Directory UMM :Data Elmu:jurnal:P:Precambrian Research:Vol105.Issue2-4.2001:

(1)

A new fragment of the early earth crust: the Aasivik terrane

of West Greenland

Minik T. Rosing

a,b,

_{*, Allen P. Nutman}

_{, Linea Løfqvist}

a,b

a_Kobenha

6ns Uni6ersitet,Geologisk Museum,Øster Voldgade5-7,DK-1350Kobenha6n K,Denmark b_{Danish Lithosphere Center}_,_{Øster Voldgade}_10,_DK_-1350_Kobenha

6n K,Denmark c_RSES_,_{The Australian National Uni}

6ersity,Canberra ACT0200,Australia

Received 12 May 1999; accepted 27 August 1999

Abstract

The Aasivik terrane is a 1500 km2_{complex of gneisses dominated by} _{3600 Ma components, which has been} discovered in the Archaean craton of West Greenland, 20 – 50 km south of the Paleoproterozoic Nagssugtoqidian orogen. The Aasivik terrain comprises granulite facies tonalitic to granitic gneisses with bands of mafic granulite, which include disrupted mafic dykes. Four gneiss samples of the Aasivik terrain have given imprecise SHRIMP U – Pb zircon ages of 3550 – 3780 Ma with strong loss of radiogenic lead and new growth of zircon probably associated with a granulite facies metamorphic event(s) at 2800 – 2700 Ma. To the Southeast, the Aasivik terrane is in tectonic contact with a late Archaean complex of granitic and metapelitic gneisses with apparently randomly distributed mafic and ultramafic units, here named the Ukaleq gneiss complex. Two granitic samples from the Ukaleq gneiss complex have U – Pb zircon ages of 2817910 and 2820912 Ma andtzirconoNdvalues of 2.3 – 5.4. Given their composition and positiveo

Ndvalues, they probably represent melts of only slightly older juvenile crust. A reconnaissance SHRIMP U – Pb study of a sample of metasedimentary rock from the Ukaleq gneiss complex found 2750 – 2900 Ma zircons of probable detrital origin and that two or more generations of 2700 – 2500 Ma metamorphic zircons are present. This gneiss complex is provisionally interpreted as a late Archaean accretionary wedge. A sample of banded granulite facies gneiss from a complex of banded gneisses south of the Aasivik terrain here named the Tasersiaq gneiss complex has yielded two zircon populations of 32129 11 and 3127912 Ma. Contacts between the three gneiss complexes are mylonites which are locally cut by late-post-kinematic granite veins with SHRIMP U – Pb zircon ages of 2700 Ma. The isotopic character and the relationships between the lithologies from the different gneiss complexes suggest the assembly of unrelated rocks along shear zones between 2800 and 2700 Ma. The collage of Archaean gneiss complexes were intruded by A-type granites, here named the Umiatsiaasat granites, at 2700 Ma, later than the tectonic intercalation of the gneiss complexes. © 2001 Elsevier Science B.V. All rights reserved.

Keywords:Aasivik terrane; SHRIMP U – Pb; Ukaleq gneiss complex; Umiatsiaasat granites

www.elsevier.com/locate/precamres

* Corresponding author. Tel.: +45 3532 2345; fax: +45 3532 2325. E-mail address:minik@savik.geomus.ku.dk (M.T. Rosing).

(2)

1. Introduction

Rock complexes with ages in excess of 3600 Ma have been identified in all the continents (Black et al., 1971, 1986; McGregor, 1973; Moorbath et al., 1972, 1973; Baadsgaard, 1973; Bowring et al., 1989; Bridgwater and Schiøtte, 1991; Myers, 1988; Nutman et al., 1991; Kinny and Nutman, 1996; Krylov et al., 1989; Song et al., 1996; Compston and Kro¨ner, 1988). However, the ancient compo-nents of the conticompo-nents are in general found as small enclaves, which are commonly migmatised by younger felsic components. The largest contin-uous \3600 Ma complex is the West Greenland Itsaq Gneiss Complex, which underlies 3000 km2_{. Friend et al. (1988) and Nutman et al. (1996)} showed that the \3500 Ma rocks were confined to the so-called Akulleq terrane, which is bounded to the Northwest by the Akia terrain and to the Southeast by the Tasiusarssuaq terrain (Friend et al., 1988; Nutman et al., 1996). The Akulleq terrain is dominated by tonalite – trondhjemite – granodiorite (TTG) gneisses metamorphosed at amphibolite facies during the early Archaean. The Early Archaean rocks of the Godtha˚bsfjord area thus represent allochtonous mid-crustal segments from an early Archaean continent. Further insight into the thickness structure and composition of this early Archaean continental mass can be gained by finding further segments of \3600 Ma rocks within the West Greenland Archaean block. The three terrains of the Godtha˚b region are separated by Archaean mylonites and have shared the same geologic history since2700 Ma, which can be regarded as the age of final accretion of that cratonic block. Currently, the northern limit of the Akia terrain is uncertain. Further north-wards, Archaean rocks are penetratively deformed in the southern part of the Paleoproterozoic Nagssugtoqidian orogen (Hanmer et al., 1997 and references therein) (Fig. 1).

The region bordering the inland ice 50 km south of the Nagssugtoqidian front and 100 km north of the northern most exposure of the Akulleq terrain can be divided into a number of domains, based on Landsat image texture, stream sediment geochemistry (Steenfelt, 1994), occur-rence and deformation state of Paleoproterozoic

Fig. 1. Geologic map of central West Greenland showing the locations of early Archaean terranes in the Akulleq belt and the inferred extend of the Aasivik terrane shown in dark grey.

mafic dykes (Escher et al., 1970) and occurrence of Mesozoic kimberlite and carbonatite intrusions (Larsen et al., 1983). A large domain shown as the Aasivik terrane in Fig. 2, shows contrasts in all

Fig. 2. Map showing the distribution of topographic linea-ments, kimberlite intrusions and the inferred outline of the Aasivik terrane.

(3)

Fig. 3. Preliminary geologic map of the early Archaean Aasivik terrane and adjoining gneiss complexes.

these characters relative to the surrounding mains, and it is separated from bordering do-mains by pronounced topographic lineaments. We suggest that this domain is a tectonic terrane dominated by early Archaean rocks.

Through most of this millennium the area has been favoured by Eskimo caribou hunters, and traces of their summer camps are frequently en-countered. We have therefore chosen the name Aasivik (= summer camp) for the terrane.

2. Geology

We have identified five crustal components in the studied area (Figs. 2 and 3) (1) The Aasivik terrane, (2) the Ukaleq gneiss complex, (3) The Tasersiaq gneiss complex, (4) The Umiatsiaasat

granite plutons and (5) Mafic dykes tentatively correlated with the Kangaˆmiut dykes in the ad-joining areas, but here referred to as ‘‘metadoler-ite dykes’’.

2.1. Aasi6ik terrain

The Aasivik terrain is dominated by garnet bearing two pyroxene granulite facies or-thogneisses. The gneisses have a layered appear-ance defined by 1 – 50 m thick, planer felsic macro units with slightly varying colour index, interca-lated with boudinaged mafic units ranging from a few centimetres to 50 m thick. At one locality, the gneisses are crosscut by fractures each rimmed by up to 2 m in wide haloes of amphibolite facies retrogression. In these zones, hornblende and bi-otite outline a mm – cm scale banding, which

(4)

re-veal very complicated fold patterns internally in the macro units. The retrogression was not ac-companied by any deformation other than frac-turing. We therefore assume that the large gneiss units were generally strongly deformed relative to their protolith, but that the granulite facies meta-morphic recrystallisation obscured complex struc-tures formed before or during metamorphism. This has implications for geochemical sampling, since even apparently homogeneous granulite fa-cies gneiss samples may represent a mixture of several chemical and chronostratigraphic end members.

