The existence of radial anisotropy in Earth's upper inner core revealed from seismic normal mode observations

Abstract

As we strive to understand the most remote region of our planet, one critical area of investigation is the uppermost inner core since its structure is related to solidification of outer core material at the inner core boundary (ICB). Previous seismic studies have used body waves to show that the top ∼100 km of the inner core is isotropic. However, radial anisotropy cannot be uniquely determined by body wave observations. Alternatively, normal mode center frequencies are sensitive to spherically symmetric Earth structure, therefore may provide a constraint on the existence of radial anisotropy in the inner core. Here we show that normal mode center frequency measurements are compatible with 2–5% radial anisotropy in the top ∼100 km of the inner core with a fast direction radially outward and a slow direction along the ICB. Given the uncertainties in the mineral physics and processes that produce anisotropy, the observed radial anisotropy may be reconciled with predictions based on either solidification processes or from texturing due to anisotropic growth.

Anisotropy at the Inner Core Boundary

“…Hence, the effect of the anisotropy on travel times would be undetectable. Lythgoe and Deuss (2015) also noted this in their study of center frequency shifts of inner core-sensitive normal modes.…”

“…The form of anisotropy we chose, VTI, yields PKPdf travel times in the shallow inner core that do not vary with spin axis orientation. Hence, the ICB surface that, to travel times, appears to be isotropic (Garcia & Souriau, 2000;Ozounis & Creager, 2001) might not in fact be so, an insight that motivated Lythgoe & Deuss's (2015) study. Rather, solid crystallites might be oriented by compaction, flow, or electromagnetic forces uniformly upon adhesion to the ICB that could impart a uniform VTI fabric.…”

“…We also note that these are the best fit values, but anisotropy levels as low as 5% also rank among the lowest 0.1% of the misfit values in the grid search range explored. An adequate model could have significantly lower anisotropy than 10%, in the 2-5% range compatible with inner core normal mode frequency shifts (Lythgoe & Deuss, 2015).…”

Effective‐Medium Models of Inner‐Core Anisotropy

“…The inner core is radially layered with a zone ~100‐km thick near its surface, where the axial anisotropy may be weaker (Ouzounis & Creager, ; Song & Helmberger, ; Waszek & Deuss, ) or more radially oriented (Lythgoe & Deuss, ), and an innermost inner core (IMIC) with a radius of 300–600 km (Beghein & Trampert, ; Ishii & Dziewoński, ), where the anisotropy may be stronger (Romanowicz et al, ; Sun & Song, ) and perhaps tilted with respect to the rotation axis (Niu & Chen, ; Wang et al, ). Inner‐core structure also varies with longitude (Creager, ; Irving et al, ; Tanaka & Hamaguchi, ), having an eastern hemisphere that shows substantially less anisotropy than its western hemisphere.…”

Grain Boundary Sliding in High‐Temperature Deformation of Directionally Solidified hcp Zn Alloys and Implications for the Deformation Mechanism of Earth's Inner Core

“…Since then, the inferred pattern of anisotropy has grown more complex, to include depth variations (Shearer, 1994), hemispherical variations (Tanaka & Hamaguchi, 1997), and regional variations (Koper et al, 2004). Some studies (Shearer, 1994) have found that the top of the inner core is isotropic, but Lythgoe and Deuss (2015) have instead suggested radial rather than cylindrical anisotropy. There are also suggestions of an innermost inner core with a different axis of anisotropy (Ishii & Dziewonski, 2003).…”