Faulting in Banks Peninsula: tectonic setting and structural controls for ...

1 downloads 0 Views 5MB Size Report
between extension and volcanism or intrusions has been widely envisaged (Moore ... most margin of the volcanic edifice of Banks Peninsula in the southeastern ...
2012

research-articleResearch Article169X10.1144/jgs2011-167U. Ring & S. HamptonBanks Peninsula Volcanism

Downloaded from http://jgs.lyellcollection.org/ at Stockholm University on February 11, 2015 Journal of the Geological Society, London, Vol. 169, 2012, pp. 773–785. doi: 10.1144/jgs2011-167.

Faulting in Banks Peninsula: tectonic setting and structural controls for late Miocene intraplate volcanism, New Zealand UWE RING 1,2* & SAM HAMPToN 1 1Department of Geological Sciences, University of Canterbury, Christchurch 8140, New Zealand 2Present address: Department of Geological Sciences, Stockholm University, 106 91 Stockholm, Sweden *Corresponding author (e-mail: [email protected]) Abstract: An analysis of faulting in the late Miocene volcanic rocks of Banks Peninsula, South Island, New Zealand, shows that the formation of the volcanic edifice was largely controlled by NE–SW-striking dextraloblique strike-slip faults. The data show a variable component of west–east- or NW–SE-oriented shortening and north–south or NE–SW extension. Synvolcanic faults reactivated Cretaceous normal faults and are interpreted to have formed a local pull-apart basin that controlled volcanism. Further east, the geometry of Akaroa Harbour is controlled by a north–south-striking oblique reverse fault. Limited fault-slip data collected from sub-recent loess deposits are not significantly different from the data collected in the volcanic rocks and appear to show that the kinematic field did not change significantly over the last c. 10 Ma. The overall kinematic field causing the recent series of earthquakes in the greater Christchurch region is also not fundamentally different from the one that controlled the eruption of the volcanic rocks. We conclude that the inherited Cretaceous faults controlled the development of the late Miocene volcanism on Banks Peninsula and largely provided a major anisotropy along which the recent faults ruptured.

Pre-existing basement structures are known to greatly influence the location and eruptive sequence of volcanic centres (Toprak 1998; Patanè et al. 2006; Lavallée et al. 2009). As lavas feeding volcanic vents and dykes ascend subvertically, strike-slip faults and other subvertical structures are most effective in aiding volcanism. Many strike-slip faults, such as the Great Sumatran fault zone, are indeed accompanied by volcanoes (Chakraborty & Khan 2009). Extensional structures forming local pull-apart basins are common along strike-slip faults (Sylvester 1988). In such areas, extensional features are often associated with single volcanoes, links between volcanoes and pull-apart structures are commonly observed (Aydin & Nur 1982; Van Wyk De Vries et al. 1998), and a causal relationship between extension and volcanism or intrusions has been widely envisaged (Moore 1979; Hutton & Reavy 1992). Pre-existing structures are also known to play an important role in guiding subsequent large-scale deformation (Ring 1994; Schumacher 2002; Di Luccio et al. 2010). The volcanic edifice of Banks Peninsula just south of Christchurch, New Zealand (Fig. 1), serves as a natural laboratory for studying the effect that pre-existing structures have on the eruption of volcanic centres and perhaps also on guiding recent earthquakes in this region. Christchurch and its surroundings was hit by four major earthquakes recently: the Mw 7.0 Darfield earthquake east of Christchurch occurred on 4 September 2010 and was followed by an Mw 6.1 event on 22 February, an Mw 6.0 earthquake on 13 June and an Mw 5.8 event on 23 December 2011 at the northernmost margin of the volcanic edifice of Banks Peninsula in the southeastern suburbs of Christchurch (Fig. 1) (GeoNet website). The earthquakes originated on different faults and it appears that pre-existing structures in the pre-mid-Cretaceous (Torlesse) basement played an important role in focusing earthquake activity (Ghisetti & Sibson 2012; Sibson et al. 2012). Stramondo et al. (2011) argued, using Coulomb stress triggering theory (King et al. 1994), that the mainshock on 4 September 2010 triggered subsequent earthquakes in the region, which tend not to occur within the Banks Peninsula volcanic complex (Fig. 1).

The aim of this paper is to explore the relationships between the Canterbury basement faults, the emplacement of the Banks Peninsula volcanic rocks and faults activated during the recent earthquakes. We argue that the basement faults controlled the late Miocene Banks Peninsula volcanic complex and were also important for the recent Christchurch earthquakes. Especially since the February 2011 earthquake, but also before, the aftershocks occurred along faults in and along the margins of the Banks Peninsula volcanic complex.

Overview Tectonic setting Up to the mid-Cretaceous New Zealand was part of a subduction system at the eastern margin of Gondwana (Bradshaw 1989). Most of the eastern part of the South Island is made up of accreted rocks of this subduction system, the Torlesse greywacke (Mortimer 2004), which represent the basement. Subduction-related shortening caused east–west-striking thrusts and mélange belts (in presentday coordinates; it should be noted that the eastern part of the South Island rotated almost 90° counterclockwise in the Tertiary (Furlong & Kamp 2009; Fig. 1 inset)). The northern margin of the east–westoriented Chatham Rise is inferred to be aligned with this Cretaceous subduction zone (Mortimer et al. 2006). After subduction along the New Zealand sector of the east Gondwana subduction system ceased at 110–100 Ma, widespread extensional deformation affected the South Island. Normal faults developed at this time have variable strike, with the dominant directions being east–west and NE–SW (Fig. 1). The first phase of extensional tectonism was roughly north–south oriented, commenced at c. 110 Ma (Deckert et al. 2002; Gray & Foster 2004), and is recorded by numerous roughly east–west-striking faults (Fig. 1, Fig. 2 inset), which parallel the former subduction-related thrusts (Barnes 1994). Arguably, north–south extension was followed by NW–SE extension in the late Cretaceous at c. 85 Ma (Eagles et al. 2004; Mogg et al. 2008; Kula et al. 2009). Rifting ceased in the latest Cretaceous 773

Downloaded from http://jgs.lyellcollection.org/ at Stockholm University on February 11, 2015 U. RING & S. HAMPToN

ƒ



774

N Lyttleton 43.60 S 172.72 E

€

‚



 

    … 40-35 25

2-0

20

5 10

15







30

 ­

35-30 10 mm/yr

   



20 mm/yr 30 mm/yr



40 mm/yr

„

  



 









„

‡

 ­  





 



€



 ‚



 

„



 

 

ˆ

†

­ „  



 



€  



„‰ 

Fig. 1. Plate-tectonic setting of the South Island of New Zealand showing the transition from the Hikurangi subduction zone in the north of the South Island to strike-slip faulting along the Alpine Fault. Also shown is the slip vector of the Pacific plate and the averaged shortening direction across the Southern Alps (from Ghisetti & Sibson 2012). Banks Peninsula and the recent Canterbury earthquake sequence in relation to the main plate boundary fault system and mid- to late Cretaceous extensional faults in the South Island of New Zealand are also shown. It should be noted that the strike of the faults on Banks Peninsula is subparallel to the strike of mid- to late Cretaceous high-angle normal faults in the Canterbury basin to the east and south of the peninsula, and that the Greendale Fault is also parallel to the strike of mid- to late Cretaceous normal fault offshore just east of Christchurch. Inset: motion of the Pacific plate for a point at Lyttelton in an Australia plate fixed reference frame. The direction of motion is given by the vector from the centre to the point (i.e. the azimuth of the point on the plot) and the velocity is the distance from the centre, so, for example, at 20 Ma the relative plate motion between Australia and Pacific at Lyttelton would be in a direction of 242° with a relative velocity of 23 mm a−1. The graph shows that the azimuth changed from about 240° at 15 Ma to about 246° at 8 Ma and relative plate motion increased by about 10 mm a−1.

