An official website of the United States government
Abstract Regularization of seismic inversions has a strong imprint on tomographic images. We analyze recorded and spectral‐element S, Sdiff, and SS waveforms to evaluate the benefit of body‐wave amplitudes in global tomography. L‐curve analysis for S40RTS models with recorded and synthetic waveforms show that SS‐S traveltimes and SS/S amplitude ratios have minima within the same damping parameter range. SS/S ratios for S40RTS and model GLAD‐M25 show the trade‐off between scale‐length and strength of lowermost‐mantle heterogeneities. The recorded SS/Sdiff ratios are lower than predicted by 3D mantle models which may be explained by a decrease in the mean shear velocity by at the lowermost 200 km of the mantle. Our results suggest that SS/S amplitude measurements made with 3D waveforms can be used to constrain damping in linearized inversions, and amplitudes are essential for studying the size of heterogeneities.
more »
« less
Data-space cross-validation of mantle structure in global tomographic models underneath the Pacific Ocean
SUMMARY Seismic tomography is a principal method for studying mantle structure, but imaging of Earth’s wave speed anomalies is conditioned by seismic wave sampling. Global models use misfit criteria that may strive for balance between portions of the data set but can leave important regional domains underserved. We evaluate two full-waveform global tomography wave speed models, GLAD-M25 (Global adjoint tomography model) and SEMUCB-WM1 (whole-mantle tomography model derived from fully numerical spectral element method forward modelling), in the mantle below the Pacific Ocean. The region of the South Pacific Superswell contains multiple hotspots which may be fed by plumes anchored in the Large Low Shear-Velocity Province at the base of the mantle. The uneven distribution of seismic receivers worldwide leaves several candidate plumes beneath various hotspots poorly resolved. We assess the regional quality of GLAD-M25 relative to its global performance using a partition of the seismic waveform data used in its construction. We evaluate synthetic waveforms computed using the spectral-element method to determine how well they fit the data according to a variety of criteria measured across multiple seismic phases and frequency bands. The distributions of traveltime anomalies that remain in GLAD-M25 are wider for trans-Pacific paths than globally, suggesting comparatively insufficiently resolved seismic velocity structure in the region of interest. Hence, Pacific-centred regional inversions, based on (augmented) subsets of the global data set have the potential to enhance the resolution of velocity structure. We compare GLAD-M25 and SEMUCB-WM1 by cross-validation with a new, independent, data set. Our results reveal that short- and long-wavelength structure is captured differently by the two models. Our findings lead us to recommend focusing future model iteration on and around the Pacific Superswell and adding data that sample new corridors, especially using ocean sensors, to better constrain seismic velocity structure in this area of significant geodynamic complexity.
more »
« less
- Award ID(s):
- 2341811
- PAR ID:
- 10571969
- Publisher / Repository:
- Oxford University Press
- Date Published:
- Journal Name:
- Geophysical Journal International
- Volume:
- 241
- Issue:
- 1
- ISSN:
- 0956-540X
- Format(s):
- Medium: X Size: p. 241-259
- Size(s):
- p. 241-259
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Our understanding of the tectonic development of the African continent and the interplay between its geological provinces is hindered by unevenly distributed seismic instrumentation. In order to better understand the continent, we used long‐period ambient noise full‐waveform tomography on data collected from 186 broadband seismic stations throughout Africa and surrounding regions to better image the upper mantle structure. We extracted empirical Green's functions from ambient seismic noise using a frequency‐time normalization method and retrieved coherent signal at periods of 7–340 s. We simulated wave propagation through a heterogeneous Earth using a spherical finite‐difference approach to obtain synthetic waveforms, measured the misfit as phase delay between the data and synthetics, calculated numerical sensitivity kernels using the scattering integral approach, and iteratively inverted for structure. The resulting images of isotropic, shear wave speed for the continent reveal segmented, low‐velocity upper mantle beneath the highly magmatic northern and eastern sections of the East African Rift System (EARS). In the southern and western sections, high‐velocity upper mantle dominates, and distinct, low‐velocity anomalies are restricted to regions of current volcanism. At deeper depths, the southern and western EARS transition to low velocities. In addition to the EARS, several low‐velocity anomalies are scattered through the shallow upper mantle beneath Angola and North Africa, and some of these low‐velocity anomalies may be connected to a deeper feature. Distinct upper mantle high‐velocity anomalies are imaged throughout the continent and suggest multiple cratonic roots within the Congo region and possible cratonic roots within the Sahara Metacraton.more » « less
