Ability to Forecast Quakes Shaky at Best, Experts Say : Seismology: Most scientists dismiss a widely reported prediction of a major Midwest temblor in December.

“Only fools, charlatans and liars predict earthquakes.”

That unkind assessment of those who claim to know when the Earth is most likely to shudder has been attributed to the legendary pioneer in seismology, Charles Richter. The quote has become part of the folklore of the dark art of predicting earthquakes, and many would say it is as true today as when it was allegedly muttered by the Caltech scientist who gave the world its earthquake magnitude scale.

For despite decades of research and enormous progress in understanding the mechanics of earthquakes, experts agree that no one has come up with a formula that is likely to lead to a reliable prediction of exactly when and where an earthquake is about to strike. Even Caltech’s Clarence Allen, the dean of Southern California seismologists and an eternal optimist in the field of earthquake predictions, admitted recently that science is still a long way from being able to do that.

“I don’t think I’m going to live long enough to see any kind of routine short-term predictions,” said Allen, 66. Then, yielding to the optimism that has marked his long career, he added: “But I hope in my lifetime I will see some intriguing signals.”

That, right now, is about the most that any reputable scientist is hoping for, despite highly publicized claims by some who say they have successfully predicted earthquakes. And that includes a climatologist who has predicted a catastrophic earthquake in early December for the New Madrid fault that runs through southeastern Missouri, using a theory that has been examined and discarded by numerous experts.

Most seismology experts are instead focusing on two expected earthquakes, one in Japan south of Tokyo and one in Central California east of Paso Robles, and either of those quakes would provide clues that will enhance the chance of predicting earthquakes elsewhere. Both projects are based on the belief that some earthquakes strike on fairly regular intervals.

But scientists in Japan are not certain they will detect evidence that their quake is imminent in the hours before it hits, despite a high degree of instrumentation throughout the region. And the Central California quake is already past due.

So for now, scientists believe that the best they can do is issue “forecasts” of the probability of a specific fault rupturing within a given number of years.

The problem is that increased understanding of seismology has led to a greater awareness of the complexity of earthquakes and a keen sense that no two earthquakes are exactly alike. Thus what “triggers” one earthquake may not trigger another, and events leading up to one quake may be absent from the next.

That has left scientists with a grab bag of clues but no formula that will tell them which events, such as smaller temblors, will precede a major quake. This has made seismologists a humble group, and many are more than a little outraged by the New Mexico climatologist’s claims that a catastrophic earthquake will strike the nation’s heartland on Dec. 2 or 3.

Iben Browning based his claim on an alignment of the Earth, sun and moon, and his forecast sent shock waves throughout much of the nation--from Chicago to New Orleans--even though a special panel of experts set up by the U.S. government concluded “there is absolutely no scientific basis” for Browning’s prediction.

And although the federal government has spent hundreds of millions of dollars on earthquake research, it has issued only one official prediction of an earthquake, and that temblor is now past due. Based largely on an average period of 22 years between quakes on a Central California segment of the San Andreas Fault, the U.S. Geological Survey predicted in 1985 that an earthquake of about 6 magnitude would strike near the community of Parkfield around 1988.

The government has spent $15 million setting up instruments to record every event leading up to this quake, and at least 50 experts have devoted chunks of their careers to preparing for the temblor. But Parkfield continues to doze.

The prediction said the quake could come within five years on either side of that date, so it is not yet invalid, but many scientists are beginning to wonder whether nature has played a cruel trick on them.

And new evidence is not encouraging, according to Evelyn Roeloffs of the U.S. Geological Survey, who is chief scientist on the Parkfield experiment. Recent work suggests that past earthquakes in that area differed considerably from each other, and those quakes may not have hit with the regularity experts had thought.

So Parkfield, once thought to be the safest of adventures into earthquake predicting, is demonstrating just how difficult the field can be.

The real world of earthquake forecasting, it turns out, bears little similarity to a recent television movie in which some experts became convinced that the Big One was about to strike Los Angeles because conditions here were so similar to events leading up to a catastrophic quake near Mexico City in 1985.

In reality, the Big One could come with no warning whatsoever. All the money in the world and the best experiments that scientists can create may fail to warn the people of Southern California that disaster is just hours away.

On the other hand, there might be ample clues so that even the dullest of seismologists will figure out something is about to happen.

History teaches that lesson quite clearly. In 1975, clues were so persuasive that China evacuated the city of Haicheng in the hours before a 7.3 quake, which leveled the city. Residents throughout the region had been alarmed by frequent foreshocks that rumbled through the region even as residents were being evacuated, as well as by peculiar animal behavior for weeks before the quake.

