Eric Debayle

High seismic velocity keels extending to depths greater than 200 km underlie the oldest parts of continents, the Precambrian cratons. These keels have probably been formed early in Earth’s history, and the preservation of these deep, cold, and highly viscous roots in a convective mantle remains enigmatic. A classical view is that the excess density due to colder temperatures is compensated for by a light composition. Here, we map the magnesium number (Mg#, a proxi for mantle depletion) within cratonic keels, based on the thermochemical interpretation of a global shear velocity model. Our interpretation suggests that depletion is strong above 150 km (Mg#>92), and decreases with depth down to the lithosphere-asthenosphere boundary (LAB). However, changing only the major element composition and the Mg# does not fully meet the requirements implied by seismology or gravity observations. We explore various hypotheses, varying the temperature of the lithosphere, considering the presence of eclogite, accounting for chromium or for a minute quantity of diamonds. We also discuss the uncertainties in seismological observations of velocity and radial anisotropy as well as in mineralogical models. Changes in temperature, the presence of eclogite, of Cr incorporated in garnet, or radial anisotropy are inconsistent with shear velocity and gravity constraints when considered individually. Below the graphite/diamond transition, the combination of depletion with a low volume fraction of diamond (< 1% at 150 km) appears to be the simplest way to explain the very high shear velocities, while maintaining the cratonic lithosphere close to neutral buoyancy. This small amount of diamond accounts for only 0.5% of the mantle’s carbon and can be formed within a reasonable geological timescale (> 180 Ma) through the subduction of organic carbon.

High seismic velocity keels extending to depths greater than 200 km underlie the oldest parts of continents, the Precambrian cratons. These keels have probably been formed early in Earth’s history, and the preservation of these deep, cold, and highly viscous roots in a convective mantle remains enigmatic. A classical view is that the excess density due to colder temperatures is compensated for by a light composition. Here, we map the magnesium number (Mg#, a proxi for mantle depletion) within cratonic keels, based on the thermochemical interpretation of a global shear velocity model. Our interpretation suggests that depletion is strong above 150 km (Mg#>92), and decreases with depth down to the lithosphere-asthenosphere boundary (LAB). Below the graphite/diamond transition, the combination of depletion with a low volume fraction of diamond (

In this presentation, I will review some of our recent global tomography results, that provide constraints on the Earth mantle structure and mantle convection.In the upper mantle, we have recently constructed global tomographic models of SV wave velocity, 𝑉𝑠𝑣, and radial anisotropy, 𝜉, using the same tomographic approach, with similar regularization and smoothing for the Rayleigh and Love wave data. We also use Rayleigh waves to constrain the azimuthal anisotropy, the quality factor 𝑄 and the melt content. We find that a 1-D model of radial anisotropy, close to PREM, but including a 3D crustal structure, explains the Love/Rayleigh differences almost everywhere, except in oldest parts of the continents and youngest parts of the Pacific ridge. No age dependence of the radial anisotropy 𝜉 in the oceanic upper mantle is required, while age is the main parameter controlling 𝑉𝑠𝑣, melt content and azimuthal anisotropy. In the asthenosphere, azimuthal anisotropy aligns on a large scale with present plate motion only for fast plates (> ∼4 cmyr−1), suggesting that only fast-moving plates produce sufficient shearing at their base, to organize the flow on the scale of the entire tectonic plates. Part of the azimuthal anisotropy is also frozen in the shallow oceanic lithosphere. The presence of a small amount of partial melt, by reducing mantle viscosity, facilitates plate motion and large-scale crystal alignment in the asthenosphere.We have also built global shear tomographic models of the whole mantle for the shear velocity (SEISGLOB2) and attenuation (QL3D). In the lower mantle, SEISGLOB2 has revealed a change in the shear velocity spectrum at around 1000 km depth. The spectrum is the flattest (i.e. richest in "short" wavelengths corresponding to spherical harmonic degrees greater than 10) around 1000 km depth and this flattening occurs between 670 and 1500 km depth. QL3D combines various S-phase measurements, including surface waves, direct (S, SS, SSS, SSSS), core-reflected (ScS, ScSScS, ScSScSScS), diffracted (S𝑑𝑖𝑓𝑓) and their depth phases (e.g., sS, sScS, sS𝑑𝑖𝑓𝑓), providing extensive depth and spatial coverage. A high attenuation zone highlights the peculiar nature of the mantle around 1000 km depth. This may indicate the presence of a global low-viscosity layer, in a region that roughly corresponds to the upper boundary of the Large Low Shear Velocity Provinces (LLSVPs), and where various changes in the continuity of slabs and mantle plumes have been observed. Our 3D shear quality factor model also confirms that the LLSVPs are attenuating, at least for body waves with periods near 35 s. The correlation between strong attenuation and low shear velocities within these regions suggests that the shear quality factor mostly captures the thermal signature of the LLSVP.

