11:00 AM - 11:15 AM
[SSS03-08] Broadband dispersion analysis of Love waves: a case study in the oldest Pacific Ocean
Keywords:radial anisotropy, Love waves, Pacific plate, lithosphere-asthenosphere system
Both Rayleigh- and Love-wave phase velocities are essential for estimating radial anisotropy, which is controlled by the formation of the mantle fabrics and/or structural layering. Combining radially and azimuthally anisotropic and isotropic structures, we can achieve a comprehensive understanding of the lithosphere-asthenosphere system. Although there are well-established broadband array analysis methods of measuring the phase velocity of Rayleigh waves recorded by arrays of ocean bottom seismometers, the measurement of long-period (>10–30 s) Love-wave phase velocity has been rare for technical difficulties (e.g., Takeo et al., 2016).
Due to the nature of the oceanic crust and mantle structures (thin crust and the well-developed low-velocity zone) (Thatcher and Brune, 1969; Takeo et al., 2015, SSJ), different modes of Love waves arrive at the observation points with similar travel times and overlap with each other, resulting in the mode interference. Thus, if Love waves are analyzed in the same manner as Rayleigh waves without any special care, which typically assumes the observed waveforms as the fundamental mode, the higher modes may bias the measurement of the fundamental-mode Love-wave (0T-mode) phase velocity. We develop a new array analysis method and attempt to reduce such bias by treating the observed Love waves as a summed wave of the fundamental and the first higher modes that have relatively large amplitudes compared to the other modes. The bias in the 0T-mode phase velocities due to the presence of the higher modes is successfully reduced for the synthetic data with the remaining bias of ~1 %, which is small compared to the bias obtained without such care (~3 %).
We apply the new method for the teleseismic Love waves recorded by the Oldest-1 Array (http://eri-ndc.eri.u-tokyo.ac.jp/PacificArray/Oldest/) situated on the oldest Pacific seafloor ~1,000 km off Mariana Trench. The array has set along the 170- and 180-Ma isochrones with the eastern edge overlapping an ancient ridge-ridge-ridge (RRR) triple junction. We measured 0T-mode phase velocity at a period range of 33–100 s. Combined with the short period (5–10 s) 0T-mode and broadband (5–200 s) Rayleigh-wave phase velocities measured using ambient noise and teleseismic waveforms, we constrained the radially anisotropic structure, RAS=(VSH/VSV-1)×100, in two layers: Moho–60 km and 60–240 km. RAs is significantly different at shallow depths between the eastern (~7.5 %) and western (~3.4 %) areas of the array. At the deeper part, RAS is 2–8 %. The radial anisotropy was jointly interpreted with azimuthal anisotropy, and we constrained olivine fabric types. A-type olivine fabric is likely dominant beneath the Oldest-1 Array assuming the horizontal shearing, except the shallow eastern part, which showed anomalously strong radial anisotropy. The variation observed within the array may reflect the complicated evolution process of the early Pacific plate, which involved the mantle upwelling at the RRR triple junction.
Due to the nature of the oceanic crust and mantle structures (thin crust and the well-developed low-velocity zone) (Thatcher and Brune, 1969; Takeo et al., 2015, SSJ), different modes of Love waves arrive at the observation points with similar travel times and overlap with each other, resulting in the mode interference. Thus, if Love waves are analyzed in the same manner as Rayleigh waves without any special care, which typically assumes the observed waveforms as the fundamental mode, the higher modes may bias the measurement of the fundamental-mode Love-wave (0T-mode) phase velocity. We develop a new array analysis method and attempt to reduce such bias by treating the observed Love waves as a summed wave of the fundamental and the first higher modes that have relatively large amplitudes compared to the other modes. The bias in the 0T-mode phase velocities due to the presence of the higher modes is successfully reduced for the synthetic data with the remaining bias of ~1 %, which is small compared to the bias obtained without such care (~3 %).
We apply the new method for the teleseismic Love waves recorded by the Oldest-1 Array (http://eri-ndc.eri.u-tokyo.ac.jp/PacificArray/Oldest/) situated on the oldest Pacific seafloor ~1,000 km off Mariana Trench. The array has set along the 170- and 180-Ma isochrones with the eastern edge overlapping an ancient ridge-ridge-ridge (RRR) triple junction. We measured 0T-mode phase velocity at a period range of 33–100 s. Combined with the short period (5–10 s) 0T-mode and broadband (5–200 s) Rayleigh-wave phase velocities measured using ambient noise and teleseismic waveforms, we constrained the radially anisotropic structure, RAS=(VSH/VSV-1)×100, in two layers: Moho–60 km and 60–240 km. RAs is significantly different at shallow depths between the eastern (~7.5 %) and western (~3.4 %) areas of the array. At the deeper part, RAS is 2–8 %. The radial anisotropy was jointly interpreted with azimuthal anisotropy, and we constrained olivine fabric types. A-type olivine fabric is likely dominant beneath the Oldest-1 Array assuming the horizontal shearing, except the shallow eastern part, which showed anomalously strong radial anisotropy. The variation observed within the array may reflect the complicated evolution process of the early Pacific plate, which involved the mantle upwelling at the RRR triple junction.