Issuu

Geological Society of Australia

ABSTRACTS Number 65

From Basins to Mountains: Rodinia at the turn of tine century

Perth 30 September - 2 October 2001

Geological Society of Australia Abstracts Number 65 From Basins to Mountains: Rodinia at the turn of the century

Editors: K.N. Sircombe and Z.X. Li

ISSN 0729-011X ©

Geological Society of Australia Incorporated, 2001

Citation for this volume: Sircombe, K.N. & Li. Z.X (Editors) 2001. From Basins to Mountains: Rodinia at the turn of the century, 30 September-2 October 2001, Geological Society of Australia Abstracts 65, 120p. Example of citations for papers in this volume: Sircombe K. 2001. Promise and peril: microanalysis, provenance and reconstruction in Rodinia and beyond. In: Sircombe K. & Li. Z.X. (editors). From Basins to Mountains: Rodinia at the turn of the century, pp. 101-105. Geological Society of Australia Abstracts 65. OR Sircombe K. 2001. Promise and peril: microanalysis, provenance and reconstruction in Rodinia and beyond. Geological Society of Australia Abstracts 65. Copies of this publication may be obtained from: The Business Manager Geological Society of Australia Incorporated Suite 706 Thakral House, 301 George Street Sydney N S W 2000 Australia www.gsa.org.au

Printed by: Fineline Print and Copy Service, 11 Bramall Street, East Perth 6004.

Geological Society of Australia

ABSTRACTS Number 65 From Basins to Mountains: Rodinia at the turn of the century Chris Powell Memorial Symposiunfi Tectonics Special Research Centre IGCP 440 Department of Geology and Geophysics University of Western Australia Perth, Australia 30 September - 2 October 2001

Editors: K.N. Sircombe & Z.X. Li

Geological Society of Australia Abstracts 65

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

This symposium is dedicated to Professor Chris McA, Powell, the founding Director of the Tectonics Special Research Centre, IGCP 440 co-leader, and a champion ofRodinia research.

Ill

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Foreword about this Symposium because he saw the opportunity to gather together many of the key researchers and their ground-breaking ideas and new data from around the world. He was relishing the prospect of a lively academic debate. The organisers trust that this dynamic forum and the results of this symposium will be a long-lasting tribute to his vision.

It has been ten years since three landmark papers of Dalziel, Hoffman and Moores launched a worldwide debate on a late Precambrian supercontinent called Rodinia, the supercontinent from which the more familiar Gondwanaland and Pangea were to form. During the course of this debate, the Tectonics Special Research Centre, funded by the Australian Research Council and directed by Professor C. McA. Powell, was established in 1997. Its mission is to discover the supercontinents of which Australia has been a part, with special emphasis on Rodinia. An IGCP project (No. 440), first led by Professors C.McA. Powell and R. Unrug, was established in 1999 to examine the assembly and breakup of Rodinia. Over 300 scientists from every comer of the world participate in the project. Ten years on, a burgeoning body of literature has been generated testing the hypothesis, and alternative configurations have been proposed to accommodate the rapidly growing quantity of new information. Where to next? Will the hypothesis survive the scrutiny of the 21st century as the Gondwanaland hypothesis did in the 20th? It is the aim of this symposium to reflect on the scientific progress we've made in the past decade regarding the Rodinia hypothesis, present the latest data and ideas on both Rodinia and the dynamics of the Neoproterozoic world, and plot a course for the future. There will be one very empty seat at the Symposium because Chris Powell passed away suddenly in July. As many of the tributes that flowed in attest, he had vast enthusiasm and inspiration for many projects, be they structural, stratigraphic, paleomagnetic, geochronologic, tectonic, or viticultural. He was particularly excited

The convenors of the Rodinia Symposium would like to acknowledge and thank all who helped organise the symposium, and the following people in particular: Rosemarie Powell, TSRC Administrator, for organising the registration, accommodation, and symposium dinner, as well as for proofreading and keeping track of the minutiae that accompanies such an event. Geoff Wood, TSRC Business manager, for helping keep everything on the financial straight and narrow. Cath Wetherley, Research Assistant, for proofreading and answering anxious calls about translating various graphic formats. Bregje Hulscher, PhD student, for logistical efforts in ensuring the success of the oral program. Robin Swindell and R.C. Parker, Documentation Specialists, for proofreading and critical tips on MS Word features. We hope you all enjoy the symposium. Z.X. Li and K. Sircombe Co-convenors

Geological Society of Australia Abstracts 65

Table of Contents

Rodinia Symposium, Perth October 2001

The significance of Neoproterozoic deformation and magmatism in the Itremo group, central Madagascar: evidence for the break-up of Rodinia? J. Abello & S.P. Johnson 1 Rodinia provenance of Devonian sandstones in Buller and Takaka terranes, New Zealand C.J. Adams, M. Barley, M.A. Bradshaw & A.L. Pickard ^ Preliminary indications for Snowball Earth in the East African Orogen M. Beyth ^ AMCG-magmatism as an indicator of major tectonic events in Proterozoic Baltica S. V. Bogdanova ^ A statistical approach to testing Proterozoic continental reconstructions C. Burrett & R. Berry 10 The structural and kinematic analyses of the Mesoproterozoic Yanbian Group, South China: implications to the Cathaysia-Yangtze collision G.D. Carter. H Terra Australis orogen: Rodinian breakup and subduction initiation in the Pacific Ocean P.A. Cawood & E. Leitch Structural, isotopic and geochemical constraints on the evolution of the Leeuwin Complex, southwest Australia A.S. Collins & I.e. W. Fitzsimons 16 The structure, geochemistry and geochronology of eastern Madagascar: the remains of a strand of the Mozambique Ocean 20 A.S. Collins, I.C.W. Fitzsimons, T.S. Brewer, T. Razakamanana, B. Hulscher & B.F. Windley The birth, life and death of the Mozambique Ocean H. Cutten ^^ Amazonian Proterozoic poles: implications to Rodinia paleogeography M.S. D'Agrella-Filho, LLC. Pacca, R. Siqueira, S-A. Elming, W. Teixeira, J.S. Bettencourt & M.C. Geraides Evidence from South America on the formation and disruption of Rodinia M.S. D'Agrella-Filho, U.G. Cordani, B.B. de Brito Neves, R.I.F. Trindade & I.G. Pacca 31 Rodinia 2001: Balkanization or reunification? I.Dalzie l ^^ Implications of true polar wander for Rodinia reconstructions D.A.D. Evans 36 The Neoproterozoic evolution of Australia's western margin I.C.W. Fitzsimons 39 Systematic assessment of deformation in the hinge region of a chevron fold M.J. Gallen, S.M. Reddy & P.A. Cawood 43 Surviving the Snowball Earth: the acritarch record K. Grey 45 The assembly of Rodinia and Gondwana: a view from East Antarctica S.L. Barley 48 Was the Kalahari Shield attached to Laurentia-Baltica during the Grenvillean orogeny? C. Hatton The Neoproterozoic Snowball Earth hypothesis: trials and tribulations P.F. Hoffman 53 Landsat TM and digital elevation models as regional mapping and exploration tools: the step from 2D to 3D D.A. Hollingsworth & P.A. Cawood 55 Reconstruction of the Archaean Jimblebar greenstone belt, Sylvania Inlier, WA. D.A. Hollingsworth, P.A. Cawood & B. Monek 57 Tectonic framework of the Itremo sheet at -800 Ma: implications for the position of central Madagascar in Rodinia B. Hulscher, C.McA. Powell & I.C. W. Fitzsimons 60

Geological Society of Australia Abstracts 65

Rodinia Symposium, Perth October 2001

Paleoproterozoic basement in NW Sonora, Mexico: a key portion of the SW margin of Laurentia to reconstruct the Rodinia supercontinent A. Iriondo, W.R. Premo & M.J. Kunk 64 Magnetic anomahes along the southern margin of the Kalahari plate - Grenvillian or Pan-African? J. Jacobs, R.J. Thomas & A. Golynsky 68 High /O2 metasomatism during whiteschist metamorphism in the Zambezi Belt, northern Zimbabwe: amalgamation of the African cratons during the formation of Gondwanaland and not Rodinia S.P. Johnson 69 East Africa and Madagascar were not part of Rodinia, and east Gondwana was not a coherent block A. Kroner. 77 Does it take a superplume to breakup a supercontinent? — a case for Rodinia Z.X. Li, X.H. LU P.O. Kinny & H. Zhou 74 Grenville-age continental collision in south China: new SHRIMP age constraints and implications to Rodinia configuration 78 Z.X. Li, X.H. Li, H. Zhou & P.D. Kinny New U-Pb SHRIMP ages from the Tugela terrane, Natal Belt, South Africa: insights into the assembly of Rodinia S. McCourt, R.A. Armstrong & S.T. Johnson 80 Tectonic evolution of the Mesoproterozoic Albany-Fraser orogen and Neoproterozoic Leeuwin gneiss complex, southwestern Australia: Australian portions of Antarctic - Australian orogens SI J.S. Myers & D.R. Nelson Reconstructions of continents during the Proterozoic- a way towards Rodinia L.J. Pesonen, S. Mertanen & S.-A. Elming 82 An animated history of Rodinia C.McA. Powell, S.A. Pisarevsky & M.T.D. Wingate 85 Neoproterozoic extensional events in South Australia in the prelude to Rodinia breakup W.V.Preiss S8 Denudation of the Grenville orogen recorded by detrital zircon provenance studies of early Neoproterozoic sandstones on the margins of Laurentia R.H. Rainbird. 91 The high pressure belt in the Grenville province - tectonic setting, evolution, preservation and possible correlations with the Sveco-Norwegian beh in Sweden T. Rivers 94 Palaeomagnetic evidence for low-latitude glaciation: the geocentric axial dipole hypothesis for the Proterozoic? P.W.Schmidt 95 A re-examination of the 'Yanbian ophiolite suite': evidence for western extension of the Mesoproterozoic Sibao orogen in south China J.A. Sinclair 99 Promise and peril: microanalysis, provenance and reconstruction in Rodinia and beyond K.N. Sircombe 101 Vorogovian failed rift - evolution of typical sedimentary basin during of the Rodinia break up: late Neoproterozoic of Siberian craton, Russia J.K. Sovetov 105 The paradox of Proterozoic glaciation at sea level and strong seasonality near the palaeoequator: evidence and implications G.E. Williams 108 AUSMEX: a new Rodinia reconstruction at 1070 Ma M.T.D. Wingate, S.A. Pisarevsky & D.A.D. Evans 113 Positions of the north China block in Neoproterozoic Rodinia: a palaeomagnetic constraint S. Zhang & Z.X. Li 117

Geological Society of Australia Abstracts 65

Rodinia Symposium, Perth October 2001

THE SIGNIFICANCE OF NEOPROTEROZOIC DEFORMATION AND MAGMATISM IN THE ITREMO GROUP, CENTRAL MADAGASCAR: EVIDENCE FOR THE BREAK-UP OF RODINIA? J. Abello and S.P. Johnson Tectonics Special Research Centre, Department of Geology and Geophysics, The University of Western Australia, 35 Stirling Highway, Crawley 6009. W.A.

RATIONALE During the amalgamation of Gondwanaland at c. 550 Ma, palaeocontinental reconstruction models place Madagascar centrally within the collision zone (the East African Orogen or EAO). The Betsimisaraka Shear Zone in eastern Madagascar has been identified as a major, Pan African-aged, oceanic suture (Collins et. al., 2000) that separates the main block of Madagascar with a thin strip of lithologies along the east coast, which are correlated with Greater India. Hence, prior to the formation of Gondwanaland, there is no reason to place Madagascar next to Greater India and it is possible that it was either part of the African cratons or existed as a disparate, microcontinental block within the Mozambique Ocean. Madagascar's connection and location within the Rodinia supercontinent at c. 1000 Ma are now open to speculation. An investigation into the Neoproterozoic geology of Madagascar may provide critical information on global-scale tectonics at this time.

THE ITREMO GROUP DEFORMATION The Itremo Group of central Madagascar is a sequence of metasedimentary rocks comprising quartzites, pelites and carbonates whose age of deposition was established at between 1855 Ma and -^800 Ma (Cox et al. 1998). Deformation of the Itremo Group is constrained by the intrusion of a bimodal plutonic suite of granitoids and gabbros dated between 804 - 779 Ma (Handke et. al, 1999) and 820 - 740 Ma (Kroner et. al., 2000). A recent study of the structural and metamorphic relationships over an E-W traverse of the Itremo Group has identified at least two major deformation events. The 25km traverse region encompasses both the Ambalafampana granitoid (on the western margin) and the Itsindro Gabbro (on the eastern margin), allowing precise correlation and interpretation at the contact regions. The first deformation event, Di, formed north-south trending, west-verging isocHnal folds. These folds are best preserved in the centre of the region and have a

well-defined Si foliation that is almost layer parallel in the limbs of the isoclines but is axial planar in the hinge zones. D] folds formed prior to the emplacement of the igneous intrusives. The second deformation event, D2, produced rounded, open folds that locally refold Di in the eastern-most margin of the traverse. D2 folding is interpreted as syntectonic with the intrusion of the Itsindro gabbro.

CONTACT METAMORPHISM Contact aureoles associated with emplacement of both the granitoid and gabbro vary between 5m 1000m. Contact metamorphism around the Itsindro gabbro is well developed within a 1000m zone with country rocks comprising graphitic schists and dolomitic marbles. Within fine grained graphitic schists up to ~200m from the contact, poorly aligned andalusite porphyroblasts are abundant, indicating that intrusion of the gabbroic body occurred at pressures less than 3.8 kbars (based on the aluminium silicate triple point of Holdaway, 1971) or less than --^lOkm depth. Microstructural analysis of the andalusite-bearing schists indicates the development of a strong syn-deformation foliation (S2 axial planar fabric) that is overprinted by the poorly aligned andalusite porphyroblasts. These structures indicate that thermal metamorphism continued after D2 deformation had ceased. The presence of periclase in contact metamorphosed dolomitic marbles, -lOOm from the contact, also indicates that contact metamorphic temperatures exceeded a minimum of 700° C at relatively low XCO2 concentrations (< 0.4). At the Ambalafampana granite contact, the sediments show little evidence for intense thermal metamorphism, where pelitic lithologies are comprised of a low-grade, greenschist facies metamorphic assemblage of muscovite-biotite. Given the relatively thin nature of these contact aureoles and the strong foliation observed along the margins of the granite, syn-emplacement shearing and attenuation of the contact aureole is inferred. Analysis of Al concentrations of amphiboles within this granite indicate emplacement at -4.2 ± 0.6 kbars (using the thermometer of Anderson and Smith,

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

1995) and is thus consistent with independent pressure estimates from the Itsindro Gabbro contact aureole. Contact relationships between the Itremo metasediments and the bi-modal plutonic suite indicates intrusion at relatively low pressures of less than 3.8 kbar or 10km depth.

REGIONAL SYNTHESIS

Preliminary investigations indicate that Di and D2 deformation were part of a single, progressive, high-level deformation event that culminated with the intrusion of a bimodal gabbro and granitoid suite at c. 790 Ma. Since the geochemistry of this bimodal suite is interpreted to have a subductionrelated component (Handke et. al., 1999; Tucker et. al, 1999; Brewer et. al., 2001), and that Madagascar at this time occupied a marginal position to Rodinia, it has been proposed that Madagascar was part of a continental arc with the Mozambique Ocean undergoing eastward subduction beneath its western margin. However, the lack of intermediate compositions - and juvenile components to - the plutonics, high level of intrusion which accompanied isoclinal folding and, synchronicity with the breakup of Rodinia led Powell et. al., (2001) to propose that this c. 790 Ma event was extensional-related. A similar period of bimodal plutonism is recorded in the Seychelles at c. 750 Ma, ranging from 748 - 808 Ma (Tucker et. al., 2001), and is interpreted by these authors to be part of the same continental arc. Stephens et. al., (1995 and 1997) have re-interpreted the geochemistry of these magmas and relate them to a period of extensional tectonics similar to that proposed by Powell et. al., (2001). Clearly there are two contrasting tectonic models for the Neoproterozoic evolution of central Madagascar (and the Seychelles) both of which have significant implications for the reconstruction and break-up of the Rodinia Supercontinent. The first model indicates that Madagascar and the Seychelles

were part a continental arc located on the fringe of Rodinia, facing an already open Mozambique Ocean. The second model allows the positioning of Madagascar centrally within Rodinia and that the extensional tectonics recorded at c. 790 - 750, relates to the break-up of Rodinia and the birth of the Mozambique Ocean. Some middle point of view may be proposed that relates both the subductionand extension-related geochemical components of the plutonics to back-arc extension. Further work on the structural, metamorphic and geochemical evolution of the these regions and a robust geotectonic model is needed before we can begin to reconstruct this part Rodinia. References

Anderson, J. L. & Smith, D. R., 1995. American Mineralogist 80, 549-559. Brewer, T. S., Collins, A., Kroner, A., Windley, B. & Razakamanana, T. 2001. Abstract. EUG XI, Strasbourg Meeting. Collins, A. S., Razakamanana, T. & Windley, B. F. 2000. Geology Magazine 137,39-51. Cox, R., Armstrong, R.A. & Ashwal, L.D., 1998. Journal of the Geological Society of London 155, 1009-1024. Kroner, A., Hegner, E., Collins, A. S., Windley, B. F., Brewer, T. S., Razakamanana, T. and Pidgeon, R. T. 2000. American Journal of Science 300, 251 - 288. Handke, M.J., Tucker, R.D. & Ashwal, L.D. 1999. Geology 27, 351-354 Holdaway, M. J. 1971. American Journal of Science 271, 97 131. Powell, C. McA., Dahl, K., Hulscher, B., Johnson, S.P., Passmore, A.R., Collins, A.S., Fitzsimons, I.C.W., and Jonsson, M. K. 2001. Abstract. Specialist Group in Structural Geology meeting, Tazmania 2001. Stephens, W. E., Fallick, A. E. and Ellam, R. M. 1995. Abstract. Terra Nova 7 (1995) pl45. Stephens, W. E., Jemielita, R. A. and Davis, D. 1997. Abstract. Terra Nova 9 (1997) pl66. Tucker, R. D., Ashwal, L. D., Handke, M. J., Hamilton, M. A., Grange, M Le. and Rambeloson, R. A. 1999. The Journal of Geology 107, 135- 153. Tucker, R. D., Ashwal, L. D. and Torsvik, T. H. 2001. Earth and Planetary Science Letters 187, 27 - 38.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

RODINIA PROVENANCE OF DEVONIAN SANDSTONES IN BULLER AND TAKAKA TERRANES, NEW ZEALAND C J . Adams\ M. Barley^ M.A. Bradshaw^ and A . L Pickard^ ^ Institute of Geological & Nuclear Sciences, P.O. Box 30368, Lower Hutt, New Zealand. ^ Dept. of Geology of Geology & Geophysics, University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia. ^ Dept. of Geological Sciences, University of Canterbury, Private Bag 4800, Christchurch, New Zealand.

INTRODUCTION

Devonian sediments of Reefton Group and Baton Formation occur in South Island, New Zealand (Fig. 1) in two small areas, 130 km apart (Bradshaw 1999, 2000), within the Buller and Takaka Terranes respectively (Cooper & Tulloch 1992). The Devonian overlies voluminous uniform Ordovician greywacke suites (Greenland Group) in the Buller Terrane, and more variable, Cambrian volcanics, Ordovician limestones and Silurian sandstones in the Takaka Terrane. Different provenances, recorded as U-Pb SHRIMP detrital zircon age patterns would be expected in the Devonian sandstones. The shelly faunas of the Reefton Group indicate a Pragian to Emsian (Early Devonian) age, while Lokhovian conodonts from the base of the Baton Formation and Siegenian brachiopods from the middle of the sequence suggest a slightly older age (see Bradshaw 1999, for references). The two areas show no clear faunal similarity.

DETRITAL ZIRCON AGES

U-Pb zircon ages were determined on SHRIMP II at Curtin Institute, Perth. Technical details are given in Pickard et al (2000). Ages <1000 Ma are 206p|^/238u data and >1000Ma, data, shown as histograms 200-1300 Ma (Fig. 2) and 03500 Ma (Fig. 3). Full analytical data are available from the authors. Also shown in Fig. 3 are analogous data for Early Paleozoic, Onekaka Schist at Onekaka, Takaka Terrane (Wyszoczanski et al. 1997) and Greenland Group, Reefton, Buller Terrane (Ireland & Gibson 1998) (see Fig. 1). Baton Formation, Takaka Terrane. Sandstone (BRX14) yielded a complex population (96 grains) of pink 40-200 micron grains, mostly rounded, but some euhedral and subhedral ones. The data (Figs. 2a, 3b) show a principal late Neoproterozoic to Early Cambrian component at 620-520 Ma (28% of total), and minor components: Early Ordovician to latest Cambrian, 463-498 Ma (8%); Neoproterozoic, 646668 Ma (4%) and 847-905 Ma (5%); late Mesoproterozoic, 1011-1086 Ma (10%); early Mesoproterozoic, 1404-1490 Ma (9%) and Paleoproterozoic, 2036-2097 Ma (5%). The 3 oldest

ages are Archean (2777-2961 Ma); the youngest 437±6 Ma, late Ordovician-early Silurian. Reefton Group, Buller Terrane. Sandstones, RFT6-7, combined, yielded a similarly complex population (76 grains) of fine 40-100 micron, mostly rounded grains, with a few subordinate subhedral ones. The data (Figs. 2b, 3c) again show a principal component at Early Cambrian to late Neoproterozoic, 528-615 Ma (22%), but less well defined Neoproterozoic, 630-980 Ma, (26%) and Mesoproterozoic, 1003-1166 Ma (14%) components. Significantly, the minor early Ordovician component (466-498 Ma) is absent. There are 6 Archean ages, 2689-3400 Ma; the youngest age is 461 ±7 Ma, mid-Ordovician.

PROVENANCE OF BULLER AND TAKAKA TERRANES

Baton River and Reefton Devonian detrital zircon age patterns are similar, suggesting a common provenance. Each is also similar to underlying Paleozoic (compare Figs. 3a and 3b, and Figs. 3c and 3d). The Baton Formation appears to rest conformably on Silurian to early Devonian Ellis Formation, a relatively deep-water sequence (Bradshaw 2000). The base of the Reefton Group, a near shore to upper shelf sequence, is not visible, but is likely to be unconformable on Greenland Group, based on cleavage ages for the latter and the absence of Silurian rocks. Local sediment sources for both areas seem unlikely. Cambrian, intermediate-basic Devils River, and Anatoki Volcanics, occur only in westernmost Takaka Terrane (Cooper & Tulloch 1992), but are volumetrically minor. No Precambrian basement occurs in either Buller or Takaka Terranes, so the abundant NeoproterozoicEarly Cambrian detrital zircons, quite characteristic of NZ Paleozoic sediments (and their counterparts in the Lachlan Fold BeU of SE Australia), must be recycled, through older sediments and granitoids, from elsewhere. Granitoid clasts occur in Cambrian and Devonian conglomerates in the Takaka Terrane (Cooper & Tulloch 1992), but their age and origin are uncertain.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

A more distant source, in North Victoria Land (Antarctica) and South Australia, could have been at the contemporary Gondwana margin, as Late Cambrian-earliest Ordovician granitoid complexes (Black & Sheraton 1990), or Early Cambrian volcanics, of the Ross-Delamerian Fold Belt. However, the dominant, late Neoproterozoic (c. 540-620 Ma) detrital zircons, seen in all NZ Paleozoic sediments (Fig. 3), seem too old to be derived from these rocks. The Devonian zircon ages in particular, match poorly with RossDelamerian sources. The Early to Middle-Ordovician age component (466-498 Ma), present at Baton River, but absent at Reefton, is important, suggesting that the two New Zealand areas probably occupied different sites along the Gondwana margin. The NZ Paleozoic detrital zircon data (Fig. 3), compare with SE Australian (Lachlan Fold Belt) counterparts (Williams 1998), in having a main latest Proterozoic to Middle Cambrian, 625-510 Ma component, then successively smaller ones in late Mesoproterozoic, 1000-1200 Ma, and late Paleoproterozoic, 1600-1800 Ma. The proportion of Neoproterozoic ages is surprisingly high (>20%), and particularly so in the NZ Devonian (>30%), given that igneous rocks of this age are rare and volumetrically insignificant in east/central Australia and Antarctica (Stump et al. 1986), and absent in New Zealand. Similarly, Mesoproterozoic (10111096 and 1404-1490 Ma) zircon age components are difficult to match with the adjacent Australian craton, which is dominated by Paleoproterozoic, >1600 Ma (Wybom et al. 1987, Fanning et al. 1988, Page 1988), and Archean plutonism. In the East Antarctic craton further south (mostly beneath Polar icecap), the same situation applies (Goodge et al. 1993), but distant Paleoproterozoic and Mesoproterozoic terranes are known in Dronning Maud Land (Moyes et al. 1997), Haag Nunataks, Antarctic Peninsula (Clarkson & Brook 1977) and the Falkland Islands (Rex & Tanner 1982). However, Neoproterozoic igneous rocks are insignificant. To the north, in NE Australia, the Delamarian Fold Belt narrows (Anakie Inlier, Queensland) and the Lachlan Fold Belt is replaced by the Thomson Orogen (Day et al. 1983). The latter includes Cambrian acid-intermediate volcanics and Late Ordovician-earliest Silurian granitoid complexes (Ravenswood Province) which are good potential zircon sources for Cambrian and Ordovician zircon components in NZ Devonian sediments. The Precambrian inliers of this region (Georgetown, Coen and Yambo), include the acid-intermediate Croydon and Holroyd Volcanics, which could have contributed Mesoproterozoic (1400-1600 Ma) or Paleoproterozoic components (Black & McCulloch 1990). However, Neoproterozoic sources are again

absent. At this point a Rodinia perspective becomes important. Li et al. (1996) suggested crustal blocks of SE China, which include important Neoproterozoic volcanic sequences (and Mesoproterozoic terranes), formed part of Rodinia adjacent to NE Australia in late Neoproterozoic to Early Paleozoic times. Adams (1999) then suggested that these SE China crustal blocks could have been a potential source of Neoproterozoic zircons. These would be cycled, firstly with Cambrian detritus into Ordovician to Silurian sediments in Queensland depocentres and then secondly, with late Ordovician-early Silurian granitoid detritus, into Devonian sediments such as the NZ Baton Formation. This implies that the Takaka Terrane originated in a lower latitude, distant from its present mid-latitude (NZ) position. The Reefton Group, Buller Terrane detrital zircon age patterns are less distinctive and a provenance remains uncertain. Different relative positions of the Baton and Reefton areas in the early Devonian are suggested by their contrasting faunas despite the age similarity. The Baton Formation shows strong faunal links with eastern Australia, especially New South Wales, and was part of the Tasman Subprovince, which had affinities with the Old World Realm of Europe. In comparison, the Reefton Group fauna contains several endemic elements indicating a Reefton Subprovince, distinct from the Tasman Subprovince. This shared some genera with Victoria, although not with Baton River. Elements of the Reefton fauna show significant Malvinokaffric Realm influence and links to South American and Antarctic (Ohio Range) Lower Devonian faunas along the contemporary Gondwana coastline. The Reefton Group contains small coral reefs in one of its limestone formations that suggest a paleolatitude of about 30®(Fagerstrom & Bradshaw, in press).

