Influence of Subcloud-layer Structures on the Transition to Deep Convection

Modeling the transition from shallow to deep cumulus convection remains a very challenging problem in atmospheric modeling. Numerical Weather Prediction and climate models, in which deep convection is parametrized, tend to predict the onset of deep convection too early in the day. However, even Cloud Resolving Models are sensitive to resolution requirements and the representation of microphysics.

In this study we use the Dutch Atmospheric Large Eddy Simulation (DALES) model to conduct detailed simulations of deep convection and investigate to what extent the spatial growth of potential temperature and humidity structures below cloud base influence the transition. When precipitation starts, density currents trigger cloud formation at the edge of areas where rain evaporates. The appearance of such cold pools coincides with a rapid growth of length scales. Previous studies in which the sensitivity of LES simulations to microphysics parameterization was investigated suggest that the evaporation of rain may indeed have a dramatic effect on the development of deep convection (e.g. Khairoutdinov, and Randall, 2006). In the present study we try to directly influence the developing spatial structures in the thermodynamical fields by manipulating the contributions in spectral space, i.e. by selectively suppressing/promoting the spectral contribution of particular wavelengths, without modifying the overall (mean) thermodynamical budgets. This enables one to study the effects of organised structures - such as cold pools - separately from the behavior of the model above the cloud layer.

Khairoutdinov, M. F., and D. A. Randall, 2006: High-resolution simulation of shallow-to-deep convection transition over land. J. Atmos. Sci., 63, 3421–3436.