Nature: The formation of a deep mantle structure below Africa

In recent decades, a very important observation in seismology has been the discovery of two large low shear velocity provinces (LLSVPs) in the deep mantle below the African and Pacific plates, which stretch horizontally for thousands of kilometers and hundreds of kilometers high at the nuclear mantle boundary, with reduced velocity as the shear waves pass through (He and Wen, 2012; Ni et al., 2002; Wang and Wen, 2007), this phenomenon is caused by thermal anomalies in temperature and even differences in chemical composition in the region (Garnero et al., 2016).

For the dynamic evolution of these two LLSVPs, the existing research is divided into two categories of views:

(1) The locations of igneous provinces and kimberlites since the restoration of 320 Ma by paleomagnetic data have found that their positions are almost all located inside or at the edge of the two large shear wave low-velocity bodies observed today (Torsvik et al., 2010; Torsvik et al., 2008), and speculates that large shear wave low-velocity bodies since 320Ma may be relatively stationary. Analytical studies based on physical models have shown that centrifugal forces dating back to 2500 Ma may lock large shear wave low-velocity bodies at low latitudes near the equator (Dziewonski et al., 2010; Torsvik et al., 2014）。 Due to the axis symmetry of the geomagnetic field, paleomagnetic data can recover the latitude information of the paleomagnet, but the longitude information is almost uncertain. Therefore, if the large shear wave low-velocity bodies are fixed for a long time, then they can be used to establish a reference frame to restore the absolute position of the paleocontinental.

(2) There are also geodynamic models based on plate reconstruction constraints and self-consistency that find it difficult for large shear-wave low-velocity bodies to remain stationary until 320Ma (Langemeyer et al., 2020; Zhang et al., 2010; Zhong and Rudolph, 2015), and believe that large shear wave low-velocity bodies are associated with the formation of the Pangea supercontinent, further proposing a model that large shear-wave low-velocity bodies may be associated with supercontinent rotations (Li and Zhong, 2009). Although before 320Ma, there was controversy about whether large shear wave low-speed bodies were fixed and immobile, but some studies believed that large shear wave low-speed bodies since 250Ma were fixed (Conrad et al., 2013).

In view of the above problems, Dr. Flament and his collaborators from the University of Wollongong in Australia tested the deep mantle structure (high density geochemical primitive mantle material source region, forming a large shear wave low-velocity body for tomography observation) within the range of mantle physical parameters through a geodynamic model constrained by plate reconstruction results since 1Ga, and compared with the restored volcanic eruption location, it was found that there is a deep connection between the fast-moving deep mantle structure and the location of the volcanic eruption. Even where the volcano erupts, it can fall on the edge of the deep mantle structure. This suggests that by the location of the volcanic eruption, it is impossible to determine whether the large shear wave low-velocity body is stationary. This finding, which questions the fixed model, provides a new geodynamic mechanism for the geodynamic model to conform to paleomagnetic observations, and the results were recently published in Nature.

The study obtained a deep mantle structure consistent with the large shear wave low velocity body observed by today's seismological tomography, and the location of the volcanic eruption obtained by paleomagnetic reconstruction is located at the edge of the deep mantle structure predicted by the model and the large shear wave low velocity body predicted by the model. In order to reduce the uncertainty of the model, the study also designed 22 mantle convection models to test the effects of different plate reconstruction models and different mantle material properties. The model results show that the structure of the deep mantle below Africa is not a continuous structure before 160Ma, and a continuous structure is formed at 60Ma instead of the previously thought 200Ma (Figure 1e) (Young et al., 2019; Zhong and Rudolph, 2015）。 In this model, continental crustal material is mainly involved in the material cycle in the African mantle region, while continental crustal material is rarely found in the Pacific mantle region, which is consistent with geochemical observations (Doucet et al., 2020). The study further explains that the geochemically observed deep mantle structure below Africa contains continental material while the lower Pacific Ocean contains continental material, mainly because the subduction zone of the African continental hemisphere of the past 1Ga is closer to the continent, so the sinking subduction plate can bring continental material into the deep mantle structure below Africa (Figure 2).

Figure 1 Mantle structure and eruption location since 320Ma (Flament et al., 2022). Gray areas are low-speed areas. The first column is a tomography model, the second column is a mantle convection model; different colored dots represent volcanic eruption locations of different ages

Figure 2 Structural evolution of deep mantles below Africa and distribution of continental crust material in mantles (Flament et al., 2022)

Although the numerical model uses the latest and longest plate reconstruction results to constrain the upper surface boundary state of the model, and it can also be a good fit with some observations, there are some things that need to be further improved.

There are 1Ga plate reconstruction results constraints on the surface, but there are no constraints on the deep mantle structure at the beginning of the model (1Ga), so it is assumed that the 1Ga deep mantle structure is composed of a uniform layer of primitive mantle material. At that time, there were many uncertainties about whether there was a layer of primitive mantle material in the deep mantle at that time, and the shape of the primitive mantle material. If the 1Ga deep mantle structure is not a uniform layered structure, but has formed two large blocks, the density of the deep mantle structure required for the effect in the reference model is higher, and the increase in density will improve the stability of the deep mantle structure, thus affecting the aggregation time of the LLSVP below Africa predicted by the model.

