Using SWAT to Identify Critical Source Areas of Nutrient Loads and Model Reduction Strategies for the Upper Scioto Watershed, Ohio.

Excessive nutrient loads from agronomic watershed have been blamed for the water quality impairment in the last three decades. A number of nutrient load reduction strategies have been proposed to mitigate nutrient pollution. However, most of the currently adopted strategies are not effective enough because of failure to quantify the diffuse nutrient loads and locate the critical source areas. This study attempts to simulate the nutrient loads, identify the critical areas and predict the effects of various nutrient reduction strategies for the Upper Scioto Watershed by using Soil and Water Assessment Tool (SWAT). The Upper Scioto Watershed, located in central Ohio, is an agronomic watershed with mainly corn and soybean planted. SWAT model was run in the daily basis between 1993 and 2010 to simulate stream flow and nutrient flux in the Upper Scioto Watershed. Multiple sites are selected for model calibration and validation based on the data availability. The Nash-Sutcliffe coefficients (E) for daily, monthly and yearly stream flow are ranging from 0.6 to 0.9, from 0.8 to 0.9, and over 0.9, respectively. Nutrient calibration is based on crop yield calibration, with E ranging from 0.4 to 0.8, and regression analysis for limited number of measured nutrient loads. These E numbers along with the regression results indicate that the simulation output from SWAT model is satisfactory. SWAT model was then used to locate the critical source areas of nutrient loads at the HUC-12 level. Within the critical source areas, a list of conservation measures, including conservation tillage, filter strip, and nutrient management, was put on the model to predict nutrient load reduction and the water quality improvement at the watershed outlet. It is shown that the adoption of conservation measures to the critical source areas would significantly reduce nutrient loads to the water bodies.