Disaster Deferred – Better Predicting the Next Midwestern Quake


For more than twenty years, Seth Stein, William Deering Professor of Earth and Planetary Sciences at Northwestern, and his team of researchers have been studying the complex system of faults and plates beneath our Earth’s surface. What they’ve found, detailed in Stein’s new book “Disaster Deferred,” is changing the way we think about predicting earthquakes. We spoke with Stein for a preview.

Your book describes a new way of predicting earthquakes in the Midwest. How have we predicted earthquakes in the past?
Scientists have never been able to predict earthquakes on a short-term basis. That is, within years, much less days. But, in the last forty years, we’ve learned that most earthquakes happen on what we call the boundaries between plates, things like the San Andreas Fault (the boundary between the Pacific and North American plates). We also know about how fast the plates move – we do that with a global positioning system (GPS).

Seth Stein (photo courtesy of Seth Stein)Seth Stein (photo courtesy of Seth Stein)For example, we can tell you that the Pacific and North American plates on either side of the San Andreas Fault are moving 36 millimeters a year. We know that a big earthquake, like the one that happened in 1906, slipped about 4 meters. If we divide those out, we can say that a big earthquake like that should happen about every 100-200 years.

What we don’t understand are earthquakes that happen in the middle of plates, like in the Midwest. That’s what this project has been focusing on for the last 20 years or so. There are lots of examples; [two] particularly famous ones are the earthquakes that happened in 1811 and 1812 in the central United States, near the town of New Madrid, Missouri. We’ve been trying to understand what causes these earthquakes, how often they happen, and by extension, how dangerous they are as a problem for society. We started making measurements from the GPS in 1990 and we found some really amazing surprises.

What have you found?
We know the ground stores up energy between quakes. It’s kind of like a rubber band – you pull it for a while and then snap, it moves. So we can see the ground deforming as it stores up energy for the next earthquake. If you look across the San Andreas fault, or the earthquake zones in Utah, or Oregon, or Washington, you’ll see that. At every earthquake zone in the world we can see the ground storing up energy for the next quake.

We took a bunch of measurements in 1990, 1993, and 1997 in the New Madrid earthquake zone, and, to our complete surprise, the ground isn’t moving at all. It’s moving less than the thickness of a piece of fishing line per year – that’s how good GPS has gotten these days – so the motion’s either tiny or zero. What that means is the ground isn’t storing up energy for the next big earthquake, so there’s no reason to believe one is on the way soon.

The scientific bottom line is that we now think, instead of one fault going on and on, having an earthquake every few hundred years, like the San Andreas, that places like the Midwest, the interior of China , or northwest Europe – the insides of the continents – have lots and lots of old faults. They’ve accumulated over the last billion years or so, and they never really heal. Every so often, one of them will pop off, then maybe do nothing for 10,000 or 100,00 years, and then an earthquake will happen somewhere else.

We call this idea episodic clustering and migrating. It’s a very different view, but it fits into a general idea that we have in a lot of areas of science about what we call complex systems. Complex systems are ones where you have a lot of interacting parts, and you can’t just pull out one piece and try to understand it. We can think about the San Andreas fault as more or less its own thing because it’s at the boundary between two plates. But in the Midwest (inside of a continent), you have to think about all of these faults interacting. It’s a very different picture, and it’s giving us a lot of new insights.

Is there an easy way to visualize this?
There’s a kid’s game called “Booby Trap,” which is really fun. You have a pusher piston, and these little round disks. You load this piston up – there’s a spring pushing it – and then pull out some of the disks. Most of the time nothing happens at all, occasionally you’ll get small motions, and very occasionally you’ll get a big motion as the whole system slides. The idea is that you have lots of faults, and as motion happens at one of them, it loads some of the other ones – it changes the forces in the system. And then every so often you’ll get a big earthquake.

How will this change the way we think about the New Madrid Seismic Zone?
There were the big earthquakes in 1811 and 1812, from the geological record we think there were some in about 1450 (probably like what happened in 1811 and 1812), and also in about 900 AD. So people have started to think, it’s been about 200 years since 1811, maybe in a few hundred years we’ll see a big earthquake. But right now, there’s no sign of any such thing on the way.

However, the government is convinced that earthquakes in the New Madrid Zone are even more dangerous than California, so they’re telling the states and cities in the area that they have to do very expensive earthquake-resistant construction. We’re talking billions and billions of dollars over the next few hundred years. The problem is that the government says the next big earthquake is coming, but it doesn’t look like there’s any good reason to believe that. At this point, it seems we’d be much better off spending that money on other societal needs. As a friend of mine said, you’re probably better off hiring more teachers than putting steel into your schools.

What about the smaller earthquakes that have occurred in the New Madrid Zone since the 1800’s?
That’s another kind of amazing thing. Every time one of these happens, the government says “See? This proves a big earthquake is coming.” We’ve looked at that very carefully, and it looks like an awful lot of those earthquakes are actually aftershocks of what happened in 1811 and 1812. So they don’t tell you a big earthquake is coming; they tell you there was a big earthquake 200 years ago.

Once we realized this, we looked into our understanding of the physics of how rocks break, and how aftershocks work, and that’s exactly what you’d expect. That, in the middle of the plates, where motions are very, very slow, aftershocks should go on for hundreds of years. We’ve started looking for this pattern in other parts of the world and can find the same thing. In a sense, it’s like after a traffic accident – if you have a hard-working clean-up crew, they can clean it up in an hour; if you have very slow-moving people, it can take them all day.  

How do you know these smaller earthquakes are actually aftershocks?
There’s four different lines of evidence. The first is that they happen on what we think were the faults that broke on 1811 and 1812. Small earthquakes that happen on the fault plane of a past big earthquake are aftershocks; that’s how we define them. Second, the size of them seems to be decreasing with time. Third, the biggest ones are at the edges of the fault, or the kinks in the fault, where the stresses were concentrated, and that’s also a pattern of aftershocks. Fourth, and perhaps most importantly, is that rock physics theory tells us that the duration of an aftershock sequence should be inversely proportional to the rate at which you rewove the fault. So that’s the clean-up crew argument. And that’s because, in the San Andreas fault, let’s say ten years after a big earthquake, the motion of the plates is what’s controlling what’s happening at the fault. But because very little is happening at the middle of the plate, the biggest effect two hundred years after the New Madrid earthquakes is still the New Madrid earthquakes. Nothing else has swamped it out yet. 

There’s a lot of publicity about the bicentennial of the 1811-1812 earthquakes. What do you think of it?

A lot of it is pretty silly. The government will be doing lots of PR and holding events, including drills, conferences, and field trips, to convince communities that disaster is coming. It’s like a few years ago when they told us to go out and buy duct tape to protect our houses against bioterrorists. Since there’s no reason to believe the disaster warnings, my advice is to enjoy the bicentennial – but not take it too seriously.

Seth Stein’s new book, “Disaster Deferred: How New Science is Changing our View of Earthquake Hazards in the Midwest” is available from Columbia University Press. Visit his website for more information.

Add new comment

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.
Refresh Type the characters you see in this picture. Type the characters you see in the picture; if you can't read them, submit the form and a new image will be generated. Not case sensitive.  Switch to audio verification.