MAHESH MOHAN AND P.K. OMANA
Abstract
Water samples from Vembanadu backwaters of south west coast of India were measured for physico chemical parameters over different seasons from September 2003 to March 2005. The whole data has been factorised using Principal Component Analysis (PCA) for extracting the total variability and linear relationships existing among a set of different physico chemical parameters of the backwater system. The scores generated by the PCA of different seasons were graphically represented by box plot. The principal Components (PCs) governing backwater system is Chloride, Salinity and Calcium hardness, total hardness, Biochemical Oxygen Demand and pH. BOD as a principal component due to urban sewage and industrial effluents along with decreased dilution rate during the non-monsoon seasons has significant effect on the backwater system. The box plots explained variation in PC scores season wise. Hierarchical Cluster Analysis (HCA) and the graphical display obtained the homogenous groups of variables are same as in first principal component (PC 1) of PCA, which simplified the interpretations.