Tectono-metamorphic Events in the North Atlantic Region in the Palaeoproterozoic from the View Point of High-grade Metamorphic Rocks in the Lewisian Complex, South Harris, NW Scotland

Gondwana Research, 5, No. 4, pp. 757-770. 02002 International Association for Gondwana Research, Japan. ISSN: 1342-93ix

Gondwana Research

Tectono-metamorphic Events in the North Atlantic Region in the Palaeoproterozoic from the View Point of High-grade Metamorphic Rocks in the Lewisian Complex, South Harris, NW Scotland Sotaro Baba* Department of Crustal Studies, National Institute of Polar Research, Kaga 1-9-10, ltabashi-ku, Tokyo, 173-8515,Japan, E-mail: [email protected] *Presentaddress: Department of Natural Environment, University of the Ryukyus, Senbaru 1, Nishihara, Okinawa 903-0213, Japan, E-mail: babaC3edu.u-ryukyu.ac.jp (Manuscript received July 27,2001; accepted May 9,2002)

Abstract The tectono-thermal history of the Lewisian Complex in South Harris (South Harris Complex) was inferred from its geologic and metamorphic characteristics. The lithological assemblages and geochemical features of the complex suggest that its precursory rocks were composed of the subduction-related accretionary complex formed in the palaeo convergent margin. The complex has suffered the ultra-high temperature (UHT) metamorphism that was contemporaneous with the igneous activity to make the South Harris Igneous Complex (SHIC) and the subsequent continent-continent collisional activity A similar complex recording the geological processes of the subduction, the UHT metamorphism and the collision has been recognized in the Lapland-Kola belt and New Quebec in the Palaeoproterozoic. This suggests an assembly of micro-continents to form the Palaeoproterozoic supercontinent in the North Atlantic region.

Key words: Lewisian Complex, accretionary prism, anticlockwise P-T path, UHT metamorphism, Palaeoproterozoic.

Introduction The Palaeoproterozoic collisional event in the North Atlantic region has been one of the important keys to understand evolving continents in early earth history. Baba (1997, 1998, 1999) pointed out the Palaeoproterozoic collisional event in the Lewisian Complex in South Harris (South Harris Complex) in the Outer Hebrides based on geological and petrological studies. Whitehouse and Bridgewater (2001) and Park et al. (2001) have focused on the event in the Lewisian Complex in the Scottish mainland based on their geochronological and geochemical studies of the Loch Maree Group. The mainland Lewisian records two episodes of granulite-faciesmetamorphism at 2.7 and 2.5 Ga, a 1.7 Ga kyanite-forming amphibolite-facies metamorphism (Whitehouse, 1989; Cohen et al., 1991; Corfu et al., 1994; Friend and Kinny, 1995; Zhu et al., 1997). In

addition, geochronological study has revealed that the central and northern parts of the mainland Lewisian derived different protoliths (Kinny and Friend, 1997). These data presumably imply the existence of plural microcontinents and their collision during late Archaean to Palaeoproterozoic. Tectonothermal events at 2.1 to 1.8 Ga in the Lewisian Complex have been documented in the South Harris Complex in the central part of the Outer Hebridean Isles with the Sm-Nd isotopic study by Cliff et al. (1983). Cliff et al. (1998) summarized the evolutionary history of the northern Outer Hebrides and concluded that the Leverburgh belt in the South Harris Complex was a postArchaean component including an Archaean supracrustal sequence. However, an evolutionary history of South Harris has not been documented well from the viewpoint of useful geological and petrological data. The author proposed the following new insights based mainly on the

S. BABA

758

geological and petrological study for South Harris (Baba, 1997, 1998, 1999): (1) the precursor rocks of the Leverburgh belt are formed in an accretionary prism. (2) the rocks have suffered a high-grade metamorphism and experienced an anticlockwise P-T path in the Palaeoproterozoic (2.1-1.8 Gal. (3) ultra-high temperature metamorphism is related to Proterozoic igneous activity. These geological and metamorphic features in South Harris are different from those on the mainland but provide useful information for the Palaeoproterozoic tectono-thermal history of the North Atlantic region. This paper reviews the previous studies in the South Harris Complex and discusses Palaeoproterozoic tectonometamorphic events in the North Atlantic region using the previous and new data.

Geological Background The Lewisian Complex in NW Scotland is a well-known, high-grade gneiss terrane composed largely of TTG (tonalite-trondhjemite-granodiorite)gneisses (Fig. 1). I7"lO'W Leverburgh belt BennObbe

On the Scottish mainland, many geological, petrological and isotopic studies have been undertaken by numerous researchers over several decades. Well-summarized review papers on the Lewisian Complex, despite lacking new age data, have been published by Park et al. (1994). The South Harris Complex, exposed in the southern part of the island of Harris (Fig. l ) , is composed of metasedimentary and metabasic rocks of the granulitefacies Leverburgh belt, the amphibolite-facies Langavat belt and the Palaeoproterozoic South Harris Igneous Complex (granulite-facies). Recently, Cliff et al. (1998) performed Sm-Nd age determination for the rocks in the northern Outer Hebrides part. They concluded that the Leverburgh belt originally consisted of post-Archaean sediments, and that the post-metamorphic cooling age of the granulite-faciesmetamorphism was Palaeoproterozoic (1.87-1.83 Ga). Detailed geological information on the Lewisian Complex in the Outer Hebrides and the South Harris Complex is summarized by Fettes et al. (1992) and Baba (1997, 1998, 1999). In this paper, I propose an evolutionary model for the South Harris Complex based on the metamorphism and tectonic setting of precursory rocks of the Leverburgh belt.

17'05'W

17"00'w

n

A

Lengavat belt

South Herrls Igneous Complex

Yetagabbro

Fig. 1. Simplified geological map of the Lewisian Complex in North-West Scotland (after Park et al., 1994), and the South Harris Complex (after Dearnley, 1962; Baba, 1997).

Gondwana Research, V. 5, No. 4,2002

PALAEOPROTEROZOIC EVENTS IN SOUTH HARRIS, NW SCOTLAND

Tectonic Setting of Precursor Rocks Lithological assemblage The constituent rocks of the Leverburgh belt are divisible into four lithological series, namely, Benn Obbe Series (A), Benn Obbe Series (B), Rodel Series and Chaipaval Pelitic Series (Baba, 1997). Their lithological characteristics are summarized below. The Benn Obbe Series (A) consists of quartzofeldspathic, pelitic and mafic gneisses, and rare ultramafic rocks. The quartzo-feldspathic gneiss is subdivided into two types, well-banded and layered, reflecting their field occurrences. The well-banded gneiss is locally interbedded with kyanite-bearing pelitic gneiss, and hence is of probable sedimentary origin. The pelitic gneiss is also divisible into two types, Fe-Al-rich and Na-K-rich; the former is typicallyrich in iron-oxidesand in Fe-rich minerals such as hercynite and staurolite. Several types of sapphirinebearing gneiss (Mg-rich) were found in this series (Baba, 1999; Baba et al., 2000; Baba, in submitted, 2002). The Benn Obbe Series (B) consists of layered quartzofeldspathic, pelitic, banded quartzo-feldspathic and mafic gneisses, with a small amount of leucocratic gneiss (Kfs-rich garnet-bearing quartzo-feldspathic gneiss). An occurrence of intercalated Fe-Al-rich pelitic gneiss in the banded quartzo-feldspathic gneisses is absent in this series. Layered quartzo-feldspathic gneisses are particularly well exposed. Along the western boundary, these gneisses are mylonitized, show a narrowly-spaced foliation and have retrograde assemblages. The Chaipaval Pelitic Series comprises pelitic, leucocratic, quartz-biotite and mafic gneisses, with rare, interbanded quartzo-feldspathic gneiss. Pelitic gneiss belongs to a Na-K-rich type and is best developed in the Northton - Toe Head region. Leucocratic gneiss occurs widely as sheets in the pelitic gneiss on the west side of Chaipaval. The Rodel Series consists mainly of alternating assemblages of strongly foliated garnet-biotite gneiss, biotite-hornblende quartzose gneiss, mafic gneiss (garnet+ clinopyroxene orthopyroxene, garnet + amphibole and clinopyroxene amphibole assemblages), orthoamphibolebiotite gneiss, sulphide-bearing quartzite and pelitic gneiss. This series, in contrast to the other main series, shows the following distinctive characteristics: (1) it is finely foliated, with common alternations of rock-types; (2) it consists of a wide variety of metamorphic rocks, such as pelitic gneiss, mafic gneiss, orthoamphibolebearing gneiss, sulphide-bearing quartzite and marble/ calc-silicate rocks, all of which occur as lenses or thin layers; (3) garnet in pelitic gneiss is generally finer-grained than that in the other series; (4) the presence of marble

