基于多源遥感数据的黄河数字流域模型改进  

Improvement of the Digital Yellow River Integrated Model based on multisource remote sensing

作　　者：薛源[1,2,3] 覃超 吴保生 张玍[1,2,3] 李丹 傅旭东 XUE Yuan;QIN Chao;WU Baosheng;ZHANG Ga;LI Dan;FU Xudong(State Key Laboratory of Hydroscience and Engineering,Tsinghua University,Beijing 100084,China;Key Laboratory of Hydrosphere Sciences of the Ministry of Water Resources,Tsinghua University,Beijing 100084,China;Department of Hydraulic Engineering,Tsinghua University,Beijing 100084,China;Emergency Science Research Academy,China Coal Research Institute,China Coal Technology&Engineering Group Co.,Ltd.,Beijing 100013,China)

机构地区：[1]清华大学水圈科学与水利工程全国重点实验室,北京100084 [2]清华大学水利部水圈科学重点实验室,北京100084 [3]清华大学水利水电工程系,北京100084 [4]中国煤炭科工集团有限公司煤炭科学研究总院有限公司应急科学研究院,北京100013

出　　处：《水利学报》2023年第8期930-941,共12页Journal of Hydraulic Engineering

基　　金：国家自然科学联合基金项目(U2243218,U2243222);清华大学水沙科学与水利水电工程国家重点实验室课题(2023-KY-02,sklhse-2023-B-03)。

摘　　要：山区河流水系众多,往往因地势险要难以现场量测获得河流基础信息,是典型的缺资料地区。本文结合多源遥感提取的68个河流断面及34个水文站实测断面,建立了黄河中游各级别河流的断面概化模型。融合传统线状DEM水系及河流表面、概化断面构建河道边界形态。针对黄河数字流域模型,基于不同级别河流的河道边界条件建立了断面判别模块,改进其汇流模型。以黄河一级支流皇甫川流域为例,以2010—2015年的资料率定模型,以2016年汛期日径流及年内3场典型洪水过程验证模拟效果。结果表明,改进后的模型输入参数均可由遥感观测获取,且日流量模拟的纳什效率系数高于0.8,流速及水深的模拟误差分别为7%、9%,全时期洪峰流量的模拟误差平均为15%、峰现时间误差约1.4 h,均较改进前明显提升。研究成果可为山区河流水文过程模拟及河流地貌演化等研究提供重要的基础数据和技术参考。Many rivers are developed in mountainous areas,but these rivers are difficult to measure on site due to the complex topography,and it is impossible to obtain sufficient basic river information,which makes mountain rivers a typical poorly gauged basin.This study combines 68 river cross sections extracted from multisource remote sensing data and 34 river cross sections measured by hydrological stations to establish a generalized model of river cross sections in the middle Yellow River.The generalized model of river cross sections,river surfaces and traditional DEM river networks forms a relatively complete river boundary geometric morphology of the entire watershed.Based on the river boundary geometric data,the confluence model of the Digital Yellow River Integrated Model(DYRIM)was improved.Subsequently,in the Huangfuchuan River basin,which is the first-order tributary of the Yellow River,based on the improved DYRIM,we simulated and verified the daily runoff process in the 2016 flood season and 21 typical flood processes from 2010 to 2016 after using data from 2010 to 2015 to calibrate the model.The results show that the input parameters of the improved model can be obtained from remote sensing observations,and the simulated Nash-Sutcliffe efficiency coefficient of daily discharge is higher than 0.8.The simulated errors of flow velocity and water depth are 7%and 9%respectively,the error of peak discharge of flood is 15%on average,and the error of peak present time was approximately 1.4 h,all of which were significantly improved compared to before.This research can provide basic data and technical references for the simulation of hydrological processes of mountain rivers and research on river geomorphology evolution.

关 键 词：山区河流 多源遥感 河流断面 黄河数字流域模型 

分 类 号：TV211[水利工程—水文学及水资源]