Stomatal response to certain environmental factors: a comparison of models for subalpine trees in the Rocky Mountains

Abstract

Using the data of Kaufmann, we compared five mathematical models of stomatal response to photosynthetic photon flux density, leaf-to-air vapor pressure difference, relative humidity and temperature. Employing non-linear least squares techniques, each model was fitted to stomatal conductance data taken with a gas-exchange system for each of four Rocky Mountain subalpine trees (aspen, lodgepole pine, Engelmann spruce and subalpine fir). The resulting conductance models were then assessed on how well they reproduce measured rates of transpiration. The stomatal conductance models have the same general structure g 1=g n+g xƒ Q p ƒ Dƒ r where f represents a modifier that describes the stomatal response to a particular driving variable such that generally (but not always) 0⩽ f⩽1. The Michaelis-Menten equation is used for the f Q p modifier, while the f T and f D modifiers vary with each of the five different models considered. The first of these five models is adapted from Jarvis. The second is from Lohammer et al. The third is Farquhar's feedforward model. The fourth is derived from Farquhar (personal communication, 1989) and the fifth is the model of Ball et al. The results show that: (1) regardless of the specific form of f D or f T , the photosynthetic photon flux density is of primary importance, the leaf-to-air vapor pressure difference and/or relative humidity is of moderate importance, and the temperature is of minimal importance; (2) all models fit each of the four data sets well, but some models perform better than others; (3) all five models appear to be slightly biased toward underpredicting transpiration rates; (4) stomatal conductance of shade-tolerant species may be more strongly coupled to the leaf-to-air vapor pressure difference and less strongly coupled to the radiation environment than that of shade-intolerant species.