Japan’s ‘black swan’: Scientists ponder the unparalleled dangers of unlikely disasters

REUTERS - An aerial view of the No. 4 reactor t the Fukushima Daiichi nuclear power complex.

The disaster bureaucrats talk about black swans: calamities from out of the blue, terrible and strange. The world is now transfixed by the black swan disaster of Japan — an earthquake larger than seismologists thought could happen in that part of the country, leading to a tsunami too big for the sea walls, and now a nuclear crisis that wasn’t supposed to be possible.

“People talk about the Big One. This is it,” said Tom O’Rourke, a Cornell professor of civil and environmental engineering and a member of the federal Advisory Committee for Earthquake Hazard Reduction. “This is what the Big One looks like. We’ve had an imaginative idea of what the Big One would be like if it struck a major, populated, modern society.”



Track the status of the nuclear crisis in Japan.
Click Here to View Full Graphic Story

Track the status of the nuclear crisis in Japan.

More on this Story

View all Items in this Story

Japan’s nightmare comes in the wake of two other events that scientists found surprising in their location and intensity: the highly destructive earthquake in Christchurch, New Zealand, on Feb. 22, which was triggered by a little-regarded fault; and the tsunami-spawning Sumatra earthquake Dec. 26, 2004, on a trench not considered likely to cause such a “mega-quake.”

It may seem as if there are more natural disasters these days, but the real issue is that there are more people and more property vulnerable to the violent forces of Earth. Natural disasters are supplemented by technological disasters — last year’s Deepwater Horizon oil spill in the Gulf of Mexico being one example. Disaster planners in the United States have to ask themselves how they would deal not only with the obvious types of calamities — Gulf Coast hurricanes, for example — but also the events that are of low probability but come with high consequences.

“You don’t get to pick the next disaster. You don’t necessarily know where the threats are,” W. Craig Fugate, director of the Federal Emergency Management Agency, said this week as he contemplated Japan’s horrific combination of catastrophes. “We plan for the things we know, but we also plan for the things we don’t know.”

‘The highly improbable’

The term “black swan” was coined and popularized by Nassim Nicholas Taleb, a New York University professor of risk engineering and author of “The Black Swan: The Impact of the Highly Improbable.”

People debate what qualifies as a black swan. Most alleged black swans turn out to have obvious precursors and warning signs — the Sept. 11 attacks included. Nothing comes out of the blue, truly.

The next big disaster could be something off the radar of most Americans. A solar flare, for example, could trigger a geomagnetic storm that could knock out much of the nation’s power grid. Or an earthquake could hit an East Coast city not generally considered vulnerable to a major temblor. That sounds like paranoia, but mainstream scientists and government officials research such things.

“South Carolina’s got a very significant seismic history,” Fugate noted. “There’s a fault that runs through Charleston, South Carolina, that has devastated that area before.” That 1886 event, with an estimated magnitude of 7.3, killed 60 people and was felt as far away as Wisconsin, Boston and Cuba.

That was what geologists call an intraplate earthquake, an event within one of the planet’s major tectonic plates rather than along the margin, where earthquakes are easier to understand and anticipate. What causes these intraplate earthquakes is a thorny scientific question. Some intraplate faults might have a major quake every few thousand years, so infrequently that they are not in sync with the human time scale and do not factor significantly in the U.S. Geological Survey’s seismic hazard maps.

Conversely, some hazards are well publicized at this point but highly unlikely for centuries to come, such as a full-blown eruption of the Yellowstone caldera, sometimes referred to as a supervolcano. If Yellowstone were to explode, it would be an event thousands of times more violent than the Mount St. Helens eruption, and its effect would be felt across much of the western United States. But it has been 640,000 years since the last such event and, although the caldera is active and generates swarms of small earthquakes, there is no sign that a major event will happen in the lifetime of anyone around today.

Some hazards become more prominent with new scientific research. Only in the past couple of decades have scientists come to appreciate the threat posed by the Cascadia subduction zone, a tectonic plate boundary running just off the coast of the Pacific Northwest, from Northern California to Vancouver Island. It could generate a mega-quake like the one that just hit Japan.

