Moist deep convection is a critical aspect of weather and climate, but its parameterization is challenging and process-level understanding remains uncertain. This presentation will give an overview of the important dynamical factors affecting the structure and evolution of deep convective updrafts in unsheared environments based on theory and numerical simulations. The first part of the talk will focus on the role of perturbation pressure, which is treated in most convection parameterizations by an ad-hoc constant scaling of updraft buoyancy. Theoretical and numerical results indicate a scaling of vertical velocity in convective updrafts by a factor that depends on the square of the updraft aspect ratio (width divided height). A simple analytic model of the effects of perturbation pressure is proposed for both 2D and 3D updrafts that gives results consistent with fully dynamical updraft simulations, and explains the tendency for weaker updrafts in 2D models compared to 3D.

The remainder of the talk will focus on extending this analytic model to include the effects of entrainment and dilution in a growing updraft, based on approximate solutions of the governing equations. In this model, the effects of entrainment on buoyancy and vertical velocity depend on the updraft radius, height, and environmental relative humidity, consistent with numerical simulations. The analytic model also links entrainment with the updraft-scale circulation and perturbation pressure, leading to enhanced lateral mixing lower in the updraft because of the engulfment of environmental air associated with organized horizontal convergence. A conceptual model is developed from these results that shares elements of previous models (thermal, entraining plume, buoyancy sorting). Several interesting hypotheses emerge from this model, supported by the numerical simulations, including the role of updraft size and environmental relative humidity on whether moist convection has a more plume-like or thermal-like structure.

The talk will conclude by discussing implications of this work for improving convection parameterizations, and for understanding biases in nonhydrostatic “gray zone” models with a horizontal grid spacing of order 1-10 km that cannot explicitly resolve deep convective updrafts.