Earth's Solid Inner Core

In the last few years, we have been investigating lateral and radial seismic heterogeneities in the Earth's solid inner core using seismic phases sensitive to this region. By illuminating these heterogeneities, our goal is to infer core dynamics that is playing a critical role in how the Earth is evolving over geologic time. In our work, we try to understand how the solid inner core is freezing from the liquid outer core, dynamic coupling between the solid inner core and core-mantle boundary, and the role of lateral temperature/impurity concentration variations. Off late, with the emergence of inner core melting and freezing hypotheses, our focus has shifted to detecting seismic signals that may result from melting and freezing. Whether the inner core is melting and/or freezing simultaneously is not known precisely. If it does, we might be able to tell the story of melting and freezing of Earth's solid inner core by mapping lateral heterogeneities and spatial correlation of seismic velocity and attenuation.

Fun Facts

Enigma surrounding Earth's core has inspired even a Hollywood creation! Remember the Sci-Fi movie The Core from 2003? While there are many factual inconsistencies, this movie serves as an intriguing preamble to both the scientifically inclined and the casual movie goer. If, however, you are more serious about learning the (scientific) basics about Earth's core, I recommend the TV series The Core (2011-2012) produced by BBC Two. Trust me, you will love it to the core!

Earth's Solid Inner Core

If you were able to drill roughly 5153 km into the Earth from the surface, your drill bit will suddenly start making a clunking noise -- you have reached the boundary separating the liquid outer core from the solid inner core. This, we call the Inner Core Boundary (ICB). Radius of the Earth's solid inner core (1217 km) is slightly smaller than our moon by about 520 km. Although we do not know exactly when it came to existence, recent hypotheses suggest that it may have formed roughly 3.5 to 1 billion years ago. Ever since, its progressive growth, about 1 mm/year, has been an important process, by which energy needed to maintain the geodynamo is provided. Geodynamo is simply the churning (technically convection) of liquid metal in the outer core that generates the geomagnetic field. The liquid in the outer core is kept in motion primarily by two mechanisms; (a) compositional convection (release of impurities/smaller elements to the outer core near the ICB as the inner core freezes). (b) thermal convection (heat given out at the time of freezing near the ICB). The freezing process together with internal dynamics (e.g. deformation, phase changes) determines the nature of the mechanical structure of the inner core, which can be studied using seismic waves.

Seismic Structure

Since its discovery in 1936 by Inge Lehman, the Earth’s solid inner core has posed a multitude of problems to be solved by the geophysical community, exhibiting heterogeneity and anisotropy at all length scales. An early model of global axisymmetric cylindrical anisotropy was later revised to include an isotropic uppermost layer of variable thickness, depth dependent anisotropy, and hemispherical differences in the degree of anisotropy. A more perplexing feature is the hemispherical heterogeneity observed in isotropic velocities and attenuation deduced from travel times and amplitudes of seismic waves that propagate along equatorial paths in the inner core. In this model, velocity and attenuation show a positive correlation with the Eastern Hemisphere being faster and more attenuating than the Western Hemisphere.

A New Structural Configuration of the Inner Core

We discovered a new structural configuration in the Earth's inner core, in which we determined the correlation between velocity and attenuation in the inner core by constraining regional models of velocity and attenuation simultaneously. These regional models are represented as 45˚ wide longitudinal bins. Our results show that the central Pacific region is anomalous in that a mantle-like relationship between velocity and attenuation is observed, whereas a positive correlation between the two seismic properties exist elsewhere as has been reported previously. We interpreted the central Pacific anomaly as arising due to higher than average homologous temperature in this region. Recent geodynamic models predict that, if the outer core convection is coupled to the thermal heterogeneity near the core-mantle boundary, such an increase in homologous temperature can be expected. On the other hand, alternative mechanisms such as grain size differences, variations in impurity concentrations along with lateral temperature variations are needed to explain seismic structure observed in the rest of the inner core. We cannot reconcile geodynamic models predicting convective translation in the inner core with our results.

To download velocity and attenuation models by bins of this study click here. Please site the following paper if you use these models. These models are also available as supplementary material of the same paper.

Attanayake, J., V. F. Cormier and S. De Silva, 2013. Uppermost inner core seismic structure - New insights from body waveform inversion, Earth Planet. Sci. Lett., 385, 49-58. doi:10.1016/j.epsl.2013.10.025.

