Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

In The Science of the total environment ; h5-index 0.0

This study used non-supervised machine learning self-organizing maps (SOM) in conjunction with traditional multivariate statistical techniques (e.g., hierarchical cluster analysis, principle component analysis, Pearson's correlation analysis) to investigate spatio-temporal patterns of eutrophication and heavy metal pollution in the water supplying lakes (i.e., the Gao-Bao-Shaobo Lake, GBSL) of the eastern route of China's South-to-North Water Diversion Project (SNWDP-ER). A total of 28 water quality parameters were seasonally monitored at 33 sampling sites in the GBSL during 2016 to 2017 (i.e., 132 water samples were collected in four seasons). The results indicated that: 1) spatially, the western and south-western GBSL was relatively more eutrophic and polluted with heavy metals; and 2) temporally, the lakes suffered from high risks of heavy metal contamination in spring, but eutrophication in summer while water quality in winter was the best among the four seasons. Two main potential sources of pollution and transport routes were identified and discussed based on the pollution patterns. These findings contributed considerably to providing in-depth understanding of water pollution patterns, as well as potential pollution sources in the water-supplying region. Such understanding is crucial for developing pollution control and management strategies for this mega inter-basin water transfer project.

Guo Chuanbo, Chen Yushun, Xia Wentong, Qu Xiao, Yuan Hui, Xie Songguang, Lin Lian-Shin


Eutrophication, Heavy metals, Multivariate analysis, Self-organizing Map, South-to-North Water Diversion Project, Water quality safety