By Stephen Budiansky

THANKS TO HURRICANES LIKE CHARLIE, FRANCES, AND IVAN, ENGINEERING STUDENTS IN FLORIDA ARE GETTING REAL-WORLD RESEARCH EXPERIENCE—AND HELPING TO MAKE HOUSES SAFER IN THE PROCESS.

"We sit in classrooms and teach students about structural analysis and draw little pictures on the board and say here are the forces and can you calculate the stresses," says Kurt Gurley, associate professor in the Department of Civil and Coastal Engineering at the University of Florida. "But there's enormous benefits in seeing the enormous complexity of what goes on in real life—and getting a sense of how little we really do know."

If you live in Florida and are looking for a lesson in reality and complexity, nature can oblige. There are few things more real than a hurricane—and few things more complex than the pattern of destruction one leaves in its wake. For the past five years Gurley and his students have taken advantage of what nature has provided to carry out an innovative research project on hurricanes and the damage they do. This year there has certainly been no shortage of educational opportunities.

Every time a storm approaches Florida, the engineers set out in a caravan of trucks and trailers to set up portable wind-monitoring towers and install pressure sensors on pre-selected houses close to the storm's predicted path. Not only does this project provide a hands-on lesson in real-world
research for engineering majors and graduate students, it also answers some fundamental questions about how houses stand up to the destructive forces of a full-blown hurricane.

The research has already yielded practical recommendations on how building codes and construction practices need to be revised for hurricane-prone areas. Along the way, they have also provided meteorologists with a wealth of detailed data on how hurricane winds behave at ground level—data that simply has not been available to researchers before.

When surveying the damage wrought by a hurricane, the most difficult—and most basic—thing that needs to be determined is whether the damage that occurred to houses was caused by local, extreme wind gusts that exceeded the design specifications of the structures, or by a flaw in the building codes or workmanship that left the building unable to meet its design specifications.

One of the difficulties in making that determination is that although meteorologists have become proficient at measuring and forecasting the maximum sustained winds in a hurricane, they still have almost no way of predicting the gusts that typically appear in local, short-lived, and chaotic patterns and cause the most damage. A gust might last just a few seconds, but reach speeds 50 to 60 percent higher than the maximum sustained winds. In a typical suburb, which is filled with obstacles that chop the wind up into turbulent patterns, the gusts become even more extreme and localized.

Gurley and his co-workers in the Florida Coastal Monitoring Project (FCMP) realized that the first challenge was to establish "ground truth": determining what actual, local wind speeds were responsible for the observed damage that occurred in a storm.

Engineers at Clemson University—where the project was initiated in 1999—designed the portable towers used in the project. They are sturdy enough to withstand 200 mph winds but light enough to be towed on a regular trailer behind a van or pickup truck. They are easy enough to set up that a crew can assemble them in about 20 minutes and leave well before the storm strikes. "This is a safe operation, I should stress," Gurley says. "We retreat to a safe location during the storm. We will not sacrifice safety to get a piece of data."

The towers are instrumented with anemometers and computers that record wind speed data at 5 and 10 meters above ground level every 1/100th of a second—providing an astonishingly detailed look at wind gusts in one precise location. It's vastly more detailed data than is provided by the National Oceanic and Atmospheric Administration's (NOAA) fixed weather stations at airports, or by the radiosondes that hurricane chasers drop by parachute into a storm. And because the FCMP's towers are mobile, specifically designed to withstand high winds, and provided with backup power, they are more likely to avoid destruction or having their power knocked out—the typical fate of airport weather stations that have the fortune, or misfortune, of ending up in a hurricane's path.

Still, mishaps do occur. During Hurricane Charley this year, the engineers set up two towers along the stretch of Florida's east coast where landfall was predicted. The eye of the storm, as Gurley ruefully puts it, "used our towers as goal posts" and passed right between them, causing them to miss recording the maximum winds. And during Hurricane Frances, flying debris took out some of the instruments on one of the towers, though the tower itself survived.

