Knowing When to Run

In the wake of history’s deadliest tsunami last week, scientists say an early-warning system would have saved many lives. But if preventing disasters were easy, everyone would be doing it.

Geoscientists have long struggled to find reliable predictors of earthquakes -- the main cause of tsunamis -- without success. Studies of possible signs, such as small tremor patterns, electric ground currents and even odd behaviors of dogs, snakes or clams, have produced few usable results. The closest thing to “early” warnings: new systems in Japan and Mexico designed to detect initial quake waves and transmit alarms 10 to 30 seconds ahead of the main shock -- perhaps enough time to stop trains, close bridges and get people under desks.

Tsunami warnings are more feasible -- it took about three hours for last week’s waves to reach India -- but most of the world lacks the hardware. The only system, serving Pacific nations, was started by the United States in 1949 and expanded in 1967 after tsunamis from Alaskan quakes killed people as far away as Hawaii and California. Alarms can go out within 15 minutes of a quake, but 75% of the alerts have been mistakes because other factors, including seafloor movement and water depth, also matter.

“False alarms are terrible -- people lose confidence,” said Frank Gonzalez, leader of the tsunami research program for the National Oceanic and Atmospheric Administration. One 1986 evacuation cost $40 million -- for a wave that turned out to be a foot high.

In the Indian Ocean, where last week’s disaster occurred, “a system is critical, and it’s doable,” said Kerry Sieh, a geology professor at California Institute of Technology who studies the Indonesian zone that produced the killer wave. He says a warning system to cover the whole ocean region could be created for a fraction of what the United States spends in Iraq per day, and he and others are scrambling for donors. Asian newspapers were abuzz last week with rumors of a planned system, but it remains to be seen whether one pans out once the news fades -- and whether it would survive.

“We routinely create early warning systems after all kinds of disasters, then we starve them in favor of the next big threat,” said Michael Glantz, a senior scientist at the National Center for Atmospheric Research in Boulder, Colo. The last 20 years have seen new systems for every hazard under the sun: typhoons, wildfires, droughts, emerging diseases. Many U.S. emergency managers say homeland security expenditures have largely wiped out natural-hazards programs.

Many systems fail because of technologies that fall short, or simple disorganization. Science is useless unless nations can receive and transmit warnings and get people moving fast, says Laura Kong, director of the International Tsunami Information Center in Hawaii. Many poor areas hit hard last week lack good transportation, communications and, in the case of civil war-ridden Somalia and parts of Indonesia, functional governments. How would countries that can’t guarantee daily security keep citizens safe in a disaster, warning system or not?

One of the most haunting images from the Indian Ocean shores was that of children scampering to retrieve fish stranded by the ominous recession of water that preceded the first big wave. Many of those children were swept away. Here, a gut-level public education message might have worked best: When the sea acts strangely, drop everything and run.