Effects of Sediment Layer and Shallow Portion of the Oceanic Crust on Waveforms of Broadband Ocean Bottom Seismometers in Northwest Pacific Ocean

Abstract:

Earthquake Research Institute, The University of Tokyo and Japan Agency for Marine-Earth Science and Technology have conducted seismic observation in the northwest Pacific Ocean with broadband ocean bottom seismometers (BBOBSs), for understanding the structure of the Earth’s interior and the mechanism of plate motion (Normal Mantle Project). We have performed receiver function (RF) analyses using the waveform data, for detecting velocity discontinuities in the upper mantle, and have understood that it is essential to reveal shallower structure (especially structure of sediment) for elucidating the upper mantle structure using RFs (Abe et al., 2014, SSJ meeting; 2015, JpGU meeting). Therefore, we attempted to estimate the shallower structure by using power spectrum and auto correlation function (ACF) of ambient noise in addition to RFs.

Power spectrum of horizontal seismogram of a BBOBS has several peaks due to resonances of S wave in the sediment. Godin & Chapman (1999, J. Acoust. Soc. Am.) introduced a method to estimate a 1-D velocity distribution in the sediment from the resonance frequencies. From the location of spectral peaks of a station (NM14), we estimated the velocity distribution to be Vs(z) = 0.519z^0.473 (Vs: S wave velocity (km/s), z: depth (km)), assuming a sediment layer thickness of 0.3 km. Two way S wave travel time in this sediment corresponds to the arrival time of a prominent negative ACF peak of horizontal seismogram of the station. On the other hand, for P-wave RFs (0.4-2.0 Hz) of the station, the arrival time of the first positive peak is not explained only by the estimated sediment structure, and another discontinuity located a few hundred meters deeper than the bottom of the sediment is necessary to explain it.

We attempt to constrain the structure of the sediment and shallow portion of the oceanic crust by analyzing RF waveforms in more detail that also explains power spectrum and ACF of ambient noise.