Studies Detail Possible Precursors to Loma Prieta Earthquake

A collection of new U.S. Geological Survey reports on the 1989 Loma Prieta earthquake have uncovered two surprising geologic phenomena in the months, weeks and hours preceding the 7.1-magnitude quake. The research reveals increased ultra-low-frequency magnetic noise and dramatic increases in stream flow as possibly significant precursors of the quake.

The powerful temblor provided the first opportunity in the history of fault monitoring in the United States to gather such a wide range of pre-quake data near the epicenter. The scientists hope that the observations recorded close to the epicenter in the Santa Cruz Mountains may help in the long-sought quest for ways to predict earthquakes.

Malcolm J. S. Johnston, a member of the Geological Survey staff at the agency’s West Coast headquarters in Menlo Park and the editor of the 85-page collection of 10 papers, called the magnetic noise increase “the most intriguing observations” of conditions before the Loma Prieta quake. However, he cautioned in an interview that the findings need to be confirmed by more instrument observations at other probable quake locations.

The noise was recorded by ultra-low-frequency recording instruments that had been installed at Corralitos, very close to the Loma Prieta epicenter. Johnston said that similar devices have been placed at Parkfield in Central California, where a strong earthquake is believed overdue, and at two Southern California sites, Pinyon Flats in the San Jacinto Mountains, and near San Bernardino.

The cause of the magnetic changes remains in doubt, he wrote, but could relate to several phenomena, including changes in electrical conductivity due to earth strains, hydrodynamic or gas-dynamic processes, or pressures from liquids working their way through subterranean pores.

The stream flow observation was made fortuitously by Dan Friend of Los Osos, Calif., a former park ranger who was hiking in the area an hour before the temblor struck. Friend reported a sudden three-to-five-fold increase in flow over Berry Creek Falls in Big Basin Redwoods State Park near Santa Cruz.

“Every aspect of the report that could be verified later was found to be correct,” wrote Evelyn Roeloffs, a Geological Survey scientist stationed at the Cascade Volcanic Observatory at Vancouver, Wash. Roeloffs noted that, as with many large earthquakes of the past, stream flows remained much higher than normal in the period immediately after the quake.

In an interview, Roeloffs said that just as with the magnetic noise, further study is needed to prove that the stream flow change was directly related to the earthquake. But, she observed, “if you look at the records of stream discharge in that area, the large discharges are really unique.”

The phenomenon was observed primarily in streams north of the Loma Prieta epicenter. “Probably, they come from a pressure increase” below the surface, Roeloffs said.

Roeloffs said instruments monitoring water flows have been placed in the Mojave Desert and near Parkfield in an effort to observe flows before the earthquakes expected there.

Regarding the magnetic changes, scientists wrote in the USGS report that the absence of similar magnetic anomalies before lesser quakes in the area may indicate that the magnetic fluctuations are particularly good indicators of impending large quakes.

If further research bears this out, the science of earthquake prediction could be advanced, some experts suggested.

One paper also discussed the already known fluctuations in geyser activity at Calistoga, Calif., north of San Francisco, before several large Northern California quakes, including Loma Prieta.