References

Adams, C.J. 1998: A provenance in northeast Australia and south China for New Zealand Paleozoic terrane sediments. In Almond, J., Anderson, J., Booth, P., Chinsamy-Turan, A., Cole, D., De Wit, M.J., Rubridge, B., Smith, R., Van Bever Donker, J. {eds) Gondwana 10: Event stratigraphy of Gondwana. Journal of African Earth Sciences 27: 218-219. Black, L.P., Sheraton, J.W. 1990: The influence of Precambrian source components on the U-Pb zircon age of a Paleozoic granite from Northern Victoria Land, Antarctica Precambrian Research 46: 275-293. Black, L.P., McCulloch, M.T. 1990: Isotopic evidence for recurrent felsic magmatism on new crust formation: An example from the Georgetown Inlier of Northeastern Australia. Geochimica et Cosmochimica Acta 54: 183-196. Bradshaw, M.A. 1999: Lower Paleozoic bivalves from the Reefton Group, New Zealand. Association Australasian Palaeontologists, Memoir 20, 171 pp. Bradshaw, M.A. 2000: Base of the Devonian Baton Formation and the question of a pre-Baton tectonic event in the Takaka Terrane, New Zealand. New Zealand Journal of Geology & Geophysics 43: 601-610. Clarkson, P.D.; Brook, M.: Age and position of the Ellsworth Mountains crustal fragment, Antarctica. Nature 265: 615616.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Cooper, R.A., Tulloch, A.J. 1992: Early Paleozoic terranes in New Zealand and their relationship to the Lachlan Fold Belt. Tectonophysics 214: 129-144. Day, R.W., Whitaker, W.G., Murray, C.G., Wilson, I.H., Grimes, K.G. 1983: Queensland Geology (A companion volume to the 1:2500000 scale geological map (1975), Geological Survey of Queensland Publication 383: 194pp. Fagerstrom, J.A.: Bradshaw, M.A. {in press): Early Devonian reefs at Reefton, New Zealand: guilds, origin and paleogeographic significance. Lethaia. Fanning, C.M., Flint, R.B., Parker, A.J., Ludwig, K.R., Blissett, A.H. 1988: Refmed Proterozoic evolution of the Gawler Craton, South Australia, through U-Pb zircon geochronology. Precamhrian Research 40/41: 363-386. Goodge, J.W., Walker, N.W., Hansen, V.L. 1993: Neoproterozoic basement-involved orogenesis within the Antarctic margin of Gondwana, Geology 21: 1087-\090. Ireland, T.M.; Gibson, G.M. 1998: SHRIMP monazite and zircon geochronology of high-grade metamorphism in New Zealand. Journal of Metamorphic Geology 16: 149-167. Li, Z.X., Zhang, L., Powell, C.McA. 1996: Positions of the East Asian cratons in the Neoproterozoic supercontinent Rodinia, Australian Journal of Earth Sciences 43: 593-604. Moyes, A.B.; Knoper, M.W.; Harris, P.D. 1997: The age and significance of the Uracil Group, western Droiming Maud Land. In Ricci, C-A. {ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antarctica PubHcation, 31-36. Page, R. W. 1988: Geochronology of Early to Middle Proterozoic fold belts in Northern Australia: A review. Precamhrian Research 40/41: 1-19. Pickard, A.L.; Adams, C.J.; Barley, M.E. 2000: Australian provenance for Upper Permian rocks forming accretionary New Zealand

complexes on the New Zealand sector of the Gondwanaland margin. Australian Journal of Earth Sciences 47: 987-1007. Rex, D.C.; Tanner, P.W.G. 1982: Precambrian age for gneisses at Cape Meredith in the Falkland Islands. In Craddock, C. {ed) Antarctic Geoscience, University of Wisconsin, Madison, Wisconsin, 107-108. Stump, E.; Smit, J.H.; Self, S. 1986: Timing of events of the Late Proterozoic Beardmore Orogeny, Antarctica: Evidence from the La Gorce Mountains. Geological Society of America Bulletin 97: 953-965. Williams, I.S. 1998: The Lachlan Fold Belt, southeastern Australia: SHRIMP's eye view of crustal growth in eastern Gondwana. In Almond, J., Anderson, J., Booth, P., Chinsamy-Turan, A., Cole, D., De Wit, MJ., Rubridge, B., Smith, R., Van Bever Donker, J., Gondwana 10: Event Stratigraphy of Gondwana, Spec. Abstr. Issue, Journal of African Earth Sciences 27: 211-212. Wybom, L.A.I., Page, R.W., Parker, A.J. 1987: Geochemical and geochronological signatures in Australian Proterozoic igneous rocks. In Pharaoh, T.C., Beckinsale, R.D., Rickard, D. (eds.) Geochemistry and mineralization of Proterozoic volcanic suites. Geological Society of London Special Publication 33: 377-394 . Wyszoczanski, R.J., Gibson, G.M., Ireland, T.R. 1997: SHRIMP dating of detrital zircons from the Takaka Terrane of New Zealand and Beacon Supergroup, Antarctica: Implications for source components and terrane accretion. In Bradshaw, J.D. and Weaver, S.D. {eds.) Terrane Dynamics 97, University of Canterbury, Christchurch, N.Z., 170-172.

J • y

T A S M A N

SEA

50 kms

Figure 1: Geological sketch map of Buller (pale grey) and Takaka (dark grey) Terranes, South Island, New Zealand and Devonian sediments (black) at Baton River, BRX and Reefton, RFT. Greenland Group, Reefton is GG, and Onekaka Schist, Onekaka, OS. Stipple denotes younger granitoids.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

(a)

brx14(R20893) Devonian; Baton River

(a)

Onekaka Schist

I 10 5

500

in Ifnillllllh n 200

400

600

800

1000

1200

nJlllm

n n n f i m n> n

1000

2000

1500

[

Age (Ma)

(b) rft6-7 (R21254-5) Devonian; Reefton

3000

3500

(b) brx14 (R20893) Devonian; Baton River

2500

Age (Ma)

c ®

L9 L?

m ^ l l r Hin. nm

r k i 0

500

1000

1500

2000

2500

3000

3500

Age (Ma)

ni 0200

400

I 600

I . M

JL.i 800

1000

1200

(c)

rft6-7 (R21254-5) Devonian; Reefton

Age (Ma)

Figure 2. U-Pb detrital zircon age histograms (2001300 Ma) for Devonian sandstones (a) BRX14 (IGNS No. R20893) Baton Formation (Takaka Terrane) at Baton River (New Zealand 1:50 000 Map M27, Grid Ref.: 820894) and (b) RFT6-7 (IGNS Nos. R21254-5), Murray Creek Formation (Buller Terrane) at Stony Creek, south of Reefton (New Zealand 1:50 000 Map L30, Grid Ref: 200952).

M M 500

:mni n i . l

1000

1500

-T

2000

2500

3000

3500

Age (Ma)

(d) Greenland Group Reefton

10 1

ilinlnll

f 0

500

1000

1500

2000

nn n 2500

3000

3500

Age (Ma)

Figure 3. U-Pb detrital zircon age histograms (03500 Ma) for (a) Onekaka Schist (?Early Paleozoic), Onekaka (Takaka Terrane), data from Wyszoczanski et al 1997, and Devonian sandstones (b) BRX14 (Devonian), see Fig. 2, (c) RFT6-7 (Devonian), see Fig. 2, and (d) Ordovician greywacke, Greenland Group, Reefton (Buller Terrane), data from Ireland & Gibson 1998.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

PRELIMINARY INDICATIONS FOR SNOWBALL EARTH IN THE EAST AFRICAN OROGEN M. Beyth Geological Survey of Israel, 30 Malkei Yisrael str., Jerusalem, Israel

The East African Orogen (EAO), which stretches from southern Israel in the in the north to Madagascar in the south, was formed between east and west Gondwanaland. The EAO manifested a Wilson Cycle orogen at the Late Neoproterozoic between -850-532 Ma. The potential expression of Snowball earth in the sedimentary record of the EAO of southern Israel and Jordan and Eastern Eritrea and northern Ethiopia was preliminary studied. The Saramuj Conglomerate from southern Jordan and the Elat conglomerate from southern Israel are diamictites, part of the later E-W volcanoclastic graben fills. The sequence indicates one of two strong erosional phases dated approximately 600 Ma which might stimulate by glacial processes. The widespread peneplain of very early Cambrian age (532 Ma) with the extensive chemical weathering may have formed by post glacial acid torrential rains.

Early Cambrian and Varanger glacial events were documented in north Africa, Algeria at the transition period and at the peneplain. Stromatolitic limestones, with microbial structures, were reported from the Mai Kenetal Synclinorium of northern Ethiopia. Laminated dolomite and Polymict conglomerate were reported from the Negash Synclinorium of northeast Ethiopia. Both are of Late Neoproterozoic age probably older than 600 Ma. Stromatolitic dolostones and algal mats were found in the metasediments of the Bizen Domain of eastern Eritrea, the northeastern continuation of the Negash sedimentary sequence. The sequence consists of greenschist metamorphic facies with light 6130 (-7.26 to -6.74) and 6180 (12.48 to -5.46) which are characteristic for Capcarbonates. Similar sedimentary sequence, with polymictic conglomerates overlain by Stromatolitic carbonates, was also reported from the Arabian Peninsula.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

AMCG-MAGMATISM AS AN INDICATOR OF MAJOR TECTONIC EVENTS IN PROTEROZOIC BALTICA S.V. Bogdanova Dept. of Geology, Lund University, Solvegatan 13, S- 223 62 Lund, Sweden Svetlana.Bogdanova@geol.lu.se

The East European Craton (EEC) is the core of Rodinian Baltica. It consists of the three cmstal segments Fennoscandia, Sarmatia and Volgo-Uralia which evolved separately during the Archaean and early Palaeoproterozoic (e.g. Bogdanova et al., 1996) but began to amalgamate at c. 2.1 Ga. By 1.7 Ga, the formation of the EEC was completed and at 1.1-1.0 Ga that craton became part of Rodinia. While most of the EEC is composed of Archean crust, its western part is dominantly juvenile Palaeoproterozoic. That is where the Proterozoic AMCG (=Anorthosite-mangerite-chamockite (rapakivi) granite) association and related supracrustal rocks are best developed. Altogether, there are five major AMCG provinces: (1) The Ukrainian (1.79-1.72 Ga), (2) the Finnish (1.65-1.56 Ga), (3) the South Baltic (1.54- 1.42 Ga), (4) the SW Swedish (1.40-1.37 Ga), and (5) the South Norwegian (0.95-0.92 Ga).

1. The Ukrainian AMCG: A consequence of the 1.8-Ga collision of minor Palaeoproterozoic terranes with Sarmatia plus Volgo-Uralia

After Sarmatia and Volgo-Uralia had docked at 2.1-2.0 Ga, the northwestern edge of the new continent became an active continental margin marked by the 2.0-1.95 Ga Osnitsk-Mikashevichi Igneous Beh. Around 1.8 Ga, several outboard Palaeoproterozoic collided with and accreted to that margin. This led to the crustal thickening and the attendant formation of AMCG plutons and associated supracrustal rocks between 1.79 and 1.72 Ga. The best studied AMCG complex is the Korosten Pluton which is a multiphase intrusion associated with a collisional mantle suture only recently detected (EUROBRIDGE Seismic Working Group, 2000).

2. The Finnish AMCG: Collision at 1.7 Ga between Fennoscandia and the Sarmatia / Volgo-Uralia protocontinent, and the birth of the East European Craton The distribution of the 1.65-1.45 Ga AMCG magmatism in space and time has been much

discussed. Usually, that magmatism is thought to represent one single protracted process which affected the west of Pre-Rodinian Baltica. The currently most favoured model is based on geological-geophysical interpretation of the largest AMCG plutons (e.g. the Wiborg, Riga and Aland) around the Gulf of Finland. These associate with the thickest Palaeoproterozoic crust in the region, which also shows signs of extensional tectonics and uppermantle disturbance. An overall post-collisional setting is here thought to have succeeded the assembly of Svecofennian crust which terminated at c. 1.85 Ga (e.g. Haapala and Ramo, 1992; Korja and Heikkinen, 1995; Nironen, 1997; Puura and Floden, 2000). However, the time span of 150-200 Ma between the Svecofennian colhsions and the AMCG magmatism appears much too long to allow a causal relationship. An alternative explanation may therefore be that the thickening of the crust to 55-65 km and the attendant AMCG event occurred in response to the definite formation of the EEC c. 1.7 Ga ago by collision between Fennoscandia and the combined Sarmatia - Volgo-Uralia protocontinent.

3. AMCG plutons around the southern Baltic Sea: Association with the 1.54-1.42 Ga Dano-Polonian collisional orogeny

Together with the older, 1.65-1.56 Ga AMCG intrusions, the intrusions of 1.54-1.42 Ga age have been considered to define a pattern of westwards younging of the AMCG magmatism in the northwestern EEC. That younging has been explained either as a consequence of crustal drift across the top of a long-lived mantle plume (Korja and Heikkinen, 2000; Puura and Floden, 2000) or in terms of causal relationships between orogenic events along the margins of ProtoBaltica and Laurentia, and the various phases of AMCG magmatism within the EEC (Ahall et al., 2000). However, both these models meet difficulties particularly in regard to the siting of the 1.54-1.42 Ga intrusions. These appear irregularly distributed in regard to the loci and age patterns of the earlier AMCG rocks. A completely different approach is suggested by recent work demonstrated that the 1.54-1.42 Ga AMCG plutons associate closely with major EWtrending lineaments/shear zones (Bogdanova and

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Cecys, 1999; Bogdanova, 2001). The kinematics of these lineaments are not completely understood yet, but new studies of the region around the southern Baltic Sea in Sweden, Denmark, Poland and Lithuania allow the tentative proposal that the AMCG emplacements between 1.54 and 1.42 Ga were connected with N-S (in present-day terms) compressional/transpressional tectonics. The large sizes and regular distribution of the E-W lineaments evidence that ProtoBaltica had undergone substantial deformation at that time, apparently during an orogeny (the "Dano-Polonian" of Bogdanova, 2001) which involve its pre-Grenvillian collision with another continent, possibly ProtoAmazonia (e.g. Sadowski, 1997).

4. Post-collisional 1.4-1.37 Ga AMCG intrusions in SW Sweden After or during the late stages of the Dano-Polonian collisional orogeny, the Varberg and other c. 1.41.35 Ga AMCG-hke intrusions in SW Sweden (Hubbard, 1975; Ahall et al., 1997; Andersson et al., 1999) were accommodated in extensional structures, subsequently to be much reworked during the Sveconorwegian orogeny at the end of the Mesoproterozoic.

5. The South Norwegian AMCG: A postcollisional response to the Grenvillian /Sveconorwegian orogeny The Rogaland AMCG complex of 0.93-0.92 Ga age is the foremost example of post-collisional intrusions succeeding the Grenvillian/Sveconorwegian orogeny. These intrusions appear related to rejuvenated terrane boundaries which channelled the AMCG melts on their way upwards from the lower crust and the upper mantle (e. g. Duchesne et al., 1999). To conclude, most of the AMCG magmatism in Baltica (1, 2, 4 and 5) appears to have been caused by post-collisional processes almost immediately (within 10 to 80 Ma) following upon major crustal collisions at c. 1.8, 1.7, 1.54-1.42 and 1.1-1.0 Ga. Only the 1.54-1.42 Ga AMCG around the southern Baltic Sea appear to have been synkinematically emplaced into large-scale lineaments formed during

the Dano-Polonian collisional orogeny which strongly affected the south-westernmost EEC. References Ahall, K.-L, Samuelsson, L. and Persson, P.-O. (1997). Geochronology and structural setting of the 1.38 Ga Torpa granite: implications for chamockite formation in SW Sweden. OFF, 119, 37-43. Ahall, K.-I., Connely, J.N. and Brewer, T.S. (2000). Episodic rapakivi magmatism due to distal orogenesis?: Correlation of 1.69-1.50 Ga orogenic and inboard, "anorogenic" events in the Baltic Shield. Geology, 28(9), 823-826. Andersson, J., Soderlund, U., Cornell, D., Johansson, L. and Moller, C. (1999). Sveconorvegian (-Grenvillian) deformation, metamorphism and leucosome formation in SW Sweden, SW Baltic Shield: constraints from a Mesoproterozoic granite intrusion. Precambrian Research, 98, 151-171. Bogdanova, S.V. (2001). Tectonic settings of 1.65-1.4 Ga AMCG magmatism in the western East European Craton (western Baltica). J. Conference Abstracts, 6(1), 769. Bogdanova, S.V. and Cecys, A. (1999^ Mesoproterozoic active shear zones controlling anorogenic magmatism in the western part of the East European Craton. J. Conference Abstracts, 4, 688. Bogdanova, S. V., Pashkevich, 1. K., Gorbatschev, R. and Orlyuk, M.(1996). Riphean rifting and major Palaeoproterozoic boundaries in the East European Craton: geology and geophysics. Tectonophysics, 268, 1-22. Duchesne, J.C., Liegeois, J.P., Vander Auwera, J. and Longhi, J. (1999). The crustal tongue melting model and the origin of massive anorthosites. Terra Nova, 11: 100-105. EUROBRIDGE Seismic Working Group (2000). EUROBRIDGE'97: Modelling of S-waves field on the EB'97 seismic profile. Geophysical Journal, 22, 4, 87-88. Haapala, I. and Ramo, T. (1992). Tectonic setting and origin of the Proterozoic rapakivi granites of Southeastern Fennoscandia. Transactions of the Royal Society of Edinburgh, Earth Sciences 83, 165-171. Hubbard, F.N. (1975). The Precambrian crystalline complex of south-western Sweden The geology petrogenetic development of the Varberg Region. Geol. Foren Stockholm Forh.(97), 223-236. Koija, A. and Heikkinen, P. (1995). Proterozoic extensional tectonics of the central Fennoscandian shield: results from the Baltic Bothnian echoes from the lithosphere experiment. Tectonics, 14, 504-517. Nironen, M., 1997. The Svecofennian Orogen: a tectonic model. Precambrian Research, 86: 21-44. Puura, V. and Floden, T.(2000). Rapakivi-related basement structures in the Baltic Sea area; a regional approach. GFF, 122, 257-272. Sadowski, G.R. (1997). The fir between Amazonia, Baltica and Laurentia during the Mesoproterozoic assemblage of the supercontinent Rodinia. Terra Nova, 9 (Abstract Supplement 1): 168.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

A STATISTICAL APPROACH TO TESTING PROTEROZOIC CONTINENTAL RECONSTRUCTIONS C. Burrett and R. Berry School of Earth Sciences, University of Tasmania, Box 252-79, Hobart, Tasmania, Australia 7001

The inclusion of the Nevada-Califomian Mojave and the San Gabriel Terranes in an otherwise Australian superprovince that includes Broken Hill and Mt Isa Terranes, strongly supports the AUSWUS Laurentia-Australia reconstruction rather than the SWEAT reconstruction. Low statistical similarities between western Laurentia and eastern Antarctica fail to support the SWEAT hypothesis whilst high similarities between Canadian and north Australian terranes provides weak support for AUSWUS. Inclusion of the Indian and Chinese terranes provides clues to the placement of the Chinese terranes in eastern Gondwana. Terranes in southern China have high similarities to those in India and support the Cambrian placement of South China near India as suggested by us based on paleobiogeographic and paleomagnetic data (Burrett et al 1990). Late Proterozoic paleomagnetic data from India and southern China suggest that their constituent terranes did nor amalgamate with Antarctica and Australia until the early Paleozoic. Our methodology and database are being updated in order to include terranes from other continents and to include qualitative data from each terrane.

A new statistical method is proposed to compare crustal terranes and to cluster terranes into crustal provinces, regions and realms. Geochronological data on mafic igneous rocks, felsic igneous rocks, deformation history and Nd model age were collected from the recent literature for over 100 terranes in Antarctica, Australia, China, India, Korea, Laurentia, Mexico and Vietnam. Our aims are: • To develop a statistical method for clustering terranes into crustal provinces. • Define Proterozoic crustal provinces for Laurentia and eastern Gondwana. • To use the quantitative inter-terrane comparisons and the newly defined provinces to test Proterozoic paleocontinental reconstructions particularly the very influential SWEAT (of Professors Dalziel, Moores and Hoffman) and the new AUSWUS (of Karlstrom et al and Burrett and Berry) hypotheses. • Assess models of the paleocontinental placement of the Chinese terranes. Although our results may be suggestive, our aim is NOT to use quantitative comparisons and clustering in order to develop palinspastic terrane reconstructions. The 54 selected Laurentian terranes cluster into 9 provinces including a previously well recognized very distinctive SW USA province, region and realm. The 38 selected Australian terranes cluster into six provinces including a distinctive Gawler Province. A combined dendrogram of the 100 terranes from Laurentia, Australia and Antarctica resuhs in 8 superprovinces and 11 provinces. Five of the superprovinces contain both Laurentian and Australian terranes.

References

Burrett, C., and Berry, R., 2000 Proterozoic Australia-Western United States (AUSWUS) fit between Laurentia and Australia, Geology 28, 103-106. Burrett, C., Long, J., and Stait, B., 1990 Early to Middle Paleozoic biogeography of Asian terranes derived from Gondwana, Geological Society London Memoir 12, 163-174.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

THE STRUCTURAL AND KINEMATIC ANALYSES OF THE MESOPROTEROZOIC YANBIAN GROUP, SOUTH CHINA: IMPLICATIONS TO THE CATHAYSIA-YANGTZE COLLISION G.D. Carter Tectonics Special Research Centre, Department of Geology and Geophysics, The University of Western Australia, Crawley, Perth, WA 6009. Australia

INTRODUCTION

Folding styles in the Yanbian Group range from close to tight, whilst fold wavelengths vary fi-om local cm scale to km scale regionally. Sinian system folding, in contrast to that of the Yanbian Group, dominantly trends northwesterly and suggests a second deformational event (D2). Polyphase folding is observed in outcrop and through stereographic analysis of structural data from the Yanbian Group. Oblique Type 1 fold interference patterns is evident throughout the Yanbian Group as suggested by bidirectional (east and west) plunging of fold axes. Rarer northwest trending folds occur locally amongst the E-W folding in the Yanbian Group. Further evidence in support of a D2 event includes type 2 fold interference patterns observed in the overturned units of the Zhagu formation. Brittle and brittle-ductile faulting striking E-W throughout the Yanbian Group generally dips north or south, with most southerly dipping faults having a reverse sense of displacement. Rarer conjugate fault sets are interpreted as belonging to the N-S compressional regime associated with Di. Thrust faulting, as a result from progressive deformation associated with Di, manifests in localised overturned limbs, typically developed in less competent rock units where tight folding was developed.

