Big Bay Area Quake Likely Within 20 Years, Experts Say

Brian Handwerk
for National Geographic News

October 11, 2005

Saturday's South Asia earthquake serves as a tragic reminder of Earth's unpredictability. Now a computer simulation suggests that the San Francisco Bay Area may experience a large quake within the next two decades.

The computer program, dubbed Virtual California, estimates a one-in-four chance that the Bay Area will be rocked by an earthquake measuring between magnitude 7 and 7.1 sometime in the next 20 years.

"The USGS has estimated a 62 percent chance of a magnitude 6.7 or larger earthquake in the Bay Area within the next 28 years," said John Rundle, director of the Center for Computational Science and Engineering at the University of California, Davis. "So we're not talking about something that's all that different."

Over the next 45 years the projected chance of such a quake rises to 50 percent. In 80 years the odds reach 75 percent, according to the UC Davis-based program.

A quake of that size could be similar to the October 1989 Loma Prieta earthquake, which famously disrupted a World Series baseball game at San Francisco's Candlestick Park. That quake killed 63 people and caused some 5.9 billion U.S. dollars in damages.

Orders of Magnitude

Geologists measure the energy released by an earthquake using a logarithmic scale, which increases by powers of ten. This means a magnitude 7 quake is ten times stronger than a magnitude 6.

Using this scale, even increases of a few tenths of a point can indicate a dramatic difference in quake strength.

"We don't know exactly how big the [infamous] 1906 earthquake was, but I think that the consensus would be an estimate of 7.8," explained Donald Turcotte, a geologist at UC Davis. "Of course, that is much bigger than a 7 on this logarithmic scale."

The probabilities expressed in the Virtual California study fall off quickly after a magnitude 7 event, so that quakes larger than 7.1 are considerably less likely.

In 2000 the Virtual California simulation was used for a combined NASA/U.S. Department of Energy study to identify the California regions with the highest probabilities of near-future quakes.

The simulation's forecast map has proven uncannily accurate in succeeding years.

Between 2000 and 2003, five of California's quakes measuring magnitude 5 or greater occurred within about 7 miles (11 kilometers) of the sites identified by the program. The probability of this occurring randomly is about 1 in 100,000.

Virtual California's most recent projections are reported in this week's issue of the Proceedings of the National Academy of Sciences.

Interconnected Faults

Rundle's program simulates how Northern California's hundreds of interrelated fault lines interact and collectively influence earthquake activity.

"Our method takes into account the present and past interactions between the San Andreas fault and the other faults in the Bay Area explicitly, and no other method does that," he explained.

Using seismic data from the 1930s to present, Rundle's team modeled earthquake activity back in time some 40,000 years.

The program rocked the Bay Area with 395 simulated earthquakes of magnitude 7 or higher. The recreated quakes occurred on an average of every 101 years.

"Obviously there have been many quakes in the past that we know nothing about," Rundle said.

"The simulation can give us information on the nature of those earthquakes and how they influenced succeeding earthquakes. We can use that information to develop forecasting methods for future earthquakes."

Earthquake Eye in the Sky

Could seismologists someday achieve the kind of simulations routinely employed in weather forecasting?

"Weather is chaotic, but forecasts for 48-hour periods are pretty good," Turcotte said. "Usually hurricane forecasts are pretty good. Weather forecasting has made remarkable progress over the last few decades."

"In principle the Earth is not any more complicated than the atmosphere," he added.

But as with any model, Virtual California is only as good at the data it receives.

"We're extremely data-poor right now, but hopefully in five or ten years that will change," said Andrea Donnellan, principal investigator of the QuakeSim Computational Technology Project at NASA's Jet Propulsion Laboratory.

The next few years could see the launch of a satellite that would improve the program's effectiveness and provide data many earthquake researchers anxiously await.

Satellites can measure Earth surface displacement with uncanny accuracy using a technology known as interferometric synthetic aperture radar (InSAR).

InSAR creates a highly accurate map of a surface by combining simultaneous radar data from two sources into one image.

"Instead of finding out where your car is to within a few meters, you can find out where a position on the Earth is to a millimeter or two," Turcotte explained. "It's a remarkable leap forward in technology."

While GPS (global positioning systems) can provide information for specific points, InSAR can create a dynamic picture of the Earth's surface and map changes and movements for an entire area.

"As new data become available, new observations, [weather forecasters] incorporate them and do what they call model steering, continually updating the model," Rundle said. "That's what we do as well."

But no one knows how much the models might be improved, or if they could someday lead to more accurate short-term quake forecasting.

"Until we start seeing the data, we don't know," Turcotte said. "Maybe yes, maybe no."

In the meantime, Virtual California gives citizens and planners an idea of what they might expect—and some time to prepare.

"The future will never be identical to the past," Rundle said. "But what you'd like to do is develop a feeling about the types of earthquakes that could possibly hit on the San Andreas Fault."

Free E-Mail News Updates
Sign up for our Inside National Geographic newsletter. Every two weeks we'll send you our top stories and pictures (see sample).