25 August 2011


Posted by Callan Bentley

Since Tuesday’s big earthquake, we’ve had 5* aftershocks in the same area (and possibly on the same fault). The most recent one popped off last night at 1am. Here’s a plot showing the size of the events (moment magnitude) relative to the passing of time:

Note that the quakes that came after “the big one” are smaller in their size (the amount of energy that they release into the surrounding rock). We call these “aftershocks” – they are smaller-scale readjustments in the rocks immediately adjacent to the main rupture (including further along on the same fault, potentially) as those rocks shift to come to equilibrium under the newly-modified stress regime.

Notice that the aftershocks are smaller, and they get to be less and less common. Chris Rowan explained this lucidly on Highly Allochthonous in the aftermath of the big Sendai (Tohoku) earthquake in Japan earlier this year. A key point is that while we cannot predict the exact timing or place of any earthquake, including aftershocks, we can make broad, general statements about the way a population of aftershocks looks, given the vantage provided by the passage of time. They decrease in severity and in frequency as time goes by.

Okay, but what if that wasn’t how things had actually played out? What if we had a slightly different outcome?

What if, at 1am this morning, we didn’t have a 4.5 quake, but instead had a 7.6?

How does this hypothetical scenario change things? In terms of our language, our descriptors for these same events needs to change, given the new information about which quake is really the biggest in the series. We must re-evaluate them in the new context that is provided by this huge event (7.6) that is way bigger than the one (5.8) that started things off on Tuesday. Now Tuesday’s quake and its “aftershocks” are all in the shadow of this much larger event. Their relative position has changed. Though they are the same events, they have been demoted in relative rank.

If my hypothetical scenario were to have played out, the events of Tuesday and Wednesday would now be seen as precursors which loaded additional stress on a more-critically-stressed section of the fault (or an adjacent fault), and so they should be called “foreshocks” instead. Again, there is no way of knowing whether this could be the case until enough time has passed that we can look back and see the pattern of the quakes in aggregate.

Compare these two graphs (one real, one totally fake) to the Japanese quake data that Chris graphed back in March. You’ll see the patterns are broadly similar.

Whether Virginia’s earthquakes this week represent a 5.8 and aftershocks or a series of foreshocks (including a 5.8) leading up to something bigger is impossible to know at this time, while it’s all still playing out. Come back in a year, and it will be more obvious which they were. That being said, Virginia’s tectonic situation is far from any plate boundary, in particular from a subduction zone (where the largest magnitude earthquakes typically occur). So I’m not losing sleep worrying about whether this week’s activity presages some big one, only acknowledging that it is a rational possibility (remember Charleston in 1886, a 7.3 rupture), and that the relevant terminology would change if such a thing were to happen here in Virginia.


* Drat! They just posted another one. Oh, well, I’m not going back to these graphs to update them. Deal with it.