Résumé : Due to climate change, overgrazing, and deforestation, arid ecosystems are vulnerable to desertification and land degradation. As aridity increases, vegetation cover loses spatial homogeneity and self-organizes into heterogeneous vegetation patterns, a step before a catastrophic shift to bare soil. Several studies suggest that environmental inhomogeneities in time or space are crucial to understand these phenomena. Using a unified mathematical model and incorporating environmental inhomogeneities in space, we show how two branches of vegetation patterns create a hysteresis loop as the mortality level changes. In an increasing mortality scenario, one observes an equilibrium branch of high vegetation biomass that forms self-organized hexagonal-like patterns. However, when the mortality trend is reversed, one observes a branch with low biomass and no periodicity, where vegetation spots form disordered clusters instead of a hexagonal lattice. This behavior is supported by remote sensing and field observations and can be linked to climate change in arid ecosystems.