Prioritizing pollution sources and targeted potentially toxic elements in soils using a probabilistic source-specific...

Soil potentially toxic elements (PTEs) pose significant ecological and health risks in mining areas with multiple pollution sources. However, accurate identification of priority pollution sources and PTEs remained unclear. To address this gap, this study developed an comprehensive framework integrating positive matrix factorization (PMF), geo-detector model (GDM), scanning electron microscopy and energy-dispersive spectroscopy (SEM-EDS) with probabilistic eco-health risk assessment. PMF model assisted with SEM-EDS and GDM revealed four potential sources: Cr (89.27 %) and Ni (88.34 %) were predominantly contributed by electroplating activities; As (86.26 %) was primarily attributed to gold leaching activities; Pb (84.60 %), Cd (65.14 %), and Zn (64.05 %) mainly originated from Pb/Zn smelting activities; while Cu (74.02 %) principally came from copper mining activities. Regarding ecological risk, Pb/Zn smelting accounting for 53.09 % of total risk was identified as the highest contributor, primarily driven by Cd (88.05 %) due to its high toxicity. While non-carcinogenic risks were acceptable for all population groups, carcinogenic risks reached cautionary level. Gold leaching activities, strongly linked to As, constituted the largest contributor to carcinogenic risks (children: 47.38 %; adult females: 49.45 %; adult males: 48.89 %). These findings enhance the accuracy of source-specific eco-health risk quantification for soil PTEs, providing actionable insights for optimized risk management under resource constraints.