Paleo-Salinity Changes
in the
Lower Everglades and Florida Bay Ecosystem
A Project of NOAA's
Coastal Ocean Program
Atlantic Oceanographic and Meteorological Laboratories
Terry A. Nelsen
Ginger Garte
Charles Featherstone
Rosential School of Marine and Atmospheric Science
Patricia Blackwelder
Terri Hood
Carlos Alvarez-Zarikian
Peter Swart
University of Miami
Hal Wanless
Florida Institute of Technology
John Trefry
Simone Metz
Woo Jun Kang
Indiana University Purdue University Indianapolis
Lenore Tedesco
Martha A. Capps
Michael A. O'Neal
This
report summarizes the paleo-salinity portion of a chapter entitled “Linkages Between
the South Florida Peninsula and Coastal Zone: A Sediment-based History of Natural and
Anthropogenic Influences” in “Linkages Between Ecosystems in the South
Florida Hydroscape: The River of Grass Continues” (Eds. K. Porter & J. Porter, in
press).
-
-
-
-
-
-
-
-
-
-
-
The south Florida ecosystem is characterized by a low-relief transition between peninsular Florida and the adjacent shallow coastal zone. Natural features such as the Shark River Slough form a major coastward conduit for freshwater and terrestrial materials while major storms that impact the area, such as hurricanes, can provide energy for landward transport of marine sediments. Coastal sediments serve as reservoirs of historical information critical to understanding the natural and anthropogenic impacts on this ecosystem. To decipher this history we investigated the physical, biological and chemical nature of these sediments, at locations indicative of the continental-coastal transition as well as central Florida Bay. Interpretation of results was facilitated by comparison with existing regional salinity, rainfall and freshwater flow data. Placed within the context of geochronology, sediment sequences were interpretable to about the turn of the century, thus making them ideal for contrasting time of minimal anthropogenic impact up to present conditions.
From historical data, salinity values
were shown to vary on decadal to monthly time scales and correlated with changing patterns
in regional rainfall. Moreover, regional
rainfall, represented by the 80+ year record at Homestead, Florida, indicated high
correlation with flow into Shark River Slough prior to major watershed construction which
began in the early-1960s. During subsequent
periods of water management strategies, enacted from the mid-1960s to present, results
indicate essentially no correlation between regional rainfall and flow during the Monthly
Allocation Plan. In contrast, correlations
most closely paralleled pre-construction, apparently more natural conditions, during the
subsequent Rainfall Plan.
Temporally coherent with the early-1950s creation of the Everglades Agricultural Area, recovered sediments adjacent to the mouth of Shark River underwent a dual transition with a statistically significant upcore reduction in organic carbon content to present, and a concurrent significant increase in microfaunal abundance. This was not observed in contemporaneous sediments from Florida Bay. Investigated parameters for the benthic microfaunal communities (foraminifera and ostracods) population characteristics such as stable isotopic compositions, abundance and community diversity exhibited changes and trends that more closely paralleled natural, rather than anthropogenic influences. Changes in the stable isotopic values of the microfauna indicated, within the limits of our geochronology, direct responses to regional rainfall. Such responses more closely paralleled rainfall as opposed to freshwater runoff, even adjacent to the outflow of the Shark River Slough. Furthermore, long-term trends for both stable isotopes and relative abundance of salinity-sensitive species indicated a statistically valid upcore trend toward less-fresh, more marine conditions at both study sites. This trend was contemporaneous with a weak decline in regional rainfall over the same time span. Moreover, crashes in microfaunal abundances at Oyster Bay and more gradual declines at Jimmy Key correlated not with changes in mean salinity, but rather with changes in the standard deviation of salinity (DSs). An order of magnitude reduction in abundance corresponded with increased DSs @3 in Oyster Bay. This abundance drop was concurrent with an equally dramatic decline in community diversity characterized by survivor-type dominance by two microfaunal species and occurred over a period of drought at both sites as well as a period of reduced flow from Shark River Slough.
