The present study focused on investigating the contamination and risk assessment for 16 metals in street dust from Ha Noi highway, Ho Chi Minh City. The results indicated that the concentrations of metals (mg/kg) were found, in decreasing order, to be Ti (676.3 ± 155.4) > Zn (519.2 ± 318.9) > Mn (426.6 ±113.1) > Cu (144.7 ± 61.5) > Cr (81.4 ± 22.6) > Pb (52.2 ± 22.9) > V (35.5 ± 5.6) > Ni (30.9 ± 9.5) > Co (8.3 ± 1.2) > As (8.3 ± 2.5) > Sn (7.0 ± 3.6) > B (5.7 ± 0.9) > Mo (4.1 ± 1.7) > Sb (0.8 ± 0.3) > Cd (0.6 ± 0.2) > Se (0.4 ± 0.1). The geo-accumulation index (Igeo) showed moderate contamination levels for Pb, Cd, Cu, Sn, Mo, and Zn. The enrichment factor (EF) values revealed moderate levels for Cd, Cu, Mo, and Sn but moderate–severe levels for Zn. The pollution load index of the heavy metals was moderate. The potential ecological risk (207.43) showed a high potential. Notably, 40.7% and 33.5% of the ecological risks were contributed by Zn and Mn, respectively. These findings are expected to provide useful information to decision-makers about environmental quality control strategies.
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.