Antarctica in the Late Precambrian Early Palaeozoic c 600450 Ma Evolution of Gondwana

The connections between Antarctica and the other southern hemisphere continents at the time of Gondwana are clearly documented by the so-called ''Gondwanian'' sequences which correlate with analogous stratigraphic successions in South America, Africa, India, Sri Lanka, Australia and New Zealand and by palaeontological evidence, such as the significant first appearance in Triassic time of the herbivorous reptile Lystrosaurus found in all Gondwanian continents, Antarctica included (Collinson, 1991; Retallack et al., 2005). Additional geological evidence for Gondwana's reconstruction is provided by a number of older Antarctic geological provinces (such as the Archean Cratons and the Palaeozoic orogenic belts, e.g. the Ross-Delamerian Orogen) which fit tightly across a closed Southern Ocean.

In comparison to this well-established evidence, the reconstruction of tectonic models for the initial stages of the formation of Gondwana is more problematic. The amalgamation phase necessarily involved the aggregation of various continental fragments which derived from the fragmentation and dispersal of a former supercontinent - variously named Ur-Gondwana (Hartnady, 1991), Katania (Young, 1995), Palaeopangea (Piper, 2000) or Rodinia (McMenamin and McMenamin, 1990). However, the precise configuration and modality of break-up of this supercontinent are yet not completely known and these uncertainties obviously propagate to the formulation of tectonic models for the constructive phase of Gondwana.

Since the focus of this paper is on the Cretaceous-Cenozoic record, we will here avoid reviewing Rodinia models (e.g. SWEAT: Dalziel, 1991; Hoffman, 1991; Moores, 1991; AUSWUS: Karlstrom et al., 1999; see also reviews by Dalziel, 1997 and Meert and Torsvik, 2003) or alternative models (e.g. Palaeopangea, Piper, 1982, 2000) and we will briefly summarize the main recent results in the reconstruction of the Late Precambrian-Early Palaeozoic tectonic evolution of Antarctica in Gondwana. From Rodinia to Gondwana

In spite of significant uncertainties about the precise reconstruction of the global palaeogeography in Neoproterozoic time and the still incomplete geochronological framework, most authors agree that the break-up of Rodinia led to the development of extensive passive continental margins which are documented in the late Neoproterozic (c. 750-600 Ma) record of most present-day continents (Dalziel, 1991, 1992, 1997; Powell et al., 1993; Meert and Van der Voo, 1997; Cawood, 2005).

In Antarctica, following Dalziel's hypothesis (1991), the first stage of Rodinia break-up involved a rifting phase which started at c. 750-725 Ma leading to the separation of Laurentian from the East Antarctica+Australia block and the formation of the intervening proto-Pacific ocean. This process was accompanied by the drift of the cratonic blocks, presently exposed in South America and Africa (Amazonian, Rio de la Plata and Western African) (i.e. West Gondwana) which eventually collided with India, Sri Lanka and East Antarctica (East Gondwana).

Most workers agree that key evidence of this evolution is stored in the Mozambique Belt (Holmes, 1951) or East African Orogen (Stern, 1994) and that this extensive orogenic belt formed as a result of the closure of a ''Mozambique Ocean'' and subsequent collision and amalgamation of East and West Gondwana during the Pan-African event (Dalziel, 1992; Stern, 1994; Shackleton, 1996). A review of structural and geochronological data from East Antarctica (Dronning Maud Land and Lutzow Holm Bay) and comparison with the adjacent (in Gondwana) Falkland Microplate and south-eastern Africa led Jacobs and Thomas (2002) to corroborate the proposal by Jacobs et al. (1998) of a southward continuation of the Mozambique belt into Dronning Maud Land in Antarctica.

A Mozambique suture zone in the Shackleton Range was suggested by Grunow et al. (1996) and conclusive evidence was found by Talarico et al. (1999) who described relics of ophiolites consisting of serpentinites and amphibolites with N-type MORB to OIB geochemistry and a maximum

Sm-Nd age of c. 900 Ma. Metamorphic reworking of these rocks occurred under variable high P to medium P conditions in the eastern Shackleton Range; a stage of eclogite-facies metamorphism in the area has been recently reported by Schmadicke and Will (2006) in the central Shackleton Range. On the basis of these discoveries and thrust patterns in both the Shackleton Range and Western Dronning Maud Land, Kleinschmidt et al. (2002) proposed that the ophiolites may have formed part of a connection between the Palaeo-Pacific and the Mozambique oceans, in the way the Drake Passage links the present Pacific and Atlantic Oceans.

