The subduction zone’s leading edge sits about seventy miles offshore, at the seam in the seafloor where oceanic and continental plates meet. But the three-dimensional interface between the tectonic plates—the fault—is like a giant ramp that extends underneath the Northwest’s urban corridor. If the portion of the fault that ruptures is mostly offshore, shaking will be muted by the time it reaches the cities. If the rupture is wider, continuing under the Olympic Peninsula and Oregon’s Coast Range, cities will be hit harder. The USGS estimates a wide rupture would intensify shaking in Seattle by about 50 percent over current forecasts and double the level in Portland.
Heaton and a graduate student developed a computer model to simulate the effects of different Cascadia scenarios on hypothetical steel high-rises in Seattle. In the worst-case, widest rupture, all the buildings collapsed. The level of uncertainty is so high that when Heaton served on a federal advisory panel, he argued that the USGS shouldn’t produce a hazard map for high-rise designers to use. “For tall buildings in Seattle, we’re just shooting in the dark.”
High-rises survive earthquakes by being flexible. Designed to cope with strong wind, they have no trouble with moderate ground shaking. People on the upper floors may experience a wild ride, but the structures barely register the motion in their bones. Megaquakes are more dangerous beasts. A magnitude 9 quake doesn’t shake the ground that much harder than a magnitude 7, but it shakes a lot longer. In a thirty-second quake, a tall building might whip back and forth ten or fifteen times. In a four-minute quake, the building can gyrate through sixty cycles. Like a paperclip, even steel can bend only so many times before it breaks.
The potential for damage is magnified because the slow, rolling ground motions that dominate in a subduction zone quake can trigger an effect called resonance in tall buildings. The British Army experienced the consequences of resonance in 1831, when a suspension bridge near Manchester collapsed under a detachment of soldiers marching four abreast. The synchronized clomping of 148 boots was perfectly in tune with the structure’s natural frequency—the point at which it vibrates like a tuning fork. As they felt the bridge begin to sway, the soldiers stomped their feet harder to amplify the motion. It was hilarious until they found themselves tumbling into the river.
The same thing can happen in an earthquake when seismic waves synchronize with a building’s natural frequency. It’s as if the building is a swing and the earthquake delivers perfectly timed shoves that send it flying farther and farther. A 1985 subduction zone quake off the coast of Mexico surprised engineers by knocking down new high-rises and leaving older, shorter buildings untouched.
Structural engineers try to plan for resonance effects. But the degree to which the basins under Seattle and other cities will amplify the shaking is not factored into building codes, nor is the fact that subduction zone quakes will rock the ground for minutes, not seconds. “Almost none of the buildings in Seattle were designed to withstand three to five minutes of shaking,” Yanev said.
Most tall buildings in Japan and Chile weathered recent megaquakes, which is encouraging. But neither country has as many questionable structures as the Northwest, Heaton pointed out. Engineers in Chile and Japan have been designing for great quakes much longer than their American counterparts. Frequent quakes have culled out the weak structure. Chilean laws hold a building’s original owner liable for ten years if code violations lead to earthquake damage. The premium on seismic robustness is so high in Japan that owners voluntarily equip their buildings with dampers and isolation bearings. Japanese high-rises also contain more steel, and the codes require sturdier connections. Given a choice, Heaton said he’d much rather ride out a big earthquake in a Japanese building than an American one.
High-rise construction in the Northwest isn’t going to stop just because geologists are uncertain what will happen in a megaquake. “We’re not going to live in tents,” Heaton said. “At the end of the day, we need buildings and we would like those buildings to be well engineered.” Given all the uncertainty, Heaton said it behooves engineers to be humble and skeptical: humble about Cascadia’s power and skeptical that the USGS or anyone else knows what to expect.
Like Heaton, John Hooper has devoted much of his career to building a bridge between engineers and earth scientists, but from the other side of the gap. His biggest challenge is to drag science into the real world.
