Interaction of strong lower and weak middle crust...

Interaction of strong lower and weak middle crust during

lithospheric extension in western New Zealand

Keith A. Klepeis,1 Daniel King,1 Mathew De Paoli,2 Geoffrey L. Clarke,2

and George Gehrels3

Received 8 June 2006; revised 23 April 2007; accepted 25 May 2007; published 30 August 2007.

[1] Exhumed sections of the middle and lower crust inwestern New Zealand reveal how deformation waspartitioned within a thermally and rheologicallyevolving crustal column during Cretaceous continentalextension. Structural data, P-T determinations, and U-Pbgeochronology from central Fiordland and the PaparoaRange in Westland show that extension initiated in thelower crust by �114 Ma as a period of arc-relatedmagmatism waned. Initially, deformation waslocalized into areas that were weakened by heat andmagma. However, these hot,weak zoneswere ephemeral.During the period 114–111 Ma, lower crustal fabricsrecord a rapid progression from magmatic flow tohigh-temperature deformation at the garnet-granulitefacies (T > 700�C, P = 12 kbar) to cooler deformation atthe upper amphibolite facies (T = 550–650�C, P = 7–9kbar).Lower crustal cooling andcompositional contrastsbetween mafic granulites and hydrous metasedimentarymaterial resulted in amiddle crust thatwasweak relative tothe lower crust. Between circa 111 and circa 90 Ma,focused subhorizontal flowandvertical thinning in aweakmiddle crust led to the collapse of the upper crust and theunroofing of midcrustal material. During this period,arrays of conjugate-style shear zones transferreddisplacements vertically and horizontally through thecrust, resulting in a structural style that resemblescrustal-scale boudinage. The New Zealand example ofcontinental extension shows that a weak middle crustand a relatively cool, highly viscous lower crust canresult in a localized style of extension, including theformation of metamorphic core complexes thatexhume the middle crust but not the lower crust.Citation: Klepeis, K. A., D. King, M. De Paoli, G. L. Clarke,

and G. Gehrels (2007), Interaction of strong lower and weak

middle crust during lithospheric extension in western New

Zealand, Tectonics, 26, TC4017, doi:10.1029/2006TC002003.

1. Introduction

[2] One of the most debated aspects of continentaltectonics centers on how a thermally and rheologicallyevolving lower crust influences the behavior of deformingcontinental lithosphere [Hopper and Buck, 1998; Westaway,1998; McKenzie et al., 2000; Clark and Royden, 2000;Abers et al., 2002]. The strength and rheological behavior ofthe middle and lower crust (as defined in section 3) areespecially important for understanding the origin of highlylocalized (<100 km wide) zones of extension called meta-morphic core complexes [Coney and Harms, 1984; Friedmanand Armstrong, 1988; Martınez et al., 2001; Wijns et al.,2005]. Many published models have suggested that corecomplexes form where efficient flow in a weak lower crustallows crustal thinning to focus into narrow zones, eventhough some strain is distributed over a broad area [Gans,1987; Buck, 1991;Wernicke, 1992; Hopper and Buck, 1996,1998; McKenzie and Jackson, 2002; Abers et al., 2002;Wijns et al., 2005]. This focusing effect may permit largedisplacements to occur on relatively few normal faults,leading to the collapse of the upper crust and the unroofingof the deep crust. However, despite evidence that flow in aweak lower crust promotes core complex-style extension,there is considerable uncertainty surrounding the mecha-nisms of weakening, how long they last, and how theychange as deformation proceeds. Some alternative modelsof extension [e.g., Nagel and Buck, 2004; Jolivet et al.,2004], including postconvergent extension in orogens[Jamieson et al., 2002], emphasize the important role offlow in a weak middle crust in controlling the behavior ofextending continental lithosphere.[3] In this paper, we present the results of a field-based

investigation that addresses differences in the behavior oflower and middle crust during the formation of Mesozoicmetamorphic core complexes in western New Zealand. TheFiordland region exposes over 3000 km2 of lower andmiddle crustal rocks (Figure 1) that record evidence forseveral mechanisms controlling lower and midcrustalweakening during extension. Among the interactions weinvestigated were relationships among extension, magma-tism, elevated temperatures (lower crustal temperatures ofT > 700�C), and granulite facies metamorphism. Thishistory enabled us to test whether flow in a hot, weak lowercrust controlled the formation of core complexes in thissetting. This hypothesis is especially relevant to Fiordlandgeology because there have been conflicting views on thethermal and tectonic evolution of this region and the ageof lower crustal fabrics. One view suggests that high

TECTONICS, VOL. 26, TC4017, doi:10.1029/2006TC002003, 2007ClickHere

for

FullArticle

1Department of Geology, University of Vermont, Burlington, Vermont,USA.

2School of Geosciences, University of Sydney, New South Wales,Australia.

3Department of Geosciences, University of Arizona, Tucson, Arizona,USA.

Copyright 2007 by the American Geophysical Union.0278-7407/07/2006TC002003$12.00

TC4017 1 of 27

http://dx.doi.org/10.1029/2006TC002003

Figure 1. Present configuration (top of inset) and Cretaceous reconstruction (bottom of inset and maindiagram) of western New Zealand after Tulloch and Challis [2000]. General geologic relationships inFiordland are from Wood [1972], Oliver and Coggon [1979], Bradshaw [1989a, Daczko et al. [2002a],Tulloch and Kimbrough [2003], and Klepeis et al. [2004]. Area near Resolution Island is from Turnbull etal. [2005], Allibone et al. [2005], and Milan et al. [2005]. Westland geology is after the compilation ofTulloch and Challis [2000]. Abbreviations are as follows: SP, Separation Point, MS, Milford Sound; GS,George Sound; CS, Caswell Sound; DS, Doubtful Sound; RI, Resolution Island; LTA, Lake Te Anau;LM, Lake Manapouri; CF, Cape Foulwind; WHC, White Horse Creek. Metamorphic pressures fromFiordland represent the peak of Early Cretaceous metamorphism at �120 Ma and are from Bradshaw[1985, 1989a, 1989b], Brown [1996], Klepeis et al. [1999], Clarke et al. [2000], Daczko et al. [2001a,2001b, 2002a, 2002b], Oliver [1977], Gibson and Ireland [1995], Davids [1999]; Hollis et al. [2004], andMilan et al. [2005]. Pressures from Westland show shallower early mid-Cretaceous (125–105 Ma) plutonemplacement depths [after Tulloch and Challis, 2000]. Profile A-B across Paparoa Range shown inFigure 3a. Boxes show areas of study.

TC4017 KLEPEIS ET AL.: CONTINENTAL EXTENSION IN NEW ZEALAND

2 of 27

TC4017

temperatures in excess of 700�C, induced by magmatism,weakened the lower crust when extension initiated duringthe mid-Cretaceous [Gibson and Ireland, 1995]. An alter-native view suggests that the lower crust had cooledsignificantly and was relatively strong and highly viscousprior to the onset of the extension [Daczko et al., 2002c;Klepeis et al., 2004; Marcotte et al., 2005]. This conflictpartially reflects models built on field observations made indifferent areas of Fiordland and uncertainties surroundingthe age and tectonic significance of deep crustal fabrics.[4] Here, we establish the role of the lower and middle

crust during the formation of Mesozoic core complexes inFiordland through the following steps: (1) determining theage and sequence of lower crustal deformation, metamor-phism, and magmatism; (2) establishing space-time relation-ships between mechanisms of lower crustal weakening andextensional deformation; and (3) determining where andwhen linkages between extensional structures in the lower,middle and upper crust were established. This latter part ofthe test is enabled by pre-Cenozoic markers, including theMedian Batholith (Figure 1 inset), that allow the restorationof the relative positions of crustal blocks representingdifferent Mesozoic paleodepths. A well-established Creta-ceous reconstruction places the lower crust of the Gond-wana magmatic arc next to rocks that represent the middleand upper crust of the same arc in Westland [Gibson, 1990;Oliver, 1990; Tulloch and Challis, 2000]. The originaldepth-stratified geometry of the Fiordland section has beenreconstructed using metamorphic pressure data and U-Pbisotopic age determinations [Klepeis et al., 2003, 2004].Similar methods have been used in orogenic belts elsewhere[Miller and Paterson, 2001; Karlstrom and Williams, 2006].

2. Tectonic and Geologic Framework

[5] Western New Zealand records a mid-Cretaceous tran-sition from a continental margin dominated by contraction,crustal thickening, and arc magmatism to one dominated bylithospheric extension and the formation of metamorphiccore complexes. By�120 Ma, convergence and arc magma-tism along the Gondwana margin had thickened the crust to>45 km [Bradshaw, 1989a;Clarke et al., 2000;Daczko et al.,2001a]. The magmatism culminated in the emplacement of amafic-intermediate batholith called the Western FiordlandOrthogneiss (WFO). Portions of the batholith emplaced atmidcrustal levels (P < 8 kbar, T > 700�C) were subsequentlyburied and experienced granulite facies metamorphism atP = 11–16 kbar and T > 750�C [Blattner, 1976; Clarke etal., 2000, 2005; Daczko et al., 2001a]. The batholith recordsages in the range 126–120 Ma in northern Fiordland[Mattinson et al., 1986; Hollis et al., 2003]. However,crystallization ages as young as 116–113 Ma have beenobtained from central Fiordland [Gibson and Ireland, 1998;Tulloch and Kimbrough, 2003; Hollis et al., 2004].[6] Daczko et al. [2002c] reported isobaric cooling of the

WFO and adjacent rocks to T = 650–700�C and P = 11–15 kbar by circa 110 Ma. Flowers et al. [2005] refined thisage estimate and concluded that lower crustal cooling fromT � 700�C through to 550–650�C occurred without sig-

nificant exhumation between �113.5 and �111 Ma.Marcotte et al. [2005] showed that cooling began as earlyas circa 119 Ma in northern Fiordland.[7] Extensional structures began to form in the upper and

middle crust of the orogen by circa 110 Ma [Tulloch andKimbrough, 1989; Spell et al., 2000]. The initiation of thesestructures appears to have been contemporary with theemplacement of the last arc plutons [Waight et al., 1998]and a reorganization of plate boundaries that ended subduc-tion along the margin [J. D. Bradshaw, 1989; Sutherland etal., 2000]. Geochemical data and geochronology fromupper crustal granitoids indicate that subduction-relatedmagmatism within the margin lasted until circa 105 Ma[Tulloch and Kimbrough, 2003]. During the period 110–90 Ma several extensional metamorphic core complexesjuxtaposed middle and upper crust in rocks now exposed inWestland. The best known of these is the Paparoa corecomplex [Tulloch and Kimbrough, 1989; Spell et al., 2000].[8] One of the largest extensional structures in lower

crustal rocks is the Doubtful Sound Shear Zone, whichwas originally mapped as a thrust by Oliver and Coggon[1979] and Oliver [1980] and later interpreted as a ductilenormal fault by Gibson et al. [1988] and Oliver [1990]. AtMount Irene (MI, Figure 2a), Scott [2004] identified amylonitic, amphibolite to greenschist facies shear zone thatseparates Paleozoic sequences (the Mount Irene Complex)from underlying orthogneiss of probably Cretaceous age(the Robin Gneiss). A syntectonic dike from the shear zoneyielded a U-Pb zircon age of �108 Ma [Scott and Cooper,2006]. Finally, bedrock mapping on Resolution Island byI. Turnbull and others at the Institute for Geological andNuclear Sciences in Dunedin (I. Turnbull, personal commu-nications to K. A. Klepeis, 2005 and 2006) has revealedanother major extensional shear zone. Kula et al. [2005]describe the Late Cretaceous Sisters Shear Zone on StewartIsland (Figure 1), which records extension of the upper andmiddle crust following emplacement of the Paparoa corecomplex.[9] By circa 84 Ma, the proto-New Zealand continent had

rifted away from Australia and Antarctica and seafloorspreading initiated in the Tasman Sea [Gaina et al.,1998]. Western New Zealand thus records a type of conti-nental extension that involved all crustal levels and even-tually led to lithospheric failure and the formation of anocean basin.

3. Definition of the Middle and Lower Crust

[10] The division of continental crust into upper, middleand lower crustal layers is a common approach in models oflithospheric-scale processes [e.g., Hopper and Buck, 1998;Westaway, 1998; McKenzie et al., 2000; Martınez et al.,2001]. These divisions are inherently approximate andtypically are made on the basis of variations in seismicvelocity, composition, and/or mechanical properties withdepth. Here, we summarize the characteristics commonlyassigned to these layers at continental margins and definethem in western New Zealand.


3 of 27

TC4017

Figure

2


4 of 27

TC4017

[11] The upper 10–15 km of the continental crust typi-cally displays seismic velocities in the range 5.9–6.3 km s�1

and a mean composition between that of granodiorite anddiorite [Christensen and Mooney, 1995; Mooney et al.,1998; Rudnick and Gao, 2003]. Laboratory experimentsand rheological models of quartz-dominated rocks suggestthat crustal yield strengths initially increase with depth andthen are sharply reduced after some 15 km [Kohlstedt et al.,1995] where rocks deform mostly by solid-state creep. Thevelocity range of ‘‘normal’’ middle crust generally is takento be 6.4–6.7 km s�1 [Mooney et al., 1998], but it is poorlydefined. Global averages suggest that it is about 11 km thickwith a depth range of 12–23 km [Rudnick and Fountain,1995; Gao et al., 1998]. Most commonly, the middle crust isdistinguished from the lower crust on the basis of bothvelocity variations and composition. The typical velocityrange of the lower crust (6.8–7.7 km s�1) suggests that thislayer is more mafic and dense than overlying middle crust[Mooney et al., 1998]. It also is commonly assumed to beweaker than the middle crust due to the elevated temper-atures found at these depths [e.g., Royden, 1996; Handy andBrun, 2004; Burov and Watts, 2006], although this rheo-logical model has been challenged [Maggi et al., 2000;Jackson, 2002; Klepeis et al., 2003, 2004; Afonso andRanalli, 2004]. In tectonically active rifts and rifted mar-gins, the lower crust is much thinner than the global averageof 17 km. In orogenic belts where substantial crustalthickening has occurred, it may be up to 25 km thick[Rudnick and Fountain, 1995]. Both dry and wet rock typesoccur in the lower crust. For dry crust, high-pressuregranulites ranging in composition from gabbro to granodi-orite and diorite and containing abundant plagioclase andpyroxene appear representative; for wet lower crust,amphibolites mixed with silicic material appear appropriate[Christensen and Fountain, 1975; Oliver, 1980; Clarke etal., 2000; Baldwin et al., 2003].[12] The Mesozoic architecture of Fiordland and the once

contiguous Westland provides a nearly complete section ofcontinental lithosphere with a depth range of 8–45 km(Figure 1). In Westland, granitoids of the �126–105 MaSeparation Point Suite (Figure 1) record Early Cretaceousemplacement depths of 8–27 km (P = 2–7 kbar) [Tullochand Challis, 2000]. In central Fiordland, Cretaceous lowercrust is represented by exposures of high-P (10–15 kbar)granulite facies orthogneiss [Oliver, 1977;Gibson and Ireland,1995; Davids, 1999; Hollis et al., 2004]. These high-Pgneisses mostly occur in the footwalls of two major exten-sional shear zones (‘‘g,’’ Figure 2c), the Doubtful SoundShear Zone and the Resolution Island Shear Zone (newname). Beneath the former they occur within the WesternFiordland Orthogneiss, beneath the latter they form part of a

mafic unit called the Breaksea Gneiss. Metagabbro andmetadiorite compositions are represented, including dry(granulite) and wet (amphibolite) varieties. Plagioclase,clinopyroxene, and garnet are common. Paragneiss is pres-ent but minor.[13] Metasediment and orthogneiss of mixed Paleozoic

and Mesozoic ages structurally overlie the WFO. Theserocks display upper amphibolite and greenschist faciesmineral assemblages and record pressures in the range5–9 kbar [Oliver, 1977; Gibson and Ireland, 1995, 1996;Ireland and Gibson, 1998; Scott, 2004]. These latter, mostlypre-Cretaceous, rocks are interpreted to represent the middlecrust, which is composed of paragneiss, quartz-biotiteschist, marble, granite, diorite and minor amphibolite.Exposures of the middle crust north of Doubtful Soundinclude relatively shallow (midcrustal) parts of the WFOand other units interpreted to be related to the WFO in ageand setting. These and other igneous units are hydrous,containing mostly biotite, clinozoisite, and hornblende.They lack clinopyroxene, garnet is rare but present, andgranulites generally are absent.

