INTEGRATING GEOLOGY, GEOPHYSICS, AND MICROMECHANICS IN THE STUDY OF SEISMIC ANISOTROPY OF THE LITHOSPHERE

Direct geologic observations are limited by the accessibility of rock exposures. Geophysical observations can provide indirect information about inaccessible parts of the lithosphere. Integrating geologic and geophysical studies has been essential to investigating the structure and deformation processes of Earth. Seismic anisotropy, which describes the directional dependence of the velocity and polarization of seismic wave propagation through an elastically anisotropic medium, is of particular interest. However, seismic anisotropy investigation requires quantitative information on the elastic stiffnesses of the lithosphere which are generally not available. On the scale of a rock sample, the elastic stiffnesses depend on the rock’s mineral composition and fabrics (both Lattice-Preferred Orientations, LPOs, and Shape-Preferred Orientations, SPOs). As the elastic property of a rock is generally anisotropic, its elastic stiffnesses are commonly represented by a fourth-order tensor (or Voigt matrix) quantity. So far, the elastic stiffnesses of a rock are estimated from the elastic properties of the constituent minerals by simplistic averaging based on the Voigt and Reuss methods. These methods are known to provide only upper and lower bounds of the stiffnesses which can be far different from the actual elastic stiffnesses. We have developed more rigorous homogenization methods, based on continuum micromechanics, to numerically compute the bulk elastic stiffnesses from the elastic stiffnesses and the fabric data (LPOs and SPOs) of the constituent phases. Earth’s lithosphere is made of heterogeneous materials, and the bulk elastic stiffness tensor depends on the selection of a Representative Volume Element (RVE) that is relevant for seismic anisotropy study. Where the RVE is beyond a rock sample, variations in rock types and fabrics emerging at that RVE scale will affect the elastic stiffnesses. Multiscale structural geology studies are critical for homogenization at various RVE scales – structural analysis results are used to build numerical models for the RVE (composition of distinct phases, their shapes, preferred orientations, etc.) from which homogenization is performed.