Transform Fault and Its Effect on Ophiolite Obduction, New Caledonia

RESEARCH

Fabric development in the mantle section of a paleotransform fault and its effect on ophiolite obduction, New Caledonia

Sarah J. Titus1*, Stephanie M. Maes2*, Bryn Benford3*, Eric C. Ferré4*, and Basil Tikoff3* 1DEPARTMENT OF GEOLOGY, CARLETON COLLEGE, ONE NORTH COLLEGE STREET, NORTHFIELD, MINNESOTA 55057, USA 2DEPARTMENT OF PHYSICAL AND BIOLOGICAL SCIENCES, THE COLLEGE OF SAINT ROSE, 432 WESTERN AVENUE, ALBANY, NEW YORK 12203, USA 3DEPARTMENT OF GEOSCIENCE, UNIVERSITY OF WISCONSIN–MADISON, 1215 W. DAYTON STREET, MADISON, WISCONSIN 53706, USA 4DEPARTMENT OF GEOLOGY, SOUTHERN ILLINOIS UNIVERSITY, CARBONDALE, ILLINOIS 62901, USA

ABSTRACT

The Bogota Peninsula shear zone has been interpreted as a paleotransform fault in the mantle section of the New Caledonia ophiolite. New, detailed fi eld measurements document the rotation of foliation, lineation, and pyroxenite dikes across a 50-km-wide region. Deformation intensity recorded by folding and boudinage of dikes increases toward a central, 3-km-wide mylonitic zone. We used several additional methods to characterize fabric patterns across the shear zone. The shape-preferred orientation of orthopyroxene grains, computed from outcrop tracings, closely parallels fi eld fabrics, with increased alignment and fl attening near the center of the shear zone. The lattice-preferred orientations of olivine are consistent with high-temperature fabrics; the a axes within the mylonitic core were used to constrain the orientation of shear zone boundaries. Seismic anisotropy calculations, based on the lattice-preferred orientation of olivine, indicate 5%–11% shear-wave anisotropies, with increased values in the center of the shear zone. The magnetic silicate fabric in the rocks, determined from anisotropy of magnetic susceptibility techniques, broadly matches fi eld fabrics but provides less consistent information across the shear zone than other fabric methods. This suite of fi eld and laboratory data provides a unique and detailed view of strain and fabric patterns across a shear zone in oceanic mantle lithosphere. Because the primary mantle fabrics seem to be related to the present distribution of ophiolitic rocks in New Caledonia, we propose that ophiolite obduction and Neogene extension may have been controlled by preexisting fabrics and structures in the oceanic lithosphere.

LITHOSPHERE; v. 3; no. 3; p. 221–244. doi: 10.1130/L122.1

INTRODUCTION patterns preserved in deeper lithospheric levels. within peridotites from the mantle section of Well-exposed transform faults, not just shear the ophiolite. We used several ﬁ eld and petro- Studies of modern oceanic transform faults zones, are relatively rare in ophiolites. The best- fabric techniques to characterize strain patterns and fracture zones have provided an increasingly known examples include the Arakapas fault across the shear zone and to develop a concep- detailed picture of the topography, geometry, zone in the crustal section of the Troodos ophi- tual model for its evolution during progressive and evolution of these important plate bound- olite in Cyprus, where sheeted dikes are rotated deformation. We then linked these strain pat- ary structures (e.g., Menard and Atwater, 1969; 90° into alignment with the fault zone (e.g., terns to ophiolite obduction and subsequent Bonatti, 1978; Choukroune et al., 1978; Fox Moores and Vine, 1971; Simonian and Gass, extension, suggesting that primary mantle fab- and Gallo, 1984; Lawson et al., 1996). How- 1978; MacLeod et al., 1990), and the Coastal rics and structures may have controlled, at least ever, actual in situ observations are difﬁ cult, complex in the Bay of Islands complex, New- in part, these major tectonic events. and, although steep fracture zone walls provide foundland, which has been interpreted as a frac- some insight into deeper rocks and structures ture zone or transform fault that may have facili- TECTONIC HISTORY (e.g., Fox et al., 1976; Prinz et al., 1976), it is tated ophiolite obduction (Karson and Dewey, impractical to study structures developed across 1978; Casey et al., 1983; Suhr and Cawood, The island of New Caledonia represents the the entire transform system at anything but the 2001). In both of these examples, most of the exposed portion of the Norfolk Ridge (Fig. 1), shallowest lithospheric depths. ophiolitic material is located on one side of the a microcontinental ribbon that rifted from the Where oceanic transform faults are pre- inferred plate boundary, providing a glimpse of eastern Gondwanaland margin during the Late served on land, such as within ophiolites, they half of the transform system deformation. Cretaceous (e.g., Dubois et al., 1974; Crawford can be investigated by more direct methods and The Bogota Peninsula shear zone in the et al., 2003). This rifted Mesozoic-age material provide invaluable insight about deformation New Caledonia ophiolite also has been inter- forms the basement in New Caledonia (Aitchi- preted as a paleotransform fault (Prinzhofer son et al., 1995; Cluzel et al., 2001). Younger *E-mails: [email protected]; [email protected] and Nicolas, 1980; Nicolas, 1989). Unlike geologic structures on the island developed in .edu; [email protected]; [email protected]; the Arakapas fault and the Coastal complex, two phases of postrifting activity, which are [email protected]. however, this shear zone is exposed entirely summarized next.

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/3/3/221/3050183/221.pdf by guest on 01 October 2021 TITUS ET AL.

Coral Sea d’Entrecasteaux Ridge Vanuatu foliation lineation S. Loyalty Basin Fiji Fig. 2A N. Loyalty Belep NewNew Basin CaledoniaCaledonia Norfolk Ridge

Lord Howe Ri Australia a e c

N = 302 N = 107 O

c 20°S i f i c a P dge foliation

Tasman Sea New N = 616 Zealand

Koumac Continental Continental Volcanic Oceanic Active Fossil crust shelf arc crust crust subduction subduction lineation foliation lineation N = 187 21°S

Bogota 100 km Peninsula Fig. 2C N = 70 N = 58

Detachment fault WCFZ Fold hinges in/near Pouebo terrane Foliation with dip direction Massif du Sud foliation lineation Lineation 22°S Ophiolite nappe Pouebo terrane (high pressure) Poya terrane (basalts) N = 375 N = 325 Noumea Undifferentiated basement rocks

163°E 164°E 165°E 166°E 167°E

Figure 1. Simplifi ed geologic map of New Caledonia; inset shows a basic map of the SW Pacifi c near New Caledonia. Maps are modifi ed from Cluzel et al. (2001), with ophiolite fabric trajectories from Prinzhofer and Nicolas (1980) and fold hinges in the Pouebo terrane from Rawling and Lister (2002). Lower-hemisphere, equal-area projections for foliation and lineation from select portions of the ophiolite illustrate the fabric inside and outside of pos- tulated transform faults. Contours for these projections are in percent area; the contour intervals vary (for details, see Nicolas, 1989). WCFZ—western Caledonia fault zone.

Eocene Convergence During convergence, several thrust sheets crust and associated sediments (Spandler et al., were emplaced onto New Caledonia, includ- 2004, 2005), making it a likely metamorphic Based on dike ages in New Caledonia and ing the basaltic Poya terrane along the south- equivalent of the Poya terrane (Cluzel et al., plate reconstructions of the southwest Paciﬁ c, ern coast, the high-pressure Pouebo terrane in 2001; Whattam et al., 2008). It is difﬁ cult to a phase of convergence affecting New Caledo- the northeast, and the ophiolite nappe (Fig. 1). determine the age of the Pouebo terrane pro- nia began at 53–55 Ma (Crawford et al., 2003; The Poya terrane is interpreted as allochtho- tolith, but peak metamorphism in these rocks Cluzel et al., 2006). Most tectonic models sug- nous material, such as seamounts, scraped off occurred at ca. 44 Ma, followed by rapid cool- gest that the South Loyalty Basin was subducted the downgoing slab during subduction and ing and exhumation of a coherent block from beneath the Loyalty arc during this time (Eissen accreted to the forearc before emplacement 40 to 34 Ma (Baldwin et al., 2007). The ophi- et al., 1998; Cluzel et al., 2001; Crawford et al., onto the Norfolk Ridge (Cluzel et al., 2001). olite nappe is predominantly composed of 2003; Spandler et al., 2005). This northeast- Radiolarians within the Poya terrane are typi- mantle material and presently drapes into the facing convergence eventually ceased when the cally Late Cretaceous to Paleocene in age (Clu- South Loyalty Basin from the Belep Islands Norfolk Ridge reached the subduction zone at zel et al., 2001). The Pouebo terrane represents to the southern part of New Caledonia, based ca. 37 Ma (Cluzel et al., 2006). the subduction and metamorphism of oceanic on gravity data (Collot et al., 1987, 1988). The

222 www.gsapubs.org | Volume 3 | Number 3 | LITHOSPHERE

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/3/3/221/3050183/221.pdf by guest on 01 October 2021 Fabric development in the mantle section of a paleotransform fault, New Caledonia | RESEARCH

ophiolite is considered either Late Cretaceous Massif du Sud Previous Tectonic Interpretation of (e.g., Prinzhofer, 1981) or Eocene (What- Structures tam et al., 2008) in age, and was the last to be The Massif du Sud has a well-developed emplaced onto New Caledonia between 37 and mantle section that is ~1–3 km thick, is composed Both the Bogota Peninsula and Belep shear 34 Ma (Cluzel et al., 2006). primarily of harzburgite and dunite, and has a few zones have been interpreted as paleotransform layered gabbros and mafi c lavas (Avias, 1967; faults by previous workers and therefore have Neogene Extension Prinzhofer et al., 1980). The contact between been used to infer ridge-transform geometries for the mantle and crustal sections of the ophiolite is rocks in the Massif du Sud and ophiolite klippen. Regional extension, which modifi ed both observed at a few locations, where it is subhor- We describe the data and reasoning for each inter- onshore (Lagabrielle et al., 2005; Chardon and izontal (Prinzhofer et al., 1980). Regional map- pretation next and attempt to place each shear Chevillotte, 2006; Lagabrielle and Chauvet, ping within the mantle section (Prinzhofer and zone into its original tectonic context (Fig. 2). 2008) and offshore (Dupont et al., 1995; Lafoy Nicolas, 1980; Prinzhofer et al., 1980) demon- For the Bogota Peninsula shear zone, sev- et al., 1996; Auzende et al., 2000; Chardon et strates generally E-W– to NW-SE–striking folia- eral fi eld observations are important: (1) A al., 2008) structures, followed Eocene con- tions with <30° southerly dips and subhorizontal strain gradient is observed on both sides of the vergence and has lasted until the present day. N-S–trending lineations (Fig. 1). Bogota Peninsula toward a central mylonitic Based on a study of small-scale faults through- On the Bogota Peninsula along the northeast zone (Fig. 2B), showing consistent dextral shear out the island, Lagabrielle et al. (2005) sug- coast, foliation becomes subvertical and N-S– sense everywhere, (2) foliation in the shear zone gested that there have been two phases in this striking, while lineation remains subhorizontal is perpendicular to that in the Massif du Sud extensional history. and N-S–trending (Fig. 2). Fabrics are rotated (Fig. 1), and (3) lineation remains consistently The fi rst phase, beginning in the Oligo- over a 50-km-wide region around a central N-S–trending and subhorizontal inside and out- cene, had extension directions from 90° to 3-km-wide high-strain zone. Fabric strength, as side the shear zone (Fig. 1). Based on these data, 140° and primarily affected peridotites in the defi ned by both the lattice-preferred orientation Prinzhofer and Nicolas (1980) interpreted the ophiolite nappe. The western Caledonian fault (LPO) of olivine and stretched orthopyroxene Bogota Peninsula shear zone as a N-S–striking, zone (known as the Sillon in older literature), grains with aspect ratios up to 25:1, increases dextral paleotransform fault. which forms the abrupt southwestern boundary toward the center of the shear zone (Prinzhofer Nicolas (1989) used these same fi eld data of the Massif du Sud, is interpreted as a large and Nicolas, 1980). Shear sense indicators are to suggest that the Massif du Sud formed along detachment fault that facilitated extension dur- consistently dextral across the Bogota Peninsula the south fl ank of an E-W–striking ridge (rela- ing this phase (Fig. 1; Lagabrielle and Chauvet, and surrounding coastline. We refer to the entire tive to present geographic coordinates). The 2008). This phase of postorogenic collapse region of rotated fabrics as the Bogota Peninsula E-W–strike is based on (1) the assumption that likely facilitated unroofi ng of the high-pressure shear zone. the N-S–trending lineations record relative plate Pouebo terrane (Rawling and Lister, 2002) and motion and (2) that a ridge should be perpendic- has been linked to possible slab break-off (Clu- Ophiolite Klippen ular to plate motion. The south fl ank interpreta- zel et al., 2005). tion is based on a shear sense inversion within The second phase, affecting Upper Plio- The ophiolite klippen have a different com- the Massif du Sud (Nicolas, 1989), a pattern that cene– through Quaternary-age rocks, has positional and fabric character than the Massif has been used to infer fl anking directions in the variable extension directions of 90°, 40°, and du Sud. Plagioclase lherzolites are exposed in Oman ophiolite (see Fig. 2.2 in Nicolas, 1989). 0°–10° (Lagabrielle et al., 2005). The obliq- many of the klippen, with local diopside harz- These interpretations of ridge-transform uity between these directions and the trend burgites and spinel lherzolites (Moutte, 1982). geometry imply that the Massif du Sud formed of the Norfolk Ridge has resulted in sinistral The fabric patterns and orientations are gener- in an inner corner environment bound on transtension (Chardon and Chevillotte, 2006). ally less consistent than those in the Massif the west by the Bogota Peninsula shear zone Lagabrielle et al. (2005) suggested that this du Sud, both within each klippe and between (Figs. 2D and 2E). From the available data, it extension is due to the modern plate-tectonic klippen. This inconsistency has been attributed is diffi cult to determine whether the shear zone setting, perhaps related to the fl exure of the to late deformation and rotation of blocks (Leb- records deformation only on the Massif du Sud oceanic lithosphere currently subducting under lanc et al., 1980; Nicolas, 1989). Lagabrielle plate, or whether it records deformation across the Vanuatu island arc. and Chauvet (2008) interpreted these klippen as the entire transform portion of the system. In isolated allochthonous remnants of the ophiolite the former case, both ridge segments would be NEW CALEDONIA OPHIOLITE nappe displaced along the western Caledonia located in the South Loyalty Basin, as illustrated fault zone in response to regional extension. in Figure 2E. In the latter case, the western ridge The ophiolite sheet has two primary expo- Several of the northernmost klippen are also segment should be south of the present-day sures on New Caledonia. The large continuous deformed by the 120-km-long Belep shear zone Bogota Peninsula and therefore onshore. sheet in the south is known as the Massif du (Fig. 2). Outside the shear zone, fabrics are sim- Similar fi eld fabric patterns are used to sug- Sud and includes the Bogota Peninsula shear ilar to those in the Massif du Sud with E-W– to gest that the Belep shear zone represents a zone, which is the focus of this study. The NW-SE–striking foliations and subhorizontal transform fault (Sécher, 1981). However, the series of klippen along the northwestern coast N-S–trending lineations (Fig. 1; Nicolas, 1989). noncontinuous nature of exposures and con- are in fault contact with the Poya terrane; sev- Inside the shear zone, the NW-striking foliation fl icting shear sense indicators complicate this eral klippen are deformed in the Belep shear becomes vertical, while lineation remains hori- argument. The shear zone affects the western zone (Fig. 1). In this section, we compare the zontal. In the Tiebaghi Massif, the subhorizontal sides of Poum and Tiebaghi, all of Yandé, and compositional and fabric patterns between foliations in harzburgites outside the shear zone the eastern sides of Art and Pott, requiring a these two exposures as well as the previous can be traced to vertical mylonitic fabrics in change from NW-striking on New Caledo- tectonic interpretations of the two shear zones. lherzolites (Moutte, 1982). nia to more N-striking near the Belep islands

LITHOSPHERE | Volume 3 | Number 3 | www.gsapubs.org 223

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/3/3/221/3050183/221.pdf by guest on 01 October 2021 TITUS ET AL.

