How earthquakes work, and how science makes us safer


The very powerful earthquake that hit Japan today was even more powerful than everybody first thought. US Geological Survey seismologists upgraded it to a magnitude 9.0—making it the most powerful earthquake to ever hit Japan. It's not unusual for scientists to revise their calculations on the strength of an earthquake, reports New Scientist. In fact, we're likely to see more recalculation on this quake.

To find out more about the science happening behind the scenes of this disaster, I spoke this morning with geophysicist Brian Shiro, who works out of the NOAA Pacific Tsunami Warning Center in Ewa Beach, Hawaii. He gave me some good background information on the science of seismology and plate tectonics, and the changing technology that's making people safer in the face of powerful natural forces.

Maggie Koerth-Baker: My first thought, when I heard about this earthquake, was to wonder whether it had any connection to the recent earthquake in Christchurch, New Zealand. They're both part of the Pacific Rim, could what happens on one fault line affect what happens on another?

Brian Shiro: It's doubtful that there's a connection to the New Zealand earthquake, but the question isn't unreasonable. There's actually a lot that seismologists have been learning about the ways that earthquakes could possibly affect one another. If you'd asked this five years ago, they'd have said, "No way." But now, particularly after studying the Sumatra earthquake and tsunami [from 2004], there's been a lot of people looking into that question. The idea of one earthquake triggering another seems to have some legs to it. It's not a crazy idea. But the jury is still out on how often it happens.

MKB: How is the Pacific Rim set up? In the maps from grade school, it looks like one, continuous gap that circles the Pacific Ocean, but I'm assuming there's more to it than that.

BS: It's a bunch of distinct faults that, taken as a whole, create this so-called Ring of Fire where most earthquakes and volcanoes and other seismic activity originates. I like to think of the Earth like hard boiled egg. If you crack it a little bit, that's like the plates. The plate boundaries are riding on mushy mantle. People also use the baseball analogy, which is maybe even better because of the stitching. You can see the seam on a baseball where it's connected all the way around, but you can also see the stitching. The Pacific Rim is like that. Even though the fault systems are distinct, it forms a continuous band. It's a series of subduction zones that mark areas in the Pacific Ocean where the Pacific plate is sliding underneath another one.

MKB: Why are there so many more earthquakes and volcanoes in this region. Is there something that makes the seismic activity in the Pacific Rim different?

BS: It's a matter of numbers really. The Pacific Rim covers a huge portion of the Earth. Really, there's just that much more opportunity for things to happen. Look at the various seas and oceans. The Mediterranean, for instance, has one subduction zone. The Atlantic has none. The Indian Ocean has two. The Pacific has on the order of 10.

MKB: What is the magnitude rating of an earthquake actually measuring?

BS: Seismologists stopped using Richter Scale 20-30 years ago. That was based on ground displacement. The public still uses the name because it is so embedded in consciousness. Today, we still use a logarithmic scale like the Richter Scale, but now it's based on the amount of energy that the earthquake released. Bigger magnitude means more energy released. And that's really related to the type of rupture in the fault.

MKB: There seems to have been a lot of confusion this morning about how big the tsunami waves would be by the time they hit places like Hawaii and the American West Coast. What factors make that difficult to predict? I think that, to a lot of laypeople, it seems like it must be a fairly simple distance/force physics problem. What complicates it?

BS: You have to separate this out into two questions: Time, and how big it will be. Those are two very different things. Estimating when a tsunami wave will reach a place is a simple thing to do from fundamental physics. You can do it with a pen and paper, and a calculator. The key thing is that you have to know the depth of the water, which we do pretty well throughout the ocean. Then you can predict time with great deal of accuracy. We've been doing that for several decades.

But knowing the size of the wave—or even if the wave will still exist after traveling a certain distance—that's another thing. There are lots of nonlinear effects. It's not trivial. This is serious fluid dynamics. Until the last 5 years, maybe, it wasn't possible to make those predictions in reasonable amount of time. Today, we have the computing power to do it in a matter of seconds. We're advancing because of computer speed.

But there are still conflicting estimations, and those have to do with different computer models asking the same question. There's a lot of variation between models on methodology and assumptions—for example fault orientation and magnitude go into it. It all boils down to conservation of energy, in the end. That's what the numbers are crunching. And that gets particularly difficult when you reach the coast. Out in the open ocean this isn't a big deal. But when you get down to the nitty gritty of a coastline, then you have to take into account a lot of nonlinear effects like reflecting or refracting. That's where small differences in the models can lead to a big difference in the end.

