An AI-Driven Prototype for Groundwater Level Prediction: Exploring the Gorgovivo Spring Case Study

Water is a vital yet increasingly endangered resource, that stands on the precipice of depletion and degradation, threatened by pollution, overexploitation, habitat alteration, and the looming spectre of climate change. Growing demand for water from various productive sectors and the escalating shifts in weather patterns are among the primary factors contributing to resource strains. Such factors lead to abrupt deterioration in terms of quantity and quality of water. Recently, technological intervention is playing a significant role in the mitigation of the stress on water resources leading to a better understanding of the dynamics behind consumption and replenishment. The substantial rise in the utilization of novel sensors is actively enhancing monitoring capabilities and facilitating data collection. There remains an ongoing requirement to enhance and advance methodologies for effectively analyzing the exponential surge in data. In this regard, the use of Artificial Intelligence (AI) towards supporting the management of water reserves can bring significant benefits in terms of the protection and sustainable utilization of the resource that is the basis of all life. In this paper, we propose an AI-based system for predicting Groundwater Level (GWL) focusing on Gorgovivo spring, located in province of Ancona, Italy as a case study. The predictive model was evaluated using the following criteria: Mean Average Error (MAE), Mean Squared Error (MSE) and correlation. Our case study concludes that the implementation of the prototype system produces valuable prediction results for GWL compared to other state-of-the-art approaches.