On 3 October 1979 a violent F4 tornado struck without warning at Windsor Locks, Connecticut just before 1900 UTC. It was the most destructive storm ever to occur in Connecticut. Our case study of the storm was motivated by the rarity of tornadoes in Connecticut and the unfavorable time of the year for such storms in New England.
The history of the storm is traced by means of synoptic and subsynoptic scale analyses. Vigorous midtropospheric ascent occurred just in advance of a cold-core cutoff vortex that was opening up due to synoptic scale deformation while moving northeastward toward southern New England.
The probable destabilization mechanism was the intersection of the northeastward moving pocket of cool air aloft with a tongue of high equivalent potential temperature surface air that originated over the relatively warm ocean water and was channeled northward through the Connecticut River Valley. The convective cell that eventually became associated with the Windsor Locks tornado was detected over the Atlantic Ocean south of Long Island. Tornadogenesis occurred with the merger of this cell with an antecedent northeastward moving subsynoptic scale surface cyclone and surface warm frontal boundary. Immediately preceding and accompanying tornadogenesis the aforementioned radar cell turned to the left and moved northward across Long Island and through the north-south oriented Connecticut River Valley. The absence of quantitative radar data and off-time soundings precludes us from making any definitive statement about left moving supercell development-versus orographic channeling as an explanation for the leftward jog in the storm track during the time of the tornado.
A probable factor in the absence of severe weather watches was that the radar indicated thunderstorm cloud tops were not penetrating the conventionally reported tropopause. Evidence is presented that the actual tropopause was much lower near the vortex and that the lapse rate definition of the tropopause needs to be modified.