Short Wave - Why Can't We Predict Earthquakes?

Episode Date: February 8, 2023

In the wake of the massive earthquake in Turkey and Syria, many scientists have been saying this area was "overdue" for a major quake. But no one knew just when: No scientist has "ever predicted a major earthquake," the U.S. Geological Survey says. Even the most promising earthquake models can only offer seconds of warning. In this episode, host Emily Kwong talks to geologist Wendy Bohon and NPR science correspondent Geoff Brumfiel about why earthquake prediction can be so difficult, and the science that fuels these models. See pcm.adswizz.com for information about our collection and use of personal data for sponsorship and to manage your podcast sponsorship preferences.NPR Privacy Policy

You're listening to Shortwave from NPR. Hey, shortwavers, Emily Kwong here. So you may have heard about the 7.8 magnitude earthquake that struck southeastern Turkey near the Syrian border in the early hours of Monday morning. It was felt hundreds of miles in all directions. Aftershocks, leveled buildings that had already been damaged. And a few hours later, there was a rare 7.5 magnitude aftershock. The death toll quickly climbed into the thousands.

To make sense of all of it, I'm here with Jeff Brumfield, MPR Science Correspondent. And Jeff, you told me earlier that an earthquake of this size had been expected in this region for a long time. Yeah, Emily, several scientists I spoke to yesterday said that this particular part of Turkey was overdue for an earthquake, a large earthquake, meaning it hadn't had a really big one for well over 100 years. And what makes this part of Turkey and Syria such an active earthquake zone? Yeah, I mean, I think to zoom out for a minute, we have to look at the big geological picture. And that is basically a tectonic pile up that's happening in this region. What's going on is that we have a plate under the Arabian Peninsula. That's known as the Arabian plate.

And it's driving its way north towards another large plate, the Eurasian plate. And I talked to Michael Steckler of Columbia University's Lamont-Dordy Earth Observatory, and he described kind of what's going on. Arabia is slowly moving north and has been colliding with Turkey, and Turkey is moving out of the way to the west. So basically, the nation of Turkey sits on top of what's known as the Anatolian plate, and that's getting squeezed in a vice between the Arabian plate and the Eurasian plate as they come together.

This tectonic shift has been behind earthquakes in this part of the world for millennia. Really? Okay. Which earthquakes? Well, I mean, one of the most famous ones was a major quake in 1138 that pretty much leveled the Syrian city of Aleppo. And then more recently, quakes such as the 1999 one

that struck the city of Ismeat killed many thousands. Monday's quake is believed to be the most powerful one to hit Turkey in more than 80 years. Yeah, okay. So it sounds, though long overdue, kind of impossible to predict the U.S. Geological Survey has said no scientist has ever predicted a major earthquake. even the most promising models only give you a few seconds of warning. And today we're going to revisit an episode about why that is. But first, Jeff, I want to focus on this region and specifically where the earthquake happened along the eastern Anatolian fault zone. Can you tell me about that?

Right. So there's two major fault lines in Turkey. There is the North Anatolian fault and the eastern Anatolian fault. And most of the action for the past century has been along that north. fault. That has seen most of Turkey's major earthquakes, in part because it's moving very quickly, relatively speaking. But seismologist I'd spoken to had said that they've been watching the Eastern Anatolian fault. I spoke to this guy named Fatea Balut, Boazache University in Istanbul, and he told me that basically that fault has also been building up stress for a long time

at the rate of about a centimeter a year. And they just haven't seen a movement. So they knew the pressure was building. They knew they were due for a big earthquake. This is not surprised for us. And in fact, Balut said to me that his modeling had predicted a 7.4. A colleague's modeling had predicted a 7.7. So they were expecting something in this ballpark.

But of course, as you just said, that doesn't mean they know when it's going to happen. It could happen tomorrow. It could happen next year. It could happen in a day. decade. And so they had no real way to warn people that this particular disaster was going to come do. Yeah. Well, Jeff, as we watch what's happening in Turkey, I just want to thank you for explaining what's been happening underneath Turkey. Thank you, Emily. We're going to spend the rest of the show talking about why predicting when and where earthquakes happen is such a difficult problem and what goes

into detecting them in the first place. I'm Emily Kwong, and you're listening to Shortwave, the Daily Science podcast from NPR. Okay, so to delve further into earthquake science today, I want to revisit this conversation with geologist and science communicator Wendy Bohan. She works for a consortium of over 100 universities in the U.S. gathering seismic data from all over the world. It's called the Incorporated Research Institutions for Seismology, or Iris for short. And this global seismographic network, as well as local, regional and international seismic networks, All of that was critical for detecting the earthquake in Turkey, the main shocks, and all of the aftershocks, giving people information about its size and depth. And no matter where you go in the world, these seismic networks are using the same tool, a seismometer. The instruments are unbelievably cool. They're called seismometers, and there are thousands of them all over the world. We place them in basically anywhere that we can. We do it strategically, usually around areas that have active faults. But there's also caches of instruments that can be used after an earthquake happens.

