Full-Rip 9.0: The Next Big Earthquake in the Pacific Northwest

by Sandi Doughton

  Art Frankel, who is still active in the USGS mapping program after leading it for more than ten years, is well aware of the uncertainties. He ties himself in knots trying to minimize them. But there’s also a lot that scientists do know about earthquakes, he pointed out. The maps are the best vehicle anyone has come up with to pull those insights together in a way that can help people prepare. “It allows you to make intelligent decisions,” Frankel said.

  Seismic hazard maps were the fallback for the USGS after earthquake prediction fizzled. Even though it wasn’t possible to pinpoint when and where a quake would hit, it was clear that some areas were at higher risk than others. The earliest hazard maps essentially drew circles around places that had been hit by big quakes in the past. The next generation of maps tried to identify the biggest possible earthquake that could strike an area and estimate how hard it would shake the ground. This “maximum credible earthquake” approach is still used for critical facilities, like dams and nuclear power plants.

  But planning for the worst case is costly. If the Seattle Fault unleashes a giant quake every five thousand years, should office buildings and apartment towers that might last a century be constructed like fortresses, just in case? The probabilistic approach the USGS uses was developed to balance cost and benefits.

  Frankel and his colleagues feed earthquake scenarios into their computer models, weighting each one based on likelihood. It’s kind of like loading the dice to reflect the fact that Mother Nature more often rolls a moderate quake than a monster. In the next fifty years, the scientists estimate it’s almost a slam dunk that another deep, Nisqually-style earthquake will hit. A Cascadia megaquake gets about 14 percent odds without factoring in the possibility of more frequent quakes on the southern Oregon coast. A medium-size quake on the Seattle Fault scores a 5 percent probability.

  The most commonly used maps account for the lineup of quakes expected to strike in either a 500- or 2,500-year period. The 2,500-year map for the Northwest includes a magnitude 9 Cascadia megaquake and a magnitude 6-plus shallow fault quake. But the USGS considers a massive Seattle Fault quake like the one that struck the region in 900 AD to be a 5,000-year quake—such a long shot that it gets scant consideration.

  And therein lies the Achilles’ heel of probabilistic mapping: It discounts the rarest quakes, which are also the most deadly. When calamities do strike, scientists dismiss them as “black swans,” statistical outliers too far-fetched to consider. But in 2010 Russian scientists compared the ground shaking forecast by hazard maps to levels actually recorded during big quakes around the world. They found that the 500-year maps underestimated the intensity half the time, often by more than a factor of two. “Our crystal ball is proving mostly cloudy around the world,” said Seth Stein, a geophysicist at Northwestern University.

  Stein argues that the maps are just as likely to overestimate the danger. Business leaders in Memphis cursed a blue streak when the latest USGS maps ranked the earthquake hazard in the Midwest on a par with California. And in parts of California, the hazard calculations yield such high numbers that the state caps them so as not to shut down all construction.

  The USGS maps for the Pacific Northwest and the rest of the country have yet to be tested by a really big quake. But so far they’ve done a good job of forecasting where small quakes will hit. The magnitude 5.8 quake that cracked the Washington Monument in August 2011 may have stunned folks inside the beltway, but it occurred in a seismic hazard area that’s been marked on the national map for years.

  The maps’ shaking forecasts are most accurate in places like California, where quakes are common and there’s a lot of data on how seismic waves move through the ground, Heaton said. That’s one of the main reasons he doesn’t have much confidence in the maps for the Pacific Northwest, where no one knows how the shaking from a megathrust off the coast will ripple across the region. “When there’s another big Cascadia earthquake, nobody has any idea what’s really going to happen.”

  Building a map requires one choice after another, with uncertainty at every turn. How often does the Seattle Fault break? What’s the biggest possible Cascadia quake? How much will the Seattle basin amplify the shaking?