and is marked by an angular unconformity that separates synrift and post-rift sequences (Mogg et al. 2008). In the late Miocene the Southern Alps of New Zealand started to form in the collision zone between the two recent subduction systems (Fig. 1). Initially the orogen grew mainly to the west (Ghisetti & Sibson 2006). At about 5 Ma a distinct and sustained rainshadow effect set in. As a result of high erosion rates and associated large-scale removal of material from the western side of the orogen, deformation in the west became largely pinned to the Alpine Fault and the orogen started to grow significantly to the east (Chamberlain & Poage 2000; Ring & Bernet 2010). The propagation of deformation to the east is now in the process of affecting the Canterbury Plains and the recent earthquakes are arguably a manifestation of this process. The motions of the Pacific plate for a point at Lyttelton in an Australia plate fixed reference frame are shown in Figure 1 (Furlong & Kamp 2009). The current tectonic regime in New Zealand was established by about 28 Ma and then changed slightly at about 10 Ma (Furlong & Kamp 2009) (Fig. 1). The graph shows

that the azimuth changed from about 240° at 15 Ma to about 246° at 8 Ma and relative plate motions increased by about 10 mm a−1. It is important to note here that at this time Banks Peninsula was some 110 km east of the plate boundary. None the less, there would have been a change in kinematics at this time, which largely coincided with the onset of mountain building in the Southern Alps.

Magmatism Since New Zealand started to drift away from Gondwana in the late Cretaceous (84–82 Ma; Gaina et al. 1998), widespread products of intraplate volcanism formed nearly continuously throughout the late Cretaceous and Cenozoic. There are various expressions of this volcanic activity throughout the South Island (Hoernle et al. 2006; Timm et al. 2009), with large composite volcanoes, such as the Banks Peninsula volcanoes, being the most impressive type of volcanism (Weaver & Smith 1989). Banks Peninsula is predominantly made up by the Lyttelton and Akaroa composite volcanoes (Fig. 2), with smaller volumes

Downloaded from http://jgs.lyellcollection.org/ at Stockholm University on February 11, 2015 BANKS PENINSULA VoLCANISM

775

Fig. 2. Simplified volcanic geology of Banks Peninsula after Weaver & Sewell (1986) and Sewell et al. (1992). The inferred faults, the various volcanic vents (Hampton 2010) and localities for fault-slip analysis are shown. Inset: tectonic setting of Banks Peninsula and Canterbury Plains, showing recent faults (fine continuous lines), late Miocene to recent(?) faults on Banks Peninsula (bold continuous lines on the peninsula) and late Cretaceous normal faults (bold dashed lines), the last of which caused regional gravity anomalies (Bennett et al. 2000). The gravity anomalies delineate the Mid Canterbury Horst, a first-order structure that focused late Miocene volcanism (Ghisetti et al. 2012).

erupted from intervening and late-stage volcanism (Mt Herbert Volcanic Group and Diamond Harbour Volcanic Group respectively; Sewell et al. 1992). Drilled lavas offshore of Banks Peninsula and beneath the surrounding Canterbury Plains suggest

that the volcanoes had original outcrop area of c. 35 km2 for Lyttelton and c. 50 km2 for Akaroa, and that the volcanic rocks were more than 2–3 km thick (Weaver & Smith 1989). In a revision of all age datasets, periods of volcanism on Banks Peninsula

776

Downloaded from http://jgs.lyellcollection.org/ at Stockholm University on February 11, 2015 U. RING & S. HAMPToN

are Lyttelton Volcanic Group 12.4–9.7 Ma, Mt Herbert Volcanic Group 9.7–8.0 Ma, Akaroa Volcanic Group 9.4–8.0 Ma and Diamond Harbour Volcanic Group 8.1–5.8 Ma (Stipp & McDougall 1968; Sewell 1985, 1988; Sewell et al. 1992; Forsyth et al. 2008; Timm et al. 2009). The onset of volcanism largely coincides with the onset of the Southern Alps orogeny but it is unclear whether there is a genetic relationship between the two events. Volcanism on Banks Peninsula has been primarily basaltic, with limited trachytic intrusive rocks. Hampton & Cole (2009) and Hampton (2010) identified seven main eruptive centres within the Lyttelton Volcano. Further eruptive centres on Banks Peninsula have been identified from mapped vents and plugs, and stratigraphic relationships (Fig. 2; Sewell et al. 1992). The Banks Peninsula volcanoes sit on Mesozoic greywacke of the Torlesse accretionary wedge and, to a much lesser degree, on intermediate to silicic, late Cretaceous volcanic rocks (Fig. 2, inset). This basement forms a local NE–SW-trending horst that is characterized by the strongest positive gravity anomaly in the region (Bennett et al. 2000). Another, less well-expressed gravity high occurs in the centre of the Akaroa Volcano (Fig. 2, inset; Bennett et al. 2000). To constrain the tectonic boundary conditions of the emplacement of the Banks Peninsula volcanic rocks, fault-slip data have been collected from the western part of Banks Peninsula, especially around Lyttelton Harbour, and from the eastern Banks Peninsula (Fig. 2).

Structure of Canterbury Plains and Canterbury Basin The mid- to late Cretaceous extensional deformation left a distinct structural grain in the greywacke basement (Figs 1 and 2). We will argue below that this inherited fault pattern was important for the extrusion of the Banks Peninsula volcanic rocks. one major structure is the Chatham Rise and the Bounty Trough in the offshore region to the east, both of which strike east–west (Fig. 1). The Bounty Trough is a late Cretaceous failed rift (Eagles et al. 2004). The rifting process created numerous east–west- to WNW–ESE-striking normal faults (Grobys et al. 2007). The Chatham Rise narrows westward and gravity data (Bennett et al. 2000) suggest that part of the Chatham Rise continues onland as an east–west-striking basement horst just south of Christchurch, the Mid Canterbury Horst (Ghisetti & Sibson 2012; Fig. 2, inset). This horst is internally segmented by east–west-striking normal faults and is bounded to the north and south by Cretaceous–Tertiary basins, which are covered by a few hundred metres of late Pleistocene gravels (Mogg et al. 2008). The narrowing of the general horst structure would have been accommodated by ENE– WSW- or NE–SW-striking faults. It is the east–west-striking Chatham Rise–Mid Canterbury horst structure that focused late Cretaceous to Recent magmatism in the region (Mortimer et al. 2006; Panter et al. 2006; Ghisetti & Sibson 2012). on Banks Peninsula the late Cretaceous Mt Somers volcanic rocks erupted and later the late Miocene Banks Peninsula volcanic rocks formed within the horst block. South of Banks Peninsula, the NE–SW-trending Canterbury Basin formed in the mid- to late Cretaceous. Its formation is largely concurrent with the much larger Great South Basin further south (Eagles et al. 2004; Kula et al. 2009). The formation of these basins left strong NE–SW-oriented anisotropies. This structural grain is locally modified by north–south- and WNW–ESE-striking graben just east of Banks Peninsula (Figs 1 and 2; Mogg et al. 2008). In the foothills of the Southern Alps west of Banks Peninsula, NE–SW-striking thrusts and anticlines formed (Fig. 2, inset) as a result of Pliocene to Recent WNW–ESE-oriented shortening (Fig. 1). In addition, ENE–WSW- to east–west-striking dextral strike-slip faults occur (Ghisetti & Sibson 2012).

