.png){kind=icon}
.png){kind=icon}
Return
Application of Hybrid Machine Learning for Groundwater Level Prediction: A Comprehensive Review
.png){kind=icon}
PRE
AI
DOI：10.1007/s11831-025-10447-w.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
One of nature’s most precious commodities in many countries is groundwater, which meets a significant portion of the needs for water in many countries. Groundwater dynamics is significantly influenced by long-term interaction of environmental, climatic and anthropogenic activities. Changes in rainfall patterns and temperature influencing groundwater recharge, increase in temperature causing increase in the evaporation rates and reducing the water available to replenish the groundwater supplies. At the same time, population expansion, agricultural development and urbanization are continuously increasing the demand for water and causing over-extraction, pollution potentially leading to groundwater depletion. These evolving conditions present complicated, site-specific problems requiring a comprehensive knowledge of groundwater movement in both existing and future setting. Predicting future ground water behaviour requires these varying factors to ensure water security and ecosystem and community resiliency. The review includes all Machine Learning model types used for Groundwater Level modelling from 2000 to 2024 (334 publications) and provides an overview of the elements of the papers that were investigated, encompassing the models’ kinds, data lengths, time intervals, input and output parameters, and performance standards applied. In addition, suggestions for potential future study topics are offered to further the understanding and accuracy of GWL prediction models.
Journal
IF:
12.1
Papers:
1.8K
Citations:
1.2W