagord. A 3 42.94 4.05 15.38
39 1963.0 ib. matagord. A 3 48.80 4.05 12.16
39 1963.0 ib. matagord. A 3 46.63 3.74 S\S7/7/ 41.16 4.09 15:29)
OSTRACODE SHELL CHEMISTRY: DWYER AND CRONIN 261
Table 3.—Continued.
Depth Mg/Ca Sr/Ca Na/Ca
(cm) Year Genus Species Molt VPI Mg/Ca Sr/Ca Na/Ca (avg.) (avg.) (avg.)
39 1963.0 IL matagord. A 4 43.68 3.97 17.39
41 1961.4 i. matagord. A 2 41.97 3.94 17.50
41 1961.4 Ibe matagord. A 3 40.36 4.25 13.08
43 1959.8 I, matagord. A 3 33.40 4.17 IIS)7/5)} 41.29 3.92 16.19
43 1959.8 I, matagord. A —- 49.18 3.66 16.63
45 1958.1 IY, matagord. A 3 35.52 3.82 14.83 35.52 3.82 14.83
47 1956.5 Ibe matagord. A 3 39.51 3.27 712 42.27 3.77 15.51
47 1956.5 IU, matagord. A 3 48.59 4.00 13.49
47 1956.5 IL. matagord. A 3 38.70 4.02 15.93
49 1954.8 Ie matagord. A 3 36.08 3.93 15.54 42.15 3.81 14.81
49 1954.8 I. matagord. A 3 46.31 3.44 14.75
49 1954.8 1b. matagord. A 3 44.06 4.04 14.15
51 1953.2 IE. matagord. A 3 28.07 3.65 17.61 33.74 3.90 15!53)
51 1953.2 iL matagord. A 4 28.57 3.98 13.90
51 1953.2 Ik matagord. A 3 44.57 4.05 15.09
53 1951.6 ike matagord. A 3 31.80 3.48 15.48 37.71 3.80 14.54
38 1951.6 IE: matagord. A 3 40.79 4.20 14.00
53 1951.6 Ie: matagord. A 3 40.54 Sh9/5} 14.15
Si) 1949.9 jt matagord. A 355 36.35 3.88 13.88 35.38 3.86 14.96
55 1949.9 IE matagord. A 35) 35.20 3.92 16.28
55 1949.9 Ee matagord. A 3h5) 34.60 S/7/ 14.73
Si 1948.3 L. matagord. A 3.5 42.42 4.05 13.05 37.10 3.98 15225
57 1948.3 iE matagord. A She) 31.49 3.80 14.80
57 1948.3 6% matagord. A 355 37/39 4.09 17.92
59 1946.6 L. matagord. A 3.5 44.98 3133 14.41 41.53 3.65 1Sal2
59 1946.6 ik. matagord. A 355 36.70 3.88 16.87
59 1946.6 ik. matagord. A 35) 42.90 3.76 14.08
6l 1945.0 16 matagord. A 355 34.42 3.89 13372 41.67 3.57 14.99
61 1945.0 ibs matagord. A 3:5 48.91 325 16.26
63 1943.4 1B matagord. A 3 33.84 BP? 18.16 32.01 3.86 16.41
63 1943.4 Ik, matagord. A 385 26.60 3.93 15.67
63 1943.4 ie matagord. A 315) 35.59 3:93 15.39
65 1941.7 JE. matagord. A 35 42.19 4.19 15.56 35.44 3.70 15.63
65 1941.7 jks matagord. A 355 PIAA 3.22 15.40
65 1941.7 Ie matagord. A 3.5 38.35 3.68 15.92
67 1940.1 L. matagord. A ys) 26.85 375 14.89 33.16 3.91 15.45
67 1940.1 ik. matagord. A B35 39.48 4.08 16.00
69 1938.4 ik. matagord. A 3.5 31.24 3.94 16.06 38.52 3.62 15.15
69 1938.4 If, matagord. A 3.5 36.77 3.60 17.20
69 1938.4 1b. matagord. A Shp) 47.54 3.31 12.18
71 1936.8 b- matagord. A 35 35.45 3.94 20.43 32.76 4.06 E27,
71 1936.8 bs matagord. A Sha) 32.04 4.15 14.21
71 1936.8 Ee matagord. A BD 33.75 4.25 17.10
71 1936.8 Ih. matagord. A Sh5) 29.80 3.89 17535
73 1935.2 ib. matagord. A 3:5 43.92 S/T) 15.55 39.82 3.74 15.44
73 1935.2 Jb, matagord. A 3:5 29.10 3.94 18.92
73 1935.2 Ik matagord. A Si) 46.43 3.50 11.85
75 1933.5 ies matagord. A 3}5) 33.46 3.66 13.69 34.40 3}5y/ 14.66
75 1933.5 Ik. matagord. A 3}.5) 83555) 3.39 15.08
75 1933.5 Ib. matagord. A 3) 36.18 3.67 15222
77 1931.9 Ib matagord. A 35) 34.49 4.19 16.18 39.61 3.97 14.41
77 1931.9 Ibe matagord. A 3.5 44.74 ae) 12.64
79 1930.2 Ib matagord. A BS 36.38 3.85 14.29 38.18 3.85 15.18
79 1930.2 iE. matagord. A 3h5) 29.62 3.81 15.96
79 1930.2 Ib. matagord. A 3) 48.54 3.89 15.30
81 1928.6 Ee matagord. A 3.5 32.32 4.21 15.21 39.08 3.85 14.61
81 1928.6 ile matagord. A 3.5 42.56 Sh Si1/ 14.32
81 1928.6 E. matagord. A 315) 42.35 3.77 14.30
83 1927.0 ik. matagord. A 3.5 32.54 3.91 14.71 35.98 3.96 15.62
83 1927.0 ibs matagord. A 3.5 42.88 3.98 16.96
262 BULLETIN 361
Table 3.—Continued.
Depth Mg/Ca St/Ca Na/Ca
(cm) Year Genus Species Molt VPI Mg/Ca Sr/Ca Na/Ca (avg.) (avg.) (avg.)
83 1927.0 Is matagord. A 355 32.53 4.00 15.20
85 1925.3 L. matagord. A 355) 34.11 3.47 17.69 32.88 3.73 15.55
85 1925.3 IL matagord. A 355) 31.65 3.99 13.42
87 1923.7 IW matagord. A 3'5 33559) 3.83 16.18 39.36 3.70 16.40
87 1923.7 L: matagord. A 3S) 41.05 3.80 17.83
87 1923.7 IE: matagord. A 3.5 43.44 3.47 Heyl)
89 1922.0 Ie matagord. A 35) 37.86 4.20 14.07 35.43 4.12 14.22
89 1922.0 IL matagord. A 35) 34.23 4.05 14.43
89 1922.0 Ib matagord. A 3h) 34.21 4.12 14.16
93 1918.8 IL matagord. A 35 34.39 3.84 15.34 38.23 3.67 15.30
93 1918.8 1: matagord. A 3.5 32:25) 3.40 15.10
93 1918.8 IE matagord. A 355 28.03 S77 15.48
95 1917.1 ES matagord. A 3}5) 32.87 4.12 14.68 32.87 4.12 14.68
97 1915.5 Ih, matagord. A 3°) 28.97 3.60 14.84 28.97 3.60 14.84
99 1913.9 JE. matagord. A 3 32.60 3.98 16.64 32.60 3.98 16.64
101 1912.2 TE matagord. A 35 36.37 3.59 15.51 36.37 3559 15.51
103 1910.6 Ib: matagord. A 3:5 36.49 3.65 14.09 83222: 3.68 13.85
103 1910.6 Ib; matagord. A BS 32.29 3.79 13.60
103 1910.6 IL, matagord. A 3°5 30.87 3.60 13.84
105 1908.9 je. matagord. A Br 32.82 3.98 13.66 37.66 3.84 15.23
105 1908.9 IE matagord. A Shp) 45.68 3.70 16.77
105 1908.9 iL; matagord. A 355 34.47 3.84 15.25
107 1907.3 Es matagord. A 35 31.31 4.03 16.13 36.32 4.11 15.14
107 1907.3 Es matagord. A BES 41.34 4.20 14.15
109 1905.7 Ib. matagord. A 355 36.09 3.76 16.67 33.07 3.73 16.11
109 1905.7 is matagord. A Bh) 35.36 3.80 IAEQil
109 1905.7 I 6 matagord. A 3}5) 27.76 3.64 13.76
113 1902.4 IL. matagord. A 35) S255 3.66 13.96 32°38) 3.67 14.69
113 1902.4 ES matagord. A 35) 32°92. 3.68 15.43
115 1900.7 ib. matagord. A 35 32.88 3.88 MAST, 32.88 3.88 NZ/E3i7/
119 1897.5 es matagord. A Bi) 48.35 3.43 15.21 48.35 3.43 15.21
121 1895.8 js matagord. A 35) 33.64 3.64 15.28 33.64 3.64 15.28
123 1894.2 iE matagord. A 3.5 37.04 3.84 15.89
123 1894.2 ih matagord. A 3.5 35.54 3.60 15.10
125 1892.5 Ib matagord. A Sh) 36.19 4.05 13:33) 85:35) 3.80 16.59
125 1892.5 iL. matagord. A SiS) 32.70 4.08 22.80
125 1892.5 Ii matagord. A 355 37.16 3.26 13.63
129 1889.3 es matagord. A BES) 38.87 3152 15.02 33.40 3.70 15.61
129 1889.3 Ike matagord. A B3e5 29.21 3.84 i>yils)
129 1889.3 L. matagord. A 335) 32.13 3.74 16.69
131 1887.6 ik matagord. A 355) 34.92 3.82 18.10 37.87 3.86 18.63
131 1887.6 Ik, matagord. A 315) 40.82 3.90 19.15
137 1882.7 Ik matagord. A 39) 47.33 3.78 15.79 STPSi, 3.74 14.52
137 1882.7 Ik; matagord. A 3h5) 28.12 3:99) 14.54
137 1882.7 ik. matagord. A 3'5 37.26 3.45 13.22
137 1882.7 Ee matagord. A 35 35.56 3572 177-333}
141 1879.4 ik, matagord. A 3.5 34.89 3.87 17.01 34.89 3.87 17.01
143 1877.8 IE, matagord. A 3h) 39.58 3.99 15632 40.26 3.84 14.80
143 1877.8 1E; matagord. A 355 42.35 3.89 13.19
143 1877.8 L. matagord. A 3.5 38.87 3.65 15.90
Bob Allen Keys Core 6A
! 1994.0 12% Setipunct. A or A-1 1 18.70 4.35 11.43 18.70 4.35 11.43
1 1994.0 12 setipunct. A or A-1 2 31.30 32011 12.76
] 1994.0 12 setipunct. A or A-1 2) 30.50 4.03 13.05
1 1994.0 12 setipunct. A-2? 2 25.22 4.11 12.52
3 1992.1 1 setipunct. A 2; 20.27 3198 Se 20.27 3.93 LIS
3 1992.1 Ie setipunct. A-1 or A-2 1 20.88 4.58 12.25
3 1992.1 P. setipunct. A-3 1 20.12 3.99 12.55
3) 1992.1 (P setipunct. A-2 1 22.82 4.16 12.61
Table 3.—Continued.