Fault zones on Banks Peninsula In general, the fault zones in the volcanic rocks of Banks Peninsula are not well exposed and the fractured rocks are severely weathered. In Figure 2 we have distinguished from west to east the NE–SWstriking Gebbies Pass fault system (consisting of the Gebbies Pass Fault in the west and two minor faults that are subparallel to it), and the Mt Herbert, Port Levy, Little River and Le Bons faults. on the western side of Akaroa Harbour occurs the north–south-striking Akaroa Fault. At the southern end of Akaroa Harbour this north– south-striking fault either swings into a NNW–SSE orientation or is cut by a NNW–SSE-striking fault. There are a few north–southstriking minor faults between the main NE–SW-striking faults. We did not observe robust cross-cutting relationships between the fault sets in the field. In addition, there are a few east–west-striking faults mainly in the Lyttelton Harbour area. We also show an inferred east–west-striking fault in the south of the volcanic edifice. The fault zones are characterized by anastomosing zones of gouge, cataclasite, breccia and hematite-clay-coated fractured rock (Fig. 3). In the centres of the fault zones grain-size reduction commonly produced fine-grained layers of non-cohesive rock, which we refer to as gouge zones. Between the gouge zones fractured and partly brecciated volcanic rocks occur, which we refer to as cataclasite. The fault zones are interlinked sets of gouge zones with minor step-over faults and small intervening blocks of intact volcanic rocks. Subsidiary fault zones show a relatively simple progression from non-fractured country rock into severely fractured cataclasite and thin gouge zones (less than c. 2 m) in their centres. Heterogeneous brittle deformation produced arrays of blocks whose surfaces have a wide distribution of orientations. Single blocks are separated by thin, slickensided surfaces with fibres and striae, which make these fault zones suitable for brittle strain analysis. The slickensided surfaces occur in the vicinity of the main fault zones and increase in number towards them. This spatial relationship is considered to imply a genetic connection of the mesoscale fault zones with the mapped main fault zones. Therefore, fault-slip analysis of the mesoscale fault zones allows us to infer the kinematics of the main fault zones.

Fault-slip analysis To evaluate the kinematics of a fault, the orientation of primary and secondary fault planes, the trend and plunge of striations, and the sense of relative displacement on these planes have been mapped. In this study a simple graphical method is used to determine principal strain axes, using the program ‘Fault Kinematics’ written by Allmendinger (Marrett & Allmendinger 1990). This method graphically constructs the principal incremental shortening and extension axes for a given population of faults. Each pair of axes lies in the movement plane of the fault (a plane perpendicular to the fault plane that contains the unit vector parallel to the direction of accumulated slip and the normal vector to the fault plane). Furthermore, each pair of axes is at an angle of 45° to each of the vectors (Fig. 3, inset). To distinguish between the shortening and extension axes it is necessary to have information on the sense of slip. Bingham distribution statistics for axial data will be used to optimize clusters of kinematic axes of a fault array (Mardia 1972). The linked Bingham distribution is equivalent to an unweighted moment tensor summation (a moment tensor sum in which all faults are weighted equally). The direction and sense of shear on these surfaces have been deduced from the orientation of fibres, striae and fractures associated with the fault (Hancock 1985). Fibre and striae orientations on slickensides from the subsidiary faults are usually simple and consistent, and are readily interpretable with the geometry of the mapped faults at a regional scale.

Downloaded from http://jgs.lyellcollection.org/ at Stockholm University on February 11, 2015 BANKS PENINSULA VoLCANISM

a

777

  



Trachyte dike



   



E

b

50 cm

E

Fault zones

W

c

SW

W NE

Riedel shear

E Normal fault

Fault zone 20 cm Fault-slip data We present our fault-slip data for the volcanic rocks of Banks Peninsula in three regions defined according to the predominant volcanic group, age and eruptive footprint: western Banks Peninsula (Lyttelton Volcanic Group), central Banks Peninsula (Mt Herbert and Diamond Harbour Volcanic Groups), and eastern Banks Peninsula (Akaroa Volcanic Group). It should be noted that within this analysis there is some overlap of volcanic groups.

Western Banks Peninsula: Lyttelton Volcanic Group (12.4–9.7 Ma) The fault-slip data from various outcrops along the Gebbies Pass Fault are dominated by roughly east–west-striking small-scale normal faults resulting from north–south extension (Fig. 4). Some datasets show a strong east–west shortening component (Fig. 4e). overall, the fault-slip data from mesoscale faults yielded fairly consistent results, which are compatible with dextral oblique-slip (transtensional) faulting along the NE–SW-striking Gebbies Pass Fault.

20 cm

Fig. 3. Photographs of fault zones along the Gebbies Pass Fault. (a) Several meso-scale fault zones cutting through Lyttelton volcanic rocks near Bridle Pass, Summit Road of Port Hills. Crosscutting trachyte dyke at the left is hardly faulted. (b) Close-up of the fault zone shown in (a), showing pronounced cataclasis of basalt in fault zone. The development of near-vertical secondary Riedel shears associated with normal faulting should be noted. (c) Small-scale normal faulting in basalt, Summit Road of Port Hills. The inset in the upper right of (a) shows a graphical construction of the principal incremental shortening and extension axes for a given fault; the movement plane of the fault is perpendicular to the fault plane and contains the unit vector parallel to the direction of accumulated slip and the normal vector to the fault plane; the shortening and extension axes are at 45° to the fault plane.

The fault-slip data from the Lyttelton Harbour area (Fig. 5) between the Gebbies Pass and Mt Herbert faults in general show patterns that are compatible with dextral oblique-slip faulting along both faults. Datasets H2 and H4 show a well-defined NNE–SSWoriented extension direction resulting from a pronounced component of normal faulting (Fig. 5d). The other two datasets are more mixed and show a strong reverse faulting component resulting from NW–SE- to WNW–ESE-oriented shortening. Taken together, the data suggest a mixture between strike-slip faulting and almost pure normal faulting.

Central Banks Peninsula: Mt Herbert and Diamond Harbour Volcanic Groups (9.7–5.8 Ma) The data from the Mt Herbert Fault depict a slightly less pronounced normal faulting component compared with the data from the Gebbies Pass Fault (Fig. 6). The extension directions of the various datasets are relatively consistent and trend NE–SW, with NW–SE-trending shortening axes. overall, the data again suggest dextral strike-slip faulting along the Mt Herbert Fault. Compared

Downloaded from http://jgs.lyellcollection.org/ at Stockholm University on February 11, 2015 U. RING & S. HAMPToN

778



a

G1 n=8

b G2 n=16



X

c G3a n=14

Y Z

Z

Z

Y Y

X

X

e G4 n=8

d G3 n=9

f G5 n=6

Y

X

Y Z

Z X

Y

Z

X

h G7 n=10

g G6 n=8

i G8 n=10 Z

X

Z

Z

Y Y

X Y X

with the Gebbies Pass Fault, the Mt Herbert Fault appears to have a less pronounced oblique normal faulting component. The fault-slip data from the Port Levy Fault (Fig. 7) are of limited use as most outcrops did not expose many fault planes. However, the data are compatible with the data from the Gebbies Pass and Mt Herbert faults and support dextral strike-slip kinematics along the NE–SW-striking faults in western Banks Peninsula.

Eastern Banks Peninsula: Akaroa Volcanic Group (9.4–8.0 Ma) The fault-slip data from the Akaroa Volcanic Group are not fundamentally different from those of western and central Banks Peninsula but overall appear to be characterized by a more pronounced NW–SE shortening component. The two NE–SW-trending faults in Akaroa Volcanic Group have almost identical fault sets to the NE–SW-trending faults in the older Lyttelton Volcanic Group. The Little River Fault (Fig. 8) and the Le Bons Fault (Fig. 9) both have mixed datasets dominated by strike-slip and oblique-slip faults that resulted from NW–SE-directed shortening and NE–SW extension. The Akaroa Fault is a north–south-striking oblique reverse fault. The fault-slip data for small-scale faults indicate a pronounced WNW–ESE- to NW–SE-trending shortening direction (especially datasets AB1 and AB2 in Fig. 10) with minor NNE–SSW

