Study of variations in water quality of Mumbai coast through multivariate analysis techniques

Abstract

Multivariate statistical techniques, such as Cluster Analysis (CA), Discriminant Analysis (DA), and Principal component analysis (PCA) were applied to evaluate the temporal/spatial variations in marine water quality of Mumbai and to identify pollution sources. Hierarchical CA grouped 12 sampling sites into three clusters of similar water quality characteristics. DA gave the best results both spatially and temporally. It provided an important data reduction as it used only four parameters (DO, Total coliform, Ammonical nitrogen and pH) affording 100% correct assignment in temporal analysis. For spatial DA, DO and temperature; Feacal strptococii, DO and Total coliform; temperature and phosphate were used for summer, monsoon and winter seasons respectively. DA gave 100% correct assignment in spatial analysis except for summer season, step wise mode DA rendered 91.6% correct assignment. PCA resulted in four factors explaining 81.4% of the total variance. The first factor obtained represents organic pollution from domestic waste water. The second factor represents natural pollution which includes the surface run off. The third factor represents nutrient pollution whereas the fourth factor represents seasonal effects of temperature.