Scholarlink

<p>Heavy metal contamination in drinking water from artisanal mining areas poses significant public health risks, yet conventional deterministic assessments often underestimate risk and uncertainty and neglect bioavailability considerations. This study conducted a comprehensive probabilistic health risk assessment of trace metals in groundwater from Ghana’s Upper East Region mining district using integrated geochemical speciation and hierarchical Bayesian exposure modelling. Water samples (<i>n</i> = 69) were analyzed for pH, total dissolved solids, arsenic, cadmium, chromium, mercury, lead, and other metals using inductively coupled plasma mass spectrometry (ICP-MS). Entropy-weighted pollution indices (heavy metal pollution index [HPI]: 0.68–0.81; potential ecological risk index [PERI]: 19.1–22.9) indicated uniformly low contamination, with cadmium dominating weight allocation (44.8%). PHREEQC thermodynamic modelling served as a critical interpretive module, confirming that metals existed predominantly as highly toxic, bioavailable free divalent ions (Cd<sup>2+</sup>, Pb<sup>2+</sup>, Ni<sup>2+</sup> &gt; 95%) and arsenate (HAsO<sub>4</sub><sup>2−</sup>, 53%). This validated the conservative approach of basing ingestion risk on total dissolved concentrations. Hierarchical Bayesian modelling with Markov chain Monte Carlo (MCMC) sampling demonstrated negligible non-cancer hazard (hazard index [HI]: 2.5 × 10<sup>−3</sup> to 1.0 × 10<sup>−2</sup>, well below the threshold of 1.0), yet lifetime cancer risk (CR) exceeded the minimum benchmarks by three orders of magnitude (median CR: 2.9 × 10<sup>−4</sup> to 7.6 × 10<sup>−3</sup>). Sobol′ global sensitivity analysis identified arsenic concentration as the dominant uncertainty driver, explaining 48%–60% of total risk variance. Blood lead predictions remained below Centers for Disease Control and Prevention (CDC) reference values (median ~2&#xa0;µg/dL), though 25% of populations showed exceedance probability. These findings reveal that chronic low-level arsenic exposure generates regulatory-relevant carcinogenic risk despite benign screening indices, necessitating arsenic-targeted interventions and speciation-aware monitoring protocols in mining-impacted aquifers. The study supports incorporating probabilistic exceedance-based criteria alongside conventional total metal thresholds in water safety planning and monitoring.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Probabilistic health risk assessment of trace-metal contamination in multiple water sources from illegal mining areas, Upper East Ghana

  • 期刊论文

  • 发表日期：

    2026年5月25日

查看全文
  • Marian Sitsofe Kretchy,
  • Samuel Yao Ganyaglo,
  • Abass Gibrilla,
  • Anita Asamoah

摘要

Heavy metal contamination in drinking water from artisanal mining areas poses significant public health risks, yet conventional deterministic assessments often underestimate risk and uncertainty and neglect bioavailability considerations. This study conducted a comprehensive probabilistic health risk assessment of trace metals in groundwater from Ghana’s Upper East Region mining district using integrated geochemical speciation and hierarchical Bayesian exposure modelling. Water samples (n = 69) were analyzed for pH, total dissolved solids, arsenic, cadmium, chromium, mercury, lead, and other metals using inductively coupled plasma mass spectrometry (ICP-MS). Entropy-weighted pollution indices (heavy metal pollution index [HPI]: 0.68–0.81; potential ecological risk index [PERI]: 19.1–22.9) indicated uniformly low contamination, with cadmium dominating weight allocation (44.8%). PHREEQC thermodynamic modelling served as a critical interpretive module, confirming that metals existed predominantly as highly toxic, bioavailable free divalent ions (Cd2+, Pb2+, Ni2+ > 95%) and arsenate (HAsO42−, 53%). This validated the conservative approach of basing ingestion risk on total dissolved concentrations. Hierarchical Bayesian modelling with Markov chain Monte Carlo (MCMC) sampling demonstrated negligible non-cancer hazard (hazard index [HI]: 2.5 × 10−3 to 1.0 × 10−2, well below the threshold of 1.0), yet lifetime cancer risk (CR) exceeded the minimum benchmarks by three orders of magnitude (median CR: 2.9 × 10−4 to 7.6 × 10−3). Sobol′ global sensitivity analysis identified arsenic concentration as the dominant uncertainty driver, explaining 48%–60% of total risk variance. Blood lead predictions remained below Centers for Disease Control and Prevention (CDC) reference values (median ~2 µg/dL), though 25% of populations showed exceedance probability. These findings reveal that chronic low-level arsenic exposure generates regulatory-relevant carcinogenic risk despite benign screening indices, necessitating arsenic-targeted interventions and speciation-aware monitoring protocols in mining-impacted aquifers. The study supports incorporating probabilistic exceedance-based criteria alongside conventional total metal thresholds in water safety planning and monitoring.