Friday, November 21, 2008
Can Earthquakes Be Predicted?
U.S. Geological Survey, Pasadena, CA
Earthquake prediction has long been the holy grail of earthquake seismology. Much research and money has been spent in its pursuit, using methods from seismicity pattern recognition to satellite-based earth imaging. But with the exception of a single claimed success in China, earthquake prediction in the classic sense has never been achieved. Is this because the problem is really difficult, or because it is impossible? Meaningful prediction will always be impossible if earthquakes propagate via a stochastic growth process such that earthquakes of all magnitudes occur under the same initial conditions and the magnitude of any earthquake cannot be known until the earthquake is over. This process may be compared to flipping a row of coins; getting 2 heads in a row (small earthquake) vs. getting 20 heads in a row (large earthquake) occurs under the same conditions and cannot be predicted until the flipping has taken place. I will argue that the preponderance of evidence supports the stochastic growth model, and thus the ultimate futility of earthquake prediction.
Refreshments at 4:00 PM. Seminar begins at 4:10 PM.
Building 8, Room 241
For further information, please call (909) 869-4014