Rumbles along the San Andreas.

Rumbles along the San Andreas 

Some offices conduct baby pools, others wager on professional sports. But the U.S. Geological Survey (USGS) has bet $15 million on a prediction that Parkfield, Calif., will feel a strong earthquake before the end of 1993. Judging from the behavior of the San Andreas fault in the last two months, scientists suspect that quake may be on its way.

In 1985, USGS issued the Parkfield prediction -- its only official prediction to date -- based on research showing that magnitude 6 shocks have occurred near the farming town about every 22 years throughout the last century. Because the most recent tremor struck in 1966, the survey concluded that the San Andreas segment running through Parkfield would generate another magnitude 6 quake within 5 years of 1988 (SN: 5/5/90, p.278). Seismologists have outfitted the region with a battalion of different devices to study the quake in unprecedented detail.

On Sept. 9 and 10, two magnitude 3 shocks rumbled through the northern end of the Parkfield fault segment, near the region where the predicted quake is expected to begin. This area has been particularly quiet since 1986, and the two jolts prompted officials to issue a "C" level alert, signifying a 1.4 to 6 percent chance the predicted shock would occur within 24 hours of the alert. In the past five years, the USGS has issued 22 "C" level alerts for Parkfield but no "B" or "A" level alerts, which indicate even greater chances of an impending quake.

In August, two magnitude 3 tremors rattled the southern end of the Parkfield segment. These shocks, in conjunction with the northern ones last month, have heightened the suspense for many geoscientists. Evelyn A. Roeloffs, chief scientist on the Parkfield Prediction Experiment, says, "My personal opinion is that we may be entering a phase when the earthquake is months away, instead of years away."