In this case, though, the models were right on for Hawaii, which means that they'll likely be good for California, as well.

MKB: Cory Doctorow, one of our editors at BoingBoing, moved to California not long after the Loma Prieta earthquake and remembers people telling him that nobody really knew, until it happened, how well earthquake-proof building designs would work. Not sure if this is outside your area, but can you tell me a little about how researchers test those designs before nature tests them?

BS: It is outside my area, but I can tell you a little about it. They do computer simulations, they also do shake tables that can shake the models and see what happens. Also, especially in last few years, they've installed instruments in structures like bridges and skyscrapers. Seismometers and accelerometers are both a lot cheaper now. There's an accelerometer in every iPhone. So now they can basically implant them in a building as you build it. That helps you really understand how the earth moves.

One other thing: The shaking that a building experiences really depends on the type of ground it's sitting on. Bedrock won't shake as much as landfill or alluvial soil. That's why sometimes one neighborhood may be OK and in the next neighborhood over everything topples. An earthquake can liquefy soil but it can't liquefy solid rock.

MKB: How well do these designs hold up when the time comes? From the little I've seen this morning, it's sounding like most of the deaths in Japan were tsunami related, rather than earthquake related. Is that a win for engineering? [NOTE: After further research, it's not clear at this time whether the earthquake or the tsunami has been responsible for more deaths than the other. We probably won't know the answer to that for a long time.]

BS: If that's true then that's really interesting. It says a lot about the engineering. Japan has the most sophisticated earthquake and tsunami warning system anywhere. They have hundreds of seismic stations throughout Japan and they're all connected to an earthquake early warning system, like what California is currently developing. When these centers first detect a seismic wave, they can send a signal out ahead of the wave to dams, power plants, and subways—places that need to shut down to avoid damage. Before the earthquake even hits they can be shut down.

[NOTE: This is actually why Japan, a country with 30% nuclear-powered electricity, is only having problems with two of its nuclear power plants today. The early warning system shut down 15 nuclear plants for safety. 11 of those are already back online, operating as normal. The two you've been hearing about today were those nearest the epicenter. —MKB]

MKB: Is it possible to engineer tsunami safety the way we engineer earthquake safety?

BS: There are. In fact, Japan again leads the way on this. They have these huge metal gates that can close off rivers, which are conduits for tsunamis traveling inland. The tsunami early warning system triggers the gates, the gates crash down, and it saves a lot of lives. There are a number of cabled offshore buoys and detectors—that gives people a few minutes notice. There's also a lot of proposals out there, too. I'm not aware of any serious efforts to implement this, but building huge artificial reefs could help because those are one of the natural protections against tsunamis. As are mangrove forests. Putting up those barriers is possible, but it would take a lot of money.

You can read about some other aspects of the science of the Sendai Earthquake on


  1. “making it the most powerful earthquake to ever hit Japan”

    Can I be a total pedant and point out that it is highly unlikely to be the most powerful earthquake ever. It is the most powerful recorded earthquake to to hit Japan.

  2. This was really interesting, especially the bit about the unpredictability of wave size at the opposite end of the ocean from the earthquake. Here on the southern Oregon coast, our sirens keep going off, we keep hauling up the bluff to higher ground, and then watching as a not-so-scary 2 foot swell rolls in.

    I imagine that some measuring device somewhere triggers these warnings to clear the coastline, yet the devices are unable to say whether the freakout is really warranted or not.

    1. I’m pretty sure it’s always warranted. It’s a mild inconvenience to have to go to high ground when it gets a false positive, but the alternative is you possibly drowning or being bashed to death by a wave.

  3. Science may make us safer but economics will always trump science. If a safeguard is expensive it might as well not exist for most people. Office towers may stay up but the worker’s homes collapse.

    1. If a safeguard is expensive it might as well not exist for most people. Office towers may stay up but the worker’s homes collapse.

      Yes, but quake-proofing lightweight low-rise buildings is much simpler and cheaper than quake-proofing large high-rise towers, so unless you’re housing your workers in shoddily-built high-rise tenements (or cheap-ass prefab concrete Khrushchyovkas), worker housing can be protected at very little additional cost.

      LA is often derided for its stick-built, stucco-walled, asphalt-shingled single-family homes and one-to-two-story courtyard apartments, but such structures are inherently (almost) quake-safe.