Fig. 4. Fault-slip data from the Gebbies Pass Fault. The diagrams show great circles of fault planes and the projected trace of the associated slickenside lineation in a lower-hemisphere equal-area projection. The principal strain axes (X > Y > Z) are shown. The deduced extension directions (X) are indicated by white arrows, and the shortening directions (Z) by grey arrows. The outcrop number is indicated at the upper left and locations are shown in Figure 2. (a) outcrop G1 is dominated by east–west-striking normal faults; consequently, the calculated extension direction is north–south. (b) G2 contains more data but shows similar characteristics to G1; in addition, a few north–south-striking sinistral strike-slip faults occur. (c) outcrop G3a is again similar to G2. (d) G3 has a limited dataset that again shows north–south extension. (e) outcrop 4 shows a number of north–south-striking reverse faults and a few NW–SE-striking oblique-slip normal faults. The fault pattern resulted from east–west shortening; there is minor north–south extension in the Y direction of the strain ellipsoid; the maximum extension direction X is subvertical. (f) outcrop G5 shows a mixture of east–west-striking normal faults and north–south-striking reverse faults. (g) outcrop G6 (see also (a) and (b)) depicts a more complicated dataset with east–weststriking normal and oblique normal faults, as well as NNE–SSW-striking oblique reverse faults; one NNE–SSW-striking fault is a normal fault. (h) outcrop G7 shows a number of NNW– SSE-striking faults, which have either reverse slip kinematics (NE-plunging striations) or normal kinematics (ENE-plunging striations). In addition, strike-slip faults and a NE–SW-striking normal fault occurs. (i) G8 shows also a mix of reverse, normal and strike-slip faults that overall combine to a kinematic pattern of shortening and extension axes similar to that of G7.

extension. Further along strike of the Akaroa Fault, the Pigeon Bay Fault records a relatively strong NNW–SSE-directed shortening component and minor ENE–WSW extension (Fig. 11).

Data from loess deposits Large parts of Banks Peninsula are covered with several metres of loess deposits. The loess is not well dated but appears to be 40– 60 ka in age (Goh et al. 1977; Berger et al. 2001). The loess covers the mapped fault zones and is commonly unfaulted. However, at a few outcrops the loess covering the fault zones has been faulted, attesting to a recent reactivation of the faults. We collected fault-slip data from the sub-recent loess deposits (Fig. 12) at two outcrops along the northern part of the Akaroa Fault. The kinematic data derived from the loess datasets are similar to the data from the Pigeon Bay Fault and thus dominated by WNW–ESE- to NW–SE-trending shortening.

Interpretation of fault-slip data The NE–SW-striking faults dominate the structure of Banks Peninsula, especially the western part of it. It appears that these faults are largely dextral strike-slip faults that have a variable component of roughly north–south to NE–SW extension. This extension component is most pronounced for the Gebbies Pass Fault bordering

Downloaded from http://jgs.lyellcollection.org/ at Stockholm University on February 11, 2015 BANKS PENINSULA VoLCANISM

a H1 n=7

 b H2 n=11

 Z

X

779



a K1 n=6

b K2 n=6

 X

Y

Y Y

X Y

X

Z Z Z

c H3 n=5

c K3 n=14

d H4 n=5

Y

d K4 n=7 X Y

Y

X

Z X Z Y

Z

Z

X

Fig. 5. Fault-slip data from the Lyttelton Harbour region. (a) outcrop H1 is dominated by ENE–WSW-striking reverse faults; in addition two north–south-striking normal faults occur. The NW–SE-trending shortening axis and the near-vertical extension axis for this fault pattern is well defined. The intermediate Y-axis is oriented ENE–WSW. (b) H2 shows a straightforward pattern of normal and oblique normal faults that strike roughly east–west. Consequently, the NNE–SSW-trending extension direction is well constrained. (c) outcrop H3 has three reverse faults, one normal fault and one NNE–SSW-striking sinistral normal fault. The fault pattern results basically from WNW–ESE shortening and near-vertical extension. (d) H4 has a limited dataset that resembles H2, with a well-defined NNE–SSW-oriented extension direction.

Lyttelton Harbour at its NW side. Further SE, the geometry of Akaroa Harbour is controlled by the north–south-striking sinistral-reverse Akaroa Fault. overall, there is a general tendency from dextral transtensional faults in the west to faults with a more transpressional component in the Akaroa Harbour area in the east. The few collected fault-slip data from the sub-recent loess deposits do not show significantly different kinematics from the data from the volcanic rocks, especially not from those collected in the Akaroa Volcanic Group. We interpret this to indicate that the overall kinematic field did not change to any significant degree since c. 10 Ma. The fact that there is hardly any evidence for displacements of more than 50–100 m of the volcanic rocks suggests that the young reactivation was modest.

Dextral pull-apart model for Lyttelton Harbour Figure 13 summarizes the inferred tectonic model for western Banks Peninsula, Lyttelton Volcano. The fault-slip data are consistent with dextral strike-slip faulting along the NE–SW-striking faults in western Banks Peninsula. The Gebbies Pass Fault shows a relatively strong north–south- to NE–SW-trending extension component associated with dextral strike slip. Between the dextral Gebbies Pass and Mt Herbert faults are segments, near Diamond Harbour for instance, which are characterized by almost pure normal faulting owing to NNE–SSW-oriented extension (Fig. 13).

Fig. 6. Fault-slip data from the Mt Herbert Fault. (a) outcrop K1 is dominated by east–west-striking dextral oblique normal faults and two east–west-striking dextral oblique reverse faults. The fault pattern is compatible with dextral strike-slip faulting along the NE–SW-striking Mt Herbert Fault (Fig. 2) and a NNE–SSW-trending extension and a NW–SEtrending shortening axis. (b) K2 shows a rather scattered dataset with three NE–SW-striking reverse faults and two NE–SW-striking oblique normal faults; in addition, one NW–SE-striking sinistral strike-slip fault occurs. The fault pattern resulted from NW–SE-trending shortening; the X- and Y-axes have very similar positive eigenvalues indicating a flattening strain type with two extension directions. (c) outcrop K3 has a larger dataset than K2 but the data are similar indicating WNW–ESE-directed shortening and NNE–SSW extension. (d) K4 has a limited dataset with oblique reverse and oblique normal faults that resulted from ENE–WSW extension.

It is proposed that the entire fault-slip dataset from the Lyttelton Volcano area represents a dextral pull-apart basin. Figure 13 shows east–west-striking releasing bend faults in the Lyttelton Harbour area, which transfer the extensional strain between the major dextral oblique normal faults.

Tectonovolcanic interactions Spatial relationships of faults and vents The age of fault activity is not directly constrained. The extrusion of the volcanic rocks took place between 12.4 and 5.8 Ma (see above). The faults in the volcanic rocks appear to be largely synvolcanic as they occur in the volcanic rocks but have, in general, not significantly displaced them. on the other hand, the mapped Gebbies Pass and Mt Herbert faults do not show up prominently on aerial photographs and are, at least in part, masked by volcanic flows. These relationships suggest that faulting was largely concurrent with the volcanic activity and has probably controlled the location of the volcanic vents (Fig. 2). As previously indicated there is differing fault kinematics within the various sectors of Banks Peninsula, defined by its volcanic region. The vents of Lyttelton Volcano concentrate along the

Downloaded from http://jgs.lyellcollection.org/ at Stockholm University on February 11, 2015 U. RING & S. HAMPToN

780



a L1 n=4

b L2 n=6



c L3 n=5

X X

X Z

Z Z Y Y

Y

  

  

   

 





 

 



  

  



  



 

 







Gebbies Pass and Mt Herbert faults, and associated faults that delineate the envisaged Lyttelton Harbour pull-apart (Fig. 13). In overlaying vent locations onto this dextral pull-apart basin model a relationship between intersection of faults and sources of volcanism can be proposed, indicating the structural control of magmatism (compare Figs 2 and 13). The present topography of Lyttelton Harbour is erosionally formed (Hampton & Cole 2009; Hampton et al. 2012) but erosion has been influenced by the underlying volcanic and faulting relationships. The vents of Mt Herbert Volcanic Group primarily occur along the Mt Herbert Fault, whereas the Akaroa Volcanic Group eruptive vents are focused along the Akaroa and Little River faults. Diamond Harbour Volcanic Group formed as discrete vents almost over the entire peninsula and these vents appear to align along north–southstriking faults (Fig. 2). These faults yielded oblique reverse fault kinematics in Akaroa Harbour. We argue that the younger age relationship of the Diamond Harbour vents reflects the subtle changes in the regional kinematic field.

