High-resolution 3-D S-velocity structure in the D″ region at the western margin of the Pacific LLSVP: Evidence for small-scale plumes and paleoslabs

https://doi.org/10.1016/j.pepi.2020.106544Get rights and content

Highlights

  • We use data from a new seismic array in Thailand to infer 3-D Vs structure in the D″ region beneath the western Pacific.

  • We found small-scale low-Vs anomalies which we interpret as a cluster of small-scale plumes southeast of a high-Vs anomaly.

  • The inferred high-Vs anomalies are interpreted as the subducted Izanagi paleo-slab.

Abstract

Although previous tomographic studies found a large low S-velocity province (LLSVP) in the lowermost mantle beneath the Pacific, due to a lack of resolution it was unclear whether the LLSVP consists of clusters of small-scale low-velocity anomalies or large-scale anomalies. We recently deployed a seismic-array in Thailand which provides a dataset with wide azimuthal coverage of the western Pacific LLSVP. We analyze the new dataset using waveform inversion, and find high-velocity anomalies extending vertically to a height of ~400 km above the core-mantle boundary (CMB) beneath the Philippine Sea and small-scale low-velocity patches with a diameter of ~300 km at the CMB beneath New Guinea. The locations of the high-velocity anomalies are consistent with the past Izanagi-plate subduction boundary, and the low-velocity anomalies can be interpreted as a small-scale plume cluster. Hence we conclude that vertical flow (upwelling plumes and downwelling of slabs) is dominant in the lowermost mantle beneath the western Pacific region.

Introduction

The lowermost several hundred kilometers of the Earth's mantle immediately above the core-mantle boundary (CMB), called the D″ region, is the thermal boundary layer (TBL) of mantle convection; the D″ region is thought to be both thermally and compositionally heterogeneous. Previous global-scale tomographic studies (e.g., Ritsema et al., 2011; French and Romanowicz, 2014; Bozdağ et al., 2016) have revealed two separated large-scale low shear-velocity provinces at the CMB beneath the Pacific and Africa (LLSVPs; e.g., Garnero and McNamara, 2008; Garnero et al., 2016) surrounded by high-velocity regions that are interpreted as cold material associated with downward convection of former oceanic plates (e.g., Domeier et al., 2016; Shephard et al., 2017). Although the LLSVPs occupy ~30% of the area of the CMB, their origin and composition have been controversial (Garnero et al., 2016). Torsvik et al. (2006) suggested that the original large igneous province (LIP) eruption sites of the past ~200 Myr appear to be located possibly above the edges of the LLSVPs. A recent geodynamical study showed that plumes preferentially form at the edges of LLSVPs (Li and Zhong, 2017). The relation between eruption sites of LIPs and the margins of LLSVPs is still being debated (Doubrovine et al., 2016). In order to better understand the thermal and chemical evolution of the Earth it is therefore essential to investigate the fine seismic velocity structure of LLSVPs, especially around the margin of the LLSVPs.

Several geodynamic simulations suggest that the LLSVPs represent small-scale clusters of (relatively chemically homogeneous) thermal plumes seen as continuous structures due to a lack of resolution of seismic tomography (e.g., Bull et al., 2009; Bunge et al., 1998; Schubert et al., 2004; Schuberth et al., 2009; Davies et al., 2012). On the other hand, thermochemical effects, such as enrichment in dense material including iron as well as incompatible elements, were suggested to be a dominant cause of large-scale low-velocity anomalies associated with large chemically distinct piles which are also considered to be a primordial geochemical reservoir (e.g., Zhang et al., 2010; Li et al., 2014). The former and latter models are called ‘plume clusters’ and ‘thermo-chemical piles,’ respectively. It has been desired to obtain high-resolution imaging of the structure inside and at the edge of the LLSVP in order to gain a better understanding of whether the low-velocity anomaly is due to thermal clustering or thermochemical piles (Davies et al., 2012).

Due to the relatively favorable locations of seismic sources and receivers, D″ beneath the western Pacific is one of the most suitable regions for inferring the detailed seismic structure within and at the edge of the LLSVP. However, the detailed seismic velocity structure in D″ beneath the western Pacific remains controversial, because of the limitations of previous imaging methods and data coverage. For example, the location of the western boundary of the LLSVP differs from model to model (see Fig. 1a).

