Pollution Characteristics and Health Risks of Heavy Metals in the Groundwater of a Chemical Industrial Park

Heavy-metal contamination of groundwater in chemical industrial parks threatens water resources and human health; however, conventional health risk assessments often use inconsistent averaging times and unsupported toxicity values. This study assessed the pollution characteristics and age-specific health risks of heavy metals in groundwater from a chemical industrial park in southern China. A total of 126 groundwater samples were collected, and 11 heavy metals were determined. Single-factor and Nemero pollution indices, Pearson correlation, and a probabilistic health risk assessment coupling the USEPA model with Monte Carlo simulation were employed; complete concentration vectors were resampled at the sample level to preserve inter-metal correlations, and partial rank correlation coefficients identified independent risk drivers. All mean heavy-metal concentrations except those of As and Cu exceeded the Class III values of GB/T 14848-2024, which specify groundwater quality for drinking-water sources after appropriate treatment, and the Nemero index (42.78) indicated severe pollution. Zn, Al, Cd, Ni and Co were strongly correlated (r generally > 0.8), indicating shared geochemical associations and/or common industrial inputs rather than proving a single source. After correcting the averaging time and restricting the carcinogenic assessment to metals with documented oral slope factors, the mean hazard index was 15.4 for adults and 45.9 for children; the probability of HI > 1 was 85.9% and 98.9%, respectively. Co, Mn, Tl and As dominated the non-carcinogenic risk in both adults and children (Co 30.2%, Mn 29.8%, Tl 20.1% and As 11.1% of HI), whereas the lifetime total carcinogenic risk (4.57 × 10−3; all simulated scenarios exceeded 1 × 10−4) was dominated by As and Cd (83.7% and 16.3%, respectively). A model-choice sensitivity check showed that using a lifetime non-carcinogenic averaging time would reverse the adult–child risk ranking, while using the legacy oral cancer slope factor for As at 1.5 instead of the current IRIS value at 32 would change the dominant carcinogen from As to Cd. Partial rank correlation analysis showed that As and Mn were the main independent metal drivers for both groups, while As controlled the carcinogenic risk. These findings support risk-based priority control of As, Cd, Co, Mn and Tl, with children as the sensitive population.