Regional tectonics and volcanism We argue that the entire Banks Peninsula volcanism was controlled by a regional horst structure crosscut by faults, with intersections becoming the focus of volcanic activity and eruptive vents.



Fig. 7. Fault-slip data from the Port Levy Fault. (a) L1 shows two NE–SW-striking reverse faults and two oblique normal faults. The strain field is flattening with extension in X and Y directions; the well-defined shortening axis is NW–SW-trending. (b) L2 shows a heterogeneous dataset with two NNW–SSEstriking normal faults, a NW–SE-striking dextral strike-slip fault, and two WNW–ESEstriking oblique reverse faults. The pattern resulted from east–west shortening and north–south extension. (c) outcrop L3 has four normal and oblique normal faults and one single reverse fault. The pattern is resulting from NE–SW extension. (d) PB3 is again dominated by normal and dextral-oblique normal faults resulting from dominantly NNE–SSW extension. (e) PL1 shows mainly north–south- to NE–SW-striking reverse faults plus a WNW–ESE-striking normal fault. This fault set is due to WNW–ESE-directed shortening. (f) outcrop PL2 has three NNW– SSE-striking normal faults and a NNE–SSWstriking sinistral-oblique normal faults. This fault pattern indicates ENE–WSW-directed extension and subvertical shortening. Fig. 8. Fault-slip data from the Little River Fault. (a) outcrop LR1 is characterized by roughly east–west-striking oblique normal faults; in addition, one NNE–SSW-striking reverse fault occurs. The NNE–SSW-trending shortening axis is well defined. (b) AB6 shows NNW–SSE-striking sinistral faults and SE–NW-striking dextral faults resulting from NW–SE shortening and NE–SW extension. (c) PB2 has a set of north–southstriking oblique normal faults resulting from ENE–WSW extension and near-vertical shortening.

A similar horst setting controls eruptive centre sites on Mt Etna in southern Italy (Patanè et al. 2006). on Banks Peninsula an eastward trending migration of activity is recognized (Shelley 1987; Sewell 1988). Subtle differences between the mapped faults and vent distributions supply some constraints on the timing of faulting and migration of volcanism. In examining age relationships, periods of volcanism are contemporaneous or overlap (i.e. Mt Herbert Volcanic Group and Akaroa Volcanic Group), suggesting related but separate controls on magmatic source and eruptive location. Here we propose that the control on the location of eruptions on Banks Peninsula is through regional fault kinematics, which changed through time. The magmatic plumbing system is proposed to follow pathways and fault intersections within the east– west-striking Mid Canterbury Horst (Fig. 2). Tibaldi (2008) showed that dykes and fractures feeding magma to the surface align perpendicular to the regional extension direction, an aspect highlighted in the NE–SW elongation of Banks Peninsula (volcanic surface expression indicating the extent or morphology of the feeding system beneath) and the regional tectonic system. Stages of Banks Peninsula volcanism with specific faulting kinematics can be highlighted within the development of the plate boundary in the South Island. Lyttelton volcanism occurred during a stage of transtensional fault kinematics (c. 12–10 Ma). The onset of the Mt Herbert and Akaroa volcanism (9.7 and 9.4 Ma) coincides

Downloaded from http://jgs.lyellcollection.org/ at Stockholm University on February 11, 2015 BANKS PENINSULA VoLCANISM

 

 

 



Fig. 9. Fault slip data from the Le Bons Bay Fault. outcrop oK1 has NW–SE-trending sinistral strike-slip faults resulting from NW–SE shortening and NE–SW extension.

with a subtle change in Pacific plate motion (Furlong & Kamp 2009) (Fig. 1, inset), and the termination of the Mt Herbert and Akaroa volcanism at 8.0 Ma coincides with a more significant change in the rate of plate convergence and plate motion (Furlong & Kamp 2009). The Diamond Harbour Volcanic Group did not follow the previous trend of large-scale eruptive series from localized sources, with eruptive vents occurring all over the peninsula. This stage of volcanism occurred during a time when the developing plate boundary underwent a change towards greater transpressional kinematics. This is hypothesized to have two effects: (1) the propagation of magma along reactivated reverse faults, intersecting at low points within the structure of Banks Peninsula, resulted in discrete eruptive locations, a trend that follows that proposed and modelled by Galland et al (2007) and Tibaldi (2008); (2) as the plate boundary further developed, transpression limited propagation of magma to the surface, and resulted in the complete shut-off of Banks Peninsula’s volcanism.

Earthquake data Figure 14 shows the mapped and assumed fault planes that caused the earthquake sequence in Canterbury (Beavan et al. 2010; Quigley et al. 2010; Beavan 2011; Sibson et al. 2012; and GeoNet website). The deduced shortening axis for the 4 September 2010 Darfield earthquake was N c. 135°E and the corresponding extension direction would be N c. 45°E (USGS website). Near the hamlet of Greendale, the dextral Greendale Fault is terminated in the west by a NW–SE-striking segment (Fig. 14). Given N c. 45°E-trending extension the orientation of this NW–SE-striking fault would suggest almost pure normal slip on this segment. However, global positioning system (GPS) data indicate that displacement on the NW–SEstriking segment is dextral strike slip as well (B. Duffy, pers. comm.).

781

The 22 February 2011 earthquake occurred on the blind Port Hills Fault in the south of Christchurch (Fig. 14). The deduced shortening axis was N c. 125°E (USGS website). The slip was a mixture of reverse faulting and right-lateral strike slip. The deeper part of the fault and the westernmost 5–6 km of the fault show predominantly strike-slip faulting, whereas the eastern and shallower parts of the fault show a greater proportion of reverse faulting (Beavan et al., 2011). These relationships suggest a curving fault trace. Intense aftershock activity was concentrated in the hanging wall of the Port Hills Fault, extending especially out to sea to the north and east of Banks Peninsula (Sibson et al. 2012). The 13 June 2011 earthquake was a sinistral transpressive event at the eastern end of the inferred rupture of the February earthquake. The deduced shortening axis was N c. 115°E (USGS website). We refer to this fault here as the Godley Head–Port Levy Fault. The most recent major earthquake occurred on 23 December 2011 and was almost a pure thrust event on an offshore ENE– WSW-striking north-dipping fault. The shortening axis was N c. 122° (USGS website). Between the Greendale Fault and the faults in Christchurch further east near Rolleston is a diffuse corridor with no identified or assumed fault(s) (Fig. 14). Geometrically this corridor would classify as a releasing step-over and most of the focal-plane solutions are compatible with extension (GeoNet website). Sibson et al. (2012) speculated that this corridor is controlled by a subvertical 145 ± 5° striking sinistral strike-slip fault, the Norwood–Doyleston– Ellesmere lineament, which follows the approximate trend of a distinct aftershock lineament south of the Greendale rupture running SE towards the mouth of Lake Ellesmere. The c. 150° striking Norwood–Doyleston–Ellesmere lineament and Godley Head–Port Levy Fault both have sinistral kinematics and appear to act as accommodation structures between the Greendale and Port Hills faults, both of which have different strikes and dips and have no common origin. The NW–SE-striking segment near the hamlet of Greendale strikes similarly but has totally different kinematics. The two sinistrally displacing faults fit perfectly with the current kinematic field (Sibson et al. 2012), but the dextral fault near Greendale does not. It thus appears that dextral strike-slip faulting along the NW–SE-striking segment is inhibiting propagation of faulting along the dextral Greendale Fault to the west. This view is consistent with the fact that the majority of the aftershocks, including many of the larger events, are concentrated at the eastern end of the Greendale Fault (Gledhill et al. 2011) and that subsequent earthquake activity progressed eastwards. The seismically active faults are avoiding the main part of the Banks Peninsula volcanic edifice (Fig. 14). The deduced kinematic axes for the recent earthquakes are similar to the kinematic axes we obtained from our fault-slip analysis in Banks Peninsula (Fig. 14). This indicates that the kinematic field has not changed in any significant way since c. 10 Ma.