Tectonic evolution of the South China Block (SCB) during the Proterozoic has been a controversial topic. The SCB comprises two Precambrian continental blocks, the Yangtze and the Cathaysia Blocks (Li, 1998). It is believed that the two blocks sutured during the late Mesoproterozoic to earhest Neoproterozoic (Li, 1998). Chinese geologists have long recognised the Mesoproterozoic orogenic belt in central-southern South China possibly related to such a collision, however, no systematic analysis of the structure and kinematics of such a belt have ever been carried out. The focus of this work is to study the Mesoproterozoic Yanbian Group in the south west of Sichuan, China. Kinematic and structural analyses, in conjunction with a stratigraphic analysis, suggest that the Yanbian Group may have evolved in a foreland basin and foreland fold and thrust belt setting.

STRATIGRAPHY

Four formations comprise the Yanbian Group, these being the Huangtian, Yumen, Xiaoping, and the Zhagu Formations. Unconformably overlying the Yanbian Group is the Neoproterozoic Sinian System formations. The Yanbian Group stratigraphy is a thick sequence dominated by finegrained siliciclastic sedimentary units, with the exception of the Huangtian Fm., which is a thick succession of volcanics and volcaniclastics interlayered with fine-grained siliciclastic sedimentary units. Stratigraphically, the sediments display distinct coarsening upward sequences. Compositional and textural immaturity of turbiditic sequences of poorly sorted sandstones lack structures associated with fluvial or shallow marine environments. Facies analysis supports deposition within a deep-water submarine fan environment.

DISCUSSION AND CONCLUSION

To assess the tectonic setting of the Yanbian Group and its structural development, its depositional environment, the structures and kinematics must all be considered. Facies analysis of the stratigraphic sequence suggests sediment deposition in a deep-water environment with evidence of some turbiditic activity. The dominance of volcanic lithic grains in sandstones suggests provenance from an uplifted volcanic terrain. Based on kinematic analysis and structural evidence throughout the study area the terrane displays geological features and attributes typically associated with foreland fold and thrust belts. With the SCB loosely being placed in convergent setting during the Mesoproterozoic, the evidence strongly suggests that deposition and structural evolution of

STRUCTURE

The Yanbian Group is characterised by east-west trending folds, which have been interpreted as belonging to a north-south contractional event (Di).

Geological Society of Australia Abstracts 65

Rodinia Symposium, Perth October 2001

the Yanbian Group took place in a foreland basin. Progressive evolution of the basin involved the fold and thrust belt cannibalising this succession. The collision and suturing of the Yangtze and Cathaysia blocks to form the SCB, as proposed by Li (1998), would provide a suitable tectonic basis for the

formation and development of a foreland basin as suggested here. References

Li, Z.X., 1998. Tectonic History of the Major East Asian Uthospherk Blocks Since the Mid- Proterozoic - A synthesis, Geodynamics, 27, p. 221 -243

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

TERRA AUSTRALIS OROGEN: RODINIAN BREAKUP AND SUBDUCTION INITIATION IN THE PACIFIC OCEAN P.A. Cawood^ and E. Leitch^ ^ Tectonics Special Research Centre, Department of Applied Geology, Curtin University, GPO Box U1987, Perth, WA 6001, Australia, p.cawood@info.curtin.edu.au ^ School of Environnnental Science, University of Technology, Sydney, PO Box 123, Broadway, NSW 2007, Australia. Evan.Leitch@uts.edu.au

The Pacific Ocean formed through Neoproterozoic rifting of Rodinia and was bounded by the conjugate Laurentian and Gondwanan margins. Despite a long history of plate convergence this ocean has never subsequently closed. The record of inception of convergent plate interaction in the Pacific is preserved in a Neoproterozoic to late Paleozoic orogenic belt here termed the Terra Australis Orogen. The orogen forms a fundamental crustal element along the Pacific margin of Gondwana and extends from the northeast coast of Australia south through Tasmania, New Zealand and the Trans Antarctic Mountains across southern Africa and into South America (Fig. 1). It has an along strike, pre-dispersal length of approximately 10,000 km and an across strike width of up to 1500 km. The orogen has traditionally been divided onto a series of separate structural units on the basis of the timing and nature of orogenic activity and the geographic disposition of units. It incorporates the Adelaide fold belt and its along strike equivalent the Ross fold belt, the Lachlan, Thompson and Tuhua fold belts and the New England fold belt, the Cape Basin of Southern Africa, and the Cordillera Frontal, Precordillera, and Pampeanas of South America. The formation, distribution and character of the Terra Australis Orogen are intimately linked to breakup and dispersal of Rodinia, and the subsequent assembly of Gondwana. Breakup of Rodinia in the mid-Neoproterozoic resulted in establishment of a rift and drift continental margin sequence along East Australia and East Antarctica following separation from its inferred conjugate margin in Laurentia and opening of the Pacific Ocean. Continued Neoproterozoic dispersal of Rodinian continental fragments through closure of the Mozambique Ocean and opening of the lapetus Ocean resulted in amalgamation of Gondwana and the propagation of the Terra Australis Orogen along the Pacific margin of Gondwana into South America. Subduction of Pacific Ocean lithosphere along and outboard of the Gondwana margin throughout the Paleozoic resulted in growth of the Terra Australis Orogen through magmatic addition and terrane accretion. The history of the Terra

Pacific H - A

1—divergent

plate margin

r '•'•.I

orogen

convergent plate margin

Figure 1. Schematic paleogeographic reconstructions at 760 Ma, 530 Ma and 260 Ma, showing position of Terra Australis Orogen and its superposition Gondwanide Orogen.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

on their relative isolation from the continental margin. The volcanic assemblage developed at or near the continent/ocean boundary and may have included crust of transitional affinities. The oceanmargin assemblage is linked to cratonic Gondwana by widespread siliciclastic detritus, but the intraoceanic elements show no evidence for a continental influence until the late Paleozoic. Ocean margin and intra-oceanic elements were progressively incorporated onto the Gondwana margin during Paleozoic orogenic movements (e.g. Tabberabberan, Tuhua) culminating in the Permian Hunter-Bowen event.

Australis Orogen terminated at about 300 Ma +/- 20 Ma associated with merging of Gondwana and Pangea and is represented by a pan-East Gondwana margin orogenic event (e.g. Hunter-Bowen orogeny). This marked a stepping out in the position of the plate boundary and commencement of the classic Late Paleozoic to Mesozoic Gondwanide Orogen. The Terra Australis orogen can be divided (Fig. 2) into a series of basement blocks of either continental or oceanic character that can be further subdivided on the basis of their geographic affinities (e.g. Laurentian vs Gondwanan) and proximity to inferred continental margin sequences (e.g. periGondwanan vs intra-oceanic). These divisions reflect initial tectonic setting and provide an insight into the character of the orogen through time. +

+ Cordillera-^

+ Pacific Ocean

+ -u

Relations within and between the various assemblages indicate that the end Neoproterozoic marks a fundamental change in tectonic style along the Terra Australis orogen. Available data indicate that subduction was established at or close to the Gondwana margin by 530 Ma but an earlier and perhaps ephemeral phase may have commented at around 570-590 Ma. The presence of magmatic arc rocks within the continental margin assemblage as well as the presence of siliciclastic strata in stratigraphic continuity with the oceanic substrate in the ocean margin sequences indicates that at least the main phase of convergent plate margin likely formed close to the Gondwana continent where relatively old, dense oceanic lithosphere was susceptible to subduction. Disruption and deformation of the continental margin successions and the initiation of subduction and convergent plate margin activity within the Terra Australis orogen between 590 and 530 Ma corresponds with a period of global plate reorganization. This involved opening of the lapetus ocean, final assembly of Gondwana through closure of the Mozambique ocean, and rifting of Siberia off northern Laurentia. The temporal equivalence of this series of end Neoproterozoic Wilson cycle stages suggests they may be interdependent and that their far-field effects could have influenced subduction initiation in the proto-Pacific. The margins of the Mozambique Ocean were aligned approximately orthogonal to the trend of the East Gondwana margin preserved in the Terra Australis orogen. With collision of East and West Gondwana along the Mozambique Belt, continued plate motion may have been transferred to the East Gondwana Pacific ocean margin resulting in strike slip deformation and margin decoupling. The initiation of subduction along the Pacific margin of Gondwana by 530 and possibly as early as -590-570 Ma argues against models which suggest breakup of Rodinia and generation of conjugate Laurentian and Gondwana margins did not occur until the end Neoproterozoic at around 560 Ma. The initiation of subduction in the late Neoproterozoic to Early Cambrian marks the

Laurentia

Appalachians

lapetus Ocean + + Laurentian + intra-oceanic _ continental terranes assemblage Id intra-oceanic" (/) 0) peri-Gondwanan - + peri-Gondwanan 5af) -oceanic assemblage" ^co|itinent^l terranes coo .0) •Gondwanan continental margin assemblage .

c o 5

o CD

c CD g •D C

West ^ East + Gondwana + Gondwana + (west Laurentia o ^ l (east Laurentia conjugate margin] §.§-vconjugate margin) +

o + + + + + + +/

\+ + + +

^ 1 + ++ + + +

Figure 2. Distribution of major lithotectonic units in Terra Australis Orogen with respected to internal units of the Gondwana craton and external units of the Pacific and lapetus oceans and Laurentia. Older rocks within the Australia-East Antarctic segment of the of the Terra Australis orogen are divisible, from west to east, into four main lithotectonic assemblages, a continental margin assemblage, a volcanic assemblage, an ocean margin assemblage and an intra-oceanic assemblage. The continental margin assemblage developed on continental crust stabilized within Rodinia, whereas the latter two developed in an oceanic realm formed following breakup of Rodinia and are differentiated

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

an example as the declining stage of the Wilson cycle of ocean basins, its protracted history of ongoing subduction, and by inference oceanic crust generation, contrasts with the clear evidence for opening and closing of oceans preserved in the lapetus/Atlantic and Tethyan realms.

inception of the Pacific 'ring of fire', yet throughout the Phanerozoic the Pacific has remained a major ocean basin. This indicates that the longevity of the Pacific and its antecedents is a resuU of continued production of oceanic hthosphere throughout the Phanerozoic, rather than a delayed onset of subduction. Although the Pacific has been cited as

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

STRUCTURAL, ISOTOPIC AND GEOCHEMICAL CONSTRAINTS ON THE EVOLUTION OF THE LEEUWIN COMPLEX, SOUTHWEST AUSTRALIA A.S. Collins and I.C.W. Fitzsimons Tectonics SRC, Department of Applied Geology, Curtin University of Technology, GPO Box U1987, Perth, WA 6845, Australia

STRUCTURE OF THE NORTHERN LEEUWIN COMPLEX

INTRODUCTION

The Leeuwin Complex is one of three basement inliers that decorate the western margin of Western AustraHa and collectively form the Pinjarra Orogen (Fig. 1; Myers 1990; Wilde and Murphy 1990; Wilde 1999). It is bounded to the west by the Indian Ocean and the east by the Dunsborough Fault, which separates crystalline Leeuwin basement from Phanerozoic sedimentary rocks of the Perth Basin (Fig. 2). The Leeuwin Complex is dominated by a variety of felsic orthogneiss units that lie at the heart of East Gondwana, a continental block comprising India, Australia, and Antarctica. East Gondwana was traditionally believed to have assembled at the close of the Mesoproterozoic, and gneisses of the Leeuwin Complex were thought to have been emplaced along a late Neoproterozoic intracontinental mobile belt developed within this late Mesoproterozoic landmass (Harris and Beeson 1993). However, geochronological data indicate that granitic protoliths of the Leeuwin Complex were emplaced throughout much of the Neoproterozoic (Nelson 1996, 1999), consistent with recent suggestions that East Gondwana was not a coherent entity until the end Neoproterozoic (Meert and van der Voo 1997; Fitzsimons 2000a,b). Fitzsimons (2000a) recognised that the late Neoproterozoic to Cambrian Pinjarra Orogen is part of a much larger orogenic belt that can be traced along the western margin of Australia into East Antarctica, and postulated that this orogenic belt (the Prydz-Denman-Darling Orogen) is a late Neoproterozoic suture zone between India and Australia that developed during the final assembly of Gondwana (Fitzsimons, this volume). We describe here a structural transect through the northern Leeuwin Complex and present new UPb SHRIMP age data that constrain the deformation history of these rocks. Our data are combined with a reinterpretation of previously published geochemical and geochronological data, to provide some new constraints on the evolution of this small but critical segment of Western Australia.

Banded tonalitic orthogneiss and foliated microgranite of the northern Leeuwin complex are

Figure 1. Part of Australia, India, and Antarctica in a Gondwana reconstruction, showing the inferred Neoproterozoic Prydz-Denman-Darling suture zone (dark grey) between older cratons (light grey). The Leeuwin Complex is located at the southwestern tip of Australia, and was presumably contiguous with rocks in Greater India that have since been subducted beneath Tibet. deformed into a 10 km-scale north-plunging antiform with an axial trace that crops out 500 m east of Cape Naturaliste, at the northwestern tip of the complex. This axial trace trends roughly north-

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

deformed and metamorphosed to granulite/upper amphibolite facies in one or more events sometime after their emplacement to produce rocks that are locally highly strained and compositionally banded. Nelson (1996) interpreted new zircon growth at 630-600 Ma as dating this metamorphism. Unlike the older gneisses, a younger suite of granitoids emplaced at 580-500 Ma are not compositionally banded, but are still foliated and folded.

south, and the northeast coast of the complex comprises parasitic S-type folds on the eastern limb of this regional structure. This folding deforms the compositional gneissic banding and microgranite foliation and is apparent in stereographic plots of the foliation data (Fig. 3). This earlier foliation is itself axial planar to isoclinally folded amphibolite and pegmatite dykes that crosscut the metagranite, consistent with two stages of folding in these rocks. Similar fold styles and geometries for these two events suggest that they formed during a single progressive deformation.

Sugarioaf Rock

Poles to foliation

702±7 M a

S H R I M P zircon sample site with inferred granite crystallization age

540±6 5 2 4 ± 1 2 M a *\

INDIAN OCEAN 1091 ± 8 M a

Leeuwin Complex Aegirine-augite syenite gneiss Massive granite gneiss Hbl-Bt Granite gneiss

Figure 3. Equal area stereographic projection of poles to foliation measured in the northern Leeuwin Complex (n = 740), illustrating the north-plunging folds that characterize this part of the complex.

779+23

Anorthosite Layered granite gneiss ^ G r t - B t granite gneiss

Ma 681±10Ma

We have obtained new SHRIMP U-Pb zircon data (Fig. 4) for two samples from Sugarioaf Rock in the northwest of the Leeuwin Complex (Fig. 2), where a foliated microgranite is folded into an antiform cored by banded tonalitic gneiss with abundant granitic melt veins. Both samples have a similar range of zircon ages between 750 and 520 Ma (Fig. 4), but we have used field relationships and textural features of the zircon grains under cathodoluminescence to postulate different histories for these two rocks. U-Pb zircon data from the tonalitic gneiss (sample SLl; Fig. 4) reveal a spread of concordant zircon cores stretching back in age to 730 Ma, which is taken as the emplacement age of this rock. These cores are resorbed and overgrown by younger zircon rims dated at 521 ±9 Ma. Zircons from the enveloping foliated microgranite (sample SL5; Fig. 4) have cores that cluster at 735 ±14 Ma (identical age to the oldest cores in SLl) and rims that range in age from 612 to 527 Ma. Based on local field evidence that the foliated microgranite intrudes through the foliation in the tonalitic gneiss, we interpret zircon cores in the microgranite as xenocrysts, and suggest that the microgranite was

10 k m 34°30'S ,114°58'30"E

Figure 2. Map of the Leeuwin Complex showing the approximate outcrop extent of the main lithologies distinguished by Wilde and Murphy, 1990). Also shown are the sampling sites for rocks dated by Nelson (1996, 1999) using SHRIMP U-Pb zircon techniques. Age data marked are the inferred protolith crystallization ages for each of the orthogneiss samples (Bt = biotite, Grt = garnet and Hbl = hornblende).

GEOCHRONOLOGY OF THE LEEUWIN COMPLEX U-Pb SHRIMP zircon data have identified three dominant periods of granitic magmatism in the Leeuwin Complex (Nelson, 1996, 1999). Late Mesoproterozoic garnet-bearing orthogneiss was emplaced at -1100 Ma, and a second extended period of mid-Neoproterozoic magmatism took place from 800 to 650 Ma. These rocks were

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

prolonged magmatic history. A reassessment of the geochemical data published by Wilde and Murphy (1990), Nelson (1995) and Wilde (1999) indicates that there are significant geochemical differences between some of the units identified by Wilde & Murphy (1990) and that these differences correlate well with the protolith age data (Fig. 5). Orthogneisses with protolith ages of 800-650 and 580-500 Ma, and rocks assumed to have protoliths of this age on the basis of similar mineralogy and fabric, have the high Zr contents and high Ga/Al ratios typical of A-type magmas, and plot in the A-type field on the discrimination diagram of Whalen et al. (1987). This is in contrast to gamet-biotite orthogneiss emplaced at 1100 Ma, which does not have A-type characteristics (Fig. 5a). Similarly, the 1100 Ma protoliths plot within the syn-collisional field on the RbAfb+Ta granite discrimination diagram of Pearce et al. (1984), whereas the 800-650 Ma units cluster in the withinplate field and the 580-500 Ma units show a wide scatter across several fields (Fig. 4b).

derived at least in part by partial melting of the tonalite. The spread of rim ages in the microgranite rock is thought to reflect a mixture of new zircon growth and partial recrystallization of the xenocrysts during intrusion of the microgranite, with the youngest concordant rim age of 527 ±10 Ma taken as the best estimate for age of intrusion. This latter age is identical to the age of zircon rims in the tonalitic gneiss, which are though to have grown during emplacement of the microgranite. data-point error ellipses are 68.3% conf

SL5

850

Foliated microgranite: Sugarloaf

^Pb corrected data 735+/-14Ma 8 grains, MSWD=1.71 612-538 Ma

Ages quoted as 206p,^238pjj gggg

527+/-10Ma 50 . 0.9

1.1

A-type granites

1.3

data-point error ellipses are 68.3% conf.

SL1

;

Tonalitic gneiss: Core of Sugarloaf antiform

850

^Pb corrected data

(

Discrimination diagram for A-type granites (Whalen et al. 1987)

1.0

1000Ga/AI

550-525 Ma foliated granite and syenogranite and similar undated rocks 0.9 207

Pbl

1.1 ^ ^

235

Figure 4. U-Pb SHRIMP zircon data for two samples from Sugarloaf Rock, northeastern Leeuwin Complex. Data are ^^"^Pb corrected and ages are presented as ^^^Pb/^^^Pb ages. Field relationships show that sample SL5 intrudes through SLl.

ocean ridge granite

800-680 Ma orthogneiss and similar undated layered Hbl gneiss 1100 Ma orthogneiss and similar undated layered Grt-Bt gniess

Granite discrimination diagram (Pearce et al. 1984) 1

GEOCHEMISTRY OF GRANITOIDS FROM THE LEEUWIN COMPLEX

10 Yb+Ta (ppm)

Figure 5. A reassessment of the geochemical data published by Wilde and Murphy (1990), Nelson (1995) and Wilde (1990X illustrating good correlations between geochemistry and protolith age. Minerals abbreviations as in Fig. 2. Note that each of the three data groups includes samples that have been directly dated by SHRIMP techniques (Nelson 1996, 1999) and samples whose age we have inferred on the basis of mineralogy and fabric.

A number of different units of granitic orthogneiss have been identified in the Leeuwin Complex, based on primarily on their mineralogy (Fig. 2; Wilde and Murphy, 1990). Previous studies have interpreted all of these units as being derived from a single magmatic suite on the basis of geochemical variation diagrams (Wilde and Murphy, 1990; Wilde, 1999). It is, however, hard to reconcile this model with the geochronological evidence for a

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

References

EVOLUTION OF THE LEEUWIN COMPLEX

Fitzsimons, I.C.W., 2000a. Grenville-age basement provinces in East Antarctica: evidence for three separate collisional orogens. Geology, 28: 879-882. Fitzsimons, I.C.W., 2000b. A review of tectonic events in the East Antarctic Shield, and their implications for Gondwana and earlier supercontinents. Journal of African Earth Sciences, 31,3-23. Fitzsimons, I.C.W., this volume. The Neoproterozoic evolution of Australia's western margin, Rodinia symposium, Perth. Harris, L.B. and Beeson, J., 1993. Gondwanaland significance of Lower Palaeozoic deformation in central India and SW Western Australia. Journal of the Geological Society of London, 150: 811-814. Meert, J.G. and van der Voo, R., 1997. The assembly of Gondwana 800-550 Ma. Journal of Geodynamics, 23: 223235. Myers, J.S., 1990. Pinjarra Orogen, Geology and Mineral Resources of Western Australia. Western Australia Geological Survey, Memoir 3: pp. 265-274. Nelson, D.R., 1995. Field Guide to the Leeuwin Complex, ACOG 3 Conference, Perth, 24 pp. Nelson, D.R., 1996. Compilation of SHRIMP U-Pb zircon geochronology data, 1995. Western Australia Geological Survey Record 1996/5. Nelson, D.R., 1999. Compilation of SHRIMP U-Pb zircon geochronology data, 1998. Western Australia Geological Survey Record 1999/2. Pearce, J.A., Harris, N.B.W. and Tindle, A.G., 1984. Trace element discrimination diagrams for the interpretation of granitic rocks. Journal of Petrology, 45: 956-983. Whalen, J.B., Currie, K.L. and Chappell, B.W., 1987. A-type granites: geochemical characteristics, discrimination and petrogenesis. Contributions to Mineralogy and Petrology, 95:407^19. Wilde, S.A., 1999. Evolution of the western margin of Australia during the Rodinian and Gondwanan supercontinent cycles. Gondwana Research, 2: 481-499. Wilde, S.A. and Murphy, D.M.K., 1990. The nature and origin of the Late Proterozoic high-grade gneisses of the Leeuwin Block, Western Australia. Precambrian Research, 47: 251270.

Our reassessment of previously published lithological, geochronological and geochemical data, coupled with our new structural and isotopic data from the area around Sugarloaf Rock, allow us to make a number of conclusions regarding the evolution of the Leeuwin Complex. (1) There is a broad correlation between age and degree of deformation, with the older rocks preserving higher strain. In particular, the late Mesoproterozoic orthogneiss is highly deformed, with local evidence of considerable non-coaxial strain. Conversely, although the 580-520 Ma plutons are deformed, they record much less strain than the older rocks. (2) Late Mesoproterozoic orthogneiss has a trace element geochemistry consistent with formation during collisional orogenesis, whereas younger 800650 Ma A-type granitoids most likely developed in a within plate setting. Geochemical characteristics of the youngest pulse of magmatism at 580-500 Ma do not indicate a clear tectonic setting. This youngest event produced a number of rock compositions including A-type alkaline granite and syenite that have been taken previously as evidence of a rift setting (Wilde and Murphy, 1990). (3) At Sugarloaf Rock there is evidence that at least some of the younger 580-520 Ma intrusions were derived by partial melting of the midNeoproterozoic 800-650 Ma intrusions. (4) The last stage of pervasive foliation and upper-amphibolite metamorphism at Sugarloaf Rock must have developed after the intrusion of microgranite sheets at 530 Ma, as did the younger N-S oriented fold hinges that dominate the regional structure of the northern Leeuwin Complex.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

THE STRUCTURE, GEOCHEMISTRY AND GEOCHRONOLOGY OF EASTERN MADAGASCAR: THE REMAINS OF A STRAND OF THE MOZAMBIQUE OCEAN A . S . C o l l i n s \ L C . W . F i t z s i m o n s \ T . S . B r e w e r ^ , T . R a z a k a m a n a n a ^ B . Hulscher"^ a n d B.F. W i n d l e y ^ ^ TSRC, Dept of Applied Geology, Curtin University, GPO Box U1987, Perth WA 6845, Australia. ^Department of Geology, University of Leicester, University Road, Leicester, U.K. ^ Departement des Sciences de la Terre, Universite de Toliara, Toliara, Madagascar "^TSRC, Department of Geology & Geophysics, University of Western Australia, Crawley WA 6009, Australia.

not part of the EAO. Southem Madagascar is isolated from these tectonic units by the cmstal-scale Ranotsara shear zone. Many similarities exist between the Antongil block of eastemmost Madagascar and the Dharwar craton in India (Peucat et al. 1993; Tucker et al. 1999; Chadwick et al. 2000; Jayananda et al. 2000). These include: 1) ortho- and paragneiss dating back to over 3.0 Ga (Peninsular Gneiss and Sargur Group of India); 2) granitoid intmsion at 2.51-2.55 Ga (the Dharwar batholith); 3) preservation of Archaean Rb/Sr wholerock ages. In contrast, the Antananarivo Block and Itremo Sheet of central Madagascar contain no known rocks older than 2.6 Ga (Tucker et al. 1999; Kroner et al. 2000) and metasediments of the Itremo sheet preserve detrital zircons more consistent with derivation from Africa than India (Cox et al. 1998). Using a tight-fit Gondwana reconstmction, a belt of late Archaean to early Proterozoic rocks can be identified passing north from the Antananarivo block through Somalia and east Ethiopia to the Yemen and Saudi Arabia. In northem Somalia, the Yemen and Saudi Arabia this ancient terrane is bound at either side by juvenile Neoproterozoic rocks (Fig. 1). These correlations imply that the boundary zone between the Antananarivo block and the Antongil block separates continental cmst with considerably different histories, with no evidence of 3.2 Ga cmst in the Antananarivo block, and no evidence in the Antongil block of the of 825-740 Ma or 630-530 Ma magmatism that is common in the Antananarivo block (Tucker et al. 1999; Handke et al. 1999; Kroner et al. 2000). This boundary zone is today a highly strained paragneiss belt with emerald mineralisation and entrained podiform ultramaficmafic bodies. Deformation along this zone is bracketed between 630 and 527 Ma.