As director of earthquake engineering at Magnusson Klemencic Associates, one of the country’s premier structural engineering firms, Hooper had a hand in many of the region’s architectural landmarks, including Columbia Center—the tallest skyscraper—and the stadiums where the Seattle Mariners and Seahawks play. Designing buildings to handle earthquakes requires a lot of number crunching, but it’s not too technically difficult, he said. “If you know the amount of shaking, it’s relatively straightforward.” Hooper works closely with geologists in the Northwest, so he knows more than most engineers about how the USGS makes the sausage. At the 2012 hazard map workshop in Seattle, he quipped that pure science doesn’t seem so pure to him anymore. The geologists laughed. Then Hooper made his pitch for the status quo. “If we’re just tweaking something for the sake of change, we don’t want that. We’d like to minimize the changes.”
Hooper vividly recalls the day Brian Atwater, Craig Weaver, and a team of USGS scientists first met with him and other engineers to brief them on something called the Cascadia Subduction Zone. Another memorable meeting followed a few years later, when the geologists sprang the news about the Seattle Fault. Hooper used to think he had it easy compared to engineers in California. Now, he envies them. “It’s much more complicated here,” he said.
With his rapid-fire delivery, Hooper could pass for a New Yorker. He grew up in Everett, though, returning to his home state after graduate school at UC Berkeley. Hooper serves on several of the national committees that develop new building codes and set seismic standards. Since he started in the business more than twenty-five years ago, the changes have been dramatic. “We’re constantly, I mean constantly, looking at what we do. We’ve come from A to Z, but I think there are four more alphabets to go.”
Structural engineers take architects’ visions and flesh them out to create designs that will stand up to gravity, wind, and seismic forces. Some engineers in the Northwest factor basin effects and duration of shaking into their designs for skyscrapers, even though it’s not required, Hooper said. For really big projects, engineers submit their work to other firms for peer review. Significant structures are subjected to simulated earthquakes when they’re still on the drawing board. But for nine out of ten new buildings, engineers don’t consider what could happen if shaking exceeds the level in the code, derived from the USGS maps. “It’s assumed that a building designed to the code will result in acceptable performance,” Hooper said.
He’s confident few, if any, modern high-rises will collapse in a megaquake. But like a scientist, he never says never. “There’s a chance a brand-new building could come down. It’s a very low likelihood.” Modern building codes are couched in probabilities, not absolutes. The goal is “life-safety:” protecting people from injury or death by preventing collapse. A building that’s so damaged it has to be torn down is considered a success as long as all the occupants make it out safely.
“The term ‘earthquake-proof’ is not in our lexicon,” Hooper said. A well-designed building that meets all requirements still stands as much as a 10 percent chance of collapse if it’s hit by the maximum earthquake the code considers, roughly a two-thousand-year quake in the Northwest. Slam the same building with something bigger and the risk shoots up. In an off-the-charts event—one that shakes the ground more than twice as hard as the code anticipates—statistics suggest a 50-50 chance of collapse. “That’s for a five-thousand- or ten-thousand-year event,” Hooper said. “The kind of thing that’s just too large to consider.”
For older buildings the prospects are worse. But just because a building is old doesn’t neces
sarily mean it’s dangerous. So much hinges on design, era, and materials. Long before seismic standards were cranked up, tall buildings in Seattle were built to cope with strong winds and moderate quakes. The Structural Engineers Association of Washington (SEAW) was formed after the 1949 Olympia earthquake and helped institute the region’s first seismic codes. With fewer earthquakes Oregon didn’t adopt significant seismic standards until 1994.
Cities all along the West Coast share the problem of older steel high-rises, which can be especially vulnerable to quake damage. Most steel-frame office towers and apartments erected between the 1960s and the mid-1990s have flanges fastened with welds rather than with bolts. Welds were cheaper, and lab tests at the time suggested they were stronger.
But after California’s 1994 Northridge quake, engineers were shocked to discover fractured welds in more than one hundred buildings, including some still under construction. When they tested the welds in more detail, it became clear they were dangerously brittle. Building codes were changed. But retrofits are costly, and few existing structures have been upgraded. When Heaton and his student simulated the impact of Cascadia quakes on hypothetical high-rises in Seattle, those with brittle welds were more prone to collapse in every scenario.
