The venerable series of studies known as VORTEX is about to enter its third chapter in 22 years, this time with a change in venue. VORTEX Southeast
gets an official kick-off with a media day on February 29 in Huntsville, Alabama, the experiment’s home base. As its name implies, VORTEX-SE is focused on the southeastern U.S., where tornadoes can ravage the heavily forested, densely populated landscape at just about any time of year--including February, as highlighted in this week’s deadly outbreak on Tuesday
VORTEX-SE is targeting a relatively small area in its first field phase this March and April, relying more on fixed instruments and radiosondes (weather balloons) than on the mobile radars that crisscrossed the southern Great Plains during the earlier experiments. The new project also has a beefed-up social science component: investigators will be looking closely at how meteorologists present information about tornadic threats, and how people use and interpret that information. Figure 1.
Photographed here in southeast Mississippi, this violent tornado raged across more than 122 miles of Mississippi and Alabama
during the Super Outbreak of April 27, 2011. Rated an EF4, it killed seven people. Image credit: NWS/Jackson, MS
. Rolling through a different alley
The original VORTEX project (Verification of the Origins of Rotation in Tornadoes, 1994-95)
was a landmark in many ways. It was the first major use of the NSF-supported Doppler on Wheels
mobile radars, which have become mainstays of weather research. Data from the first VORTEX and its successor, VORTEX2 (2009-10)
, helped lead to a much more detailed and nuanced picture of how supercell thunderstorms gather and organize rotational energy (vorticity) and why only a small number of supercells manage to parlay their energy into making tornadoes.
The southern-Plains focus of the first two VORTEX projects emerged naturally, given that supercell tornadoes are more frequent and often more violent across this region than anywhere else on Earth. But deadly tornadoes also plague other parts of the nation. Geographer Michael Frates outlined four distinct tornado “alleys”
during his graduate studies at the University of Akron. As shown in Figure 2, these include the traditional Tornado Alley as well as Hoosier Alley (centered on Indiana), Carolina Alley (spanning parts of the North Carolina Piedmont and Coastal Plain), and Dixie Alley (extending from far east Texas to north Georgia). Figure 2.
The four “alleys” of peak U.S. tornado activity, as analyzed by geographer Michael Frates. Colored boxes denote the relative frequency of long-track F3, F4, and F5 tornadoes between 1950 and 2006. Image credit: Michael Frates, University of Akron.
It’s easy to quibble over exactly what constitutes a tornadic alley, or where its boundaries should be drawn, but it’s also clear that the southeast U.S. is especially tornado-vulnerable. That point was hammered home by the 2011 Super Outbreak
. Concentrating its fury in Mississippi, Alabama, Georgia, and southern Tennessee, the outbreak took more than 300 lives on April 27. It was the nation’s largest single-day tornado toll since the Super Outbreak of April 3, 1974
, which also hit the same area hard (as well as points further north).
With their constituents suffering, regional lawmakers decided to work toward bringing tornado research from the Plains into the hard-hit Southeast. Congress provided guidance in 2014 to NOAA and NSF to collaborate in organizing and conducting a VORTEX-SE research study, and more than $5 million was allocated by Congress in fiscal year 2016. VORTEX-SE organizers have put in place a planning process for a succession of research projects that, if funding support continues, are designed to lead to improvements in tornado forecasts and responses in the Southeast. A grab bag of threats
Tornadoes obey the same laws of physics no matter where they develop, but some types of tornadoes are especially problematic in the Southeast. Violent “wedge” tornadoes can be more difficult to see than on the Plains because of lower cloud bases produced by richer low-level moisture and because of the widespread hills and trees. The region’s high frequency of severe weather year-round leads to a perennial threat from weaker, short-lived (but still dangerous) tornadoes produced by quasi-linear convective systems (QLCSs)
, or what are traditionally known as squall lines. Thunderstorms are often moving quickly, sometimes at more than 50 mph. Nighttime tornadoes are also more common here than in many areas, which adds to public safety concerns.
