A multivariate analysis of the spatial variations of water quality during high-flow period in the Chaobai River (Beijing, China) restored by reclaimed water

Abstract

Reclaimed water has demonstrated its broad applications in social construction to alleviate the contradiction of water shortage in Beijing, China. Using multivariate statistical analysis, the current study investigated the spatial variations of water quality in the Chaobai River restored by reclaimed water during the high-flow period. Hierarchical cluster analysis (CA) classified the 11 sampling sites into four clusters, namely most polluted, highly polluted, moderately polluted, and lowly polluted sections. The Kruskal-Wallis test showed that pH, TDS, EC, Ca2+, Mg2+, Cl~, SO42~, NO3 -N, and TN had significant spatial differences among four clusters (p < 0.05). Mean value of total nitrogen (TN) in the most polluted site exceeded the guideline (15 mg/L) of the Water Quality Standard for Scenic Environment Use, reaching 22.3 mg/L. Principal component analysis (PCA) extracted three principal components (PCs) accounting for 81.5% of the total variance in the data set of water quality. The three PCs reflected the chemical characteristics of reclaimed water, mineral pollution, and nutrient pollution, respectively. With the ordination biplot of sampling sites defined by the first and second PCs, PCA provided a classification of sampling sites based on the similarity of pollution sources, which supported the results of CA. The results revealed that water quality of the Chaobai River restored by reclaimed water was affected by untreated domestic and agricultural sewage with nitrogen and minerals being the main pollutants along the river basin. This study showed rivers restored by reclaimed water had significant spatial variations of water quality, demonstrating effectiveness of multivariate statistical methods on water quality analysis.