Stomatal Decoupling From Photosynthesis Under High Temperatures Is Consistent With Stomatal Optimisation

ABSTRACT

Stomatal pores on plant leaves regulate the gain of carbon through photosynthesis and the loss of water through transpiration. Through their responses to environmental conditions, stomata can constrain plant productivity and transpiration fluxes, exerting a strong control on climate feedbacks over land. Although mechanistic modelling of stomata remains a challenge, semi‐empirical and optimisation models have been successfully applied to improve the simulation of land‐atmosphere fluxes of water and carbon. Optimisation approaches assume that some aspect of plant function, such as photosynthesis or growth rate, is optimised with respect to an environmental constraint, such as available soil water. Both optimisation models and semi‐empirical models predict that stomatal conductance will increase in concert with rising photosynthetic rates as temperatures approach a thermal optimum, beyond which declines in both photosynthesis and stomatal conductance are expected. However, a growing number of experiments have found that while photosynthesis declines beyond its thermal optimum, stomatal conductance often continues to increase at high temperatures. Early modelling work suggests that this phenomenon can be captured and explained by an optimal thermoregulation strategy via increased evaporative cooling at the leaf surface. However, many stomatal conductance models that are embedded within climate and Earth System Models do not correctly account for this feedback and so cannot capture observed decoupling. Here, we demonstrate that if leaf temperature is calculated iteratively outside the optimisation scheme, as is commonly done in Earth System Models, stomatal decoupling will not be captured. However, by calculating leaf temperature in parallel with optimal stomatal conductance, we find that we are able to capture observed decoupling and improve predicted leaf temperature and gas exchange. Correctly implementing the leaf energy balance equation within stomatal optimisation models will be essential for capturing high temperature responses of forests across the globe.