Stephen G. Gaddy

Senior, The Pennsylvania State University, B.S. Candidate in Meteorology

(Manuscript received in final form 21 April 1995)


On 27 March 1994, an outbreak of severe thunderstorms and tornadoes affected portions of the Southeastern United States. This paper presents an overview of the outbreak, which emphasizes the role of mesoscale meteorology in forecasting storm initiation and intensity. The synoptic weather environment is described, with an emphasis on the lack of the usual mechanisms present before a major severe weather outbreak. The mesoscale environment, its evolution, and its role in storm formation is then presented. A section on the most intense and most deadly storm presents an analysis of Doppler radar imagery at large distances from the radar; distances many believe to generally be ineffective for the analysis of supercell features. Finally, issues relating to reducing the loss of life and property from such disasters are discussed.


On Sunday 27 March 1994, one of the most deadly tornado outbreaks in recent history occurred in the Southeastern United States. The deadliest storms occurred in northern Alabama and northern Georgia, mainly during the late morning and early afternoon hours. Many of the deaths occurred as the largest tornado (ranked F4 on the Fujita Scale) move northeastward from Calhoun County AL into Cherokee County AL, striking the Goshen United Methodist Church at 11:39 AM CDT (1739 UTC) during a Palm Sunday service. As a result, the roof of the building collapsed, killing 20 people and injuring 90 (U.S. Dept of Commerce 1994).

These storms were unique not only due to the unusually high death toll, but also due to the unusual meteorological situation which prompted them. The synoptic weather situation, as seen well in advance of the outbreak, showed no signs which would lead to such strong tornadoes. There was little, if any, upper air support at the time of the outbreak. No organized areas of positive vorticity advection (PVA) were present, and the jet stream was located well northwest of Alabama. The surface analysis showed only a stationary front crossing through northwest Alabama, with little movement forecasted due to the upper level pattern.

It was not until early on the morning of the outbreak when the potential for strong tornadoes became clear. Many of the features which came together to cause the outbreak were subtle, low- level features which could not be forecasted well in advance of the storms. Furthermore, the exact location of the storms would depend on mesoscale focusing mechanisms which would develop during the morning hours.

This paper will address the synoptic and mesoscale situation which led to the outbreak; then it will focus on the largest storm and its imbedded tornado which hit the church in northeastern Alabama, Finally, the conclusion will address how the loss of life could have been reduced from this outbreak. (References are provided with original paper.)

Selected Figures

A very intense supercell thunderstorm is shown over southern Cherokee County in the base reflectivity display for 1737 UTC 27 March 1994 from the Maxwell AFB WSR-88D Doppler radar. An F4 tornado struck the Goshen United Methodist Church about 2 min after this image. A classic tornado hook echo is not apparent, probably due to the rather large distance of the supercell from the radar site. A flanking line of echoes extends southwestward into northeast Calhoun County.

An extremely strong velocity couplet with at least 100 kts of gate-to-gate shear is shown in the storm relative velocity display for 1737 UTC 27 March 1994 from the Maxwell AFB WSR-88D Doppler radar. An F4 tornado struck the church at Goshen AL about 2 min after this image.