The structures outlined by the alternation of mafic and felsic macro units is much simpler, and define large open folds with km-scale amplitude. This may suggest that the mafic bodies represent dykes intruded into an already strongly deformed gneiss complex.

2.2. Ukaleq gneiss complex

The Ukaleq gneiss complex forms a package of granulite facies rocks dominated by garnet – bi-otite – feldspar – quartz9sillimanite rocks span-ning greywacke – pelite composition, interrupted by 0.5 – 1 m thick layers of garnet bearing leuco-granite, and rare thin diorite sheets. The felsic rocks contain abundant pods of hornblende – augite – hypersthene – plagioclase – quartz metaba-sites and olivine – tremolite – diopside – spinel ultramafic rocks. Due to deformation and exten-sive anatexis of the felsic rocks, the complex has the appearance of a me´lange, and no sedimentary structures have been observed. Based on its com-position, we interpret this gneiss complex as mainly supracrustal in origin, and dominated by metasediments.

2.3. Tasersiaq gneiss complex

The Tasersiaq gneisses are dominated by banded granulite facies orthogneisses and are sep-arated from the Aasivik gneiss complex and the Ukaleq gneiss complex by E – W and NW – SE trending topographic lineaments. The gneisses have been retrogressed along the E – W lineaments that separate them from the Aasivik gneiss

com-plex. Tasersiaq gneisses are found within 100 m from a Umiatsiaasat granite dome, but it has not been demonstrated that the granite intruded the Tasersiaq gneisses. The Tasersiaq gneisses are in-truded by the metadolerite dykes.

2.4. Umiatsiaasat granites

The supracrustal rocks of the Ukaleq gneiss complex were intruded by granite plutons exposed in large topographic domes typically 3 – 5 km long and 1 – 2 km across. The axises of the domes trend 140°, which is also the strike of foliations and lithologic layering in banded basement gneisses found in the cusps between domes. In most cases, the granites are hosted by gneisses which can be identified as belonging to the Ukaleq gneiss com-plex, but in a number of outcrops the banded gneiss host cannot be assigned to any particular defined unit. One granite dome extends across the lineament that separates the Aasivik and Ukaleq gneisses, which suggests that the granites were emplaced later than the assembly of the gneiss complexes.

2.5. Lineaments and Archaean mylonites

Marked topographic lineaments dissect the en-tire region in a nearly orthogonal pattern (Fig. 2). These lineaments coincide with the boundaries between the Aasivik gneiss complex, the Ukaleq gneiss complex and the Tasersiaq gneiss complex. Where examined, these lineaments also coincide with zones of high strain, including mylonites. The mylonites have synkinematic growth of gar-net suggesting that they were formed at granulite facies. The mylonites are generally annealed, and quartz and feldspar grains show no strain, but recrystallisation at lower metamorphic grade is uncommon in the studied examples. In a number of cases, the mylonites have experienced partial melting at various stages in the deformation pro-cess. One prominent set of mylonites with folia-tions of 140°/30°NE and a sinistral sense of displacement, and a less prominent set with folia-tion of 80°/70°SE and steeply plunging lineafolia-tion have been observed. However, systematic struc-tural studies of the high strain zones have not yet

(5)

been carried out. The high strain zones are hosts of a diffuse network of red granitic veins. In some places, these veins consist of mylonite or fine grained annealed mylonite. In other places, the veins are distinctly magmatic and consist of isotropic garnet bearing myrmekitic granite. Gneisses in the vicinity of the veins are affected by metasomatic infiltration, seen as a red colouration of the gneisses, which only partially overprint the compositional banding. We interpret the red granitic rocks assyn- to post-kinematic relative to the high strain zones. They could have been formed by flux melting within these zones as a response to fluid introduction during deforma-tion, or be preferentially intruded into these zones from an external source.

2.6. Metadolerite dykes

The gneiss complexes were intruded by rare E – W trending metadolerite dykes. Such dykes have not been observed in the post-tectonic gran-ites. However, the dykes show no signs of dis-placement along lineaments within the Aasivik gneiss complex. We interpret the dykes as em-placed later than the juxtaposition of the two complexes, but preferentially intruded into the gneissic roks for structural and rheological

rea-sons. The dolerite dykes show no signs of penetra-tive deformation, and the igneous mineral assemblage is widely preserved, although amphi-bolite facies metadoleritic domains are common. In places, the metamorphic hornblende – plagio-clase assemblage is replaced by garnet and clinopyroxene along fractures. We interpret the two metamorphic assemblages as formed during one event of static metamorphism in which am-phibolite facies metamorphism was followed by dehydration, possibly in response to decompres-sion. Based on their field occurrence and petrog-raphy we correlate these mafic dykes with the E – W component of the Paleoproterozoic Kangaˆmiut dyke swarm (Ramberg, 1949).

3. Geochemistry

At present, geochemical characterisation of the Aasivik gneisses is only at a reconnaissance stage. Major element compositions of apperantly homo-geneous lithologies from the Aasivik and Ukaleq gneiss complexes and the Umiatsiaasat granites are given in Table 1. Sample 430476 is a trond-hjemite, while samples 430441, 430451 and 94-0000 are granites. These granites are slightly peraluminous and alkaline, and project along the Table 1

Chemical compositions of representative lithologies from the Aasivik terrain determined by XRF at Geological Survey of Greenland (GEUS)a

430441 430451 430476

Sample c 430498 940000 940012

72.71 74.12

SiO2 67.59 65.67 74.11 74.25

0.16 0.18

TiO2 0.15 0.93 0.17 0.3

12.68 13.66

14.48

Al2O3 18.37 14.55 13.87

0.79

Fe2O3 0.5 0.47 0.31 1.45 0.35

3.83 0.73 1.14

0.76

FeO 0.84 0.73

0.02

MnO 0.022 0.02 0.02 0.07 0.02

MgO 0.48 0.33 0.56 1.25 0.47 0.29

1.01 3.27

3.27 1.17

1.79 2.25

CaO

2.87

Na2O 3.83 3.14 6.34 3.21 3.95

3.52 4.6 1.66

K2O 4.48 4.33 5.37

0.16 0.04

0.05

P2O5 0.23 0.06 0.07

0.27 0.2

Volat 0.29 0.2 0.44 0.44

Sum 99.14 99.39 99.36 99.36 99.5 99.14

a₄₃₀₄₄₁₌_{felsic gneiss cut by red veins, 430451}₌_{charnockite, 430476}₌_{trondhjemitic gneiss, 430498}₌_{2730 Ma}

(6)

1 – 2 kbar water saturated cotectic in the haplo-granite system. Petrographically they are sub-solvus granites and the excess Al is expressed as garnet in the mode. The granites are thus formed at high pressure. Compositionally they resemble the :2800 Ma Ikattoq gneisses of the Akulleq terrain in Godtha˚bsfjord (McGregor et al., 1991).

4. Zircon geochronology

4.1. Analytical technique and data assessment U – Th – Pb isotopic ratios and concentrations of zircons were determined using SHRIMP I and were calibrated to the Australian National Uni-versity standard zircon SL13 (572 Ma; 206

Pb/ 238_U₌_0.0928). _Descriptions _of _analytical procedure and data assessment were given by Compston et al. (1984), Nutman (1994) and Claoue´-Long et al. (1995). Comparative isotope dilution and SHRIMP analyses of zircons from several well-preserved Proterozoic and Archaean samples (Roddick and van Breemen, 1994; Ire-land, 1995) demonstrate that SHRIMP 207_Pb/ 206_{Pb ratios are accurate.}

As further cross-check on the accuracy of 207

Pb/206

Pb ratios obtained with SHRIMP, zir-cons from the Paleoproterozoic norite QGNG were analyzed interspersed with some of the un-knowns. From isotope dilution thermal ionisation analysis, different QGNG zircon fractions have 207_Pb/206_{Pb ages of 1850}₉_{2 Ma (C.M. Fanning,} personal communication, 1995), and 1850 Ma to as low as 1810 Ma (T. Skjo¨ld, personal communication, 1996). QGNG analyses run with samples presented here yielded a 207

Pb/206 Pb weighted mean age of 18639 18 Ma.

