Goldfinger delivered his presentation to the crew then stumbled off to bed, feeling like an idiot. But his funk didn’t last long. There was nothing to do but press on and follow the data where it led. Goldfinger was more experienced at shaking off mistakes than most beginning scientists. It was a lesson he had learned building boats for eight years after college. He would fabricate railings, metal plates for rigging or engine mounts, and inevitably have to do it all over again. Nothing on a boat ever fits the first time, he explained. “It’s the same in science. Every idea you think is great—the next day it might turn out to be nonsense.”
That willingness to change gears is a hallmark of Goldfinger’s personality. He was nine years old when he set out to become the first sailor in his family, inspired by a California teenager’s round-the-world voyage. He agreed to paint the family home in Palo Alto if his dad would front him $75 to buy a secondhand Sunfish sailboat.
Goldfinger’s father was a radio engineer for the Apollo missions, and the son dreamed of being an astronaut. He reconsidered when he found out fighter pilot training was a prerequisite. The Vietnam War was raging, and Goldfinger couldn’t stomach the idea of shooting at people. Geology captured his attention in high school, thanks to a teacher who ran field trips in his spare time. “The biology students were picking apart frogs, but the geology students were loading camping gear into trucks,” Goldfinger recalled. “I said, ‘I’m going with them.’ ”
Having a good time ranks high on Goldfinger’s list. He wasn’t ready to be what he called a “nose-to-the-grindstone working unit” when he graduated from Humboldt State University in Northern California. So he took the boat-building job, the most appealing aspect of which was delivering yachts around the Pacific for rich people. It was only the prospect of even sweeter boondoggles that lured him back to science.
On a run to Fiji Goldfinger met a geologist with the best gig on the planet: surveying island beaches for minerals. “I said, ‘That’s the kind of job I want.’ ” The irony of winding up in the soggy Northwest still makes him wince. But with his mediocre grades and long hiatus from school, Goldfinger barely made the cut for Oregon State’s graduate program. Bob Yeats, then the chairman of the Geology Department, sensed promise. “There was just something about the guy,” Yeats recalled. “He marches to his own drummer.” Yeats still considers it one of his luckiest calls. “I’ve learned so much more from him than he ever learned from me.”
Like Yeats, Goldfinger was an early skeptic about Cascadia’s threat. His views put him at odds with Atwater, and the two scientists have been sparring for much of their professional careers.
During his dissertation defense, Goldfinger ran down a Letter-man-style list of the top ten reasons the subduction zone couldn’t produce a magnitude 9 quake. At scientific meetings Goldfinger was the guy who kept questioning Atwater’s logic while other researchers rolled their eyes. But Atwater’s case kept getting stronger and stronger. “And I didn’t have any evidence,” Goldfinger recalled cheerfully. “That settled it.”
Goldfinger’s 1999 cruise turned him into a true believer. After he and Nelson got back to the lab at OSU they reviewed all their cores and agreed that simultaneous landslides were the best explanation for the layers they saw. “Big earthquakes were the only thing that could cause that.”
Since 1999 Goldfinger has mounted another four coring cruises. He has also led or participated in thirty-five other seagoing research trips. But the work has landed him at the receiving end of skepticism. Atwater and others argue that he’s reading too much into the mud. The debate is more than academic, particularly for the millions of people who live along the coasts of Southern Oregon and Northern California. Goldfinger’s evidence suggests that stretch of the subduction zone gets shaken nearly twice as often as the northern stretch.
Rain was sheeting as Goldfinger ducked into his lab on the OSU campus on a fall morning in 2011. When the Northwest weather turns foul, he’s always on the lookout for scientific conferences in warm places. One winter he slipped off to Hawaii in his sailboat, answering work e-mails as if he were down the hall. Goldfinger may be a working unit these days, but his nose isn’t always to the grindstone.
Shaking the water off his jacket, he popped open the door of a walk-in cooler the size of a classroom. Racks that reached almost to the ceiling cradled hundreds of cores he and other OSU scientists collected over the years. Griggs’s originals were there, though too shriveled to be of much use. Nestled in white plastic casings, the cores were about four inches wide and split down the middle like hot dog buns. Goldfinger slid out one of the tubes and carried it into the adjoining lab.
