Rapid Environmental Changes Observed by Remote Sensing Systems in the Local Vicinity of an Unusual Colorado Tornado

An EF3 tornado devastated the town of Windsor, Colorado on 22 May 2008. This tornado was unusual for a number of reasons: this region is unaccustomed to such strong tornadoes, especially one that formed in the late morning in the month of May and took a track to the northwest. This study extends previously published research on this event by focusing on how the local pre-storm environment changed as detected with 5-6 min sampling by a ground-based 35-channel microwave radiometer and a 449-MHz Wind Profiler (both being within 90 km of Windsor), and using a 1-km resolution variational Local Analysis and Prediction System (vLAPS) mesoanalysis with coupled WRF model forecasts. vLAPS assimilated data from many observing systems, including reflectivity and radial wind data from WSR-88D radars; satellite data soundings, Cloud-Track Winds, and Total Precipitable Water; GPS-Met sensors; and conventional observations.

The combined Wind Profiler and VLAPS datasets indicated that the storm-relative helicity in the 0–1 km AGL layer jumped from 50 to 350 m2 s-2 in the two hours prior to tornado touchdown. This extremely rapid and enormous change was associated with deepening and descent of a lower tropospheric layer of strong vertical wind shear. The radiometer showed a gradual increase of SBCAPE (Surface-Based Convective Available Potential Energy) throughout the morning hours, despite thick cloud cover, to 3100 J/kg at 1815 UTC, just prior to passage of a pronounced dryline attendant the explosive development of the Windsor supercell storm. Even more interesting was a sudden increase in potential instability in the very low levels of the atmosphere (below 200 m AGL) in concert with the aforementioned rapid increases in vertical wind shear and storm-relative helicity. This could not be attributed to near-surface warming/moistening, but rather to the appearance of a region of cooling in the 700 – 800 hPa layer in association with the deepening of the shear layer. The origin of this cooling was traced to air that had been adiabatically cooled as it was advected and lifted over the Palmer Lake Divide, a mesoscale region of ~1.0 km higher terrain just south of Denver and Windsor.

Another interesting finding from this study concerns the likely lifting mechanism that created a long-lasting supercell from this very moist, unstable air with exceedingly high vertical shear. One factor may have been the orographic generation of elongated streamers of isentropic potential vorticity seen at 500–700 hPa in the VLAPS analyses. These streamers were aligned parallel to the south-southeasterly flow regime existing across the region, and as the most intense streamer propagated slowly eastward past the Front Range to the immediate west of Windsor, strong ascent was diagnosed. This process was preceded by ~90 min by a rapid increase in low-level frontogenesis in the Windsor region that correlated even better in time with observed convection initiation. Frontogenesis developed along a northward bulge in the warm front associated with flow around a pronounced low-level “Denver cyclone” as a dryline advancing northward toward this region from the Palmer Lake Divide merged with the front. The confluence of all these factors in a very short period of time and juxtaposed over the proximate Windsor region was likely critical in explaining how this rare tornadic event could have happened so suddenly and precisely where it did.