基于SWAT模型的观澜河流域城市面源污染负荷量化及影响效应评估

In recent years, with the improvements of sewage collection, interception systems in urban region of China and effluent standards of water purification plants, point source pollution has been gradually controlled, and the impact of urban non-point source pollution on water quality of river has becoming more and more significant. Aiming at the problems such as being difficult to quantify the current urban non-point source pollution load in river catchment areas of cities and to analyze the impact on water quality, Shenzhen Guanlan river basin was taken as the research object, site survey and investigation of land surface accumulation were performed in the river basin, then the land use type data of the river basin and the urban database of SWAT model were revised to build an urban non-point source pollution assessment model. The results showed that in the urban database of SWAT model, maximum accumulation of land surface sediments (DIRTMX), TN in land surface sediments (TNCONC), half saturation time of land surface dust accumulation (TNALF) and TP in land surface dust (TPCONC) were most sensitive to the simulation of TN and TP loads. Nash efficiency coefficients ( E _NS) of the built SWAT model for the runoff of the Guanlan river basin, TP and TN simulation verification periods were 0.79, 0.7 and 0.67, the corresponding coefficients of determination ( R ²) were 0.81, 0.77 and 0.81, respectively. The model simulation results were good. The built SWAT model was used to analyze the spatiotemporal distribution characteristics of urban non-point source pollution of TN and TP in the Guanlan river basin in 2018. The highest values of TN and TP export were 13.31~14.91 t·km⁻² and 1.69~1.86 t·km⁻². In August, rainfall runoff pollution in the Guanlan river basin was the most serious, and TN and TP loads were 2.15 t·km⁻² and 0.24 t·km⁻², respectively. The result can provide reference for load quantification and effect evaluation of regional urban non-point source pollution.