Palaeomagnetic Restraints on the Proterozoic Tectonic Evolution of Australia

Australia: The Land Where Time Began

Palaeomagnetic Restraints on the Proterozoic Tectonic Evolution of Australia

According to the authors¹ assembly of Proterozoic Australia by the processes of tectonics involving horizontal movements on a large scale have been advocated by recent models of plate tectonics, though earlier models suggested that the evolution of Australia was as an essentially intact lithospheric block. They suggest the testing of whether or not the major cratonic blocks comprising Australia were together or separated widely during the Proterozoic cannot be determined by geological and geochemical observations alone, palaeomagnetic studies are required to provide the quantitative constraints on the movement of the plates relative to each other in the Precambrian. Groups of palaeopoles that overlap for 1.7-1.8 Ga and 1.5-1.6 Ga allow the assemblages of North and West Australian Cratonic Assemblages to have occupied their relative positions since at least 1.7 Ga, joining the South Australian Cratonic Assemblage since at least 1.5 Ga, in spite of the deficiencies of the palaeomagnetic record of Proterozoic Australia. To test whether there were large oceans that closed between any of the continental blocks still requires additional, geochronological and palaeomagnetic data.

The authors¹ suggest it is generally accepted that in the Proterozoic plate tectonic processes are similar to those of the present time.

Tectonic Summary

The assembly of the Australian continent has been portrayed by recent geological syntheses as being a series of events that were complex and protracted (e.g. Myers et al., 1996; Tyler et al., 1998). Australian rocks of Precambrian age comprise 3 main regions, the North Australian Cratonic Assemblage, West Australian Cratonic Assemblage and the South Australian Cratonic Assemblage, the blocks comprising these assemblages being of Archaean and /or Palaeoproterozoic crust, bounded by mobile belts of Palaeoproterozoic-Mesoproterozoic age, that in part lie beneath younger sedimentary basins. An interpretation of these 3 regions (Myers et al., 1996) suggested they had formed independently in the Palaeoproterozoic, their final amalgamation taking place along the orogenic belts of Mesoproterozoic age, 1.3-1.0 Ga. Cohesion of these regions since at least the Late Palaeoproterozoic has received support from several geological and palaeomagnetic observations. The simplified overview that follows, of the tectonic evolution in the Proterozoic of each cratonic assemblage focuses mainly on observations and interpretations that are suggestive of tectonic environments.

West Australian Cratonic Assemblage (WAC)

North Australian Cratonic Assemblage (NAC)

South Australian Cratonic Assemblage (SAC)

Mesoproterozoic amalgamation (?)

A record of 2 major tectonothermal events are preserved in the Albany-Fraser Mobile Belt and the Musgrave Mobile Belt - Collision at about 1.3 Ga, then at 1.2 Ga intracratonic reactivation of sutures (Clarke et al., 1995; Nelson et al., 1995; White et al., 1999; Clark et al., 2000). The authors¹ suggest that these observations imply that the Albany-Fraser Orogen and the Musgrave Orogen are continuous, as was suggested previously (Myers et al, 1996). It is also suggested that the NAC and WAC could have been combined before the collision between the the South Australian-Antarctic Craton, and along the eastern edge of the combined Yilgarn-Pilbara Block the Miles Orogeny was intracratonic.

It is suggested by limited data, that is equivocal, from the Albany-Fraser Belt and the Musgrave Belt that felsic orthogeneses of about 1.3 Ga are related to collision, but the granitoids from 1.2 Ga have a more intraplate signature (Nelson et al., 1995; Sheraton & Sun, 1995). In the southeast part of the Yilgarn Craton the absence of plutons from the Late Mesoproterozoic led to the suggestion that the initial collision at about 1.3 Ga involved oceanic crust subduction beneath the South Australian-Antarctic Craton (Clark et al., 2000). The Fraser complex, that is situated within the eastern Albany-Fraser Orogen is comprised of imbricated slices of a mafic intrusion (Myers, 1985) that was emplaced, metamorphosed to granulite grade, following which it was uplifted then thrust onto the southeast margin of the Yilgarn Craton at 1.3 Ga within a period of about 30 My (Fletcher et al., 2000). The interpretation of geochemical data to indicate the Fraser Complex represents the remains of oceanic arcs that accreted before collision (Condie & Myers, 1999). The presence within the Albany-Fraser Belt and the Musgrave Belt of crustal components from the Archaean and/or Palaeoproterozoic (e.g. Nelson et al., 1995; Comacho & Fanning, 1995; Clark et al., 2000) implies that at least in parts these orogens may be floored by continental crust. Emplaced subparallel to the margins of the Yilgarn Craton at about 1.2 Ga were extensive dyke swarms (Evans, 1999; Wingate et al., 2000; Wingate unpublished data).

