Integrating spatial heterogeneity and speciation dynamics in source apportionment of toxic metal(loid)s at an abandoned hydrometallurgical zinc smelting site.

Zinc hydrometallurgy sites are critical hotspots for combined toxic metal(loid)s (TMs) pollution, yet the integration of spatial heterogeneity and migration dynamics into source apportionment remains underexplored. This study investigated the concentrations, speciation, and spatial distribution of nine TMs (As, Cd, Cu, Hg, Mn, Ni, Pb, Sb, Zn) in soils at an abandoned zinc smelter in southwest China. Multivariate statistical methods and the Positive matrix factorization (PMF) model were applied to disentangle primary sources and secondary redistribution. Spatial analysis revealed that As, Cd, Cu, Pb, Sb, and Zn shared similar contamination patterns, concentrated in slag storage and comprehensive recovery areas, whereas Hg and Mn exhibited distinct hotspots near sulfuric acid production and electrolysis zones. Vertical migration was most pronounced for Cd and Zn (> 8 m depth), followed by Hg and Mn (4-8 m), while As, Cu, Pb, and Sb were restricted to 0-4 m due to adsorption in clay-rich layers. Speciation analysis indicated high mobility of Cd and Zn (acid-soluble fraction: 66.96 and 52.10%, respectively), contrasting with reducible Pb and Mn (51.59 and 48.32%) and residual As, Hg, Ni, Sb (60.74-76.64%). The results from PMF model identified aqueous-phase (Cd, Zn, Mn) and solid-phase (As, Cu, Pb, Sb) migration pathways, validated by spatial correlations with topography and functional zones. Aqueous-phase contributions dominated low-lying areas, while solid-phase contributions aligned with elevated regions, reflecting topography-driven redistribution. This study advances source apportionment of TM in soil by unifying spatial heterogeneity, speciation dynamics, and receptor modeling, offering a framework for targeted risk assessment and remediation of industrial sites.