Natural features such as the Shark River Slough (SRS) form a major coastward conduit for freshwater and terrestrial materials that influence and contribute to marine sediments. These coastal sediments serve as reservoirs of historical information critical to understanding the paleo-salinity record of this ecosystem.
Within the south Florida ecosystem one of our objectives was to understand the changes in salinity that have influenced both the areas adjacent to SRS and in central Florida Bay along with understanding the forcing mechanisms of such change. In the absence of long-termed monitoring studies and archived documentation of such change, we have examined the sedimentary record to decipher the area’s depositional history. The laminae that constitute these sequences can be compared to the pages of a history book upon which nature and man potentially can record their influences. To deconvolute this record we must understand the story told by biological, chemical and physical evidence contained within these sediments. To accomplish this we chose regions representative of the south Florida coastal zone: 1) Oyster Bay, adjacent to Shark River, the seaward outflow of the dominant regional freshwater source, and 2) Jimmy Key, located in central Florida Bay, distal from direct freshwater runoff from terrestrial sources. Oyster Bay and Jimmy Key (Figure 1) are sites of rapid recent sediment accumulation and preserved finely laminated stratification. Radiometric age dating indicated that the sedimentary sequences at each site recorded approximately the past 100 years making them ideal sites for this study.
· Cores were collected and opened to verify that the site was an area of active sediment accumulation, had well-preserved stratification (minimal bioturbation activity) and maintained uniform macro-benthic environment through the sequence (e.g., no sea grass colonization).
·
After site evaluation, additional cores were recovered at the
sites selected for focussed study. X-radiographs for each core were made and evaluated for sediment
disruption and lamination.
· Samples were collected and allocated for age dating, isotopes, chemistry, and biostratigraphy from the same stratigraphic interval to ensure subsequent comparability.
·
Sediment accumulation rates and age determinations were
measured by 210Pb and 137Cs techniques.
· Biogenic community structure shifts were evaluated using statistical methods to characterize the ostracod and benthic foraminifera communities. These included relative and absolute abundance, potential effects of rarefaction and the Shannon-Weiner diversity index (SWDI).
· For a better understanding of sediment-based observations, historical salinity, freshwater flow in SRS, and rainfall data were analyzed. Databases were sorted for stations proximal to our sediment coring sites.
Results
Geochronology - In the Oyster Bay core, the sediment accumulation rate from the well-defined profile for excess 210Pb was about 1.1 cm/y (0.6 g/cm2/y). The high-resolution profile in this core shows maximum 137Cs activity levels at about 31 cm and no detectable 137Cs below 50 cm (Figure 2a,b). In central Florida Bay at Jimmy Key, the excess 210Pb profile for the upper 39 cm of the core is well-defined and yields a calculated sediment accumulation rate of about 1.0 cm y-1 (~0.78 g cm-2 y-1), again in good agreement with the 137Cs profile (Figure 2c,d).
Microfauna - Sediments adjacent to Shark River in Oyster Bay and in central Florida Bay at Jimmy Key were analyzed for foraminifera and ostracods with the results shown in Figure 3a,b for Oyster Bay and Figure 3c,d for Jimmy Key. For both sites, total abundances and the Shannon-Weiner Diversity Index (SWDI) are shown. The SWDI expresses community diversity numerically and is sensitive to the presence/absence of trace species, such that larger positive values reflect a more diverse community while lower numbers indicate a community dominated by fewer species.
