Although the catastrophic tsunami that swept across the Indian and eastern Pacific oceans at the end of 2004 killed as many as 200 000 people and prompted wide calls for improving warning systems, those systems continue to suffer from serious shortcomings. Last July, when another tsunami hit the Indonesian island of Java, again there was too little warning, and in one coastal area at least 600 individuals died. Unfortunately, two warning buoys in the immediate area where the tsunami had originated came unmoored—but even if they had been working properly, the system in place probably would have provided adequate alerts only to people living far from the tsunami’s origin.

Could a system relying on signaling between Global Positioning System satellites and ground stations provide prompter warnings? A group of scientists led by space geodesist Geoffrey Blewitt say they have developed a concept for such a system and that it could detect deadly tsunamis in as little as 15 minutes.

Their system works by ­measuring GPS satellite radio signals as they are picked up during an undersea ­earthquake by GPS ground stations positioned around the globe [see diagram, “GPS Displacement,” below]. Such earthquakes happen when two opposing tectonic plates on the sea floor push against each other until one plate slips suddenly beneath the other. The upper plate then lifts the ocean upward, ­generating a tsunami.

Illustration: Matt Zang

GPS displacement: By accurately knowing the location of Global Positioning System satellites, scientists can measure how far GPS ground stations move during an earthquake. This allows them to calculate how big the earthquake is and how large a tsunami might be produced by it.

Blewitt and his team at the Nevada Bureau of Mines and Geology and at the University of Nevada, Reno, got the idea of using GPS signals after analyzing GPS measurements from the 26 December 2004 killer tsunami. The data showed that 2000 kilometers from the originating earthquake’s epicenter, the ground had shifted about 1 centimeter.

“By knowing the exact locations of the GPS satellites, you could measure how much the GPS ground stations had moved relative to the earthquake,” says Blewitt. With his proposal, the data from the GPS ground stations would be used to gauge the precise strength of the earthquake, and that in turn would help predict the size of a resulting tsunami.

Earthquake magnitude assessments from seismometers can take several hours to calculate accurately for the largest temblors. For example, the underwater earthquake that caused the 2004 tsunami was initially estimated to be about a magnitude of 8.0 on the Richter scale and was thought to be too small to generate a large tsunami. However, after nearly 5 hours had passed, refined calculations showed that the true size of the earthquake was about magnitude 9. By the time this information was distributed, it was too late to warn people in the endangered area, which stretched from Indonesia all the way to East Africa.

To be sure, even if Blewitt’s GPS displacement approach proves to be the best way of improving warning times, implementing the necessary ­infrastructure, along with proper ­testing of the method, cannot be done ­overnight. “It might take a couple of years to ­perform ­further research on the ­technique, using ­simulated data from models of ­previous great earthquakes, and to develop the interface to real-time tsunami ­models currently being developed by [the National Oceanic & Atmospheric Administration],” says Blewitt.

What’s more, a minimum of five GPS satellites would be required to be in view of the ground station to accurately and swiftly predict the size of an earthquake, though in practice this condition is almost always satisfied, according to Blewitt. And while new GPS ground stations cost about US $10 000 each, there are already ­stations in place that could contribute. “I’m no expert at estimating system costs,” Blewitt says, “but altogether the cost should be very small in comparison to what already exists in the tsunami warning systems, which include remote ocean buoys that are relatively costly to deploy and operate.”

Richard Anthes, president of the University Corporation for Atmospheric Research, in Boulder, Colo., pioneered the use of GPS satellites to map atmospheric temperature and humidity [see “Roundabout Way of Profiling Earth’s Atmosphere,” IEEE Spectrum, News, April]. He says that the tsunami sensor “is just another one of the many uses of GPS to sense the earth and use the information for scientific studies and applications of great benefit to people.”

But not everyone thinks Blewitt’s concept, as formulated, is practical. Yehuda Bock, a research geodesist at the Scripps Institution of Oceanography, in La Jolla, Calif., who has been working on a similar idea, says: “To put it in perspective, a large earthquake in the Pacific Northwest could generate a ­tsunami that would hit Vancouver Island within 15 minutes—not much time to gather and process data from the global GPS network.”