根据可解释的机器学习,识别关键特征变量和预测水分转移湖中的有害藻类繁殖
Yundong Wu1, Bo Xian1, Xiaowei Xiang1
1Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan, 430072, PR China; University of Chinese Academy of Sciences, Beijing, 100049, PR China.
Environmental research
|March 30, 2025
概括
预测有害藻类繁殖 (HABs) 对湖泊管理至关重要. 一个新的机器学习模型准确地预测了水分转移期间的HAB,有助于有效的缓解策略和湖泊健康.
科学领域:
- 环境科学 环境科学
- 生态生态学 生态生态学
- 机器学习 机器学习
背景情况:
- 有害的藻类繁殖 (HABs) 对全球的湖泊构成越来越大的环境威胁.
- 转移水是缓解HAB的关键策略,但由于环境条件的变化,其有效性变得复杂.
研究的目的:
- 开发一个强大的机器学习模型来预测湖泊中的HAB,在非水转移和水转移场景下都能准确执行.
- 确定影响HAB发生的关键环境特征,并解释模型的预测.
主要方法:
- 收集了2008-2022年的湖泊监测数据,包括非水分流和水分流期间.
- 与10个机器学习模型进行了比较,选择了CatBoost,因为它在预测HAB方面具有卓越的性能 (AUC=0.948).
- 利用SHapley添加式解释 (SHAP) 来解释模型可解释性和部分最小平方路径建模 (PLS-PM) 来分析水分转移影响.
主要成果:
- CatBoost模型,缩小到8个关键特征 (包括TP,TN,CODCr),在预测HAB方面表现出高准确度.
- 模型的解释揭示了特定的环境因素如何导致HAB形成.
- 发现转水间接地减轻了HABs,主要是通过营养稀释.
结论:
- 开发的可解释的CatBoost模型有效地预测了在不同水分转移条件下,在宜龙湖的HAB.
- 该模型为湖泊管理提供了实际应用的好处,并作为水分转移策略的指南.
- 这项研究为全球类似的环保湖泊管理和水分转移策略提供了宝贵的见解.
相关概念视频
01:18Steps in Outbreak Investigation
In the ever-evolving field of public health, statistical analysis serves as a cornerstone for understanding and managing disease outbreaks. By leveraging various statistical tools, health professionals can predict potential outbreaks, analyze ongoing situations, and devise effective responses to mitigate impact. For that to happen, there are a few possible stages of the analysis:
01:24Freshwater Microbial Ecology
Freshwater systems such as streams, rivers, and lakes exhibit distinct physical and biological characteristics that influence their microbial communities. These environments are broadly categorized into lotic systems—those with flowing waters like streams and most rivers—and lentic systems, which include still or slow-moving waters such as lakes, ponds, and marshes.In lentic systems, phytoplankton drive primary production, generating autochthonous organic carbon. In contrast, lotic systems...