Ecologic and precursor success criteria for south Florida ecosystem restoration

A Science Sub-Group Report to the Working Group of the South Florida Ecosystem Restoration Task Force

[ Next Chapter ] [ Previous Chapter ] [ Table of Contents ]




The salinity success criteria team consists of Drs. Joan Browder, Robert Fennema, Robert Halley, and Michael Robblee. We met as a group on January 23, 1996 and followed up our meeting with phone conversations, faxes, and a draft manuscript. An original draft was produced during February and restructured to fit the requested format during the first week in March. The draft was distributed to team members for review and revision on March 7. Revision and comments were incorporated into this draft during the last week in March.

The approach taken is to concentrate on Florida Bay because of pre-existing information available from the Bay and because of the limited time and resources of the team. It is recognized that similar criteria should be established for the Gulf Coast estuaries of ENP, Biscayne Bay, Rookery Bay, St. Lucie-Indian River Estuarine System, Estero Bay, and possibly Charlotte Harbor. Most of these areas need additional monitoring, modeling, and analysis before criteria can be established. These information needs are addressed in the document South Florida Ecosystem Restoration: Scientific Information Needs" produced by the Science Subgroup, September 1994.


The salinity success criteria we propose are based on the haloperiod, defined here as a climatically modeled range of annual salinity variation at a particular locality in the Bay. The haloperiod is normally presented as a graph showing monthly salinity averages during a year. The haloperiod is analogous to the hydroperiod used in wetland ecology. As the hydroperiod describes depth and duration of standing water in a wetland for a year, so the haloperiod describes the annual variation of salinity in an estuary. Because South Florida is subject to substantial inter-annual climatic variation, it is important to relate haloperiod in Florida Bay to actual climatic measurements.


Ideal haloperiods should be established for geographically representative locations within Florida Bay. We suggest these be at existing ENP monitoring stations and chosen to monitor fresh water inflow (Long Sound, Trout Cove, Little Madiera Bay, McCormick Creek, and Garfield Bight), and salinity in the central and southwest Bay (Whipray Basin and Peterson Keys Basin). At this time only the haloperiod at Little Madiera Bay is available and is based on modifications of previous work.

A. Statement of success criteria

The success criteria are derived from the difference between a modeled (ideal) haloperiod, derived from the best available climatic and hydrodynamic models using climatic data for the previous twelve months, and an measured haloperiod for a location in the Bay. Each month is scored individually by comparison with monthly measurements and the sum for twelve months is taken as a score from A ( or 1, being most improved) to E (or 5, being least improved) following the methodology described in part IIIE below.

B. Reference baseline

In the context of success criteria, a reference baseline for Florida Bay salinity does not exist. Salinity in the Bay is constantly varying in response to natural fluctuations in rainfall, freshwater inflow, evaporation, sea level, and circulation. Wide inter-annual (long term) variations in climate make characterization of the salinity difficult, if not impossible.

C. Historic conditions including predrainage, if known

The predrainage (prior to the turn of the century) salinity variation in Florida Bay is not directly known. However several sources of information about past salinity of the Bay are, or will be, available to researchers. These sources of information are: 1) historical salinity data; 2) data from on-going salinity monitoring; 3) salinity predicted, or hindcast, from several modeling approaches; 4) geological records from the sediments of the Bay; and 5) historical observations related to salinity of the Bay. These sources are of varying usefulness and each source has its strengths and weaknesses when used to establish success criteria.

  1. Salinity data has been collected intermittently in Florida Bay since the late 1930s. By the 1960s, many ecological and geological research projects in the Bay collected some accompanying salinity data. Some of these studies monitored salinity for as much as several years. These data have recently been assembled by Robblee (unpublished data) in a database. The database provides information to develop basic salinity relationships between different sub-basins in the Bay and between salinity and climatic variations. From the data base it is clear that hypersalinity has occurred during low rainfall periods since the late 1930s. The highest salinity recorded is 70 ppt on July 25, 1956 in Snake Bight.

  2. Beginning in the late 1980s several agencies established water quality monitoring programs that collect salinity at regular intervals and locations in Florida Bay. The National Park Service, South Florida Water Management District (through contract to the Southeastern Environmental Research Program at Florida International University), and Dade County conduct surveys at different frequencies and varying density of data collection. The USGS collects salinity data every other month in order to construct a map showing the relationships between mudbanks and resulting salinity contours.

