Atmospheric Boundary Layer Processes and Their Parameterization in Climate Models

Abstract

The thermal and dynamic interaction between the atmosphere and underlying surface occurs through the boundary layer. While these interactions have generally been ignored for short-range prediction of large-scale atmospheric circulations, they are quite important for long-range forecasting and in studies related to general circulation and climate dynamics. In this case a consideration must be given not only to the supply of energy but also to the dissipation of kinetic energy, as well as to the vertical transport of heat and moisture within the boundary layer. Since the intensity of small-scale processes is affected by large-scale processes, incorporation of the boundary-layer dynamics constitutes an essential part in studying the physical principles of long-range forecasting of large-scale processes and climatic changes. Numerical climate models which have been adopted as important tools for understanding the physical basis of climatic changes vary greatly in many respects. This is especially true with respect to the degree of detail in their treatment of the physical processes which cannot be resolved by the grid spacing of the model. These unresolved processes (which include boundary layer processes) are incorporated in climate models through parameterization which involves empirically related parameters determined on the basis of observations or theoretical considerations. The boundary-layer parameterization in climate models is usually dependent on the vertical resolution of the model and is related to the determination of four factors in terms of the variables predicted by the model. These are: • surface fluxes of momentum, heat, and moisture, • vertical profiles of the turbulent fluxes within the boundary layer, • height of the boundary layer, and • vertical velocity at the top of the boundary layer. There are several approaches to the boundary-layer parameterization aimed at incorporating the boundary-layer processes in climate models. Some of these techniques are based on the so-called K-theory while others are based on the similarity theory. Considering the varying degrees of vertical resolution of different climate models, the determination of the surface fluxes is perhaps the most important aspect of the boundary-layer parameterization in a general circulation model. At present there is not sufficient evidence to determine which particular boundary-layer parameterization scheme is the most satisfactory. It is recommended that a systematic sensitivity test of various schemes be carried out to determine the best approach.