Application of Apex and Remm in Evaluating Water Quality Benefits of Riparian Buffers.

Calibrating and validating field and hillslope scale models presents special challenges, especially in areas dominated by extreme and unpredictable hydrologic events. Some of these challenges will be illustrated using water quality/hydrologic models and field data to estimate the water, nitrogen, and phosphorus fluxes from an agricultural field and riparian buffer to a mangrove wetland in Jobos Bay Watershed, Puerto Rico.  We used the Agricultural Policy/Environmental eXtender (APEX) model and the Riparian Ecosystem Management Model (REMM) sequentially to simulate the water quality and hydrology of the agricultural fields and an adjacent riparian buffer, respectively.  The basic data used for hydrologic calibration were monthly and daily water table depths in the field and riparian area. Depth to the water table surface  was measured monthly at numerous sites in both field and riparian areas and were integrated with recording well data from outside the field to estimate daily water table depths in the field and  riparian zone and to evaluate field-scale hydrologic processes. Calibration and validation of the models was done using the daily water table depths. Although calibrations and validations were successful based on goodness of fit statistics (R2 0.41 to 0.95 and Nash-Sutcliffe 0.33 to 0.80) better calibrations could have probably been achieved if surface runoff hydrology was monitored. This system, as are many, was dominated by large unpredictable hydrologic events. These events, due to tropical storms and hurricanes, would have been very difficult to monitor due to the amount of overbank flow and the weather conditions during the flow event.