Scholarlink

<p>Red mud storage sites pose significant environmental and health risks due to the accumulation and dispersion of toxic heavy metals (HMs). This study presents a novel framework integrating Self-Organizing Maps (SOM) with the APCS–MLR model for source apportionment and health risk assessment of soil heavy metal contamination near a red mud site in Shandong, China. Unlike traditional receptor models or SOM alone, this approach combines SOM’s ability to reveal nonlinear spatial patterns with the quantitative strength of APCS–MLR to estimate source contributions. Additionally, we incorporate a source-oriented health risk assessment to directly link pollution sources with health impacts. These were further coupled with Monte Carlo simulation-based probabilistic health risk assessment and Getis–Ord Gi* spatial hotspot analysis to attribute health risks to specific sources and localize high-risk zones. Four major sources were identified: natural sources (46.34% ± 3.21%), red mud storage (20.08% ± 1.54%), coal combustion (2.84% ± 0.56%), and agricultural–industrial activities (2.81% ± 0.45%), along with 27.93% (± 4.12%) from unknown sources. Although red mud contributed less to total heavy metal loads, it dominated health risks, accounting for 63.84% of carcinogenic and 58.91% of non-carcinogenic risks, primarily due to its high arsenic content. This finding highlights the disproportionate health impact of red mud despite its lower contribution to the overall contamination. Spatial risk clusters corresponded strongly with red mud proximity and As concentrations. This study demonstrates the effectiveness of a multi-model integrative approach for resolving complex source–risk–space linkages and informing precision pollution control strategies.</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

Coupling self-organizing map and APCS-MLR for source apportionment and health risk assessment of soil heavy metals near a red mud storage site in Shandong, China

  • 期刊论文

  • 发表日期：

    2025年12月4日

查看全文
  • Mengqi Liu,
  • Dingming Xue,
  • Dong Xu,
  • Yaqi Jia,
  • Xiaofei Yan,
  • Mengcheng Wu,
  • Congcong Sun

摘要

Red mud storage sites pose significant environmental and health risks due to the accumulation and dispersion of toxic heavy metals (HMs). This study presents a novel framework integrating Self-Organizing Maps (SOM) with the APCS–MLR model for source apportionment and health risk assessment of soil heavy metal contamination near a red mud site in Shandong, China. Unlike traditional receptor models or SOM alone, this approach combines SOM’s ability to reveal nonlinear spatial patterns with the quantitative strength of APCS–MLR to estimate source contributions. Additionally, we incorporate a source-oriented health risk assessment to directly link pollution sources with health impacts. These were further coupled with Monte Carlo simulation-based probabilistic health risk assessment and Getis–Ord Gi* spatial hotspot analysis to attribute health risks to specific sources and localize high-risk zones. Four major sources were identified: natural sources (46.34% ± 3.21%), red mud storage (20.08% ± 1.54%), coal combustion (2.84% ± 0.56%), and agricultural–industrial activities (2.81% ± 0.45%), along with 27.93% (± 4.12%) from unknown sources. Although red mud contributed less to total heavy metal loads, it dominated health risks, accounting for 63.84% of carcinogenic and 58.91% of non-carcinogenic risks, primarily due to its high arsenic content. This finding highlights the disproportionate health impact of red mud despite its lower contribution to the overall contamination. Spatial risk clusters corresponded strongly with red mud proximity and As concentrations. This study demonstrates the effectiveness of a multi-model integrative approach for resolving complex source–risk–space linkages and informing precision pollution control strategies.