Study On Runoff Forecasting Of Honganjian River Basin Based On TIGGE Data

With the progress of human beings and the development of society,water resources play a more and more important role in economic development and social stability.However,due to the unreasonable use of human beings,the amount of water resources is gradually decreasing.The serious pollution of water resources leads to the gradual deterioration of the relationship between human and water resources,which is not conducive to the sustainable development of ecology.Runoff prediction plays an important role in improving the utilization rate of water resources,the rational utilization and exploitation of water resources,and the improvement of water environment quality.This paper takes the Honganjian River basin as the research object.Firstly,the rainfall set forecast data published by ECMWF and UKMO meteorological center in TIGGE data are evaluated.Based on the rainfall forecast information and Xinanjiang model,the runoff prediction of the basin is carried out.The results of the study are as follows:(1)The TS score,BS score,Brier score and Talagrand distribution were used to evaluate the rainfall ensemble forecast data published by ECMWF and UKMO centers.The results showed that,compared with the control forecast,the effect of the set forecast was better than that of the control forecast as a whole.With the extension of the forecast period,the forecast accuracy of the forecasting center tends to decrease,the short-term UKMO,the short-term rainfall ensemble forecast and the short-term set forecast of ECMWF are not enough members to diverge,the long-term set forecast of UKMO and the medium of ECMWF,There is systematic deviation in long-term ensemble prediction.(2)The precision analysis of the ensemble average forecast data of UKMO and ECMWF shows that with the increase of the rainfall level and the extension of the forecast period,the accuracy of the forecast is gradually reduced,the accuracy rate of the no-rain forecast is above 80%,and the accuracy rate of the first 6 days in the small rain forecast is above 50%.The accuracy of the rainfall collection forecast information is more accurate,can be used in the runoff forecasting,and the ECMWF prediction effect is better than that of the UKMO.(3)The Xinanjiang model is chosen as the runoff forecast model and the model parameters are calibrated with the data from 1991 to 2003.The results show that the simulation results are ideal and the qualified rate is more than 80%,and the simulation accuracy is verified by the data from 2004 to 2013.The deterministic coefficient DC and the correlation coefficient R2 are above 0.8,the fitting degree is high,and the model has good applicability.(4)The aggregate rainfall forecast information and average rainfall forecast information released by ECMWF in 2010 flood season are used to drive the Xinanjiang River model to simulate the runoff process.The results show that the aggregate runoff forecast is similar to the measured runoff,and the results show that there is a certain similarity between the aggregate forecast runoff and the measured runoff.With the extension of forecast period,the effect of runoff forecast using rainfall ensemble forecast will decrease gradually.The determinacy coefficient DC and the correlation coefficient R2 obtained from the 5-day and 10-day set average rainfall forecast are above 0.7,indicating that the runoff forecast can be carried out by using the rainfall set forecast.