Large Eddy Simulations of Shallow to Deep Convective Cloud Transitions during HI-SCALE

Poster PDF

Description

The Holistic Interactions of Shallow Clouds, Aerosols, and Land-Ecosystems (HI-SCALE) field campaign conducted from April 24 to May 21 and August 28 to September 24 near the ARM SGP site provided a detailed set of aircraft and surface measurements needed to obtain a more complete understanding and improved parameterizations of the lifecycle of shallow clouds. On August 30, 2016 shallow cumulus formed over the SGP site around 1145 LST and two G-1 aircraft missions were conducted to obtain in situ measurements of cloud and boundary layer properties. Satellite images indicated that shallow convection initially formed over eastern and southeastern Oklahoma around 0945 LST and spread over the entire state and southeastern Kansas over the next two hours. As the day progressed, shallow clouds transitioned to short-lived deeper convection that produced precipitation in some areas, forming cold pools that suppressed surrounding shallow convection. A complex population of convective clouds was observed over Oklahoma by the late afternoon, suggesting that this day would be useful to evaluate the performance of explicit and parameterized representations of convective clouds in models. In this study, we utilize two sets of Large Eddy Simulations (LES) made using the Weather Research and Forecasting (WRF) model for August 30, 2016. One set of simulations are taken from the LES ARM Symbiotic Simulation and Observation (LASSO) product. The LASSO LES domain is ~14 km wide, uses periodic boundary conditions, and has an ensemble of simulations that use different larger-scale forcing. In contrast, the other set of simulations use a much larger domain that is ~120 km wide and a nested grid approach so that the model can represent a more complex evolution of the environmental forcing and cloud population. Routine in situ and remote sensing ARM observations and HI-SCALE aircraft observations including vertical velocity distribution, droplet size distribution, liquid water content, cloud depth, and cloud fraction are used to evaluate both sets of simulations. The LASSO LES was not able to represent the observed transition of shallow convection to deeper convection that was observed near the SGP site. While the larger LES domain contained regions where shallow convection transitioned to deep convection, the transition occurred later than observed. We also use new metrics that quantify cloud population properties, which include 1) a cloud-tracking algorithm to determine the lifetime and size evolution of shallow clouds over the LES domain and 2) a new data product that quantifies the size of shallow clouds near the SGP site from the ARM total sky imager (TSI). Mesoscale simulations are also performed and compared to the explicit LES simulations to illustrate how well simulations with grid spacings ranging from 1 to 36 km represent the evolving convective cloud population on this day.