At both sites, benthic foraminifera and ostracods exhibit dramatic shifts in abundance and diversity. At Oyster Bay abundances range from <25 to >350 foraminifer tests and ostracod valves per gram of sediment while at Jimmy Key they ranged from 0-180 and 0-230 respectively. At Oyster Bay a period of benthic foraminifer and ostracod high abundance extended from the early-1950s to late-1970s (Figure 3a). A similar, though less distinct period also occurred at Jimmy Key (Figure 3c). For convenience, a working definition of these abundance changes are as follows: 1) Zone B = high-abundance period ~1950-1980); 2) Zones A and C are periods after and before Zone B respectively. Two periods of reduced diversity can be seen at Oyster Bay. The first constitutes a period of more than a decade, extending from the mid-1930s to late-1940s (Figure 3b). The second reduction starts in the late-1970s and extends to the time of core recovery in 1995 (Figure 3b). At Jimmy Key, a zone devoid of both microfauna was encountered which was centered at the mid- to late-1930s. As at Oyster Bay, a second more gradual decline in the SWDI starts in the late-1970s and extends upward to core recovery in 1997 (Figure 3b,d). Summary statistics of these data are presented in Table 1. During the low diversity period of the late-1970s to the present (Figure 3b,d), both benthic foraminifer and ostracod communities exhibit dominance by one species (Ammonia parkinsoniana typica and Peratocytheridea setipunctata respectively). The absolute abundance data of Figure 4 indicates a survivor-type behavior rather than opportunism.
Stable Isotopic Compositions - Summary data for ostracod and foraminifer d18O and d13C, from both Oyster Bay and Jimmy Key sites, are listed in Table 2.
On a sample-by-sample basis, d13C and d18O data from Oyster Bay are generally positively correlated for both foraminifera and ostracods suggesting a normal interaction between fresh and marine waters. When viewed on a longer time scale, a linear fit to the entire data set of both d13C and d18O showed a consistent upcore trend to more marine for ostracods and mixed results for foraminifera (Table 2). Ostracod d18O trends from Oyster Bay (Figure 5a) and Jimmy Key (Figure 5b) are compatible with the declining trend (Figure 5c) of the low-salinity tolerant foraminifer species Elphidium gunteri (Parker et al., 1953; Phleger, 1966a,b; Poag, 1978; Murray, J. 1991) thus reinforcing an interpretation of upcore (to present) more marine conditions.
For Jimmy Key the ostracod oxygen isotopic data for Malzella floridana show negative values in the lower portion of the core (pre-1960s) compared to more positive values from there to present (Figure 5b). The same deviation is evident in the benthic foraminifer Ammonia parkinsoniana typica suggesting a period of possible lower salinity in the middle of Florida Bay during the first half of this century and an increase of salinity thereafter.
Because of the generally
positive correlations between the d13C
and d18O
for both foraminifera and ostracods, suggesting a normal interaction between freshwater
and marine waters, the data also allows evaluation of the relative importance of
freshwater sources (rainfall vs flow in SRS). Rainfall
can account for input of freshwater to both the Oyster Bay and Jimmy Key sites. However, due to its location adjacent to the
outflow from the SRS, the influx of continentally derived freshwater runoff can be assumed
to be greater at Oyster Bay than at Jimmy Key (Figure 1). In order to evaluate which may play a dominant
role at Oyster Bay, we compared the d18O
data for Peratocytheridea
setipunctata with historical local rainfall (Flamingo, Figure 1) and SRS flow.
This time period was chosen for the best temporal match between absolute (rainfall,
freshwater flow) and estimated (sediment geochronology) data. Direct comparison of the d18O
record indicates that the inverse co-variance is best with rainfall (Figure 6).
Discussion
Natural Influences – Based on long-term historical salinity data available from Oyster Bay and Jimmy Key, decadal to monthly time series showed high degrees of variability (Figure 7a,b). On the decadal scale, at both sites, this variability was linked to regional rainfall (Figure 8a,b) indicating long-term salinity changes were strongly controlled by long-term, regional-scale precipitation patterns. On the monthly scale, salinity variability was synchronized between sites with short-term local rain events (Figure 7b, red arrow at June, 1997: 54.6 cm rain Þ DS=33.6 to 10.0) accounting for intra-site divergence.
