Title: Physical Mechanisms Controlling Self-Aggregation of Convection in Idealized Modeling Simulations

Abstract: Convective cloud clusters are responsible for much of the rainfall and cloudiness over the tropics, allowing them to modulate the radiative heating and cooling rates of the surface and atmosphere and influence the large-scale circulation and moisture distribution. Therefore, understanding how and why tropical convection organizes is important for understanding both tropical and global climate variability. In this study, the problem is approached through the context of idealized modeling of convective organization in radiative convective equilibrium using a cloud system resolving model. Previous studies have investigated interactions between the environment and the convection that allow convection to self-aggregate into a single cluster, and have found this self-aggregation to be dependent on a sea surface temperature threshold. In this study, the System for Atmospheric Modeling is used to perform 3-d cloud system resolving simulations with a doubly periodic horizontal domain, interactive radiation and surface fluxes, and no rotation or external forcing other than solar insolation. Simulations are run at fixed SST.

We quantify the magnitudes of the various feedbacks that control self-aggregation within the framework of the budget for the spatial variance of column frozen moist static energy. The absorption of shortwave radiation by atmospheric water vapor is found to be a key positive feedback in the early stages of aggregation, while the longwave radiation feedback term can be either positive or negative. In addition, we find a positive wind speed-surface flux feedback whose role is to counteract a negative feedback due to the effect of air-sea enthalpy disequilibrium on surface fluxes. Finally, we note that the self-aggregation processes begins as a dry patch that expands, eventually forcing all the convection into a single clump. Thus, when investigating the temperature dependence of self-aggregation we focus on processes that can amplify this initial dry patch.

Speaker's website: http://www.mit.edu/~awing/home/