Paleoseismology of the 2010 Mw 7.1 Darfield (Canterbury) earthquake source, Greendale Fault, New Zealand
Introduction
Despite significant scientific advances in the detection and mapping of active faults worldwide, many historical earthquakes have caused surface rupture on faults that were previously unknown due to a paucity or absence of evidence of prior surface rupture. Recent examples include the 2001 Bhuj (India) Mw 7.7 (McCalpin and Thakkar, 2003), 2010 El Cucapah Mw 7.2 (Oskin et al., 2012), and the 2010 Darfield (Canterbury) Mw 7.1 earthquakes (Quigley et al., 2010a, Quigley et al., 2010b). Characterizing the earthquake history of previously undetected faults and understanding why they evaded detection is important for assessing the completeness of active fault catalogues contributing to seismic hazard models. It is also important for understanding the maximum earthquake Mw potential for areas where surface rupturing faults have not been identified (e.g., Stirling et al., 2012).
Sedimentation or erosion may obscure or remove evidence for surface faulting in alluvial settings and increase the challenge of detecting active faults. Strike-slip faults with typically low-relief rupture traces are particularly susceptible to burial or erosion. Undersampling of active faults at the ground surface is exacerbated when fault slip rates are slow relative to the rates of surface processes (e.g., Gold et al., 2013). Fault detection is likely to be most difficult at the peripheries of active plate boundary zones, where rapid rates of surface processes due to proximal orogenic activity may overlap with areas of lower strain rates and longer earthquake recurrence intervals. Furthermore, when rupture occurs through thick packages of unconsolidated sediments, the total displacement may be expressed as a combination of discrete surface faulting and broad wavelength folding (Quigley et al., 2012, Van Dissen et al., 2011), with the latter typically difficult to recognize in the geologic record (Bray and Kelson, 2006, Fielding et al., 2009, Oskin et al., 2012, Rockwell and Klinger, 2013, Rockwell et al., 2002, Wesnousky, 2008). For this reason, the use of displaced geomorphic features to estimate the slip and Mw of paleoearthquakes relies upon the careful documentation of single-event coseismic slip and slip variability from historic earthquakes for which slip and Mw were recorded (e.g., Wells and Coppersmith, 1994, Wesnousky, 2008). Discrete surface ruptures typically account for 50–60% of the slip of their subsurface equivalent (e.g., Dolan and Haravitch, 2014).
The 2010 Mw 7.1 Darfield (Canterbury) earthquake triggered the 2010–2011 Canterbury earthquake sequence, which includes three earthquakes of Mw 6 or greater (Bannister and Gledhill, 2012). The 22 February 2011 Mw 6.2 Christchurch earthquake caused 185 fatalities and the greatest damage (e.g., Bradley et al., 2014, Kaiser et al., 2012) (Fig. 1). Of the faults that accrued slip during the Canterbury earthquake sequence only the Greendale Fault generated ground-surface rupture (Fig. 1, Fig. 2A) (Beavan et al., 2012, Elliott et al., 2012, Quigley et al., 2010a, Quigley et al., 2010b). The Greendale Fault surface rupture morphology and associated coseismic displacements have been extensively studied using combined field, lidar, InSAR, and geodetic techniques (Barrell et al., 2011, Duffy et al., 2013, Elliott et al., 2012, Litchfield et al., 2014a, Quigley et al., 2012, Van Dissen et al., 2011, Van Dissen et al., 2013, Villamor et al., 2011, Villamor et al., 2012). Abandoned river meanders and terrace patterns have been tentatively interpreted to suggest fault-related pre-2010 Holocene uplift at the western end of the Greendale Fault (Campbell et al., 2012). However, neither interpretation of ortho-photographs predating the Darfield earthquake nor analysis of post-Darfield imagery provides unequivocal evidence that the Greendale Fault ruptured the ground surface prior to 2010 (Villamor et al., 2012). In the absence of a clear pre-2010 surface trace, sub-surface information is required to constrain the paleoearthquake history of the fault. The Greendale Fault paleoseismic history has not been studied prior to this investigation.
This paper summarizes the tectonic, geologic and geomorphic setting of the Greendale Fault together with the surface rupture morphology and displacements obtained from the fault trace following the Darfield earthquake. New ground penetrating radar (GPR) and trenching data from two sites on the central Greendale Fault constrain the subsurface fault geometry and displacements. The timing of the penultimate event in the trenches has been constrained by new optically stimulated luminescence (OSL) dating of faulted stratigraphic units. Our results illuminate some of the challenges of detecting and studying active faults in alluvial landscapes at the comparably low strain rate fringes of tectonic plate boundaries. We illustrate how robust paleoseismic information for long-recurrence interval faults with diffuse and complicated patterns of surface rupture can be obtained by combining subsurface displacement measurements with multi-method high resolution surface displacement measurements.