The East African Orogen (EAO; Stem 1994) is a belt of deformed and metamorphosed rock that stretches from the Middle East, through Arabia, East Africa, Madagascar and southem India to East Antarctica (Fig. 1). It has been regarded as the principal collision zone developed during the late Neoproterozoic assembly of Gondwana (Shackleton 1986, 1996; Hoffman 1991; Powell et al. 1993) and although new data have identified sutures of similar age elsewhere in Gondwana (Fitzsimons 2000) there is little doubt that the EAO records the closure of a major Neoproterozoic ocean, commonly referred to as the Mozambique ocean. There is no conclusive evidence for continental collision in the northern EAO. Rocks deformed and metamorphosed during Neoproterozoic times extend to the eastern edge of the Arabian-Nubian Shield, where they crop out as basement to the Oman ophiolite (Gass et al. 1990) and on the island of Socotra (Beydoun and Bichan 1970), but there is no trace of an eastern colliding continent. This could indicate that the ArabianNubian Shield formed solely by accretionary processes and lay open to a northem Mozambique Ocean throughout the Neoproterozoic. Madagascar is the most northem part of the EAO where there is clear evidence for a colliding continent at the eastern margin of the orogen. The Dharwar craton of India was unaffected by pervasive Neoproterozoic deformation or high-grade metamorphism and lies immediately east of Madagascar in Gondwana reconstructions. Isotopic data indicate that part of this craton is exposed at the extreme eastern margin of Madagascar (Fig. 1). Collins et al. (2000) have split the EAO of central and northem Madagascar into 4 tectonic units (Fig. 2; also see Collins 2000; Kroner et al. 2000). These units are: 1) the Antananarivo block; 2) the Tsaratanana sheet; 3) the Itremo sheet; and 4) the Bemarivo belt. A fifth unit, the Antongil block; is regarded as part of the Dharwar craton and thus

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Neoproterozoic suture zones Shear zones/thrusts Margin of East African Orogen (Limit of 700-500 Ma upr-annphibolite grade metamorphism and deformation)

Region affected by 700-500 Ma upr-amphibolite grade or higher metamorphism Pre~1000 Ma crust in the EAO Present day water - either sea, or lakes

Figure 1. The relationships between Madagascar and surrounding regions of Gondwana. Tight-fit Gondwana reconstruction after Lawver et al. (1998). Ab = Abdulkadir terrane. A = Achankovil shear zone. A-B = Al-Bayda terrane. A-M = Al-Mahfid terrane. An = Antananarivo. ASZ = Aswa shear zone. DML = Dronning Maud Land. H = Highland Complex. I-A = Inda Ad complex. If = Ifanadiana shear zone. KK = Karur-Kambam-PainavuTrichur shear zone. L-H = Liitzow-Holm complex. M&Q = Mora and Qabri Bahar terrane. NC = Napier complex. P-C = Palghat-Cauvery shear zone system. R = Ranotsara shear zone. RC = Rayner complex. T = Tranomaro shear zone. V = Vohibory belt. W = Wanni complex. Y-B = Yamato-Belgica complex. Betsimisaraka = Betsimisaraka suture.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Shear zone Thrust Extensional ^ ^ shear zone

Bemarivo belt Itremo sheet Tsaratanana sheet Antananarivo block with structural trend Betsimisaraka suture Antongil block southern Madagascar (not subdivided)

Figure 2. The tectonic units of central and northern Madagascar (after CoUins 2000; note that southern Madagascar is not subdivided), a = Antananarivo, t = Toamasina, B = Betsileo shear zone, If = Ifanadiana shear zone, R = Ranotsara shear zone. Shear zones south of the Ranotsara shear zone (after Windley et al. 1994; Martelat et al. 2000; de Wit et al. 2001): A = Ampanihy shear zone; Be = Betroka shear zone; Ts = Tranomaro shear zone. In a traverse through eastern Madagascar, directly west of the Betsimisaraka suture, four structural domains are picked out in the eastern Antananarivo block and the Beforona unit of the Tsaratanana sheet. All four domains show evidence of east-west contraction and one of these domains is a >10 km wide top-to-the-east mylonite zone, which is interpreted as a ductile thrust that has displaced central Madagascar eastwards towards the Antongil foreland. Correlation of structures between the four domains has identified four deformation events. These are: 1) a D1 gneissic foliation preserved sporadically in low-strain areas; 2) north-south plunging upright D2 folds associated with extension parallel to fold hinges and local north-south D2

Detrital zircons from metasedimentary rocks within this boundary zone date back to 3.26 Ga, suggesting that at least a component of the protolith sediment was eroded from the Dharwar craton. Detrital age populations occur in these rocks at 3.13.5 Ga, 2.5-2.7 Ga and 700-900 Ma. The youngest detrital zircon core (>95% concordant) has a 206Pb/238U age of 713 ±5 Ma and provides a maximum possible age for protolith deposition. Available structural, geochronological and lithological data suggest that this boundary is a Neoproterozoic suture zone, which we name the Betsimisaraka suture after the region in which it crops out.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

References

high-strain zones developed by coaxial flow during east-west shortening; 3) cm- to km-scale top-to-theeast D3 mylonite zones developed by non-coaxial flow during east-west shortening; and 4) late open D4 folds. D2 and D3 produced the dominant pervasive structures on outcrop and regional scales, and most likely developed at 550-515 Ma. Neoproterozoic granitoids that are found along the traverse described above are chemically divided into two groups, one linked to active subduction processes (Group 1), and the other a product of infracrustal melting (Group 2). Some of the Group 1 rocks have high Sr/Y and La/Yb ratios typical of Cenozoic adakites, suggesting that slab melting was an important process in this arc. In contrast, the Group 2 rocks have peraluminous A-type and within-plate signatures, suggesting that these were products of cmstal melting probably triggered by the underplating and intrusion of mantle-derived melts within the continental arc. We present a model where the eastern Antananarivo block is interpreted as the leading edge of a microcontinental block separated from the Dharwar craton by a strand of the Mozambique Ocean that was being subducted beneath central Madagascar between 825 and 740 Ma. This ocean closed after 713 Ma as detrital zircon grains of this age are found within sediments of the Betsimisaraka suture zone. The Tsaratanana sheet structurally overlies the Antananarivo block and shares a history of relatively late east-west contraction with the Antananarivo block. Deformation continued until after intrusion of the 532 Ma (Meert et al. in press) Carion granite.

Beydoun Z.R. & Bichan H.R., 1970. Quart. J. Geol. Soc. Lond. 125,413-446. Chadwick B., Vasudev V.N. & Hegde G.V., 2000. Precambrian Res. 99,91-111. Collins A.S., 2000. Gondwana Res. 3, 549-552. Collins A.S., Kroner A., Razakamanana T. & Windley B.F., 2000a. J. African Earth Sci. 30 (4A), 21. Cox R., Armstrong R.A. & Ashwal L.D., 1998. J. Geol. Soc. Lond. 155, 1009-1024. de Wit M.J., Bowring S.A., Ashwal L.D., Randrianasolo L.G., Morel V.P.I. & Rambeloson R.A., 2001. Tectonics 20, 1-45. Fitzsimons I.C.W., 2000. Geology 28, 879-882. Gass I.G., Ries A.C., Shackleton R.M. & Smewing J.D., 1990. In: Robertson A.H.F., Searle M.P. & Ries, A.C. (Eds.), The Geology and Tectonics of the Oman Region. Geol. Soc. Lond. Spec. Publ. 49, 585-599. Handke M., Tucker R. & Ashwal L.D., 1999. Geology 27, 351354. Hoffman P.F., 1991. Science 252, 1409-1412. Jayananda M., Moyen J.-F., Martin H., Peucat J.-J., Auvray B. & Mahabaleswar B., 2000. Precambrian Res. 99, 225-254. Kroner A., Hegner E., Collins A.S., Windley B.F., Brewer T.S., Razakamanana T. & Pidgeon R.T., 2000. Amer. J. Sci. 300, 251-288. Lawver L.A., Gahagan L.M. & Dalziel I.W.D., 1998. Mem. Nat. Inst. Polar Res. Tokyo 53, 214-229. Martelat J.-E., Lardeaux J.-M., Nicollet C. & Rakotondrazafy R., 2000. Precambrian Res. 102, 1-20. Meert J.G., Nedelec A., Hall C., Wingate M.T.D. & Rakotondrazafy M., in press, Tectonophysics. Peucat J.J., Mahabaleshwar B. & Jayananda M., 1993. J. Metamorphic Geol. 11, 879-888. Powell C.M., Li Z.X., McElhinny M.W., Meert J.G. & Park J.K., 1993. Geology 21,889-892. Shackleton R.M., 1986. In: Coward M.P. & Ries A.C. (Eds.) Collision Tectonics. Geol. Soc. Lond. Spec. Publ. 19, pp. 329-349. Shackleton R.M., 1996. J. African Earth Sci. 23, 271-287. Stem, R.J. 1994. Ann. Rev. Earth Planet. Sci. 22, 319-351. Tucker R.D., Ashwal L.D., Handke M.J., Hamilton M.A., Le Grange M. & Rambeloson R.A., 1999. J. Geol. 107, 135153. Windley B.F., Razafmiparany A., Razakamanana T. & Ackermand D., 1994. Geol. Rundschau 83, 642-659.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

THE BIRTH, LIFE AND DEATH OF THE MOZAMBIQUE OCEAN H. Cutten Tectonics Special Research Centre, University of Western Australia, 35 Stirling Highway, Crawley WA 6009, Australia

INTRODUCTION

surrounded Rodinia, as represented by (Unrug 1997). Determining which of these altematives, requires accurate dating of any rift-related volcanics or at best, constraints on the deposition of the passive margin sequence. Evidence for rifting is scarce and certainly not unequivocal. Bimodal volcanics are an essential part of a rift succession as are coarse clastic units. In Kenya, (Mosley 1993) noted the lack of preserved or recognised bimodal volcanics but cites unpublished (M.G.D.) reports that apparently show that metamorphosed acid volcanics may be a more common component of the quartzo-feldspathic gneisses than previously recognised. These possible rift-related volcanics have been tectonically imbricated with passive margin sediments and reworked cratonic material in the mid crust during continental collision. Low grade coarse clastic volcano-sedimentary sequences (the Embu and Ablun formations) occur as high-level thrust sheets over the passive margin sequence, and may be part of the rift-related sedimentary sequence. However, there are no age constraints on any of these units. In central Kenya the presence of ubiquitous, variably metamorphosed and deformed, quartzites, schists and marbles have been interpreted by (Mosley 1993) and (Key 1989) to represent a passive margin sequence. (Mosley 1993) correlated some of these lithologies (the Mara-Loita quartzites) with the less metamorphosed Kisii Group, preserved, unconformably on the Tanzanian craton. The Kisii Group contains abundant bimodal volcanics with Rb-Sr ages of 906±35 and 964±35 Ma (Briden 1971) which would appear to represent pre-Mozambique Ocean rifting. However, Pinna et al. (2000) determined single zircon Pb-Pb evaporation ages of 2531±3 for the Ikonge Ignimbrite indicating that the Kisii Group is in fact a Neoarchaean rift-related sequence. (Mosley 1993) considered that the passive margin sediments are bracketed in age between -1200 Ma and -800 Ma based on a Rb-Sr age of 1206 ± 96 Ma for the Mukugodo migmatite which forms basement underlying the sediments and 818 ± 48 Ma, the Rb-Sr age for metamorphism of the tectonically intercalated Kotim Gniess, a para-gneiss within the Mozambique Beh in Kenya. This is in

The Proterozoic Mozambique Belt is a northsouth trending orogenic beU exposed on the eastern flank of the Tanzanian (Congo) Craton and extends from northern Kenya, Uganda and Ethiopia, southwards to Mozambique. Its contains elements relating to the opening of the Mozambique Ocean including passive margin sediments and possible evidence of the original rifting. It also represents the closure suture (or one of them) of the Mozambique Ocean (interpreted by many authors as the collision of sub-supercontinents, East and West Gondwanaland). The Mozambique Belt appears to represent a crustal section resulting from collision and varying uplift with supracrustal rocks including para-gneisses and ophiolites in the north (Kenya and the Arabian Nubian Shield) and mid to lower crustal rocks including ortho-gneisses and granulites in the south (Tanzania and Mozambique). The transition appears gradational, located between southern Kenya and northern Tanzania. In Kenya the supracrustal rocks include para-gneisses, quartzites, schists, amphibolites, migmatites and metacalcareous rocks with felsic, mafic and ultramafic intrusions and post kinematic granites and pegmatites. These supracrustal rocks are largely amphibolite grade (kyanite and sillimanite). In Tanzania, in the western part of the Mozambique Belt mid-crustal ortho-gniesses predominate with extensive reworking of the underlying Archean basement. Lower crustal rocks, granulites, anorthosite and high grade gneisses make up the 'Eastem Granulites' extensively exposed in eastern Tanzania but extending further west. These are klippe remnants with a basal thrust retrogressively metamorphosed to amphibolite facies. This study aims to unify observations about the evolution of the Mozambique Ocean, across political boundaries, and to provide a synthesis of tectonometamorphism related to the opening, life and final closure of the Mozambique Ocean.

RIFTING AND DEVELOPMENT OF THE PASSIVE MARGIN

Inception of the Mozambique Ocean may have occurred with breakup of Rodinia or it may have formed earlier as part of the Mirovoi Ocean which 24

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

CONCLUSIONS

conflict with data from (Muhongo 2001) who dated detrital zircons (Pb-Pb) from granulites in the Uluguru Mountains (700-745 Ma) and eastern Usambara Mountains (737-780 Ma). Assuming these are part of the same depositional basin as the Kenyan para-gneisses they provide the oldest possible age for the passive margin sequence since Rb-Sr ages are rather unreliable and prone to isotopic resetting during later metamorphism. Locating and dating of rift-related volcanics could provide evidence of the age of initial rifting that ultimately resulted in the opening of the Mozambique Ocean and thus whether it coincided with the breakup of Rodinia or was part of an older rifting event within the Congo craton. The identity of the rifted eastern continent is unknown. A hint of its identity might be provided by the comparison of detrital zircon ages and morphologies within the passive margin sequence of the Mozambique Belt with that of potential cratons.

The scenario proposed for the life of the Mozambique Ocean begins with rifting of an unknown craton from the eastern margin of the Tanzanian (Congo) Craton. The Congo Craton at this time may have been part of Rodinia or may have existed as an independent craton. Bimodal volcanics associated with this rifting have not been identified; however, intercalated acid volcanics and coarse clastic units exposed in Kenya may be part of the rift sequence. Deposition of the developing passive margin sequence may have commence post -780 Ma. Contraction of the ocean may have occurred, in part, as an east dipping subduction zone, although neither the arc volcanics nor their roots have been identified. Some Congo passive margin sediments may have been subducted to lower crustal levels. Thermally induced granulite facies metamorphism occurred at 640 Ma. Initial collision of the Congo Craton and 'East Gondwanaland' was immediately followed by west directed thrusting of the passive margin sequences (granulite and amphibolite grade). The original east-dipping subduction zone was reactivated as a west-verging thrust system and parts of the Congo Craton basement were effectively transferred from the footwall to the hanging wall. This exhumed the complex array of west verging, thrusted, and recumbently folded lithologies now observed in the Mozambique Belt. Exhumed lithologies reached upper crustal levels about 500 Ma. At most the Mozambique ocean existed for 280 my, but likely at least over 140 my. Continental collision occurred at the southern end of the Mozambique Belt exhuming middle and lower crustal rocks while in the north, oceanic and supracrustal elements are preserved. Recent evidence suggest Gondwanaland amalgamated from a series of independent cratons rather than a simple collision between East and West Gondwanaland (Kroner 2001).

METAMORPHISM AND CONTINENT COLLISION The timing of Tan African' collision of East and West Gondwanaland has traditionally been considered to be the age of peak granulite facies metamorphism in the granulite complexes of eastern Tanzania. However, Appel et al. (1998) derived an anti-clockwise PTt path for these granulites, indicating prograde heating at moderate pressures followed by a near-isobaric retrograde cooling path and as such is more consistent with intrusion and magmatic underplating of the crust, than with continental collision. Peak PT's for the granulite facies metamorphism are calculated at 810 ± 40° C and 9.5 to 11 kbar and has been dated by Muhongo et al. (2001) from metamorphic zircons within the Wami River (641.2 ± 0.9 Ma), Usambara Mountains (641.4 ± 0.9 Ma) and Uluguru Mountains (638±1.0 Ma). This consistent 640 Ma age is also confirmed by Moller et al., (2000) from U-Pb, monazite ages from the Pare Mountains. The near isobaric cooling portion of the PT path is dated by the U-Pb closure temperature of titanite of 650°C at 619 ± 2 Ma while the later portion of the decompression and uplift part of the path is dated at 500 ± 5 Ma from the 430 ± 30' C closure temperature of rutile. It is possible that this is a composite PTt path representing granulite facies, magmatic underplating, static residence in the mid - lower crust, and subsequent cooling between C.640 and c. 620 Ma followed by the exhumation of these rocks, presumably during continental collision, prior to c.500 Ma.

References Appel, P., A. Moeller, et al. (1998). "High-pressure granulite facies metamorphism in the Pan-African Belt of eastern Tanzania; P-T-t evidence against granulite formation by continent collision." Journal of Metamorphic Geology 16(4): 491-509. Briden, J. C., Piper, J.D.A., Henthom, D.I., Rex, D.C. (1971). "New Paleomagnetic results from Africa and related potassium-argon age determinations." 15th Annu. Rep. Res. Inst. Afr. Geol.. Univ. of Leeds.: pp46-50. Key, R. M., Charsley, T.J., Hackman, B.D.,Wilkinson, A.F., Rundle, C.C., (1989). "Superimposed Upper Proterozoic collision-controlled orogenies in the Mozambique orogenic belt of Kenya." Precambrian Research 44: 197-225. Kroner, A., Kehelpannala, W., Collins, A.S., Muhongo, S. (2001). "Are all 1000-1100 Ma rocks related to formation of the supercontinent Rodinia ? The view from Sri Lanka, Madagascar and Southeastern Africa." Abstract HUG XI Strasburg. Moller, A., Mezger, K., Schenk, V (2000). "U-Pb dating of metamorphic minerals: Pan-African metamorphism and

Geological Society of Australia Abstracts 65

Rodinia Symposium, Perth October 2001

prolonged slow cooling of high pressure granulites in Tanzania. East Africa." Precambrian Research 104: 123-146. Mosley, P. N. (1993). "Geological evolution of the late Proterozoic "Mozambique Belt" of Kenya." Tectonophysics 221(2): 223-250. Muhongo, S., Kroner,A., Nemchin, A.A. (2001). "Single zircon evaporation and SHRIMP ages for granulite-facies rocks in the Mozambique Belt of Tanzania." Journal of Geology 109(2).

Pinna, P., A. Cocherie, et al. (2000). "The Kisii Group of western Kenya; an end-Archaean (2.53 Ga) late orogenic volcano sedimentary sequence." Journal of African Earth Sciences and the Middle East 30(1): 79-97. Unrug, R. (1997). "Rodinia to Gondwana; the geodynamic map of Gondwana supercontinent assembly." GSA Today 7(1): 1-6.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

AMAZONIAN PROTEROZOIC POLES: IMPLICATIONS TO RODINIA PALEOGEOGRAPHY M.S. D'Agrella-Filho\ I.I.G. Pacca\ R. Siqueira\ S.-A. Elming^ W. Teixeira^ J.S. Bettencourt^ and M.C. Geraldes^ ^ Institute of Astronomy, Geophysics and Atmospheric Sciences, University of Sao Paulo, Sao Paulo, Brazil. ^Department of Applied Geophysics, Lulea University of Technology, Lulea, Sweden. ^ Institute of Geosciences, University of Sao Paulo, Sao Paulo, Brazil.

magmatism and sedimentary deposition in a continental margin environment (sediments from the Sunsas and Vibosi groups in Bolivia and the Aguapei group in Brazil) represent this distensive phase. This basin was closed during the development of the Sunsas/Aguapei Orogenic Belt. During the same orogenic episode the Nova Brasilandia volcano-plutonic sedimentary sequence evolved with syn-tectonic magmatism. The 1.11±0.01 Ga age for a metamorphosed (amphibolite facies) subophitic metagabbro is based on four zircon fractions, and was interpreted as the age of the metamorphism that affected the sequence (Rizzoto, 1999). Samples were collected from 10 levels of the undeformed Fortuna Formation of the Aguapei Group, and from a basic dyke cutting these sediments, close to the town of Rio Branco (Mato Grosso State). In regions close to Alta Floresta d'Oeste and Colorado d'Oeste, samples from 19 metabasic dykes of the Nova Brasilandia Group were also collected. Rapakivi rocks in the Brazilian States of Rondonia and Mato Grosso represent the Serra da Providencia Intrusive Suite, which comprises the Serra da Providencia granite. Gabbros, chamockites and mangerites form parts of the suite. U/Pb geochronological data constrain the age of this batholith to the interval between 1.606±0.024 Ga (the age of the oldest dated granitoid) and 1532±0.024 Ga. Our sampling was concentrated in mafic rocks found in the Serra da Providencia area. Samples from 19 sites were collected close to Cacoal, Presidente Medici, Ouro Preto d'Oeste and Ariquemes. The Sunsas Orogeny affected the northern region of Rondonia (rocks that belong to the Rio NegroJuruena and Rondonian-San Ignacio Provinces) during the period 1.15-0.97 Ga (Tassinari et al., 2000). This resulted in metamorphic overprinting and deformation at 1.156 to 1.1 Ga and the emplacement of Rapakivi granite intmsives, mafic dykes and granitic plutons from 1.08 to 0.97 Ga. Biotite and amphibole from a chamockite from Ouro Preto (our sampling site R037) yielded "^^Ar/^^Ai ages of ca. 1.10 Ga, which is an evidence of this

Geological and geochronological (and a few paleomagnetic) evidences have been provided for the possible link between the Amazonian Craton and southern Laurentia (Rodinia Supercontinent), although some small variations among the proposed models are quite evident (e.g., Sadowski and Bettencourt, 1996, Bettencourt et a l , 1996, D'AgrellaFilho et a l , 1998, Dalziel, 1997). This link possibly existed up to late Neoproterozoic/ early Paleozoic, when the ephemeral Pannotia Supercontinent was formed (-580-570 Ma), and the lapetus Ocean began to open (McCausland and Hodych, 1998). Unfortunately, Meso-Neoproterozoic to early Paleozoic paleomagnetic data for the Amazonian Craton are very sparse, and therefore no paleomagnetic basis can be provided for a test of these models. Trying to rapidly improve the Amazonian MesoNeoproterozoic paleomagnetic database, a joint research project of geochronological and paleomagnetic studies has been developed by the Institute of Astronomy, Geophysics and Atmospheric Sciences and the Institute of Geosciences (University of Sao Paulo, Brazil), with the collaboration of the Depart-ment of Applied Geophysics (Lulea University, Sweden). The western part of the Amazonian Craton is a multi-orogen region formed between 1.8 and 1.0 Ga where successive magmatism, metamorphism and deformation events regionally affected and reworked previous provinces, producing new complexes, as well as new juvenile continental crust (Fig.l). Three major geochronological and tectonic provinces have been identified in the western part of the craton: the Rio Negro-Juruena Province (1.8-1.55 Ga), the Rondonian-San Ignacio Province, (1.5-1.3 Ga) and the Sunsas Province (1.3-1.0 Ga). Our main purpose was to collect samples related to the evolution of the Sunsas Province, the youngest tectonic unit of the Amazonian Craton. According to Tassinari et al. (2000), the start of the Sunsas tectonic evolution, which has been chrono-correlated with the Grenville Orogenic Cycle (1.3 to 1.0 Ga) in Laurentia and Baltica, was marked by an important phase of continental distension (rifting). Basaltic

