Evaluation of River Health Status Based on Water Quality Index and Multiple Linear Regression Analysis

Abstract

The present study was abstracted to investigate the health status of the river Damodar, India, for drinking purpose using water quality index (WQI) method. Eleven water quality parameters, i.e., pH, DO, BOD, TDS, Cl−, NO3−, SO42−, TC, F−, Fe, and Pb were analyzed from ten monitoring sites designated on the river stretch. The multiple linear regression (MLR) model was also anticipated as an efficient methodology for the prediction of WQI. The accuracy of the MLR model was assessed by using the coefficient of determination (R2), root mean square error (RMSE), and mean absolute error (MAE). The results obtained from the WQI method revealed that the river health diverse between good (26–50) and unfit (>100) categories, whereas it was not found excellent for drinking purpose in any monitoring site. The values of WQI were primarily affected by the high concentration of BOD (>2 mg/L), TC (>50 MPN/100 ml), and Fe (>0.5 mg/L) in the river water. Moreover, the study disclosed that MLR models performed well (R2 = 0.999, RMSE = 0.110 and MAE = 0.085) for the estimation of WQI. Therefore, the established MLR model can be applied for affordable, quick, and effective evaluation of river health for efficient river management practices.