Impact of climate change on groundwater levels in the Iberian Peninsula

Groundwater represents a strategic freshwater resource for multiple sectors, including drinking water, agriculture production, and ecosystem services. The Mediterranean Basin is a well-known water-scarce region that is increasingly relying on groundwater use, especially during drought periods. Many areas in the Mediterranean region are already facing water stress due to increasing demand and limited resources. Climate change is likely to exacerbate these issues, as it is expected to lead to more frequent and severe drought conditions in some areas as well as irregular rainfall in others. Due to the growing availability of data and computational processing capabilities nowadays, deep learning models are seeing an increase in popularity. In this study, we attempted to create 92 location-specific Convolutional Neural Network (CNN) models in wells spatially distributed over the Iberian Peninsula to estimate groundwater levels until the end of the century. Our models use monthly precipitation and temperature data as input variables. Specifically, we considered cumulative precipitation for 3, 6, 12, 18, 24, and 36 months to account for the recharge time lag between precipitation and groundwater changes. Once trained using historical precipitation and temperature records, the CNNs were applied to assess the influence of climate change on groundwater levels. For future climate projections, an ensemble of six combinations of distinct General Circulation Models (GCMs) and Regional Climate Models (RCMs) was considered under two Representative Concentration Pathways (RCPs): the RCP4.5 and RCP8.5. Our preliminary results revealed a more consistent decline in groundwater levels in the southwest region of the Iberian Peninsula under the RCP8.5 scenario, while a general more constant groundwater level under the RCP4.5 scenario has been detected towards the end of the century. Detailed results of this study will be shared and discussed during the event.