+ +

Gondwana Research, V. 5, No. 4,2002

759

or calc-silicate blocks and lenses, which occur at several horizons, but appear to be restricted to the Rodel Series. The characteristics of each series are summarized as follows: The Benn Obbe Series (A) is characterized by quartzo-feldspathic gneiss with small amounts of two distinct types of pelitic gneiss (Na-K-rich and Fe-Al-rich) with high-Mg metasediment; the Benn Obbe Series (B) is dominated by layered quartzo-feldspathic gneiss, but lacks Fe-Al-rich pelitic gneiss. Pelitic gneisses are notably well developed in the Chaipaval Pelitic Series; these pelitic gneisses are rich in aluminosilicates (kyanite or sillimanite) and contain leucocratic gneiss sheets. The Rodel Series is characterized by a wide range of rock types such as marble/calc-silicate rocks, pelitic gneiss, mafic gneiss, quartzite and ultramafic pods, which generally occur as discontinuous layers. Boundaries between the four series are locally marked by shear zones.

Geochemical characteristics of the basic metamorphic rocks The whole-rock chemistry of mafic gneisses was determined with the aim of estimating the bulk composition of precursor rocks. The characteristic major and trace elements of the meta-basic rocks give valuable information for estimating the tectonic environment of the protolith in comparison with modem plate-tectonic systems. The plate-tectonic environments and related lithological associations in the Proterozoic are considered to be very similar to those of the Phanerozoic (see, Condie, 1989). Based on trace-element patterns as expressed in discrimination diagrams of TiO, / Zr and Ni / Ti (Figs. 2a and 2b), primitive mantle-normalized abundance patterns (Fig. 3), and MORB-normalized spider diagrams (Baba, 1997), the igneous protoliths of the basic metamorphic rocks of the Rodel Series appear to have been derived from MORB (mid-ocean ridge basalt) and/or OIT (oceanisland tholeiite) (E-MORB: generally occurring near seamounts or oceanic islands), and those of the Benn Obbe Series (A) and (B) from rocks which have subductionrelated magmatic products such as IAB (island arc basalt) and/or CAB1 (calc-alkaline basalts from island arcs). The mafic rocks in the Chaipaval Pelitic Series may be derived from OIT or WPB (within-plate basalt), in view of the differing affinities given by the discrimination diagrams and the MORB-normalized patterns (Baba, 1997).

Tectonic setting of precursor rock The tectonic setting of the precursor rock types to the Leverburgh belt can be interpreted from their lithological assemblages combined with the geochemical characters of the basic metamorphic rocks. In the Benn Obbe Series (A) and (B), garnet-pyroxenebearing quartzo-feldspathic gneiss is the most common

760

S. BABA

1c

I

I

9

0"

i=

1

0 Rodel Series

0 Rodel Series

I Chaipavai Pelitic 0 Benn Obbe Serles(A) 8 Benn Obbe Serles(B)

0 Benn Obbe Series(A) 8 Benn Obbe Series(B)

0.1

5000

I

1

100 Zr PPm

10000

15000

20000

I

1000

Ti ppm

Fig. 2. Chemical variations of basic metamorphic rocks (modified after Baba, 1997 and with unpublished data). (a) Zr-TiO, diagram. Fields are after Pharaoh and Pearce (1984). (b) Ti-Ni diagrams. Fields are from Ishizuka (1981). WPB-within plate basalt, MOM-mid-ocean ridge basalt, ACB-arc basalt, OIT-ocean-island tholeiite, IAB-island arc basalt.

[ Rodel Series

'Oo0

1000

Benn Obbe Series

100

100

10

10

1

1

I

0.1

RbBaThUFbKLafePbPrSrP MZrSmEuTiDyY

y b h

Fig. 3. Multi-element plots (primitive mantle-normalized) of the composition of basic metamorphic rocks. N-MOM,E-MOM, OIT, and normalized values are after Sun and McDonough (1989).

Gondwana Research, V 5, No. 4,2002

PALAEOPROTEROZOICEVENTS IN SOUTH HARRIS, NW SCOTLAND

rock type. The precursors of these quartzo-feldspathic gneisses are considered to be Ca-rich greywackes, by comparing high-temperature experimental work using greywackes as starting materials (Vielzeuf and Montel, 1994) with the observed mineral assemblages in both series. The well-banded quartzo-feldspathic gneisses with minor intercalations of Fe-Al-rich pelitic gneiss occur adjacent to Na-K-rich pelitic gneisses, suggesting that these rocks were formerly greywacke turbidites alternating with pelites (shales). Greywacke turbidites with a minor pelagic component are considered to have been trench sediments (Condie, 1989).Al-Fe-rich pelitic gneiss, which only occurs in the Benn Obbe Series (A), was most likely deposited as a proximal facies at the trench. The basic metamorphic rocks from the Benn Obbe Series (A) and (B) show island arc tholeiitic affinities; therefore, the turbidite sequence was intruded by basic intrusion (sheet, dyke or sill) in an island arc or continental-margin arc setting. In the Chaipaval Pelitic Series, kyanite- or sillimanitebearing pelitic gneiss and K-feldspathicgneiss (leucocratic gneiss) are dominant, and these gneisses alternate with subsidiary biotite- and hornblende-bearing quartzose gneiss. The precursors of the Chaipaval Pelitic Series are considered to be K-Al-rich pelites locally alternating with a small amount of psammitic material. The sedimentary precursors may have been deposited as clastic sediments in the proximal facies of a trench or as hemipelagic sediments on oceanic crust. Basic metamorphic rocks are rare, but the few sampled rocks from this series show T-MORB (transitional MORB) and E-MORB(enriched MORB) affinities. The basic rocks occur as lenses and pods, suggesting that they may have been taken up tectonically by sediments as small blocks during the subduction of oceanic crust. The Rodel Series is characterized by a lack of continuous lithological layering, and by numerous inclusions of fragments of impure limestone (marble/calc-silicate rocks), greywacke (quartzo-feldspathic gneiss), mudstone (pelitic gneiss), chert (sulphide-bearing quartzite?), basalt (basic gneiss) and ophiolite (meta-ultramafic rock and garnetiferous mafic gneiss). This lithological association and occurrence are typical of melange assemblages. Therefore, the Rodel Series is interpreted as a possible accreted mdange in an accretionary prism. The meta-basic rocks show N-MORB (normal MORB), T-MORB-and E-MORB-like affinities. Rock assemblages and the geochemistry of the meta-basic rocks from the Rodel Series suggest that it was formed from oceanic materials related to the subduction of oceanic lithosphere. The possible tectonic settings of the precursor rocks of each series are summarized in table 1. Gondwana Research, V. 5, No. 4,2002

761

Table 1. Summary of precursor rock of the Leverburgh belt. Rodel Series Lithology: Lack of continuous bedding, and various inclusion of rock fragments of mafic gneiss (basalt), meta-ultramafic gneiss (ophiolite fragment?), marble/calc-silicate (impure limestone), quartz-feldspathic gneiss (greywacke). Basic rock MORB or OIT (oceanic island or seamount). Possible tectonic setting: Melange. Benn Obbe Series (A) and (B) Lithology: Quartzo-feldspathic gneiss with alternating pelite (greywacke turbidites with a minor pelagic component). Basic rock IAT (island arc tholeiitic basalt) or C h I (island arc calcalkaline basalt). Possible tectonic setting: Trench sediment. Chaipaval Pelitic Series Lithology: Aluminosilicate-rich pelitic gneiss and K-feldspar rich pelitic gneiss (pelite). Basic rock: OIT or WPB (within-plate basalt). Possible tectonic setting: Proximal facies of trench sediment or hemipelagic sediment.

