The death of thousands of people following Monday's earthquake in Sichuan, China, has once again called attention to the need for technology that provides better warning of such devastating natural events. While China lacks an earthquake early-warning system, it is hardly unique. Such systems are still very much works in progress: Japan, Mexico, and Taiwan are deploying them, but most other nations, including the United States, are still in the research stages, says a leading seismologist, Haroo Kanamori.
Earthquake alert systems furnish only a few seconds of warning, but that can be enough time for people to take protective measures. And more research and investment would improve the systems' speed and the efficiency with which the warnings are distributed, Kanamori says.
The principle behind earthquake early warnings is fairly straightforward. Earthquakes begin with tremors called P-waves that have short wavelengths and generally do little damage; they may not even be felt by humans. Seconds later--the exact number of seconds depends on distance from the quake's epicenter---longer-wavelength tremors known as S-waves arrive and inflict damage as they shake buildings and landmasses. Early-warning systems consist of seismographs that detect the P-waves and send out electronic alerts. Networks of such sensors improve the accuracy of the systems.
In general, such systems are in their infancy, says Kanamori, a Caltech professor emeritus of geophysics and former director of Caltech's seismological laboratory. "It's really exploratory right now, but this is the only way at this moment to make some kind of prediction of what is going to happen in the short term," he says.
Japan has the most advanced earthquake early-warning system in the world; last October, it launched a system that delivers earthquake warnings via mass media, including television and the Internet. Japan had long used a system specifically geared to notifying and slowing down its bullet trains; the new system is intended to enable countermeasures by the general public, including clearing out elevators, taking cover under desks, or moving away from rocky walls. A few seconds' warning could also save lives on construction sites, allowing, for example, workers to grip girders or move cranes to safer positions. Taiwan and Mexico are developing similar systems; seven or eight other countries are researching them, says Kanamori.
One problem with early-warning systems is that they require many sensors distributed over a wide geographic area; the greater the number of sensors, the more precise the calculation of the quake's epicenter and magnitude, and the earlier the warning. To fill in gaps left by conventional seismic instruments, some researchers have proposed exploiting sensors inside laptop computers to allow them to serve as distributed P-wave sensors. While laptop sensors are not very sensitive, their great numbers may have value, Kanamori says. "It is a matter of engineering, how you put together the laptops. If you have a large quantity, maybe it can be helpful."
Had a warning system been in place in China before Monday's earthquake, the benefits may have been blunted by the poor quality of some of the buildings that collapsed. "There is no early-warning system implemented there, but even if it was, I don't know how effective it might be in that circumstance, when you have so many people living in bad buildings," Kanamori says.
Even a sophisticated system like Japan's is, of course, limited by the short warning time. For example, in 2004, a 6.5-magnitude earthquake struck in central Japan. A bullet train was running in the area, and thanks to the train-warning system, it shut off its power and hit the brakes. But the quake struck the tracks three seconds later, before the train had lost much velocity. It derailed anyway, the first such derailment in the bullet train's 40-year history.
Several European countries and the United States are conducting research on earthquake early-warning systems. But while a group of universities in California has developed and tested a warning network, no governmental system is in place to alert the public. "There hasn't been a major seismic event, and that is very lucky," Kanamori says. "But without that, there is no real incentive to start doing" public warnings.