San Francisco--Swarms of barely perceptible tremors could provide the best glimpse yet into deformations in the Earth's deep crust--the root cause of earthquakes--two seismologists announced here this week at the annual meeting of the American Geophysical Union. They have found that these microearthquakes sometimes travel in waves and suggest that the waves could eventually provide a warning signal of larger quakes to come.
The new finding stems from studies of a stretch of the San Andreas fault near Parkfield in southern California, perhaps the most closely monitored seismic hot spot in the world. Peter Malin and Eylon Shalev of Duke University examined the record of slight twitches with magnitudes of 0.1 which give off the energy of just one stick of dynamite. They found that during two periods of heightened seismic activity in 1990 and 1993, these microearthquakes rippled along the fault in waves, at the snail's pace of about 30 kilometers per year.
Malin and Shalev used a computer to simulate the seismic processes that might lead to these microearthquakes. Their scenario suggests that the waves are generated by the motion of the crust at depths greater than 12 kilometers, where the rock is warm enough to deform ductilely, like a piece of pastry dough. This deformation is transferred to the brittle crust above it, which consists of fractured blocks that grind together and shudder as they move around. Malin and Shalev have proposed that porous, fluid-filled pieces of rock between these blocks dampen the jostling and transfer the motion from one block to the next, generating swarms of microearthquakes. When one block--strained by the motion of the lower crust--suddenly moves, it knocks into its neighbor and passes the momentum down the chain of blocks, creating a microearthquake at each step. Malin likens this effect to the transfer of signals from car to car in a traffic jam. "Imagine someone in front of you slamming on the brakes," he says. "You'd see a wave of red lights coming at you."
If the mechanism is indeed correct, the researchers say they could figure out how much strain might be producing measurable swarms of microearthquakes. Strainmeters in use at present only measure deformation near the Earth's surface. Tracking more subtle temblors, therefore, might someday lead to methods to locate regions of heavy seismic strain that could result in larger earthquakes, the scientists say.
But prediction is still far off: Neither wave of microearthquakes observed at Parkfield ended with a large earthquake. Says Steve Miller, a seismologist at the Geology Institute of Eth-Zentrum in Zurich, Switzerland, who studies earthquakes at Parkfield, "It's an interesting result, but not the end of the story."