Improving Numerical Weather Prediction through Planetary Boundary Layer and Moist Convection Parameterizations

Physics parameterizations are one of the major components in numerical weather prediction (NWP) models. Moist convection is mainly represented by the cumulus parameterization (CP) while conflicts between CP and the planetary boundary layer parameterization (PBLP) might occur when there are convective clouds within the planetary boundary layer. To make the PBLP better couple with the CP, a PBLP named CVEDMF was recently developed based on the Eddy Diffusivity Mass Flux (EDMF) framework. The eddy diffusivity is parameterized with the first order K-profile method and the mass flux is parameterized with a convective plume model. The convective plume model only focuses on dry convection, leaving all moist convection to the CP. The results in a global weather model show it improves the global forecast skills.

The E-epsilon PBLP was officially released within the Weather Research and Forecasting Model (WRF) for version 4.3. Both the Turbulence Kinetic Energy (TKE) and TKE dissipation rate are prognostic and can be advected in the model dynamics. Therefore, mixing lengths are no longer required. Numerous improvements have been implemented since the first release. WRF testing with a regional 1 km grid spacing highlighting improved fog, derecho, and precipitation forecasts will be presented.