SWAT对大沙河水库流域径流模拟研究

Simulation of Runoff Parameters with SWAT in Dashahe Reservoir Watershed

广东开平大沙河水库为开平市主要供水源,其入库河流都受到不同程度的污染,水库呈富营养化状态需控制外源营养盐的输入。为了探究SWAT(Soil and Water Assessment Tool)模型在亚热带地区的适用性,为流域污染管理提供参考,运用ArcSWAT2009对大沙河水库流域进行了径流模拟和评价研究。用LH-OAT方法进行参数敏感性分析,找出对径流模拟影响较大的15个参数,采用2003—2012年的水库总入流数据对模型进行率定和验证。结果表明,LH-OAT敏感性分析方法可准确地找出模型的敏感参数;将时间尺度分别为日和月的率定和验证结果进行比较,月模拟效果明显较好,其与实测数据的相关系数R2分别为0.884和0.911,纳什系数(NSE)分别为0.879和0.907;模型的自动率定SCE-UA方法在大沙河水库流域的水量模拟中较适用,值得进一步研究应用,为水质水量管理工作提供参考。

Dashahe Reservoir is the main source of drinking water for Kaiping City, Guangdong Province. With increasing pollution of inflow waters arising from developments of animal farming, industry and agriculture, the reservoir is now eutrophic and the external loadings need to be reduced for water quality reg ulation. For testing its availability for subtropical area and water environment management, ArcSWAT2009 was applied to simulate and evaluate the river inflows of Dashahe Reservoir. LH-OAT method was used for sensitivity analysis. 15 parameters were sensitive to the simulation of river inflows. Auto-calibration and validation were conducted based on observed inflows data during 2003--2012. The results indicated that the LH-OAT method could effectively find out the sensitive parameters. The monthly simulations had a better result than the daily simulations, the monthly coefficients of determination （Re） were 0. 884 and 0. 911 and Nash-sutteliffes coefficients were 0. 879 and 0. 907, respectively, which suggests that SCE-UA is effective for SWAT calibration at Dashahe Reservoir catchment and SWAT model might be externally applied for water environment management in this area.