Science / Medicine : Scientists Await Overdue Quake : Seismology: Millions have been spent setting up and maintaining instruments at the Central California site of regular temblors. But the quake hasn’t come, and some experts ask whether the continued cost is justified.
Almost like clockwork, about every 22 years a moderately strong earthquake has struck the Central California community of Parkfield, and scientists seized upon that pattern several years ago to create the most sophisticated natural seismological laboratory in the world.
Today, no other spot on the planet is as thoroughly instrumented as Parkfield. The hills around the rural hamlet are covered with instruments that can measure even the slightest change in the Earth’s crust along the San Andreas Fault, which bisects the tiny town, and seismologists around the world are waiting to learn what happens just before, during and after an earthquake. It is an extraordinary opportunity, because it means that for the first time scientists will be able to analyze an earthquake almost as if they had created it in their labs.
The last quake hit in 1966, so seismologists had expected one to strike around 1988, give or take a few years.
But so far, zilch.
Now a lot of scientists are wondering if Parkfield is as reliable as they had thought.
“We could be off as much as a decade,” said one top official with the U.S. Geological Survey, which is funding most of the experiments that have been set up around Parkfield.
John Langbein of the agency’s Menlo Park office, the scientist in charge at Parkfield, said the original forecast in April, 1985, predicted that the earthquake would strike sometime between 1985 and 1993, so there is still time left in the window.
“But it’s pretty quiet,” said Langbein. “There’s nothing happening.”
In fact, the only excitement so far was a false alarm last March when instruments detected movement along the fault, called “creep events,” thus activating an alarm that sent scientists racing to Parkfield. But it turned out that the movement was caused by heavy rains in the area.
“That was the only serious one (alert) we’ve had,” Langbein said. “It was a learning experience.”
Several million dollars have been spent setting up and maintaining instruments around Parkfield, and some scientists are beginning to wonder how long they can afford to hold on if something doesn’t happen soon. Langbein said the Geological Survey will continue to support experiments that are already there, but he does not expect new projects to be started.
“We’re in a holding pattern,” he said. “Mostly, the funding today is just for maintenance. Trying to deploy new experiments would be very expensive.”
Usually, no one knows when or where an earthquake is most likely to occur, so the Parkfield area offered scientists an opportunity to set up their instruments with some assurance that the quake would strike reasonably soon. That would give them a clearer understanding of exactly what kind of pre-quake signals the Earth may emit in the form of crustal deformation, creep along the fault, changes in the depth of the water table or possibly even radio emissions.
A better understanding of what happens at Parkfield might help them predict earthquakes when similar patterns are found elsewhere. However, some scientists doubt that the project will make the general field of earthquake prediction any easier because earthquakes differ so much from one area to the next that what happens at Parkfield may not be repeated elsewhere. Still, many believe it is the best show in the 34-person town. In fact, it’s the only show.
It is not practical to set up experiments all along the San Andreas and wait for decades for an earthquake, so Parkfield, with its history of producing earthquakes of 5.5 to 6 magnitude on an average of every 22 years, at least gives scientists a chance of recording every aspect of the event before the quake strikes.
So no other spot on the planet is as fully examined as Parkfield. There are 18 state-of-the-art seismometers within a dozen miles to record the rate of ground motion during the earthquake. These instruments are linked with computers in Menlo Park so that scientists can be notified within three to five minutes that a shock as small as magnitude 0.8 has struck.
Additional seismometers have been installed in wells around Parkfield, including 116 in one well, which will detect even smaller tremors. Scientists expect the main shock to be preceded by swarms of smaller quakes.
In expectation that the fault will slip slightly before the quake, 13 “creepmeters” have been installed, and they relay data to Menlo Park every 10 minutes via satellite or telephone.
One method for determining whether an earthquake is imminent is to measure the strain in the rocks along the fault, and that can be done in several ways. When a rock is stressed almost to the point of breaking, microcracks and voids occur and the rock increases in volume. Seven “dilatometers” have been installed at Parkfield to measure such changes and determine the strain.
Four “tiltmeters” have been placed in shallow wells around Parkfield to determine if the ground is tilted slightly as strain in the system accumulates.
Scientists believe that before an earthquake strikes, strain forces some geological formations to expand, thus pushing the water table toward the surface. Water levels are determined every 15 minutes.
Magnetometers have been placed at seven sites in the area to detect changes in the Earth’s magnetic field caused by stress-induced changes in the crust. Other experiments will measure changes in how an electric current flows through rocks just before the quake, and other instruments will look for electromagnetic emissions.
Some scientists believe geochemical changes precede an earthquake, so the soil around Parkfield is being monitored for changes in the level of radon and hydrogen gases.
Scientists also want to know more precisely how the soil behaves during an earthquake, because the ability of any structure to withstand the shock depends largely on what happens to the soil upon which it stands. Some soils, for example, become liquefied during a quake, causing buildings to topple. This phenomenon accounted for much of the damage in the Loma Prieta earthquake that ravaged the San Francisco Bay Area in 1989.
For years after the quake strikes, experts will analyze the data to see how structures might be designed to better withstand ground failure during an earthquake.
But scientists need Parkfield to come through with its historic promise.
Parkfield, however, is proving to be fickle.
More than a year ago, the region had grown so seismically quiet that scientists believed the San Andreas was storing up its energy and they would not have to wait much longer. Some went so far as to make a formal prediction.
Scientists at the University of Colorado claimed in an article in the British journal Nature that Parkfield would surely shimmy within two years. They even estimated the month the quake most likely would occur:
March, 1991, nearly four months ago.
Plotting the Pattern
The tiny Central California community of Parkfield, which straddles the San Andreas Fault, has experienced moderate earthquakes with some regularity for more than a century. Seismologists from around the world have wired the area with equipment, hoping to learn whether there are seismic warning signals before a quake.
Parkfield, a town of 34 residents, is located on the San Andreas Fault in the Cholame Valley, midway between Los Angeles and San Francisco.
The Cycle A powerful earthquake occurs on a segment of the San Andreas near Parkfield about every 22 years. The U.S. Geological Survey believes a quake of magnitude 6 or larger will hit before 1993. Some quake history: Year: Magnitude* Jan. 9, 1857: 6.0 Feb. 2, 1881: 6.0 March 3, 1901: 6.0 June 8, 1934: 6.0 June 28, 1966: 6.0 * estimated magnitudes in some cases The Equipment No other spot on the planet is as thoroughly instrumental as Parkfield. There are 18 state-of-the-art seismometers within a dozen miles of Parkfield to record the rate of ground motion during an earthquake. These instruments are linked with computers in Menlo Park so scientists can be notified within three to five minutes that a shock as small as magnitude 0.8 has struck. Among the instruments in Parkfield: Two-Color Laser A. Sends out laser beams to 18 reflectors distributed radially over a several-mile area around Car Hill. B. Reflectors mounted on a tripod reflect the laser beam back to the instrument. C. The amount of time it takes for the laser beam to return is compared to previous times, enabling seismologists to make precis measurements of movement in the Parkfield area.
