Modeling stomatal and nonstomatal effects of water deficits on CO2 fixation in a semiarid grassland

[1] The confidence with which we can model water deficit effects on grassland productivity is limited by uncertainty about the mechanisms, stomatal and nonstomatal, by which soil water deficits reduce CO2 uptake. We propose that these reductions can accurately be modeled from a combination of stomatal effects on gaseous CO2 diffusion and nonstomatal effects on biochemical CO2 fixation. These effects can be combined through a solution for the intercellular CO2 concentration (Ci) at which rates of diffusion and fixation are equal for each leaf surface in the canopy. In this model, both stomatal and nonstomatal effects are driven by a common indicator of plant water status calculated in a hydraulically-driven scheme of soil-plant-atmosphere water transfer. As part of the ecosystem model ecosys, this combined model simulated concurrent declines in latent heat effluxes and CO2 influxes measured by eddy covariance during soil drying in a drought-affected semiarid grassland. At the same time, the model simulated the declines in Ci at which CO2 fixation occurred during soil drying as calculated from seasonal measurements of phytomass δ13C. Alternative model formulations based on stomatal or nonstomatal effects alone simulated these declines in CO2 influxes and in Ci less accurately than did the formulation in which these effects were combined. We conclude that modeling water deficit effects on CO2 fixation requires the concurrent simulation of stomatal and nonstomatal effects. As part of a larger ecosystem model, this combined model can be used to assess climate effects on grassland productivity.