  3. A variety of models have been and are being constructed to predict salinity in the Bay. Among the earliest were models driven by correlation between groundwater elevations in southern Dade County and salinity in the northern and eastern sub-basins (Tabb and others, 1967; Sculley, 1986). McIver et al. (1994) predicted salinity in portions of the Bay using Tabb's method, but substituted ground water elevations derived from the natural System Model for measured groundwater elevations. Currently more sophisticated models are being developed including a salt balance model and a hydrodynamic circulation model linked to fresh water inflow and climatic conditions. It will be some time before these latter models are thoroughly developed and tested.

  4. Cores of sediments and coral skeletons from the Bay can provide proxy measures of pre-historical salinity (Smith et al., 1989; Wingard et al., 1995). Sediment cores must be carefully analyzed to demonstrate that relatively undisturbed sediments have accumulated at appropriate rates to provide annual to decadal time resolution during the past century. Once appropriate chronology is established, geochemical analyses of the sediment or paleoecological analyses of associated fossils can be used to approximate past salinity. Cores of coral skeletons show annual variation of skeletal density which is used to establish chronology, much like tree-ring dating. Again, once the chronology is established, geochemical estimates of salinity can be made from skeletal chemistry. It is important to verify the results of these studies by comparing analyses from multiple cores.

    5) The historical literature of South Florida is replete with observations of ecological and hydrologic conditions that relate indirectly to the salinity of Florida Bay. Many of these deal with elevated water tables in the Everglades (Parker, 1974). Example are photographs of the freshwater springs of Biscayne Bay, the falls in the Miami River (Parks, 1977), and the ability of Willoughby to canoe over the Rocky Glades during the last week of January, 1896 (Willoughby, 1898). Another important set of observations in the timing of human alteration such as the completion of the Flagler railway in 1912, the Homestead Canal in 1922 (Will, 1967), and the Tamiami Trail in 1928. Although it is difficult to relate these events quantitatively to the salinity of the Bay, they serve as important temporal benchmarks for the history of landscape alteration in South Florida. Many of these events are summarized in Cantillo et al. (1995).

    The historical record of human alteration in South Florida provides us with the temporal perspective of possible anthropogenic influence on Florida Bay. It suggests the possibility that major changes in Everglades hydrology and freshwater flow into the Bay may have taken place prior to the first salinity measurements in the Bay. Thus, although the historical salinity data and salinity monitoring provide invaluable insight into the range and spatial relationships of salinity in the Bay, these data do not describe the Bay prior to significant upstream and downstream alterations. Geological analyses do cover the appropriate timespan, but such proxies are indirect, not available in some critical geographic locations of the Bay, and dating resolution may be on the order of decades.

    D. Expectation from restoration (trend or number)

    It is expected that restoration will reduce the frequency of hypersalinity in the central Bay and that salinity will be reduced in those areas of the Bay directly affected by freshwater inflow. The numerical score is expected to decrease from 5 to 1.

    E. Scoring and scoring methodology: Example from Little Madiera Bay

    For the task at hand, we have used a modified method of Tabb (1967) that correlates salinity in the Bay with the groundwater elevation as measured in well S-196A (Homestead well). Tabb's (1967) correlation of salinity at Little Madiera Bay with the Homestead well is shown in Figure 1. The modification used here is that groundwater elevation is derived from rainfall using a relationship between precipitation and S-196A 1" elevation for the period 10/1932 - 12/1942 (Figure 2). For the initial exercise, the criterion is calculated for Little Madiera Bay, a location studied by Tabb (1967) and a site that reflects freshwater inflow to the Bay from Taylor Slough.

