A New TKE-Based Parameterization of Atmospheric Turbulence in the Canadian Global and Regional Climate Models

A new semi-empirical turbulence parameterization is presented. Key features of the scheme include representation of turbulent diffusivities in terms of the turbulent kinetic energy that is determined by solving a quasi-equilibrium form of the equation representing the turbulent kinetic energy budget. The new parameterization is innovative in the treatment of turbulent transfer in stably stratified conditions and the representation of nonlocal contributions to the vertical transport of heat, moisture, and scalar prognostic variables in convectively active boundary layers. A key element in the modeling of turbulence in stably stratified conditions is the formulation of the turbulent Prandtl number based on the results of recently published theoretical, modeling, and observational studies of stratified turbulence in the atmospheric boundary layer. The new parameterization has been implemented in the CanAM4 single column model. Its performance in comparison with that of the operational CanAM4 turbulence parameterization is documented in terms of selected results from case studies for clear-sky conditions based on meteorological observations from the KNMI-mast at Cabauw, Netherlands, and the Second Dynamics and Chemistry of Marine Stratocumulus case study of stratocumulus-topped marine boundary layers. The performance of the new and operational schemes is qualitatively similar in clear-sky conditions in both convective and stable boundary layer regimes. However, they perform differently for the extended simulations for the Second Dynamics and Chemistry of Marine Stratocumulus case study. The new scheme maintains an elevated stratocumulus layer throughout a 30-hr simulation, but peak liquid water contents are larger than large eddy simulations. Plain Language Summary This paper presents a new mathematical formulation to account for the effects turbulent motions in comprehensive global climate models. The new formulation is based on recently published theoretical advances and results of high-resolution numerical model simulations for specialized atmospheric turbulence regimes. The new formulation is tested and evaluated using a simplified model configuration designed to represent a single grid volume of a global climate model.