Microphysical controls on clouds, radiation, and tropical circulation

In global storm-resolving models that resolve convection explicitly instead of parameterizing it, microphysical processes are now fundamentally linked to their controlling factors, i.e., the circulation. While in conventional climate models the convective parameterization is one of the main sources of uncertainties (and a popular tuning parameter), this role might be passed on to the microphysical parameterization in global storm-resolving models. In an ongoing project we investigate how microphysical choices affect the albedo, radiative fluxes, and the heat budget in more general.


Vertical structure of water vapor and clouds in moisture space

Global storm-resolving models come along with methodological challenges: because of the large cost of the simulations, new ways to explore sensitivities using only very short runs are needed. In this study, we refine a method based on moisture space, where observational limitations can be taken into account and carried over to the simulated moisture space thereby allowing for a fair comparison between short-term simulations and limited observations. Applying this method to airborne lidar measurements, we illustrate the fragility of simulated cloudiness at kilometer-scale resolution and how a model's ability to properly capture the water vapor distribution does not need to translate into an adequate representation of shallow cumulus clouds that live at the tail of the water vapor distribution.


New theory for radiatively driven shallow circulations

By simplifying a physical system to its fundamental processes, conceptual models help us to investigate and understand more complex behavior. To understand how radiative cooling induces shallow circulations, we develop a conceptual model of the lower part of the tropical troposphere. According to our theory, flows may be driven as much by differential longwave cooling in the lower troposphere, as by surface temperature gradients arising from differences in near-surface wind speeds or insolation changes. In other words, the radiative effects of water vapor in the lower troposphere may play a very active role in the coupled dynamics of lower, tropical troposphere. In a follow-up study, we extend the framework to the moist convective boundary layer of the trade wind regime. We show that shallow circulations driven by radiative differences suppress convection in the descending branch and enhance it in the ascending branch, which resembles shallow cloud patterns.


A Lagrangian perspective on warm rain formation

Understanding precipitation remains one of the major challenges in numerical weather prediction as well as climate modeling. To investigate warm rain microphysical processes on a particle-based level, we have developed a Lagrangian drop (LD) model to simulate raindrop growth in shallow cumulus. The LD model is part of the UCLA-LES and represents all relevant rain microphysical processes such as evaporation, accretion and selfcollection among LDs as well as dynamical effects such as sedimentation and inertia. Sensitivities of the LD model are small compared to the uncertainties in the assumptions of commonly used bulk rain microphysics schemes.

We have applied the LD model to study the development of the raindrop size distribution in shallow cumulus clouds and show that the shape of the raindrop size distribution depends on the stage of the lifecycle of the cloud. The study suggests that two-moment schemes with a diagnostic parameterization of the shape parameter, i.e., a local closure in space and time, are not sufficient, especially when being applied across different cloud regimes. One way to overcome this issue may be a prognostic shape parameter, i.e., a triple-moment warm-rain microphysics scheme.

Using the LD model, we have also investigated the growth process of raindrops and the role of recirculation of raindrops for the formation of precipitation in shallow cumulus. Recirculation of raindrops is found to be common in shallow cumulus, especially for those raindrops that contribute to surface precipitation. The fraction of surface precipitation that is attributed to recirculating raindrops differs from cloud to cloud but can be as large as 50 %. This implies that simple conceptual models of raindrop growth that neglect the effect of recirculation disregard a substantial portion of raindrop growth in shallow cumulus.