4. Paparoa Core Complex: Middle and Upper

Crust

[14] Ductilely deformed granitoids and paragneiss in thecore of the Paparoa Range are separated from overlyingPaleozoic and Mesozoic cover rocks by two low-anglenormal faults called the Pike and Ohika detachments(Figure 3a). These faults dip in opposite directions anddisplay opposite senses of shear. Quartz and epidote striaeon the faults show symmetrical, SW and NE directeddisplacements (Figure 3b).[15] Below the Pike fault, a 50 m thick zone of mylonite,

referred to here as the Paparoa Shear Zone, occurs (PSZ,Figures 3a and 3c). Previous work suggests that this shearzone formed at depths of 15–20 km [White, 1994] by�110Ma[Tulloch and Kimbrough, 1989; Spell et al., 2000]. Wedetermined the sequential evolution of ductile structureswithin and below this zone of mylonite.[16] The upper and lower boundaries of the Paparoa

Shear Zone are well exposed (Figure 3c). Below the shearzone, deformed granitoids of the Charleston MetamorphicGroup display several gneissic foliations, collectivelyreferred to here as S1. This composite fabric is folded anddips mainly to the south, SE, and east (Figure 3d). Downdipquartz-biotite mineral lineations (L1) plunge gently to theSE (Figure 3d). The L1-S1 fabric is cut discordantly bypenetrative mylonitic and ultramylonitic foliations (S2) thatdefine the Paparoa Shear Zone. These mylonites displayquartz ribbons and aligned biotite and muscovite crystals

Figure 2. (a) Generalized geologic map showing location of section shown in Figure 2c. (b) Geologic map of theDoubtful Sound region showing location of kinematic indicators and sites of detailed vorticity analyses. Mapping from thisstudy and data from Oliver and Coggon [1979]. (c) Cross section of central Fiordland showing location of high-strainzones. DSSZ, Doubtful Sound Shear Zone; RISZ, Resolution Island Shear Zone (new name); SRSZ, Straight River ShearZone; KPG, Kellard Point Gneiss; DCG, Deep Cover Gneiss. Mineral abbreviations are as for Figure 6. U-Pb dates onzircon (white boxes) are from the following data sources: 1, this study; 2, Hollis et al. [2004]; 3, King et al. [2005]; 4,Tulloch and Kimbrough [2003]; 5, Scott and Cooper [2006].


5 of 27

TC4017

that form a fine-grained matrix surrounding rotated, frac-tured feldspar porphyroclasts. Dynamically recrystallizedquartz (regime 3 subgrain rotation and strain-induced grainboundary migration) and brittle fractures in feldspar areconsistent with deformation temperatures in the range of400–550�C [Tullis et al., 2000]. Penetrative mineral line-ations (L2) defined by quartz rods and stretched quartz-micaaggregates plunge to the SWandNE on S2 planes (Figure 3e).These L2 lineations are easily distinguished from the olderL1 lineations on the basis of discordant geometries (compareFigures 3d and 3e). A few L1 lineations in Figure 3d plot inthe NE and SW quadrants and appear to have rotated intoparallelism with L2 during deformation. Several generationsof folds occur in the Paparoa Shear Zone. The S2 foliation isfolded by tight to isoclinal, recumbent folds (F2) that displaysheared out limbs. These are refolded about an axis (F3) thatplunges gently to the SW parallel to the L2 lineations(Figure 3f). The F3 folds are asymmetric, open to tight,and display both rounded and chevron-shaped hinges. TheF3 axial planes dip moderately to the NE. Brittle normalfaults, including the Pike fault, cut all ductile structures inthis shear zone.[17] High strain zones also occur, locally, below the

Ohika detachment fault (Figure 3g). However, these shearzones are subordinate in size (<10 m thick) and lack thepenetrative character of the Paparoa Shear Zone. Instead,they form a series of subparallel S2 shear bands that dip

gently to the NE, parallel to the Ohika fault in Cretaceousgranitoids of the Rahu Suite (Figure 1). Open folds of thisolder foliation occur between the S2 shear bands. Quartzand biotite mineral lineations (L2) plunge gently to the NEand SW (Figure 3h).[18] The predominance of mylonitic fabrics on the SW

side of the range suggests that deformation in the PaparoaShear Zone and, later, on the Pike fault was responsible formost of the unroofing of the midcrustal core. Spell et al.[2000] reached a similar conclusion on the basis of ayounging of Cretaceous cooling ages toward the SW. Wealso conclude that the directions of extension and similarage of D2 deformation in the Paparoa Shear Zone and in theCretaceous fabrics located below the Ohika fault indicatethat there was kinematic compatibility between ductiledeformation on the northern and southern sides of the range.Kinematic compatibility between the brittle upper crust andthe ductile middle crust also is indicated by the parallelismbetween L2 stretching lineations and striae orientations inthe Pike and Ohika faults.

5. Central Fiordland: Middle and Lower

Crust

5.1. Crustal Structure

[19] We determined the structure and progressive evolu-tion of lower and midcrustal fabrics within an �80 km

Figure 3. (a) Cross section of the Paparoa range (modified from Tulloch and Kimbrough [1989]).Location of profile is shown in Figure 1. Boxes show areas of detailed study illustrated in Figures 3c and3g. U-Pb date on zircon (white box) is from this study. (b) Orientation and slip directions on Pike andOhika faults. (c) Cross section of the Paparoa shear zone (PSZ) below the Pike detachment. Profilelocation shown in Figure 3a. Note top-down-to-the-SW sense of displacement parallel to F3 folds axes(out of plane of cross section). (d)–(f) Equal-area, lower hemisphere stereoplots showing poles tofoliations, lineations, and F3 fold axes from section shown in Figure 3c. (g) Cross section below theOhika detachment. Profile location shown in Figure 3a. Note top-down-to-the-NE sense of displacement.(h) Equal-area, lower hemisphere stereoplots show orientation data from section shown in Figure 3g.(i) Sketch of shear indicators observed beneath the Ohika detachment.


6 of 27

TC4017

composite section across central Fiordland (Figures 2a and 2c).The sections were constructed parallel to the general NE-SWdirection of tectonic transport in major lower crustal shearzones. For simplicity, we divide rock units into two maingroups: (1) the WFO and related early Mesozoic plutonsand (2) Paleozoic metasedimentary, granitic, and metavol-canic rocks that host the Mesozoic intrusions. Two majorgently dipping shear zones dominate the section: theDoubtful Sound Shear Zone and the Resolution IslandShear Zone (new name). These shear zones are defined bythin (10–100 m) bands of upper amphibolite faciesmylonite that form a branching network of high-strain zones(Figures 2b and 2c). The bands display two dominantorientations: one set dips to the SW, the other to the NE.Intermediate orientations also are present. The macroscopicstructure of the largest splays dip gently to the northeast. Forthis reason we refer to these main northeast dipping splaysas synthetic and the southwest dipping splays as antithetic tothe main shear zone. The symmetric style implies that theshear zones record a significant component of pure shear.[20] The main splays of the Doubtful Sound and Reso-

lution Island shear zones locally form the contacts betweenlower crustal rocks of the granulite facies WFO andoverlying midcrustal host rock. South of Doubtful Sound,branches of mylonite cut deep into coherent bodies ofgabbroic and dioritic WFO (Figures 2b and 2c). A similargeometry occurs on Resolution Island where the ResolutionIsland Shear Zone cuts deep inside mafic granulite faciesorthogneiss containing eclogite pods in the BreakseaGneiss. These branching zones, enclose oblate-shaped loz-enges of deformed plutonic and host rock that patchilypreserve older relics of garnet granulite (Figures 4a and 4b).The amount of displacement on these shear zones isunknown but appears to be at least 15 km, and may beconsiderably more, on the basis of contrasts in metamorphicpressures across the high-strain zones. The garnet-granulitemineral assemblages attest to an early high-T history in thelower crust.[21] Above the main splays of the Doubtful Sound and

Resolution Island shear zones are two antithetic, upperamphibolite facies shear zones that dip gently to the westand southwest (Figure 2c). Open synclines deform gneissiclayering (S1) in host rocks above the splays (Figure 2c) andtight, recumbent folds of S1 occur near them (Figures 4cand 4d). The upper part of the WFO is exposed betweenDoubtful Sound and Mount Irene. Most of this section ofthe WFO also lacks evidence of penetrative subsolidusdeformation, except along its margins. The two mainexceptions to this include the antithetic shear zone atBradshaw Sound and high-strain zones that occur belowand west of Mount Irene along the contact between theWFO and overlying Paleozoic host. These high-strain zonesdip to the SW and NE, mimicking the internal structure ofthe Doubtful Sound and Resolution Island shear zones.Some high-strain zones are folded about NE plunging foldaxes [Scott, 2004] and penetrate into the footwall of thebatholith indicating that they are Cretaceous in age. Thelack of penetrative deformation in large parts of the section

shown in Figure 2c indicates that extension affecting thelower crust was highly localized.

5.2. Evolution of Lower Crustal Fabrics

[22] Mineral fabrics within the footwalls of the DoubtfulSound and Resolution Island shear zones (i.e., the deepestpart of the lower crustal section) comprise three groups thatare distinguishable on the basis of crosscutting relation-ships, mineral assemblage, texture, and spatial distribution.In sequence, these include (1) a heterogeneous, but wide-spread magmatic fabric (SWFO) composed of hydrous, two-pyroxene igneous assemblages; (2) a penetrative progradegarnet granulite facies foliation (SGG); and (3) thin strandsof retrogressive upper amphibolite facies mylonite (SSZ) ofthe Doubtful Sound and Resolution Island shear zones.These three groups of fabrics correspond to a generalpattern of zoning within the lower crustal part of theWFO first identified by Oliver [1977]. Oliver [1977]provides excellent petrologic descriptions of these fabrics,which he refers to as the Malaspina Gneiss, the Turn PointGneiss, and Waipiro Cove Gneiss, respectively (Figure 2b).Here, we report the sequential evolution of these fabrics andshow that they record a progressive change in temperature,pressure, and fluid conditions in the lower crust. They alsorecord a progressive localization of strain toward the top ofthe lower crust and into the middle crust.[23] Two-pyroxene, hornblende-bearing igneous assemb-

lages (SWFO) display weak to moderate grain shape alignmentand lack textural evidence of subsolidus recrystallization.Plagioclase grains are tabular and compositionally zoned,and hornblende commonly is anticlustered. Foliation planesare highly variable in orientation. Garnet- and hornblende-bearing dikes within this fabric cut one another at highangles and are unfoliated (Figure 5a). These textures reflectdeformation prior to the complete crystallization of theWFO batholith.[24] Below and within 1 km of the Doubtful Sound and

Resolution Island shear zones the SWFO fabric is folded,recrystallized and transposed parallel to a penetrative garnetgranulite facies foliation (SGG) (Figure 5b). Thin, 1–4 cmwide veins and dikes that cut SWFO and are bordered byhaloes of garnet granulite [Oliver, 1977] also are transposedparallel to SGG. Garnet- and clinopyroxene-bearing coronassurround the older igneous SWFO assemblages. The SGGfoliation planes are subhorizontal and gently dipping to theNE and SW, and are defined by coarse ribbons of recrystal-lized plagioclase, sheared clinopyroxene + garnet aggre-gates, and minor quartz (Figures 5c and 5d). Stretchedplagioclase forms a penetrative, gently plunging minerallineation (LGG) on foliation planes in high-strain zones.Garnet and clinopyroxene overgrowths on garnet porphyr-oblasts are aligned parallel to this lineation. Plagioclaseexhibits evidence of dynamic recrystallization via subgrainrotation, grain boundary migration, and the formation ofcore-mantle structures (regime 2 dislocation creep, Tullis etal. [2000], Figure 5d). A strong lattice preferred orientation(LPO), first reported by King [2006], occurs in recrystal-lized plagioclase, suggesting deformation occurred by dis-location creep at high-T (>550�C) and under dry conditions.


7 of 27

TC4017

[25] Upper amphibolite facies foliation planes (SSZ) ofthe Doubtful Sound and Resolution Island shear zones aredefined by aligned clusters of plagioclase, hornblende,clinozoisite and, locally, biotite. Stretched hornblende andplagioclase ribbons define penetrative mineral lineations(LSZ) on foliation planes where boudinage indicates thatthey mark true stretching directions. Plagioclase grainsshow recrystallization through grain boundary migration(Figure 5f). Plagioclase grains also contain exsolutionlamellae and form a shape preferred orientation (SPO).These textures reflect deformation under hydrous high-T(T > 550�C) conditions (regime 2 dislocation creep [Tullis etal., 2000]). Most of the hydrous minerals exhibit evidenceof extensive neoblastic growth. Near the boundaries ofhigh-strain zones, garnet granulite facies mineral assemb-lages are recrystallized and transposed into parallelism withthe upper amphibolite facies SSZ foliations (Figure 5e).

Structurally below this transition zone, within at least thefive kilometers of exposed section, the LGG and the LSZ

lineations remain parallel. This parallelism of the lineationsin zones where the two fabrics do not interact suggests thatthe deformations that produced them were kinematicallycompatible.[26] In the hanging walls of the Doubtful Sound and

Resolution Island shear zones Paleozoic metasediment andorthogneiss display several regional upper amphibolitefacies gneissic foliations that may be parallel or oblique toSGG and SSZ (Figures 2b and 2c). Metamorphic mineralassemblages that define these foliations (collectively referredto as S1) may be Paleozoic or Mesozoic [Gibson andIreland, 1998; Hollis et al., 2004]. Within paragneiss, tight,recumbent folds of S1 occur within several hundred metersof the Doubtful Sound Shear Zone (Figures 4c and 4d),

Figure 4. (a) Field photograph and (b) sketch of the conjugate-style geometry of upper amphibolitefacies high-strain zones that define the Doubtful Sound Shear Zone. Note flattened lozenges and oldergranulite foliation (SGG) surrounded by high-strain zones. (c) Sketch of a syntectonic felsic dike that bothcuts mylonitic foliation (SSZ) in the Doubtful Sound Shear Zone and is folded isoclinally within it. U-Pbisotopic data on zircon collected from dike are presented in Figure 11. (d) Sketch of recumbent fold of S1in cover rocks within the Doubtful Sound Shear Zone. Site locations (020, 060, and 0422) are shown inFigure 2b.


8 of 27

TC4017

implying that a component of near vertical shorteningaccompanied deformation in the shear zone.

