Cloud Resolving Simulations with a Simplified Higher-Order Turbulence Closure

Description

At some point in the future it is likely that all atmospheric general circulation models (AGCMs) will be cloud resolving, with horizontal grid sizes of about 1-5 km. In addition, most numerical weather prediction models are currently cloud resolving. While models with grid spacings at the kilometer scale can resolve deep convective circulations it remains that boundary layer turbulence, shallow convection, and turbulence associated with deep convection must continue to be parameterized. Unfortunately, most cloud resolving models (CRMs) use subgrid-scale (SGS) parameterizations that were originally designed for large eddy simulations (LES) and are not applicable for models at the kilometer scale. This poster presents simulations where a CRM is coupled with a computationally efficient SGS turbulence and cloud parameterization. The parameterization is known as Simplified Higher-Order Closure (SHOC) and is a diagnostic higher-order turbulence closure centered around an assumed double-Gaussian probability density function. CRM results are presented from various field campaigns, such as TWP-ICE. In addition, results are presented for the super-parameterized Community Atmosphere Model (SP-CAM), which is a unique GCM that uses an embedded CRM in each GCM grid column to replace traditional cloud parameterizations. SP-CAM simulations using SHOC in the embedded CRM show marked improvements in tropical and sub-tropical cloud representation compared baseline SP-CAM simulations.