A U-Pb zircon age of 1.31 Ga (Nelson, 1996) and Rb-Sr ages of 1.3 and 1.1 Ga (Chin & De Laeter, 1981) for foliated granite in the Rudall Block of Palaeoproterozoic age provides evidence for activity between the NAC and WAC in the Late Mesoproterozoic. The Cape River Province and the Anakie Inlier contain inherited zircons with ages from 1.3-1.2 Ga suggesting there was a belt in northeastern Australia that may have extended to the west beneath a younger cover, of Grenville age, and connecting with the Albany-Fraser-Musgrave Orogen (Blewett et al., 1998; Hutton et al., 1998; Ferguson et al., 2001).

In the Musgrave Block deformation of a late phase has been recognised, though not dated, from about 1.06 Ga (Sun et al., 1996; White et al., 1996) in the Albany-Fraser Belt (Clarke et al., 2000) that is synchronous with deformation and magmatism in the Darling Mobile Belt and the Bangemall Basin with ages of 1.09-1.06 Ga (Bruguier et al., 1999; Wingate et al., 2002) It has been suggested that in the Darling Mobile Belt, blocks from the Mesoproterozoic, Northampton Block and Leeuwin Block, were accreted to the margin of Western Australia at some time following their metamorphism and deformation further to the south (present coordinates) and before the Northampton block was 'stitched' to the margin of Western Australia that took place at about 755 Ma by the Mundine Well Dykes (Fitzsimons, 2001) (Fig.4 in Wingate & Giddings, 2000).

Events of the Neoproterozoic

In the Proterozoic, Australia had essentially stabilised by 1.0 Ga, and in the Centralian Superbasin that extended across the continent, sedimentation began between 850-830 Ma (Walter et al., 1995). During initial rifting in central and eastern Australia NE-SW extension is reflected by the Gairdner Dykes at about 825 Ma (Wingate et al., 1998). The emplacement of the Mundine Wells Dykes parallel to the continental margin have been suggested to have been emplaced before the separation of an unknown continental fragment, suggested by the authors¹ to possibly be Kalahari, from the margin of Western Australia (Wingate & Giddings, 2000; Powell & Pisarevsky, 2002). In the latest Neoproterozoic there were several intracratonic events, though Australia remained essentially intact during the events of the breakup along its eastern margin some time after 780 Ma (Powel et al., 1994; Preiss, 2000). The Petermann Ranges Orogeny, about 550 Ma, involved reactivation of the Miles-Musgrave Orogeny, thrusting being both to the north and the south, and the Musgrave Block underwent considerable exhumation (Maboko et al., 1992; Preiss and Krieg, 1992; Camacho & Fanning, 1995; Scrimgeour & Close, 1999), and at the eastern edge of the Pilbara Craton and in the King Leopold Orogen there was thrusting to the southwest (Tyler et al., 1998). An anticlockwise rotation of the NAC, with respect to the remainder of the continent is suggested to have taken place at this time (Powell et al, 1994).

Palaeomagnetism

The relative positions of continental fragments in the Precambrian can only be determined quantitatively by the use of palaeomagnetism. The latitude and orientation of a sampling locality relative to the palaeomagnetic pole at the time the rocks was magnetised can be indicated by palaeomagnetic directions. During the palaeoproterozoic there were many changes of latitude and orientation of the Australian continent relative to the pole between 1.8 and 0.7 Ga, according to the positions of palaeopoles relative to Australia. Joining palaeopoles, backwards through time, by the shortest possible segments, is used to construct the apparent pole wandering path (APWP). The use of this simplest approach results in a solution that is not unique. Ambiguity in choosing a pole versus its antipole results from the lack of a continuous record of field reversals back to the Proterozoic, and it is inevitable there will be some features of the path that are overlooked because of large age gaps between adjacent poles. The lack of palaeopoles that are well dated is another uncertainty that is critically important. The dating of many are imprecise, and/or their magnetization cannot be demonstrated to be primary. In spite of this all palaeopoles from a single block will help to define the APWP for that block, as long as later deformation events have not rotated the rocks; those that reflect secondary overprints will be located on a younger segment than primary poles for the same rock. Most intracratonic deformation, including formation or destruction of small (≤1000 km) will not be detectable within typical uncertainties (A₉₅) of 5-15 in palaeopole determination. Being aware of these constraints there are 3 ways the date can be explored.

