Seismic anisotropy in the Earth's innermost inner core: Testing structural models against mineral physics predictions

Abstract

Using an updated data set of ballistic PKIKP travel time data at antipodal distances, we test different models of anisotropy in the Earth's innermost inner core (IMIC) and obtain significantly better fits for a fast axis aligned with Earth's rotation axis, rather than a quasi‐equatorial direction, as proposed recently. Reviewing recent results on the single crystal structure and elasticity of iron at core conditions, we find that an hcp structure with the fast c axis parallel to Earth's rotation is more likely but a body‐centered cubic structure with the [111] axis aligned in that direction results in very similar predictions for seismic anisotropy. These models are therefore not distinguishable based on current seismological data. In addition, to match the seismological observations, the inferred strength of anisotropy in the IMIC (6–7%) implies almost perfect alignment of iron crystals, an intriguing, albeit unlikely situation, especially in the presence of heterogeneity, which calls for further studies.

Effective‐Medium Models of Inner‐Core Anisotropy

“…We have studied the seismic anisotropy in the outer 600 km of inner core by comparing seismological constraints on anisotropy with stochastic, monomineralic models in which the stiffness tensor C ( x ) is a random rotation of a single‐crystal hcp‐Fe tensor . Our results support the inferences of recent studies that compare seismological data with mineralogical models (Martorell et al, ; Romanowicz et al, ; Song & Jordan, ). In particular, ab initio calculations of that take into account premelting softening (Martorell et al, ) can account for the seismic data on inner‐core anisotropy if the hexagonal axes of tensors are moderately aligned with Earth's rotation axis.…”

“…An example is , obtained by Vočadlo et al () at midcore conditions (316 GPa, 5500 K) and used by Romanowicz et al () to model the PKP data. The local P wave anisotropy ratio is negative ( ), so the fast propagation direction is in the plane perpendicular to the c axis.…”

“…Stiffness tensors calculated ab initio for body‐centered cubic iron (bcc‐Fe) crystals have been used to fit body‐wave data for the IMIC in previous studies (Romanowicz et al, ; Wang et al, ). In Appendix B, we apply our effective‐media models to a hexagonal stiffness tensor derived by averaging a bcc‐Fe simulation at 6000 K and 360 GPa (Belonoshko et al, ) about its fast [111] crystallographic axis.…”

The influence of carbon on the seismic properties of solid iron

“…The seismic anisotropy is greatly enhanced, with the fastest axis of Vp propagation tilted away from the hexagonal axis of Fe. In particular, this aspect helps in relaxing the strong textural constraints put today on pure iron aggregates [ Romanowicz et al , ]. But the increase of anisotropy of Vs is opposite to the current observations.…”

Polymorphic Nature of Iron and Degree of Lattice Preferred Orientation Beneath the Earth's Inner Core Boundary

“…According to the cylindrically anisotropic model (Song, ; Stixrude & Cohen, ; Thomsen, ), the polar‐equatorial velocity difference for is 2.64%, while for the hexagonal‐close‐packed aggregates such variation amounts to 1.95% (Mattesini et al, ). Therefore, differences in seismic wave velocities along the polar and equatorial paths are significantly dissimilar for the two types of aggregates, unlike those presented in Figure of Romanowicz et al (). It is certainly worth remarking that the cylindrically averaged cubic iron is the mineral physics model with the largest anisotropy in the compressional wave velocity (V p ) (Mattesini et al, ).…”