HR: 08:00h
AN: S31A-01 INVITED [PDF]
TI: A Fresh Look at the Triggering of Earthquake Pairs, Such as the Landers-Big Bear, Landers-Hector Mine,
Izmit-Duzce, and Nenana-Denali, and March-May 1997 Kagoshima Events
AU: * Toda, S
EM: s-toda@aist.go.jp
AF: Active fault Research Center (AIST), Higashi 1-1, Tsukuba, CA 305-8567
Japan
AU: Stein, R S
EM: rstein@usgs.gov
AF: U.S. Geological Survey, MS 977, Menlo Park, CA 94025 United States
AB:
Recent precise hypocentral information provides opportunities to study how secondary earthquakes are spatially and temporally
triggered by mainshocks. Most studies regard the second shock in the pair as an aftershock triggered exclusively by the
mainshock, with its likelihood decreasing with time after the first event. Here we argue that small aftershocks occurring
near the future hypocenter of the second shock play an important role in enhancing the likelihood of the pair. For such
multiple-order triggering calculations, we incorporate ate/state friction of Dieterich (1996) into stress transfer. The
results may answer two riddles: The first is to ask why there is a net probability gain during the aftershock sequence when
there must be equal areas of stress enhancement (trigger zones) and stress depression (shadows) near the mainshock. The Omori
law describes only the behavior in the trigger zones. The answer appears to be that the net probability gain occurs because
the seismicity rate increases exponentially in response to the stress change, so the seismicity rate gain in the trigger
zones dwarfs the seismicity rate drop in the shadows. The second riddle is to ask how small aftershocks triggered by the
mainshock affect the likelihood of the second large earthquake in the pair. The answer is that because the state variable
(gamma) in the Dieterich earthquake formulation has already plummeted as a result of the mainshock, the effect of small
aftershocks on the nearby seismicity rate is highly amplified relative to the effect of the mainshock, even if the mainshock
and the small aftershock change the stress by the same amount. The implication is that that probability of the second shock
is not necessarily highest immediately after the mainshock, as we typically assume. Instead, it might be highest when the
distribution of first-order aftershocks expands (as also predicted by rate/state friction) to a potential earthquake source,
such as an active fault. A more complete probability calculation would thus need to include the effects of smaller
aftershocks, as advocated by Felzer et al (2002), a very challenging task.
UR: http://quake.wr.usgs.gov/research/deformation/modeling/refs/ross_refs.html
DE: 7209 Earthquake dynamics and mechanics
DE: 7223 Seismic hazard assessment and prediction
DE: 8123 Dynamics, seismotectonics
SC: Seismology [S]
MN: 2003 Fall Meeting