A unique reconstruction of their relative positions can be achieved, in principle, by matching APWP for 2 or more tectonic blocks over an interval of time. The authors¹ suggest the data are insufficient [at the time the source¹ was published] to construct adequate APWP for different blocks, though many studies had been conducted in the previous decade. The McArthur Basin, 1.73-1.59 Ga, is the only block for which there is a well-defined path (Fig. 5, Idnurm et al., 1995; Idnurm, 2000). A consistently directed APWP direction for the Yilgarn Craton, at about 1.36 Ga (ML on Fig. 5), appears to be yielded, based on preliminary poles reported for the Morawa Lavas and the sedimentary rocks that enclose the Lavas.
The authors suggest that the palaeopoles form all the different blocks would be on a single APWP if the constituent Australian terranes assembled by large horizontal motions in the Proterozoic, such as those that are characteristic of plate tectonic regimes of the Phanerozoic. The closing of large oceans, in particular, between the NAC, WAC and SAC at 1.3 Ga or earlier, it would be expected that the APWP would be dissimilar before that time, and they having converged to produce a common path after the blocks were joined. The authors¹ suggest it is possible to construct a single APWP for all Proterozoic poles, though they are derived from different tectonic bocks, as noted in previous palaeomagnetic analyses (e.g. Idnurm & Giddings, 1988; Plumb, 1993). They also say that all recent results plot on the APWP produced by Idnurm & Giddings (1988), which is consistent with the different Australian regions having evolved in what is essentially their relative positions of the present since at least 1.8 Ga, though significant movement between the crustal blocks cannot be ruled out (Plumb, 1993), as a result of inadequacies of the APWP described above.
If for a particular time 2 or more blocks are in their correct relative positions the palaeopoles from the different blocks should overlap. Longitude is unconstrained when matching individual poles, or in this case average pole positions, instead of the APWP, hence the E-W separation between the blocks cannot be determined. According to the authors¹ the data from the different blocks overlap for 4 segments of the APWP.

Segment 1: 1.8-1.7 Ga

1.73-1.7 Ga McArthurBasin palaeopoles (Idnurm et al., 1995; Idnurm, 2000) are similar to

1.79-1.7 Ga Elgee Formation (McNaughton et al., 1999; Li. 2000a) and

1.79 Ga Hart Dolerite (McElhinny & Evans1976) of the Kimberley Block,

consistent with geological evidence of accretion of the Kimberley Block to the NAC by 1.82 Ga (Bodorkos et al., 1999; Sheppard et al., 1999a). A Pilbara syn-folding overprint pole (JO in Fig. 5; Schmidt & Embleton, 1985), as indicated by structural and palaeomagnetic studies, was acquired during the Capricorn Orogeny (Li, 2000a). That the NAC and WAC were in their present positions since at least 1.7 Ga (Li, 2000a) is implied by the proximity of this and similar poles from the iron ore deposits of the Pilbara (Porath & Chamalaun, 1968; Li et al., 1993; Schmidt & Clark, 1994) to the McArthur Basin and Kimberley Poles.

Segment 2: 1.6-1.5 Ga

In the McArthur Basin there are many 1.59-1.5 Ga overprint poles (Idnurm et al., 1995; Idnurm, 2000) and a 1.55-1.5 post-metamorphic cooling, pole (IM on Fig.5) from Mt Isa (Tanaka & Idnurm; 1994) overlap with overprint poles that are poorly dated from iron-ore deposits in South Australia (Chamalaun & Porath, 1968) and mafic dykes (GA and GB, Fig.5; Giddings and Embleton, 1976. The NAC and SAC were joined by at least 1.5 Ga, as suggested by the collective data, though the current age constraints are poor. At about 1.8 Ga the Tournefort Dykes, (Parker et al., 1987) from which the GA and GB palaeopoles (Fig.5) were obtained, were intruded into the 1.85 Ga Lincoln Batholith at about 1.81 Ga (Schaefer, 1998), subsequently being metamorphosed during the Kimban Orogeny at about 1.72 Ga (Bendall, 1994) and Rb-Sr ages have been obtained from them of 1.6-1.55 Ga (Mortimer et al., 1988b). The magnetisations could be overprints, the ages of which could possibly be the results of the Rb-Sr dating, as suggested by the lack of filed tests to verify the stability of the magnetisation of the GA and GB dykes, and by these observations.

