Critical review of groundwater quality analysis techniques using multivariate statistical and soft computing approaches

The continuous monitoring and assessment of water quality is an important aspect of the management of groundwater resources, substantially affected by the exponentially increasing demand for water to meet irrigation and industrial requirements. Various researchers have studied the variation in groundwater quality and applied different techniques to quantify, evaluate, predict and model the flux in groundwater quality parameters. The multivariate statistical analysis techniques including cluster analysis, factorial method and principal component analysis considerably used for the identification of spatio-temporal variation in groundwater characteristics. Several groundwater quality indices (GWQI) have been used to quantify the level of pollution and give an in-depth to improve the quality. In addition to the above approaches, various machine learning techniques have been developed increasingly due to their accuracy. The applications of traditional and present methods such as multivariate statistical models, groundwater quality indices and machine learning techniques through programming languages, probability and uncertainty analysis are analysed in this proposed study.