The Mozambique Belt and its continuation in Antarctica shows a general N-S trending (Jacobs and Thomas, 2002). East of Dronning Maud Land, structural and petrological data indicate that the Lutzow Holm Complex (Kriegsman, 1995) can be interpreted as representing the western end of a c. E-W, subperpendicular branch of the Mozambique Belt that might have extended to Prydz Bay, where Pan-African high-grade rocks have also been reported (Dirks and Wilson, 1995; Fitzsimons, 1997).

The Lutzow Holm-Prydz Bay Pan-African orogenic belt has provided sound evidence in contrast to the classical assumption that Eastern Gondwana formed during the consolidation of Rodinia in the Mesoproter-ozoic time and it remained tectonically stable until the modern continents rifted from Gondwana in the Mesozoic. The Lutzow Holm-Prydz Bay-Pan-African structures were initially considered part of a wider belt termed the Kuunga Orogen by Meert et al. (1995) or Kuunga Suture by Boger and Miller (2004), and interpreted as the result of the collision between East Antarctica (+Australia) and India (+Madagascar+Sri Lanka) at c. 535-520 Ma, after the amalgamation of India with the rest of Gondwana along the Mozambique suture (Fig. 7.4).

More recent data actually suggest that there are three orogenic belts formed at about the same Pan-African period of 500-600 Ma in East Antarctica (Fig. 7.2): (1) the belt in the Shackleton Range-Dronning Maud Land-southern S0r Rondane-Lutzow-Holmbukta region (''East Antarctic Orogen'' or ''East Antarctic Belt'' of Jacobs et al., 1998, renamed ''Lutzow Holm Belt'' by Fitzsimons, 2000b); (2) the belt in the southern Prince Charles Mountains-Grove Mountains (''Kuunga Suture'' according to Boger et al. 2002); and (3) the belt in the Denman Glacier region, interpreted as prolongation of the Leeuwin Complex of Australia's Pinjarra Orogen (e.g. Fitzsimons, 2000b). However, the exact extent of these orogens is doubtful because of the extensive ice cover, and this applies especially to the Denman Glacier belt.

The Lutzow Holm Belt is characterized by extensive thrust and nappe tectonics (Shackleton Range), by widespread and distinct late-orogenic collapse

Figure 7.4: Main tectonic stages of the amalgamation of Gondwana (modified after Boger and Millar, 2004, with permission from Elsevier). Abbreviated cratons - D: Dwarhai craton; G: Gawler craton; K: Kalahary craton; P/Y: Pilbara/Yilgarn craton; sPCM: southern Prince Charles


Figure 7.4: Main tectonic stages of the amalgamation of Gondwana (modified after Boger and Millar, 2004, with permission from Elsevier). Abbreviated cratons - D: Dwarhai craton; G: Gawler craton; K: Kalahary craton; P/Y: Pilbara/Yilgarn craton; sPCM: southern Prince Charles


structures (Shackleton Range), by thick molasse formations (Blaiklock Glacier Group: Shackleton Range; Urfjell Group: Dronning Maud Land), and by syn- and post-orogenic magmatism (Dronning Maud Land; Paech, 2004). Therefore, these Antarctic portions, formed during late Neoproterozoic to Cambrian, show typical characteristics of a collisional orogen.

The three belts indicate the amalgamation of West Gondwana (South America, Africa and Grunehogna Craton) and East Gondwana (India, Australia and main East Antarctica). As proposed by Boger et al. (2001. 2002), Gondwana's amalgamation may have taken place in two steps: the first before 550 Ma and the second after 550 Ma. The first step involved the amalgamation of West Gondwana with ''Indo-Antarctica'' (i.e. India and the northern Prince Charles Mountains+Napier Complex) documented by the Mozambique Belt and as its Antarctic prolongation, the Lutzow Holm Belt. The second step led to the aggregation of these terranes to the rest of East Gondwana, i.e. the rest of East Antarctica and Australia and thus producing the Kuunga Suture. This model could explain the diacronous development of the Lutzow Holm Belt, mainly older than 550 Ma in Dronning Maud Land and < 550 Ma in the Shackleton Range, as well as the existence of an interleaved old alien element - the Grenvillian granophyres of Coats Land - maybe just an exotic terrane, or a mini-craton (Kleinschmidt, 2007). The Antarctic record of the late precambrian-early palaeozoic evolution of the palaeo-pacific margin of Gondwana

The Transantarctic Mountains, at the margin of the East Antarctic shield, represent a key element in providing an important but still cryptic record of Proterozoic and Early Palaeozoic supercontinent history in the period from c. 800 to 500 Ma. The existing dataset and data that can be provided by future research are from sedimentary, plutonic and volcanic assemblages that potentially reflect different tectonic events integral to the Rodinian-Gondwanan transformation: from the break-up of Rodinia to the development of a transitional margin and plate-margin activity during Gondwanan assembly.

