Research On Non-Point Source Pollution Management Measures Of Zhuxi River Basin Based On SWAT Model

China ’ s watershed water management basically realizes the prevention and control of point source pollution,but non-point source pollution needs to be solved urgently.Compared with point source,non-point source pollution has the characteristics of wide source range,strong randomness,and difficult real-time monitoring and management.Best Management Practices(BMPs)have become one of the most effective non-point source pollution control measures,which can prevent or reduce water pollution.Based on SWAT model,this paper analyzes the characteristics of nonpoint source pollution in the Pearl River Basin,analyzes the spatial and temporal distribution of pollutants(N,P,sediment,etc.)in different land use,and uses spatial interpolation to obtain the spatial distribution of river inflow coefficients,so as to select typical small watersheds for measures.The effects of single and combined measures on pollutant reduction were analyzed,and the BMPs benefits of each measure were evaluated.The measures with environmental and economic benefits were proposed,which provided a theoretical basis for the prevention and control of non-point source pollution in the Pearl River Basin.The main research results are as follows :(1)Explore the applicability of SWAT model in the study area.Through field investigation and data collection,the data of DEM,land use,meteorology,farmland management,soil types and their properties in the Zhuxi River Basin are sorted out.The study area was divided into 33 sub-basins and 203 hydrological response units.Based on the monitoring data of Yuexi Station in the lower reaches of the Pearl River from 2010 to 2018,the SWAT model is calibrated and verified.The calibration and verification results meet the requirements of R2 > 0.6 and Ens > 0.5,indicating that the model meets the applicable requirements of the study area.(2)Analyzing the characteristics of non-point source pollution under different land use conditions.According to the remote sensing image,the land use in 2010 and2018 was obtained by using ENVI 5.3 software for supervised classification,and the simulation analysis was carried out based on the two periods of different land use.The distribution of sediment load is similar to that of water production,mainly concentrated in areas with large water production.It is mainly distributed in 12,18 subbasins in the east and 26,32 subbasins in the southeast.The results of land use sediment load in the two periods are mainly different in sub-basins 1,17,16,19 and 21.The total nitrogen load was mainly concentrated in the sub-basins 1 and 3 in the north,14 and 18 in the middle and east,and 23 and 25 in the southwest.The total nitrogen load of land use in the sub-basins 1 and 15 was quite different.The intensity of total phosphorus load was higher in 14,18,25 and 26 sub-basins in the middle east,but lower in 4,5,6 sub-basins in the north and 15,16 and 21 sub-basins in the west.The total phosphorus load of land use was mainly different in 1,15 and 16 sub-basins.The trend of total nitrogen load was consistent with rainfall and sediment load in the early stage(2010-2016),and negatively correlated in the later stage(2016-2018).The trend of total phosphorus load was consistent with rainfall and sediment load.(3)Scenario analysis was used to evaluate the reduction effect of BMPs on pollutants.Through the statistical output results,the river inflow coefficients of each subbasin outlet are obtained,and the Kriging spatial interpolation is carried out to obtain the spatial distribution of river inflow coefficients in the whole study area.Combined with the spatial distribution of pollution load,the typical sub-basins are selected.Five single measure scenarios and four combined measure scenarios are simulated from nonengineering measures and engineering measures,and the SWAT model is used to evaluate the effect.The results showed that engineering measures such as vegetation buffer zone(Fs)and grass-planting river had better pollutant reduction effect,and the reduction rate reached 50 %-70 %,while non-engineering measures such as chemical fertilizer reduction(FR)and no-tillage(NT)had limited pollutant removal effect less than 10 %.At the same time,it was found that combined measures had better pollution control effect than single measures,and FR20 + NT + Fs had the best effect.(4)Benefit evaluation of BMPs.The cost-benefit ratio(CE)and cost-benefit ratio(CB)were mainly used to evaluate BMP measures.The results showed that engineering measures such as vegetation buffer zone and grass-growing river had good environmental and economic benefits,while the CE and CB values of non-engineering measures such as chemical fertilizer reduction and no-tillage were significantly lower than those of engineering measures.