A coherent approach of Water Quality Indices and Multivariate Statistical Models to estimate the water quality and pollution source apportionment of River Ganga System in Himalayan region, Uttarakhand, India

River Ganga covers around 26% of India’s land area and sustains diverse ecosystems in this overly populated area. The globally accepted coherent approach of water quality indices (WQIs) and multivariate statistical models (principal component analysis (PCA) and cluster analysis (CA)) were applied on the dataset to evaluate the spatial-temporal variation and pollution source identification and apportionment. Twenty-two hydro-chemical parameters were analyzed by collecting the samples from 20 different vertically elevated monitoring locations for different seasons. The CA evaluation of data, grouped the monitoring locations into five clusters of varied water quality with human perturbations and geo-genic inputs. The PCA analysis of an extensive dataset indicated the seven significant principal components (PCs) explaining 93.0% of the total variance and finalized 8 water quality parameters out of preselected 22 to represent good aspects of the water quality. The seasonal variation in river water quality by the Canadian Council of Ministers for Environment Water Quality Index (CCMEWQI) showed the quality class at a marginal level in summer (62.16), monsoon (59.96), and post-monsoon (60.20) season, whereas in winters (71.18), water quality was in fair condition. The response of National Sanitation Foundation Water Quality Index (NSFWQI) classified the river water in medium quality class for summer, monsoon, post-monsoon, and winter season, respectively. The present observations contribute in the usefulness of these statistical methodologies to interpret and understand large dataset and also provide reliable information to reduce the tedious and cost of water quality monitoring and assessment programs.