Non-point Source Pollution Simulation Of Multi-spatial Watershed Scale And Its Response To Climate Change

Non-point source pollution has become the main source pollution in China,and due to the over-exploitation and destruction of water resource,it has become more and more serious.In addition,climate warming has significantly changed the temporal and spatial distribution of the water cycle,leading to frequent extreme rainfall-runoff events,thus driving the changes of the non-point source pollution in the basin areas.More and more pollutants migrate with the rainfall-runoff processes,aggravating the deterioration of water quality in rivers,lakes,reservoirs and other water bodies,seriously threatening the water security and ecological environment,and even affecting the sustainable development of human and nature.Therefore,it is of urgent practical significance to carry out simulations of non-point source pollution in the river basin and explore the impact of climate change on it,which can not only theoretically improve the methodology of the response of non-point source pollution to climate change,making the research more scientific and rich,but also provide reliable support for the refined prevention and control of non-point source pollution under climate change.Based on the simulation and evaluation of non-point source pollution at multiple spatial scales,including the typical watershed,reservoir area watershed and Feilaixia catchment,this study uses the regional climate model to carry out the research about the impact of climate change on non-point source pollution.The main conclusions are as follows:（1）Using the simultaneous monitoring results of rainfall-runoff and water quality in the typical watershed,the characteristics between rainfall-runoff and non-point source pollution were explored in the Beijiang River Basin.Based on the average concentration method,the average concentration of pollutants in each rainfall event was approximated as the pollutant concentration of a typical day.Therefore,the scale mismatch problem between the rainfall events and the daily scale of the SWAT model was solved.Then,the calibration and validation of the SWAT model were carried out.The results show that soluble substance such as NH₃-N and TN has a very obvious initial flushing effect.The SWAT model has a good simulation effect on the runoff process and pollutant loads in the typical watershed.The NS coefficients of the runoff simulation in calibration and validation period are 0.85 and 0.73,respectively,and the relative errors are-5.28%and-14.28%,respectively.In the non-point source pollution simulation,the SWAT model has the best effect for a typical day with high rainfall intensity,with the relative errors within 10%,while the model performs worse when the rainfall intensity is low.（2）Aiming at the non-point source pollution in the Feilaixia reservoir area,a comprehensive evaluation was carried out using the equivalent pollution load method and cluster analysis method to obtain the load ratio of each pollution source in the reservoir area and the degree of non-point source pollution in each township.The results show that the load ratios of the four pollution sources are ranked as domestic sewage（57.79%）>livestock and poultry breeding（38.13%）>planting（2.38%）>aquaculture（1.69%）.The pollutant load ratios of the 10 towns are ranked as Shihuipu Town>Dazhan Town>Yinghong District>Wangbu Town>Lixi Town>Xiniu Town>Lianjiangkou Town>Shuibian Town>Feilaixia Town>Yingcheng Town.Among them,the load ratios of Shihuipu and Dazhan Town are both close to 20%.The cluster analysis also show that pollution degree of the towns in the upper reaches of the reservoir area is very high,such as Dazhan Town,Shihuipu Town,Wangbu Town and Yinghong District,whose non-point source pollution should be prevented and controlled.（3）Based on the SWAT model,the daily and monthly runoff was simulated and the non-point source pollution was analyzed to explore the temporal and spatial distribution characteristics of the non-point source pollution in the Feilaixia catchment.Combined with the environmental quality standards of surface water,the priority control areas for non-point source pollution were effectively identified.The results show that the NS coefficients of daily scale runoff at the catchment outlet are 0.85 and 0.77 in the calibration and validation period,respectively,and as high as 0.96 and 0.80 at the monthly scale,reflecting the high accuracy and high adaptability of the SWAT model,so as the performance of the pollutants simulation.The NS coefficients of NH₃-N at the outlet are 0.775 and 0.633 during the calibration and validation period,and that of the TN are both 0.589,the relative errors of both NH₃-N and TN can be controlled within 25%.The areas with high NH₃-N concentration are mainly located in the main river areas with high population density and large domestic sewage discharge;while the areas with high TN concentration are located in areas with a lot of arable land and a large amount of fertilization,such as the Wushui catchemnt,the upper Zhenshui catchment,the upper Lianjiang catchment,and the middle Wengjiang catchment,showing strong regional differences.Seasonally,the pollution degree of NH₃-N and TN are in the same order as spring>winter>autumn>summer.The highest level control areas are located in the upper Wushui catchment,the eastern part of the upper Lianjiang catchment and the upper Zhenshui catchment.The first-level control areas are located in the lower Wushui catchment,Nanshui catchment and the middle to lower Lianjiang catchment.The secondary control areas are located in Wengjiang catchment,central part of the whole catchment and Jinjiang catchment.（4）The improved quantile mapping method was used to correct the output rainfall of the regional climate model,including the correction of the number of wet days and the rainfall frequency.The changes of extreme rainfall-runoff events in the catchment were explored.The results show that the improved quantile mapping method has a very significant effect on the bias correction of the RegCM output,effectively eliminating the false micro rainfall,and also enhancing the ability of GCMs to capture the precipitation characteristics of the catchment.Additionally,it is expected that extreme precipitation events in the catchment will increase in the future,as can be seen from the performance of extreme precipitation indices,such as RX5day and SDII under RCP4.5 scenario in the mid-to-long future,R20 and CWD under RCP4.5 scenario in the near future.Moreover,the average annual runoff in the catchment are also expected to increase by more than 10%in the future,and the mid-to-long future RCP4.5scenario will be the scenario with the largest increase.（5）Comprehensive impacts of climate change,land use changes and socio-economic development changes were considered to explore the changes of non-point source pollution in the future.Combined with the priority control area results in the base period,the future priority control area of non-point source pollution was effectively identified,and the temporal and spatial distribution characteristics of annual and seasonal non-point source pollution were analyzed from the catchment and administrative unit scale.The results show that non-point source pollution in the catchment will be aggravated both in the near and mid-to-long future.The degree of aggravation in the near future is higher than that in the mid-to-long future,and that under the RCP4.5 scenario is higher than that under the RCP8.5 scenario.Areas with high pollutant concentration change rates are mainly located in the Jinjiang catchment,the lower Wushui catchment,the Nanshui catchment,and the upper Lianjiang catchment.On the catchment scale,S-level and A-level control areas are mostly located in the Nanshui catchment,lower Wushui catchment,right bank of the Jinjiang catchment,and the upper Lianjiang catchment,while on administrative unit scale,the S-level and A-level control areas are mostly located in Ruyuan County,the southwestern part of Renhua County,the southwest of Renhua County,Yizhang County,Lianzhou County,and Shaoguan City.