SHRIMP zircon geochronology results on nine samples are reported here. For some of them, many zircon analyses were undertaken, whereas for others fewer were done. The results are suffi-cient to demonstrate that there is an important early Archaean component to the crust in the area, and has determined the age of late Archaean rocks and some metamorphic zircon growth with precisions of 10 Ma (2s).

4.2. Aasi6ik gneiss complex

4.2.1. 94-0338 and 94-0000

94-0338 and 94-0000 are migmatitic gneisses. 94-0338 and 94-0000 yielded similar mixed zircon populations. Brown prismatic grains are domi-nant, up to 200 mm long, commonly with pro-nounced mm-scale euhedral zoning and locally metamict. In a few cases, the centres of these grains consist of clearer, non-zoned zircon. These grains show deep embayments on prismatic faces and rounding of pyramidal terminations. Less commonly, there are overgrowths of clear zircon on their terminations. Also present are a few small, clear to brown oval to multifaceted grains. The heavy mineral separates also contained mon-azites, which were not analyzed.

In 94-0338, three analyses of an oval grain gave close to concordant ages, with a207_Pb/206_{Pb mean} age of 2712 912 Ma (Fig. 4a). The dominant poorly preserved prismatic grains of 94-0338 have moderate to high U content (up to 750 ppm) and yielded mostly discordant ages with 207_Pb/206_Pb ages between 3200 and 3600 Ma. Several grains have 207_Pb/206_{Pb ages of} _{3600 Ma, one of} which is concordant, within error. These grains are interpreted to be early Archaean in age, and to have suffered variable loss of radiogenic Pb in a zircon-growth event at 2700 Ma, also more recently. Three analyses with the oldest 207

Pb/ 206_{Pb ages yield a mean age of 3596}₉_{9 Ma.} However, given the disturbed nature of the zir-cons, this is interpreted by us as a minimum age, but probably close to the true value.

In 94-0000, metamorphic overgrowths and a brown oval grain yielded close to concordant ages with 207_Pb/206_{Pb ages between 2628}₉_{70 and} 2711 944 Ma (2s), with a mean value of 2688 912 Ma, indistinguishable at the 95% confidence level from the 2712 912 Ma age de-termined on a metamorphic grain in 94-0338. Analyses of the dominant brown prismatic grains yielded an array of older ages, of which two with the oldest 207_Pb/206_{Pb ages of} _\_{3500 Ma are} close to concordant (Fig. 4b). This indicates an important early Archaean component in this sample.

(7)

Fig. 4. U – Pb concordia diagrams for SHRIMP analysis of zircons from a) 94-0338: Migmatitic gneiss from the Aasivik terrain. b) 94-0000 red myrmekitic granite/migmatite bordering the Aasivik terrain. c) 430476 Felsic orthogneiss from the Aasivik terrain, d) 430475 Felsic orthogneiss from the Aasivik terrain.

4.2.2. 430441

430441 is a banded gneiss migmatised by a fine network of red coloured granitic veins just north of the E – W trending lineament separating the Aasivik and the Ukaleq gneiss terrains. It yielded mostly pale yellow to clear prismatic grains up to 300 mm long with aspects ratios of 2 – 5 and have well-developedmm-scale euhedral zoning. Pyrami-dal terminations are slightly rounded, and one has a metamorphic overgrowth, too small to be ana-lyzed with the 40-mm spot used in the analytical session. The grains are interpreted as a magmatic population, slightly modified during metamor-phism. Amongst the 50 mounted hand-picked grains one was devoid of euhedral zoning, but with a possible rind of new zircon, a few microns thick. Analyses of two of the dominant well-zoned prismatic grains showed high U contents (\1000 ppm), close to concordant ages, with a207_Pb/206_Pb mean age of 2817910 Ma, which is interpreted as the timing of magmatic crystallisation of the granitic neosome (Fig. 4b). Two analyses of the non-zoned zircon gave a low U content (B100 ppm), close to concordant ages, with a207_Pb/206_Pb

mean age of 3784 918 Ma. This probably indi-cates an early Archaean age of the banded gneiss protolith.

4.2.3. 430476and 430475

The zircons in 430476 and 430475 are similar to the main population in 94-0338, but are slightly larger on average, and less strongly coloured. Six analyses of 430476 grains yielded a discordant array on a concordia diagram, like the 94-0338 zircons (Fig. 4c). In this case, no analyses yielded concordant ages and the oldest 207_Pb/206_{Pb age} was 3559948 (2s). Five analyses of grains from 430475 yielded discordant points (Fig. 4d), with the least discordant result having the oldest207_Pb/ 206_{Pb age of 3602} ₉_{8 Ma (2}_s_{). As for samples} 94-0000 and 94-0338 the (moderate to high U) protolith zircons suffered Pb-loss in a late Ar-chaean, generating the observed discordant array of data. These results are not sufficient to define accurate and precise ages, but they do indicate the zircons are probably a strongly disturbed 3600 Ma population.

(8)

4.3. Ukaleq gneiss complex 4.3.1. 94-0006

The supracrustal rock 94-0006 is a garnet – bi-otite – feldspar – quartz gneiss interpreted as a semipelitic metasediment. It gave a low yield of zircons. Dominant are equant to oval grains, B 100 mm across, strongly zoned and brown in colour. Some of these grains form twins. Also present are prismatic and rounded anhedral grains up to 150 mm long. These are brown to yellow and also generally strongly zoned. The termina-tions of these grains are rounded. The equant, oval and twinned grains have on average higher U contents and lower Th/U than the rounded pris-matic grains. A reconnaissance study of sixteen grains was undertaken on this sample. All analy-ses yielded close to concordant ages (Fig. 6) with 207_Pb/206_{Pb ages from 2530}₉ _{46 – 2905}₉_{20 Ma} (2s). The oval, equant and twinned grains yielded 207_Pb/206_{Pb ages of up to 2726} ₉_{24 Ma and if} metamorphic in origin probably grew or were recrystallised in several high-grade events. The rounded prismatic grains yielded the oldest ages (1-1, 8-1, 10-1, 11-1 in Table 2). If they are detrital in origin, their ages indicate that the sediment must have been deposited in the late Archaean. 4.3.2. 430451

430451 is massive granulite facies orthogneiss or charnockite just south of the E – W trending lineament separating the Aasivik terrane and the Ukaleq gneiss complex and within tens of metres from the contact to a Umiatsiaasat granite dome. It yielded a population very similar in morphol-ogy to those in 430441 (just north of the linea-ment, and within the Aasivik gneiss complex) and likewise is interpreted as a magmatic population, slightly modified during metamorphism. However, neither inherited grains nor metamorphic over-growths were observed. Six analyses were under-taken, all of which showed moderate to high U contents (100 – 987 ppm) and yielded close to con-cordant ages (Fig. 5). Grain 3-1 (Table 2) yielded the oldest age of 28779 42 Ma (2s) and is most discordant. However, from optical microscopy, there is no evidence it is an inherited core. With no rejections, all analyses yielded a 207_Pb/206

Pb-weighted mean age of 28129 8 Ma, with analy-sis 3-1 indistinguishable from the mean. Rejecting 3-1, and 6-1 (interpreted as having undergone slight ancient Pb-loss) yielded an indistinguishable age of 2820 912 Ma. These results indicate the magmatic crystallisation of 430451 shortly before 2800 Ma.

4.3.3. 430455

440455 is a thin diorite sheet which form an early component in the gneisses hosting the Umi-atsiaasat granites. It yielded a population of clean colourless to pale yellow prismatic grains, up to 200 mm long, with some faint euhedral zoning. Neither inherited cores nor metamorphic over-growths were observed. The grains show only slight rounding from metamorphic corrosion. Five grains were analyzed, which had low U content (59 – 154 ppm) and close to concordant ages (Fig. 5d). All analyses yielded a 207_Pb/206 Pb-weighted mean age of 2701 916 Ma, interpreted as the magmatic age of the sample.

