skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


Title: Shear properties of MgO inferred using neural networks
Abstract. Shear properties of mantle minerals are vital for interpreting seismic shear wave speeds and therefore inferring the composition and dynamics of a planetary interior. Shear wave speed and elastic tensor components, from which the shear modulus can be computed, are usually measured in the laboratory mimicking the Earth's (or a planet's) internal pressure and temperature conditions. A functional form that relates the shear modulus to pressure (and temperature) is fitted to the measurements and used to interpolate within and extrapolate beyond the range covered by the data. Assuming a functional form provides prior information, and the constraints on the predicted shear modulus and its uncertainties might depend largely on the assumed prior rather than the data. In the present study, we propose a data-driven approach in which we train a neural network to learn the relationship between the pressure, temperature and shear modulus from the experimental data without prescribing a functional form a priori. We present an application to MgO, but the same approach works for any other mineral if there are sufficient data to train a neural network. At low pressures, the shear modulus of MgO is well-constrained by the data. However, our results show that different experimental results are inconsistent even at room temperature, seen as multiple peaks and diverging trends in probability density functions predicted by the network. Furthermore, although an explicit finite-strain equation mostly agrees with the likelihood predicted by the neural network, there are regions where it diverges from the range given by the networks. In those regions, it is the prior assumption of the form of the equation that provides constraints on the shear modulus regardless of how the Earth behaves (or data behave). In situations where realistic uncertainties are not reported, one can become overconfident when interpreting seismic models based on those defined equations of state. In contrast, the trained neural network provides a reasonable approximation to experimental data and quantifies the uncertainty from experimental errors, interpolation uncertainty, data sparsity and inconsistencies from different experiments.  more » « less
Award ID(s):
2009935
PAR ID:
10438238
Author(s) / Creator(s):
; ; ; ;
Date Published:
Journal Name:
European Journal of Mineralogy
Volume:
35
Issue:
1
ISSN:
1617-4011
Page Range / eLocation ID:
45 to 58
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; (, American Mineralogist)
    null (Ed.)
    Abstract We investigated the structure, equation of state, thermodynamics, and elastic properties of tremolite amphibole [Ca2Mg5Si8O22(OH)2] up to 10 GPa and 2000 K, using first principles simulations based on density functional perturbation theory. We found that at 300 K, the pressure-volume results can be adequately described by a third-order Birch-Murnaghan equation of state with bulk moduli K0 of 78.5 and 66.3 GPa based on local density approximation (LDA) and generalized gradient approximation (GGA), respectively. We also derived its coefficients of the elastic tensor based on LDA and GGA and found that the LDA result is in good agreement with the experimental results. At 300 K, the shear modulus G0 is 58.0 GPa based on LDA. The pressure derivative of the bulk modulus K′ is 5.9, while that of the shear modulus G′ is 1.3. The second Grüneisen parameter, or δT = [–1/(αKT)](∂KT/∂T)P, is 3.3 based on LDA. We found that at ambient conditions, tremolite is elastically anisotropic with the compressional wave velocity anisotropy AVP being 34.6% and the shear wave velocity anisotropy AVS being 27.5%. At higher pressure corresponding to the thermodynamic stability of tremolite, i.e., ~3 GPa, the AVP reduces to 29.5%, whereas AVS increases to 30.8%. To evaluate whether the presence of hydrous phases such as amphibole and phlogopite could account for the observed shear wave velocity (VS) anomaly at the mid-lithospheric discontinuity (MLD), we used the thermoelasticities of tremolite (as a proxy for other amphiboles), phlogopite, and major mantle minerals to construct synthetic velocity profiles. We noted that at depths corresponding to the mid-lithosphere, the presence of 25 vol% amphibole and 1 vol% phlogopite could account for a VS reduction of 2.3%. Thus based on our thermoelasticity results on tremolite amphibole, it seems that mantle metasomatism could partly explain the MLD. 
    more » « less
  2. ; (, The Journal of the Acoustical Society of America)
    Guided ultrasonic wave localization systems use spatially distributed sensor arrays and wave propagation models to detect and locate damage across a structure. Environmental and operational conditions, such as temperature or stress variations, introduce uncertainty into guided wave data and reduce the effectiveness of these localization systems. These uncertainties cause the models used by each localization algorithm to fail to match with reality. This paper addresses this challenge with an ensemble deep neural network that is trained solely with simulated data. Relative to delay-and-sum and matched field processing strategies, this approach is demonstrated to be more robust to temperature variations in experimental data. As a result, this approach demonstrates superior accuracy with small numbers of sensors and greater resilience to spatially nonhomogeneous temperature variations over time. 
