Because of the recent news, these Friday links will deal with earthquake prediction. As in the L’Aquila case and in the Judgment Day “prophecy”, a reliable prediction is not possible until now. Repeat: NOT POSSIBLE! Of course this situation is far from being satisfactory. So, what do we know about future earthquakes?
1) The location
Okay, we only know the approximate location of most future earthquakes. We know about the seismic active areas, the plate boundaries, active volcanoes, and most faults. Most earthquakes will happen there. Our information is based on recent seismicity (c. 100 yrs), historical reports (several thousands of years), archeoseismology (several thousands of years), paleoseismology (several tens of thousands of years) and neotectonic studies. We do not know all locations. Some faults have very long recurrence intervals and we have no idea that they exist. Or we know they exist, but consider them inactive. All in all, not too bad.
2) Date of earthquake occurrence
Nope. No exact date. Sorry. We cannot predict earthquakes, we can only forecast. We know about the recurrence intervals of some faults, which gives a statistic value only. We try to estimate how likely an earthquake is, mainly based on earthquake frequency and latest event. However, there are a lot of factors that play a role: What do neighbouring faults do? If they are activated, does this increase or decrease the probability for our fault to create an EQ? What about the stress? Is there a connection to volcanoes, other nearby earthquakes, strong earthquakes far away, rain, snow, glaciers, tides, mass movements, high pressure areas, the filling of reservoirs, nuclear test sites …? Faults can be born and die as well. Of course it is no problem to predict an earthquake in Greece tomorrow. You will be right with this. But that’s not useful, there are dozens of quakes daily. Where exactly, come on?
3) Magnitude
Puh, even harder. We know about the potential of many faults (well, we think we know the maximum possible magnitude…). We can say: This fault is capable of a M7.0 event. But will the next event be that strong? No idea. There are also enough examples of underestimated faults.
4) Intensities
That’s complicated. You’d need to know about exact location, depth, magnitude and (as if this wouldn’t be enough) the geology of the surroundings. Additionally, it might be useful to know some facts about the fault mechanics, rupture propagation etc. Really, really hard.
So, a precise earthquake prediction seems not to be possible today. Estimation, statistics – okay. But remember that a real prediction needs to work in two ways: You have to be able to say where, when and how an earthquake will occur. The other way round, you must know where and when no earthquake will occur.
As Kris Vanneste said, let toads do this job: Toads can ‘predict earthquakes’ and seismic activity.
Or trust the other guys:
I would really love to hear your opinion about that. I might be wrong or too pessimistic. What do you think about earthquake prediction? Which papers can you recommend?
Have a nice weekend!
parclair | 2011-05-27|19:17 (UTC)
Thank you for writing this, it’s been recommended at a number of sites I visit. I like the tone, as well. I’ll be sending this link off to a number of people.
I recommend the Susan Elizabeth Hough books. She’s a geologist with the USGS, and writes about earthquakes. I’m currently reading “Earthshaking Science: What We Know (and Don’t Know) about Earthquakes”. I’m finding it very informative.