4.4. Tasersiaq gneiss complex

A granulite facies gneiss 430428 from the Taser-siaq gneiss complex yielded prismatic and more rarely equant colourless to very pale yellow zir-cons, up to 250 mm long and mostly with low aspect ratios of 1 – 3. They are non-zoned or show weak mm-scale euhedral zoning, more prominent towards, the exterior of the grains. Inclusions of apatite and opaques are common. Pyramidal ter-minations are slightly rounded, and some pris-matic faces are slightly embayed. There are no obvious overgrowths of metamorphic zircon. Twelve grains were analyzed. They have low U contents (73 – 250 ppm), yielded concordant or close to concordant ages (Fig. 5a) with a range in 207

Pb/206

Pb ages from 3226936 Ma – 2971941 (2s). Four analyses (4-1, 8-1, 9-1 and 11-1) yielded a 207_Pb/206_{Pb-weighted mean age of} 3212 911 Ma. Five analyses (1-1, 2-1, 6-1, 7-1, 12-1) yielded a distinctly younger age of 3127 9 12 Ma, and the remaining three somewhat younger ages. There are no obvious morphologi-cal differences between grains of different age, but on average, the oldest ones have the lowest U

(9)

Table 2

SHRIMP U–Pb zircon analyses

Spot _{U (ppm)} Th/U f206% 206_Pb_/238_{U ratio} 207_Pb_/206_{Pb ratio} 207_Pb_/206_{Pb (age)} _{% (disc)}

94-0338

1.76

215 0.11 0.737922 0.3202942 3571920 0

1-1

0.05 0.708914 0.3297914

1.88 3616907

215 −5

1-2

0.18

707 0.04 0.633944 0.2723909 3319950 −5

2-1

0.19

394 0.03 0.652913 0.2928907 3433904 −6

3-1

0.11 0.695917 0.3220914

0.46 3580907

4-1 144 −5

0.16

548 0.04 0.620918 0.2822913 3375907 −8

4-2

0.03 0.556908 0.2439905

5-1 667 0.14 3145903 −9

0.20 0.516913 0.1883922

1.76 2727919

62 −2

6-1

0.45

170 0.02 0.527930 0.1859917 2706915 +1

6-2

0.01

6-3 89 1.24 0.474921 0.1854929 2702925 −7

0.01 0.636911 0.3000908

0.82 3470904

365 −9

7-1

0.14

638 0.06 0.608909 0.2636913 3269908 −6

8-1

0.07 0.603921 0.3158922

9-1 164 0.81 3550911 −14

B0.01 0.559910 0.2430916

0.17 3140911

568 −9

10-1

0.31

366 0.09 0.621918 0.3073968 3507934 −11

11-1

0.01 0.668915 0.3049927 3495914 −6

12-1 479 0.23

0.06 0.677915 0.3122939

1.07 3532920

13-1 342 −6

0.19

435 0.02 0.626911 0.2747921 3333912 −6

14-1

0.05 0.666915 0.2892933 3413918 −4

15-1 443 0.28

0.03 0.629913 0.2831914 3380908 −7

0.62 748

16-1

94-0000

485 0.64 0.14 0.552 913 0.2027936 2848929 0

1-1

0.39 0.716915 0.3134915 3538907 −2

2-1 218 1.33

0.20 0.692915 0.3164918

1.89 3552916

3-1 227 −5

1.00 0.628912 0.2785909

4-1 680 0.68 3354905 −6

0.27 0.598910 0.2745910

0.61 3332906

5-1 522 −8

0.22

253 0.01 0.514911 0.1773938 2628935 +2

6-1

B0.01 0.516911 0.1838908

0.55 2687907

128 0

7-1

0.41

161 0.05 0.505911 0.1831914 2681913 −2

8-1

9-1 315 0.28 0.01 0.515913 0.1791917 2644916 +1

B0.01 0.504924 0.1864925

0.41 2711922

10-1 388 −3

11-1 822 0.338 0.05 0.548911 0.2400907 3120905 −10

430475

0.17 0.531 914

1-1 581 0.26 0.2916924 3426913 −20

0.06 0.616918 0.3055907

0.25 3498904

289 −12

2-1

0.61

210 0.12 0.717918 0.3267908 3602904 −3

3-1

0.59

183 0.06 0.648917 0.3011924 3476912 −7

4-1

0.11 0.633915 0.2746912 3333907 −5

0.31 200

5-1

430476

48 0.06 0.79 0.594 957 0.2712913 3313979 −9

1-1

0.15 0.655921

3-1 88 0.29 0.3177949 3559924 −9

0.01 0.678916 0.3086909

0.17 3514905

644 −5

4-1

0.42

189 0.12 0.699918 0.3170920 3555910 −4

5-1

0.26 0.667924

6-1 60 0.07 0.2604966 3249940 +1

0.18 0.550914 0.2251913 3018909 −6

0.11

7-1 204

430428

73 0.76 0.24 0.619 936 0.2418954 3132936 −1

1-1

2-1 147 0.63 0.18 0.606915 0.2417915 3131910 −2

0.38 0.573917 0.2187929

0.54 2971921

109 −2

(10)

Table 2 (Continued)

Spot U (ppm) Th/U f206% 206_Pb_/238_{U ratio} 207_Pb_/206_{Pb ratio} 207_Pb_/206_{Pb (age)} _{% (disc)}

0.51 0.45

4-1 95 0.645917 0.2566913 3226908 −1

0.82 0.25 0.606917

149 0.2450923

6-1 3153915 −3

142

7-1 1.04 0.07 0.580913 0.2386914 3111909 −5

145

8-1 0.57 0.09 0.634915 0.2524933 3200921 −1

0.60 0.06 0.613913

116 0.2523918

9-1 3200911 −4

176

10-1 1.05 0.03 0.568923 0.2287915 3043911 −5

11-1 83 0.63 0.31 0.591915 0.2521920 3198912 −6

0.85 0.06 0.600927 0.2421932 3134921

250 −3

12-1

430441

0.13 0.03 0.574 914

1-1 1452 0.1978912 2808910 +4

0.21 0.02 0.528913

1969 0.1998911

2-1 2825909 −3

3-1 0.67 1.00 0.814927 0.3648930 3770913 +2

0.90 0.38 0.793932 0.3720932 3800913

3-2 77 −1

430451

0.91 0.04 0.557 922 0.1978923 2808919 +2

1-1 747

0.49 0.04 0.560914

987 0.1993908

2-1 2820906 +2

282

3-1 0.55 0.14 0.516914 0.2063927 2877921 −7

0.39 0.17 0.557918

100 0.1999932

4-1 2825926 +1

191

5-1 0.37 0.22 0.535914 0.1993925 2821920 −2

6-1 128 0.63 0.19 0.548914 0.1957910 2791908 +1

430455

1.20 0.07

1-1 154 0.514 915 0.1842920 2691918 −1

0.51 0.05 0.492913 0.1822935 2673931 −4

2-1 54

1.08 B0.01 0.521911

152 0.1860912

3-1 2707910 0

4-1 0.71 0.09 0.499920 0.1853948 2701943 −3

0.63 0.48 0.488916 0.1859946 2706941

5-1 59 −5

94-0006

0.44 0.25 0.561 925 0.1938919 2775916 +3

1-1 213

0.17 0.19 0.520915

787 0.1798912

2-1 2651911 +2

0.14 0.04

3-1 617 0.513915 0.1729939 2586938 +3

0.42 0.21 0.545920

466 0.2100913

4-1 2905910 −3

860

5-1 0.20 0.03 0.508915 0.1698905 2556905 +4

0.15 0.11 0.507915

6-1 690 0.1777926 2628924 +1

0.16 0.17 0.506916

488 0.1672923

7-1 2530923 +4

182

8-1 0.49 0.06 0.564920 0.2083913 2892910 0

9-1 194 0.20 0.21 0.511916 0.1882913 2726912 −2

0.12 0.14 0.516915

501 0.1824910

9-2 2675909 0

461

10-1 0.81 0.25 0.571916 0.19799109 2809993 +4

0.31 0.13 0.555916

11-1 366 0.1981928 2811923 +1

0.17 0.21 0.489914

501 0.1737923

12-1 2593922 −1

643

13-1 0.22 0.11 0.500914 0.1691927 2549926 +2

0.16

14-1 674 0.12 0.513914 0.1795926 2649924 +1

0.24 1.00 0.530916

324 0.1911939

15-1 2751933 0

574

16-1 0.15 0.04 0.505916 0.1720919 2577919 +2

0.14

16-2 614 0.08 0.491914 0.1793919 2647918 −3

content. The rock is interpreted as having two age components of 32129 11 Ma and 31279 12 Ma, followed by subsequent disturbance. From the zircon data alone, it is uncertain