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

path, suggesting that they represent the effect of a very slow cooling of the sampled area. Some of these VGPs fall on (or near to) the 1.1 Ga part of the APW path for Laurentia. Component A yielded poles AS and AN for directions with positive and negative inclinations, respectively. This component was disclosed for several samples from the northern studied area, which was affected by the Sunsas event. Several "^Âr/^Âr datings yielded ages around 1.15 Ga (see above), which suggest the possible age for AS and AN poles. These poles are located close to the NF pole obtained for the Nova Floresta Formation, dated at about 1.0 Ga by K-Ar method (Tohver et al., 2001). However, since K-Ar ages are usually affected by Ar excess (or loss), they may not indicate the age of magnetization. Therefore, "^Âr/^Âr determinations are needed to better constrain the age of this volcanism. Components B and C yielded poles BS, BN and CN, that fall approximately along the APW path for Laurentia (between 1.1 and 0.82 Ga) apparently suggesting younger ages for these components. Our paleomagnetic results, although very preliminary, do not deny a possible link between the Amazonian Craton and Laurentia, at about 1.1 Ga (Fig. 3). However, a more trustworthy interpretation awaits new ^ÂY/^ÂT determinations.

metamorphic event. Another geochronological evidence is presented by Bettencourt et al. (1996) from rocks collected close to the town of Ariquemes (sites R012 and ROB in their article; site R012 is very close to our sites R038, R039 and R042). Their rocks (Augen gneisses and granitic gneisses) yielded 40Ar/39Ar ages on hornblendes of 1156±36 Ma (R012) and 1149±35 Ma (R013) and on biotites of 1001±33 Ma (R012) and 912±30 Ma (R013), respectively. These data suggest a Sunsas orogen related metamorphic overprint. The best paleomagnetic results are from the samples of the Aguapei Group and the dyke sampled in the Mato Grosso State, where northern (southern) directions with moderate to steep negative (positive) inclinations were isolated, after alternating field (AF) and thermal demagnetization (Fig. 2a). Magnetization is carried by magnetite in the dyke samples and by both magnetite and hematite in the sediments. The paleomagnetic results from the Rondonia area are more complicated, probably due to a more complex geological evolution. AF and thermal treatments revealed generally multicomponent behavior. However, basically three groups of magnetic directions could be isolated. The most representative group is formed by northwestern (southeastern) directions with moderate to steep negative (positive) inclinations (Fig. 2b). This component (A) was disclosed for samples from three sites collected close to Ariquemes, two sites close to Presidente Medici, some sites from the Cacoal region and some dykes from the Nova Brasilandia Group (mainly for the less metamorphosed dykes). A northern (southern) component (B) with low inclination appears mainly in the northern area, although it was also found in a few dykes from the Nova Brasilandia Group (Fig. 2c). A third northeastern, moderate positive inclination, direction (component C) was disclosed for dykes from the Nova Brasilandia Group, but it is also present in some samples of the northern area (Fig. 2d). Figure 3 shows the preliminary paleomagnetic poles calculated from components A, B and C described above, together with the virtual geomagnetic poles (VGPs) obtained from the sediments from the Aguapei Group (discriminated by AG symbols). Also shown in this figure are some VGPs (TU symbols) obtained for some crystalline basement rocks from the Rio Negro-Juruena Province (Bettencourt et al., 1996), and the paleomagnetic pole (NF) obtained for basalts and gabbros from the Nova Floresta Formation (Tohver et al., 2001). These poles were rotated to the configuration of Rodinia presented in D'Agrella-Filho et al. (1998), and they are compared with the APW path for Laurentia in the time interval: 1.24 Ga - 0.82 Ga. The VGPs from the sediments from the Aguapei Group fall along an apparent polar wander (APW)

The authors thank Gilmar J. Rizzotto (CPRM/RO) for his help during field work.

References

Bettencourt, J.S., Onstott, T.C., De Jesus, T., and Teixeira, W., 1996. Tectonic interpretation of 40Ar/39Ar ages on Country rocks from the Central Sector of the Rio Negro-Juruena Province, Southwest Amazonian Craton. International Geology Review, 38: 42-56. D'Agrella-Filho, M.S., Trindade, R.I.F., Siqueira, R., Ponte-Neto, C.F., and Pacca, I.G., 1998. Paleomagnetic constraints on the Rodinia Supercontinent: Implications for its Neoproterozoic break-up and the Formation of Gondwana. International Geology Review, 40: 171-188. Dalziel, I.W.D., 1997. Neoproterozoic-Paleozoic geography and tectonics: Review, hypothesis, environmental speculation. GSA Bulletin, 109: 16-42. McCausland, P.J.A., and Hodych, J.P., 1998. Paleomagnetism of the 550 Ma Skinner Cove volcanics of western Newfoundland and the opening of the lapetus Ocean. Earth Planet. Sci. Lett., 163: 15-29. Rizzoto, G.J., 1999. Petrologia e ambiente tectonico do Grupo Nova Brasilandia-RO. Master Thesis, UFRS, 136p. Sadowski, G.S., and Bettencourt, J.S., 1996. Mesoproterozoic tectonic correlations between eastern Laurentia and the western border of the Amazon Craton. Prec. Res., 76: 213227. Tassinari, C.C.G., Bettencourt, J.S., Geraldes, M.C., Macambira, M.J.B., and Lafon, J.M., 2000. The Amazonian Craton. In: Cordani, U.G., Milani, E.J., Thomaz-Filho, A. and Campos, D.A. (eds.) Tectonic Evolution of South America, p. 41-95, Rio de Janeiro. Tohver, E., Van der Pluijm, B., Scandolara, J., and Rizzoto, G.J., 2001. A reassessment of the tectonics and paleogeography of the Grenville-aged Sunsas-Aguapei Belt, SW Amazon Craton : New paleomagnetic and 40Ar/39Ar data . State of the art of the SW Amazonian Craton Geology Symposium, August, 1012, Sao Paulo, Brazil (in press).

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Sampled Areas

Geologicai Units

Geochronologlcal Provinces

Phanerozoic covers

Central Amazonian >2.3 Ga

Granitoid plutons

Maroni- Itacaiunas 2.2-1.95 Ga

I Precambrian sedimentary cover Acid - intermediate volcanic cover

VentuariTapajos 1,95-1,8Ga R x ^ ^ ^ ^

Basic volcanism

Rio Negro-Juruena 1.8-1.55Ga

I I111II

Greenstone belts

m H

Granulitic complex

Rondonian-San Ignacio 1.5-1.3Ga

Neoproterozoic mobile belt ^

Sunsas 1.25-1.0Ga

Basement structural high

Figure 1. Geological/geotectonic map of the Amazonian Craton (adapted from Tassinari et al., 2000)

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

R038-F1 (a)

AG19-C1 N/UP r 0.6x10'

^ ^ NRM \l50'

300°

.600° 700° S/DOWN R021-E1

NRM

Figure 2. Examples of AF and thermal demagnetizations. Zijderveld projections: full (empty) symbols represent horizontal (vertical) projections (see text for details).

Figure 3. Configuration of Amazonian Craton and Laurentia, as presented in D'Agrella-Filho et al. (1998) (Laurentia is in its present position). The present paleomagnetic data for the Amazonian Craton (full squares) is compared with the APW path for Laurentia (empty circles) in the time interval: 1.24-0.82 Ga. AG: VGPs from the Aguapei sediments; TU: VGPs from crystalline basement rocks (Rio NegroJuruena Province); NF: paleomagnetic pole for the Nova Floresta Formation; AS, An, Bs, Bn and Cn: paleomagnetic poles from the Nova Brasilandia Group and the Serra da Providencia Intrusive Suite.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

EVIDENCE FROM SOUTH AMERICA ON THE FORMATION AND DISRUPTION OF RODINIA M.S. D'Agrella-Filho\ U.G. Cordani^ B.B. de Brito Neves^ R.I.F. Trindade^ and I.G. Pacca^ ^ Institute of Astronomy, Geophysics and Atmospheric Sciences, Sao Paulo University, Brazil ^ Institute of Geosciences, Sao Paulo University, Brazil

INTRODUCTION

paleomagnetic data (D'Agrella-Filho et al., this volume). Based on paleomagnetic data, Weil et al. (1998) and D'Agrella-Filho et al. (1998) suggest a possible link between Kalahari and Laurentia. Dalziel et al. (2000) reinforce this hypothesis suggesting that the collision occurred along the Namaqua-Natal (Kalahari craton) and the Llano (Laurentia) orogenic belts, which have similar structural and metamorphic evolution. The position of the Congo-Sao Francisco Craton is less constrained, although a possible link between this block and Kalahari along the Zambezi mobile belt at ca. 0.80 Ga suggests a close position between them (Weil et al., 1998; Dalziel et al., 2000). The position of West Africa and Rio de la Plata cratons is yet less constrained since no reliable paleomagnetic data for Meso- and Neoproterozoic times are available. The only evidence of a possible link between the Rio de La Plata craton and Laurentia, based on stratigraphical and faunal similarities between Early Paleozoic sections in northwestern Argentina and the eastern United States is presently contested (Hoffman, 1999). In addition to the large fragments cited above, it is worth to mention the huge Potiguar-NigeriaHoggar block, which corresponds to the present-day NE Brazil and its counterparts in Africa (Fig. 1). This block was accreted to the Congo-Sao Francisco craton along the Cariris Velhos orogen (1.0 - 0.97 Ga). This event is well characterized in Brazil (Brito Neves et al., 1999). It initiated with rifting and bimodal volcanism, followed by deep-water sedimentary sequences. Collisional, arc-related intrusives, well dated at 0.97 Ga, close the regional development. Several other smaller cratonic nuclei (ArequipaAntofalla, Pampia, Paranapanema, Luis Alves, Central Goias, Serra do Mar and Juiz de Fora; Fig. 1) were also part of Rodinia (see Brito Neves et al., 1999 and Almeida et al., 2000 for a review). In the Andean zone, a Mesoproterozoic signature can be identified in the high-grade Garzon-Santa Marta orogenic belt and the Occidentalia terrain, despite a vigorous policyclic reworking. In central Brazil, the longitudinal Urua^uano belt, located at the western margin of the Central Goias fragment is thought to be formed by Mesoproterozoic (-1.4-1.2 Ga)

Since the seminal works of Dalziel (1991), Hoffman (1991), and Moores (1991), the paleogeography of Rodinia has received some paleomagnetic support from several continental fragments, such as Laurentia, Baltica, Siberia, South-China, Congo-Sao Francisco, India, Kalahari, Australia and East Antarctica (e.g., Weil et al., 1998; D'Agrella-Filho et al., 1998; Pisarevski et al., 2000; Evans et al., 2000). Other large cratonic areas, such as Amazonia, Rio de la Plata and West Africa, which form most of the Precambrian basement of South America, are usually placed on the basis of geological correlations with the neighboring continents. In these reconstructions, the smaller pieces, such as microcontinents and island arcs situated in-between the Amazonia, Kalahari, CongoSao Francisco and Rio de la Plata blocks, as well as a major piece, the Potiguar-Nigeria-Hoggar shield, are always omitted. In this work we will examine the available evidence for the position of the major and minor blocks from South America that eventually formed the Rodinia supercontinent, the orogenic events that led to their assembly and the timing of their dispersion.

RODINIA ASSEMBLY

The Rodinia assembly is characterized by a long time of accretionary and collisional orogenic processes (1.4 - 0.97 Ga). The Amazonia craton is a typical example of this evolution. Its western part is a multi-orogenic region formed between 1.8 and 1.0 Ga where successive magmatism, metamorphism and deformation events affected and reworked previous provinces, producing new complexes, as well as new juvenile continental crust (Tassinari et al., 2000). Three major geochronological/tectonic provinces have been identified in the Amazonia craton (Van Schmus et al., 1998; Geraldes, 2000): Sta. Helena (-1.45 Ga), Rondonian-San Ignacio (-1.35 Ga), and Sunsas-Aguapei (-1.1 Ga). The similar geological evolution of Amazonia and Laurentia suggest their connection in Neoproterozoic times (Sadowski & Bettencourt, 1996). This connection is not denied by the available Amazonian and Laurentian Neoproterozoic

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

no longer after its fmal assembly, at the time of the Grenville orogeny. In South America, this process led to the dispersion of that supercontinent into smaller masses, down to microcontinents and continental fragments. These blocks were later reassembled into other supercontinents (Gondwana, Pannotia) through the collision of allochtonous terrains of all sizes within the large areas affected by the so called Brasiliano/Pan-African orogenic cycle. These orogenies are related to the closure of several oceanic branches (Adamastor, Goianides, Phamsian), usually assembled under the name of Brasiliano ocean, and were formed in association with a geodynamic mantle/crust pattern quite coherent during a few hundred million years. Figure 1, modified after Almeida et al. (2000), shows a very large collage of cratonic masses and fragments, which makes up a branched system of orogenic belts of Neoproterozoic age. The main cratonic areas were outlined: Amazonia, West Africa, Sao Francisco-Congo, Kalahari, and Rio de La Plata. They exhibit marginal basins that were transformed into Neoproterozoic mobile belts. The extension of the respective cratonic area is traced as reworked basement. The Arequipa-Antofalla, Pampia, Paranapanema and Luiz Alves terranes are supposed to be cratonic fragments, not greatly disturbed by the Brasiliano/Pan-African tectonomagmatic events, although the first two were tectonically affected by the Phanerozoic events connected to the evolution of the Andes. In contrast, the Central Goias massif is formed by old crustal material, with Archean to Mesoproterozoic ages, but reworked by the Neoproterozoic orogenies. The same is true for the Borborema terrane, which is the extension, in South America, of the very large Hoggar massif in Africa. A relatively extensive ocean was formed in central Brazil, at about 800-900 Ma. Its closure is well marked by the existence of juvenile material, the Goias magmatic arc, that was added to the continental crust between 800 and 650 Ma. This was the only large episode of Neoproterozoic continental accretion in South America. In this way, the growth of continental crust for that part of Gondwana was minimal in the Neoproterozoic, basically restricted to the mentioned area of the Goias magmatic arc and the small Sao Gabriel block in southem Brazil. Although juvenile oceanic crust must have been produced by the disruption of Rodinia, only small volumes of it survived as ophiolitic fragments, added as tectonic slices within the Brasiliano/PanAfrican orogens. The very large amount of granitoid material with ages between 635 and 530 Ma, and intrusive into all Neoproterozoic mobile belts, was formed by reworking of older crust, with some contribution of mantle material, suggesting ensialic settings for the magmatic arcs.

orogenic mafic/ ultramafic rocks, and anorogenic Paleoproterozoic volcano-sedimentary rocks. The Espinha90 fold system is interpreted as an eastward extension of these Mesoproterozoic orogenic belts. It is an inland orogen, whose main deformation phase is dated at~1.3 Ga.

l l i ^ ^ i CSF f" /

Neoproterozoic Cratons (+microcontinents) Marginal belts I E 3

I Volcano-sedimentary belts Ophlolites

I N H Magmatic arcs I

Present shorelines

I Covered and unknown areas

Shear zones. Major faults -jii

t, I

Thrust zones

Figure 1. The new Proterozoic blocks (large, intermediate and small) and the different types of Neoproterozoic fold belts surrounding and among them. AM - Amazonian; WA - West Africa/Sao Luis; CSF - Congo/ Sao Francisco; KA - Kalahari; RP - Rio de la Plata; AA - Arequipa/Antofalla; PA Pampia; PP - Paranapanema; LA - Luis Alves; CG Central Goias; PR - Pamaiba; JF - Juiz de Fora; SM - Serra do Mar; Po-Ho - Potiguar/Hoggar (adapted from Almeida et al., 2000). The above described orogenic events that led to the assembly of Rodinia fragments in South America, are diachronic, beginning at -1.45 Ga and lasting up to the early Neoproterozoic at 0.97 Ga (the Cariris Velhos orogeny).

RODINIA DISRUPTION Several authors argue for the stability of Rodinia until about 750-700 Ma, when break-up would have occurred (Dalziel, 1991; Hoffman, 1991). From our data, however, the break-up of Rodinia has started

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

metamorphic age also affects the granitoid rocks of the Goias magmatic arc, formed earlier in a large oceanic environment. There are other evidences for events occurring in the 650-850 Ma. age interval, related to rocks from the Neoproterozoic belts. They are less documented, some of them indicate coUisional and others indicate distensive episodes. They also emphasize the already mentioned diachronism, when contemporaneity is observed between compressional and tensional events occurring in different parts of the same supercontinent.

Apparent polar wander (APW) paths, or at least parts of them, can be traced for the Sao FranciscoCongo and Rio de La Plata continental masses between 700 and 500 Ma. These APW paths suggest that they collided and joined around 600-650 Ma. The APW path of Australia, however, indicates that this continent joined West Gondwana a little later, around 550-530 Ma (Kuunga orogeny, Meert et al., 1995). The positioning of Amazonia can be assessed if we consider this block joined to Laurentia. The Laurentia path indicates a connection of these blocks with the Sao Francisco-Congo at about 580 Ma. For West Africa and Kalahari there is no reliable paleomagnetic evidence. The collisional belts within the Borborema and Damara provinces suggest that these large continental masses joined West Gondwana along a wide time span from 650 to 550 Ma. Moreover, geochronological and paleomagnetic data indicate that Kalahari was always close to Congo-Sao Francisco, at least since about 800 Ma, after the Zambezi orogeny. As a consequence of the above reasoning. West Gondwana was already formed at around 550 Ma. The limited amount of accretionary juvenile material within the Brasiliano/Pan-African mobile belts seems to indicate that only small oceans were formed between the fragments resulting from Rodinia disruption. This suggests the action of a series of Wilson Cycle episodes, with minor formation of oceanic lithosphere that left only traces within the mobile belts formed later. In our view, time and duration for the Rodinia break-up shall be re-examined, because within the Brasiliano/PanAfrican mobile belts there is ample evidence of collisional events that occurred prior to the final assembly of Gondwana at about 550 Ma. Such events are already driven by the geodynamic pattern that is related to the formation of the Neoproterozoic supercontinent. In many cases, such earlier collisional events, in the general time frame between 850 and 650 Ma, are difficult to identify, because of the usually stronger imprint by the tectonomagmatic events and associated granitoid magmatism which are typical of the Brasiliano/Pan-African belts. The main geochronological evidence in this respect are: the ca. 800 Ma age of the Zambezi orogeny (Hanson et al., 1998), which indicates a link between the Sao Francisco-Congo and the Kalahari cratons since that time; the ca. 800 Ma age of the Embu Complex, within the Ribeira belt (Cordani et al., 2001), and the 750 to 900 Ma ages for granitoids related to the Passinho and Sao Gabriel orogenies in southern Brazil (Hartmann, et al., 2000), which are constraints for the junction of the Rio de La Plata and the Kalahari cratons; the ca. 760 Ma. metamorphic age of the high grade gneisses of the Central Goias Complex (Pimentel et al., 2000), indicating a collisional episode between a microcontinent and the Amazonian craton. This

CONCLUSIONS

From the previous items, a few main conclusions

can be drawn:

a) When the reconstruction of Rodinia is taken into account, it is necessary to include into the picture the smaller continental masses, microcontinents and the like, such as the Central Goias terrain, the PotiguarNigeria-Hoggar, the Luiz Alves, the Pampia, etc. The same is true for the marginal zones of the major cratonic masses, where basement inliers are present as tectonic slices. b) For the break-up of Rodinia, it is necessary to consider a large time interval of disruption. Signals of distension are as old as 1000 Ma., when basic dike swarms are observed in the Sao FranciscoCongo craton. The formation of the large Goias ocean occurred at about 900 Ma., and other fragmentation episodes were recorded successively until about 630 - 640 Ma. c) For the agglutination of Gondwana (and Pannotia), the first continental collisions are reported at about 800 Ma. However, the most important episodes are those associated to the Brasiliano/PanAfrican orogenic cycle, between 650 and 530 Ma. It is apparent that the fragmentation of Rodinia is more or less synchronous with the accretion of Pannotia and Gondwana. d) If Laurentia is central for the Rodinia supercontinent, the Sao Francisco-Congo is central for Gondwana, because most of the Brasiliano/PanAfrican mobile belts are disposed all around it. Its APW path, between 750 and 500 Ma., seems to indicate a rotational movement which could be explained by its accommodation to the successive collisional events that affected it. The last one may well have been the accretion of East Gondwana, by the collisional episodes recorded in the Mozambique belt, and related to the Kuunga orogeny.

References

Almeida, F.F.M., Brito Neves, B.B., Cameiro, C.D.R. 2000. The origin and evolution of the South American Platform. Earth Sci. Rev., 50: 77-111. Brito-Neves, B.B., Campos-Neto, M.C., Fuck, R.A. 1999. From Rodinia to Western Gondwana: an approach to the BrasilianoPan African Cycle and orogenic collage. Episodes, 22: 155-166.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Meert, J.G., Van der Voo, R., Ayub, S. 1995. Paleomagnetic investigation of the Neoproterozoic Gagwe lavas and Mbozi complex, Tanzania and the assembly of Gondwana. Prec. Res., 74: 225-244. Moores, E.M. 1991. Southwest U.S.-East Antarctica (SWEAT) connection: a hypothesis. Geology, 19: 425-428. Pimentel, M.M., Fuck, R.A., Jost, H., Ferreira-Filho, C.F. and Araujo, S.M. 2000. The basement of the Brasilia Fold Belt and the Goias Magmatic Arc. In: Cordani, U.G., Milani, E.J., Thomaz-Filho, A. and Campos, D.A. (Eds.). Tectonic Evolution of South America, Rio de Janeiro, 31 st International Geological Congress, pp. 195-229. Pisarevsky, S.A., Komissarova, R.A., Khramov, A.N. 2000. New palaeomagnetic result from Vendian red sediments in Cisbaikalia and the problem of the relationship of Siberia and Laurentia in the Vendian. Geophys. J. Int., 140: 598-610. Sadowski, G.R. and Bettencourt, J.S. 1996. Mesoproterozoic tectonic correlations between eastern Laurentia and the western border of the Amazon Craton. Prec. Res., 76: 213-227. Tassinari, C.C.G., Bettencourt, J.S., Geraldes, M.C., Macambira, M.J.B., Lafon, J.M. 2000. The Amazonian craton. In: U.G. Cordani, J. Milani, A. Thomaz Filho, D.A. Campos. Teconic Evolution of South America, Rio de Janeiro. 41-95pp. Van Schmus, W.R., Geraldes, M.C., Kozuchi, M., Fetter, A.H., Tassinari, C.C.G., Teixeira, W. 1998. U/Pb and Sm/Nd constrains on the age and origin of proterozoic crust in southwestern Mato Grosso, Brazil: evidence for a 1450 Ma magmatic arc in SW Amazonia. Intemation Symposium on Tectonics, Ouro Preto, MG, Abstracts Vol., 121-125. Weil, A.B., Van der Voo, R., Niocaill, C.M. and Meert, J.G. 1998. The Proterozoic supercontinent Rodinia: paleomagnetically derived reconstructions for 1100 to 800 Ma. Earth Plan. Sci. Lett., 154: 13-24.

