A research team based in California at the U.S. Geological Survey’s Menlo Park Science Center has successfully built and tested a system in Chile that uses lightly modified smartphones and is capable of detecting earthquakes as small as magnitude 5.
“The biggest surprise was how sensitive the devices are,” says Benjamin Brooks, Ph.D., who is a research scientist with the U.S. Geological Survey and is leading the project, which is being funded by the United States Agency for International Development. “We’ve actually been detecting earthquakes a lot smaller than we had designed for. We were designing for magnitude seven and above. It turns out we are detecting magnitude five earthquakes.”
The system uses the coarse/acquisition codes generated by the phone’s Global Positioning System receivers, and a special chip provides access to the raw data. Each phone has a small photovoltaic panel for recharging. The phone is placed in a box that is then mounted on the roof of a building. Each unit costs between $300 and $350.
The team first confirmed the concept by using well-tested simulation data from an earthquake on the Hayward Fault in the San Francisco Bay Area. Next, they took actual coarse/acquisition code data from the Tohoku earthquake, which had a magnitude of 9.1 and struck northeastern Japan in March 2011, and ran a simulation. The results indicated that the system would have triggered an alert 77 seconds after the event began.
“That was 23 seconds before the S wave hit Tokyo,” Brooks says. “That would’ve been more than five minutes before the tsunami hit. Five minutes is a lot of time, especially when you think about populated regions where vertical evacuation is often the mitigating strategy. Rather than just running [toward a hill], you can just walk up the steps.”
“It’s a warning network. It’s good enough for warnings,” Brooks says. “But it’s not good enough for doing the amazing science that seismologists have been doing for years. The value of scientific-grade networks is unquestioned for understanding earthquakes.”
Eventually, an early warning system of cellphones could tap into the power of the so-called Internet of things, a network formed by linking computers in cars, appliances, smart buildings, and other items. The promise of such a network is widespread data sharing from a vast number of sources, which would eventually cut operational costs by a significant amount. “So, we envision the same thing for earthquake warnings, that your operational costs effectively go away,” Brooks says. “But there are numerous issues to be worked out.”
One of those issues is population density. In Chile, for instance, where the Centro Sismológico Nacional, part of the Universidad de Chile, is a partner in the project, there are large areas in which residents face significant seismic risk, but the population density in these areas is so low that crowdsourced data wouldn’t be available in large enough quantities to provide a reliable result. In such areas, Brooks says, special cellphones would have to be put in place.
In the meantime, the network of deployed cellphones in Chile is growing as the project continues. “What we know well is that we can detect events that matter. What remains to be learned is how quickly we can get the warnings to people and, frankly, how people react to certain types or grades of warnings,” Brooks says. “That’s the next step.”
Brooks says this type of system would mostly likely be deployed in countries that face a significant seismic risk but don’t have the resources to deploy a more sophisticated network.