In Situ Characterization of the Lithosphere‐Asthenosphere System beneath NW Pacific Ocean Via Broadband Dispersion Survey With Two OBS Arrays

Abstract

We conducted broadband dispersion survey by deploying two arrays of broadband ocean bottom seismometers in the northwestern Pacific Ocean at seafloor ages of 130 and 140 Ma. By combining ambient noise and teleseismic surface wave analyses, dispersion curves of Rayleigh waves were obtained at a period range of 5–100 s and then used to invert for one‐dimensional isotropic and azimuthally anisotropic βV (VSV) profiles beneath each array. The obtained profiles show ~2% difference in isotropic βV in the low‐velocity zone (LVZ) at a depth range of 80–150 km in spite of the small difference in seafloor ages and the horizontal distance of ~1,000 km. Forward dispersion‐curve calculation for thermal models indicates that simple cooling models cannot explain the observed difference and an additional mechanism, such as sublithospheric small‐scale convection, is required. In addition, the fastest azimuths of azimuthal anisotropy in the LVZ significantly deviate from the current plate motion direction. We infer that these observations are consistent with the presence of small‐scale convection beneath the study area. As for azimuthal anisotropy in the Lid, the peak‐to‐peak intensity is 3–4% at the depth from Moho to ~40 km. The fastest direction is almost perpendicular to magnetic lineation in area A at 130 Ma and oblique to magnetic lineations in area B at 140 Ma, suggesting complex mantle flow beneath the infant Pacific Plate surrounded by three ridge axes. The intensity of azimuthal anisotropy in the LVZ is ~2%, indicating that radial anisotropy is stronger than azimuthal anisotropy therein.

High‐Resolution Constraints on Pacific Upper Mantle Petrofabric Inferred From Surface‐Wave Anisotropy

“…The fast propagation direction of Rayleigh‐waves, Ψ G , is a proxy for the direction of shear strain in the mantle and is typically thought to be parallel to the FSD in the oceanic lithosphere (Nicolas & Christensen, ). While many seismic observations in the Pacific support this notion (e.g., Beghein et al, ; Debayle & Ricard, ; Eddy et al, ; Forsyth, ; Forsyth et al, ; Hess, ; Lin et al, ; Nishimura & Forsyth, ; Raitt et al, ; Smith et al, ; Weeraratne et al, ), other observations of fast wave speeds rotated from fossil spreading in the lithosphere challenge this simple model of spreading‐controlled fabric (Keen & Barrett, ; Morris et al, ; Shintaku et al, ; Takeo et al, , ; Toomey et al, , Vanderbeek & Toomey, ). Additionally, some global studies suggest that the correlation between Ψ G and fossil spreading breaks down for older aged seafloor (Debayle & Ricard, ), perhaps due to reheating processes at >80 Ma (Becker et al, ).…”

“…The Pacific is especially well‐suited for investigating plate evolution and MOR processes due to its broad range of plate ages, excellent distribution of seismic sources, and the recent proliferation of onshore and offshore data. Radial and azimuthal anisotropy have been extensively studied in the Pacific upper mantle at regional (Forsyth, ; Forsyth et al, ; French & Romanowicz, ; Gaherty et al, ; Lin et al, ; Nishimura & Forsyth, ; Tan & Helmberger, ; Takeo et al, , , , ; Weeraratne et al, ) and global scales (Beghein et al, ; Debayle & Ricard, ; Ekström & Dziewonski, ; Montagner & Tanimoto, , ; Moulik & Ekström, ; Montagner, ; Nettles & Dziewoński, ; Schaeffer et al, ; Yuan & Beghein, ), providing a comprehensive picture of the upper mantle LPO and flow field.…”

“…The strength of azimuthal anisotropy in the lithosphere is less well constrained, however, appearing weaker than the asthenosphere in some models (Beghein et al, ; Schaeffer et al, ; Yuan & Beghein, ) and stronger in others (Lin et al, ; Nishimura & Forsyth, ; Rychert & Harmon, ; Takeo et al, , ). Observations of radial anisotropy in the low velocity zone (LVZ) beneath the Pacific plate show ξ > 1, suggesting horizontal fabric interpreted as low‐viscosity channels of flow (Beghein et al, ; Nettles & Dziewoński, ).…”

“…Equation is only valid for Love waves when r is much larger than the wavelength of the waves being measured, which is true for this study. Previous studies have included amplitude information, A , to extract dispersion from multiple mode branches (Nishida et al, ; Takeo et al, , , , ). However, we have chosen to reduce the number of free parameters by letting A = 1, measuring dispersion for a single mode branch on each component, which is justified by our relatively tight group velocity window (Figure ).…”

“…These early findings are corroborated by a growing body of seismic observations from the oceanic lithosphere documenting variations in anisotropic fabric that appear to depend on seafloor‐spreading rate (e.g., Gaherty et al, ), differences in relative versus absolute plate motion (APM) at the ridge (e.g., Toomey et al, ; Takeo et al, ; Vanderbeek & Toomey, ), and/or the nature of upwelling beneath the ridge (e.g., Delorey et al, ; Gaherty, ). In the oldest reaches of the Pacific, the lithospheric anisotropy varies over relatively short length scales and does not always correlate with the direction of seafloor spreading (e.g., Shintaku et al, ; Takeo et al, , ). These observations suggest a rich diversity of dynamic processes beneath MORs that go well beyond the simple symmetrical‐spreading models explored to date (e.g., Blackman et al, ; Blackman & Kendall, ; Blackman et al, ).…”