We are pleased to highlight a new scientific publication involving researchers from the AI4Water Algeria and Egypt teams, focusing on the sustainable assessment and management of groundwater resources in the Mediterranean region.
The article, entitled:
“Precision Mapping of Groundwater Potential in the Middle Seybouse Basin, Northeastern Algeria, via a Hybrid Geospatial and Predictive Modeling Approach”
has been published in Earth (2026, Vol. 7, Issue 5, Article 168).
Authors: Lamia Zitoune, Hamza Bouguerra, Salah Eddine Tachi, Jazia Arrar, Osama Elsherbiny and Salah Elsayed.
The study focuses on the Middle Seybouse Basin in northeastern Algeria and develops a high-resolution approach for mapping groundwater potential by combining GIS-based geospatial information with advanced Machine Learning techniques.
Starting from 18 environmental conditioning factors, the researchers identified 10 key predictors related to topography, climate, land use, soil and hydrogeomorphological characteristics. Four Machine Learning approaches were then evaluated: Random Forest (RF), Support Vector Machine (SVM), Light Gradient Boosting Machine (LGBM), and a stacking ensemble.
The results show that the LGBM model achieved the highest predictive performance, with an AUC of 0.944, closely followed by the stacking ensemble. The resulting 30 m-resolution groundwater potential map classifies the study area into five categories, ranging from very low to very high groundwater potential.
Supporting sustainable groundwater management
The study provides a data-driven methodology for identifying areas with different groundwater potential and demonstrates how Artificial Intelligence, geospatial analysis and hydrogeological knowledge can be combined to support water-resource planning and decision-making.
The research is particularly relevant to AI4Water activities in Algeria, where the Coastal Constantinois and Seybouse Basins represent one of the project’s Mediterranean study areas.
It also highlights the value of international collaboration between Algerian and Egyptian researchers within the AI4Water community, strengthening the exchange of expertise in hydrogeology, geospatial analysis and predictive modelling.
These results contribute to the wider AI4Water objective of developing innovative and transferable approaches for smarter and more sustainable water management across Mediterranean regions.
Read the full open-access article
AI4Water – Artificial Intelligence for Smarter Water Management in the Mediterranean