Response of stomatal conductance of two tree species to vapor pressure deficit in three climate zones

Abstract

Stomatal behavior is a central topic of plant ecophysiological research under global environmental change. However, the physiological mechanism controlling the response of stomata to vapor pressure deficit (VPD) or relative humidity (RH) has been inadequately understood till now. In this study, responses of stomatal conduc- tance (gs) to VPD in two species of trees (Fraxinus chinensis Roxb., Populus alba L. var. pyramidalis Bge.) in three different climate zones (Jinan with typical warm humid/semi-humid climate, Urumqi with temperate continental arid climate and Turpan with extreme arid desert climate) were measured. Levels of two phytohormones (abscisic acid, ABA; indole-3-acetic acid, IAA) in the leaves of the two tree species at these three sites were also measured by high performance liquid chromatography. The results showed that the responses of gs to an increasing VPD in these two tree species at the three sites had peak curves which could be fitted with a Log Normal Model (gs=a·exp(-0.5(ln(D/c)/b) 2 ). The VPD/RH values corresponding to the maximum gs can be calculated using the fitting models for the two tree species in the three sites. We found that the calculated gs-max-VPD correlated nega- tively with relative air humidity in the three sites during the plant growth period (April to October 2010), which showed the values of gs-max-VPD were related to the climate conditions. The prevailing empirical stomatal model (Leuning model) and optimal stomatal behavior model could not properly simulate our measured data. The water use efficiency in the two tree species did not show obvious differences under three very different climatic conditions, but the highest gs, photosynthetic and transpiration rates occurred in P. alba var. of Turpan. The sensitivity in re- sponse of gs to VPD in leaves of the two trees showed positive correlations with the concentration of ABA, which implied that ABA level could be used as an indicator of the sensitivity of stomatal response to VPD. Our results confirmed that the prediction of the response of gs to VPD might be incomplete in the two current popular models. Therefore, an improved gs model which is able to integrate the results is needed. Also, the stomatal response mechanism of single peak curves of gs to VPD should be considered.