  Frankel and his USGS colleagues try to make conservative choices, within reason. Unlike in Japan where mappers underestimated the subduction zone that snapped in 2011, the Northwest maps allow for magnitudes as high as 9.2 on Cascadia. The scientists add in a background level of hazard to account for undiscovered faults. “Don’t think you’ve seen everything that nature can throw at us,” Frankel cautioned the participants at the Seattle workshop.

  When he took over the USGS mapping program in 1993, it was an insular operation. Frankel’s predecessor scoffed at the notion that the Cascadia Subduction Zone might be active, so he left it off the maps. Decisions that affected businesses and lives across the country were made by a tiny group of experts. One of the first things Frankel did was open the doors to engineers and other scientists, who were starting to wonder what went on behind the curtain. “They basically wanted to be assured that we didn’t go in a back room and smoke cigars and draw lines on a map,” Frankel recalled.

  The Seattle meeting was one of several Frankel and his colleagues convened around the country to gather input on a new set of maps to be issued in 2014. With several recent discoveries to consider, the schedule for the two-day meeting was as packed as the room. Item number one was a question with major implications: how often does the Cascadia Subduction Zone snap?

  Oregon State University scientist Chris Goldfinger kicked things off with pictures of the cores he extracts from the seafloor. Goldfinger explained how he distinguishes layers formed by earthquake-triggered landslides. Then he walked the group through his evidence that cores from the southern coast of Oregon contain double the number of earthquake layers as cores from elsewhere. The scientists listened politely as he laid out his interpretation. In addition to serving up a full-rip 9 every 500 years or so, Goldfinger said, Cascadia also produces smaller quakes that rupture only the southern part of the subduction zone. So on average, the region gets a hard pounding every 250 years.

  After Goldfinger wrapped up his presentation, Frankel took the podium again. Factoring the additional quakes into the maps would raise the hazard significantly, he explained. Did the other scientists think the evidence was strong enough to justify the change? “Now we get into the making of the sausage,” Frankel said. “Let’s get the discussion started.”

  A few years earlier, when Goldfinger first suggested southern Cascadia rips more frequently, he got a chilly reception. A lot of scientists weren’t convinced all the layers in his cores were caused by megaquakes. At the Seattle meeting, though, more heads were nodding than shaking. Goldfinger had more data and some of his “extra” quakes seemed to be corroborated by tsunami layers in coastal lakes. Still, Frankel was reluctant to wholly embrace the story. Changing the seismic hazard maps has economic consequences, he explained after the meeting, so it’s important to vet new evidence carefully. Frankel was leaning toward a compromise that would factor in only those southern quakes corroborated by tsunami layers.

  A lot of hazard mapping comes down to choices like that, which drives the critics crazy. “It’s completely guessing in the sense that you can come up with a range of options and you could say they all represent a reasonable interpretation of the data,” Stein said. “Which choice you claim is best is based purely on your gut feeling.”

  Those who aren’t in on the map-making process—including the public—don’t realize how many assumptions are folded into the process, Stein said. He likens hazard mapping to credit default swaps and the other convoluted financial deals Wall Street cooked up before the 2008 financial meltdown. Few people on the outside understood the risks or even how the deals were structured. “In hindsight, the problem is that present hazard mapping practices were accepted and adopted without rigorous testing to show how well they actually worked,” Stein wrote in the journal Tectono
physics.

  Max Wyss, of the World Agency of Planetary Monitoring and Earthquake Risk Reduction based in Geneva, is among those calling for a return to worst-case scenario planning. “It’s a moral question of who do we want to protect,” he said. If countries design dams and other critical facilities to stand up to the maximum credible earthquake, why not apply the same standard to the places where people live?

  But defining the worst-case earthquake is also fraught with uncertainty. Even if it could be done, it’s unlikely societies would accept the cost. What’s next, builders might ask? Bracing for an asteroid strike? “It’s a very hard issue,” Wyss conceded.

  If nothing else, he said, the public should have an opportunity to weigh in as it does on air pollution rules, siting of nuclear power plants, and other decisions that affect the collective welfare. Popular opinion can’t wish away the threat of earthquakes. But the people of the Northwest have never been asked if they’re comfortable with an approach that seeks the sweet spot between cost and benefit, or whether they would prefer to pay extra for an extra margin of safety. “Ultimately, this is going to affect everybody in the Pacific Northwest,” Heaton said.

