A deep learning-based hybrid model of global terrestrial evaporation

Abstract

Terrestrial evaporation ( E) is a key climatic variable that is controlled by a plethora of environmental factors. The constraints that modulate the evaporation from plant leaves (or transpiration, E _t ) are particularly complex, yet are often assumed to interact linearly in global models due to our limited knowledge based on local studies. Here, we train deep learning algorithms using eddy covariance and sap flow data together with satellite observations, aiming to model transpiration stress ( S _t ), i.e., the reduction of E _t from its theoretical maximum. Then, we embed the new S _t formulation within a process-based model of E to yield a global hybrid E model. In this hybrid model, the S _t formulation is bidirectionally coupled to the host model at daily timescales. Comparisons against in situ data and satellite-based proxies demonstrate an enhanced ability to estimate S _t and E globally. The proposed framework may be extended to improve the estimation of E in Earth System Models and enhance our understanding of this crucial climatic variable.

Abstract

Global evaporation is a key climatic process that remains highly uncertain. Here, the authors shed light on this process with a novel hybrid model that integrates a deep learning representation of ecosystem stress within a physics-based framework.