HR: 10:35h
AN: S31G-02 INVITED     [PDF]
TI: Conditional Probabilities for Earthquake Triggering in Global and Regional Datasets
AU: * Main, I
EM: ian.main@glg.ed.ac.uk
AF: University of Edinburgh, School of Geosciences West Mains Road, Edinburgh, EH9 3JW United Kingdom
AU: Huc, M
EM: mhuc@nordnet.fr
AF: University of Edinburgh, School of Geosciences West Mains Road, Edinburgh, EH9 3JW United Kingdom
AU: McKernon, C
EM: conor.mckernon@glg.ed.ac.uk
AF: University of Edinburgh, School of Geosciences West Mains Road, Edinburgh, EH9 3JW United Kingdom
AB: We define earthquake triggering as the existence of a pair of similar sized events that are close in position and time, above chance occurrence in time, but allowing for the spatial correlation of epicentres. The global CMT catalogue contains a finite triggering signal with a similar correlation function to that predicted in a critical point system. No strong directional dependence is found, but the Coulomb triggering hypothesis cannot be tested quantititively due to the fault plane ambiguity in many of the focal mechanisms. The correlation length and mean triggering distance are on the order 10-20km, similar to the seismogeneic thickness, and increase only gradually with time after a given event, in the form of a power law of exponent H $\sim 0.06$. This compares with H $\sim 0.5$ expected for Fickian stress diffusion, but is quantitatively consistent with delayed stress corrosion failure in a random heterogeneous medium. The regional study is based on ISC data with a smaller magnitude threshold. We do not detect a strong bias in the results based on the choice of region for concentric circles centred on the Gulf of Corinth area as a test case. The results of an analysis of the most active Flinn-Engdahl-defined regions are consistent with the model formed from the global study, but show significant regional variation in the model parameters that depend on the local tectonic style. The parameters of the model may be used directly to calculate a conditional probability of further similar-sized events, once one has occurred.
DE: 3220 Nonlinear dynamics
DE: 7209 Earthquake dynamics and mechanics
DE: 7223 Seismic hazard assessment and prediction
DE: 7230 Seismicity and seismotectonics
SC: Seismology [S]
MN: 2003 Fall Meeting