Hydrological connectivity-mediated spatial vegetation patterns and regime shifts in drylands

• An improved index of connectivity is proposed to evaluate hydrological connectivity of watershed. • Regime shifts model is built to explore effectiveness and critical threshold of hydrological connectivity. • Empirical research displays two distinct high and low modes of hydrological connectivity. • Positive feedback loop between reduced vegetation cover and increased hydrological connectivity accelerates regime shifts. Hydrological connectivity (HC) provides essential information about ecosystem structure and function at large spatial scales, playing a pivotal role in predicting catchment ecosystem degradation in drylands. However, the relationships between HC and spatial vegetation patterns, and their interactions that contribute to regime shifts in catchment ecosystems, remain unclear. This study selects the Tarim River Basin in Northwest China as a typical study area to reveal the mutual effects and relationships between HC and vegetation patterns using the improved index of connectivity (IIC), spatial vegetation pattern indexes, potential analysis, and partial least squares path modeling (PLS-PM). The regime shifts model is used to explore the effectiveness and critical threshold of HC as an early warning indicator for ecosystem degradation. The findings show that: (1) the hydrological connectivity exhibits significant spatial variability due to the influences of the terrain characteristics and vegetation cover; (2) the basin displays two distinct high and low modes of hydrological connectivity. As vegetation cover decreases to approximately 0.23, the ecosystem undergoes a sudden and discontinuous shift from the low to high hydrological connectivity mode; and (3) the increased hydrological connectivity leads to marked changes in spatial vegetation pattern metrics, indicating the effectiveness of hydrological connectivity as an early warning signal of ecosystem degradation. The positive feedback loop between reduced vegetation cover and increased hydrological connectivity accelerates regime shifts, potentially leading to irreversible ecosystem degradation if not properly mitigated. This study lays an important foundation for identifying potential degradation hotspots at larger scales, and explores the potential of hydrological connectivity as an early warning indicator for regime transitions.