Skip to main content

<i>Science Advances</i>: Crowdsourcing System Can Detect Earthquakes

Smartphone data may be used in the future to send out early warnings about major earthquakes such as the 2011 Tohoku quake and tsunami in Japan. | William Saito/ CC-BY-NC-ND 2.0

The GPS sensors in smartphones could be used as part of a crowdsourcing system to issue early earthquake and tsunami warnings, scientists report in the 10 April issue of Science Advances. Combining data from hundreds of smartphones could lower the cost of earthquake early detection methods and potentially reduce deaths caused by medium-to-large-sized earthquakes.

While previous studies have investigated earthquake early warning systems based on scientific-grade GPS data, most regions at risk for earthquakes don't have the infrastructure needed to monitor earthquakes at this level. Consumer-quality GPS data collected by smartphones are less accurate than data collected by scientific-grade instruments, but could be used as a cheaper and easier way to issue warnings in areas without scientific-grade coverage.

"Most of the world does not receive earthquake warnings mainly due to the cost of building the necessary scientific monitoring networks," said Ben Brooks, co-author of the study and a U.S. Geological Survey (USGS) research geophysicist.

Brooks, along with USGS colleague Sarah Minson and others show that a smartphone's ability to take continuous measurements of its location with GPS can be harnessed to detect and measure the large ground movement that occurs during an earthquake. This information could be sent to a central server that could rapidly estimate earthquake magnitude. If enough smartphones were triggered by unusually large ground motion, a warning could be issued a few seconds before the strongest seismic waves from the earthquake began.

The number of smartphones needed to trigger such a warning would depend on the type of the smartphone, the size of the earthquake, and the number of phones needed to minimize false alarms, the authors say. In their simulations, the authors chose to declare an event after a hundred devices were triggered.

The researchers performed two simulations to test the ability of consumer-grade GPS sensors to locate earthquakes. One simulation relied on a standard earthquake model of the Hayward fault zone, which runs through the San Francisco Bay area, and the other used real data from the 2011 Tohoku earthquake in Japan, one of the largest earthquakes ever measured at magnitude 9.0.

In the Tohoku earthquake example, the researchers used raw GPS data called C/A-code (the same kind of data that would come from smartphones) to show that earthquake early warnings could have been issued before the seismic waves from the offshore event reached Tokyo, and that a tsunami warning could have been issued before any inundation began along coastal Japan.

For the Hayward fault simulations, the authors show that GPS data combined with a technology called SBAS (satellite-based augmentation system) could be used to issue warnings five seconds before strong seismic waves from a magnitude 7 earthquake reached major population centers such as San Francisco and San Jose.

However, in order for a crowdsourced warning system to become a reality, the authors say, smartphone manufacturers would have to allow consumer smartphones to access C/A-code data — a minor software fix. The researchers also envision a mobile app that people could use to opt-in to the detection network.

"In our paper we discuss smartphones because they are so prevalent. But this is really more about harnessing the observational power of any type of connected device with positioning information," said Brooks.