Source apportionment and health risks of heavy metals in agricultural soils near mining areas: APCS-MLR and Monte Carlo approaches

Abstract

Soil contamination is a significant threat to global food security and public health. Accurate apportionment of pollutant sources is a prerequisite for developing science-driven pollution control protocols. This research was undertaken in Huanren Manchu Autonomous County, located in Northeast China. With a resident population of approximately 216,000, the county boasts abundant natural resources including mineral deposits, biodiversity, and water reserves. Data were preprocessed using Principal Component Analysis (PCA) to enhance interpretability for subsequent modeling. Abbreviated Principal Component Score Multilinear Regression (APCS-MLR) and Positive Matrix Factorization (PMF) were cross-validated to ensure robust source attribution, thereby addressing the limitations of single-method uncertainty. This triangulation approach, combined with probabilistic Monte Carlo Simulation and health risk assessment, enabled a multi-dimensional evaluation of contamination pathways and risks. This aspect has been underexplored in heavy metal (HM) studies of mining-impacted agricultural soils. The average concentrations of eight heavy metals were as follows: Cr (74.0 mg/kg), Ni (32.1 mg/kg), Cu (118.9 mg/kg), Zn (541.7 mg/kg), Cd (2.2 mg/kg), Pb (202.0 mg/kg), Hg (0.3 mg/kg), and As (12.0 mg/kg). Quantitative pollution source analysis revealed three primary contributors to soil HMs: industrial point sources (contributing 46.1%), which is the most significant contributor to pollution; agricultural sources (contributing 22.2%) and natural sources (contributing 31.7%). Industrial sources, as the primary local pollution contributors, will effectively guide relevant government departments in formulating targeted management policies and measures. Probabilistic risk evaluation yielded two crucial findings: (1) Non-carcinogenic hazard indices for adults and children remained below 1, indicating acceptable risks from the presence of HMs in agricultural soils, however, (2) Carcinogenic risks surpassed the 1 × 10⁻⁴ cancer risk benchmark for 100% of children and 32.3% of adults. Carcinogenic risks to the human population arising from individual HMs showed the following sequence: Cr > Ni > As > Zn > Cd. This research has not only revealed an alarmingly high risk of cancer in the study region due to HMs accumulation in its agricultural soils but also, by identifying the crucial sources, provided a scientific basis for controlling this harmful pollution.

Keywords: Heavy metals; Human health risk assessment; Monte Carlo simulation; Pollution source apportionment; Soil contamination.