      Make sure the wall panels have some cross-bracing, bolt the wall frames to the foundation, strap down the water heater, and install an automatic seismic shutoff on the gas line, and you’re good to go.

      It’s no accident that so much of LA’s housing stock is built this way.

      Those measures add only a negligible amount to the cost of building housing, and are even affordable as a retrofit.

  4. Great interview. As a geologist I already knew all this stuff, but I enjoyed reading it anyway :)

    Regarding engineering earthquake-proof structures – it will always be true to some extent that you don’t know if a new design will hold up the way you expect until you get a large earthquake. However, the science and engineering (and testing) that goes into such designs is incredible, and they do actually know quite well how these things should react.

    In fact I’d go so far as to say it’s a solved problem – for new construction in earthquake-prone areas, there’s no reason to expect structural failure in an earthquake, except in circumstances that are way out of what’s normally expected (and while rare and exceptional, a 9.0 off the coast of Japan is not totally unexpected).

    The problem is of course paying for it, and retrofitting existing structures. In Japan they’ve done this quite well, but even in California this isn’t the case – there are still a lot of older buildings that are not “up to code”, although there is plenty of money spent on retrofitting in California.

    In the rest of the world’s earthquake-prone areas, there is little to no money spent on retrofitting, and new construction is, realistically, unregulated (bribes and corruption ensure this continues despite what the law may say) – and done on the very, very cheap.

    Wood-frame construction as is common in the US is actually very earthquake-safe; I’ve seen videos of full-scale wood houses on shake tables being hit with massive shaking with no structural damage. Most of Asia and the Pacific does not extensively use wood-frame construction, and you see a lot of concrete structures especially. This requires a lot more engineering to survive a large earthquake, which simply doesn’t get done in many places.

    Just as an example, I did research on tectonics in Thailand in grad school – it’s actually not at all an earthquake-prone place, despite its formation being related to the India-Asia convergence, so this isn’t surprising – but the construction there is awful. I’ve seen new structures going up that look like they’re made out of concrete playing cards and toothpicks. The old stuff (and most of it is old, even in Bangkok where everything is constantly under construction) is even worse.

    Further, tsunami warning systems and education about what to do is entirely lacking in the pacific, except in Japan, including the places that were hit by the tsunami in 2004.

    I mean heck, in many of the videos coming out of Japan, you see lots of people who weren’t doing the right thing while the earthquake was happening. Really basic stuff. And this is easily the places with the most awareness and education regarding earthquakes in the entire world. So it certainly isn’t an easy problem.

    1. “In the rest of the world’s earthquake-prone areas, there is little to no money spent on retrofitting, and new construction is, realistically, unregulated (bribes and corruption ensure this continues despite what the law may say) – and done on the very, very cheap.”

      I think its a bit over the top to claim that USA and Japan are the only two earthquake prone countries that don’t suffer from unregulated construction practices, bribery and corruption? New Zealand would be one country that springs to mind as having a similar standard of building codes and a regulated construction industry

    2. The reason concrete is used extensively in many parts of the world is that hurricanes are a more frequent problem than massive earthquakes. Wood frame construction fares very poorly in a hurricane.

  5. When I first heard about the earthquake this morning I remember my Mother telling me, in the 1960’s, that she thought we should expect to see more earthquakes here on the west coast after quakes in Japan. I’ll be paying attention now.

  6. This might sound like fringe science as I’m unaware of any geological study on the subject, but it’s a very basic thermodynamic principle: if something is heated up, it dilates.

    Seems to me quite an overlook to not consider what effects climate change can have on the planet’s crust. If its temperature goes up, its surface will experience expansion. Such expansion would be translated into more friction between the tectonic plates, volcanic activity and so on — precisely what we have been experienced in the past decade.

    1. Both, I would say.
      Obviously, poverty prevented Haiti from making the kind of earthquake preparations that Japan (or Chile, or even the Dominican Republic) made, but science still allowed the international community to get aid to Haiti almost immediately, making the impact much less than it would have otherwise been.

  7. More than “Are the New Zealand and Japan quakes related”, my question would be, is the erupting volcano Shinmoedake (other end of Japan, the one from YOU ONLY LIVE TWICE) directly related to this quake, or not really?

    Similar to the way that, around the time of the Haiti quake, that Caribbean volcano Monserrat was more active.

    1. Yeah, I’ve been thinking that. And Kilauea in Hawai’i, sort of in the middle between New Zealand, Japan and California is unusually active since last weekend. Had some minor quakes too.

Comments are closed.