Building an Explicit Representation of Sub-Grid Variability with Subcolumns in CAM

Abstract:

In March 2014, the Community Atmosphere Model (CAM) issued a new release supporting subcolumns in physics calculations. A set of subcolumns represents profiles of model state and properties on scales smaller than the model grid width. This study utilizes this new feature and a new approach to modeling convective variability that couples convective and microphysical processes more explicitly than traditional convection parameterizations. Our subcolumn scheme uses the Subgrid Importance Latin Hypercube Sampler (SILHS) to generate profiles of moisture, temperature, vertical velocity and cloud condensate through sampling the multivariate PDF used by the Cloud Layers Unified By Binormals (CLUBB) parameterization.

The subcolumn profiles created by the SILHS generator are an explicit representation of the sub-grid variability predicted by the CLUBB moist turbulence parameterization. The CAM microphysics scheme runs on each subcolumn, and the resulting heat and moisture tendencies are averaged and returned to the large scale model. We will show results of global atmospheric simulations with CAM, where the convection and microphysics schemes are coupled through this explicit representation of sub-grid variability.