As a result of the evacuation, there were few deaths. But only a year later a 7.6 quake destroyed the city of Tangshan, China. About a quarter of a million people died. There had been no warning signs. There had been no prediction. Earthquakes are like that.

The humbling nature of seismology left many experts chuckling to themselves when they first heard that an obscure climatologist had predicted a major quake on the New Madrid fault system, which ruptured in the early 1800s with a series of quakes that rank among the most powerful ever to strike North America. But some Midwestern officials took it seriously, vowing to close schools and other facilities for the days of “geological danger” cited by Browning.

Browning publishes the Browning Newsletter, which is sold to industrial customers and has warned of everything from the probability that the U.S. government will be overthrown next year to the imminent beginning of an ice age just as most scientists are worrying about global warming.

Browning’s prediction of a New Madrid quake snowballed in credibility when some of his subscribers claimed that he had successfully predicted such things as the Loma Prieta earthquake that devastated the San Francisco Bay Area last year and the eruption of Mt. St. Helens.

Faced with growing public alarm, the U.S. Geological Survey established an “ad hoc working group” to evaluate Browning’s prediction.

“We took it seriously and looked at it carefully, even though we were very skeptical (of the prediction),” said Thomas Heaton, a scientist in charge of the Geological Survey’s Southern California office and a member of the working group. But after examining the evidence, the prediction “just didn’t make any sense to us,” Heaton said.

On Oct. 18, the group issued a report challenging virtually every aspect of Browning’s prediction, including the claim that he had predicted the Loma Prieta quake.

Browning, who is recovering from an operation, has declined all interviews. A woman who answered the phone at his home said simply that the 72-year-old business consultant would stand on his record, despite a strong attack from the seismological community.

But did Browning really predict the catastrophes he claims to have foreseen?

The working group reviewed a videotape and a transcript of two speeches Browning gave before the Loma Prieta quake, in which he claims to have made the prediction. But according to the report, Browning never mentioned California and said only that “there will probably be several earthquakes around the world, Richter 6-plus, and there may be a volcano or two.”

The report concluded that his track record in predicting volcanoes is even less impressive. He did predict Mt. St. Helens, the report says, during a speech in Portland, Ore., six days before the eruption. But by then the area around the volcano had already been evacuated, and geologists had issued numerous warnings “that a major eruption was imminent,” the report said. Browning’s predictions are particularly frustrating to experts because they are based on a phenomenon that has been extensively researched and discarded as the “triggering” mechanism for earthquakes. He believes major quakes are triggered by the alignment of the Earth and the sun and the moon--in other words, when there is either a full moon or a new moon.

Years ago, many scientists thought the same thing because during such a period the stress on the Earth is particularly high. The ground bulges and contracts, just as the seas swell in tidal actions, so it seemed reasonable that such immense stresses could cause earthquakes.

Ironically, Heaton himself once thought that might be true, and his early work has been cited by Browning to support his prediction.

“I wrote a paper I would just as soon forget,” Heaton said.

Heaton later repudiated his own work after conducting more extensive research. When he looked at earthquakes on a global scale, Heaton found that 79 out of 80 quakes did not fit the pattern because they did not occur at times of maximum tidal stress.

“If you have 79 out of 80 occurring randomly, there’s not much predictive value in a full or new moon,” Heaton said.

The working group’s report expressed puzzlement over why Browning zeroed in on early December. Even if tides do play a role, Heaton said, there is no reason to believe quakes should occur at one particular peak tide.

In fact, the tidal conditions in early December will not be significantly different than they were as recently as three years ago, and no earthquake struck then.

Tidal stresses on Jan. 17, 1988, were only a little less than they will be on Dec. 2 and 3, and the difference is so slight that even subtle changes in atmospheric pressures would make the stresses indistinguishable for the two dates.

So the report concludes that the danger on Dec. 2 and 3 will be no greater--and no less--than on any other day.

Unfortunately, despite the condemnation of Browning’s method, no seismologist can say there won’t be an earthquake because no one knows exactly when one is going to strike. And strike it will, someday.

Asked what he will say if a major earthquake hits the Midwest in early December, Heaton winced. With his face nearly contorting in agony, he said, “I guess we will just say, ‘damn.’ ”

If it does hit, even though Iben Browning’s method could still be as flawed as the experts think, the world would never forget a cantankerous climatologist who caused a lot of folks in the Midwest to schedule their vacation trips for early December.