    Selection of the 1932-42 period for precipitation/groundwater correlation has advantages and drawbacks An important advantages is that the first ten years of the record most closely represents the natural system. A long term change is apparent in the S- 196A record that probably results from water management modifications in the region. Although some shift in the record is apparent after 1945, a prominent change takes place after 1968 when the Central and Southern Florida Project was completed. Another advantage is that groundwater e}elevations tend to integrate precipitation over a broad area. Local precipitation events may cause short-term, localized increases in the water table, but on the time-scale of monthly averages used here, these brief events are undetected (Parker et al,, l 1955). An * additional advantage noted by Tabb (1967) is that the location of S-196A monitors the headwaters of Taylor Slough (prior to the installation of C-111 and L-3IW) which flows into little Madiera Bay. As Figure 2 illustrates, the relationship between groundwater elevation (E) and precipitation is not strong (only 44% of the variance is explained) and precipitation (P) is a particularly poor predictor of elevations above 6 ft. The regression of precipitation against elevation yields:

    (1) E=2322+0.361 xP

    Substituting 1994 monthly average precipitation in equation 1 provides a calculated monthly elevation that can then be used in Tabb's (1967) relationship between groundwater elevation at S-196A and Madiera Bay Salinity (S)

    (2) S = 49.496 - 6.889 x E (from Tabb, 1967)

    Figure 3 compares the predicted haloperiod of Little Madiera Bay, derived from equations l and 2, with measured salinity for 1994. The added error bars of +5 ppt account for most of the variation in Tabb's (1967) relationship. Only two months, February and December, show agreement within error.

    On order that the criterion meet the prescribed scoring scale (1 to 5), months that show agreement within error are scored 1/12 (0.0833) and months that do not are scored 5/12 (0.4167). All months are then summed for the year. In the Little Madeira Bay case for 1994 the score is:

    (3) (2 x 0.0833) + (10 x 0.4167) = 4.334

    If the scoring required is a letter grade, then the "report card" receives a D (actually a D-) for the salinity success criterion for 1994.

    IV. Supporting information needed

    Tabb (19676) recognized that a simple linear correlation is a relatively poor model relating salinity in Florida Bay to groundwater elevation or precipitation. In Little Madiera Bay the approximation is useful because fresh water from Taylor Creek mixes with Bay water at that site. Tabb (1967) showed that linear correlations become weak at localities further into the Bay because factors such as circulation and evaporation become more important than simple mixing. Salinity stratification has been shown to introduce nonlinearities between fresh water flow and salinity in estuaries of western Everglades National Park (Browder et al., 1989). The simple method used here needs vast improvement based on more sophisticated relationships and modeling methods. One approach is to use the Natural System Model to generate groundwater elevations as was done by Van Lent et al. (1993) and McIver et al. (1994). An even better approach is to integrate groundwater and surface water models of inflow from the surrounding mangrove shores into a complete hydrodynamic circulation model of the Bay. The latter must include evaporation and precipitation routines which may require additional monitoring in the Bay.


    Success criteria haloperiods need to be established for a minimum of seven locations in the Bay. As described above, these should be selected to monitor freshwater flow into the Bay and frequency of hypersaline events in the central and western Bay.

    The salinity criteria used here are based on physical parameters but biological indices for salinity may also be appropriate. In a sense, salinity is a precursor index for ecological criteria. If species can be identified that have known responses to salinity variation and are desired in the Bay, then salinity criteria based on their geographic distributions may offer the potential for a biologically based salinity success criteria.


    A. Prepare, improve, and test recommended success criteria

    Important improvements to the criteria have been outlined above and require the successful completion of many ongoing projects in the Interagency effort for restoration. Valuable sources of information listed in section IIIc will become available as the studies underway are brought to completion. These include Florida Bay circulation modeling, monitoring freshwater flow into the Bay, groundwater and surface water modeling in the areas between Taylor Slough/C-111 and the Bay, improved climatic monitoring, continued salinity and water quality monitoring, quantification of sedimentation, paleoecology and geochemistry. All of the modeling to date needs to be carefully tested and verified to determine the appropriateness of the models for answering management questions. This will require significant investment in the gathering of data specifically to test models.

    B. Interpret Results

    An interesting challenge for the near term is to interpret and integrate the result's from various studies in different agencies into improved success criteria. Currently models are being developed that are cite specific (the expertise to run them lies in one agency). These models will somehow need to be made available to a variety of scientists and managers as community tools if general agreement on success criteria is to be established.

    C. Develop potential success criteria

    Although our team is able to outline an approach to success criteria using haloperiods, the methods generating ideal haloperiods are grossly oversimplified. As extensively integrated models are created, the task for developing success haloperiods should be directed to the modeling community. Integrated models should be designed so that haloperiods are a direct output of complex relationships that can replace the simple correlations presented here.


