Correlation and Linear Regression Analysis of Water Quality Parameters of Mahi River, Gujarat

Abstract

Statistical analysis, such as correlation and regression analysis were applied to the data on water
quality of Mahi River, generated during 2014-2015 year at seven different sites for 14 parameters.
Establishing an empirical equations of regression analysis has been found to be a highly useful tool for
correlating different parameters and to find monitoring the status of surface water quality. The physicochemical parameters were measured such as pH, Total Dissolved Solids (TDS), Total Solids (TS), Total
Suspended Solids (TSS), Total Hardness (TH), Calcium Hardness (Ca
2+
), Magnesium Hardness (Mg
2+
),
Chloride (Cl
-
), Total Alkalinity (TA), Dissolved Oxygen (DO), Biochemical Oxygen Demand (BOD),
Chemical Oxygen Demand (COD) Temperature and Ammonia (NH3N). Pearson’s correlation coefficient (r)
value is determined using correlation matrix to identify the highly correlated and interrelated water
parameters. The regression equation established between these parameters to assess the extent of pollution in
river. Correlation coefficient find between highly positive correlated parameters such as Total Dissolved
Solids (TDS)-Total Solids (TS), Magnesium Hardness (Mg+2) -Total Hardness (TH), Chemical Oxygen
Demand (COD) - pH, pH-Total Alkalinity, Total Hardness (TH) -Total Alkalinity (TA), Calcium Hardness
(Ca+2) - Total Hardness (TH), Biochemical Oxygen Demand (BOD)- Dissolved Oxygen (DO), pH - Ammonia
(NH3N) Total Suspended Solids - pH. Based on correlation coefficient (r), regression equation establish
which has R > 0.6 – 0.8 which shows parameters are dependent or independent on each other and measure
standard error for validation