This paper presents the interpretation of voluminous and complex dataset of Vishav stream, using different environmetric techniques, acquired during 1 year monitoring program of 21 parameters at five different sites. General linear regression model depicted better and higher positive and negative relationship among various analyzed water quality parameters (p < 0.0001), indicating that the water temperature, dissolved oxygen, conductivity, TDS, NO3-N, and ortho-phosphorus (ortho-P) appears to be important parameters to predict overall water quality. Hierarchical cluster analysis classified similar water quality sites into two significant clusters (Cluster II and Cluster I). Cluster II (sites I and II) is characterized by excellent water quality while cluster I (sites III, IV and V) is considered to be of moderately poor water quality, thereby reflecting the different physico-chemical characteristics and quality status. Principal component analysis has allowed identification of a reduced number of mean three varifactors (VF1, VF2 and VF3), pointing out 88.3 % of both spatio-temporal changes. Factor analysis showed that the first factor (VF1) explained 63.9 % of the total variance comprising of WT, discharge, free carbon dioxide, NO3-N and total phosphorus and strong negative loading on pH, DO, total hardness, calcium hardness and magnesium hardness. VF2, explaining 14.13 % of total variance, has a strong positive loading on ortho-P and sulphate. VF3, explaining the lowest variance (10.12 %), has strong positive loading on ammonical-nitrogen and dissolved silica. Thus, the present study demonstrates the importance of environmetric techniques for reliable characterization and evaluation of surface water quality, over a short period for effective management.
Read full abstract