Scholarlink

<p>To investigate the sources and distribution characteristics of groundwater pollutants in an abandoned mine in Zhangqiu City, Shandong Province, 27 groundwater samples collected from the site were tested for potential toxic elements (PTEs). Pollution characteristics and levels were assessed using ion concentration statistical analysis and the Nemero index method. Combined with correlation analysis, principal component analysis, Positive Matrix Factorization (PMF) model, and Self-Organizing Map (SOM) model to establish a multi-model fusion analytical framework for quantitative source identification. Results indicate: (1) All PTEs in the study area exceeded Class III groundwater quality standards (GB/T 14848–2017), with 59.3% of samples exhibiting severe contamination. Ni and Pb demonstrated high single-factor pollution indices. (2) The PMF model identified three primary pollution sources: pollution accident sources (39.97% contribution rate), industrial and transportation emissions (30.65%), and historical mining activities (29.37%); (3) The SOM model partially corroborated the PMF findings, further revealing spatial distribution patterns: pollution incident sources exhibited a continuous gradient distribution, industrial and transportation emissions formed localized hotspots, and historical mining activities showed a dual-center distribution pattern. These results provide scientific basis for targeted remediation of groundwater pollution in abandoned mines.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Identification of Potential Toxic Element Pollution Sources in Mine Groundwater Based on the PCA-PMF-SOM Model

  • 期刊论文

  • 发表日期：

    2026年3月13日

查看全文
  • Ding Guantao,
  • Han Yu,
  • Wei Shanming,
  • Li Zhe,
  • Liu Yuxiang,
  • Chen Liang,
  • Jiang Yumin,
  • Gao Shuai,
  • Xiao Jun

摘要

To investigate the sources and distribution characteristics of groundwater pollutants in an abandoned mine in Zhangqiu City, Shandong Province, 27 groundwater samples collected from the site were tested for potential toxic elements (PTEs). Pollution characteristics and levels were assessed using ion concentration statistical analysis and the Nemero index method. Combined with correlation analysis, principal component analysis, Positive Matrix Factorization (PMF) model, and Self-Organizing Map (SOM) model to establish a multi-model fusion analytical framework for quantitative source identification. Results indicate: (1) All PTEs in the study area exceeded Class III groundwater quality standards (GB/T 14848–2017), with 59.3% of samples exhibiting severe contamination. Ni and Pb demonstrated high single-factor pollution indices. (2) The PMF model identified three primary pollution sources: pollution accident sources (39.97% contribution rate), industrial and transportation emissions (30.65%), and historical mining activities (29.37%); (3) The SOM model partially corroborated the PMF findings, further revealing spatial distribution patterns: pollution incident sources exhibited a continuous gradient distribution, industrial and transportation emissions formed localized hotspots, and historical mining activities showed a dual-center distribution pattern. These results provide scientific basis for targeted remediation of groundwater pollution in abandoned mines.