LINKING PALEOECOLOGY AND LINEAR REGRESSION MODELS TO ESTIMATE NATURAL HYDROLOGIC CONDITIONS FOR RESOURCE MANAGEMENT

Understanding historic hydrologic conditions is an important issue for resource management, particularly in systems undergoing restoration. A 30-year effort is currently underway to restore the Everglades ecosystem of south Florida. Success of this effort depends on re-establishing more natural hydrologic patterns. Everglades resource managers initially relied on complex hydrologic systems models to establish salinity targets and performance measures for restoration; however, these models were not in agreement with paleoecologic and/or anecdotal data about conditions in the near-shore estuaries prior to alteration of the natural system (~1910 CE). To fill this information gap, we have developed a 3-phase process that links paleoecologic information on salinity with linear regression models (LRMs) developed using observed hydrologic data (salinity, stage, flow). In phase 1, paleosalinity estimates are derived from molluscan assemblages in 5 radiometrically-dated sediment cores using a modern analog approach. The second phase is the development of LRMs that enable the prediction of salinity, stage, and flow based on observed data from hydrologic stations located throughout the Everglades ecosystem. To produce a pre-drainage simulated time series, a natural system model (NSM) that produces stage time series for the 1965-2000 period of record is used as input to the LRMs. The paleosalinity estimate produced in phase 1 is used to adjust the NSM salinity to yield a paleo-based salinity time series. In the final phase, this paleo-based salinity time series replaces the observed salinity in the statistical models to estimate the upstream flow and stage necessary to achieve the paleosalinity estimates. The results show that in the absence of water management, Florida Bay salinity would be approximately 3 to 9 practical salinity units (psu) lower than current conditions. In order to achieve these salinities, freshwater flow into the upstream marshes would need to be 2.1 to 3.7 times greater than existing flows and the upstream stage would be approximately 0.25m higher than current conditions. Scientists and resource managers working in coastal ecosystems with available paleoecologic information and empirical hydrologic data can use this method to establish targets for effective and sustainable restoration.