Regional implications of U/Pb SHRIMP age constraints on the tectonic evolution of New Caledonia

ELSEVIER Tectonophysics 299 (1998) 333–343

Regional implications of U=Pb SHRIMP age constraints on thetectonic evolution of New Caledonia

J.C. Aitchison a,Ł, T.R. Ireland b, G.L. Clarke c, D. Cluzel d, A.M. Davis a, S. Meffre e

a Department of Earth Sciences, University of Hong Kong, Pokfulam Road, Hong Kong, Chinab Department of Geological and Environmental Sciences, Stanford University, Stanford, CA 94305-2215, USA

c Department of Geology and Geophysics, University of Sydney, Sydney, NSW 2006, Australiad Laboratoire de Geologie, Universite Francaise du Pacifique, BP 4477 Noumea, Nouvelle Caledoniee Department of Geology, University of Tasmania, GPO Box 252C, Hobart, Tasmania 7001, Australia

Received 2 December 1997; accepted 12 August 1998

Abstract

U=Pb SHRIMP ages for zircons from plagiogranites indicate a Late Carboniferous age of formation for ophioliticbasement of the Koh terrane in the Central Chain Mountains of New Caledonia. Samples from ophiolites at Koh and Kouayielded ages of 302 š 7 Ma and 290 š 5 Ma, respectively. The similarity of these ages to those of plagiogranites in theDun Mountain Ophiolite Belt=Maitai terrane of New Zealand, and the comparable structural positions of the two terranes,potentially constrain reconstructions of the Cretaceous SW Pacific margin. Detrital zircons from late Mesozoic sedimentsthat overlap the Koh terrane also provide potential constraints on the location of New Caledonia prior to its break-up anddispersal from eastern Australia. 1998 Elsevier Science B.V. All rights reserved.

Keywords: Koh terrane; Late Carboniferous; ophiolite; Cretaceous; Precambrian; inheritance; zircon; U=Pb; New Zealand;New Caledonia; Maitai terrane

1. Introduction

The islands of New Zealand and New Caledo-nia present the best exposures of the eastern mar-gin of the Indo–Australian Plate, which mostly liessubmerged beneath the SW Pacific Ocean. The col-lage of terranes that form the islands, and correla-tions between these and equivalent rocks in Aus-tralia and Antarctica, provide the basis for our un-derstanding of the Paleozoic to Tertiary evolution ofthe eastern Pacific region. Four main tectonic phases

Ł Corresponding author. Fax: C852 2517 6912; E-mail:[email protected]

are reflected in the onland geology of New Caledonia(Paris, 1981; Aitchison et al., 1995): (1) late Paleozoicto early Mesozoic development of subduction-relatedterranes that were accreted to the Gondwana marginto form the ‘basement’ terranes of New Caledonia(Fig. 1); (2) Cretaceous passive-margin developmentand sea-floor spreading during Gondwana breakup;(3) foundering of an oceanic basin and the Eocenearrival of thinned Gondwana margin crust at a south-west-facing subduction zone, resulting in collisionalorogenesis and obduction of an ophiolitic nappe fromthe northeast; (4) detachment faulting during exten-sional collapse, resulting in unroofing of metamor-phic core complexes (Clarke et al., 1997).

0040-1951/98/$ – see front matter 1998 Elsevier Science B.V. All rights reserved.PII: S 0 0 4 0 - 1 9 5 1 ( 9 8 ) 0 0 2 1 1 - X

334 J.C. Aitchison et al. / Tectonophysics 299 (1998) 333–343

Fig. 1. Map of New Caledonia showing the distribution of Koh terrane ophiolitic rocks and the locations from which plagiogranites weresampled.

The basement of New Caledonia (Fig. 1) incor-porates three major terranes of late Paleozoic toMesozoic age, each of which shows evidence of hav-ing developed in convergent-margin settings. Theseterranes developed along, or were accreted to, theeastern Gondwana margin prior to the Early Creta-ceous. The Upper Permian to Jurassic Teremba ter-rane, on the western side of the island, includes fos-siliferous, proximal, calc-alkaline island arc-derivedstrata (Paris, 1981; Campbell et al., 1985). Whereaswestern portions of the terrane include calc-alkalinevolcanic rocks, further east it is dominated by shal-

low to deep marine volcaniclastic sedimentary rocks.Enigmatic, predominantly fine-grained volcaniclas-tic sedimentary rocks lie to the east of what areunequivocal Teremba terrane rocks. These may com-prise yet another terrane, the ‘Moindou terrane’, butfurther work is required to assess this hypothesis.

Little-studied schistose rocks of the Boghen ter-rane are faulted against the eastern margin of theTeremba terrane. These mostly metafelsic sedimen-tary rocks have previously been referred to as the‘ante-Permian’ schists (Paris, 1981) because theycontain a foliation that is absent in the Teremba ter-

J.C. Aitchison et al. / Tectonophysics 299 (1998) 333–343 335

rane. Metabasic units (Cluzel, 1996) that are tectoni-cally interleaved within the sedimentary assemblagepossibly represent fragments of accreted seamounts.Despite numerous attempts to extract a microfauna,the age(s) of formation of rocks within this terraneremains uncertain. Late Jurassic metamorphism lo-cally reached blueschist facies (Blake et al., 1977),indicative of a subduction event of, as yet, indetermi-nate polarity.

