Data-Driven Assessment of Seawater Intrusion and Salinization in Coastal Aquifers under Climate and Human Pressures

Coastal aquifers are critical freshwater resources that are increasingly threatened by seawater intrusion driven by the combined impacts of climate change and intensive human activities. The introduction of harmful substances, rising sea levels and groundwater overextraction increase salinization processes. Coastal Groundwater contamination is a complex environmental challenge that requires robust, scalable tools for reliable assessment and prediction. This talk presents recent advances in data-driven approaches for groundwater contamination analysis, with a particular focus on the application of probabilistic, supervised and unsupervised learning techniques in coastal aquifer systems. Drawing on case studies from arid to semi-arid regions of Mexico and Peru, the presentation demonstrates how Bayesian networks, clustering algorithms and random forest models can be applied to multi-parameter hydrogeological and hydrochemical datasets to identify sources of salinization and improve the characterization of seawater intrusion processes. These approaches have proven effective in handling data scarcity and uncertainty. By incorporating artificial intelligence and probabilistic frameworks into hydrogeological assessments, the proposed methods enhance contaminant source identification, support the development of early-warning tools, and enable more informed decision-making. The talk concludes with recommendations for advancing groundwater sustainability through interdisciplinary, data-driven strategies.