Holistic approach towards pollution abatement of groundwater in major industrial belts of Jharsuguda District, Odisha, India and its modeling.

The major aim of this study is to assess the level of contamination of the groundwater quality in the major industrial zone of Jharsuguda district, Odisha, using multivariate statistical techniques (Principal component analysis [PCA], cluster analysis [CA], and multivariate analysis of variance [MANOVA]) and a feed-forward artificial neural network (ANN) model. From physicochemical analysis, it was observed that although most of the parameters of water are beyond the permissible limit (World Health Organization [WHO]), total suspended solids (TSS), total dissolved solids (TDS), electrical conductivity (EC), and turbidity values are comparatively much higher. The result obtained from happy planet index (HPI), Nemerow's Pollution Index (NPI), and water quality index (WQI) shows that water is not suitable for human use. The WQI values range from 366.32 to 430.96 (class "E"). Class "E" indicates that the water is heavily contaminated and not suitable for human use. HPI values range from 454.02 to 1962.21, indicating high contamination of water. The feed-forward ANN model is used to determine the level of modeling performance, and the data are completely fitted by the regression predictions, as indicated by the values of R = 0.99 and R = 1. PRACTITIONER POINTS: The level of contamination and quality of the groundwater in major industrial zones of the Jharsuguda district is estimated through WQI, NPI, and HPI. Pearson's correlation shows the dynamic relationship among various parameters, whereas in one-way ANOVA, P<0.05 indicates high contamination of the groundwater. Multivariate statistical analyses such as CA, MANOVA, and PCA are used to detect the variability of the different parameters. Feed-forward ANN modeling is used to know about the correlation among various water parameters and the comparison of the experimental values with predicted values. This study helps in achieving sustainable development goals related to clean water and sanitation, specifically in industrial areas.