The April 2011 earthquake in Tuscaloosa, Alabama, that killed dozens and destroyed parts of the city, shown here, as well as the more devastating tornado in Joplin, Missouri, that May prompted researchers at the University of Arkansas to explore the relationship between twisters and the slopes of terrain they ravage. Wikimedia Commons/Silosarg
Researchers are examining how terrain influences the strength and path of devastating cyclones to help inform community planning.
September 10, 2013—Researchers at the University of Arkansas are exploring the relationship between tornadoes and the terrain they ravage in an effort to determine if there is a feasible way to develop communities that will be less vulnerable to the devastating storms in the future.
The research is led by R. Panneer Selvam, Ph.D., P.E., F.ASCE, a professor of civil engineering at the University of Arkansas. Selvam has researched tornadoes for more than 30 years, first examining their interaction with structures before moving on to their interactions with terrain.
“We started looking into this matter about two or three years ago,” Selvam says. “Now I have three Ph.D. students working on it.” Selvam and his students recently presented research papers at the 12th Americas Conference on Wind Engineering, held in Seattle this summer.
Graduate student Piotr Gorecki worked with Selvam in developing a three-dimensional computational fluid dynamics (CFD) model of a tornado and then simulating collisions with three rectangular hills of different sizes. The tornado in the CFD model had a vortex diameter of 36 m. The heights of the hills were 12 m, 24 m, and 36 m.
The models show that barriers that are nearly equal to or exceed the diameter of a tornado’s vortex have a significant mitigating effect on the cyclone’s force, causing the vortex to split and weakening wind velocity by as much as 50 percent. Smaller hills have little effect on vortex strength. Researchers began looking for a way to validate the findings in the field.
“When there is not reasonable verification in some form, models are always models,” Selvam says. “People don’t look at them very seriously.”
To develop this validation, graduate student Nawfal Ahmed worked with Selvam to analyze satellite images of tornado damage available via Google Earth. The researchers examined the aftermath of the deadly 2011 tornadoes in Joplin, Missouri, and Tuscaloosa, Alabama.
The Tuscaloosa tornado destroyed parts of the city and neighboring Birmingham on April 27, causing an estimated $2.2 billion in damage and killing 64. The Joplin tornado cut a swath of destruction nearly 1 mi wide on May 22. The tornado reached level 5 on the Enhanced Fujita scale, the winds exceeding 200 mph. The tornado destroyed 2,000 homes, killed approximately 160 people, and caused an estimated $2.8 billion in damage.
Satellite images available in the days following the tornadoes showed pronounced patterns of damage that relate to the terrain, the researchers found.
“We went in and looked at what was happening up the hill and down the hill—the damages,” Selvam says. “With our fluid mechanics background, what we thought we [would be] able to see was there. It was documented.” The tornadoes created more damage when climbing up the hill and less damage when going down a hill.
Selvam explains that tornadoes expend more energy in climbing a hill. Increased friction causes increased damage. After the tornado peaks a hill, momentum causes it to glide slightly before it returns solidly to the ground.
The satellite images also revealed that whenever possible, it appears the tornadoes climb to higher elevations. In fact, when a valley area is surrounded by hills, tornadoes often largely skip over the valley to a surrounding hill. Although the satellite research adds validity to work from the CFD model, more research is needed, Selvam says.
“Still they are sitting as apples and oranges,” Selvam explains. He plans to develop a research grant he intends to submit to the National Science Foundation for funding to enter GPS and elevation data from Joplin and Tuscaloosa into the CFD model. If the simulations result in damage patterns similar to what was observed, it will provide greater validation.
Selvam’s goal is to determine if it is possible, on the basis of a better understanding of a tornado’s likely movement along terrain, to develop cost effective solutions to protect communities. He notes that because the tornado vortex diameter and strength are both factors, it might not be possible to mitigate the largest, strongest storms.
“Hopefully, in the coming years, we will come to that point. That is the idea that has been floating in my mind for several years,” Selvam says. “This is only the beginning. Can we …create engineered structures that can protect certain regions? The cost [would be] far cheaper.”