Initiation and Evolution of the South Dakota Tornado Outbreak of 24 June 2003

The outbreak that occurred in eastern South Dakota on 24 June 2003 provides an excellent opportunity for the study of tornadoes because they were significant in number, yet concentrated in a small area over a short period of time. Precursors were examined, particularly the differential advection of the air masses involved. Similarities were found between this and previous tornado outbreaks in South Dakota. Specifically, low-level parcels often originated in Arkansas, Louisiana, or Texas ≤ 72 h before such outbreaks, and midlevel parcels came from Colorado, Utah, Arizona, or New Mexico ≤ 24 b before outbreaks. This movement of air masses was in contrast to other cases in which a tornado outbreak failed to materialize. A post-analysis of radar, computer model, and observational data pertinent to the outbreak was performed to determine what meteorological processes contributed to sustaining tornado production and behavior during the outbreak. This analysis provided documentation of differences between the movements of tornadoes in comparison to their parent storms. A subset exhibited highly anomalous movement, with damage paths veering to the left of the translational motion of the parent storm. A review of published accounts of tornado paths in previous outbreaks shows leftward tornado movement motion is not uncommon, but it usually occurs as leftward trochoidal curls at the end of a long-track tornado\u27s life cycle, and often as a component of cyclic tomadogenesis. But in this case tornado damage paths were wholly oriented southeast-to-northwest. A candidate mechanism for this anomalous movement, based upon spotter reports and radar data, was cyclonic rotation of the tornado vortexes around a parent mesocyclone circulation in a highly unstable, weakly-sheared environment