Use of Geographic Information Systems and Remote Sensing for Crop Analysis in a Saharan Environment: A Case Study of El Oued Province

Abstract

The Oued Souf region in southeastern Algeria has experienced significant agricultural expansion relying on center pivot irrigation, leading to severe groundwater depletion (a 30 m drop in the water table) and increased soil salinization. Although the Oued Righ region no longer administratively belongs to Oued Souf, it remains one of southern Algeria’s most important agricultural areas due to its oasis system and date palm groves. This study aimed to employ remote sensing and GIS techniques to monitor barley as a strategic crop and to analyze the spatiotemporal dynamics of vegetation cover over the period 2016–2026, while testing the reliability of the Normalized Difference Vegetation Index (NDVI) for agricultural decision support. The methodology integrated weekly field measurements of barley plant height from an experimental plot in Debila (February–May 2026) with NDVI values derived from Sentinel-2 imagery, together with a long-term time-series analysis of vegetation cover using seven spectral classes. The results showed a strong relationship between NDVI and barley vertical growth (R² = 0.905). NDVI peaked at 0.67 in week nine, coinciding with maximum vegetative development, then declined as physiological maturity began. At the regional level, the 10-year analysis revealed that drought periods (2017–2025) caused a contraction of higher NDVI classes (>0.4) and an expansion of lower classes, while the exceptionally wet year 2026 showed marked recovery. The moderate vegetation class (0.2–0.4) peaked in 2021 with 6.15 million pixels, and the dense vegetation class (0.4–0.6) peaked in 2025 with 2.05 million pixels. The study concludes that remote sensing and Geographic Information System GIS constitute an accurate and reliable alternative to costly field surveys. It recommends their adoption for agricultural and environmental monitoring, and the use of class-based frequency