Map of Belep shear zone A Map of Bogota Peninsula shear zone B Pott Foliation trajectories Peridotites Expanded maps show Main shear zone compositional variations: Art Plagioclase and spinel lherzolites Harzburgite/ dunite N Secondary shear zone Yande Canala N

Foliation trajectories

Poum Mylonites Upper unit Thio Lower unit

20 km Tiebaghi 10 km

C Map view geometry of Belep shear zone D Block diagram geometry for Bogota Peninsula shear zone ridge

lithospheric shear (low T, high strain)

fracture ttransformransform ffaultault zone

shear sense inversion inner outer asthenospheric corner corner shear (high T, low strain) Crust Lithospheric mantle Asthenospheric mantle ridge

Schematic interpretation of shear zones E in New Caledonia ophiolite

N Bogota Peninsula Belep shear zone shear zone Massif du Sud

100 km

Figure 2. Maps of the (A) Belep and (B) Bogota Peninsula shear zones, showing foliation trajectories and shear sense indicators. (C) Interpretation of the Belep shear zone from Nicolas (1989), where the diffuse fabric patterns and opposite shear sense in the Belep Islands are linked to asthenospheric fl ow patterns. (D) Interpretation of fabric patterns from Bogota Peninsula shear zone and Massif du Sud, where the shear zone records dextral motion along a transform fault and the Massif du Sud records spreading-related fabrics on the south fl ank of an E-W–trending ridge. Nicolas (1989) suggested that the Bogota Peninsula shear zone may represent only one half of the transform fault system, meaning that the present exposure on New Caledonia refl ects rocks presently separated by a fracture zone (and not rocks between two ridge segments). Diagrams A–D were modifi ed from Nicolas (1989). (E) Using these interpretations of the Belep and Bogota Peninsula shear zones, we show possible ridge-transform geometries if both refl ect transform faults. For the Bogota Peninsula shear zone, the distance between ridge segments is unknown and used solely to illustrate the expected sense of shear.

224 www.gsapubs.org | Volume 3 | Number 3 | LITHOSPHERE

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/3/3/221/3050183/221.pdf by guest on 01 October 2021

Fabric development in the mantle section of a paleotransform fault, New Caledonia | RESEARCH

′ (Fig. 2A). Dextral shear sense indicators are

05 66°

1 observed everywhere except for one sinistral

4 measurement from olivine LPO in the Belep Islands (Sécher, 1981). The fabric is more dif- N fuse than fabrics in the Bogota Peninsula shear 21°′S Massif du Sud Thio

30 = 375 zone (Nicolas, 1989).

5 49

10 km The plate-boundary geometry responsible

for fabrics in the klippen and Belep shear zone

′

05 is illustrated in Figure 2E. Again, the transform 166° = 51

N strike would parallel the lineation in the sheared ed from Lozes and Yerle (1976), (1976), Yerle and Lozes ed from portions of the klippen. Based on the sinistral

shear sense, Nicolas (1989) suggested that 81

27 rocks in the Belep Islands experienced an anti- N thetic shear sense due to diapiric upwelling at an outer corner (Fig. 2C). This inference requires one ridge segment between the Belep Islands 77 and the Yandé massif. To maintain dextral shear

13 sense elsewhere in the system, the other ridge 44

ell as data from the Massif du Sud and limited data from data from the Massif du Sud and limited ell as data from

4 2

24 would presumably be adjacent to the last evi-

57 62

= 23 L N

° 71 dence of shearing in the klippen. 39 21 ′S 46

25 K Mesozoic sedimentary rocks The obliquity between these ridge-transform

systems suggests that it would be difﬁ cult to link

15 J

77 them within a single ocean basin without some I

N rotation of the ophiolite sheet during obduction;

15 74

10 this rotation, however, would destroy the simi- H

= 125

64 64 6 N

G lar strike of foliation and lineation throughout

82 8

72 7

72 the nappe. We revisit the interpretations of these

D 4

C two systems later herein after presenting new

7 74 74

E 6

′ A

21 and detailed fabric and strain data across the 4

00 64 6 64

166° 68

12 0

14 Bogota Peninsula shear zone.

80 8 80 76 are used for all stereographic projections. See Nicolas (1989) for details on the Massif du Sud for See Nicolas (1989) projections. all stereographic used for are

68 4

= 34 84

ve domains within the Bogota Peninsula shear zone. Map is modiﬁ shear zone. Peninsula domains within the Bogota ve FIELD MEASUREMENTS

Canala 6

56 5 56

11 21

Mafic volcanic/plutonic rocks

71 7

71 Our ﬁ eld measurements from the Bogota

72 5

Peninsula shear zone are primarily from coastal

9 N outcrops, where wave-cut terraces provide 81 excellent three-dimensional exposures and rela- 82 tively fresh rocks (Fig. 3). Most of these coastal

53 sites are in the northern structural domain of Bogota Peninsula, where steeply dipping 21°′S 20 foliations are observed. In contrast, rocks in Ultramafic rocks

5 the southern domain (the highlands) are often 64 deeply weathered and have moderately inclined and inconsistent fabrics. Prinzhofer and Nicolas = 67

N (1980) ﬁ rst noted these two structural domains secondary mylonites

28 and interpreted the contact between them as a 74 51 low-angle, E-W–striking thrust fault.

Laterite

E far NW near NW center near SE far SE N 29

′ Field Fabrics 55

165° 74

Kouaoua The alignment of spinel and/or pyroxene

83 24 grains deﬁ nes foliation and lineation in the 12 peridotites along the Bogota Peninsula (Fig. 4).

78 Careful analysis of foliation and lineation ori-

N N

= 12 entations across the 50-km-wide study region N in Figure 5 demonstrates a more complex pat-

Quaternary deposits

lt 66

Cape Koua tern than is evident when fabric data are simply

Lineation Lineation

Foliation (bold) Foliation Fau Fabric station

Poro combined into shear zone and non–shear zone

13 N KA04-6 fabrics (Fig. 1). Instead of observing N-S–strik- Figure 3. Geologic map of the Bogota Peninsula and surrounding coastline showing ﬁ coastline showing and surrounding Peninsula Geologic map of the Bogota 3. Figure domain as w each from foliation poles to show projections stereographic Equal-area and Guy (1982). and Paris (1979), Guy et al. at 2 Kamb contours the Massif du Sud, for Except of the shear zone). (northwest Poro data. The location of station KA04-6, which is used for reference in Figure 5 and Table 1, is indicated near Cape Kouaoua. is indicated 1, Table 5 and in Figure reference is used for which The location of station KA04-6, data. ing vertical foliations in the shear zone, the

LITHOSPHERE | Volume 3 | Number 3 | www.gsapubs.org 225

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/3/3/221/3050183/221.pdf by guest on 01 October 2021 TITUS ET AL.

A B

C D

Figure 4. Field photographs of fabric development across the shear zone including (A) moderate folding of compositional banding common in the far-fi eld areas, (B) near-vertical banding common in the near-fi eld and central regions, where a dunite compositional band is parallel to foliation, (C) typical foliation within the shear zone defi ned by aligned orthopyroxene grains, and (D) orthopyroxene clots in the center of the shear zone that are larger than normal orthopyroxene grains, which we interpret as former dikes boudinaged beyond recognition. The outcrop in A is approximately 2 m tall.

strike of foliation rotates from NW-SE outside these fi ve domains or, if fewer data are available, we focus on pyroxenite dikes and diabase dikes, Bogota Peninsula to NNE-SSW in the center of in terms of only three domains (center, near which are present in great enough quantity to the shear zone; this rotation is accompanied by fi eld, and far fi eld), given the overall symmetry determine robust patterns across the shear zone. increasing dips, from ~60° on the margins of the across the shear zone. study area to 75°–90° in the center. Instead of Pyroxenite Dikes observing N-S–trending lineations everywhere, Dikes Two sets of pyroxenite dikes can be identi- the subhorizontal lineation directions rotate fi ed in each of the fi ve shear zone domains based clockwise with foliation from NW-SE– to NNE- Several generations of dikes intrude the on their deformation behavior. Poles illustrated SSW–trending in the center of the shear zone. ultramafi c rocks along Bogota Peninsula. as solid circles in Figure 6 represent dikes that Based on these patterns of rotation, we divide Crosscutting relationships indicate the rela- are necked or boudinaged, whereas open circles the shear zone into fi ve domains: a center with tive timing from earliest to latest as pyroxenite, represent dikes that display constant thickness. NNE-striking foliations, two near-fi eld regions feldspathic pyroxenite, hornblende gabbro, and Like foliation and lineation, both sets rotate with N-striking foliations, and two far-fi eld diabase dikes (Prinzhofer and Nicolas, 1980). clockwise from the far fi eld toward the center regions with NW-striking foliations (stereo- While Nicolas (1989) reported that all dikes on of the shear zone. graphic projections in Figs. 3 and 6). The stip- the Bogota Peninsula shear zone were vertical The stretched dikes rotate from NW-striking pling in Figure 5 highlights these fi ve domains. with an average strike of 30°, we demonstrate with a variety of dips in the far fi eld to NNW- This interpretation is noted here because the that dike orientations depend on position across striking and steeper in the near fi eld; dikes from remaining fi eld data are either presented from the shear zone, as well as composition. Here, the center have similar orientations to near-

226 www.gsapubs.org | Volume 3 | Number 3 | LITHOSPHERE

240 A Foliation KEY 180 Field fabrics Foliation strike/dip 120 Lineation trend/plunge Strike (°) Data from Prinzhofer 60 and Nicolas (1980) 80 These are based on average orientations at 60 each station. 40 Dip (°) SPO measurements 20 S S SPO 1- 2 plane S 0 SPO long axis ( 1) 0 10 20 30 40 50 ε s - natural strain ν - Lode’s parameter 240 B Lineation Note: no dip direction 180 or plunge direction information is plotted in (A) or (B) graphs. Trend (°) Trend 120

60 Shear zone regions 60 Center/secondary shear zone 45 Near field 30 Far field Plunge (°) 15 Regions are based on 0 0 10 20 30 40 50 consistency of fabric data at each station. C 0.3 SPO Comparisons 0.2 strain s Average orientation ε or value from 0.1 Massif du Sud (width reflects ~ 0.0 standard deviation) Oblate 0.5

ν 0.0

-0.5 Prolate -1.0 0 10 20 30 40 50 Distance east from KA04-6 (km) Figure 5. Graphs showing the orientation of (A) foliation and (B) lineation across the shear zone, where distance is measured relative to the westernmost station studied (KA04-06). Foliation strikes and lineation trends were also added from Prinzhofer and Nicolas (1980).

For comparison, the shape-preferred orientation (SPO) foliation (S1-S2 plane) and lineation (S1) have been added to A and B, respec- tively. (C) The magnitude and shape of the SPO ellipsoid. The stippled regions in each chart show the shear zone center and near-ﬁ eld regions based on ﬁ eld fabric orientations. The gray horizontal bar denotes the average value calculated from four stations within the main Massif du Sud.

fi eld dikes, but this set includes dikes with true these dikes include subhorizontal orientations, folds within the center of the shear zone, folded N-S strikes. Except for within the center, these WNW-striking surfaces with a variety of dips, pyroxenite dikes were exclusively found in stretched dikes are subparallel to foliation. Quali- and N-S–striking steep surfaces. In the near the far-fi eld regions. We plot the orientations tatively, the stretching recorded by this dike set fi eld, the dikes either parallel the boudinaged of folded dikes of any composition, including increases toward the center of the shear zone, set or are steeply dipping and NNE-striking (in pyroxenite dikes, in Figure 6 because so few as illustrated by the greater separations between the NW) or gently dipping and E-W–striking (in folds were observed across the region. Because boudins in the shear zone interior. Further, boudi- the SE). In the center, there are few dikes with the axial planes were often diffi cult to mea- nage is observed on both horizontal and vertical constant thickness; those present strike NE with sure accurately in the fi eld, we plot the poles surfaces in the center, as opposed to primarily on a variety of dips. to fold limbs. Since fold shapes are typically horizontal surfaces in far-fi eld regions (Fig. 7). Folded pyroxenite dikes were occasionally isoclinal, these demonstrate that fold axial The dikes with constant width also change observed with axial planar foliations and iso- planes are subparallel to foliation for each of orientation across the shear zone. In the far fi eld, clinal fold shapes (Fig. 7E). Except for three the three shear zone domains. These data also

LITHOSPHERE | Volume 3 | Number 3 | www.gsapubs.org 227

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/3/3/221/3050183/221.pdf by guest on 01 October 2021 TITUS ET AL.

far NW near NW center near SE far SE Foliation

N = 67 N = 34 N = 125 N = 23 N = 51

N = 97 N = 95 N = 70 N = 42 N = 104 Joints

N = 229 N = 129 N = 438 N = 191 N = 84 Diabase dikes N = 2 N = 10 N = 61

Figure 6. Lower-hemisphere, equal-area projections from the ﬁ ve (or three) shear zone regions highlighted in Figure 3. These regions were deﬁ ned empirically based on consistent foliation and lineation orientations. Many of the other

Folded dikes types of data (dike orientations, joint orienta- N = 35 N = 9 N = 14 tions, shape-preferred orientation [SPO]) support these divisions as they also rotate with the changing foliation orientation. For smaller data sets, we combined data from the two near-fi eld and two far-fi eld regions to better illustrate the patterns across the shear zone. For pyroxenite dikes, fi lled circles represent dikes showing boudinage or necking while open circles repre-

Lineation Pyroxenite dikes sent dikes with constant thickness.