Metamorphism and Inferred Tectonic History Metamorphic P-T path The metamorphic history of the South Harris Complex has been determined from the following mineral textures and compositions observed in the pelitic, quartzofeldspathic and mafic gneisses, with particular emphasis on pelitic gneisses from the Leverburgh belt. (1) Some coarse-grained garnets in the pelitic gneiss include biotite and quartz in the inner core, sillimanite in the outer core, and have overgrowths of kyanite in the rims. (2) Garnet grains in the pelitic gneiss show a progressive increase in grossular content from outer core to rim. (3) The Aim/ Allv ratio of clinopyroxene in both mafic gneiss and metagabbro in the SHIC increases from core to rim. (4) Reaction coronas of cordierite and hercynite+ cordierite are formed between garnet and kyanite, and orthopyroxene+cordierite and orthopyroxene+ plagioclase reaction coronas occur between garnet and quartz in the Fe-Al-rich pelitic gneiss and quartzofeldspathic gneiss. (5) In the Fe-Al-rich pelitic gneiss, an early stage of the P-T path is deduced from inclusion assemblages in garnet and from staurolite-breakdown reactions to produce garnet +sillimanite and garnet+ sillimanite+ hercynite with increasing temperature. (6) In sheared and strongly foliated rocks, hydrous minerals such as biotite, muscovite and hornblende form a foliation that modifies pre-existing textures.

762

S. BABA

(7) In Opx-Ky granulites, the orthopyroxene and kyanite are intergrown in a stable mineral assemblage, and indicates metamorphic conditions > 12 kbar at 800900°C. Sillimanite inclusions within orthopyroxene, which is surrounded by kyanite, suggest that sillimanite was stable earlier and lyanite was formed later (Fig. 4a).

(10) Sapphirine inclusions within garnet contain inclusions of tiny, rounded spinel, cordierite and corundum. These textural relationships indicate that sapphirine was formed at the expense of these inclusion minerals at high-T and relatively low-P conditions, and that is subsequently broken down to form garnet at high-P conditions (Fig. 4b; Baba, in submitted, 2002). The inferred metamorphic history of the Leverburgh belt is divided into 4 stages, ( M l ) prograde metamorphism with increasing temperature, (M2) metamorphism with increasing pressure, (M3) retrograde decompressional metamorphism characterized by decreasing pressure and (M4) retrograde metamorphism accompanied by shearing. The P-T conditions of the M 1 and M2 stages are <980°C, 8-10 kbar and 800-900 "C, <14 kbar, respectively. The maximum temperature conditions of 95O+3O0Cat 10 kbar are estimated from orthopyroxene isopleths (Harley, 1998) for aluminous-orthopyroxeneporphyroblasts (<9.7 wt. %). The two types of orthopyroxene-aluminosilicate granulite indicate that the peak metamorphic conditions were over 900"C, and this is comparable with ultra-high temperature metamorphism. These granulites were formed during the Palaeoproterozoic metamorphism'due to the emplacement of the SHIC. The M3 stage metamorphic condition is considered to be <800"C, < 8 kbar. A pressure increase from M 1 to M2 and the P-T conditions at each stage indicate that the metamorphic P-T path was anticlockwise. Further detailed discussions were summarized in Baba (1998, 1999) and Baba et al. (2000).

Tectonothermal history

Fig. 4. Photomicrograph of mineral relation. (a) Orthopyroxene includes sillimanite, and forms an intergrowth with kyanite in Opx-Ky granulite (Baba, 1999). (b) Sapphirine includes spinel (sometimes together with corundum and cordierite), and is enclosed by garnet (Baba, in submitted, 2002). Width of the photomicrographs is (a) 1 mm and (b) 2.2 mm. Plane-polarized light.

(8) In Opx-Sil granulites, the orthopyroxene+sillimanite +garnet +sapphirine assemblage is stable at a peak metamorphic stage, indicating P-T conditions of 930-950°C, > 8 kbar, according to the FMAS petrogenetic grid (Hensen and Harley, 1990). Similar conditions were obtained by using orthopyroxene-garnet geothermobarometers (Baba, 1999). (9) Surinamite [(Mg,Fe2+),A1,BeSi,0,,] includes orthopyroxene, sillimanite, and Si-rich sapphirine (khmaralite), suggesting that the surinamite was formed as a result of increasing pressure at a high temperature (Baba et al., 2000).

The P-T path deduced for the metamorphic evolution of granulites in the Leverburgh belt is shown in figure 5. The P-T path is determined from the evidence for an early prograde history deduced from mineral textures and mineral compositions, and from estimates of P-T conditions for the two stages of prograde and two stages of retrograde metamorphic assemblages. The granulites in the Leverburgh belt initially experienced a prograde path with increasing temperature within the sillimanite stability field, and attained a thermal peak at M1. The prograde part of the inferred P-T path is indicated by prograde reactions of staurolite-breakdown and other dehydration reactions seen in inclusions within garnet and clinopyroxene porphyroblasts. The M 1 metamorphism, probably an isobaric heating metamorphism, was most likely caused by the emplacement of the SHIC as a heat source (Baba, 1998, 1999). The ultra-high temperature conditions, 950k 30"C, characterized by the aluminous orthopyroxene + sillimanite sapphirine quartz assemblage, were

+

+

Gondwana Research, V. 5, No. 4,2002

PALAEOPROTEROZOIC EVENTS IN SOUTH HARRIS, NW SCOTLAND

763

P (kbar) 14

12

10

8

6

4 400

500

600

700

800

900

1000 1100 T

("C)

Fig. 5. Metamorphic P-T path of the Leverburgh belt (after Baba, 1998, 1999 and in submitted, 2002). Al,SiO, triple point is after Holdaway and Mukhopadhyay (1993). Maximum temperature conditions of M1 were estimated from Opx-isopleths (H and H: Hensen and Harley, 1990; H: Harley, 1998) and petrogenetic grids (ornamented field) (Harley, 1998). M2 conditions were calculated by Grt-Cpx-P1-Qtz (Ec: Eckert et al., 1991) and Grt-Ky-P1-Qtz (Nand H: Newton and Haselton, 1981) equilibria and Grt-Cpx thermometry (E and G: Ellis and Green, 1979). Opxisopleths for orthopyroxene-kyanite association are also shown. The temperature conditions of the early part of the retrograde path are based on sapphirine-spinel exchange thermometry (Baba in submitted, 2002). M3 conditions (an oblique line field) are after Baba (1998).

