气候变化对元江流域水文过程的影响  

The Impact of Climate Change on Hydrological Processes in Yuanjiang River Basin

作　　者：葛路 詹良通[2] 江衍铭 GE Lu;ZHAN Liang-tong;CHIANG Yen-ming(Institute of Hydrology and Water Resources Engineering,Zhejiang University,Hangzhou 310030,China;Institute of Geotechnical Engineering,Zhejiang University,Hangzhou 310030,China)

机构地区：[1]浙江大学建筑工程学院水文与水资源工程研究所,杭州310030 [2]浙江大学建筑工程学院岩土工程研究所,杭州310030

出　　处：《中国农村水利水电》2022年第4期41-49,56,共10页China Rural Water and Hydropower

基　　金：国家自然科学基金国际(地区)合作与交流项目(NSFCTRF:41961144018);浙江省重点研发计划择优委托项目(2021C03017);浙江省属基本科研业务费专项资金资助(2021XZZX015)。

摘　　要：为探讨气候变化对元江流域水文气象的可能影响,并为防洪减灾提供参考依据,本研究基于CMIP6(Coupled Model Intercomparison Project Phase 6)的5个GCMs(Global Climate Model)模型,通过M-BCSD(Monthly Bias-Corrected Spatial Disaggregation)降尺度方法,获得元江流域2041-2100年在SSP126、SSP245、SSP370、SSP585四种路径下的月降雨和月气温,并利用交叉验证于倒传递神经网络(Back-Propagation Neural Network,BPNN)预测气候变化对流域出口断面径流的影响。研究结果表明:M-BCSD方法在元江流域适用性好;建立的BPNN可有效预测元江流域月径流;总体而言,元江流域在2041-2100年的年均降雨量、温度和径流都增加,最大增幅分别为24.6%、20.4%、10.2%,就季节尺度而言,春季径流减少而秋季径流增加,月尺度下,4月径流减少而11月径流增加;不同路径情景下未来水文气象要素变化一致,仅幅度有所差别。To explore the possible impact of climate change on the hydrometeorologicalprocesses and provide the reference for flood prevention and disaster reduction in Yuanjiang River Basin,this paper obtains the outputs of five GCMs announced by CMIP6 from 2041 to 2100.Then,monthly precipitation and temperature under SSP126,SSP245,SSP370 and SSP585 paths are generated through the M-BCSD downscaling method. Moreover,the impact of climate change on the runoff at the basin outlet is forecasted by back-propagation neural network(BPNN)with cross validation skill. The results indicate that the M-BCSD method has better applicability in Yuanjiang River Basin and the BPNN is able to predict the monthly runoff effectively. In general,the mean annual rainfall,temperature and runoff have an increasing trend within the periods of 2041-2100 with the maximum increase of 24.6%,20.4% and 10.2% respectively. As for the seasonal scale,the runoff decreases in spring and increases in fall. As for the monthly scale,the runoff decreases in April and increases in November. Under different path scenarios,the changes of the hydrometeorological elements are basically consistent but their amplitudes are different.

关 键 词：降尺度 全球气候模式 交叉验证 倒传递神经网络 径流模拟 

分 类 号：TV213[水利工程—水文学及水资源] P339[天文地球—水文科学]