Periodic Reporting for period 4 - TRANSCALE (Reconciling Scales in Global Seimology)
Berichtszeitraum: 2021-10-01 bis 2023-03-31
For more than 30 years, seismologists have used seismic waves to produce 3D images of the structure of the Earth. Despite many successes, a number of key questions still remain, which are of the uttermost importance to understand plate tectonics. What is the nature of the Lithosphere-Asthenosphere Boundary? What is the structure and history of the continental lithosphere?
The problem is that different seismic observables sample the Earth at different scales; they have different sensitivity to structure, and are usually interpreted separately. Images obtained from short period converted and reflected body waves see sharp discontinuities, and are interpreted in terms of thermo-chemical stratification, whereas seismic models constructed from long period seismograms depict a smooth and anisotropic upper mantle, and are usually interpreted in terms of mantle flow. However, sharp discontinuities may also produce effective anisotropy at large scales, and only a combination of different observations, together with geodynamical models, can allow to fully understand the patterns of deformation in the mantle.
This project consists in developing and applying new approaches to geophysical data interpretation, where different type of constraints on different scales are combined to infer the structure of the crust and mantle. This project focuses on theoretical, algorithmic and computational advances needed for a new generation of tomographic models. The development of such techniques is by nature multidisciplinary, and the techniques themselves have many domains of application. While this project is motivated by specific disciplinary goals, the developed algorithms are applicable to other problems in geophysics and in other fields.
The problem is that different seismic observables sample the Earth at different scales; they have different sensitivity to structure, and are usually interpreted separately. Images obtained from short period converted and reflected body waves see sharp discontinuities, and are interpreted in terms of thermo-chemical stratification, whereas seismic models constructed from long period seismograms depict a smooth and anisotropic upper mantle, and are usually interpreted in terms of mantle flow. However, sharp discontinuities may also produce effective anisotropy at large scales, and only a combination of different observations, together with geodynamical models, can allow to fully understand the patterns of deformation in the mantle.
This project consists in developing and applying new approaches to geophysical data interpretation, where different type of constraints on different scales are combined to infer the structure of the crust and mantle. This project focuses on theoretical, algorithmic and computational advances needed for a new generation of tomographic models. The development of such techniques is by nature multidisciplinary, and the techniques themselves have many domains of application. While this project is motivated by specific disciplinary goals, the developed algorithms are applicable to other problems in geophysics and in other fields.
1- A new misfit function for inversion of scattered body waves.
Analysis of scattered body waves is widely used to make quantitative inferences about the structure below a seismic station. As these observables are mainly sensitive to travel-times of phases converted and reflected at interfaces, the solution non-unique, and there are strong trade-offs between the depth of discontinuities (small scales) and absolute velocities (large scales). To overcome this difficulty, we measure the misfit between the predicted and observed data with an optimal transport distance. A joint inversion of P-wave receiver functions (high frequency information) and surface wave dispersion curves (low frequency information) has been performed at the Hyderabad station in India, where a low velocity zone has been observed.
2- Estimating the effect of small scales in smooth tomographic models
It is well known that the small-scale heterogeneities that cannot be resolved by long-period seismic waves may be mapped in terms of effective anisotropy in tomographic models. In a theoretical study, we have investigated the relation between the amplitude of seismic heterogeneities and the level of induced S-wave radial anisotropy as seen by long-period seismic waves. We showed that a non-negligible part of the observed anisotropy in tomographic models may be the result of unmapped small-scale heterogeneities in the mantle, mainly in the form of fine layering, and that caution should be taken when interpreting observed anisotropy in terms of mantle deformation. This effect may be particularly strong in the lithosphere where chemical heterogeneities are assumed to be the strongest.
3- A new regional probabislitic 3D anisotropic model for the Alps
Our team is involved in the AlpArray project, which is a European initiative to deploy a large number of seismic stations across the Alpine region. In collaboration with colleagues from Grenoble (Anne Paul, Helle Pedersen, Laurent Stelhy), we have used ambient noise measurements to construct a 3D anisotropic model of the Alpine region where this trade-off between small-scale heterogeneities and anisotropy is accounted for. We observe a region of strong radial anisotropy at the base of the crust in the Italian peninsula. This strong anisotropic signal can be associated with extension in the Adriatic plate, which is disappearing between the oceanic subduction in the west, and the continental collision below the Balkanides-Dinarides.
4 – A new method to combine short and large scale information in tomographic models.
A powerful approach to image structure in the crust is full waveform inversion (FWI), where the full seismogram is inverted for. In order to account for the effect of small-scale heterogeneities mapping into effective anisotropy, we have proposed an approach where the seismic imaging problem is broken down in a two-stage multi-scale approach. In the first step, the observed waveforms are inverted for a smooth and fully anisotropic effective medium. In a second inversion, this resulting image is used as data, and the goal is to recover micro-scale parameters. We have illustrated the method with a synthetic cavity detection problem: we search for the position, size and shape of void inclusions in a homogeneous elastic media, where the size of cavities is smaller than the resolving length of the seismic data.
5-Geodynamical tomography.
Finally, we have developed an approach where constraints from geodynamics are used to infer anisotropic structure. A recurring issue in anisotropic imaging is that seismic anisotropy is fully described by the 21 parameters of the elastic tensor, which cannot be resolved independently at every location. In this work, instead of inverting for anisotropy, we parametrized our model in terms of a scalar temperature field. The forward problem consists of three steps: (1) calculation of mantle flow induced by temperature anomalies, (2) modelling of the strain-induced texture development of pure olivine, and (3) computation of azimuthally-varying surface wave dispersion curves. We show how a fully nonlinear Bayesian inversion of surface wave dispersion curves can retrieve the temperature field, without having to explicitly parameterize the elastic tensor.
