S41B-2748
Using Ambient Seismic Noise to Monitor Post-Seismic Relaxation After the 2010 M7.1 Darfield Earthquake, New Zealand

Thursday, 17 December 2015
Poster Hall (Moscone South)
Rachel Heckels1, Martha K Savage1 and John Townend2, (1)Victoria University of Wellington, Wellington, New Zealand, (2)Victoria University of Wellington, School of Geography, Environment, and Earth Sciences,, Wellington, New Zealand
Abstract:
Quantifying seismic velocity changes following large earthquakes can provide insights into the crustal response of the earth. The use of ambient seismic noise to monitor these changes is becoming increasingly widespread. Cross-correlations of long-duration ambient noise records can be used to give stable impulse response functions without the need for repeated seismic events.

Temporal velocity changes were detected in the four months following the September 2010 Mw 7.1 Darfield event in South Island, New Zealand, using temporary seismic networks originally deployed to record aftershocks in the region. The arrays consisted of stations lying on and surrounding the fault, with a maximum inter-station distance of 156km. The 2010-2011 Canterbury earthquake sequence occurred largely on previously unknown and buried faults. The Darfield earthquake was the first and largest in a sequence of events that hit the region, rupturing the Greendale Fault. A surface rupture of nearly 30km was observed. The sequence also included the Mw 6.3 February 2011 Christchurch event, which caused widespread damage throughout the city and resulted in almost 200 deaths.

Nine-component, day-long Green's functions were computed for frequencies between 0.1 – 1.0 Hz for full waveform seismic data from immediately after the 4th September 2010 earthquake until mid-January 2011. Using the moving window cross-spectral method, stacks of daily functions covering the study period (reference functions), were compared to consecutive 10 day stacks of cross-correlations to measure time delays between them. These were then inverted for seismic velocity changes with respect to the reference functions. Over the study period an increase in seismic velocity of 0.25% ± 0.02% was determined proximal to the Greendale fault. These results are similar to studies in other regions, and we attribute the changes to post-seismic relaxation through crack-healing of the Greendale Fault and throughout the region.