The Koh terrane, discussed further below, cropsout in the rugged Central Chain of the island. TheLate Cretaceous ‘formation a charbon’ provides asedimentary assemblage that overlaps all the base-ment terranes and post-dates their accretion to themargin of Gondwana (eastern Australia).

2. Koh terrane

The Koh terrane comprises four main structuralblocks (Fig. 1) containing gabbros, quartz gabbros,diorites, dolerites, tholeiitic and boninitic pillowbasalts, and felsic volcanics that represent frag-ments of a supra-subduction zone (SSZ) ophiolite(Cameron, 1989; Meffre, 1991, 1995; Meffre et al.,1996). Whereas most mafic fragments formed in re-sponse to a single tholeiitic magmatic episode, theproducts of two tholeiitic magmatic episodes sepa-rated by boninites are preserved at Koh. The firstmagmatic episode produced cumulate gabbros, do-lerite, plagiogranites and a sequence of pillowedtholeiites. These rocks are overlain by a high-Caboninitic unit, which includes a basal section of boni-nite pillows, flows and breccias and an upper sectionof boninitic dacites and tuffs. The last magmaticepisode involved the eruption of evolved tholeiiticpillow basalts and the intrusion of tholeiitic dykesand sills into older plutonic and volcanic sectionsof the ophiolite. The petrogenesis and geochemistryof these rocks (Meffre et al., 1996), are consistentwith their formation in a SSZ environment similar tomodern backarc basins.

The upper tholeiites are conformably overlainby red and green pelagic siliceous siltstones thatdrape the uppermost pillows. These siltstones areup to 130 m thick and are succeeded by tuffaceoussiltstones, volcaniclastic sandstones and conglomer-ates. Approximately 1000 m above the ophiolite,

the volcaniclastic rocks are interbedded with mica-ceous black siltstones. An ammonoid fauna withinthese siltstones is indicative of a mid-Triassic age(H.J. Campbell, IGNS, New Zealand, written com-mun., 1990). The black siltstones are themselvesapproximately 1000 m thick and are overlain byfurther volcaniclastic sandstones and conglomeratesthat continue up section and form a Middle Triassicto Upper Jurassic sedimentary sequence reportedlyseveral km thick (Guerange et al., 1977; Paris, 1981;Maurizot et al., 1985).

Outside of their main occurrence at Koh, sev-eral other ophiolitic sequences are exposed alongthe Central Chain of New Caledonia (Meffre, 1995;Fig. 1). The northernmost body is the Cantaloupaıophiolite (Fig. 1), which contains both igneous andsedimentary rocks. A strong foliation containinglawsonite and fine-grained blue amphibole reflectsthe effects of an Eocene collisional event (Aitchisonet al., 1995). The Sphinx ophiolite, 50 km to thenorth of Koh, includes cumulate plutonic gabbros,plagiogranites, dolerite intrusions and a pillowedsection, up to 850 m thick, that is similar to the lowertholeiites at Koh. The Tarouimba ophiolite is a frag-ment that is offset 10 km to the northwest from themain Sphinx ophiolite by a sinistral strike slip fault.South of Koh, the Pocquereux, Koua and Nassirahophiolite outliers are volcanic and shallow intrusivesections that were disrupted into small fragments(1 to 3 km in length) during the Eocene collision(Aitchison et al., 1995). Well-developed dyke com-plexes are present in the three southern segments.Most of the ophiolite fragments are overlain by astratigraphic succession of red pelagic cherts andvolcaniclastic sediments similar to those at Koh.

The age of ophiolite fragments in the Koh ter-rane has not previously been determined, thoughthey must be older than the Anisian ammonoid-bearing strata, which crop out some distance abovethe igneous section. U=Pb SHRIMP dates from co-magmatic plagiogranites within the ophiolite at twolocalities are presented in this paper.

3. Late Mesozoic–Early Cenozoic overlap rocks

A transgressive sequence of Upper Cretaceouscoal measures and conglomerates (la formation a


charbons, Paris, 1981) overlies all three basementterranes with angular unconformity. Fluvial sedi-ments grade upsection into mineralogically matureshallow-marine sandstones containing Inoceramusfossils. This unit provides a minimum (pre-LateCretaceous) age constraint for the timing of amal-gamation of New Caledonian basement terranes toGondwana.

In the Central Chain area, near the village ofKoh, the Cretaceous and Lower Eocene sedimen-tary sequence has been examined in detail whereit overlies the Koh ophiolite on Mt Rembaı, andat Table Unio where it overlies the Boghen terrane(Meffre, 1991, 1995). Sedimentary rocks depositedabove the basal unconformity consist of a 20–30m thickness of sandy matrix-supported conglomer-ate containing rounded pebble-sized clasts of meta-morphic, volcanic and volcaniclastic rocks largelyderived from the three pre-Cretaceous terranes. Thematrix contains monocrystalline quartz, plagioclase,opaque minerals, epidote and carbonate cement. Theconglomerate is overlain by 50–180 m of quartz-richsandstone containing small bivalves from a marinedepositional environment. Turonian to Campanianinoceramids (Paris, 1981) and Coniacian to Campa-nian ammonites (Collignon, 1977) in shallow ma-rine sediments constrain deposition to an intervalof approximately 20 million years during the LateCretaceous. At the top of the sandstone unit thereis an increase in carbonate content and the rocksgrade into marls containing planktonic foraminifers.These marls are overlain by 100–170 m of fine-grained black chert (typically referred to as ‘phtanite’by French authors) containing radiolarians, spongespicules and rare diatoms in a fine-grained siliceousgroundmass.

