Structure and dynamics of the Earth's inner core

Abstract:

Seismic observations, using both body wave and normal mode data, provide strong evidence that the Earth's inner core is anisotropic (e.g. Morelli at al, 1986, Woodhouse et al, 1986), finding cylindrical anisotropy with the fast axis aligned with the Earth's rotation axis. More recently, hemispherical anisotropy variations have been found in both body wave and normal mode data, with the west being more strongly anisotropic than the east (e.g. Tanaka & Hamaguchi, 1997, Deuss et al, 2010). However, it has proven difficult to reconcile both data types. Here, we try to reconcile body waves and modes, making a comprehensive model.

We find that a top layer with isotropic velocity of about 70km thick fits both the body wave and normal mode data. Hemispherical variations in isotropic velocity, with the west being slow and the east being fast, exist in the top 275 km of the inner core. Strong anisotropy starts below 70km depth, and only exists in a wedge in the western hemisphere, increasing in strength with depth. The location of the anisotropic wedge seems unrelated to the isotropic hemispheres found at the top of the inner core.

The isotropic hemispheres are most likely due to variations in solidification rate at the inner core boundary. The deeper anisotropic alignment of crystals is most likely due to a deformation process. Recent work fitting normal mode frequencies suggests that the top layer of the inner core may in fact be radially anisotropic with the fastest velocity perpendicular to the inner core boundary (Lythgoe & Deuss, 2015). Such radial anisotropy would be in line with the model proposed by Yoshida et al (1996) where the inner core grows preferentially near the equator, leading to a flow from the equator to the poles generating both the shallow radial anisotropy and deeper cylindrical anisotropy.