IE 11 is not supported. For an optimal experience visit our site on another browser.

Predicting quakes? Seconds may be all we have

A hundred  years after the San Francisco temblor that gave birth to earthquake science, seismologists still don’t have a way of saying when a major quake will hit a given area in the same way meteorologists can predict snowstorms, tornadoes and hurricanes. NBC's George Lewis reports.
/ Source: NBC News

BERKELEY, Calif. — Bounding up the steps of Memorial Stadium, a University of California-Berkeley structure that straddles the Hayward earthquake fault, Assistant Professor Richard Allen of the university’s seismological laboratory led an NBC News team on a brief field trip. 

“There’s creep going on here,” Allen said, pointing to a growing gap of several inches between two sections of the stadium, the gap covered by a bent steel plate, “and we’re seeing the effects of that creep.”

To ease our creepy feelings, Allen, who hails from Great Britain, immediately assured us that engineers believe the stadium is safe and that the university is spending millions of dollars retrofitting it and other structures on the Berkeley campus to withstand the next “Big One,” referring to the huge 1906 San Francisco earthquake.

Like most other scientists, though, Allen confesses he has virtually no idea when that next “Big One” might be. 

Basically, no way to predict earthquakes
Yes, 100 years after the event that gave birth to earthquake science, seismologists still don’t have a way of saying when a major temblor will hit a given area in the same way meteorologists can predict snowstorms, tornadoes and hurricanes.

“There’s no method in existence,” Allen said, “that’s been consistently able to predict earthquakes.”

Indeed, there’s a great debate going on among seismologists about whether it will ever be possible to make such predictions. Some believe that earthquake faults behave in a completely random manner that defies any attempt, no matter how sophisticated, to determine how they’ll act in the future. Others contend that there may be predictable patterns in the behavior of earthquake faults that science has not yet discovered.

“As a community, seismologists are working on this problem,” Allen said, measuring his words carefully, “and at some point in the future, that might lead to the ability to predict earthquakes, but we’ll have to see.”

An early warning system
In the meantime, without any means of long or medium prediction, Allen and his colleagues are working with what they have — the few seconds between the first indication of an earthquake and its destructive main arrival — to develop a system that would alert key officials (and, soon, automated systems) just ahead of future major earthquakes.

The goal is to pick up the first rumblings on a network of seismic sensing devices and immediately sound the alarm on computers, pagers and radios tuned to emergency frequencies. Allen and fellow researcher Erik Olson recently reported in the journal Nature that they’ve developed computer formulas that rapidly analyze the initial jolt and calculate the earthquake’s magnitude.

Though the warning would come only seconds before the quake hits, Allen said that is potentially life-saving information. “You can get under a desk,” he noted, “but also in terms of larger complexes and transportation systems, they can actually take action to reduce the impact of the earthquake.”

He explained that railroads would have time to slow down or stop trains; that airports could prevent planes from landing during an earthquake and that chemical plants could shut down valves to minimize hazardous spills from ruptured pipes.

“It also allows these companies to get up and running more quickly after the event,” Allen added, “so it has less of a long-term impact on society.”

Allen’s system is based on the two kinds of shock waves — “P” (for primary) waves and “S” (for shear) waves — sent out by an earthquake. The relatively harmless “P” waves radiate from a quake’s epicenter more rapidly than the slower-moving “S” waves, which cause the most violent earth motions. The seismic networks would pick up the first “P” waves and rapidly determine how big an area might be affected by violent shaking, thus enabling automatic systems such as cutoff valves to spring into action.

Lessons from a deadly tsunami
Meanwhile, researchers are using information from more recent earthquakes, such as the huge temblor that caused the 2004 Indian Ocean tsunami, in an attempt to nail down longer term predictions — and, they say, are making some progress.

California Institute of Technology scientist Kerry Sieh, for instance, believes there are discernable patterns to earthquake behavior.  

Part of a team that’s been studying the 9.1-magnitude quake and tsunami centered off the Indonesian coast that killed hundreds of thousands — the 1906 quake was at least 10 times weaker — Sieh and his colleagues have been focusing on another temblor that hit the area just to the south of the original rupture, killing another 2,000 people.

Now Sieh and others believe that a third section of the same fault is on the verge of letting go. “It could devastate the coastal communities of southwestern Sumatra,” Sieh said, “It could happen tomorrow or it could happen 30 years from now, but I’d be surprised if it were delayed much beyond that.” 

According to Sieh, the behavior of the Indonesian fault shows that earthquakes are not just random events. The challenge for science is to see if those patterns can someday lead them to develop a system for earthquake predictions.

Relying on rough estimates
In the meantime, people living in earthquake zones around the world will have to live with today’s rough predictions about temblors to come. 

In the San Francisco Bay area, scientists say the odds of a major earthquake along the San Andreas or the Hayward fault lines are fairly high. “We can say that there is a 60 percent chance of an earthquake in the Bay Area in the next 30 years, Allen said, “and that allows us to build buildings that are less likely to be damaged.”

The U.S. Geological survey agrees with that prediction, putting the chances of a magnitude 6.7 or greater quake in the Bay Area at 62 percent between now and the year 2031. 

As for San Francisco itself, scientists don’t believe that pressure on the San Andreas Fault —the one that lies directly beneath the city — has built up to the point that there will be another great quake like the one in 1906 for many years, though they can’t say so with absolute certainty. 

Such is the state of the art of earthquake prediction 100 years after the “Big One.”