6. Temperatures and Pressures of Deformation

[27] Estimates of T > 700�C for peak assemblages (SGG)in the footwall of the Doubtful Sound Shear Zone [Oliver,1977;Gibson and Ireland, 1995;Hollis et al., 2004;De Paoli,

2005] are in agreement with T estimates of 750–850�Cderived from similar garnet granulite mineral assemblagesexposed near Milford Sound in northern Fiordland [Clarkeet al., 2000; Daczko et al., 2002b]. The evidence of highgrain boundary mobility and subgrain formation in SGGplagioclase we observed is consistent with deformation atthese high temperatures.[28] Quantitative constraints on P-T conditions accompa-

nying the formation of amphibolite facies (SSZ) strands of

Figure 5


9 of 27

TC4017

the Doubtful Sound Shear Zone have not been discussedpreviously. Plagioclase microstructures indicate that thedeformation occurred at T > 550�C [Olsen and Kohlstedt,1985; Pryer, 1993]. The high variance mineral assemblagesdeveloped during the formation of the Doubtful SoundShear Zone make it is nearly impossible to obtain preciseP-T estimates using conventional thermobarometry. To overcome this problem, we calculated P-T pseudosections(Figure 6) using THERMOCALC software to help definethe P-T history of the WFO metabasites. P-T pseudosectionsdisplay the multivariant mineral assemblages that areencountered by a particular bulk rock composition. Theyare an appropriate way of assessing changing mineralparagenesis in mafic systems, which typically contain highvariance mineral assemblages [e.g., Carson et al., 1999].THERMOCALC is an excellent method for handling thesolid solution series between complex minerals, especiallyamphibole [Powell et al., 1998]. Nevertheless, some fieldscould change by 0.5–1 kbar if alternative end-membermixing models and software techniques are used, as dis-cussed by Pattison [2003].[29] Pseudosections are presented for both gabbroic

(Figure 6a) and dioritic (Figure 6b) rock types of theWFO, using the NCKFMASH (Na2O-CaO-K2O-FeO-MgO-Al2O3-SiO2–H2O) system for H2O-saturated condi-tions. We used version 3.25 of THERMOCALC [Powell etal., 1998], the most recent internally consistent thermody-namic data set (Holland and Powell [1998], upgrade 5.5,November 2003) and the amphibole mixing model of Daleet al. [2005]. Limitations of the pseudosections involve themodeled equilibria being water saturated, and an inability ofthe software to model calcic silicate liquid. These twolimitations make the assemblages shown for T > 650�Cmetastable with respect to melt-present equilibria. To par-tially overcome these limitations, we prepared a T-watercontent (MH2O) diagram [Guiraud et al., 1996; Carson etal., 2000] for the gabbroic rock type (Figure 6c). There islimited evidence for leucosome development contemporarywith slip in the footwall of the Doubtful Sound Shear Zonein these rocks, consistent with metamorphic conditionshaving involved water activities less than one [e.g., Whiteand Powell., 2002]. The H2O content has been normalizedso that MH2O = 100% corresponds to the H2O content at thelowest temperature in the pseudosections. Bulk rock com-positions used in the construction of the pseudosectionswere based on XRF whole rock analyses – SiO2:Al2O3:CaO:MgO:FeO:K2O:Na2O = 52.56: 12.07: 10.80: 9.41:

8.52: 0.40: 4.87 (gabbro) and = 59.49: 11.38: 7.69: 6.79:7.60: 1.22: 4.99 (diorite). The pseudosections were con-structed for metamorphic conditions between T = 550–800�C and P = 4–16 kbar, to adequately model the P-Trange inferred by previous workers.[30] Themodeled water-present pseudosections (Figures 6a

and 6b) indicate that with cooling at P > 8 kbar muscovitebecomes stable first and then clinozoisite. Garnet is unstablefor both diorite and gabbro at moderate P < 7 kbar andmoderate T < 620�C. There is a general spatial progressionin the Doubtful Sound Shear Zone from granulite faciesassemblages in the footwall with limited shear strain, toepidote-hornblende-plagioclase-dominated assemblages inthe hanging wall [De Paoli, 2005]. Intensely foliatedepidote-amphibole-dominated shear strands also cut lessintensely deformed rocks dominated by garnet, diopside,hornblende, and plagioclase; indicating a temporal evolutionof the Doubtful Sound Shear Zone from granulite toamphibolite facies conditions. On the basis of the presenceof clinozoisite, textural evidence for the resorption of garnetand evidence for the dynamic recrystallization of plagio-clase we suggest that deformation in the Doubtful SoundShear Zone occurred at temperatures of T > 550�C andprobably close to 600�C. This temperature range corre-sponds to a conservative pressure range of 6.2–9.4 kbar with aprobable range of 7–9 kbar in the fields involving horn-blende, plagioclase, clinopyroxene (diopside), biotite, clino-zoisite, and quartz for dioritic gneiss (Figure 6a) andhornblende, plagioclase, clinopyroxene, muscovite, clinozoi-site, and quartz for gabbroic gneiss (Figure 6b). This inter-pretation is reasonable because extensional displacementwithin the shear zone would have resulted in exhumationof the lower crust from pressures representing the peak ofgranulite facies metamorphism involving P < 12 kbar in thefields plagioclase-garnet-diopside-biotite-quartz (diorite,Figure 6a) and hornblende-plagioclase-garnet-diopside(gabbro, Figure 6b). Estimates of P � 12 kbar and T �750�C representing the peak of metamorphism are derivedfrom overlapping fields for the assemblages highlighted inFigures 6a and 6b. The P-T path indicated in Figures 6aand 6b (bold black lines) represents cooling and decom-pression from peak garnet granulite facies conditionsthrough to epidote amphibolite (Doubtful Sound ShearZone) conditions. This range of calculated pressures issimilar to those recorded in midcrustal rocks exposed atGeorge Sound [Bradshaw, 1985, 1990] and in the Paparoacore complex [Tulloch and Kimbrough, 1989]. The T-MH2O

Figure 5. (a) Field photograph of two pyroxene magmatic foliation (SWFO) in lower crustal dioritic gneiss. Note the highangles between felsic veins that cut this fabric, which is characteristic of fabric outside the DSSZ. (b) Garnet granulite faciesfoliation (SGG) affecting the Western Fiordland Orthogneiss at Crooked Arm. In this zone, veins and mafic pods aretransposed parallel to the high-T foliation. (c) Field sketch showing crosscutting relationships between the igneous LWFO-SWFO fabric and the granulite facies LGG-SGG fabric in the DSSZ. (d) Thin section (crossed polars) view of deformedgarnet-granulite facies foliation (SGG) in dioritic gneiss at Crooked Arm. Note oblique foliations formed by plagioclase andcuspate-lobate grain boundaries (arrows) indicative of high grain boundary mobility during deformation. Also note opx fishindicating top-to-the-left sense of shear. (e) Photograph showing how garnet granulite facies foliation (SGG) is deflected,recrystallized and transposed by the mylonitic DSSZ. (f) Thin section (crossed polars) view of upper amphibolite faciesfoliation (SSZ) in the Doubtful Sound Shear Zone. Arrows indicate areas of cuspate-lobate plagioclase grain boundaries.


10 of 27

TC4017

Figure 6


11 of 27

TC4017

diagram (Figure 6c) indicates our preferred interpretation ofhowwater activity would have evolved during evolution of theDoubtful Sound Shear Zone: prior to development of the shearzone, water activity would have been less than unity and thegarnet-hornblende-plagioclase-clinopyroxene field wouldhave persisted to lower T conditions than indicated on thewater-saturated pseudosections (Figure 6a). Fluid ingresscontemporary with slip would enable the development ofhornblende-rich, then muscovite and clinozoisite-bearing butfluid-deficient assemblages until water-saturated conditionswere reached at upper amphibolite facies conditions. Weenvisage fluid ingress as having occurred in several stagesduring cooling resulting in a stepped-style T-MH2O path asindicated in Figure 6c.

7. Middle and Lower Crustal Flow Patterns

[31] To characterize the kinematics of flow within themiddle and lower crust, we applied several analyticaltechniques to ductile fabrics in the Paparoa and DoubtfulSound shear zones, respectively. The conjugate geometry ofhigh-strain zones, the orientation of shear zone boundaries,the degree of strain localization, and an analysis of sense ofshear indicators provided boundary conditions. Applicationof the Rf/f/q technique allowed us to evaluate the differentbehavior of feldspar in the two shear zones. This methodhas been used extensively to measure variations in thedegree of alignment and shape of ellipsoidal objects indeformed rocks [e.g., Robin, 1977; Lisle, 1985; De Paor,1988; Simpson and De Paor, 1993; Ramsey and Huber,1983; Klepeis et al., 1999; Bailey and Eyster, 2003; Law etal., 2004]. Distinctive mechanical behaviors are expected onthe basis of differences in the temperatures of deformationin the (midcrustal) Paparoa and the (lower crustal) DoubtfulSound shear zones. The analysis of feldspar also is impor-tant because it is a dominant phase in both shear zones, andit is the mineral commonly assumed to be weakest in thelower crust and therefore control lower crustal rheology[Kohlstedt et al., 1995]. Measurements of quartz andfeldspar shape fabrics across strain gradients in threedimensions were used to examine the rotational historyand behavior of the clasts, rather than to determine truestrains. The magnitudes of finite strain we measured are

underestimates of the true strains because of the combinedeffects of grain size reduction and recrystallization.[32] To determine the finite vorticity of flow, we

employed the porphyroclast hyperbolic distribution methodof Simpson and De Paor [1993, 1997]. This method alsohas been used successfully by others [Klepeis et al., 1999;Bailey and Eyster, 2003; Law et al., 2004]. We exploitedsystems of rotated feldspar grains to determine and comparefinite flow parameters, including the bulk shear direction,the relative contribution of pure shear and simple shear (i.e.,the degree of noncoaxiality), and the degree of crustalthinning in a vertical plane. Analyses at multiple lengthscales were used to obtain spatially averaged estimates ofthese parameters. We cross checked the results of thekinematic vorticity analyses at the scale of the entire middlecrust using the methods of Gapais et al. [1987] andChoukroune et al. [1987], which involve evaluating thesymmetry of shear zones in a coordinate system defined bythe finite strain axes of the deformation. In all cases ourreference frame was the upper and lower boundaries of thePaparoa and Doubtful Sound shear zones. Our P-T deter-minations and the depth-stratified geometry of the sectionindicate that these shear zones formed in shallow and deepparts of the middle crust, respectively.

7.1. Sense of Shear Indicators

[33] Sense of shear indicators in the upper amphibolitefacies bands of the Doubtful Sound Shear Zone includeasymmetric clasts and lozenges (Figures 7a and 7b), sys-tems of rotated plagioclase clasts (Figure 7c), C-S fabric,shear bands (C’-type, Figure 7d), and asymmetric horn-blende and clinozoisite fish. Maximum asymmetry occurson surfaces oriented perpendicular to foliation and parallelto LSZ mineral lineations. The two dominant orientations ofmylonite (SSZ) display opposite senses of shear. Where theSSZ foliation dips gently to the SW the sense of shear is top-down-to-the-SW, where it dips gently to the NE the sense ofshear is top-down-to-the-NE. The two sets of shear bandsoccur together and appear contemporaneous. The anglebetween the two sets ranges from <10� to about 30�.[34] Sense of shear indicators in granulite facies rocks

typically are rare because high-T conditions promote anneal-ing and grain growth that destroys delicate microstructures.

Figure 6. (a) and (b) P-T pseudosections in NCKFMASH, with H2O in excess. Bulk rock composition is specified in text.The univariant reactions experienced by this bulk rock composition, as well as those that bind the higher variance fields thatwere experienced are shown as heavy solid lines and include, in Figure 6b, (1) hb + mu + o + cz + q = pl + g + di, (2) pa + g+ o + q = hb + pl + mu + cz; and in Figure 6a, (1) hb + mu + di = pl + bi + g + cz + q, (2) hb + mu + o + cz + q = pl + g + di,(3) pa + g + o + q = hb + pl + mu + cz, (4) gl + mu + cz = hb + pa + g + o + q, (5) gl + pa + o + cz = hb + pl + mu + q, (6) hb+ mu + o + q = pl + bi + g + di. The following abbreviations are used: opx, orthopyroxene; hb, hornblende; mu, muscovite;o, omphacite; cz, clinozoisite; q, quartz; pl, plagioclase; g, garnet; di, diopside; pa, paragonite; bi, biotite; gl, glaucophane;Ky, kyanite; Sill, sillimanite. Divariant fields are unshaded seven-phase regions, trivariant fields are shaded six-phase fields,quadrivariant fields are deeply shaded five-phase regions, and quinivariant fields are intensely shaded four-phase regions,etc. The heavy bold black line represents the P-T trajectory inferred for the rock types. Bold solid lines highlight the peakassemblage fields in the respective pseudosections, while the heavy dashed black line represents kyanite to sillimanitephase transition. (c) T-MH2O diagram calculated at P = 10 kbar for the rock type shown in Figure 6b. Diagram showsreduced water activities during granulite facies metamorphism prior to development of the Doubtful Sound Shear Zone(white star). Black dashed line is an inferred T-MH2O path showing fluid infiltration in several stages during cooling.


12 of 27

TC4017

We found the following types of asymmetric structureswithin the prograde SGG-LGG fabric: asymmetric garnetand clinopyroxene coronas around hornblende fish, obliqueplagioclase foliations, oblique LPO in plagioclase, ortho-pyroxene fish (Figure 5d), and the deflection of dikes andthe SWFO fabric in high-strain zones. Like those of theLSZ-SSZ fabric, maximum asymmetry occurs within verticalplanes that contain the LGG mineral lineations. Similarly,where the SGG foliation dips gently to the SW the sense ofshear is top-down-to-the-SW. Where it dips gently to theNE, the sense of shear is top-down-to-the-NE. This sym-metric pattern suggests that a component of pure shearaccompanied deformation in the lower crust during granu-lite facies metamorphism as well as later during laterdeformation at the upper amphibolite facies.[35] The directions and sense of shear recorded by the

ductile and brittle components of the Paparoa core complexalso are consistent with regional NE-SW extension and acomponent of subvertical pure shear. In the Paparoa ShearZone, sense of shear indicators include, C-S fabric, micafish, asymmetric tails of biotite and quartz around feldspar

porphyroclasts, microfaults in feldspar, and lattice preferredorientation (LPO) in dynamically recrystallized quartz.These structures reveal a normal, top-down-to-the-SWsense of shear in the Paparoa Shear Zone parallel to L2

mineral stretching lineations. Folding between shear bandslocally has resulted in an apparent reverse sense of shearbetween shear bands that display a dominantly top-down-to-the-NE shear sense (Figure 3i).

7.2. Strain and Vorticity Measurements

[36] We used three sites in the Charleston MetamorphicGroup near White Horse Creek to characterize quartz andfeldspar shape fabrics in the Paparoa Shear Zone using theRf/f/q method [Lisle, 1985; Ramsey and Huber, 1983].Single grains and aggregates of recrystallized grains derivedfrom initial single grains were measured. Site WHC6 islocated structurally below and outside the shear zone(Figure 3b) where the L1-S1 fabric dominates. Site WHC5is located several tens of meters inside the shear zone andsite WHC2 is from a zone of L2-S2 ultramylonite (Figure 8a).Two-dimensional measurements of >80 quartz grains from

Figure 7. (a) Asymmetric, recrystallized plagioclase aggregate in gabbroic gneiss in the DSSZ. Notehornblende rims surrounding garnet and clinopyroxene symplectites, indicative of retrogressed garnetgranulite. (b) Macroscopic asymmetric pod in host gneiss (S1) in the DSSZ. (c) Mylonitic strand of theDSSZ showing asymmetric, rotated plagioclase porphyroclasts used for finite strain and kinematicvorticity analyses. (d) C’ shear band offsetting an amphibolite layer in the DSSZ.


13 of 27

TC4017

each site show an increase in minimum strain ratio (Rs)from Rs = 1.9 outside the shear zone to at least Rs = 2.9 inthe zone of ultramylonite (Figures 8b, 8c and 8d). Incontrast, measurements of >90 feldspar grains at each siteindicated no significant change in strain ratio across thegradient, with Rs values consistently <1.5 (Figures 8e, 8fand 8g) and commonly near 1.0. These patterns illustrate theductility of quartz and minimal shape change in feldspar.They also define low, intermediate and high-strain zones inthe field and confirm a similar, qualitative result indicated

by variations in the degree of F2 fold tightness and thetransposition of S1.[37] A three-dimensional analysis of measured quartz and

feldspar grains from the three sites confirmed the distinctivegeometries of the L1-S1 and L2-S2 fabrics, indicating thatthe Rf/f/q technique is sensitive to these different deforma-tions. To obtain the ellipsoids, we combined two-dimensionalinformation on strain ellipses obtained from three mutuallyperpendicular faces following the methods of Owens [1984]and Bhattacharyya and Hudleston [2001]. A comparison ofseveral solutions at different strain ratios indicates an

Figure 8. (a) Polished slab of L2-S2 mylonite from the Paparoa Shear Zone showing abundant rotatedplagioclase porphyroclasts in a fine-grained biotite-quartz matrix used in finite strain and kinematicvorticity analyses. Rf/f/q data for (b)–(d) quartz and (e)–(g) feldspar clasts in the Paparoa Shear Zoneare shown. Plots combine data from three perpendicular planes. Lower hemisphere equal-area stereoplotsof the best fit finite strain axes and principle planes of strain derived from (h)–(j) quartz and (k)–(m)feldspar discussed in the text. Plots show the best fit three-dimensional ellipsoid calculated for eachsample by recombining the harmonic mean of ellipses found for each of the three perpendicular planes(recommended by Lisle [1985]) according to the least squares method described by Owens [1984].Computer program was provided by B. Tikoff. (n)–(q) Two-dimensional and three-dimensional Rf/f/qdata for feldspar clasts in the DSSZ. Note the wide range of clast orientations (f) and aspect ratios (R)used. See text for explanation.


