Home > Highlights > Shelter and the Storm

Shelter and the Storm

AUTHOR: Nina Welding

PUBLISHED: October 2, 2017

ND engineers lead assessment of structural damage from hurricanes

A few days after Hurricane Harvey struck Houston and a week after classes started at Notre Dame, Tracy Kijewski-Correa, a structural engineering professor, called about an urgent email while she waited in the car during her son’s tutoring session.

Chase the storm, assess the damage, determine how to improve future construction.

That was the request of GEER, an association of geologists and earth scientists that organizes immediate response teams for earthquakes and other disasters. The group wanted Kijewski-Correa to organize a rapid reconnaissance of Harvey’s massive winds on different types of building structures in a state that does not mandate adoption and enforcement of its residential building code statewide.

While GEER had organized other scientific teams before, it had never done so for structural engineers. Kijewski-Correa would later expand this first-ever, multi-university coordinated response to include hurricanes Irma and Maria, some of the largest ever to hit U.S. territory.

“Reconnaissance brings home why you’re an engineer,” she said. “You’re trying to channel that call to service. We can study math and models, but the human face is different. You look at this house that’s blown apart like toothpicks and it reminds you what nature can do. There’s a certain awe and respect; then as engineers, we pivot to ‘How can we build better?’”

Kijewski-Correa dropped everything for the unique opportunity to learn how buildings, often on the same block, either withstood the storm or tumbled into scattered shards. She spent Labor Day weekend assembling teams, planning travel, and creating policies and procedures on the fly. Ten days later, she was coordinating three teams near Rockport, Texas, where the fiercest winds made landfall south of the record rains that turned parts of Houston into a biblical lake.

The teams arrived as the residents returned—after the humanitarian disaster was over but before cleanup destroyed the evidence. They joined survivors to eat sloppy Joes and dilly bars from the Red Cross. For four days, they climbed over the rubble to take measurements, and they heard bereft homeowners share their stories of loss.

"We can study math and models, but the human face is different."

“The levels of destruction were breathtaking,” she said. “One of the hardest parts of a ground team is that you want to give a homeowner the dignity of talking about losing everything they own. You’re trying to cover as much ground as you can and collect photos and forensics, but that was someone’s home, their kid’s bedroom. You have to suspend your engineering for a little and just listen.”

They explored homes that rocked and swayed, gaps opening and closing, but only lost some shingles. Some residents who rode out the storm said the wind battered them from changing directions, like a surgeon probing for a weak point from different angles. Kijewski-Correa said those houses likely had metal hurricane straps that reinforce the connection of the roof rafters to the wall studs.

“Engineers are looking at load path,” she said. “In some houses, there’s nothing left because the builder used only traditional construction. There was nothing sufficient tying the roof to the walls or the walls to the foundation that could transfer the load.”

Bad:
•  With traditional construction, there is nothing sufficient tying the roof to the walls or the walls to the foundation.
•  Unprotected windows and doors that open inward are vulnerable to envelope penetration that can result in rapid pressurization and the roof blowing off.

Good:
•  Hurricane straps are metal ties that reinforce the connection of the walls to the roof rafters.
•  Hurricane shutters and building the door to open outward can protect the building envelope and prevent the roof blowing off.


Sometimes, the force of sustained winds above 100 miles per hour left behind only an elevated platform: the walls and roof had blown away, the interior possessions scattered for several blocks. One man told them, “This is what $400,000 gets you when you don’t build to code.”

After Hurricane Andrew ravaged Florida in 1992, the state stepped up the level and enforcement of its building codes. Texas and other southern states did not. In particular, mobile homes, older construction and do-it-yourself structures were shredded, Kijewski-Correa said.

Hurricanes like Harvey often spin off tornadoes, she said. Houses can’t be built to withstand that much force or they would feel like bunkers.

Still, smaller steps can make a big difference. Engineers can educate home builders and owners about best construction practices. Simple changes such as quality hurricane shutters and building doors to open out rather than in can prevent the “envelope penetration” that results in rapid pressurization and the roof blowing off.

“Sweeping code reform is a big political process,” she said. “We don’t expect most states to be like Florida, though what they did was admirable. In a comparison between Harvey and Irma, it’s night and day in terms of the performance of structures.”

Kijewski-Correa said another study she is conducting shows that adoption of strict building codes can depend on more than just economic levels, including everything from people’s religious convictions to their feelings about the role of the federal government. Tailoring construction education for different audiences can make it more effective.

“I laugh when people criticize others for not knowing about their home’s construction standards,” she said. “I mean, do you know what kind of roof you have? The system is broken. It doesn’t protect home buyers.”

Disaster reconnaissance missions are standard practice in civil engineering because it’s the only way to validate structural design, she said. Buildings are too large to prototype first, the way car companies crash multiple vehicles to test for safety. So disasters provide the real-life opportunity to evaluate load performance.



“We saw some isolated victories where structures were able to withstand 130-mile-per-hour winds,” she said. “But we saw more instances of heartbreak, houses that didn’t have consistent load path and couldn’t transfer the wind force, so the roof and walls peeled apart.”

She said structural engineering professors and other building experts often work on their own or with a close circle of collaborators to produce their publications. This effort was different, soliciting unfamiliar participants nationwide. GEER told Kijewski-Correa that they reached out to her because she coordinated a difficult but successful reconnaissance last year in Haiti after Hurricane Matthew.

Compared to going it alone, a coordinated response can collect far more data, standardize it, and make it widely available. The goal, she said, was to avoid duplication of effort, where structural engineers go off “like cowboys riding on their own, and you end up with five reports on the same building collapse.”

The Harvey teams include Andrew Kennedy and Alexandros Taflanidis, two other civil engineers from Notre Dame, as well as engineers from Princeton, Rutgers, UMass Amherst, LSU, Nebraska and Texas, among others. Irma teams include engineers from universities in Florida and Puerto Rico.

Kijewski-Correa created a data-sharing policy that requires fast action — publishing data within 30 days and a comprehensive data paper that credits all collaborators by the end of the year. She also organized distinct responsibilities.
Palm trees

For more scenes from Kijewski-Correa's on-site assessments, head to stories.nd.edu.


In Texas, a coastal team studied the combination of storm surge and wind, two wind teams surveyed hard-hit neighborhoods, and a liDAR (land-based radar) team created 3-D maps of storm damage. As coordinator, she was supposed to coordinate this work remotely but ended up leading a wind team when someone else had to cancel.

For Irma, she divided the storm’s vast distance into four regions. Irma’s path provided a special opportunity for study because it moved from the U.S. Virgin Islands to Puerto Rico to Florida, diminishing from a record-powerful level 5 to 1 along the way.

Reconnaissance trip costs are surprisingly low – about $40,000 for Harvey and $80,000 for Irma. The National Science Foundation funds Kijewski-Correa’s project, and she reimburses the collaborators for travel expenses only, plus some stipends for grad students who curate data.

Engineers across the country jumped at the chance even though they were volunteering their time. Seeing a disaster first-hand can re-anchor structural experts, reminding them why they do calculations and run complex models, Kijewski-Correa said.

“We’ve never done a coordinated effort before,” she said. ‘This is the first time we’ve created a process. People who don’t have an incentive to play nice or play together are being asked to do so. It’s really testing the waters in how universities can collaborate across a large geographic region to respond to an event. But we became engineers for this reason.”

Ready to respond, her teams will soon be off to Puerto Rico and the U.S. Virgin Islands to study the impact of Hurricane Maria, then back to Haiti to follow up on rebuilding stronger.

— Originally produced and published by the Office of Public Affairs and Communications