This rainfall linkage implies that salinity patterns should reflect rainfall forcing, with the potential exception of sites adjacent to the outflow from SRS. When viewed on a seasonal basis for low- and high-salinity years (Figure 7c), Oyster Bay can be evaluated in conjunction with contemporary rainfall and related freshwater flow patterns (Figure 8c). The bimodal (July, September highs) nature of south Florida rainfall presented by Duever et al. (1994) correctly represents average conditions. During non-average conditions, such as low-salinity years at Oyster Bay (Figure 7c, open symbols), seasonal rainfall typically increases such that ~50% of these increases occur during June with ~95% of the total yearly increase occurring before August. This not only modifies freshwater outflow timing (Figure 8c), but can also account for difference patterns between high- and low-salinity periods in Oyster Bay (Figure 7c). Specifically, at Oyster Bay, the August and December-March salinity-difference maxima suggest a ³one-month phase lag with summer and fall outflow modes (Figure 8c, July-August and October). These periods are separated by a flow reduction (September)/salinity increase (October). Although the data provide a less compelling argument at Jimmy Key, minimum salinity patterns during September-January (Figure 7d, open symbols) suggest delayed linkage to terrestrial freshwater outflow patterns which may influence central Florida Bay as well.
As shown in Figure 6, rainfall, rather than freshwater runoff, is the dominant driver of salinity changes, even in Oyster Bay adjacent to the mouth of SRS. On a longer time scale, the microfaunal d18O record (Figure 5a,b) indicated a trend toward more marine conditions since about the turn of the century that was concordant with the changes in salinity-sensitive foraminifera (Fig 5c, Elphidium gunteri). Regional rainfall during this period (Figure 5d) has declined from 155 cm/yr (mean of first decade of data) to 150 cm/yr (most recent decade). Although these rainfall means are not statistically different (95% confidence level) major departures from the microfaunal trends for both d18O and E. gunteri support a strong linkage to regional rainfall at essentially the yearly level.
The data in Figs. 5a-c were reexamined for major departures from the long-term linear trends. Results indicate that these divergent data correlate with the regional rainfall record (Figure 5d). It is important to restate here that geochronology-based downcore ages have associated error estimated such that core dates in the 1940s and 1920s have a range of ±3 and ±4 years at Oyster Bay and ±6 and ±8 at Jimmy Key respectively. Within the limit of age estimates, the d18O and Elphidium gunteri records in the early- to mid-1920s and in the mid-1940s (Figure 5a-c) indicated excursions toward less-marine conditions contemporaneous with increased regional rainfall during both periods (Figure 5d). Similarly, a mid-1980s extreme excursion to more marine conditions at Oyster Bay (Figure 5a) coincided with record low regional rainfall (Figure 5d). Although a point-by-point matchup of rainfall/microfaunal data cannot be accomplished with these data, we believe that when viewed together, d18O and E. gunteri patterns at both sites provide compelling evidence for short-term, essentially yearly coupling of microfaunal variations to regional rainfall and the attendant salinity changes. Accordingly, these observations support rainfall/microfaunal response at essentially yearly time scales and imply validity of overall longer-term trends.
Anthropogenic Influences – Two major and concurrent shifts occurred in the early-1950s that were recorded in the Oyster Bay sediments. The first was a decline in the sediment’s organic C content relative to pre-1940s sediment (Figure 9a) and the second was the onset of major increases in foraminifer and ostracod abundances (Figure 3a, Zone B). Moreover, the transition from above- to below-average organic C (Figure 9a, arrow) temporally coincided with the onset of the late-1940s to early-1950s resurgence in watershed modification and impoundment of 700,000 acres of organic-rich soils during the creation of the Everglades Agricultural Area (EAA). Evidence from regional rainfall and freshwater flow data (Figure 10) indicates that a high degree of coherence still existed between these parameters from before this transition period until 1960 and thus water-flow trends appear unaltered. It is important to note that despite these changes in the concentration of organic C at Oyster Bay, there was no co-occurring change in the d13C (Figure 9b, D‰<1) indicating that the source of the organic C had not changed.
