The threat of a 9.0-magnitude earthquake along the Cascadia Subduction Zone has hung over the Pacific Northwest for decades.
Seismologists and emergency managers say “The Big One” could be one of the worst natural disasters in the region’s history, but it may not be the worst-case scenario. New research from Oregon State University suggests the Cascadia Subduction Zone may be linked to the San Andreas Fault in California, with seismic activity on one triggering corresponding activity on the other.
Chris Goldfinger is a professor emeritus at OSU and the study’s lead author. He joins us with more details on what the findings could mean for our region.
Note: The following transcript was transcribed digitally and validated for accuracy, readability and formatting by an OPB volunteer.
Dave Miller: From the Gert Boyle Studio at OPB, this is Think Out Loud. I’m Dave Miller. The threat of a 9.0 magnitude earthquake along the Cascadia Subduction zone has hung over the Pacific Northwest for decades. Seismologists and emergency managers say the big one could be one of the worst natural disasters in U.S. history, but it may no longer be the worst case scenario. New research from Oregon State University suggests the Cascadia subduction zone may be linked to the San Andreas fault in California. That means one massive earthquake could trigger another one, potentially affecting an enormous portion of the Pacific coast from Vancouver down to San Francisco. Chris Goldfinger is a professor emeritus at OSU and the study’s lead author. He joins us now. Welcome back to the show.
Chris Goldfinger: Thanks for having me on. Good afternoon.
Miller: Good afternoon. I understand that you first collected a soil core that suggested a link between these two areas ‒ San Andreas and Cascadia Seduction zone faults ‒ back in 1999. How did that come about?
Goldfinger: You know, that’s right. We were on our first project to work on Cascadia earthquakes, and we were at the very far south end near Eureka and Humboldt County in California, near Cape Mendocino. We made a little navigational error during the nightwatch and wound up off Northern California at Fort Bragg and Noyo Canyon. So since we were there, we stopped and took a core there as well, thinking maybe it might be useful sometime in the future and it turned out that it was.
Miller: Did you have a hypothesis back then that they were connected?
Goldfinger: Oh, no. Not at all, not at all. We treated them as separate systems as all of us have for many years. So we were just hoping for a decent earthquake record from the San Andreas.
Miller: So what led you to actually investigate the question of linkage?
Goldfinger: About 2004, 2005 or so, we started to notice that the radiocarbon ages for Cascadia earthquakes, both from the workers onshore and our own, seem to match the earthquake record for the San Andreas. Radiocarbon has a lot of uncertainty, so matching is maybe not the best word. But they seemed to be very similar and we just couldn’t understand why they would be similar. It didn’t make sense. And it took us a while to think of a hypothesis that would make sense for this and one was that the faults had gone off together, or one had triggered the other, or they’d gone off within some period of time that we couldn’t resolve.
Miller: Can you explain then the different kinds of data sources that, put together, gave you confidence to say now that these two faults can be linked, have earthquakes that in recent times over the last some number of thousands of years have happened at around the same time?
Goldfinger: Right, right. So, in the offshore geology that we do, we take core samples and we can correlate one bed to another, to another. And we can radiocarbon date them. That gives us the basic framework of time and space for these earthquakes. And we can do that on both the Cascadia side and the San Andreas side with the boundary at Cape Mendocino. And we wrote a paper in 2008 suggesting this hypothesis based on that coincidence of timings. But with all the uncertainty in radiocarbon, you can’t really nail it down to any sort of precision. It just remains a nice, interesting coincidence, a hypothesis, if you will.
But there was something really odd about these beds. They’re called turbidites. They’re submarine landslide beds that are triggered by earthquakes and sometimes other things. And there was something really odd about them. The ones near Cape Mendocino seemed to be upside down. Normally these sedimentary beds have all the sand at the bottom just as if you had a bucket of sand at the beach and you’re swishing it around. All the sand goes right to the bottom, and the fine grain stuff is at the top. But these appeared to be upside down with all the sand at the top. And so that kind of nagged at us for years, and we couldn’t think of an explanation, couldn’t think of one, couldn’t think of one.
Then quite a few years later, 2016 or 2017, somewhere in there, we started to realize that there might be an explanation for this. It was that there wasn’t a single upside down bed. It was actually two beds. And the lower one was weaker and finer grained. And the upper one was sandy and coarser grained. So gravity hadn’t reversed its normal direction. It wasn’t upside down, it was just two beds. And then it began to make sense. We had two beds stacked on top of each other. And they occurred very closely spaced in time, because the sediment that accumulates like snow in between events wasn’t there. There was just nothing there.
