Forecast implications of regional variations in storm modes and environments that produce significant tornadoes in the United States

Several climatological studies comparing tornado occurrence, storm modes, and environmental characteristics for predetermined regions in the United States have been performed in the past few years. Although regions of higher tornado occurrence outside of the Plains states have been identified in these studies (particularly in the Southeast), statistically significant differences separating these regions have not been shown to exist. However, preliminary research suggests that significant differences in storm modes may exist, particularly between predominantly supercellular storm modes in Plains states and higher proportions of tornadoes associated with mesoscale or quasi-linear convective systems (MCS/QLCSs) in the Southeast and Midwest. Therefore, the main objective of this study is to determine whether storms producing significant tornadoes (rated EF2 or higher on the Enhanced-Fujita scale) exhibit statistically significant differences in storm modes, structures, and forcing mechanisms for certain regions relative to the mean climatological distribution by analyzing approximately 900 significant tornadoes identified in Storm Data between February 2007- January 2012. Secondary objectives include investigating forecasting and societal implications of these differences, including their potential effect on tornado warning-lead times, path length and width, and loss of life and property.

Each tornadic storm's convective mode is classified as a discrete supercell, MCS/QLCSs, or other. Discrete supercells and MCS/QLCSs are further subdivided into non-mergers and mergers based on whether they were isolated cells or merged with another cell or line within one hour of tornadogenesis. Interannual, seasonal and diurnal variations associated with each storm type and variability in storm structures (i.e., CL, HP, LP, or mini for supercells and TS, PS, LS, TS-AS, QS-BB, or other QLCSs for MCSs) and forcing mechanisms (i.e., dryline, outflow boundary, prefrontal troughs or confluence zones, cold front, warm front, or other) for each storm type will be investigated. In addition, variations in tornado warning lead-times (using Iowa State's archive of unofficial NWS Storm Based Warning Verifications) will be analyzed preliminarily to help forecasters and emergency managers better understand tornadic threats and warning capabilities for their respective regions when they experience specific storm modes. Finally, composite synoptic conditions for significant and outbreak (five or more EF2 tornadoes or greater) tornado events will be compared for each region if time permits.