14 of 27

TC4017

increase in the preferred alignment of the long axes of bothquartz and feldspar grains toward the SSW with increasingstrain (Figures 8h–8m). The calculated directions of max-imum finite extension (X) (Figures 8f and 8m) for bothquartz and feldspar parallel measured L2 mineral lineationsat high strains (Figures 8j and 8m). The cluster of strainaxes from the 5–8 iterations, suggests a precise result. Thepatterns also are consistent with the progressive rotation offeldspar clasts with increasing strain ratio. Axes of maxi-mum finite shortening (Z) plunge steeply in directions thatare subperpendicular to foliation planes (S2) (compareFigures 3e and 8j). The range of plunges is interpreted toreflect folding (F3) of S2 about axes that parallel the Xdirection (Figure 3e). Slightly prolate and slightly oblatefinite strain ellipsoids are represented. However, micro-structural evidence of neoblastic mineral growth (mostlybiotite and muscovite) and fluid transfer within the shearzone raises the possibility of some volume loss duringdeformation, indicating that the significance of these shapedifferences cannot be resolved with certainty. The possibil-ity of significant volume loss suggests that some of thestrain ellipsoids fall within the field of true constriction.[38] For the Doubtful Sound Shear Zone, we used three

sites (0421, 0427 and 0429) from First Arm to investigatefeldspar behavior in the shear zone (Figure 2b). Site 0429 islocated within weakly deformed diorite of the WFO outsideand structurally below the shear zone. This locality isdominated by SWFO. Site 0427 is from several tens ofmeters inside the shear zone and site 0421 is from a zoneof SSZ mylonite at Joseph Point (JP, Figure 2b). Two-dimensional measurements of 30 plagioclase grains fromeach site indicate an increase in minimum strain ratio (Rs)from Rs = 1.8 outside the shear zone to at least Rs = 3.8 atJoseph Point (Figures 8n, 8o and 8p). These patternsconfirm the presence of a strain gradient identified in thefield and they illustrate that shape change in feldspar wasimportant in accommodating strain in the Doubtful SoundShear Zone. However, the strain ratios are much lowerthan those observed in other major detachment faults (e.g.,Rs = 6.4 from a detachment fault measured by Bailey andEyster [2003] in the Pinaleno mountains metamorphic corecomplex). This probably reflects the fact that other phases,such as hornblende and biotite also played a role inaccommodating the strain.[39] A three-dimensional (3-D) analysis of feldspar shape

fabrics from the Doubtful Sound Shear Zone shows the axesof six possible strain tensors that fit the data (Figure 8q).The plot shows a cluster of X axes toward the NE, parallelto LSZ mineral lineations in the DSSZ. However, thecalculated axes also show a larger degree of scatter thanthose in the Paparoa Shear Zone. This relatively poorclustering reflects subsolidus creep in feldspar.[40] The results of a comparison of strain measurements in

the Paparoa and Doubtful Sound shear zones reveal thefollowing patterns: (1) Different feldspar rheologies charac-terize the mylonitic portions of the two shear zones. (2) Despitethe different rheologies, both shear zones display similarsubhorizontal X directions and similar subvertical Z directions.(3) The clasts in shear zones at different depths rotated toward

similar subhorizontal NE-SW directions with increasingstrain.[41] The orientations and patterns of asymmetric tails on

rotated feldspar porphyroclasts (e.g., Figures 9b–9d)allowed us to obtain estimates of the finite vorticity of flowin the Paparoa and Doubtful Sound shear zones. Wemeasured the distribution of rotated clasts at six localitiesat Crooked Arm (Figure 2b) and three localities at WhiteHorse Creek (Figure 3b). This approach requires that thegrains rotated freely in a ductilely deforming matrix[Passchier, 1987]. Our strain measurements (Figure 8)suggest that feldspar clasts in both the Paparoa and DoubtfulSound shear zones rotated toward the direction of maximumfinite extension (X) with increasing strain. Those in thePaparoa Shear Zone behaved approximately rigidly. Wediscuss the implications of some shape change in feldsparon the results below.[42] The distribution of rotated clasts within a ductile

flow field provides a measure of the degree of noncoaxialityof the deformation. This parameter reflects the relativecontributions of the stretching rate and the rotation rate ofmaterial lines in the flow field. The instantaneous rate ofrotation is called the vorticity of flow and can be describedby a vector (w) that reveals the direction and magnitude ofthe rotation. This vector lies perpendicular to the vorticitynormal section (VNS), in which material lines and objectsrotate. In our application, measurements of shear zoneboundaries (SZB, Figure 9a), stretching lineations, differenttypes of shear indicators, and the X direction of the finitestrain ellipsoids were used to identify the finite VNS, whichin both shear zones is approximately vertical (Figures 9eand 9g). The bulk shear direction lies at the intersectionbetween the shear zone boundary and the VNS. Theparallelism among this bulk shear direction, the minerallineations, and the X directions of the 3-D strain calcula-tions indicate the compatibility of the observations(Figures 9e and 9g).[43] A useful means of describing the degree of non-

coaxiality for specific planes of observation is a dimension-less quantity called the kinematic vorticity number (Wk). Intwo dimensions this quantity is a measure of the relativecontributions of the simple shear and pure shear compo-nents such that Wk = 0 for wholly pure shear and Wk = 1 forwholly simple shear [Lister and Williams, 1983; Passchier,1987; Passchier and Urai, 1988]. We used this 2-D measurein specific planes of a 3-D flow field. This is equivalent tothe sectional kinematic vorticity number of Robin andCruden [1994] and Jiang and Williams [1998]. Becausepure shear deformation and simple shear deformation accu-mulate at different rates, equal contributions of pure shearand simple shear occurs at a value of Wk = 0.71 (see Tikoffand Fossen [1995] and Bailey et al. [2007] for a generaldiscussion of kinematic vorticity). Exposure of the shearzone boundaries and the measures of mean kinematicvorticity (Wm) allowed us to determine whether the shearzones were thinning or thickening in the vertical plane andthe relative contributions of pure and simple shear in thisprocess.


15 of 27

TC4017

[44] In both the Paparoa and Doubtful Sound shear zones,a wide range of clast orientations, shapes and tail types,including forward rotated (s-type, d-type) and back-rotated(b-type) clasts (Figures 9b–9d), allowed us to measuremean kinematic vorticity (Wm) at three different scales:the scale of individual thin sections (10 cm2), the outcropscale (10–100 m2), and at the scale of multiple outcrops(10–50 km2). The Paparoa Shear Zone yielded 0.74 < Wm

< 0.79, the Doubtful Sound Shear Zone yielded 0.71 < Wm

< 0.74. These results are derived from the orientation andsize of the fields of back rotation and the eigenvectors offlow (A1 and A2) in the shear zones. Theoretical treatments[Ghosh et al., 2003] indicate that for large deformations,such as those associated with most shear zones, there isalways a stable orientation. The angular width (a) of thisfield is related to the kinematic vorticity number as shownin the equation in Figure 9a. Within each shear zone,estimates of the width of the field of back rotation madeat all three scales were consistent within 2�. This corre-sponds to errors in Wm of ± 0.03. For this reason, allmeasurements have been combined onto single hyperbolicplots (Figures 9f and 9h), which represents a scale-independentresult.[45] The kinematic vorticity data indicate that both the

Paparoa and the Doubtful Sound shear zones are character-ized by subhorizontal (NE-SW) stretching within a flowregime that involved 40–50% pure shear in a vertical plane.The similarity of the results from both shear zones suggestthat the degree of noncoaxiality, the shear zone orientations,and the bulk shear directions are independent of temperatureand shear zone rheology. This may reflect similar styles ofdeformation (subhorizontal extension) and the fact that allthe sites we used include large feldspar porphyroclasts in a

fine-grained mylonitic matrix (compare Figures 7c and 8a).It also may reflect the fact that the rotational component ofdeformation in the Doubtful Sound Shear Zone is wellrecorded by the motion of feldspar grains, allowing us todiscern the mean finite kinematic vorticity.

7.3. Kinematics at the Scale of the Crustal Section

[46] A convenient graphical method for evaluating kine-matic patterns at the crustal scale is described by Gapais etal. [1987] and Choukroune et al. [1987]. This methodinvolves plotting shear zone data on an equal-area stereo-graphic projection in a coordinate system defined by thefinite strain axes of the deformation. In general, the sym-metry of the pattern on the plot provides a measure ofthe relative contributions of pure shear and simple shear tothe deformation and provides a good qualitative test of theresults of our kinematic vorticity analysis. If the deforma-tion is dominated by simple shear, then the shear zonesshould show an asymmetric pattern in the X-Z strain axiscoordinate system. Alternatively, if the bulk strain is coaxialthen the shear zones will display a symmetric pattern in theX-Z plane.[47] The convention used in constructing a strain sym-

metry plot involves describing the geometry of each shearzone using three parameters (Figure 10a). Following theoriginal terminology of Gapais et al. [1987], the sheardirection (L) is taken to be parallel to the stretching lineationin zones of high strain. Our strain measurements andkinematic vorticity analyses (Figures 8, 9, and 10b) indicatethat this approximation works well in both the DoubtfulSound and the Paparoa regions. The normal to the shearingplane (N) is represented by the poles to foliation planes inhigh-strain zones. In our study these planes approximately

Figure 9. Results of kinematic vorticity analysis of L2-S2 mylonites from the Paparoa Shear Zone andLSZ-SSZ mylonites from the Doubtful Sound Shear Zone. (a) Reference model for kinematic analysisshowing the vorticity normal section (VNS). (b) Forward (clockwise) rotated d-type, (c) forwardrotated s-type, and (d) backward (anticlockwise) rotated b-type tails used in the analysis. (e) Equal-arealower hemisphere stereoplot and (f) hyperbolic plot of rotated clast data for the Paparoa Shear Zone.(g) Equal-area lower hemisphere stereoplot and (h) hyperbolic plot of rotated clast data for the DoubtfulSound Shear Zone. See text for explanation and Figures 2b and 3b for sample locations.


16 of 27

TC4017

parallel the boundaries of each individual shear zone. Thenormal to the plane of shear (M) also provides a usefulmeasure of shear zone geometry because its orientationdepends on both the shear direction and the shearing plane[Gapais et al., 1987]. The finite strain axes (X, Y, Z) weobtained for the Doubtful Sound and Paparoa regions(Figure 8) show similar orientations, allowing us to com-bine the data from both regions onto the same plots.[48] The results of this analysis reflect the pattern of

lozenge-shaped, midcrustal pods that are bounded by twosets of conjugate shear zones recording opposite senses ofshear (Figure 10b). Shear directions (L) lie within or closeto the X-Z plane and cluster near the X axis of finite strain(Figure 10c). The two conjugate sets display an approxi-mately symmetric pattern, suggesting a strong component ofpure shear flattening at the scale of the middle and lowercrust. Poles to shearing planes (N) tend to form low anglesto the X-Z plane and cluster near the Z axis (Figure 10d).These patterns reflect high finite strains and the approxi-mately symmetric geometry of shear zones that intersectabout the Y axis. They also conform to patterns that areexpected for approximate plane strain where shear surfacestend to form conjugate planes at low angles to the foliationplanes and shear directions occur mostly in or near theX-Z plane so that they accommodate stretching along X butnot Y [Gapais et al., 1987].

[49] The cluster of M poles (Figure 10e) near the Y axis(i.e., the center of each stereoplot in Figures 10c–10e) andperpendicular to the shear direction (L) reflects the devel-opment of a preferred orientation of shear directions onlens-shaped shear surfaces [Gapais et al., 1987]. Theclustering of N poles near the shortening axis (Z) indicatesthat the lenses are strongly flattened within the regionalfoliation at the scale of the middle crust. This pattern ispredicted for areas of high flattening strains and supportsour interpretation of a significant component of vertical pureshear shortening accompanying extension of the middle andlower crust using kinematic vorticity data (Figure 9) andfold geometries (Figures 4c and 4d). The approximatelysymmetric pattern suggests that the pure shear componentmay be more strongly recorded at the scale of the middlecrust than at the local scale, where the vorticity data indicateapproximately equal components of pure and simple shear(Figure 9). This possibility highlights the importance ofevaluating kinematic data at a variety of scales and suggeststhat the simple shear component could preferentially local-ized into narrow shear zones while the pure shear compo-nent is distributed across a much larger portion of themiddle and lower crust. In this view the kinematic patternsindicated by the kinematic vorticity data correspond to thestrands of high strain shown in Figure 10b (bold, blacklines). The overall patterns displayed by the shear zone dataat both the local and the regional scales are consistent with

Figure 10. (a) Conventions used to describe shear zone geometry and kinematics by three perpendiculardirections [after Gapais et al., 1987]. L, shear direction; N, pole to shearing plane; M, pole to plane ofshear. (b) Block diagram summarizing the three-dimensional conjugate-style shear zone pattern and theorientation of the finite strain axes (X, Y, Z) for the Doubtful Sound and Paparoa regions. Note oblatelozenge-shaped domains. (c)–(e) Equal-area stereographic projections showing distribution of L, N andM directions in a coordinate system defined by the finite strain axes shown in Figure 10b. Black dots areshear zones that record top-down-to-the-northeast displacement, white dots are shear zones that recordtop-down-to-the-southwest displacement.


17 of 27

TC4017

an extensional style that resembles crustal-scale boudinageand involved a high degree of strain localization.

8. U-Pb Geochronology

[50] To determine the absolute age of ductile fabrics inthe Doubtful Sound and Paparoa shear zones, we analyzedzircon using laser ablation (LA) inductively coupled plasmamass spectroscopy (ICPMS) at the University of Arizona.Analytical methods and data tables (Table 1) are presentedin Appendix A. In most cases the LA-ICPMS data providedinformation that was difficult or impossible to obtain withother (Thermal Ionization Mass Spectrometry, TIMS, orSensitive High Resolution Ion MicroProbe, SHRIMP) tech-niques. Here, we report 206Pb/238U ages derived from twosamples (0422 and 0412) from the Doubtful Sound regionand one sample (03CH2) from the Paparoa core complex.[51] A pegmatitic dike (sample 0422) cuts the SSZ mylo-

nitic foliation of the Doubtful Sound Shear Zone and also istightly folded within it (Figure 4c). This syntectonic rela-tionship allows us to place limits on the ages of the SGG andSSZ fabrics. We also analyzed zircon from a diorite (sample0412) on Secretary Island (Figure 2b) that intruded parallelto the S1 gneissic foliation. This sill is one of several that aredeformed by a syncline that occurs between two mylonitic(SSZ) shear zones exposed along Doubtful Sound andBradshaw Sound (Figure 2b).[52] Analyses of most zircons from sample 0422 yielded

an age of 102.1 ± 1.8 Ma (Figures 11a and 11b). A fewgrains give inheritance ages scattered within the ranges 320–380Ma and 450–550Ma (Figure 11a). High uranium contentsand high U/Th ratios in the zircon grains (Figure 11c)suggest that metamorphism occurred during igneous crys-tallization of the dike. This interpretation is consistent withits position within the shear zone. These young agesindicate that deformation in the shear zone must have begunbefore and outlasted 102.1 ± 1.8 Ma. The inherited agesmost likely reflect the age of host gneiss in the hanging wallof the shear zone.[53] Sample 0412 yielded no sign of inheritance and

gave an igneous crystallization age of 120.2 ± 2.2 Ma(Figures 11d and 11e). Zircon tips (Figure 11f) and cores(Figure 11g) have identical ages. These ages confirm thatthe syncline and shear zones that deform Paleozoic hostgneiss in the hanging wall of the Doubtful Sound andBradshaw Sound shear zones are Cretaceous and notPaleozoic features.[54] Isotopic data obtained by Spell et al. [2000] from

the Paparoa range using the 40Ar/39Ar method indicaterapid cooling of the core from midcrustal temperatures ofT > 500�C at �110 Ma to T = 170�C by �90 Ma at a rate of�110�C per million years. These results suggest that ductiledeformation in the Paparoa and Doubtful Sound shear zoneswas contemporaneous. To test this interpretation, we ana-lyzed zircon from a granitic dike (sample 03CH2) that cutsacross the S1 foliation in biotite schist of the CharlestonMetamorphic Group and is folded. This syntectonic rela-tionship places an upper limit on the age of S1 and a lowerlimit on the age of the folding.

