Using multiple normalization procedures and contamination indices in the assessment of trace metal(loid) contents of the surface sediments of Lake Xingyun, southwestern China

Yang Zhang; Fengqin Chang; Donglin Li; Lizeng Duan; Xiaonan Zhang; Qi Liu; Haoyu Li; Youhong Gao; Fengwen Liu; Hucai Zhang

doi:10.1016/j.scitotenv.2023.164812

Using multiple normalization procedures and contamination indices in the assessment of trace metal(loid) contents of the surface sediments of Lake Xingyun, southwestern China

Sci Total Environ. 2023 Oct 1:893:164812. doi: 10.1016/j.scitotenv.2023.164812. Epub 2023 Jun 12.

Authors

Affiliations

¹ Institute for Ecological Research and Pollution Control of Plateau Lakes, School of Ecology and Environmental Science, Yunnan University, Kunming, Yunnan 650504, China.
² Institute for Ecological Research and Pollution Control of Plateau Lakes, School of Ecology and Environmental Science, Yunnan University, Kunming, Yunnan 650504, China. Electronic address: changfq@ynu.edu.cn.
³ Institute for Ecological Research and Pollution Control of Plateau Lakes, School of Ecology and Environmental Science, Yunnan University, Kunming, Yunnan 650504, China; Southwest United Graduate School, Kunming 650500, Yunnan, China. Electronic address: zhanghc@ynu.edu.cn.

PMID: 37315608
DOI: 10.1016/j.scitotenv.2023.164812

Abstract

Trace metal(loid) (TM) contamination, especially of aquatic ecosystems, is a global ongoing environmental problem. Fully and accurately determining their anthropogenic sources is a key requirement for formulating remediation and management strategies. Herein, we developed a multiple normalization procedure, combined with principal component analysis (PCA) to assess the influence of data-treatment and environmental factors on the traceability of TMs in surface sediments of Lake Xingyun, China. Multiple contamination indices, i.e., Enrichment factor (EF), Pollution Load Index (PLI), Pollution Contribution Rate (PCR) and Exceeded multiple discharge standard limits (BSTEL) suggest that contamination is dominated by Pb with the average EF exceed 3, especially within the estuary aeras with the PCR >40 %. The analysis demonstrates that the mathematical normalization of data, which adjusts it for various geochemical influences, has a significant effect on analysis outputs and interpretation. Routine (Log) and extreme (outlier removing) transformations may mask and skew important information contained within the original (raw) data, which create biased or meaningless principal components. Granulometric and geochemical normalization procedures can obviously identify the influence of grain size and other environmental impact on TM contents in principal components, but incorrectly explains the potential sources and contamination on different sites. Reducing the influence of organic matter by normalization allowed the mineralogy, bio-degradation, salinity, and anthropogenic sources associated with local sewage and anthropogenic smelting to be identified and interpreted more clearly. Moreover, the co-occurrence network analysis also confirms that the influence of grain size, salinity, and organic matter content are the primary factors controlling the spatial variability in the type and concentrations of TMs.

Keywords: Contamination indices; Data normalization; Geochemical tracing; Sediment; Trace metal(loid)s.