So, about that earthquake advisory...

At this point, it seems that the whole internet has heard the story about how a swarm of earthquake activity near the Salton Sea in Imperial County, California, has raised the chances of a large earthquake on the San Andreas Fault, and an earthquake advisory was issued detailing all of this. If many headlines and reactions to them are to believed, the Big One is imminent, and large swathes of southern California are about to bite the big one.

Well, the raised probability aspect is, in fact, true - or it is according to our best understanding of how faults interact with each other. But we’re talking a raise to a maximum of a 1% chance in the next seven days. One percent! Or less! I don’t think I’ve ever seen a 1% chance of something bad happening get magnified into It Is Imminent And About To Happen, but that’s how this story is increasingly being spun. It’s misleading at best, and harmful at worst, when one major initial purpose of the earthquake advisory was simply to acknowledge that scientists are paying attention to what’s going on down there instead of just ignoring it.

Faults are actually a lot like people. They undergo a lot of stress from day to day. For people, stress can come from a lot of things, but for faults, it mainly comes from the long-term motion of tectonic plates. People have a whole lot of different coping mechanisms to not break under the stress, but all faults share the same coping mechanism: friction. Just like you may have a breakdown when you can’t handle the stress in your life, faults break when the amount of stress they’ve accumulated is too much for friction to resist. That breaking process is an earthquake.

Most of the stress on a fault comes from plate motion, but other earthquakes on nearby faults can also add stress to the picture. The bigger the earthquake, the more stress it inflicts on everything around it. But, unlike stress accumulation from plate motions, the stress change from another earthquake is pretty instantaneous. It’s possible - and has happened - for an earthquake to set off another one within seconds, hours, or even a couple of days, but the longer you go since that earthquake, the more apparent it becomes that the stress change wasn’t enough to send any other faults into breakdown mode.

So - that increase in earthquake likelihood on the San Andreas, up to maybe as high as one percent last week, came from the stress changes from this earthquake swarm on top of the regular tectonic stresses. The odds are lowering again now that the swarm itself is slowing down.

I’d say not particularly, no, and I’m not alone among seismologists here.

Firstly - there are lots of swarms like this near the Salton Sea. They happen a couple of times per decade, and they tend to last for a few days. Swarms in this area in 2008 and 2012 had earthquakes with magnitudes in the low 5s; the largest one this time around has been a 4.3. These earthquakes all occur on faults that are not the San Andreas, and so far, there is zero historic precedent for one of these swarms directly preceding earthquake on the San Andreas, or any other fault. All of these swarms also raised stresses on the San Andreas, raised the odds of a larger quake to something like what they were last weak, just without a formal earthquake advisory issued - and nothing happened. This is enough precedent that, basically, if there IS a Big One soon, it was very close to happening anyway regardless of the current swarm. It’s entirely possible that a swarm event could eventually be the proverbial straw that broke the camel’s back, but there just isn’t enough energy involved in one of these swarms to bring the fault from Not Particularly Close to Big One. It won’t be the cause in and of itself.

Secondly - earthquakes that aren’t parts of swarms also affect stresses on nearby faults. The larger the earthquake, the more substantial the stress change, and the longer the period of increased hazard. Things like the 1992 M7.3 Landers or the 2010 M7.2 El Mayor Cucapah earthquakes kicked the San Andreas a lot harder than this week’s swarm of 4s did, but that didn’t set it off, either. If a similar earthquake advisories has been published after those, the percentages would’ve looked scarier than 1%. Earthquake scientists were actually pretty worried after Landers and its M6.4 aftershock near the San Andreas in Big Bear, and that’s not the case right now. We’re observing, but nobody’s particularly worried. Basically, faults affect each other all the time, it’s just that those effects are usually not the immediate triggering of another large earthquake.

Thirdly - the southern San Andreas is considered the highest hazard fault in the state already, but we’ve been living with it that way for a long time. The last earthquake on the part of the San Andreas was sometime in the late 1600s, and the average time between earthquakes on that section of the fault is more like 200 to 250 years. This is a section scientists have been saying is due for a while, which is what brings on the extra attention when there’s a swarm, and the extra nervousness. The most recent USGS statistical earthquake forecast - based on everything we know about every fault in the state - puts the odds of a M6.7 or greater on the southern San Andreas Fault at 19% within the next 30 years and the odds of a M7.5 or greater at 17.3%. That’s the baseline we’re living with, and this swarm hasn’t changed it by very much or for very long. And for another southern California point of comparison, your odds of getting into an accident on a Los Angeles freeway in any given year are higher than the yearly odds of a San Andreas earthquake, let alone your odds of being hurt in that earthquake.

This kind of thing shows that scientists acknowledge the swarm, acknowledge it’s in an area that has already had media attention for being high hazard, and acknowledge that people are nervous. (Which, hooboy, the advisory did fail as an effective acknowledgment of nerves, looking at the reaction it’s caused.) It’s also some insurance against anyone who can claim later that “nobody said anything about the risk” if this becomes the unprecedented case where the swarm does lead to a damaging earthquake. That was the case with a swarm in central Italy in 2009 which eventually lead to a M6.4. The L’Aquila earthquake killed a bunch of people, and the scientists involved ended up on trial for manslaughter for ostensibly not acknowledging the change in risk. The whole L’Aquila situation was a huge mess of bad communication and bad policy, and that part of Italy didn’t have the precedent of lots of swarms without a bigger quake like the Imperial Valley does, but hey. It happened. We want to avoid that here in California.

Another important goal of advisories like that is to remind people, yes we’re on an active plate boundary and we have earthquakes. Yes, preparedness kits are a good idea. A lot of folks seem to misread “check your emergency kit” as “you’re gonna need it really soon” though, which is…a problem that may be hard to work around. But if you can’t use small earthquakes as a motivator to prepare for big ones, that’s a problem, too. If people are only motivated to prepare in hindsight after getting wrecked by a large earthquake, that’s not an effective mitigation strategy. I absolutely always tell people to use small ones as a reminder to prepare for big ones, but media storms like this are just a sign that scientists need to watch our phrasing, too…

I have been meaning to reblog this since it appeared over the weekend. At least I remembered tonight. Share widely - seismologist commenting on the recent “Earthquake alerts” in California