Cordani, U.G., Coutinho, J.M.V. and Nutman, A.P. 2001. Geochronological constraints on the evolution of the Embu Complex. J. South Amer. Earth Sci, (in print). D'Agrella-Filho, M.S., Trindade, R.I.F., Siqueira, R., Ponte-Neto, C.F. and Pacca, I.G. 1998. Paleomagnetic constraints on the Rodinia Supercontinent: Implications for its Neoproterozoic break-up and the formation of Gondwana. Int. Geol. Review, 40: 171-188. D'Agrella-Filho, M.S., Pacca, I.G., Siqueira, R., Elming, S-A., Teixeira, W., Bettencourt, J.S., Geraldes, M.C. (this volume) Amazonian proterozoic poles: implications to Rodinia paleogeography. Dalziel, I.W.D. 1991. Pacific margins of Laurentia and East Antarctica-Australia as a conjugate rift pair: evidence and implications for an Eocambrian supercontinent. Geology, 19:598-601. Dalziel, I.W.D., Mosher, S., Gahagan, L.M. 2000. LaurentiaKalahari collision and the assembly of Rodinia. J. Geol, 108: 499-513. Evans, D.A.D., Li, Z.X., Kirschvink, J.L., Wingate, M.T.D. 2000. A high-quality Mid-Neoproterozoic paleomagnetic pole from South China, with implications for ice-ages and the breakup configuration of Rodinia. Prec. Res., 100: 313-334. Geraldes, M.C. 2000. Geocronologia e geoquimica do plutonismo Mesoproterozoico do SW do estado de Mato Grosso (SW do craton Amazonico). PhD thesis, USP, Sao Paulo. 187pp. Hanson, R.E., Hargrove, U.S., Martin, M.W., Bowring, S.A., Krol, M.A., Hodges, K.V., Munyanyiva, H. and Bleckinsop, T.G. 1998. New geochronological constraints on the tectonic evolution of the Pan-African Zambezi Belt, south central Africa. J. Afr. Earth Sci., 27: 104-105. Hartmann,L.A., Leite, J.A.D., Silva, L.C.da, Remus, M.V.D., McNaughton, N.J., Groves, D.I., Fletcher, I.R., Santos, J.O.S. and Vasconcellos, M.A.Z. 2000. Advances in SHRIMP geochronology and their impact on understanding the tectonic and metallogenic evolution of southern Brazil. Aus. J. Earth Sci., 47: 829-844. Hoffman, P.F. 1991 Did the breakout of Laurentia turn Gondwanaland inside-out? Science, 252: 1409-1412. Hoffman, P.F. 1999. The break-up of Rodinia, birth of Gondwana, true polar wander and the snowball Earth. J. Afr. Earth Sci., 28: 17-33.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

RODINIA 2001: BALKANIZATION OR REUNIFICATION? I. Dalziel Institute for Geophysics, University of Texas at Austin, 4412 Spicewood Springs Road, Austin, Texas 78759 (Also associated with the Tectonics Special Research Centre, Department of Geology and Geophysics, University of Western Australia)

supercontinent, with special emphasis on the state of the concept when Chris Powell proposed the establishment of the Tectonics Special Research Centre (TSRC) in the University of Western Australia to elucidate the supercontinents of which Australia has been a part over the past 3 billion years. I suggest that a credible pre-Pangaea palaeogeography, which is critical, as Chris Powell appreciated, for understanding Earth system evolution, can emerge only very gradually from a painstaking multi-disciplinary global approach to the problem. During its remaining four years, the TSRC can play a major role in moving the Earth science community in this direction.

Current ideas concerning 'Rodinia' developed with stratigraphic analysis of the thermal subsidence of late Precambrian continental margins around the globe by Bond et al. (1984). They concluded that Laurentia had 'broken out' from within a supercontinent. Moores (1991) identified possible matching 'piercing points' along the proto-Pacific margins of Laurentia and East Antarctica-Australia, suggesting the now well-known South-West United States-East Antarctica (SWEAT) model. Dalziel (1991) suggested additional matching piercing points and demonstrated the nearly identical lengths of the two rifted margins. Hoffman (1991) provided the first model for the break-up of a Rodinia supercontinent juxtaposing these margins, the opening of the Pacific Ocean basin, and the amalgamation of Gondwanaland. In the intervening decade, several modifications of the SWEAT fit have been suggested on the basis of isotope geochemistry and regional geology, and Gose et al. (1997) demonstrated that Dalziel's critical East Antarctic piercing point was probably part of the Kalahari craton until the amalgamation of Gondwanaland at the end of the Precambrian. Radically different interpretations of a possible Rodinia supercontinent have been also been proposed. One places the South-West United States against South-East Australia (AUSWUS), another revives a pre-1990's hypothesis that the Siberian craton rifted from the proto-Pacific margin of Laurentia. New paleomagnetic and geochronologic data to be presented by Wingate et al. (this volume) suggest, if taken at face value, that both SWEAT and AUSWUS are incorrect for ca. 1070 Ma. In this introductory contribution, I will review the history of geologic thought regarding a Rodinian

References Bond G.C., Nickeson P.A, Kominz M.A. 1984. Breakup of a supercontinent between 625 Ma and 555 Ma; new evidence and implications for continental histories, Earth and Planetary Science Letters, 70, 325-345. Dalziel, I.W.D., 1991. Pacific margins of Laurentia and East Antarctica-Australia as a conjugate rift pair; evidence and implications for an Eocambrian supercontinent. Geology, 19, 598-601. Gose W.A, Helper M.A., Connelly J.N., Hutson F.E., Dalziel L W.D., 1997. Paleomagnetic data and U-Pb isotopic age determinations from Coats Land, Antarctica; implications for late Proterozoic plate reconstructions. Journal of Geophysical Research, B, Solid Earth and Planets. 102, 7887-7902. Hoffman, P.F., 1991. Did the breakout of Laurentia turn Gondwanaland inside-out? Science, 252, 1409-1412. Moores, E.M., 1991. Southwest U.S.-East Antarctic (SWEAT) connection; a hypothesis. Geology, 19,425-428. Wingate, M.T.D., Pisarevsky, S.A. Evans, D.A.D., 2001. AUSMEX: a new Rodinia reconstruction at 1070 Ma (this volume).

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

IMPLICATIONS OF TRUE POLAR WANDER FOR RODINIA RECONSTRUCTIONS D.A.D. Evans Tectonics Special Research Centre, Department of Geology and Geophysics, University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia.

and therefore only a brief exposition of the model is presented here. A long-lived and slowly drifting supercontinent prevents subduction of cool slabs under its interior, a preclusion that ultimately develops a hot upwelling zone (superplume) beneath the supercontinent. Such an upwelling zone deforms all solid-Earth density discontinuities, creating excess inertia thus drawing the supercontinent to the equator via a single episode of TPW (the first of two "flavors" experienced throughout the supercontinent cycle). The "Pangean" upwelling is encircled by a "ring of fire" of subduction zones that also circumscribe an external "Panthalassan" ocean, which itself may overlie a superplume antipodal to its "Pangean" counterpart. The "Pangean" superplume causes the supercontinent to fragment, but the general convective pattern persists even long after breakup. Today's geoid, lower-mantle seismic structure, and hotspot distribution reflect such a legacy of the Pangean world, 200 Myr after initial rifting of that supercontinent (Anderson, 1982). The two antipodal mantle upwellings or superplumes, encircled by the downwelling "ring of fire," create a prolate nonhydrostatic axis of figure that is represented to first order by the degree-2 geoid. This prolateness, of amplitude ±200m, is dwarfed by the Earth's 20-km hydrostatic bulge, but the latter feature relaxes on short timescales and thus may be considered stationary in the celestial reference frame: TPW drives the solid Earth incrementally through the bulge (Goldreich and Toomre, 1969) as if through a standing wave. Because prolate objects are rotationally unstable, the TPW response of minor inertial fluctuations superimposed on the system will be greatly amplified. Thus we deduce a second "flavor" of TPW, entirely following the first, oscillating quasichaotically within a single great circle-the same "ring of fire" encircling the fragmenting supercontinent. The system is extremely sensitive to relatively minor inertial adjustments, such as initiation of a single subduction zone or development of a single, isolated mantle plume, so the TPW oscillations may be difficult to relate to specific geological events. Depending on mantle viscosity structure, plausible rates of TPW may reach 90° per few Myr, or ca.30cm/yr along the great circle (Steinberger and O' Connell, 1997). The

INTRODUCTION Paleomagnetic data from the terminalProterozoic to Cambrian interval are not easily reconciled with geologically based continental reconstructions. For example, the commonly accepted Laurentian dataset appears to show rapid drift of that continent from the South pole to the equator, between 565 and 550 Ma (Torsvik et al., 1996; McCausland and Hodych, 1998), whereas the nascent "West" Gondwanaland apparently remained at high southerly latitudes until the dawn of the Cambrian, ca.545 Ma. However, the ca.590-550-Ma separation of these cratons from a Pannotia assemblage (see Dalziel, 1997) would need to occur at astonishingly rapid rates to achieve a wide lapetus Ocean as suggested by the paleomagnetic data (McCausland et al., 1998). Another geodynamic puzzle is the entirely counterintuitive system of passive margins on the leading edges of northwardfleeing Laurentia during the terminal Proterozoic and the Amazon sector of Gondwanaland during the Early Cambrian (Dalziel, 1997); and the EarlyMiddle Cambrian Ross-Delamerian orogenic belt along Gondwanaland's trailing margin (see Li and Powell, 2001). There are several solutions to these dilemmas. First, one may ascribe special geodynamic circumstances for that interval of Earth history, perhaps unlike those operating today (Gumis and Torsvik, 1994). Second, one may question certain portions of the widely accepted paleomagnetic database and thus permit geologically based reconstructions (e.g., the entirely low-latitude model for terminal Proterozoic Laurentia by Pisarevsky et al., 2000). Finally, there is the possibility that rapid true polar wander (TPW) occurred throughout the Proterozoic-Cambrian transition (Evans, 1998, in press), rendering most paleomagnetic poles too imprecisely dated for robust intercontinental comparisons. This last hypothesis is not entirely defeatist, however, and it leads to a radically different approach for working with paleomagnetic data from this interval, with numerous ramifications for Rodinia reconstructions.

THE MULTIPLE-TPW HYPOTHESIS The Proterozoic-Cambrian TPW controversy has been summarized exhaustively by Evans (in press).

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Laurentia's terminal Proterozoic polar excursion has traditionally been considered to be a southern venture (see Dalziel, 1997), with rapid TPW there is the possibility of two polar excursions, the older perhaps being in the northern hemisphere (Evans and Kirschvink, 1999).

prolate axis persists even after assembly of the next supercontinent, which may experience rapid latitudinal changes until it eventually develops its own mantle upwelling axis, starting the cycle anew.

MESOPROTEROZOIC-PALEOZOIC PALEOMAGNETIC DATA

RECONSTRUCTIONS

Many broad features of the paleomagnetic database can be explained most easily by the TPW model, beginning with the late Mesoproterozoic interval in which Rodinia formed. Note that regardless of any precise Rodinia configuration, Laurentia must have lain near the center because that craton is surrounded on all sides by terminal Proterozoic passive margins. Laurentia's 1100-800 Ma apparent polar wander (APW) path, the best defined of any Rodinian craton, thus approximately constrains the motion of the supercontinent's centroid. Rapid and oscillatory changes of paleolatitude are evident during this interval and the preceding 200 Myr (McElhinny and McFadden, 2000). I propose that these abrupt latitudinal shifts result from oscillatory TPW about the relict prolate axis inherited from Nuna, Rodinia's predecessor (Hoffman, 1989, 1997). A second interval of rapid and oscillatory APW, observed among all the well constrained cratons, is the terminal Proterozoic-Cambrian transition accompanying the final assembly of Gondwanaland. The best constrained cratons are Laurentia and Australia. Laurentia's most reliable poles are primarily derived fi-om isotopically dated igneous rocks, whereas Australia's most reliable results are from poorly dated but incontrovertibly ordered stratigraphic successions. Both datasets suggest rapid oscillatory rotations, with Laurentia shifting latitudes (at rates possibly as high as --Im/yr!) and Australia pivoting about a nearby Euler pole. The Australian rotations are co-axial with subsequent oscillations of Gondwanaland, some of which have been attributed to TPW (Van der Voo, 1994). If these oscillations are all due to TPW about a vanished Rodinian axis, then large-scale mantle convective patterns may lag behind supercontinental transitions for as much as 300-400 Myr (Evans, 1998).

If TPW had occurred as fast as 90° per few million years, there may be little hope in precise enough magnetization ages for comparing poles to test proposed cratonic reconstructions. One solution to this problem would involve a new paleogeographic reference frame altogether, the long-lived and equatorial, common TPW axis (Evans, in prep.). For times when TPW occurred at a rate fast enough to dwarf between-plate motions yet slow enough for adequate calibration by precise isotopic dating or biostratigraphic control, then both (orthogonal) reference frames could be used in conjunction to yield an absolute global paleogeography in latitude and longitude (e.g., Kirschvink et al., 1997). This method is somewhat model-driven, however, and will likely not be utilized by most paleomagnetists until the multipleTPW hypothesis survives rigorous testing with new data. In the meantime, traditional methods applied to the current paleomagnetic database require rejection or modification of apparently reliable data in order to achieve geographically sound and platetectonically feasible reconstructions for the interval 615-500 Ma. In other words, none of the traditionally based reconstructions for that period (e.g., Torsvik et al., 1996; Smethurst et al, 1998; Pisarevsky et al., 2000) satisfy all paleomagnetic data passing consistent and reasonable quality filters.

INDEPENDENT OBSERVATIONS Leading and trailing edges of continents may provide independent constraints on these issues. We might expect leading edges of continents to record active tectonic settings in general, whereas trailing margins would generally be passive (see Dalziel, 1997). Exceptions exist, notably if slab pull on a plate draws a continent toward a subduction zone such as the northward pull of India in the Cretaceous, with its leading passive margin. Nonetheless, the situation is more simple in times of supercontinental assembly into a single land-bearing plate (Marcano et al, 1999). During the final assembly of Rodinia (1140-ca.l000 Ma), Laurentia drifted rapidly away from the pole, with the Grenville belt at the leading edge. The Ottawan orogeny, however, occurs only near the end of this interval (1080-1020 Ma), and there is no "hairpin" in the APW path signalling the onset of this tectonism (see Powell et al., 2001). Another possible independent test of the TPW hypotheses concerns apparent changes in sea level

POLARITY AMBIGUITIES One of the implications of these possibilities is that all paleomagnetic poles older than Cambrian have uncertain polarity. Because TPW shifts as rapid as 90° in a few Myr are geodynamically feasible (Steinberger and O' Connell, 1997), and because most of the Neoproterozoic-Cambrian cratons' APW paths show great-circle distributions of paleomagnetic poles showing very rapid rates of APW (see Torsvik et al., 1996; Evans, 1998, in press; Hofftnan, 1999), it is difficult to assign polarity to older poles with any certainty. Although

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Dalziel, I.W.D., 1997. Neoproterozoic-Paleozoic geography and tectonics: Review, hypothesis, environmental speculation: Geol. Soc. Am. Bull., v. 109, p. 16-42. Evans, D.A., 1998. True polar wander, a supercontinental legacy. Earth Plan. Sci. Lett., v. 157, p. 1-8. Evans, D.A.D., 2002. True polar wander and supercontinents. Tectonophys., in press. Evans, D.A.D., and Kirschvink, J.L., 1999. Multiple episodes of rapid true polar wander in Vendian-Cambrian time? Geol. Soc. Am., Abst. with Prog., v.31, no.7, p.318. Goldreich, P., and Toomre, A., 1969. Some remarks on polar wandering. J. Geophys. Res., v.74, p.2555-2567. Gumis, M., and Torsvik, T.H., 1994. Rapid drift of large continents during the late Precambrian and Paleozoic: Paleomagnetic constraints and dynamic models. Geology, V.22, p. 1023-1026. Hoffman, P.P., 1989. Speculations on Laurentia's first gigayear (2.0-1.0 Ga). Geology, v.l7, p.135-138. Hoffman, P.P., 1997. Tectonic genealogy of North America. In: Van der Pluijm, B.A., and Marshak, S., eds.. Earth Sructure: An Introduction to Structural Geology and Tectonics. McGraw-Hill, p.459-464. Hoffman, P.P., 1999. The break-up of Rodinia, birth of Gondwana, true polar wander and the snowball Earth. J. Afir. Earth Sci., v.28, p. 17-33. Kirschvink, J.L., Ripperdan, R.L., and Evans, D.A., 1997. Evidence for a large-scale Early Cambrian reorganization of continental masses by inertial interchange true polar wander. Science, v.277,p.541-545. Li, Z.X. & Powell, C.McA., 2001. An outline of the palaeogeographic evolution of the Australasian region since the beginning of the Neoproterozoic. Earth-Science Reviews, V.53, p.237-277. Marcano, M.C., Van der Voo, R., and Mac Niocaill, C., 1999. True polar wander during the Permo-Triassic. J. Geodynam., v.28, p.75-95. McCausland, P.J.A. and Hodych, J.P., 1998. Paleomagnetism of the 550 Ma Skinner Cove volcanics of western Newfoundland and the opening of the lapetus Ocean: Earth Plan. Sci. Lett., v. 163, p. 15-29. McElhinny, M.W., and McFadden, P.L., 2000. Paleomagnetism: Continents and Oceans. Academic Press, International Geophysics Series v.73, 386pp. Mound, J.E., Mitrovica, J.X., Evans, D.A.D., and Kirschvink, J.L., 1999. A sea-level test for inertial interchange true polar wander events. Geophys. J. Int., v. 136, p.F5-F10. Pisarevsky, S.A., Komissarova, R.A., and Khramov, A.N., 2000. New palaeomagnetic result from Vendian red sediments in Cisbaikalia and the problem of the relationship of Siberia and Laurentia in the Vendian. Geophys. J. Int., v. 140, p.598-610. Powell, C.McA., Jones, D.L., Pisarevsky, S., and Wingate, M.T.D., 2001. Palaeomagnetic constraints on the position of the Kalahari craton in Rodinia. Precambr. Res., v.l 10, p.3346. Smethurst, M.A., Khramov, A.N., and Torsvik, T.H., 1998. The Neoproterozoic and Palaeozoic palaeomagnetic data for the Siberian Platform: From Rodinia to Pangea. Earth-Sci. Rev., v.43,p.l-24. Steinberger, B., and O' Connell, R.J., 1997. Changes of the Earth's rotation axis owing to advection of mantle density heterogeneities. Nature, v.387, p. 169-173. Torsvik, T.H., Smethurst, M.A., Meert, J.G., Van der Voo, R., McKerrow, W.S., Brasier, M.D., Sturt, B.A., and Walderhaug, H.J., 1996. Continental break-up and collision in the Neoproterozoic and Palaeozoic~A tale of Baltica and Laurentia. Earth Sci. Rev., v.40, p.229-258. Van der Voo, R., 1994. True polar wander during the middle Paleozoic? Earth Plan. Sci. Lett, v.l22, p.239-243.

due to continental flooding. Mound et al. (1999) modelled the sea-level response to rapid TPW events and predicted transgressive or regressive events with amplitudes of ca.lOOm, depending on lithospheric thickness and mantle viscosity models. Importantly, these apparent sea-level swings would change sign depending on location relative to the TPW rotation; given precise enough intercontinental stratigraphic correlations, one could distinguish between a TPW signal and a eustatic one. Mound et al. (1999) determined that Cambrian flooding records from Laurentia, Baltica, and Australia conformed to the TPW hypothesis of Kirschvink et al. (1997), at least to first order. The final independent test of TPW hypotheses involves geologically reasonable continental reconstructions. If TPW had occurred, then APW paths can be superimposed to generate reconstructions in both latitude and longitude. These paleogeographies must not include any significant continental overlaps. In addition, the paleomagnetically based reconstructions should be reasonable according to geological constraints. The Kirschvink et al. (1997) model produced a Cambrian paleogeography that is strikingly similar to that generated primarily from tectonic information by Dalziel (1997), with a narrow lapetus Ocean at that time.

IS THERE ANY HOPE? If the paleomagnetic results are not due to TPW, then the geoscience community must find another mechanism to explain the apparently rapid shifts in latitude for the terminal Proterozoic-Cambrian continents. One may select between nonuniformitarian plate tectonics or perhaps a nonuniformitarian geomagnetic field, both of which require careful testing (see Evans, in press). But at least the paleomagnetic data from well constrained cratons (such as Laurentia) appear to be well behaved at about 800-750 Ma, a time of major anorogenic magmatism related to Rodinia's rifting. These data, in conjunction with tectonic information, can produce an end-Rodinian paleogeography which in turn may serve as the basis for testing geomagnetic field models. The independent tests outlined above may help bolster or reject the multiple-TPW hypothesis, and with detailed sampling of volcanic-sedimentary successions we might just be lucky enough to catch one of the APW shifts "in the act," possibly simultaneously on several cratons. References Anderson, D.L., 1982. Hotspots, polar wander, Mesozoic convection and the geoid. Nature, v.297, p.391-393

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

THE NEOPROTEROZOIC EVOLUTION OF AUSTRALIA'S WESTERN MARGIN I.C.W. Fitzsimons TSRC, Dept of Applied Geology, Curtin University, GPO Box U1987, Perth, WA 6845, Australia

subducted beneath the Himalaya and the southern parts are largely buried beneath the Antarctic ice sheet. The eastern margin of the suture is exposed as the Pinjarra Orogen of Western Australia (Fig. 2), which truncates the Archaean Yilgam Craton, the Palaeoproterozoic Capricorn Orogen, and the Mesoproterozoic Albany-Fraser Orogen. It comprises three gneissic basement inliers (the Northampton, Mullingarra and Leeuwin complexes), which are draped by Phanerozoic sedimentary rocks of the Carnarvon and Perth sedimentary basins, and a series of low-grade metasedimentary rocks exposed along the Darling Fault (the Cardup, Yandanooka, Moora, and Badgeradda groups and the Nilling Formation). Like many Precambrian orogens, the Pinjarra Orogen has traditionally been regarded in terms of orthogonal continental collision, but new data and a reassessment of published relationships imply that gneissic inliers in the Pinjarra Orogen are allochthonous terranes. These terranes were formed some distance from their present locations adjacent to the Darling Fault, and were accreted to the Australian craton during at least two Neoproterozoic episodes of continent-scale transcurrent tectonics.

INTRODUCTION Early models for the assembly of East Gondwana assumed that Australia, India and Antarctica were amalgamated in the latest Mesoproterozoic and remained a coherent block throughout the Neoproterozoic, but increasing evidence for Neoproterozoic tectonism in East Antarctica indicates that the component parts of East Gondwana did not assemble until the end Neoproterozoic along a suture zone that cuts across East Antarctica and passes to the west of the Australian craton (Fig. 1; see review by Fitzsimons 2000).

LATE MESOPROTEROZOIC TECTONISM Evidence for 1100-1000 Ma tectonism is preserved in all three gneissic complexes of the Pinjarra Orogen (Bruguier et al. 1999; Cobb et al. 2001). Metamorphic zircon in granulite-facies (6 kb, 850°C) and amphibolite-facies (6 kb, 670°C) gneisses from the Northampton and Mullingarra complexes respectively have SHRIMP U-Pb ages of 1079 ±3 and 1058 ±83 Ma, and garnet-bearing granitic orthogneiss in the Leeuwin Complex has SHRIMP U-Pb zircon crystallization ages of 1091 ±8 and 1091 ±17 Ma (Nelson 1999). These data also correlate with 1040 ±50 Ma tectonism in the Obruchev Hills of the Denman Glacier region of East Antarctica (Sheraton et al. 1995), which would have been immediately along-strike from the Pinjarra Orogen before Gondwana break-up. This tectonism is younger than the last recorded tectonism in the Albany-Fraser Orogen (1130 Ma) and the crustal blocks preserving this younger event are aligned in a distribution that truncates the western margin of the Albany-Fraser Orogen.

Figure 1. Australia, India, Antarctica in a Gondwana reconstruction, showing the Neoproterozoic suture zone (dark grey) between older cratons (light grey). This suture zone is very poorly exposed and consequently poorly understood. The western half (present-day coordinates) in Greater India was

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

500 km

-35°S

115°E.

120°E

ALBANY-FRASER OROGEN

Figure 2. The geology of south-western Austraha. Gneissic basement inhers of the Pinjarra Orogen marked in black (1, Northampton Complex; 2, Mullingarra Complex; 3, Leeuwin Complex) and low-grade metasedimentary units of the Pinjarra Orogen in dark grey (4, Badgeradda Group and Nilling Formation; 5, Moora and Yandanooka groups; 6, Cardup Group). Phanerozoic sedimentary cover in stippled pattern. Northsouth dashed line at the western margin of the Yilgam Craton is a mylonite zone.

AN ALLOCHTHONOUS ORIGIN FOR THE NORTHAMPTON AND MULLINGARRA COMPLEXES

This north-south distribution of Grenville-age basement has been interpreted as an in situ metamorphic beh recording collision of the Australian Craton with India at 1100-1000 Ma (Myers et al. 1996; Bruguier et al. 1999), in which case the western margin of the combined CapricomYilgam-Albany-Fraser craton must have been truncated in the short time interval between 1130 and 1100 Ma. However, available structural, metamorphic and isotopic data support an allochthonous origin for these Grenville-age blocks, which were accreted to the craton margin some time after they were deformed and metamorphosed.