[55] Analyses of zircon from sample 03CH2 yielded anage of 113.2 ± 4.3 Ma (Figures 11h and 11i). These youngages form a good cluster and display a range of U/Th ratios(Figure 11j), indicating that they could be metamorphic origneous in origin. Older grains reflect inheritance. Zircongrains that cluster near 300 Ma and are older than 600 Marecord low U/Th ratios, suggesting that they are of igneousorigin. The cluster of ages in the range 400–500 Ma hashigh U/Th ratios, suggesting that they may reflect Paleozoicmetamorphism. The young ages, which are most importantfor our study, indicate that folding in the core is Cretaceous.This latter result proves that Mesozoic extension locallyreactivated older Paleozoic fabrics that comprise much ofthe middle crust.

9. Discussion

9.1. Evolution of the Lower and Middle Crust DuringExtension

[56] The age and tectonic significance of superposedlower and middle crustal fabrics has been an unsolvedproblem in Fiordland for many years. Prior to our study,only one U-Pb zircon date (119 ± 5 Ma, 1s) had beenobtained from an upper amphibolite facies sample in theDoubtful Sound region [Gibson et al., 1988] and none fromthe hanging wall of the Doubtful Sound Shear Zone. Theinterpretation presented by these authors is that this agecorresponds to movement along the Doubtful Sound ShearZone. However, this is inconsistent with crosscutting rela-tionships and the 116–112 Ma age of the WFO batholith inthe area [Ireland and Gibson, 1998; Tulloch and Kimbrough,2003; Hollis et al., 2004].[57] The structural, petrologic, and isotopic data reported

in this paper allow us to place new constraints on the agesand progressive evolution of lower crustal fabrics in centralFiordland. Crosscutting relationships and U-Pb data onzircon reveal a rapid progression from magmatic flow(SWFO) to high-T (T > 700�C) subsolidus deformation atthe garnet granulite facies (LGG-SGG) and, finally, to cooler(T = 550–650�C) deformation in networks of narrow upperamphibolite facies shear zones (LSZ-SSZ). The dates fromsample 0412, together with those reported by Hollis et al.[2004] from Crooked Arm, indicate that emplacement of theWFO and its satellite intrusions began at �120 Ma, and thatmost bodies in the lower crust had crystallized by 114 ±2.2 Ma. The ages (102.1 ± 1.8 Ma) from sample 0422 andour P-T determinations link the formation of the upperamphibolite facies shear zones to a period (114–111 Ma)of lower crustal cooling following magmatism and granulitefacies metamorphism (Figure 12). This result agrees wellwith data reported by Flowers et al. [2005], who showedthat the WFO and its host rock cooled from T > 750�Cthrough 550–650�C by �111 Ma without involving sig-nificant exhumation. Our work indicates that extension-induced exhumation and decompression occurred aftercirca 111 Ma when deformation was localized in narrowmidcrustal shear zones. Crosscutting relationships betweenthe Doubtful Sound Shear Zone and a pegmatite that yieldedan age of 88.4 ± 1.2 Ma [King et al., 2005; King, 2006]


18 of 27

TC4017

Table 1. U-Pb (Zircon) Geochronologic Analyses by Laser Ablation Multicollector ICP Mass Spectrometrya

Sample-Grain

U,ppm

Isotopic Ratios Apparent Ages, Ma

206Pb/204Pb U/Th

207Pb*/235U ±1s, %

206Pb*/238U ±1s, %

ErrorCorrection

206Pb*/238U

Error ±,Ma

207Pb*/235U

±1s,Ma

206Pb*/207Pb*

±1s,Ma

03CH2-3C 16 2486 1.5 0.41840 24.3 0.06013 7.1 0.29 376 28 355 98 217 26903CH2-4T 348 2989 3.8 0.12047 27.6 0.01744 4.6 0.16 112 5 116 33 200 31603CH2-4T2 259 2511 2.8 0.14869 19.6 0.01788 3.1 0.16 114 4 141 29 615 20903CH2-5C 262 11982 2.1 0.50758 7.6 0.06677 5.3 0.69 417 23 417 39 418 6203CH2-6C 553 9131 18.8 0.65939 7.0 0.08217 6.2 0.88 509 33 514 46 537 3603CH2-8T 158 1366 1.8 0.13080 27.0 0.01929 5.2 0.19 123 7 125 35 157 31003CH2-8T 158 12667 1.8 0.15005 21.8 0.01878 5.1 0.23 120 6 142 33 527 23203CH2-8T2 185 3158 2.5 0.18168 17.3 0.01838 3.4 0.20 117 4 170 32 977 17303CH2-10C 372 13371 2.9 1.09432 4.9 0.11522 3.8 0.77 703 28 751 53 895 3203CH2-11T 396 925 3.1 0.12113 22.7 0.01766 2.9 0.13 113 3 116 28 183 26203CH2-11T2 464 1314 7.2 0.13608 16.6 0.01737 3.5 0.21 111 4 130 23 484 17903CH2-11C 113 1922 2.3 0.57308 14.9 0.07320 7.1 0.48 455 34 460 83 483 14403CH2-12T 579 585 28.4 0.13440 29.6 0.01736 3.8 0.13 111 4 128 40 458 32503CH2-12T2 914 1052 21.9 0.10145 35.7 0.01749 13.1 0.37 112 15 98 36 �222 41803CH2-12C 130 4226 7.8 0.67472 8.1 0.08004 4.3 0.53 496 22 524 54 644 7303CH2-13T 148 13623 2.3 0.16663 51.0 0.01877 4.9 0.10 120 6 157 83 754 53503CH2-13T2 190 7585 14.9 0.83314 7.0 0.07705 2.4 0.34 479 12 615 57 1158 6503CH2-13C 473 37346 9.9 0.75802 4.6 0.08795 3.9 0.84 543 22 573 35 692 2603CH2-13C2 62 6862 2.3 3.73290 7.1 0.24949 3.5 0.49 1436 56 1579 239 1775 5603CH2-14T 1130 2077 106.5 0.13611 14.0 0.01782 2.4 0.17 114 3 130 19 428 15403CH2-14T2 561 1423 7.4 0.11751 19.1 0.01674 2.2 0.11 107 2 113 23 237 21803CH2-15T 82 3215 17.6 0.44790 16.1 0.05530 4.1 0.25 347 15 376 71 557 16903CH2-15C 91 6402 19.9 0.41205 19.1 0.05655 3.7 0.19 355 13 350 77 322 21303CH2-16T 338 1885 6.2 0.11701 21.8 0.01851 2.5 0.11 118 3 112 26 �10 26203CH2-16T2 382 2809 3.9 0.12328 12.3 0.01806 4.3 0.35 115 5 118 15 172 13403CH2-16C 42 1203 2.1 1.64181 14.5 0.15506 5.7 0.39 929 56 986 217 1116 13303CH2-17T 480 3210 12.2 0.10977 15.4 0.01930 1.6 0.10 123 2 106 17 �272 19503CH2-17T2 501 2888 9.9 0.15156 10.7 0.01819 3.3 0.31 116 4 143 16 619 11003CH2-17C 129 26905 4.8 0.80481 9.8 0.08592 3.5 0.36 531 19 600 77 867 9503CH2-18T 1126 539 22.8 0.11353 33.2 0.01707 8.8 0.26 109 10 109 38 111 37803CH2-18C 192 4673 24.4 0.55007 10.2 0.06956 6.3 0.62 434 28 445 55 505 8803CH2-19T 809 6915 19.3 0.13140 10.2 0.01878 3.4 0.33 120 4 125 14 229 11103CH2-19T2 550 6221 10.9 0.13922 11.5 0.01855 4.6 0.39 119 5 132 16 389 11903CH2-19C 45 6445 3.6 1.55511 23.2 0.15190 15.3 0.66 912 148 953 313 1048 17604-12-1T 166 780 1.2 0.13636 88.5 0.01907 3.9 0.04 121.8 4.7 130 108 279 71704-12-1T2 276 973 1.0 0.13733 21.4 0.01874 2.4 0.11 119.7 2.8 131 26 335 48704-12-2T 234 869 0.9 0.12860 24.3 0.01885 2.0 0.08 120.4 2.4 123 28 171 57304-12-3T 179 493 1.5 0.12503 96.0 0.01913 4.1 0.04 122.2 4.9 120 109 69 93904-12-4T 122 470 1.2 0.12075 31.6 0.01921 3.3 0.11 122.6 4.0 116 35 �24 77604-12-5T 210 723 0.9 0.13180 20.3 0.01904 3.0 0.15 121.6 3.6 126 24 204 46904-12-6T 130 1035 1.1 0.17445 32.3 0.01928 5.0 0.16 123.1 6.2 163 49 794 68704-12-7C 170 559 0.9 0.17181 125.3 0.01829 3.6 0.03 116.8 4.1 161 189 872 56704-12-7T 265 859 1.1 0.14476 25.2 0.01843 3.7 0.15 117.8 4.3 137 32 490 55904-12-8T 159 596 0.8 0.17830 41.5 0.01837 4.1 0.10 117.4 4.7 167 64 940 88504-12-8C 163 257 0.8 0.09220 37.8 0.01857 4.0 0.11 118.6 4.7 90 32 �631 105804-12-9T 155 507 1.2 0.13202 41.5 0.01879 4.4 0.10 120.0 5.2 126 49 239 99104-12-10T 127 618 0.9 0.13708 46.1 0.01916 6.2 0.13 122.4 7.5 130 56 280 109904-12-11T 209 1168 0.9 0.15347 29.4 0.01940 3.3 0.11 123.9 4.1 145 40 506 65504-12-11C 86 368 1.0 0.11446 39.0 0.01931 6.6 0.17 123.3 8.1 110 41 �168 98904-12-13C 117 747 0.9 0.14011 32.7 0.01907 5.2 0.16 121.8 6.3 133 41 341 74804-12-14T 125 831 1.4 0.16618 46.1 0.01888 7.2 0.16 120.6 8.6 156 67 736 101604-12-14C 308 1138 0.7 0.11461 44.3 0.01853 4.2 0.10 118.4 4.9 110 46 �64 112504-12-15T 147 707 1.0 0.16699 36.0 0.01863 6.1 0.17 119.0 7.2 157 52 775 76904-12-16T 174 1014 0.8 0.11955 61.2 0.01905 4.0 0.06 121.6 4.8 115 66 �28 162704-12-16C 228 848 0.7 0.15211 21.6 0.01831 3.4 0.16 117.0 3.9 144 29 613 46504-22-1T 197 824 12.1 0.14034 42.9 0.01525 3.9 0.09 97.6 3.8 133 54 830 93304-22-1C 66 740 35.0 0.12802 92.2 0.01618 9.8 0.11 103.5 10.1 122 107 506 58304-22-2T 224 927 14.2 0.14851 38.9 0.01537 5.0 0.13 98.3 4.9 141 51 931 82204-22-3T 265 762 20.5 0.12360 21.2 0.01650 2.4 0.11 105.5 2.5 118 24 384 47704-22-4T 298 960 19.6 0.10567 37.0 0.01678 2.4 0.07 107.2 2.6 102 36 �19 92204-22-4C 389 1252 18.8 0.09733 26.9 0.01588 2.1 0.08 101.5 2.1 94 24 �85 66804-22-5T 321 683 17.3 0.10945 27.2 0.01537 2.4 0.09 98.3 2.4 105 27 270 63104-22-5C 315 1372 17.4 0.12205 12.1 0.01566 2.3 0.19 100.1 2.3 117 13 474 26404-22-6T 199 732 17.0 0.12414 60.0 0.01623 2.8 0.05 103.8 2.9 119 67 431 147204-22-6C 100 931 39.1 0.10994 70.5 0.01630 5.0 0.07 104.2 5.2 106 71 144 190004-22-7T 6430 74187 3.0 0.57610 0.8 0.07416 0.6 0.79 461.2 2.7 462 3 466 11


19 of 27

TC4017

indicate that all deformation in the lower crustal shear zoneterminated prior to circa 90 Ma (Figure 12).[58] The 113.2 ± 4.3 Ma ages we obtained from sample

03CH2 show that Cretaceous deformation in the midcrustalcore of the Paparoa range, as well as along the detachments,occurred contemporaneously with ductile flow in the middleand lower crustal exposures in central Fiordland. The ageswe obtained are in good agreement with previous ages fromthe Charleston Metamorphic Group reported by Tulloch andKimbrough [1989], Ireland and Gibson [1998], and Spell etal. [2000]. They are also similar to 109–111 Ma crystalli-zation ages for the Buckland granite (Figure 3a) obtained byMuir et al. [1997] and Ireland and Gibson [1998]. Thesedata establish the central Fiordland-Paparoa extensionalsystem as one of the oldest in New Zealand, a conclusionalso reached by Kula et al. [2005].[59] In addition to resolving previously conflicting inter-

pretations on the age of extensional deformation in thelower crust, our data show that the mechanism of lowercrustal cooling in central Fiordland differs from that whichaccompanied cooling in northern Fiordland. In northernFiordland, lower crustal cooling occurred within a regimeof contraction as arc-related magmatism waned and partialmelts migrated out of the lower crust [Klepeis et al., 2004;Marcotte et al., 2005]. Our results indicate that the initiationof crustal extension, as recorded by the LGG-SGG fabric,occurred in central Fiordland between 114 and 111 Ma(Figure 12). The fact that this extensional fabric does notoccur in northern Fiordland, strongly suggests that exten-sion preferentially localized into central Fiordland crust,which was hot and weak during this interval. It alsoindicates that zones of lower crustal weakening due to theinjection of magma and heat are spatially extremely hetero-geneous and transient.[60] The dates from sample 0422 and crosscutting rela-

tionships between the LGG-SGG and LSZ-SSZ fabrics indicatethat the period of high-T deformation and crustal thinningassociated with granulite facies metamorphism was short-lived, lasting 3–4 Ma. The high-T LGG-SGG fabric formedbefore �111 Ma at deep (35–40 km) levels within the lowercrust, the cooler LSZ-SSZ fabric formed after �111 Ma atshallower depths (25–30 km). This trajectory of rapid

cooling (Figure 12) implies that the effective viscosity ofthe lower crust also increased and its mechanical behaviorchanged as the extension proceeded. This interpretation isbased partly on the dependence of lower crustal viscosity ontemperature, which is independent of the specific rheologyused [Turcotte and Schubert, 2002]. Numerical studies alsoshow a similar interdependence between temperature andthe mechanical behavior of deforming middle and lowercrust [Jamieson et al., 2002; Vanderhaeghe et al., 2003].The lack of migmatite or evidence of widespread partialmelting in the middle and lower crust during the period111–90 Ma also supports a relatively cool, strong lowercrust beneath the developing core complexes. The paucityof migmatite reflects the infertility of the mafic orthogneiss[Daczko et al., 2001b; Antignano, 2002] and contrasts withmigmatite domes, such as the Naxos gneiss dome in Greece[Vanderhaeghe, 2004], the Central Gneiss Complex incoastal British Columbia [Andronicos et al., 2003], andthe Fosdick gneiss dome in West Antarctica [Siddoway etal., 2004]. This conclusion is similar to one reached byDaczko et al. [2002c] and Klepeis et al. [2004] in northernFiordland. The removal of heat by fluid transfer andconductive heat loss as cool middle crust was juxtaposedagainst warmer lower crust during crustal thinning andextension appears to have aided lower crustal cooling.Our interpretation also agrees well with the conclusions ofMcKenzie and Jackson [2002], who showed that the char-acteristics of lower crustal flow are a consequence of thetimescale over which changes in viscosity occur and thatthis timescale may be controlled by the conductive coolingof intrusions and/or by the removal of melt.[61] At the same time as the lower crust cooled, the

inherently weak quartz and biotite assemblages that domi-nate the middle crust accentuated the strength contrastbetween the middle and lower crust. These patterns suggestthat the combination of rapid cooling and strength differ-ences created by lithologic contrasts resulted in a lowercrust that was strong relative to the middle crust during theinterval 111–90 Ma. Accompanying these changing con-ditions, deformation became increasingly partitioned intothe middle crust and along its boundary with the lowercrust. This localization of strain into arrays of narrow shear

Table 1. (continued)

Sample-Grain

U,ppm

Isotopic Ratios Apparent Ages, Ma

206Pb/204Pb U/Th

207Pb*/235U ±1s, %

206Pb*/238U ±1s, %

ErrorCorrection

206Pb*/238U

Error ±,Ma

207Pb*/235U

±1s,Ma

206Pb*/207Pb*

±1s,Ma

04-22-8T 419 1381 16.5 0.12625 17.9 0.01653 2.6 0.15 105.7 2.8 121 20 428 39804-22-9T 487 3396 12.2 0.10353 25.8 0.01590 2.0 0.08 101.7 2.0 100 25 60 62304-22-9C 390 2916 12.7 0.12947 16.7 0.01577 2.1 0.13 100.8 2.1 124 19 588 36204-22-10T 300 1343 13.7 0.11866 29.8 0.01577 2.7 0.09 100.9 2.7 114 32 395 67904-22-11T 166 2272 3.8 0.63751 9.4 0.08477 3.4 0.36 524.5 17.1 501 37 394 19704-22-12T 269 1123 3.0 0.44296 17.6 0.05897 2.1 0.12 369.4 7.6 372 55 391 39404-22-13T 764 5109 1.7 0.62420 4.2 0.07935 2.4 0.56 492.3 11.2 492 16 493 7604-22-14T 1540 4881 25.7 0.40016 5.0 0.05372 4.4 0.87 337.3 14.3 342 15 372 56

aC, core of grain analyzed; T, tip or rim of grain analyzed. All errors are reported at the 1s level and incorporate only uncertainties from measurement ofisotopic ratios. U concentration and U/Th have uncertainty of �25%. Decay constants are 235U = 9.8485 � 10�10, 238U = 1.55125 � 10�10, 238U/235U =137.88. Isotope ratios are corrected for Pb/U fractionation by comparison with standard zircon with an age of 564 ± 4 Ma. Initial Pb composition interpretedfrom Stacey and Kramers [1975], with uncertainties of 1.0 for 206Pb/204Pb and 0.3 for 207Pb/204Pb.