Rebecca Morss (National Center for Atmospheric Research) is shepherding the VORTEX-SE social science component as a member of the project’s scientific steering committee. “The Southeast has unique issues related to tornadoes--meteorologically, operationally, and societally,” Morss told me in an interview. One key example: about half of the nation’s mobile/manufactured homes are located across the Southeast, where they constitute more than 20% of housing units
in many counties. Thousands of manufactured homes are located on small acreages, which complicates the task of getting to shelter.
Along with challenges unique to the Southeast, Morss and colleagues are teaming up with NWS forecasters to advance a longer-term, national-scale effort to improve forecast and warning communication. Those warnings aren’t received in a vacuum, she stressed. “What do people do when they get a warning? Do they look at social media? Do they look outside? We often think of individual decision making, but it’s really a social process. People are usually making decisions about how to act in conjunction with other family members and what friends are doing.” Figure 3.
A soda machine is seen in debris in the Holt neighbourhood of Tuscaloosa, Alabama, on May 1, 2011, following the devastating tornado of April 27. Many trailer homes in the area were destroyed by the tornado. Image credit: Mandel Ngan/AFP/Getty Images. Spinning up an experiment at tornadic speed (or close to it)
Most major atmospheric research campaigns--including the first two VORTEX projects--emerge after several years of proposals (typically to NSF and/or NOAA), which means the scientific goals tend to be fully developed by the time the campaign gets under way. In the case of VORTEX-SE, researchers have funding in hand while they continue to work out the priorities and strategies that are most relevant to the study area. A workshop held in Huntsville last November drew more than 100 on-site participants, with virtual collaborations boosting the total number of researchers involved to around 150.
“It’s turning into a really good science program,” said coordinator Erik Rasmussen. A veteran tornado researcher involved with the two prior VORTEX projects, Rasmussen is steering VORTEX-SE from NOAA’s National Severe Storms Laboratory, the official organizing agency.
Much of the funding thus far has gone into an initial set of NOAA-allocated grants
to scientists scattered across the Southeast and beyond. Some of these research threads will extend beyond the spring, such as one aimed at improving prediction of tornadoes in landfalling hurricanes and tropical storms. Others are being woven into a set of intensive observing periods this March and April. These IOPs will be conducted from an operations center at the University of Alabama in Huntsville’s new SWIRLL facility
(Severe Weather Institute & Radar Lightning Laboratory). Local researchers will figure heavily into VORTEX-SE. Along with a number of scientists and facilities from UAH’s Atmospheric Science Department
, the project will draw on the capabilities of the Huntsville-based NASA Marshall Space Flight Center
and its expertise in lightning and storm electricity
This year’s field campaign will focus on the boundary layer, the lowest mile or so of the atmosphere. Many of the Southeast’s severe weather episodes are intense but localized, with high wind shear and relatively low instability (or CAPE
). “Some people have hypothesized there are little pockets of CAPE in these situations,” says Rasmussen. “Are models keeping up with such developments? I think that’ll end up being the main focus this year.” Figure 4.
A StickNet unit is deployed in far western Nebraska ahead of a tornado-producing thunderstorm on June 5, 2009, during the VORTEX2 experiment. Image credit: Bob Henson.
Out in the field, there’ll be radiosonde-launching mobile units from the University of Louisiana at Monroe as well as Mississippi State, North Carolina State, Purdue, and Texas Tech. Fixed-site launches will be conducted by UAH and NOAA’s Air Resources Laboratory. A dozen compact StickNet instrument packages
from Texas Tech will also be deployed. To see how much the StickNet data improves model performance, NOAA’s Earth System Research Laboratory will carry out post-storm runs using the high-resolution HRRR
models. The extra data may help better characterize pools of cool surface air left behind by thunderstorms. This, in turn, could improve short-term forecasts of storm rotation and longevity.
Rather than moving along with major supercells for dozens or even hundreds of miles, as was the case during prior VORTEX campaigns, the observing platforms this spring will be concentrated in northern Alabama and southern Tennessee, largely in the warning area served by the National Weather Service office in Huntsville.