N = 22 N = 11 N = 47 SPO axes

N = 11 N = 6 N = 11

228 www.gsapubs.org | Volume 3 | Number 3 | LITHOSPHERE

A B

C D

foliation trace

E F

Figure 7. Photographs of pyroxenite (A–E) and diabase (F) dikes from the ophiolite. (A) Subhorizontal, weakly boudinaged dike from the Massif du Sud can be compared with boudinage for dikes in the far-fi eld (B), near-fi eld (C), and central (D) regions of the shear zone. Note the near total transposition into the foliation plane and vertical boudins in D from the center of the shear zone. (E) Folded dikes are common in the far-fi eld regions but were rarely observed elsewhere. (F) An undeformed diabase dike.

LITHOSPHERE | Volume 3 | Number 3 | www.gsapubs.org 229

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/3/3/221/3050183/221.pdf by guest on 01 October 2021 TITUS ET AL.

demonstrate that folds are most common in the lected at each station on outcrop surfaces that the SPO lineations are often steeper than those far ﬁ eld. were approximately mutually perpendicular. of ﬁ eld lineations in the center of the shear

To summarize, both the boudinaged and Data were also collected from four stations in zone (Fig. 5). The orientations of S1, S2, and

folded dikes support our division of the shear the Massif du Sud to facilitate comparison with S3 are also plotted from the three shear zone zone into fi ve domains. The low-strain far- fabrics developed outside of the shear zone. domains in Figure 6. These data illustrate the fi eld domains preserve pyroxenite dikes with a clockwise rotation of fabrics from the far fi eld variety of orientations, including folded dikes. Determining the Shape-Preferred to the center; their better clustering in the cen- The high-strain center domain preserves a nar- Orientation Ellipsoid ter highlights the increased fabric strength in row range of dike strikes, and most are quite the high-strain portion of the shear zone. boudinaged. We also note here that orthopy- Tracings were digitized using a fl atbed roxene blebs, signifi cantly larger than the aver- scanner; the resulting images were processed MICROSTRUCTURAL FABRICS age orthopyroxene grain size, were observed using ImageJ (Rasband, 2009). For each trac- (Fig. 4D). We interpret these as relict boudins ing, a two-dimensional fabric ellipse was calcu- We measured olivine lattice-preferred ori- from pyroxenite dikes, which are so extended lated using the Intercept method (Launeau and entations (LPO) across the shear zone to better that it is no longer possible to trace them as con- Robin, 1996) from the software package Ellip- characterize the temperature and style of defor- tinuous planar structures. These blebs may help soid 2003 (Launeau et al., 1994; Launeau and mation in peridotites (Nicolas and Christensen, explain why boudinaged dikes from the center Robin, 1996, 2005; Robin, 2002). For each sta- 1987; Drury and Fitz Gerald, 1998). We also of the shear zone are not parallel to foliation, tion, these two-dimensional ellipses were com- computed predicted seismic anisotropy values unlike the pattern from the other four domains. bined mathematically into a three-dimensional based on the fabric strength and orientation of These dikes would have become so extended fabric ellipsoid using Ellipsoid2003. This ellip- olivine LPO. Both are described here. that they could no longer be recognized. soid represents the shape-preferred orientation (SPO) of the population of pyroxene grains. Olivine LPO Diabase Dikes A signifi cant advantage of SPO analysis is Diabase dikes are mostly restricted to the that, in addition to orientation information, the Olivine LPO was measured at 12 sites across center of the shear zone, with consistent ~30° method provides quantitative constraints on the shear zone (A–L in Fig. 3). In each sample,

strikes and steep dips to the NW (Fig. 6). These strain magnitude. Using the principal axes (S1, 102 grains were oriented using a universal

dikes commonly crosscut fi eld fabrics in harz- S2, S3) of each SPO ellipsoid, we can deter- stage mounted on a petrographic microscope. ε burgites. The dikes may show slight changes mine the natural shear s and shape parameter Because of the moderate degree of serpentiniza- ν ε of orientation relative to position in the shear . When s = 0, the SPO ellipsoid is spherical; tion in these rocks, this technique was invalu- ε zone—those from near-fi eld regions have more ellipticity increases with increasing values of s able for identifying and orienting individual northerly strikes than the main NE-striking set (Nadai, 1963; Hsu, 1966; Hossack, 1968). The grains. The LPO patterns are presented in their from the center of the shear zone. Boudinage values of the shape parameter have the range in situ geographic orientation (and not the more of diabase dikes was not observed (Fig. 7), −1 < ν < 1, where ν < 0 corresponds to con- standard rock fabric reference frame) in Fig- although several dikes showed minor warping strictional fabrics, ν = 0 is plane strain, and ν ure 8. When possible, we also show the orienta- and bending. > 0 represents fl attening fabrics (Ramsay and tions of fi eld foliation and lineation, which were Huber, 1983). independently measured at the time of sample Joints collection. To test the repeatability of these mea- SPO Ellipsoid Results surements, LPO values were measured from Joint orientations were measured on marine two different samples at the same site (site D). terraces at select stations along the coastline. We The results of our SPO analysis are reported All sites show point distributions for the collected data about the representative orienta- in Table 1. To examine spatial changes across three crystallographic axes consistent with the ε ν tions of nonhorizontal, systematic joint sets, but the shear zone, the values of s and are plot- high-temperature fabrics expected in mantle we did not record information about joint spac- ted graphically in Figure 5C. The biggest dif- conditions. In detail, sites from the center of ing or density. ference between values from the Massif du Sud the shear zone (A–H) have subhorizontal a axes The poles to joints for the fi ve shear zone (in gray) and those from the shear zone occurs that are ~10°–15° clockwise of the fi eld linea- ε regions are plotted in Figure 6. Two distinct in the center, where s is elevated, between 0.2 tion, consistent with dextral shear sense (Nico- joint sets are present in each region. These sets and 0.3, and fabrics change from prolate/plane las, 1989). The b and c axes form more weakly are mutually perpendicular and become steeper strain (in the far fi eld) to plane strain/oblate (in defi ned point distributions, where b axes often in the center of the shear zone. The poles to the center). A secondary shear zone originally lie near the pole to fi eld foliation. Sites from the joints mimic the rotation of foliation, so that the noted by Prinzhofer and Nicolas (1980) is also near fi eld (I and J) and far fi eld (K and L) have two sets are always parallel and perpendicular evident northwest of the central zone, as indi- slightly weaker fabric patterns than those from ε ν to foliation. cated by higher s and values, steeper folia- the center. The a axes in these regions rotate tions, and S-plunging lineations. with the changing foliation strike. ORTHOPYROXENE MACROSCOPIC The orientations of the fabric ellipsoids are

FABRICS also plotted in Figure 5 as an SPO foliation (S1- Seismic Anisotropy

S2 plane) and lineation (S1) to facilitate direct In the fi eld, tracings of orthopyroxene grains comparison with fi eld fabrics on a station-by- Using the LPO data, we computed seismic were made on sheets of clear plastic at thirty station basis. The consistency between these velocities and shear-wave anisotropy values stations along the coast. Three oriented tracings, calculated SPO fabrics and the measured fi eld with the Ani2K software (based on Mainprice, each with 80–300 individual grains, were col- fabrics is quite clear, although the plunges of 1990). In these calculations, seismic velocities

230 www.gsapubs.org | Volume 3 | Number 3 | LITHOSPHERE

TABLE 1. SUMMARY OF ORTHOPYROXENE SHAPE-PREFERRED ORIENTATION (SPO) DATA Station Latitude Longitude SZ distance SPO long axis SPO intermediate axis SPO short axis ε n s

(°S) (°E) (km) Mag Trend Plunge Mag Trend Plunge Mag Trend Plunge (°) (°) (°) (°) (°) (°) KA04-06 21.29848 165.75627 0.0 1.058 258 12 0.996 353 23 0.949 142 63 0.08 –0.1 KA04-20 21.34460 165.80430 5.0 1.088 299 1 0.994 207 47 0.924 30 43 0.12 –0.1 KA04-22 21.35942 165.83210 7.9 1.137 334 30 0.967 92 39 0.910 218 36 0.16 –0.5 KA04-18 21.37538 165.86497 11.3 1.122 334 8 0.971 87 71 0.918 242 18 0.15 –0.4 KA04-17 21.38165 165.87892 12.7 1.154 116 60 1.049 351 18 0.826 253 23 0.24 0.4 KA04-23 21.35670 165.88508 15.1 1.113 272 24 1.044 36 51 0.861 168 28 0.19 0.5 KA04-25 21.41602 165.92443 17.4 1.097 315 6 0.976 194 78 0.934 46 10 0.12 –0.5 BG04-41 21.47915 165.98143 23.3 1.114 342 15 0.981 154 75 0.915 251 2 0.14 –0.3 BG04-13 21.42695 165.92970 24.5 1.052 144 39 0.994 326 51 0.957 235 1 0.07 –0.2 BG04-15 21.44048 166.00655 25.9 1.098 220 44 1.014 4 40 0.898 111 19 0.14 0.2 BG04-04 21.45832 166.02588 27.9 1.126 19 2 1.029 110 31 0.863 285 59 0.19 0.3 BG04-05 21.45721 166.02901 28.3 1.112 195 1 0.965 102 69 0.932 286 21 0.13 –0.6 BG04-06 21.46030 166.03332 28.7 1.108 200 16 0.997 23 74 0.906 290 1 0.14 0.0 BG04-21 21.45885 166.03592 29.0 1.178 195 3 0.994 303 82 0.854 105 8 0.23 –0.1 BG04-20 21.46457 166.04077 29.5 1.183 183 16 0.989 30 72 0.854 275 8 0.23 –0.1 BG04-07 21.47682 166.04618 30.0 1.219 200 19 1.004 22 71 0.817 290 1 0.28 0.0 BG04-18 21.46140 166.04618 30.0 1.166 177 40 1.027 6 50 0.836 271 5 0.23 0.4 BG04-17 21.46338 166.04828 30.3 1.154 183 30 1.049 24 58 0.826 279 9 0.24 0.2 BG04-08 21.47652 166.04830 30.3 1.147 195 37 1.042 66 39 0.837 309 29 0.24 0.4 BG04-8.5 21.47565 166.04882 30.3 1.235 184 9 0.972 22 80 0.833 274 3 0.28 –0.2 BG04-09 21.47382 166.05242 30.7 1.18 328 49 1.04 161 40 0.82 66 7 0.27 0.3 BG04-12 21.46577 166.05567 31.0 1.092 343 42 1.053 132 43 0.870 238 16 0.17 0.7 BG04-10 21.46473 166.05825 31.3 1.173 208 2 1.116 321 85 0.764 118 4 0.10 0.2 BG04-44 21.48633 166.05833 31.3 1.067 22 13 1.009 271 57 0.929 120 29 0.33 0.8 BG04-45 21.49280 166.06808 32.3 1.059 353 25 1.005 145 62 0.940 258 11 0.08 0.1 BG04-39 21.53002 166.07717 33.2 1.046 83 56 1.015 263 34 0.942 353 0 0.19 0.5 BG04-47 21.48903 166.07967 33.5 1.11 125 73 1.04 348 13 0.86 256 11 0.08 0.4 BG04-38 21.52807 166.08510 34.1 1.026 97 69 1.017 324 15 0.958 231 15 0.05 0.7 TH04-02 21.51960 166.12472 38.2 1.137 261 25 1.029 58 64 0.855 167 9 0.20 0.3 TH04-05 21.58053 166.20155 46.1 1.099 313 36 1.030 203 24 0.884 88 44 0.16 0.4 RB04-01 22.14783 166.66437 – 1.141 285 3 0.975 15 8 0.899 176 82 0.172 –0.31 RB04-03 22.15677 166.93503 – 1.085 257 8 1.019 166 2 0.905 60 82 0.130 0.31 RB04-04 22.16637 166.88272 – 1.063 340 11 0.995 248 11 0.945 114 75 0.083 –0.12 RB04-05 22.22368 166.62780 – 1.106 257 2 1.008 167 11 0.897 359 79 0.149 0.12 Note: Station names reﬂ ect general geographic locations: KA stations are near Kouaoua, BG stations are near the Bogota Peninsula, TH stations are near Thio, and RB stations are in the Massif du Sud and used for baseline comparisons. SZ distance indicates the distance from KA04-06. The magnitude and orientation of each principal axis of the shape-preferred orientation (SPO) ellipsoid are also provided. The magnitude of these three axes are used to compute ( ( ) − ( )) 1 2 2 2 2 lnSS23 ln ε = ((ln(SS) − ln( )) + ( ln( S) − ln( S)) + ( ln( S) − ln(S )) ) , from Nadai (1963) ν= from Ramsay and Huber (1983). s 12 23 3 1 ( ) − ( ) 3 lnSS13 ln

are determined using the LPO, density (Crosson the true values, but they are still useful for com- MAGNETIC FABRICS and Lin, 1971), and elastic stiffness coefﬁ cient parison with shear-wave splitting results from of olivine (Abramson et al., 1997). The Voight- other lithospheric-scale shear zones. For the same subset of sites across the shear Reuss-Hill averaging technique was used for The maximum, minimum, and average com- zone (A–L), we applied a combination of high-

these calculations; the choice of averaging tech- pressional and shear-wave velocities (Vp, Vs) ﬁ eld and low-ﬁ eld anisotropy of magnetic sus-

nique affects the absolute velocities but not the and anisotropies (AVp, AVs) for each LPO mea- ceptibility (AMS) techniques. These two types anisotropy values (Mainprice and Silver, 1993). surement are reported in Table 2 (see defi nitions of AMS measurements are used in tandem to Samples were assumed to be 100% dunite of these values in Mainprice and Silver, 1993). isolate the primary magnetic silicate fabric because serpentinization makes it diffi cult to Select seismic properties are illustrated graphi- from olivine and pyroxene. The methods and determine the original modal composition, cally in Figure 8. There is no statistical difference results of this analysis are presented here as a and there were too few orthopyroxene grains between velocities or anisotropies for sites in the test case for the applicability of this technique to determine their statistically signifi cant LPO center of the shear zone (A–H) and those fl anking to mantle rocks. pattern. Typically, the addition of pyroxene LPO the shear zone (I–L). The compressional wave decreases the overall seismic anisotropy param- velocities are typical of those calculated from AMS Methodology eters (Christensen and Lundquist, 1982). Fur- LPO patterns in mantle xenoliths, but most other

ther, calculated anisotropy values are typically parameters (especially AVp and AVs) are slightly For mantle rocks, secondary minerals like higher than laboratory measurements by ~1% higher than values calculated from xenoliths (see magnetite typically dominate the low-ﬁ eld (Christensen, 2002). Given these caveats, our table 4 in Titus et al., 2007), which is not surpris- AMS signal (MacDonald et al., 1988; Bina and seismic anisotropy results likely overestimate ing given the aforementioned caveats. Henry, 1990; Richter et al., 1996; Borradaile

LITHOSPHERE | Volume 3 | Number 3 | www.gsapubs.org 231

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/3/3/221/3050183/221.pdf by guest on 01 October 2021 TITUS ET AL.

