摘要：Large-scale modes of climate variability (or teleconnection patterns), such as the El Niño Southern Oscillation and the North Atlantic Oscillation, affect local weather worldwide. However, the response of terrestrial water and energy fluxes to these modes of variability is still poorly understood. Here, we analyse the response of evaporation to 16 teleconnection patterns, using a simple supervised learning framework and global observation-based datasets of evaporation and its key climatic drivers. Our results show that the month-to-month variability in terrestrial evaporation is strongly affected by (coupled) oscillations in sea-surface temperature and air pressure: in specific hotspot regions, up to 40% of the evaporation dynamics can be explained by climate indices describing the fundamental modes of climate variability. While the El Niño Southern Oscillation affects the dynamics in land evaporation worldwide, other phenomena such as the East Pacific–North Pacific teleconnection pattern are more dominant at regional scales. Most modes of climate variability affect terrestrial evaporation by inducing changes in the atmospheric demand for water. However, anomalies in precipitation associated to particular teleconnections are crucial for the evaporation in water-limited regimes, as well as in forested regions where interception loss forms a substantial fraction of total evaporation. Our results highlight the need to consider the concurrent impact of these teleconnections to accurately predict the fate of the terrestrial branch of the hydrological cycle, and provide observational evidence to help improve the representation of surface fluxes in Earth system models.
