Soil and Atmospheric Drought Explain the Biophysical Conductance Responses in Diagnostic and Prognostic Evaporation Models Over Two Contrasting European Forest Sites.

Diagnosing and predicting evaporation through satellite‐based surface energy balance (SEB) and land surface models (LSMs) faces challenges due to the non‐linear responses of aerodynamic (ga) and stomatal conductance (gcs) to concurrent soil and atmospheric drought. Despite a soaring popularity to refine gcs formulation in LSMs by integrating soil‐plant hydraulics, SEB models often overlook the utility of gcs. This oversight is attributed to the overriding emphasis on reducing ga uncertainties and the lack of coordination between these two modeling communities. This disengagement between modeling communities poses a persistent challenge in understanding divergent evaporation estimates during intense soil‐atmospheric drought. Here we conducted a theoretical experiment over two contrasting European forest sites to examine the sensitivity of conductances and evaporative fluxes to a water‐stress factor (β‐factor), coupled with land surface temperature (LST) and vapor pressure deficit (representing soil and atmospheric drought proxy). Utilizing a non‐parametric diagnostic model (Surface Temperature Initiated Closure, STIC) and a prognostic model (Community Land Model, CLM5.0), the analysis revealed that the β‐factor, alongside different functional forms of conductances and the loose coupling of CLM5.0 conductances to LST, significantly influenced the response of the two models to soil and atmospheric drought. These discrepancies propagated in the estimates of evaporative fluxes between STIC and CLM5.0. The analysis reaffirms the need for a consensus on theory and models capturing the sensitivity of biophysical conductances to soil‐atmospheric drought interplay. It emphasizes the need for fostering collaboration between modeling communities to enhance the prediction of evaporation in complex environmental conditions. Plain Language Summary: The regulation of plant water loss through evaporation is governed by two key physical and biological attributes: aerodynamic and stomatal conductance. The magnitude and variability of these conductances and their degree of regulation on evaporation is heavily dependent on how the conductances respond to the conjugate dryness from the soil and the atmosphere. Because these conductances cannot be typically measured at a large scale, the majority of the global evaporation models use different mechanistic functions to estimate them, which involves many empirical parameters. Such methods do not fully capture the evaporation variability of ecosystems during water stress, leading to large errors in water cycle monitoring. Our model‐based synthetic experiment shows how two structurally different models with different functional forms of the conductances respond very differently to emerging soil‐atmospheric water stress and produce divergent estimates of evaporation in a variety of dry and wet conditions. This study not only provides insights into the role of conjugate effects of soil and atmospheric drought in explaining the conductances and evaporation variability but also suggests a novel perspective for reconciling predictive and remote sensing evaporation models, benefiting water management, testing plant water use theories, and understanding land‐atmosphere interactions. Key Points: Diagnostic (STIC) and prognostic (CLM5.0) evaporation models show distinct levels of sensitivity to water stressSTIC and CLM5.0 evaporation models better agree in simulating the energy fluxes as compared to underlying biophysical conductanceMajor differences in the simulated stomatal conductance are due to divergences in the physiological assumptions of the two evaporation modeling approaches [ABSTRACT FROM AUTHOR]