The overall character of this unit is similarthroughout New Caledonia, although internal de-tails of units may vary. Evidence for development oflocalised basins or highs is seen in thickness vari-ations (e.g. 200–300 m at Mt Rembaı and TableUnio, to 1000–2000 m at the Col de la Boghenbetween Moindou and Bourail). In some areas thebasal conglomerate is up to 700–800 m thick (Mau-rizot et al., 1986) and the sandstone is overlain byfine-grained siltstone. Deltaic and possibly fluvial fa-cies, including cross-bedded sandstone interbeddedwith coal lenses, occur both in the Moindou area

and in the Central Chain (Gonord, 1970; Paris, 1981;Maurizot et al., 1986). In many areas the black chertrests directly on the pre-Cretaceous terranes with-out sandstone or conglomerate, and is interbeddedwith pelagic micrite containing pelagic foraminifersin a fine-grained calcareous groundmass. These ar-eas may represent former basement highs. The ageof the black chert has not been directly deter-mined, although radiolarians from reworked blackchert clasts collected from Eocene breccias suggestthat the cherts were deposited between the Campa-nian and the Late Paleocene (Bodorkos, 1994; Paul,1994). This is slightly older than the age generallyascribed to these rocks, based on their close associ-ation with micrites containing Paleocene to middleEocene planktonic foraminifers (Paris, 1981; Mau-rizot et al., 1986). The black chert may be olderthan the micrites, although this has not yet beenconfirmed from field observations.

In the Noumea and Diahot regions Upper Cre-taceous sandstones are interbedded with mafic andfelsic volcanics, including basalt, diabase, rhyolitesand ignimbrites (Paris, 1981; Maurizot et al., 1989;Black, 1995) which denote a short-lived series ofvolcanic events in the Late Cretaceous. Three dis-tinct volcanic horizons have been recognised in thelower part of the Pilou Formation in the NoumeaBasin (Noesmoen and Tissot, 1970; Black, 1995).Rhyolitic and basaltic lavas together with associ-ated volcaniclastic sediments constitute a bimodaligneous association. Geochemical analyses (Black,1995) show a bimodal association dominated byhigh-Si rhyolites with minor accompanying calc-alkaline basalts. Mafic rocks exhibit an evolutionarytrend towards shoshonitic compositions.

4. Samples

Small volumes of plagiogranite, widely consid-ered to represent late stage differentiates of morebasic magmas, are common in most ophiolites. Assuch granites commonly contain minor amounts ofzircon they are useful for dating ophiolite forma-tion. Plagiogranites from the Koh terrane are com-posed predominantly of granophyric-textured inter-growths of albitic plagioclase and quartz, with vari-able abundances of amphibole. Accordingly, they


include quartz diorites, tonalites and trondhjemites.Clear intrusive relationships can be demonstratedbetween the plagiogranites and more basic igneouscomponents of the ophiolite. Two samples contain-ing zircons were collected. One is from the Kohophiolite at a locality within the Koh River (GR58397613); the other is from the Koua ophiolite, and wascollected from within a small tributary of the KouaRiver (GR6124 7594).

Aronson and Tilton (1971) reported Precambriandetrital zircons in the Cretaceous formation a char-bons which could provide important informationabout the former position of New Caledonia alongthe eastern margin of Australia. Aronson and Tiltonanalyzed gram-size samples from 1000 lb of col-lected sample and so it is difficult to assess thediscordant data from this sandstone. Therefore, weextracted and analysed age spectra for detrital zir-cons from a sandstone sample collected from thisunit at Table Unio.

5. Experimental technique and results

Zircons were separated, mounted in epoxy, andpolished to expose interior sections. U–Th–Pb anal-yses were carried out on the SHRIMP I ion mi-croprobe at the Australian National University, fol-lowing standard operating procedures (e.g. Williamsand Claesson, 1987). Pb=U analysis of the sampleswas calibrated using an empirical quadratic rela-tionship between 206PbC=238UC and UOC=UC, andnormalised to 206Pb*=U of 0.09279 for the 572 MaSL13 zircon standard. U and Th concentrations wereestimated relative to their concentrations in the SL13standard and are accurate to š20%. The commonPb contribution to the analyses was determined fromthe 207Pb=206Pb ratio, and hence age information isprovided by the 206Pb=238U intercept. The composi-tion of common Pb used is that of Cumming andRichards (1975; 300 Ma model Pb). The method isdocumented in more detail by Muir et al. (1996).