ABCD1D2 EF WEST shear zone center axes a axes b axes c (km/s) p V 8.8 9.0 9.3 8.8 9.1 8.7 9.0 7.9 7.9 7.8 8.0 7.9 8.0 8.0 (%) s AV 7.4 8.8 11.4 5.9 9.2 5.9 8.0 0.2 0.2 0.3 0.3 0.4 0.1 0.2 directions Polarization

Figure 8. Results of olivine fabric analysis. Each column represents data from a single station labeled A–L on Figure 3. The stations are arranged from west to east, and the portion of the shear zone (center, near fi eld, far fi eld) from which samples were collected is also indicated. The fi rst three rows show lower-hemisphere, equal-area projections for the olivine lattice-preferred orientation (LPO) data. Data have been contoured as percent area, and N = 102 for each site. The bottom three rows show calculated seismic anisotropy parameters from these data, including compressional wave velocity

(Vp), shear-wave anisotropy (AVs), and polarization directions. Contours for Vp are in 0.2 km/s, and the contours for AVs are in 1% delay time; the maximum and minimum values are denoted next to each plot (and also in Table 3). (Continued on following page).

232 www.gsapubs.org | Volume 3 | Number 3 | LITHOSPHERE

GH I JKLEAST shear zone center near eastern margin far eastern margin a axes b axes c axes V p k/)AV (km/s)

9.2 9.2 8.8 8.9 8.8 8.9 8.0 8.0 8.0 7.9 8.0 8.0 s (%)

9.0 9.2 6.6 7.8 6.9 7.2 0.1 0.3 0.1 0.1 0.2 0.2 Polarization directions

Figure 8 (continued).

LITHOSPHERE | Volume 3 | Number 3 | www.gsapubs.org 233

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/3/3/221/3050183/221.pdf by guest on 01 October 2021 TITUS ET AL.

TABLE 2. SUMMARY OF SEISMIC ANISOTROPY VALUES High-Field AMS

Sample Vp max Vp min AVp Vs1 max Vs1 min Vs2 max Vs2 min AVs AVs (km/s) (km/s) (%) (km/s) (km/s) (km/s) (km/s) (%) (s) We might expect the long axis, k1, of the A 8.84 7.93 10.9 5.02 4.82 4.86 4.61 7.4 0.36 high-fi eld AMS ellipsoid to parallel the lineation B 9.04 7.93 13.2 5.07 4.84 4.88 4.56 8.8 0.43 direction and the short axis to parallel the pole C 9.29 7.81 17.2 5.19 4.86 4.88 4.50 11.4 0.56 to foliation (Borradaile and Henry, 1997; Gré- D1 8.79 8.02 9.1 4.98 4.83 4.88 4.64 5.9 0.28 goire et al., 1998). However, our results, which D2 9.09 7.92 13.8 5.07 4.88 4.95 4.55 9.2 0.44 represent 4–6 specimens per site, show no clear E 8.74 8.03 8.6 5.00 4.81 4.86 4.67 5.9 0.29 pattern in the high-fi eld ellipsoid orientation. F 8.99 7.97 12.0 5.01 4.85 4.91 4.58 8.0 0.38 G 9.15 7.96 13.9 5.11 4.85 4.87 4.55 9.0 0.44 We attribute this scatter, in part, to the presence H 9.19 7.96 14.4 5.11 4.86 4.87 4.54 9.2 0.45 of hematite, which cannot be saturated even at I 8.79 7.95 10.1 5.03 4.81 4.85 4.66 6.6 0.32 high-fi eld conditions. Further, fewer specimens J 8.87 7.93 11.2 5.05 4.82 4.86 4.61 7.8 0.38 were measured in high-fi eld conditions because K 8.76 8.04 8.6 5.01 4.78 4.86 4.66 6.9 0.33 these measurements were more time-consuming L 8.92 8.00 10.9 4.99 4.87 4.9 4.60 7.2 0.34 (1 h per sample) than low-fi eld measurements Avg. (all) 8.96 7.96 11.8 5.05 4.84 4.88 4.59 7.9 0.38 (5 min). Last, the low intrinsic degree of anisot- Avg. (in SZ) 9.03 7.95 12.7 5.07 4.85 4.89 4.58 8.4 0.41 ropy for Fo (Belley et al., 2009) may contrib- Avg. (out SZ) 8.87 7.96 10.8 5.03 4.82 4.87 4.62 7.4 0.36 92 ute to the weakly anisotropic high-fi eld AMS. Note: AV —P-wave anisotropy; AV —S-wave anisotropy, where subscript 1 denotes the fast wave and p s High-fi eld susceptibilities, calculated from subscript 2 denotes the slow wave. See text for details. SZ—shear zone. the high-fi eld slope, vary from 350 to 615 μSI. The degree of anisotropy ranges from 1.02 to 1.13. The shape factor varies from −0.66 to 0.68. and Lagroix, 2001). These minerals form during In the laboratory, 20 mm cubic specimens Similar to the low-fi eld AMS results, there are

serpentinization at temperatures <500 °C (e.g., were cut from oriented hand-samples with hor- no consistent patterns for Pj or T relative to posi- Saad, 1969), so that their shape anisotropy is izontal, vertical N-S, and vertical E-W faces. tion within the shear zone. often unrelated to mantle fl ow fabrics (MacDon- A Kappabridge KLY-3S was used to measure ald et al., 1988). We wanted to separate the pri- low-fi eld AMS at a low alternating fi eld of 300 COMPARISON OF FABRIC mary, mantle fabric due to paramagnetic compo- A/m. A Vibrating Sample Magnetometer was MEASUREMENTS nents (olivine, orthopyroxene, chromian spinel) used for high-fi eld AMS measurements up to from the secondary, serpentine-related fabric. fi elds of ~1 T. High-fi eld AMS was measured To facilitate comparisons between our shear This separation is possible in high magnetic on a subset of low-fi eld cubic specimens. Low- zone measurements, we compiled fi eld fabric fi elds (>0.7 T), where the saturation magneti- fi eld and high-fi eld AMS data are presented and dike orientations, SPO from macroscopic zation of ferromagnetic minerals is reached (at graphically in Figure 9, and site value averages orthopyroxene, LPO of olivine, low- and high- least in the absence of hematite or maghemite), are reported in Table 3. fi eld AMS data, and joint measurements for because the slope of the hysteresis loop becomes stations A–L in Figure 9. constant and represents the contribution from Low-Field AMS There is broad consistency between data sets paramagnetic minerals only (Kelso et al., 2002; that record high-temperature fabrics, including Martín-Hernández and Ferré, 2007). This tech- The low-fi eld AMS axes, representing foliation, lineation, SPO, LPO, and high-fi eld nique has been applied successfully to other 14–53 cubic specimens per site, are clustered AMS. As discussed previously, the fi eld and

mantle rocks (Ferré et al., 2005). tightly at the site level. Low-ﬁ eld k1 axes SPO lineations are typically parallel. Olivine a Both low- and high-ﬁ eld AMS results can be always form point distributions that plunge axes are clockwise from this orientation, which represented geometrically by an ellipsoid with moderately to steeply toward many different is expected for dextral systems (Tommasi et al.,

three mutually perpendicular principal axes: k1 azimuths across the shear zone. The k2 and k3 1999). Although the high-fi eld AMS patterns ≥ ≥ k2 k3 (Jelinek, 1981). The bulk susceptibil- axes either form point distributions or together are less well defi ned, k1 is parallel to fi eld linea- ity, k, is the average value of the principal axes form a girdle along the plane perpendicular to tion for sites with the most prolate ellipsoids (E

and relates the induced magnetization in a sam- k1. Like k1 orientations, k2 and k3 have incon- and G). For several other sites (B, D, and F), k2 ple to an applied ﬁ eld. By using the separation sistent orientations across the shear zone. For parallels lineation; these sites have less prolate method, the low-ﬁ eld AMS ellipsoid character- most stations, a great circle connecting two ellipsoid shapes, meaning that the distinction

izes the bulk rock fabric, whereas the high-fi eld of the low-fi eld AMS axes is approximately between k1 and k2 is less important. AMS ellipsoid characterizes the primary silicate parallel to fi eld foliation, but the correlation The records of lower-temperature fabrics fabric of the peridotites. Two other parameters is not perfect. from low-fi eld AMS and joints are consistent are also useful for our analysis (Jelinek, 1981): The bulk susceptibility k ranges from 350 on a station-by-station basis. The poles to joints, μ (1) the degree of anisotropy Pj quantifi es the to 2350 SI, although 70% of specimens have which are always gently plunging due to the sub- μ strength of the magnetic fabric, where Pj = 1 k < 1000 SI. The degree of anisotropy Pj horizontal coastal exposures, typically parallel

for a perfect sphere, and Pj > 1 for increasing ranges from 1.009 to 1.34, with ~90% speci- low-ﬁ eld AMS axes that are also gently plung-

ellipticity; and (2) the shape parameter T char- mens showing Pj < 1.1. The shape factor varies ing. The orientation of joints (Figs. 6 and 9) also acterizes the shape of the AMS ellipsoid, vary- from –0.94 to 0.85, although 70% of speci- seems to mimic the rotation of foliation across ing such that −1 ≤ T ≤ 1. When T = −1, the AMS mens fall in the ﬂ attening ﬁ eld (T > 0). There the shear zone, suggesting that original ductile

ellipsoid is perfectly prolate; when T = 1, it is are no consistent patterns for Pj or T relative to fabrics may control the orientation of later-stage perfectly oblate. position within the shear zone. brittle deformation and serpentinization.

234 www.gsapubs.org | Volume 3 | Number 3 | LITHOSPHERE

Field fabrics Orthopyroxene SPOLPO a axes LPO b axes LPO c axes High-field AMS Low-field AMS Poles to joints A

N N = 4 N = 6 N = 14 N = 32 B

N = 13 N = 6 N = 22 N = 41 C

N = 10 N = 33 N = 26 D

N = 4 N = 39

N = 24 N = 47

N = 6 N = 17 E

N = 5 N = 6 N = 32 N = 30 F

N = 5 N = 4 N = 32 N = 27

Figure 9. Comparison of all data sets for select stations across the Bogota Peninsula. The lower-hemisphere, equal-area projections are organized by data type (in columns) and station location (rows A–L; see Fig. 3 for map locations). Columns show (1) fi eld measurements; (2) orthopyroxene shape-preferred orientation (SPO); (3) olivine lattice-preferred orientation (LPO); (4) high- and low-fi eld anisotropy of magnetic susceptibility (AMS) measurements; and (5) poles to joints measured in the fi eld. The last row shows details about the different types of data shown on each projection. Occasionally, the different data sets were not collected at the same locations; this situation is indicated by offsets in the different rows (I–L).

LITHOSPHERE | Volume 3 | Number 3 | www.gsapubs.org 235

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/3/3/221/3050183/221.pdf by guest on 01 October 2021 TITUS ET AL.

Field fabrics Orthopyroxene SPOLPO a axes LPO b axes LPO c axes High-field AMS Low-field AMS Poles to joints G

N = 5 H N = 6 N = 53

N = 41 N = 6 I

N = 3 J

N = 6 N = 26 N = 45

N = 10 K

N = 4 L

N = 4 N = 22 N = 40

N = 5

Field foliation Field foliation Field foliation Confidence ellipse Poles to joints Field lineation Field lineation Field lineation Mean k1 Poles to dikes Long axis Contours of % area Mean k2 Intermediate axis Mean k3 N = # of poles Dotted contour is 1% area Short axis Specimen k1 to dikes N = 102 for all samples N = total # of SPO foliation Specimen k2 specimens Specimen k3

Figure 9 (continued).

236 www.gsapubs.org | Volume 3 | Number 3 | LITHOSPHERE

MODEL OF SHEAR ZONE EVOLUTION occurred in the shear zone center, suggesting that this was the last active part of the system. T Using our extensive data set recording different components of strain and deformation Shear Zone Evolution

j across the shear zone, and existing fabric data P

HF parameters from the Massif du Sud (Prinzhofer et al., Figure 10 shows our simpliﬁ ed, conceptual 1980), we constructed a conceptual model for model for shear zone localization as four dis- (°) 3

Plunge the development of the Bogota Peninsula shear crete time steps, although in reality, the defor- ) for both the LF (low-ﬁ eld) ) for both the LF (low-ﬁ

3 zone over time (Fig. 10). This model is based on mation may have varied smoothly in space and (°) the interpretation of Nicolas (1989) illustrated in time. The top row shows foliation trajectories

HF-AMS k Figure 2D, where the Massif du Sud represents that are active in a given step (solid lines) and

Mag Trend material formed at a spreading center and the those trajectories that have been frozen (dotted Bogota Peninsula records dextral motion due to line). The bottom row shows dike orientations

(°) transform motion, either refl ecting deformation and their associated stretches exclusively within 2 ), and short axes (k Plunge 2 recorded on one plate or across both plates. the actively deforming portion of the system. We purposely have not labeled this diagram (°) Localizing Deformation with specifi c directions in space; instead, we HF-AMS k discuss the orientation of the system in the fol- Mag Trend For our conceptual model, we must decide lowing section. ), intermediate (k 1 the order in which strain was accommodated in In the fi rst step, fabrics develop due to sea- (°) g (k

1 the far fi eld, near fi eld, and center of the shear fl oor spreading. Based on mapping in the Mas- Plunge zone. Two models for shear zone evolution sif du Sud (Prinzhofer, 1981), the orientations of

(°) are common (Wojtal and Mitra, 1986; Means, pyroxenite dikes seem to vary widely, but their 1995; Horsman and Tikoff, 2007): (1) localiza- orientations were not sorted according to spa- tion, where deformation is fi rst accommodated tial location. Thus, we cannot determine if dike 1.024 346 13 1.000 167 77 0.977 76 0 1.048 0.000 across a broad region, i.e., the far fi eld in New orientations vary systematically in space (e.g., Caledonia represents the fi rst increment of with the foliation trajectories from the Massif T Mag Trend deformation, but becomes focused in a narrower du Sud). The few dikes that we observed in the high-strain zone (Passchier, 1986; West and Massif du Sud were typically subhorizontal and j P Hubbard, 1997); or (2) delocalization, where only weakly deformed (Fig. 7A), if at all. the shear zone width increases, i.e., from the In the second step, deformation across a

[SI] center of the Bogota Peninsula shear zone to the wide shear zone causes steepening and rota- K –6 far fi eld, during progressive deformation (e.g., tion of foliation preserved in the two far-fi eld LF parameters HF-AMS k x10 Aoya and Wallis, 2003). regions. The fabrics developed in this phase Modern oceanic transform systems are may be slightly stronger than those in Massif du (°) 3 Plunge instructive for choosing an appropriate model of Sud based on the SPO patterns (Fig. 5). Two sets shear zone development, and most data support of dikes are commonly observed in the far-fi eld (°) a localizing model. Deformation at ridge-trans- regions. One set is steeply dipping and not quite LF-AMS k form intersections, often inferred by the pres- parallel to foliation. This set shows evidence of