Depth
(cm)
iS)
mMyNy hd td to
nn
~
101
101
105
107
107
109
113
113
117
121
125
125
127
133
Year
1986.3
1986.3
1986.3
1980.6
1980.6
1980.6
1978.7
1976.7
1976.7
1974.8
1974.8
1972.9
1971.0
1971.0
1969.0
1969.0
1965.2
1965.2
1963.3
1963.3
1961.3
1961.3
1959.4
1959.4
1955.6
1953.7
1953.7
1949.8
1949.8
1938.3
1934.4
1934.4
1930.6
1928.7
1926.7
1926.7
1921.0
1919.0
1919.0
1915.2
1L9I3"3
1911.3
1911.3
1907.5
1907.5
1905.6
1897.9
1897.9
1894.0
1892.1
1892.1
1890.2
1886.3
1886.3
1882.5
1878.7
1874.8
1874.8
1872.9
1867.1
Genus
UNO MO URC UN URU NU RTUNC UN UR IU UUs UI Ul UUs, Ul. Un Ua 0) Ui UN Ul Ul Un sUrU: ss Un Un Uli 0! 5.20) 50 AUS 20) FU, AU) sU) FU) GUIrU FU 0 80 Ful Fu! Te) Tue! ro itd
Species
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
Setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
Setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
setipunct.
Setipunct.
setipunct.
setipunct.
setipunct.
OSTRACODE SHELL CHEMISTRY: DWYER AND CRONIN 263
Mg/Ca St/Ca Na/Ca
Molt VPI Mg/Ca Sr/Ca Na/Ca (avg.) (avg.) (avg.)
A or A-1 D 21.97 4.14 13.92 21.81 4.16 14.15
A or A-1 2 21.65 4.18 14.39
A-2 2 31.15 4.11 11.82
A I 22.22 4.15 14.27 22.22 4.15 14.27
A-l I 17.69 4.14 11.27
A-4 l 27.89 3.98 8.85
A-2 1 32.08 4.44 13.65
A-2? 2 26.26 3.99 11.79
A-3 2 19.93 3.94 11.08
A? I 22.57 4.30 12.83 26.53 4.39 13.19
A 1 30.49 4.48 13.54
A I 23.03 4.11 13.00 23.03 4.11 13.00
A 2 17.63 4.11 14.17 17.97 4.13 14.24
A 2 18.31 4.16 14.31
A l 22.56 4.12 11.35 21.82 3.82 10.94
A p 21.08 3.51 10.53
A 2 24.66 4.55 13.69 24.41 4.38 13.24
A or A-l 2 24.15 4.21 12.79
A 2 21.70 4.41 14.40 21.42 4.44 14.38
A 2 21.15 4.46 14.36
A 2 24.04 4.24 12.34 23.27 4.18 12.11
A 2 22.50 4.13 11.87
A 2 20.44 4.00 10.83 20.35 4.05 12.25
A 2 20.26 4.11 13.67
A 2 16.36 4.09 12.47 16.36 4.09 12.47
A 2 17.39 3.82 12.50 17.54 4.02 12.45
A 2 17.69 4.22 12.40
A D 19.61 4.36 13.72 19.46 4.37 13.96
A 2 19.32 4.38 14.20
A 2 24.36 3.83 10.39 4.36 3.83 10.39
A l 24.49 4.12 12.59 23.34 4.17 12.91
A 2 22.19 4.21 13:23
A 2 25.86 4.26 13.93 25.86 4.26 13.93
A I 20.71 4.39 14.09 20.71 4.39 14.09
A 2 29.29 4.22 13.29 26.33 4.23 12.66
A 2 23.36 4.25 12.02
A 2 18.58 4.07 13.04 18.58 4.07 13.04
A D 18.91 3.94 12.13 19.93 4.13 12.48
A l 20.96 4.32 12.82
A 2 21.36 4.14 13.46 21.36 4.14 13.46
A 2 23.06 4.11 13.03 23.06 4.11 3.03
A 2 21.83 4.11 12.43 20.96 4.23 12.29
A 2 20.09 4.35 12.16
A 2 24.06 4.73 12.07 20.53 4.31 12.40
A 2 17.00 3.90 12.73
A 2 16.90 4.27 13.58 16.90 4.27 13.58
A 2 17.82 4.03 11.15 18.64 4.18 11.06
A D) 19.45 4.34 10.96
A 2 23.10 4.26 12.42 23.10 4.26 12.42
A 1 22.67 4.43 11.92 22.79 4.37 11.95
A I 22.90 4.32 11.98
A 2 19.20 4.02 12.49 19.20 4.02 12.49
A 2 15.02 4.08 11.43 17.44 4.09 11.89
A 19.87 4.11 12.35
A 2 22.47 4.29 13.09
A 2 19.04 4.05 i17/
A 2 16.69 3.31 10.27 17.43 3.89 11.93
A 2 18.18 4.48 13.59
A 2 20.75 4.03 12.62
A 1 16.55 4.18 12.81 17.07 3.99 12.14
264
Table 3.—Continued.
Depth
(cm) Year Genus Species
133 1867.1 12, setipunct.
135 1865.2 Ps setipunct.
137 1863.3 12 setipunct.
137 1863.3 je, setipunct.
139 1861.3 P: setipunct.
141 1859.4 iP: setipunct.
141 1859.4 P: setipunct.
141 1859.4 ie setipunct.
143 1857.5 iP: setipunct.
143 1857.5 P: setipunct.
143 1857.5 12, setipunct.
145 1855.6 IZ setipunct.
145 1855.6 P: setipunct.
147 1853.7 P: setipunct.
151 1849.8 PR: setipunct.
153 1847.9 len setipunct.
153 1847.9 PB. setipunct.
155 1846.0 ie setipunct.
155 1846.0 12 setipunct.
157 1844.0 12 setipunct.
Little Madeira Core T24
1 1991.4 IE setipunct.
1 1991.4 iP: setipunct.
11 1955.7 P. setipunct.
11 1955.7 lem setipunct.
27 1898.6 P: setipunct.
33 1877.1 iP setipunct.
37 1862.9 P. setipunct.
37 1862.9 12 setipunct.
39 1855.7 12 setipunct.
49 1820.0 1e. setipunct.
53 1805.7 P. setipunct.
59 1784.3 12 setipunct.
79 1712.9 P: setipunct.
Park Key Core 23A
Site 11 1998.0 E: matagord.
1 1993.7 IBS matagord.
1 1993.7 L. matagord.
1 1993.7 ib matagord.
3 1991.2 Ig; matagord.
3 1991.2 Ik; matagord.
3 1991.2 L. matagord.
5 1988.6 Ik, matagord.
5 1988.6 JE matagord.
5 1988.6 ibs matagord.
7 1986.0 Te matagord.
7 1986.0 ths matagord.
7 1986.0 IE matagord.
9 1983.5 | Be matagord.
9 1983.5 1B matagord.
9 1983.5 Ihe matagord.
11 1980.9 Ibs matagord.
11 1980.9 Ibs matagord.
11 1980.9 bs matagord.
13 1978.3 Ee matagord.
13 1978.3 E. matagord.
15 1975.8 E. matagord.
15 1975.8 Ee matagord.
17 1973.2 Ibs matagord.