We present a summary of all our seismological models based on surface wave observations. First, wen have updated our global SV model that is based on the waveform modelling of more than 3 millions Rayleigh waves recorded since 1976. The tomographic model is built using the same automated scheme as was presented in Debayle et al., GRL 2016, while the number of data has increased by a factor larger than 2. Second, we present new SV wave velocity and radial anisotropy, ξ, global model obtained from the joint inversion of multi-mode Rayleigh and Love wave data. The Rayleigh dataset has been also used in previous studies to constrain the azimuthal anisotropy, the quality factor Q and the melt content. All these models are based on self-consistent datasets and have been obtained with similar regularization and smoothing. All together they confirm that (1) there is no age dependence of the radial anisotropy ξ in the oceanic lithosphere while age is the main parameter controlling Vsv, (2) Positive (ξ > 1) radial anisotropy extends deeper (down to 300 km depth) beneath oceans than beneath continents (3) melt content and azimuthal anisotropy have a significant dependence with age under oceans, particularly under fast plates, (4) cratons appear fast, melt-free and anisotropic, down to 250-300 km depth for the azimuthal anisotropy and down to 150-200 km for the radial anisotropy, a depth coinciding with the detection of a positive velocity gradient, (5) ξ ≤ 1 anomalies are present in the deeper part of the cratonic roots (200-250 km) that could reflect past delamination or compression of the deep lithospheric roots. Out of these observations, we propose a conceptual model that enables reconciling a number of seismic observations. Finally, will present our new online tool (https://fascil.univ-lyon1.fr/) available to explore these tomographic models and to compare with existing ones.

The details of heterogeneities in the lower mantle have increased considerably during the last decades thanks to seismic imaging revealing ULVZs, D” layer, PERM anomaly and now mega-ULVZs. However, the origin of these anomalies is actively debated, as seismic velocities alone cannot disentangle between thermal or compositional origins. Seismic attenuation can provide an additional perspective to seismic velocity for constraining physical properties of the lower mantle heterogeneities. In this study, we aim to develop the first 3D global model of body wave attenuation (Q) in the lower mantle using various S-phase measurements. To maximize the depth and spatial coverage, we incorporate multiple phases (S, SS, SSS, SSSS), core phases (ScS, ScSScS, ScSScSScS), Sdiff and their depth phases (e.g., sS, sScS, sSdiff). We process > 80, 000 seismic data recorded on more than 2000 global seismic stations from earthquakes occurring during 2009-2023. We measure differential anelastic delay times between the observed S phases and the same phases on 3D synthetics using the instantaneous frequency matching method in period range of 100 ~ 10 seconds. These synthetic seismograms are computed in 3D mantle model S40RTS and crust model CRUST1.0 using SPECFEM3D-globe, which can fully account for the effect of 3D heterogeneities, allowing for reliable attenuation measurements. The differential anelastic time delays exhibit abnormally large variations for all S phases, reflecting the complexity of the data potentially brought by the elastic effects. Despite this, the average differential anelastic time delays for all S phases remain consistently negative across all epicentral distances and generally decrease with increasing epicentral distances, suggesting that the Earth is, on average, less attenuating than the PREM model. We further find that the scattering of differential anelastic time delays can be significantly reduced, and abnormal measurements effectively excluded, when the waveform similarity of observed and synthetic phases is high. This is likely because, in such cases, uncertainties arising from factors like source mechanisms and heterogeneity are substantially minimized. We perform a 1D tomographic inversion using high-similarity data. The preliminary 1D attenuation model we obtain is similar to PREM model, with lower attenuation in the lower mantle and the highest attenuation in the upper mantle, roughly corresponding to the depth range of the low velocity zone. However, the Q values in our model are approximately 1.5 times larger than the Q values in PREM model. Next, we will perform 3D tomographic inversion and subsequently integrate this 3D Q model with global 3D shear wave models to jointly invert for the thermo-chemical state of the lower mantle.

