Wait, I Thought We Couldn’t Predict Earthquakes
Oh, right, we can't
Last week, the headlines were particularly frightening. ‘Massive Earthquake Along the San Andreas Fault Is Disturbingly Imminent‘, ‘Risk of big earthquake on San Andreas fault rises after quake swarm at Salton Sea‘, all related to a swarm of earthquakes in the Salton Sea that researchers feared might trigger an epically large earthquake in the area. That earthquake hasn’t materialized (yet) but it raises the question: how well can we actually predict earthquakes?
The swarm in an area known as the Brawley Seismic Zone featured over 96 earthquakes in four days, in an area right near a portion of the infamous San Andreas fault. That particular section of the San Andreas hasn’t moved in about 330 years, which researchers say makes it long overdue for a major quake. Last week, the U.S. Geological Survey stated that the possibility of a 7.0 or larger earthquake in the area was raised to between one and 3,000 and one in 100, much higher than the background risk of about one in 10,000.
The increase in risk triggered an earthquake warning from the California Office of Emergency Services, which led to San Bernardino shutting down its earthquake-vulnerable City Hall , and a lot of anxiety from residents and visitors alike.
Why Was There So Much Concern?
Even though a major earthquake didn’t materialize this time, given the public response, it’s worth looking into why the researchers thought that there was a heightened risk.
Ken Hudnut is a geophysicist and the Science Advisor for Risk Reduction for the USGS, and has been studying the area for decades, in particular a 1987 earthquake sequence, known as the Elmore Ranch-Superstition Hills sequence, which was one of the influences for raising the risk of a big earthquake happening in the area last week.
In that case, a quake along a smaller fault happened first, shaking the area. Then, 11 hours later, the larger Superstition Hills fault released, causing a much larger earthquake. The smaller fault was perpendicular to the Superstition Hills fault, a structure known as a cross fault. The sequence of the two faults located so close to each other both in time and place, led researchers like Hudnut to predict that the smaller earthquake had led into the larger one.
“It was an understandable pre-shock, and it made sense. Then, we applied that knowledge to the future and we said ‘If in the future if we have a cross fault earthquake, that might unclamp the San Andreas fault, and you could have a similar pre-shock with a delay’” Hudnut says.
Because of that, when a swarm occurs on a cross fault of the San Andreas, like it did last week, seismologists sit up and take notice, even when the swarms are much smaller than the quake that set off the 1987 quake. They also were paying close attention in November of 2001 and March of 2009, when similar swarms struck the area. But no large quake has been born out of the shaking swarms.
“We’re seeing that kind of behavior repeatedly without escalation to a bigger event on the San Andreas. In a way, we’re becoming a little less concerned each time now because we’ve seen it back off three times.” Hudnut said.
“It’s not like we’re on the brink of being able to actually predict earthquakes.” Hudnut says. Despite decades of research, seismologists are still in the very beginning stages of understanding earthquake processes. They’re still looking at case studies like the 1987 quake, gathering data, observing, and doing their best to understand the complicated physics of major earthquakes, but it isn’t enough to be able to forecast earthquakes like you would a hurricane.
“It’s not like we can say for sure something more is going to happen, this sequence is going to escalate, or not. Unfortunately, we don’t have that ability. We don’t have that crystal ball all polished up and ready to see the future” Hudnut says.
Just because the huge quake didn’t happen right after the swarms doesn’t mean that it isn’t going to happen at all. It could happen tomorrow. It could happen right after we publish this article. It could happen several years from now. But the bottom line is that we still don’t know exactly when ‘The Big One’ will hit.
Earthquake Warning Systems vs Earthquake Prediction
While panic is never productive, it probably made a lot of sense for the Southern California region to stay alert, and take the opportunity to prepare for a large quake while the risk was raised. While the risk has since declined to almost normal levels, being prepared for an earthquake is never a bad idea in Southern California. There is always a chance of a large earthquake happening in this area, even on a normal day without a recent swarm.
Precautions to take vary depending on whether or not you’re at the individual, community, or city level. On an individual level, the best way to prepare for an earthquake is to participate in drills, and make sure that your home is earthquake ready, with enough supplies to last all family members a few days. At a community level, Hudnut says that the recommendations are to talk with your neighbors and have a plan to check in with each other if an earthquake occurs. The city and state levels get much more complicated, and involve retrofitting infrastructure like buildings, roads, and water and electricity supplies to withstand severe shaking.
Because earthquakes can happen so suddenly, and can be so devastating to human communities, researchers are actually moving away from trying to predict earthquakes and towards preparing for them.
“Most empirical, observational, trying to detect precursors, that kind of work has really died out, at least in this country.” Hudnut says. Instead, investments have gone into sophisticated seismic sensors and a GPS network that can track movement of the plates. The seismic sensors measure vibrations in the ground, and the GPS measures the location of the Earth as it moves.
“We’re processing huge amounts of data in real time, all the time. As soon as something happens we’re on top of it.” Hudnut says. “That’s really where we’re going with the technology right now, is towards having a fully functional earthquake early warning system that’s capable of observing earthquake ruptures as they grow in real time.”
Unlike earthquake prediction, earthquake early warning systems are considered to be an achievable goal. The heavy tremors of an earthquake moving through the rock and soil and debris of the Earth’s crust travel more slowly than text messages or other communications. If one of the sensors that Hudnut and his colleagues monitor notices a large quake, it could in theory give people living miles away a precious few seconds to drop, cover, and hold on.
Earthquake early warning systems are catching hold in Japan, and last week, California Governor Jerry Brown signed legislation that directed funding towards California’s nascent earthquake early warning system.
When a large earthquake strikes California, it is extremely unlikely that we we would know about it a week in advance, but in the future, California residents will at least be prepared, and might even have a few seconds of warning to take cover before the big one arrives.