Scholarlink

<p>Climate change disrupts groundwater levels (GWL) by modifying precipitation patterns, reducing recharge rates, and limiting water availability. Rising temperatures and evolving weather patterns further degrade surface and groundwater quality. These changes exacerbate competition for water resources, heightening allocation challenges and ecological disruptions. Groundwater fluctuations adversely affect ecosystems, causing habitat disturbances and biodiversity loss. This study explores the impacts of climate change on GWL using machine learning techniques to analyze 9,430 time series data points (1993–2021) from Northern China. Four distinct classes of top-performing machine learning models were evaluated. The CNN model demonstrated superior performance, achieving an R² value of 0.9924 and an RMSE of 0.1832, highlighting its efficacy in processing complex patterns. Pearson correlation analysis revealed that Average Annual Precipitation (AAP), Average Soil Moisture (ASM), and Evapotranspiration (EV) positively influence GWL, while Severe Wet Potential (SWP), Severe Drought Potential (SDP), and Temperature (T) exhibit negative correlations. Feature ranking identified AAP as the most critical factor for groundwater recharge, followed by ASM and EV, which also play significant roles in groundwater dynamics. These findings provide a robust understanding of the key drivers influencing groundwater recharge and storage, offering valuable insights to inform sustainable water resource management in the context of climate 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

Data-Driven Insights into Climate Change Effects on Groundwater Levels Using Machine Learning

  • 期刊论文

  • 发表日期：

    2025年1月27日

查看全文
  • Xinyong Lu,
  • Zimo Wang,
  • Menghao Zhao,
  • Songzhe Peng,
  • Song Geng,
  • Hamzeh Ghorbani

摘要

Climate change disrupts groundwater levels (GWL) by modifying precipitation patterns, reducing recharge rates, and limiting water availability. Rising temperatures and evolving weather patterns further degrade surface and groundwater quality. These changes exacerbate competition for water resources, heightening allocation challenges and ecological disruptions. Groundwater fluctuations adversely affect ecosystems, causing habitat disturbances and biodiversity loss. This study explores the impacts of climate change on GWL using machine learning techniques to analyze 9,430 time series data points (1993–2021) from Northern China. Four distinct classes of top-performing machine learning models were evaluated. The CNN model demonstrated superior performance, achieving an R² value of 0.9924 and an RMSE of 0.1832, highlighting its efficacy in processing complex patterns. Pearson correlation analysis revealed that Average Annual Precipitation (AAP), Average Soil Moisture (ASM), and Evapotranspiration (EV) positively influence GWL, while Severe Wet Potential (SWP), Severe Drought Potential (SDP), and Temperature (T) exhibit negative correlations. Feature ranking identified AAP as the most critical factor for groundwater recharge, followed by ASM and EV, which also play significant roles in groundwater dynamics. These findings provide a robust understanding of the key drivers influencing groundwater recharge and storage, offering valuable insights to inform sustainable water resource management in the context of climate change.