A unified probabilistic model of dry, shallow, and deep convection

Description

We will present the probabilistic bulk coupled model (PBCM), a boundary-layer scheme based on a probabilistic framework. In the dry and shallow convection case an ensemble of plumes generated at the surface permit the growth of the boundary layer. The advantage of the probabilistic approach is twofold: the entrainment velocity of the mixed layer top and the cloud base mass flux are described through a complementary relationship, based on the fraction of plumes overshooting a dry convective inhibition (corresponding to the subcloud layer) and the convective inhibition, generating active cloud cover.

The probabilistic framework accurately describes the diurnal course of convection and the triggering of deep convection. Once rain occurs, a probabilistic ensemble of cold pools organizes the mixed-layer turbulence and provides a mechanical and thermodynamic lifting to the environmental updrafts. The cold pools also increase the size of the updrafts, thus reducing their lateral entrainment.

The model favorably compares to large-eddy simulation results in various cases (ARM Southern Great Plains, Bomex, Amazon...), which will be presented.