Source-oriented landscape apportionment of heavy metals by using an integrated approach in typical mining urban...

Heavy metals (HMs) contamination in agricultural soils presents potential environmental risks and has therefore attracted widespread public concern. Precise identification of pollution source landscapes is essential to mitigate soil pollution. In this research, the principal component analysis-absolute principal component score-multiple linear regression model, along with correlation analysis, bivariate spatial autocorrelation analysis, and the Geodetector model, was used to identify source landscapes. Pollution radius, diffusion coefficient, positive matrix factorization, and atmospheric input fluxes were used to quantify their contributions in a typical mining city. Results revealed that Cd and Cu were the most serious contaminants among eight HMs. The natural (38 %), agricultural (19 %), and industrial (22 %) sources were identified as the dominant pollution sources. Compared with existing models, the coupled source landscape apportionment model could identify the locations of source landscapes and quantify their contribution. The industrial source landscapes result indicated that Source Landscapes 2, 3, 4, 5, and 6 were the major source landscapes, which accounted for 13.96 %, 27.93 %, 27.35 %, 9.98 %, and 20.78 %, respectively. Cu-related mining and smelting activities were identified as the most important sources of anthropogenic emissions, which were the major contributors to the accumulation of Cu in SLs 2, 3, and 4. Overall, our findings provide precise location and contribution information for key source areas, which enables decision-makers to formulate highly effective regional policies for preventing and controlling soil HM-pollution.