| Based on the ArcSWAT model,the surface runoff and non-point source NH3-N simulation of the Benxi section of the Prince River Basin were clarified,and the distribution of non-point source NH3-N pollution in the basin was clarified,and the management measures for prevention and treatment were proposed.The main research and conclusions are:(1)Construction of the ArcSwat model,applyed a global sensitivity analysis method,determining the more sensitive parameters in the runoff and NH3-N simulation;the calibration and verification of runoff and NH3-N by the SUFI-2 algorithm,the evaluation results of periodic and verification period show these correlation coefficient R2,Nash-Sutcliffe efficiency coefficient Ens and percent bias PBIAS are all standard,and the model is suitable for the research area.It provides simulation experience for SWAT simulation in other basins.(2)Analysis of the characteristics of non-point source NH3-N pollution load:non-point source NH3-N load reached maximum in 2010,minimum in 2017;non-point source NH3-N load is positively related to precipitation,strong rainfall data during August and September reached maximum.According to the research area 2018 non-point source NH3-N space distribution analysis:where higher load was distributed in 13 sub-basins such as 3,6 and 70,which was the northwest and east of the study area,which is more farmland.It provides a clear direction for local management of key ammonia-nitrogen polluted areas(3)Exploring the reduction of non-point source NH3-N reduction under different management measures in 2018,after implementing"returning farmland to forest",the reduction ratio was 25.5%,due to the impact of high plant coverage,the reduction effect is lower than the average level;After decreasing,the reduction ratio was 0.0087%,the reduction is very poor;"less fertilizer deep farm"management measures are not suitable for all research zones,the reduction ratio was 0.3856%,and the reduction is very poor;""vegetation buffer",cut the non-point source NH3-N load effect and the actual situation of the study area,3m of"vegetation buffer"width is more appropriate,and the reduction ratio was 51.34%;"returning farmland"and"vegetation buffer belt"integrated management measures reduction percentage was 65.64%.The best management practice is combined by"returning farmland"and"vegetation buffer belt".It provides a theoretical basis for local control of non-point source ammonia nitrogen pollution in watershed.This paper has 30 figures,20 tables and 118 references. |