Scholarlink

<p>Aquatic ecosystems—critical to biodiversity and human health—face escalating threats from pollution, eutrophication, and climate change. Aquatic bioindicators, such as macroinvertebrates, phytoplankton, and microbial communities, are critical tools for assessing ecosystem health and detecting anthropogenic pressures. By reviewing the literature over&#xa0;the past decade, we systematically explored how artificial intelligence (AI) methods such as machine learning (ML), deep learning (DL), and hybrid models have transformed the detection, prediction, and management of aquatic ecosystems. By synthesizing cutting-edge applications and addressing existing barriers, this review highlights AI’s potential in achieving intelligent aquatic ecosystem management. However, significant gaps remain, particularly regarding the lack of standardization in data processing, model development, and validation frameworks across studies. Future advancements in multi-omics integration, climate-resilient modeling, and participatory AI systems are exceped to in transitioning from reactive monitoring to proactive stewardship, ensuring the sustainability of water resources in an era of global environmental change.</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

Recent Advances of Artificial Intelligence in Aquatic Bioindicators and Ecological Assessment

  • 期刊论文

  • 发表日期：

    2025年12月18日

查看全文
  • Kaiming Hu,
  • Qingyu Xu,
  • Kangyun Zhu,
  • Zhe Wang,
  • Yingxue Chu,
  • Yifan Qian,
  • Xianwu Zhang,
  • Binhao Wang,
  • Hangjun Zhang

摘要

Aquatic ecosystems—critical to biodiversity and human health—face escalating threats from pollution, eutrophication, and climate change. Aquatic bioindicators, such as macroinvertebrates, phytoplankton, and microbial communities, are critical tools for assessing ecosystem health and detecting anthropogenic pressures. By reviewing the literature over the past decade, we systematically explored how artificial intelligence (AI) methods such as machine learning (ML), deep learning (DL), and hybrid models have transformed the detection, prediction, and management of aquatic ecosystems. By synthesizing cutting-edge applications and addressing existing barriers, this review highlights AI’s potential in achieving intelligent aquatic ecosystem management. However, significant gaps remain, particularly regarding the lack of standardization in data processing, model development, and validation frameworks across studies. Future advancements in multi-omics integration, climate-resilient modeling, and participatory AI systems are exceped to in transitioning from reactive monitoring to proactive stewardship, ensuring the sustainability of water resources in an era of global environmental change.