Rb-Sr biotite ages decrease westwards across the northern Yilgam Craton towards the Darling Fault (Libby et al. 1999), but they are Palaeoproterozoic in the region adjacent to the Northampton and Mullingarra complexes. Given that these two complexes underwent pervasive high-grade tectonism at 1080 Ma and the Mullingarra Complex is only 15 km from the craton margin, it is unlikely that the Northampton and Mullingarra complexes were metamorphosed in their present position, and this argument is reinforced by an undeformed granite body in the Mullingarra Complex, which crystalHzed at 2181 ±10 Ma and has a faulted

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

dykes that was emplaced at 755 Ma (Wingate & Giddings 2000). The orientations of these dykes and later brittle-ductile shear zones are consistent with north-south dextral wrenching along the Pinjarra Orogen (Byrne & Harris 1993). North-south trending mylonite is also common along the western edge of the Yilgam Craton within 20 km of the Darling Fault, and is best studied in the region between the Moora Group and the Albany-Fraser Orogen (Fig. 2; Bretan 1986) where the mylonites show evidence for at least three stages of movement. Steep dip-slip lineations probably reflect Phanerozoic reactivation of these structures during Gondwana break-up, but earlier sub-horizontal lineations record a Precambrian strike-slip history. Kinematic data predominantly indicate sinistral strike slip movement (see below), but there is limited evidence for an earlier stage of dextral wrenching. Although this dextral movement was assumed to be Archaean since Archaean Rb-Sr whole-rock ages are preserved within mylonite 100 km south of Perth (Blight et al. 1981), the Rb-Sr data need not have been reset by the deformation and this dextral movement along the Yilgam Craton could well be related to the Neoproterozoic dextral wrenching recorded in the Northampton Complex. Dextral wrenching must have commenced sometime after 1100-1000 Ma metamorphism in the Northampton and Mullingarra complexes, and commenced with or was preceded by tmncation of the Yilgam Craton, Capricom Orogen and AlbanyFraser Orogen along their present-day westem margins. This tmncation could have been focused along a pre-existing terrane boundary in the Yilgam Craton. The Northampton and Mullingarra complexes were then translated 750-1000 km northwards by dextral strike-slip until they were accreted onto the northwest Yilgam Craton. Palaeomagnetic data require that the Northampton Complex has not moved significantly relative to the Australian Craton since dyke emplacement at 755 Ma (Wingate & Giddings 2000), implying that the dykes and brittle-ductile shears represent the final stages of dextral movement as the Northampton Complex was accreted to the northwestem Yilgam Craton. 780-680 Ma felsic magmatism in the Leeuwin Complex (Nelson 1996) may also reflect this accretion event.

contact with overlying metasedimentary rocks (Cobb et al. 2001). This granite must have been juxtaposed against the gneiss some time after the 1080 Ma metamorphic event. Further evidence for an allochthonous origin comes from detrital zircon grains in the metasedimentary gneisses of the Northampton and Mullingarra complexes. Psammitic gneisses from the two complexes have indistinguishable detrital zircon populations with major age peaks at 19001600 Ma and 1400-1150 Ma (Bruguier et al. 1999; Cobb et al. 2001) indicating that they derive from the same sedimentary package. The dominant detrital populations closely match the age of basement rocks in the Albany-Fraser Orogen (Nelson et al. 1995; Clark et al. 2000), and are lacking the 2000 Ma population characteristic of the southern Capricorn Orogen. An Albany-Fraser source is consistent with the age of the youngest detrital grain (1113 ±26 Ma), which indicates that the sediments were deposited shortly after the cessation of tectonic activity in the Albany-Fraser Orogen and were presumably eroded from the new mountain range produced by this activity. However, only four of the 132 zircon grains analysed had an Archaean age, which is difficult to explain if the sediments were deposited in their present location since this would involve northwards transport of the detritus for 500-1000 km across the Yilgam craton. It is likely that sedimentary protoliths in the Northampton and Mullingarra complexes were eroded and deposited south of the watershed in the Albany-Fraser mountains to account for the lack of Yilgam detritus. These sedimentary rocks were deformed and metamorphosed at 1079±3 Ma, within 50 Myr of their deposition. Evidence for the metamorphic setting comes from the geochemistry of 1090 Ma granite protoliths in the Leeuwin Complex (Wilde & Muiphy 1990; Nelson 1995; Wilde 1999), which is consistent with syncollisional magmatism. A collisional setting would also account for burial of the Northampton and Mullingarra sediments to depths of 20 km, and implies that the original sedimentary basin was located near a continental margin.

EVIDENCE FOR DEXTRAL WRENCHING ALONG THE PINJARRA OROGEN AT 750 MA

EVIDENCE FOR SINISTRAL WRENCHING ALONG THE PINJARRA OROGEN AT 550 MA

The detrital and metamorphic data suggest that sedimentary protoliths in the Northampton and Mullingarra complexes were deposited and metamorphosed some distance from their present locations, and the detrital data imply that their original location was to the south, closer to the Albany-Fraser Orogen. Independent structural evidence for northwards transport of these blocks comes from the Northampton Complex, which is dissected by a NNE-SSW trending swarm of dolerite

The model described above does not account for widespread evidence of sinistral strike-slip movement along the Pinjarra Orogen (Harris 1994). The most obvious evidence is the regional-scale drag of the Albany-Fraser Orogen against the Pinjarra Orogen (Fig. 2), but there are also sinistral mylonite zones along the westem margin of the Yilgam

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Lambert graben of Antarctica (Fig. 1; Fitzsimons 2000; Soger et al. 2001), indicating that Precambrian orogens need not be simple orthogonal collision zones between two continental blocks. Precambrian sutures are clearly just as complex as those that are forming today.

Craton that largely obliterate any evidence for earlier dextral movements (Bretan 1986) and minor sinistral SSE-NNW shear zones that displace dolerite dykes in the Northampton Complex. K-Ar ages of 550 Ma from Northampton dykes adjacent to shear zones (Embleton & Schmidt 1985) and similar Rb-Sr biotite ages from the south-westem Yilgam Craton (Libby & De Laeter 1998) imply that the sinistral displacement occurred in the latest Neoproterozoic. This was synchronous with 570-525 Ma granitic magmatism in the Leeuwin Complex (Nelson 1996), where all plutons older than 530 Ma have penetrative fabrics consistent with sinistral transcurrent deformation (Harris 1994). The lack of penetrative sinistral wrenching at 550 Ma in the Northampton and Mullingarra complexes indicates that this later phase of ductile deformation was focused to the west of these terranes, which remained attached to the Yilgam Craton, whereas the Leeuwin Complex was probably translated southwards to its present position during the 550 Ma event. New palaeomagnetic data from India (Torsvik et al. 2001) indicate that India was about 30° of latitude north of its Gondwana position relative to Australia at 750 Ma, and it seems likely that the sinistral displacement observed along the Pinjarra Orogen at 550 Ma reflects the oblique collision of India with the westem Australian margin during the fmal assembly of Gondwana.

References Blight, D.F., Compston, W., & Wilde, S.A. 1981. West. Austral. Geol. Surv. Ann. Report 1980, 72-80. Boger S., Wilson C.J.L. & Fanning C.M., 2001. Geology 29, 4636 Bretan P.O., 1986. Unpubl. PhD Thesis, Imperial College, University of London. Bruguier O., Bosch D., Pidgeon R.T., Byrne D.R. & Harris L.B. 1999. Contr. Min. Pet. 136, 258-72. Byrne D.R. & Harris L.B. 1993. Ore Geol. Rev. 8, 89-115. Clark D.J., Hensen B.J. & Kinny P.D. 2000. Precam. Res. 102, 155-83. Cobb M.M., Cawood P.A., Kinny P.D. & Fitzsimons LC.W. 2001. Geol. Soc. Aust. Abstracts 64, 21-2. Dentith M.C., Bruner L, Long A., Middleton M.F. & Scott J., 1993. Explor. Geophys. 24, 455-62. Embleton B.J.J. & Schmidt P.W. 1985. Aust. J. Earth Sci. 32, 279-86. Fitzsimons LC.W. 2000. Geology 28, 879-82. Harris L.B. 1994. J. Geol. Soc. Lond. 151, 901-4. Libby W.G. & De Laeter J.R. 1998. Austral. J. Earth Sci. 45, 623-32. Libby W.G., De Laeter J.R. & Armstrong R.A. 1999. Austral. J. Earth Sci. 46, 851-60. Myers J.S., Shaw R.D & Tyler LS., 1996. Tectonics 15, 1431-46 Nelson D.R. 1995. Field Guide to the Leeuwin Complex, ACOG 3 Conference, Perth. Nelson D.R. 1996. Westem Austral. Geol. Surv. Record 1996/5. Nelson D.R. 1999. Westem Austral. Geol. Surv. Record 1999/2. Nelson D.R., Myers J.S. & Nutman A.P. 1995. Austral. J. Earth Sci. 42, 481-495. Sheraton J.W., Tingey R.J., Oliver R.L. «fe Black L.P. 1995. Aust. Geol. Surv. Org. Bull. 244. Torsvik T.H, Carter L., Ashwal L., Bhushan S, Pandit M. & Jamtveit B. 2001. Precamb. Res. 108, 319-33. Wilde S.A. 1999. Gondwana Res. 2, 481-99 Wilde S.A. & Murphy, D.M.K. 1990. Precamb. Res. 47, 251-70. Wingate M.T.D. & Giddings J.W. 2000. Precamb. Res. 100, 3 3 5 57.

CONCLUSIONS The Pinjarra Orogen comprises a number of allochthonous terranes accreted to the westem Australian margin during two stages of Neoproterozoic strike-slip tectonics, consistent with seismic evidence that the Darling Fault has reactivated a older strike-slip structure (Dentith et al. 1993). The second episode of sinistral transcurrent movement was driven by the oblique collision of India with Australia. An earlier episode of dextral transcurrent movement is less well understood, but probably involved displacements of at least 1000 km. A collage of terranes has also been identified at the westem margin of the suture zone adjacent to the

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

SYSTEMATIC ASSESSMENT OF DEFORMATION IN THE HINGE REGION OF A CHEVRON FOLD MJ. Gallen, S.M. Reddy and P.A. Cawood Tectonics Special Research Centre, Department of Applied Geology, Curtin University, G.P.O. Box U1987, Perth, Western Australia, Australia 6845

structures at a small scale, includes: bedding-parallel fibrous quartz vein formation, multiple folding events, dissolution of quartz and faulting. The bedding-parallel fibrous quartz veins are the oldest deformation related structures in the sample. The veins are particularly common in the hinge region but they are also present in the limbs. The veins are interpreted to have grown antitaxially between micro-bedding layers and the sense of curvature in some veins indicates a south over sense of movement, whereas in others it indicates a north over south sense of movement. This is interpreted to be a product of different amounts of compression during layer-parallel shortening in early stages of the D2 deformation event. The sense of curvature in any individual vein, and hence the movement direction, does not change across small-scale F2 folds hinges, indicating that the veins are older than the F2 folds. The veins were folded by the F2 folds of the D2 deformation event, which are represented in the sample by vertical to sub-vertical axial surfaces. Some axial surface can not be traced through zones of quartz dissolution indicating that the folds were older than the dissolution and cleavage development. Quartz dissolution was typically axial planar and was identified by discontinuities in bedding and axial surfaces and by the concentration of magnetite ± haematite. Significant dissolution of quartz post-dated the quartz vein formation. Two small faults, indicating movement consistent with the regional south over north sense of movement are located on the southern edge of the sample. These faults are identified as planes that define discontinuities in bedding layers and in the axial surfaces of F2 folds. Two small faults located on the northern edge of the sample indicate material on the northern side of the faults moved downwards. These brittle faults show mm offset of bedding layers and layer-parallel fibrous quartz veins. A later stage of folding that refolds all previously described structures, except for the faults, produces angular to open folds with shallowly dipping axial planes.

Folding in the Hamersley Province of Western Australia is associated with mineralisation but a clear understanding of folding at all scales has not been developed. Here we present a detailed study of fold formation from the Hamersley Province which has implications for chevron fold development and for models of ore body formation. The sample, from the Joffre Member of the Hamersley Group, is composed of layers of chert and banded iron (magnetite ± haematite) with sub-mm to cm scale bands. The sample is approximately 17.5 cm high and 13 cm wide and was taken from a close fold with a wavelength of 85 cm and an amplitude of 120 cm. The axial surfaces strikes at 260°/86°S and the fold axis plunges at 10° towards 256°. The orientation of the fold and the geometry of structures nearby suggests that it developed during the Palaeoproterozoic Ophthalmian Orogeny (Gallen, 1998), Tyler's (1991) regional D2 deformation event related to the collision of the Yilgam and Pilbara Cratons. A systematic assessment of age relationships (Potts & Reddy, 1999) has been used to systematically analyse the development of this chevron fold. The technique shows that deformation developed sequentially with no evidence for repetition or synchronous development of major structures. The fold has a well developed angular hinge but the shape of individual layers varies with lithology. Deformation structures are concentrated in the hinge regions of the fold with only minor strain in the limbs. A competent chert layer near the base of the sample in the inner arc of the hinge shows a more rounded form and would be classified as a Class IB fold in Ramsay's (1967) classification scheme. This band overlies a layer of fine-grained magnetite and minor quartz, which is also rounded. The top layer of the sample is a magnetite dominated layer with minor fine-grained quartz and is typically 6 mm thick although it has been locally thickened to more than 25 mm. The central portion of the sample contains a combination of chert and magnetite bedding layers and bedding-parallel fibrous quartz veins. The thickness of this portion of the sample ranges from 30 mm in an area of significant dissolution to >120 mm in the hinge of the antiform. The deformation history of the sample, which documents a complex series of interactions between

A deformation history was constructed using a technique developed by Potts & Reddy (1999). All of the data was entered into a single large younging table (Potts & Reddy, 199) and all observed relative age relationships were noted. The structures were

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

the layer-parallel fibrous quartz veins as indicated by the optical discontinuity of some bent fibres, evidence for grain boundary migration and the consistent sense of movement indicated in some veins on both limbs of small-scale F2 folds. Dissolution of both primary quartz in bedding layers and vein quartz followed this episode of folding, and becomes more significant with increasing distance from the axial surface of the fold. Evidence for dissolution is concentrated in the limbs of the fold, and in the limbs of small-scale folds in the hinge region. The final folding event, which folds all previously described structures, may or may not be related to the Ophthalmian deformation and may reflect a late stage local deformation event. The faulting on the northern limb of the fold is younger than the bedding parallel fibrous quartz veins whereas the faulting on the southern limb of the fold is younger than the F2 folds, but no relative age relationships have been observed with any other structures. The systematic development of deformation in the hinge region of the fold, with no evidence for repetition or synchronous development of structures, indicates that rocks in the Hamersley Province may respond to stress in a predictable and sequential manner. Models of ore body formation in the Hamersley Province incorporate elements of quartz dissolution and concentration of iron oxides (Powell et al., 1999). The precipitation of quartz during fibrous quartz vein formation prior to the dissolution of quartz during cleavage development has important implications for models of ore body formation, such models being highly influenced by the solubihty of quartz (Powell, et al., 1999).

ordered from oldest (at the bottom) to youngest (at the top). The size of the younging table was reduced by temporarily removing the 41 structures that had no observed relationships (Potts & Reddy, 1999). The reduced younging table was divided into seven mutually exclusive groups of structures. No relative age relationships were observed for any two structures of the same structural type (eg. the fibrous quartz veins). Consequently, the relative ages between different fibrous quartz veins are unknown. However, all of the relative relationships observed between structures from other types were consistent (i.e. F2 folds were always younger than fibrous quartz veins). A consequence of this observation was that the fibrous quartz veins could be placed in any order in the younging table (eg. ascending; descending; random), with no effect on the validity of the younging table (all younging structures would still point in the same direction (Potts & Reddy, 1999)). This relationship was true for fibrous quartz veins, F2 folds, zones of quartz dissolution and the late stage folds. Because of this consistent relationship between structures of different types it was interpreted that the structures within each type were synchronous (i.e. all fibrous quart veins were synchronous, all F2 folds were synchronous etc.) and that each type of structure occurred systematically with no repetition or synchronous development of structural types. A number of deformation related structures can be observed in the hinge region of the fold at a grain scale and sub-grain scale. These include sutured grain boundaries, optical discontinuity of fibrous quartz veins and bedding parallel quartz fibres. All of these features are observed in the central portion of the sample and are consistent with the sequence of deformation recognised on the basis of mesoscale features. The presence of sutured grain boundaries and preferential growth of grains at the expense of their neighbours is indicative of grain boundary migration and is most commonly observed in fibrous quartz veins that have been folded by F2 folds. Undeformed fibres that grew in a bent shape are optically continuous whereas deformed, or mechanically bent, fibres may or may not have been bent before deformation. The mechanical bending of fibrous quartz grains results in bent fibres that are optically discontinuous. A model of deformation has been developed that incorporates all of the structural features described above in a systematic deformation sequence. Early compression, possibly related to the Ophthalmian D2 deformation event, led to the formation of the layer-parallel fibrous quartz veins. Continued compression led to the folding of bedding layers and

References

Gallen, M. J. 1998. Structural styles in the southern Hamersley Province in the vicinity of Coondawanna Ridge. BSc. (Hons) thesis, Curtin University of Technology (unpublished). Potts, G. J. & Reddy, S.M. 1999. Construction and systematic assessment of relative deformation histories. Journal of Structural Geology 21: 1,245-1,253. Powell, C.M., Oliver, N.H.S., Xheng-Xiang, L., Martin, D. M. & Ronaszeki, J. 1999. Synorogenic hydrothermal origin for giant Hamersley iron oxide ore bodies. Geology 27: 175178. Tyler, I.M. 1991 The geology of the Sylvania Inlier and southeast Hamersley Basin. Geological Society of Western AustraHa Bulletin 138. Watters B.R. 1977. The Sinclair-group: Definition and regional correlation. Transactions of the Geological Society of South Africa 80, 9-16.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

SURVIVING THE SNOWBALL EARTH: THE ACRITARCH RECORD K. Grey Geological Survey of Western Australia, 100 Plain Street, East Perth, 6004, Western Australia

middle Fertatataka Formation (Amadeus Basin), and from more limited data, the lower Bunyeroo Formation (ARC). These formations are correlatives ft-om lithostratigraphic evidence, isotope chemostratigraphy, and seismic interpretation. ECAF assemblages are taxonomically diverse and demonstrate patterns of secular variation that allow effective correlation between Australian basins and have potential for global application. Fre-glacial and the ELF (post-Marinoan glacial) populations are almost identical. Species are morphologically simple and long ranging (some extend back to the Mesoproterozoic). Early Cryogenian populations (c.850-700 Ma) are taxonomically diverse but dominated by simple leiospheres. Rare species of phytoplankton are more complex, have short ranges, and are markers for specific horizons; e.g. Cerebrosphaera buickii restricted to c.750-700 Ma (Hill et al, 2000). Falynology of the interval between the Sturtian and early Marinoan glaciations (c.700-600 Ma) is poorly documented, with sparse records of poorly preserved and possibly reworked fossils (including some from glacial sediments). Assemblages are similar to preglacial ones but less diverse. Likewise, the ELF resembles preceding palynofloras, but is more impoverished and lacks acanthomorphs. Samples immediately above Marinoan-glacial horizons are barren essentially barren. This interval, from the top of cap dolomite to the first identifiable leiosphere, is c.l30 m thick in Wallara 1 (Amadeus Basin), and 440 m thick in SCYW la (ARC). The first productive samples above the Marinoan glaciation resemble Tonian and Cryogenian ones in consisting of simple, small-diameter leiospheres and rare Octoedryxium truncatum. The ELF in the Wallara 1-Rodinga 4 composite section is c.1520 m thick. In SCYW la, the ELF interval is c. 650 m thick, and is similar to that in the WWD-MJ 1 composite section. The Marinoan glaciation is not known in the Officer Basin but may be coeval with the Tarlina Sandstone. In Munta 1, the ELF is at least 680 m thick, and could be c.1800 m thick. By contrast, the ECAF is only c.80 m thick in Rodinga 4 (first acanthomorphs to the top of Fertatataka Formation) and 90-120 m thick in Munta 1 (first acanthomorphs to the canyon cutting event). In Munta 1, sedimentological studies identified several transgressive and regressive cycles

In discussing the 'Snowball Earth Hypothesis' (Hoffman et ai, 1998), Hoffman & Schrag (2000) proposed that 'the climate recovery following a huge Neoproterozoic glaciation paved the way for the explosive radiation of multicellular animal life soon thereafter'. They suggested that the early Marinoan glaciation was a major, and probably global, crisis that caused a fall in sea level, and depleted the biota by causing a fall in temperature and reduction in photosynthesis. '[C]reatures that survived glacial episodes may have taken refuge at hot springs both on the seafloor and near the surface of the ice where photosynthesis could be maintained'...'the steep and variable temperature and chemical gradients endemic to ephemeral hot springs would preselect for survival in the hellish aftermath to come' (Hoffman & Schrag, 2000). Climatic stress during the spread of the ice may have annihilated most organisms, and then given rise to an explosion of new life forms in the aftermath of Snowball Earth. The Hoffman & Schrag scenario is readily tested by examining the phytoplankton-diversity record for the Australian terminal Proterozoic, based mainly on palynological studies of nearly 2000 samples from c. 30 drillholes in the Adelaide Rift Complex (ARC), and Officer and Amadeus Basins (Damassa & Knoll, 1986; Zang & Walter, 1992; Jenkins et al, 1992; Zang 1995, 1997; Gravestock et al, 1997; Grey & Cotter, 1996; Grey & Stevens; 1997; Cotter, 1997, 1999; Hill et al, 2000; Grey, 1998, & in prep., and unpublished data). Five assemblage zones and two palynofloras are present in the post-Marinoan-glacial interval. Zones are recognisable despite palaeoenvironmental and taphonomic complications, and are independent of lithology, lithostratigraphy, and sequence stratigraphy. Transition from the older 'Ediacarian Leiosphere-dominated Palynoflora' (ELF) to the younger 'Ediacarian Complex Acanthomorph-dominated palynoflora' (ECAF) is abrupt and apparently synchronous in all drillholes studied. The biotic change is radical, and characterised by a marked and rapid increase in abundance, size, morphological complexity, and taxonomic diversity. Nearly 50 species and 20 genera of phytoplanktonic green algae (spiny acritarchs or acanthomorphs) appear for the first time, and show a major radiation that indicates a significant evolutionary change. The transition is in the lower Dey Dey Mudstone (Officer Basin),

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

appears unrelated to changes in sedimentology or sequence stratigraphy as described from Munta 1 (Arouri et al, in press), and did not take place until the second major marine excursion after the glaciation. However, the actual or presumed position of the transition shows a remarkable coincidence between the first appearance of the acanthomorph palynoflora, a negative S^^C^j-g excursion (Calver

characterised by slope aprons (Arouri et al, 2000). Subsidence rates varied during deposition of each apron. Rapid subsidence was followed by slower subsidence, and then terminated in extremely slow sedimentation during the formation of a terminal carbonate. However, overall subsidence rates were probably slow and generally constant (Calver & Lindsay, 1998). There is little evidence for a drastic change in sedimentation rate, so the relative thicknesses give a good approximation of relative duration. The thickish barren interval, thick ELP, and thin ECAP palynofloral intervals are consistent with estimates of c.595 Ma for the Marinoan cap dolomite, 578 Ma for the Acraman impact event, 549 Ma for the climax the Ediacara biota, and 544 Ma for the Neoproterozoic-Cambrian boundary (Walter et al, 2000). Combining the Walter et al estimates, and relative thicknesses, it appears that the period between the glaciation and the first appearance of the ECAP was substantial and of at least 20 m.y. duration. Diversification of the acanthomorphs did not begin until at least 580 Ma, and the ECAP terminated shortly after the canyon cutting event, at c. 565 Ma. The rapid radiation took place in less than 15 m.y., but not until c. 20 m.y. years after the end of the glaciation. Diversification rates remained slow or nonexistent throughout the ELP. In Australia at least, the dominant assemblage components were longranging, simple leiospheres and filaments that are typical of shelfal, shallow-water environments. This environment is one most likely to suffer from extreme glaciation. The post-glacial leiospheredominated population is not consistent with recolonisation by rapid diversification from hotspring refligia colonies. No evidence was found of new, rapidly diversifying taxa from a different type of ecological niche after the glaciation. Taxonomic affinities in both the ELP and ECAP do not indicate evolution from extremophiles that might have survived in hot-spring or ice-dominated refugia. Climatic conditions were severe during the Marinoan glaciation, and probably resulted in an impoverishment in species diversity, but there does not seem to be any major change in the nature of the acritarch populations at this time. Animal evolution probably mirrored plant evolution. Available evidence suggests that rapid expansion did not take place until at least 20 m.y. after the glaciation. The biotic evidence does not support the more extreme conditions predicted by a Snowball Earth, and extreme glaciation does not seem to have been the principal cause of biotic change, because substantial changes did not occur until much later. In contrast to the ELP, the ECAP is dominated by large, complex planktonic acanthomorphs that show rapid diversification. Rare elements of the older leiosphere palynoflora persist, but appear to be derived by reworking. The increase in diversity

and Lindsay, 1998), and the Acraman impact ejecta layer. This layer forms a significant synchronous marker horizon that links the Officer Basin and the Adelaide Rift Complex, and was used as an appropriate datum for plotting the relative distribution of species. Evidence for a relationship between the impact event and changes observed in acritarch assemblages is still largely circumstantial, but there is a possibility that the late Ediacarian diversification represents a recovery event following a large bolide impact. Whether the events are related or purely coincidental remains to be investigated, but the patterns observed are sufficiently consistent to justify further study. The aftermath of a large bolide impact would siq)ply a plausible explanation for dramatic palynofloral changes that are otherwise difticult to explain. The sharp palynofloral boundary does not correspond to a lithostratigraphic boundary, change in lithology, or sequence boundary of any type. Instead, it occurs in a uniform succession of mudstone in the middle of a transgressive systems tract, and the increase in diversity continues through a lowstand into a regressive phase. Moreover, the palynofloral transition is documented at the same relative stratigraphic position in three different tectonic units (ARC, Officer and Amadeus Basins). The rapid radiations and extinctions that have been recorded in the terminal Proterozoic biotas, could be explained by the following model, which invokes a cumulative effect of several closely spaced crises. An already depauperate bacterial and algal population on continental shelves suffered a further decline during the Marinoan glaciation as a result of extreme conditions. Initial recolonisation was slow, but increased as climatic warming followed the glacial maximum, a rapid marine transgression flooded the continental shelves, and nutrient input to the oceans increased as glaciers melted. Reestablished populations showed little variation from pre-glacial ones. Recovery was characterised by the development of blooms and opportunistic colonisers, but as the climate warmed and sea level rose, the depauperate biota was beginning to diversify. However, before the biota could properly recover, it was further depleted by a bolide impact of a magnitude sufficient to induce a global biotic crisis. The subsequent reduction in light intensity had a devastating effect on many pre-Marinoan species, including benthic and non-encysting assemblages, but had a lesser impact on encysting taxa, such as

