Speaker
Description
The Skoura oasis, located in the Ouarzazate region of southern Morocco, represents a fragile agro-ecosystem increasingly affected by land degradation processes. It is currently facing growing water stress due to climate change (frequent droughts, rising temperatures, declining groundwater levels) and the gradual abandonment of the traditional irrigation system (khettara).
This study aims to assess the influence of Land Surface Temperature (LST), precipitation, and soil moisture on desertification dynamics in the oasis between 1984 and 2024. Annual climate data (LST from Landsat, precipitation from CHIRPS, and soil moisture from ERA5) were extracted using Google Earth Engine. Each time series was binarized based on the median and compared to desertification classes derived from a desertification index calculated using spectral indices (non-desertified and severely desertified zones), through three similarity indicators: coincidence rate, Jaccard index, and Pearson correlation coefficient.
For non-desertified areas, LST shows a moderate inverse correlation (Jaccard ≈ 0.63, coincidence ≈ 70%, r ≈ –0.64), while soil moisture exhibits strong co-occurrence (Jaccard ≈ 0.79), and precipitation shows a weak influence (Jaccard ≈ 0.29). In severely desertified areas, LST is highly synchronized (Jaccard ≈ 0.69, coincidence ≈ 72%), and soil moisture coincides with degradation phases in 81% of the years, whereas precipitation plays a much weaker role (Jaccard ≈ 0.37, coincidence ≈ 21%, r ≈ –0.19). These findings suggest that the combination of warming and soil drying is the main climatic driver of desertification in Skoura, while precipitation exerts a much more indirect influence.