Under the fluorescent lights, the four-foot-long half cylinder of mud was as impressive as, well, mud. “It doesn’t really look like much, does it?” Goldfinger said. He pointed to a layer at the top of the core. “That’s the 1700 quake.” The band of sediment was about eight inches thick and a little darker than the rest of the mud. “From a magnitude 9 quake, you’d expect something as tall as a house with boulders at the bottom, but that’s not what you see.”
The segment of core held three more dark layers, each representing ancient earthquakes. Goldfinger dates the layers by carbon 14 analysis of the shells of tiny marine creatures. “It’s like a telescope looking backward,” he said.
It took him a long time to learn how to interpret the messages in the muddy bands. After he did he realized that the turbidite layers revealed something the record on land only hinted at: not all Cascadia’s quakes were created equal. “It’s not rocket science, but we think a bigger earthquake makes a bigger turbidite,” Goldfinger said, fetching another tube from the cold room. He pointed out a layer nearly a foot and a half thick. In cores from the tip of Vancouver Island to Northern California, that layer—number 11 in the sequence—is always huge. “It’s our biggest event,” Goldfinger said.
What might have been the Godzilla of Cascadia megaquakes struck about 6,000 years ago. Goldfinger estimated its size at magnitude 9.1. That may not sound much bigger than a magnitude 9, but a tenth of a point increase on the logarithmic scale represents more than a 40 percent increase in energy. Goldfinger’s cores also hold evidence of some quakes that appear to be smaller than magnitude 9, and at least two others that were clearly bigger. “What we’re seeing is that the 1700 event was only average, which is not good news.”
Geologists accustomed to working on land were initially skeptical of the seafloor evidence. But it takes only a glance at the underwater topography off the Northwest coast to grasp the basics, Goldfinger explained, flipping open his laptop and punching up a three-dimensional view.
From the water’s edge, the continental shelf extends out in a gently sloping plain wider than the Oklahoma panhandle. About fifty miles offshore, the ground falls off steeply in a rumpled mass of canyons whose feet stand at the edge of the abyssal plain. These are the cliffs, some nearly ten thousand feet high, that Goldfinger surveyed during his Alvin dive. Drain the water and the view would be dizzying to a person standing on the ocean floor gazing up at the palisade. “It would make most of the Cascades look puny,” Goldfinger said.
Currents continually sweep sediment from the Columbia and other rivers to the edge of the continental shelf. During the quiet centuries between quakes, this mass of terrestrial detritus piles up along the canyon heads like cornices of snow. When the subduction zone rips and the ground convulses, the piles collapse.
The sand and pebbles funnel down canyons with names like Rogue, Trinidad, and Quinault, gathering speed and force until they reach the stage turbidite experts call “ignition.” That’s when the slides transform into something akin to the pyroclastic flows that sweep down volcanoes and obliterate everything in their paths. “You’ve got this sandy, silty mass riding on a water cushion, and it can go for hundreds of kilometers,” Goldfinger said.
The biggest Cascadia landslides spill out on the abyssal plain. That’s where Goldfinger found the best earthquake record, unmuddled by
smaller, storm-triggered slides closer to land. By 2012 he and his team had evidence of nineteen quakes that ruptured the entire Cascadia margin in the past 10,000 years. There’s not much controversy about those anymore. The layers turn up consistently along the coast, and their dates match up with land-based records.
But Goldfinger also reports another twenty-three thinner layers in cores collected off Southern Oregon and Northern California. Some of the wisps are so faint even he didn’t notice them at first. But when he looked closer with CT scans and instruments that measure magnetic properties, the bands jumped out. Goldfinger is convinced they represent smaller quakes—maybe magnitude 8—that rip only the southern half of the subduction zone and strike in between the full-rip 9 monsters. That would translate into a major quake every 250 years on average.
Seafloor cores show evidence of nineteen full rip 9 Cascadia quakes in the past 10,000 years, and twenty-three smaller quakes on the southern portion of the subduction zone (twenty-two shown in this diagram). (image credits 4.1)
It’s not unusual for subduction zones to rupture in segments. Indonesia’s killer quake and tsunami in 2004 unzipped only the northern end of the Sumatran subduction zone. A quake that followed three months later broke an adjacent segment. But many scientists, including Atwater, question Goldfinger’s interpretation. It’s possible some of the thin layers were from storm-triggered landslides, or slides that simply sloughed off underwater slopes.
Geologists working in coastal lakes in Oregon found corroborating evidence for some of Goldfinger’s smaller quakes in the form of sand layers flung inland by tsunamis at roughly the same time. The search is on for more. In 2011 Goldfinger and his students started pulling cores from lakes in Oregon’s Coast Range. They found a record there of landslides that closely mirrors the marine layers.
If the southern part of the subduction zone has been hammered as frequently as Goldfinger believes, it changes the risk equation for the entire region. A magnitude 8 quake anywhere on the coast will have far-reaching effects. Based on the standard view that Cascadia uncorks every five hundred years on average, there’s a 10 to 15 percent chance the region will get clobbered in the next five decades. Goldfinger’s interpretation raises the odds to 37 percent.
“You say 10 or 15 percent, and to most people that’s kind of like the chance of getting hit by an asteroid,” Goldfinger said. “But when you say 37 percent, that starts to sound like a real number.”
The Pacific Storm was “mowing the lawn,” motoring back and forth off the central Oregon coast while instruments scanned the ocean bottom. It was August 2011, at the tail end of a ten-week cruise. One of the objectives was to refine the seafloor maps that are so crucial to Goldfinger’s work.
A geologist without a topo map is like a cell biologist without a microscope. Finding the right place to dig and core, to sample and scan, means the difference between discovery and a pile of dirt. As if underwater geology weren’t hard enough, the field was long hampered by military secrecy. Goldfinger’s group has devoted almost as much time to mapping as it has to coring. That’s why he was onboard the Pacific Storm, an eighty-four-foot converted crabber. The federal government had recently lifted some of the last restrictions on the Northwest coast, and Goldfinger was eager to complete as many surveys as possible before the Navy had a change of heart.
But he was in a foul mood that morning after waking to the news that a key instrument was on the fritz. Goldfinger huddled in the deckhouse with two students, trying to troubleshoot the problem. It took four hours of fiddling before the subsurface profiler was back in action, chirping like a demented robin. The high-pitched pulses penetrated the seafloor and bounced back, revealing the layers below. A multibeam sonar simultaneously scanned the bottom, recording every bump and ridge.
Such sophisticated instruments used to be off-limits for civilians, Goldfinger said, settling into a plastic chair on the Pacific Storm’s stern. The boat was passing Yaquina Head near the town of Newport. Nuclear submarines still travel the coast regularly, coming and going from a base on Puget Sound and hiding in the same underwater canyons Goldfinger studies. Especially during the Cold War, the Navy didn’t want the Soviets to know the detailed topography. “It’s all very Hunt for Red October.” Goldfinger said.
Multibeam sonar, developed in the 1980s, revolutionized underwater mapping with its ability to scan wide swaths of the seafloor. But the Navy kept a tight lid on the technology for years. Goldfinger’s graduate school adviser held a security clearance and was allowed to peek at some of the classified maps. But he couldn’t share the information. Even after the new sonar was commercialized, the Navy—which owns many university research vessels—leaned on scientists not to publish their results. “They held a kind of government club over everyone’s heads.”
It was frustrating because sketchy maps hampered the science. The researchers needed to zero in on the best spots to core and know which areas to avoid. Goldfinger pushed the boundaries, winding up in his dean’s office once with a furious Navy oceanographer on the phone.
Portions of Washington’s underwater landscape stayed secret long after Oregon’s was declassified. Over the years, Goldfinger has been working to fill in the remaining gaps in the bathymetry, as he was doing on the Pacific Storm. “It’s taken us nearly two decades to get the level of detail we have now.”