6-Discovery of melt below oceanic tectonic plates.
We combined together information from two global tomographic models (shear wave velocity and attenuation models). We used results from mineral physics to jointly interpret these two tomographic models in terms of temperature and level of partial melt. In this way, the discovery of melt in the low velocity zone below oceanic plates has been obtained by combining constraints from two different disciplines: seismology and mineralogy.
Analysis of scattered body waves is widely used to make quantitative inferences about the structure below a seismic station. As these observables are mainly sensitive to travel-times of phases converted and reflected at interfaces, the solution non-unique, and there are strong trade-offs between the depth of discontinuities (small scales) and absolute velocities (large scales). To overcome this difficulty, we measure the misfit between the predicted and observed data with an optimal transport distance. A joint inversion of P-wave receiver functions (high frequency information) and surface wave dispersion curves (low frequency information) has been performed at the Hyderabad station in India, where a low velocity zone has been observed.
2- Estimating the effect of small scales in smooth tomographic models
It is well known that the small-scale heterogeneities that cannot be resolved by long-period seismic waves may be mapped in terms of effective anisotropy in tomographic models. In a theoretical study, we have investigated the relation between the amplitude of seismic heterogeneities and the level of induced S-wave radial anisotropy as seen by long-period seismic waves. We showed that a non-negligible part of the observed anisotropy in tomographic models may be the result of unmapped small-scale heterogeneities in the mantle, mainly in the form of fine layering, and that caution should be taken when interpreting observed anisotropy in terms of mantle deformation. This effect may be particularly strong in the lithosphere where chemical heterogeneities are assumed to be the strongest.
3- A new regional probabislitic 3D anisotropic model for the Alps
Our team is involved in the AlpArray project, which is a European initiative to deploy a large number of seismic stations across the Alpine region. In collaboration with colleagues from Grenoble (Anne Paul, Helle Pedersen, Laurent Stelhy), we have used ambient noise measurements to construct a 3D anisotropic model of the Alpine region where this trade-off between small-scale heterogeneities and anisotropy is accounted for. We observe a region of strong radial anisotropy at the base of the crust in the Italian peninsula. This strong anisotropic signal can be associated with extension in the Adriatic plate, which is disappearing between the oceanic subduction in the west, and the continental collision below the Balkanides-Dinarides.
4 – A new method to combine short and large scale information in tomographic models.
A powerful approach to image structure in the crust is full waveform inversion (FWI), where the full seismogram is inverted for. In order to account for the effect of small-scale heterogeneities mapping into effective anisotropy, we have proposed an approach where the seismic imaging problem is broken down in a two-stage multi-scale approach. In the first step, the observed waveforms are inverted for a smooth and fully anisotropic effective medium. In a second inversion, this resulting image is used as data, and the goal is to recover micro-scale parameters. We have illustrated the method with a synthetic cavity detection problem: we search for the position, size and shape of void inclusions in a homogeneous elastic media, where the size of cavities is smaller than the resolving length of the seismic data.
5-Geodynamical tomography.
Finally, we have developed an approach where constraints from geodynamics are used to infer anisotropic structure. A recurring issue in anisotropic imaging is that seismic anisotropy is fully described by the 21 parameters of the elastic tensor, which cannot be resolved independently at every location. In this work, instead of inverting for anisotropy, we parametrized our model in terms of a scalar temperature field. The forward problem consists of three steps: (1) calculation of mantle flow induced by temperature anomalies, (2) modelling of the strain-induced texture development of pure olivine, and (3) computation of azimuthally-varying surface wave dispersion curves. We show how a fully nonlinear Bayesian inversion of surface wave dispersion curves can retrieve the temperature field, without having to explicitly parameterize the elastic tensor.
6-Discovery of melt below oceanic tectonic plates.
We combined together information from two global tomographic models (shear wave velocity and attenuation models). We used results from mineral physics to jointly interpret these two tomographic models in terms of temperature and level of partial melt. In this way, the discovery of melt in the low velocity zone below oceanic plates has been obtained by combining constraints from two different disciplines: seismology and mineralogy.
As shown above, the outcomes of this project are both methodological and observational. We develop new methods that allow us to discover new structures, in order to better understand processes in the earth.
To summarize the obtained results:
New methods:
1) An optimal transport approach for inversion of scattered body waves.
2) A novel approach to invert for scales smaller than the minimum wavelength in Full Wave Inversion
3) A method to include geodynamic modelling in tomographic inversion
New results:
1) A zone of strong anisotropy at the base of the Adriatic plate
2) The discovery that long wavelength anisotropy might be due to small scale chemical heterogeneities in the mantle.
3) The discovery of partial melt below oceanic tectonic plates.
Other expected results by the end of the project:
1)A global Bayesian model.
2)Applications to some real earth problems of methods developed above.
To summarize the obtained results:
New methods:
1) An optimal transport approach for inversion of scattered body waves.
2) A novel approach to invert for scales smaller than the minimum wavelength in Full Wave Inversion
3) A method to include geodynamic modelling in tomographic inversion
New results:
1) A zone of strong anisotropy at the base of the Adriatic plate
2) The discovery that long wavelength anisotropy might be due to small scale chemical heterogeneities in the mantle.
3) The discovery of partial melt below oceanic tectonic plates.
Other expected results by the end of the project:
1)A global Bayesian model.
2)Applications to some real earth problems of methods developed above.