Importance of inherited structures In the following we argue that the inherited mainly Cretaceous structural grain played a dominant role in guiding late Miocene volcanism on Banks Peninsula and in turn also controlled the recent earthquake activity in the greater Christchurch area.

Pre-existing faults and their possible control on volcanism and synvolcanism faulting The major faults on Banks Peninsula strike NE–SW. These faults are parallel to the Cretaceous normal faults in the Canterbury Basin

Downloaded from http://jgs.lyellcollection.org/ at Stockholm University on February 11, 2015 U. RING & S. HAMPToN

782

  





  

  





Fig. 10. Fault-slip data from the Akaroa Harbour Fault. (a) AB1 depicts a straightforward dataset with NE–SW-striking reverse and oblique reverse faults resulting from WNW–ESE shortening. (b) AB2 is rather similar with north–south- to NE–SWstriking reverse and oblique reverse faults resulting from WNW–ESE shortening. (c) AB3 shows a more heterogeneous dataset with NE–SW-striking reverse faults, a single dextral strike-slip fault and east–weststriking normal faults. overall, the pattern again resulted from WNW–ESE shortening. (d) North–south-striking reverse and NW– SE-striking normal faults characterize the AB4 outcrop. A NW–SE-oriented shortening direction is apparent. (e) AB5 is dominated by strike-slip faults, with NNW–SSE-striking ones being sinistral and NW–SE-striking ones being dextral. one WSW–ENE-striking fault is also dextral. As for AB3 and AB4, the shortening direction is roughly NW–SE with a NE–SW-trending extension component.

   



 

  

  



  





   





  





  

   

 

 













Fig. 11. Fault-slip data from the Pigeon Bay Fault. (a) PB5 has NW–SEstriking sinistral and NW–SE-striking dextral strike-slip faults, and roughly NNE–SSW- and east–west-striking oblique reverse faults. The fault set resulted from NNW–SSE-directed shortening. (b) PB4 shows NW–SEstriking oblique reverse faults and NNW–SSE-striking oblique normal faults resulting from east–west shortening and north–south extension.

Fig. 12. Fault-slip data from loess deposits. (a) PB1 shows a simple set of east–west- to NE–SW-striking oblique reverse faults resulting from NW–SE shortening. (b) outcrop PB6 has mainly NNE–SSWstriking sinistral-oblique reverse faults resulting from NW–SE-directed shortening.

to the south. The north–south-striking oblique reverse faults parallel Cretaceous faults mapped just east of Banks Peninsula. We therefore suggest that the faults that controlled the volcanic vents are largely inherited basement faults that were reactivated in the late Miocene and controlled the emplacement of the Banks Peninsula volcanic rocks. A Cretaceous east–west-striking fault system is aligned along the Chatham Rise and controlled the opening of the Bounty Trough (Grobys et al. 2007). Barnes (1994) argued that at least some of the east–west-striking faults were extensionally reactivated in the late Miocene. This late Miocene extensional reactivation is related to the encroachment of the incoming Chatham Rise, as part of the Pacific plate, against the evolving transpressive plate boundary zone. The advance of the Chatham Rise would have been associated with an anticlockwise rotation and would have and possibly still is interacting with the North Island. overall, this scenario caused a complex kinematic regime and the ENE–WSW-striking faults formed as a

result. The latter faults might simply be rotated and reactivated east– west faults and/or newly formed late Miocene faults. The NE–SW-striking faults are oblique to the trend of the elongate structure of Banks Peninsula (Fig. 13). Barnes (1994) and Mogg et al. (2008) showed roughly east–west-striking inherited faults to the east of the peninsula all along the Chatham Rise. East– west-striking faults are also inferred on-land from gravity data (Bennett et al. 2000) (Fig. 2, inset). We argue that the intersection of these two inherited fault trends controls the location of the eruptive centres (Fig. 13), similar to what has been observed in the Cappadocian Volcanic Province.

Pre-existing structures and the recent earthquakes Ghisetti & Sibson (2012) and Sibson et al. (2012) have argued that the recent earthquakes were, at least in part, occurring along faults that reactivated older structures. For some of the new faults it is

Downloaded from http://jgs.lyellcollection.org/ at Stockholm University on February 11, 2015 BANKS PENINSULA VoLCANISM



783









 







   







  



  



 







     



    





 

Fig. 13. Digital elevation model of the northeastern part of Banks Peninsula showing the major faults and the maximum horizontal extension direction inferred from the fault slip data shown in Figures 4–7. In those cases where the maximum shortening direction is near-horizontal, the shortening direction has been projected into the horizontal and is also shown. It should be noted that both the maximum extension and the shortening directions have been projected into the horizontal (SHmax and SHmin) and therefore do not coincide with the strain axes shown in Figures 4–7. Also shown is our interpretation of the kinematics of the major faults along which the data have been collected. The envisaged releasing bend model for the formation of Lyttelton Harbour is indicated by the large white arrows. The east–west-oriented depocentre is interpreted as a result of a complex dextral releasing bend structure between the NNE–SSW-striking dextral strike-slip to oblique-slip faults.

hard or impossible to show that they were reactivating older structures as most of them are blind structures or cut through late Pleistocene gravels that had not previously been deformed. Here we provide some tentative arguments on possible reactivation based mainly on geometric grounds. The east–west-striking Greendale Fault formed right above a late Cretaceous normal fault in the subsurface that has been robustly inferred from the gravity data. This late Cretaceous fault is probably dipping at about 60° to the south, whereas the largely dextral Greendale Fault is subvertical with a south-side-up component of movement (Quigley et al. 2011). Similar cases where the strike of a new fault has been inherited from older structures but the dip has not have been reported by Ring (1994) from the East African Rift. According to Ring (1994), the new fault needs a nucleation point, which is provided by the pre-existing structure. Above and below the pre-existing structure are numerous fracture surfaces that are subparallel to the pre-existing fault and also subvertical extension fractures. After having nucleated along a pre-existing structure the

new fault propagates, at least in part, in its own way and would use the pre-existing fault-parallel and vertical fractures wherever convenient. We suggest that a similar mechanism operated in Canterbury when the Greendale Fault formed. The initial anisotropy may have been provided by Jurassic–Cretaceous thrusts and reverse faults in the Torlesse accretionary wedge that guided the development of the late Cretaceous normal faults (Barnes 1994), which then in turn helped the nucleation of the recent faults. South of Christchurch marks an area where the recent strike-slip deformation is changing into oblique thrusting along ENE–WSWstriking faults and the new Port Hills Fault developed in the late Pleistocene gravel along the northern interface of the volcanic rocks and gravel. We mapped similarly ENE–WSW-striking faults in and near Lyttelton Harbour (Fig. 13) and argued above that they either are rotated east–west faults or formed when the Chatham Rise encroached. This reasoning would again suggest that older faults may have guided the recent faults and we propose a similar

Downloaded from http://jgs.lyellcollection.org/ at Stockholm University on February 11, 2015 U. RING & S. HAMPToN

784

     

    

­ 



 

€



  

       



‚ƒ   ƒ

                  



nucleation mechanism to that envisaged for the Greendale Fault. We further speculate that the upward propagation of the Port Hills Fault was, at least in part, controlled by the contact between the volcanic rocks and the basement and late Cretaceous–Tertiary sediments into which the volcanic rocks intruded. The sinistral Godley Head–Port Levy Fault also seems to have reactivated older structures as it is dipping at 70° and not vertical (R. Sibson, pers. comm.). However, these NW–SE-striking accommodation structures do not run parallel to any obvious pre-existing structural grain, so further speculations on possible reactivation mechanisms are not warranted. The loading of the about 1000 km3 volcanic edifice of Banks Peninsula elevates the vertical stress and therefore increases the strength on any non-vertical fault. This might be the reason why the Port Hills Fault slipped at the northern periphery of the volcanic edifice, where the latter is covered by gravel and its thickness is significantly reduced. We also note that most of the reverse slip on the Port Hills Fault occurred near the surface and along the eastern segment of the fault (Beavan et al., 2011), where the vertical stress was less.