Hooper works in the real thing. The 514-foot-tall Rainier Tower in downtown Seattle was built in 1977 with a welded steel frame. Perched atop a twelve-story pedestal that looks like a giant golf tee, the skyscraper has always drawn nervous looks from locals. Some call it the wine glass. The Seattle Post-Intelligencer’s architecture critic described the tower as both an engineering tour de force and “a teetering colossus that makes any normal human jittery.” Hooper’s firm did the engineering. He doesn’t feel any qualms about reporting for work on the 32nd floor. In a quake the building will sway, but that’s what it was designed to do. Hooper was part of a team that investigated brittle welds in California, and he pointed out that no steel buildings collapsed in the twenty-second-long Northridge quake. A building like the Rainier Tower would have to lose more than half its welds to cause major damage, Hooper said. With the amount of steel and number of columns that support the skyscraper, he’s betting that’s not very likely to happen.
Just as they learned from Northridge, engineers are schooled by every major quake. The SEAW and other groups dispatch teams around the world to study damage and glean lessons that can be applied at home. Codes aren’t perfect but they’re a lot better than they used to be, said Hooper.
But in order to design buildings for earthquakes, he and other engineers need a starting point. For all their uncertainty, the USGS maps are the only game in town. “We engineers are scratching our heads, asking ourselves, ‘Should we believe everything the scientists tell us?’ But at some point you’ve got to believe your smart kid brother.”
When Hooper first heard about Cascadia, he was skeptical. The evidence won him over. Despite objections from many in the construction industry, he championed code changes to account for this new seismic threat. Hooper continues to put his trust in science. “There’s a handshake and a handoff. But it’s not just a blind handoff, like in track and field. We actually look each other in the eye.”
Heaton hopes he’s wrong, but he can imagine a day when the relationship turns sour. If many new buildings fail when Cascadia lets loose, engineers will point their fingers at geologists and say, “We designed this based on what you told us, and look what happened.” Geologists will reply that they were always up-front about what they didn’t know. “We’ll probably end up in congressional hearings some day,” Heaton said. “People will be asking, ‘How did it turn out like this?’ ”
The maxim that new buildings are better took a beating in Chile’s magnitude 8.8 megaquake in 2010. Some of the most heavily damaged high-rises were only a few years old. A fifteen-story condominium that toppled onto its side, killing eight people, wasn’t even fully occupied yet. In one city close to the epicenter, nearly one in five tall buildings—many of them new—were damaged beyond repair.
Engineers say the surprising failures raise red flags for the U.S. West Coast. “This earthquake caused severe damage to many buildings that are much better than the typical building in Seattle,” said Yanev.
Before he sold his company in 2000, Yanev led or directed field investigations of more than ninety earthquakes around the world. Since then, he has traveled the globe as an adviser to the World Bank. In Chile Yanev trekked from leaning apartment towers to cracked hotels and sagging office complexes. Chilean codes generally require stronger construction than do codes in the United Sates. “If our buildings were in Chile, they would be all over the ground,” he said.
But in Chile as in the United States, architects and builders have been pushing the boundaries with fewer and thinner shear walls, the structures that protect buildings from damaging side-to-side motion. “Ten years ago most of the big construction companies in Chile were run by engineers who understood the risks. Now they’re run by financial people, MBAs who just want to know how much they can shave off the cost,” Yanev said. Only a handful of buildings in Chile collapsed, but enough were seriously damaged to leave tens of thousands of people homeless and to throw thousands of businesses into chaos.
Yanev said that the design shift in Chile is similar to what’s been going on in California and the Pacific Northwest. Steel construction, considered the most earthquake resistant, is increasingly rare. The typical new high-rise is a slender concrete tower with panoramic windows and few bulky columns to interfere with the views. Earthquake resistance is provided by a heavy central core that contains the elevator shaft. “All the strength is in those four walls.” Yanev said. If an earthquake shakes the building hard enough and long enough to damage the core, there’s little backup to prevent collapse.
