The Emilia-Romagna region (Italy) hosts extensive agricultural and industrial activity, and densely populated urban areas. Groundwater serves as a crucial freshwater source, particularly during droughts, which are expected to become more frequent and intense.
This study estimates the evolution of groundwater conditions in part of Emilia-Romagna, considering climate change and human impacts, to assess the resilience of the regional multi-layered aquifer system to droughts and outline potential guidelines for long-term sustainable groundwater management. A numerical groundwater flow model and a random forest algorithm, implemented in MODFLOW 6 and R respectively, are applied to compare the performance of a physics-based and a machine learning method in simulating past and future groundwater levels, and to explore the benefits of their combination. Input data are sourced from the regional groundwater model by Arpae (Regional Agency for Prevention, Environment and Energy of Emilia-Romagna) and publicly available datasets on the Emilia-Romagna Region and Arpae repositories.
Both techniques are then used to analyze scenarios of reduced precipitation and altered pumping, focusing on their combined effects on the aquifer system. Results show the aquifer system’s vulnerability to future droughts. Increased pumping amplifies precipitation reduction effects, while lower abstraction partly mitigates them. Critical hotspots are identified, emphasizing the need for multi-scale approaches to develop effective mitigation and adaptation strategies.
The random forest algorithm provides insights into factors influencing groundwater head distribution, enhancing the groundwater model results interpretation and potential improvement. However, its lack of physical grounding limits its generalization potential. These findings highlight the value of integrating physics-based and machine learning methods to improve their performance, making a significant contribution to groundwater modeling.