Extended Abstract (2.5M)12.8
The Anatomy of the Big Event That Never Happened—The Grand Finale of the May 2003 Tornado Outbreak
Neil A. Stuart, NOAA/NWS, Wakefield, VA
During the week from 3-10 May 2003, over 400 tornadoes were reported across the central and eastern United States, more tornadoes then in any other week on record. Most of the tornadoes occurred west of the Appalachian Mountains, many were rated F2 or greater on the Fujita scale. A series of strong upper level vorticity centers tracked out of the northern plains and into the Midwest and Great Lakes through the week, coupled with strong upper level jet stream dynamics, contributing to this historic tornado outbreak. The final in the series of strong upper level vorticity centers formed over southern Minnesota on 10 May, tracking into Wisconsin by early 11 May. This impulse was forecasted to affect the eastern U. S., triggering one last day of tornadoes, before exiting the northeast late on 11 May. The National Weather Service (NWS) Storm Prediction Center (SPC) and all local NWS Weather Forecast Offices (WFOs) in the mid-Atlantic were predicting the formation of a severe squall line and a potentially significant wind damage and tornado outbreak. The threat for wind damage and tornadoes was emphasized on the SPC daily severe weather outlook, and local WFO statements and Hazardous Weather Outlooks. Despite the extreme instability present, no thunderstorms and no severe weather were observed from Maryland through North Carolina. Considering the unanimous agreement within the meteorological community (NWS and media), that a significant severe weather outbreak was imminent, this can be viewed as a significant forecast error, highly visible to the user community. The 11 May non-event was compared with two similar synoptic events, which did produce severe weather: 28 April 2002 (F4 tornado in Maryland) and 2 May 2002 (widespread wind damage and hail in Virginia and North Carolina). Several important distinctions were identified between the non-event of 11 May 2003 and the observed events of 28 April 2002 and 2 May 2002. By comparing the evolution of the synoptic scale and mesoscale features associated with the 11 May non-event, with the two past events substantial differences can be seen in the evolution of synoptic features as well as differing forcing mechanisms. One important difference in the events that was noted, was the movement of the upper level vorticity center. On 11 May, the upper level vorticity center was nearly stationary, moving from Wisconsin to Michigan in 12 hours. In the other 2 events, the upper level energy tracked from the western Great Lakes into New England and Southeast Canada in 12 to 18 hours. Another important factor was the lack of low level forcing or convective trigger on 11 May. The surface and 850 hPa cold fronts did not track east of the Appalachian Mountains until the surface based instability had exited. Detailed analysis of wind profiles, RUC, Eta and GFS model data will be presented for all three events, which will illustrate the forcing mechanisms at both the synoptic scale and mesoscale levels. By comparing expected events that do not occur, to events with similar synoptic and mesoscale characteristics that did produce severe weather, forecasters can improve their forecasting skills in the future. Based on the comparisons, evidence in the data for the 11 May case showed that the potential for thunderstorms was much less than was forecasted.
Session 12, Studies of the May 2003 tornado outbreaks
Thursday, 7 October 2004, 8:00 AM-10:00 AM
Previous paper
Recorded presentation