Runoff Simulation Using SWAT Model In The Middle Reaches Of The Shule River Basin

The hydrological model is a tool for the rational allocation of water resources. It playsan important role during the hydrological laws and production of practical problems. Inrecent years, distributed hydrological model is the focus and hot spots in the field ofHydrological Sciences Modeling, which describe the precipitation, model parameters, thespatial variation of the model structure and spatial variability of underlying surface and soilmoisture. It is an effective tool to study the hydrological cycle. The soil and waterassessment tool(SWAT) as a representative of the distributed hydrological model has a widerange of applications of runoff simulation, changes in watershed surface conditions,climate change response simulation, pollutant transport modeling and so on.With the rapid development of economic and social, there are increasingly seriouswater shortages especially in arid and semi arid areas. Northwest landscape is complex andlarge undulating terrain. The climate of the northwest is dry and the vegetation is low. Inthe trend of global warming, studying runoff variation has important implications andsignificance on grasping the runoff variation to prevent disasters. The ShuLe River basin isone of the three major rivers in Gansu Province, which is more than130,000squarekilometers. In recent decades, there is shrinking natural vegetation and serious deteriorationof grasslands in ShuLe River. Land desertification was a worsening trend. The excessivedevelopment of the middle reaches of water resources lead to environmental degradation ofdownstream ecosystems and exacerbating water supply and demand.So this paper selects the middle reaches of the Shule River basin as the study area anduses the distributed hydrological model of SWAT(Soil and Water Assessment Tool), whichis developed by the Agricultural Research Center of the United States Department ofAgriculture. On the basis of the collecting basic data of the watershed, we simulate therunoff in the middle reaches of Shule River in support of GIS and also simulate the effectof test to determine the feasibility of the SWAT model in the application of this basin,which provide an effective means of simulation, evaluation and analysis of water resourcesand lay the foundation for further development of the basin water quality modeling.(1) Firstly we discuss the background and significance of this research, then elaborateon the progress of distributed hydrological models and the SWAT model at home andabroad. We last discuss the technical route and the major research methods. (2) Analyzing and processing the basic data needed in the model runs including Landuse data, Meteorological data and Soil attribute data, which established the database of landuse, meteorology and soil properties. We divide the sub-basin and generate the network ofthe river to determine the watershed boundaries and hydrological response units. Then weextract some parameters which related to the topography with hydrology significance. Thearea of Shule River basin extracted from DEM was33885.09km2and the whole study areawas divided into55sub-watersheds,137hydrological response units.(3) We apply the existing data to simulate runoff and then use the measured runoff datato calibrate and validate the model, which determine its applicability in the study area.However, the uncertainty of parameters will impact the results of the model simulation, sowe need to calibrate the parameter and analysis the sensitivity. Using measured annualrunoff data of1988-2007was calibration and validation. The correlation coefficient R2was0.85and0.78, the Nash coefficient Ens was0.77and0.72, the relative error Re was5.71%and3.53%; using monthly runoff data of1988-1993was calibration and validation. Thecorrelation coefficient R2was0.82and0.86, the Nash coefficient Ens was0.75and0.82,the relative error Re was2.51%and5.33%, indicating the good precision.(4) Based on these findings, we establish different land use/land cover and climatechange scenarios by BNU-SWATV1.0, which is developed by Digital WatershedLaboratory in Beijing Normal University. At the same time, we investigate the hydrologicalresponse under the land use cover change and climate fluctuations in the middle reaches ofthe Shule River and then forecast the change trend of runoff. The results reflected that therunoff effect under precipitation change was larger than that under temperature change.They also two double impact the runoff. In the changes of land use type, the increase ofarable land may lead to reduced runoff. The effect of forest was better than grass forconservation of water and the mitigation of runoff. Increase in bare land cause runoffincreased and urban area increases and decreases was small the impact on surface runoff.(5) We analyze the results and summarize the main conclusions, innovation andshortcoming. Then we point out the issues which need to further explore in the follow-upstudies to expect more perfect results.