Evaluating the impact of agricultural BMPs on riverine nutrient export to Lake Erie

We used the Soil and Water Assessment Tool (SWAT) applied to seven watersheds draining into Lake Erie to simulate implementation of various agricultural best management practices (BMPs). The SWAT watershed models were calibrated and validated for stream discharge and water quality parameters using data from 1998-2005 by using time series plots and statistical measures to verify model predictions.  The models were calibrated and validated against daily flow measurements and near-daily sediment and nutrient concentration measurements near the watershed outlets.  Simulated hydrology and water quality parameters closely resembled observed data overall.  For a number of BMP scenarios we predict changes in daily and annual streamflow and sediment and nutrient loads in river export.  SWAT scenarios showed that the implementation of agricultural BMPs had a marked effect on sediment and nutrient export from all watersheds.  Furthermore, certain BMPs were shown to be much more effective at reducing riverine sediment and nutrient fluxes than others.  For example, we predicted that adding 10 m grass filter strips to agricultural land in one watershed would result in 22%, 27%, and 3% reductions in annual river loads for total phosphorus, total nitrogen, and sediment respectively.