“These concrete buildings scare the living daylights out of me and a lot of other engineers.” Many of the new towers are residential. High-ceilinged shops, restaurants, and lobbies on the ground floor, what engineers call a soft story, add to the risk.
U.S. codes used to require more redundancy. Buildings were designed with strong frames and shear walls. If one failed, the other could pick up the load. Now engineers rely on computer simulations to whittle away at the margins and design structures that just meet the minimum standards. “The overreliance on these computer analyses is just crazy,” said California-based engineer Kit Miyamoto. “People do these simulations and feel like they know everything.”
With five offices in California and one in Oregon, Miyamoto International works on projects around the world. The company designed a 1.2-million-pound seismic damper to protect the Los Angeles airport’s famous Theme Building without spoiling its futuristic lines. In Haiti Miyamoto and his team are designing low-cost retrofits for schools and houses damaged in the disastrous 2010 quake. He’s on a crusade in the Northwest and in California to convince developers and engineers that it makes good business sense to go beyond the bare-bones seismic code.
To make his case, Miyamoto points to Christchurch, New Zealand. As in Chile few buildings collapsed in the magnitude 6.3 quake that struck in 2011. One that did was the six-story concrete Canterbury Television Building, where more than one hundred people died. Most offices and apartments performed to code, staying upright long enough for occupants to flee. Nevertheless, 80 percent of buildings across a square mile of the city’s core were slated for demolition because of damage. Economic losses totaled more than $15 billion.
Superimpose that square mile on downtown Seattle and almost everything from the waterfront to the top of Capitol Hill would be gone. “Looking at Christchurch, you can see what’s going to happen to Seattle,” Miyamoto said. “I wouldn’t be surprised if the city is shut down for more than twelve months.” The same could happen in Portland and Vancouver.
Christchurch has top-notch engineers. New Zealand has strict buildings codes and inspectors to enforce them. But just as in the U.S., the codes don’t consider the economic impact to communities, business
es, and residents when buildings are rendered unusable. In Chile and Japan, many buildings that rode out the shaking with little structural damage were a shambles inside, with water lines and power supplies severed, ventilation systems wrecked, and expensive finishes shattered. Japanese shelters were filled with high-rise refugees forced to move out of their apartments. Nearly two million people were left homeless by Chile’s quake and tsunami, many of them urban condo dwellers. “There’s a huge gap between what society expects and what is really practiced by engineers,” Miyamoto said. “We need to do better. It’s not like we don’t know how.”
Designing buildings that are seismically sustainable—able to remain standing and usable after a quake—adds only 5 to 10 percent to the total cost. But when engineering firms compete for big jobs, fortunes rise and fall on margins as slim as 1 percent. High-tech businesses like Microsoft and Amazon are among the minority willing to pay extra to ensure their operations won’t be disrupted for weeks or months following a big quake. Most new buildings meet the code and nothing more.
Modern building codes may not always be able to achieve even their modest life-safety goals. “It’s not like we’ve got it all solved,” said University of British Columbia engineering professor Perry Adebar. Adebar visited Chile and came back concerned about some of the tall buildings in British Columbia’s biggest city. The Vancouver metro area has more high-rises than Seattle and Portland combined. Many sit on a river delta and other loose soils guaranteed to amplify shaking.
Adebar noticed that the damaged buildings in Chile all seemed to have six-inch concrete shear walls between rooms, instead of the eight-inch walls that used to be standard. Developers pushed for the switch to cut costs, and building regulators allowed it.
Adebar knew some Canadian buildings have the thinner walls, too. When he got home, he decided to run a series of lab tests to see just how much shaking six-inch walls could take. He used a hydraulic press to squeeze wall sections up to six feet tall, and found they shattered more easily and much more suddenly than expected.
Full-Rip 9.0: The Next Big Earthquake in the Pacific Northwest Page 21