Stochastic framework for source quantification and health risk evaluation of soil potentially toxic elements in the Maluwe–Tinga–Wasipe Corridor, Savannah Region, Ghana

Abstract This study applies a stochastic framework to quantify Potentially Toxic Elements (PTEs) and related risks in soils of the Maluwe–Tinga–Wasipe corridor in the Savannah Region of Ghana. A total of 729 B-horizon samples were analysed by Energy Dispersive X-ray Fluorescence. Chromium ranged from 11 to 512 mg/kg, nickel from 1.9 to 216 mg/kg, and cobalt from 2.9 to 183 mg/kg respectively. Arsenic averaged 4.7 mg/kg, and peak levels reached 17.4 mg/kg, far above the United States Environmental Protection Agency residential threshold of 0.39 mg/kg. Barium reached 3610 mg/kg, which far exceeded its Upper Continental Crust value of 425 mg/kg. Pollution indices showed that 97.5% of samples fell within the low-risk class of the Nemerow integrated risk index. However, 87.8% showed enrichment on the geoaccumulation scale. Thirty-seven per cent were moderately polluted, and up to three per cent were heavily polluted. Positive Matrix Factorisation (PMF) resolved three sources. Factor 1 (19.8%) was dominated by strontium and barium from lithological weathering. Factor 2 (60.4%) was defined by nickel, copper, vanadium, manganese, and cobalt from mafic and ultramafic bedrock and sulphide oxidation. Factor 3 (19.8%) was enriched in arsenic, lead, zinc, and chromium from artisanal mining and agrochemical inputs. Gradient Boosting models reproduced source contributions with R2 values of 0.97, 0.97, and 0.89. Probabilistic health risk assessment showed low adult risks but elevated child risks. A majority of the child hazard index simulations (61.3%) and cancer risk simulations (92.9%) exceeded acceptable limits.