Diffraction image of subduction of the Hikurangi Plateau along the SAHKE transect, lower North Island, New Zealand
Abstract
As part of the Seismic Array HiKurangi Experiment (SAHKE) project, we acquired wide-angle reflection / refraction seismic data using ocean bottom seismometers (OBSs) along a transect across the southern North Island of New Zealand, where the Hikurangi Plateau, an early Cretaceous large igneous province, subducts westward beneath Wellington, the capital city of New Zealand. The SAHKE project was designed to investigate the physical parameters controlling locking at the plate interface beneath the southern North Island and characterize slip processes in a major segment of the Hikurangi system. We deployed 16 OBSs with 5 km spacing off the east coast. Airgun sources were shot at every 100 m along an onshore-offshore transect. Although data from OBSs at shallow depths (~100 m) contain large amplitude ambient noise, first arrivals from the airgun sources can be traced up to over 100 km offset on record sections of most OBSs. We applied first-arrival travel-time inversion in order to obtain P-wave velocity structure along the 80 km-long OBS profile. Starting with a simple stratified velocity model including subduction structure, we iteratively revise the initial model utilizing constraints from the first arrival picks. The velocity structure to ~25 km depth is well resolved, with the down going slab marked by a velocity gradient from 5 km/s to 7.5 km/s. In order to visually compare the velocity structure and the geometries of reflection interfaces in depth, we calculated a diffraction migration image section for the OBS array. First, we calculated P-wave travel times from each shot or each OBS to grid points of 100 m interval within the 2D model space using our derived velocity structure. The diffraction imaging condition is the summed source and receiver travel times of P-wave at every grid point. Here we have used 16 (OBSs) times 800 (shots) travel time tables. Seismic traces have been deconvolved with filters designed to transform the source signatures to the 15 Hz Ricker wavelet. We stacked a window segment of deconvolved waveforms for each shot-receiver pair. The length of the waveform segment is 0.14 seconds for the 15 Hz Ricker wavelet. If a diffracted or scattered energy exists at a grid point, waveforms of shot traces at their calculated travel times should be constructively stacked, while the diffraction angles are limited within the range of critical angles. Finally, we calculate a cross correlation coefficient between the stacked waveform and the Ricker wavelet at each grid point to represent the total diffracted amplitude. The resulting image successfully depicts fine structure at shallow depths, the subducting interface and reveals internal structure within the subducting Hikurangi Plateau. Previously, we had identified later arrivals with very fast apparent velocities (8.5 km/s) from beneath the Hikurangi Plateau crust. The depth image shows a rather constant thickness of ~13 km of the subducting oceanic crust dipping to the west from offshore near the trough axis. Such arrivals are not reversed but fast velocities can be a common characteristic beneath the oceanic crust of the Hikurangi Plateau.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2013
- Bibcode:
- 2013AGUFM.T43D2691M
- Keywords:
-
- 3025 MARINE GEOLOGY AND GEOPHYSICS Marine seismics;
- 3060 MARINE GEOLOGY AND GEOPHYSICS Subduction zone processes;
- 7230 SEISMOLOGY Seismicity and tectonics