基于CMADS+SWAT驱动的新疆甘河子河流域水文过程模拟及径流成分研究

柴文光; 杨广; 田浩; 李小龙; 张衡; 徐文韬; 任杉; 张诚; 倪灿帮

doi:10.13928/j.cnki.wrahe.2026.04.001

水利水电技术（中英文） ›› 2026, Vol. 57 ›› Issue (4) : 1 -12. DOI: 10.13928/j.cnki.wrahe.2026.04.001

复杂灾害链与干旱巨灾风险评估专栏

基于CMADS+SWAT驱动的新疆甘河子河流域水文过程模拟及径流成分研究

柴文光 ¹^,² ,
杨广 ¹^,² ,
田浩 ¹^,² ,
李小龙 ¹^,² ,
张衡 ¹^,² ,
徐文韬 ¹^,² ,
任杉 ³ ,
张诚 ³ ,
倪灿帮 ³

作者信息 +

Hydrological process simulation and runoff component analysis of Ganhezi River Basin, Xinjiang based on CMADS+SWAT

Wenguang CHAI ¹^,² ,
Guang YANG ¹^,² ,
Hao TIAN ¹^,² ,
Xiaolong LI ¹^,² ,
Heng ZHANG ¹^,² ,
Wentao XU ¹^,² ,
Shan REN ³ ,
Cheng ZHANG ³ ,
Canbang NI ³

Author information +

文章历史 +

PDF (16219K)

摘要

【目的】由于融雪、降雨入渗等因素综合影响，寒区流域水文过程复杂。西北高寒山区水文气象站点稀少，数据缺失制约着寒区径流模拟分析，气象数据成为我国西北高寒山区水文规律研究的主要瓶颈。【方法】针对我国西北高寒山区水文规律研究过程中气象数据缺失问题，以新疆甘河子河流域为研究区域，结合SWAT中国大气同化驱动数据集(CMADS)探讨了CMADS+SWAT对高寒山区径流模拟适用性并分析了流域不同径流成分产水量，运用SWAT-CUP进行模型参数的敏感性分析、率定和验证，以决定系数(R²)和纳什效率系数(NSE)评价SWAT模型的适用性。【结果】研究表明：(1)运用CMADS+SWAT模型在月尺度能够较好的模拟甘河子河流域径流过程，决定系数(R²)和纳什效率系数(NSE)在率定期和验证期分别为0.87、0.72和0.84、0.64;(2)不同径流成分中壤中流多年平均产水量占比最大，为43.7%,地表径流产水量占比次之为43.3%,地下水产水量占比最小仅为13%。【结论】运用CMADS+SWAT模型在月尺度能够较好的模拟甘河子河流域径流过程，研究结果可以为我国西北高寒山区径流模拟与分析提供基础依据。

Abstract

[Objective] The hydrological processes in the river basins of cold regions are complicated due to the combined effects of snowmelt, rainfall infiltration, and other factors. Hydrometeorological stations in the alpine mountains of Northwest China are scarce, and the lack of data limits the simulation and analysis of cold-region runoff. Meteorological data has become the main bottleneck in studying hydrological processes in the alpine mountains of Northwest China. [Methods] To address the lack of meteorological data in studying hydrological processes in the alpine mountains of Northwest China, the Ganhezi River Basin in Xinjiang was selected as the study area. Using the China Meteorological Assimilation Driving Datasets for the SWAT model(CMADS), the applicability of CMADS+SWAT for runoff simulation in alpine mountains was explored, and the water yield of different runoff components in the river basin was analyzed. SWAT-CUP was used to perform sensitivity analysis, calibration, and validation of model parameters, and the coefficient of determination(R²) and Nash-Sutcliffe efficiency coefficient(NSE) were used to evaluate the applicability of SWAT model. [Results] The result showed that:(1) the CMADS+SWAT model could effectively simulate the runoff process of the Ganhezi River Basin at the monthly scale, and R² and NSE were 0.87 and 0.72 during the calibration period, and 0.84 and 0.64 during the validation period, respectively.(2) Among the different runoff components, interflow had the highest multi-year average water yield proportion at 43.7%, followed by surface runoff at 43.3%, and groundwater had the smallest proportion at only 13%. [Conclusion] The CMADS+SWAT model can effectively simulate the runoff process in the Ganhezi River Basin at the monthly scale. The findings can provide a basis for runoff simulation and analysis in the alpine mountains of Northwest China.