FREQUENCY-DOMAIN INVERSION OF STRONG MOTIONS - APPLICATION TO THE 1992 LANDERS EARTHQUAKE

Citation
F. Cotton et M. Campillo, FREQUENCY-DOMAIN INVERSION OF STRONG MOTIONS - APPLICATION TO THE 1992 LANDERS EARTHQUAKE, J GEO R-SOL, 100(B3), 1995, pp. 3961-3975
Citations number
38
Categorie Soggetti
Geosciences, Interdisciplinary
Journal title
JOURNAL OF GEOPHYSICAL RESEARCH-SOLID EARTH
ISSN journal
21699313 → ACNP
Volume
100
Issue
B3
Year of publication
1995
Pages
3961 - 3975
Database
ISI
SICI code
2169-9313(1995)100:B3<3961:FIOSM->2.0.ZU;2-D
Abstract
We present a frequency domain inversion in which the observed earthqua ke strong ground motions are used to constrain the space-time dependen ce of slip on a fault. Green's functions are numerically evaluated and the parameters describing the rupture are the local slip, rupture tim e and rise time. These parameters are simultaneously evaluated without additional constraints. This procedure allows for large variations in the local rupture velocity. The June 28, 1992 Landers earthquake (M(W ) = 7.3) offers an exceptional opportunity to apply this technique to a major strike-slip event. We model the rupture evolution, including l ocal differences in slip durations and variations in rupture velocity. Our final results are in good agreement with other inversion studies, geodetic and surface observations. The main discrepancies occurred at depth and at the end of the Johnson Valley fault. We show that a relat ively low resolution could be an explanation for these differences. Ru pture velocity and slip are extremely heterogeneous, both along strike and with depth. A moment of 0.90 x 10(20) N m was found. The slip dis tribution shows that this event consists of a series of regions of hig h slip (subevents) separated by regions of relative low slip. Approxim ately 50% of the moment was released on the Homestead Valley fault; in this region of large slip, the rupture velocity inferred by our inver sion is well constrained and is equal to 3.0 km/s at depth and 2.5 km/ s near the surface. Our inversion favors the hypothesis that the durat ion of the slip at each point of the fault is of the order of the dura tion of rupture of each subevent.