Modeling Ecosystem Interactions in Florida Bay

Topical Area: Algal Blooms, Zooplankton and Phytoplankton Ecology

George A. Jackson and Adrian B. Burd, Department of Oceanography, Texas A&M University,

College Station, TX

The aim of this project is to study the nutrient and trophic dynamics of the planktonic ecosystem in Florida bay using two complementary approaches. First, we are analyzing existing data for representative basins to estimate the nutrient and carbon flows between different planktonic trophic groups and their interaction with the benthos. The second approach is the development of models to explore the dynamics of interactions among the different trophic groups in the ecosystem.

To understand the movement of carbon and nitrogen through the planktonic ecosystem of Florida Bay we must first estimate the rates at which they pass through and are transformed by the food webs. Unfortunately, marine food webs are complicated systems having the potential for a myriad of interactions that are not easily sampled or understood. Typical ecological studies of marine food webs are able to measure only a few of the many interactions known, or suspected to be occurring within them.

The application of inverse techniques to the analysis of marine ecosystems allows estimates to be made of all the interactions taking place within the food web using a limited set of measurements. The technique (introduced by Vezina and Platt) incorporates information about known assimilation and production efficiencies of organism groups together with a description of the flows in the food web and results of field measurements.

We have applied this technique to various basins within Florida Bay using measurements made at different times of year and supplied by M. Dagg and P. Ortner. Many of the basins that make up Florida Bay have depths of less than 3m, giving the benthos a disproportionate influence on the system relative to most coastal systems. Preliminary results from our inverse model show that about half of the primary production is being exported to the benthos, either by filter feeders or by deposition and subsequent degredation. This result appears to hold for different basins and at different times of the year.

Present measurements require little flow to dissolved organic carbon (DOC) and bacterial components of the food web. However, this could be an artifact of limited measurements of the cycling of these components. At present, seagrasses are not explicitly included in the food web being modeled, so detrital and DOC release from the plants is not directly accounted for. If measurements of DOC cycling and bacterial metabolism indicate that they are large components of the system, we will have better estimates of benthic interactions.

The analysis results indicate large scale spatial homogeneity within the Bay. Food webs for Rankin and Twin Keys using data taken during September 1997 show a considerable similarity. Comparison between results for September and July 1997 for Twin Key also indicate a shift of meso-zooplankton grazing from phytoplankton in July to micro- and proto-zooplankton in September.

Investigating the structure of a food web using inverse analysis is a data intensive activity. We will be incorporating more data into the analysis as they become available. This will enable us to include more compartments allowing a better representation of the interactions of the system with the benthos. In addition we will be able to add and subtract compartments.

The analysis of the data using the inverse techniques gives a static picture of the food web. To investigate the dynamics of the system, we are using the results from the inverse method to

Using the results from this analysis of the data we are developing a dynamic model of the trophic interactions seen in Florida Bay and hope to present initial results from this model at the meeting.