EVALUATION OF NON-POINT SOURCE POLLUTION REDUCTION BY APPLYING BEST MANAGEMENT PRACTICES IN DONGLIAO RIVER WATERSHED USING SWAT MODEL

The purpose of this study is to identify the critical areas of Dongliao River watershed and evaluate the reduction effect of non-point source pollution by applying the best management practices (BMPs) using a physical process-based watershed scale model-SWAT (Soil and Water Assessment Tool). At first, the SWAT model's performance was tested for the watershed with observed data. Data from 2005 to 2007 was used for calibration while data from 2008 to 2009 was used for validation. The calibration and validation results indicated that the SWAT model was able to simulate well the stream-flow and non-point source pollutant loads, with coefficient of determination and Nash-Sutcliffe efficiency both about 0.7 for calibration and validation, respectively. Then, the model was used to identify the critical sub-basins and to assess the impacts of seven BMP scenarios including contour farming, conversation tillage, parallel terraces, filter strips, grassed waterways, grade stabilization structures, and residue management on river loads at the outlet. The results showed that among the seven BMPs simulated, filter strip had the best effects on the reduction of sediment. The reduction rate could reach 63.24%. The conversation tillage, parallel terrace, contour farming, filter strip and residue management all showed significant TN and TP load reduction, and the reduction rate could reach about 30% both for TN and TP loading. The grassed waterway had minimal benefit on TP and TN loads but had significant effect on sediment reduction. So, BMPs have significant effects on the reduction of non-point source pollutant loads and the improvement of water quality, and the results will provide a scientific basis for the protection and improvement of water quality in Dongliao River.