Concurrent with this transition of decreased organic C and increased anthropogenic modification of the watershed was a dramatic increase in foraminifera and ostracod abundance at Oyster Bay (Figure 3a, Zone B). The abundance increase ca. 1950 indicates only a biomass increase of the pre-1935 communities (i.e. – similar diversities, Figure 3b) and implies the onset of conditions favorable to the existing communities of microfauna. The temporal correlation with decreased organic C and impoundment of the EAA suggests a possible water quality change that may have influenced these communities, but this cannot be verified with our current data.
The abundance increase was followed by an equally dramatic decline during the late-1970s to early 1980s (Figure 3a, Zone A). Existing data strongly indicate the ensuing abundance and diversity drop is related to the changing nature of the salinity field at Oyster Bay after ~1980. Specifically, Zones A and B (Figure 3a) were not only statistically different for both abundance and diversity but concurrently the standard deviation of salinity (Ss) had changed by an amount observed by others to reduce macrobenthic biomass by nearly an order of magnitude (Tables 1 & 3). Montague and Ley (1993), in their study of seagrass and macrobenthic fauna in Florida Bay, showed that benthic biomass declined by an order of magnitude when Ss increased by 3. Our data parallels this with a 9:1 decline in abundance associated with a DSs = 2.7 between Zones A and B (Table 3). These observations provide a cogent argument for salinity-related changes as the driving factor for both decreased abundances and diversity between Zones A and B, but do not identify the cause of this salinity change. Although this abundance decline takes place approximately at a time of changing water-management strategy, more regional natural influences, such as regional rainfall, may also play a role, probably a dominant one, as noted earlier for Oyster Bay ostracod d18O data (Figure 6).
Nearly concurrent with Zones A-B changes at Oyster Bay (Figure 3a), are less abrupt but similar transitions at Jimmy Key (Figure 3c) that also link changing salinity fields with variability in microfaunal abundance. The Jimmy Key data indicate a statistically valid reduction in abundance which coincides with a rise in the Ss that parallels similar changes at Oyster Bay in both direction and approximate magnitude (Table 3). Moreover, the gradual decadal (Figure 3c, ~1970s+) abundance change observed for both foraminifera and ostracods at Jimmy Key appears to be a long-term, more natural transition also related to environmental changes such as salinity and rainfall. Comparison of these abundance changes to rainfall trends (Figure 10) confirms this hypothesis. Specifically, rapid declines in rainfall during the late-1960s through early- to mid-1970s (Figure 10) correspond to above average salinities (Figure 8b) and the onset and peak of foraminifer and ostracod abundances at Jimmy Key. Increasing and above average rainfall thereafter, until approximately the mid-1980s, corresponds to declining abundances (Figure 3c) and salinities (Figure 8b).
Embedded within the abundance data for Oyster Bay (Figure 3a) is a subfield of information derived from two critical species, the foraminifer Ammonia parkinsoniana typica and the ostracod Peratocytheridea setipunctata. Both field and culture studies (Phelger, 1956, 1966a,b; Keyser, 1976; Poag, 1978; Garbett & Maddocks, 1979; Murray, 1991) have shown these species to be tolerant of wide fluctuations in salinity that tend to reduce community diversity through stress. Specifically, between ~1980 and core recovery (1995) both microfaunal abundance and diversity are significantly lower than in the preceding two decades (Figure 3a, Zones A vs B). Comparison of the absolute abundances for these two species (Figure 4a) mirrors the changes in abundance for the entire microfaunal community (Figure 3a). In contrast, their relative abundance, on the average, doubles upcore from Zone B to A reaching intervals throughout the 1980s of >50% of the total population (Figure 4b). This indicates a stressed microfaunal community dominated by species, which survived during prolonged periods of reduced flow and drought (Figure 10). When viewed as a whole, microfaunal abundance and diversity, salinity, flow in SRS, and rainfall data for both Oyster Bay and Jimmy Key, suggest that anthropogenic influences play a secondary role to natural influences such as regional rainfall.
Summary
and Conclusions.
