Spatially Explicit Assessment of Natural Capital Condition in Arid Mediterranean Ecosystems: A Multi-Source Geospatial Framework for Detecting Climate-Induced Tipping Points in Argania spinosa Woodlands

Quantifying the resilience of arid socio-ecosystems under intensified climatic and anthropogenic pressures remains a critical challenge for global ecosystem accounting. This study develops and applies a spatially explicit Natural Capital Index (ICN) for the Argan-dominated landscapes of Essaouira Province, Morocco, over a 25-year period (2000–2025). The framework is fully aligned with the System of Environmental-Economic Accounting for Ecosystem Accounting (SEEA-EA) and integrates twelve biophysical and anthropogenic indicators processed at 300 m resolution on Google Earth Engine. Our results reveal a pronounced intra-provincial heterogeneity, with ICN scores spanning over 37 units between preserved inland woodlands and degraded coastal belts. The provincial trajectory displays a distinct three-phase temporal dynamic: progressive improvement (2000–2011), drought-induced regression (2012–2016), and a post-drought partial stabilization phase characterized by ecological hysteresis. Random Forest analysis identifies Mean NDVI (35.2%) and Summer Land Surface Temperature (22.1%) as the dominant predictors of ecosystem condition, underscoring the acute vulnerability of Argania spinosa to thermal and hydric stress. Mann-Kendall trend analysis highlights a widening ecological divergence, with 19.3% of communes exhibiting statistically significant declines, primarily along the coastal corridor, contrasting with 40.4% showing improvement. This operational, reproducible workflow offers a transferable template for national SEEA-EA reporting and restoration prioritization, providing critical insights into the non-linear response of Mediterranean drylands to climate change scenarios.