High seismic velocity keels extending to depths greater than 200 km underlie the oldest parts of continents, the Precambrian cratons. Although mantle xenoliths suggests that these keels formed early in Earth’s history, the preservation of deep, cold, and highly viscous roots for hundreds of millions of years in a convective mantle remains enigmatic. A classical view is that the excess density due to colder temperatures is compensated for by a light composition that stabilizes high viscosity roots. Here, we map the magnesium number (Mg#, a proxi for mantle depletion) and the Modal content of olivine within cratonic keels, based on the thermochemical interpretation of a global shear velocity model. This thermochemical interpretation requires mineralogical depletion and a decrease of compositional density beneath Precambrian cratons. We observe that depletion is strong above 150 km (Mg#>92), and decreases with depth down to the lithosphere-asthenosphere boundary (LAB). These conditions ensure cratons’s preservation for hundred to billions of years in a convective mantle, in agreement with mantle xenoliths suggesting that high viscous keels formed early in the history of cratons

Craton formation and evolution remains enigmatic because observations from long and short period seismic waves and geochemical data are inconsistent. For example, both internal layering and radial anisotropy are poorly constrained. By inverting cratonic Rayleigh and Love surface wave dispersion curves for shear-wave velocity and radial anisotropy using a flexible Bayesian scheme, we show that these inconsistencies can be reconciled. Our methodology does not require any vertical smoothing and only includes anisotropic layers where necessary to fit the data. Results show all cratons possess a positively radially anisotropic upper lithospheric layer that is best explained by Archean underplating. An isotropic layer lies beneath, likely indicative of two-stage craton formation. We find a variable amplitude low velocity zone (LVZ) may exist within the upper anisotropic layer of up to 9 of 12 cratons studied. This LVZ is well correlated to observed Mid-Lithospheric Discontinuities (MLDs). Our results suggest the MLD is best explained by post formation modification within cratons.

We present the latest update of the global SV model developed by our team, in Lyon. It is based on the waveform modeling of more than 3 millions Rayleigh waves recorded since 1976. The tomographic model is built using the same automated scheme as was presented in Debayle et al., GRL 2016, while the number of data has increased by a factor larger than 2. For each seismogram, we obtain a path average shear velocity and quality factor model, and a set of fundamental and higher-mode dispersion and attenuation curves from 40s to 250s. We incorporate the resulting set of path average shear velocity models into a tomographic inversion. Due to the drastic increase of data, this second step of inversion became too computationally costly. We rewrote it so that the largest matrix to invert has now a size (number of geographical point)**2 instead of (number of data)**2. Thanks to these improvements we reduced the correlation lengths from 4 deg down to 1deg. We will focus in several geographical areas and geological objects, to emphasize the improvement in precision of this new model. We will also present our new online tool (https://fascil.univ-lyon1.fr/) available to explore this tomographic model and to compare with existing ones.