Analysis of water quality using multivariate statistical methods in Duliujian River, China

Water quality assessments are essential for providing information to water resource management processes. The objective of this study was to investigate the internal correlation among water parameters, to analyze main water contamination and to identify the most polluted section along a seagoing river. Five water quality parameters (WT, EC, NH4+-N, NO3-N, NO2--N) were chosen to assess the water quality using multivariate statistical methods (Spearman's correlation, PCA, CA). PCA results showed 2 PCs can explain 78% of the total variation. The main water contamination was related to anthropogenic activities (domestic wastewater, industrial effluents and livestock operations) as well as natural condition (seawater intrusion). CA results gave 3 clusters by analyzing similarities of each section, indicating that the middle-downstream was the most polluted and the last section was mainly influenced by seawater intrusion.