Stomatal optimization based on xylem hydraulics (SOX) improves land surface model simulation of vegetation responses to climate

Abstract

Land surface models (LSMs) typically use empirical functions to represent vegetation responses to soil drought. These functions largely neglect recent advances in plant ecophysiology that link xylem hydraulic functioning with stomatal responses to climate.We developed an analytical stomatal optimization model based on xylem hydraulics (SOX) to predict plant responses to drought. Coupling SOX to the Joint UK Land Environment Simulator (JULES) LSM, we conducted a global evaluation of SOX against leaf-and ecosystemlevel observations. SOX simulates leaf stomatal conductance responses to climate for woody plants more accurately and parsimoniously than the existing JULES stomatal conductance model. An ecosystem-level evaluation at 70 eddy flux sites shows that SOX decreases the sensitivity of gross primary productivity (GPP) to soil moisture, which improves the model agreement with observations and increases the predicted annual GPP by 30% in relation to JULES. SOX decreases JULES root-mean-square error in GPP by up to 45% in evergreen tropical forests, and can simulate realistic patterns of canopy water potential and soil water dynamics at the studied sites.SOX provides a parsimonious way to incorporate recent advances in plant hydraulics and optimality theory into LSMs, and an alternative to empirical stress factors.

Plant hydraulics play a critical role in Earth system fluxes

“…Eller et al . (2020) and Sabot et al . (2020) implement slightly different versions of a hydraulic‐based stomatal optimization put forward by Wolf et al .…”

“…Two recent studies published in this issue of New Phytologist – Eller et al . (2020; pp. 1622–1637) and Sabot et al .…”

“…Eller et al . (2020) use an analytical approximation of this stomatal algorithm in the JULES land‐surface model and test the model at 70 eddy flux sites around the world. Sabot et al .…”

“…Another drawback of incorporating plant hydraulics can be increased computation time, but this is not as important due to rapid development of computation capacity and approaches such as followed by Eller et al . (2020) that leverages an analytical approximation. In addition, the failure of plant hydraulic transport has been broadly observed as crucial to predicting mortality risk during drought and ecosystem resilience (Adams et al , 2017; Anderegg et al , 2018a; Venturas et al , 2018), suggesting that improved hydraulic physiology might improve multiple important aspects of biosphere models.…”

Optimal carbon partitioning reconciles the apparent divergence between optimal and observed canopy profiles of photosynthetic capacity

“…where  controls the slope of the function at L = 50 (large  = steep slope). The formulation represented by Eqns 3-4 is very similar to those adopted in recent optimization based models of stomatal conductance (e.g., Wolf et al, 2016;Sperry et al, 2017;Eller et al, 2020). However, Eqn 4 is not the hydraulic vulnerability curve; instead, it represents the risk of catastrophic desiccation posed by allowing L to reach a given value.…”