Groundwater Arsenic Distribution in India by Machine Learning Geospatial Modeling

Abstract

Groundwater is a critical resource in India for the supply of drinking water and for irrigation. Its usage is limited not only by its quantity but also by its quality. Among the most important contaminants of groundwater in India is arsenic, which naturally accumulates in some aquifers. In this study we create a random forest model with over 145,000 arsenic concentration measurements and over two dozen predictor variables of surface environmental parameters to produce hazard and exposure maps of the areas and populations potentially exposed to high arsenic concentrations (>10 µg/L) in groundwater. Statistical relationships found between the predictor variables and arsenic measurements are broadly consistent with major geochemical processes known to mobilize arsenic in aquifers. In addition to known high arsenic areas, such as along the Ganges and Brahmaputra rivers, we have identified several other areas around the country that have hitherto not been identified as potential arsenic hotspots. Based on recent reported rates of household groundwater use for rural and urban areas, we estimate that between about 18-30 million people in India are currently at risk of high exposure to arsenic through their drinking water supply. The hazard models here can be used to inform prioritization of groundwater quality testing and environmental public health tracking programs.

Keywords: India; arsenic; geospatial modeling; groundwater; machine learning; random forest.

Figure 1

Groundwater arsenic data points and simplified geology of the Indian subcontinent. (a) Spatially averaged arsenic data points used in modeling along with topography in India and neighboring countries. (b) Lithology of the Indian subcontinent.

Figure 2

Arsenic hazard maps. (a) Probability of arsenic concentration in groundwater exceeding 10 µg/L. (b) High hazard areas in India based on probability cutoffs of 0.49 and 0.55.

Figure 3

Normalized variable importance in terms of mean decrease in accuracy and mean decrease in Gini as calculated on the test dataset. Both decrease in accuracy and decrease in Gini were normalized by their respective greatest values (see Table S1).

Figure 4

Correlations of predictor variables (a–x) with percentages of arsenic data points exceeding 10 µg/L in 16 equally sized bins. Kendall correlations (τ_B) with a statistically significant p value (95% confidence level) are shown in bold.

Figure 5

Analyses of model performance using full modeling dataset. (a) Sensitivity and specificity were found to be equivalent at a probability cutoff of 0.49 with a corresponding accuracy of 96%. (b) Positive predictive value (PPV) and negative predictive value (NPV) were found to be equivalent at a probability cutoff of 0.55 also with a corresponding accuracy of 96%.