Sensitivity of Vortex Production to Small Environmental Perturbations in High-Resolution Supercell Simulations

Past work by Cintineo and Stensrud (2012) explored the predictability of simulated supercells at 1 km horizontal resolution in the presence of typical environmental errors observed in a database of 1-, 2-, and 3-hr RUC forecasts from severe convective events. The sensitivity of storm development to uncertainty in the initial conditions has implications for initiatives such as Warn-on-Forecast that seek to anticipate storm-scale hazards, such as tornadoes, through short-range numerical modeling. This study will focus on the relative impact of different magnitudes of environmental perturbations on the production of low-level vortices in supercells.

After the method in CS12, errors from the 1-hr RUC forecast database were scaled to 10%, 25%, and 50% of the original magnitude and randomly applied throughout the depth of a control sounding. A total of 60 different perturbed soundings were generated, 20 for each error magnitude, and were used to initialize a horizontally homogeneous model environment. Simulations were run for three hours with 100 m horizontal grid resolution, which allowed low-level mesocyclone-scale vortices to be resolved. Vortices were retrieved from the model wind field using the Vortex Detection and Classification (VDAC) algorithm described in Potvin (2013).

The spatial distribution, time of occurrence, and vortex characteristics will be compared between the three ensembles. A case study of a subset of 50% perturbation ensemble members that did and did not produce vortices will investigate the role of the environment on storm processes that affect the development of low-level rotation.