Prrrrrrrrrrrrrrrrrre
| >>>
rrr rrrrr sd
~)
rrr rrrrrrrrrrrrrrrrrerre
Molt
BULLETIN 361
VPI
NN
wn
NNNNNNNMNNNNNN VN
Bo ey eo © CCE || eFNWN
34.569
S522
36.764
Sr/Ca
3.80
4.05
4.23
4.05
4.30
4.09
4.14
4.18
4.25
3.61
4.59
4.56
3.48
4.31
4.02
4.39
3.97
Na/Ca
11.47
9.88
12.76
12.68
13.07
13.10
12.24
10.84
12.39
9.63
12.36
11.44
9.98
10.31
11.66
12.71
10.46
12.68
12.26
12.64
8.64
10.54
10.00
10.26
9.96
8.64
8.23
11.05
10.19
6.74
10.03
8.92
8.49
14.895
15.444
17.609
23.407
16.441
16.880
17.019
14.916
14.189
18.559
19.459
14.429
14.765
17.307
17.574
15.879
15.149
13.350
19.202
14.738
20.084
14.296
13.189
19.534
Mg/Ca
(avg.)
19.82
20.21
22.82
17.94
23.29
20.11
20.48
26.66
30.70
NNN
= WwW
fOr
wn
~_
mnnv
wn
in
Nn
sot
D Ww
AN
35.69
37.42
40.07
36.89
37.56
41.27
34.92
38.09
Sr/Ca
(avg.)
4.14
4.15
3.96
373.
3.78
Na/Ca
(avg.)
11.46
9.59
10.13
9.96
8.64
9.64
10.19
6.74
10.03
8.92
8.49
14.895
18.82
16.78
15.89
16.22
16.92
15.90
17.41
13.74
16.40
Table 3.—Continued.
OSTRACODE SHELL CHEMISTRY: DWYER AND CRONIN
Depth
(cm)
17
17
Ow =
NNNNNNDN LV
NIX NUNN WwW
tN
~
45
59
Sy)
Year
1973.2
1973.2
1970.6
1970.6
1970.6
1968.1
1968.1
1968.1
1965.5
1965.5
1965.5
1962.9
1962.9
1960.4
1960.4
1960.4
1957.8
1957.8
1955.3
1955.3
1955:3
1952.7
1950.1
1950.1
1950.1
1947.6
1947.6
1947.6
1945.0
1942.4
1942.4
1939.9
1939.9
1939.9
1937.3
1937.3
1937.3
1934.7
1934.7
1934.7
1932.2
1932.2
1929.6
1929.6
1927.1
1927.1
1927.1
1924.5
1924.5
1921.9
1921.9
1919.4
1919.4
1919.4
1916.8
1916.8
1911.7
1960.5
Genus
Species
JSC URC SE Teer Sst fest [est fest dese lest jest fest fest Jose Jesh de=Falentolesialesu[estlest [St lente dgsd testitest deal leatalead dss tiles testes Tiles il este let lel ctl estal cfc] alesis etilea lls tal osted eas lel ss eilss lest ll sal esol Il ca)
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
matagord.
rrr PrPrrrrrrrrrerrrrrrrrrPrrPrPrPrPrPrPerPrPrrPrrrrrrrrrrrrrrrrrrrrrrrrd>d
Mg/Ca Sr/Ca Na/Ca
39.529 3.834 13.869
37.99] 3.627 15.795
40.801 3.941 16.625
46.646 3.505 17.558
34.976 4.248 19.266
43.454 3.955 18.219
38.430 3.916 18.228
31.993 3.815 15.805
41.083 3.730 13.934
44.853 3.805 17.531
44.190 3.815 12.135
43.813 3.832 14.306
27.063 3.099 10.673
37.161 3.845 15.234
35.081 3.762 14.894
39.754 3.661 18.617
42.313 4.079 14.168
34.934 4.025 15.124
38.633 3.834 18.052
36.882 3.639 15.555
47.863 4.055 13.613
38.465 35712 18.343
32.028 3.720 14.424
35.132 3.618 15.580
46.068 3.852 15.419
39.966 3.877 17.063
36.811 3.928 17.330
49.168 4.240 23.390
37.609 3.710 16.161
39.738 3.832 11.676
43.387 4.214 14.597
36.224 3.436 14.897
40.645 4.357 19.004
33.668 3.645 14.923
40.307 3.804 16.482
38.335 3.540 16.605
38.279 4.019 15.730
SESH 31935 16.233
34.743 3.170 13.040
34.881 3.667 16.925
33.969 3.978 15.483
46.949 4.500 13.614
34.572 3.568 14.314
35.948 4.032 18.054
30.785 375i 20.072
46.299 3.431 14.939
31.636 3.900 13.483
40.095 4.165 16.014
37.119 3.855 17.116
39.917 3.587 17.332
37.298 3.763 16.494
44.702 S725 13.998
42.021 3.638 15.739
43.679 4.002 16.583
37.332 4.188 16.794
47.322 3.608 17.898
32.995 SV IBY) 19.137
31.379 3.681 16.232
Mg/Ca
(avg.)
40.81
37.96
38.62
41.13
37.74
41.98
37.61
41.56
36.85
38.97
35.59
33.00
35.54
Sr/Ca
(avg.)
3.90
3.90
4.01
Na/Ca
(avg.)
17.82
17.42
14.65
15.74
19.26
16.16
13.14
16.27
16.27
15.40
16.91
19.14
16.12
Table 3.—Continued.
BULLETIN 361
Depth
(cm) Year Genus Species
69 1906.5 Ib matagord.
69 1906.5 Ib matagord.
7p 1904.0 Ie: matagord.
71 1904.0 Te, matagord.
71 1904.0 E. matagord.
73 1901.4 os matagord.
73 1901.4 L. matagord.
73 1901.4 IG, matagord.
75 1898.8 Ibs matagord.
WY 1896.3 1B matagord.
77 1896.3 E. matagord.
77 1896.3 1B matagord.
79 1893.7 1B matagord.
19) 1893.7 E. matagord.
a2) 1893.7 Es: matagord.
83 1888.6 ibe matagord.
83 1888.6 Ih matagord.
83 1888.6 1B matagord.
Manatee Bay Core MB1
1 — EE. matagord.
1 — 1B matagord.
1 — | es matagord.
3 — EE. matagord.
3 — E matagord.
3 = E. matagord.
5 a Es matagord.
5 — hs matagord.
=) —_— iD matagord.
7 —- ES matagord.
7 —_ jb matagord.
7 — ES matagord.
9 — E. matagord.
9 — j be matagord.
9 —_— L. matagord.
11 — TE matagord.
11 — Ii matagord.
11 —_— E. matagord.
13 — 1B matagord.
13 — ike matagord.
13 — E. matagord.
15 —_ Be matagord.
15 — E. matagord.
15 — ib; matagord.
19 — Ib, matagord.
19 — 1B matagord.
19 — Ik. matagord.
21 — Jb matagord.
21 — / Bs matagord.
21 — ] Bs matagord.
25 — 1b matagord.
25 — iL. matagord.
25 — LE. matagord.
27 — 18s matagord.
27 — Ibs matagord.
AG] — [hs matagord.
29 — Ik. matagord.
29 — iE. matagord.
29 — Ibs matagord.
31 — J be matagord.
>PrPrPrrrrrrrrrrrrerre
>SPPPr rr rrPrPrPrrrrEESeerrrrrrrrrrrrrrrrePrrrrrr
VPI
Wh PWW PW WW W
es)
WNONWH WWE HK KWH HE
WN hy WwW
WW WwW WwW
oe WW
Www
Sr/Ca
3.825
3.495
3.837
3.877
S92
3.648
4.142
4.164
3.740
3.959
3.910
3.676
3.973
4.064
3.700
3.954
3.702
4.178
3.62
3.89
3.91
4.09
4.23
Si
3.71
3.85
4.19
3.92
3.83
4.26
4.31
4.22
3.78
3.91
4.13
4.10
3.61
3.60
4.38
4.31
4.13
4.00
3.59
3.61
4.07
3:93
S77
4.09
3.64
3.72
4.23
3.68
3.40
3.58
4.00
3.47
B07)
3152
Na/Ca
15.165
16.964
14.760
18.267
14.161
16.474
13.580
17.895
15.468
19.349
15.632
16.025
16.350
15.975
14.722
13.319
17.053
18.043
17.07
14.76
14.18
12.61
24.29
14.87
14.47
15.87
12.98
13.68
15.56
16.03
16.77
15.25
14.10
1539
16.93
13.79
14.47
10.24
17.05
16.24
SESS)
13.69
12.09
13.66
12.93
18.29
itil}
12.34
13.36
16.79
15.85
15.78
14.11
14.34
13221
14.85
14.76
12.85
Mg/Ca
(avg.)
36.84
33.69
35.02
36.78
38.91
33.91
37.54
36.26
37.46
40.02
40.75
31.66
36.51
40.31
Sr/Ca
(avg.)
us
(or)
ie)
eS)
\o
i)
4.00
4.10
4.15
3.76
WwW
\o
Ww
3.47
Na/Ca
(avg.)
15273)
15.68
15.19
15.34
17.26
14.44
15.09
15.37
15.37
13.92
15.09
15.33
14.75
14.28
13.22
OSTRACODE SHELL CHEMISTRY: DWYER AND CRONIN 267
Table 3.—Continued.
Depth Mg/Ca Sr/Ca Na/Ca
(cm) Year Genus Species Molt VPI Mg/Ca St/Ca Na/Ca (avg.) (avg.) (avg.)
