200 Characterizing the Degree of Convective Clustering Using Radar Reflectivity and Its Application to Evaluating Model Simulations

Thursday, 19 April 2018
Champions DEFGH (Sawgrass Marriott)
Daehyun Kim, Univ. of Washington, Seattle, WA; and W. Y. Cheng, A. K. Rowe, Y. Moon, and S. Park

Despite the impact of mesoscale convective organization on the properties of convection (e.g., entrainment rates, vertical velocity), parameterizing the degree of convective organization has only recently been attempted in cumulus parameterization schemes (e.g., Unified Convection Scheme UNICON). Additionally, challenges remain in determining the degree of convective organization from observations and in comparing directly with the organization metrics in model simulations. This study addresses the need to objectively quantify the degree of mesoscale convective organization using high quality S-PolKa radar data from the DYNAMO field campaign.

One of the most noticeable aspects of mesoscale convective organization in radar data is the degree of convective clustering, which can be characterized by the number and size distribution of convective echoes and the distance between them. We propose a method of defining contiguous convective echoes (CCEs) using precipitating convective echoes identified by a rain type classification algorithm. Two rain type classification algorithms, Steiner et al. (1995) and Powell et al. (2016), are tested and evaluated against WRF simulations to determine which method better represents the degree of convective clustering. Our results suggest that the statistics of CCEs based on Powell et al.’s algorithm better representing the dynamical properties of the convective updrafts and thus provide a reliable metric for convective organization. Furthermore, through a comparison with the observational data, the WRF simulations driven by the DYNAMO large-scale forcing the same way as UNICON Single Column Model simulations will allow us to evaluate the ability of both WRF and UNICON to simulate convective clustering based on the physical processes that are explicitly represented in both models, including the mechanisms leading to convective clustering, and the feedback to the convective properties.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner