One of the best science stories you'll read all year is Kathryn Schulz's unnerving piece in The New Yorker about the Cascadia subduction zone, a little-known fault line along the US Pacific Northwest that has produced tremendous earthquakes in the distant past — and is all but certain to strike again someday.
If another truly large earthquake ever hit, it could ravage buildings and bridges throughout Seattle, Tacoma, Portland, and Eugene — and, much more worryingly, produce a giant tsunami capable of killing thousands of people along the West Coast. And we're not really prepared for this event. Not nearly as well as we should be.
Schulz's piece (which you should read in full, not least because it's beautifully written) doesn't include diagrams, so here's a helpful visual showing the subduction zone in question. The Cascadia fault stretches 700 miles from Vancouver Island in Canada down to Cape Mendocino in northern California:
The basic story here is that the Juan de Fuca Plate is being shoved beneath the North American Plate. Periodically, the two plates get stuck by friction, building up enormous strain over many hundreds of years until the rocks suddenly slip past each other, releasing that pent-up energy in seconds — and creating a staggeringly large earthquake.
When that happens, the land in the Pacific Northwest will suddenly drop several feet, while land to the west will rebound under the ocean, producing a large tidal wave heading for the West Coast. Terrible news.
As Schulz recounts, scientists didn't even realize earthquakes occurred along the Cascadia subduction zone until the 1980s. Careful sleuthing by two scientists determined that a tsunami known to have slammed Japan in 1700 was actually caused by a giant earthquake in the Pacific Northwest. And since then, Schulz writes, scientists have deduced that there have been roughly 41 earthquakes along the Cascadia subduction zone in the last 10,000 years.
But what does that number mean, really? Do the earthquakes occur regularly? Could another devastating quake hit tomorrow? Is there any way to predict this? To find out more, I called up Evelyn Roeloffs, a scientist at the US Geological Survey who studies the Cascadia fault.
Huge Cascadia earthquakes occur every 500 years, on average — but averages can be deceiving
Roeloffs' bottom line was that we can't say precisely when the next gigantic earthquake will hit Cascadia — all the more reason we should start preparing now.
She did offer a bit more insight into frequency, however. As best we can tell, the really enormous subduction zone earthquakes that run the full length of the Cascadia fault line — magnitude 9.0 or more — seem to occur every 500 to 600 years, on average. The catch? That's only an average, and earthquakes don't seem to occur at predictable intervals. The timeline below helps illustrate:
We don't have any official records of past subduction zone earthquakes in Cascadia, so scientists have had to reconstruct history from indirect evidence. In 2012, a team led by Chris Goldfinger of Oregon State University looked at marine sediment cores for signs of underwater landslides thought to be linked to earthquakes. (As it turns out, there's some debate over whether these landslides correspond perfectly with earthquake occurrences, but it's the best data we have for now.)
Goldfinger's results, illustrated on the timeline above, suggest there have been roughly 41 earthquakes along the Cascadia fault over the last 10,000 years of magnitude 8.0 or more. Those are all big earthquakes, but the precise size varied quite a bit. About half were on the smaller end and only occurred along the southern half of the fault.
About 19 of those Cascadia earthquakes, however, were truly enormous events, running the full length of the fault — magnitude 9.0 or greater. (The magnitude scale is logarithmic, so a 9.0 earthquake releases about 31 times as much energy as an 8.0.) If one of these hit today, it could prove incredibly destructive. "It's the difference between a big event and a really big event," Roeloffs says.
On average, these "full-margin ruptures" have occurred once every 526 years, but Roeloffs cautions that this is only an average. In other regions on Earth where we can observe earthquakes more regularly, seismologists have found that they don't necessarily recur in predictable intervals. So the next subduction zone quake could hit Cascadia tomorrow. Or it might not hit for centuries. We honestly don't know. Which means, rather than stay complacent, it makes sense to prepare for anything.
We still can't predict earthquakes — but we can build early warning systems
Why can't we predict these things? In order to do so, seismologists would need to find some sort of telltale sign that regularly pops up before an earthquake hits. And, right now, we just don't have that.
Only about 5 percent of earthquakes have smaller "foreshock" quakes that occur beforehand — so that's only mildly helpful. More recently, GPS and other instruments have allowed scientists to track plate motion more precisely, but it's not yet clear how all this data might be used to predict earthquakes. Roeloffs says this is an active area of research but will require "many decades of observations in subduction zones before we have a better understanding."
