Défense de Thèse de Doctorat en “Artificial Intelligence” by Mr. Musa MUSTAPHA

The Euromed University of Fes (UEMF) is pleased to inform the public of

the doctoral thesis defense in “Artificial Intelligence”

The thesis defense will take place on Tuesday, December 9, 2025, at 09:00 am at the UEMF

Location: The Incubator's Main Hall (LOC001994)

The thesis will be presented by Mr. Musa MUSTAPHA

under the topic:

“Advancing Sustainable Agriculture with Artificial Intelligence and Remote Sensing Techniques: Applications in Irrigation Water Quality, Drought Assessment, and Crop Yield Prediction”

General Summary

The global population grows, and climate change intensifies, water scarcity and frequent droughts increasingly threaten agricultural productivity and food security, particularly in arid and semi-arid regions. This thesis leverages Artificial Intelligence (AI), Machine Learning (ML), and Remote Sensing (RS) to improve water resource management and promote sustainable farming in drought-prone areas. The research focuses on three key areas: drought assessment, irrigation water quality and quantity, and crop yield prediction. A comprehensive review of AI applications in irrigation water quality (IWQ) assessment identified critical research gaps and opportunities. Building on these insights, an advanced ML framework was developed for classifying irrigation water quality, achieving up to 98.5% accuracy. For water quantity, a model was designed to predict reference evapotranspiration (ETo) in data-scarce regions using both partial and complete climate datasets (ground and gridded (AgERA5)), with results showing high accuracy (R² > 0.95). However, the AgERA5 dataset slightly underestimated ground-based measurements. Drought severity was assessed using Sentinel-2 and MODIS data combined with ML techniques, revealing strong relationships between declining agricultural land, drought severity (via the Vegetation Health Index), and rainfall variability in semi-arid regions. A crop-specific drought stress assessment framework was developed based on remote sensing data and a hybrid 1D-CNN-GRU model to predict the Wheat Water Requirement Satisfaction Index (WRSI) using SMAP soil moisture and gridded climate data (AgERA5 and CHIRPS) for crop health monitoring. The model achieved an R² of 0.98. Field experiments validated these frameworks for wheat yield prediction, employing an attention-based CNN–GRU model that integrated AgERA5 data with in-situ measurements, yielding an R² of 0.95. These experiments confirmed that supplemental irrigation in semi-arid regions can significantly boost productivity while conserving freshwater. This research demonstrates the transformative potential of AI and RS for precise, scalable, and cost-effective agricultural water management. The methodologies advance precision agriculture and support the United Nations Sustainable Development Goals (SDGs), offering practical solutions for water-scarce regions to enhance food security and sustainability.

. This thesis will be presented to the jury members: Full name Grade Institution Role Prof. Abdelghafour MARFAK Full Professor Université Euromed de Fès Jury Chair Prof. Nabil El Akkad Full Professor Université Sidi Mohamed ben Abdellah Reviewer Prof. Omar ELOUTASSI Full Professor Moulay Ismail University, Errachidia Reviewer Prof. Zerhoune MESSAOUDI Full Professor École Nationale d'agriculture de Meknès Reviewer Prof. Mohammed OUANAN Full Professor Moulay Ismail University, Meknes Examiner Prof. Alaoui Chakib Assoc. Professor Université Euromed de Fès Examiner Prof. Ahmed El Hilali ALAOUI Full Professor Université Euromed de Fès Thesis Director Prof. Mhamed ZINEDDINE Asst. Professor Université Euromed de Fès Thesis co-Director