probably formed by emplacement of the SHIC prior to the formation of kyanite and surinamite (Baba, 1999; Baba et al., 2000). The idea of emplacement of SHIC at M 1 and the thermal effect of UHT metamorphism are consistent with the previous observations. As igneous equilibration temperatures of 1135 to 1315°C (+ 7OOC) were obtained from the SHIC by Witty (in Fettes et al., 1992), the age of M 1 might coincide with the age of intrusion of the SHIC. However, the emplacement of the SHIC was thought to have occurred in several stages during 2.2 Ga to 1.8 Ga, with the anorthosite at first, followed by the gabbro and diorite (Cliff et al., 1983). The intrusion age of about 2.2 to 1.8 Ga may be regarded as the age of the M 1 metamorphism in response to continuous heating by multiple intrusions. In the M2 stage of metamorphism, the pressure increased progressively. M2 pressure conditions are consistent with a crustal depth of 45-50 km, and the pressure increase from the M 1 to M2 metamorphism might reflect thrusting of a continental crust over the South Harris belt. According to Sm-Nd studies of the granulitefacies mineral isochrons, high-pressure granulite facies Gondwana Research, V. 5, No. 4,2002

metamorphism (M2 stage) took place at least earlier than 1.87-1.83 Ga (Cliff et al., 1983; Cliff et al., 1998). Unfortunately, precise age determinations of the M 1 and M2 stages are not possible from existing data. The retrograde path after peak-M2 metamorphism involved a decompression path associated with the development static recrystallization. Isothermal decompression paths have been generally considered to be due to rapid exhumation of overthickened crust as a result of isostatic rebound (Ellis, 1987), or to the extensional thinning of thickened crust (Harley, 1989). Based on mineral textures, Baba (1998) concluded that the M 3 stage probably underwent non-deformational static conditions reflecting isostatic rebound. However, the M3 stage extends from M2 down to conditions of 600+ 50°C and 5-7 kbar, with the formation of different types of corona and the replacement of kyanite by sillimanite. In sheared and foliated rocks, biotite and hornblende crystals were produced by further retrograde hydration during M4, and they form a distinct new foliation, which truncates the M3 metamorphic textures (e.g., orthopyroxene +plagioclase corona). This foliation

764

S. BABA

probably resulted from crustal scale deformation after the M3 stage. In the zones where deformation was enhanced, hydrous minerals, which characterize the M4 stage, were developed. Further detailed study of the interaction of the M3 and M4 stages is needed. Nevertheless, the M4 stage is considered to be part of a later collisional event (e.g., second orogenic event; Ellis, 1987) associated with deformation, which enhanced exhumation and uplift of the South Harris high-pressure granulites.

Tectonic Model of the South Harris Complex The rock assemblages in the Leverburgh belt can be classified into the following types: melange (Rode1Series) with oceanic-type volcanics; trench sediments with arctype volcanics [Benn Obbe Series (A) and (B)] and trench or hemipelagic sediments with ocean-type volcanics (Chaipaval Pelitic Series). The various rocks showing different protoliths are observed in a narrow region, and are similar to those of modern arc-trench systems. The idea that Archaean and Palaeoproterozoicrocks are similar to those in modern arc-trench systems has been well documented by several workers (e.g., Helmstaedt and Scott, 1992; Kimura et al., 1993; Windley, 1993, 1995). Archaean oceanic assemblages have comparable HFS/FEE relative to MORE3 and oceanic plateaus, except for being generally richer in FeO than modern MORB (e.g., Storey et al., 1991). Previous studies support a tectonic setting of precursory rocks of the South Harris Complex deduced from the chemical compositions of mafic gneiss and the lithological constitutions. Cliff et al. (1998) concluded that the Leverburgh metasediments belonged to the Proterozoic supracrustal belt of the Loch Maree Group, rather than to an Archaean metasedimentary component, as exposed in the Scottish mainland. In addition, the SHIC is tholeiitic - calc-alkaline and alkaline in nature, resembling that of igneous rocks in magmatic arcs on continental margins (Fettes et al., 1992). Therefore, it is concluded that the Leverburgh belt represents a Palaeoproterozoic accretionary complex, which formed during the subduction of an oceanic plate to form magmatic arcs prior to regional metamorphism and deformation. Recently, Whitehouse and Bridgewater (2001) suggested that a psammite from the Chaipaval Pelitic Series was derived from the high-level equivalent of the presently exposed SHIC. However, the psammite in the Leverburgh belt described by Whitehouse and Bridgewater (2001), presumably identical with massive quartzo-feldspathic gneiss in this paper, is very similar in appearance to the fine-grained meta-tonalite exposed in the marginal part of the meta-anorthosite body in terms

of its mineral assemblage. The possibility is raised tentatively that the Chaipaval Pelitic Series is younger than the other series, and accreted at a later stage. However this possibility requires further detailed investigation. The anticlockwise P-T path for the Leverburgh belt is different from anticlockwise paths that involve isobaric cooling at pressures of 5-8 kbar (Bohlen, 1991; Harley, 1989), because in the South Harris Complex pressure increased along the prograde path up to 14 kbar. Furthermore, the observed retrograde reaction coronas are similar to those of typical overthickened granulite terranes formed by continental collision (Harley, 1989). The inferred P-T path can be explained by significant or substantial loading (collision) after magmatism, resulting in the South Harris granulites preserving mineral textures consistent with an isothermal decompression, which is typical of overthickened granulite terranes formed by continent-continent collision. An anticlockwise P-T path in a granulite terrane is likely to be characteristic of magmatically thickened and heated crust, such as a continental arc (Bohlen, 1987). I propose that the South Harris anticlockwise metamorphism reflects magmatism at a continental margin or an island arc, followed by overthrusting during continental collision. The evolutionary history of South Harris deduced from the nature of sedimentary precursors, the character of igneous activity and the metamorphic history is as follows: (1) The South Harris Complex was formed in a magmatic arc on a continental margin (Fig. 6 , Model 1) or an island arc (Fig. 6 , Model 2), and the precursory rocks were subduction-related accretionary complex in the palaeo convergent margin. In these tectonic settings, the SHIC intrudes, and gives thermal affection to make UHT metamorphism (Fig. 6, subduction-magmatism).There is no room here for detailed discussion about which models to adopt, i.e., whether the magmatism occurred on a continental margin or on an island arc, because the tectonic setting of the magmatism of the SHIC was estimated mainly on the characteristics of the major elements, not on the geochemistry of the trace and rare-earth elements. (2) The continental blocks eventually collided along the South Harris granulite belt after a temperature increase at M 1 stage. During the collision, thrusting of continental crust over the South Harris granulites caused the M2 high-pressure metamorphism at around 1.83-1.87 Ga (Fig. 6, collision). (3) Following the collision, exhumationof overthickened crust occurred as a result of isostatic rebound or extensional uplift (Fig. 6 ) . At some later stage (M4), a second collisional event occurred, and exhumation and uplift of South Harris Complex then took place (Fig. 6, shearing). Gondwana Research, V. 5, No.4,2002

PALAEOPROTEROZOIC EVENTS IN SOUTH HARRIS, NW SCOTLAND

765

Subduction magmatism

-

Model 1: pokm

'

_

Maamatism

..

P

Model 1:

'

SHIC

Collision-

Collision of small continent

I

45-5Okm

K-rich granitic magma?

Model 2:

Collision of small continent

I

45-5Okm

K-richgranitic magma?

U p l i e

axfordian granite ?