“We have greatly limited mobility, and to complicate that, we have fast-moving storms,” said Rasmussen. “We’re going to throw a network up in a good dual-Doppler coverage area and take most of the obervations in that network.”
Another VORTEX veteran, Yvette Richardson (Pennsylvania State University), is heading up the physical science wing of the VORTEX-SE steering committee. “We’re putting a great deal of thought into which problems can be addressed in the Southeast, given the topography and tree cover. It's a challenging part of the country to do an observing campaign.” Rounding out the steering committee is Kevin Laws, science and operations officer at the NWS/Birmingham office, who will shepherd the operational science elements.
Tornadoes and terrain
The role of landscape will get increasing emphasis in 2017 and beyond. One of the current grant-funded projects, led by NOAA/ARL, will use high-resolution, 3-D observations to assess the variations in the low-level atmosphere as well as the roughness of the land surface. “This knowledge will lead to a better understanding of how local conditions can lead to more favorable environments for tornadoes when conventional understanding would indicate tornadoes are unlikely,” states the grant abstract.Figure 5.
Rotational tracks corresponding to tornadoes during the Super Outbreak of April 27, 2011. This map was produced by the On-Demand system of NOAA’s National Severe Storms Laboratory. Image credit: NOAA/NSSL
The extent to which tornadic storms can be shaped by landscape and topography has long been controversial and difficult to assess. Generations of Tornado Alley dwellers grew up hearing about towns that were “protected” from tornadoes by hills or ridges. (Some residents of Topeka, KS, thought that a local feature called Burnett’s Mound would steer tornadoes away, until a devastating twister
plowed through town on June 8, 1966, killing 17 people.) It’s now clear that tornadoes can maintain their strength while climbing and descending hills, a message forecasters have been stressing for years. Yet this doesn’t rule out the possibility that some tornadic storms get a boost from landscape features that help to channel low-level winds.
At Penn State, graduate student Branden Katona is looking into such possibilities. He’s built a climatology of where environmental characteristics that could influence storm rotation are maximized, based on three years of HRRR forecasts across the Southeast. Katona focused on two variables: significant tornado parameter
(STP, which incorporates both instability and wind shear) and storm-relative helicity
(SRH, a measure of potential storm rotation). “Quite a few local maxima in both SRH and STP follow terrain contours,” Katona pointed out in an email. “In particular, the SRH maxima are aligned with local ridge axes in both northeastern Alabama and in the Great Smoky Mountains in southeastern Tennessee. Additionally, a local STP maximum stretching from northeast Mississippi down into western Alabama is aligned with a local valley axis.” Katona will soon be comparing the HRRR data to actual tornado paths: “I expect at least some of the model perturbations to coincide with clusters of tornadoes.”
Bob Henson PS from Dr. Jeff Masters and Bob Henson: Help us re-name this blog!
As regular readers know, we have been co-authoring this blog since early 2015. It’s time for a name change from “Dr. Jeff Masters’ WunderBlog” to a new name that reflects our dual authorship. We have some ideas in mind, but we’re hoping that you, our readers, can help us expand our pool of nominees. We’re looking for a cool, pithy name--ideally just two or three words--that reflects the spirit of this blog in covering both weather and climate, with particular emphases on tropical meteorology, severe weather, and climate change. Please feel free to chime in with your suggestions in the comments section of this post. If you’re a WU member, you can drop us a line via WU Mail if you prefer. All suggestions made by March 10 will be considered. Thanks in advance for your creativity!Figure 6
. Climatological locations of storm-relative helicity (STH) in the lowest kilometer of the atmosphere (left) and elevation (right), generated by the high-resolution HRRR model and averaged across 358 days with thunderstorms during the period Feb. 1 through Sept. 30, 2013-2015. All STH calculations are at 2100 UTC (4 PM or 5 PM local time, around peak solar heating, when most surface-based thunderstorms would be developing or ongoing). Storm-relative helicity indicates the likelihood that updrafts in a right-moving supercell storm will rotate. Image credit: Branden Katona, Penn State University.