Hidden threats beneath: uncovering the bio-accessible hazards of chromite-asbestos mine waste and their impacts on rice components via multi-machine learning algorithm.
Journal:
Environmental geochemistry and health
PMID:
40381117
Abstract
The chromite-asbestos mining leaves behind tonnes of toxic waste, contaminating nearby agricultural fields with potentially toxic elements (PTEs). Over time, wind and water erosion spread these pollutants, severely impacting the ecosystem, food chain, and human health. This study evaluates the bioaccessible (stomach and intestinal phases) and leachable forms of PTEs, emphasizing the health and dietary risks associated with PTE pollution in this region. The study result indicates that the leachable and bio-accessible PTEs concentrations in agricultural soil, mainly Cr and Ni, were higher in zone 1 (mine tailings dumping area) and zone 2 (tailings contaminated soil) than zone 3 (uncontaminated soil). PTEs content in rice parts, mainly in boiled rice, showed moderate risk in the SAMOE model from Cr (0.011) and Ni (0.013) while in rice (without husk), it indicated high (class 5) dietary risk. The Fuzzy-TOPSIS, artificial neural network, and Monte-Carlo simulation models all demonstrated that Cr was the major contributor to anthropogenic risk. Compared to adults (5.08E-05), children (1.88E-03) were more vulnerable to total carcinogenic risk via ingestion pathway. Machine learning methods have been implemented to forecast the effects of leachable PTEs on soil-rice systems and possible health hazards associated with consuming food from the chromite-asbestos waste-contaminated zone. The survey-based Fuzzy-DEMATEL technique also showed that consumption of starch and cooked rice were the most crucial factors influencing the population's health risk. Overall, the implications of the statistical model may aid in assessing potential health hazards and enhancing regulations for ecosystem preservation.