Groundwater Pollution Source Identification And Apportionment Using PMF And PCA-APCS-MLR Receptor Models

Groundwater is the main source of water for producting and living in many cities,and the water quality is influenced by both nature and human.Once it is polluted,it is difficult to restore the original state.By detecting the physical and chemical properties of groundwater samples,the receptor model can be used to identify the pollution sources in groundwater,and provide theory for improving the quality of groundwater and protecting groundwater.At present,the results of single receptor model are one-sided in groundwater pollution sources analysis.This study combined PMF model and PCA-APCS-MLR model,put forward the concept of model error,unexplained variable and the calculation method of contribution rate.Moreover,carried out qualitative identification and quantitative analysis of groundwater pollution sources.Through the comparative analysis of the results of the two models,a more reliable analytical result is provided for the sources of groundwater pollution.Applying the improved model to study area can better reveal the application significance of the model.In this study,the improved model was applied to Jingyang District,Deyang City,Sichuan Province.This study area is rich in groundwater for production and living.In recent years,with the rapid development of economy and society in the study area,the groundwater resources in the study area have been seriously damaged and the water quality continues to decline.Based on the improved model,this study comprehensively analyzed the hydrogeological conditions,pollution status and physical geographical conditions of the study area,and traced the source of groundwater pollution in the study area through sampling,experimental research and relevant data collected,so as to understand the applicability of the improved model.The results of the two models show that there are five main sources of groundwater pollution in the study area,including domestic pollution,geological process,agricultural activities,natural source,and industrial pollution.The contribution rates of unexplained variables and agricultural activities in PMF model were 13.35%and 22.1%,respectively,and 32.87%and 9.35%in PCA-APCS-MLR model,respectively.The value of R~2of PMF model is better than that of PCA-APCS-MLR model,and the contribution rate of unexplained variables is lower than that of PCA-APCS-MLR model,indicating that PMF model can provide a more reasonable pollution sources in this study.Compared with the analytical results of PCA-APCS-MLR model,it can truly reflect the local groundwater pollution sources.