Identifying sources of groundwater contamination in a hard-rock aquifer system using multivariate statistical analyses and GIS-based geostatistical modeling techniques

Abstract

The study area is Udaipur district, which is situated in hard-rock hilly terrain of Rajasthan, India. In this study, spatio-temporal variations of fifteen groundwater quality parameters are explored by box–whisker plots, trends are detected and quantified, and GIS-based groundwater quality index (GQI) is computed. For the first time, scores of principal component analysis (PCA) are combined with GIS-based geostatistical modeling by following a sound methodology in comprehensive manner to identify sources of groundwater contamination. Box–whisker plots revealed linkages between rainfall and groundwater quality, which were further verified by GQI ranging from 69 to 76 in Cluster I and from 73 to 78 in Cluster II. Cluster analysis identified two clusters of sites based on groundwater contamination controlled by geology. Significantly increasing trends are indicated (p < 0.05) at most sites in fluoride, sodium, EC and TDS, but significantly decreasing trends in silica at 40% sites indicate a possibility of replacement of older groundwater with recent rainfall recharge. Spatial distribution of increasing trends is affected by anthropogenic processes. Sen's method indicated increasing rates for calcium, magnesium, sodium, iron, bicarbonate, sulphate, fluoride, TDS, hardness and EC. PCA results indicated occurrence of groundwater contamination in Cluster I by anthropogenic sources and presence of natural/geogenic processes in Cluster II. Significant PCs, viz. major ion and soil leaching pollution factors, govern overall evolution of geochemical processes.