  The USGS’s job is to do the science. It’s up to society to make the call about acceptable risk, Frankel pointed out. But few people are beating down the door to join the conversation. Technically, the USGS workshops are open to everyone, but they’re not a genuine forum for the public. Anyone without a PhD would be baffled by discussions about aleatory variability, Gaussian smoothing, and b-values.

  The public can weigh in on building codes. But it’s been years since the United States was rocked by a big earthquake, and only a few private citizens are motivated to tackle such a dense topic. James Bela is one of them.

  The founder of a one-man organization called Oregon Earthquake Awareness, Bela was the sole public representative at the Seattle workshop. He made the drive from his Portland home in a 1968 Dodge Dart with more than half a million miles on the odometer. This time the car didn’t die as it had a few months earlier, stranding him at the border after a session on earthquakes in British Columbia. At his own expense, Bela travels to science meetings and building code hearings across the country. At the Seattle workshop, he taped up a homemade poster listing the death toll from major quakes around the world—deaths he believes could have been averted if countries had just prepared for the worst-case quake.

  Bela stood alone in the back of the room, videotaping the proceedings and occasionally asking questions or making comments. He urged the USGS not to water down the threat of more frequent quakes on the Oregon coast. He reminded the scientists of the lessons from Japan’s 2011 earthquake and other recent seismic surprises. “I think we ought to look at the maximum event to protect public safety,” he said.

  A former geologist for the state of Oregon, Bela isn’t intimidated by the scientific jargon. But that didn’t change the fact that nobody took him seriously. Every time he walked to the microphone, there was a sense of eye rolling in the room. What Bela wanted wasn’t on the agenda.

  CHAPTER 11:

  SHAKE, RATTLE-PORTLAND, VANCOUVER, AND SEATTLE

  TO APPRECIATE HOW MIRED IN DENIAL many Northwesterners remain about big earthquakes, consider the reaction to a Washington State Department of Transportation (WSDOT) video that simulates the impact on Seattle’s waterfront. The animation opens with a panoramic shot across Elliott Bay and zooms in as the initial jolt hits and cars spin slowly out of control.

  The Alaskan Way Viaduct, a double-decker monstrosity built in the 1950s, starts to lurch. Thirty-five seconds into the quake, sections of the concrete roadway come crashing down as supporting columns snap and the ground turns to goo. The city’s seawall, nearly a century old, slumps into Puget Sound, taking piers, restaurants, and curio shops with it. (Ivar’s on Pier 54 remains standing. Apparently, even the WSDOT scrooges couldn’t bear to trash the venerable fish house.)

  Fires flare from broken gas lines. Half the city goes dark as power poles come crashing down. The simulated quake is like the 2001 Nisqually quake, but bigger—magnitude 7 instead of magnitude 6.8. Its epicenter is ten miles closer to Seattle and the imaginary shaking goes on twenty seconds longer.

  WSDOT kept the video under wraps for two years, releasing it only when forced by a public records request. The bureaucrats feared it was too sensational. A lot of people agreed. One letter to the editor dismissed the animation as Disney-caliber fantasy. Others called it a scare tactic. “Perhaps the department wants to bolster a request for funding,” one writer sniffed.

  Geologists and engineers just shook their heads. They knew the video presented a plausible glimpse of the future, and a sanitized one at that. There’s no audio, so viewers don’t hear the roar of collapsing concrete, screeching tires, or screams. As the camera pans out for the closing shot, the only visibly damaged building is an office block with broken windows. The skyline is untouched. Compared to what the city would look like after a direct hit from the Seattle Fault or a Cascadia megaquake, WSDOT’s video really is Disney-esque. “We’re talking about a different ballgame,” said Peter Yanev, the outspoken cofounder of EQE, a leading earthquake engineering firm. “It’s going to be bad.”

