Determination of the Principal Factors of River Water Quality of Mahanadi Basin, Odisha using Multivariate Techniques

In this work, cluster analysis (CA) and principal component analysis (PCA)/factor analysis (FA) were used to analyse the quality of water to detect and measure the causative factors affecting the surface quality of the Mahanadi River in Odisha. 20 water quality characteristics comprising 19 sampling stations were employed in a 1-year (2020–2021) dataset for this purpose. To reflect the various water quality criteria of the river system, hierarchical CA was used to arrange the 19 sampling locations into three comparable groups. Highly polluted Cluster III is claimed to include Cuttack D/s (R-8), Choudwar D/s (R-19), and Paradeep (R-9). The data matrix's total variance was explained by the PCA/FA method's five most significant variables, which accounted for 93.9 % of it. TSS, TKN, EC, TDS, B, TH, Cland SO4 2- have all been correlated with (loading greater than 0.7) in the 1st PC, which accounted for 43.133 % of the total variance. COD, NH3-N, Free NH3, SAR; pH, F were all linked to the 2nd PC, which accounted for 23.055 % of total variation, whereas the 3rd, 4th, and 5th PCs, while accounting for 12.866 %, 8.603 %, and 6.241 % of total variation, respectively. Five separate VFs with eigenvalue greater than 1 were justified by the principle of Maximum Variance Rotation of the PC i.e., VF-1 (40.191%), VF-2 (21.358%), VF-3 (14.593%), VF-4 (9.082%), and VF-5 (8.675 %) of total variance (93.899%). The factors responsible of river water contamination/deterioration were discovered to be point source contamination discharges from human impacts caused by the discharge of agricultural residues, home and industrial wastewater. Additionally, it comprised of nutrients (agricultural runoff), alkalinity, hardness, EC, solids as well as soil weathering, leaching from landfills and industrial waste disposal sites. Comparing the findings of the CA and PCA/FA approaches revealed some notable disparities in the estimated contributions for each probable source of pollution. The PCA/FA procedure was found to be more physically reasonable in the end for the current investigation. According to expectations, the findings could be highly helpful to the local government in managing and controlling pollution and better safeguarding vital riverine water quality.