HR: 16:15h
AN: S42H-02 [PDF]
TI: A Quantitative Evaluation of SCEC Community Velocity Model Version 3.0
AU: * Chen, P
EM: pochen@usc.edu
AF: Department of Earth Sciences, University of Southern California, Los Angeles, CA 90089 United States
AU: Zhao, L
EM: zhaol@usc.edu
AF: Department of Earth Sciences, University of Southern California, Los Angeles, CA 90089 United States
AU: Jordan, T H
EM: tjordan@usc.edu
AF: Department of Earth Sciences, University of Southern California, Los Angeles, CA 90089 United States
AB:
We present a systematic methodology for evaluating and improving 3D
seismic velocity models using broadband waveform data from regional
earthquakes. The operator that maps a synthetic waveform into an
observed waveform is expressed in the Rytov form $D(\omega) =
\rm{exp}[\rm{i} \omega \delta \tau_{\rm{p}}(\omega) - \omega \delta
\tau_{\rm{q}}(\omega)]$. We measure the phase delay time
$\delta\tau_p(\omega)$ and the amplitude reduction time
$\delta\tau_q(\omega)$ as a function of frequency $\omega$ using Gee
\& Jordan's [1992] isolation-filter technique, and we correct the
data for frequency-dependent interference and frequency-independent
source statics.
We have applied this procedure to a set of small events in Southern
California. Synthetic seismograms were computed using three types of
velocity models: the 1D Standard Southern California Crustal Model (SoCaL)
[Dreger \& Helmberger, 1993], the 3D SCEC Community Velocity Model, Version
3.0 (CVM3.0) [Magistrale et al., 2000], and a set of path-averaged 1D models
(A1D) extracted from CVM3.0 by horizontally averaging wave slownesses along
source-receiver paths. The 3D synthetics were computed using K. Olsen's
finite difference code. More than 1000 measurements were made on both P and
S waveforms at frequencies ranging from 0.2 to 1 Hz. Overall, the 3D model
provided a substantially better fit to the waveform data than either
laterally homogeneous or path-dependent 1D models. Relative to SoCaL, CVM3.0
provided a variance reduction of about 64% in $\delta\tau_p$, and 41% in
$\delta\tau_q$. Relative to A1D, the variance reduction is about 46% and
20%, respectively.
The same set of measurements can be employed to invert for both seismic
source properties and seismic velocity structures. Fully numerical methods
are being developed to compute the Fr\'{e}chet kernels for these
measurements [L. Zhao et. al., this meeting]. This methodology thus provides
a unified framework for regional studies of seismic sources and Earth
structure in Southern California and elsewhere.
DE: 7203 Body wave propagation
DE: 7205 Continental crust (1242)
DE: 7212 Earthquake ground motions and engineering
DE: 7260 Theory and modeling
SC: Seismology [S]
MN: 2003 Fall Meeting