Competition Between Organized and Unorganized Convection Regions in Large-Domain CRM Simulations

Abstract:

Multicellular organization of deep convection is commonly observed, but ignored in contemporary global circulation model parameterizations. Is it important, and if so how? In an atmosphere destabilized by homogenous forcing, would a region of more organized convection out-compete regions with spotty convection for the resulting moist convective instability?

A set of idealized simulations is built to address those questions using large-domain simulations with Cloud Model 1 (CM1), a 3D cloud-resolving model with explicit representations of both convective-scale and domain-scale circulations. A double-periodic domain as large as we can afford is uniformly destabilized with a homogeneous cooling of -4K/day and a prescribed surface flux. Experimental runs use a vertically sheared zonal wind profile or mesoscale flux patterning in an east-west strip covering only part of the y-domain. In the sheared or mesoscale-patterned flux area, convection organize into mesoscale clusters, following Robe and Emanuel (2000) for instance. If organized convection on average heats the atmosphere more (or less) than the unorganized convection in the unsheared area, zonal-mean meridional circulations develop, with rising branches in whichever regime is most efficient at converting the uniform destabilization into latent heat release. Moisture transport reinforces this development of east-west banding, cleanly isolating and then amplifying the most positive of the convection-organizing effects of a prescribed nondivergent shearflow or pattern of surface flux.