20 of 27

TC4017

Figure

11


21 of 27

TC4017

zones contrasts with styles of penetrative, wholesale ductileflow of the lower crust in regions where lower crustaltemperatures are high [e.g., Wernicke, 1992; McKenzie andJackson [2002]; Amato and Miller, 2004].

9.2. Model of Core Complex-Style Extension

[62] The results of our study show that any model of mid-Late Cretaceous extension in western New Zealand mustexplain the following observations: (1) The initiation ofsubhorizontal flow and vertical thinning of the lower crustin central Fiordland at garnet granulite facies conditions(T > 700�C) at or shortly after 114 Ma. (2) Cooling of thelower crust to T = 550–650�C by circa 111 Ma [Flowers etal., 2005]. (3) The formation of thin (�0.5 km) networks ofupper amphibolite facies shear zones that cut through themiddle and lower crust. (4) Collapse of the upper crust andexhumation of the middle crust but not the lower crustabove networks of symmetric, conjugate-style midcrustalshear zones during the period 111–90 Ma. Below we presenta three-stage tectonic model of extension (Figure 13) thatincorporates transient lower crustal weakening followed byductile deformation that progressively localized into arelatively weak middle crust. This model is consistent witha ‘‘bottom-driven’’ system where subhorizontal flow beginsin the lower crust and induces the collapse and dismember-ment of the upper crust above midcrustal shear zones.Similar types of models where continental deformation iscontrolled by flow in the deep crust and mantle aredescribed by Tikoff et al. [2002, 2004] and Giorgis et al.[2004].[63] Emplacement of mafic-intermediate magma into a

layered lower crust occurs during the interval 120–114 Ma(Figure 13a). The initiation of subhorizontal flow andvertical thinning of the lower crust begins at �114 Maand is localized into areas weakened by magmatism andheat (Figure 13b). Granulite facies shear zones form in thelower crust during the period 114–111 Ma. At these timesthe lower crust is weaker than the middle crust due to theelevated temperatures (T > 700�C) and the residual effectsof magmatism. By �111 the lower crust cools to T < 650�Cas a result of extension-induced exhumation. Between �111and �90 Ma, when the lower crust is relatively cool (T =550–650�C), deformation progressively localizes in themiddle crust and along the boundary between the middleand lower crust where strength contrasts due to lithologicdifferences control deformation partitioning (Figure 13c).Extension during this third stage results in the formation ofthe thin upper amphibolite facies shear zones. These narrowzones relay displacements through a ductilely deforming

middle crust along conjugate-style shear zones that eventu-ally exhume midcrustal material. Although most of thedeformation is localized in the middle crust a few narrowshear zones penetrated the lower crust (Figures 2c and 13c).These latter shear zones are required for kinematic compat-ibility among the deforming layers of the lithosphere.However, the lower crust is not exhumed at this time,implying that the magnitude of displacement on individualshear zones was moderate but not extreme, with displace-ments distributed among many shear zones. Estimates basedon differences in metamorphic pressures across the Doubt-ful Sound [cf. Oliver, 1977] and Mount Irene [Scott andCooper, 2006] shear zones suggest displacement magni-tudes on the order of 10–20 km. The resulting lower crustalstructure (Figure 13c) resembles large-scale boudinage.[64] The occurrence of this highly focused, conjugate

style of upper amphibolite facies deformation in Fiordlandand the Paparoa Range invites comparison to previouslypublished models of lithospheric-scale extension at othercontinental margins. Nagel and Buck [2004] describe anumerical simulation of continental rifting that explainsstructural features at magma-poor margins such as theGalicia margin west of the Iberian Peninsula and the ApuliaMargin in the Alps. In their model, symmetric styles ofextension result from the collapse of the upper crust above ahorizontal midcrustal shear zone on top of a relativelystrong lower crust. Thermal perturbations locally aid exten-sion and the lower crust is not exhumed within the conti-nental margin. These characteristics and their causes aresimilar to those displayed by the New Zealand example.Our data augment this model by suggesting that changingthermal conditions in the lower crust and a compositionallystratified crustal column can enhance weakness in themiddle crust and lead to a highly localized form of exten-sion. These factors, combined with a flow regime involvingboth pure and simple shear components, led to a symmetricstyle of shear zone development. The similar kinematics,ages, and styles of mylonitic deformation exhibited bycentral Fiordland (the Doubtful Sound and ResolutionIsland shear zones) and the Paparoa core complex (thePaparoa Shear Zone) during the 111–90 Ma period alsosuggest that shear zones forming at different depths withinthe middle crust involved coordinated, kinematically cou-pled flow. Our results illustrate the transient nature of lowercrustal weakening and that midcrustal processes controlledthe formation of core complexes in New Zealand.[65] The progressive localization of extensional deforma-

tion into a weak middle crust and the style of crustal-scaleboudinage that results is similar to processes observed in

Figure 11. U-Pb isotopic data from zircon collected using laser ablation ICPMS. (a), (d) and (h) Concordia plots thatshow all analyses from samples 0422, 0412, and 03CH2, respectively. See Figures 2b and 3a for sample locations. Errorellipses on these plots are at the 1s level, ages are shown at the 2s level. An explanation of methods is provided inAppendix A. (b), (e), (f), (g), and (i) show the distribution of 206Pb/238U ages measured. Two uncertainties are shown; thoselabeled ‘‘age’’ include all errors, whereas the ‘‘mean’’ includes only the random (measurement) errors. The largeruncertainty is reported as the age of the grains in the text. Note that these uncertainties are shown at 2s levels, whereas theerror bars for individual analyses are shown at 1s levels. (c) and (j) U/Th versus 206Pb/238U ages for samples 0422 and03CH2, respectively. These plots helped us to discriminate between igneous and metamorphic zircon. See text forexplanation.


22 of 27

TC4017

some extensional gneiss domes in the Aegean Sea. Jolivet etal. [2004] describe the extensional unroofing of middlecrust in the Tinos gneiss dome as a result of the progressivelocalization of deformation into detachment faults duringstretching and crustal-scale boudinage. In this case a seriesof subparallel detachment faults that root into the middlecrust formed in high-strain zones between individual bou-dins. Strain localization, and the boudinage, was a directconsequence of the original heterogeneity of the deformedmaterial and a rheological stratification of the lithosphere[see also Karlstrom and Williams, 1998]. Fluid infiltrationinto the middle crust enhanced the localization process.Displacements on the faults caused the extensional unroof-ing of the middle crust but not the lower crust. Lower crustwas exhumed only in areas that record extreme extension ina hot lower crust. The similarities among the Aegean andFiordland examples suggest that relatively cool lower crustaltemperatures and ductile flow that is focused into a weakmiddle crust may inhibit amplification of displacements indetachment faults, leading to exhumation of middle but notlower crust. This result, and the predominantly symmetricstyle of extension at the crustal-scale, also conforms to thepredictions of numerical models that incorporated layers ofvariable strength. For example, Huismans et al. [2005]showed that symmetric, pure shear-dominated styles ofextension are promoted by fast rates of extension of cold,high-viscosity lithosphere whereas slow rates of extensionof weak lithosphere tend to promote asymmetric, simpleshear-dominated styles.

[66] The results of our analysis of flow patterns in themiddle crust during the extensional collapse and dismem-berment of the upper crust (111–90 Ma interval) allows usto characterize this process. The kinematic data suggest asymmetric to slightly asymmetric style of extension at thescale of the middle crust. Subhorizontal (NE-SW) flow ofthe middle and lower crust was accommodated by verticalthinning, involving approximately equal amounts of pureshear and simple shear, and a localization of strain intonetworks of thin, conjugate-style shear zones. The predom-inance of horizontal fabrics in central Fiordland is expectedfor areas where vertical thinning and lateral flow accom-modate extensional deformation. High rates of lateral flowalso explain why rapid cooling during the 114–111 Mainterval (stage 2) was not accompanied by significantexhumation (Figures 11 and 12b). Only after circa 111 Mawhen midcrustal shear zones that cut across the middle andlower crust formed did exhumation of the middle crustbegin. This implies that cooling prior to circa 111 Ma wascontrolled by vertical thinning (necking) of the lower crustor by some other mechanism.

10. Conclusions

[67] The results of a field-based investigation of exhumedmiddle and lower crust illustrate the important role of aweak middle crust in controlling the behavior of extendingcontinental lithosphere and the formation of metamorphic

Figure 12. Temperature-time-deformation (T-t-D) path for the lower plate of the Doubtful Sound ShearZone, central Fiordland. Isotopic age data sources are as follows: 1, Hollis et al. [2004]; 2, King et al.[2005]; 3, Flowers et al. [2005]; 4, Gibson et al. [1988]. Results from this study are marked with blacksquares. Deformation history is based on field data reported in this study.


23 of 27

TC4017

core complexes. Central Fiordland exposes a 10–15 kmthick section of lower crust composed of both wet (am-phibolite) and dry (granulite) varieties of metagabbro andmetadiorite. A 15 km thick section of middle crust exposedin Fiordland and the Paparoa core complex in Westland iscomposed mostly of paragneiss, quartz-biotite schist, mar-ble, granite and various other igneous units.

[68] Superposed lower crustal fabrics in the footwalls oftwo major upper amphibolite facies shear zones, the Doubt-ful Sound and Resolution Island shear zones, record atransition from magmatic flow to high-temperature deforma-tion at the garnet granulite facies (T > 700�C; P = 12 kbar)to cooler deformation at the upper amphibolite facies(T = 550–650�C, P = 7–9 kbar). Kinematic data, P-Tdeterminations, and U-Pb isotopic analyses on zircon indi-cate that crustal thinning and subhorizontal stretching initi-ated in parts of the lower crust that were weakened bymagma and heat by �114 Ma. However, as the lower crustrapidly cooled and strengthened over a period of 3–4 Ma,deformation progressively localized into a middle crust thatwas weak relative to the lower crust.[69] The progressive localization of strain into the middle

crust after circa 111 Ma resulted in the formation ofconjugate-style arrays of mylonitic shear zones that dissectedthe middle crust and, in the case of the Doubtful Sound andResolution Island shear zones, penetrated into the lowercrust. Displacements within these arrays resulted in astructural style that resembles crustal-scale boudinage.Strain and kinematic vorticity data indicate that the upperand lower parts of the middle crust (depths of 15–20 kmand 25–30 km, respectively) record different rheologiesbut similar flow patterns involving pure shear-dominatedvertical thinning and subhorizontal stretching. In the uppercrust, the Pike and Ohika detachment faults dip in oppositedirections and display opposite senses of shear. Ductile flowin the middle crust and displacements on these detachmentfaults were kinematically compatible and resulted in theexhumation of midcrustal material and formation of thePaparoa metamorphic core complex during the interval111–90 Ma. The lower crust was not exhumed during thisperiod.[70] Our results provide direct confirmation that horizon-

tal stretching and vertical thinning in a middle crust that isweak relative to the lower crust can result in the formationof metamorphic core complexes. The data support modelssuggesting that core complex formation occurred after thelower crust had cooled significantly and was strong andhighly viscous relative to the middle crust. Initial weakeningof the lower crust by magma and heat was followed byfluid-enhanced weakening in narrow hydrous shear zones inthe middle crust. Time-space relationships support a ‘‘bot-tom-driven’’ model of extension where subhorizontal flowinitiated in the lower crust prior to the collapse anddismemberment of the upper crust. Comparisons betweennorthern and central Fiordland suggest that rheologicaltransitions linked to lower crustal magmatism and theireffects on the mechanical behavior of continental litho-sphere may be much more heterogeneous and short-livedthan previously believed.

Appendix A: U-Pb Analytical Methods

[71] Zircons were analyzed with a Micromass Isoprobemulticollector ICPMS equipped with 9 faraday collectors,an axial Daly detector, and four ion-counting channels. The

Figure 13. Three-stage tectonic model of Cretaceousextension leading to the formation of metamorphic corecomplexes in western New Zealand. (a) Stage 1, emplace-ment of mafic-intermediate magma (the WFO) into thelower crust (lc) of central Fiordland. (b) Stage 2,subhorizontal stretching and vertical thinning begins athigh-T (Tlc = 700�C) in areas of the lower crust weakenedby magma and heat. Note that northern Fiordland,represented by right side of diagram, does not recordextension because it was cool. (c) Stage 3, deformationinvolving a cooling (Tlc = 550–650�C) lower crust incentral Fiordland. Formation of metamorphic core com-plexes that exhume midcrustal but not lower crustal materialform above a middle crust that is weak compared to thelower crust. See text for explanation. Abbreviations (forlocations, see Figures 1, 2, and 3): BG, Buckland granite;BS, Bradshaw Sound; CA, Crooked Arm; DS, DoubtfulSound; MI, Mount Irene; PSZ, Paparoa Shear Zone; RI,Resolution Island; VR, Victoria range core complex; WFO,Western Fiordland Orthogneiss; WJ, Wet Jacket Arm.