whether the ages represent different phases in a composite sample or whether the protolith is 3127 912 Ma, but carries 3212911 Ma in-herited material.

(11)

5. Discussion

U – Pb age determination has been carried out on zircons extracted from a number of units. Samples 430428, 430475 and 430476 are regional banded gneisses, 94-0006 is a paragneiss, 94-0338 and 430441 are red-coloured metasomatised gneisses, 430451 is an isotropic charnockite, probably a diatexite, 94-0000 is a red myrmekitic granite and 430455 is an undeformed diorite dyke intruded into the supracrustal package. As can be seen from Table 2 and Figs. 3 – 6, the regional gneisses give disturbed ages in the range 3600 – 3780 Ma for the samples within the Aasivik terrain as outlined in Fig. 1. Sample 430428 south of the bounding lineament contains :3100 and :3200 Ma com-ponents. These ages have not been represented in the zircon populations in other samples from the region. The ages found in this sample are within the range of ages of gneisses from the Akia terrane to the south (Friend et al., 1988; Nutman et al., 1996) and it is tentatively suggested that the Tasersiaq gneiss complex forms a part of the Akia terrane.

The supracrustal sample from the Ukaleq gneiss complex contains 2750 – 2900 Ma zircons of prob-able detrital origin, which gives the maximum age of deposition. This along with a positiveo_Ndof the anatectic granites (tzircon oNd=2.3 – 5.4; Løfqvist, unpublished data) formed within the metasedi-ments suggests that the sources of the sedimetasedi-ments were dominated by juvenile or only slightly older material at 2800 – 2700 Ma. The chaotic intercala-tion of metapelic rocks, mafic granulites and meta-peridotites together with the occurrence wedged between a late Arechaean and an early Archaean gneiss complex suggests that the Ukaleq gneiss complex form part of a late Archaean accretionary wedge.

The red granites and the charnockitic diatexite found in the lineaments separating the different gneiss complexes are all dominated by 2700 – 2800 Ma zircons. We suggest that this age range is bracketing the time of deformation in the zones, and thus the time of tectonic juxtaposition of the gneiss complexes.

Fig. 5. U – Pb concordia diagrams for SHRIMP analysis of zircons from a) Felsic gneiss from the Tasersiaq terrain. b) 430428 Red coloured banded gneiss bordering the Aasivik terrain. c) 430451 Diatexite near contact of Umiatsiaasat granite dome. d) 430455 Diorite sheet in gneisses along contact of Umiatsiaasat granite dome.

(12)

Fig. 6. U – Pb concordia diagram and relative probability histogram for SHRIMP analysis of zircons from a sample of Ukaleq paragneiss.

The nature of the contact between this gneiss complex and the Aasivik terrane has not been investigated in this study, but a pronounced topo-graphic lineament coincides with contrasts in stream sediment geochemistry (Steenfelt, 1994), topographic grain, occurrence of klimberlite in-trusions (Larsen et al., 1983). This lineament sepa-rates a region where no early Archaean zircons have been identified in the gneisses (Kalsbeek and Nutman, 1996) to the north from the early Ar-chaean gneisses to the south, and is here taken as the terrane boundary.

We suggest that the gneiss complexes were as-sembled at :2700 Ma and formed a stable cra-ton during Nagssugtoqidian orogeny which caused penetrative deformation and metamor-phism of all rock units only 20 – 50 km to the north.

Kimberlite intrusions which are common in the adjoining terrains have not been identified within the Aasivik terrane, which in turn is partly defined by the apparent absence of kimberlite intrusions. By applying the well-proven method of circular reasoning, we can deduce that the terrane con-straint on the distribution of kimberlite intrusions indicates that the Aasivik terrane has deep litho-spheric roots. The occurrence of diamond in a kimberlite south of the Aasivik terrane (Nunaoil A/S press release, 1996) indicate that the kimber-lite originated at \150 km depth. This is accord with the barometric estimates by Larsen and Rønsbo, 1993) for kimberlites along the inferred western and northern boundaries of the Aasivik terrane. If this interpretation is correct the early Archaean terrane is the exposed lower crustal level of a continent which had already existed for one billion years at the time of terrane assembly at 2700 Ma, and not merely a tectonic panel such as a fault block or nappe. This interpretation is in accord with Abbott (1996), who suggested that only accreted material with deep lithospheric roots could have been stabilised as parts of the continental crust during the Archaean. In this respect, the Aasivik terrane differs from the Akulleq terrane of the Godtha˚bsfjord region, which represent tectonic slices of an old gneiss complex intercalated with younger gneiss units. The Akulleq terrane is dominated by mid-crustal The diorite dyke 430455 is associated with the

post-kinematic granites, and indicate that mem-bers of the :2700 Ma magmatic suite intruded the supracrustal rocks of the Ukaleq gneiss com-plex. The age of 2701916 Ma for the diorite is in agreement with U – Pb TIMS analysis of zircon from a granite pluton which gave 271794 Ma (J. Connelly, personal communication, 1995). These ages support the field observation that the granites were emplaced late in the 2800 – 2700 Ma tectono-metamorphic event responsible for the as-sembly of the gneiss complexes.

This study shows that the Aasivik gneiss com-plex has a major early Archaean component, and is separated from rock complexes of different ages and origin by zones of high strain. We suggest that the Aasivik gneiss complex is a tectonometa-morphic terrane for which we propose the name Aasivik terrane. The area immediately north of the Aasivik terrane is dominated by :2800 Ma felsic gneisses (Kalsbeek and Nutman, 1996), which are overprinted by the Paleoproterozoic Nagssugtoqidian orogeny further to the north.

(13)

amphibolite facies rocks, while the Aasivik ter-rane is at high-pressure granulite facies. The pos-sibility that the Aasivik terrane represents the roots of the allochthonous Akulleq terrane should be investigated further by detailed geochemistry and chronometry.

6. Conclusions

The region south of the Nagssugtoqidian front in West Greenland comprises three Archaean gneiss complexes, which were assembled at 2800 – 2700 Ma. The early Archaean Aasivik terrane gives U – Pb zircon ages of 3500 – 3600 Ma for the formation of gneiss protoliths, but probably com-prise gneisses formed back to3800 Ma, judging from the occurrence of zircons with an age of 3780 Ma. This gneiss terrain is dominated by granitic gneisses with a complicated deformation history, intruded by large mafic dykes of un-known age, and now represented by pods and layers of mafic granulite. No large supracrustal units have been identified in the early Archaean gneiss complex at present. The Aasivik terrane probably underlies 1500 km2

. This terrane is in contact to the Southwest with the Tasersiaq mid-Archaean gness comples, which we provisionally interpret as the northern extension of the Akia terrane (eg. Nutman et al., 1996) of the Nuuk region. One sample of banded gneiss from this gneiss complex has given ages of 3127912 Ma and 3212911 for the formation of gneiss pro-toliths. The late Archaean Ukaleq gneiss complex to the Southeast is dominated by supracrustal rocks with similarities to modern accretionary complexes and intruded by post-kinematic A-type granite plutons, which probably also stitch the tectonic boundaries between the early Archaean and the mid-Archaean and late Archaean terranes.

