基于SWAT模型耦合精细化格点降水预报产品的径流预测研究

Study on runoff prediction based on SWAT model coupled with refined grid precipitation forecast product

为探究复杂地形下水文预报方法,以贵州省平寨流域为例,采用阳长站2017~2019年逐日和2009~2018年逐月实测径流量数据,利用分布式SWAT水文模型进行建模、率定和验证,并耦合2021年6~9月流域气象关键点降水预报数据开展入库流量预报,最后采用水文模型评价指标分析了径流量预报效果。结果表明:SWAT模型在贵州省平寨流域有较好的适用性,能够较好地模拟实际径流量变化趋势与洪峰过程,率定期和验证期R 2和NSE系数均高于0.69;预报降水未订正时,预报入库流量在6~7月基本达标,但是8~9月效果不佳;预报降水订正后,预报入库流量在6~9月的NSE系数和R 2明显提高。

In order to explore hydrological forecasting methods under complex terrain,this paper used the daily measured runoff data of Yangchang Station from 2017~2019 and monthly measured runoff data from 2009~2018,and used the distributed hydrological model SWAT to simulate,calibrate and verify the hydrological cycle process of Pingzhai Basin,in order to provide technical reference for the reasonable operation of hydropower stations in Pingzhai Basin.The results showed that the model can simulate the change trend of actual runoff,and the simulated daily and monthly peak discharge process was consistent with the actual process,the R 2 and NSE of the regular rate and verification period were higher than 0.69.The model had good applicability in Pingzhai Basin.The calibrated SWAT model was used to couple the meteorological forecast data from June to September 2021 in Pingzhai Basin,and the runoff change in this period was simulated.When the predicted rainfall was used to simulate the runoff change,the requirements were basically met from June to July,but the simulation effect was not good from August to September.After the rainfall was revised,the runoff change was simulated again.The NSE coefficient and R 2 in June to September were significantly improved compared with those before the rainfall revised,and the predicted rainfall in several flood peak processes would also be closer to the actual situation.Therefore,using SWAT data coupled with rainfall forecast to forecast the inflow of Pingzhai reservoir on the basis of successful hydrological model modeling,improving the accuracy of rainfall forecast is the key factor to improve the accuracy of inflow forecast.