Mag Trend ence of active fault scarps, is typically expressed boudinage on horizontal surfaces; these dikes

, and T are also reported. T , and across a wider region near the intersection and also may form the limbs of isoclinal folds, the j

(°) becomes narrower within the transform val- axial planes of which are subparallel to folia- 2 Plunge TABLE 3. SUMMARY OF ANISOTROPY OF MAGNETIC SUSCEPTIBILITY (AMS) DATA OF MAGNETIC SUSCEPTIBILITY ANISOTROPY OF 3. SUMMARY TABLE ley away from the ridge (Fox and Gallo, 1984; tion. The second set of dikes is oblique to the

(°) Gallo et al., 1986; Kastens et al., 1986; Macdon- foliation strike, gently dipping, and relatively ald et al., 1986). The transform fault also marks undeformed. In reality, fabric development and LF-AMS k a major thermal boundary between oceanic dike injection in the Massif du Sud (fi rst step) Mag Trend lithosphere of different ages (Parker and Olden- and far-fi eld regions (second step) may have burg, 1973), and many thermal models (Phipps occurred at the same time. The steeper orien- (°) 1 Plunge Morgan and Forsyth, 1988; Shen and Forsyth, tations in the far fi eld may simply refl ect the 1992; Furlong et al., 2001) support the narrow- changing stress fi eld near the ridge-transform (°) ing of the shear zone as the oceanic lithosphere intersection (e.g., Allerton, 1989; Gudmunds- moves away from the mid-ocean ridge (see, son, 1995). however, Behn et al. [2007], where the opposite In the third step, deformation becomes local- pattern is predicted). ized in the narrower shear zone preserved in the The diabase dikes from the Bogota Pen- near-fi eld regions. Continued deformation in insula shear zone also support a localizing this phase causes foliation to rotate and steepen. model. These dikes are undeformed and essen- One set of pyroxenite dikes is now subparallel Station numbers across the shear zone are shown in Figure 1. The magnitude (mag) and orientation (trend plunge) of the lon Station numbers across the shear zone are shown in Figure 1. tially restricted to the center of the shear zone. with foliation and quite stretched; the separation number Mag Trend

Note: Thus, the last pulse of magmatism in the region between boudins is observed on both horizon- K–L BG02_22A 1.022 234 75 1.008 69 14 0.971 338 4 0.65 1.055 0.440 1.035 304 19 0.996 43 25 0.971 181 57 1.067 –0.198 I BG02_10B 1.029 262 36 0.998 12 26 0.974 130 44 0.69 1.057 –0.108 1.043 70 60 0.994 316 13 0.965 220 26 1.081 –0.240 C BG02_2BE 1.035 158 BG02_8A 34 1.103 305 1.006 309 61 52 1.013 130 0.960 58 29 0.896 14 39 0.86 2 1.079 0.231 1.7 1.233 0.174 1.040 37 7 0.986 130 27 0.975 294 62 1.072 –0.661 and HF (high-ﬁ eld) AMS data are included. Values for k (in SI), P AMS data are included. Values eld) and HF (high-ﬁ J BG02_21A 1.016 270 28 1.002 163 28 0.982 37 49 0.71 1.034 0.183 D1D2 BG02_1AF BG02_1B 1.012G 1.010 288 164 BG02_17A 56 BG02_15A 1.027 22 1.036 198 0.996 1.006 123 13 35 282 33 0.996 81 50 340 0.992 0.980 0.985 166 29 48 60 0.978 8 7 32 93 0.967 257 0.92 20 0.81 1.022 1.028 4 –0.632 0.92 1.071 0.670 1.051 1.026 284 1.5 –0.263 33 1.041 34 1.071 119 –0.077 1.040 47 1.028 174 0.996 10 32 153 0.997 27 211 0.898 18 26 0.993 0.978 13 55 185 261 0.963 44 351 70 32 1.208 0.980 1.049 73 0.675 299 –0.239 1.081 –0.120 9 1.051 –0.465 H BG02_12A A BG02_6A 1.009 176 36 1.003 286 25 0.989 42 43 0.72 1.021 0.436 1.010 10 70 1.000 124 8 0.990 216 18 1.020 0.058 B BG02_4A 1.019 141 34 1.001 240 14 0.980 348 52 0.96 1.040 0.055 1.053 59 48 1.013 178 24 0.938 285 32 1.125 0.340 Station Sample LF-AMS k Station Sample LF-AMS tal and vertical faces. The second set of dikes

LITHOSPHERE | Volume 3 | Number 3 | www.gsapubs.org 237

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/3/3/221/3050183/221.pdf by guest on 01 October 2021 TITUS ET AL.

t = 1 t = 2 t = 3 t = 4 Spreading related fabrics Far-field fabrics formed Near-field fabrics formed Central fabrics formed fossil foliation trajectories active foliation trajectories

Foliation dips gently away Foliation steepens and Shear zone narrows; Deformation concentrated from spreading ridge rotates across farfield width foliation continues to within high-strain core steepen & rotate diabase pyroxenite dike dikes

Dike orientations vary; Subhorizontal dikes steepen; Most dikes nearly transposed Dikes parallel to foliation no longer dikes mostly undeformed subvertical dikes rotate towards into foliation with horizontal observed except as individual boudins; foliation and stretch horizontally; and vertical stretching; late diabase dikes are undeformed folded dikes have axial planes few folds observed parallel to foliation shear zone localization

Figure 10. Schematic diagram illustrating how the Bogota Peninsula shear zone may have evolved as a localizing shear zone over time. The data along the top row show foliation relationships (active and fossilized), whereas data along the bottom row show dike orientations and behaviors relative to foliation solely within the actively deforming zone. At t = 1, fabrics in the Massif du Sud were formed due to seafl oor spreading, followed by shear zone localization at t = 2 far-fi eld, t = 3 near-fi eld, and t = 4 central domains. Late-stage diabase dikes were intruded in this last stage of deformation.

is relatively undeformed, steeply dipping, and data. Recall that the previous interpretation of boundary must be oriented 15°–35° east of N. oblique to foliation. ophiolite fabrics (Nicolas, 1989) had an E-W– Two independent data sets suggest that a NE- In the fourth and fi nal step, deformation is striking ridge responsible for the Massif du oriented value (30°–35°) is more appropriate. concentrated in the 3-km-wide center of the Sud and a N-S–striking transform fault, result- First, numerical modeling of LPO developed shear zone. Foliation is subvertical, and only ing in perpendicularity between the ridge and in transpression, transtension, and simple shear one set of subvertical dikes is still observed. transform (Fig. 2). However, our more detailed (Tommasi et al., 1999) shows that olivine a axes These dikes are not parallel to foliation, but they account of fabrics across the shear zone (Fig. 6), are closer to the shear direction than the fi eld are boudinaged. We infer that the set of dikes and closer examination of the spatial patterns of lineation (i.e., the long axis of the fi nite strain that was most parallel to foliation in the previ- foliation and lineation in the Massif du Sud sug- ellipsoid). Second, the consistent orientation of ous phase is now absent because the stretches gest that this interpretation of the ridge-trans- diabase dikes from the center of the shear zone were so large as to make these dikes unrecog- form geometry requires revisions. (~35°) may be due to strong anisotropy devel- nizable in the fi eld. The evidence for this comes The fi rst change we suggest is that the trans- oped in the transform system. Instead of refl ect- from small, lozenge-shaped orthopyroxene form fault was striking NNE to NE instead of ing the local extension direction at the time of blebs observed in the fi eld, which were larger due N. As noted earlier, fi eld foliation from intrusion, the dikes may have intruded parallel than normal orthopyroxene grains distributed in the high-strain center of the shear zone (with to the shear zone boundaries. Similar patterns harzburgites (Fig. 4D). its subhorizontal lineation) strikes 15°, while have been observed in the Husavik-Flatey fault, olivine a axes trend 30°. This magnitude and a modern transform fault system in Iceland Orientation of the Shear Zone Boundaries sense of angular discrepancy between fi eld lin- (Garcia et al., 2002). Therefore, 30°–35° is our eation and LPO are expected in dextral systems best estimate of the shear zone boundaries. The cartoon in Figure 10 implicitly assumes (e.g., Darot and Boudier, 1975; Nicolas, 1989). The second revision involves the spread- that the spreading ridge, which caused gently Given any reasonable model for fabric develop- ing direction versus ridge segment orientation dipping foliation in the fi rst time step, is perpen- ment in a shear zone, and assuming relatively responsible for fabric development in the Mas- dicular to the shear zone boundaries, but this is high-strain values in a dominantly simple shear sif du Sud. If spreading were perpendicular to actually diffi cult to determine from the available history, these data suggest that the shear zone the ridge, the foliation strike and lineation trend

238 www.gsapubs.org | Volume 3 | Number 3 | LITHOSPHERE

should also be mutually perpendicular. Even This orientation is also consistent with the broad we discuss whether the shear zone represents though the stereographic projections for the trends in magnetic (Bitoun and Recy, 1982; a whole or a half of a transform fault. We then Massif du Sud, on average, show perpendicular- Lafoy et al., 1996) and gravity data (Collot et al., examine how mantle fabrics and their associated ity between these fabrics (Fig. 1), the fabric tra- 1987) from the South Loyalty Basin. mechanical anisotropies may have controlled jectories demonstrate “left-leaning” lineations Figure 11B illustrates our revised interpre- processes of ophiolite emplacement and later relative to foliations (Fig. 11). This pattern sug- tation of the ridge-transform geometry, now Neogene extension. gests that spreading was oblique enough to the oriented in space and accounting for the non- ridge segment to cause relative sinistral motion, perpendicularity between ridge and transform Is the Bogota Peninsula Shear Zone a which any model for the ridge-transform geom- segments. We show where the rocks currently True Transform Fault? etry must take into account. exposed on the Bogota Peninsula may have The third revision involves the actual orienta- originated, either preserving the true transform The high-temperature olivine LPO pat- tion of the ridge segments. This orientation is the fault between two ridge segments, or half of a terns support formation of the Bogota Penin- most diffi cult to constrain with the available fi eld transform fault preserved on the Massif du Sud sula shear zone within the oceanic lithosphere data. Our best option is to assume that relative plate. These two possibilities for the history of and not due to later reactivation of a preexist- plate motion was parallel to the transform fault, the transform fault are discussed next. ing structure. The stretched pyroxenite dikes, a common observation in modern plate recon- which rotate clockwise with foliation and show structions. If spreading were oriented ~30°, the DISCUSSION larger stretches in the center of the shear zone, ridge segment would need to be clockwise from suggest that the shear zone records progressive 120° to result in the sinistral sense of motion for Our rich and varied data set across the deformation. We used our detailed fi eld data the Massif du Sud. We suggest the most likely Bogota Peninsula shear zone provides detailed (Fig. 6) and fabric measurements (Figs. 5, 7, 8, orientation is parallel to the 140° trend of New information about fabric evolution within a and 9) from the Bogota Peninsula shear zone Caledonia and, locally, of the Norfolk Ridge. large-scale mantle shear zone. In this section, to develop a conceptual model for its evolution

Data constraining original Tectonic context for Bogota Peninsula shear zone A ridge-ridge-transform geometry B Previous geometry Revised geometry

E-W-striking Norfolk Ridge- ~15° 30–35° spreading ridge parallel spreading ridge N N

Sinistral shear sense N-S-striking in inner corner transform fault

Possible locations lineation LPO NE-trending of Bogota Peninsula transform fault Fabrics from high-strain center N shear zone of Bogota Peninsula shear zone

Foliation and lineation trajectories not perpendicular in Massif du Sud

Norfolk Ridge 100 km N

Figure 11. (A) Data relevant for determining the orientation of the ridge-ridge-transform geometry, including lower-hemisphere, equal-area projections from the center of the shear zone and foliation/lineation trajectories from Prinzhofer et al. (1980). The rotated lattice-preferred orientation (LPO) data were compiled by combining the 1% area contours for stations A–H from Figure 8. (B) The previous and revised interpretations for ridge-ridge-transform geometry responsible for fabrics in the Massif du Sud and the Bogota Peninsula shear zone.

LITHOSPHERE | Volume 3 | Number 3 | www.gsapubs.org 239

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/3/3/221/3050183/221.pdf by guest on 01 October 2021 TITUS ET AL.