31 — Ee. matagord. A 2 34.09 3.43 13.58
33 — Ib. matagord. A 3 33.24 3332 10.93 38.04 3.69 12.56
33 — Ibs matagord. A 3} 42.84 4.06 14.19
35 — EE matagord. A 4 31.02 3.56 14.80 34.18 Shi/il 15.46
35 — T.. matagord. A 37.34 3.86 16.12
37 _ L. matagord. A 3 42.76 3.80 14.69 40.60 3.78 13.48
37 a IE matagord. A 3 38.44 S77 12.27
39 — L. matagord. A 3) 36.24 3.58 14.14 41.04 3.63 14.30
39 = iB; matagord. A 3 40.93 355 15.08
39 —_— Ee. matagord. A 3 45.96 SET) 13.68
41 — Ike matagord. A 3 41.12 3.30 11.88 37.72 3.71 12.72
41 — IES matagord. A 2 39.08 3.91 14.16
41 os EE matagord. A 2 32.97 3.93 12:12
43 — Ib matagord. A 3 35.84 3193 14.50 35.84 3.93 14.50
45 — E. matagord. A 3 36.79 3.64 12.61 Slat 3.66 Ne Igi7/
45 _ Jb matagord. A 4 25:95 3.67 13.74
47 — IE matagord. A 3 39.23 S313 13.96 42.06 3:38 14.17
47 — Ibs matagord. A 3 41.76 3.38 13.80
47 — Te matagord. A 3 45.18 3.49 14.75
49 = IE. matagord. A 3 33.74 Sail 113'58) 39.42 3.45 13:33)
49 os jE. matagord. A 3 39.95 3.09 13.72
49 _- jb. matagord. A 3 44.58 3.47 12.73
Sl oS IE. matagord. A 4 40.64 3.65 13.92 35.47 3.62 14.16
Sil — Te matagord. A 3 29.99 3.80 EV 78}
51 a Ibe matagord. A 4 35579) 3.41 14.82
53 — jE matagord. A 3} 36.93 3.42 14.52 36.95 3.64 14.60
53) _ 1% matagord. A 3 35.61 3.61 13.45
53 — Es matagord. A 2 38.31 3.90 15.82
55 — IL, matagord. A 3 32.62 3.60 17.79 37.04 3.61 16.09
55 — cE matagord. A 4 38.62 3.48 15.51
55 — E. matagord. A 3) 39.87 375) 14.98
57 —_ 1b matagord. A 4 26.06 6.83 143.25 33.10 4.83 57.34
Sy/ — Ee: matagord. A 3 35.01 3.98 12.33
57 — IL. matagord. A 3 38.22 3.67 16.44
59 — IE. matagord. A 3 40.17 3.79 15.41 39.89 3.90 15.02
59 —_— 1B, matagord. A 3 3750 4.04 15.78
59 — IE. matagord. A 2 42.00 3.88 13.87
61 — Es matagord. A 2 26.79 3.46 14.73 32.40 3.49 14.30
61 — Ibs matagord. A 3 36.57 3.98 14.76
61 — IES matagord. A 3 33.83 3.05 13.40
63 — iE matagord. A 4 35.64 3.54 12:39 36.70 Br? 14.21
63 — 183 matagord. A 2 33.18 3.60 15.80
63 os EL. matagord. A 3 41.26 3.42 14.44
65 — JE matagord. A 3) 37725) 3.43 14.57 37.25 3.43 14.57
67 — EE matagord. A 3) 43.09 4.29 15.92 40.62 S107 14.41
67 — ih matagord. A 3} 35.03 3.75 eats
67 — IE: matagord. A 4 43.74 3.86 14.19
69 — iL. matagord. A 5 34.67 3.36 14.97 34.67 3.36 14.97
71 — IL, matagord. A 2 33.00 3.60 11.71 34.40 S12: 13.59
71 os ibs matagord. A 2 35.79 3.85 15.48
73 a Ihe matagord. A 3 35.69 3.64 14.28 37.16 3.81 14.20
73 — Jb matagord. A 3 47.17 3.78) 14.32
73 — Ib matagord. A 28.62 4.00 14.01
75 — IE matagord. A 36.67 3.99 15.10 41.07 4.00 16.15
75 — IE. matagord. A 45.46 4.02 17.20
117 = Je setipunct. A 5 24.49 3.91 8.94 23.96 3.61 9.07
117 — P: setipunct. A 5 21.25 B35 8.44
117 — P: setipunct. A 5 26.15 Shei) 9.84
115 — P. setipunct. A 5 24.22 3.97 6.64 23.79 3.67 7.71
Table 3.—Continued.
BULLETIN 361
Molt
Depth
(cm) Year Genus Species
115 — PR setipunct. A
115 — 1p setipunct. A
113 — Ie setipunct. A
113 = P. setipunct. A
107 — P. setipunct. A
107 — P setipunct. A
105 — P setipunct. A
103 — 12 setipunct. A
103 = 12 setipunct. A
103 — 1? setipunct. A
101 — 1 setipunct. A
101 — P, setipunct. A
99 — Pp setipunct. A
9) = P. setipunct. A
97 — 12 setipunct. A
97 — Pp setipunct. A
97 — Pp setipunct. A
95 — Pp: setipunct. A
95 — P. setipunct. A
93 — 12% setipunct. A
93 — 1% setipunct. A
91 — Ps setipunct. A
91 — 12 setipunct. A
89 — iP setipunct. A
89 — / 2; setipunct. A
89 = 12. setipunct. A
87 — P: setipunct. A
85 — 12 setipunct. A
83 —_— 12. setipunct. A
83 — 1 setipunct. A
83 — 12 setipunct. A
81 — Be setipunct. A
81 _ P. setipunct. A
81 — lez setipunct. A
79 Pp: setipunct. A
79 — RP: setipunct. A
79 — le setipunct. A
77 — 1? setipunct. A
WAL — P: setipunct. A
75 — 12 setipunct. A
75 — IP. setipunct. A
Ws) — PB: setipunct. A
73 — Bs setipunct. A
73 so RB: setipunct. A
73 — P. setipunct. A
71 — Je setipunct. A
71 — Je, setipunct. A
67 — Jee setipunct. A
67 — P; setipunct. A
63 — PR: setipunct. A
61 — P: setipunct. A
61 a IP: setipunct. A
55 — P: setipunct. A
333) —- 12 setipunct. A
33 _— Pp: setipunct. A
51 — ied, setipunct. A
VPI
Nan h BOAAKHKHKHNANnaAnnnnnnnnnn
wh RUAN HL
WNWNA LA A
n
WwwWwn kW
WWNN WwW
WWnN WwW ty
Sr/Ca
Na/Ca
8.66
7.82
11.28
8.80
6.60
6.68
5.10
10.90
7.59
15.73
4.54
7.90
5.58
6.24
9.30
8.88
10.01
7.01
7.19
9.08
8.46
8.74
8.18
6.03
8.16
6.29
WAS)
9332
8.92
9.88
6.79
7.00
13.08
12.69
9.94
9.96
9.71
6.13
8.23
7.86
S)5)9)
8.96
8.14
UT
8.08
8.31
2.92
8.89
8.63
8.92
9.09
8.21
6.82
8.44
q95
8.57
Mg/Ca
(avg.)
29,11
Sr/Ca
(avg.)
4.00
3.85
Na/Ca
(avg.)
7.10
8.46
6.83
Vas)
932
8.53
7.00
12.89
9.87
7.18
8.78
8.00
OSTRACODE SHELL CHEMISTRY: DWYER AND CRONIN 269
15 20 25 30 35 40 45
Salinity (ppt)
Salinity (ppt) Water Temperature (°C)
e Febraury 1998 + July 1998 o February 1999
Text-figure 6.—Mg/Ca and Sr/Ca ratios of shells of Loxoconcha matagordensis from modern vegetation samples plotted versus salinity and
temperature at time of collection.
270
Text-figure 7—Mg/Ca, Sr/Ca, and Na/Ca ratios of fossil ostracode shells versus depth from sediment cores collected from Russell Bank, Park
Key, Bob Allen Keys, Little Madeira Bay, and Manatee Bay. Loxoconcha matagordensis (filled circles); Pereatocytheridea setipunctata (open circles).
Russell Bank Core 19B
Mg/Ca (avg)
15 20 25 30 35 40 45 50
Depth (cm)
Park Key Core 23A
Mg/Ca (avg)
15 20 25 30 35 40 45 50
0
10
20
30
40
50
Depth (cm)
60
70
80
90
Bob Allen Core 6C
Mg/Ca (avg)
15 20 25 30 35 40 45 50
Depth (cm)
160
BULLETIN 361
Sr/Ca (avg)
3.0 3.5
0
10
20
4.0
Sr/Ca (avg)
3.5 4.0
Sr/Ca (avg)
3.5
4.0
140
160
4.5
4.5
Na/Ca (avg)
5.0
80
Depth (cm)
100
110
120
130
140
150
Na/Ca (avg)
Depth (cm)
Na/Ca (avg)
12
4.5 15 18 20
pth (cm)
De
oh
(2)
oO
160
OSTRACODE SHELL CHEMISTRY: DWYER AND CRONIN 271
Manatee Bay Core MB1
Mg/Ca (avg) Sr/Ca (avg) Na/Ca (avg)
15 20 25 30 35 40 45 25S Ol 3-0) 4:0) 425) 50 5 10 15 20
Depth (cm)
Depth (cm)
Depth (cm)
Little Madeira Bay Core T24
Mg/Ca (avg) Sr/Ca (avg) Na/Ca (avg)
15 20 25 30 35 40 45 3.0 35 40 45 5.0 5 10 15 20
10
20
30
40
Depth (cm)
Depth (cm)
Depth (cm)
50
60
60
70 70
O--------0,
ce)
i
!