Uplift (shearing) S

Secondarycollision

S

Preexisting continent 1

Granite and migmatitecomplex

Pre-existing continent 2

[rm Pre-existing continent 3 PalaeoproterozoicTectonothermal Events in the North Atlantic Region The Lewisian Complexhas been considered to be closely comparable in structural history with the Nagssugtoqidian of both East and West Greenland, and the West Nain province of Labrador (Park et al., 1994; Myers, 1987; Korstgird e t al., 1987). The Palaeoproterozoic tectonothermal events in the Lewisian have recently been the subject of several studies (Park, 1995; Droop et al., Gondwana Research, V: 5, No. 4, 2002

N

Fig. 6. Two plate-tectonic models for the evolution of the Lewisian Complex in the Outer Hebrides (Baba, 1999). Model 1:Leverburgh belt underlies the continental margin and SHIC emplaced in the continental margin. Model 2: Leverburgh belt formed by the accretion of the island arc and SHIC emplaced in the island arc. Pre-existing continent consisting of grey gneiss: 1= grey gneiss complex which has U-Pb zircon age of 2.77 Ga (Pidgeon and Aftalion, 1972) dominant in Lewis, 2 = grey gneiss complex Pb-Pb age of 2.64k0.12 Ga (Moorbath et al., 1975), Sm-Ndt CHUR age of 2.62-2.68 Ga (Whitehouse, 1990) and U-Pb zircon age of ca. 2.83 Ga (Whitehouse and Bridgewater, 2001) dominant in South and North Uist, 3 = speculative grey gneiss complex not exposed now by erosion.

1999; Park et al., 2001). Park (1995) reviewed the relationships of the Lewisian Complex and neighbouring Palaeoproterozoic belts of Laurentia and Baltica. He discussed four stages of geological processes for the regions between ca. 2.6 and 1.5 Ga. (1) 2.6-2.4 Ga: development of a conjugate shear-zone system in the North Atlantic Craton. (2) 2.4-2.0 Ga: rifting and dyke emplacement in the older craton; creation of oceanic and intracontinental basins.

766

S. BABA

External belts

IIIIIJI]-1.9 -1.85 Ga

-1.85 -1.5 Ga

Fig. 7. Simplified reconstruction of Laurentia and Baltica for 1.2 Ga (modified after Buchan et al., 2000; Jackson and Berman, 2000). The stars mark the localities of possible high-P granulite-facies metamorphism and UHT metamorphism at Palaeoproterozoic age (see text). WNagWest Nagssugtoqidian belt, ENag-East Nagssugtoqidian belt.

(3) 2.0-1.8 Ga: subduction at active margins of older cratons with creation of magmatic arcs; collision of cratons accompanied by closure of intra-cratonic basins and accretion of arc terranes. (4) 1.8-1.5 Ga: development of a new active margin discordant to the previous ones, with significant change in convergence direction within the amalgamated continental assembly. A recent reconstruction of Laurentia and Baltica for the period at ca. 1.2 Ga is shown in figure 7 (modified after Buchan et al., 2000; Jackson and Berman, 2000).

Origin of precursor rocks in the 2.0-1.8 Ga collisional suture zone With regard to the 2.4-2.0 Ga precursory rocks in the 2.0-1.8 Ga collisional suture, ophiolites and mafic volcanics (ocean-floor material) and passive-margin sequences were recognized in the New Quebec and Lapland-Kola belts.

In North Quebec, the MORB and OIB character of the ca. 2.0 Ga Watts Group was reported by Scott et al. (1992) and St-Onge et al. (2000). This unit is interpreted as an ophiolitic sequence. Tholeiitic basalt and ultramafic rocks from Baffin Island were interpreted to have an origin similar to the Watts Group (Jackson and Berman, 2000). Palaeoproterozoic pelite-dominated metasedimentary sequences (Tasiuyak gneiss) in the Torngat Orogen are considered to have been formed in an accretionary prism setting, and their protoliths to have been deposited at 2.0-1.9 Ga (Scott, 1998). Connelly (2001) reported that the tectonic environment was formed by the convergence of the Archaean Rae and Nain Cratons. In the Kola-Lapland belt, detailed studies have been made of Palaeoproterozoic geological processes. Several notable insights are as follows: (1) Berthelsen and Marker (1986) suggested that 2.4-2.0 Ga metavolcanic and metasedimentary sequences Gondwana Research, V: 5, No. 4,2002

PALAEOPROTEROZOIC EVENTS IN SOUTH HARRIS, NIV SCOTLAND

in the Kola suture were originally part of the attenuated margin of the North Atlantic Craton. (2) A Palaeoproterozoic supracrustal sequence involving a greenstone belt with komatiitic pillow lava (2.1-2.0 Ga: Often, 1985) in the Karasjokgreenstone belt might be a palaeo-accretionary complex. (3) Daly et al. (2001) proposed that juvenile protoliths of the Lapland Granulite belt, Umba and Tersk terranes developed as the products of arc magmatism, as a result of subduction of the ‘Lapland- Kola’ Ocean, which formed by the continental drift following rifting and terrane dispersal initiated at ca. 2.45 Ga. (4) A metaflysch sequence in the Lapland Granulite belt is interpreted as having been deposited in a back-arc basin to the southwest of the Kola Ocean (Marker, 1985). The period 2.4-2.0 Ga is considered to be associated with a widespread extension (Balagansky et al., 2001). As described above, the formation of subduction-related sediment associated with oceanic crustal material has been reported from several North Atlantic regions at 2.4-2.0 Ga. They imply existence of similar geological process of subduction and formation of accretionary complex to the Lewisian at South Harris.

Igneous activity and metamorphism at 2.0-1.8 Ga The 2.0-1.8 Ga events have previously been interpreted in terms of subduction and collision in Laurentia and Baltica (see, Park, 1995). It is indicated by the presence of subduction-related calc-alkaline plutonism and collision-related deformation and plutonism. The distribution of anorthosite suites and UHT metamorphism in Laurentia and Baltica are summarized here, because both are characteristic features of the South Harris Complex. In the Kola region, a gabbro-anorthosite suite emplaced at ca. 2.45-2.46 Ga has been found in the Kolvista belt in the Kola Peninsula (Frisch et al., 1995). Meta-gabbroic and anorthositic rocks are also recognized in the Tanaelv belt, which is considered to be a northern extension of the Kolvista belt (Barbey et al., 1980). The large ca. 1.9 Ga Vaskojoki anorthosite also occurs in this belt (Bernard-Griffiths et al., 1984). In the Torngat Orogen (NE. Quebec and N. Labrador), Scott and Machado (1995) reported the Hutton anorthosite and related anorthositic orthogneiss which has an emplacement age of > 1.8 Ga. Similarly, the Ammassalik Intrusive Complex (Angmagssakik Charnockite Complex) in SE. Greenland, although lacking anorthosite, formed at 1.89 Ga (Hensen and Kalsbeek, 1989) and profoundly affected the thermal history of country rock (Kalsbeek et al., 1993). In the SHIC, anorthosite occurred at 2.2-1.8 Ga (Sm-Nd data, Cliff et al., 1983). The thermal affect associated with emplacement of this anorthosite is similar to the Gondwana Research, V. 5, No. 4,2002

