Simulation Of Non-point Source Pollution In Karst Region Based On Modified SWAT Model

With the in-depth study and effective control of point source pollution,non-point source pollution simulation in watershed has become a hot research topic in recent years.As a typical ecologically fragile area,compared with non-karst area,the development of economy and society in karst area aggravates the shortage of water resources supply and demand.However,land use change caused by human activities and global climate change are the main driving factors affecting hydrology and water resources.Therefore,the study of non-point source pollution load in karst areas and the impact of land use change and climate change on non-point source pollution load in watershed is of great significance to the healthy development of water cycle system and water environment protection in karst areas,and can provide effective guarantee for realizing sustainable utilization of water resources.From the perspective of karst geomorphic features,this paper combined RS,GIS,SWAT model and NCC/GU-WG,and took Tongzi River,the largest tributary of Chishui River,as the research area,introduced the hydrological processes and non-point source pollutant transport processes of the fall hole and underground river,and modified the original hydrological cycle process and related algorithms based on SWAT open source code.Combined with field hydrology and water quality data,non-point source pollution simulation in karst area is taken as the research object,and a non-point source pollution model suitable for karst basin is established.The modified SWAT model is applied to study the impact of land use change and climate change on non-point source pollution load in karst area.The NCC/GU-WG weather generator is used to predict the trend of non-point source pollution load under future climate change scenarios,and the main conclusions were as follows:（1）Land surface temperature was retrieved from Landsat8 remote sensing images,vegetation coverage was extracted,and the first principal component P_FLCwas extracted by principal component analysis combined with cirrus band.The feature space was established between P_FLCand NDVI,and the karst specific gravity index of vegetation in the study area was obtained through calculation.Combined with the hydrogeological map of the study area,it was found that the method of extracting karst information of karst basin by using karst specific gravity index of vegetation was reasonable.Finally,6 karst points of slocker and 23 karst points of underground river were extracted in the sub-basins 13,17,34,36 and 39.（2）The R²and NS coefficients of runoff simulation were improved before and after the model modification,but the R²and NS coefficients of water quality simulation were not changed much before and after the model modification.The simulated results of total nitrogen were closer to the observed values at the flood peak（June and July 2020）.The revised simulated surface runoff and water yield showed a decreasing trend,especially at the flood peak,while the underground runoff increased significantly.The revised pollution load of total nitrogen and total phosphorus increased,indicated that the revised model could more accurately describe the phenomenon of runoff reduction and non-point source pollution load increase caused by the increase of karst variables in karst areas.（3）The overall coverage areas of forest and grassland in the study area were large,and the ability of water and soil conservation was strong.The insignificant impact of land use change on non-point source pollution load of the basin was also small.The impact of climate change on hydrology and water quality of watershed was mainly manifested in precipitation and temperature.Under the climate change scenario,the area with large surface runoff was mainly concentrated in the middle of the basin,and the area with large underground runoff was in the lower reaches of the basin.The distribution of sub-basins with high water yield was consistent with that of runoff,and most of them were distributed in the area with high precipitation in the south.The distributions of total nitrogen and total phosphorus from non-point source pollution were consistent with that from runoff.The spatial distribution trend of each factor under the co-change scenario was generally consistent with that under the climate change scenario,indicated that under the current land use pattern,the insignificant change of land use type had little impact on the hydrological and water quality processes of the basin,while the climate change had a greater impact on the hydrological and water quality processes of the basin.（4）NCC/GU-WG weather generator was used to predict the climate change in tongzi River Basin from 2030 to 2050,and the relative error and NS coefficient were used to evaluate the accuracy of the simulated value and measured data of the weather generator.The results showed that the NS coefficients of daily precipitation,maximum temperature and minimum temperature simulated by the weather generator were all above 0.95.The relative errors were between 3.45%-5.03%,and the simulation results were reliable.The simulation of future climate change showed that annual precipitation increased from 2030 to 2050,but temperature increased slightly.The simulation effect of weather generator showed that precipitation was greater than temperature.Under the future climate change scenarios,runoff would increase obviously.In 2038 and 2049,the average annual surface runoff,average annual underground runoff,total phosphorus and total nitrogen showed a significant increase trend,with the average annual surface runoff increased by 137.86 mm and 133.83 mm,and the average annual underground runoff increased by 275.42 mm and 230.28mm.The annual increases of total nitrogen were 3.957 million tons and 5.201 million tons,while the annual increases of total phosphorus were 1.862 million tons and 2.114million tons,with the largest increase in 2049.Under the future climate change scenarios,runoff,total nitrogen load and total phosphorus load of the basin would generally increase,and they were mainly concentrated at the outlet of the basin.