  A realistic simulation of a powerful quake would show historic buildings collapsing into piles of bricks. No surprise there. Engineers have been warning for decades about the dangers of unreinforced masonry. But people might be shocked to see multistory concrete buildings pancake as shaking severs the flimsy supports that were common until the early 1970s. The tide of liquefied soil that caused the seawall to slump in the WSDOT animation would also undermine the footings of port terminals and industrial tank farms, knocking cranes askew and spilling fuel. An aerial view across the surrounding suburbs would show flattened warehouses and industrial parks, collapsed bridges, and heaved pavement.

  But what about the skyline? Will the skyscrapers that dominate the urban core of Seattle—or Portland, or Vancouver—remain upright? The fate of the region’s tall buildings has the potential to eclipse all other threats to the built environment. “If a couple of high-rises fall down in Seattle, you could be talking about two or three thousand people,” Yanev said. He thinks it’s possible, and suggested as much in a 2010 New York Times op-ed headlined “Shake, Rattle, Seattle.” Many local engineers dismiss Yanev, who is based in California, as an alarmist. But he’s not the only one who argues that buildings across the Pacific Northwest could be more vulnerable than most people realize.

  To the USGS geologists who prepare the region’s seismic hazard maps, the biggest unanswered question is what a Cascadia megaquake will do to tall buildings. But most engineers insist there’s no question at all. They’re confident modern skyscrapers will do what they were designed to do and roll with the punches. The disconnect is almost as pronounced as the one thirty years ago between the nuclear consultants who insisted the Cascadia Subduction Zone was dead and the geologists who suspected it was alive and poised to kick hard. Tom Heaton was in the middle of the fray then, so there’s a kind of symmetry to finding him back at it again.

  In the 1980s Heaton was the young hotshot who swept aside the nuclear industry’s assertions and zeroed in on the subduction zone’s essential nature, finding no reason to dismiss it as harmless. Nearly thirty years later, he’s pushing Medicare age, and his wavy hair has faded from blond to white. He still writes music and plays guitar, performing with a Christian band called Leap of Faith. Professionally, he hasn’t mellowed a whit. He continues to rankle the establishment, focusing much of his attention these days on the building industry.

  When engineers say high-rises in the Northwest are designed for a magnitude 9 quake, Heaton snorts. “As far as I’m concerned, we don’t know whether or not tall buildings in Seattle will survive a megaquake,” he said. “For anyone to say otherwise is deceptive.”

  Of roughly 1,300 high-rise buildings in Seattle, Vancouver, and Portland, nearly half were built before 1990. Codes for most urba
n areas weren’t upgraded to account for Cascadia until the mid-1990s. But Heaton isn’t convinced even the latest codes adequately consider the unique nature of subduction zone quakes, which rattle tall structures the hardest. No skyscraper in the United States has been through that roller-coaster ride yet. “It’s going to be a big experiment, and I’m not sure I want to be there for it,” he said.

  When Heaton left the USGS almost twenty years ago to take a professorship at Caltech, his mission was to bridge the gap between geologists and engineers. It hasn’t gone as well as he had hoped. “I discovered there are no bridges that long. Earth scientists and earthquake engineers are so different I think they must have been born that way.”

  Engineers crave certainty. Geologists get bored if their work doesn’t have an element of mystery. Discovery is messy, full of doubt, debate, and about-faces. Science couldn’t progress any other way. If many ideas turn out to be wrong, that’s just part of the process. Engineers operate in a higher-stakes world. If they get things wrong, people can die and businesses can be ruined. So when engineers ask geologists what kind of seismic shaking to expect, the last thing they want to hear is, “I don’t know.”

  But Heaton says that’s the only honest answer to the question of what will happen when Cascadia quake meets urban core. Of all the seismic hazard maps the USGS prepares, none is more fraught with uncertainty than the forecast for shaking in Seattle during a subduction zone quake. The basin that sits under the city and its suburbs will undoubtedly amplify the shaking, but by how much? Barely? By a factor of two or more? Nobody knows the answer. The other major unknown is how close the quake will come.