Temporal and spatial changes in full-waveform inversion velocity models along the northern Hikurangi subduction margin

Abstract:

The northern Hikurangi subduction margin experiences shallow slow-slip events (SSE) (< 5km) with a relatively frequent interval of 18 – 24 months. The physical mechanism behind these shallow SSEs is poorly understood but is often hypothesized to be related to high pore-fluid pressure and/or transitional frictional properties within the megathrust fault zone. Recent IODP drilling has revealed that the fault zone may be highly heterogeneous in zones of SSEs, and seismic and magnetic data has suggested the presence of a subducted seamount.

In 2018, a 3D marine seismic multi-channel streamer dataset (NZ3D) was collected across the northern Hikurangi margin extending for ~15 km along-strike offshore of Gisborne, with shots also being recorded on ocean bottom and onshore seismometer stations. We are applying full-waveform inversion (FWI) to these data to recover a high-resolution P-wave velocity model in order to help understand the driving processes of shallow SSEs at the north Hikurangi subduction zone. The ultimate aim is to perform 3D FWI using all the NZ3D seismic datasets, but, prior to this, we are applying 2D FWI to several streamer lines taken from across the whole width of the acquisition area to reveal the shallow (upper 1-2 km) velocity structure. In this presentation, we will discuss variations in the shallow velocity structure within a 15 km along-strike distance along the margin.

Furthermore, we will compare our results with an FWI velocity model obtained using a seismic streamer profile acquired in 2005 (Gray et al. 2019, JGR) that lies within the 3D seismic acquisition area. This enables an investigation of how the velocity structure of this part of the margin has changed over the 13 year time interval. The velocity model from the 2005 data shows low-velocity zones along thrust faults within the accretionary prism potentially indicating the presence of fluids and high pore-fluid pressure. We will investigate the temporal and spatial evolution of these low-velocity zones as well as other significant changes in the velocity structure beneath bottom-simulating reflectors within the accretionary prism.