Study On Medium And Long-term Runoff Forecast Of Chengbi River Basin Based On MSWX Precipitation Products

There is a serious shortage of water resources in our country,and the contradiction between supply and demand of water resources is prominent.The medium and long-term runoff forecast is of guiding significance for alleviating the contradiction between supply and demand of water resources and optimizing water resources scheduling decisions.The development of high-resolution seasonal scale precipitation forecast products provides data support for medium and long-term runoff forecast by process-driven model method.Therefore,this paper takes Chengbi River Basin in Guangxi as the study area,and uses MSWX precipitation forecast data to drive SWAT model to carry out medium and long-term runoff forecast research.The main research contents and results of this paper are as follows:(1)Based on the relative deviation BIAS,critical success index CSI and other indicators,the accuracy of MSWX precipitation forecast products in 120seasonal scale precipitation forecasts was evaluated.The results show that the accuracy of the five modes data of MSWX precipitation forecast products is better in summer and autumn than in spring and winter.BIAS is up to 2.87 in spring and winter,and generally below 0.5 in summer and autumn.CSI ranges from 0.00 to 0.36 in spring and winter,and increases to 0.26 to 0.63 in summer and autumn.Based on the distribution of all indicators,the accuracy of the 00mode of the MSWX precipitation forecast data is highest,and it is more suitable as the precipitation input data of hydrological model for medium and long-term runoff forecast in the basin.(2)Based on the relevant data of Chengbi River Basin,the SWAT hydrological model was constructed,and the applicability of the model in the basin was analyzed according to the determination coefficient R~2,the relative deviation BIAS and the Nash efficiency coefficient NSE.The results show that the SWAT hydrological model is applicable in Chengbi River Basin,and the runoff simulation results of each station reach qualified grade or above.In Xiajia Station,NSE is 0.61 in the calibration period,and is 0.68 in the validation period.In Pingtang Station,NSE is 0.80 in the calibration period,and is 0.79 in the validation period.In Bashou Station,NSE is 0.76 in the calibration period,and is 0.71 in the validation period.On the seasonal scale,the runoff simulation accuracy of the three stations in spring is lower than that in summer,autumn and winter.(3)Based on MSWX 00 precipitation forecast data and the optimal parameter SWAT model,the medium and long-term runoff was forecasted in Chengbi River Basin,and the runoff results were compared with the site data simulation scenario and the lag forecast scenario.The results show that the runoff forecast based on MSWX precipitation forecast data can reflect the trend of runoff increase or decrease.In the initial stage of the forecast,MSWX forecast runoff has a good fitting degree with the station simulation runoff,and in the middle and late stage of forecast,the deviation is more serious.With the increase of the forecast period,the determination coefficient R~2 of MSWX forecast runoff gradually decreases,and when the forecast period exceeds 7 days,the deviation of forecast runoff results is large,so it is necessary to further improve the forecast accuracy if it is used as a reference for reservoir dispatching decisions.