The inner and outer cores are the most remote layers of Earth, yet have wide-ranging impacts on its evolution. Here, we detail current structural models of the core, deliberate dynamic interpretations, and highlight limitations and potential targets for future research to address.
At the centre of the Earth, and under extreme conditions, the core is physically inaccessible and difficult to observe remotely. Its structures, dynamics, and composition influence processes throughout our planet, including plate tectonics and the geomagnetic field. More fundamentally, improved comprehension of the core’s structural properties and their dynamical origins helps shed light on Earth’s formation and ongoing evolution.
The core is an iron-nickel alloy with still-debated quantities of lighter elements, constituting approximately 15% of the Earth’s volume, and separated from the bulk of the Earth by the core-mantle boundary (CMB)1,2,3,4,5. It is partitioned into two layers: the solid inner core surrounded by the fluid outer core, a convecting molten metal alloy extending to about half the Earth’s radius. The inner core grows slowly (~1 mm/yr) as material from the outer core freezes onto its surface at the inner core boundary (ICB); this mechanism helps drive convection in the outer core, and records a history of the core’s changing environments into the fabric of the inner core. Probing the inner core’s layers therefore provides insights into Earth’s past, whilst its current processes inform regarding its future.
Heterogeneities across multiple lengthscales are observed in the core’s seismic properties1,2,3 (Fig. 1), attributed to compositional or rheological variations5, and linked to geodynamical processes such as convection, uneven growth, or differential rotation1,2,3. Well-established structures of the inner core include compressional wave anisotropy—a directional dependence of velocity—aligned to Earth’s rotation axis, an east-west quasi-hemispherical asymmetry, and smaller-scale features such as a further internal layer termed the “innermost inner core” (IMIC). Structurally, anisotropy is linked to intrinsic mineral anisotropy and crystal alignment, thereby informing about constituent materials5 and solidification or texturing processes. The low-viscosity outer core displays fewer spatial variations, with stratification in the so-called E’ and F-layers at its top and base3.
Our knowledge of the core relies on a combination of remote sensing seismological techniques, experimental modelling, and computational simulations. New high-quality seismic measurements provide fundamental inputs for models of dynamics and composition, the plausibility of which must be evaluated within the context of our broader understanding of constraints on the state of the core and its known physical parameters.
Advancements in seismic techniques
Seismology’s limited observational resolution results from geographical constraints on the geometrical sampling of core-sensitive seismic waves and difficulties in isolating these small amplitude signals from noise2. New discoveries have employed both ever-increasing numbers of seismic stations and novel observational methods6, 7. These techniques deliver information beyond traditional seismic methodologies, which primarily employ the arrival times and amplitudes of direct seismic waves1,2,3 or information about whole-Earth vibrational frequencies4.
A significant recent development introduced the concept of the global coda-correlation wavefield6,8,9,10,11. Its realisation, a correlogram, is constructed by cross-correlating and stacking late hours of seismic records following large earthquakes. The long-lasting global reverberations dramatically improve spatial coverage, and the method boosts the amplitude of weak signals, manifesting as cross-correlation peaks in correlograms6. They have proved ideal for new observations of the core structure, including measurements of inner core shear properties6,8, anisotropy of the IMIC9,10, and stratification of the outer core11.
Another emerging approach uses transdimensional statistical tomography to produce 3D models of the core’s seismic properties, accompanied by uncertainty, simultaneously mapping quasi-hemispherical asymmetry, anisotropy, and regional-scale features7,12. Tomographic inversions for inner core structures are not new, but remain more challenging and less well-developed than mantle tomography. The most up-to-date inner core models contain large uncertainties, uneven spatial parameterisation, and theoretical errors from ambiguity in mantle structure13, so limitations on the resultant interpretations must be acknowledged12.
Multiscale seismic structures and interpretations
We highlight here seismic structures emerging into focus within contemporary observational models, which provoke ongoing dynamical and compositional interpretations. Notable topics currently attracting discourse are inner core shear properties3,6,8, the IMIC9,10,12,13, and boundary layers4,11,14,15,16,17, yet inconsistencies remain in their reported properties.
Inner core shear properties
Seismic observations confirmed the inner core as solid in the 1970s, some 30 years after mineral physics arguments proposed its solidity1,2. The elusiveness of inner core shear waves has restricted their ability to constrain its shear properties3. Recent observations of shear waves in the coda-correlation wavefield have provided improved estimates of inner core shear velocity, anisotropy, and attenuation6,8, proposing lower velocity and, perhaps, stronger attenuation than previous values. Updated constraints on shear velocity anisotropy near the core’s centre find the direction of the highest wavespeeds oblique to Earth’s rotation axis, in contrast to compressional velocity anisotropy8.
