In this study, the authors develop and validate an approach to calculate a canopy conductance that can successfully be implemented in an atmospheric model. The approach is based on plant physiology approaches that have been developed recently. However, it includes a new analytic formulation to scale the conductance up from leaf to canopy. Furthermore, a new expression that accounts for the effect of soil moisture on the canopy conductance is proposed. The parameters are not optimized locally; rather, values are assigned to them as a function of vegetation type. This approach is validated for three experimental sites: the First International Satellite Land Surface Climatology Project Field Experiment (FIFE)-KANSAS area, the Hydrological Atmospheric Pilot Experiment-Mod\'elisation du Bilan Hydrique (HAPEX-Mobilhy) site, and the Cabauw area. The vegetation at these sites is representative for large areas with low vegetation in the world. The results of the plant physiological approach are based on a distinction in $\mathrm{C_3}$ and $\mathrm{C_4}$ plant types, and these results are found to be better (FIFE-KANSAS) and more consistent (HAPEX-Mobilhy) than estimates obtained by a traditional Jarvis-Stewart approach. The parameters of the latter are also obtained from a vegetation classification. For the Cabauw area, both approaches perform comparablely. Furthermore, the new soil moisture content response function is found to work well, as compared with previous formulations.