-
Abstract The shapes of mantle plumes are sensitive to mantle viscosity, density structure, and flow patterns. Increasingly, global tomographic models reveal broad plume conduits in the lower mantle and highly tilting conduits in the mid and upper mantle. Previous studies mostly relied on 2D slices to analyze plume shapes, but fully investigating the complexity of 3D plume structures requires more effective visualization methods. Here, we use immersive headset‐based virtual reality (VR) to visualize the full‐waveform global tomographic models SEMUCB‐WM1 and GLAD‐M25. We develop criteria for the identification of plume conduits based on the relationship between the plume excess temperature and theVSanomaly (δVS). We trace 20 major plume conduits, measure the offsets of the conduits in azimuth and distance with respect to the hotspots, calculate the tilt angle, and evaluate theδVSalong all traced conduits. We compare our traced conduits with the conduits predicted by global mantle convection models and vertical conduits. The wavespeed variations along conduits traced from each tomographic model are slower than modeled or vertical conduits, regardless of which tomographic model they are evaluated in. The shapes of traced conduits tend to differ greatly from modeled conduits. Plume ponding and the emergence of secondary plumes, which could result from a combination of compositional variations, phase transitions, small‐scale convection, and variations in viscosity, can contribute to the complex observed plume shapes. The variation ofδVSalong the traced conduits and complex plume shapes suggest a thermochemical origin of many plumes.more » « less
-
null (Ed.)Seismology provides important constraints on the structure and dynamics of the deep mantle. Computational and methodological advances in the past two decades improved tomographic imaging of the mantle and revealed the fine-scale structure of plumes ascending from the core-mantle boundary region and slabs of oceanic lithosphere sinking into the lower mantle. We discuss the modeling aspects of global tomography including theoretical approximations, data selection, and model fidelity and resolution. Using spectral, principal component, and cluster analyses, we highlight the robust patterns of seismic heterogeneity, which inform us of flow in the mantle, the history of plate motions, and potential compositionally distinct reservoirs. In closing, we emphasize that data mining of vast collections of seismic waveforms and new data from distributed acoustic sensing, autonomous hydrophones, ocean-bottom seismometers, and correlation-based techniques will boost the development of the next generation of global models of density, seismic velocity, and attenuation. ▪ Seismic tomography reveals the 100-km to 1,000-km scale variation of seismic velocity heterogeneity in the mantle. ▪ Tomographic images are the most important geophysical constraints on mantle circulation and evolution.more » « less
-
SUMMARY Differences between P- and S-wave models have been frequently used as evidence for the presence of large-scale compositional heterogeneity in the Earth's mantle. Our two-step machine learning (ML) analysis of 28 P- and S-wave global tomographic models reveals that, on a global scale, such differences are for the most part not intrinsic and could be reduced by changing the models in their respective null spaces. In other words, P- and S-wave images of mantle structure are not necessarily distinct from each other. Thus, a purely thermal explanation for large-scale seismic structure is sufficient at present; significant mantle compositional heterogeneities do not need to be invoked. We analyse 28 widely used tomographic models based on various theoretical approximations ranging from ray theory (e.g. UU-P07 and MIT-P08), Born scattering (e.g. DETOX) and full-waveform techniques (e.g. CSEM and GLAD). We apply Varimax principal component analysis to reduce tomography model dimensionality by 83 percent, while preserving relevant information (94 percent of the original variance), followed by hierarchical clustering (HC) analysis using Ward's method to quantitatively categorize all models into hierarchical groups based on similarities. We found two main tomography model clusters: Cluster 1, which we called ‘Pure P wave’, is composed of six P-wave models that only use longitudinal body wave phases (e.g. P, PP and Pdiff); and Cluster 2, which we called ‘Mixed’, includes both P- and S-wave models. P-wave models in the ‘Mixed’ cluster use inversion methods that include inputs from other geophysical and geological data sources, and this causes them to be more similar to S-wave models than Pure P-wave models without significant loss of fitness to P-wave data. Given that inclusion of new data classes and seismic phases in more recent tomographic models significantly changes imaged seismic structure, our ML assessment of global tomography model similarity may improve selection of appropriate P- and S-wave models for future global tomography comparative studies.more » « less