(Fig. 10). What we have not discussed, however, Whattam et al. (2008), ophiolite emplacement recorded progressive deformation as material is whether the shear zone records both halves tends to occur shortly after formation instead within it moved away from the spreading ridge. of a transform fault or half of a transform fault of 30–50 m.y. later. When ophiolites are more The shear zone either recorded all of the trans- translated along an oceanic fracture zone. than 10 m.y. old, they become diffi cult to obduct form deformation (left column) or half of the Two main shear zone features support the because their lithosphere is too thick (Dewey, deformation preserved on one side of a fracture interpretation as a true transform fault. First, 2003). Other suprasubduction zone ophiolites zone (right column). foliations and dike orientations from the two seem to share the pattern of obduction shortly In the next panel (Fig. 12B), plate motion far-fi eld and two near-fi eld regions are parallel after formation, including the nearby Papuan became convergent in the Eocene, and the ophi- (Fig. 6). If the shear zone only recorded half of ophiolite (Lus et al., 2004) and the Semail ophiolite was obducted. We have shown the ophio- the total transform deformation, a more asym- olite in Oman (Hacker et al., 1996). Further, the lite sheet extending over all of New Caledonia metric fabric pattern might be expected. Sec- Troodos (Varga and Moores, 1985) and Oman based on exposures in the Massif du Sud, ophi- ond, the scale of the shear zone, where fabrics ophiolites (Ceuleneer et al., 1988; MacLeod and olite klippen, and small remnants preserved in and dikes rotate over a 50-km-wide region, is Rothery, 1992; Le Mée et al., 2004) both pre- the Pouebo terrane. Some models suggest that consistent with the observations from currently serve former spreading ridges within the sheet. the emplacement process was diachronous, active transform fault systems in oceanic envi- Thus, the data and obduction models from other beginning in the northern part of the island and ronments. In these modern systems, the trans- ophiolites suggest that the Bogota Peninsula continuing southward (e.g., Cluzel et al., 2001; form valley can be between 10 and 50 km wide shear zone represents a true transform fault. Baldwin et al., 2007). We have no fabric or (e.g., Choukroune et al., 1978; Garfunkel, 1981; structural evidence to support or refute this idea, Gallo et al., 1986; Parson and Searle, 1986; Fabric Controls on Ophiolite and as such we have simply shown the posi- Lagabrielle et al., 1992). Within this valley, there Emplacement and Neogene Extension tion of the ophiolite after obduction was com- may be multiple fault strands, or a single fault, plete. Because the ophiolite sheet drapes into that make up the principal transform displace- Because joint orientations across the Bogota the South Loyalty Basin (Collot et al., 1987), it ment zone, which can be as narrow as 300 m Peninsula shear zone rotate with the changing probably did not experience signifi cant rotation and up to 1–5 km in width (e.g., Fox and Gallo, orientation of foliation (Figs. 6 and 9), we sug- during emplacement. 1984; Parson and Searle, 1986; Mamaloukas- gest that mechanical anisotropies from the origi- After emplacement, we suggest that a fea- Frangoulis et al., 1991). Thus, the 3-km-wide, nal mantle fabrics controlled late-stage brittle ture from the non–Massif du Sud spreading high-strain zone on the Bogota Peninsula, as deformation of these rocks. In this section, we ridge was reactivated as a dextral strike-slip or well as the smaller mylonitic zone in the north- propose that this reactivation may be true on a transpressional shear zone (Fig. 12C). This fea- western far fi eld (Fig. 3), can be thought of as much larger scale as well, where inherited man- ture could have been the spreading ridge itself the mantle continuations of the fault systems tle fabrics controlled, at least in part, ophiolite or a spreading ridge–parallel structure such as a that would be observed at the surface within the obduction and much later Neogene extension. normal fault or extensional core complex, which transform valley. We also return to the interpretation of the Belep are both observed in other ophiolites (Varga and Reconstructions of the tectonic develop- shear zone, which deforms several klippen in Moores, 1985; Cann et al., 2001) and in modern ment of the southwest Pacifi c may also provide northwest New Caledonia. mid-ocean-ridge environments (e.g., Blackman insight on interpretation of the transform fault. et al., 1998; Ildefonse et al., 2007). In fact, this Schellart et al. (2006) suggested that the South Time Line of Deformation structure may have been the original discontinu- Loyalty Basin, where the ophiolitic sheet likely Our proposed tectonic model is summa- ity that initiated ophiolite emplacement. formed, was once 750 km wide, compared to rized in Figure 12, which shows a simplifi ed Reactivation of an inherited structure may its present width of <100 km. The symmetry of map-view history of the ophiolite. This model explain two structural features on the island. seafl oor spreading requires all ridge segments in is based primarily on fabrics and structures in First, the high-pressure Pouebo terrane experi- the middle of the basin to be subducted with this New Caledonia and does not explicitly rely on a enced shortening before its exhumation (Raw- amount of shortening. Thus, these features could particular tectonic reconstruction for the region ling and Lister, 2002), causing folding that not be preserved within the South Loyalty Basin (e.g., Aitchison et al., 1995; Cluzel et al., 2001; includes ophiolitic material. These folds hinges and on New Caledonia because of the geometry Crawford et al., 2003; Sdrolias et al., 2003; are asymptotic toward the present-day position required for a true transform fault (see Fig. 2). Schellart et al., 2006), since these models differ of the western Caledonia fault zone (Fig. 12C). However, Whattam et al. (2008) suggested that with regard to the location where the ophiolite Second, reactivation might account for the for- the proto-ophiolite lithosphere formed in the formed, the age of the ophiolite, the direction mation of mylonites in the Belep shear zone, forearc above a NE-dipping subducting slab east of subduction, and the original size of the South which are also along strike with the western of New Caledonia. These authors argued that the Loyalty Basin. Caledonia fault zone. Dextral motion is consis- ophiolite is much younger than its commonly In the fi rst panel (Fig. 12A), the material tent with both features, except for the one sinis- assumed Late Cretaceous age, forming instead that would become the New Caledonia ophio- tral shear sense indicator from the Belep Islands in the Eocene and emplaced not long after its lite formed in a ridge-ridge-transform environ- (Fig. 2A). Instead of using the antithetic shear formation. Given this tectonic environment, the ment. As discussed previously, we suggest that sense to place the Belep Islands in the outer cor- original ridge-transform geometry is unlikely to the spreading ridge segments were more or less ner of a transform fault (Fig. 2B), perhaps the have been subducted and could therefore have parallel with the Norfolk Ridge (Fig. 11). The Belep shear zone refl ects reactivation of mate- been emplaced directly onto New Caledonia. transform fault orientation forms a slightly acute rial during or after ophiolite emplacement. In Based on the available fi eld evidence and angle with the Norfolk Ridge, resulting in the the top diagram in Figure 12C, note the align- tectonic reconstructions, we favor the model sinistral shear sense for the inner-corner envi- ment of the ophiolite klippen in the shear zone, where the Bogota peninsula shear zone pre- ronment that becomes the Massif du Sud. Dur- which is different from their modern position serves the full transform fault. As noted by ing this time, the Bogota Peninsula shear zone (Fig. 12E). This arrangement allows the western

240 www.gsapubs.org | Volume 3 | Number 3 | LITHOSPHERE

Formation of Bogota Peninsula shear zone A Half transform fault Full transform fault Left-leaning fabrics Shear zone is part of larger plate- recorded in proto- boundary system High-strain core marks Massif du Sud location of transform fault Massif du Sud Bogota Peninsula shear zone localizes Massif du Sud

Bogota Peninsula Bogota Peninsula

N B Obduction of ophiolite Ophiolite sheet extends over all of New Caledonia and Younger drapes into South lithosphere Loyalty basin

Older lithosphere Younger lithosphere Older lithosphere

C Postobduction deformation: dextral reactivation of pre-existing structure N

Two main events: 1. Folding in HP terrane Belep SZ 2. Deformation of PtA Y Pm T Belep shear zone klippe Five klippe involved in shear zone - locations are approximate but based on portions that are sheared and aligned with folded section of HP terrane

Belep islands displaced relative to New Caleodnia? Belep SZ

D Neogene extension: normal faulting throughout New Caledonia N Normal faults in HP terrane

WCFZ

Normal motion along detachment fault drops klippe into place

E Today - present level of ophiolite expsoure Features aligned with HP fabrics N

Ophiolite preservation controlled by lithospheric WCFZ age? - Massif du Sud represents older Klippe in Belep shear zone deformed lithosphere compared to by reactivation of older structure and ophiolite klippe Klippe displaced on WCFZ, do not reflect an original lithospheric feature which may have been original structure Sharp boundary of Massif du Sud that allowed ophiolite obduction suggests Bogota transform fault/FZ acted as a tear fault during obduction

Figure 12. Cartoon illustrating how (A) fabric patterns formed within the ophiolite may have inﬂ uenced both (B) ophiolite emplacement and (C–D) later deformation. (E) The modern setting. See text for details. The klippen deformed by the Belep shear zone in C are abbreviated Pott (Pt), Art (A), Yandé (Y), Poum (Pm), and Tiebaghi (T). Fold hinge orientations in C and normal faults in D within the Pouebo terrane are from Rawling and Lister (2002). HP—high pressure; WCFZ—western Caledonia fault zone.

LITHOSPHERE | Volume 3 | Number 3 | www.gsapubs.org 241

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/3/3/221/3050183/221.pdf by guest on 01 October 2021 TITUS ET AL.

margins of Poum and Tiebaghi and the eastern same oceanic basin unless the klippen (or Mas- detachment fault that facilitated the movement margins of Art and Pott to deform. Continued sif du Sud) experienced major rotation during of ophiolite klippen into their present position motion along the shear zone may explain why emplacement. However, fabrics in the klippen along the northern portion of New Caledonia. the Belep Islands are so far offshore in relation away from the Belep shear zone are parallel to Thus, inherited mantle fabrics may explain to the other klippen, as illustrated by the lower those in the Massif du Sud (Fig. 1), making sig- the present distribution of ophiolitic material portion of the diagram (Fig. 12C). nifi cant rotation during emplacement unlikely. across New Caledonia and provide an expla- We suggest that the Belep shear zone was Thus, reactivation of a spreading ridge–parallel nation for a secondary shear zone within the reactivated as the western Caledonia fault zone feature may explain why the Belep shear zone klippen, the orientation of which is nearly per- during Neogene extension (Fig. 12D). This is nearly perpendicular to the Bogota Peninsula pendicular to the Bogota Peninsula shear zone. extension may have displaced most of the ophi- shear zone and parallel to our inferred spreading olite klippen into their present position; a simi- ridge orientation. ACKNOWLEDGMENTS lar model has been proposed for postorogenic Last, the ridge-transform geometry from collapse preserving the Leka ophiolite in Nor- our model (Fig. 12A) provides anisotropies Christopher Gordon was invaluable in the fi eld, way (Titus et al., 2002). This motion accounts within the ophiolite nappe that could have been and we thank the Ouasee Tribe, especially Chief for the misalignment of the ophiolite klippen, exploited during obduction and Neogene exten- Toussi, for permitting us to live and work on the which were displaced along the detachment sion. The Bogota Peninsula shear zone could Bogota Peninsula. We also thank Gerard Bes- fault, relative to the Belep Islands, which were have acted as a tear fault during obduction, son, Pierre Maurizot, and Erika Peterson for not. This may account for the apparent bend in allowing the continuous portion of the Massif du logistical support. Seismic anisotropy calcula- the Belep shear zone illustrated in Figure 2A. Sud to be emplaced with relative ease. The sharp tions were made using software from David This later phase of extension affected rocks western margin of the Massif du Sud, which is Mainprice, University of Montpelier. This mate- across the island. aligned with the southward continuation of the rial is based upon work supported by a National Bogota Peninsula shear zone, supports this sug- Science Foundation (NSF) Graduate Research Implications for the Tectonic Development gestion (see Fig. 12E). Both the Belep shear Fellowship (Titus), the Class of ’49 fund from of New Caledonia zone and Neogene extensional structures, such Carleton College (Titus), NSF grant EAR- This proposed reconstruction of ophiolite as the western Caledonian fault zone, may have 0337458 (Ferré and Tikoff), and a Packard Fel- formation, obduction, and Neogene extension exploited the ridge-parallel anisotropy. lowship (Tikoff). ties together four additional features about the ophiolite nappe, shear zones, and younger geo- CONCLUSIONS REFERENCES CITED logic structures on New Caledonia. First, our model may explain compositional The Bogota Peninsula shear zone provides Abramson, E.H., Brown, J.M., Slutsky, L.J., and Zaug, J., 1997, The elastic constants of San Carlos olivine to differences within the ophiolite sheet. The Mas- an unparalleled record of fabric development 17 GPa: Journal of Geophysical Research, v. 102, sif du Sud is primarily harzburgitic, whereas the across the mantle section of a transform fault. p. 12,253–12,263, doi:10.1029/97JB00682. Aitchison, J.C., Clarke, G.L., Meffre, S., and Cluzel, D., 1995, ophiolite klippen include spinel- and plagio- Field foliations, lineations, and pyroxenite dikes Eocene arc-continent collision in New Caledonia and clase-bearing lherzolites (Moutte, 1982; Ulrich show systematic rotation across a 50-km-wide implications for regional southwest Pacifi c tectonic et al., 2010). If the Bogota Peninsula shear zone region. Folded dikes are observed on the margins evolution: Geology, v. 23, p. 161–164, doi:10.1130/0091 -7613(1995)023<0161:EACCIN>2.3.CO;2. represents a dextral transform fault, separating of the deforming zone, and boudinage of dikes Allerton, S., 1989, Distortions, rotations and crustal thin- two island-parallel ridge segments (Fig. 12A), increases toward a central 3-km-wide mylonite ning at ridge-transform intersections: Nature, v. 340, the rocks that become the Massif du Sud and zone. Olivine LPO patterns demonstrate that a p. 626–628, doi:10.1038/340626a0. Aoya, M., and Wallis, S.R., 2003, Role of nappe boundaries in klippen would have formed at two different axes are slightly clockwise of fi eld lineations subduction-related regional deformation; spatial varia- spreading centers, thereby explaining their com- within the high-strain core, consistent with dex- tion of meso- and microstructures in the Seba eclogite unit, the Sambagawa Belt, SW Japan: Journal of Struc- positional differences. tral motion along the shear zone. The shear zone tural Geology, v. 25, p. 1097–1106, doi:10.1016/S0191 Second, the style of ophiolite exposure is interpreted as recording the deformation on -8141(02)00147-5. across the island may be linked to the original both sides of a transform fault. The Massif du Auzende, J.-M., van de Beuque, S., Regnier, M., Lafoy, Y., and Symonds, P., 2000, Origin of the New Caledonian oceanic plate-boundary geometry. If the Mas- Sud is attributed to rocks forming in an oceanic ophiolites based on a French-Australian seismic tran- sif du Sud formed in an inner corner, the rocks inner corner adjacent to a transform fault. sect: Marine Geology, v. 162, p. 225–236, doi:10.1016/ S0025-3227(99)00082-1. within it would be older and presumably thicker The strong material anisotropy from the Avias, J., 1967, Overthrust structure of the main ultraba- than rocks northwest of the Bogota Peninsula inherited ridge-ridge-transform geometry may sic New Caledonian massives: Tectonophysics, v. 4, shear zone, which became the klippen. Given have played an important role in the obduc- p. 531–541, doi:10.1016/0040-1951(67)90017-0. Baldwin, S.L., Rawling, T.J., and Fitzgerald, P.G., 2007, Ther- its age and thickness, the Massif du Sud could tion of the ophiolite: The initiation of obduc- mochronology of the New Caledonia high-pressure have been obducted more easily as a continu- tion may have been facilitated by a former terrane: Implications for middle Tertiary plate bound- ous sheet. ridge segment (or ridge-parallel fabric), and ary processes in the southwest Pacifi c, in Cloos, M., Carlson, W.D., Gilbert, M.C., Liou, J.G., and Sorensen, Third, our reinterpretation of the Belep shear the Bogota Peninsula shear zone may have S.S., eds., Convergent Margin Terranes and Associ- zone, not as an original transform fault but as acted as a tear fault during emplacement. The ated Regions: A Tribute to W.G. Ernst: Geological Society of America Special Paper 419, p. 117–134, a reactivated ridge-parallel structure, helps to inherited material anisotropy also may have doi:10.1130/2006.2419(06). explain its orientation relative to other ophiol- infl uenced subsequent Neogene extension, Behn, M.D., Boettcher, M.S., and Hirth, G., 2007, Thermal itic structures. The current strike of the Belep allowing a spreading ridge-parallel feature to structure of oceanic transform faults: Geology, v. 35, p. 307–310, doi:10.1130/G23112A.1. shear zone is 60°–80° oblique to our revised be reactivated twice: fi rst as a dextral shear Belley, F., Ferré, E.C., Martín-Hernández, F., Jackson, M., estimate for the strike of the Bogota Peninsula zone linked to the formation of folds in the Dyar, M.D., and Catlos, E.J., 2009, The magnetic

properties of natural and synthetic (Feχ, Mg1–χ)2 SiO4 shear zone. It is difﬁ cult to imagine how both high-pressure Pouebo terrane and mylonitiza- olivines: Earth and Planetary Science Letters, v. 284, shear zones could be transform faults from the tion in the Belep shear zone, and second as a p. 516–526, doi:10.1016/j.epsl.2009.05.016.

