Predicting Heavy Metal Concentrations in Shallow Aquifer Systems Based on Low-Cost Physiochemical Parameters Using Machine Learning Techniques.

Monitoring ex-situ water parameters, namely heavy metals, needs time and laboratory work for water sampling and analytical processes, which can retard the response to ongoing pollution events. Previous studies have successfully applied fast modeling techniques such as artificial intelligence algorithms to predict heavy metals. However, neither low-cost feature predictability nor explainability assessments have been considered in the modeling process. This study proposes a reliable and explainable framework to find an effective model and feature set to predict heavy metals in groundwater. The integrated assessment framework has four steps: model selection uncertainty, feature selection uncertainty, predictive uncertainty, and model interpretability. The results show that Random Forest is the most suitable model, and quick-measure parameters can be used as predictors for arsenic (As), iron (Fe), and manganese (Mn). Although the model performance is auspicious, it likely produces significant uncertainties. The findings also demonstrate that arsenic is related to nutrients and spatial distribution, while Fe and Mn are affected by spatial distribution and salinity. Some limitations and suggestions are also discussed to improve the prediction accuracy and interpretability.

Contamination, source attribution, and potential health risks of heavy metals in street dust of a metropolitan area in Southern Vietnam.

This study investigates distribution, pollution indices, and potential risk assessment for human health and ecology of eight heavy metals in twenty-five street dust samples collected from metropolitan area-Ho Chi Minh City, Vietnam. Results showed that Zn was of the highest concentration (466.4 ± 236.5 mg/kg), followed by Mn (393.9 ± 93.2 mg/kg), Cu (153.7 ± 64.7 mg/kg), Cr (102.4 ± 50.5 mg/kg), Pb (49.6 ± 21.4 mg/kg), Ni (36.2 ± 15.4 mg/kg), Co (7.9 ± 1.9 mg/kg), and Cd (0.5 ± 0.5 mg/kg). The principal component analysis revealed that three sources of heavy metals measured in street dust include vehicular activities (32.38%), mixed source of vehicular and residential activities (26.72%), and mixture of industrial and natural sources (20.23%). The geo-accumulation index values showed levels of non-pollution to moderately pollution for Mn and Co; moderately pollution for Ni; moderately to strongly pollution for Cd, Cr, and Pb; and strongly pollution for Cu and Zn. The potential ecological risk values of all sampling sites were close to the high-risk category. Zn (28.9%), Cu (25.4%), and Mn (24.4%) dominantly contributed to the ecological risk. For non-carcinogenic risk, the hazard quotient values for both children and adults were within a safety level. For carcinogenic risk, the TCRChildren was about 3 times higher than TCRAdults, but still within a tolerable limit (1 × 10-6 to 1 × 10-4) of cancer risk. Cr was a major contribution to potential risks in humans. Such studies on heavy metal in street dust are crucial but are still limited in Vietnam/or metropolitan area in Southeast Asia. Therefore, this study can fill the information gap about heavy metal contaminated street dust in a metropolitan area of Vietnam.