Conclusions We have shown that the eruption of the early volcanic rocks (Lyttelton Volcanic Group) on Banks Peninsula was controlled by NE–SW-striking dextral oblique-slip faults and we have speculated that the Lyttelton Harbour region represents the eroded remnants of a dextral pull-apart basin. Later volcanic vents tend to align with roughly north–south-striking oblique reverse faults, suggesting a more pronounced ESE–WNW-trending shortening component after about 10 Ma. Both the NE–SW- and the north–south-striking faults appear to be inherited structures that formed as a response to late Cretaceous extension of a former subduction-related accretionary wedge. Lyttelton volcanism was focused at releasing bends in the dextral pull-apart basin, with further volcanism on Banks Peninsula proposed to be concentrated at fault intersections in a regional horst structure. Late Cretaceous east–west-striking normal faults also represent an important anisotropy that controlled the recent series of

Fig. 14. Local setting of the Christchurch earthquake sequence. Surface rupture along the Greendale Fault is from Quigley et al. (2011) and other faults with inferred kinematic axes are from Sibson et al. (2012). Also shown are main faults in Banks Peninsula with kinematic axes inferred from the data presented in Figures 4–12. Because the method used for constructing principal shortening and extension axes for a given population of faults is similar to that used to determine infinitesimal strain directions for earthquake focal mechanisms (P- and T-axes, respectively), the kinematic axes can be compared. Fault pattern and deduced kinematic axes in Banks Peninsula do not differ significantly from those for the recent faults. The basement in shown in dark grey and the Tertiary sediments in light grey.

Canterbury earthquakes. Rupturing that caused the earthquakes did not occur parallel to pre-existing faults but the latter rather served as nucleation points for the recent faults. In addition to the east– west-striking faults, ENE–WSW-striking faults were important for earthquake generation. The ENE–WSW-striking faults also occur in the Lyttelton Harbour area and are considered to be either rotated late Cretaceous east–west faults or faults that formed in the late Miocene when the encroaching Chatham Rise interacted with the developing plate boundary zone on the South and North Island of New Zealand. We thank two anonymous reviewers and editor R. Phillips for helpful comments, K. Furlong for providing the sketch showing the relative plate motions, and F. Ghisetti and R. Sibson for discussions.

References Aydin, A. & nur, A. 1982. Evolution of pull-apart basins and their scale independence. Tectonics, 1, 91–105. BArnes, P.M. 1994. Continental extension of the Pacific Plate at the southern termination of the Hikurangi subduction zone: the North Mernoo Fault Zone, offshore New Zealand. Tectonics, 13, 753–754. BeAvAn, J. 2011. Preliminary report on the location and slip distribution of the fault that caused the February 2011 Christchurch earthquake, as derived from geodetic data. GNS Science, Lower Hutt, internal report. BeAvAn, J., sAmsonov, S., motAgh, M., WAllAce, L., ellis, S. & PAlmer, N. 2010. The Mw 7.1 Darfield (Canterbury) earthquake: geodetic observations and preliminary source model. Bulletin of the New Zealand Society for Earthquake Engineering, 43, 228–235. Bennett, d., BrAnd, r., FrAncis, d., lAngdAle, s., mills, c., morris, B. & tiAn, X. 2000. Preliminary results of exploration in the onshore Canterbury Basin. In: Proceeding of the NZ Petroleum Conference, Auckland, 19–22 March 2000. World Wide Web address: http://www.nzpam.govt.nz/cms/ pdf-library/petroleum-conferences-1/2000-conference-proceedings/ bennett-378-kb-pdf (accessed 28 April 2012). Berger, G.W., PillAns, B.J. & tonkin, P.J. 2001. Luminescence chronology of loess–paleosol sequences from Canterbury, South Island, New Zealand. New Zealand Journal of Geology and Geophysics, 44, 89–94. BrAdshAW, J.D. 1989. Cretaceous geotectonic patterns in the New Zealand region. Tectonics, 8, 803–820, doi:10.1029/TC008i004p00803. chAkrABorty, P.P. & khAn, P.K. 2009. Cenozoic geodynamic evolution of the Andaman–Sumatra subduction margin: current understanding. Island Arc, 18, 184–200.

Downloaded from http://jgs.lyellcollection.org/ at Stockholm University on February 11, 2015 BANKS PENINSULA VoLCANISM chAmBerlAin, C.P. & PoAge, M.A. 2000. Reconstructing the paleotopography of mountain belts from the isotopic composition of authigenic minerals. Geology, 28, 115–118. deckert, H., ring, U. & mortimer, N. 2002. Tectonic significance of Cretaceous bivergent extensional shear zones in theTorlesse accretionary wedge, central otago Schist, New Zealand. New Zealand Journal of Geology and Geophysics, 45, 537–547. di luccio, F., venturA, G., di giovAmBAttistA, R., Piscini, A. & cinti, F.R. 2010. Normal faults and thrusts reactivated by deep fluids: the 6 April 2009 Mw 6.3 L’Aquila earthquake, central Italy. Journal of Geophysical Research, 115, doi:10.1029/2009JB007190. eAgles, G., gohl, K. & lArter, R.D. 2004. High-resolution animated tectonic reconstruction of the South Pacific and West Antarctic Margin. Geochemistry, Geophysics, Geosystems, 5, doi:10.1029/2003GC000657. Forsyth, P.J., BArrell, D.J.A. & Jongens, R. 2008. Geology of the Christchurch Area. Institute of Geological and Nuclear Sciences 1:250,000 geological map 16, 1 sheet. GNS Science, Lower Hutt. Furlong, K.P. & kAmP, P.J.J. 2009. The lithospheric geodynamics of plate boundary transpression in New Zealand: initiating and emplacing subduction along the Hikurangi margin, and the tectonic evolution of the Alpine Fault system. Tectonophysics, 474, 449–462. gAinA, C., müller, D.R., royer, J.-Y., stock, J., hArdeBeck, J. & symonds, P. 1998. The tectonic history of the Tasman Sea: a puzzle with 13 pieces. Journal of Geophysical Research, 103, 12413–12433. gAllAnd, o., coBBold, P.R., de Bremond d’Ars, J. & hAllot, E. 2007. Rise and emplacement of magma during horizontal shortening of the brittle crust: insights from experimental modelling. Journal of Geophysical Research, 112, doi:10.1029/2006JB004604. ghisetti, F.c. & siBson, r.h. 2006. Accommodation of compressional inversion in north-western South Island (New Zealand): old faults versus new? Journal of Structural Geology, 28, 1994–2010. ghisetti, F. & siBson, R. 2012. Compressional reactivation of E–W inherited normal faults in the area of the 2010–2011 Canterbury earthquake sequence. New Zealand Journal of Geology and Geophysics, 55, doi:10.1080/00288 306.2012.674048. gledhill, K., ristAu, J., reyners, M., Fry, B. & holden, C. 2011. The Darfield (Canterbury, New Zealand) Mw 7.1 earthquake of September 4 2010: a preliminary seismological report. Seismological Research Letters, 82, 378–386. goh, K.M., molloy, B.P.J. & rAFter, T.A. 1977. Radiocarbon dating of Quaternary loess deposits, Banks Peninsula, Canterbury, New Zealand. New Zealand Journal of Geology and Geophysics, 7, 177–196. grAy, D.R. & Foster, D.A. 2004. 40Ar/39Ar thermochronologic constraints on deformation, metamorphism and cooling/exhumation of a Mesozoic accretionary wedge, otago Schist, New Zealand. Tectonophysics, 385, 181–210. groBys, J.W.G., gohl, K., dAvy, B., uenzelmAnn-neBen, G., deen, T. & BArker, D. 2007. Is the Bounty Trough off eastern New Zealand an aborted rift? Journal of Geophysical Research B: Solid Earth, 112, B03103, doi: 10.1029/2005JB004229. hAmPton, S. 2010. The volcanic history of Lyttelton Volcano. PhD thesis, University of Canterbury. hAmPton, S.J. & cole, J.W. 2009. Lyttelton Volcano, Banks Peninsula, New Zealand: primary volcanic landforms and eruptive centre identification. Geomorphology, 104, 284–298. hAmPton, S.J., cole, J. & Bell, D. 2012. Syn-eruptive alluvial and fluvial volcanogenic systems within an eroding Miocene volcanic complex, Lyttelton Volcano, Bank Peninsula, New Zealand. New Zealand Journal of Geology and Geophysics, 55, 53–66, doi:10.1080/00288306.2011.632424. hAncock, P.L. 1985. Brittle microtectonics; principles and practice. Journal of Structural Geology, 7, 437–457. hoernle, K., White, J.D.L., et al. 2006. Cenozoic intraplate volcanism on New Zealand: upwelling induced by lithospheric removal. Earth and Planetary Science Letters, 248, 335–352. hutton, D.H.W. & reAvy, R.J. 1992. Strike-slip tectonics and granites petrogenesis. Tectonics, 11, 960–967. king, G.C.P., stein, R.S. & lin, J. 1994. Static stress changes and the triggering of earthquakes. Bulletin of the Seismological Society of America, 84, 935–953. kulA, J., tulloch, A.J., sPell, T.L., Wells, M.L. & zAnetti, K.A. 2009. Thermal evolution of the Sisters shear zone, southern New Zealand: formation of the Great South Basin and onset of Pacific–Antarctic spreading. Tectonics, 28, doi:10.1029/2008TC002368. lAvAlee, Y., de silvA, S.L., sAlAs, G. & Byrnes, J.M. 2009. Structural control on volcanism at the Ubinas, Huaynaputina, and Ticsani Volcanic Group (UHTVG), southern Peru. Journal of Volcanology and Geothermal Research, 186, 253–264. mArdiA, K.V. 1972. Statistics of Directional Data. Academic Press, London. mArrett, r. & Allmendinger, r.W. 1990. Kinematic analysis of fault-slip data Journal of Structural Geology, 12, 973–986.