Poles for the Gawler Range Volcanics, 1592 ± 2 Ma (GR on Fig.5; Chamalaun & Dempsey, 1978) and upper Balbirini Dolomite in the McArthur Basin (BDu on Fig.5; Idnurm, 2000) are about 60° apart, though the rocks are of identical age (Fanning et al., 1988; Page et al., 2000). The NAC and SAC would not be in their present relative positions at 1.59 Ga if both poles are primary, though the authors¹ consider it possible that either pole might be significantly younger than the rocks themselves. It has been argued, e.g., that the GR magnetisation is an overprint as palaeomagnetic directions from the lower parts of the unit dip steeply, are similar to previously obtained palaeomagnetic directions from upper flows that are flat-lying. A negative fold test could be suggested by combined datasets. Both successions are essentially of the age and the upper flow was erupted at temperatures of 950-1000°C (Creaser & White, 1991), hence there is the possibility that the entire unit did not cool through its magnetic blocking temperatures until the younger lava had been erupted, or possibly that the lower flows were reheated and overprinted by the overlying rocks. Magnetisation should have been acquired during cooling of the upper flows not long after 1592 Ma. The authors¹ therefore favour primary age for the GR pole, though they suggest the Gawler Ranges should be investigated further by additional field and lab tests on the stability of samples from the upper succession. A pre-folding age for the magnetisation (Idnurm et al., 1995; Idnurm, 2000) is indicated by the combined data from the upper and lower Balbirini Dolomite, and they say they tentatively regard the BDu pole as primary.

Segment 3: 1.36-1.32 Ga

An APWP vector has been obtained by for Yilgarn Craton at about 1.36 Ga (Idnurm & Giddings, 1988) from preliminary results for Morawa Lavas (ML on Fig.5) and sedimentary rocks that beneath and above the lavas. At about 1320 Ma dolerite was intruded (preliminary SHRIMP U-Pb baddeleyite age; Cloué-Long, pers. comm. to the authors¹) and this intrusion was probably responsible for overprints in the Roper Group (RG on Fig. 5) sedimentary rocks (Plumb, 1993; Idnurm et al., 1995). According to the authors¹ these results are consistent with the NAC and WAC being connected before 1.3 Ga.

Segment 4: 1.07 Ga

In the Giles Complex (GC on Fig.5) palaeopoles from the Late Mesoproterozoic have been obtained, as well as from the Stuart Dykes (SDS on Fig. 5) and the Kulgera Sills (KDS in Fig.5) in central Australia (Facer, 1971; Idnurm & Giddings, 1988; Camacho et al., 1991). The Giles Complex is well dated at 1078 Ma (Glickson et al., 1996), but the palaeopole (GC on Fig.5; Facer, 1971; recalculated by Tanaka & Idnurm, 1994) was obtained using outdated techniques so is of low reliability. Sm-Nd isochron ages of 1076 ± 33 Ma for the Stuart Intrusion and 1090 ± 32 Ma for the Kulgera Intrusion (Zhao & McCulloch, 1993). It is difficult to assess the reliability of the SDS palaeopole (Idnurm & Giddings, 1988) as there are no published analytical details. For the Giles Intrusion, the Kulgera Intrusion and the Stuart Intrusion (no tectonic corrections were applied), reliable constraints on palaeohorizontal are not available, and all of the 3 suites are on crustal blocks that were deformed and probably reoriented during the latest Neoproterozoic (Petermann Ranges) and/or Carbonaceous (Alice Springs) tectonothermal events (Tanaka & Idnurm, 1994; Wingate et al., 2002). The authors¹ suggest the Giles Complex may have also been deformed during the later stages of the Musgrave Orogeny (White et al., 1999). In dolerite sills in the Bangemall Basin, a new palaeopole (BBS on Fig. 5) disagrees with the previous results at about 1070 Ma, but is considered significantly more reliable (Wingate et al., 2002). The BBS pole is inferred to be primary and dates precisely to 1070 ± 6 Ma, and in the adjacent sedimentary rocks the structural control is well defined. Discrepancies between the BBS pole and about 1070 Ma poles from central Australia has been suggested by the authors¹ to possibly be to tectonic corrections that are inadequate for poles from central Australia, overprints that are unrecognised (GC and SDS on Fig. 5), and/or age differences.

The palaeomagnetic dates are not adequate to demonstrate that by the latest Mesoproterozoic Australia was amalgamated, though this interpretation is supported by the crude groupings of the poles about 1070 Ma (Fig.5). On the Kimberley Block a palaeopole for the dolomite cap of the Walsh Tillite has been proposed to indicate an age in the Sturtian, about 750-700 Ma, for that unit (Li, 2000b), though a correlation with the Marinoan glacial interval, about 600 Ma, is indicated by other evidence (Grey & Corkeron, 1998). Both interpretations have had compelling arguments in support of them. The authors¹ have proposed an alternative argument that is provocative "could similarity between the WTC pole and other Australian poles for 1070 Ma indicate a Late Mesoproterozoic age for some of the glaciogenic rocks in the Kimberley Block?