Observed shear properties of the inner core represent critical constraints for mineral physics experiments and calculations. Low shear velocity indicates softening of the inner core5, and the orientation of anisotropy helps ascertain candidate mineral phases of iron1,2,3. Improved geographical seismic coverage is needed for ongoing refinements to these measurements, including mapping radial and regional variations, fully calculating attenuation properties, and establishing the link to compressional velocity properties6,8.
Innermost inner core (IMIC)
Until recently, the IMIC was considered a possible artefact1, but current work is iterating towards a consensus for its properties. A few hundred kilometres in radius, the IMIC is characterised by different anisotropy from the bulk of the inner core, with a gradual transition between the two regions. Specifically, the IMIC has stronger velocity anisotropy, and a different direction of slowest wavespeed9,10,13. Although it has been proposed that the IMIC displays hemispherical asymmetry1, recent work argues that it may instead be shifted off-centre12,13. However, these two analyses define the IMIC with different criteria, using the polar-equatorial difference in wavespeed13 versus the direction of slowest wavespeed12, and subsequently reach opposing conclusions for its position.
Dynamically, the IMIC may mark a gradual transition in the growth processes of the inner core, observed as a change in the crystalline structure of iron3,10; the relic of two-stage formation, also detected in paleomagnetic data18; or the onset or cessation of external mechanisms creating texturing13. We anticipate ongoing improvements to the current IMIC models will provide the underpinning measurements to orient the next generation of core evolution models.
Inner core boundary (ICB)
We have considerably more data from the uppermost inner core than for deeper layers due to greater volumetric coverage and comparatively straightforward techniques to isolate the relevant signals1,2. The upper inner core demonstrates a distinct east-west dichotomy in its seismic velocity, the quasi-western hemisphere being slower for compressional waves by about 1%. Anisotropy is negligible in the outermost layers1,2,3, but increases considerably ~50 km beneath the ICB. It is thought to be confined to the western hemisphere12, although the current spatial coverage is far from complete. The ICB itself sustains localised topography of a few kilometres2,14.
As the locus of inner core growth, the ICB region provides insight into current dynamical processes of the inner core and its interaction with the outer core1,2,3,4. Its features are ascribed to various dynamical mechanisms recorded in the inner core structure as it freezes3. For example, asymmetric heat flow could generate hemispherical differences in the inner core, with localised enhanced flux creating dynamical topography on the ICB and compositional heterogeneity in the F-layer1,2,3,4,14. Temporal variation of ICB topography and the uppermost IC may also arise from the differential rotation of the inner core with respect to the mantle19,20. This is difficult to detect seismically, hence there is currently no consensus on rotation and/or oscillation rate1,2,3,14,19,20.
Outer core layering
Layering may exist at both boundaries of the outer core. The ~100 km thick F-layer at its base exhibits a decreasing seismic velocity gradient4,11, and lateral variability15. In the uppermost outer core, the E’-layer displays a lower velocity anomaly relative to the well-mixed bulk of the outer core3,4,11, although studies disagree on its thickness16,17. Fluctuations in geomagnetic field strength and length-of-day measurements support the presence of stratification in the outermost outer core21. These data further indicate additional small-amplitude lateral variations in temperature and composition, but which are seismically undetectable3. Linking seismic models of the E’-layer to specific compositions is also not straightforward5.
The existence of the F-layer is attributed to inner core growth mechanisms causing chemical stratification; for example, a partially-frozen slush snowing onto the ICB3. Localised melting or freezing of the ICB could create regional enrichment of iron or light elements. There are multiple options for generating a compositionally stratified E’-layer: remnants of core formation, interaction at the CMB, or light elements released upon inner core growth3. At the CMB, heat is extracted unevenly by the silicate mantle; seismic investigations of this dynamically important interface are an active research area22.
As we face an unprecedented expansion of data sensitive to the deep Earth, seismic studies of the core are at a pivotal stage. We anticipate that the recent discoveries discussed here will precipitate further advancements in knowledge of core structures and dynamics. Potential near-future targets for core seismology include inner core shear wave attenuation, anisotropy, and tomography, high-resolution observations of boundary and transition layers, and improved tracking of temporal changes. Performing many of these analyses will be challenging, with several likely beyond the scope of present seismic detection techniques. Continual improvements in observational infrastructure and methodologies will therefore remain integral to the development of our understanding of the core. Coda-correlation6 and machine learning algorithms23 are emerging as valuable tools in studies of Earth’s inner layers, for seismic measurements, and their integration with outputs from other disciplines24. We anticipate that such data driven approaches will become prevalent for investigations into the core in coming years. Overall, interdisciplinary and multidisciplinary collaborations will be essential as we move forward in our attempts to understand Earth’s enigmatic core.
Deuss, A. Heterogeneity and anisotropy of Earth’s inner core. Annu. Rev. Earth. Planet. Sci. 42, 103–126 (2014).