What we can build, however, is an early-warning system:
When an earthquake occurs, it produces both fast-moving P-waves — rolling waves of compression — and slower-moving S-waves that cause shaking. If we built systems to detect those P-waves, that would give us 10 to 15 seconds advance notice before the S-waves start doing damage. That may not sound like a lot of time, but it would allow train operators to put on the brakes, for elevators to stop, for electric utilities to prepare for potential disruptions, and so on.
Japan has an early-warning system of this sort, and the Pacific Northwest is currently developing one, although it's not yet fully funded. "It needs more testing and full funding before it is ready to be released to the public," said John Vidale of the Pacific Northwest Seismic Network.
The best preparation? Better building codes and steering clear of tsunami zones
Since the Pacific Northwest can't possibly know when the Big One will strike, Roeloffs notes, the smart moves would be to fortify buildings, improve building codes, try to reduce exposure to tsunami zones, and improve emergency preparedness.
Buildings: Seismologists often say that "Earthquakes don't kill people, buildings do." When an earthquake hits, a great many of the deaths occur because buildings collapse — as happened in Christchurch, New Zealand, in 2011, when a quake flattened a seven-story office building, killing 115 people.
In Washington and Oregon, unfortunately, most buildings were constructed before anyone realized that massive subduction zone quakes were a possibility in Cascadia. Schulz cites a FEMA analysis suggesting that 1 million buildings in the Pacific Northwest could either collapse or be compromised in the event of a major earthquake.
But we're not resigned to that fate. Over at Dot Earth, Andrew Revkin has long tracked the (slow but steady) progress in retrofitting vulnerable schools, a particular concern. Meanwhile, in a recent Reddit AMA chat, three earthquake experts explored various nuances about infrastructure throughout the Pacific Northwest. In Washington state, hundreds of bridges have been retrofitted, although hundreds still haven't. Most of Seattle's skyscrapers would likely sway but remain standing in a major earthquake. But Portland's downtown buildings are much more vulnerable to shaking, since no one there began planning for quakes until the 1990s.
The experts stressed that the Pacific Northwest wouldn't necessarily be reduced to smoking rubble in the event of a 9.0 quake (in Schulz's piece, she quotes one expert saying everything west of I-5 would be "toast"). Still, it would be bad. "Communications may black out, transportation may grind to a halt, stores conceivably could run out of goods for a while," Vidale said in the AMA. "But that doesn’t constitute 'toast' in one’s mind."
Tsunamis: Perhaps more importantly, there's the issue of minimizing people's exposure to tsunami-inundation zones. This, after all, is where the vast majority of casualties are likely to happen — it was the cause of most of the deaths in Japan after the 2011 Tohoku earthquake.
The next step is using this information to guide better planning. "Often that means rethinking land-use decisions — such as facilities located in tsunami-inundation zones," says Roeloffs. Even so, it's likely impossible to keep these areas completely free of people (beachgoers in the summer are a particular concern, Schulz writes in her piece).
Emergency readiness: Then there's day-to-day preparation. In the Reddit AMA, Debbie Goetz of Seattle's Emergency Management Office noted that Seattle has a recovery plan at the ready, but she advises people to prepare for 7-10 days of outages and disruptions in case a big earthquake hits.
"For a major quake," Goetz says, "life won’t be back to 'normal' after just three days. I’ve got enough at home to make it through a week, and also keep a stash of stuff in my car as well as at work. Beyond supplies, I always encourage people to talk about their plans – especially around communication, which we know will be affected. Where will they be? How can they get back together? Where could they meet if not at home?"
-- Read Kathryn Schulz's piece on the Cascadia fault.
-- Here's the Reddit AMA chat with three earthquake experts on the piece. They all agreed the piece was good and accurate, although they didn't want to oversell the devastation that would occur. Here's a good selection of highlights from the chat that might be useful for Washington/Oregon residents.
-- Andrew Revkin has been doing excellent writing around a Pacific Northwest earthquake for a long time, with a particular focus on schools. See here, but also see this insightful piece on some lessons for safer construction after the Nepal earthquake.
-- Here's a very good piece at Slate by Eric Holthaus on the earthquake and the AMA.
-- Sandi Doughton, a journalist for the Seattle Times, wrote an entire book about the Cascadia subduction zone, Full-Rip 9.0: The Next Big Earthquake in the Pacific Northwest. Worth checking out.