767

Ammassalik Intrusive Complex, but precise information from other regions is not available. Possible 2.0-1.8Ga high-pressure granulites, recording similar conditions to the M2 stage metamorphism in South Harris, have been reported from the Tanaelv belt in the Kola region (Bylinski et al., 1977 in Marke, 19851, the South Torngat Orogen in Labrador (Mengel and Rivers, 1991) and the Ammassalik area in Greenland (Kalsbeek et al., 1993). The textures of Grt-Opx-Sil association and high-Al orthopyroxene, which are indicative of possible UHT metamorphism, were reported from the Umba block granulite-facies metasediments of the southeastern extension of the Lapland Granulite belt (Alexejev, 1997; Glebovitsky et al., 2001). In addition, rocks from these terranes also record a decompressional tectonothermal history that is very similar to that described from the South Harris M3 stage. In North America, the Boothia-Somerset granulite terrane and its possible extension of the 1.9-2.0 Ga Taltson - Thelon Orogen have several localities of UHT assemblage such as high-Al Opx in sapphirinebearing rocks and a hercynite/spinel-quartz associations (Grover et al., 1997; Berman and Bostock, 1997; Kitsul et al., 2000). At these localities,the metamorphic pressures, up to 7 kbar, was somewhat lower than that of South Hams. It is generally accepted that the enormously high geothermal gradient as a result of magmatic heat: source such as magmatic underplating or emplacement of anorthosite, is required for UHT metamorphism (e.g., Sandiford et al., 1987). Therefore, magmatic arcs could be a possible tectonic setting for such high temperature metamorphism. The 2.0-1.8 Ga UHT metamorphism in the North Atlantic region could therefore be a product of subduction-relatedmagmatism. In the case of the Lewisian Complex at South Harris, a suite of anorthosite and gabbro occur adjacent to the UHT metamorphic rocks, and they are interpreted as the heat source for UHT metamorphism (Baba, 1998, 1999). These igneous bodies have a calcalkali affinity, and thus provide strong evidence in support of magmatic arc setting. Possible heat source and a suitable tectonic model to create other UHT rocks in the North Atlantic region have not been proposed, although these rocks were products of a magmatic arc setting similar to Lewisian at South Harris. Further petrological studies and tectonic modeling that take into account involving tectonothermal history are needed for useful comparisons. Apart from UHT metamorphism, high-pressure granulitefacies metamorphism and subsequent isothermal decompression are recognized from several areas of the North Atlantic region, and both indicate the presence of overthickened crust caused by continental collision and subsequent uplift. Based on such geological evidence, it is tempting to speculate the existence of a huge mountain

768

S. BABA

chain and supercontinent caused by assembly of microcontinents throughout the North Atlantic region at this time.

2.0 Ga Tectonothermal Event in Gondwana A ca. 2.0 Ga event involving continental collision, UHT metamorphism and formation of the subduction-related accretionary complex have been proposed from the Gondwana supercontinent. For example, the Capricorn Orogen (2.0-1.6 Ga) in southwestern Australia has been interpreted as continental collision (Myers, 1990; Tyler and Thorne, 1990), and 2.0 Ga UHT metamorphism up to 1050°C caused by crustal thickening has been reported from In Ouzzal, Hogger (Ouzegane and Boumaza, 1996). Similarly, the transamazonic orogenic belt in South America which contains oceanic volcanism (back arc extension), calc-alkalic volcanic arcs and a batholithic granite suite of 2.1-2.0 Ga age (Santons et al., 2000), is considered to represent a subduction-related tectonic setting. Their possible extension in West Africa is the Birmian orogenic belt, which consists of a greenstone belt, containing pillow basalt, volcanic turbidite and subduction-related andesite and dacite (Sylvester and Attoh, 1992). As described above, the geological processes of subduction, collision, and UHT metamorphism, which are characteristic of the North Atlantic region, can be also recognized in several regions of Gondwana. However, the huge igneous Bushveld Complex in South Africa intruded at ca. 2.0 Ga Won Gruenewaldt and Harmer, 1992) is indicative of a possible extensional setting and should be noted. Further studies on 2.0 Ga geological processes in Gondwana would contribute to the reconstruction of a pre-existing supercontinent in Palaeoproterozoic times (Hoffman, 1990).

Acknowledgments This paper is partly based on my doctoral and master theses completed under supervision by M. Yoshida at Osaka City University and M. Komatsu at Ehime University. B.F. Windley, I. Buick, T. Itaya, Y. Osanai and Y. Hiroi are thanked for their constructive comments and improvement of an early version of the manuscript. Parts of this study are funded by “Fukada Geological Institute” and “Japan Society for Promotion of Sciences”.

References Alexejev, N.L. (1997) Reaction textures in intrusive and metamorphic rocks as indicator of P-T path of metamorphic processes (an example from the Kandalaksha-Kolvitsa Belt, Baltic Shield). Ph.D. dissertation extended abstract. St. Petersburg, p. 26.

Baba, S. (1997) Geology and geochemical characteristics of the Leverburgh Belt in South Harris, Outer Hebrides, northwest Scotland. J. Geosci., Osaka City Univ., v. 40, pp. 119-143. Baba, S. (1998) Proterozoic anticlockwise P-T path of the Lewisian Complex of South Harris, Outer Hebrides, NW Scotland. J. Metam. Geol., v. 16. pp. 819-841. Baba, S. (1999) Sapphirine-bearing orthopyroxene-kyanite/ sillimanite granulites from South Harris, NW Scotland: evidence for Proterozoic UHT metamorphism in the Lewisian. Contrib. Mineral. Petrol., v. 136, pp. 33-47. Baba, S. (2002) Two stages of sapphirine formation during prograde and retrograde metamorphism in the Palaeopmterozoic Lewisian Complex in South Harris, NW Scotland. J. Petrol., (submitted). Baba, S., Grew, E.S., Shearer, C.K. and Sheraton, J.W. (2000) Surinamite: a high-temperature metamorphic beryllosilicate from Lewisian sapphirine-bearing kyanite-orthopyroxenequartz-potassium feldspar at South Harris, N.W. Scotland. h e r . Mineral., v. 85, pp. 1474-1484. Balagansky V.V., Timmerman, M.J., Kozlova, N. Ye. and Kislitsyn, R.V. (2001) A 2.44 Ga syn-tectonic mafic dyke swarm in the Kolvitsa Belt, Kola Peninsula, Russia: implications for early Palaeoproterozoictectonics in the north-eastem Fennoscandian Shield. Precamb. Res., v. 105, pp. 269-287. Barbey, P., Convert, J., Martin, H., Moreau, B., Capdevila, R. and Hameurt, J. (1980) Relationships between granite gneiss terrains, greenstone belts and granulite belts in the Archaean crust of Lapland (Fennoscandia). Geol. Rundschau, V.69, pp. 648-658. Berman, R.G. and Bostock, H.H. (1997) Metamorphism in the northern Taltson Magmatic Zone, northwest territories. Can. Mineral., v. 35. pp. 1069-1091. Bernard-Griffiths, J., Peucat, J.J., Postaire, B., Vidal, Ph., Convert, J. and Moreau, B. (1984) Isotope data (U-Pb, Rb-Sr, Pb-Pb and Sm-Nd) on mafic granulites from Finnish Lapland. Precamb. Res., v. 23, pp. 325-348. Berthelsen, A. and Marker, M. (1986) Tectonics of the Kola collision suture and adjacent Archaean and early Proterozoic terrains in the northeastern region of the Baltic Shield. Tectonophys., v. 126, pp. 31-55. Bohlen, S.R. (1987) Pressure-temperature-time paths and a tectonic model for the evolution of granulites. J. Geol., V. 95, pp. 617-632 Bohlen, S.R. (1991) On the formation of granulites. J. Metam. Geol., v. 9, pp. 223-229. Bylinski, R., Glebovitski,V., Dolivo-Dobrovolski,A. and Porotova, G. (1977) The major Belomorian deep-fault zone. In: Kortman, C. (Ed.), Fault tectonics in the eastern part of the Baltic Shield. Proc. Finnish-Soviet Symposium, Helsinki. pp. 49-62. Buchan, K.L., Mertanen, S., Park, R.G., Pesonen, L.J., Elming, S.-A., Abrahamsen, N. and Bylund, G. (2000) Comparing the drift of Laurentia and Baltica in the Proterozoic: the importance of key palaeomagnetic poles. Tectonophys., V. 319, pp. 167-198. Cliff, R.A., Gray, C.M. and Huhma, H. (1983) A Sm-Nd isotope study of the South Harris Igneous Complex, the Outer Hebrides. Contrib. Mineral. Petrol., v. 82, pp. 91-98. Cliff, R.A., Rex, D.C. and Guise, P.G. (1998) Geochronological Gondwana Research, V. 5, No. 4,2002