    The ambitious federal and state program to preserve and restore the Everglades and associated ecosystems in South Florida has attracted national attention in the scientific community (Holloway, 1994; Culotta, 1995). Much attention has been focused on changing water quality in the Bay, particularly the episodes of hypersalinity in the central and eastern Bay. Observations of hypersalinity during the last, local, climatic dry period (1985-1990) together with subsequent environmental alteration have resulted in a salinity paradigm. The Florida Bay salinity paradigm is that Florida Bay has become more hypersaline in response to onshore water management practices. Beginning in the 1880's, drainage canals, dikes and pumping have been built to recover land for urban and agricultural development and to protect it from floods. Much of the water management infrastructure was in place by the 1950's and projects were completed in the 1960s. A significant portion of water that historically flowed through the Everglades into Florida Bay is now routed to the Atlantic Ocean and Gulf of Mexico. The degree to which changes in freshwater flow to the Bay are known has been reviewed by McIver et al. (1994).

    Land management agencies have taken the arguable approach of requiring guidelines by which to measure the success of their efforts to restore South Florida ecosystems. One is a salinity criterion for Florida Bay. Managers have asked scientists to provide quantitative guidelines for the salinity of Florida Bay. Unfortunately this is not an entirely objective exercise. Many complex and partially unknown processes control the salinity of Florida Bay. The process of artificially quantifying success criteria is inherently risky. The fact that subjective judgments can be quantified should not be allowed to obscure the fact that they are judgments, and that final decisions should rest on sound application of these judgments rather than upon uncritical application of data. However, we recognize that in order to make progress in ecosystem management activities, managers and engineers require objective goals. The criteria proposed here are preliminary and subject to change as more becomes known about the processes in the Bay, particularly the results of developing circulation models. The restoration process benefits from the managerial view that the South Florida Ecosystem restoration program is an adaptive management effort and that new scientific results can be incorporated as the program progresses.

    In this report the authors describe an approach to salinity criteria that is in large part non-ecological. The authors recognize that an equally valid approach might be to establish the salinity requirements from the autecology of component members of a biological community which must be predetermined ideal for the Bay. However, in the absence ecological criteria, we have adapted a previously used approach in order to prescribe salinity success criteria for the Bay.

    In the absence of human intervention, Florida Bay is ultimately controlled by climatic, biologic, and geologic processes that cause salinity to vary widely in space and time. Climatic factors were evident in the very first salinity measurements documented from the Bay in 1936to 1938 when it was noted that salinity less than that of average sea water was prevalent during wet years and hypersaline conditions existed in dry years (Davis, 1940). During the 1950's it was pointed out that mudbanks in the Bay restrict circulation and that the banks are the accumulated remains of organisms which live in the Bay (Ginsburg, 1957). These and many subsequent studies indicate that success criteria for salinity need to be variable in space and time and derived from climatic conditions. The criteria may also have to be modified to account for long term geologic and biologic processes.

    Florida Bay restoration is not meant to connote restoration of the Bay to a particular time, rather the restoration is aimed at restoring a state of the Bay, a state of less radical ecological change than has occurred during the past decade. What the Bay was in the past, is not necessarily what it should be. Natural, long term processes, have permanently changed the Bay during the past 100 years. In some places almost a yard of sediment has accumulated since the turn of the century causing significant shoalling. During the same time sea level has risen more than 6 inches, which might be considered insignificant except for the fact that that large areas of the mud banks are less than one foot deep and exposed at low tide. Managers are not likely to undertake lowering sea level or altering natural patterns of sedimentation to "restore" the Bay. However, to the extent that past states of the Bay are considered the natural state of the Bay, they should be targets for restoration.

    During 1994 and 1995, no major structural changes had yet been made to change water delivery to Florida Bay. Therefore, evidence of progress toward restoration is not to be expected. The scoring of 4.3 reflects this. Although any score less than ~ might be considered improvement, 4.3 probably reflects an improvement (reduced salinity) caused by the dramatic increase in precipitation in late 1994. Although the success criteria are designed to take climatic change into account, the simple correlations used here are insufficient to completely buffer salinity criteria from such large climatic variation


    Browder, J. A., Tashiro, J., Colman-Duffie, E., Rosenthal, A., and Wang, J. D., 1989, Documenting estuarine impacts of fresh water flow alterations and evaluating proposed remedies. Proceedings of the International Wetland Symposium, Wetland and River Corridor Management, 300-312, Charleston, South Carolina.Berne, N.Y., The Association of Wetland Managers, Inc.

