Multivariate statistical analysis of water quality data in Narmada river

Abstract

Multivariate statistical techniques are gaining popularity in the analysis of water quality data due to its simplicity when handling a large number of variables simultaneously and capable of producing more easily interpretable results for the evaluation of observed quality data. In this study, factor analysis and cluster analysis are applied to water quality data set obtained from Narmada River and an attempt has been made to present a strategy that reduces the measured parameters, locations and frequency without compromising the quality of the original data. Factor analysis shows that river water quality data consists of three significant components viz., animal waste discharges, sewage discharges and industrial discharges whereas Chloride, Bicarbonate Alkalinity and pH are the respective indicator variables. Cluster analysis also suggests that number of sampling sites as well as the sampling frequency can be consolidated. This reduced set of parameters, sampling sites and sampling frequency could be monitored over larger areas within the watershed to provide more detailed spatial information about sources and processes.