As fog enveloped the boat, Goldfinger ducked back into the cramped laboratory packed with gear and lined with plywood tables. The ship’s engineer was cracking jokes with the students, showing off his photo gallery of seasick passengers doubled over the railing. He punched the buttons on his Puke Master key ring to add for a soundtrack of retching sounds.
Goldfinger shrugged and paraphrased the eighteenth-century essayist Samuel Johnson: “Going to sea is like being in prison, with the added possibility of drowning.” Bathymetry cruises, with their back-and-forth trajectories, are particularly monotonous. Goldfinger used the downtime to prepare for an upcoming conference.
Among his newest results are hints that Cascadia quakes come in clusters. On his laptop he pulled up a graph of the nineteen full-rip 9 quakes from the last ten thousand years. The quakes appear to clump together in knots of three to five, separated by a few hundred years. In between the clusters, all is quiet for about a thousand years. On the graph, the 1700 quake looks like the latest in a cluster of five.
Some scientists dismiss the patterns as nothing more than a statistical illusion. Goldfinger isn’t sure, but thinks there’s a possibility the clusters represent a real phenomenon. But the information isn’t very useful yet. Does the pattern mean the next magnitude 9 megaquake is a thousand years away? Or could a sequence of five average-sized quakes in a row indicate the fault is overdue for a magnitude 9-plus monster? Even with a ten-thousand-year record, there’s no way to tell. “We’re still at the blind-man-feeling-the-elephant’s-butt stage of plate tectonics,” Goldfinger said. “Nobody likes to think of it that way, but it’s true.”
The groping extends to the question of whether the world experiences periodic flurries of megaquakes—and whether one of those periods started with the 2004 Sumatra disaster. Before Sumatra, seven of the ten biggest quakes on record struck between 1950 and 1965. The lineup included Chile’s magnitude 9.5 record-holder and the 1964 Alaska megaquake and tsunami. The next four decades were strangely quiet, without a single quake of magnitude 8.5 or greater. Since 2004, there have been five, including an 8.8 in Chile and Japan’s cataclysmic 2011 quake and tsunami. “Everybody has noticed there’s something going on,” Goldfinger said.
Two USGS scientists analyzed the very biggest quakes and concluded there was only a 2 percent chance the clustering was random. “It’s very statistically significant,” lead author Charles Bufe said at a 2011 meeting of the Seismological Society of America. “We think we’re in an increased hazard situation for these very large earthquakes.” Bufe calculated a 63 percent chance another megaquake will strike by 2017, though he can’t say where.
<
br /> But the majority of scientists see no connection between giant quakes. Even coin tosses can produce a string of consecutive heads. Several other statistical analyses attribute the apparent clumps to chance and a short historic record. Another reason most scientists are skeptical is that they can’t explain how one megaquake could trigger another on the opposite side of the globe.
It’s well-known that earthquakes can beget other earthquakes. That’s what explains aftershocks. It makes intuitive sense that those effects are largely localized. When a fault ruptures it’s like a cog slipping, and the motion will ricochet through the adjacent geologic machinery. But geologists were caught off guard in 1992, when an earthquake in the Mojave Desert was followed almost immediately by more than a dozen quakes as far away as Wyoming. How could triggering occur over such long distances?
There’s still no good answer, even though the list of examples is growing. One of the most dramatic was a 2002 quake in Alaska that set off rumbles more than two thousand miles away in Yellowstone National Park. “The Earth is like a puzzle,” Goldfinger said. “Any time one piece moves, it interacts with the other pieces.”
Giant earthquakes jolt the Earth so hard its axis shifts slightly. Vibrations sweep over the entire planet and cause it to ring like a bell. The thrusting of plates sends slow waves through the taffylike mantle that underlies the planet’s crust. “Just because geologists haven’t identified a mechanism that links megaquakes, that doesn’t rule it out,” Goldfinger said. Geologists ridiculed Alfred Wegner’s theory of continental drift because there was no way at the time to explain how continents could move. “Almost everything we discover in nature, we make the observation first and the explanation comes later,” Goldfinger said. “It’s crazy to say something doesn’t exist because we can’t explain it.”
Full-Rip 9.0: The Next Big Earthquake in the Pacific Northwest Page 8