We used two datasets of differential travel times to constrain inner core velocity to a depth of about 550 km from the inner core boundary. The first dataset was obtained from differencing travel times of PKIKP and PKiKP phases in the epicentral distance of 129^o- 141^o. This dataset is strongly sensitive to the uppermost 140 km of the inner core. The second dataset we used was assembled from differencing travel times of PKP_BC and PKP_Cdiff waves with respect to PKIKP waves in the 149^o-161^o distance range, and is strongly sampling the region 140 - 550 km below the inner core boundary and the bottommost 200 km of the outer core known as the F region. In this study, we showed that the seismic velocity in the inner core is laterally varying in the depth range considered and that a clear hemispheric signal is present to a depth of about 550 km from the inner core boundary although its intensity appears to diminish with depth. We measured roughly a 1% velocity contrast across hemispheres at the top of the inner core, whereas at a depth of 550 km from the inner core boundary, this contrast was about 0.4%. Our best-fitting velocity models indicated that the velocity profile in the F region is best described by that given in AK135-F reference model. We also made crude measurements of attenuation in the inner core using amplitude ratios of previously mentioned phases. Note that information contained in amplitude ratios is incomplete (compared to full waveforms) and these results should be interpreted with great care. The above figures show differential travel time residuals measured with respect to AK135-F model. Regions having blue columns are faster, whereas regions with red columns are slower relative to the reference model.

In the above work, we took great care to minimise the effect of mantle and source on waveforms by introducing a novel spectral deconvolution technique. In this technique, we inverted P waveforms in the 30^o- 90^o distance range for an Effective Source Time Function (ESTF) for each earthquake. To our knowledge, no study to-date has considered individual effects of sources and mantle attenuation as we have done here and hence, our attenuation measurements are arguably the most precise ones made hitherto.

The above figure shows the Effective Source Time Function (ESTF), focal mechanism, and the best fitting depth obtained by inverting 23 seismograms in the epicentral distance range 30^o and 90^o for the Earthquake that occurred on November 3rd, 1997 at latitude -30.74^o and longitude -71.22^o. We call the Source Time Function the Effective Source Time Function because it includes average mantle attenuation in addition to source effects.

Cormier, V. F., and J. Attanayake, 2013. Earth’s solid inner core: Seismic implications of freezing and melting, J. Earth. Sci., 24, 683-698. doi10.1007/s12583-013-0363-9

Uniformity of the F-Layer in the Outer Core

F region is approximately the bottommost 200 km of the liquid outer core characterised by a reduced P velocity gradient with respect to PREM. Recently Yu et.al. (2005) found hemispherical differences in the F region P velocity structure where the western hemisphere is AK135-like and the eastern hemisphere is PREM-like. It is difficult to explain this result considering known geodynamic constraints. We tested their hypothesis by checking the uniformity of this layer. Because the predicted velocity gradient in the F region of the eastern hemisphere differs from previous studies, we obtained the PKiKP-PKIKP and PKPbc-PKIKP differential travel times relevant for this particular region from their study and attempted to fit an AK135-like model to the F region by varying the structure in the inner core under reasonable assumptions. The results are shown in the above figure. E1 is the model predicted by Yu et.al. (2005) and JV2x is what is predicted by us for the inner core. We minimised an L₁-norm while fitting models to data and observed that JV2x explains data better than E1. JV2x has a small non unique velocity jump 140 km below the inner core boundary. We conclude that F region is a relatively uniform global layer that has a reduced velocity gradient comparable with AK135. The reduction of velocity may be due to iron enrichment from inner core solidification.

Cormier, V. F., J. Attanayake, and K. He, 2010. Inner core melting and freezing: Constraints from seismic body waves, Phys. Earth Planet. Int., 188, 3–4, 163-172. doi:10.1016/j.pepi.2011.07.007

Yu, W., L., Wen, and F., Niu, 2005. Seismic Velocity Structure in the Earth’s Outer Core, J. Geophys. Res., 110: B02302. doi:10.1029/2003JB002928

A Transition Layer in the Inner Core

Antipodal seismic waves interacting with the inner core boundary (ICB) sample almost the entire surface of the ICB. Hence, models inverted from antipodal observations of PKIKP, PKIIKP, and PKP_Cdiff waveforms can provide an estimate of the global averages of velocity and density discontinuities and their vertical gradients at the inner core boundary. We have provided evidence of the presence of a seismic transition layer between the outer core and the inner core, in which rigidity of the inner core gradually increases from zero to that of the inner core.

Attanayake, J., C. Thomas, V.F. Cormier, K. Koper, and M.S. Miller, 2018. Irregular Transition Layer Beneath The Earth’s Inner Core Boundary from Observations of Antipodal PKIKP and PKIIKP Waves, Geochemistry, Geophysics, Geosystems, 19(10), 3607-3622, https://doi.org/10.1029/2018GC007562

Attanayake, J., V.F. Cormier and K. He, 2008. Modeling the inner core boundary with Antipodal Seismic Waves, Eos Trans. AGU, 89(53), Fall Meet. Suppl., Abstract DI43A-1767