Modification and optimization of DRASTIC model for groundwater vulnerability and contamination risk assessment for Bhiwadi region of Rajasthan, India

The objective of the study is to estimate groundwater vulnerability against contamination in Bhiwadi region of Rajasthan by applying geographical information system (GIS)-based DRASTIC model which considers seven hydrogeological parameters of an aquifer: depth to water (D), net recharge (R), aquifer media (A), soil media (S), topography (T), impact of vadose zone (I), and hydraulic conductivity (C). The groundwater vulnerability (intrinsic vulnerability) and risk assessment (specific vulnerability) is done using original DRASTIC as well as modified DRASTIC to find the best-suited model for the study area. Groundwater risk assessment is done by integrating land use map with appropriate weight and ratings with vulnerability map. The study methodology includes modification of DRASTIC parameter ratings based on the mean chromium (Cr) concentration of each parameter range through simple statistical technique and DRASTIC parameter weight modification by two different methods: (1) maximizing coefficient of correlation between vulnerability index and chromium (Cr) concentration using generalized reduced gradient (GRG) solver package in excel (2) single parameter sensitivity analysis (SPSA) for evaluating effective weight based on influence of individual parameter on vulnerability index. The DRASTIC parameter ratings modified on the basis of chromium concentration gives very low correlation coefficient (r = 0.24) due to less correlation of ratings of depth to the water table, slope and vadose zone to chromium concentration, hence no revision of parameter ratings is required for the study area. Compared to the correlation coefficient (r = 0.3) of original DRASTIC, weight modified DRASTIC, based on SPSA method gives r = 0.35 and GRG optimizing solver gives r = 0.37, indicating improvement in correlation coefficient due to weight modification. The result also shows that GRG optimizing solver gives better correlation coefficient than SPSA method. The best correlation coefficient (r = 0.41) between vulnerability index and chromium concentration is obtained by weight modification (using GRG optimizing solver) and including land-use layer in the original DRASTIC model. The result indicates that integration of DRASTIC with land use improved the correlation coefficient between chromium concentration and vulnerability index for original as well as ratings and weight modified DRASTIC models. The sensitivity analysis test performed indicates that recharge is the most sensitive parameter whereas soil is the least sensitive parameter in the study area. The study confirms that modification and optimization of DRASTIC model by means of weight modification as well as the integration of land use layer in original DRASTIC increases the correlation coefficient and gives a more accurate assessment of groundwater vulnerability to chromium contamination in the study area.