1
1
70 '
I
I
fe)
80 80
Text-figure 7.—Continued.
272 BULLETIN 361
ligible which is further supported by the overlapping
values obtained for down-core shells (below).
DETERMINATION OF PARTITION COEFFICIENTS
From the water and shell chemistry results, we de-
termined the partition coefficients for Mg ands Sr for
Loxoconcha matagordensis. For these determinations
we used the grand average of Mg/Ca and Sr/Ca ratios
for all of the Florida Bay water samples (Table |) and
the grand average of Mg/Ca and Sr/Ca ratios measured
on 171 modern shells (Table 2). The determinations
also assume an average water temperature of 26.5°C
(average calculated from over 100 February and July
water temperature measurements from 1997 and 1998;
Brewster-Wingard er al., this volume). Under these
conditions the following K, are derived:
Kp.me = (36.98 mmol/mol Mg/Ca)
= (4980 mmol/mol Mg/Ca)
= 0.00743
Ks, = (3.81 mmol/mol Sr/Ca)
= (8.46 mmol/mol Sr/Ca)
0.4499
Il
The Kj, 1. of 0.00743 is similar to the value of 0.00756
calculated from the Mg data reported for shells of liv-
ing specimens of open-marine Loxoconcha by Cadot
and Kaesler (1977). Kx, is not determined because
Na is apparently a non-lattice bound foreign ion within
calcite (Busenberg and Plummer 1985). The calculated
K, values are used below, in conjunction with the wa-
ter me/Ca-salinity relationship in Florida Bay, to cal-
culate estimates of past salinity of Florida Bay.
METAL/CA TRENDS IN FossiL SHELLS FROM
SEDIMENT CORES
Down-core shell me/Ca ratios are listed in Table 3
and summarized in Text-figure 7, which include data
from two genera. Although changes in average salinity
and temperature are considered the primary factors in
controlling ostracode shell chemistry, some of the high
frequency variability in Mg/Ca ratios may result from
a combination of other factors. These include: (a) se-
cretion of shells during different seasons, reflecting
strong seasonal fluctuations in temperature and salinity
of Florida Bay, (b) bioturbational mixing of penecon-
temporaneous shells, and/or, (c) short-term (hours to
days) variability in salinity and temperature at a site,
which may not be representative of longer term con-
ditions. In order to evaluate inter-annual- to decadal-
scale variability, we minimized the high-frequency sig-
nal by calculating intrasample means for each me/Ca
ratio and these are shown in Text-figure 7 versus core
depth.
At Russell Bank, Park Key, and in Manatee Bay,
Mg/Ca ratios for Loxoconcha fluctuate downcore be-
tween about 30 and 45 mmol/mol. Overall, the values
for down-core specimens of Loxoconcha overlap with
values obtained for modern Loxoconcha shells.
At Bob Allen Key and Little Madeira Bay, Loxo-
concha was not as abundant downcore so we analyzed
shells of the genus Peratocytheridea. Metal/Ca ratios
for Peratocytheridea, however, appear to show sys-
tematic differences from those of Loxoconcha. In gen-
eral, Peratocytheridea Mg/Ca ratios are offset toward
lower values, averaging between 20 and 25 mmol/mol,
values that are comparable to the lowest values ob-
tained for modern Loxoconcha shells.
Sr/Ca ratios in Peratocytheridea shells generally
show the opposite trend, yielding average values from
4.1 to 4.3 mmol/mol, approximately 15% higher than
average values for Loxoconcha. Like Mg/Ca ratios,
Peratocytheridea Na/Ca ratios are offset toward lower
values relative to Na/Ca for Loxoconcha. The reason(s)
for the differences observed between Loxoconcha and
Peratocytheridea are unclear. The differences could be
related to phylogenetic effects or might possibly indi-
cate that these two genera preferentially secrete their
adult shells at different times of the year, 7.e., Perato-
cytheridea primarily during the winter months.
Oscillations in the ostracode Sr/Ca ratio record from
Russell Bank core generally show good correspondence
with the pattern of variability observed in the Mg/Ca
record (Text-fig. 7). However there are clear differences
in the magnitude of the shifts in Sr/Ca and Mg/Ca and
some of the changes appear to be out of phase. The
reason for the variability between Sr/Ca and Mg/Ca re-
cords is not known. Possibly, it results from a differ-
ential response to changes in water temperature. Unlike
the co-precipitation of Mg into CaCO , which is posi-
tively correlated with temperature, Sr uptake may de-
crease with increasing temperature (reviewed in Morse
and Mackenzie). Another possible factor is the effect of
calcium carbonate production within Florida Bay. Bio-
genic aragonite, the dominant mineral produced in Flor-
ida Bay (Stockman et al., 1967), co-precipitates large
amounts of Sr at a Sr/Ca ratio that is similar to slightly
lower than the Sr/Ca ratio of seawater. Thus, assuming
continued high production of aragonite in Florida Bay,
the Sr/Ca ratio of evaporatively concentrated, hypersa-
line bay waters would likely show only a slight increase
in Sr/Ca, whereas the Mg/Ca ratio of the same waters
would increase more sharply as Ca is removed by ara-
gonite formation and Mg is virtually unaffected because
it is all but excluded from the aragonite lattice. A third
possible factor is the relatively small shift in Sr/Ca ratio
OSTRACODE SHELL CHEMISTRY: DWYER AND CRONIN 273.
Russell Core 19B
Mg/Ca (mmol/mol)
L. matagordensis
2000
1990
1980
1970
1960
1950
1940
1930
1920
Year
1910
1900
1880
1870
1860
1850
--o-- Rainfall
1840
170 160 150 140 130 120 110 170 160
Annual Rainfall (cm)
NOAA Area 5
5-year running mean
Park Core 23A
Mg/Ca (mmol/mol)
L. matagordensis
Annual Rainfall (cm)
NOAA Area 5
5-year running mean
Bob Allen Core 6A
Mg/Ca (mmol/mol)
P. setipunctata
5 10 15 20 25 30
2000
1990
1980
1970
1960
1950
1940
1930
1920
1910
1900
1890
1880
1870
1860
1850
1840
140 130 120 110 170 160 150 140 130 120 110
Annual Rainfall (cm)
NOAA Area 5
5-year running mean
Text-figure 8 —Ostracode Mg/Ca ratios versus age from Russell, Park, and Bob Allen Keys sediment cores plotted along with 5-year running
mean of annual rainfall. Solid bars along left axis mark years of strong El Nino events.
with salinity compared to that of Mg/Ca (Text-fig. 5).
As a result of these potential temperature and water
chemistry effects on Sr/Ca ratios, ostracode Mg/Ca ra-
tios may be a more a sensitive indicator of salinity
changes in Florida Bay. This contention is generally
supported by the higher dynamic range displayed by
the down-core Mg/Ca records and the apparent corre-
lation between shell Mg/Ca ratios and rainfall (below).
COMPARISON OF MG/CA AND HISTORICAL RAINFALL
To investigate whether the ostracode Mg/Ca ratios
might be related to salinity variability, we compared
downcore trends to historical rainfall records. We fo-
cus our discussion on the records from Russell, Bob
Allen, and Park cores as they have the most robust
chronostratigraphy. Text-figure 8 shows the down-core
records of ostracode Mg/Ca ratios versus year. Because
rainfall derived runoff from peninsular Florida is such
an important factor controlling the salinity of Florida
Bay, we also show an instrumental record of south
Florida annual rainfall which is available back to 1895
(NOAA Data Archive). The rainfall data are shown as
a 5-year running mean to approximate the sampling
frequency of the sediment cores.
In the Russell core, except for a single-specimen
spike in the late 1890’s, Mg/Ca ratios show a gradual
decrease with little variation from the 1870’s until
around 1915, although the sampling resolution and age
model are not as good as those for the upper part of
this core. The succeeding interval, from 1915 to 1998
(the top most data point in the Russell record is an
average of 4 living specimens collected from this site
in 1998), is characterized by a slight overall increase
in Mg/Ca that is punctuated by moderate to large-scale
fluctuations. Eight periods of high Mg/Ca ratios are
observed between 1900 and 1998 (Text-figure 8).
274 BULLETIN 361
Russell Bank Core 19B
100:
110:
120:
Annual ,
Rainfall
(cm) 140
oO
i=}
150:
160: ;
comnans Rainfall
170
Ostracode Mg/Ca - Based Salinity vs. Rainfall
Russell Salinity
1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
Text-figure 9.—Ostracode Mg/Ca-based salinity estimates for Russell core 19B and 5-year running mean of annual rainfall. Salinity derivation
discussed in text. SM designates intervals of relative salinity maximum.
A comparison of the Mg/Ca record from the Russell
core to the rainfall record reveals many similarities
(Text-figure 8). For example, the rainfall data indicate
that since the late 1950’s there have been 3 quasi-de-
cadal oscillations in precipitation over southern Flori-
da. While the amplitude correspondence varies, within
the resolution of sediment chronology, the three rain-
fall oscillations are matched one-for-one by oscilla-
tions in ostracode Mg/Ca ratios. Over this time inter-
val, peaks in Mg/Ca occur during dry intervals in the
early 1960’s, mid-to-late 1970's, and the late 1980’s,
all periods of known hypersalinity within Florida Bay.
The decadal-scale correlation between rainfall and
Meg/Ca persists in the earlier part of the records as
well. Based on the rainfall record, this was a period
of low amplitude rainfall oscillations relative to the
last 50 years, a pattern that also appears in the Mg/Ca
record. Thus, Mg/Ca and rainfall show a correspon-
dence over multi-decadal timescales.