Recovered sediments from Oyster Bay, adjacent to the outflow from the Shark River Slough and central Florida Bay near Jimmy Key recorded nearly a century of chemical, biological, and physical data. When these data were interpreted, results allowed the following conclusions:
· Regional rainfall and freshwater flow in Shark River Slough. Rainfall at Homestead proved representative of the study area and when compared to Shark River Slough flow indicated a strong positive correlation during periods preceding construction of the current Water Conservation Areas and subsequent regulated flow. After construction of the Water Conservation Areas, correlations between regional rainfall and flow in Shark River Slough varied with water management practices, ranging from essentially no correlation during the Monthly Allocation Plan to correlation approaching pre-management levels for the Rainfall Plan.
· Salinity trends and microfaunal responses. Changes in salinity, both near the outflow of the Shark River Slough at Oyster Bay, and in central Florida Bay near Jimmy Key, show a direct response to regional rainfall from the decadal to the monthly time scales. At both Jimmy Key and Oyster Bay, foraminifer and ostracod data also indicate direct correlation to rainfall patterns for temporal scales ranging from decadal down to the limit-of-resolution of our geochronology. At Oyster Bay, ostracod stable isotope (d18O) trends correlated better with variations in regional rainfall than with freshwater outflow from the adjacent Shark River Slough. An order-of-magnitude drop in foraminifera and ostracod abundances in the late-1970s at Oyster Bay, and a more gradual decrease in the mid-1970s at Jimmy Key correlated not to changes in mean salinity but rather to increases in the standard deviation of salinity. Stable isotope (d18O, d13C ) trends for ostracods and foraminifer at Oyster Bay and Jimmy Key showed mixed signals with most data suggesting upcore (to present) trends to less fresh, more marine conditions.
Acknowledgements.
Salinity data were provided by Drs. J. Boyer and R. Jones at Florida International University as funded by the South Florida Water Management District and Everglades National Park. The HisSal05 data set was provided by Dr. M. Robblee. T. Ross, of the National Climate Data Center, provided rainfall data and the USGS provided flow data for the Shark River Slough.
Duever, M. J., Meeder, J. F., Meeder,
L. C. and McCollom, J. M., 1994, The climate of south Florida and its role in shaping the
Everglades ecosystem, in Everglades: The Ecosystem and Its Restoration, Davis,
S. M. and Ogden, J.C., Eds., St. Lucie Press, Delray Beach, Florida, 225.
Garbett, E. C. and Maddocks, R. F., 1979, Zoogeography of Holocene
Cytheracean Ostacodes in the Bays of Texas, J. Paleontol., 53, 841.
Keyser, D., 1976, Ecology and
zoogeography of recent brackish-water Ostracoda (Crustacea) from southwest Florida, Sixth
International Ostracod Symposium, Saalfelden, 207.
Montague, C. L. and Ley, J. A., 1993,
A possible effect of salinity fluctuation on abundance of benthic vegetation and
associated fauna in northeastern Florida Bay, Estuaries, 16, 703.
Murray, J. W., 1991, Ecology and
Paleoecology of Benthic Foraminifera, John Wiley and Sons, New York.
Parker, F. L., F. B. Phleger, and J.
F. Peirson. 1953. Ecology of foraminifera from San Antonio Bay and environs, southwest
Texas. Contributions for the Cushman Foundation for Foraminiferal Research, special paper
2:1.
Phleger, F. B., 1956, Significance of
living foraminiferal populations along the central Texas coast, Contributions for the
Cushman Foundation for Foraminiferal Research, 7, 106.
Phleger, F. B., 1966a, Living
foraminifera from coastal marsh, southwestern Florida, Boletin de la Sociedad
Geologica Mexicana, 28, 45.
Phleger, F. B., 1966b, Patterns on
living marsh foraminifera in south Texas coastal lagoons, Boletin de la
Sociedad Geologica Mexicana, 28, 1.
Poag, C. W., 1978, Paired
foraminiferal ecophenotypes in Gulf coast estuaries: Ecological and paleoecological
implications, Trans.,
Gulf Coast Assoc. Geol. Soc., 28, 395.
