Contrasting Ecosystem Responses to the Dynamics of Compound Climate Extremes

The ecological impact of compound climate extremes often exceeds that of individual events; however, the cumulative responses of vegetation to the dynamics of these extremes remains unexplored. In this study, we utilized the Normalized Difference Vegetation Index (NDVI) as a proxy for vegetation to investigate the cumulative responses of global vegetation greenness to the duration, frequency, and magnitude of three types of events: compound hot-dry extremes (CHDE), extreme heat (EHE), and extreme drought (EDE) from 1982 to 2020. Our results reveal a pronounced increasing trend in CHDE across transitional climate zones, where more persistent and stronger events occur in densely vegetated regions. This was characterized by a strong positive correlation between CHDE and vegetation dynamics in these zones, likely driven by intensified land–atmosphere feedbacks, with vegetation response maximized at a 4-month timescale, identified from cross-correlation analysis. Meanwhile, the Amazon and Congo basins emerge as hotspots for heat-related extremes, where EDE and CHDE exhibit greater persistence. At high latitudes of the Northern Hemisphere, vegetation exhibits a robust sensitivity to temperature-driven events, particularly under EHE dominance, where the response shows no temporal lag, indicating an immediate physiological reaction to thermal relief or stress. Furthermore, we observe a global divergence in climate risk and response: while arid regions are experiencing a significant warming trend, humid regions are increasingly threatened by desiccation (drying). Notably, vegetation productivity in humid ecosystems shows a predominant temporal response to EDE, suggesting a higher initial resistance but eventual vulnerability to prolonged water deficits. Finally, we employed machine learning to identify the primary drivers behind these temporal vegetation responses and elucidate how their interactions shape ecosystem sensitivity. These findings underscore the critical role of compound extremes in modulating vegetation dynamics and provide insights for enhancing ecosystem resilience under future climate scenarios.