PALAEOPROTEROZOIC EVENTS IN SOUTH HARRIS, NW SCOTLAND

studies of Proterozoic crustal evolution in northern Outer Hebrides. Precamb. Res., v. 91, pp. 401-418. Cohen, A.S., ONions, R.K. and O’Hara, M.J. (1991) Chronology and mechanism of depletion in Lewisian granulites. Contrib. Mineral. Petrol., v. 106, pp. 142-153. Condie, K.C. (1989) Plate tectonics and crustal evolution. Pergamon, Oxford, 476p. Connelly, J.N. (2001) Constraining the timing of metamorphism: U-Pb and Sm-Nd ages from a transect across the Northern Torngat Orogen, Labrador, Canada: J. Geol., v. 109, pp. 57-77. Corfu, E, Heaman, L.M. and Rogers, G. (1994) Polymetamorphic evolution of the Lewisian Complex, NW Scotland, as recorded by U-Pb isotopic compositions of zircon, titanite and rutile. Contrib. Mineral. Petrol., v. 117, pp. 215-228. Daly, J.S., Balagansky,V.V., Timmerman, M.J., Whitehouse, M.J., de Jong, K., Guise, P., Bogdanova, S., Gorbatschev, R. and Bridgwater, D. (2001) Ion microprobe U-Pb zircon geochronology and isotopic evidence for a trans-crustal suture in the Lapland-Kola Orogen, northern Fennoscandian Shield. Precamb. Res., v. 105, pp. 289-314. Dearnley, R. (1962) An outline of the Lewisian Complex of the Outer Hebrides in relation to that of the Scottish Mainland. Q. J. Geol. SOC.London, v. 118, pp. 143-176. Droop, G.T.R., Fernandes, L.A.D. and Shaw, S. (1999) Laxfordian metamorphic conditions of the Palaeoproterozoic Loch Maree Group, Lewisian Complex, NW Scotland. Scott. J. Geol., V. 35, pp. 31-50. Eckert, J.O., Newton, R.C. and Kleppa, O.J. (1991) The AH of reaction and recalibration of garnet-plagioclase-quartz geobarometers in the CMAS system by solution chemistry. Amer. Mineral., v. 76, pp. 148-160. Ellis, D.J. (1987) Origin and evolution of granulites in normal and thickened crust. Geology, v. 15, pp. 167-170. Ellis, D.J. and Green, D.H. (1979) An experimental study of the effect of Ca upon garnet-clinopyroxene Fe-Mg exchange equilibria. Contrib. Mineral. Petrol., v. 71, pp. 13-22. Fettes, D.J., Mendum, J.R., Smith, D.I. and Watson, J.V. (1992) Geology of Outer Hebrides. Sheets (solid edition) Lewis and Harris, Uist and Barra (Scotland). British Geol. Surv. Mem., pp. 1-175. Friend, C.R.L. and Kinny, P.D. (1995) New evidence for protolith ages of Lewisian granulites, northwest Scotland. Geology, V. 11, pp. 1027-1030. Frisch, T., Jackson, G.D., Glebovitsky, V.A., Yefimov, M.M., Bogdanova and Parrish, R.R. (1995) U-Pb ages of zircon from the Kolvitsa gabbro-anorthosite complex, southern Kola Peninsula, Russia. Petrology, v. 3, pp. 219-225. Glebovitsky,V., Marker, M., Alexejev, N., Bridgwater, D., Sedova, I., Salnikova, E. and Berezhnaya, N. (2001) Age, evolution and regional setting of the Palaeoproterozoic Umba igneous suite in the Kolvitsa-Umba zone, Kola Peninsula: constraints from new geological, geochemical and U-Pb zircon data. Precamb. Res., v. 105, pp. 247-267. Grover, T.W., Pattison, D.M.R., McDonough, M.R. and McNicoll, V.J. (1997) Tectonometamorphic evolution of the southern Taltson Magmatic Zone and associated shear zones, northeastern Alberta. Can. Mineral., v. 35, pp. 1051-1067. Harley, S.L. (1998) On the occurrence and characterisation of ultrahigh-temperature crustal metamorphism. In: Treloar, PJ. and O’Brien, P. (Eds.), What drives metamorphism and

Gondwana Research, V. 5, No. 4,2002

769

metamorphic reactions? Geol. SOC.London, Spl. Publ. No. 138, pp. 75-101 Harley, S.L. (1989) The origins of granulites: a metamorphic perspective. Geol. Mag., v. 126, pp. 215-247. Helmstaedt, H.H. and Scott, D.J. (1992) The Proterozoic ophiolite problem. In: Condie, K.C. (Ed.), Proterozoic crustal evolution. Elsevier, Amsterdam, pp. 55-95. Hensen, B.J. and Harley, S.L. (1990) Graphical analysis of P-T-X relations in granulite-facies metapelite. In: Ashworth, J.R. and Brown, M. (Eds.), High temperature metamorphism and crustal anatexis. Unwin Hyman, London, pp. 19-56. Hensen, B.T. and Kalsbeek, F. (1989) Precise age for the Ammassalik Intrusive Complex. Rapp. Grplnl. Geol. Unders., v. 146, pp. 46-47. Hoffman, P. E (1990) Dynamics of the tectonic assembly of northeast Laurentia in geon 18 (1.9-1.8 Ga). Geosci. Canada, V. 17, pp. 222-226. Holdaway, M.J. and Biswajit Mukhopadhyay (1993) A reevaluation of the stability relations of andalusite: thermochemical data and phase diagram for the aluminum silicates. Amer. Mineral., v. 78, pp. 298-315. Ishizuka, H. (1981) Geochemistry of the Horokanai ophiolite in the Kamuikotan tectonic belt, Hokkaido, Japan. J. Geol. SOC. Japan, v. 87. pp. 17-34. Jackson, G. and Berman, R.G. (2000) Precambrian metamorphic and tectonic evolution of northern Baffin Island, Nunavut, Canada. Can. Mineral., v. 38, pp. 399-421. Kalsbeek, F., Austrheim, H., Bridgwater, D., Hansen, B.T., Pedersen, S. and Taylor, P.N. (1993) Geochronology of the Ammassalik area, South-East Greenland, and comparison with the Lewisian of Scotland and the Nagssugtoqidian of West Greenland. Precamb. Res., v. 62, pp. 239-270. Kimura, G., Ludden, J.N., Desrochers, J.P. and Hori, R. (1993) A model of ocean-crust accretion for the Superior province, Canada. Lithos, v. 30, pp. 337-355. Kinny, P.D. and Friend, C.R.L. (1997) U-Pb isotope evidence for the accretion of different crustal blocks to form the Lewisian Complex of northwest Scotland. Contrib. Mineral. Petrol., V. 129, pp. 326-340. Kitsul, V.I., Glebovitsky, V.A., Yapnik, Y.A. and Frisch, T. (2000) Gneisses from the granulite terrane of the Central Boothia Uplift, Arctic Canada. Can. Mineral., v. 38, pp. 443-454. Korstgird, J., Ryan, B. and Wardle, R.J. (1987) The boundary between Proterozoic and Archaean crustal blocks in Central West Greenland and northern Labrador. In: Park, R.G. and Tarney, J. (Eds.), Evolution of the Lewisian and Comparable Precambrian high grade terrains. Geol. SOC.London, Spl. Publ. NO. 27, pp. 247-259. Marker, M. (1985) Early Proterozoic (ca. 2000-1900 Ma) crustal structure of the northeastern Baltic Shield: tectonic division and tectogenesis. Nor. Geol. Unders. Bull., v. 403, pp. 5.547. Mengel, E and Rivers, T. (1991) Decompression reactions and P-T conditions in high-grade rocks, northern Labrador: P-T-t paths from individual samples and implications for early Proterozoic tectonic evolution. J. Petrol., v. 32, pp. 139-167. Moorbath, S., Powell, J.L. and Taylor, P.N. (1975) Isotope evidence for the age and origin of the “grey gneiss” complex of southern Outer Hebrides, Scotland. J. Geol. SOC.London, V. 131, pp. 213-222.