In general, short-wavelength low-velocity anomalies are difficult to image with traveltime tomography, because low-velocity anomalies have only a small effect on traveltimes, due to wavefront healing effects (Nolet and Dahlen, 2000). Previous studies using both P- and S-waves reported both amplitude effects such as focusing/defocusing due to finite-frequency and waveform complexity due to multi-pathing, implying the existence of laterally heterogeneous low-velocity structure in our study region (He et al., 2006; Sun et al., 2009; Tanaka et al., 2015). In order to better image the detailed seismic structure in D″ beneath the western Pacific, it is necessary to apply a method which can take into account finite frequency effects, using a dataset of broadband seismic waveforms with wide azimuthal and epicentral distance coverage.

In this study we image the structure using waveform inversion, which is able to make full use of all the information in waveforms, such as amplitudes and complex interacting phases, which is difficult to handle using either traveltime tomography or waveform forward modeling. Our group recently applied localized waveform inversion techniques (Kawai et al., 2014) to much larger datasets to invert for the 3-D S-velocity structure in D″ beneath the northern Pacific (Suzuki et al., 2016) and Central America (Borgeaud et al., 2017) using relatively shorter period data (up to 12.5 s period), and in and around the mantle transition zone beneath Central America (Borgeaud et al., 2019), with high resolution (150–250 km laterally; ~50 km vertically) for both low- and high-velocity anomalies.

Section snippets

Data and methods

In order to improve data coverage in D″ beneath the western Pacific we recently deployed a seismic array of 40 portable broadband stations throughout Thailand (Thai Seismic Array—TSAR; Tanaka et al., 2019), which operated from late 2016 to early 2019. The Australian National Seismograph Network (AU) also provides excellent quality waveform data which sample our target region in the North-South direction. We combined waveforms recorded by TSAR and AU stations as well as other temporary and

3-D S-velocity structure at the western margin of the Pacific LLSVP

Map views of the S-velocity model obtained by our inversion using data in the period range between 12.5 and 200 s are shown in Fig. 3(a). The final model achieved a variance reduction of more than 30% in the ‘improvement window’ (from 20 s before to 30 s after the expected ScS arrival for reference model, PREM), which is the part of the seismogram that is most sensitive to the structure in D″ (Table 1). We note the following key features. (i) There are significant (up to ~3.5%) high-velocity

Discussions and geophysical implications

Our inferred model (Fig. 3, Fig. 4) shows distinct high-velocity anomalies beneath the Philippine Sea and vertically continuous (to a height of at least 400 km above the CMB) low-velocity anomalies, to the southeast of the high-velocity anomalies, which is generally consistent with previous tomographic models beneath the western Pacific (e.g., Grand, 2002; Takeuchi, 2012; French and Romanowicz, 2014). Since the S-velocity anomalies can be primarily attributed to effects of temperature, the

Conclusions

We recently deployed a seismic array in Thailand (TSAR) which provides a dataset with wide azimuthal coverage at the western edge of the Pacific LLSVP. In this study we analyze the new dataset using localized waveform inversion, which has better resolution, especially for low-velocity anomalies, than traveltime tomography. We find a high-velocity anomaly extending vertically to a height of at least 400 km above the CMB beneath the Philippine Sea and small-scale low-velocity anomalies

CRediT authorship contribution statement

Yuki Suzuki:Methodology, Formal analysis, Validation, Writing - original draft.Kenji Kawai:Data curation, Methodology, Writing - review & editing.Robert J. Geller:Methodology, Writing - review & editing.Satoru Tanaka:Project administration, Data curation, Writing - review & editing.Weerachai Siripunvaraporn:Data curation, Writing - review & editing.Songkhun Boonchaisuk:Data curation, Writing - review & editing.Sutthipong Noisagool:Data curation, Writing - review & editing.Yasushi Ishihara:Data

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This research was partly supported by grants from the Japanese Ministry of Education, Culture, Sports, Science and Technology (Nos. 16K05531, 15K17744, 15H05832, and 18K03797), Earthquake Research Institute of the University of Tokyo (ERI JURP 2016-F2-07, 2017-F2-07, 2018-F2-07) and a grant for Excellent Young Researchers at The University of Tokyo to K.K. The authors thank all the people in the TSAR project for installing and servicing the seismic array, especially Koji Miyakawa, Benjawut

Data availability

All data needed to evaluate the conclusions in the paper are presented in the paper and/or the Supplementary materials. The seismic waveforms except observed at TSAR stations can be downloaded freely from the various data centers noted above. The TSAR data analyzed during the current study will become publicly available in spring 2021 in the Ocean Hemisphere Research Center (OHRC; network name is “Pacific21” and network code is “PS”), the Earthquake Research Institute (ERI), the University of

References (66)

  • N. Takeuchi

    Detection of ridge-like structures in the Pacific large low-shear-velocity province

    Earth Planet. Sci. Lett.