In addition to the longer-term correlation, it also ap-
pears that single-year climate events may be recorded
in the Mg/Ca record. For example, four of the extreme
minima over the last century closely correspond with
years when the El Nino southern oscillation index
(SOD was strongly negative (Text-fig. 8). Typically,
south Florida receives unusually high dry-season rain-
fall under such climate conditions (Douglas and En-
glehart, 1981), presumably leading to significantly
lower salinity in Florida Bay.
Meg/Ca records from the Park and Bob Allen cores
also correspond with the rainfall record, though the
similarities are not as strong as in the Russell core.
This may be a result of the lower resolution sampling
within these cores or in the case of Bob Allen, due to
an ecological effect related to Peratocytheridea. An-
other possibility is that these sites have stronger local
effects than Russell Bank. The Bob Allen and Park
core sites appear to be located on the leeward side of
complexes of islands and broad, shallow mudbanks
which often experience wide temperature and salinity
shifts relative to more open waters of Florida Bay. In
contrast, the Russell Bank core site appears to be lo-
cated in a much smaller lee on a narrow mudbank that
extends out into open waters of the largest and deepest
sub-bay within the eastern sector of Florida Bay. Thus,
the Russell Bank site may be better situated to monitor
large-scale changes in Florida Bay.
A PALEOSALINITY CURVE FOR CENTRAL FLORIDA BAY
Using the down-core Loxoconcha Mg/Ca ratios and
the provisional Kp. for Loxoconcha, we calculated
estimates of past Mg/Ca ratios of Florida Bay water
and converted them to estimates of past salinity based
on the correlation between salinity and Mg/Ca,,,,.,
(Text-fig. 5). Text-figure 9 shows ostracode Mg/Ca-
based salinity estimates derived for Russell Bank and
the south Florida rainfall record. Two extreme values
at intervals equivalent to 1963 (91 ppt) and 1897 (113
ppt) are excluded from this plot. Calculated salinity
estimates range from 13 to over 50 ppt. The overall
average is 32 ppt with around two thirds of the values
falling between 20 and 43 ppt.
The salinity estimates are quite reasonable, strongly
overlapping with the range of instrumental salinity
measurements taken at or near this site over the last
50 years which range from 10 to 55 ppt (Swart et al.,
1996, Halley et al., 1995). One notable example is the
salinity during the late 1980’s which has been widely
discussed in the context of Florida Bay seagrass
dieotfs. Another is the high salinity recorded during
the mid 1970s (Swart et al., 1996). It is important to
note that calculated salinity estimates that are greater
OSTRACODE SHELL CHEMISTRY: DWYER AND CRONIN AY)
than ~42 ppt lie outside the range of our modern cal-
ibration data set and thus are unconstrained. These es-
timates may be high because the slope of best-fit curve
to the Mg/Ca,,.,.. versus salinity data increases contin-
uously above 42 ppt, even though it appears that water
Meg/Ca ratios begin to plateau near the seawater value.
Thus, we might expect that ostracode shell Mg/Ca ra-
tios would also plateau at a value around 38 mmol/
mol, a value that is actually exceeded in a number of
intervals in the Russell core, perhaps suggesting that
Meg/Ca ratios of Florida Bay water increases further
with increasing salinity.
An alternative interpretation of the trends in shell
Meg/Ca is that the signal may be all or partially related
to temperature. Temperature and salinity measure-
ments collected seasonally for the last few years
(Brewster-Wingard er al., 2001) near the Russell core
site suggest that temperature and salinity covary at this
site, more so than at Bob Allen and Park Key. Given
the thermodependence on Mg uptake in Loxoconcha
and other ostracode genera (Cadot and Kaesler, 1977;
Dwyer et al., 1995), the increased uptake of Mg re-
sulting from higher water Mg/Ca ratios at higher sa-
linity is likely further enhanced by an accompanying
increase in water temperature. This may also help to
explain the better correspondence between Mg/Ca and
rainfall in the Russell core.
Until additional experiments in which adult ostracode
shells are grown under controlled salinity and temper-
ature conditions, the paleosalinity curve in Text-figure
9 must remain preliminary in nature. Still, the scale of
variability in Mg/Ca ratios within all five cores and the
estimated salinities are consistent with the hypothesis
that salinity fluctuations occur over decadal timescales
in Florida Bay. Moreover, the correspondence between
interdecadal trends in rainfall and paleosalinity ex-
tremes indicates a strong climatic control on Florida
Bay salinity. The shift to high amplitude oscillations in
salinity after about 1950 seems to reflect a correspond-
ing increase in rainfall extremes, but it may also reflect
an alteration of the salinity regime due to human alter-
ation of freshwater flow into Florida Bay.
REFERENCES
Berner, E.K., and Berner, R.A.
1996. Global environment. Prentice Hall, New Jersey, 376 pp.
Berner, R.A.
1966. Chemical diagenesis of some modern carbonate sedi-
ments. American Journal of Science, vol. 264, pp. 1—36.
Bodergat, A.
1985. Spark-source mass spectrometry of carapaces of Cypri-
deis torosa (Crustacea, Ostracoda); a model to reconstruct
multifarious saline environments. in D.E. Caldwell, J.A.
Brierley, C.L. Brierley, eds., Planetary ecology, Van Nos-
trand Reinhold, New York, pp. 178-185.
Boesch, D.F., Armstrong, N.E., DiElia, C.F., Maynard, N.G.,
Paerl, H.W., and Williams, S.L.
1993. Deterioration of the Florida Bay ecosystem: an evaluation
of the scientific evidence. Report of Working Group on
Florida Bay, National Fish and Wildlife Foundation,
South Florida Water Management District, 27 pp.
Brewster-Wingard, G.L., Ishman, S.E., Willard, D.A., Edwards,
L.E., and Holmes, C.W.
1997. Preliminary paleontological report on cores 19A and 19B
from Russell Bank, Everglades National Park, Florida Bay.
U.S. Geological Survey Open-file Report 97-460, 29 pp.
Brewster-Wingard, G. L., Ishman, S. E., and Holmes, C. W.
1998. Environmental impacts on the southern Florida coastal
waters: A history of change in Florida Bay. Journal of
Coastal Research, vol. 26, pp. 162-172.
Brewster-Wingard, G. L., and Ishman, S. E.
1999. Historical trends in salinity and substrate in central Flor-
ida Bay: A paleoecological reconstruction using modern
analogue data. Estuaries, vol. 22, pp. 369-383.
Brewster-Wingard, G.L., Stone, J.R., and Holmes, C.W.
2001. Molluscan faunal distribution in Florida Bay, past and pre-
sent: and integration of down-core and modern data. Bul-
letins of American Paleontology, no. 361, pp. 199-231.
Burns, S.J., and Swart, P.K.
1992. Diagenetic processes in Holocene carbonate sediments:
Florida Bay mudbanks and islands. Sedimentology, vol.
39, pp. 285-304.
Burton, E.A., and Walter, L.M.
1991. The effects of PCO, and temperature on magnesium in-
corporation in calcite in seawater and MgCl,-CaCl, so-
lutions. Geochimica et Cosmochimica Acta, vol. 55, pp.
777-785.
Busenberg, E., and Plummer, L.N.
1985. Kinetic and thermodynamic factors controlling the distri-
bution of SO,?~ and Na® in calcites and selected arago-
nites. Geochimica et Cosmochimica Acta, vol. 49, pp.
713-725.
Cadot, H.M., and Kaesler, R.L.
1977. Magnesium content of calcite in carapaces of benthic ma-
tine Ostracoda. University of Kansas Paleontological
Contributions, Paper 87, pp. 1—23.
Chave, K.E.
1954. Aspects of the biogeochemistry of magnesium 1. calcareous
marine organisms. Journal of Geology, vol. 62, pp. 266-283.
Chester, R.A.
1990. Marine Geochemistry. Unwyn Hyman, London, 698 pp.
Chivas, A.R., De Deckker, P., and Shelley, J.M.G.
1983. Magnesium, strontium, and barium partitioning in non-
marine ostracode shells and their use in paleoenviron-
mental reconstructions—a preliminary study. in R.E Mad-
docks, ed., Applications of Ostracoda, University of
Houston Geoscience, pp. 238-249.
Chivas, A.R., De Deckker, P., and Shelley, J.M.G.
1986. Magnesium content of non-marine ostracod shells: a new
paleosalinometer and paleothermometer. Palaeogeogra-
phy, Palaeoclimatology, Palaeoecology, vol. 54, pp. 43-
61.
Cronin, T.M., Holmes, C.,Wingard, L., Ishman, S. Dowsett, D.,
and Waibel, N.
276 BULLETIN 361
2001. Historical trends in epiphytal ostracodes from Florida
Bay: implications for seagrass and macro-benthic algal
variability. Bulletins of American Paleontology, no. 361,
pp. 159-197.
De Deckker, P., Chivas, A.R., and Shelley, J.M.G.
1999. Uptake of Mg and Sr in the euryhaline ostracod Cyprideis
determined from in vitro experiments. Palaeogeography,
Palaeoclimatology, Palaeoecology, vol. 148, pp. 105-116.
Douglas, A.V., and Englehart, P.J.
1981. On a statistical relationship between autumn rainfall in
the central equatorial Pacific and subsequent winter pre-
cipitation in Florida. Monthly Weather Review, American
Meteorological Society, November, pp. 2377-2382.
