Retired senior lecturer in the Department of Meteorology at Penn State, where he was lead faculty for PSU's online certificate in forecasting.
By: 24hourprof , 3:23 PM GMT on May 28, 2013
On Tuesday, May 21, in the aftermath of the EF-5 tornado that devastated Moore, Oklahoma, the previous day, I watched Dr. Greg Forbes of The Weather Channel give a live interview standing in front of some of the damage. At the time, the media referred to the twister as an EF-4 tornado, but Dr. Forbes pointed to a heavy propane tank that had been ripped from its moorings and then thrown a distance of a quarter mile, providing evidence, Dr. Forbes stated, that supported a rating of EF-5. The damage survey by the National Weather Service later confirmed Dr. Forbes' assessment, officially upgrading the tornado to EF-5 (it appears that the National Weather Service based their rating on the damage to one of the schools (photograph below).
EF-5 damage at one of the schools in Moore, OK, struck by the killer tornado on May 20, 2013. Courtesy of the National Weather Service.
The issue with the airborne propane tank at Moore, Oklahoma, reminded me of the F-5 tornado that struck Wheatland, Pennsylvania, and the Ohio towns of Newton Falls and Niles on May 31, 1985. The F-5 tornado was one of a swarm of twisters over parts of the eastern Great Lakes region and southern Canada on this date (tornado tracks). I remember the F-4 tornado that rapidly knocked down thousands of trees in Black Moshannon Park (north of State College, Pa.), shaking the ground so much that "tremors" registered on seismometers on the campus of Penn State about 20 miles away. At any rate, Dr. Forbes' statement about the propane tank thrown a quarter mile in Moore, OK, reminded me of the 1985 outbreak because, while ripping through Niles, Ohio, the F-5 twister crumbled and tossed large storage tanks as it they were plastic play toys (see photograph), qualifying as F-5 damage.
I will never forget the depth of the human tragedy inflicted by the Moore tornado. I was also deeply saddened by the suffering of animals hurt and killed by the tornado (I cannot really bring myself to look at this photograph). As a scientist, I will remember the rapid intensification of the parent supercell's low-level mesocyclone (in other words, rapid mesocyclogenesis) and the resulting rapid intensification of the Moore tornado.
The goal of my blog today is to focus on the rapid intensification of the supercell's low-level mesocyclone in the hopes of giving you a better sense for why the supercell's tornado intensified so quickly.
The bulk vertical wind shear between the ground and an altitude of six kilometers was more than sufficient for supercells to develop on May 20 (radar loop, courtesy of SPC). To get a better sense of bulk shear and its role in determining thunderstorm mode (type), please review my previous blog on the tornado that struck Granbury, Texas, on May 16, 2013. Although the bulk shear between the ground and six kilometers favored supercells, the low-level wind shear in the lower troposphere was not particularly impressive on this day. As it turns out, the magnitude of the low-level wind shear plays a role in tornadogenesis. Let's investigate.
Spin around a horizontal axis (a streamline drawn parallel to the storm-relative inflow of the supercell) gets tilted into the vertical by a convective updraft, paving the wave for a rotating updraft. Courtesy of Jessica Higgs
Remember the annotated photograph of a supercell (shown above) that I've already highlighted in a couple of my blogs? By way of review, low-level vertical wind shear helps to create the spin around the horizontal orange line. By convention, the orange horizontal line represents a streamline of the storm-relative wind. By definition, the storm-relative wind is the observed wind vector minus the storm-motion vector. In plainer language, we can think of the storm-relative wind as the movement of air relative to a stationary storm (the storm is moving, of course, but we can better visualize the storm-relative wind by imagining the storm at a standstill). So the horizontal orange line you see in the image above indicates the storm-relative inflow, which ultimately gets tilted into the vertical by a developing convective updraft. The bottom line here is that the spin around the orange horizontal streamline (drawn parallel to the storm-relative wind) gets tilted into the vertical, creating the rotating updraft (mesocyclone), which is the hallmark of the supercell. For the record, the spin around the horizontal streamline for the storm-relative inflow is called streamwise vorticity. Again, streamwise vorticity is induced by low-level wind shear, which brings us back to the Moore tornadic supercell.
To a large degree, the strength of a supercell's mesocyclone depends on the rate at which the rotating updraft "ingests" streamwise vorticity. In this context, you can think of the mesocyclone ingesting streamwise vorticity as Lee slurping up spinning spaghetti noodles at the dinner table ("Grenci" is Italian). How's that for visualization!! One way to measure the rate at which a supercell ingests streamwise vorticity is storm-relative helicity (SRH). Essentially, storm-relative helicity is computed by multiplying the magnitude of the storm-relative inflow (the speed at which Lee slurps up his spaghetti noodles) multiplied by the magnitude of the streamwise vorticity (how fast the air is spinning about the local horizontal storm-relative streamline). Then we simply add up all the contributions from the ground to a designated altitude in the boundary layer (usually one kilometer or three kilometers). Ladies and gentlemen, I give you storm-relative helicity.
The 18Z Rapid Refresh model analysis of storm-relative helicity from the ground to an altitude of one kilometer. Courtesy of the Storm Prediction Center.