    more » « less
  3. ; ; (, Geochemistry, Geophysics, Geosystems)
    Abstract To quantitatively convert upper mantle seismic wave speeds measured into temperature, density, composition, and corresponding and uncertainty, we introduce theWhole‐rockInterpretativeSeismicToolboxForUltramaficLithologies (WISTFUL). WISTFUL is underpinned by a database of 4,485 ultramafic whole‐rock compositions, their calculated mineral modes, elastic moduli, and seismic wave speeds over a range of pressure (P) and temperature (T) (P = 0.5–6 GPa,T = 200–1,600°C) using the Gibbs free energy minimization routine Perple_X. These data are interpreted with a toolbox of MATLAB® functions, scripts, and three general user interfaces:WISTFUL_relations, which plots relationships between calculated parameters and/or composition;WISTFUL_geotherms, which calculates seismic wave speeds along geotherms; andWISTFUL_inversion, which inverts seismic wave speeds for best‐fit temperature, composition, and density. To evaluate our methodology and quantify the forward calculation error, we estimate two dominant sources of uncertainty: (a) the predicted mineral modes and compositions, and (b) the elastic properties and mixing equations. To constrain the first source of uncertainty, we compiled 122 well‐studied ultramafic xenoliths with known whole‐rock compositions, mineral modes, and estimatedP‐Tconditions. We compared the observed mineral modes with modes predicted using five different thermodynamic solid solution models. The Holland et al. (2018,https://doi.org/10.1093/petrology/egy048) solution models best reproduce phase assemblages (∼12 vol. % phase root‐mean‐square error [RMSE]) and estimated wave speeds. To assess the second source of uncertainty, we compared wave speed measurements of 40 ultramafic rocks with calculated wave speeds, finding excellent agreement (VpRMSE = 0.11 km/s). WISTFUL easily analyzes seismic datasets, integrates into modeling, and acts as an educational tool. 
    more » « less
  4. ; ; ; (, American Mineralogist)
    Abstract Thermoelastic properties of mantle candidate minerals are essential to our understanding of geophysical phenomena, geochemistry, and geodynamic evolutions of the silicate Earth. However, the lower-mantle mineralogy remains much debated due to the lack of single-crystal elastic moduli (Cij) and aggregate sound velocities of (Al,Fe)-bearing bridgmanite, the most abundant mineral of the planet, at the lower mantle pressure-temperature (P-T) conditions. Here we report single-crystal Cij of (Al,Fe)-bearing bridgmanite, Mg0.88Fe0.1Al0.14Si0.90O3 (Fe10-Al14-Bgm) with Fe3+/ΣFe = ~0.65, up to ~82 GPa using X-ray diffraction (XRD), Brillouin light scattering (BLS), and impulsive stimulated light scattering (ISLS) measurements in diamond-anvil cells (DACs). Two crystal platelets with orientations of (–0.50, 0.05, –0.86) and (0.65, –0.59, 0.48), that are sensitive to deriving all nine Cij, are used for compressional and shear wave velocity (νP and νS) measurements as a function of azimuthal angles over 200° at each experimental pressure. Our results show that all Cij of singe-crystal Fe10-Al14-Bgm increase monotonically with pressure with small uncertainties of 1–2% (±1σ), except C55 and C23, which have uncertainties of 3–4%. Using the third-order Eulerian finite-strain equations to model the elasticity data yields the aggregate adiabatic bulk and shear moduli and respective pressure derivatives at the reference pressure of 25 GPa: KS = 326 ± 4 GPa, µ = 211 ± 2 GPa, KS′ = 3.32 ± 0.04, and µ′ = 1.66 ± 0.02 GPa. The high-pressure aggregate νS and νP of Fe10-Al14-Bgm are 2.6–3.5% and 3.1–4.7% lower than those of MgSiO3 bridgmanite end-member, respectively. These data are used with literature reports on bridgmanite with different Fe and Al contents to quantitatively evaluate pressure and compositional effects on their elastic properties. Comparing with one-dimensional seismic profiles, our modeled velocity profiles of major lower-mantle mineral assemblages at relevant P-T suggest that the lower mantle could likely consist of about 89 vol% (Al,Fe)-bearing bridgmanite. After considering uncertainties, our best-fit model is still indistinguishable from pyrolitic or chondritic models. 
    more » « less
  5. ; ; ; ; (, American Mineralogist)
    Abstract The influence of Al substitution on the elastic properties of stishovite and its transition to post-stishovite is of great importance for interpreting the seismic wave velocities of subducted mid-ocean ridge basalt (MORB) within the mantle transition zone and the lower mantle. However, atomistic mechanisms of Al substitution effects on the transition and its associated elasticity remain debated. Here synchrotron single-crystal X-ray diffraction measurements have been performed at room temperature on Al1.3-SiO2 (1.3 mol% Al in the chemical formula of Si0.965(3)Al0.041(1)O2H0.017(4)) and Al2.1-SiO2 (2.1 mol% Al in Si0.948(2)Al0.064(1)O2H 0.018(3)) crystals in diamond anvil cells with Boehler-Almax designed anvils up to 38.0 GPa and 28.5 GPa, respectively. Refinements of the diffraction patterns show that a transformation from stishovite (space group P42/mnm; No. 136) to CaCl2-typed post-stishovite (space group Pnnm; No. 58) is accompanied by splitting of O coordinates. The Al substitution in stishovite results in a faster decrease in the O coordinate, softer apical (Si,Al)-O bonds, and a softer and less distorted (Si,Al)O6 octahedron under compression. This leads to reduced adiabatic bulk modulus (KS), shear modulus (G), shear wave velocity (VS), and compressional wave velocity (VP) in the stishovite phase, explaining seismic wave perturbations in the mantle transition zone. Together with Raman data, Landau theory modeling shows that Al substitution increases the order parameter and excess free energy, stabilizing the post-stishovite phase at lower pressures. Correlation between elasticity and octahedral distortion index (D) reveals that at certain D, the Al substitution reduces KS, G, VS, and VP of the stishovite phase while increasing G, VS, and VP of the post-stishovite phase. Importantly, the maximum shear reduction is slightly enhanced at D = 0.00620(9) at the transition point. Our results help explain the seismically observed small-scale VS anomalies beneath subduction regions in the shallow lower mantle where Al,H-bearing stishovite undergoes the post-stishovite transition. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email