A parameterization of land surface processes to be included in mesoscale and large-scale meteorological models is presented. The number of parameters has been reduced as much as possible, while attempting to preserve the representation of the physics which controls the energy and water budgets. We distinguish two main classes of parameters. The spatial distribution of primary parameters, i.e., the dominant types of soil and vegetation within each grid cell, can be specified from existing global datasets. The secondary parameters, describing the physical properties of each type of soil and vegetation, can be inferred from measurements or derived from numerical experiments. A single surface temperature is used to represent the surface energy balance of the land/cover system. The soil heat flux is linearly interpolated between its value over bare ground and a value of zero for complete shielding by the vegetation. The ground surface moisture equation includes the effect of gravity and the thermo-hyd... Abstract A parameterization of land surface processes to be included in mesoscale and large-scale meteorological models is presented. The number of parameters has been reduced as much as possible, while attempting to preserve the representation of the physics which controls the energy and water budgets. We distinguish two main classes of parameters. The spatial distribution of primary parameters, i.e., the dominant types of soil and vegetation within each grid cell, can be specified from existing global datasets. The secondary parameters, describing the physical properties of each type of soil and vegetation, can be inferred from measurements or derived from numerical experiments. A single surface temperature is used to represent the surface energy balance of the land/cover system. The soil heat flux is linearly interpolated between its value over bare ground and a value of zero for complete shielding by the vegetation. The ground surface moisture equation includes the effect of gravity and the thermo-hyd...