Receptor Modeling Application on Surface Water Quality and Source Apportionment

Abstract

There is a need for regular monitoring of river water quality to determine specific pollutants in order to aid amelioration schemes. In this study, Principal Component Analysis (PCA) was applied to eighteen water quality parameters; pH, conductivity, dissolved oxygen(DO), turbidity, temperature, total dissolved solids (TDS), total solids (TS), total hardness (TH), biochemical oxygen demand (BOD), carbon dioxide (CO2), ammonia (NH3), nitrate (NO3-), chloride (Cl-), lead (Pb), iron (Fe), chromium (Cr), copper (Cu) and manganese (Mn) to identify major sources of water pollution of river Asa. The generated Principal Components (PCs) were used as independent variables and the water quality index (WQI) as the dependent variable to predict the contribution of each of the sources using the multiple linear regression model (MLR). The PCs results showed that the sources of pollution are storm water runoff, industrial effluent, erosion and municipal waste, while MLR identified storm water runoff (0.786) and industrial effluent (0.241) as the respective major contributors of pollution. The study showed that the PC-MLR model gives a good prediction (R2=0.8) for the water quality index.