Using aftershocks to Image the Subducting Pacific Plate in a Region of Deep Slow Slip, Hikurangi Margin, New Zealand

Wednesday, 17 December 2014
Katrina M Jacobs1, Hamish Hirschberg1, John N Louie2, Martha K Savage1 and Stephen C Bannister3, (1)Victoria University of Wellington, Wellington, New Zealand, (2)Univ of Nevada, Reno, NV, United States, (3)GNS Science, Lower Hutt, New Zealand
We present seismic migrations using aftershocks of two M>6 earthquakes as sources. The Southern Cook Strait earthquake sequence, beginning on 19 July 2013, included the 21 July M=6.5 and 16 August M=6.6 2013 earthquakes, which were the largest shallow earthquakes to strike the Wellington region since 1942. Following the two largest earthquakes we began the Seddon Earthquake Aftershock Structural Investigation (SEASI) and deployed a line of 21 seismometers stretching approximately 400 km along the strike of the Hikurangi subduction zone in order to use aftershocks to illuminate the structure of the subducted Pacific slab. The SEASI line ties into the SAHKE line, which was an array of up to 900 seismometers that recorded air gun and explosion shots in deployments from 2009-2011. The SAHKE project characterized the structures perpendicular to the strike of the subduction zone. Our results use the SAHKE line as a starting point and look for strike-parallel variations in the depth of the Moho and other structures. Previous studies have suggested potential changes along strike in this region, and deep slow slip events (> 35 km) are also observed north of Wellington, further indicating that variation in properties exists along slab strike.

We have used 246 M > 3 earthquakes that occurred from September 2013 through January 2014 to create common receiver gathers. Multicomponent prestack depth migration of these receiver gathers, with operator antialiasing control and prestack coherency filtering, produces reflectivity sections using a 1-D velocity model derived from the SAHKE project. Relocation of aftershocks of the Seddon earthquakes using the deployment of a temporary array by New Zealand GeoNet facilitates the migration. An initial P-P migration shows a north-dipping reflector at 15-25 km depth under the earthquake sequence, and suggests the Moho at 20-25 km depth. From Wellington, a reflector dips very gently south from 25-35 km depth, which is probably the slab interface. These results are helping to build 3-D information about the plate interface. We hope that this will help us to understand future hazards posed by subduction thrust earthquakes in this region, and the feeding system for deep slow-slip earthquakes.