Lagged precipitation effects on plant production across terrestrial biomes.

Precipitation effects on plant carbon uptake extend beyond immediate timeframes, reflecting temporal lags between rainfall and plant growth. Mechanisms and relative importance of such lagged effects are expected to vary across ecosystems. Here we draw on an extensive collections of productivity proxies from long-term ground measurements, satellite observations and model simulations to show that preceding-year precipitation exerts a comparable influence on plant productivity to current-year precipitation. Statistically supported lagged precipitation effects are detected in 13.4%-19.7% of the grids depending on the data source. In these sites, preceding-year precipitation positively controls current-year plant productivity in water-limited areas, while negative effects occur in some wet regions, such as tropical forests. While aridity emerges as the main driver of this spatial variability, machine learning-based spatial attribution also indicates interactions among plant traits, climatic conditions and soil properties. We also show that soil water dynamics, plant phenology and foliar structure might mediate lagged precipitation effects across time. Our findings highlight the role of preceding-year precipitation in global plant productivity.