Results for both plagiogranite samples are listedin Table 1 and illustrated in Fig. 2 in Tera–Wasserburg concordia plots. Fourteen analyses weremade on both samples. Replicate analyses of Kohgrains were required because of a paucity of suitablematerial free from cracks and inclusions. Common

Pb contributions range up to 6.6% of the total 206Pbfor Koua and up to 1.7% for Koh. All data lie within2¦ uncertainty of the radiogenic–common Pb mix-ing line (dashed in Fig. 2). Ages are reported asweighted means of the extrapolated 206Pb=238U agesat concordia. For both samples, all data are includedin the weighted means, and the MSWD is consistentwith a single magmatic population with a small pro-portion of common Pb. The ages, 290 š 5 Ma forKoua and 295 š 7 Ma .2¦/ for Koh, agree withinanalytical uncertainties. Koua zircons have lower Thand U concentrations and more uniform Th=U ratiosthan Koh zircons (67–220 ppm U, 21–99 ppm Th,Th=U 0.29–0.45 for Koua versus 140–950 ppm U,180–1620 ppm Th, Th=U 0.75–2.82 for Koh.

Results for analyses of 50 detrital zircons from theformation a charbons arkose sample are illustratedon a cumulative probability histogram presented inFig. 3. These ages comprise concordant 206Pb=238Uages for the younger clear grains, and the 207Pb=206Pbages for the older darker grains because these twograins are discordant (both have 206Pb=238U ages ofaround 700 Ma). The younger clear and euhedralgrains show a range of ages with two distinct pop-ulations, one around 90–140 Ma, the other 170–240Ma. Over half the grains are 90–105 Ma old.

6. Discussion

Previous workers had considered the Koh ophi-olite, a significant and widely distributed portionof the basement terranes of New Caledonia, to beLate Permian or Triassic. However, the Late Car-boniferous SHRIMP U=Pb age reported herein forconstituent plagiogranites substantially revises ourknowledge of the time of formation for this ophiolite.The revised age demands a reassessment of plausiblecorrelations of terranes from the continental frag-ments that once constituted portions of the easternmargin of Gondwana. The new dates are the oldestknown for any rocks in New Caledonia, though sig-nificantly older detrital zircons occur in sedimentsof the Cretaceous overlap sequence (Aronson andTilton, 1971; results published herein).

In any reconstruction of the Mesozoic SW Pacificit is tempting to investigate potential correlations be-tween the few areas of continental crust presently lo-


Table 1U–Th–Pb data for Koua and Koh plagiogranites

No. U Th Th=U f 206Pb 207Pb=206Pb 238U=206Pb Age(ppm) (ppm) (%) (Ma)

Koua1.1 67 21 0.31 6.57 š 0.47 0.1055 š 0.0038 19.71 š 0.78 298.6 š 11.72.1 174 75 0.43 1.44 š 0.16 0.0635 š 0.0013 22.02 š 0.40 282.3 š 5.13.1 125 46 0.36 1.91 š 0.42 0.0676 š 0.0033 21.26 š 0.48 290.7 š 6.64.1 137 53 0.39 1.54 š 0.33 0.0647 š 0.0026 20.84 š 0.52 297.6 š 7.45.1 111 35 0.31 2.44 š 0.22 0.0719 š 0.0018 21.01 š 0.50 292.6 š 6.96.1 218 99 0.45 0.96 š 0.16 0.0599 š 0.0013 21.38 š 0.45 292.0 š 6.07.1 138 48 0.35 2.21 š 0.25 0.0699 š 0.0020 21.47 š 0.45 287.1 š 5.98.1 128 51 0.40 2.32 š 0.28 0.0708 š 0.0023 21.46 š 0.46 287.0 š 6.19.1 116 33 0.29 1.82 š 0.39 0.0669 š 0.0032 21.15 š 0.67 292.6 š 9.1

10.1 164 67 0.41 2.09 š 0.26 0.0690 š 0.0021 21.27 š 0.52 290.1 š 7.011.1 109 32 0.30 1.89 š 0.28 0.0673 š 0.0023 21.43 š 0.50 288.6 š 6.612.1 118 36 0.31 2.42 š 0.30 0.0718 š 0.0024 20.85 š 0.38 294.9 š 5.413.1 195 80 0.41 1.53 š 0.27 0.0644 š 0.0022 21.44 š 0.41 289.5 š 5.414.1 124 53 0.43 2.53 š 0.20 0.0725 š 0.0016 21.26 š 0.59 288.9 š 7.8

Koh1.1 226 215 0.95 0.60 š 0.15 0.0571 š 0.0012 22.11 š 3.69 283.5 š 46.41.2 952 719 0.75 0.33 š 0.07 0.0550 š 0.0006 20.57 š 0.48 305.0 š 7.02.1 355 943 2.65 1.71 š 0.33 0.0661 š 0.0027 20.55 š 0.68 301.2 š 9.83.1 947 1125 1.19 0.98 š 0.09 0.0602 š 0.0007 22.27 š 0.93 280.5 š 11.44.1 268 328 1.22 0.99 š 0.18 0.0603 š 0.0014 21.65 š 1.06 288.3 š 13.95.1 157 262 1.66 0.43 š 0.22 0.0558 š 0.0018 21.74 š 0.66 288.6 š 8.56.1 347 685 1.97 1.06 š 0.16 0.0608 š 0.0012 21.26 š 0.49 293.2 š 6.66.2 374 772 2.06 0.55 š 0.16 0.0567 š 0.0013 21.31 š 0.60 294.0 š 8.07.1 515 828 1.61 1.64 š 0.14 0.0654 š 0.0011 19.97 š 0.40 309.9 š 6.18.1 136 180 1.32 1.01 š 0.25 0.0604 š 0.0020 21.70 š 0.74 287.6 š 9.69.1 574 1619 2.82 1.59 š 0.22 0.0651 š 0.0017 21.37 š 0.89 290.3 š 11.99.2 411 1105 2.69 1.59 š 0.16 0.0651 š 0.0012 20.86 š 0.93 297.2 š 13.0

