Vegetation water content mediates decoupling between leaf-out and rainfall onset in the African dry tropics

Vegetation phenology and productivity in water-limited ecosystems are tightly coupled to plant hydraulic functioning, particularly the capacity to access and store water across seasonal dry periods. Across the African dry tropics, many woodland ecosystems initiate leaf green-up prior to the onset of seasonal rainfall, suggesting complex water-use mechanisms that are not directly observable through precipitation or surface soil moisture data alone. Understanding the hydraulic basis of these phenological strategies is critical for interpreting remotely sensed vegetation water signals and for predicting ecosystem responses to shifting rainfall regimes.

Here, we integrate satellite observations and reanalysis data to investigate how vegetation phenology and ecosystem productivity are mediated by plant water status across Africa. We identify three water-stress regimes based on the sensitivity of gross primary productivity (GPP) to rainfall frequency, intensity, and rainy season length, and assess the extent to which these regimes explain the widespread decoupling between rainfall onset and vegetation green-up across dry tropical woodlands. Furthermore, using observations of vegetation optical depth (VOD) as an integrative proxy for vegetation water content, we evaluate the role of plant-stored water in facilitating pre-rain leaf-out. We find that 64% of Africa's terrestrial ecosystems are subject to chronic water stress, and another 22% experience acute water stress. These acutely water-stressed regions initiate green-up when soil moisture is lower relative to chronically water-stressed regions, indicating decoupling between onset of rainfall and leaf-out. Notably, seasonal trajectories of LAI and VOD are asynchronous in regions with pre-rain green-up, consistent with the mobilization of plant-stored water to support early leaf-out. 

Our results demonstrate how satellite-derived vegetation water content metrics can reveal hydraulic strategies that decouple vegetation dynamics from surface moisture forcing. This work highlights the value of microwave-based observations for diagnosing plant hydraulic functioning at ecosystem scales and underscores vulnerabilities of water-limited ecosystems to shifts in rainfall timing and seasonality under climate change.