Groundwater pollution is controlled by many anthropogenic processes, and source apportionment is particularly difficult in cases with multiple superimposed anthropogenic influences. Forty-three shallow groundwater samples in the Hun River alluvial fan were collected and 15 parameters comprising Ca2+, Mg2+, Fe2+, Mn2+, NH4+, NO3−, NO2−, F−, SO42−, PO43−, Cl−, COD, TDS, CN− and HCO3− were studied using principal component analysis (PCA) and factor analysis (FA), coupled with the absolute principal-component-score multiple-linear-regression (APCS-MLR) receptor model. The relationship among land-use types, hydrochemical composition and change of the groundwater quality from natural and anthropogenic sources was demonstrated. The results show that the four major potential pollution sources were water–rock interaction, agricultural fertilizer pollution, geological background and domestic and industrial wastewater; the contribution was 36.4%, 24.1%, 14.7% and 11.8%, respectively. The high-concentration areas were mainly located in the western and northwestern areas, especially in the downstream portion of the Hun River. Rapid economic development and the acceleration of urbanization led to an increase in industrial sewage effluent with complex and changeable composition. The variation of land-use type and evolution of the spatial distribution of the pollution sources in the groundwater show good consistency, which demonstrate that PCA/FA coupled with APCS-MLR is a versatile tool for comprehensive source apportionment of groundwater.