Elevated temperature altered the reaction norms of stomatal conductance in field-grown grapevine

We measured the effect of elevated temperature, and its interaction with fruit load (exp. 1) and water supply (exp. 2), on the stomatal conductance (g(s)) of Vitis vinifera, cv. Shiraz. Thermal regimes (control vs. elevated temperature using open-top chambers) were initiated in spring, thus affecting leaf development, and were maintained during the whole growing season. We used a top-down approach where reaction norms were derived that relate gs for each treatment and the mean g(s) across treatments: the slopes of reaction norms quantify phenotypic plasticity. Stomatal conductance responded to neither the interaction between temperature and fruit load (P = 0.37) nor the interaction between temperature and water regime (P = 0.33). We therefore dealt with each factor separately. Reaction norms of g(s) under elevated temperature and control treatments diverged, i.e. elevated temperature had no effect on g(s) under conditions conducive to low conductance (below similar to 100 mmol m(-2) s(-1)) but increased g, in relation to controls under conditions favouring high conductance. Stomata density was 159 +/- 6.7 mm(-2), and was unaffected by temperature. Stomata length increased from 20.6 mu m in controls to 23.2 mu m in the heated treatment (P < 0.0001) and width increased from 14.4 mu m in controls to 16.8 mu m in the heated treatment (P < 0.0001). Thus, longer and wider stomata contributed to the enhanced plasticity of stomatal conductance under elevated temperature. Consistently, reaction norms of light-saturated photosynthesis were divergent, i.e. elevated temperature did not affect photosynthesis under conditions conducive to low g(s) and photosynthesis, but increased photosynthesis in relation to controls under more favourable conditions. A high source:sink ratio reduced the plasticity of stomatal conductance. Water regime had a minor effect on the plasticity of stomatal conductance, but the reaction norms for irrigated and water deficit treatments were offset by 60 mmol m(-2) s(-1) thus indicating a consistent effect of water deficit across environmental conditions. We conclude that the responses of stomata derived from transient temperature treatments in fully expanded leaves are unsuitable to model the dynamics of gas exchange in response to projected warming. Our experiments support a conceptual model where current stomatal conductance is a function of (i) the prevailing environmental conditions during early stages of leaf development that set the upper limit of conductance through the modulation of stomatal density and size, (ii) the current environmental conditions and (iii) the source:sink ratio. (c) 2012 Elsevier B.V. All rights reserved.