785

mogg, W.G., Aurisch, K., o’leAry, R. & PAss, G.P. 2008. offshore Canterbury Basin—beyond the shelf edge. In: Proceedings of PESA Easterm Australasian Basins Symposium III. Crown Minerals Division, Ministry of Economic Development, 369–378. moore, M.J. 1979. Tectonics of the Najd transcurrent fault system, Saudi Arabia. Journal of the Geological Society, London, 136, 441–454. mortimer, N. 2004. New Zealand’s geological foundations. Gondwana Research, 7, 261–272. mortimer, N., hoernle, K., hAuFF, F., PAlin, J.M., dunloP, W.J., Werner, R. & FAure, K. 2006. New constraints on the age and evolution of the Wishbone Ridge, southwest Pacific Cretaceous microplates, and Zealandia–West Antarctica breakup. Geology, 34, 185–188, doi:10.1130/ G22168.1.2002/2003. PAnter, K.S., BeluztAn, J., hArt, S.R., kyle, R., esser, S.R. & mcintosh, W.C. 2006. The origin of HIMU in the SW Pacific: evidence from intraplate volcanism in southern New Zealand and subantarctic islands. Journal of Petrology, 47, 1673–1704, doi:10.1093/petrology/egl024. PAtAnè, G., lA delFA, S. & tAnguy, J.-C. 2006. Volcanism and mantle– crust evolution: the Etna case. Earth and Planetary Science Letters, 241, 831–843. Quigley, M., vAn dissen, R., et al. 2011. Surface rupture of the Greendale Fault during the Mw 7.1 Darfield earthquake, New Zealand: initial findings. Bulletin of the New Zealand Society of Earthquake Engineering, 43, 242–246. ring, U. 1994. The influence of preexisting structure on the evolution of the Cenozoic Malawi rift (East African rift system). Tectonics, 13, 313–326. ring, U. & Bernet, M. 2010. Fission-track analysis unravels the denudation history of the Bonar Range in the footwall of the Alpine Fault, South Island, New Zealand. Geological Magazine, 147, 801–813, doi:10.1017/ S0016756810000208. schumAcher, M.E. 2002. Upper Rhine Graben: role of preexisting structures during rift evolution. Tectonics, 21, 61–76. seWell, r.J. 1985. The volcanic geology and geochemistry of central Banks Peninsula and relationships to Lyttelton and Akaroa volcanoes. Unpublished PhD thesis, University of Canterbury, 493pp. seWell, R.J. 1988. Late Miocene volcanic stratigraphy of central Banks Peninsula, Canterbury, New Zealand. New Zealand Journal of Geology and Geophysics, 31, 41–64. seWell, R.J., WeAver, S.D. & reAy, M.B. 1992. Geology of Banks Peninsula Scale 1:100,000. Geological Map 3. Institute of Geological & Nuclear Sciences, Lower Hutt. shelley, D. 1987. Lyttelton-1 and Lyttelton-2, the 2 centers of Lyttelton Volcano. New Zealand Journal of Geology and Geophysics, 30, 159–168. siBson, R.H., ghisetti, F.C. & crookBAin, R.A. 2012. Andersonian wrench faulting in a regional stress field during the 2010–2011 Canterbury, New Zealand, earthquake sequence. In: heAly, D., Butler, R.W.H., shiPton, Z.K. & siBson, R.H. (eds) Faulting, Fracturing and Igneous Intrusion in the Earth’s Crust. Geological Society, London, Special Publications, 367, 7–17. stiPP, J.J. & mcdougAll, I. 1968. Geochronology of the Banks Peninsula volcanoes, New Zealand. New Zealand Journal of Geology and Geophysics, 11, 1239–1260. strAmondo, S., kyriAkoPoulos, C., BignAmi, C.M., moro, M., PicchiAni, M., sAroli, M. & enzo Boschi, E. 2011. Did the September 2010 (Darfield) earthquake trigger the February 2011 (Christchurch) event? Scientific Reports, 1, doi:10.1038/srep00098. sylvester, A.G. 1988. Strike-slip faults. Geological Society of America Bulletin, 100, 1666–1703. tiBAldi, A. 2008. Contractional tectonics and magma paths in volcanoes. Journal of Volcanology and Geothermal Research, 176, 291–301, doi:10.1016/j. jvolgeores.2008.04.008. timm, C., hoernle, K., vAn den BoogArd, P., BindemAnn, I. & WeAver, S. 2009. Geochemical evolution of intraplate volcanism at Banks Peninsula, New Zealand: interaction between asthenospheric and lithospheric melts. Journal of Petrology, 50, 1–35. toPrAk, V. 1998. Vent distribution and its relation to regional tectonics, Cappadocian Volcanics, Turkey. Journal of Volcanology and Geothermal Research, 85, 55–67. vAn Wyk de vries, B. & merle, o. 1998. Extension induced by volcanic loading in regional strike-slip zones. Geology, 26, 983–986. WeAver, S.D. & seWell, R.J. 1986. Cenozoic volcanic geology of the Banks Peninsula. South Island igneous rocks. In: houghton, B.F. & WeAver, S.D. (eds) South Island Igneous Rocks: tour guides A3, C2 and C7. New Zealand Geological Survey Record, 13, 39–63. WeAver, S.D. & smith, I.E.M. 1989. New Zealand intraplate volcanism. In: Johnson, R.W., knutson, J. & tAylor, S.R. (eds) Intraplate Volcanism in Eastern Australia and New Zealand. Cambridge University Press, Cambridge, 157–188.

Received 3 January 2012; revised typescript accepted 25 June 2012. Scientific editing by Richard Phillips.