Acknowledgements

This study was carried out in collaboration with the Danish Lithosphere Centre Nagssug-toqidian project funded by the Danish National

Research Fond. Funding by the Carlsberg Foun-dation and RSES at the Australian National Uni-versity, Canberra is gratefully acknowledged. We thank David Bridgwater, Vic McGregor, Jim Connelly, Flemming Mengel, Jeroen van Gool and Clark Friend for discussions and suggestions. Heidi Thomsen, Nick Rose, Kyle Mayborn and Esbern Hoch are thanked for company in the field.

References

Abbott, H.H., 1996. Plumes and hotspots as sources of green stone belts. Lithos 37, 113 – 127.

Baadsgaard, H., 1973. U – Th – Pb dates on zircons from the early Precambrian Amıˆtsoq gneisses, Godtha˚b district, West Greenland. Earth. Planet. Sci. Lett. 33, 261 – 267. Black, L.P., Gale, N.H., Moorbath, S., Pankhurst, R.J.,

Mc-Gregor, V.R., 1971. Isotopic dating of very early Precam-brian amphibolite facies gneisses of the Isukasia area, southern West Greenland. Earth. Planet. Sci. Lett. 12, 245 – 259.

Black, L.P., Sheraton, J.W., James, P.R., 1986. Late Archaean granites of the Napier Complex, Enderby Land, Antarc-tica: a comparison of Rb – Sr, Sm – Nd and U – Pb isotopic systematics in a complex terrain. Precambrian Research. 32 (4), 343 – 368.

Bowring, S.A., Williams, I.S., Compston, W., 1989. 3.96 Ga gneisses from the Slave Province, Northwest Territories, Canada. Geology 17, 971 – 975.

Bridgwater, D., Schiøtte, L., 1991. The Archaean gneiss com-plex of northern Labrador: a review of current results, ideas and problems. In: Pulvertaft, T.C.R. (Ed.), The Early History of the Earth and its Continental Crust. Bull. Geol. Soc. Denmark, 39, 153 – 166.

Claoue´-Long, J.C., Compston, W., Roberts, J., Fanning, C.M., 1995. Two Carboniferous ages: a comparison of SHRIMP zircon dating with conventional zircon ages and

40_Ar_/39_{Ar analysis. In Geochronology, Time scales and}

Global Stratigraphic Correlation. Society for Sedimentary Geology Special Publication 54, 3 – 21.

Compston, W., Williams, I.S., Myers, C., 1984. U – Pb geochronology of zircons from lunar breccia 73217 using a sensitive high mass-resolution ion microprobe. In: Boyn-ton, W.V., et al. (Eds.), Proceedings of the Fourteenth Lunar and Planetary Science Conference; part 2. J. Geophys. Res. 89B, 525 – 534

Compston, W., Kro¨ner, A., 1988. Multiple zircon growth within early Archaean tonalitic gneiss from the Ancient Gneiss Complex, Swaziland. Earth and Planetary Science Letters 87, 13 – 28.

Escher, A., Escher, J.C., Watterson, J., 1970. The Nagssug-toqidian boundary and the deformation of the Kangaˆmiut dyke swarm in the Søndre Strømfjord area. Rapp. Grønlands geol. Unders 28, 21 – 23.

(14)

Friend, C.R.L., Nutman, A.P., McGregor, V.R., 1988. Late Archaean terrane accretion in the Godtha˚b region, south-ern West Greenland. Nature 335, 535 – 538.

Hanmer, S., Mengel, F., Conelly, J.N., van Gool, J., 1997. Significance of crustal-scale shear zones and synkinematic mafic dykes in the Nagssugtoqidian orogen, SW Green-land: a reexamination. Journ. Structural Geol. 19, 59 – 75. Ireland, T.R., 1995. Ion microprobe mass spectrometry: tech-niques and application in Cosmochemistry, Geochemistry and Geochronology. In: Hyman, M., Rowe, M. (Eds.), Advances in Analytical Geochemistry. JAI press. Kalsbeek, F., Nutman, A.P., 1996. The anatomy of the Early

Proterozoic Nagssugtoqidian Orogen, West Greenland, ex-plored by reconnaisance SHRIMP U – Pb zircon dating. Geology 24, 515 – 518.

Kinny, P.D., Nutman, A.P., 1996. Zirconology of the Meeber-rie Gneiss, Yilgarn Craton, Western Australia: an early Archaean migmatite. In: Kro¨ner, A. (Ed.), The Oldest Rocks on Earth. Precambrian Research 78, 165 – 178. Krylov, D.P., Levchenkov, O.A., Beliatsky, B.V., Komarov,

A.N., Yakovleva, S.Z., 1989. New isotope-geochronologi-cal evidence of ancient events in Enderby Land (Antarc-tica). Dok. Akad. Nauk SSSR 304 (2), 433 – 436 (in Russian).

Larsen, L.M., Rex, D.C., Secher, K., 1983. The age of carbon-atites, kimberlites and lamprophyres from southern West Greenland: recurrent alkaline magmatism during 2500 mil-lion years. Lithos 16, 215 – 221.

Larsen, L.M., Rønsbo, J., 1993. Conditions of origin of kim-berlites in West Greenland: new evidence from the Sarfar-toq and Sukkertoppen regions. Rapp. Grønlands Geol. Unders 159, 115 – 120.

McGregor, V.R., 1973. The early Precambrian gneisses of the Godtha˚b district, West Greenland. Phil. Trans. Roy. Soc. London A273, 343 – 358.

McGregor, V.R., Friend, C.R.L., Nutman, A.P., 1991. The late Archaean mobile belt through Godtha˚bsfjord, south-ern West Greenland: a continent – continent collision zone? Bull. Geol. Soc. Denmark 39, 179 – 197.

Moorbath, S., O’Nions, R.K., Pankhurst, R.J., Gale, N.H., McGregor, V.R., 1972. Further rubidium – strontium age determinations on the very early Precambrian rocks of the Godtha˚b district.

Moorbath, S., O’Nions, R.K., Pankhurst, R.J., 1973. Early Archaean age for the Isua iron formation, West Green-land. Nature 245, 138 – 139.

Myers, J.S., 1988. Early Archaean Narryer Gneiss Complex, Yilgarn Craton, Western Australia. In: Kro¨ner, A. (Ed.), Development and Evolution of the Continental Crust. Precambrian Research 38, 297 – 307.

Nutman, A.P., 1994. Anatomy of an early Archaean gneiss complex: 3900 to 3600 Ma crustal evolution in southern West Greenland: Reply. Geology 22, 571 – 572.

Nutman, A.P., Kinny, P.D., Compston, W., Williams, I.S., 1991. SHRIMP U – Pb zircon geochronology of the Nar-ryer gneiss complex, Western Australia. Precambrian Re-search 52, 275 – 300.

Nutman, A.P., McGregor, V.R., Friend, C.R.L., Bennet, V.C., Kinny, P.D., 1996. The Itsaq Gneiss Complex of southern West Greenland: the World’s most extensive record of early crustal evolution (3900 – 3600 Ma). Precambrian Re-search, 177,

Ramberg, H., 1949. On the petrogenesis of the gneiss com-plexes between Sukkertoppen and Christianshaab, West Greenland. Meddr. Dansk Geologisk Forening 11, 312 – 327.

Roddick, J.C., van Breemen, O., 1994. U – Pb zircon dating: a comparison of ion microprobe and single grain conven-tional analyses. In Radiogenic age and isotopic studies: Report 8. Geological Survey of Canada Current Research 1994-F, 1 – 9.

