Sendai/Tohoku earthquake round-up

A post by Chris RowanIt’s hardly surprising that my browsing this week has been focussed largely on the aftermath of the earthquake and tsunami in Japan (which is now officially being referred to as the Tohuku earthquake, rather than the Sendai earthquake; I’d complain this was harder to pronounce but I’m hardly in a position to on this blog). Given this, in place of our normal Twitter links round up, we have an earthquake link-fest instead.


  • For those who are still trying to understand exactly how the plate boundaries that intersect near Japan are structured, this figure, which I put together for my post on the Scientific American guest blog, may be helpful.

    The location of Friday's earthquake, with respect to the numerous plate boundaries that intersect near Japan. Base map generated by GeoMapApp (

  • What are the forces at work at subduction zones that turn constant and relatively smooth plate motions into the jerky stick-and-slip behaviour of earthquake-generating faults? It’s all about friction, as Matt Kuchta demonstrates with an ‘Earthquake Machine’ constructed from a brick, spring and sandpaper.
  • As Dave Petley argues at The Landslide Blog, the Tohuku earthquake is entirely consistent with our understanding of how subduction zones work. However, the size of this particular earthquake does seem to have been a little beyond what was expected from this particular subduction zone. Old, cold oceanic crust, such as the Pacific plate being subducted beneath north Honshu, was thought to lead to smaller megathrust earthquakes than younger, more buoyant oceanic crust (such as what is being thrust beneath Cascadia, for example) because the lower buoyancy would lead to reduced friction at the plate boundary. Clearly, this is not the case, and may lead to us having to rethink the seismic hazards at “safer” subduction zones.
  • Sumatra, Chile, and now Japan, all in less than a decade: could there be a causal link between large megathrust events around the world? New Scientist explores the idea that great earthquakes could beget more great earthquakes. Sadly, it is very hard to do meaningful statistics on the small numbers and timescales represented by our instrumental records. Perhaps paleotsunami research can eventually provide some insight into this question.
  • We can’t predict earthquakes: any claims to the contrary either involve “precursors” that throw up more false positives than hits, or predictions so general that they can hardly fail to be “fulfilled”. Progress, however, is being made in earthquake forecasting – identifying areas that are particularly at risk from damaging earthquakes, even if one can’t say exactly when.



  • Below are updates of the earthquake plots that I produced for my original post, and my discussion of the aftershocks. So far there have been 336 earthquakes of greater than magnitude 5 following the main shock.
  • Magnitude of earthquakes (M5-6=small yellow circles, M6-7 orange circles, M7+ large red circles) off the coast of Honshu, 9-20 March.

  • The number of aftershocks continues to decline approximately according to the inverse of the time since the earthquake, as expected.
  • Frequency of >M5 aftershocks following the main shock on March 11. Black dashed line: 'predicted' number of >M 5 aftershocks if their frequency is inversely proportional to the time since the main shock. Red dashed line: 'predicted' frequency for >M 6 aftershocks.

  • Jascha Polet has also produced some really nice visualisations showing the distribution of the aftershocks in space and time. Contrary to some reports, the aftershocks are not currently showing any migratory trend, seemingly being randomly distributed along the length of the rupture.
  • For an animation of the earthquake sequence, Paul Nicholl’s Japan Quake Map is the place to go.


Categories: earthquakes, geohazards, links
Tags: , , , ,

Comments (7)

  1. PeakVT says:

    The link to the BBC video points back to this post.

    The video from Crescent City is nice, but it doesn’t seem like such a great idea for someone to have stood there to take it.

    • Chris Rowan says:

      Fixed the link – thanks.

      It seems that the natural response of many in the Western US to a tsunami alert is to go down to the beach to have a look – which is a definite concern.

  2. Lyle says:

    On the video of displacements the wave of vertical displacements that propagates south from the epicentral area is interesting. It will be interesting when all the strong motion records from the southern part of Japan are recorded to see the effects of this. The displacements imply that there might have been some pretty big acceleration events in that region.

  3. Nina says:

    Thanks, Chris, for organizing so much information. An excellent reference.

  4. Monika Landy-Gyebnar says:

    Chris, thank you a lot for the superb info on the quake! I have a question now regarding the aftershock tendencies. Yesterday (21.03.) there were quite small amount of them but today (22.03.) there were much more and even stronger ones. So today seems to overcome the decreasing tendencies. Is this normal to have such jumps in the aftershocks or might this be a sign of a forthcoming increase in them? Thank you for your reply in advance!

    • Chris Rowan says:

      The time-inverse decay of aftershock frequency is based on probabilities, which means there is always going to be some variation – some days you might get more, sometimes you might get less, than a smooth predictive curve. I’d suspect that we’re just seeing some of that spiky-ness.

      Location is also important; if you get more aftershocks closer to populated areas then you might perceive an increase in activity, even when the aggregate behaviour over the whole area doesn’t change much.

  5. Monika Landy-Gyebnar says:

    Thank you a lot, Chris!