To qualitatively identify the potential sources of groundwater pollution in Tongchuan City and quantitatively evaluate the contribution of each source to groundwater quality, 59 groundwater samples were collected from the study area, and 14 key water quality indicators were measured. PMF model and PCA-APCS-MLR model were used to analyze the pollution sources of groundwater samples in Tongchuan City, and their applicability in groundwater pollution assessment in Tongchuan City was tested. Results showed that both of the models identified four sources of groundwater contamination with similar main load species of each potential source. Agricultural activities and human activities were the important reasons for groundwater pollution in the study area. The spatial distribution of each pollution source is also different in the two models, however, the areas with more serious pollution caused by agricultural pollution and human activities were mainly distributed in the central and northern parts of the study area. The fitting degree of each water quality parameter of the two models was good, but the R2 value parsed by the PMF model are mostly higher than that of the PCA-APCS-MLR model, indicating that the analytical results of the PMF model were more accurate and could better reflect the groundwater pollution in the study area.
 

Groundwater pollution, Source analysis, Positive matrix factorization, Principal component analysis, Tongchuan City