HR: 15:25h
AN: S43C-08    [Abstracts]
TI: Stochastic Description of Seismic Anisotropy in the Lithosphere and Upper Mantle
AU: * Browaeys, J T
EM: browaeys@usc.edu
AF: University of Southern California, Department of Earth Sciences, 3651 Trousdale Parkway, Los Angeles, CA 90089-0740
AU: Becker, T W
EM: twb@usc.edu
AF: University of Southern California, Department of Earth Sciences, 3651 Trousdale Parkway, Los Angeles, CA 90089-0740
AU: Jordan, T H
EM: tjordan@usc.edu
AF: University of Southern California, Department of Earth Sciences, 3651 Trousdale Parkway, Los Angeles, CA 90089-0740
AB: Shear wave splitting data recorded at the Earth surface sometimes appear to be spatially variable, even at a regional scale. We attempt here to extract the characteristic parameters of the anisotropy heterogeneity by using parametric statistics. A suitable two-point correlation function was introduced by Von Karmàn (1948) for the characterization of a random velocity field in a turbulent fluid. This function has since been used with success for random fields implied in wave scattering theoretical studies (Chernov, 1960) and to describe the seafloor topography (Goff & Jordan, 1988). The covariance function depends on the distance r between two points and is of the form r^νK_ν(r) where K_ν(r) is the modified Bessel function of the second kind and ν lies in [0,1]. This random field has a Hausdorff (fractal) dimension of 4-ν at small scale. The statistical description for our problem is derived from the stochastic modeling of small scale anisotropic structures in three dimensions with hexagonal symmetry. Random fields are produced by a Gaussian probability combined with the previous correlation function. The model is characterized by the horizontal wave number of the heterogeneity, the aspect ratio of the anisotropy, the aspect ratio of the heterogeneity and the fractal dimension of the field. In the limit of a stochastic horizontal laminate, this description produces the second-order approximation of Backus (1962) for a layered medium. To inspect the homogeneity of the shear wave splitting records, the rms angular difference depending on the distance between two stations is calculated. This approach is applied to the Western US which provides a statistically significant amount of seismic data to retrieve the parameters of the distribution heterogeneity. The typical range of the horizontal correlation length for the splitting directions is a hundred of kilometers, corresponding to the dimensions of the different tectonic settings. A local correlation between the shear wave splitting intensity and the directions variability is found as expected by the theory. References: Backus, G.E. Long-wave elastic anisotropy produced by horizontal layering. J. Geophys. Res., 67, 4427--4440, 1962.
Chernov, L.A. Wave Propagation in a Random Medium. 168 pp., McGraw-Hill, New-York, 1960.
Goff J.A. and T.H. Jordan. Stochastic Modeling of Seafloor Morphology: Inversion of Sea Beam Data for Second-Order Statistics. J. Geophys. Res., 93, B11, 13,589--13,608, 1988.
Von Karmàn, T. Progress in the statistical theory of turbulence. J. Mar. Res., 7, 252--264, 1948.
DE: 1236 Rheology of the lithosphere and mantle (7218, 8160)
DE: 7203 Body waves
DE: 7205 Continental crust (1219)
DE: 7208 Mantle (1212, 1213, 8124)
DE: 7260 Theory
SC: Seismology [S]
MN: Fall Meeting 2005