Plant hydraulics: a key for appropriate accounting of evapotranspiration in a warmer world

Abstract:

Transpiration, the dominant component of terrestrial evapotranspiration (ET), directly connects the water, energy, and carbon cycles. It is typically restricted by soil moisture availability and vapor pressure deficit (VPD), both of which are often correlated (lower soil moisture is generally accompanied by drier air) thus making it difficult to disentangle their individual impacts. Given that VPD is projected to rise globally with increased air temperature while the projected soil moisture changes are heterogeneous and uncertain, it is crucial for Land surface and Earth system models (LSEMs) to correctly account for the relative role of these stresses on transpiration fluxes.
Using a model–data fusion approach, here, we demonstrate that the common empirical approach used in most LSEMs to evaluate the ET response to soil moisture and VPD underestimates ET sensitivity to VPD and compensates by overestimating the sensitivity to soil moisture stress. In contrast, a hydraulic model that describes water transport through the plant better captures ET under high VPD conditions for wide-ranging soil moisture states. Our results highlight that the current generation of models, likely, underestimate the future impairment of ecosystem productivity and biosphere-atmosphere interactions by warming temperatures and increasing heatwaves. Furthermore, they also overestimate the reduction in soil moisture during droughts. The study underscores the need to incorporate plant hydraulics in the next generation of earth system models.