    Cantillo, A. Y., Pikula, L., Beattie, J., Collins, E., Hale. K., and Schmidt, T., 1995 Natural and anthropogenic events impacting Florida Bay, 1910-1994 time line, NOAA Technical Memorandum NOS ORCA 90, 510p.

    Culotta, Elizabeth, 1995, Bringing back the Everglades. Science. 268:1688-1690

    Fennema, R. J, Neidrauer, C. J., Johnson, R. A., MacVicar, T. J., and Perkins, W. A., 1994, A computer model to simulate natural Everglades hydrology, in: Davis, S. M., and Ogden, J. C., Everglades. the Ecosystem and its Restoration. St. Lucie Press, Delray Beach, FL, pp. 117-146.

    Ginsburg, R. N. 1964 South Florida Carbonate Sediments, Guidebook for Field Trip No. 1, Geological Society of America Convention, Miami, Florida 71p. (Republished in 1972 as Sedimenta II, Comparative Sedimentology Laboratory, Rosenstiel School of Marine and Atmospheric Sciences, University of Miami.

    Holloway, Marguerite, 1994, Nurturing Nature. Scientific American. 270:98-108

    Lloyd, R. M., 1964, Variation in the oxygen and carbon isotope ratios of Florida Bay mollusks and their environmental significance: Jour. Geology, v. 72, p. 84-111

    McIver, C. C., Ley, J. A., and Bjork, R. D., Changes in freshwater inflow from the Everglades to Florida Bay including effects on biota and biotic processes: a review. in: Davis, S. M., and Ogden, J. C., Everglades: the Ecosystem and its Restoration. St. Lucie Press, Delray Beach, FL, pp. 117-146

    Parker, G. G. 1974. Hydrology of the pre-drainage system of the Everglades in south Florida. Pages 18-27 in P.J. Gleason, ed. Environments of south Florida: present and past. Memoir 2: Miami Geological Society. Miami, Florida.

    Perker, G. G., Ferguson, G. E., and Love, S. K., 1955, Water resources of Southeastern Florida, USGS Water Supply Paper 1255, 965p.

    Parks, A. M., 1977, The Forgotten Frontier - Florida Through the Lens of Ralph Middleton Munroe, Banyan Books, Inc., Miami, FL, 178p.

    Robblee, M. B., (unpublished), A historical salinity database for Florida Bay

    Science Subgroup, 1994 (unpublished), South Florida Ecosystem Restoration: Scientific Information Needs, 661p., Everglades National Park.

    Sculley, S., P., 1986, Florida Bay salinity concentration and groundwater stage: correlation and regression, South Florida Water Management District unpublished report

    Smith, T.J. III, J.H. Hudson, M.B. Robblee, G.V.N. Powell, and P.J. Isdale. 1989.Freshwater flow from the Everglades to Florida Bay: a historical reconstruction based on fluorescent banding in the coral bands in the coral Solenastrea boumoni Bull. Mar. Sci., v. 44. p. 274-282.

    Tabb, D. C., 1967, Prediction of estuarine salinities in Everglades National Park, Florida, by the use of ground water records. University of Miami, Ph. D Dissertation, 107p.

    Will, L. E., 1967, A Dregeman of Cape Sable, Great Outdoors Publishing Co., St. Petersburg, FL, 156p.

    Willoughby, H. L., 1898. Across the Everglades, reprinted (1992) by Florida Classics Library, Port Salerno, FL, 192p.

    Wingard, G. L., Ishman, S. E., Cronin, T. M., Edwards, L. E., Willard, D. A., and Halley, R. B., 1995, Preliminary analysis of down-core biotic assemblages: Bob Allen Keys, Everglades National Park, Florida Bay, USGS Open-File Report 95

    [ Next Chapter ] [ Previous Chapter ] [ Table of Contents ]

    This document is designed and maintained for the Everglades Digital Library by:

    Project Director
    Everglades Information Network & Digital Library
    Florida International University Libraries

    Mark-up © Florida International University, 1998. All rights reserved.