Font Size: a A A

Test And Analysis Of Precipitation Ensemble Prediction And Its Application Study In Flood Prediction

Posted on:2020-01-14Degree:MasterType:Thesis
Country:ChinaCandidate:Y ZhangFull Text:PDF
GTID:2370330590958530Subject:Hydraulic engineering
Abstract/Summary:PDF Full Text Request
As the basic link of non-engineering measures in flood control and disaster alleviation system,timely and accurate flood prediction is very important.Aiming at extending the flood prediction period and ensuring the accuracy of flood prediction,the ensemble numerical weather prediction technology is introduced into the Jinqian River Basin as the research object.In the study of Hydrometeorological coupling,the precipitation ensemble prediction system under different initial disturbances and different model conditions is tested and analyzed,and the precipitation ensemble forecast data is revised to ensure the accuracy of flood forecasting while prolonging the forecasting period.The main contents and results of this study are as follows:(1)Based on the measured precipitation data,the Xin'anjiang model of Jinqian River Basin is constructed.Flood simulation results are obtained by parameter calibration using SCE-UA algorithm and verified by the flood in the validation period.The simulation results show that the Xin'anjiang model with three water sources has good applicability in the study of flood forecasting in the basin.(2)The precipitation ensemble forecast lattice data in the study basin are transformed into station rainfall and surface rainfall in the basin by using point data and surface data estimation method.The results of test and evaluation show that each estimation method has certain applicability,and the improved inverse distance weighted averaging method has better effect.(3)The deterministic prediction,probability prediction and reliability of ECMWF,NCEP,CMA,BoM,ECCC and UKMO ensemble prediction systems are comprehensively evaluated and analysised by TS,Brier and Talagrand distribution methods at two levels of point rainfall and surface rainfall.The evaluation results show that each ensemble prediction system has certain advantages for different precipitation levels,and NCEP has a better forecasting effect for the study basin as a whole.(4)The control member prediction and ensemble member prediction of NCEP precipitation ensemble prediction system are used to drive the hydrological model.At the same time,the precipitation ensemble forecast revised by the ensemble average method,the super ensemble average method and the super ensemble correction method based on different precipitation grades are used to forecast flood respectively.The simulation results show that the revised methods have improved the accuracy of flood prediction to varying degrees.
Keywords/Search Tags:Flood prediction, Hydrometeorological coupling, Ensemble numerical weather prediction, Xin'anjiang model, Correction method
PDF Full Text Request
Related items