10.1 429 323 0.75 0.40 š 0.15 0.0555 š 0.0012 23.02 š 0.71 273.1 š 8.311.1 670 1123 1.68 0.47 š 0.11 0.0561 š 0.0009 20.95 š 0.53 299.3 š 7.4

All analytical errors are 1¦ .238U=206Pb and 207Pb=206Pb are the measured values uncorrected for common Pb.f 206Pb is the proportion of common Pb relative to total measured Pb.Age is based on 238U=206Pb corrected for common Pb.

cated east of Australia. The most likely area in whichextensions of New Caledonian terranes might occurlies to the south in New Zealand. Previous studieshave demonstrated similarities in fossil content, sed-imentation, age and inferred former tectonic settingsbetween terranes in New Caledonia, and those ofsimilar ages in New Zealand (Benson, 1928; Parisand Bradshaw, 1977; Cawood, 1984; Campbell et al.,1985; Waterhouse and Sivell, 1987). In particular,the Teremba terrane exhibits numerous similaritieswith the Murihiku terrane (of N.Z.). These includefauna (Campbell and Grant-Mackie, 1984; Ballanceand Campbell, 1993), palynomorphs (de Jersey and

Grant-Mackie, 1989), a volcaniclastic source dom-inated by andesite (Roser and Korsch, 1988), anda similar age range, although slightly older rocksare exposed in the Teremba terrane (Campbell andGrant-Mackie, 1984). Notable differences includethe presence of granitic and other continentally de-rived clasts in a thicker stratigraphic sequence in theMurihiku terrane (Boles, 1974; Frost and Coombs,1989; Ballance and Campbell, 1993). Calc-alkalinevolcanic rocks such as those at the base of theTeremba terrane are unknown from the Murihikuterrane, although they do provide a potential sourcefor many of the sediments there. Variations in the


Fig. 2. Tera–Wasserburg concordia plots for SHRIMP analysesof U=Pb from zircons extracted from Koh and Koua plagiogran-ites. Data are uncorrected for common Pb contributions and alldata from their respective samples plot on a mixing line be-tween concordant radiogenic Pb and common Pb indicating thata single magmatic component is required.

stratigraphic thicknesses of rocks in the two terranesmay be related to local facies control or section at-tenuation resulting from local faulting. Despite theseminor differences, the evidence for the correlation ofthe two terranes is compelling.

It has been further suggested (Harrington, 1983;Waterhouse and Sivell, 1987) that the volcanic rocksat Teremba may be correlated with those of theBrook Street terrane in New Zealand. However,this correlation is not supported by geochemicaldata, which indicate a distinctly different island arctholeiitic composition for the Brook Street terranevolcanics. More likely equivalents of these rocksoccur in the Gympie district of eastern Queenslandwhere the Highbury Volcanics (Sivell and Water-house, 1988) are petrographically and geochemicallysimilar and are of comparable age. No likely correla-tives are known from New Caledonia.

The correlation of other terranes in New Zealandand New Caledonia, which lie outboard of theTeremba=Murihiku terrane, is more difficult. Thisis in part related to a lack of detailed study of the lessfossiliferous basement terranes in the interior of NewCaledonia, and the poorly exposed basement terranesof northernmost New Zealand. Until recently rocksof the Koh terrane were not regarded as an ophioliticassemblage. Furthermore, because of their inferredTriassic or possibly Permian age they did not appearas obvious candidates for correlation with any NewZealand terrane. Recognition of the supra-subduc-tion zone affinities of rocks at the stratigraphic baseof the Koh terrane (Meffre et al., 1996) and the LateCarboniferous age data presented herein (which con-strain the age of formation) provide the basis for thereconsideration of possible correlations.

Three large-scale terranes to the south of Auck-land, New Zealand (from west to east) are the Muri-hiku, Dun Mountain=Matai and Waipapa terranes(Fig. 4). These can be traced, at least in subsurface,through Northland to the northern tip of New Zealand(Sporli, 1978; Black, 1994). More recently, Black(1997), has recognised several further terranes withinrocks previously mapped (Black, 1994) as Waipapaterrane in the Northland region. The structural dis-position of terranes in Northland may be repeatedin New Caledonia because late Paleozoic ophioliticrocks lie east of the Teremba=Murihiku terranes inboth places. Late Carboniferous fragments of theKoh ophiolite are potential equivalents of the DunMountain=Maitai terrane. U=Pb analyses of zirconfrom plagiogranites of the Dun Mountain ophiolitehave yielded marginally younger ages of around 285Ma (Kimbrough et al., 1992). Geochemical data in-dicate that both the Koh and Dun Mountain=Maitaiterranes are likely to have formed in similar tectonicsettings (Moore, 1991; Meffre et al., 1996).