A breakthrough in understanding that fault line came when American scientist Brian Atwater found signs of coastal cedar forests that had been swamped after the growing season of 1699. In a major earthquake, the land can drop several feet instantly and the tidal line can change. Atwater consulted with Japanese scientists and found records of an “orphan tsunami” in the year 1700. Scientists were able to determine that around 9 p.m. on Jan. 26, 1700, the Cascadia subduction zone generated a mega-quake.

“What we’re watching in Japan is possible off the Cascadia subduction zone,” said O’Rourke, the Cornell professor.

Disaster preparation requires a careful calibration of risk and a strong sense of what’s a reasonable level of caution. Society cannot protect itself from everything that conceivably could go wrong. Even with nuclear power, where safeguards are piled on top of safeguards, there is a point at which the operation becomes too expensive for anyone to attempt.

David Lochbaum, director of the Union of Concerned Scientists’ Nuclear Safety Program, said this week, “You can design a reactor to be bulletproof, but no one would want to pay for it.”

Tom Hunter, former head of Sandia National Laboratories and a leader of the federal government’s scientific team that responded to the Deepwater Horizon oil spill — the most recent black swan in the United States — said it is up to America’s political and business leaders to find the right balance between safety and efficiency.

“I think many of our systems do not operate as if things could go wrong,” Hunter said. “They operate as if everything will go right.”

Scientists have put together what they call a Probablistic Seismic Hazard Assessment that seeks to map the probability of a certain amount of shaking in any given time window. But “probabilistic” isn’t the same thing as deterministic.

“The problem is that, by design, you leave yourself open to the low-probability, high-impact event,” said Susan Hough, a U.S. Geological Survey seismologist who has written about the difficulty of predicting earthquakes. In Boston, for example, she said, “hazard maps say the hazard is low, and rightly so, but the potential risk could be enormous. Seismologists know this, but I don’t think the point is widely appreciated outside of the scientific community.”

Preparing for the worst

In May, FEMA will run an exercise in the central United States to help prepare for any earthquakes along the New Madrid seismic zone. This is the 200th anniversary of the first of three powerful earthquakes — estimated at magnitude 7.7 — that struck along the Mississippi River Valley.

Three years ago, 5,000 emergency responders in Southern California participated in an earthquake drill, known as the “ShakeOut,” that pondered the effects of a magnitude-7.8 quake on the San Andreas Fault. But now officials are talking about something that they say is just as likely and possibly would cost three times as much in property damage: a flood of biblical proportions that would turn California’s Central Valley into an inland sea.

The U.S. Geological Survey prepared what it called the ARkStorm scenario (for “Atmospheric River 1,000”), in which California experiences rainfall similar to what occurred in the winter of 1861-62, when rain fell for 45 consecutive days. The flooded area could be 300 miles long and 20 miles wide, and property damage could reach $400 billion, with an additional $325 billion in business interruption costs.

The Oak Ridge National Laboratory issued a warning this past fall about the possible disastrous effects of an electromagnetic pulse (EMP), caused by a solar flare or a terrorist attack, that would knock out the electrical grid for up to 130 million people. A solar storm damaged a Quebec power system in 1989 and caused a blackout that affected 6 million.

The question for policymakers is to what extent it makes sense to take mitigation steps for low-probability, high-consequence hazards.

The Oak Ridge report suggests that paying now is a lot cheaper than paying later. “The cost of damage from the most extreme solar event has been estimated at $1 [trillion] to $2 trillion with a recovery time of four to 10 years,” the report said, “while the average yearly cost of installing equipment to mitigate an EMP event is estimated at less than 20 cents per year for the average residential customer.”

The disaster experts have a buzzword: resilience. You can’t stop the disaster from happening — the very nature of a black swan is that it catches you off-guard — but you can increase the speed and grace with which society bounces back.

“Think of resilience in terms of the old Timex commercial,” said Jack Hayes, director of the National Earthquake Hazard Reduction Program. “It can take a licking and keep on ticking.”

Read what others are saying