The epicenter of a major earthquake is someplace you really don’t want to be. Not only does the shaking tend to be most violent there, but there’s also no time to look for shelter. Even a few seconds of warning can make the difference between life and death. And now, scientists say, there's an app for that.
Writing in the journal Science Advances, Sarah Minson of the U.S. Geological Survey (USGS) and California Institute of Technology and several colleagues claim that plain old smartphones can detect quakes, pinpoint their location, and calculate when the shaking will arrive.
The USGS is spending tens of millions of dollars developing an early warning system called ShakeAlert that will use texts and emails to alert people that a quake has just let loose.
But some earthquake-prone nations—in Central and South America, for example, or the Caribbean, or south Asia—don’t have tens of millions of dollars to spare, or networks of seismographs. That’s why Minson and the USGS Earthquake Hazards Program are working on a much cheaper alternative. “This is a really clever idea that leverages an existing resource," says Kristine Larson, a University of Colorado aerospace engineer who studies how GPS systems can be useful in the earth sciences. Larson wasn’t involved in this research.
Smartphones could be ideal for rapid earthquake warning, Minson and her colleagues realized, because they have built-in GPS sensors and accelerometers, which detect movement. That's how they know where they are at all times, and when they’re being shaken. Phones closer to the epicenter will start shaking first, followed by phones a little farther away, so the timing and the intensity—which drops off with distance—allows the system to calculate where and how powerful the quake is.
The phones are also part of a pre-existing data network, so it’s trivially easy to program them to send news of a quake to a central location, where the data are merged to form an overall picture of the unfolding disaster. And that same network can blast an alert out to anyone in the danger zone.
The phone signals and processing happen at nearly light speed, but it’s reasonable to wonder if even that is fast enough. If your phone is already shaking, it’s arguably too late to do anything about it.
With even a few seconds of warning, you can get under your desk, or pull your car off the road.
But in fact, says Minson, an earthquake creates two distinct types of waves through the solid earth. “The damaging shaking comes from S waves,” she says, which shake the ground back and forth. A quake also sends out P waves, which simply compress the earth in front of them as they go—and, says Minson, “P waves travel faster, so they get to you first. It’s like the midnight ride of Paul Revere,” she says. “He began riding after the British landed, but he was far enough ahead of them to warn people.”
When Seconds Matter
It’s the P waves that earthquake early warning systems pick up on, and while the dangerous S waves can arrive very quickly afterward, that can be enough time to take action. “With even a few seconds of warning,” says Minson, “you can get under your desk, or pull your car off the road. If you’re a surgeon,” she says, alluding to a scenario with particularly gruesome potential, “it’s long enough to retract your scalpel and step away from the patient.” It’s also long enough for subways and other trains to be brought to a halt and elevators to be stopped at, rather than between, floors.
The authors of the new report tested the idea with two simulations—a hypothetical magnitude-7 quake on California’s Hayward Fault, in the Bay Area, and another based on the devastating magnitude-9 quake that struck off the shore of Japan in 2011, triggering a killer tsunami. The simulations show that a system relying on just two-tenths of a percent of smartphone users could pinpoint the strength and location of a quake within just five seconds.
“In an ideal world,” says Minson, “we would test the system with actual phones and actual earthquakes.” And in fact, that’s already in process: A pilot program funded by the U.S. Agency for International Development should be in place within the year to do just that.
There are a few potential pitfalls, observes John Vidale, a seismologist at the University of Washington who wasn’t involved in this project. “A system like this would not necessarily be free of false alarms,” he says, “nor immune to pranksters or worse who wanted to cause turmoil from malicious hoaxes.”
Still, he says, “The idea is creative and sound, and we could see crude versions of such apps very soon."
Follow Michael D. Lemonick on Twitter.