So scientists will get those instruments and we'll deploy them or put them out near where a big earthquake has just happened so that we can learn more about what's happening underground. So you can think of every seismometer, almost like a pixel in a camera, right? So the more pixels you have, the more scientific instruments that we have out, the higher resolution we can see into the ground. So these instruments, can you describe like how they're installed and then who is babysitting them? Like what are they looking for? So there's lots of different kinds of seismic instruments. The scientific kind are generally putting a big hole in the ground, sometimes installed in concrete, depending on where you are. Sometimes they're just dug down.

What they're looking for is anything that shakes the ground. And these suckers are very, very sensitive. They can detect changes. in the ground that are like the size of a human hair. What's the mechanism inside these seismometers that has gotten so sensitive? It can detect things like that. Now it's electric. It's based on the idea that you have kind of like a pendulum that's hanging and then the

seismometer is attached to the ground. So the idea is that the seismometer moves with the ground, but the pendulum stays still. And so that pendulum can measure how much the ground has moved. So advances in the technology, advances in the electronics. have allowed us to really detect, you know, much more sensitively and also a lot more about, like, the different directions that the ground can move. Because an earthquake doesn't just shake the ground back and forth. You know, you get an upwards thrust, a sideways push, a downward drop, and then a sideways push in the other direction. The ground's moving in all kinds of different ways.

It's important for us to know and record all of those different aspects. What signals, well, first, so there's this network of seismometers that exist, throughout the U.S. to detect earthquakes. Is that right? Throughout the U.S., throughout the world. We have the global seismic network, which is maintained by Iris and the U.S.GS, U.S. Geological Survey, and that's like 150 to 200 instruments that are around the world. We have networks in the United States. Other countries have their own networks.

And then there are also smaller regional networks, like the Pacific Northwest network. And most of the data from most of these networks gets fed into the IRIS data management center. So it's open source for anyone in the world to use. While it is pretty much impossible to predict an earthquake, it is possible to develop an early warning system. Turkey and Syria do not have a comprehensive early alert system across the country, so people there had no warning. But other seismically active regions, including Japan, Mexico, and the west coast of the U.S. do have systems like that. Here's how they work. When an earthquake happens and it sends the waves out, if we need at least three seismometers,

to detect it. And very quickly, we can get an estimate using mathematical algorithms and things about the magnitude of the earthquake and also the level of shaking. And so all of that information gets sent from the seismometers to the data center with the USGS and their partner organizations. Then they decide automatically, it's not like a person deciding, whether or not that earthquake meets the standard that it's going to potentially cause damage or people need to know. then that will send alerts out to people in the area. And so you could get, you know, between a few seconds of warning, sometimes even as much as 30 seconds or a minute of warning.

And people will say, you know, well, who cares about 10 or 15 seconds? I can tell you, you would care if you were getting LASIC eye surgery. You know, you would care if you were in the dentist chair. So there's other things we can do with that much warning, like slow down trains to prevent derailments. we can open firehouse doors so that the emergency response equipment doesn't get stuck inside if the doors are jammed during the earthquake shaking. So there's, you know, things that we can do both as individuals and as, you know, emergency response personnel, people that are involved in public safety can use these alerts to help keep people in property safe. Okay.

Okay. Like, how would you grade the current state of global earthquake science? I do. Well, you know, we at Iris maintain the database, and that's open science. We want this science to be open and accessible to everyone. Public citizen, researcher, anywhere, anybody in the world can access that data. So that's huge, right? Open access data in science is not always the norm.

Also, researchers are really working together quite a bit, and things like Twitter have really, really helped with that. So here's an example. After the 2019 Ridgecrest earthquake that happened in California, which was actually the largest earthquake to happen in California since Hector Mine, it was magnitude 7.1 out in the desert, there were some folks from the California Geological Survey that were out surveying where the ground was broken. And some folks in France had satellite data that showed them areas where there was deformation in the ground. And so through Twitter, these French researchers were sending information to these. American geologists that are out looking for the rupture saying, okay, go to this coordinate, go to this coordinate, this is where these things are. So people are really working in real time

to coordinate responses. So I'm hopeful that as these early career scientists and mid-career scientists are reaching across the ocean and starting to talk to different research groups, that great collaborations will come out of that because science is not a solitary endeavor. Today's episode was produced by Liz Metzger, Indy Kara, and Rebecca Ramirez. It was edited by Gabriel Spitzer and Viet Leigh, and fact-checked by Anil Oza and Burley McCoy. The audio engineer for this episode was James Willits. Rebecca Ramirez is our supervising producer. Brendan Crump is our podcast coordinator.

Our senior director of programming is Beth Donovan, and the senior vice president of programming is Anya Grundman. I'm Emily Kwong. Thank you for listening to Shortwave. The Daily Science Podcast from NPR.