<p>Desertification poses a significant environmental challenge in arid and semi-arid regions, driven by a complex interaction of natural and human-induced factors that accelerate land degradation, jeopardize biodiversity, and impede socio-economic advancement. The Gabès region in southeastern Tunisia exemplifies this vulnerability, characterized by an arid climate with low annual precipitation (~ 200 mm/year) and high potential evapotranspiration (~ 2300 mm/year). The area is marked by extensive irrigated agriculture and intensive groundwater extraction, which exacerbates desertification processes. In this study, we applied a modified Mediterranean Desertification and Land Use (MEDALUS) model integrated with Geographical Information Systems (GIS) to assess the environmental sensitivity to desertification within the region. The analysis focused on four key indicators: Soil Quality Index (SQI), Groundwater Quality Index (GWQI), Vegetation Quality Index (VQI), and Climate Quality Index (CQI). We collected and analyzed soil and groundwater samples for various physicochemical properties, including electrical conductivity, pH, organic matter content, sodium adsorption ratio, and salinity. Additionally, remote sensing data and land cover maps were utilized to evaluate vegetation parameters. Each index was weighted and synthesized to derive the Desertification Sensitivity Index (DSI), which enabled a spatial representation of desertification risk across the landscape. Findings revealed that approximately 44% of the area is classified within the “very high” sensitivity category, with the remaining 56% in the “high” sensitivity class. Notably, coastal zones and intensively irrigated agricultural areas exhibited the most severe degradation, primarily due to soil salinization, groundwater overexploitation, and a lack of vegetation cover. To validate our findings, we implemented a Support Vector Machine (SVM) model, using classes derived from the DSI as training data. This model achieved an overall accuracy of 94.4% and a Kappa coefficient of 0.81, indicating a strong correspondence with the MEDALUS-based classification. Among the four indices analyzed, climate variables, vegetation type, and groundwater quality emerged as the predominant factors driving desertification dynamics in Gabès. Our study demonstrates that the integrated MEDALUS-GIS-SVM approach offers a robust and adaptable framework for quantitatively assessing desertification risk. It effectively integrates local environmental data, including soil and groundwater characteristics, facilitating the production of high-resolution spatial maps of desertification sensitivity. The insights gained from this research are crucial for policymakers and land managers, enabling targeted interventions for sustainable land use, enhanced water resource management, and land degradation mitigation in arid environments. Furthermore, this methodology holds the potential for application in other arid and semi-arid regions globally, particularly where relevant data on soil and groundwater quality are accessible.</p> Graphical Abstract <p></p>