Magnetic signal preceded October quake.

Magnetic signal preceded October quake

Weeks and hours before the Loma Prieta earthquake ruptured the San Andreas fault on Oct. 17, a nearby instrument recorded highly unusual changes in the Earth's magnetic field, according to a group of atmospheric researchers led by Antony C. Fraser-Smith from Stanford University. The finding renews hope that scientists may one day provide short-term warnings of an impending earthquake.

The Stanford magnetometer monitors ultra-low-frequency (ULF) variations, between 0.01 and 10 hertz, in the intensity of the magnetic field. Ordinarily, these variations result from solar magnetic storms and other effects high in Earth's atmosphere, says Fraser-Smith.

On Oct. 5, the instrument detected an increase in ULF intensity. Then, three hours before the main shock, the signal suddenly gained strength, reaching about 30 times the normal level. Such levels were unprecedented in two years of measurement, Fraser-Smith and his colleagues reported last week at a meeting of the American Geophysical Union in San Francisco.

Fraser-Smith says the strongest signal is undoubtedly related to the quake. He speculates that water movement underground or pressure on certain minerals could have generated voltages that would alter the magnetic field in the ULF band. For years, researchers have searched for such magnetic signals before a tremor, but most have monitored other frequencies (SN: 9/12/87, p.167).

He cautions, however, that the weaker intensification 12 days before the main shock may represent something unconnected with the quake. Unusual atmospheric processes could have have caused those variations, he says, though he has not heard of any such events occurring at that time.

The magnetometer is located near Corralitos, a remote area about 7 kilometers from the epicenter of the magnitude 7.1 quake, which broke the San Andreas fault in the Santa Cruz mountains (SN: 12/9/89, p.374). The instrument's proximity to the fault was pure chance: In order to escape electromagnetic noise from cities, Fraser-Smith had moved it to a site near the home of the sister of one of his students.

The device did not record any signal preceding the magnitude 5.2 earthquake that struck the same section of the San Andreas in August. But that quake -- which released roughly one-seven-hundredth the energy of the October main shock -- may have produced a signal too weak to show up on the monitor.