HR: 1340h
AN: T23A-0528    [Abstracts]
TI: Validate Tomography with Broadband Waveform Modeling: An Example at La Ristra Transect in the Southwestern United States
AU: * Song, T A
EM: alex@gps.caltech.edu
AU: Helmberger, D V
EM: helm@gps.caltech.edu
AB: Travel time tomography has been the main tool for seismologists in developing mantle structure and studying regional tectonics. Standard practice for geodynamists is to convert the velocity anomalies into temperature and density and fit geophysical observables such as topography and gravity. However, tomographic models produced by smooth, damped inversions usually underestimate the amplitude and sharpness of the velocity structure. To validate these tomographic models, it is important to propagate seismic waves through them and compare synthetic waveforms with obvervations directly, which enable us to enhance and sharpen these models. Here we illustrate an example using the Rio Grande Rift PASSCAL observations in the southwestern US. The La Ristra passive experiment was designed to cross the Rio Grande Rift system and study the tranition in mantle structure from Great Plain to Colorado Plateau (e.g. Gao et al. 2004; West et al. 2004, Wilson et al. 2005). Ray-based body wave travel time tomography (Gao et al. 2004) indicated a linear, south-east dipping slab-like fast velocity anomaly under the western edge of the Great Plain. They interpreted it as a downwelling lithosphere produced by a small scale convection. We take advantages of the dense linear array and examine data from two deep events in south America. After deconvolving the source wavelet from the raw data, deconvolved waveforms of both events consistently show that P and S waveforms are severely distorted at stations across the slab boundary by a factor of 2-3. The waveform amplitude also diminishs in proportional to the wavefrom broadening. The waveform shape becomes simple again for stations near the center of the rift. We implement the tomography model into the 2-D finite difference scheme (Vidale, 1985; Helmberger and Vidale, 1988). Preliminary result shows that amplifying the tomography model by a factor of 3 starts to produce the waveform distortions observed.
DE: 7290 Computational seismology
DE: 8120 Dynamics of lithosphere and mantle: general (1213)
DE: 8180 Tomography (6982, 7270)
SC: Tectonophysics [T]
MN: Fall Meeting 2005