Séminaire
The representation of stratocumulus clouds in Global Climate Models (GCM) is still problematic. Indeed, the different spatial scales at play and the complexity of the large set of phenomena that governs the Stratocumulus Topped Boundary Layer (STBL) still challenge present-day GCM parameterizations.
We here discuss two process-oriented approaches that – making use of high-resolution large eddy simulations – aims at improving our physical understanding and eventually at improving the representation of stratocumulus clouds in GCM.
First we show that the main convective coherent structures, namely updrafts, downdrafts and entrainment, are equally important to describe the STBL dynamics in terms of mass, heat and moisture fluxes. This is done with a novel approach based on two passive tracers able to detect and track each convective structure in the STBL. It is thus possible to estimate the sensitivity of the turbulent fluxes to the intensity of the cloud-top cooling, to the surface latent and sensible fluxes or to the strength of the wind shear.
In second place, we explore the role of evaporative cooling on the dynamics of a non-precipitating STBL. This is done by inhibiting the dynamical contribution of evaporative cooling to the buoyancy term in the vertical momentum equation, without directly affecting the conserved variables. By comparing the results from this simulation with a reference twin one we find that evaporative cooling is on average negligible and thus it does not influence boundary layer entrainment.
These two results are important for the representation of stratocumulus in climate models, as they suggest that in STBL 1) the turbulent transport from downdrafts is as important as the one from updrafts and should be included in any parametrization 2) only the slab-average cooling associatedwith evaporative cooling is relevant but that the exact spatial structure is not.
bruno.deremble@ens.fr