Sub‐slab anisotropy beneath the Sumatra and circum‐Pacific subduction zones from source‐side shear wave splitting observations

Abstract

Understanding the dynamics of subduction is critical to our overall understanding of plate tectonics and the solid Earth system. Observations of seismic anisotropy can yield constraints on deformation patterns in the mantle surrounding subducting slabs, providing a tool for studying subduction dynamics. While many observations of seismic anisotropy have been made in subduction systems, our understanding of the mantle beneath subducting slabs remains tenuous due to the difficulty of constraining anisotropy in the sub-slab region. Recently, the source-side shear wave splitting technique has been refined and applied to several subduction systems worldwide, making accurate and direct measurements of sub-slab anisotropy feasible and offering unprecedented spatial and depth coverage in the sub-slab mantle. Here we present source-side shear wave splitting measurements for the Central America, Alaska-Aleutians, Sumatra, Ryukyu, and Izu-Bonin-Japan-Kurile subduction systems. We find that measured fast splitting directions in these regions generally fall into two broad categories, aligning either with the strike of the trench or with the motion of the subducting slab relative to the overriding plate. Trench parallel fast splitting directions dominate beneath the Izu-Bonin, Japan, and southern Kurile slabs and part of the Sumatra system, while fast directions that parallel the motion of the downgoing plate dominate in the Ryukyu, Central America, northern Kurile, western Sumatra, and Alaska-Aleutian regions. We find that plate motion parallel fast splitting directions in the sub-slab mantle are more common than previously thought. We observe a correlation between fast direction and age of the subducting lithosphere; older lithosphere (>95 Ma) is associated with trench parallel splitting while younger lithosphere (<95 Ma) is associated with plate motion parallel fast splitting directions. Finally, we observe source-side splitting for deep earthquakes (transition zone depths) beneath Japan and Sumatra, suggesting the presence of anisotropy at midmantle depths beneath these regions.

Mid-mantle seismic anisotropy beneath southwestern Pacific subduction systems and implications for mid-mantle deformation

“…We ensured that each station had backazimuthal coverage sufficient to show consistent null or simple splitting in at least two quadrants. Any station with insufficient backazimuthal coverage or splitting that varied with backazimuth was not used (further details of receiver-side anisotropy corrections can be found in Lynner and Long, 2014). While this conservative approach to station selection may limit the number of stations we can use in our study, it also minimizes potential errors due to inaccurate receiver-side corrections, as we only use stations for which we are highly confident in our ability to correct for upper mantle anisotropy on the receiver side.…”

“…For the purpose of this discussion, we combine our results with previously published measurements obtained using identical station selection, receiver side correction techniques, and data processing procedures. These measurements come from the work of Foley and Long (2011) for Tonga, Lynner and Long (2014) for Sumatra and Japan, and Lynner and Long (2015) for Izu-Bonin, Japan-Kuriles, and South America. A schematic diagram summarizing first-order observations from different subduction zones is shown in Fig.…”

“…Several studies have documented seismic anisotropy at mid-mantle depths using both body waves (Wookey et al, 2002;Chen and Brudzinski, 2003;Wookey and Kendall, 2004;Foley and Long, 2011;Di Leo et al, 2012;Lynner and Long, 2014;Nowacki et al, 2015) and surface waves (Trampert and van Heijst, 2002;Yuan and Beghein, 2013), although they are not plentiful compared to studies of upper mantle anisotropy. Global models by Trampert and van Heijst (2002) using dispersion measurements of Love wave overtones suggested the global presence of azimuthal anisotropy in the transition zone.…”

Erratum to: Crystal preferred orientations of olivine, orthopyroxene, serpentine, chlorite, and amphibole, and implications for seismic anisotropy in subduction zones: a review

“…This pattern of seismic anisotropy may be explained by the change in olivine CPO from Type-B in the fore-arc to Type-A, -C (or -E) in the back-arc (Kneller et al, 2005;Katayama and Karato, 2006;Kneller et al, 2007;Karato et al, 2008;Jung, 2012). The trench-parallel seismic anisotropy was also observed in the subducting slab and below the slab at a depth greater than 100 km in the subduction zone (Russo and Silver, 1994;Müller et al, 2008;Long and Silver, 2009;Abt et al, 2010;Di Leo et al, 2014;Lynner and Long, 2014;Lynner et al, 2017). Because of the relatively dry conditions below the slab (Karato et al, 2008), one possible explanation of this seismic anisotropy is the Type-B olivine CPO that can be produced by high pressure in dry conditions (Jung et al, 2009b;Ohuchi et al, 2011;Soustelle and Manthilake, 2017).…”

“…Azimuthal anisotropy of Pwaves (Vp) and polarization anisotropy of S-waves are shown (AVs is a contour plot of the magnitude of shear wave polarization anisotropy and Vs1 is a plot of the polarization direction of fast Swaves along different orientations of propagation). Lynner and Long, 2014). The source of this trenchparallel seismic anisotropy was proposed as: (1) the trenchparallel mantle flow due to trench-roll back (Russo and Silver, 1994;Long and Silver, 2008;Long and Silver, 2009); (2) Type-B CPO of olivine (Jung and Karato, 2001;Kneller et al, 2005;Jung et al, 2006;Katayama and Karato, 2006;Kneller and van Keken, 2007;Karato et al, 2008); (3) the 3-D mantle flow around the slab Capitanio, 2012, 2013;Li et al, 2014; et al, 2017); (4) fault or crack-induced anisotropy in the slab (Faccenda et al, 2008;Healy et al, 2009); (5) strong radial anisotropy Kawakatsu, 2012, 2013); (6) CPO of serpentine in the slab-mantle interface and in the mantle wedge (Katayama et al, 2009;Jung, 2011;Nagaya et al, 2016); and (7) CPOs of chlorite (Kim and Jung, 2015;Kang and Jung, submitted) and amphibole in the lower part of mantle wedge (Ko and Jung, 2015;Kang and Jung, submitted).…”