Check out the 18Z Rapid Refresh model analyses of the surface-to-one-kilometer storm-relative helicity (above) and the surface-to-three-kilometers storm-relative helicity on May 20, 2013 (before the supercells were initiated in central Oklahoma; see the 19Z mosaic of composite reflectivity and compare to the 18Z mosaic of composite reflectivity). Both RR analyses indicate that, at face value, the pattern was not screaming out any loud message that supercells capable of spawning severe tornadoes were imminent. That's because the threshold values above which forecasters start to pay closer attention to supercells spawning tornadoes are 100 square meters per square second (for sfc-1 km storm-relative helicity) and 250 square meters per square second (for sfc-3 km storm-relative helicity). Representative SRH values in central Oklahoma before storm initiation were in the neighborhood of 80-90 square meters per square second (for sfc-1 km storm-relative helicity) and 125 to 150 square meters per square second (for sfc-3 km storm-relative helicity). Yes, supercells were likely (the magnitude of the vertical wind shear between the ground and an altitude of six kilometers was large), but there were no screaming messages that an outbreak of supercellular tornadoes was imminent (possible, yes, but not written in stone).
So how could the mesocyclone associated with the Moore supercell rapidly intensify the way it did? In other words, how did the already rotating updraft spin faster? Keep in mind that we've already established that the low-level wind shear was not really noteworthy in the pre-storm environment. There are a couple of additional factors we should consider. While reading my account, keep in mind that, according to the National Weather Service, the tornado formed at 1945Z (2:45 P.M. CDT) and dissipated at 2035Z (3:35 P.M. CDT).
The 1946Z (2:46 P.M. CDT) base reflectivity (left) and storm-relative velocities (right) from the Doppler radar at Oklahoma City (KTLX). Larger image. Courtesy of NOAA.
First, the hook echo of the Moore supercell initially was not very impressive. To see what I mean, check out, above, the two-panel radar image from the Oklahoma City Doppler Radar at 1946Z on May 20 (2:46 P.M. CDT; larger image). That's base reflectivity (BR) on the left and storm-relative velocities (SRV) on the right (larger image). Note the "hook" on the reflectivity panel and the rather "modest" velocity couplet marking the position of the rotating updraft (mesocyclone). Also note the more diminutive thunderstorm south of the Moore supercell. In a very short time, this smaller storm merged with the Moore supercell (1951Z BR and SRV; 1955Z BR and SRV; 1959Z BR and SRV). By 2003Z (3:03 P.M. CDT), the hook echo was much more pronounced and the mesocyclone had rapidly intensified (see the 2003Z two-panel BR and SRV below; larger image). Indeed, the circulation associated with the mesocyclone was dramatically stronger (very strong inbound and outbound velocities). More importantly, there was a debris ball on the base reflectivity, indicative that there were non-meteorological airborne targets with high reflectivity. In other words, the tornado associated with the rapidly intensifying mesocyclone had also rapidly strengthened and was ripping and tossing debris into the air. Just five minutes later at 2008Z (3:08 P.M. CDT), the debris ball was even more dramatic as the tornado entered Moore (two-panel image).
The 2003Z (3:03 P.M. CDT) base reflectivity (left) and storm-relative velocities (right) from the Doppler radar at Oklahoma City on May 20, 2013. By this time, a smaller thunderstorm had merged with the Moore supercell and its mesocyclone had undergone rapid intensification. The tornado had also intensified...note the debris ball on the reflectivity panel. Larger image. Just five minutes later at 2008Z (3:08 P.M. CDT), the debris ball was even more dramatic as the tornado entered Moore (two-panel image). Courtesy of NOAA.
When thunderstorms merged, there was a marked increase in low-level convergence and the corresponding low-level vorticity (spin around a local vertical axis). In effect, the air column housing the mesocyclone stretched vertically and increased the spin (check out the first half of this copyrighted flash animation from Penn State's online certificate program; note how horizontal convergence, vertical stretching, and increase spin about a vertical axis go hand in hand). Lesson learned: Merging thunderstorms likely played a role in the rapid intensification of the low-level mesocyclone associated with the Moore tornado. I point out that there was also a merger of thunderstorms during the initial phases of the tornado that devastated Joplin, Missouri, on May 22, 2011.
Ingesting Vorticity from the Synoptic-Scale Front
The second factor that might have played a role in the rapid intensification of the low-level mesocyclone was that the Moore supercell ingested low-level vorticity associated with the stationary front. To get a sense for the bigger picture, check out the 18Z (2 P.M. CDT) surface analysis and the 19Z mosaic of composite reflectivity. Obviously, the stationary front helped to initiate severe storms in Oklahoma. Keep in mind that fronts lie in pressure troughs and that the wind shift that typically marks the position of fronts helps to produce convergence. Let's revisit the first part of the copyrighted flash animation from Penn State's online certificate program that connects the concepts of convergence and vorticity (spin about a vertical axis in this case). Revisit the 18Z surface analysis and compare it to the 18Z Rapid Refresh model analysis of 1000-mb absolute vorticity below (1000 mb is a proxy for the surface; larger image). Can you pick out the position of the stationary front? Yes, the stationary front coincided with the elongated maximum of absolute vorticity. Given that the Moore supercell was initiated along the stationary front, a reasonable question to ask is whether the storm ingested vorticity associated with the front.
18Z The Rapid Refresh model analysis of 1000-mb absolute vorticity on May 20, 2013. Units are expressed in 10^-5 seconds^-1. Note the elongated maximum in vorticity along the stationary front (18Z surface analysis). Larger image. Courtesy of Penn State.
Whether the complete explanation of the rapid mesocyclogenesis inside the Moore supercell involves both the merging of thunderstorms and the ingestion of vorticity from the stationary front (or some other factors such as instability, moisture, etc.) will likely result from future studies. My blog is only meant to help to stimulate discussion among my faithful Wunderground readers.
Whatever the answers, it was a storm that never will be forgotten in central Oklahoma.
Stay safe, my friends.
Comments will take a few seconds to appear.