Section snippets
Tectonic, geologic and geomorphic settings
The Greendale Fault sits at the eastern periphery of the Pacific–Australian plate boundary deformation zone in New Zealand's South Island (Fig. 1). The Pacific and Australian plates converge obliquely in a west to southwest direction at ~ 35–44 mm/year (e.g., Beavan et al., 2002, DeMets et al., 2010, Wallace et al., 2007). In the central South Island, slip on the Alpine Fault accommodates ~ 75% of the plate convergence and produces uplift of the Southern Alps, with the remainder of the convergence
Geometry and slip of the Darfield earthquake rupture
InSAR imagery, GPS measurements and seismicity data indicate that the 4 September 2010 Mw7.1 Darfield earthquake was sourced from a complex rupture comprising multiple faults and fault segments. These structures included E-W striking right-lateral, NE-striking reverse, NNW-striking left-lateral, and NW-striking normal right-lateral faults (Beavan et al., 2010, Beavan et al., 2012, Elliott et al., 2012, Holden et al., 2011). Because the earthquake initiated on a steep reverse fault, the first
Pre-trenching site investigations
Prior to trenching, the surface and sub-surface structure of small sections of the central Greendale Fault were studied using terrestrial lidar and GPR to aid trench-site selection. Given the abundance of offset cultural features, the Highfield Road site (Figs. 2B, Fig. 3, Fig. 4) was the first targeted for paleoseismic investigation. Acquisition of terrestrial lidar data at this site was undertaken immediately following the earthquake (see Litchfield et al., 2014a for data capture and
Greendale fault penultimate event and recurrence intervals
Following the Darfield earthquake, questions were raised about how many destructive earthquake sources remain undetected close to New Zealand's main cities and how often these sources generate large magnitude earthquakes. The Canterbury earthquake sequence revealed a number of previously unrecognized active faults, but the slip rates and recurrence intervals on these faults remain largely unresolved. For the Greendale Fault the absence of a clear pre-2010 trace on the Canterbury Plains mapped
Discussion
The five dated OSL samples collected from the gravel-dominated alluvial sediments exposed in the Highfield and Clintons Road trenches range in age from 21.6 ± 1.5 ka to 33.3 ± 2.0 ka (Table 1, see Supplementary Information), and provide the first numerical age constraints for this portion of the Waimakariari fan and Burnham Formation. To test the repeatability of these ages, we dated two OSL samples from a nearby gravel quarry (Fig. 2). These samples were collected at 1 m and 4 m depth from sand lenses
Conclusions
- 1.
The previously unknown Greendale Fault ruptured the ground surface in the September 2010 moment magnitude (Mw) 7.1 Darfield Earthquake producing a fault zone up to 300 m wide that comprised both distributed (folding) and discrete (faulting) deformations dominated by right-lateral displacement.
- 2.
Discrete surface fracturing accommodates an average of ~ 30% of the total right-lateral displacement with the remainder taken up by broad wavelength folding about vertical hinges accompanied by a slip
Acknowledgments
This research was funded by the New Zealand Earthquake Commission (EQC) (grant number 6/4/1BIE12/624) and by a GNS Science Sarah Beanland Scholarship (2011) to SH. We thank the landowner Mr Fitzgerald for unlimited access to the Highfield Road paddock, and all the workers and managers of the Clintons Road dairy farm for site access.
References (66)
- et al.
Facies development and sequence architecture of a late Quaternary fluvial-marine transition, Canterbury Plains and shelf, New Zealand: implications for forced regressive deposits
Sediment. Geol.
(2003) - et al.
How well do surface slip measurements track slip at depth in large strike-slip earthquakes? The importance of fault structural maturity in controlling on-fault slip versus off-fault surface deformation
Earth Planet. Sci. Lett.
(2014) - et al.
Late Quaternary slip rates and slip partitioning on the Alpine Fault, New Zealand
J. Struct. Geol.
(2001) - et al.
Optically stimulated luminescence dating of glaciofluvial sediments on the Canterbury Plains, South Island, New Zealand
Quat. Geochronol.
(2012) - et al.
Three-dimensional variations in present-day tectonic stress along the Australia–Pacific plate boundary in New Zealand
Earth Planet. Sci. Lett.
(2012) - et al.
Towards a climate event stratigraphy for New Zealand over the past 30 000 years (NZ‐INTIMATE project)
J. Quat. Sci.
(2007) - et al.
Stress and crustal anisotropy in Marlborough, New Zealand: evidence for low fault strength and structure-controlled anisotropy
Geophys. J. Int.
(2005) - et al.
Evolution of the 2010–2012 Canterbury earthquake sequence
N. Z. J. Geol. Geophys.
(2012) - et al.
Strike-slip ground-surface rupture (Greendale Fault) associated with the 4 September 2010 Darfield earthquake, Canterbury, New Zealand
Q. J. Eng. Geol. Hydrogeol.
(2011) - et al.
Motion and rigidity of the Pacific Plate and implications for plate boundary deformation
J. Geophys. Res. B Solid Earth
(2002)