Isotopic age constraints notwithstanding, differ-ences do exist between the two terranes. In particular,the Dun Mountain ophiolite is overlain by a thicksuccession of volcaniclastic sandstones of the MaitaiGroup which, in their lower part, are characterisedby mappable units of atomodesmatinid limestone. Nosuch limestone is known from the Koh terrane. Earlyworkers considered the Maitai Group to be entirelyPermian but more recent work in New Zealand (Pillaiet al., 1991; Owen, in prep.) has shown that much of


Fig. 3. Histogram of detrital ages in the sample from la formation a charbons, The younger ages are 206Pb=238U ages plotted ona cumulative probability plot (each zircon is given a unit area, Gaussian, and the contributions are summed into bins). Two mainpopulations are evident 90–140 Ma and 170–220 Ma with over half the grains being consistent with a ca. 95 Ma source. The inset showsa histogram for all zircons showing two older grains approaching 1000 Ma.

it is Triassic. The Triassic volcaniclastic sections arenot dissimilar to those which overlie the Koh Ophio-lite and it is possible that lower portions of the NewCaledonia section have been removed by faulting.

In New Zealand the Dun Mountain=Maitai andMurihiku terranes are everywhere juxtaposed alonga faulted contact with no intervening terranes, butin New Caledonia the structural position of the Kohterrane differs somewhat from that of its potentialcorrelatives in New Zealand. Koh terrane rocks notonly lie to the east of Murihiku-like rocks, but alsolie east of schistose rocks of the Boghen terrane. TheBoghen terrane of New Caledonia represents an as-semblage of rocks similar to those seen to the east ofthe Dun Mountain=Maitai terrane in New Zealand.The general characteristics of rocks of the Boghenterrane appear to be similar to some of those from theOmahuta terrane (Black, 1997). Metamorphic miner-als in both terranes record high P=T metamorphismin the Late Jurassic; further study of these potentialcorrelations is warranted.

Detrital zircon data present an intriguing prob-lem. Existing tectonic reconstructions would placelate Mesozoic New Caledonia to the east of theNew England Orogen (NEO) of eastern Australia,slightly south of present-day Brisbane. The spectrum

of ages from detrital zircons is thus surprising. Ifthe reconstruction is correct, one might expect tosee populations of grains clearly derived from theNEO. However, despite the presence of voluminouslate Paleozoic=early Mesozoic igneous rocks withinthe NEO there are only a few grains with apparentTriassic ages, and no Paleozoic grains.

There are several possible explanations for thelack of NEO-derived material. Amongst these are:New Caledonia was not in the position that hasbeen suggested; overlap assemblage sandstones ofNew Caledonia were transported laterally along themargin of Gondwana from a location further to thenorth; sediment somehow bypassed the NEO; and=orthe NEO was itself masked by covering strata ofonce more extensive Sydney–Bowen and Clarence–Morton basins.

The above ages are also found in younger (Cre-taceous–Jurassic) Torlesse sandstones. The forma-tion a charbon is far more mature than the Torlessegraywackes but a similar provenance is possible. Sucha land link would need to be found if the Torlesse isindeed sourced from Queensland (Adams and Kelly,1998). The presence of grains of ca. 1000 Ma con-firms the inferences of Aronson and Tilton (1971) thata Precambrian source was contributing to the detritus.


Fig. 4. Reconstruction of the eastern margin of Gondwana at 100 Ma just after the accretion of the Late Carboniferous to EarlyCretaceous arc-related terranes and prior to the opening of the Tasman Sea and the New Caledonia Basin (after Meffre, 1995). Plateboundaries and positions after Yan and Kroenke (1993), New Zealand terranes after Sporli and Ballance (1989), New England Orogenand Carboniferous arc system after Day et al. (1978).

This age component is familiar throughout southeast-ern Australian detritus (see for example Ireland etal., 1998); however, it is always associated with amore dominant 500–650 Ma age peak which is lack-ing from the New Caledonia sandstone. This wouldsuggest that a continental source unaffected by theDalemarian Orogen must be found. The most proxi-mal location would be northern Queensland, althoughthe potential for Neoproterozoic grains is unknown.

Acknowledgements

This research was supported by the fundingthrough grants from the Australian Research Council

(JA and GC, A39600827), Australian PostgraduateAwards Scheme (SM) and the Australian Govern-ment Department of Trade, Industry and Commerce(JA).

References

Adams, C.J., Kelly, S., 1998. Provenance of Permian–Triassicand Ordovician metagraywacke terranes in New Zealand: evi-dence from 40Ar=39Ar dating of detrital micas. Geol. Soc. Am.Bull. 110, 422–432.

Aitchison, J., Clarke, G.L., Meffre, S., Cluzel, D., 1995. Eocenearc–continent collision in New Caledonia and implicationsfor regional Southwest Pacific tectonic evolution. Geology 23,161–164.


Aronson, J.L., Tilton, G.R., 1971. Probable Precambrian detritalzircons in New Caledonia and southwest Pacific continentalstructure. Geol. Soc. Am. Bull. 82, 3449–3456.

