文献类型：期刊文献

英文题名：Co-occurrence correlations of heavy metals in sediments revealed using network analysis

作者：Liu, Lili[1,2];Wang, Zhiping[1,3];Ju, Feng[1];Zhang, Tong[1]

机构：[1]Univ Hong Kong, Environm Biotechnol Lab, Hong Kong, Hong Kong, Peoples R China;[2]E China Univ Sci & Technol, State Environm Protect Key Lab Environm Risk Asse, Shanghai 200237, Peoples R China;[3]Shanghai Jiao Tong Univ, Sch Environm Sci & Technol, Shanghai 200240, Peoples R China

年份：2015

卷号：119

起止页码：1305

外文期刊名：CHEMOSPHERE

收录：;EI(收录号：20143600044261);WOS:【SCI-EXPANDED(收录号:WOS:000347739600178)】；

基金：The authors would like to acknowledge the kind support of Ying Yang for the R script and operation. This research work is supported by the open database of Environmental Protection Department (EPD) of Hong Kong. And it is financed by projects of the National Natural Science Foundation of China (41001316) and the Fundamental Research Funds for the Central Universities (WB1214059). This study was also financially supported by General Research Fund (GRF) of Hong Kong (HKU7122/10E).

语种：英文

外文关键词：Network; Co-occurrence correlations; Normalization; Sediment; Heavy metals

摘要：In this study, the correlation-based study was used to identify the co-occurrence correlations among metals in marine sediment of Hong Kong, based on the long-term (from 1991 to 2011) temporal and spatial monitoring data. 14 stations out of the total 45 marine sediment monitoring stations were selected from three representative areas, including Deep Bay, Victoria Harbour and Mirs Bay. Firstly, Spearman's rank correlation-based network analysis was conducted as the first step to identify the co-occurrence correlations of metals from raw metadata, and then for further analysis using the normalized metadata. The correlations patterns obtained by network were consistent with those obtained by the other statistic normalization methods, including annual ratios, R-squared coefficient and Pearson correlation coefficient. Both Deep Bay and Victoria Harbour have been polluted by heavy metals, especially for Pb and Cu, which showed strong co-occurrence with other heavy metals (e.g. Cr, Ni, Zn and etc.) and little correlations with the reference parameters (Fe or Al). For Mirs Bay, which has better marine sediment quality compared with Deep Bay and Victoria Harbour, the co-occurrence patterns revealed by network analysis indicated that the metals in sediment dominantly followed the natural geography process. Besides the wide applications in biology, sociology and informatics, it is the first time to apply network analysis in the researches of environment pollutions. This study demonstrated its powerful application for revealing the co-occurrence correlations among heavy metals in marine sediments, which could be further applied for other pollutants in various environment systems. (C) 2014 Elsevier Ltd. All rights reserved.