June 12, 2006
New study shows earthquake shaking triggers aftershocks
By Tim Stephens
A new analysis of earthquake data indicates that aftershocks are triggered by the shaking associated with the mainshock, rather than by the added stress on nearby faults resulting from rearrangement of the Earth's crust.
The triggering of aftershocks by shaking may seem obvious, but is in fact a surprising result, said Emily Brodsky, assistant professor of Earth sciences at UCSC.
"The problem is that it's not clear how shaking can trigger an aftershock that doesn't happen right away, but happens a day or two after the earthquake. That's why most seismologists have thought that aftershocks are triggered by static stress resulting from the movement of the crust," Brodsky said.
Brodsky is coauthor of a paper describing the new findings in the June 8 issue of Nature. The first author of the paper is Karen Felzer, who began work on the study as a postdoctoral researcher with Brodsky at UCLA and is now with the U.S. Geological Survey in Pasadena.
Felzer and Brodsky looked at the distribution of aftershocks in relation to their distance from the site of the mainshock. They observed a smooth, consistent trend, with the number of aftershocks falling off steeply with increasing distance from the mainshock over a range from 0.2 to 50 kilometers (0.12 to 30 miles).
The smooth trend suggests that the same triggering process is operating over the entire distance range. But static stress is negligible at the far end of the range, so the dynamic stress from shaking must be the trigger, Felzer said.
"No one expected small earthquakes to trigger aftershocks at these distances," she said. "The traditional idea is that the aftershock zone is one to two times the length of the fault rupture, so for earthquakes of this size you wouldn't expect to see aftershocks beyond more than one kilometer. We're seeing aftershocks all the way out to 50 kilometers."
Furthermore, the aftershocks fall off in the same relation to distance as is seen in the decay of seismic waves. In other words, the number of aftershocks and the amount of shaking show the same mathematical relation to distance from the mainshock (an "inverse power law" relation).
"That's the kicker. The aftershocks fall off with distance in the same way that seismic waves do," Brodsky said. "We propose that the chance of having an aftershock depends directly on the amplitude of the shaking."
This hypothesis is consistent not only with the researchers' measurements of how aftershock density varies with distance, but also with previous measurements of the number of aftershocks triggered by a mainshock of a particular magnitude, Brodsky said.
The data analyzed in this study were obtained from a large catalog of southern California earthquakes with precise earthquake locations, published in 2005. This research was supported in part by a grant from the National Science Foundation.