A radar network approach to characterize shallow convection at the SGP megasite to support the LASSO activity

 

Authors

Pavlos Kollias — Stony Brook University
Mariko Oue — Stony Brook University
Kirk North — McGill University
Aleksandra Tatarevic — McGill University
William I. Gustafson — Pacific Northwest National Laboratory
Andrew M. Vogelmann — Brookhaven National Laboratory
Heng Xiao — Pacific Northwest National Laboratory
Satoshi Endo — Brookhaven National Laboratory

Category

ARM next generation – Megasite and LES activities

Description

The LES ARM Symbiotic Simulation and Observation (LASSO) activity aims to bring together routine, large domain large-eddy simulations (LESs) and observations of shallow convection at the ARM SGP megasite. Here, we focus on some of the observational aspects of this modeling-observation activity. Shallow convection is characterized by large inhomogeneity in cloud properties (e.g, cloud fraction, PDF of cloud top heights, precipitation, mass flux). Furthermore, shallow clouds at the SGP pose detection challenges to radars due their weak reflectivity. These issues raise important questions: How well profiling observations of shallow convection capture these properties? How best to compare domain-average model output with profiling observations? Under what circumstances does the use of a radar-network approach that utilizes all available radar resources at the SGP megasite provide the best constraints on the modeling of shallow convection? To address these issues, we use LASSO test case model output combined with a radar simulator that accounts for all the available radar frequencies used at the SGP megasite: the profiling KAZR, the scanning SACR and the network of XSAPRs. In addition, a lidar simulator is introduced to account for the enhanced detection of clouds in the column, which is one of the inputs into the ARSCL cloud layering product. The simulations illustrate that a single vertically profiling site within a 30x30 LES km domain is insufficient to describe the average macro-scale properties of shallow cumulus convection and can lead to large differences when compare to domain-average cloud properties. The added value of scanning radar (SACR’s and XSAPR’s) observations to capture these properties is discussed here. In addition, the simulator output is used as input to a 3DVAR velocity algorithm and synthetic retrievals of the horizontal wind field in the cloud layer and of in-cloud mass flux are presented.