Drought and vegetation dynamics in dryland of arganeraie biosphere reserve in Morocco: toward an understanding of the impact of environmental shocks on the Agro-sylvo-pastoral system

Agroforestry systems are critical for building resilience to socio-environmental shocks, especially in drought-prone arid and semi-arid areas. This study the ecological impact of drought on the Mediterranean Globally Important Agricultural Heritage Systems (GIAHS) site of Ait Saoub-Ait Mansour in Morocco. Machine learning-based random forest (RF) classification and change detection were used to explore the change in land use classes such as built-up areas, agricultural areas, woodland, and bare land. Spatiotemporal analysis of temperature, precipitation, and Precipitation-Evapotranspiration Index (SPEI) is performed using Mann-Kendall and Sen’s slope test for the period 1983–2020. Additionally, the Normalised Difference Vegetation Index (NDVI) is analyzed from 2000 to 2020 for some specified crop species such as Argan tree (Argania spinosa (L.) Skeels), cereal, palm tree, and cereal mixed. The results show a decrease in the area of agricultural and woodland and a consistent increase in drought (decrease in SPEI-6 and SPEI-12) over time. Annual and seasonal trends in NDVI for Argan tree, cereal, palm tree, and mixed cereal are negative. These findings provide consistent evidence of the ongoing degradation of this ecosystem and the changing climate. The resilience implications are discussed to inform socio-economic actors and stakeholders.