Steenfelt, A., 1994. Crustal structure in West and South Greenland reflected by regional distribution patterns of Calcium and Potassium in stream sediments. Rapp. Grønlands Geol. Unders 161, 11 – 20.

Song, B., Nutman, A.P., Liu, D., Wu, J., 1996. 3800 to 2500 Ma crustal evolution in the Ashan area of Liaoning Province, northeastern China. In: Kro¨ner, A. (Ed.), The Oldest Rocks on Earth. Precambrian Research 78, 79 – 94.

(1)

Table 2

SHRIMP U–Pb zircon analyses

Spot _{U (ppm)} Th/U f206% 206_Pb_/238_{U ratio} 207_Pb_/206_{Pb ratio} 207_Pb_/206_{Pb (age)} _{% (disc)}

94-0338

1.76

215 0.11 0.737922 0.3202942 3571920 0

1-1

0.05 0.708914 0.3297914

1.88 3616907

215 −5

1-2

0.18

707 0.04 0.633944 0.2723909 3319950 −5

2-1

0.19

394 0.03 0.652913 0.2928907 3433904 −6

3-1

0.11 0.695917 0.3220914

0.46 3580907

4-1 144 −5

0.16

548 0.04 0.620918 0.2822913 3375907 −8

4-2

0.03 0.556908 0.2439905

5-1 667 0.14 3145903 −9

0.20 0.516913 0.1883922

1.76 2727919

62 −2

6-1

0.45

170 0.02 0.527930 0.1859917 2706915 +1

6-2

0.01

6-3 89 1.24 0.474921 0.1854929 2702925 −7

0.01 0.636911 0.3000908

0.82 3470904

365 −9

7-1

0.14

638 0.06 0.608909 0.2636913 3269908 −6

8-1

0.07 0.603921 0.3158922

9-1 164 0.81 3550911 −14

B0.01 0.559910 0.2430916

0.17 3140911

568 −9

10-1

0.31

366 0.09 0.621918 0.3073968 3507934 −11

11-1

0.01 0.668915 0.3049927 3495914 −6

12-1 479 0.23

0.06 0.677915 0.3122939

1.07 3532920

13-1 342 −6

0.19

435 0.02 0.626911 0.2747921 3333912 −6

14-1

0.05 0.666915 0.2892933 3413918 −4

15-1 443 0.28

0.03 0.629913 0.2831914 3380908 −7

0.62 748

16-1

94-0000

485 0.64 0.14 0.552 913 0.2027936 2848929 0

1-1

0.39 0.716915 0.3134915 3538907 −2

2-1 218 1.33

0.20 0.692915 0.3164918

1.89 3552916

3-1 227 −5

1.00 0.628912 0.2785909

4-1 680 0.68 3354905 −6

0.27 0.598910 0.2745910

0.61 3332906

5-1 522 −8

0.22

253 0.01 0.514911 0.1773938 2628935 +2

6-1

B0.01 0.516911 0.1838908

0.55 2687907

128 0

7-1

0.41

161 0.05 0.505911 0.1831914 2681913 −2

8-1

9-1 315 0.28 0.01 0.515913 0.1791917 2644916 +1

B0.01 0.504924 0.1864925

0.41 2711922

10-1 388 −3

11-1 822 0.338 0.05 0.548911 0.2400907 3120905 −10

430475

0.17 0.531 914

1-1 581 0.26 0.2916924 3426913 −20

0.06 0.616918 0.3055907

0.25 3498904

289 −12

2-1

0.61

210 0.12 0.717918 0.3267908 3602904 −3

3-1

0.59

183 0.06 0.648917 0.3011924 3476912 −7

4-1

0.11 0.633915 0.2746912 3333907 −5

0.31 200

5-1

430476

48 0.06 0.79 0.594 957 0.2712913 3313979 −9

1-1

0.15 0.655921

3-1 88 0.29 0.3177949 3559924 −9

0.01 0.678916 0.3086909

0.17 3514905

644 −5

4-1

0.42

189 0.12 0.699918 0.3170920 3555910 −4

5-1

0.26 0.667924

6-1 60 0.07 0.2604966 3249940 +1

0.18 0.550914 0.2251913 3018909 −6

0.11

7-1 204

430428

73 0.76 0.24 0.619 936 0.2418954 3132936 −1

1-1

2-1 147 0.63 0.18 0.606915 0.2417915 3131910 −2

0.38 0.573917 0.2187929

0.54 2971921

109 −2

(2)

Table 2 (Continued)

Spot U (ppm) Th/U f206% 206_Pb_/238_{U ratio} 207_Pb_/206_{Pb ratio} 207_Pb_/206_{Pb (age)} _{% (disc)}

0.51 0.45

4-1 95 0.645917 0.2566913 3226908 −1

0.82 0.25 0.606917

149 0.2450923

6-1 3153915 −3

142

7-1 1.04 0.07 0.580913 0.2386914 3111909 −5

145

8-1 0.57 0.09 0.634915 0.2524933 3200921 −1

0.60 0.06 0.613913

116 0.2523918

9-1 3200911 −4

176

10-1 1.05 0.03 0.568923 0.2287915 3043911 −5

11-1 83 0.63 0.31 0.591915 0.2521920 3198912 −6

0.85 0.06 0.600927 0.2421932 3134921

250 −3

12-1

430441

0.13 0.03 0.574 914

1-1 1452 0.1978912 2808910 +4

0.21 0.02 0.528913

1969 0.1998911

2-1 2825909 −3

3-1 0.67 1.00 0.814927 0.3648930 3770913 +2

0.90 0.38 0.793932 0.3720932 3800913

3-2 77 −1

430451

0.91 0.04 0.557 922 0.1978923 2808919 +2

1-1 747

0.49 0.04 0.560914

987 0.1993908

2-1 2820906 +2

282

3-1 0.55 0.14 0.516914 0.2063927 2877921 −7

0.39 0.17 0.557918

100 0.1999932

4-1 2825926 +1

191

5-1 0.37 0.22 0.535914 0.1993925 2821920 −2

6-1 128 0.63 0.19 0.548914 0.1957910 2791908 +1

430455

1.20 0.07

1-1 154 0.514 915 0.1842920 2691918 −1

0.51 0.05 0.492913 0.1822935 2673931 −4

2-1 54

1.08 B0.01 0.521911

152 0.1860912

3-1 2707910 0

4-1 0.71 0.09 0.499920 0.1853948 2701943 −3

0.63 0.48 0.488916 0.1859946 2706941

5-1 59 −5

94-0006

0.44 0.25 0.561 925 0.1938919 2775916 +3

1-1 213

0.17 0.19 0.520915

787 0.1798912

2-1 2651911 +2

0.14 0.04

3-1 617 0.513915 0.1729939 2586938 +3

0.42 0.21 0.545920

466 0.2100913

4-1 2905910 −3

860

5-1 0.20 0.03 0.508915 0.1698905 2556905 +4

0.15 0.11 0.507915

6-1 690 0.1777926 2628924 +1

0.16 0.17 0.506916

488 0.1672923

7-1 2530923 +4

182

8-1 0.49 0.06 0.564920 0.2083913 2892910 0

9-1 194 0.20 0.21 0.511916 0.1882913 2726912 −2

0.12 0.14 0.516915

501 0.1824910

9-2 2675909 0

461

10-1 0.81 0.25 0.571916 0.19799109 2809993 +4

0.31 0.13 0.555916

11-1 366 0.1981928 2811923 +1

0.17 0.21 0.489914

501 0.1737923

12-1 2593922 −1

643

13-1 0.22 0.11 0.500914 0.1691927 2549926 +2

0.16

14-1 674 0.12 0.513914 0.1795926 2649924 +1

0.24 1.00 0.530916

324 0.1911939

15-1 2751933 0

574

16-1 0.15 0.04 0.505916 0.1720919 2577919 +2

0.14

16-2 614 0.08 0.491914 0.1793919 2647918 −3

content. The rock is interpreted as having two

age components of 3212

9 11 Ma and 3127

9 12 Ma, followed by subsequent disturbance.