24 of 27

TC4017

Isoprobe is equipped with a New Wave DUV 193 laserablation system with an emission wavelength of 193 nm.The analyses were conducted on 35 mm spots with an outputenergy of �40 mJ and a repetition rate of 8 Hz. Eachanalysis consisted of one 20-s integration on the back-grounds (on peaks with no laser firing) and twenty 1-sintegrations on peaks with the laser firing. The depth ofeach ablation pit is �20 mm. The collector configurationallows simultaneous measurement of 204Pb in a secondaryelectron multiplier while 206Pb, 207Pb, 208Pb, 232Th, and238U are measured with Faraday detectors. All analyseswere conducted in static mode.[72] Correction for common Pb was done by measuring

206Pb/204Pb, with the composition of common Pb fromStacey and Kramers [1975] and uncertainties of 1.0 for206Pb/204Pb and 0.3 for 207Pb/204Pb. Fractionation of206Pb/238U and 206Pb/207Pb during ablation was monitoredby analyzing fragments of a large concordant zircon crystalthat has a known (isotope dilution (ID)-TIMS) age of 564 ±4 Ma (2s) (G. E. Gehrels, unpublished data, 2004). Typi-cally, this reference zircon was analyzed once for every fourunknowns. The uncertainty arising from this calibrationcorrection, combined with the uncertainty from decay con-stants and common Pb composition, contributed �1%systematic error to the 206Pb/238U and 206Pb/207Pb ages(2s level).[73] The reported ages are based on 206Pb/238U ratios

because the errors of the 207Pb/235U and 206Pb/207Pb ratios

are significantly greater (Table 1). This is due in large partto the low intensity (commonly �1 mV) of the 207Pb signalfrom these young grains. Analyses were conducted with theaid of CL images, which revealed strong compositionalvariation in most grains. Laser spots were located withincores and tips of these grains (as shown on Table 1) in aneffort to resolve both igneous and inheritance ages.[74] For each sample, the 206Pb/238U ages are shown on a

Pb/U Concordia diagram and on separate age plots (usingplotting program of Ludwig [2001]). The final age calcu-lations are based on the weighted mean of the cluster of206Pb/238Y ages, with the error expressed both as theuncertainty of this mean and as the error of the age. Theage error is based on the quadratic sum of the weightedmean error and the systematic error. Both ages are expressedat the 2s level.

[75] Acknowledgments. Funding to support this work was providedby the National Science Foundation (EAR-0087323 and EAR-0337111 toK.A.K. and EAR-0443387 to G.G.) and the Australian Research Council(A10009053, DP0342862).We thankA. Tulloch, I. Turnbull, and N.Mortimerfor many discussions and assistance and the Department of Land Conser-vation in Te Anau for permission to visit and sample localities in theFiordland National Park. W. C. Simonson and A. Claypool helped analyzestructural relationships in the Paparoa core complex and in Crooked Arm,respectively, as part of their M.Sc. theses at the University of Vermont.J. Fitzherbert provided valuable assistance in the field and with the modelingof metamorphic mineral assemblages. Art Goldstein and G. Rusty Riggprovided help with field work near Doubtful Sound. We thank ChrisAndronicos, Brad Hacker, Harold Stowell, and Bob Miller for helpfulreviews.


25 of 27

TC4017

ReferencesAbers, G. A., A. Ferris, C. Mitchell, D. Hugh, L. L.

Arthur, J. C. Mutter, and B. Taylor (2002), Mantlecompensation of active metamorphic core com-plexes at Woodlark rift in Papua New Guinea,Nature, 418, 862–865.

Afonso, J. C., and G. Ranalli (2004), Crustal and man-tle strengths in continental lithosphere; is the jellysandwich model obsolete?, Tectonophysics, 394,221–232.

Allibone, A., I. Turnbull, L. Milan, S. Carroll, andN. Daczko (2005), The granulite facies WesternFiordland Orthogneiss in southwest Fiordland, paperpresented at 50th Annual Meeting, Geol. Soc. ofN. Z., Kaikoura, 28 Nov. to 1 Dec.

Amato, J. M., E. L. Miller (2004), Geologic map andsummary of the evolution of the Kigluaik Moun-tains gneiss dome, Seward Peninsula, Alaska, inGneiss Domes in Orogeny, edited by D. L. Whitney,C. Teyssier, and C. S. Siddoway, Spec. Pap. Geol.Soc. Am., 380, 295–306.

Andronicos, C. L., D. H. Chardon, L. S. Hollister, G. E.Gehrels, and G. J. Woodsworth (2003), Strain par-titioning in an obliquely convergent orogen, pluton-ism, and synorogenic collapse: Coast MountainsBatholith, British Columbia, Canada, Tectonics,22(2), 1012, doi:10.1029/2001TC001312.

Antignano, A., IV (2002), Experimental constraints ongranitoid compositions in convergent regimes: ageochemical study, M. S. thesis, 135 pp., Univ. ofVt., Burlington, Dec.

Bailey, C. M., and E. L. Eyster (2003), General sheardeformation in the Pinaleno Mountain metamorphiccore complex, Arizona, J. Struct. Geol, 25, 1883–1892.

Bailey, C. M., et al. (2007), Pure shear dominatedhigh-strain zones in basement terranes, in The

Four-dimensional Framework of Continental

Crust, edited by R. D. Hatcher Jr. et al.,Mem. Geol. Soc. Am., in press.

Baldwin, J. A., S. A. Bowring, and M. L. Williams(2003), Petrological and geochronological con-straints on high pressure, high temperature meta-morphism in the Snowbird tectonic zone, Canada,J. Metamorph. Geol., 21, 81 – 98, doi:10.1046/j.1525-1314.2003.00413.x.

Bhattacharyya, P., and P. Hudleston (2001), Strain inductile shear zones in the Caledonides of northernSweden: a three dimensional puzzle, J. Struct.Geol., 23, 1549–1565.

Blattner, P. (1976), Replacement of hornblende by gar-net in granulite facies assemblages near MilfordSound, New Zealand, Contrib. Mineral. Petrol.,55, 181–190.

Bradshaw, J. D. (1989), Cretaceous geotectonic patternsin the New Zealand region, Tectonics, 8, 803–820.

Bradshaw, J. Y. (1985), Geology of the northern Frank-lin Mountains, northern Fiordland, New Zealand,with emphasis on the origin and evolution of Fiord-land granulites, Ph.D. thesis, Univ. of Otago, Dune-din, New Zealand.

Bradshaw, J. Y. (1989a), Origin and metamorphic his-tory of an Early Cretaceous polybaric granulite ter-rain, Fiordland, southwest New Zealand, Contrib.Mineral. Petrol., 103, 346–360.

Bradshaw, J. Y. (1989b), Early Cretaceous vein-relatedgarnet granulite in Fiordland, Southwest New Zeal-and: A case for infiltration of mantle-derived CO2-rich fluids, J. Geol., 97, 697–717.

Bradshaw, J. Y. (1990), Geology of crystalline rocks ofnorthern Fiordland; details of the granulite facieswestern Fiordland Orthogneiss and associated rockunits, N. Z. J. Geol. Geophys., 33, 465–484.

Brown, E. H. (1996), High-pressure metamorphismcaused by magma loading in Fiordland, New Zeal-and, J. Metamorph. Geol., 14, 441–452.

Buck, W. R. (1991), Modes of continental lithosphericextension, J. Geophys. Res., 96, 20,161–20,178.

Burov, E. B., and A. B. Watts (2006), The long-term strength of continental lithosphere: ‘‘Jellysandwich’’ or ‘‘creme brulee’’?, GSA Today, 16,4 –10.

Carson, C. J., R. Powell, and G. L. Clarke (1999),Calculated mineral equilibria for eclogite in Ca-Na2O-FeO-MgO-Al2O3-SiO2-H2O; application tothe Pouebo Terrane, Pam Peninsula, New Caledo-nia, J. Metamorph. Geol., 17, 9 – 24.

Carson, C. J., G. L. Clarke, and R. Powell (2000),Hydration of eclogite from the Pam Peninsula,New Caledonia, J. Metamorph. Geol., 18, 79– 90.

Choukroune, P., D. Gapais, and O. Merle (1987), Shearcriteria and structural symmetry, J. Struct. Geol., 9,525 –530.

Christensen, N. I., and D. M. Fountain (1975), Consti-tution of the lower continental crust based onexperimental studies of seismic velocities ingranulite, Geol. Soc. Am. Bull., 86, 227–236.

Christensen, N. I., and W. D. Mooney (1995), Seismicvelocity structure and the composition of the con-tinental crust: A global view, J. Geophys. Res., 100,9761–9788.

Clark, M. K., and L. H. Royden (2000), Topographicooze: Building the eastern margin of Tibet by lowercrustal flow, Geology, 28, 703–706.

Clarke, G. L., K. A. Klepeis, and N. R. Daczko (2000),Cretaceous high-P granulites at Milford Sound,New Zealand: metamorphic history and emplace-ment in a convergent margin setting, J. Metamorph.Geol., 18, 359–374.


26 of 27

TC4017

Clarke, G. L., N. R. Daczko, K. A. Klepeis, andT. Rushmer (2005), Roles for fluid and/or melt advec-tion in forming high-P mafic migmatites, Fiordland,New Zealand, J. Metamorph. Geol., 23, 557–567.

Coney, P. J., and T. A. Harms (1984), Cordilleranmetamorphic core complexes: Cenozoic exten-sional relics of Mesozoic compression, Geology,12, 550–554.

Daczko, N. R., K. A. Klepeis, and G. L. Clarke(2001a), Evidence of Early Cretaceous collisional-style orogenesis in northern Fiordland, New Zeal-and and its effects on the evolution of the lowercrust, J. Struct. Geol., 23, 693–713.

Daczko, N. R., G. L. Clarke, and K. A. Klepeis(2001b), Transformation of two-pyroxene hornble-nde granulite to garnet granulite involving simulta-neous melting and fracturing of the lower crust,Fiordland, New Zealand, J. Metamorph. Geol.,19, 547–560.

Daczko, N. R., K. A. Klepeis, and G. L. Clarke (2002),Thermomechanical evolution of the crust duringconvergence and deep crustal pluton emplacementin the Western Province of Fiordland, New Zealand,Tectonics, 21(4), 1022, doi:10.1029/2001TC001282.

Daczko, N. R., J. A. Stevenson, G. L. Clarke, and K. A.Klepeis (2002b), Successive hydration and dehy-dration of high-P mafic granofels involving clino-pyroxene-kyanite symplectites, Mt. Daniel,Fiordland, New Zealand, J. Metamorph. Geol.,20, 669–682.

Daczko, N., K. A. Klepeis, and G. L. Clarke (2002c),Kyanite-paragonite-bearing assemblages, northernFiordland, New Zealand: Rapid cooling of the lowercrustal root to a Cretaceous magmatic arc, J. Meta-morph. Geol., 20, 887–902.

Dale, J., R. Powell, R. W. White, F. L. Elmer, and T. J.B. Holland (2005), A thermodynamic model forCa-Na clinoamphiboles in Na2O-CaO-FeO-MgO-Al2O3-SiO2-H2O-O for petrological calculations,J. Metamorph. Geol., 25, 771–791.

Davids, C. (1999), A thermochronological study ofsouthern Fiordland, New Zealand, Ph.D. thesis,211 pp., Aust. Natl. Univ., Canberra.

De Paoli, M. (2005), The mechanism of corona forma-tion under high-P granulite to high- P amphiboliteconditions, Western Fiordland Orthogneiss, NewZealand, B.Sc. (Honors) thesis, Univ. of Sydney,Sydney, Australia.

De Paor, D. G. (1988), Rf/f strain analysis using anorientation net, J. Struct. Geol., 10, 323 –333.

Flowers, R. M., S. A. Bowring, A. J. Tulloch, and K. A.Klepeis (2005), Tempo of burial and exhumationwithin the deep roots of a magmatic arc, Fiordland,New Zealand, Geology, 33, 17– 20.

Friedman, R. M., and R. L. Armstrong (1988), TatlaLake metamorphic complex; an Eocene meta-morphic core complex on the southwestern edgeof the intermontane belt of British Columbia, Tec-tonics, 7, 1141 –1166.

Gaina, C., D. R. Muller, J. Y. Royer, J. Stock, J. Hardebeck,and P. Symonds (1998), The tectonic history of the Tas-man Sea: A puzzle with 13 pieces, J. Geophys. Res.,103, 12,413–12,433.

Gans, P. B. (1987), An open system, two-layer crustalstretching model for the eastern Great Basin, Tec-tonics, 6, 1 –12.

Gao, S., T.-C. Luo, B.-R. Zhang, H.-F. Zhang, Y.-W.Han, Y.-K. Hu, and Z.-D. Zhao (1998), Chemicalcomposition of the continental crust as revealed bystudies in east China, Geochim. Cosmochim. Acta,62, 1959–1975.

Gapais, D., P. Bale, P. Choukroune, P. R. Cobbold,Y. Mahjoub, and D. Marquer (1987), Bulk kine-matics from shear zone patterns: Some field exam-ples, J. Struct. Geol., 9, 635–646.

Ghosh, S. K., G. Sen, and S. Sengupta (2003), Rotationof ling tectonic clasts in transpressional shear zones,J. Struct. Geol., 25, 1083–1096.

Gibson, G. M. (1990), Uplift and exhumation of middle& lower crustal rocks in an extensional tectonicsetting, Fiordland, New Zealand, in Exposed

Cross-Sections of the Continental Crust, NATOASI Ser., Ser. C, vol. 317, edited by D. M. Fountainand M. H. Salisbury, pp. 71–101, Kluwer Acad.,Dorchecht, Netherlands.

Gibson, G. M., and T. R. Ireland (1995), Granulite for-mation during continental extension in Fiordland,Nature, 375, 479–482.

Gibson, G. M., and T. R. Ireland (1996), Extension ofDelamarian (Ross) Orogen into western New Zeal-and: evidence from zircon ages and implications forcrustal growth along the Pacific margin of Gond-wana, Geology, 24, 1087–1090.

Gibson, G. M., and T. R. Ireland (1998), SHRIMPmonazite and zircon geochronology of high-grademetamorphism in New Zealand, J. Metamorph.Geol., 16, 49 –167.

Gibson, G. M., I. McDougall, and T. R. Ireland (1988),Age constraints on metamorphism and the develop-ment of a metamorphic core complex in Fiordland,southern New Zealand, Geology, 16, 405–408.

Giorgis, S., M. Markley, B. Tikoff (2004), Vertical-axisrotation of rigid crustal blocks driven by mantleflow, in Vertical Coupling and Decoupling in theLithosphere, edited by J. Grocott et al., Geol. Soc.Spec. Publ., 227, 83– 100.

Guiraud,M., R. Powell, and J. Y. Cottin (1996), Hydrationof orthopyroxene-cordierite-bearing assemblages atLaouni, Central Hoggar, Algeria, J. Metamorph.

Geol., 14, 467–476.Handy, M. R., and J.-P. Brun (2004), Seismicity, struc-

ture and strength of the continental lithosphere,Earth Planet. Sci. Lett., 223, 427–441.

Holland, T. J. B., and R. Powell (1998), An internallyconsistent thermodynamic dataset for phases of petro-logical interest, J. Metamorph. Geol., 16, 309–343.

Hollis, J. A., G. L. Clarke, K. A. Klepeis, N. R. Daczko,and T. R. Ireland (2003), Geochronology and Geo-chemistry of high-pressure granulites of the ArthurRiver Complex, Fiordland, New Zealand: CretaceousMagmatism andMetamorphism on the Palaeo-PacificMargin, J. Metamorph. Geol., 21, 299–313.

Hollis, J. A., G. L. Clarke, K. A. Klepeis, N. R. Daczko,and T. R. Ireland (2004), The regional significance ofCretaceous magmatism and metamorphism in Fiord-land, New Zealand, from U–Pb zircon geochronol-ogy, J. Metamorph. Geol., 22, 607–627.

Hopper, J. R., and W. R. Buck (1996), The effect oflower crustal flow on continental extension and pas-sive margin formation, J. of Geophys. Res., 101,20,175–20,194.

Hopper, J. R., and W. R. Buck (1998), Styles of exten-sional decoupling, Geology, 26, 699–702.

Huismans, R. S., S. J. H. Buiter, and C. Beaumont(2005), Effect of plastic-viscous layering and strainsoftening onmode selection during lithospheric exten-sion, J. Geophys. Res., 110, B02406, doi:10.1029/2004JB003114.

Ireland, T. R., and G. M. Gibson (1998), SHRIMPmonazite and zircon geochronology of high-grademetamorphism in New Zealand, J. Metamorph.

Geol., 16, 149–167.Jackson, J. (2002), Strength of the continental litho-

sphere; time to abandon the jelly sandwich?, GSAToday, 12, 4 – 10.

Jamieson, R. A., C. Beaumont, M. H. Nguyen, andB. Lee (2002), Interaction of metamorphism,deformation and exhumation in large convergentorogens, J. Metamorph. Geol., 20, 9 –24.

