A Comparison of Approaches for Modeling Crop Growth and Transpiration Response to CO2 in Apsim, CERES, Cropgro, Cropsyst, EPIC, and Oryza Models.

Six crop models were compared for CO₂ effects on photosynthesis and transpiration.  The CO₂ effects on photosynthesis are captured via modifiers of radiation use efficiency (RUE) in the APSIM, CERES, and EPIC models, and on rubisco kinetics of leaf photosynthesis in the CROPGRO and the ORYZA models.  In CropSyst, daily DM is the minimum of transpiration multiplied by transpiration use efficiency which is dependent on vapor pressure deficit (VPD), or light interception multiplied by RUE.   RUE in EPIC is adjusted for VPD.  Functions for CO₂ effect on RUE are logistic (EPIC), Michaelis-Menten (CropSyst), user-defined input ratio (APSIM), nonlinear CO₂-sensitive factor (APSIM-wheat), and nonlinear lookup in CERES.  These functions attempt to mimic metadata on biomass/yield response to CO₂ at future versus baseline CO₂.  CROPGRO and ORYZA predict sunlit and shaded leaf assimilation scaled up to daily canopy assimilation, in which quantum efficiency and Amax in asymptotic exponential equations are based on simplified rubisco kinetics to predict response to CO₂, temperature, and light.  For transpiration, the models have daily energy balance equations, Penman-Monteith or Priestley-Taylor, to predict potential evapotranspiration and partition energy to potential plant transpiration and potential soil evaporation, using a canopy energy extinction coefficient.  Soil evaporation depends on energy reaching the soil surface and soil water content.  The CO₂-sensitivity of transpiration in CERES, CROPGRO, CropSyst, and EPIC is created by CO₂ effects on crop canopy conductance.  A ratio effect for a given CO₂ versus reference CO₂ is computed to reduce transpiration.  Transpiration of ORZYA is not CO₂-responsive.  The APSIM model predicts transpiration as a function of daily dry matter production via a species-dependent water-use efficiency coefficient that depends on VPD and a CO₂ -sensitive transpiration efficiency modifier.  We will compare behavior of the different approaches for modeling assimilation and transpiration responses to CO2.