Tkalčić, H. The Earth’s inner core: revealed by observational seismology. (Cambridge University Press, 2017).
Cormier, V. F., Bergman, M. I. & Olson, P. Earth’s core: geophysics of a planet’s deepest interior. (Elsevier, 2021).
Irving, J. C. E., Cottaar, S. & Lekić, V. Seismically determined elastic parameters for Earth’s outer core. Sci. Adv. 4, eaar2538 (2018).
Hirose, K., Wood, B. & Vočadlo, L. Light elements in the Earth’s core. Nat. Rev. Earth. Environ. 2, 645–658 (2021).
Tkalčić, H., Phạm, T.-S. & Wang, S. The Earth’s coda correlation wavefield: rise of the new paradigm and recent advances. Earth-Sci. Rev. 208, 103285 (2020).
Burdick, S., Waszek, L. & Lekić, V. Seismic tomography of the uppermost inner core. Earth. Planet. Sci. Lett. 528, 115789 (2019).
Tkalčić, H., Wang, S. & Phạm, T. S. Shear properties of Earth’s inner core. Annu. Rev. Earth. Planet. Sci. 50, 153–181 (2022).
Costa de Lima, T., Waszek, L. & Tkalčić, H. A new probe into the innermost inner core anisotropy via the global coda‐correlation wavefield. J. Geophys. Res. Solid Earth. 127, e2021JB023540 (2022).
Phạm, T.-S. & Tkalčić., H. Up-to-fivefold reverberating waves through the Earth’s center and distinctly anisotropic innermost inner core. Nat. Commun. 14, 754 (2023).
Ma, X. & Tkalčić, H. CCMOC: A new view of the Earth’s outer core through the global coda correlation wavefield. Phys. Earth. Planet. Inter. 334, 106957 (2023).
Brett, H., Hawkins, R., Waszek, L., Lythgoe, K. & Deuss, A. 3D transdimensional seismic tomography of the inner core. Earth. Planet. Sci. Lett. 593, 117688 (2022).
Frost, D., Lasbleis, M., Chandler, B. & Romanowicz, B. Dynamic history of the inner core constrained by seismic anisotropy. Nat. Geosci. 14, 531–535 (2021).
Yao, J., Tian, D., Sun, L. & Wen, L. Temporal change of seismic Earth’s inner core phases: inner core differential rotation or temporal change of inner core surface? J. Geophys. Res. Solid Earth. https://doi.org/10.1029/2019JB017532 (2019).
Ohtaki, T., Kanehsima, S., Ichikawa, H. & Tsuchiya, T. Seismological Evidence for Laterally Heterogeneous Lowermost Outer Core of the Earth. J. Geophys. Res. Solid Earth. 123, 10903–10917 (2018).
Wu, W. & Irving, J. C. E. Array-based iterative measurements of SmKS times and their constraints on outermost core structure. J. Geophys. Res. Solid Earth. 125, e2019JB018162 (2020).
van Tent, R., Deuss, A., Kaneshima, S. & Thomas, C. The signal of outermost-core stratification in body-wave and normal-mode data. Geophys. J. Int. 223, 1338–1354 (2020).
Zhou, T. et al. “Early Cambrian renewal of the geodynamo and the origin of inner core structure. Nat. Commun. 13, 4161 (2022).
Wang, W. & Vidale, J. Seismological observation of Earth’s oscillating inner core. Sci. Adv. 8, eabm9916 (2022).
Yang, Y. & Song, X. Multidecadal variation of the Earth’s inner-core rotation. Nat. Geosci. 16, 182–187 (2023).
Buffet, B., Knezek, N. & Holme, R. Evidence for MAC waves at the top of Earth’s core and implications for variations in length of day. Geophys. J. Int. 204, 1789–1800 (2016).
Russell, S., Irving, J. C. E., Myhill, R. & Cottaar, S. A more complex core-mantle boundary? Nat. Comm, this collection (2023).
Garcia, J., Waszek, L., Tauzin, B. & Schmerr, N. Automated identification of seismic wave arrivals using Convolutional Neural Networks. Geophys. Res. Lett., 48, https://doi.org/10.1029/2020GL091658 (2021).
Rijal, A., Cobden, L., Trampert, J., Marquardt, H. & Jackson, J. M. Shear properties of MgO inferred using neural networks. Eur. J. Mineral. 35, 45–58 (2023).
The authors declare no competing interests.
Peer review information
Nature Communications thanks the anonymous reviewer(s) for their contribution to the peer review of this work.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Waszek, L., Irving, J., Phạm, TS. et al. Seismic insights into Earth’s core. Nat Commun 14, 6029 (2023). https://doi.org/10.1038/s41467-023-41725-5