242 www.gsapubs.org | Volume 3 | Number 3 | LITHOSPHERE

Bina, M.M., and Henry, B., 1990, Magnetic properties, Collot, J.Y., Rigolot, P., and Missegue, F., 1988, Geologic Gudmundsson, A., 1995, Stress fi elds associated with oceanic opaque mineralogy and magnetic anisotropies of ser- structure of the northern New Caledonia Ridge, as transform faults: Earth and Planetary Science Letters, pentinized peridotites from ODP Hole 670A near the inferred from magnetic and gravity anomalies: Tecton- v. 136, p. 603–614, doi:10.1016/0012-821X(95)00164-8. Mid-Atlantic Ridge: Physics of the Earth and Planetary ics, v. 7, p. 991–1013, doi:10.1029/TC007i005p00991. Guy, B., Paris, J.P., and Prinzhofer, A., 1979, Carte géologique Interiors, v. 65, p. 88–103. Crawford, A.J., Meffre, S., and Symonds, P.A., 2003, 120 to et notice explicative de la Nouvelle Caldédonie: Koua- Bitoun, G., and Recy, J., 1982, Origine et évolution du bassin 0 Ma tectonic evolution of the southwest Pacifi c and oua sheet, scale 1:50,000. des Loyauté et de ses bordures après la mise en place de analogous geological evolution of the 600 to 220 Ma Hacker, B.R., Mosenfelder, J.L., and Gnos, E., 1996, Rapid la série ophiolitique de Nouvelle-Calédonie: Contribu- Tasman fold belt system, in Hills, R.R., and Mueller, emplacement of the Oman ophiolite: Thermal and geo- tion à l’Étude Géodynamique du Sud-Ouest Pacifi que, R.D., eds., Evolution and dynamics of the Australian chronologic constraints: Tectonics, v. 15, p. 1230–1247, Travaux et Documents de l’ORSTROM, p. 505–539. plate: Geological Society of America Special Paper doi:10.1029/96TC01973. Blackman, D.K., Cann, J.R., Janssen, B., and Smith, D.K., 372, p. 383–403, doi:10.1130/0-8137-2372-8.383. Horsman, E., and Tikoff, B., 2007, Constraints on defor- 1998, Origin of extensional core complexes: Evidence Crossen, R.S., and Lin, J.W., 1971, Voight and Reuss predic- mation path from fi nite strain gradients: Journal of from the Mid-Atlantic Ridge at the Atlantis frac- tion of anisotropy elasticity of dunite: Journal of Geo- Structural Geology, v. 29, p. 256–272, doi:10.1016/j.jsg ture zone: Journal of Geophysical Research, v. 103, physical Research, v. 79, p. 570–578. .2006.09.003. p. 21,315–21,333, doi:10.1029/98JB01756. Darot, M., and Boudier, F., 1975, Mineral lineations in Hossack, J.R., 1968, Pebble deformation and thrusting in the Bonatti, E., 1978, Vertical tectonism in oceanic fracture deformed peridotites: Kinematic meaning: Pétrologie, Bygdin area (southern Norway): Tectonophysics, v. 5, zones: Earth and Planetary Science Letters, v. 37, v. 1, p. 225–236. p. 315–339, doi:10.1016/0040-1951(68)90035-8. p. 369–379, doi:10.1016/0012-821X(78)90052-3. Dewey, J.F., 2003, Ophiolites and lost oceans: Rifts, ridges, Hsu, T.C., 1966, The characteristics of coaxial and non- Borradaile, G.J., and Henry, B., 1997, Tectonic applications of arcs and/or scrapings?, in Dilek, Y., and Newcomb, coaxial strain paths: Journal of Strain Analysis, v. 1, magnetic susceptibility and its anisotropy: Earth-Sci- S., eds., Ophiolite Concept and the Evolution of Geo- p. 216–222, doi:10.1243/03093247V013216. ence Reviews, v. 42, p. 49–93, doi:10.1016/S0012-8252 logical Thought: Geological Society of America Special Ildefonse, B., Blackman, D.K., John, B.E., Ohara, Y., Miller, (96)00044-X. Paper 273, p. 153–158. D.J., and MacLeod, C.J., 2007, Oceanic core complexes Borradaile, G.J., and Lagroix, F., 2001, Magnetic fabrics Drury, M.R., and Fitz Gerald, J.D., 1998, Mantle rheology: and crustal accretion at slow-spreading ridges: Geol- reveal upper mantle fl ow fabrics in the Troodos ophi- Insights from laboratory studies of deformation and ogy, v. 35, p. 623–626, doi:10.1130/G23531A.1. olite complex, Cyprus: Journal of Structural Geology, phase transition, in Jackson, I., ed., The Earth’s Mantle: Jelinek, V., 1981, Characterization of the magnetic fabric of v. 23, p. 1299–1317, doi:10.1016/S0191-8141(00)00186-3. Composition, Structure and Evolution: Cambridge, rocks: Tectonophysics, v. 79, p. T63–T67, doi:10.1016/0040 Cann, J.R., Prichard, H.M., Malpas, J.G., and Xenophon- UK, Cambridge University Press, p. 503–560. -1951(81)90110-4. tos, C., 2001, Oceanic inside corner detachments of Dubois, J., Launay, J., and Recy, J., 1974, Uplift movements Karson, J., and Dewey, J.F., 1978, Coastal complex, west- the Limassol Forest area, Troodos ophiolite, Cyprus: in New Caledonia–Loyalty Islands area and their plate ern Newfoundland: An Early Ordovician oceanic frac- Journal of the Geological Society of London, v. 158, tectonics interpretation: Tectonophysics, v. 24, p. 133– ture zone: Geological Society of America Bulletin, p. 757–767, doi:10.1144/0016-764900-143. 150, doi:10.1016/0040-1951(74)90134-6. v. 89, p. 1037–1049, doi:10.1130/0016-7606(1978)89 Casey, J.F., Karson, J.A., Elthon, D., Rosencrantz, E., and Dupont, J., Lafoy, Y., Pautot, G., Le Suave, R., Cluzel, D., <1037:CCWNAE>2.0.CO;2. Titus, M., 1983, Reconstruction of the geometry of Missegue, F., Grandperrin, R., Hénin, C., Voisset, M., Kastens, K.A., Macdonald, K.C., Miller, S.P., Fox, P.J., and accretion during formation of the Bay of Islands ophi- Saint Omer, L.D., Gautheron, L., Butscher, J., Mollard, Detrick, R.S., 1986, Deep-tow studies of the Vema frac- olite complex: Tectonics, v. 2, p. 509–528, doi:10.1029/ L., and Rakoia, M., 1995, Etude morphostructurale de ture zone: 2. Evidence for tectonism and bottom cur- TC002i006p00509. la zone sud des rides Nouvelle-Calédonie et Loyauté rents in the sediments of the transform valley fl oor: Ceuleneer, G., Nicolas, A., and Boudier, F., 1988, Mantle fl ow (Zone Economique Exclusive de Nouvelle-Calédonie, Journal of Geophysical Research, v. 91, p. 3355–3367, patterns at an oceanic spreading centre: The Oman Pacifi que Sud-Ouest): Comptes Rendus de l’Académie doi:10.1029/JB091iB03p03355. peridotites record: Tectonophysics, v. 151, p. 1–26, des Sciences de Paris, v. 320, p. 211–218. Kelso, P.R., Tikoff, B., Jackson, M., and Sun, W., 2002, A new doi:10.1016/0040-1951(88)90238-7. Eissen, J.-P., Crawford, A.J., Cotten, J., Meffre, S., Bellon, method for the separation of paramagnetic and fer- Chardon, D., and Chevillotte, V., 2006, Morphotectonic evo- H., and Delaune, M., 1998, Geochemistry and tectonic romagnetic susceptibility anisotropy using low fi eld lution of the New Caledonia Ridge (Pacifi c southwest) signifi cance of basalts in the Poya terrane, New Cale- and high fi eld methods: Geophysical Journal Inter- from post-obduction tectonosedimentary record: Tec- donia: Tectonophysics, v. 284, p. 203–219, doi:10.1016/ national, v. 151, p. 345–359, doi:10.1046/j.1365-246X tonophysics, v. 420, p. 473–491, doi:10.1016/j.tecto.2006 S0040-1951(97)00183-2. .2002.01732.x. .04.004. Ferré, E.C., Tikoff, B., and Jackson, M., 2005, The magnetic Lafoy, Y., Missegue, F., Cluzel, D., and Le Suave, R., 1996, Chardon, D., Austin, J.A.J., Cabioch, G., Pelletier, B., Saustrup, anisotropy of mantle peridotites: Example from the The Loyalty–New Hebrides arc collision: Effects on S., and Sage, F., 2008, Neogene history of the northeast- Twin Sisters dunite, Washington: Tectonophysics, the Loyalty Ridge and basin system, southwest Pacifi c ern New Caledonia continental margin from multichan- v. 398, p. 141–166, doi:10.1016/j.tecto.2005.02.001. (fi rst results of the ZoNéCo Programme): Marine Geo- nel refl ection seismic profi les: Comptes Rendus Geo- Fox, P.J., and Gallo, D.G., 1984, A tectonic model for ridge- physical Researches, v. 18, p. 337–356, doi:10.1007/ science, v. 340, p. 68–73, doi:10.1016/j.crte.2007.09.017. transform-ridge plate boundaries: Implications for BF00286084. Choukroune, P., Francheteau, J., and Le Pichon, X., 1978, the structure of oceanic lithosphere: Tectonophysics, Lagabrielle, Y., and Chauvet, A., 2008, The role of extensional In situ structural observations along Transform Fault v. 104, p. 205–242, doi:10.1016/0040-1951(84)90124-0. tectonics in shaping Cenozoic New Caledonia: Bulletin A in the FAMOUS area, Mid-Atlantic Ridge: Geologi- Fox, P.J., Schreiber, E., Rowlett, H., and McCamy, K., 1976, The Society Geologique de France, v. 179, p. 315–329. cal Society of America Bulletin, v. 89, p. 1013–1029, geology of the Oceanographer fracture zone: A model Lagabrielle, Y., Mamaloukas-Frangoulis, Y., Cannat, V., and doi:10.1130/0016-7606(1978)89<1013:ISSOAT>2.0.CO;2. for facture zones: Journal of Geophysical Research, Auzende, J.-M., 1992, Vema fracture zone (central Christensen, N.I., 2002, Continental mantle seismic anisot- v. 81, p. 4117–4128, doi:10.1029/JB081i023p04117. Atlantic): Tectonic and magmatic evolution of the ropy: A new look at the Twin Sisters Massif: Tectono- Furlong, K.P., Sheaffer, S.D., and Malservisi, R., 2001, Ther- Median Ridge and the Eastern Ridge–Transform inter- physics, v. 355, p. 163–170, doi:10.1016/S0040-1951(02) mal-rheological controls on deformation within oce- section domain: Journal of Geophysical Research, 00139-7. anic transforms, in Holdsworth, R.E., ed., The Nature v. 97, p. 17,331–17,351, doi:10.1029/92JB01086. Christensen, N.I., and Lundquist, S.M., 1982, Pyroxene ori- and Tectonic Signifi cance of Fault Zone Weakening: Lagabrielle, Y., Maurizot, P., Lafoy, Y., Cabioch, G., Pelletier, entation within the upper mantle: Geological Society Geological Society of London Special Publication 186, B., Regnier, M., Wabete, I., and Calmant, S., 2005, Post- of America Bulletin, v. 93, p. 279–288, doi:10.1130/0016 p. 65–84. Eocene extensional tectonics in southern New Caledo- -7606(1982)93<279:POWTUM>2.0.CO;2. Gallo, D.G., Fox, P.J., Macdonald, K.C., and Detrick, R.S., nia (SW Pacifi c): Insights from onshore fault analysis Cluzel, D., Aitchison, J.C., and Picard, C., 2001, Tectonic accre- 1986, A Sea Beam investigation of the Clipperton and offshore seismic data: Tectonophysics, v. 403, tion and underplating of mafi c terranes in the late transform fault: The morphotectonic expression of a p. 1–28, doi:10.1016/j.tecto.2005.02.014. Eocene intraoceanic fore-arc of New Caledonia (south- fast slipping transform boundary: Journal of Geophys- Launeau, P., and Robin, P.Y.F., 1996, Fabric analysis using the west Pacifi c): Geodynamic implications: Tectonophys- ical Research, v. 91, p. 3455–3467. intercept method: Tectonophysics, v. 267, p. 91–119, ics, v. 340, p. 23–59, doi:10.1016/S0040-1951(01)00148-2. Garcia, S., Angelier, J., Bergerat, F., and Homberg, C., doi:10.1016/S0040-1951(96)00091-1. Cluzel, D., Bosch, D., Paquette, J.-L., Lemennicier, Y., Mont- 2002, Tectonic analysis of an oceanic transform fault Launeau, P., and Robin, P.Y.F., 2005, Determination of fab- joie, P., and Ménot, R.-P., 2005, Late Oligocene post- zone based on fault-slip data and earthquake focal ric and strain ellipsoids from measured sectional obduction granitoids of New Caledonia: A case for mechanisms: The Húsavík-Flatey fault zone, Iceland: ellipses: Implementation and applications: Journal of reactivated subduction and slab break-off: The Island Tectonophysics, v. 344, p. 157–174, doi:10.1016/S0040 Structural Geology, v. 27, p. 2223–2233, doi:10.1016/j Arc, v. 14, p. 254–271. -1951(01)00282-7. .jsg.2005.08.003. Cluzel, D., Meffre, S., Maurizot, P., and Crawford, A.J., 2006, Garfunkel, Z., 1981, Internal structure of the Dead Sea Launeau, P., Cruden, A.R., and Bouchez, J.-L., 1994, Mineral Earliest Eocene (53 Ma) convergence in the southwest leaky transform (rift) in relation to plate tectonics: recognition in digital images of rocks: A new approach Pacifi c: Evidence from pre-obduction dikes in the ophi- Tectonophysics, v. 80, p. 81–108, doi:10.1016/0040 using multichannel classifi cation: Canadian Mineralo- olite of New Caledonia: Terra Nova, v. 18, p. 395–402, -1951(81)90143-8. gist, v. 32, p. 919–933. doi:10.1111/j.1365-3121.2006.00704.x. Grégoire, V., Darrozes, J., Gaillot, P., Nedelec, A., and Lau- Lawson, K., Searle, R.C., Pearce, J.A., Browning, P., and Collot, J.Y., Malahoff, A., Recy, J., Latham, G., and Missegue, neau, P., 1998, Magnetite grain shape fabric and dis- Kempton, P., 1996, Detailed volcanic geology of the F., 1987, Overthrust emplacement of New Caledo- tribution anisotropy vs. rock magnetic fabric: A three- MARNOK area, Mid-Atlantic Ridge north of Kane nia ophiolite: Geophysical evidence: Tectonics, v. 6, dimensional case study: Journal of Structural Geology, transform, in MacLeod, C.J., Tyler, P.A., and Walker, p. 215–232, doi:10.1029/TC006i003p00215. v. 20, p. 937–944, doi:10.1016/S0191-8141(98)00022-4. C.L., eds., Tectonic, Magmatic, Hydrothermal and

LITHOSPHERE | Volume 3 | Number 3 | www.gsapubs.org 243

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/3/3/221/3050183/221.pdf by guest on 01 October 2021 TITUS ET AL.

