Ecosystem analysis of phosphorus impacts and altered hydrology in the Everglades: A landscape modeling approach

The Everglades has undergone significant change in response to altered hydrology and water quality, which is why natural resource managers are now evaluating alternative water and nutrient management strategies for the region. Simulation models are an integral part of the process of understanding complex ecological systems, providing a means to evaluate potential ecosystem response to changes in management. To evaluate various management alternatives on the Everglades ecosystems, we developed a spatially explicit ecosystem model for Water Conservation Area 2A (the Conservation Area Landscape Model, or CALM). The CALM simulates interactions among hydrology, chemistry, and biology of the marsh systems across the landscape, synthesizing ecosystem behavior in response to changing environmental inputs. Calibration results for 1980 to 1996 indicated good hydrologic and ecological agreement with observed data. Observed and simulated water stage were well correlated (r(2) = 0.70). North-south gradients of simulated dissolved inorganic phosphorus in the surface waters (50 - 4 mu g P L-1) and in the pore waters (950 - 10 mu g P L-1) were spatially and temporally realistic. Likewise, simulated and observed data along the gradient had similar values of peat accretion (5.5 - 3.1 mm y(-1), respectively), macrophyte biomass (1100 - 300 g C m(-2), respectively), calcareous periphyton biomass (0 - 52 g C m(-2), respectively), and community shifts in periphyton and macrophytes. The model captures the feedbacks among plants, hydrology, and biogeochemistry through process-based algorithms, including the influence of spatial patterns on these processes. For example, in a 17-year (1980 to 96) simulation sensitivity analysis of increased evaporative losses, there were more pronounced regions of increased soil P remineralization associated with reduced hydroperiods, which in turn increased macrophyte growth. If future management of the area lowers water levels below those that occurred in the 1980s, the simulation under drier conditions indicates that changes in the system's biogeochemistry are likely to occur beyond those directly linked to reduced allochthonous nutrient loads. In another simulation scenario in which external phosphorus loads were reduced, the CALM indicated that some landscape measures, such as sorbed phosphorus and macrophyte biomass, would not be immediately affected (improved) in all currently-impacted areas. That scenario indicated that internal cycling of phosphorus would likely continue to maintain a eutrophic state in some impacted areas for years, with soil porewater P increasing for a decade, albeit at reduced rates compared to a simulation with actual, observed loads. In that reduced-load scenario run, appearance of calcareous periphyton within the currently-impacted zone reflected the reduction in surface water nutrients compared to the nominal run. However, macrophyte biomass land thus shading) was not greatly reduced in much of that zone, thus preventing calcareous periphyton from attaining high densities that are found in pristine areas. We are currently working on refinements and further model verification in anticipation of applying the model framework toward evaluating Everglades restoration alternatives in the entire Everglades/Big Cypress region.