This chapter discusses seismic anisotropy in the mantle, highlighting mechanisms, observations and interpretations. Although evidence of anisotropy in the Earth is now commonplace, it can be difficult to uniquely identify its underlying cause. Part of the difficulty is due to the nature of wave propagation in anisotropic media; intuition based on wave propagation in isotropic media can be misleading. Observations suggest that anisotropy is not uniform throughout the mantle, but appears to be concentrated in the boundary layers of the convecting mantle where deformation is largest. Studies of surface waves and shear-wave splitting in phases like SKS have provided abundant evidence of anisotropy in the uppermost boundary-layer. In continental settings, the magnitude of the anisotropy is quite variable and its orientation mimics trends in surface geology, suggesting the crust and mantle are coupled during orogenic deformation. In contrast, upper-mantle anisotropy in oceanic settings appears to be quite consistent over large regions and reflects large-scale flow patterns in the mantle. Deeper in the mantle, there is some evidence of anisotropy in the upper-mantle transition zone suggesting that in places a boundary layer may exist between the upper and lower mantle. The bulk of the lower mantle seems to be isotropic except for the Dʺ region, the lowermost boundary layer. Here the anisotropy exhibits regional variations on scales similar to those in the lithosphere. As it appears that there are regions of azimuthal anisotropy in Dʺ, there is a potential problem separating the effects of upper-mantle anisotropy from lower-mantle anisotropy in data interpretations. At present, observations of seismic anisotropy lack global coverage, but as this improves we are faced with the challenging task of including anisotropy in global seismic models.
