Seismic Imaging in California: A Combination of Classic and State-of-The-Art Techniques

Abstract:

Seismic velocity and attenuation of P- and S-waves (Vp, Vs, Qp, and Qs) and their respective ratios provide essential constraints on Earth properties because of their sensitivity to rock composition, fluid content, thermal effect, and other factors. Seismic tomography has been an important tool for the determination of three-dimensional (3-D) velocity structure. However, attenuation inversion has generally fallen behind velocity studies mainly because of the data scattering, inversion complexity and interpretation difficulty. Here, I combine the recent results of attenuation tomography with the existing velocity models for the interpretation of structural heterogeneity in both northern and southern California. I focus the attenuation study on frequency-independent 3-D compressional wave attenuation model (indicated by the reciprocal of quality factor Qp) for the crust and uppermost mantle in the study areas. The classic simul2000 tomographic algorithm is applied to invert the frequency-independent attenuation operator t* values measured from amplitude spectra of P-wave arrivals through recently developed 3-D velocity models and earthquake relocation catalogs. In general, the resulting Qp values increase with depth and agree with the surface geology at shallow depth layers. I also apply a state-of-the-art technique to estimate in situ Vp/Vs ratios for similar event clusters by using P- and S-wave differential times from waveform cross-correlation. This approach provides highly precise results for near-source regions because cross-correlation can measure differential times to within a few milliseconds and can achieve a precision of 0.001 in the estimated Vp/Vs ratio, corresponding to about 0.0004 in Poisson’s ratio. The availability of 3-D attenuation models and in situ Vp/Vs ratios provides a complementary definition of the fault systems in California to the existing velocity models.