Water and carbon fluxes in rain fed agricultural sites under a changing climate: The role of stomata

Abstract:

Vegetation models are needed to assess how crop productivity may be altered due to variations in climatic conditions. Stomatal conductance controls both diffusion of CO₂ from the atmosphere into the leaf and water losses from the soil-plant system to the atmosphere through transpiration (E). Despite its significance, stomatal conductance and its links to climatic variables remains empirically specified in current crop models thereby challenging their application to future climatic conditions. It has long been conjectured that stomata has evolved so as to allow terrestrial plants to assimilate CO2 in a desiccating atmosphere while minimizing water losses. Hence, the hypothesis that stomata adapt optimally to their environment so as to maximize assimilation (A) for a given amount of water loss has received significant attention over the past 4 decades. Here, a new approach to implement optimization theory of stomatal conductance into a dynamic canopy gas exchange model is introduced. A key variable in this theory is the so-called marginal water use efficiency (MWUE), which is assumed to be constant on time scales commensurate with fluctuations in stomatal aperture. However, on time scales relevant to crop productivity (daily to seasonal), the boundary conditions on the optimization problem evolve in time prompting the question of how to assign MWUE on such time scales. To address this question, MWUE was formulated as a function of time-integrated leaf-water potential and atmospheric CO2. Next, leaf water potential was linked to root and soil pressure using a soil water balance model based on a modified Richards' equation that considers vertical distribution of root water uptake. The adequacy of the new approach was tested by comparing predicted diurnal cycles of A and E as well as variability of soil moisture with long-term observations at a winter wheat (Triticum aestivum cv.Cubus) field in southwest Germany (see Figure), where transpiration and assimilation rates were derived from eddy-covariance measurements of latent heat flux and net ecosystem exchange. To place those results in the broader context of climate change and food security issues, a sensitivity analyses on water and carbon fluxes with respect to climatic variables, soil texture, and root-density distribution is also presented.