Trial Over Earthquake in Italy Puts Focus on Probability and Panic

Filippo Monteforte/Agence France-Presse — Getty Images

AFTERMATH In the months before a temblor hit L'Aquila in 2009, the area had experienced an earthquake swarm.

By HENRY FOUNTAIN

Published: October 3, 2011

The manslaughter trial of six seismologists and a government official in the central Italian city of L'Aquila, stemming from what the authorities say was a failure to warn the population before a deadly 2009 earthquake, has outraged many scientists. Thousands have signed petitions protesting the prosecution as anti-science.

But the trial, which resumed Saturday, has also focused attention on a vexing problem in earthquake-prone regions around the world: how to effectively communicate the risk of potential disaster. Whatever the merits of the L'Aquila case, scientists and government officials have difficulty conveying what they know about the risk of earthquakes in ways that help prepare the public without sowing panic.

"People are expecting much more information, in particular quantitative information," said Thomas H. Jordan, a professor at the University of Southern California and director of the Southern California Earthquake Center. "Coming clean with what you know is being demanded by the public."

Earthquakes differ from other types of natural disasters. Meteorologists can track a hurricane with precision, but seismologists cannot predict exactly when and where an earthquake will occur. Scientists have condemned the Italian prosecution for this reason, saying the defendants are on trial for failing to do something that is impossible.

What seismologists are increasingly able to do, however, is forecast the likelihood that a quake will occur in a certain area over a certain time. Statistical analysis shows, for example, that some seismic activity — a minor quake or a swarm of very small ones — increases the probability of a larger, destructive earthquake in the same area.

But the probabilities are still very small, and they become even smaller with time. Given a low-probability forecast of an event that has potentially high consequences, the problem, Dr. Jordan said, becomes "what the heck do you do with that kind of information?"

That was a question that the Italian defendants faced. In the months before a magnitude 6.3 quake hit L'Aquila on April 6, 2009, killing more than 300, the area had experienced an earthquake swarm. That probably increased the likelihood of a major earthquake in the near future by a factor of 100 or 1,000, Dr. Jordan said, but the probability remained very low — perhaps 1 in 1,000.

But there was a wild card in L'Aquila that complicated the situation. As the earthquake swarm continued over several months, a local man who is not a scientist issued several predictions of a large earthquake — specific as to date and location — based on measurements of radon, a radioactive gas that is released as rocks fracture.

The predictions, none of which proved accurate, increased public anxiety in the city — so much so that the Italian government convened a meeting of a national risk-forecasting commission, including the seismologists and the government official, in L'Aquila on March 30.

At the meeting, the seismologists noted that it was possible, though unlikely, that the seismic activity could be a sign that a larger quake was imminent. They also noted that there was always some risk in L'Aquila, which has a history of earthquakes. But in a news conference afterward, the message to the public became garbled, with the government official assuring that there was no danger.

"The government ended up looking like it was saying, 'No, there's not going to be a big earthquake,' " when the scientists had not precluded the possibility, said Dr. Jordan, who was the chairman of a commission established by the Italian government after the quake to look at the forecasting issue.

The statement by the official, who is not a seismologist, violated a cardinal rule of risk communication, which is that those involved should speak only to their expertise, said Dennis Mileti, an emeritus professor of behavioral science at the University of Colorado at Boulder. "This person should not have been speaking," said Dr. Mileti, who has studied risk communication.

In general, said Michael Lindell, a professor at Texas A & M, scientists should advise emergency managers about the likelihood of events, and then the managers should make the yes-or-no decisions about whether to order an evacuation or urge the public to make other, simpler preparations. But often the roles become confused.

"When you step over the boundary outside your area of expertise, then there aren't necessarily any warning signs," Dr. Lindell said.

The L'Aquila news conference did not fill what was essentially an information vacuum, Dr. Jordan said. "One of the principles that social science has shown is that the public wants to hear things from people they trust," he said. "They want to hear things repeated.

"You don't want to put out information just when there's a seismic crisis, because people then don't have the context for this kind of information," he added. "You want people to get used to how these things ebb and flow."

California, with its active seismic zones, has a system for communicating risks to the public on a regular basis — though it, too, has flaws, Dr. Jordan said.

Just a few weeks before the L'Aquila quake, an analysis of an earthquake swarm in Southern California showed an increased likelihood of a major earthquake near the southern end of the San Andreas fault. While the probability was small, it was high enough that a scientific group decided to advise the state's emergency management agency. (In the end, no quake occurred.)

Even if the information at the L'Aquila news conference had been correct and the public had been warned there was a slightly higher risk, Dr. Mileti said, it would probably have made little difference. "One person saying once 'You don't have to worry' is probably not why they didn't do what they might have done to protect themselves," he said. "Humans are hard-wired to deny low-probability, high-impact events."

The only way to overcome that, he continued, is through constant communication. Once-a-year earthquake drills, like those in California, are not enough. The messages have to be everywhere, repeated ad nauseam.

"If you want to sell earthquake preparation in a way that it affects human behavior," he said, "you have to sell it like Coca-Cola."