In the Transantarctic Mountains, constraints on the timing of Rodinia break-up are provided by mafic magmatism and siliciclastic sedimentation of Neoproterozoic age from two areas. In the Skelton Glacier area, basalts interlayered with Skelton Group sediments yielded a Sm-Nd model age of 700-800 Ma (Rowell et al., 1993). In the Nimrod Glacier area (Cotton Plateau), gabbros and basalts interlayered with sediments of the Beardmore

Group previously dated at 762 Ma (Sm-Nd isochron age, Borg et al., 1990) are now considered to have been emplaced at 668 Ma (U-Pb zircon age, Goodge et al., 2002). Thick siliciclastic successions were long interpreted as deep-water turbidities deposited in Proterozoic time along a rifted margin. Although the older parts of some successions may relate depositionally to the rifting process, recent investigations have demonstrated that some units are nearshore deposits and the depositional age of several major successions must be revised upward. For example, in the central Transantarctic Mountains, sandstones formerly included in the Beardmore Group and considered Neoproterozoic are now assigned to the Middle Cambrian or younger (Goodge et al., 2002).

A transformation from drifting to active subducting mode is inferred for the late Neoproterozoic, starting at c. 700 Ma and continuing, most probably as a protracted contractional tectonic cycle including several discrete deformation events, with a tectonic climax during the Ross Orogeny, until the Ordovician (Goodge, 2001). The first stage of the conversion from a passive margin to an active convergent margin is signalled by the structural inversion of craton-margin sedimentary successions in central Transantarctic Mountains.

Younger tectonic events between 560 and 480 Ma are generally referred to a broadly defined Ross Orogeny, a long-lived tectonic process which developed along the active Gondwanan margin involving episodic deformation, calc-alkaline magmatism and syn-orogenic deposition of arc-derived detritus in a sinistral-transpressive, continental-margin arc setting (Fig. 7.2). The orogenic belt is exposed from northern Victoria Land at the Pacific end up to the Pensacola Mountains at the Atlantic end. Westernmost Marie Byrd Land (Edward VII Peninsula) has to be considered as part of the same orogenic belt, from which it became isolated only much later, during a major phase of the evolution of the WARS around the end of the Cretaceous.

The Ross Orogen is characterized by folds, thrusts, very low- to high-grade metamorphism, granitoids, terranes and flysch- and molasse-type sediments. Remarkable thrusts have been reported from Oates Land, which could be traced into the Australian continuation of the Ross Orogen, the Delamerian Orogen (Flottmann et al., 1993). The systematic distribution of high- and low-pressure types of metamorphism (e.g. Talarico et al., 2004) and of S- and I-type granitoids (e.g. Vetter and Tessensohn, 1987) led to the model of subduction of the palaeo-Pacific beneath East Antarctica. The igneous activity occurred as early as 560-550 Ma ago and is considered to be related to initial subduction of palaeo-Pacific lithosphere beneath the East Antarctic cratonal margin. In southern Victoria Land, the older plutons also include a peculiar suite of highly alkaline rocks (nepheline syenites and carbonatites)

(Koettlitz Glacier Alkaline Province, Cooper et al., 1997) and a stage of Precambrian rift-related magmatism and sedimentation has been documented by Cook (2007) in the Skelton Glacier area.

In northern Victoria Land, the Ross Orogen is made up of three so-called terranes: the high- to medium-grade and granite-dominated Wilson Terrane to the west, the low-grade turbiditic Robertson Bay Terrane to the east, and the low-grade and volcanic-rich Bowers Terrane in between. These terranes are considered to be allochthonous or just adjacent palaeogeo-graphic domains including an intra-oceanic island arc (Bowers Terrane) and an accretionary wedge (Robertson Bay Terrane) (Tessensohn and Henjes-Kunst, 2005).

In northern Victoria Land, a unique occurrence of ultra-high-pressure rocks, including well-preserved mafic eclogites as lenses and pods within metasedimentary gneisses and quartzites, decorates the tectonic boundary between the inboard Wilson Terrane and the Bower Terrane in the Lanterman Range. Geological, petrological and geochronological studies indicate that mafic, ultra-mafic and felsic host rocks in this region underwent a common metamorphic evolution with an eclogite facies stage about 500 Ma ago at temperatures of up to about 850°C and pressures greater than 2.6 GPa (Di Vincenzo et al., 1997; Palmeri et al., 2003, 2007).

0 0

Post a comment