From the zircon data alone, it is uncertain

whether the ages represent different phases in a

composite sample or whether the protolith is

3127

9 12 Ma, but carries 3212

9 11 Ma

(3)

5. Discussion

U – Pb age determination has been carried out on

zircons extracted from a number of units. Samples

430428, 430475 and 430476 are regional banded

gneisses, 94-0006 is a paragneiss, 94-0338 and

430441 are red-coloured metasomatised gneisses,

430451 is an isotropic charnockite, probably a

diatexite, 94-0000 is a red myrmekitic granite and

430455 is an undeformed diorite dyke intruded into

the supracrustal package. As can be seen from

Table 2 and Figs. 3 – 6, the regional gneisses give

disturbed ages in the range

3600 – 3780 Ma for

the samples within the Aasivik terrain as outlined

in Fig. 1. Sample 430428 south of the bounding

lineament contains

:

3100 and

:

3200 Ma

com-ponents. These ages have not been represented in

the zircon populations in other samples from the

region. The ages found in this sample are within the

range of ages of gneisses from the Akia terrane to

the south (Friend et al., 1988; Nutman et al., 1996)

and it is tentatively suggested that the Tasersiaq

gneiss complex forms a part of the Akia terrane.

The supracrustal sample from the Ukaleq gneiss

complex contains 2750 – 2900 Ma zircons of

prob-able detrital origin, which gives the maximum age

of deposition. This along with a positive

o

_Nd

of the

anatectic granites (

tzircon

o

_Nd

=

2.3 – 5.4; Løfqvist,

unpublished data) formed within the

metasedi-ments suggests that the sources of the sedimetasedi-ments

were dominated by juvenile or only slightly older

material at 2800 – 2700 Ma. The chaotic

intercala-tion of metapelic rocks, mafic granulites and

meta-peridotites together with the occurrence wedged

between a late Arechaean and an early Archaean

gneiss complex suggests that the Ukaleq gneiss

complex form part of a late Archaean accretionary

wedge.

The red granites and the charnockitic diatexite

found in the lineaments separating the different

gneiss complexes are all dominated by 2700 – 2800

Ma zircons. We suggest that this age range is

bracketing the time of deformation in the zones,

and thus the time of tectonic juxtaposition of the

gneiss complexes.

(4)

Fig. 6. U – Pb concordia diagram and relative probability histogram for SHRIMP analysis of zircons from a sample of Ukaleq paragneiss.

The nature of the contact between this gneiss

complex and the Aasivik terrane has not been

investigated in this study, but a pronounced

topo-graphic lineament coincides with contrasts in

stream sediment geochemistry (Steenfelt, 1994),

topographic grain, occurrence of klimberlite

in-trusions (Larsen et al., 1983). This lineament

sepa-rates a region where no early Archaean zircons

have been identified in the gneisses (Kalsbeek and

Nutman, 1996) to the north from the early

Ar-chaean gneisses to the south, and is here taken as

the terrane boundary.

We suggest that the gneiss complexes were

as-sembled at

:

2700 Ma and formed a stable

cra-ton

during

Nagssugtoqidian

orogeny

which

caused penetrative deformation and

metamor-phism of all rock units only 20 – 50 km to the

north.

Kimberlite intrusions which are common in the

adjoining terrains have not been identified within

the Aasivik terrane, which in turn is partly defined

by the apparent absence of kimberlite intrusions.

By applying the well-proven method of circular

reasoning, we can deduce that the terrane

con-straint on the distribution of kimberlite intrusions

indicates that the Aasivik terrane has deep

litho-spheric roots. The occurrence of diamond in a

kimberlite south of the Aasivik terrane (Nunaoil

A

/

S press release, 1996) indicate that the

kimber-lite originated at

\

150 km depth. This is accord

with the barometric estimates by Larsen and

Rønsbo, 1993) for kimberlites along the inferred

western and northern boundaries of the Aasivik

terrane. If this interpretation is correct the early

Archaean terrane is the exposed lower crustal

level of a continent which had already existed for

one billion years at the time of terrane assembly

at 2700 Ma, and not merely a tectonic panel such

as a fault block or nappe. This interpretation is in

accord with Abbott (1996), who suggested that

only accreted material with deep lithospheric

roots could have been stabilised as parts of the

continental crust during the Archaean. In this

respect, the Aasivik terrane differs from the

Akulleq terrane of the Godtha˚bsfjord region,

which represent tectonic slices of an old gneiss

complex intercalated with younger gneiss units.

The Akulleq terrane is dominated by mid-crustal

The diorite dyke 430455 is associated with the

post-kinematic granites, and indicate that

mem-bers of the

:

2700 Ma magmatic suite intruded

the supracrustal rocks of the Ukaleq gneiss

com-plex. The age of 2701

9 16 Ma for the diorite is

in agreement with U – Pb TIMS analysis of zircon

from a granite pluton which gave 2717

9 4 Ma

(J. Connelly, personal communication, 1995).

These ages support the field observation that the

granites were emplaced late in the 2800 – 2700 Ma

tectono-metamorphic event responsible for the

as-sembly of the gneiss complexes.

This study shows that the Aasivik gneiss

com-plex has a major early Archaean component, and

is separated from rock complexes of different ages

and origin by zones of high strain. We suggest

that the Aasivik gneiss complex is a

tectonometa-morphic terrane for which we propose the name

Aasivik terrane. The area immediately north of

the Aasivik terrane is dominated by

:

2800 Ma

felsic gneisses (Kalsbeek and Nutman, 1996),

which are overprinted by the Paleoproterozoic

Nagssugtoqidian orogeny further to the north.

(5)

amphibolite facies rocks, while the Aasivik

ter-rane is at high-pressure granulite facies. The

pos-sibility that the Aasivik terrane represents the

roots of the allochthonous Akulleq terrane should

be investigated further by detailed geochemistry

and chronometry.

6. Conclusions

The region south of the Nagssugtoqidian front

in West Greenland comprises three Archaean

gneiss complexes, which were assembled at 2800 –

2700 Ma. The early Archaean Aasivik terrane

gives U – Pb zircon ages of 3500 – 3600 Ma for the

formation of gneiss protoliths, but probably

com-prise gneisses formed back to

3800 Ma, judging

from the occurrence of zircons with an age of

3780 Ma. This gneiss terrain is dominated by

granitic gneisses with a complicated deformation

history, intruded by large mafic dykes of

un-known age, and now represented by pods and

layers of mafic granulite. No large supracrustal

units have been identified in the early Archaean

gneiss complex at present. The Aasivik terrane

probably underlies

1500 km

. This terrane is in

contact to the Southwest with the Tasersiaq

mid-Archaean gness comples, which we provisionally

interpret as the northern extension of the Akia

terrane (eg. Nutman et al., 1996) of the Nuuk

region. One sample of banded gneiss from this

gneiss complex has given ages of 3127

9 12 Ma

and 3212

9 11 for the formation of gneiss

pro-toliths. The late Archaean Ukaleq gneiss complex

to the Southeast is dominated by supracrustal

rocks with similarities to modern accretionary

complexes and intruded by post-kinematic A-type

granite plutons, which probably also stitch the

tectonic boundaries between the early Archaean

and

the

mid-Archaean

and

late

Archaean

terranes.

Acknowledgements

This study was carried out in collaboration

with the Danish Lithosphere Centre

Nagssug-toqidian project funded by the Danish National

Research Fond. Funding by the Carlsberg

Foun-dation and RSES at the Australian National

Uni-versity, Canberra is gratefully acknowledged. We

thank David Bridgwater, Vic McGregor, Jim

Connelly, Flemming Mengel, Jeroen van Gool

and Clark Friend for discussions and suggestions.

Heidi Thomsen, Nick Rose, Kyle Mayborn and

Esbern Hoch are thanked for company in the

field.

References