Roanoke Times
Copyright (c) 1995, Landmark Communications, Inc.
DATE: FRIDAY, May 12, 1995 TAG: 9505120053
SECTION: NATIONAL/INTERNATIONAL PAGE: A-1 EDITION: METRO
SOURCE: Associated Press
DATELINE: WASHINGTON LENGTH: Medium
An earthquake may twitch before it rumbles, researchers say. And if that slight seismic signal can be detected, it might be possible to predict the big one before it happens.
Government and Stanford University seismologists who analyzed and amplified the records of ground motion from each of 30 earthquakes said they found evidence of a pattern of faint, irregular motion that immediately precedes a quake.
``An earthquake doesn't start going rapidly from the very beginning,'' Gregory Beroza, a Stanford scientist, said Thursday. ``There is a weak or tentative stage.''
Detecting this stage - from a split-second to five seconds before the actual quake - would not give time for a warning, but the discovery suggests there may be an even earlier phase that signals the start of the earthquake process.
If such early signals exist, said Beroza, there is the theoretical possibility of predicting earthquakes.
Beroza and William Ellsworth of the U.S. Geological Survey report in the journal Science that the size and duration of the slow slippage phase also may predict the magnitude of the eventual quake.
James R. Rice, a Harvard University geophysicist, said the study by Ellsworth and Beroza keeps alive the promise of one day being able to predict earthquakes.
``The fact that they see some systematic thing going on [before an earthquake] does hint that there is a preparation phase,'' Rice said. ``It does give an optimism that there is something there to be discovered.''
The research, he said, shows that an earthquake is not abrupt, but that ``there are starts and stops before it breaks out.''
In their study, Beroza and Ellsworth found cycles of slight and strengthening motion just before the main earthquake. Often this early motion is detected only faintly, and the study required that the data be amplified for the pattern to be found, they said.
``We are encouraged because we see it before every earthquake,'' Beroza said. ``There is also a lot of variation to this weak initial phase.''
Some of this premotion was a split second before the main movement, he said. For other quakes, the initial phase started and continued for up to five seconds. But the pattern was always there.
Beroza said the length and strength of this initial phase was related directly to the magnitude of the earthquake: Powerful quakes had long initial phases; weaker quakes had short, puny phases.
Ellsworth said the finding ``is a step toward answering the question about being able to predict earthquakes'' by keeping alive the hope.
``It raises that possibility, but we're a long way from that now,'' Ellsworth said. ``Whether we can ever realize it is still an open question.''
Rice said the work is valuable because it pushes back ``the terrifying'' idea that there never could be a way to predict earthquakes and that people would always be victimized by their uncertainty.
Science, which published the study, is the journal of the American Association for the Advancement of Science.
by CNB