Water quality assessment of the Karasu River (Turkey) using various indices, multivariate statistics and APCS-MLR model

Abstract

Determining the water quality status of a river and accurately identifying potential pollution sources threatening the river are pillars in effective control of pollution and sustainable water management. In this study, water quality indices, multivariate statistics and absolute principal component score-multiple linear regression (APCS-MLR) were applied to evaluate the water quality of the Karasu River, the main tributary of the Euphrates River (Turkey). For this, 19 water quality variables were monitored monthly at eight stations along the river during one year. Based on the mean dissolved oxygen (DO), electrical conductivity (EC), nitrate-nitrogen (NO3–N), orthophosphate-phosphorus (PO4–P), total phosphorus (TP), ammonium-nitrogen (NH4-N), chemical oxygen demand (COD) and total nitrogen (TN) levels, most stations of the river had “very good” water status according to surface water quality criteria. Spatial cluster analysis (CA) divided eight stations into three regions as clean region, moderate clean region and very clean region. The mean values of Nutrient Pollution Index indicated that the river was “no polluted”. Similarly, Water Quality Index and Organic Pollution Index values indicated that the river water quality was between “good” and “excellent”. A minimum water quality index (WQImin) consisted of ten crucial parameters was not significantly different with the WQI based on all the 17 parameters. Discriminant analysis (DA) results showed that water temperature (WT), EC, chlorophyll-a (Chl-a), potassium (K), calcium (Ca), NO3–N and COD are the variables responsible for temporal changes, while WT, total dissolves solids (TDS), Chl-a, K, magnesium (Mg), Ca, NH4–N and COD are the variables responsible for spatial changes in the river water quality. Principal component analysis/factor analysis (PCA/FA) identified four potential sources, including anthropogenic, natural, seasonal and phytoplankton. Source apportionment in the APCS-MLR model revealed that seasonal and anthropogenic sources contributed 35.2% and 25.5% to river water quality parameters, respectively, followed by phytoplankton (21.4%) and natural sources (17.9%).