Identifying human activities causing water pollution based on microbial community sequencing and source classifier machine learning

Zhangmu Jing; Yi Zhang; Xiaoling Liu; Qingqian Li; Yanji Hao; Yeqing Li; Hongjie Gao

doi:10.1016/j.envint.2024.109240

Identifying human activities causing water pollution based on microbial community sequencing and source classifier machine learning

Environ Int. 2024 Dec 24:195:109240. doi: 10.1016/j.envint.2024.109240. Online ahead of print.

Authors

Zhangmu Jing¹, Yi Zhang², Xiaoling Liu³, Qingqian Li³, Yanji Hao⁴, Yeqing Li⁴, Hongjie Gao⁵

Affiliations

¹ State Key Laboratory of Environmental Criteria and Risk Assessment, Chinese Research Academy of Environmental Science, Beijing 100012, China; State Environmental Protection Key Laboratory of Estuarine and Coastal Environment, Chinese Research Academy of Environmental Science, Beijing 100012, China; State Key Laboratory of Pollution Control and Resource Reuse, College of Environmental Science and Engineering, Tongji University, Shanghai, 200092, China; School of Civil and Environmental Engineering, Nanyang Technological University, 639798, Singapore.
² State Key Joint Laboratory of Environment Simulation and Pollution Control (SKLESPC), School of Environment, Tsinghua University, Beijing 100084, China.
³ State Key Laboratory of Environmental Criteria and Risk Assessment, Chinese Research Academy of Environmental Science, Beijing 100012, China; State Environmental Protection Key Laboratory of Estuarine and Coastal Environment, Chinese Research Academy of Environmental Science, Beijing 100012, China.
⁴ State Key Laboratory of Heavy Oil Processing, Beijing Key Laboratory of Biogas Upgrading Utilization, College of New Energy and Materials, China University of Petroleum Beijing (CUPB), Beijing, 102249, China.
⁵ State Key Laboratory of Environmental Criteria and Risk Assessment, Chinese Research Academy of Environmental Science, Beijing 100012, China; State Environmental Protection Key Laboratory of Estuarine and Coastal Environment, Chinese Research Academy of Environmental Science, Beijing 100012, China; State Key Laboratory of Pollution Control and Resource Reuse, College of Environmental Science and Engineering, Tongji University, Shanghai, 200092, China. Electronic address: [email protected].

PMID: 39740270
DOI: 10.1016/j.envint.2024.109240

Abstract

Identifying and differentiating human activities is crucial for effectively preventing the threats posed by environmental pollution to aquatic ecosystems and human health. Machine learning (ML) is a powerful analytical tool for tracking human impacts on river ecosystems based on high-through datasets. This study employed an ML framework and 16S rRNA sequencing data to reveal microbial dynamics and trace human activities across China. The results revealed that the microbial assembly was mainly dominated by deterministic factors (environmental factors and interactions between species), and the metacommunity partition was significantly associated with human activities in both water and sediment (Chi-square test_wP = 1.93 × 10^-22; Chi-square test_sP = 6.00 × 10^-6). Human activities increased the vulnerability of interspecific occurrence networks and the influence of environmental factors on the OTUs similarity and phylogenetic distance. Combined of microbiological indices (MBIs), microbial relative abundance (MRA), and environmental and geographical indices (EGIs), the source classifier machine learning (SCML) algorithm was used to categorize five human activities (i.e., low human-impact, agricultural inputs, domestic inputs, industrial inputs, and dam construction). The SCML optimal configuration is (MBIs + MRA + EGIs) exhibited strong performance with Test_W R² of 0.882 and Test_S R² of 0.924. This study provides valuable insights for improving ecosystem management, supporting sustainable water resource management and advancing pollution mitigation efforts.

Keywords: 16S rRNA sequencing data; Human activities; Microbial communities; Pollution source tracing; Source classifier machine learning.