Biological Segmentation of Mid-Ocean Ridges: Lon- the Bulletin of the Society of Economic Geologists, Geological Society of America Special Paper 372, don, Geological Society of London, p. 61–102. v. 77, p. 576–591, doi:10.2113/gsecongeo.77.3.576. p. 343–359, doi:10.1130/0-8137-2372-8.343. Leblanc, M., Dupuy, C., Cassard, D., Moutte, J., Nicolas, Nadai, A., 1963, Theory of fl ow and fracture of solids (2nd Sécher, D., 1981, Les Lherzolites Ophiolitiques de Nouvelle A., Prinzhofer, A., Rabinovitch, M., and Routhier, P., edition): New York, McGraw-Hill, 705 p. Calédonie et leurs Gisements de Chromite—Déforma- 1980, Essai sur la genese des corps podiformes de Nicolas, A., 1989, Structures of Ophiolites and Dynamics of tion de la Chromite [Ph.D. thesis]: Nantes, France, Uni- chromitite dans les peridotites ophiolitiques; etude Oceanic Lithosphere: Dordrecht, Netherlands, Kluwer versity of Nantes, 247 p. des chromites de Nouvelle-Caledonie et comparaison Academic Publishers, 367 p. Shen, Y., and Forsyth, D.W., 1992, The effects of temperature- avec celles de Mediterranee orientale (The genesis of Nicolas, A., and Christensen, N.I., 1987, Formation of anisot- and pressure-dependent viscosity on three-dimen- podiform chromitite complexes in ophiolitic perido- ropy in upper mantle peridotites: A review, in Fuchs, sional passive fl ow of the mantle beneath a ridge- tites; comparison of chromites in New Caledonia and K., and Froideveaux, C., eds., Composition, Structure transform system: Journal of Geophysical Research, the eastern Mediterranean), in Panayiotou, A., ed., and Dynamics of the Lithosphere-Asthenosphere Sys- v. 97, p. 19,717–19,728, doi:10.1029/92JB01467. Ophiolites: Proceedings of the International Ophiolite tem: Washington, D.C., American Geophysical Union Simonian, K.O., and Gass, I.G., 1978, Arakapas fault belt, Symposium, Cyprus, 1979: Nicosia, Cyprus Geological Geodynamics Series 16, p. 111–123. Cyprus: A fossil transform fault: Geological Society of Survey Department, p. 691–701. Paris, J.P., and Guy, B., 1982, Carte géologique et notice America Bulletin, v. 89, p. 1220–1230, doi:10.1130/0016 Le Mée, L., Girardeau, J., and Monnier, C., 2004, Mantle explicative de la Nouvelle Caldédonie: Canala-La-Foa -7606(1978)89<1220:AFBCAF>2.0.CO;2. segmentation along the Oman ophiolite fossil mid- sheet, scale 1:50,000. Spandler, C., Hermann, J., Arculus, R., and Mavrogenes, ocean ridge: Nature, v. 432, p. 167–172, doi:10.1038/ Parker, R.L., and Oldenburg, D.W., 1973, Thermal model of J., 2004, Geochemical heterogeneity and element nature03075. ocean ridges: Nature Physical Science, v. 242, p. 137– mobility in deeply subducted oceanic crust: Insights Lozes, J., and Yerle, J.-J., 1976, Carte géologique et notice 139, doi:10.1038/physci242137a0. from high-pressure mafi c rocks from New Caledo- explicative de la Nouvelle Caldédonie: Thio sheet, Parson, L.M., and Searle, R.C., 1986, Strike-slip fault styles in nia: Chemical Geology, v. 206, p. 21–42, doi:10.1016/j scale 1:50,000. slow-slipping oceanic transform faults; evidence from .chemgeo.2004.01.006. Lus, W.Y., McDougall, I., and Davies, H.L., 2004, Age of GLORIA surveys of Atlantis and Romanche fracture Spandler, C., Rubatto, D., and Hermann, J., 2005, Late Cre- the metamorphic sole of the Papuan ultramafi c belt zones: Journal of the Geological Society of London, taceous–Tertiary tectonics of the southwest Pacifi c; ophiolite, Papua New Guinea: Tectonophysics, v. 392, v. 143, p. 757–761, doi:10.1144/gsjgs.143.5.0757. insights from U/Pb sensitive, high-resolution ion p. 85–101, doi:10.1016/j.tecto.2004.04.009. Passchier, C.W., 1986, Flow in natural shear zones: The con- microprobe (SHRIMP) dating of eclogite facies rocks Macdonald, K.C., Castillo, D.A., Miller, S.P., Fox, P.J., Kas- sequences of spinning fl ow regimes: Earth and Plan- from New Caledonia: Tectonics, v. 24, 16 p. tens, K.A., Bonatti, E., and Detrick, R.S., 1986, Deep- etary Science Letters, v. 77, p. 70–80, doi:10.1016/0012 Suhr, G., and Cawood, P.A., 2001, Southeastern Lewis Hills tow studies of the Vema fracture zone: 1. Tectonics of -821X(86)90133-0. (Bay of Islands ophiolite): Geology of a deeply eroded, a major slow slipping transform fault and its inter- Phipps Morgan, J., and Forsyth, D.W., 1988, Three-dimen- inside-corner, ridge-transform intersection: Geologi- section with the Mid-Atlantic Ridge: Journal of Geo- sional fl ow and temperature perturbations due to a cal Society of America Bulletin, v. 113, p. 1025–1038, physical Research, v. 91, p. 3334–3354, doi:10.1029/ transform offset: Effects on oceanic crust and upper doi:10.1130/0016-7606(2001)113<1025:SLHBOI>2 JB091iB03p03334. mantle structure: Journal of Geophysical Research, .0.CO;2. MacDonald, W.D., Ellwood, B.B., Ellwood, B.B., Hrouda, F., v. 93, p. 2955–2966, doi:10.1029/JB093iB04p02955. Titus, S.J., Fossen, H., Pedersen, R.-B., Vigneresse, J.L., and and Wagner, J.-J., 1988, Magnetic fabric of peridotite Prinz, M., Keil, K., Green, J.A., Reid, A.M., Bonatti, E., and Tikoff, B., 2002, Pull-apart formation and strike-slip with intersecting petrofabric surfaces, Tinaquillo, Ven- Honnorez, J., 1976, Ultramafi c and mafi c dredge partitioning in an obliquely divergent setting, Leka ezuela: Physics of the Earth and Planetary Interiors, samples from the Equatorial Mid-Atlantic Ridge and ophiolite, Norway: Tectonophysics, v. 354, p. 101–119, v. 51, p. 301–312, doi:10.1016/0031-9201(88)90071-4. fracture zones: Journal of Geophysical Research, v. 81, doi:10.1016/S0040-1951(02)00293-7. MacLeod, C.J., and Rothery, D.A., 1992, Ridge axial seg- p. 4087–4103, doi:10.1029/JB081i023p04087. Titus, S.J., Medaris, L.G.J., Wang, H., and Tikoff, B., 2007, mentation in the Oman ophiolite: Evidence from Prinzhofer, A., 1981, Structure et Pétrologie d’un Cortege Continuation of the San Andreas fault system into along-strike variations in the sheeted dyke complex, Ophiolititque: Le Massif du Sud (Nouvelle-Calédonie) the upper mantle: Evidence from spinel peridotite in Parson, L.M., Murton, B.J., and Browning, P., eds., [Ph.D. thesis]: Paris, Université de Paris, 237 p. xenoliths in the Coyote Lake basalt, central Califor- Ophiolites and their Modern Oceanic Analogues: Prinzhofer, A., and Nicolas, A., 1980, The Bogota Peninsula, nia: Tectonophysics, v. 429, p. 1–20, doi10.1016/j.tecto Geological Society of London Special Publication 60, New Caledonia: A possible oceanic transform fault: The .2006.07.004. p. 39–63, doi:10.1144/GSL.SP.1992.060.01.03. Journal of Geology, v. 88, p. 387–398, doi:10.1086/628523. Tommasi, A., Tikoff, B., and Vauchez, A., 1999, Upper mantle MacLeod, C.J., Allerton, S., Gass, I.G., and Xenophontos, C., Prinzhofer, A., Nicolas, A., Cassard, D., Moutte, J., Leblanc, tectonics: Three-dimensional deformation, olivine 1990, Structure of a fossil ridge-transform intersection M., Paris, J.P., and Rabinovitch, M., 1980, Structures in crystallographic fabrics and seismic properties: Earth in the Troodos ophiolite: Nature, v. 348, p. 717–720, the New Caledonia peridotites-gabbros: Implications and Planetary Science Letters, v. 168, p. 173–186, doi:10.1038/348717a0. for oceanic mantle and crust: Tectonophysics, v. 69, doi:10.1016/S0012-821X(99)00046-1. Mainprice, D., 1990, An effi cient Fortran program to calcu- p. 85–112, doi:10.1016/0040-1951(80)90128-6. Ulrich, M., Picard, C., Guillot, S., Chauvel, C., Cluzel, D., and late seismic anisotropy from the lattice preferred ori- Ramsay, J.G., and Huber, M.I., 1983, The Techniques of Modern Meffre, S., 2010, Multiple melting stages and refertil- entation of minerals: Computers & Geosciences, v. 16, Structural Geology: San Diego, Academic Press, 307 p. ization as indicators for ridge to subduction formation: p. 385–393. Rasband, W.S., 1997–2009, ImageJ: Bethesda, Maryland, U.S. The New Caledonia ophiolite: Lithos, v. 115, p. 223– Mainprice, D., and Silver, P.G., 1993, Interpretation of SKS- National Institutes of Health, http://rsb.info.nih.gov/ij/. 236, doi:10.1016/j.lithos.2009.12.011. waves using samples from the subcontinental litho- Rawling, T.J., and Lister, G.S., 2002, Large-scale structure of Varga, R.J., and Moores, E.M., 1985, Spreading structure of sphere: Physics of the Earth and Planetary Interiors, the eclogite-blueschist belt of New Caledonia: Journal the Troodos ophiolite, Cyprus: Geology, v. 13, p. 846– p. 257–280. of Structural Geology, v. 24, p. 1239–1258, doi:10.1016/ 850, doi:10.1130/0091-7613(1985)13<846:SSOTTO>2.0 Mamaloukas-Frangoulis, Y., Auzende, J.-M., Bideau, D., S0191-8141(01)00128-6. .CO;2. Bonatti, E., Cannat, M., Honnorez, J., Lagabrielle, Y., Richter, C., Kelso, P.R., and MacLeod, C.J., 1996, Magnetic West, D.P., Jr., and Hubbard, M.S., 1997, Progressive local- Malavielle, J., Mével, C., and Needham, H.D., 1991, fabrics and sources of magnetic susceptibility in lower ization of deformation during exhumation of a major In-situ study of the eastern ridge-transform intersec- crustal and upper mantle rocks from Hess Deep, in strike-slip shear zone: Norumbega fault zone, south- tion of the Vema fracture zone: Tectonophysics, v. 190, Mével, C., Gillis, K.M., Allan, J.F., and Meyer, P.S., eds., central Maine, USA: Tectonophysics, v. 273, p. 185– p. 55–71, doi:10.1016/0040-1951(91)90354-U. Proceedings of the Ocean Drilling Program, Scientifi c 201, doi:10.1016/S0040-1951(96)00306-X. Martín-Hernández, F., and Ferré, E.C., 2007, Separation of Results, 147: College Station, Texas, Ocean Drilling Pro- Whattam, S.A., Malpas, J., Ali, J.R., and Smith, I.E.M., 2008, paramagnetic and ferrimagnetic anisotropies: A review: gram, p. 393–403. New SW Pacifi c tectonic model: Cyclical intraoceanic Journal of Geophysical Research, v. 112, doi:10.1029/ Robin, P.-Y.F., 2002, Determination of fabric and strain magmatic arc construction and near-coeval emplace- 2006JB004340. ellipsoids from measured sectional ellipses; theory: ment along the Australia-Pacifi c margin in the Ceno- Means, W.D., 1995, Shear zones and rock history: Tecto- Journal of Structural Geology, v. 24, p. 531–544, zoic: Geochemistry Geophysics Geosystems, v. 9, nophysics, v. 247, p. 157–160, doi:10.1016/0040-1951 doi:10.1016/S0191-8141(01)00081-5. Q03021, doi:10.1029/2007GC001710. (95)98214-H. Saad, A.H., 1969, Magnetic properties of ultramafi c rocks Wojtal, S.F., and Mitra, G., 1986, Strain hardening and strain Menard, H.W., and Atwater, T., 1969, Origin of fracture zone from Red Mountain, California: Geophysics, v. 34, softening in fault zones from foreland thrusts: Geo- topography: Nature, v. 222, p. 1037–1040, doi:10.1038/ p. 974–987, doi:10.1190/1.1440067. logical Society of America Bulletin, v. 97, p. 674–687, 2221037a0. Schellart, W.P., Lister, G.S., and Toy, V.G., 2006, A Late Creta- doi:10.1130/0016-7606(1986)97<674:SHASSI>2.0.CO;2. Moores, E.M., and Vine, F.J., 1971, The Troodos massif, ceous and Cenozoic reconstruction of the southwest Cyprus, and other ophiolites as oceanic crust; evalu- Pacifi c region: Tectonics controlled by subduction and ation and implications: Philosophical Transactions slab rollback processes: Earth-Science Reviews, v. 76, p. 191–233, doi:10.1016/j.earscirev.2006.01.002. of the Royal Society of London, ser. A, Mathematical MANUSCRIPT RECEIVED 5 SEPTEMBER 2010 Sdrolias, M., Mueller, R.D., and Gaina, C., 2003, Tectonic and Physical Sciences, v. 268, p. 443–467, doi:10.1098/ REVISED MANUSCRIPT RECEIVED 3 FEBRUARY 2011 evolution of the southwest Pacifi rsta.1971.0006. c using constraints MANUSCRIPT ACCEPTED 4 FEBRUARY 2011 Moutte, J., 1982, Chromite deposits of the Tiebaghi ultra- from backarc basins, in Hills, R.R., and Mueller, R.D., mafi c massif, New Caledonia: Economic Geology and eds., Evolution and Dynamics of the Australian Plate: Printed in the USA

244 www.gsapubs.org | Volume 3 | Number 3 | LITHOSPHERE

Downloaded from http://pubs.geoscienceworld.org/gsa/lithosphere/article-pdf/3/3/221/3050183/221.pdf by guest on 01 October 2021