Revealing the correlations between heavy metals and water quality, with insight into the potential factors and variations through canonical correlation analysis in an upstream tributary. (July 2018)

• Record Type: Journal Article
• Authors: Wei, Huaibin; Yu, Huibin; Zhang, Guangcai; Pan, Hongwei; Lv, Chunjian; Meng, Fansheng
• Is Part Of: Ecological indicators. Volume 90(2018)
• Page Start: 485
• Page End: 493
• Publication Date: 2018-07
• Subjects: Water quality -- Heavy metals -- Potential factor -- Canonical correlation analysis -- Hierarchical cluster analysis; Environmental monitoring -- Periodicals; Environmental management -- Periodicals; Environmental impact analysis -- Periodicals; Environmental risk assessment -- Periodicals; Sustainable development -- Periodicals; 333.71405
• DOI: 10.1016/j.ecolind.2018.03.037
• Languages: English
• ISSNs: 1470-160X

Abstract

Graphical abstract: Highlights: Insight into variations in water quality and heavy metals. Explore differences and similarities among mainstem, tributaries and reservoirs. Seek potential factors of heavy metals. Abstract: Water pollution is a worldwide problem that requires urgent attention and prevention. Multivariable analyses have been applied to water quality data for insight into potential factors, identification of pollution sources and determination of a dynamic set of physico-chemical interactions and equilibrium. Canonical correlation analysis (CCA) and hierarchical cluster analysis (HCA) were employed simultaneously to water quality data sets with 14 parameters measured at 9 sampling sites in an upstream tributary in March, June, September and December. The sampling sites were grouped into three clusters using case HCA with the rescaled squared Euclidean distance (SED) <5, proving that the decreasing order of the water quality level was approximately reservoirs > mainstem > tributaries. The potential factors of the water quality were sought using variable HCA with SED < 0.5, which included As, E. coli, F −, Zn, TP, CODCr, pH and Cd. According to correlation analysis, the heavy metals not only showed correlations with each other but stronger correlations with F − . The CCA of the sampling sites determined that E. coli, TEMP, CODCr, DO and pH were the potential factors differentiating the sites, revealing that natural processes deeply influenced the reservoirs, while anthropogenic activities deeply influenced the tributaries and mainstem. The CCA of the months indicated that the seasonal factors included E. coli, TEMP, CODCr, DO, pH and BOD5, demonstrating that June and September were considerably impacted by no-point source and natural pollution, while March and December by point source and natural pollution. The CCA of the heavy metals showed that F −, TP, E. coli and CODCr were potential factors, which could be associated with industrial activities and household wastewater.