Identifying key hydrochemical processes in a confined aquifer of an arid basin using multivariate statistical analysis and inverse modeling

Tertiary fractured permeable confined aquifer, which covered about 70 % of the studying area, played an important role in alleviating drinking water shortages. However, about 58 and 79 % of the groundwater samples exceeded the desirable limits for fluoride (1.5 mg/L) and TDS (1,000 mg/L). Two multivariate statistical methods, hierarchical cluster analysis (HCA) and principal components analysis (PCA), were applied to a subgroup of the dataset in terms of their usefulness for groundwater classification, as well as to identify the key processes controlling groundwater geochemistry. In the PCA, two principal factors have been extracted, which could explain 73 % of the total data variability. Among them, factor 1 revealed the source of groundwater salinity and factor 2 explained the elevated fluoride. Two major groups were classified by HCA and Group 1 was near the groundwater recharge zone and Group 2 was mainly distributed over the groundwater discharge zone. Inverse modeling (NETPATH) results indicated that the hydrochemical evolution was primarily controlled by (1) the dissolution of mirabilite, gypsum and halite for the source of groundwater salinity; (2) the release of the adsorbed fluoride through desorption or through competition with HCO₃ ⁻ under alkalinity condition for the elevated fluoride in the groundwater.