Harnessing machine learning for assessing climate change influences on groundwater resources: A comprehensive review.

Climate change is a major concern for a range of environmental issues including water resources especially groundwater. Recent studies have reported significant impact of various climatic factors such as change in temperature, precipitation, evapotranspiration, etc. on different groundwater variables. For this, a range of tools and techniques are widely used in the literature including advanced machine learning (ML) and artificial intelligence (AI) approaches. To the best of the authors' knowledge, this review is one of the novel studies that offers an in-depth exploration of ML/AI models for evaluating climate change impact on groundwater variables. The study primarily focuses on the efficacy of various ML/AI models in forecasting critical groundwater parameters such as levels, discharge, storage, and quality under various climatic pressures like temperature and precipitation that influence these variables. A total of 65 research papers were selected for review from the year 2017-2023, providing an up-to-date exploration of the advancements in ML/AI methods for assessing the impact of climate change on various groundwater variables. It should be noted that the ML/AI model performance depends on the data attributes like data types, geospatial resolution, temporal scale etc. Moreover, depending on the research aim and objectives of the different studies along with the data availability, various sets of historical/observation data have been used in the reviewed studies Therefore, the reviewed studies considered these attributes for evaluating different ML/AI models. The results of the study highlight the exceptional ability of neural networks, random forest (RF), decision tree (DT), support vector machines (SVM) to perform exceptionally accurate in predicting water resource changes and identifying key determinants of groundwater level fluctuations. Additionally, the review emphasizes on the enhanced accuracy achieved through hybrid and ensemble ML approaches. In terms of Irish context, the study reveals significant climate change risks posing threats to groundwater quantity and quality along with limited research conducted in this avenue. Therefore, the findings of this review can be helpful for understanding the interplay between climate change and groundwater variables along with the details of the various tools and techniques including ML/AI approaches for assessing the impacts of climate changes on groundwater.