UA team studying lay of the land, tornadoes

Focus is on topography’s effect on twisters

FAYETTEVILLE - Not many grant programs exist that fund the kind of research that the University of Arkansas’ Panneer Selvam has been doing for years concerning buildings that can better withstand tornadoes and high winds.

So about five years ago, Selvam and his graduate students at the Fayetteville campus decided to try a different approach: studying whether hills and valleys can diminish a twister’s destructive power.

Some early results, presented this summer at the 12th Americas Conference on Wind Engineering, indicate that in some cases, terrain can disrupt a tornado’s vortex or slow the twister down. It has to do with various comparisons of factors such as the width of the tornado’s vortex, the twister’s size and velocity, the size of a building, the height of a hill and the location of a building in relation to a hill.

“When the building is much larger than the tornado’s width, then the effect ismuch more like straight-line winds,” Selvam said Thursday while demonstrating the computer models his team created for the research.

The research has included 3-D computer modeling and field studies using Google Earth images of destruction from large tornadoes in places like Tuscaloosa, Ala., and Joplin, Mo. He estimated that the work will take another five years and the help of three to five students before he has strong scientific evidence - though the initialresults look promising.

“We are looking for science-based proof,” Selvam said.

One question that he hopes to answer: When a tornado moves eastward from the flat, open terrain of Oklahoma toward Arkansas, does the Fayetteville area’s mountainous terrain slow it? Or would it lose energy even if the state’s northwest corner were also flat and open?

Ultimately, the research could help cities and developers identify the safest locations for public disaster shelters, housing developments and mobile home parks, Selvam said.

Selvam is a university professor in the UA College of Engineering’s civil engineering department. He also holds an endowed professorship, the James T. Womble Professorship in Computational Mechanics and Nanotechnology Modeling.

TORNADO BEHAVIOR

Among other things, the scientists have discovered that the lower levels of a tornado’s swirling vortex are substantially disrupted when it hits a hill whose height isequal to or greater than the vortex’s radius. The research also supports a previous field study in finding that wind velocity is markedly reduced on a hill’s leeward side - or the side sheltered from the wind - determining that the wind speeds are reduced by about 40 percent.

That indicates the leeward sides of hills are the best places to build housing or shelters, but Selvam said more investigation is needed.

In a paper that Selvam co-authored with doctoral student Piotr Gorecki of Poland, they contrasted a tornado’s behavior in regardto the smallest of three hills and that of the largest. They used computational fluid dynamics, a method of studying winds’ effects on structures, to set up 3-D computer models.

The tornado flowed smoothly over a 39-foot-high hill and its shape was preserved, the paper read. But in the case of a 118-foot-high hill, the vortex was disrupted and weakened.

In another paper submitted to the conference, Selvam and doctoral student Nawfal Ahmed of Iraq looked at the effects of terrain elevation - hills and valleys - on a tornado’s path.

For that study, Ahmed and Selvam used Google Earth software to study the effect of elevation changes on a tornado’s path, vortex strength and the damage it caused.

Google Earth captured images of the 2011 tornadoes in Tuscaloosa, Ala., and Joplin, Mo. It showed the one-day and 16-day aftermath of each storm.

One of Ahmed and Selvam’s findings was that tornadoes cause more damage when they travel up a hill than traveling down. Also, tornadoes tend to move toward higher elevations.

Finally, in regions surrounded by hills, tornadoes tend to jump over valleys, inflicting more damage on the tops of hills.

“This contradicts the findings expressed in previous studies,” according to the paper.

FUNDING LACKING

Right now, Selvam is using the Womble endowed funds for wind and tornado research because he has no funding from other sources such as the National Science Foundation or the Federal Emergency Management Agency.

Collis Geren, who retired after serving as UA’s vice provost for research and dean of the Graduate School, recalled being frustrated that most funding programs are tornado-disaster reactive rather than proactive.

“Most of the funding there is for rebuilding - not for prediction,” Geren said.

There’s not a total lack of funding for proactive tornado research, but it’s generally scarce compared with some other research areas, said Jim Rankin, UA vice provost for research and economic development.

One potential grant source for this kind of research is the National Science Foundation’s Hazard Mitigationand Structural Engineering program, he said.

It supports research to mitigate disasters’ effects on civil infrastructure and to advance the reliability of buildings, according to the foundation’s website. The program covers earthquakes, tsunamis, hurricanes and tornadoes.

“The recent disasters have increased the awareness that this type of research is needed,” Rankin wrote in an email Thursday, noting that Science Daily reported in August 2011, a few months after the Joplin and Tuscaloosa tornadoes, that a research team funded by the National Science Foundation wrote a report calling for changes after examining the extent of the damage.

It called for such things as “more intensive engineering design and more rigorous, localized construction and inspection standards.” The research team said modern building codes are “kind of a bare minimum” for resisting tornadic winds. The problem is exacerbated if construction is inadequate or codes are not enforced.

“We have a cultural way of building,” Selvam said, which he considers ingrained.

Buildings are “not designed to withstand high winds, even,” he said.

Northwest Arkansas, Pages 15 on 10/20/2013

Upcoming Events