Darfield earthquake aftershocks: temporal evolution of the aftershock sequence, faulting and stress

Abstract

Data from portable deployments recording the aftershock sequence of the 4 September 2010 Mw 7.1 Darfield earthquake were used to determine high-resolution earthquake locations, seismic velocity models and stress in the region. Major findings were:

1) Using a three-dimensional velocity model to locate the earthquakes delineates eight individual faults active prior to the 22 February 2011 Mw 6.3 Christchurch earthquake, the largest aftershock of the Darfield earthquake. Two of these faults are in the Christchurch region, one of which corresponds to geodetically determined rupture planes of the Christchurch earthquake.

2) The direction of maximum compression measured by two different techniques agree with each other and show a rotation of the stress field to-wards nearly fault-parallel at stations close to the fault. We consider that the large drop in stress accompanying the Darfield earthquake caused a rotation of the stress near the fault after the earth-quake. The changes in stress angle with distance from the fault can be used to determine that the stress dropped by 40% from its original value before the rupture.

3) The movement of seismic sur-face waves across the Canterbury plains is amplified by a factor of three at a period of about 2.5 s. Therefore if large structures with natural periods of vibration of 2.5 s are built in the region, we would expect that the motion from a large earthquake would similarly amplify the motion in those structures.

---

Technical Abstract

The 3 September 2010 (UTC) Mw 7.1 Darfield (Canterbury) and 22 February 2011 Mw 6.2 Christchurch earthquakes and related aftershocks in Canterbury, New Zealand have revealed a major hazard in the Canterbury region in the form of the Greendale Fault and a number of associated faults. We used aftershocks of the Darfield earthquakes and ambient noise recorded on portable seismometers between 9 September 2010 and 11 January 2011 to study the earthquake occurrence patterns, seismic velocity structure, and stress in the region. We jointly inverted for three-dimensional isotropic P-wave and S-wave velocities and hypocentral locations, using data for 2840 aftershocks recorded at 36 temporary and permanent seismic stations within 70 km of the main shock epicenter. These relocations delineate eight individual faults active prior to the 22 February 2011 Mw 6.3 Christchurch earthquake, the largest aftershock of the Darfield earthquake. Two of these faults are in the Christchurch region, one of which corresponds to geodetically determined rupture planes of the Christchurch earthquake.

We use P-wave picks to estimate focal mechanisms and invert those mechanisms to estimate the azimuth of the axis of maximum horizontal compression (SHmax). We also use S waveforms to determine shear-wave splitting (SWS) parameters, whose fast orientations are also expected to be parallel to SHmax. Furthermore, we re-examined the paper that initially described the splitting measurement technique and published corrections to the calculation of the error bars. The tectonic stress field is remarkably uniform and has an average maximum horizontal compressive stress orientation of SHmax = 116 ± 18°, forming an angle with the average strike of the Greendale Fault of c. 25°. However, several SHmax estimates along the Greendale Fault are sub-parallel to the fault strike (93.6±13.1°, 100.8±11.5° and 100.8±12.6°), indicating that the fault may be frictionally weak, in an Andersonian sense. This variation occurs via an anti-clockwise rotation of SHmax southwards across the Greendale Fault. SWS fast directions (φ) generally match nearby SHmax, suggesting stress-aligned microcracks, but φ estimates at stations Cch3 and MQZ, which are near known and inferred faults, are subparallel to these faults and differ greatly from nearby stress orientations, indicating structure-dependent anisotropy. A lack of seismicity in the area prior to the Darfield earthquake precludes detailed analysis of time variations. However, there are two end member scenarios: if the pre-seismic stress orientation near the Greendale Fault was in the same direction as we have measured after the earthquake, then it was mis-oriented for rupture. Alternatively, if the stress rotated from the average regional orientation during the earthquake, then we can use the rotation to determine that an average of c. 40% of the pre-seismic differential stress on the Greendale Fault was released during the Darfield earthquake.

Measurement of basement seismic resonance frequencies can elucidate shallow velocity structure, an important factor in earthquake hazard estimation. Ambient noise cross correlation, which is well-suited to studying shallow earth structure, is commonly used to analyse fundamental-mode Rayleigh waves and, increasingly, Love waves. We showed via multicomponent ambient noise cross correlation that the basement resonance frequency in the Canterbury region of New Zealand can be straightforwardly determined based on the horizontal to vertical amplitude ratio (H/V ratio) of the first higher-mode Rayleigh waves. At periods of 1–3 s, the first higher-mode is evident on the radial-radial cross-correlation functions but almost absent in the vertical-vertical cross-correlation functions, implying longitudinal motion and a high H/V ratio. A one-dimensional regional velocity model incorporating a ~ 1.5 km-thick sedimentary layer fits both the observed H/V ratio and Ray-leigh wave group velocity. Similar analysis may enable resonance characteristics of other sedimentary basins to be determined.