Identification of key controlling factors of groundwater chemistry and probabilistic health risk assessment using...

Groundwater is a critical potable water source worldwide, particularly in dry and semi-arid regions. This study integrates a qualitative hydrogeochemical graphical method, semi-quantitative Self-Organizing Maps (SOM), and the quantitative Positive Matrix Factorization (PMF) model to investigate the sources and key determinants of groundwater chemistry in Baicheng City, China. Additionally, probabilistic health risks associated with direct oral consumption of groundwater were further assessed by KDE-MCs (Kernel Density Estimation based-Monte Carlo Simulations). Groundwater was generally fresh, with localized enrichment of iron, fluoride, sulfate, chloride, arsenic, and nitrate. Hydrochemical facies evolved from HCO₃-Ca·Mg and HCO₃-Na to SO₄·Cl-Ca·Mg and SO₄·Cl-Na. Groundwater samples were categorized into seven groups by the SOM method: Group I (73%)-freshwater, and Groups II-VII containing excessive total iron, fluoride, sulfate, chloride, arsenic, and nitrate, respectively. Hydrogeochemical graphical method and PMF model demonstrated that natural processes acted as the main controlling factor governing the groundwater chemistry (>66.4%), among which 27.3% accounted for the fluoride and arsenic pollution caused by geogenic processes. Anthropogenic factors also played a significant role (≥11.9%), primarily responsible for the elevated nitrate levels. Long-term ingestion of contaminated groundwater may increase the likelihood of both non-carcinogenic and carcinogenic health effects, particularly among children. Groundwater arsenic concentrations in southern Baicheng demand particular attention, as they represent both the primary contributors to health risks and the most sensitive parameters for carcinogenic/non-carcinogenic risk assessments. These insights facilitate targeted regional groundwater management and remediation strategies.