Earthquake! Your Smartphone Could Give an Early Warning

Becky Oskin
Damage in northern Chile caused by the tsunami associated with the magnitude-8.2 earthquake that struck in April 2014.

Shaking a smartphone can help you pinpoint your parked car, discover a good diner and then pay for your meal. Now, scientists have shown that a jiggling, GPS-equipped phone could someday deliver an early warning of earthquakes.

Earthquake early warning systems depend on the time delay between two sets of seismic waves. The first set, called P waves, often delivers a sharp jolt. The most damaging shaking comes from slower surface waves, called S waves. For cities built far from powerful faults, distant seismic instruments stationed on the hazardous faults can detect the P waves and send an alert before the S waves arrive. The warning may be only a few seconds to a minute, but that's enough time to stop trains and surgeries and for people to find cover.

"A few seconds can be enormously helpful," said lead study author Sarah Minson, a geophysicist at the U.S. Geological Survey in Menlo Park, California.

For major earthquakes, those of magnitude-7 and larger, GPS sensors in smartphones can also provide up to several seconds of warning, according to a study published today (April 10) in the journal Science Advances.

The researchers tested the system with computer modeling, not in the real world, though they plan to launch a pilot project in Chile this year.

The GPS warning system is deceptively simple: If the GPS receivers from just a few phones suddenly lurched in one direction, that's probably not an earthquake. But if a few thousand phones all move at once, then the motion is likely from a sudden seismic jolt. (During a major earthquake, the earth may jump horizontally and/or vertically by several feet.) [The 10 Biggest Earthquakes in History]

The phones can determine the location and magnitude of the earthquake from the amount of surface movement and then send out a warning.

The computer models simulated what actual GPS data might look like from a magnitude-7 earthquake along California's Hayward fault, and also tested GPS data from the 2011 Tohoku earthquake in Japan. The team ran the models using the sensitivity of a Google Nexus 5 smartphone. The phone picked up geologic shifts as small as 0.5 inches (1 centimeter).

Minson hopes that if enough volunteers joined in, a smartphone network could provide an early warning in remote areas of the world where smartphones are popular but earthquake warnings are nonexistent. Scientists at the U.S Geological Survey are already monitoring Twitter for earthquake keywords, which provide early confirmation of their instrumental data.

"Probably the biggest impact would be in parts of the world that don't have networks and can't afford to build them," Minson said. "We have high hopes that this could make a huge impact, because the vast majority of the world has no form of early warning or even rapid response to an earthquake. What people do increasingly have are these devices."

However, there's one big hurdle before any smartphone warning system becomes a reality. The researchers need access to raw GPS data, which smartphone manufacturers currently prevent. The processing that provides a good GPS position to smartphone users also masks earthquake signals, Minson said. It's likely that both phone software and chip firmware would need to be reprogrammed for the earthquake early warning idea to work.

Also, at least a few hundred people will need to be near the earthquake's epicenter so their phones can send out alerts to others.

For the test project in Chile, the scientists will hack about 250 smartphones so they can get the GPS data they need.

The research team is also involved in the West Coast’s $145 million ShakeAlert earthquake early warning system, which uses both GPS and seismometers.

Follow Becky Oskin @beckyoskin. Follow Live Science @livescience, Facebook & Google+. Originally published on Live Science.

Copyright 2015 LiveScience, a Purch company. All rights reserved. This material may not be published, broadcast, rewritten or redistributed.