Analyzing Future Runoff Response To Climate Change In Biliu River Basin Based On HEC-HMS Model

In recent years,China has been increasingly affected by extreme weather and climate,and global climate change has become a research hotspot.Therefore,it is of great significance to study the change of future meteorological elements and the change characteristics of runoff,which can provide theoretical basis for relevant departments to make scientific decisions on water resource allocation,reservoir operation and ecological protection.In this paper,the biliuhe reservoir basin is taken as the research object.Firstly,the hec-hms model in the research area is constructed to simulate the runoff in the historical period,and the accuracy of the simulation results is evaluated.To build the SDSM model of research area,using the observed meteorological data and NCEP reanalysis data through the screening of prediction factors,to set up the forecast and the forecast factors statistical empirical relationship between the rate model,using Nash efficiency coefficient and relative error analysis model of simulation results,and based on SDSM model Can ESM2 model under three scenarios RCP2.6,RCP 4.5 and RCP 8.5 future rainfall downscaling simulation;Finally,the future precipitation results are input into the calibrated hec-hms model,so as to analyze the response of runoff in biliuhe reservoir basin under future climate change.The conclusions of this study are as follows:（1）During the base period（1978-2011）,both precipitation and runoff showed a downward trend,and the downward trend was significant.The annual decline rate of precipitation is 1.42 mm/a;The annual decrease rate of runoff is 8 million m³/s.Through the M-K trend test,it is found that the declining trend of runoff is larger than that of precipitation.（2）The simulation results of the HEC-HMS model were evaluated by the root mean square error,the relative error of the maximum flood peak flow,the relative error of the runoff depth,the relative error of the runoff flow,and the peak occurrence time.The pass rate of peak discharge is 0.77.The qualified rate of runoff is 0.86 and the qualified rate of annual discharge is 0.82,both of which meet the forecasting standard of grade b.（3）The accuracy of hec-hms model was evaluated by efficiency coefficient（DC）,qualified rate（QR）value and n_t value of nash-suttcliffe model.Where,DC and nt values indicate that in the 22-year periodic rate,6 years’prediction is unqualified,16 years’prediction is qualified,and its QR value is 0.73,reaching the accuracy standard of class b.（4）Statistical downscaling model SDSM establishes the statistical relationship between prediction factor and prediction quantity（P）in the research area.The model was evaluated by explaining variance and standard error,and then the model’s eligibility was analyzed by DC value and QR value.Its qualified rate is 0.77,which meets the forecast standard of class b.The results show that the model can be used for scale reduction simulation of large-scale meteorological factors.（5）Based on the verified SDSM downscaling model,three scenarios of RCP 2.6,RCP4.5 and RCP 8.5 in Can ESM2 model were simulated to obtain the daily precipitation data of nine rain stations under three future climate scenarios in the basin.The results showed that in the three scenarios of RCP 2.6,RCP 4.5 and RCP 8.5,the decreased water amount increased compared with the base period,and the added value of the average annual precipitation in the three scenarios was 166.3 mm,178.1 mm and 163.7 mm respectively compared with the base period.In the future,the declining water quantity in the three scenarios shows an overall trend of decrease,in which the annual average precipitation in different modes is 877.0 mm,888.8mm and 874.4 mm respectively,and the annual declining rate is 1.14 mm/a,0.49 mm/a and0.24 mm/a respectively.Under the three models,the annual precipitation peak is concentrated in the flood season,but with the increase of CO₂ emissions,the precipitation peak has a certain backward characteristic.The precipitation peak in RCP 4.5 was half a month later than that in RCP 2.6,and the precipitation peak in RCP 8.5 was half a month later than that in RCP4.5.（6）All precipitation station will HEC-HMS model of time sequence of daily rainfall input module,after verification of HEC-HMS model for the next three scenarios basin export daily traffic data,through the analysis of data obtained in RCP 2.6,RCP 4.5 and RCP8.5 three scenarios of runoff depth are increased,three scenarios in the future runoff deep the added value of the reference period were 225.39 mm,232.58 mm and 216.81 mm.Under the three scenarios of RCP 2.6,RCP 4.5 and RCP 8.5,the runoff increased compared with the base period,and the added value of RCP 2.6 was 241 million m³ compared with the base period.RCP 4.5 increased by 234 million m³ compared with the base period.The added value of RCP 8.5 was 225 million m³ compared with the base period.In the three scenarios,the future runoff presented a decreasing trend as a whole,in which the annual decline rate of RCP2.6 mode was 2.5 billion m³/a and the annual average runoff was 759 million m³.In RCP 4.5mode,the annual decline rate was 0.32 billion m³/a,and the average annual runoff was 752million m³.Under the mode of RCP 8.5,the annual decline rate was 0.6 million m³/a,and the average annual runoff was 743 million m³.Under the three scenarios,the peak occurrence time was similar to the precipitation,and all of them had the characteristics of backward movement.The peak occurrence time under RCP 4.5 was half a month later than that under RCP 2.6,but the peak occurrence time under RCP 8.5 was longer than that under RCP 4.5.