770

S. BABA

Myers, J.S. (1987) The East Greenland Nagssugtoqidian mobile belt compared with the Lewisian Complex. In: Park, R.G. and Tarney, J. (Eds.), Evolution of the Lewisian and comparable Precambrian high grade terrains. Geol. SOC. London, Spl. Publ. No. 27, pp. 235-246. Myers, J.S. (1990) Precambrian tectonic evolution of part of Gondwana, southwestern Australia. Geology, v. 18, pp. 537-540. Newton, R.C. and Haselton, H.T. (1981) Thermodynamics of the garnet-plagioclase-All,SiO,-quartz geobarometer. In: Newton, R.C., Navrotsky, A. and Wood, B.J. (Eds.), Thermodynamics of minerals and melt. Springer-Verlag, Heidelberg, pp. 129-145. Often, M. (1985) The early Proterozoic Karasjok greenstone belt, Norway; a preliminary description of lithology, stratigraphy and mineralization. Nor. Geol. Unders. Bull., v. 403, pp. 75-88. Ouzegane, K. and Boumaza, S. (1996) An example of ultrahightemperature metamorphism: orthopyroxene-sillimanitegarnet, sapphirine-quartz parageneses in Al-Mg granulite from In Hihaou, In: Ouzzal, Hoggar. J. Metam. Geol., v. 14, pp. 693-708. Park, R.G. (1995) Palaeoproterozoic Laurentia-Baltica relationships: a view from the Lewisian. In: Coward, M.P. and Ries, A.C. (Eds.), Early Precambrian processes. Geol. SOC.London, Spl. Publ. No. 95, pp. 211-224. Park, R.G. and Tarney, J. (1987) The Lewisian Complex: a typical Precambrian high-grade terrain? In: Park, R.G. and Tarney, J. (Eds.), Evolution of the Lewisian and comparable Precambrian high grade terrains. Geol. SOC.London, Spl. Publ. NO. 27, pp. 13-25. Park, R.G., Tarney, J. and Connelly, J.N. (2001) The Loch Maree Group: Palaeoproterozoic subduction-accretion complex in the Lewisian of NW Scotland. Precamb. Res., v. 105, pp. 205-226. Park, R.G., Cliff, R.A., Fettes, D.J. and Stewart, A.D. (1994) Precambrian rocks in northwest Scotland west of the Moine Thrust: the Lewisian Complex and the Torridonian. In: Gibbsons, W. and Harris, A.L.A. (Eds.), Revised correlation of Precambrian rocks in the British Isles. Geol. SOC.London, Sp. Rep. No. 22, pp. 6-22. Pharaoh, T.C. and Pearce, J.A. (1984) Geochemical evidence for the geotectonic setting of early Proterozoic metavolcanic sequences in Lapland. Precamb. Res., v. 25, pp. 283-308. Pidgeon, R.T. and Aftalion, M. (1972) The geochronological significance of discordant U-Pb ages of ovalshaped zircons from a Lewisian gneiss from Harris, Outer Hebrides. Earth Planet. Sci. Lett., v. 17, pp. 269-274. Sandiford, M., Neall, F.B. and Powell, R. (1987) Metamorphic evolution of aluminous granulites from Labwor Hills, Uganda. Contrib. Mineral. Petrol., v. 95, pp. 217-225. Santons, J.O.S., Hartmann, L.A., Gaudette, H.E., Groves, D.I., Mcnaughton, N.J. and Fletcher, I.R. (2000) A new understanding of the provinces of the Amazon Craton based on integration of field mapping and U-Pb and Sm-Nd geochronology. Gondwana Res., v. 3, pp. 453-488.

Scott, D.J. (1998) An overview of the U-Pb geochronology of the Palaeoproterozoic Trongat Orogen, northeastern Canada. Precamb. Res., v. 91, pp. 91-107 Scott, D.J. and Machado, N. (1995) U-Pb geochronology of the northern Torngat Orogen, Labrador, Canada: a record of Palaeoproterozoic magmatism and deformation. Precamb. Res., v. 70, pp. 169-190. Scott, D.J., Helmstaedt, H. and Bickle, M.J. (1992) Purtuniq ophiolite, Cape Smith Belt, northern Quebec, Canada: a reconstructed section of early Proterozoic oceanic crust. Geology, v. 20, pp. 173-176. St-Onge, M.R., Wodicka, N. and Lucas, S.B. (2000) Granuliteand amphibolite-facies metamorphism in a convergent-platemargin setting: synthesis of the Quebec-Baffin segment of the Trans-Hudson Orogen. Can. Mineral., v. 38, pp. 379-398. Storey, M., Mahoney, J.J., Kroenke, L.W. and Saunders, A.A. (1991) Are oceanic plateaus sites of komatiite formation? Geology, v. 19, pp. 376-379. Sun, S.S. and McDonough, W.F. (1989) Chemical and isotopic systematics of oceanic basalts: implication for mantle composition and process. In: Saunders, A.D. and Norry, M.J. (Eds.), Magmatism in the ocean basins. Geol. SOC.London, Spl. Publ. NO. 42, pp. 313-345. Sylvester, P.J. and Attoh, K. (1992) Lithostratigraphy and composition of 2.1 Ga greenstone belts of the West African craton and their bearing on crustal evolution and ArcheanProterozoic boundary. J. Geol., v. 100, 377-393. Tyler, I.M. and Thorne, A.M. (1990) The northern margin of the Capricorn orogen, western Australia - an example of an early Proterozoic collision zone. J. Str. Geol., v. 12. Vielzeuf, D. and Montel, J.M. (1994) Partial melting of metagraywackes. Part 1.Fluid-absent experiments and phase relationships. Contrib. Mineral. Petrol., v. 117, pp. 375-393. Von Gruenewaldt, G. and Harmer, R.E. (1992) Tectonic setting of Proterozoic layered intrusions with special reference to the Bushveld Complex. In: Condie, K.C. (Ed.), Proterozoic crustal evolution. Elsevier, Amsterdam, pp. 181-213. Whitehouse, M.J. (1989) Sm-Nd evidence for dischronous crustal accretion in the Lewisian Complex, northwest Scotland. Tectonophys., v. 161, pp. 245-256. Whitehouse, M.J. (1990) Isotopic evolution of the south Outer Hebridean Lewisian gneiss complex: constraints on late Archaean source regions and the generation of transposed Pb-Pb palaeoisochrons. Chem. Geol., v. 86, pp. 1-20. Whitehouse, M.J. and Bridgewater, D. (2001) Geochronological constrains on Palaeoproterozoic crustal evolution and regional correlations of the northern Outer Hebridean Lewisian Complex, Scotland. Precamb. Res., v. 105, pp. 227-245. Windley, B.F. (1993) Uniformitarianism today: plate tectonics is the key to the past. J. Geol. SOC.London, v. 150, pp. 7-19. Windley, B.E (1995) The evolving continents. 3rd Ed. Wiley, London. 526p. Zhu, X.K., O’Nions, R.K., Belshaw, N.S. and Gibb, A.J. (1997) Lewisian crustal history from in situ SIMS mineral chronometry and related metamorphic textures. Chem. Geol., V. 136, pp. 205-218.

Gondwana Research, V: 5, No. 4,2002