In addition, the plate reconstruction results used in the model are based on the paleomagnetic reference system, not the mantle coordinate system. Studies have found that the position, size, and shape of the Pangea supercontinent under the mantle reference frame are better similar to those of the large shear wave low-velocity bodies below Africa (Mitchell et al., 2020). Thus reconstruction results of mantle reference frames using true polar shift constraints (Mitchell et al., 2012; The reconstruction results of Torsvik et al., 2014) instead of the paleomagnetic reference frame used in the study (Merdith et al., 2021) constrain the model surface may cause significant changes in the mantle structure in the model. It has been found that increasing the thermal viscosity difference (the ratio of surface viscosity to nuclear mantle boundary viscosity under the same pressure conditions) will increase the stability of the deep mantle structure, while the thermal viscosity difference used in this study is small, and if a larger thermal viscosity difference is used, the deep mantle structure below Africa may be more stable. It has been found that increasing the thermal viscosity difference (the ratio of surface viscosity to nuclear mantle boundary viscosity under the same pressure conditions) increases the stability of the deep mantle structure (Li et al., 2014), while the thermal viscosity difference used in this study is small, and if a larger thermal viscosity difference is used, the structure of the deep mantle below Africa may be more stable.

In any case, the study by Flament et al. allows us to continue to explore and think about the interconnections between the Earth's surface and deep structures and whether large shear wave low-velocity bodies are relatively stationary or rapidly moving with supercontinental rotations from a new perspective, which are fundamental questions related to the long-term evolution of the Earth.

Key References (Swipe up and down to view)

Conrad C P, Steinberger B, Torsvik T H. Stability of active mantle upwelling revealed by net characteristics of plate tectonics[J]. Nature, 2013, 498(7455): 479-482.

Doucet L S, Li Z X, Gamal El Dien H, et al. Distinct formation history for deep-mantle domains reflected in geochemical differences[J]. Nature Geoscience, 2020, 13(7): 511-515.

Dziewonski A M, Lekic V, Romanowicz B A. Mantle anchor structure: an argument for bottom up tectonics[J]. Earth and Planetary Science Letters, 2010, 299(1-2): 69-79.

Flament N, Bodur ？F, Williams S E, et al. Assembly of the basal mantle structure beneath Africa[J]. Nature, 2022, 603(7903): 846-851.

Garnero E J, McNamara A K, Shim S H. Continent-sized anomalous zones with low seismic velocity at the base of Earth's mantle[J]. Nature Geoscience, 2016, 9(7): 481-489.

He Y, Wen L. Geographic boundary of the “Pacific Anomaly” and its geometry and transitional structure in the north[J]. Journal of Geophysical Research: Solid Earth, 2012, 117(B9).

Langemeyer S M, Lowman J P, Tackley P J. The dynamics and impact of compositionally originating provinces in a mantle convection model featuring rheologically obtained plates[J]. Geophysical Journal International, 2020, 220(3): 1700-1716.

Li Y, Deschamps F, Tackley P J. The stability and structure of primordial reservoirs in the lower mantle: insights from models of thermochemical convection in three-dimensional spherical geometry[J]. Geophysical Journal International, 2014, 199(2): 914-930.

Li Z X, Zhong S. Supercontinent–superplume coupling, true polar wander and plume mobility: plate dominance in whole-mantle tectonics[J]. Physics of the Earth and Planetary Interiors, 2009, 176(3-4): 143-156.

Merdith A S, Williams S E, Collins A S, et al. Extending full-plate tectonic models into deep time: Linking the Neoproterozoic and the Phanerozoic[J]. Earth-Science Reviews, 2021, 214: 103477.

Mitchell R N, Kilian T M, Evans D A D. Supercontinent cycles and the calculation of absolute palaeolongitude in deep time[J]. Nature, 2012, 482(7384): 208-211.

Mitchell R N, Wu L, Murphy J B, et al. Trial by fire: Testing the paleolongitude of Pangea of competing reference frames with the African LLSVP[J]. Geoscience Frontiers, 2020, 11(4): 1253-1256.

Ni S, Tan E, Gurnis M, et al. Sharp sides to the African superplume[J]. Science, 2002, 296(5574): 1850-1852.

Torsvik T H, Burke K, Steinberger B, et al. Diamonds sampled by plumes from the core–mantle boundary[J]. Nature, 2010, 466(7304): 352-355.

Torsvik T H, Smethurst M A, Burke K, et al. Long term stability in deep mantle structure: Evidence from the~ 300 Ma Skagerrak-Centered Large Igneous Province (the SCLIP)[J]. Earth and Planetary Science Letters, 2008, 267(3-4): 444-452.

Torsvik T H, van der Voo R, Doubrovine P V, et al. Deep mantle structure as a reference frame for movements in and on the Earth[J]. Proceedings of the National Academy of Sciences, 2014, 111(24): 8735-8740.

Wang Y, Wen L. Geometry and P and S velocity structure of the “African Anomaly”[J]. Journal of Geophysical Research: Solid Earth, 2007, 112(B5).

Zhang N, Zhong S, Leng W, et al. A model for the evolution of the Earth's mantle structure since the Early Paleozoic[J]. Journal of Geophysical Research: Solid Earth, 2010, 115(B6).

Zhong S, Rudolph M L. On the temporal evolution of long‐wavelength mantle structure of the E arth since the early P aleozoic[J]. Geochemistry, Geophysics, Geosystems, 2015, 16(5): 1599-1615.

Written by: Shi Zhidong, Li Yang/Lithospheric Room

Editor: Chen Feifei

Proofreader: Zhang Tengfei Liu Qijun