FAR-FIELD PULSE SHAPES FROM CIRCULAR SOURCES WITH VARIABLE RUPTURE VELOCITIES

Authors
Citation
N. Deichmann, FAR-FIELD PULSE SHAPES FROM CIRCULAR SOURCES WITH VARIABLE RUPTURE VELOCITIES, Bulletin of the Seismological Society of America, 87(5), 1997, pp. 1288-1296
Citations number
16
Categorie Soggetti
Geochemitry & Geophysics
ISSN journal
00371106
Volume
87
Issue
5
Year of publication
1997
Pages
1288 - 1296
Database
ISI
SICI code
0037-1106(1997)87:5<1288:FPSFCS>2.0.ZU;2-9
Abstract
Recently, Sate (1994) developed a simple earthquake source model of a circular rupture expanding outward from the center of a fault with con stant stress drop. In contrast to previous models, the rupture velocit y is allowed to vary over the duration of faulting. This model is used to synthesize apparent moment-rate functions for a three-stage source process: first, the rupture starts out with a gradually increasing ve locity, then, it continues to expand uniformly until, finally, it slow s to a gradual stop. Synthetic velocity seismograms are obtained from a convolution of the apparent moment-rate functions with a causal Q-op erator and an appropriate instrument response. Comparisons with an exa mple of an earthquake signal show that, in the context of the proposed model, the observed emergent P-wave onset, which is not compatible wi th a constant rupture velocity, can be explained by a gradually accele rating rupture front. Systematic departures from the generally expecte d scaling relationship between seismic moment and rupture duration are often interpreted as evidence for a dependence of stress drop on seis mic moment. However, the tradeoff between stress drop and rupture velo city inherent in all kinematic source models implies that such deviati ons can just as well be attributed to systematic variations of rupture velocity. Whereas, in general, the total duration of the far-field di splacement pulse is shorter for P waves than for S waves, the model pr edicts that the rise time, tau(1/2), of the displacement pulse should be longer for P waves than for S waves. This feature could constitute a critical test of the model and also provide a constraint on the rupt ure velocity.