Ballance, P.F., Campbell, J.D., 1993. The Murihiku arc-relatedbasin of New Zealand (Triassic–Jurassic). In: Ballance, P.F.(Ed.), South Pacific Sedimentary Basins. Sedimentary Basinsof the World. Elsevier, Amsterdam, pp. 21–33.

Benson, W.N., 1928. The Mesozoic rocks of New Caledoniaand their relation to those of New Zealand. Third Pan PacificScience Congress, Tokyo, pp. 1701–1703.

Black, P.M., 1994. The ‘Waipapa Terrane’, North Island, NewZealand: subdivision and correlation. Geosci. Rep. ShizuokaUniv. 20, 55–62.

Black, P.M., 1995. High-Si rhyolites and shoshonitic volcanics;a Late Cretaceous bimodal association, Noumea Basin, NewCaledonia. In: Mauk, J.L., St-George, J.D. (Eds.), Proceedingsof the 1995 PACRIM congress; Exploring the Rim. Aus-tralasian Institute of Mining and Metallurgy, 9=95, pp. 55–58.

Black, P.M., 1997. Permian–Jurassic ‘basement’ terranes of theNorfolk Ridge system: Northland Peninsula (New Zealand)and New Caledonia. In: Bradshaw, J.D., Weaver, S.D. (Eds.),International Conference on Terrane Geology. University ofCanterbury, Christchurch, pp. 19–21.

Blake, M.C., Brothers, R.N., Lanphere, M.A., 1977. Radiometricages of blueschists in New Caledonia. In: International Sym-posium on Geodynamics in the South West Pacific. EditionTechnip, Paris, pp. 276–282.

Bodorkos, S., 1994. Tectonic Evolution of South Western NewCaledonia: Evidence from the Noumea–Boulouparis Basin.B.Sc. (Hons) Thesis, University of Sydney.

Boles, J.R., 1974. Structure, stratigraphy, and petrology ofmainly Triassic rocks, Hokonui Hills, Southland, NewZealand. N. Z. J. Geol. Geophys. 17, 337–374.

Cameron, W.E., 1989. Contrasting boninite–tholeiite associationsfrom New Caledonia. In: Crawford, A.J. (Ed.), Boninites.Unwin Hyman, London, pp. 314–336.

Campbell, H.J., Grant-Mackie, J.A., 1984. Biostratigraphy of theMesozoic Baie de St.-Vincent Group, New Caledonia. J. R.Soc. N. Z. 14, 349–366.

Campbell, H.J., Grant-Mackie, J.A., Paris, J.P., 1985. Geology ofthe Moindou–Teremba area, New Caledonia; stratigraphy andstructure of Teremba Group (Permian–Lower Triassic) andBaie de St.-Vincent Group (Upper Triassic–Lower Triassic).Geologie de la France, 1985, pp. 19–36.

Cawood, P.A., 1984. The development of the SW Pacific marginof Gondwana: correlations between the Rangitata and NewEngland Orogens. Tectonics 3, 539–553.

Clarke, G., Aitchison, J., Cluzel, D., 1997. Eclogites andblueschists of the Pam Peninsula, NE New Caledonia: a reap-praisal. J. Petrol. 36, 843–876.

Cluzel, D., 1996. Affinites intra-oceaniques des metavolcan-ites de l’unite de la Boghen (ex-‘ante-Permien’ de Nou-velle-Caledonie, Pacifique sud-ouest); consequences paleogeo-graphiques. C. R. Acad. Sci., Paris, Ser. II 323, 657–664.

Collignon, M., 1977. Ammonites neocretacees de la Nouvelle-Caledonie. Bull. B.R.G.M., Fr., Sect. 4, 1, 7–36.

Cumming, G.L., Richards, J.R., 1975. Ore lead isotope ratiosin a continuously changing Earth. Earth Planet. Sci. Lett. 28,155–171.

Day, R.W., Murray, C.W., Whitaker, W.G., 1978. The easternpart of the Tasman Orogenic Zone. Tectonophysics 48, 327–364.

de Jersey, N.J., Grant-Mackie, J.A., 1989. Palynofloras from thePermian, Triassic and Jurassic of New Caledonia. N. Z. J.Geol. Geophys. 32, 463–476.

Frost, C.D., Coombs, D.S., 1989. Nd isotope character of NewZealand sediments: implications for terrane concepts andcrustal evolution. Am. J. Sci 289, 744–770.

Gonord, H., 1970. Sur la presence d’olistolites et sur la miseen place probable de nappes de glissement dans le flyscheocene du bassin tertiaire de Noumea–Bouloupari (Nouvelle-Caledonie). C. R. Acad. Sci. Paris, Ser. D 270, 3010–3013.

Guerange, B., Lozes, J., Autran, A., 1977. Mesozoic metamor-phism in the New Caledonia central chain and its geodynamicimplications in relation to the evolution of the Cretaceous Ran-gitata Orogeny. In: International Symposium on Geodynamicsin South-West Pacific. Technip, Paris, pp. 265–278

Harrington, H.J., 1983. Correlation of the Permian and TriassicGympie Terrane of Queensland with the Brook Street andMaitai Terranes of New Zealand. In: Murray, C.G. (Ed.),Permian Geology of Queensland. Geol. Soc. Aust., Brisbane,pp. 431–436.