Dwyer, G.S., Cronin, T.M., Baker, P.A., Raymo, M.E., Buzas,
J.S., and Correge, T.
1995. North Atlantic Deep Water Temperature Change During
Late Pliocene and Quaternary Climatic Cycles. Science,
vol. 270, pp. 1347-1351.
Enos, P., and Perkins, R.D.
1979. Evolution of Florida Bay from island stratigraphy. Geo-
logical Society of America Bulletin, vol. 90, pp. 59-83.
Fennema, R.J., Neidrauer, C.J., Johnson, R.A., MacVicar, and
Perkins, W.A.
1994. A computer model to simulate natural Everglades hy-
drology. in Everglades, The Ecosystem and its Restora-
tion. S.M. Davis and J.C. Ogden, eds., St Lucie Press,
Delray Beach, Florida, pp. 249-290.
Ginsburg, R.N., editor.
1972. Sedimenta H, South Florida Carbonate Sediments. publi-
cation of the Rosenstiel School of Marine and Atmospher-
ic Science, University of Miami, Miami Florida, 72 pp.
Halley, R., Smith, D., and Hansen, M.
1995. Florida Bay Surface Salinities. U.S. Geological Survey
Open-file Report 95—634.
Holmes, J.-A.
1996. Trace-element and stable-isotope geochemistry of non-
marine ostracod shells in Quaternary palaeoenvironmental
reconstruction. Journal of Paleolimnology, vol. 15, pp.
223-235.
Kamiya, T.
1988. Morphological and ethological adaptations of Ostracoda
to microhabitats in Zostera beds. in Developments in Pa-
laeontology and Stratigraphy 11, Evolutionary Biology of
Ostracoda. T. Hanai, N. Ikeya, K. Ishizaki, eds., Elsevier,
pp. 303-318.
Keyser, D.
1975. Ostracoden aus den Mangrovegebieten von Sudwest-Flor-
ida. Abhandlungen und Verhandlungen des Naturwissen-
schaftlichen Vereins in Hamburg, vol. 18/19, pp. 255—
290.
King, C.E., and Kornicker, L.S.
1970. Ostracoda in Texas bays and lagoons: an ecologic study.
Smithsonian Contributions to Zoology, vol. 24, pp. 1-87.
Melvor, C.C., Ley, J.A., and Bjork, R.D.
1994. Changes in freshwater inflow from the Everglades to Flor-
ida Bay including effects on biota and biotic processes,
in Everglades, The Ecosystem and its Restoration. S.M.
Davis and J.C. Ogden, eds., St Lucie Press, Delray Beach,
Florida, pp. 117-148.
Miller, J.A.
1997. Hydrogeology of Florida. in The Geology of Florida. A.E
Randazzo and D.S. Jones, eds., University Press of Flor-
ida, Miami, pp. 69-88.
Morse, J.W., and Mackenzie, F.T.
1990. Geochemistry of Sedimentary Carbonates. Elsevier, New
York, 707 pp.
NOAA Data Archive (http://www.ncdc.noaa.gov/onlineprod/
drought/xmgr.html).
RAIL.
1973. Resource and Land Information for South Dade County,
Florida. U.S. Geological Survey Investigation I-850, 66
PPp-
Robbins, J.A., Holmes, C.W., Halley, R.B., Bothner, M., Shinn,
E., Graney, J., Keeler, G., ten Brink, M., Orlandini,
K.A., and Rudnick, D.
2001. Time constraints characterizing predeposition integration
of 137Ce and Pb fluxes to sediments in Florida Bay. Geo-
chimica et Cosmochimica Acta. (in press).
Sonntag. W.H.
1987. Chemical characteristics of water in the surficial aquifer
system, Dade County, Florida. U.S. Geological Survey
Water Resources Report no. 87-4080, 42 pp.
Stockman, K.W., Ginsburg, R.N., and Shinn, E.A.
1967. The production of lime mud by algae in south Florida.
Journal of Sedimentary Petrology, vol. 37, pp. 633-648.
Swart, P.K., Healy, G.F., Dodge, R.E., Kramer, P., Hudson, J.H.,
Halley, R.B., and Robblee, M.B.
1996. The stable oxygen and carbon isotopic record from a coral
growing in Florida Bay: a 160 year record of climatic and
anthropogenic influence. Palaeogeography, Palaeoclima-
tology, Palaeoecology, vol. 123, pp. 219-237.
Tabb, D.C.
1967. Predictions of estuarine salinities in Everglades National
Park, Florida by the use of ground water records. Unpub-
lished Ph.D. thesis, University of Miami, Coral Cables
Florida, 107 pp.
Teeter, J.W., and Quick, T.J.
1990. Magnesium-salinity relation in the saline lake ostracode
Cyprideis americana. Geology, vol. 18, pp. 220-222.
Wansard, G.
1996. Quantification of paleotemperature changes during isoto-
pic stage 2 in the La Draga continental sequence (NE
Spain) based on the Mg/Ca ratio of freshwater ostracods.
Quaternary Science Reviews, vol. 15, pp. 237-245.
White, D.
1983. Oceanographic monitoring study 1980-1983. South Flor-
ida Research Center, Everglades National Park, Home-
stead, Florida, 144 pp.
Wingard, G.L., Ishman, S.E., Cronin, T.M., Edwards,. L.E., Wil-
lard, D.A., and Halley, R.B.
1995. Preliminary analysis of down-core biotic assemblages:
Bob Allen keys, Everglades National Park, Florida Bay.
U.S. Geological Survey Open-file Report 95-628.
Xia, J., Engstrom, D.R., and Ito, E.
1997. Geochemistry of ostracode calcite: Part 2. The effects of
water chemistry and seasonal temperature variations on
Candona rawsoni. Geochimica et Cosmochimica Acta,
vol. 61, pp. 383-391.
INDEX
ADUINGANGES Mere ee create ae eee etciatoteie tae cloz ctesciete tassios el faiceiaeleterers 4,5
JV GCG 17 RRR RATA Sen oon SO mS E ORS ECO EET ROR CE TOOOCE CEE CIOE 88
ENGI OTOL AS fea a® SPSS USS TESS TCR Soe FI CRORE HERP EE ee ree 79, 80, 81
LNCTITHGE. Sac eaoa renee ou oanocne nde to eCc ree eoE mete nee caceeToneanreas 83
RARITSTITAD odeasnosacossuonoouoosDUceesadaencobasccononscoodocer 5,6
PELOPHUTCE Ia LOUIS Herter clerseisrleletelsiens sie etslsistlelsiesa settiretars ciaciersiele eerie aes 81
agnicultura Weng CHIMENG vserroctseinetosvsimaecincisicoee seceecer sie cieceeer ae 70
Haarewwiliueea irqbhoy nh eae eRe oanander oaaaAEnSeeenacessacdeenos-ScoskeeooodnG 70
Ei\eeil [DILoYeysi) sao onmenoonpedocuoasedecMarcecenaoare cecontnaaes IZA styl boe)
JNVDIS) oo cngenbp eg oc aE ROROO Ene eeee eran 88, 107, 116, 118, 128, 131
PAIL ONECL RST INS S 5 Fotcpeh at tetera cs als Siatotererer a eValeels SisTetotaleledurerelarsielainae 88, 89, 90
AILOTLELL MPR Cet a cocina tite isc eistes io aad nine NTE A Hea 89, 90
PAllLermanthera-(G OMpPHTEn@ wane eni tence ccc de eiddssiseeiae re 88
PALLY ERAT NAVIN poy etcratatchctatotejnjorchsrs cove svolata lola otajtavateratele aterstatane siatetctsretsie bela 72,74, 82
VATIELAT CULLTES yoo sooo alo Siac aie cteroh la oleic wis ate ete ToTwlatelaTe fore Sorsto oem tls Geet 83
PEATIEEIIONNALCTELELES 1h takater, (oistars yetatanieleretdisis|afs oa le ss loots ola dladlieiaisla stave veeinena) LS
INDOLE an Mae ean deeean se aecrmimecee gene laren 3
by dropernlodiee-ceaseeeneaa te 103, 105, 106, 107, 121, 130, 131
Ilex\ (dahoon holly) erect ee eee 106, 107, 123, 124, 127, 130
TLYOCHY Plus’ scoceeicetice teak Soe ae es RSE ORE SE ee 89, 90
Inductively Coupled Plasma-Optical Emission Spectrometry (ICP-
(0) SS) eAnpaotescospocesenatocoassnpaececnaonacceos 61, 70, 71, 72,74
WON theta pide tein ode ection aieceict eeiseteeer sete carer 123133 TAS Oe,
Tsoetes" «Tecan weeempeee she rece teehee eee 59, 85, 86, 94
Uoi-el Nise: qusencasoosoLocopencsseacda
JUNIPETUS -x;s\ccidadaten sie saiocete ee se
TUSULCI cates racers injeieleielovefelafeinte cloteie oietere are etereeteieta wea ata ae eerste eater