Windy Classroom

Students, both graduate and undergraduate, are directly involved in everything from designing the computer equipment, to analyzing the data afterwards, to the hands-on work of setting up the towers and instruments. With four full-blown hurricanes hitting Florida this season already—the first time that has happened in 40 years—it's been a busy time for all. "This project couldn't exist without the students," Gurley says. "It's driven entirely by the dedication of the students I work with. They have to know everything from details of computer architecture and instrumentation to how to drop a diesel fuel tank out of a Ford F250 to drain water out of it. So we need people who are mechanically inclined, electronically inclined—and people who are just gung-ho."

A recent graduate student of Gurley's, Forrest Masters—now an assistant professor at Florida International University—designed and created a system that has been added to the tower instrumentation package to transmit real-time data over a cellular telephone connection. The data is relayed directly to a website where meteorologists, utility companies, and emergency management agencies (and anyone else, too: http://grove.ufl.edu/~fcmp/) can access it in the middle of a storm. The five to six undergraduates who work on the project get paid an hourly wage and "it's a great thing to have their résumé to show they've participated in actual research," Gurley notes.

While the instrument towers provide ground truth on actual wind speeds that impinge on houses in the path of the storm, a second set of instruments is used to fill another void in data that has long vexed building engineers. To calculate how a given wind speed relates to a given structural load that a house has to be able to resist, engineers involved in setting building codes use wind tunnel studies. They place a miniature of a house in a wind tunnel and measure the pressures that result on various parts of the house. The trouble with that, Gurley observes, is that wind tunnel studies are always an approximation and involve a lot of simplifying assumptions. And there have been almost no data collected to validate or calibrate the wind tunnel results against those from the real world.

To fill that void, the FCMP team has pre-selected 32 houses located at 10 mile intervals along the coastal regions that lie in the major storm paths. The houses have been equipped with cables and connections for computer equipment and stainless steel brackets on the roofs to hold pressure sensors. When a storm is approaching, the engineers and student teams put the sensors in place and hook up the computers. (In exchange for their cooperation, the homeowners receive an analysis of their home's structural vulnerabilities and free upgrades to make them more wind-resistant.)

"This gives us the actual forces on a building as the wind flows over," Gurley explains. And then "every time we take data from one of our houses, we build a miniature of it and put it in a wind tunnel to validate what we see in the wind tunnel. So now we have a bridge between the building code and the real world."

The final step in the process is to go back once the storm has passed and see what damage has been done to nearby houses. Overall, Gurley says, the more stringent building codes put in place after Hurricane Andrew in 1994 have been noticeably effective in reducing damage, especially in the case of mobile homes. After Hurricane Charley, he says, "we could see the 1994 and later [mobile] homes a mile away, literally, because they were the only ones left standing." Newer, site-built homes also fared well. The researchers did, however, find problems. For example, clay roof tiles that had been mortared in but not fastened with screws tended to come loose and "started flying around like little 10-pound missiles."

Tim Reinhold, who helped start the project when he was an engineering professor at Clemson, is now vice-president of the Institute for Business & Home Safety, a nonprofit research organization funded by insurance companies. Reinhold emphasizes that this kind of real-world data and analysis is essential to setting effective—and cost-effective—building codes for new houses, and especially in trying to decide what retrofits need to be made to older houses. Given the current rate of new construction, he notes, "We're able to address only about 2 percent of the housing stock a year through the building codes. So you've got this vast number of older structures built to a variety of older standards, and with poorer enforcement in the past. So there are a lot of very vulnerable structures."

Being able to compare the performance of retrofitted houses with unimproved houses around them, and with actual wind and pressure data as a reference point, can point directly to what steps are effective in making older houses more hurricane-resistant.

The FCMP has been run on something of a shoestring budget that includes some funding from NOAA. The Florida state government has helped pay for equipment development, but Gurley says it's always a scramble to cover the operational costs of meals, lodging, fuel, and student pay. But it's a great example of how innovative thinking can both fill a real-world need for engineering expertise and provide students an experience nonpareil.

Stephen Budiansky is a freelance writer based in Leesburg, Va. He is the author, most recently, of Air Power: The Men, Machines, and Ideas That Revolutionized War, From Kitty Hawk to Gulf War II (Viking).