Johns Hopkins Magazine -- November 1998
Johns Hopkins 
     Magazine Home



How do hurricane-force winds tear a house apart? To help get answers, Hopkins civil engineers are taking their research into the eye of the storm.

S C I E N C E    &    T E C H N O L O G Y

Blow, Bonnie, Blow!
By Sue De Pasquale

Picture at top: Jones and Porterfield chart Bonnie's blasts.
THE NOONTIME AUGUST sun shone high in the North Carolina sky as Michelle Porterfield cruised down Route 158, the lone car in the southbound lane headed for the small oceanfront Town of Southern Shores. In the northbound lane, bumper-to-bumper traffic snaked for miles. Hurricane Bonnie was finally about to hit, and residents and vacationers had been warned to get out while there was still time.

The 25-year-old Porterfield didn't waste time worrying about the impending danger. She had been waiting months for a storm like this; the success of her doctoral dissertation was riding on it. Besides, Porterfield was confident that she and Marguerite Jeansonne '00 would have enough time to get into town, secure their Hopkins experiment, and then get out again before the full wrath of Bonnie's fury descended.

Around 2 p.m. they pulled up in front of their destination, a low-lying beach house next to Southern Shores' Town Hall, about a quarter-mile from the beach. As they unlocked the building's front door and headed up the stairs, Porterfield slid out her cell phone and called her Hopkins advisor, Nick Jones. From his Latrobe Hall office back in Baltimore, the professor of civil engineering gave Porterfield a few last bits of advice, then the duo began to check and recheck their equipment.

Let Bonnie huff and puff--the stronger the better. As long as she didn't actually blow the house down, the data the Hopkins team would glean from her gales had the potential to be unprecedented.

WHEN HURRICANE ANDREW HIT South Florida in 1992, the effect was devastating: dozens of people were killed and more than 100,000 homes were destroyed. In all, the area sustained some $25 billion worth of damage. Some of that loss of life and property could have been prevented, experts believe, through sounder construction techniques and materials.

Enter Nick Jones, Hopkins's internationally known expert on structural dynamics, and wind and earthquake engineering. After the October 17 earthquake that rocked Oakland, California, in 1989, Jones was there the next day to inspect buildings and collect data to find out why some structures stood while others collapsed. He's also installed sensors on cable-stayed bridges (like the Fred Hartman Bridge in Houston), to determine how and why bridge cables oscillate, sometimes dangerously.

During hurricanes, some of the worst damage can occur when roofs are ripped from buildings. Once windows break, air rushes inside and appears to push upward, while outside, the roiling air above the roof is like a giant hand, yanking upward. At least, that's one theory. The problem is that scientists have never really field tested the interaction between sustained high winds, like those generated by a hurricane, and low-lying structures, like the beachfront homes found in ocean resorts.

That's why Jones's eyes lit up when he learned of Southern Shores' plan--initiated by the nonprofit Blue Sky Foundation--to construct a hurricane test house of various materials. The idea was to see which materials and methods of construction were most effective in withstanding hurricane damage. Could steel rods successfully be used to anchor the roof to the building's foundation, for instance? The project would get its funding from FEMA and industrial sponsors. Town officials hoped the test house, known as the Kern P. Pitts Center, would ultimately help insurers, developers, and emergency management officials come up with new building standards that would reduce storm damage.

At Hopkins, Jones quickly recognized the proposed test house as an ideal site to assess high-wind forces, and town leaders enthusiastically welcomed his involvement. Porterfield, who'd completed her BS with Jones and was looking for a doctoral project, was intrigued by this one. So, as the building's foundation was being laid and the frame of the house was going up, in January 1997, she and Jones (as well as undergraduates Steve Kelly '98 and Joseph Main '98, and electrical technician Jack Spangler '77) drove down to Southern Shores to do some prewiring. Once construction was complete the following October, they returned to fit the gray frame house with sensors and other equipment to measure the wind's forces.

The hub of their activity was the second-floor room in the southeastern corner. There they drilled holes and painstakingly installed 13 pressure gauges in the walls and roof, which would measure how much pressure the wind exerts. They also epoxied 20 strain gauges on the studs and rafters, to chart how parts of the house would be pushed and pulled. And to find out how, and whether, the building moved, they installed a displacement transducer along the inside of the house's northern wall.

The data captured by these sensors would be fed to a computer, which they set up down the hall in a utility room. During future storms, the researchers would be able to access the data remotely, from the Homewood campus.

Outside, as the shadows lengthened that autumn afternoon, the youthful Jones climbed to the top of a 33-foot utility pole near the house. There he installed an ultrasonic anemometer, which measures windspeed in three directions. Jones also climbed up to the house's chimney to set up devices to monitor weather conditions--a rain gauge, temperature sensor, barometric pressure sensor, and a prop-vane anemometer.

