Research Of Runoff Simulation In Ebinur Lake Basin Based On SWAT Model

In recent years, under the background of the national water shortage and theserious water pollution, the water environment problem is particularly acute in thenorthwest arid areas. The largest salt water lake of Xinjiang–The Ebinur Lake,serious deterioration of ecological environment, and the lake area is constantlyshrinking. Thereby affecting the entire basin’s ecological balance. Therefore, itrequires the use of hydrological models to quantify the analog quantity of water, toassist the rational use and research water resources in the Ebinur Lake Basin.This article selected the distributed hydrological model-SWAT model tosimulate runoff on the Ebinur Lake Basin. With the ArcGIS software platform toreclassify the land-use map and soil map of the study area in order to meet therequirements of the SWAT model input data, according to the local soil and watershed1961-2011meteorological data to establish the appropriate soil database andmeteorological databases.According to the Ebinur Lake Basin’s DEM map, land use map, the rating ofslope and soil map to divided the whole basin into25sub-basins and558HRUs, inorder to simulate the study area’s monthly runoff and annual runoff, and by inputWenquan hydrological station’s2004-2011monthly observations runoff to calibrateand validate the SWAT model.Analysis the results of calibration and verification: The model calibration period(in2004-2007) and verification period (in2008-2011), the annual runoff R~2was0.939and0.853; NSE coefficient was0.837and0.709; REwas-0.226%and-0.733%; theNSE coefficients of annual runoff in calibration and validation period were greaterthan0.7, the SWAT model to simulate the effect of good. The model calibration period(in2004-2007) and verification period (in2008-2011), the monthly runoff R~2was0.0.721and0.616; NSE coefficient was0.709and0.569; REwas0.489%and -0.728%; the NSE coefficients of monthly runoff in calibration and validation periodwere greater than0.5, R~2and REreached the SWAT model requires, so we considerthe SWAT model is applicable to the Ebinur Lake Basin runoff simulation.Based on the calibrated model, and according to climate change, analyze therelationship between runoff and temperature and precipitation by setting threedifferent climate scenarios. The results show that: the runoff is more sensitive toprecipitation change, a positive correlation between runoff and precipitation changes,the runoff is less obvious response to temperature change. Runoff was negativelycorrelated with temperature changes.According to landuse change trend set two different landuse scenarios of landusechange impact on the Ebinur Lake basin runoff. The results show that: for the aridareas, grassland vegetation coverage is low, evaporation strong, grass has the lowerrunoff and storing function than arable land; arable land has higher vegetationcoverage than the unused land, can better retain moisture, and to prevent soil erosion.Therefore, we should increase efforts to govern unused land, to protecte the arableland, and appropriate increase in arable land, and can improve soil and waterconservation capacity.