    (2012)
  • T. Tsuchiya et al.

    Phase transition in MgSiO3 perovskite in the Earth's lower mantle

    Earth Planet. Sci. Lett.

    (2004)
  • A. Young et al.

    Global kinematics of tectonic plates and subduction zones since the late Paleozoic Era

    Geosci. Front.

    (2019)
  • Y. Yu et al.

    Lateral variations of crustal structure beneath the Indochina Peninsula

    Tectonophysics

    (2017)
  • A.F.E. Borgeaud et al.

    Imaging paleoslabs in the D″ layer beneath Central America and the Caribbean using seismic waveform inversion

    Sci. Adv.

    (2017)
  • A.F.E. Borgeaud et al.

    3-D S-velocity structure of the mantle transition zone beneath Central America and the Gulf of Mexico using waveform inversion

    J. Geophys. Res.: Solid Earth

    (2019)
  • E. Bozdağ et al.

    Global adjoint tomography: first-generation model

    Geophys. J. Int.

    (2016)
  • J.M. Brown et al.

    Thermodynamic parameters in the Earth as determined from seismic profiles

    Geophys. J. R. Astron. Soc.

    (1981)
  • H.-P. Bunge et al.

    Time scales and heterogeneous structure in geodynamic Earth models

    Science

    (1998)
  • M. Domeier et al.

    Global correlation of lower mantle structure and past subduction

    Geophys. Res. Lett.

    (2016)
  • P.V. Doubrovine et al.

    A failure to reject: testing the correlation between large igneous provinces and deep mantle structures with EDF statistics

    Geochem. Geophys. Geosyst.

    (2016)
  • S.W. French et al.

    Whole-mantle radially anisotropic S-velocity structure from spectral-element waveform tomography

    Geophys. J. Int.

    (2014)
  • E.J. Garnero et al.

    Structure and dynamics of Earth's lower mantle

    Science

    (2008)
  • E.J. Garnero et al.

    Continent-sized anomalous zones with low seismic velocity at the base of Earth's mantle

    Nat. Geosci.

    (2016)
  • R.J. Geller et al.

    Two efficient algorithms for iterative linearized inversion of seismic waveform data

    Geophys. J. Int.

    (1993)
  • R.J. Geller et al.

    Computation of synthetic seismograms and their partial derivatives for heterogeneous media with arbitrary natural boundary conditions using the direct solution method

    Geophys. J. Int.

    (1994)
  • S. Goes et al.

    Subduction-transition zone interaction: a review

    Geosphere

    (2017)
  • S.P. Grand

    Mantle shear-wave tomography and the fate of subducted slabs

    Philos. Trans. R. Soc., Math. Phys. Eng. Sci.

    (2002)
  • Y. He et al.

    Geographic boundary of the “Pacific Anomaly” and its geometry and transitional structure in the north

    J. Geophys. Res.

    (2012)
  • C. Houser et al.

    Shear and compressional velocity models of the mantle from cluster analysis of long-period waveforms

    Geophys. J. Int.

    (2008)
  • K. Idehara et al.

    High-velocity anomaly adjacent to the western edge of the Pacific low-velocity province

    Geophys. J. Int.

    (2013)
  • K. Kawai et al.

    Temperature profile in the lowermost mantle from seismological and mineral physics joint modeling

    Proc. Natl. Acad. Sci.

    (2009)
  • K. Kawai et al.

    Complete synthetic seismograms up to 2 Hz for transversely isotropic spherically symmetric media

    Geophys. J. Int.

    (2006)
  • Cited by (0)

    1

    Present address: Geoscience Program, Mahidol University, Kanchanaburi Campus, Saiyok, Kanchanaburi, Thailand and Thailand Center of Excellence in Physics, Ministry of Higher Education, Science, Research and Innovation, 328 Si Ayutthaya Road, Bangkok 10400, Thailand.

    View full text