The design of ecological landscape models for Everglades restoration

Restoration of the Everglades is a multi-objective, multi-scale, multi-agen cy program that requires numerous computer models to test alternatives, und erstand ecosystem processes, and evaluate restoration performance. Landscap e models used for Everglades restoration include hydrologic models, transit ion probability models, gradient models, distributional mosaic models, and individual-based models. As tools for restoration feasibility and as the ba ckbone of the policies that will drive Everglades restoration for the next 20 years, it is critical that a wide audience evaluate the strengths and we aknesses of six landscape models. Simulations of historic hydropatterns and current hydropatterns, based mostly upon sheet-flow equations and canal-fl ow equations, respectively, have been the realm of the Natural Systems Mode l (NSM) and the South Florida Water Management Model (SFWMM). Despite a lac k of biology in these two models, a comparison of their spatial output beca me the basis for the Comprehensive Everglades Restoration Plan (CERP) appro ved by the US Congress in October, 2000. SAWCAT, a transitional probability model, was based upon an analysis of the patchiness of cattail (Typha) and sawgrass (Cladium) cells in association with levees, water depth, and phos phorus. This statistical approach was used to predict the amount of sawgras s that would be converted to less desirable cattail, if phosphorus runoff p atterns to the Everglades remained constant. The Everglades Water Quality M odel (EWQM), a mass-balance gradient approach used to track phosphorus acco rding to a simple net phosphorus removal coefficient, was used to design St orm Water Treatment Areas (STA) and to evaluate where and when phosphorus ' thresholds' would be exceeded under various hydrologic restoration plans. T he Everglades landscape Model (ELM), a complex distributional mosaic model, used site-specific biogeochemical mechanisms and mass-balance to control e nergy and material flows, and to predict changes in carbon and phosphorus s tructure of the soil, water, and plant communities as a result of modified water deliveries to the Everglades. The Across Trophic Level Spatial Simula tion (ATLSS), also a distributional mosaic modeling approach, used individu al-based rules of behavior to predict animal movement and abundance in rela tion to hydrologic restoration plans. When these landscape models are combi ned, they effectively contribute to water management and policy for Evergla des restoration. To insure their effectiveness, an applied science strategy provides the framework for their integration into the restoration process. (C) 2001 Elsevier Science B.V. All rights reserved.