Preventing disaster
An earthquake expert who saw the warning signs
Phil Cummins heard about the Sumatran tsunami in much the same way as the rest of the unaffected world: by flipping on the TV. But when the 44-year-old American seismologist first saw the reports at his Canberra, Australia, home on December 26, his sinking recognition may well have been unique. "As soon as they said 'tsunami, Sumatra,' " he recalls, "I knew exactly what it was."
Come again? The Indian Ocean, we were told, was the last place you'd expect a disaster like this. Some 90 percent of tsunamis--whether kicked off by underwater earthquakes, volcanoes, or landslides--occur in the Pacific. But Cummins, who works for the Australian geological agency, had recently stumbled upon evidence that a Sumatran tsunami was all but inevitable. And he'd begun making the case to expand the international tsunami warning system to the region--a move that could have saved tens of thousands of lives.
Hindsight. Along with Australian and Indonesian colleagues, Cummins first raised the idea of a warning system for the Indian Ocean at an October 2003 meeting of the United Nations body, known as ITSU, that coordinates tsunami warnings for the Pacific. That proposal was based on relatively modest tsunamis that originated near Java in 1977 and 1994, Cummins says, and it wasn't until later that he discovered the potential for a much more severe event near Sumatra. "In retrospect, maybe I should have pursued it more urgently," Cummins says. "But even I wasn't 100 percent convinced, and I had no way of knowing that it was going to be now rather than 200 years from now."
That initial proposal for a warning system was met with interest, Cummins says, and he joined a working group set up to study the idea. "I volunteered to start looking at general hazards for the Indian Ocean" beyond Java, says Cummins. "And that's when I started seeing these signs" that the subduction zone that set off the December 26 tsunami had the potential to cause a major disaster. Surveying scientific papers, he noticed that conditions along the Sunda Arc--where a relatively young chunk of the Earth's crust is squeezing itself beneath the larger Eurasian plate--closely mirrored known sources of major tsunamis in the Pacific. And he found references to at least two previous major earthquakes along that subduction zone, one in 1833 and another in 1861. "I think you could see [the potential] just by looking at the geology," Cummins says. "I don't know why no one had made the connection to tsunami before."
Part of the reason Cummins took notice has as much to do with his biography as with his training. A native of southwest Florida, Cummins started out as a physicist. During graduate school at the University of California-Berkeley, he "just sort of gravitated over to geophysics," he says, and moved to Australia as a postdoctoral fellow in 1989 to study the deep structure of the Earth. It wasn't until 1997, though, when he moved to a geology institute in Japan, that Cummins started taking a personal interest in earthquakes. Along with his wife and three children, Cummins says, he ended up living at what had been the epicenter of an earthquake that devastated Tokyo in 1923. "I came to realize that we were sitting right on top of the rupture area that caused a tsunami in that area," he recalls. "It was a funny place for a seismologist to live." It was also the kind of place that would get a father thinking about tsunamis and the devastation they can cause, an interest Cummins maintained when the family moved back to Australia in June 2002.
Making the case. After recognizing the potential for a major Sumatran tsunami, Cummins set out to see if one had ever actually happened. There were no historical accounts of tsunami destruction in the wider Indian Ocean region, so, together with numerical modeler David Burbidge, he produced a computer simulation of the 1833 Sumatran quake. Published in September 2004, the re-creation indicated that tsunami waves would have radiated outward from the epicenter in a pattern distressingly similar to the one observed on the day after Christmas. But the simulation didn't include the much more difficult "run-up" calculations, when tsunami waves move from the open ocean to shorelines, so Cummins and Burbidge didn't realize how severe the potential damage was. When the December 26 earthquake struck, Cummins says, "I was working on the next step, which would be to find historical evidence to verify that you do get tsunami from these earthquakes. Then I would have had a strong case for a warning system."
Today, there is little question that a warning system for the Indian Ocean will go forward. Geologists from around the world are already combing through the region, looking for clues to explain why this tsunami was so devastating--and to help predict destruction patterns for any future ones. Several different warning-system proposals are being floated and will be discussed at a previously scheduled U.N. disaster reduction conference in Kobe, Japan, from January 18 through 22. India has said it will set up its own alert network, and Cummins is involved in drawing up the Australian government's plan for a $20 million Indian Ocean system that could be in place in about a year. Sen. Joseph Lieberman has called for the United States to step in and set up a system for the entire globe, including other potentially overlooked regions like the Mediterranean and Caribbean seas. Many tsunami experts support that plan, pointing out that the great increase in coastal development and settlement over the last decades means that many more people are vulnerable than at any point in history.
As for the argument that the damage is done and it's too late to put an Indian Ocean warning system in place, Cummins is emphatic. "I don't think that's true at all," he says. "These are not one-off events, and the eastern part of the Sumatran subduction hasn't ruptured yet. The danger's still there."
ADVANCE WARNING
With the death toll of the December 26 tsunami climbing toward 160,000, momentum is building to extend the tsunami early warning system beyond the Pacific Ocean to cover other vulnerable coastlines. The United Nations is set to coordinate the effort, and while the details have yet to be worked out, the final plan will need to couple new and existing detectors with a rapid, reliable communications system. Here's what is likely to be included:
DETECTING TSUNAMIS
SEISMOGRAPHS
They detect earthquakes, their size and location, but can't tell scientists whether tsunami waves have been generated. Data from seismographs around the world are relayed to regional warning centers as a first alert.
DEEP OCEAN SENSORS/BUOYS
Deep ocean sensors are the most effective detection device, picking up subtle pressure changes as tsunami waves pass by. Buoys transmit the data to warning centers via satellite.
TIDE GAUGES
These sea-level detectors measure the rise and fall of tides. They can also give the first indication of whether tsunami waves are spreading outward from underwater quakes.
A COMMUNICATIONS SYSTEM
REAL-TIME ACCESS
No matter what detectors are in place, seismologists in regional warning centers need data fast to initiate emergency response plans--or avert false alarms. Some options: VHF radio, satellite, microwave transmissions, and dedicated land lines.
GETTING OUT THE INFORMATION
Tsunami monitors halfway around the world picked up signs of the December 26 earthquake within 15 minutes but had no contacts in the region. Warning systems, from cellphone alerts to sirens and loudspeakers, are essential, as is a network of local warning centers.
PREPARING THE PUBLIC
Tsunamis travel fast, and for nearby areas "early warning" can mean minutes, or less. Public education can help citizens recognize impending danger and warning sirens, learn how to escape, and locate food, water, and shelter after the event.
[Map Labels]
9.0 earthquake December 26 epicenter
Proposed tsunami detector sites
AFRICA
Subduction zone
Earthquake zone: One plate moves under another
Arabian Sea
Maldives
India
Sri Lanka
SUNDA ARC
Indian Ocean
Bay of Bengal
Burma
Thailand
Malaysia
Indonesia
Japan
Philippines
AUSTRALIA
Pacific Ocean
Sources: DM Solutions, Dr Simon Day (University College London and University of California-Santa Cruz), NOAA, UNESCO
This story appears in the January 24, 2005 print edition of U.S. News & World Report.