There was nothing left to do then but wait for a big storm to whip up. So the Hopkins group headed home to Baltimore. In February, a series of northeasters blew through the region, enabling them to test out their system. But none of those winter storms generated sustained winds exceeding 40 mph.

So, in mid-August, when meteorologists started tracking Hurricane Bonnie off the eastern Atlantic, Porterfield and Jones couldn't help but feel a sense of anticipation. If Bonnie ended up hitting North Carolina's Outer Banks broadside, as expected, then the "hurricane house" would finally see its first real test.

THE MIDAFTERNOON SUN is still bright outside as Porterfield and Jeansonne move about checking the sensors. On a visit weeks earlier, Porterfield had discovered water and insects coming in through the holes they had drilled in the roof and walls. She spent hours cleaning them. Today, the holes and the sensors remain clear and dry.

Next, she and Jeansonne do their best to secure the computer, by attaching it to the wall with a bungie cord. They also shorten the PC's power cord. That way, should the winds knock the computer off its desk, the power would go off before it hit the floor--crucial for preserving the data being stored inside.

Then they tackle their final, and most important, task: setting up a battery pack to keep the computer running for up to four hours should there be a local power outage.

By now, it's 5 p.m. and time to get home. On their way out of the house, the two young women note that the house's specially designed hurricane shutters are in place, then they take a quick walk to the beach to clear their heads and make sure they haven't forgotten anything crucial (Jones's idea). After all, there will be no second chances.

The beach is nearly deserted, and, though the sun still shines, the surf is churning ominously. After a few moments they head back to their car and join the northbound traffic, for the 10-hour trip back to Baltimore.

The storm hits Southern Shores once Porterfield is back in Baltimore; the next 48 hours pass in a caffeine-induced state of excitement as she stays glued to her PC.
THE STORM HITS Southern Shores once Porterfield is safely back in Baltimore; the next 48 hours pass in a caffeine-induced state of excitement as she stays glued to her PC in Latrobe Hall, watching the data come in. Jones is there for much of that time, too.

To their relief, everything works as planned; they download data onto computer disks every few hours as a precaution, but the house never does lose power, and the computer terminal stays securely in place. In fact, Bonnie's blasts do not prove as formidable as originally predicted. By the time the hurricane hits Southern Shores, it has been downgraded to a tropical storm.

Nevertheless, the 60 mph-plus winds that gust through the town prove strong enough to yield some important new data. It will take Porterfield and Jones months to fully analyze the 60 hours' worth of information that comes in. But preliminary analysis turns up some promising leads:

Data from the outdoor ultrasonic anemometer show some significant vertical wind speeds- -important because these previously were often assumed to be small and were frequently neglected in design and analysis. If the findings bear out, says Porterfield, it may become necessary for engineers to think more carefully about including vertical windspeed in standard analysis procedures.

Fortuitously, throughout most of the storm, the wind bisected the southeastern corner of the house, where most of the sensors are located. The data show high levels of pressure pulling away from the roof and eaves (as opposed to pushing in toward the structure)--the "giant hand" at work?

Until now, pressure measurements on low-rise buildings have been measured during relatively low winds--leading researchers to question whether it's possible to extrapolate the results to estimate what would happen at higher windspeeds. Porterfield will need to do more analysis before she can answer the question with certainty.

In some instances the wind's pressure tended to be high on all the sensors in one area, while in other instances the sensors in an area reported strong values at different times. The significance? If one location on the roof, for instance, experienced very high pressures, that alone may or may not be enough to pull the roof away. But if several areas on the roof all experience high pressures at the same time, the roof would be more likely to blow off.

Because the storm's wind blasts were not hurricane strength, Porterfield did not expect to see the wind causing large structural movement of the house. But data collected from the strain gauges showed some small fluctuations that may have been wind-induced.

Few strain measurements of this kind have been done in the past, partly because it's often difficult to determine which components of strain are due to the wind and which are due to temperature and humidity effects. Armed with the new data, Porterfield and Jones can begin to try to separate the wind-induced strain from the other strain components. Ultimately, that could help them determine how the wind's pressure affects different parts of the house.

JONES IS ENCOURAGED by these findings, and he intends to keep the experiment running at the Pitts Center for as long as he can continue to secure funding. (Until now their work has been funded mostly through grants from the NSF and the Blue Sky Project/Southern Shores.)

Meanwhile, Porterfield will keep one eye on TV's Weather Channel, as she continues to sort out the wealth of data garnered during Bonnie's visit to Southern Shores. Should another major storm slam into North Carolina this hurricane season, she'll be ready.

For more information visit, the website created by Jeff Lowe.