And so that’s what we call a relative dating test where we can’t get the radiocarbon ages or precision linkage to absolute timing very well. But we can get the relative timing to say that one thing happened close to another, based on the sedimentation rate. You know when it’s snowing, you can see it’s accumulating at a certain rate and the same thing is true in the marine environment where the sedimentary snow, if you will, accumulates slowly and you can measure the rate. And if there’s nothing there, then you know, typically, that a very short time has passed between two events.
Miller: When a geologist says a very short amount of time has passed between two events, it can mean something different than when a regular human says that. I mean, the difference, a gap of five hours, say, and five years is the blink of an eye for a lot of geologists and a lot of geology. It makes a gigantic difference if we’re going to talk about emergency preparedness or emergency response. Do you have any sense, based on the data, for the time between these two linked earthquakes if they do happen around the same time. How closely are they actually happening?
Goldfinger: That’s a really good point. 1,000 years is almost meaningless to a geologist. So, yeah, that’s an excellent point. We have a whole series of these going back 10,000 years. And each one is different in terms of what evidence it produces, or can be used to to detect the timing difference. So the best supported case was in 1700 AD. The very well known Cascadia 1700 earthquake apparently had a San Andreas mate that happened very likely in the same year, and very likely, within even a shorter time than that.
The constraint is that on the Cascadia side, the core samples there clearly show the Cascadia event being deposited. And they also show a second event, which we believe came from the San Andreas, that was deposited before the first event bed was completed. So it was actually coming down while the first bed was settling. So the timing constraint on that could be as short as half an hour. It could be maybe a few days, possibly as much as a week, but most likely the timing is in hours.
Miller: And that’s the last major Cascadia Subduction Zone event, within perhaps a couple of hours or even shorter than that?
Goldfinger: Yes, that’s right. And so we know the timing of that event, the absolute timing we happen to know, from the Japanese tsunami. It’s January 26th, 1700 at 9 p.m. at night. And so now we think that the San Andreas earthquake followed within some number of hours of that.
The other support that works for that event is that all of the radiocarbon ages for landslides on the San Andreas going south as far as San Francisco including a place called Lake Merced also point fairly closely to 1700 to 1710, which is a big difference, but they’re consistent with it, is the best way to say it.
And then the final bit that helps that one is from a group of scientists working on dendrochronology. Tree ring damage from earthquakes have pinned down the penultimate San Andreas earthquake to either 1698 or 1700. So their data probably puts that event in at least the same year as 1700. That’s an example of the types of support, and the 1700 event is the best one of all. There are two others that aren’t quite as good as that that show the settling of one deposit into the other before it was completed. And then we have a whole group of others where it’s much less precise than that, but looks close.
Miller: You’re an earthquake geologist, not an emergency preparedness manager. But I have to ask you what you think this finding could mean for planning on the West Coast, a potentially devastating earthquake in the Pacific Northwest happening just before or soon before, a decent sized quake to the south? And you have a minute and a half to to answer that question.
Goldfinger: Yeah. Well, yeah, good question. So for people in the Pacific Northwest, we won’t even know that the San Andreas went off other than during the event. So it would have really no practical impact during the event for us. In terms of the emergency management side, trying to deal with two very large earthquakes close in time, whether it’s an hour or a day or whatever, something close, would tend to draw down the resources of the whole country in a way that nobody, I don’t think, has really considered at this time.
The other impact though, is if you’re in San Francisco or in the Bay Area, somewhere in Northern California, and a Cascadian earthquake starts, that essentially could serve as a warning that could be minutes, hours, days, something like that, for a San Andreas earthquake. Because the matching of these two things is not a black swan event. It seems to be the majority of the time. And the 1906 earthquake that we all know so well that destroyed San Francisco was the exception, not the norm. So significant changes in thinking and possibly a little bit of help actually for the Bay Area to have a warning that’s more than a few seconds.
Miller: Chris Goldfinger, thanks very much.
Goldfinger: Thank you, Dave.
Miller: Chris Goldfinger is an earthquake geologist and a professor emeritus at OSU. He’s the lead author on a new paper that found that earthquakes from the Cascadia Subduction zone and the San Andreas Fault are often linked, meaning two separate catastrophic earthquakes on the West Coast could happen, have happened in history many times at around the same time.
“Think Out Loud®” broadcasts live at noon every weekday and rebroadcasts at 8 p.m.
If you’d like to comment on any of the topics in this show or suggest a topic of your own, please get in touch with us on Facebook, send an email to thinkoutloud@opb.org, or you can leave a voicemail for us at 503-293-1983.
