基于SWAT模型的龙溪河泸县境内面源污染特征分析

An Analysis of the Characteristics of Non-point Source Pollution of Long-xi River Basin in Luxian County Based on SWAT Model

应用分布式水文模型SWAT,首次以龙溪河泸县境内的面源污染为研究对象,通过对该流域的面源污染进行调查、监测和模拟,定量计算和分析该流域面源污染的时空分布规律以及识别关键污染源和污染关键区。结果表明:①SWAT模型在该研究区具有较好适用性,径流、氮磷负荷的纳什效率系数Ens在0.6~0.75,决定系数R^2在0.75~0.85,相对误差Re在10%~30%;②年产水量、TN产量、TP产量丰水期显著多于平水期,丰水期比重大约为70%~80%,平水期多于枯水期,枯水期产量比重在5%以内;③面源污染是各个子流域最主要的污染类型,农业化肥污染与农村生活污染在面源污染中所占比重最大;④农村生活污染对TN产量影响最大,农业化肥污染对TP产量影响最大;⑤11、13号子流域是非点源污染的关键区,选定云锦镇为重点防治区。

The distributed hydrological model SWAT is firstly applied in Longxi River Basin in Luxian County as a typical area in this paper.Through the investigation, monitoring and simulation of the non-point source pollution in this basin, it has calculated quantitatively and analyzed the spatial and temporal distribution law of non-point source pollution, as well as the identification of the key pollution sources and the key pollution areas.The results show that:① The SWAT model is highly applicable in the research area.The Nash efficiency coefficient of runoff and nitrogen and phosphorus load is between 0.6 and 0.75, and the determination coefficient is between 0.75 and 0.85, and the relative error is between 10% and 30%.② The annual output of water, TN and TP are significantly higher in the wet period than in the level period, as well as in the level period than in the dry period.The yield in the wet period accounts for about 70%-80%, and the yield ratio in the dry period is less than 5%.③ The non-point source pollution is the most important type of pollution in each sub-basin, among which the agricultural fertilizer pollution and the rural living pollution account for the largest proportion.④ The rural living pollution has the largest impact on TN yield, and The agricultural fertilizer pollution has the largest impact on TP yield.⑤ Yunjin Town is selected for prevention and control emphatically, as the sub-basin 11 and sub-basin 13 are the key areas of non-point source pollution.