Home > News
  print button email button

Tuesday, Dec. 27, 2011

Quake prediction myth debunked

Professor's new book blows cover off panic-driven ¥10 trillion industry

Special to The Japan Times
First of two parts

There's a map of Japan on a wall in Robert Geller's office liberally marked with color-coded dots. Titled "hazardo mappu" (Hazard Map), it's a government-produced chart indicating areas believed to be most susceptible to earthquakes.

News photo
Different perception: Robert Geller sits in his office at the University of Tokyo's Department of Earth and Planetary Sciences in late November. ROB GILHOOLY PHOTO

"I prefer to call it the 'hazure (off-target) mappu,' " says Geller, a professor of seismology at the University of Tokyo's Department of Earth and Planetary Science. "To say that certain areas in Japan are dangerous and others are less at risk when that's actually proven not to be the case is obviously against the public interest."

This argument is at the heart of Geller's recently published book "(Nihonjin wa shiranai) Jishin Yochi no Shotai," ("The Truth About Quake Prediction (that the Japanese don't know)").

In it, he brings into question Japan's fascination with quake prediction research, a multibillion-dollar government initiative dating back to the 1970s that forms the basis of the probabilistic maps.

Of the nine quakes since 1979 that caused 10 or more fatalities, none occurred in areas designated on the maps as high-probability quake zones, he says.

He also decries the "cavalier approach" to building nuclear power plants on locations that the maps suggest are less at risk from major disasters, despite research data and historical records suggesting otherwise.

According to Geller, one example is the Fukushima No. 1 nuclear plant, which was severely damaged by an estimated 15-meter-high tsunami on March 11, causing the release of large amounts of radioactivity.

By identifying and interpreting sedimentary rocks deposited by tsunami several kilometers from the shoreline near Sendai, in Miyagi Prefecture, geologists, including those from Tohoku Electric Power Co.'s Onagawa nuclear power station, showed in the 1980s that the Jogan tsunami advanced as far as 4 km inland in 869. Historical chronicles indicate 1,000 fatalities from the disaster.

Geologists have since used "paleo-tsunami" data from this and two other comparable prehistoric tsunami to try and convince authorities and power companies that the region's nuclear power plant defenses need further strengthening, Geller says.

"After the March 11 disasters, TV pundits and officials frequently said the events in Tohoku were 'unforeseeable,' when in fact some experts have been warning of the risks for decades," he says.

As recently as 2009 a governmental hearing held to review seismic and tsunami safety at nuclear plants also was warned of the risks of a large tsunami based on the Jogan data.

"They ignored it, preferring to concentrate on 'foreseeable' quakes used to create hazard maps, which are hypothetical and based on unproven prediction theories," he says.

For 140 years, geologists worldwide have speculated that before a major quake there ought to be some precursor that can be measured using sensors, or more recently GPS analysis, to indicate well in advance that a quake is coming.

Such precursors would be most notable at "seismic gaps" — zones in active faults in the Earth's crust where no major quakes have occurred for a significant time.

This theory attracted followers in Japan, gaining momentum in 1977 when then Tokyo University professor Katsuhiko Ishibashi announced that a major quake — the Tokai Earthquake (often referred to as "the Big One") — was imminent at one of those seismic gaps in Suruga Bay off the coast of Shizuoka Prefecture.

The ensuing panic — fueled, says Geller, by an unquestioning mass media — led to the birth of Japan's ¥10 trillion-a-year quake prediction industry. This despite there being no physical theory yet established to explain how earthquakes occur, Geller says.

"A lot of money has been spent putting out a lot of instruments in the hope some precursory phenomenon or other will turn up. That's not only obviously wrong, but also very strange from a scientific point of view."

Ishibashi's report came in the same year that Charles Richter, creator of the earthquake magnitude scale, commented that prediction "provided a happy hunting ground for amateurs, cranks and outright publicity-seeking fakers."

Geller concurs, but says there is another category to add to that list: prediction experts — "some who don't even believe prediction is possible" — who continue their work purely to profit from the funds allocated each year to research here.

"As (author) Upton Sinclair once said, it's very difficult to get someone to understand something when their salary depends upon his not understanding it," Geller says.

The current system allows prediction researchers to bypass the peer review system that usually exists for scientists, he says. Instead of submitting scientific papers for scrutiny here, some researchers take their theories straight to the mass media, Geller adds.

"Over the past 30 years there have been hundreds of stories published," says Geller, who has a closet in his office piled high with weekly magazines and comic books featuring stories about quake prediction theories.

"Most of the theories just repeat what has gone before and research justified on the basis that nobody has yet proved it to be impossible.

Under normal circumstances, the funding agency would laugh you out of the room (for citing such justification)."

A crucial consequence of this system is that the "seemingly authoritative" hazard maps and prediction theories have led to a dangerous level of dependency among the Japanese public, Geller says.

This is particularly true among those living in areas designated as low risk, who are lured into a false sense of security by what Geller calls Japan's "anzen shinwa" (safety myth).

"We probably know most of the damaging quakes that occurred over the past 1,500 years, but even then that's probably too short a time sample to say for sure that this place is safe and that one isn't. It's like trying to generalize the weather from three days in July."

While most prediction researchers are "hung up on the idea that there has to be a preparation process for a big quake," Geller believes it is far more random. "We need a new paradigm and I believe that it will likely say that the quake process is not deterministic but stochastic, that any quake has some probability of just running away into your next magnitude 9," he says.

Consequently, the defenses currently in place around many of Japan's nuclear power plants are in need of urgent review, he says.

"Utility companies were a bit cavalier when plants such as those in Fukushima were built. But as more and more knowledge came along about how dangerous they were, and they didn't upgrade the defenses, they went from being cavalier to highly negligent and/or irresponsible.

"Once in a while they have to be prepared to take a hit on profits in the short run in order to move away from the anzen shinwa and guarantee public safety."

Related link

Back to Top

About us |  Work for us |  Contact us |  Privacy policy |  Link policy |  Registration FAQ
Advertise in japantimes.co.jp.
This site has been optimized for modern browsers. Please make sure that Javascript is enabled in your browser's preferences.
The Japan Times Ltd. All rights reserved.