Key Eargojllimestone sc saemcnsa- keer ceteeeenseeieeeiscisetteeetet 65
Waguna dewPetenxall (er. tke eases eee ee eee eee ereeeeee 82, 90, 92
Laguncularia (white mangrove) ... 83, 87, 88, 106, 118, 126, 130
INDEX 279
Wakes OKEeEchObDee sep -rienincaee rial cielleeeeelecicanerieeeer 59, 65, 104, 105
WAN SMU ANE sf yotafa cicresorwiarsiaersonmttosiomenvetrcioaer lesbian cecasioa senses secant 92
ATEN EIOLOCENES ccm sias-pe aaiteseteete econ ese 103, 106, 122, 132, 104
[La vnrgstelatalet ePns) nYerelt So opeanmcnaedsorenccsosunadesbnnacaeaoabat cea 915,93
MEAG ee ra caraycrctste se ste Sissieieeieletsraele eis I Cie eT=rsfere eloteiclororeceereyse 10N (2S
IPE SUTAIMOSAS, c.-jcisseisiace ists siote seis tatdja stv afolsia eo eiateistaisiclaiotaiaicicteissstatetateictorsiarevs sreieis's 88
IG Cea aeescaaniaartcnte Vaan cae as conte nrinaccrn acco rnracaacccsore acne 67 65
LOG LG TR GRE SoCo SRECE ERE Ernene Somer CoE eae Teneo oOa oa cTonCedSr 89, 90
ILTUIOXES LD a pegarceacnncueacncnaatcbeoccercnonterenaatercaantoopcocae 6
HEL TPLOTALLAIII to Tatas mse eee one. NS SIS 8 2% ala oaiale s BAR OM aRia see iaioes MA a 87
WEL GULA AMBER, oie ociejersyeic\eis sie stats G5 viv sis felsic seiciofna stone rate celelepeleie eles Sales ts 129
AS har ene ore yo ote aie ie hie ici israotes esis sew oe See one mele aloe eee V5 13; 15
IL tHe SEIU) oy obeVes Se panengoeseaanoapcaaccasdosaeosodcoonoe sagaccecuupas 92
BUTTERS HAL KRAVE Ts a rclstay erences elejntstelsislere(aretctoroselareie sistotclaictaroatate steteeloe eens 106
live oak (Quercus) .. 106, 116, 118, 119, 120, 123, 124, 128, 131
HEOKOGONGHAY, tajoccstaistctare sions statalstalattigniee letaeictea alo eratnstetGeiate wlatae om ele wiser ls 5,6
EY COP OGULINAT were fiee ceases eee een eee eeiiceete 107, 123
MEY STLOMICN WAL GtAIM ATINNG) le erate settee sessile ictesoniserversers 106, 107
IM terol ava ley (abl lee ope ace pa deqoceacunotnanecen ann ommencoanocanorccacceoen 90
IN GBR A1bS QoustacsencccsoCereooSseAsBOESaHadCHeOSaeee ee eReCOee ea sanoan 90
MECMEN OTD cutqhoosdasemadader nae aAcasn encode noOmaneroseaAaeT suse 72, 74
TALON? (SUNG OIE) | 55a s0cs0bcndssasanaboosoBdoascopeassbasooaqnon 106
imrnntelamenine (CRaATATO)) cagnoendaccoontendoaaaageosodaoebeaganeacnace 106
1 GG Miewpsouetene ocogocad i aeeCUBaaee cee cnoncesacreracmoroemarcad 5-7
SDRINERINES -ssancnapascbcesadobnseacobadbansagsanoess 70, 72, 73, 74, 75
aan PLOV Caste eericeeen ence ate rsaaniae: 72, 83, 84, 85, 86, 87, 88, 94
(ATO KONS [DEEL Gonogansasaoueobonousodosuenacasosboas 105, 122, 123, 130
mangrove swamp ... 103, 106, 116, 117, 119, 120, 121, 122, 124,
12S SIO MIS2. 7,
marl ... 59, 62, 63, 64, 65, 70, 76, 82, 84, 85, 86, 88, 91, 93, 94,
105, 122
PUTAS LIME PA ee cy atet ojc clots wie tine sic oe os cmticinemratclolen 59, 83, 84, 85, 86, 87
WIVAGSTOSIOIAN = sooo eescisiecisieisiess oisiciste 79, 80, 81, 83, 152-154, 157, 159
TEMBRAVINY soooqndosacaspoceagaococconsqce aa scoosaoDeOoAOSUGTOUSDnGGRC 69, 70
Wiramil Canal 332i vic 2/-/-teac aeietscicectesesebicicis acc e ete 2 atceietelearocetejenemeecicte 87
WTADAIPIETINES CONE! lais.tiee er aesrisesiers wate sewrase eitee cme mam sine 65, 105, 104
VIE CLOG) STIS epic t orca Cie neeree hicle se Siete eis ute sleictasieanclomteee Sues eeeeeneee 82
Mitlankovatchicy, Cledanseree rel cei ela mnatciere shir sere clasiaeeeen eee 93
LG OT GIT tape pocarrenosansanccesadacuseeeads 244, 245, 247, 249, 250
MADAME a ANALY SIS oo cstejas2 ea esate statone tate eteateTeneratotelclats wlat sya elerstolelaraamrsragynateiee 70-75
fauve} Moyel jal tee) Goaneanaenaaasenenaanpaaso seer scorn ecanoeonarcanodeeinc (3375
WA ONOCOLP GLE Sire teretotae cere 2 Ss Sass SS eI Sees sooo ero aes eas 131
VIOUS Sectrcec a Naon ites sis oie eeiae aes teint ete are 107, 116, 118, 127, 130
MUGS LONE ys olor os ciciereie seine ies saosin wieretaie ainisiete ala cloreraterehe sare 123, 130, 131
Muhlenbergia(Muhly/erass) ie s.tsca- cree eeseasseeeseeecee 76, 106
WAV IGA WAXY TUE) Waser tite ce seeecmareiaesinee ee tts 83, 88, 106
AVM ALE) » corigongbee cdSenOn nea COUSCa Aaa cMOnSER Tete ame ce bamretionaen conte 118
Myrtaceae we renner eicistcacscteclnerisece seco 83, 107, 116, 129
INE COG Gan ico Ua aas a Cocnat CoRR DE ene an: Aarrnar eee nas wane Heterinctas 82
IN AG Die esssecetcratctscsiotetavstevc mafia slo eletevsianatafarensete ciattictote(sicvaseteo)avele Rtcls ninleelars Stn Relste ae 82
IN AVEC chore osayctsitahara'aiors -Telosnrsectsaiaseeerereteleetee idlstereleroetes 77, 78, 79, 80, 81
INGA LT ue Renan s cao DD REECE OOBREEE Cac Cote aoe Cae ESOC USGOSOeTEE ace core 77, 78
PN COCOAUAUES Myer Te Se OTN ASP a a = biota ae ole elo cia Be olan shee OOE 5}, (9)
MONS a aie oy otay aro stetaleainratorelstescioteseiaereciae Gio clelots Bele wns veisramtewieterneis W335 IE)
MRT ENO SEMA apes oll atte ulate lat ctayatstst aj ctaleieersiale sa stcelafsse(olercieinlacicialeeereite 70, 82
UNEES CRG ee te rate Seite aercictoeieineoeeseisias one EER 78, 79, 80, 81, 151-159
Wodobaculartellancarcncsecmncccoen eer ceten eee een eneen aeenlee 245
INOdosarlidae s2iiscesi setae acters crate ste er olocis sioiste Date ste diclateclotwe wisi taheaeer 245
INonionellar cession ts see ee seas eee co ee ower eens 245, 250
INOMLISEEUL ee foe cee ie ee ca cei seceve nt once cae seae ecole 242
North Biscayne Bay 250
Nubecularia 245
INU DIO raetraacn nae tenes cok che mot eece eee 129
INymphacede) ae as-cidseteeteseinv me cencessemerace 77, 83, 84, 85, 86, 100
Nymphaea (water lily) .... 61, 83, 85, 86, 87, 106, 116, 117, 118,
119, 120, 122, 124, 126, 130
INDSSQ) Race ceposeasene eecprnsnke@acosteneadscene tees 107, 116, 118, 128
AKG. aise crsiotcmphisicce merece CeLenec Celie Misc oe pase CEReE Cocca eee eee Re 83
Oligohalines 3. cssicieistatmrorcmiaseiienciesiiesiis sins cateiorerese tees emacieeeeeee 679)
OEE PI areas sada e fois aro stass resets tuce(epete lee wo so as e)ahaisyaia wlels oevsiosstersterorereerers 87
Osmunda ........ 83, 107, 116, 117, 118, 121, 123, 124, 129, 131
OSTACOGES SH ato meinieidisisiet ico eaiele seizes ee Meee cela ccs cee cee eee 4
Ovoidites ....... 107, 116, 117, 118, 119, 120, 123, 124, 128, 131
ORY UTOLIG sae sas Rass ser cise ee ies ones en ate aat sare 88, 89, 90
LOMAS MIEN ACO) (oe Saactencrmascoperasecnd imap aadecodtenenos bocacactocaaor 2,
paleoecology =-...c-nee--
Palmae. a. ste: seemaeeese ee
Panicum (maidencane)
PAF QUIOy hi tanec eee
EALCULING. merin.cisa se ocaee ceclaec tt eset cet ee
by dating wroncn-nomsc cise actce oes eeneeee meee needa eee eee
peat 2. 5951625163; 64516551695, 70; 76; 717,983, 85,1075 Sono los
3
94, 95
TEC HOT ATT heoeetoner teitacan ee erieccr noua iaccsccce dascoAnadsocus TS Ths}
ReelDee Belemmite aa ccaeett cn rascme cee et ee tence eee ech e eee eee ere 4
IRENCTOPOLISS sercmcctes soe - acetic oe eee see LD RE ILA DO 2D.
PET atOGy ther ided. 2o icteric cia aetasisee sila ease eer eee see 4, 5, 7-10
Persea) (Ted Day)