Evaluation of Efficacy of Groundwater Level Monitoring Network by Graphical and Multivariate Statistical Techniques

Abstract

Knowledge of spatial and temporal patterns of groundwater levels is useful to evaluate efficacy of existing groundwater monitoring network. Present study demonstrates application of graphical and multivariate statistical techniques to evaluate monitoring network of 50 sites in hard-rock aquifers of Ahar River catchment of Udaipur, Rajasthan. The graphical analysis included plotting of time plots and box-whisker plots for monthly groundwater levels for three years (May 2006-July 2009). The result revealed that temporal pattern of the groundwater level was almost similar for majority of the sites. Box-whisker plots revealed that the spatial variations of the groundwater levels were smaller in post-monsoon compared to that in the pre-monsoon season. Furthermore, principal component analysis (PCA) of the groundwater levels revealed that the first two principal components (PC) characterized as ‘hydraulic connectivity factor’ and ‘rainfall factor’ explained more than 75% of the total variance. Based on the PCA, the monitoring sites were classified into five groups according to behavior or patterns of the groundwater levels. In every group, a single site is adequate to represent groundwater characteristics/responses. It was observed that there are possibilities of reducing number of monitoring sites as much as by 90% to make the network more cost-effective and less time-consuming.