Jiang, D., and P. F. Williams (1998), Transpression(or transtension) zones of triclinic symmetry: nat-ural example and theoretical modelling, in Con-

tinental Transpressional and TranstensionalTectonics, edited by R. E. Holdsworth, R. A.Strachan, and J. F. Dewey, Geol. Soc. Spec.

Publ., 135, 41– 57.Jolivet, L., V. Famin, C. Mehl, T. Parra, C. Aubourg,

R. Hebert, and P. Philippot (2004), Strain localiza-tion during crustal-scale boudinage to form exten-sional metamorphic domes in the Augean Sea, inGneiss Domes in Orogeny, edited by D. L. Whitney,C. Teyssier, and C. S. Siddoway, Spec. Pap. Geol.Soc. Am., 380, 185–210.

Karlstrom, K. E., and M. L. Williams (1998), Hetero-geneity of the middle crust: Implications forstrength of continental lithosphere, Geology, 26,815 –818.

Karlstrom, K. E., and M. L. Williams (2006), Natureand evolution of the middle crust: heterogeneity ofstructure and process due to pluton-enhanced tec-tonism, inEvolution and Differentiation of the Con-tinental Crust, edited by M. Brown and T. Rushmer,pp. 268 –295, Cambridge Univ. Press, New York.

King, D. S. (2006), Shear zone processes in the mid tolower crust and the structural evolution of centralFiordland, New Zealand, M. S. thesis, 150 pp.,Univ. of Vt., Burlington, Feb.

King, D., K. Klepeis, G. Gehrels, and A. Goldstein(2005), Lithologic heterogeneity, mechanical aniso-tropy and the formation of deep crustal shear zonesin Fiordland, New Zealand, Geol. Soc. Am. Abstr.Programs, 37, 60.

Klepeis, K. A., N. R. Daczko, and G. L. Clarke (1999),Kinematic vorticity and tectonic significance ofsuperposed mylonites in a major lower crustal shearzone, northern Fiordland, New Zealand, J. Struct.Geol., 21, 1385–1405.

Klepeis, K. A., G. L. Clarke, and T. Rushmer (2003),Magma transport and coupling between deforma-tion and magmatism in the continental lithosphere,GSA Today, 13, 4 –11.

Klepeis, K. A., G. L. Clarke, G. Gehrels, and J. Vervoort(2004), Processes controlling vertical coupling anddecoupling between the upper and lower crust oforogens: results from Fiordland, New Zealand,J. Struct. Geol., 26, 765–791.

Kohlstedt, D. L., B. Evans, and S. J. Mackwell (1995),Strength of the lithosphere: Constraints imposed bylaboratory experiments, J. Geophys. Res., 100,17,587–17,602.

Kula, J. L., A. J. Tulloch, T. L. Spell, and M. L. Wells(2005), Timing of continental extension leading toseparation of eastern New Zealand from West Ant-arctica; 40Ar/39Ar thermochronometry from Stew-art Island, NZ, Geol. Soc. Am. Abstr. Programs,37, 73.

Law, R. D., M. P. Searle, and R. L. Simpson (2004),Strain, deformation temperatures and vorticity offlow at the top of the Greater Himalayan Slab, Ever-est Massif, Tibet, J. Geol. Soc. London, 161, 305–320.

Lisle, R. J. (1985), Geological Strain Analysis—A Man-

ual for the Rf/f technique, 99pp., Pergamon, Oxford,U. K.

Lister, G. S., and P. F. Williams (1983), The partitioningof deformation in flowing rock masses, Tectonophy-sics, 49, 37 –78.

Ludwig, K. J. (2001), Isoplot/Ex (rev. 2.49), Spec. Publ.1a,56pp., Berkeley Geochronol. Cent., Berkeley,Calif.

Maggi, A., J. A. Jackson, D. McKenzie, and K. Priestly(2000), Earthquake focal depths, effective elasticthickness, and the strength of the continental litho-sphere, Geology, 28, 495 –498.

Marcotte, S. B., K. A. Klepeis, G. L. Clarke, G. Gehrels,and J. A. Hollis (2005), Intra-arc transpression in thelower crust and its relationship to magmatism in aMesozoic magmatic arc, Tectonophysics, 407, 135–163.

Martınez, F., A. M. Goodliffe, and B. Taylor (2001),Metamorphic core complex formation by densityinversion and lower-crust extension, Nature, 411,930 –933.

Mattinson, J. L., D. L. Kimbrough, and J. Y. Bradshaw(1986), Western Fiordland orthogneiss: Early Cre-taceous arc magmatism and granulite facies meta-morphism, New Zealand, Contrib. Mineral. Petrol.,92, 383–392.

McKenzie, D., and J. Jackson (2002), Conditions forflow in the continental crust, Tectonics, 21(6), 1055,doi:10.1029/2002TC001394.

McKenzie, D., F. Nimmo, and J. A. Jackson (2000),Characteristics and consequences of flow in thelower crust, J. Geophys. Res., 105, 11,029–11,046.


27 of 27

TC4017

Milan, L. A., N. R. Daczko, I. Turnbull, and A. Allibone(2005), Thermobarometry of an Early Cretaceoushigh-pressure contact metamorphic aureole near Re-solution Island, Fiordland, New Zealand, Geol. Soc.Aust. Abstr., 76, 79–85.

Miller, R. B., and S. R. Paterson (2001), Influence oflithological heterogeneity, mechanical anisotropy,and magmatism on the rheology of an arc, North Cas-cades, Washington, Tectonophysics, 342, 351–370.

Mooney, W. D., G. Laske, and T. G. Masters (1998),CRUST 5.1: A global crustal model at 5� � 5�,J. Geophys. Res., 103, 727–748.

Muir, R. J., T. R. Ireland, S. D. Weaver, J. D. Bradshaw,T. E. Waight, R. Jongens, and G. N. Eby (1997),SHRIMP U-Pb geochronology of Cretaceous mag-matism in northwest Nelson-Westland, SouthIsland, New Zealand, N. Z. J. Geol. Geophys., 40,453–463.

Nagel, T. J., and W. R. Buck (2004), Symmetric alter-native to asymmetric rifting models, Geology, 32,937–940.

Oliver, G. J. H. (1977), Feldspathic hornblende andgarnet granulites and associated anorthosite pegma-tites from Doubtful Sound, Fiordland, New Zeal-and, Contrib. Mineral. Petrol., 65, 111 –121.

Oliver, G. J. H. (1980), Geology of the granulite andamphibolite facies gneisses of Doubtful Sound,Fiordland, New Zealand, N. Z. J. Geol. Geophys.,1, 27– 41.

Oliver, G. J. H. (1990), An exposed cross-section ofcontinental crust, Doubtful Sound, Fiordland, NewZealand: Geophysical and geological setting, inExposed Cross-Sections of the Continental Crust,NATOASI Ser., Ser. C, 317, edited byD.M. Fountainand M. H. Salisbury, pp. 43 – 69, Kluwer Acad.,Dorchecht, Netherlands.

Oliver, G. J. H., and J. H. Coggon (1979), CrustalStruc-ture of Fiordland, New Zealand, Tectonophysics, 54,253–292.

Olsen, S. V., andD. L. Kohlstedt (1985), Natural deforma-tion and recrystallization of some intermediate plagi-oclase feldspars, Tectonophysics, 111, 107–131.

Owens, W. H. (1984), The calculation of a best-fit ellip-soid from elliptical sections on arbitrarily orientedplanes, J. Struct. Geol., 6, 571–578.

Passchier, C. W. (1987), Stable positions of rigidobjects in noncoaxial flow, a study in vorticityanalysis, J. Struct. Geol., 9, 679–690.

Passchier, C. W., and J. L. Urai (1988), Vorticity andstrain analysis using Mohr diagrams, J. Struct.Geol., 10, 755–763.

Pattison, D. R. M. (2003), Petrogenetic significance oforthopyroxene-free garnet + clinopyroxene + plagi-oclase ± quartz-bearing metabasites with respect tothe amphibolite and granulite facies, J. Metamorph.Geol., 21, 21– 34.

Powell, R., T. Holland, and B. Worley (1998), Calcu-lating phase diagrams involving solid solutions vianon-linear equations, with examples using THER-MOCALC, J. Metamorph. Geol., 16, 577–588.

Pryer, L. L. (1993), Microstructures in feldspars from amajor crustal thrust zone: the Grenville Front, Ontario,Canada, J. Struct. Geol., 15, 21–36.

Ramsey, J. G., and M. I. Huber (1983), The Techniquesof Modern Structural Geology, vol. 1, Strain Ana-lysis, 307 pp., Academic, New York.

Robin, P.-Y. F. (1977), rmination of geologic strainusing randomly oriented strain markers of anyshape, Tectonophysics, 42, T7–T16.

Robin, P.-Y. F., and A. R. Cruden (1994), Strain andvorticity patterns in ideally ductile transpressionalzones, J. Struct. Geol., 16, 447–466.

Royden, L. (1996), Coupling and decoupling of crustand mantle in convergent orogens: Implications forstrain partitioning in the crust, J. Geophys. Res.,101, 17,679–17,705.

Rudnick, R. L., and D. M. Fountain (1995), Nature andcomposition of the continental crust: A lower crus-tal perspective, Rev. Geophys., 33, 267–309.

Rudnick, R. L., and S. Gao (2003), The composition ofthe continental crust, in The Crust, vol. 3, Treatiseon Geochemistry, edited by H. D. Holland and K. K.Turekian, pp. 1–64, Elsevier-Pergamon,Oxford, U.K.

Scott, J. M. (2004), The Mt Irene Shear Zone: Meta-morphism, magmatism and structure in the Murch-ison Mountains, Fiordland, B.Sc. (Honors) thesis,130pp., Univ. of Otago, Dunedin, New Zealand.

Scott, J. M., and A. F. Cooper (2006), Early Cretaceousextensional exhumation of the lower crust of a mag-matic arc: Evidence from the Mount Irene ShearZone, Fiordland, New Zealand, Tectonics, 25,TC3018, doi:10.1029/2005TC001890.

Siddoway, C. S., S. M. Richard, C. M. Fanning, B. P.Luyendyk (2004), Origin and emplacement of amiddle Cretaceous gneiss dome, FosdickMountains,West Antarctica, inGneiss Domes in Orogeny, editedby D. L. Whitney, C. Teyssier and C. S. Siddoway,Spec. Pap. Geol. Soc. Am., 380, 267–294.

Simpson, C., and D. G. De Paor (1993), Strain andkinematic analysis in general shear zones, J. Struct.Geol., 15, 1 –20.

Simpson, C., and De D. G. Paor (1997), Practical ana-lysis of general shear zones using porphyroclasthyperbolic distribution method: an example fromthe Scandinavian Caledonides, in Evolution of Geo-logical Structures in Micro- to Macro-scales, editedby S. Sengupta, pp. 169–184, Chapman and Hall,London.

Spell, T. L., I. McDougall, and A. J. Tulloch (2000),Thermochronologic constraints on the breakup ofthe Pacific Gondwana margin: The Paparoa meta-morphic core complex, South Island, New Zealand,Tectonics, 19, 433–451.

Stacey, J. S., and J. D. Kramers (1975), Approximationof terrestrial lead isotope evolution by a two-stagemodel, Earth Planet. Sci. Lett., 26, 207–221.

Sutherland, R., F. Davey, and J. Beavan (2000), Plateboundary deformation in South Island, New Zeal-and is related to inherited lithospheric structure,Earth Planet. Sci. Lett., 177, 141–151.

Tikoff, B., and H. Fossen (1995), The limitations ofthree-dimensional kinematic vorticity analysis,J. Struct. Geol., 17, 1771–1784.

Tikoff, B., C. Teyssier, and C. Water (2002), Clutchtectonics and the partial attachment of lithosphericlayers, Eur. Geophys. Soc. Spec. Publ., 1, 57 –73.

Tikoff, B., R. Russo, C. Teyssier, A. Tommasi (2004),Mantle-driven deformation of orogenic zones andclutch tectonics, in Vertical Coupling and Decou-

pling in the Lithosphere, edited by J. Grocott et al.,Geol. Soc. Special Publ. , 227, 41 –64.

Tullis, J., H. Stunitz, C. Teyssier, R. Heilbronner (2000),Deformation microstructures in quartzo-feldspathicrocks, in Stress, Strain and Structure: A Volume inHonour of W. D. Means, edited by M. W. Jessel andJ. L. Urai, J. Virtual Explor., 2. (Available at http://www.virtualexplorer.com.au/special/meansvolume/contribs/tullis/SlideSet/deformation_micro.html)

Tulloch, A. J., and G. A. Challis (2000), Emplacementdepths of Paleozoic-Mesozoic plutons from westernNew Zealand estimated by hornblende-Al geobaro-metry, N. Z. J. Geol. Geophys., 43, 555–567.

Tulloch, A. J., and D. L. Kimbrough (1989), ThePaparoa metamorphic core complex, New Zealand:

Cretaceous extension associated with fragmentationof the Pacific margin of Gondwana, Tectonics, 8,1217–1235.

Tulloch, A. J., D. L. Kimbrough (2003), Paired plutonicbelts in convergent margins and the development ofhigh Sr/Y magmatism: The Peninsular RangesBatholith of California and the Median Batholithof New Zealand, in Tectonic Evolution of North-western Mexico and the Southwestern USA, editedby S. E. Johnson et al., Spec. Pap. Geol. Soc. Am.,374, 275–295.

Turcotte, D. L., and G. Schubert (2002), Geodynamics,2nd ed., 472pp., Cambridge Univ. Press, Cam-bridge, U. K.

Turnbull, I., A. Allibone, R. Jongens, H. Fraser, andA. Tulloch (2005), Progress on QMAP Fiord-land, paper presented at 50th Annual Meeting,Geol. Soc. of N. Z., 0Kaikoura, New Zealand,28 Nov. to 1 Dec.

Vanderhaeghe, O. (2004), Structural development ofthe Naxos migmatite dome, in Gneiss Domes in

Orogeny, edited by D. L. Whitney, C. Teyssier,and C. S. Siddoway, Spec. Pap. Geol. Soc. Am.,380, 211–227.

Vanderhaeghe, O., S. Medvedev, P. Fullsack, C. Beaumont,and R. A. Jamieson (2003), Evolution of orogenicwedges and continental plateaux: Insights fromcrustal thermal-mechanical models overlying sub-ducting mantle lithosphere, Geophys. J. Int., 153,27 –51.

Waight, T. E., S. D. Weaver, and R. J. Muir (1998),Mid-Cretaceous granitic magmatism during thetransition from subduction to extension in southernNew Zealand: A chemical and tectonic synthesis,Lithos, 45, 469–482.

Wernicke, B. (1992), Cenozoic extensional tectonics ofthe U. S. Cordillera. in The Geology of North Amer-ica: DNAG, Decade of North American Geology,edited by B. C. Burchfiel, P. W. Lipman, andM. L. Zoback, pp. 553–581, Geol. Soc. of Am.,Boulder, Colo.

Westaway, R. (1998), Dependence of active normalfault dips on lower-crustal flow regimes, J. Geol.Soc., 155, 233–253.

White, P. J. (1994), Thermobarometry of the Charlestonmetamorphic group and implications for the evolu-tion of the Paparoa metamorphic core complex,New Zealand, N. Z. J. Geol. Geophys., 37, 201–209.

White, R. W., and R. Powell (2002), Melt loss and thepreservation of granulite facies mineral assem-blages, J. Metamorph. Geol., 20, 621–632.

Wijns, C., R. Weinberg, K. Gessner, and L. Moresi(2005), Mode of crustal extension determined byrheological layering, Earth Planet. Sci. Lett., 236,120 –134.

Wood, B. L. (1972), Metamorphosed ultramafites andassociated formations near Milford Sound, NewZealand, N. Z. J. Geol. Geophys., 15, 88– 127.

��G. L. Clarke and M. De Paoli, School of Geos-

ciences, F09, University of Sydney, NSW 2006,Australia.

G. Gehrels, Department of Geosciences, Universityof Arizona, Tucson, AZ 85721, USA.

D. King and K. A. Klepeis, Department of Geology,University of Vermont, Burlington, VT 05405-0122,USA. ([email protected])

Date post:	26-Jun-2020
Category:	Documents
Upload:	others
View:	3 times
Download:	0 times

Interaction of strong lower and weak middle crust...

Documents