HR: 0830h
AN: S21E-0355 [PDF]
TI: Three-Dimensional Velocity Structure in Southern California from Teleseismic Surface Waves and Body
Waves.
AU: * Prindle-Sheldrake, K L
EM: kenton_prindle@umail.ucsb.edu
AF: Department of Geological Sciences, UC Santa Barbara-Building 526, Santa Barbara, ca 93106-9630 United States
AU: * Prindle-Sheldrake, K L
EM: kenton_prindle@umail.ucsb.edu
AF: Institute for Crustal Studies, 1140 Girvetz Hall
UC Santa Barbara, Santa Barbara, ca 93106-1100 United States
AU: Tanimoto, T
EM: toshiro@geol.ucsb.edu
AF: Department of Geological Sciences, UC Santa Barbara-Building 526, Santa Barbara, ca 93106-9630 United States
AU: Tanimoto, T
EM: toshiro@geol.ucsb.edu
AF: Institute for Crustal Studies, 1140 Girvetz Hall
UC Santa Barbara, Santa Barbara, ca 93106-1100 United States
AB:
Analysis of teleseismic waves generated by large earthquakes worldwide across the Southern California TriNet Seismic
Broadband Array has yielded high quality measurements of both surface waves and body waves. Rayleigh waves and Love waves
were previously analyzed using a spectral fitting technique (Tanimoto. and Prindle-Sheldrake, GRL 2002; Prindle-Sheldrake and
Tanimoto, submitted to JGR), producing a three-dimensional S-wave velocity structure.
Features in our velocity structure show some regional contrasts with respect to the starting model (SCEC 2.2), which has
detailed crustal structure, but laterally homogeneous upper mantle structure. The most prominent of which is a postulated
fast velocity anomaly located west of the Western Transverse Ranges that could be related to a rotated remnant plate from
Farallon subduction. Analysis indicates that, while Rayleigh wave data are mostly sensitive to mantle structure, Love wave
data require some modifications of crustal structure from SCEC 2.2 model.
Recent advances in our velocity structure focus on accommodation of finite frequency effect, and the addition of body
waves to the data. Thus far, 118 events have been analyzed for body waves.
A simple geometrical approach is used to represent the finite frequency effect in phase velocity maps. Due to concerns
that, for seismic phases between 10-100 seconds, structure away from the ray theoretical is also sampled by a propagating
surface wave, we have adopted a technique which examines a normal mode formula in its asymptotic limit (Tanimoto, GRL 2003 in
press). An ellipse, based on both distance from source to receiver and wavelength, can be used to approximate the effect on
the structure along the ray path and adjacent structure. Three models were tested in order to select the appropriate
distribution within the ellipse; the first case gives equal weight to all blocks within the ellipse; case 2 incorporates a
Gaussian function which falls off perpendicular to the ray path, allowing the amplitude to peak at the receiver; case 3 is
the same as case 2, yet removes the effect of the peak at the receiver. A major improvement is that the locale under
consideration has expanded due to the effect of ray paths spreading over a larger area than the ray theoretical. Comparison
of the three techniques yields very similar results, and all techniques show an exceptional correlation to the ray
theoretical phase velocity maps. After analyzing our data in terms of the finite frequency effect, we find that little
change has occurred as a result of employing this technique other than expanding our region of study.
P-wave measurements were obtained from the data set for 118 events. Preliminary results show systematic patterns. We
have successfully measured 30 S-wave events which we plan to incorporate into our velocity structure. Our goal is to proceed
with a joint inversion of P-waves, S-waves and Surface waves for a collective Southern California velocity structure.
UR: http://www.geol.ucsb.edu/projects/seismic
DE: 7205 Continental crust (1242)
DE: 7207 Core and mantle
DE: 7218 Lithosphere and upper mantle
DE: 7255 Surface waves and free oscillations
DE: 7260 Theory and modeling
SC: Seismology [S]
MN: 2003 Fall Meeting