Earthquakes cannot be predicted.
That right there is a fact you can take to the bank, and one that will remain good for probably a very long time. There are a lot of reasons why, but the main one is every earthquake starts the same, whether it’s a tiny one or it grows to be a big one. There’s really no way to tell which one will stay tiny and which one will grow to be a monster, and that’s probably not going to change anytime soon, if ever.
But we can forecast them and we can observe they follow some rules. If an earthquake has happened in a place before, it’ll likely happen again. If industry is injecting fluids (of any type, from plain water for geothermal to toxic waste to salt brine to frackfluid for disposal) into certain geologic strata, it’ll likely cause an earthquake. Drain a quarry? Could cause an earthquake. Fill a quarry? Same thing. Build a big reservoir, yep, can induce a quake too. There are so many mining earthquakes (because of the use of explosives) that USGS doesn't post them on their realtime maps. Incidentally, it is very, very easy to artificially create an earthquake if the conditions are right-- which they are over much of the Earth’s surface.
Beyond that, the best we can do is forecast and give probabilities and that won’t really change anytime soon, if ever.
But there’s an interesting new paper by Rebecca Bendick, from the University of Montana and Roger Bilham, of the University of Colorado, that suggests a synchronization.
Now, we only have good data since 1900 but even that 117-year record suggests, perhaps, large and mega-quakes come in pulses. It is pretty hard not to wonder, looking at the last 13 years, if there wasn’t some trigger that started the current pulse of subduction zone megaquakes, starting with Indonesia in 2004 and including Chile in 2010 and Japan in 2011, and Mexico’s M8.2 on September 7th. There was a similar pulse in the 1950s and 1960s, ending with Alaska’s 1964 monster, and there was perhaps another pulse at the beginning of the recording period 117 years ago.
Subduction zone giant quakes were not necessarily what Bendick and Bilham were looking at although I’m aware people are looking into whether there is something there about subduction zone earthquakes. What they were looking at was the Earth’s spin.
You’ll never notice it but the Earth’s spin isn’t exactly even, probably due to the core of our planet which is also rotating. There’s a lag of about five to six years where the spin speeds up and slows down. This year, it began its slowdown.
Remember how easy it is for us to create an earthquake? That’s because the Earth’s surface is constantly under stress. Now with the core putting on its brakes---there ya go. More strain transferred up into the crust, meaning perhaps more earthquakes of magnitude 7 and up, and since we know where M7s happened in the past, we can infer those regions will get wacked in the future.
Dr. Erik Klemetti explains:
Although we never notice it, the Earth’s spin does change ever so slightly. Geologists aren’t sure what drives this slight variation, but it is likely linked to the Earth’s core. The outer core is a layer of molten iron and nickel that is 2,200 kilometers thick and the whole thing is convecting (moving). This is what drives the dynamo that generates the Earth’s magnetic field, so you should remember to send that “thank you” note to the outer core. However, small changes in that convection in the mantle could be reflected in slight changes in the rate the Earth spins as well.
Just like you might feel that seatbelt pull you in if you hit the brakes (thank you, momentum), if the Earth’s spin slows even a little bit, there is a lot of rock with a lot of momentum that slows with it. You can think of it like the whole planet “sloshing” a little bit. That causes a lot of stress to rocks … and stress added to rocks can lead to earthquakes.
However, Bilham and Bendick say that they have found examples where the slight deceleration of the Earth’s spin take awhile to propagate through the Earth’s rocks. So, after that small slowing of the Earth’s spin happens, there is a 5-6 year lag before they identify an increase of up to 30% in big earthquakes (over magnitude 7) between 10-30º latitude N and S. This increase could last up to 5 years, but the duration is unclear based on the current data.
Good idea to refresh those earthquake plans, if you live in an earthquake-prone area.
(Addendum, Dr. Roger Bilham wrote this in response to other hysterical reporting last year. Many of the articles written about this paper were pretty bad, thus the reason I did not link to them. Again, this is not a prediction. We cannot predict earthquakes. But one should always have their emergency plans ready to go at any time.)