Ireland, T.R., Flottmann, T., Fanning, C.M., Gibson, G.M.,Preiss, W.V., 1998. Development of the lower-Paleozoic Pa-cific margin of Gondwana from zircon-age structure across theDelamerian orogen. Geology 26, 243–246.

Kimbrough, D.L., Mattinson, J.M., Coombs, D.S., Landis, C.A.,Johnston, M.R., 1992. Uranium–lead ages from the DunMountain Ophiolite Belt and Brook St Terranes, South Is-land, New Zealand. Geol. Soc. Am. Bull. 104, 429–443.

Maurizot, P., Paris, J.P., Feignier, D., 1985. Paleogeographie depart et d’autre de l’Accident Ouest Caledonien durant la peri-ode Cretace superieur–Paleocene: autochtonie de la Formationdes Basaltes de la cote ouest. Geologie de la France, 1985, pp.53–60.

Maurizot, P., Feignier, D., Schmid, M., Yerle, J.-J., 1986. Carteet notice explicative de la carte geologique de la Nouvelle-Caledonie a l’echelle du 1=50 000: feuille Paeoua. Bureau deRecherches Geologiques et Minieres, Paris.

Maurizot, P., Eberle, J.-M., Habault, C., Tessarollo, C., 1989.Carte et notice explicative de la carte geologique de la Nou-velle-Caledonie a l’echelle du 1=50 000: feuille Ouegoa. Bu-reau de Recherches Geologiques et Minieres, Paris.

Meffre, S., 1991. Terrane Analysis of New Caledonia with Spe-cial Reference to the Koh Area. B.Sc. (Hons) Thesis, Univer-sity of Sydney.

Meffre, S., 1995. Island Arc-Related Ophiolites and SedimentarySequences in New Caledonia. PhD. Thesis, University ofSydney.

Meffre, S., Aitchison, J.C., Crawford, A.J., 1996. Geochemi-cal stratigraphy of boninites and tholeiites from the Permo–Triassic Koh Ophiolite, New Caledonia. Tectonics 15, 67–83.

Moore, P.J., 1991. Geology and Geochemistry of Mafic Rocks


from Ophiolites of East Nelson, New Zealand. PhD. Thesis,Memorial University of Newfoundland.

Muir, R.J., Ireland, T.R., Weaver, S.D., Bradshaw, J.D., 1996.Ion microprobe dating of Paleozoic granitoids: Devonian mag-matism in New Zealand and correlations with Australia andAntarctica. Chem. Geol. (Isot. Geosci. Sect.) 127, 191–210.

Noesmoen, A., Tissot, B. 1970. Carte et notice explicative dela carte geologique de la Nouvelle-Caledonie a l’echelle du1=50 000: feuille Noumea. Bureau de Recherches Geologiqueset Minieres, Paris.

Paris, J.P., 1981. La Geologie de la Nouvelle-Caledonie, un essaide synthese. Mem. B.R.G.M. 133, 250 pp.

Paris, J.P., Bradshaw, J.D., 1977. Paleogeography and geotecton-ics of New Caledonia and New Zealand in the Triassic andJurassic. In: International Symposium of Geodynamics in theSouth West Pacific. Editions Technip, Noumea, pp. 209–216.

Paul, A.E.H. 1994. Sedimentological and Biostratigraphical Con-straints on Arc–Continent Collisions in New Caledonia. B.Sc.(Hons) Thesis, University of Sydney.

Pillai, D.D.L., Landis, C.A., Cook, N.D.J., Campbell, H.J., 1991.Permian ammonoids from the Greville Formation, Upukeroravalley, Southland. N. Z. J. Geol. Geophys. 34, 365–372.

Roser, B.P., Korsch, R.J., 1988. Provenance signatures ofsandstone–mudstone suites determined using discriminant

function analysis of major-element data. Chem. Geol. 67,119–139.

Sivell, W.J., Waterhouse, J.B., 1988. Petrogenesis of GympieGroup volcanics; evidence for remnants of an Early Permianvolcanic arc in eastern Australia. Lithos 21, 81–95.

Sporli, K.B., 1978. Mesozoic tectonics, North Island, NewZealand. Geol. Soc. Am. Bull. 89, 415–425.

Sporli, K.B., Ballance, P.F., 1989. Mesozoic oceanfloor=continent interaction and terrane configuration, south-west Pacific area around New Zealand. In: Ben-Avraham, Z.(Ed.), The Evolution of the Pacific Ocean Margins. OxfordMonogr. Geol. Geophys. 8, 176–189.

Waterhouse, J.B., Sivell, W.J., 1987. Permian evidence for trans-Tasman relationships between East Australia, New Caledoniaand New Zealand. Tectonophysics 142, 227–240.

Williams, I.S., Claesson, S., 1987. Isotopic evidence for thePrecambrian provenance and Caledonian metamorphism ofhigh grade paragneisses from the Seve Nappes, ScandinavianCaledonides, II. Ion microprobe zircon U–Th–Pb. Contrib.Mineral. Petrol. 97, 205–217.

Yan, C.Y., Kroenke, L.W., 1993. A plate reconstruction of thesouth west Pacific, 0–100 Ma. Proc. ODP, Sci. Results 130,697–709.

Regional implications of U/Pb SHRIMP age constraints on the tectonic evolution of New Caledonia

Documents