Grid spacing sensitivities of simulated mid-latitude and tropical mesoscale convective systems

 

Submitter

Ramos Valle, Alexandra — National Center for Atmospheric Research (NCAR)
Prein, Andreas Franz — National Center for Atmospheric Research (NCAR)

Area of research

Cloud Distributions/Characterizations

Journal Reference

Ramos‐Valle A, A Prein, M Ge, D Wang, and S Giangrande. 2023. "Grid Spacing Sensitivities of Simulated Mid‐Latitude and Tropical Mesoscale Convective Systems in the Convective Gray Zone." Journal of Geophysical Research: Atmospheres, , e2022JD037043, 10.1029/2022JD037043.

Science

Our study focused on determining the model resolution needed to accurately represent processes in mesoscale convective systems, focusing on cases in the U.S. Southern Great Plains and in the Amazon region.

Impact

Mesoscale convective systems (MCSs) are large and long-lasting complexes of thunderstorms known to produce extreme weather such as damaging winds, heavy precipitation, and flooding. The overarching goal of our study is to improve the simulation of mesoscale convective systems to advance predictions and projections of such storms.

Summary

In this study, we leveraged the availability of ARM MCS observations to evaluate model simulations. Observations were collected at the U.S. DOE Atmospheric Radiation Measurement (ARM) user facility's Southern Great Plains (SGP) observatory in Oklahoma and at ARM's mobile GoAmazon2014/15 site in Manaus, Brazil (MAO). We simulated observed MCSs at the SGP and MAO sites using the Weather Research and Forecasting model at 12-, 4-, 2-, and 1-km horizontal grid spacing. ARM observations from radiosondes, surface meteorology, and radar wind profilers were used to characterize MCS timing and location, cold pool strength, and convective drafts to evaluate these simulations. Improvements from increasing the model grid spacing are generally smaller at SGP than at the MAO site. Our results indicate that while model biases decrease with increasing grid spacing, deficiencies still remain in 1-km simulations, suggesting the need for further model development and even higher model resolution.