Rabat — A group of Moroccan researchers has developed an artificial intelligence system that could help solve critical water shortages in oases, offering new hope for communities facing severe groundwater depletion.
Adil Moumane, a researcher at Ibn Tofail university in Kenitra, presented the study at the second edition of the International Congress on Oases and Date Palm (CIO), held May 22-23 in Ouarzazate.
This year’s edition focused on the resilience and adaptation of oases ecosystems to global changes, particularly climate change.
The research combines machine learning with geographic technology to identify optimal locations for water collection infrastructure in arid regions.
In an interview with Morocco World News (MWN), Moumane said: “We use Geo AI to identify areas where we can build construction that can collect the rainwater.”
“We observed that in the area of Zagora, for example, there is a big decrease of the groundwater table.”
The research focused on Morocco’s Middle Draa Valley, a region facing severe aquifer stress.
Moumane’s team developed a system that analyzes ten environmental factors, including soil conditions, rainfall patterns, elevation, and vegetation coverage, to identify where water collection structures would work best.
Read also: Morocco’s ANDZOA Chief Calls for Urgent Action to Protect Argan, Oases Ecosystems
The AI system tested six machine learning approaches and found that LightGBM, a gradient boosting framework that uses tree-based learning algorithms, performed best, achieving 90% accuracy in predicting optimal groundwater recharge zones. The system identified soil permeability, elevation, and proximity to streams as the most important factors for successful water collection.
Morocco’s oasis regions face mounting pressure from climate change and population migration. Traditional communities that have depended on these water sources for generations now struggle as water tables drop to dangerous levels.
“The Oases area is under pressure of climate change and also under the pressure of different other global changes like migration of local people,” Moumane noted.
The researcher’s framework offers a data-driven solution using satellite imagery and advanced algorithms. Local authorities can make informed decisions about water infrastructure investments before building expensive projects.
“Recent extreme rainfall during the 2024-2025 season in southeastern Morocco highlighted both challenges and opportunities these regions face,” he told MWN, explaining that “while sudden floods cause damage, they also represent precious water that could be captured with proper infrastructure placement.
“This research pioneers the application of machine learning and deep learning on geospatial data for groundwater mapping in data-limited contexts,” Moumane explained.
The framework’s modular design allows adaptation for arid regions worldwide, potentially helping millions facing similar water challenges.
Conferences like CIO provide a platform for innovations like Moumane group’s AI system to offer practical solutions to preserve ancient oasis ecosystems for future generations, particularly in the backdrop of climate change pressing challenges.
Read also: Morocco-UAE Partnership Strengthens Date Palm Industry Development