Discussion

The Capricorn, Late Barramundi and Halls Creek tectonothermal events all occurred at about 1.8 Ga. Across the North Australian Cratonic Assemblage (NAC) other similarities include low-P/high-T metamorphic conditions, as well as extensive magmatism with intraplate geochemical signatures (e.g., Etheridge et al., 1987; Wyborn, 1988; Mortimer, 1988a) that are widely developed. Between about 1.8 and 1.6 Ga basin evolution across the NAC (Scott et al., 2000) was contemporaneous with deformation in the Gawler Craton and central Australia. The Kimban Orogeny in the Gawler Craton, 1.74-1.7 Ga, was contemporaneous with the Late Strangways Orogenic Event in the Arunta Inlier, though in the intervening Musgrave Block no rocks have been found older than about 1.6 Ga (White et al., 1999).

According to the authors¹ amalgamation of the North, West and South Australian Cratonic Assemblages earlier than 1.3 Ga is suggested by a number of geological observations. A combined West and North Australian block is suggested to have been present by 1.3 Ga by orogenic events in the Albany-Fraser Belt and the Musgrave Belt at 1.3 Ga and 1.2 Ga respectively, though activity between them continued in the Miles Belt between 1.3 and 1.1 Ga. It has been suggested that the eastern Gawler Craton, Curnamona Province and the Georgetown and the eastern Mt Isa Inliers that bear striking similarities among them of such features as basin formation between 1.8 Ga and 1.55 Ga, magmatism, mineralisation, deformation, and alteration styles, that these blocks formed part of a Late Palaeoproterozoic-Early Mesoproterozoic mobile belt, the Diamantina Orogen, that was regionally extensive (Page & Laing, 1992; Connors & Page, 1995; Laing, 1996; Black et al., 1998; Robertson et al., 1998).

The North Australian Cratonic Assemblage and at least the Curnamona Province of the South Australian Cratonic Assemblage are implied by this hypothesis to have been joined by 1.7 Ga at the latest. If a collisional orogen in the Late Mesoproterozoic that extended between the Musgrave Block and the Cape River Block that has been suggested (Blewett et al., 1998) this would preclude the existence of a Late Palaeoproterozoic orogen linking Mt Isa to the Curnamona Province, with these blocks in their present relative positions,

It has been proposed (Giles & Betts, 2000) that an anticlockwise rotation of 55^o and slight eastward translation of the South Australian Cratonic Assemblage relative to the North Australian Cratonic Assemblage, based on detailed tectonostratigraphic comparison would be in better alignment with what the authors¹ suggest may have been continuous, linear tectonic elements in the Palaeoproterozoic-Mesoproterozoic. In the reconstruction of Giles & Betts, e.g., mobile belts of the Arunta Block and Gawler-Curnamona Block would have formed a continuous accretionary margin that was between 1.88-1.67 Ga.

At about 1.45 Ga a cycle of extension and breakup between the SAC and NAC has been suggested (Giles & Betts, 2000), with reamalgamation occurring between 1.3-1.1 Ga, the final result being the present configuration of the cratons. The authors¹ say that there is independent support to the first order for this proposal from palaeomagnetic data. If it is accepted that about 1.59 Ga both the GR and the BDu poles represent their respective cratons rotation of the SAC craton and the GR pole, according to this model, aligns better the GR and BDu poles along the aggregate APWP from the McArthur Basin (Idnurm, 2000). The authors¹ suggest that as alignment is still not exact it is possible that the GR and/or the BDu pole may not be primary, or possibly it may be necessary to modify the model of Giles & Betts (2000).

According to the authors¹ it is possible that collisional orogeny occurred by interaction between the Australian crustal blocks, as well as that intraplate processes can occur within a context of plate tectonics, they suggest there is a lack of compelling evidence that large basins have closed between any of the Australian blocks since 1.8 Ga, at least. To test whether major cratonic blocks of Australia were together or separated widely during the Proterozoic requires more than geological and geochemical observations. They suggest it is necessary to have conclusive palaeomagnetic evidence of large horizontal movements of crustal fragments in Australia in the Proterozoic. Groups of overlapping palaeomagnetic poles permit the NAC,WAC and SAC to have been in their present positions since 1.5 Ga, at least, and it is possible that the NAC and WAC were assembled by about 1.7 Ga. It is indicated by these conclusions that consistent with the blocks of Australia not having wide gaps between them since at least 1.8 Ga, and several pre-1.3 Ga and geological correlations support this.

The authors¹ suggest that there is still an inadequate palaeomagnetic database to rule out large horizontal movements between the crustal blocks of Australia since 1.8 Ga.

Sources & Further reading