Seismic anisotropy and mantle deformation: What have we learned from shearwave splitting?

Shear wave splitting measurements now allow us to examine deformation in th e lithosphere and upper asthenosphere with lateral resolution <50 km. In an anisotropic medium, one component of a shear wave travels faster than the orthogonal component. The difference in speed causes the waves to separate; this phenomenon is called shear wave splitting. The polarization of the fa st: component and the time delay between the components provide simple meas urements to characterize the anisotropy. Strain aligns highly anisotropic o livine crystals in the mantle, which is the most likely cause of splitting measured from records of distant earthquakes. The seismic community is in t he fundamental stages of determining the relations between strain and aniso tropy, measuring anisotropy around the world, and determining how much is f ormed by past and present lithospheric deformation and how much is formed b y crustal;Ind asthenospheric sources. The mantle appears isotropic between 600 km depth and the D " layer at the top of I:he core-mantle boundary. She ar wave anisotropy of up to 4% is ubiquitous in the upper 200 km of the cru st and mantle. Evidence for stronger and deeper anisotropy is less common. Anisotropy in the transition zone between 400 and 600 km and in the D " lay er may be patchy. Transcurrent deformation at plate boundaries appears to b e one of the best mechanisms for causing splitting on nearly vertically tra veling waves by aligning foliation planes and the fast axes of olivine with in the lithosphere parallel to the boundary and in the most favorable orien tation for splitting. Similar deformation may also contribute to anisotropy observed at convergent margins. Shear wave splitting data are challenging conventional beliefs about mantle flow. Simple models of asthenosphere dive rging at spreading centers and flowing downward beneath subduction zones ap pear to be only part of the story, with significant components of flow para llel to ridges and trenches. Parallelism between fast polarizations of wave s passing through the deep mantle beneath cratons and surficial geological strain indicators has been used to suggest that the mantle at depths of sev eral hundred kilometers beneath the cratons may have been stable since the initial deformation in the Archean. New paths of investigation include test ing a wider range of anisotropic symmetry systems and more complicated mode ls by examining variations in splitting as a function of earthquake arrival angle and distance and by numerical modeling of waveforms and of proposed deformation scenarios.