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Grey, K., (in prep.). Ediacarian palynology of Australia: Association of Australasian Palaeontologists Memoirs. Grey, K., & Cotter, K.L., 1996. Palynology in the search for Proterozoic hydrocarbons: Western Australia Geological Survey Annual Review for 1995-96, 70-80. Grey, K., & Stevens, M.K, 1997. Neoproterozoic palynomorphs of the Savory Sub-basin, Western Australia, and their relevance to petroleum exploration. Western Australia Geological Survey Annual Review for 1996-97, 49-54. Hill, A.C., & Walter, M.R., 2000. Mid-Neoproterozoic (~830750 Ma) isotope stratigraphy of Australia and global correlation. Precambrian Research 100, 181-211. Hoffman, P.F., & Schragg, D.P., 2000. Snowball Earth. Scientific American, January, 68-75. Hoffman, P.F., Kaufman, A.J., Halverson, G.P., and Schrag, D.P., 1998. A Neoproterozoic Snowball Earth. Science 281, 13421346. Jenkins, R.J.F., McKirdy, D.M., Foster, C.B., O'Leary, T., & Pell, S.D., 1992. The record and stratigraphic implications of organic-walled microfossils from the Ediacaran (terminal Proterozoic) of South Australia. Geological Magazine 129, 401^10. Walter, M.R., Veevers, J.J., Calver, C.R., Gorjan, P., & Hill, A. C., 2000. Dating the 840-544 Ma Neoproterozoic interval by isotopes of strontium, carbon, and sulfur in seawater, and some interpretative models. Precambrian Research 100, 371433. Zang Wenlong, 1995. Early Neoproterozoic sequence stratigraphy and acritarch biostratigraphy, eastern Officer Basin, South Australia. Precambrian Research 74, 119-175. Zang Wenlong, 1997. Megascopic carbonaceous Chuaria and Tawuia from the late Neoproterozoic in South Australia. MESA Journal 4,37-41. Zang Wenlong & Walter, M.R., 1992. Late Proterozoic and Early Cambrian microfossils and biostratigraphy, Amadeus Basin, central Australia. Memoirs of the Association of Australasian Palaeontologists 12, 1-132.

acanthomorph acritarchs. As conditions improved following the impact, encysting phytoplankton had had a competitive advantage that allowed rapid diversification. The recovery assemblage, dominated by acanthomorphs, was short lived, and succumbed to the evolution of Ediacara-biota predators. Evolutionary shifts were probably accelerated by the relatively close spacing of three significant events, the Marinoan glaciation, the Acraman impact event, and the evolution of predators, that allowed little time for recovery between them. References Arouri, K., Conaghan, PJ., Walter, M.R, Bischoff, G.C., & Grey, K., 2000. Reconnaissance sedimentology and hydrocarbon biomarkers of Ediacarian microbial mats and acritarchs, lower Ungoolya Group, Officer Basin. Precambrian Research 100, 235-281. Calver, C.R., & Lindsay, J. F., 1998. Ediacarian sequence and isotope stratigraphy of the Officer Basin, South Australia. Australian Journal of Earth Sciences 45, 513-532. Cotter, K.L, 1999. Microfossils from Neoproterozoic Supersequence 1 of the Officer Basin, Western Australia. Alcheringa 23, 63-86. Cotter, K.L., 1997. Neoproterozoic microfossils from the Officer Basin, Western Australia. Alcheringa 21, 247-270. Damassa, S.P., & Knoll, A.H., 1986. Micropalaeontology of the late Proterozoic Arcoona Quartzite Member of the Tent Hill Formation, Stuart Shelf, South Australia. Alcheringa, 10, 417^30. Gravestock, D.I., Morton, J.G.G., & Zang, W-L., 1997. Chapter 7: Biostratigraphy and correlation. In Petroleum Geology of South Australia, Volume 3: Officer Basin. South Australia, J.G.G. Morton,& J.F. Drexel, eds., Department of Mines and Energy Resources Report Book 97/19, 87-97. Grey, K., 1998. Ediacarian acritarchs of Australia: PhD thesis Macquarie University, 669 p. (unpublished).

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

THE ASSEMBLY OF RODINIA AND GONDWANA: A VIEW FROM EAST ANTARCTICA S.L. Harley University of Edinburgh, Kings Buildings, West Mains Road, Edinburgh, EH9 3JW, United Kingdom

either side of Pan-African tectonic zones must be considered.

EAST ANTARCTICA AND RODINIA / GONDWANA RECONSTRUCTIONS

The geology of East Antarctica provides critical tests for continental reconstructions of Gondwana and Rodinia, and has been integral to the development of many of the ideas and models put forward over the past decade for these supercontinent configurations. However, recent metamorphic, structural and geochronological studies in the East Antarctic Shield, reviewed by Fitzsimons (2000a, b), demonstrate that it comprises a number of distinct crustal terranes that were not finally amalgamated until the Cambrian, at 550-500 Ma in late 'Pan-Afi-ican' tectonism. The demonstration of intense high-grade metamorphism and deformation at 550-500 Ma in a number of regions (e.g. Prydz Bay, Lutzow-Holm Bay, Central Dronning Maud Land), coupled with the recognition of distinct Meso- to Neoproterozoic crustal histories in intervening regions (e.g. Fitzsimons, 2000a) strongly suggests that there is no circum-East Antarctic 'Grenville' orogenic beh representing the ca. 1000 Ma collision of a unified East Antarctic Craton with other parts of East Gondwana. The evidence from East Antarctica does not support supercontinent models involving the existence of a unified East Gondwana prior to 550-500 Ma, and indicates that the original conception of a Rodinia supercontinent involving all the southern continents is incorrect. The key lines of evidence in support of these conclusions are reviewed in this contribution.

550-500 Ma Pan-African metamorphism and deformation: occurrence, style and nature

High-grade (granulite and amphibolite facies) metamorphism coupled with intense deformation over the time interval 500-500 Ma is now recognized from three main regions along the East Antarctic continental margin, as summarized below.

LUTZOW-HOLM BAY (LHB) AND DRONNING MAUD LAND (DML)

550-500 Ma high-grade metamorphism and deformation, first documented from the LutzowHolm Bay (LHB) region (Shiraishi et al., 1994), has now been identified from across a significant part of Central Dronning Maud Land (DML) and may even extend into western Dronning Maud Land. In the latter areas the Pan-African tectonism affects both juvenile or post-1000 Ma crustal precursors (e.g. 600 Ma intrusives) and older crust that had already undergone high-grade metamorphism at 1090-1030 Ma, ages that are compatible with those obtained in the Mesoproterozoic Maud province (Paech, 1997). Nappe structures, sheath folding, thrusts that disrupt and re-orient earlier structures, and the presence of laterally variable high-strain zones all point to the Pan-African metamorphism being accompanied by horizontal shortening and extensive displacement of crustal blocks in this probable southern extension of the East African / Mozambique / Madagascar orogenic beh. In at least two localities (Forefinger Point: Shiraishi et al., 1997; Rundvagshetta: Eraser et al., 2000) within the Liitzow-Holm Bay domain, PanAfrican metamorphism attained UHT (ultrahigh temperature) conditions of 9-11 kbar and >900°C, leading to the formation of sapphirine-bearing assemblages. Rapid near-isothermal decompression (ITD) of these UHT areas occurred at rates of ca. 3 mm/yr (Harley, 1998; Eraser et al, 2000), comparable with exhumation rates achieved today only in active collisional orogens (e.g. Himalayas Nanga Parbat) and large-scale oblique collision or transpressional fault zones. The metamorphic and deformational evidence are therefore consistent with

CAMBRIAN AMALGAMATION OF EAST ANTARCTICA

The interpretation of East Antarctica as being comprised of a number of separate and disparate continental blocks throughout the Neoproterozoic, only finally brought together and assembled in the Cambrian depends upon whether the newlyrecognised 550-500 Ma high-grade areas represent zones of suturing and collision instead of regions of reworking with limited lateral displacements. In the high-grade terrains preserved in East Antarctica it is not possible to define suture zones based on direct criteria such as ophiolite remnants. Instead, indirect criteria including the nature and style of metamorphism and deformation, and the presence of distinct crustal histories in terrains on

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

centred on defining its event chronology and the grade and style of any Cambrian tectonism.

both lateral and vertical crustal displacements during the progress of the 550-500 Ma tectonism in the LHB/DML sector of East Antarctica, compatible with coUisional orogenesis.

Juxtaposition of distinct Precambrian ('Grenviilian') crustal provinces

PRYDZ BAY REGION

The 550-500 Ma or 'Pan-African' high-grade zones summarized above juxtapose crustal terrains or provinces that preserve distinct Mesoproterozoic and Neoproterozoic geological histories and hence are unlikely to have formerly been continuous. Fitzsimons (2000a) has elegantly summarised the essential differences between the isotopic / event records of three broadly 'Grenville'-aged basement provinces that are separated by the LHB/DML, Prydz and Denman belts. Further examination of such records indicates that at least four distinct provinces should be distinguished, the three noted by Fitzsimons (2000a) - the Maud, Rayner and Wilkes provinces - and the less extensive, enigmatic but equally important Rauer / Vestfold province. The basis for this four-fold division is presented below.

This region, initially considered to be dominated by ca. 1000 Ma tectonism, is now known from a variety of isotopic studies applied to a range of metasedimentary and igneous rocks (e.g. Zhao et al., 1995; Hensen & Zhou, 1995; Carson et al, 1995; Fitzsimons et al, 1997) to have experienced its last major high-grade tectonic events at 535-510 Ma. This tectonism involved initial compression, interleaving probable basement terrains showing older histories (e.g. 960-920 Ma) with cover sequences, followed by extension accompanied by exhumation (Carson et al., 1995; Fitzsimons et al., 1997). Extensive melting in the metamorphosed cover rocks occurred both before and during extension and exhumation. As with the LHB region, crust initially was thickened and metamorphosed at considerable depths (7-8 kbar in Prydz Bay, and even up to 11 kbar if UHT in the Rauer Group can be attributed to the Pan-African events) and then underwent ITD to pressures of only 4 kbar prior to rapid cooling. By 490 Ma the presently exposed granulites of this area were already cooled below 300°C and probably resident at only <10 km depths in the crust. The metamorphic record, in particular the rapid implied exhumation at ca. 1 mm/yr or more, coupled with the deformational evidence for compression followed by extension, again point to major crustal displacements and are compatible with collisional orogenesis. Though not yet as well documented, 510-450 Ma metamorphism and intense deformation in parts of the Mawson Escarpment and in the Grove Mountains may represent continuations of the PanAfrican tectonism seen in Prydz Bay and provide further evidence for its effects on disparate crustal blocks.

THE MAUD, RAYNER AND WILKES PROVINCES

The major Proterozoic events in the Maud province in Dronning Maud Land and to the west of the LHB region are magmatism at 1150 Ma followed by regional high-grade metamorphism and deformation at 1080-1030 Ma and further thrustrelated deformation that may be as young as 990 Ma. This history contrasts dramatically with that preserved in the Rayner province, represented by the Rayner Complex of Enderby, Kemp and MacRobertson Lands, in which older crust (e.g. Archaean Napier Complex, 3800-2500 Ma; Mesoproterozoic anorthosites and other intmsives at 1450-1500 Ma), sediments and younger intmsives (e.g. chamockites at 960 Ma to 920 Ma) are variably deformed and metamorphosed in a series of events between 990 Ma and 920 Ma. Recent detailed geochronology along the Kemp and MacRobertson Land coasts support earlier indications that the main high-grade metamorphism in the west and northern Rayner Complex is 930-920 Ma in age (Kelly et al, 2000), somewhat younger than the 1000-980 Ma peak metamorphic event to the SE in the Northern Prince Charles Mountains. This age polarity could reflect development of the Rayner Complex during progressive or oblique collision between Antarctica, parts of India, and the Archaean Napier Complex. The third distinct province recognized by Fitzsimons (2000a), the Wilkes Province, occurs to the east of the Denman Glacier area notwed above and correlates very well with the Albany-Fraser belt of west Australia. The Wilkes province lacks evidence for any high-grade tectonothermal events

DENMAN GLACIER REGION

The third area in which Cambrian high-grade tectonism is likely to have been important is the Denman Glacier area, to the west of the Bunger Hills (Black et al., 1993; Sheraton et al., 1993). This area is correlated, on the basis of general continental reconstructions as well as the available geochronology of both the region itself and bounding areas to the west and east, with the Darling Mobile Belt of Western Australia. However, although a 520-500 Ma isotopic and rock record is present, the geology of the Denman Glacier region and the relationships between these ages and the principal high-grade deformation fabrics remain to be clarified. This area must be accorded priority as a major target for future geological field programmes

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

is clear that the ca. 1000 Ma isotopic record in this domain is associated with granulite facies metamorphism and deformation, and that this record is distinct fi-om that seen in the Rayner and Wilkes provinces. This, and the possibility that contrasting Archaean and Proterozoic high-grade gniesses have been interleaved and co-deformed in the Prydz Bay Pan-African tectonism, again testifies to the importance of this Cambrian event in the final amalgamation of East Antarctica. The lack of any record of 1000 Ma metamorphism and deformation in the nearby Vestfold Hills also supports the contention that these were not adjacent to the Rauers until 500 Ma.

younger than 1130 Ma, is dominated by lowmedium pressure granulite metamorphism and the emplacement of chamockitic and other intrusions at 1210-1150 Ma, and in some places (e.g. Windmill Islands: Post et al., 1997) preserves evidence for earlier tectonic events at 1340-1280 Ma. This Mesoproterozoic geological history is clearly very distinct from and bears no relation to those of the Maud and Rayner provinces.

A FOURTH DISTINCT PROVINCE: THE RAUER PROVINCE AND THE VESTFOLD DOMAIN

The fourth disparate and distinctive Precambrian crustal province that was involved in the final amalgamation of East Antarctica in the Cambrian is represented by the Rauer Islands and the neighbouring Vestfold Hills. These crustal blocks, which themselves appear to share very little in common in terms of their Precambrian geological histories (e.g. Harley et al., 1995, 1998), lie on the eastern margin of the Prydz Bay Pan-African tectonic zone. The Vestfold Hills only saw highgrade deformation and metamorphism in the latest Archaean to earliest Proterozoic 2520-2480 Ma in two phases of rapid crustal accretion (Snape et al., 1997), and even preserves Archaean to early Proterozoic cooling ages (2460 Ma) in monazites from gneisses re-deformed in younger zones of retrogression. The nearby Rauer Islands preserve evidence for distinct Archaean and Mesoproterozoic events that do not correlate in time with events in the Vestfold Hills, and the Rayner and the Wilkes provinces, variably overprinted by the 550-500 Ma tectonothermal events that dominate the areas further SW in Prydz Bay. The Archaean in the Rauer Islands is dominated, on the basis of present data, by >3300 Ma and ca. 2840-2800 Ma orthogneisses. These gneisses, extensively dyked and then deformed again in younger events, are interleaved with Mesoproterozoic metasediments and 1030-1000 Ma felsic to mafic intrusives that have also experienced high-grade metamorphism. The intensity of deformation post-dating the latter intrusives and various generations of pegmatite and ca. 1000 Ma leucogranitoids is highly variable, and many highstrain zones may reflect reactivation of the gneissic fabrics in overprinting Pan-African deformation. Despite the extensive isotopic evidence for highgrade metamorphism in the Rauer Islands at 550-500 Ma (e.g. Hensen & Zhou, 1995, 1997), monazites from three paragneisses that on structural grounds were metamorphosed prior to 1000 Ma (the age of cross-cutting leucogranites) still preserve cooling ages of nearly 1000 Ma. Although the issue of the relative grade and importance of the Pan-African event in the Rauer Islands is still a contentious one, from the perspective of the amalgamation history it

INSIGNIFICANCE OF ASSEMBLY EVENTS BETWEEN 920 Ma AND 550 Ma

Having established that the 550-500 Ma tectonic events in East Antarctica occurred at high grade and affected crustal provinces with distinct crustal histories, the possibility that there was an intervening event or set of events that stitched two or more of those provinces together after 1000-920 Ma and prior to 550 Ma must be considered. In this regard the possibility of a regional tectonic event at ca. 800 Ma, proposed by some workers on the basis of Rb-Sr isochrons / errorchrons and Sm-Nd mineral dating requires discussion. Black et al. (1987) in their study of the Rayner Complex recognized a period of igneous activity at 780-770 Ma, an age supported by recent work (Shiraishi et al., 1997). However, there are no ca. 800 Ma ages that can unequivocally be ascribed to a metamorphic episode or important penetrative deformation event in any of the four 'Grenville'aged provinces described above. Rb-Sr isochrons in these polydeformed rocks must be treated with caution and are most likely mixing lines, as proven by recent SHRIMP studies that yield Mesoproterozoic ages for orthogneisses previously assigned ca. 800 Ma ages on the basis of Rb-Sr isochrons. Likewise, gamet Sm-Nd mineral isochrons can be demonstrated from a number of studies in the East Antarctic Shield and elsewhere to yield ages considerably younger (by up to hundreds of Ma) than those of the metamorphism that they purportedly date. In view of these caveats and the absence of definitive age evidence from metamorphic zircons, for example, the case for a major period of crustal development and terrane amalgamation at ca. 800 Ma is very weak.

CONCLUDING REMARKS

This overview of the geology of East Antarctica has emphasized the key lines of evidence for its late assembly and amalgamation from disparate fragments at 550-500 Ma. Whilst earlier, pre-920

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Harley S.L., Snape I. & Black L.P., 1998. The early evolution of a layered metaigneous complex in the Rauer Group, East Antarctica: evidence for a distinct Archaean terrane. Precamb. Res. 89, 175-205. Harley S.L., Snape 1. & Fitzsimons I.C.W., 1995. Regional correlations and terrane assembly in East Prydz Bay: evidence from the Rauer Group and Vestfold Hills. Terra Antartica 2, 49-60. Hensen B.J. & Zhou B., 1995. A Pan-African granulite facies metamorphic episode in Prydz Bay, Antarctica: evidence from Sm-Nd garnet dating. Aust. J. Earth Sci. 42, 249-258. Hensen B.J. & Zhou B., 1997. East Gondwana amalgamation by Pan-African collision? Evidence from Prydz Bay, Eastern Antarctica. In: Ricci C.A. (ed.). The Antarctic Region: Geological Evolution and Processes, Terra Antartica Publication; Siena, 115-119. Kelly, N.M., Clarke, G.L., Carson, C.J. & White, R.W., 2000. Thrusting in the lower crust: evidence from the Oygarden Islands, Kemp Land, East Antarctica. Geol. Mag. 137, 219234. Kinny P.D., Black L.P. & Sheraton J.W., 1993. Zircon ages and the distribution of Archaean and Proterozoic rocks in the Rauer Islands. Antarctic Sci. 5, 193-206. Paech H.-J., 1997. Central Dronning Maud Land: Its history from amalgamation to fragmentation of Gondwana. Terra Antartica 4, 41-49 Post N.J., Hensen B.J. & Kinny P.D., 1997. Two metamorphic episodes during a 1340-1180 Ma convergent tectonic event in the Windmill Islands, East Antarctica. In: Ricci C.A. (ed). The Antarctic Region: Geological Evolution and Processes, Terra Antartica Pubhcation; Siena, 157-161. Sheraton J.W., Tingey R.J., Black L.P. & Oliver R.L., 1993. Geology of the Bunger Hills area, Antarctica: implications for Gondwana correlations. Antarctic Sci. 5, 85-102. Shiraishi K., Ellis D.J., Hiroi Y., Fanning C.M., Motoyoshi Y. & Nakai Y., 1994. Cambrian orogenic belt in East Antarctica and Sri Lanka: implications for Gondwana assembly, J. Geol. 102,47-65. Shiraishi K., Ellis D.J., Fanning C.M., Hiroi Y., Kagami H. & Motoyoshi Y., 1997. Re-examination of the metamorphic and protolith ages of the Rayner Complex, Antarctica: evidence for the Cambrian (Pan-African) regional metamorphic event. In: Ricci C.A. (ed), The Antarctic Region: Geological Evolution and Processes, Terra Antartica Publication; Siena, 79-88. Snape I.S., Black L.P. & Harley S.L., 1997. Refinement of the timing of magmatism and high-grade deformation in the Vestfold Hills, East Antarctica, from new SHRIMP U-Pb zircon geochronology In: Ricci C.A. (ed). The Antarctic Region: Geological Evolution and Processes, Terra Antartica Publication; Siena, 139-148. Zhao Y., Liu X., Song B., Zhang Z., Li J., Yao Y. & Wang Y., 1995. Constraints on the stratigraphic age of metasedimentary rocks from the Larsemann Hills, East Antarctica: possible implications for Neoproterozoic tectonics, Precamb. Res. 75, 175-188.

Ma, 'stitching' events did occur, particularly between Archaean and younger domains that show overprinting by tectonism in the Rayner province, the contrasts between the 'Grenville'-aged events recorded in the Maud, Rayner, Rauer and Wilkes provinces provide very firm evidence that, as pointed out by Fitzsimons (2000b), East Antarctica was not a single entity or shield, a unified crustal block in the edifices of East Gondwana or Rodinia, until the Cambrian. Important further goals that arise from this new view of East Antarctica include the assessment of how the three (or more) Pan-African tectonic zones link together, the documentation of the geological record in the as yet little known Denman Glacier area, and definition of where the Prydz Bay Cambrian belt and the adjacent Rauer province continued in India (or elsewhere) prior to its collision with Asia.

References Black L.P., Harley S.L., Sun S.S. & McCulloch M.T., 1987. The Rayner Complex of East Antarctica: complex isotopic systematics within a Proterozoic mobile belt. J. Metamorphic Geol. 5, 1-26. Black L.P., Sheraton J.W., Tingey R.J. & McCulloch M.T., 1992. New U-Pb zircon ages from the Denman Glacier area, East Antarctica, and their significance for Gondwana reconstruction. Antarctic Sci. 4, 447-460. Carson C.J., Dirks P.H.G.M., Hand M., Sims J.P. & Wilson C.J.L., 1995. Compressional and extensional tectonics in low-medium pressure granulites from the Larsemann Hills, East Antarctica. Geol. Mag. 132, 151-170. Fitzsimons, I.C.W., 2000a. Grenville-age basement provinces in East Antarctica: Evidence for three separate collisional orogens. Geology 28, 10, 879-882. Fitzsimons, I.C.W., 2000b. A review of tectonic events in the East Antarctic Shield and their implications for Gondwana and earlier supercontinents. J. African Earth Sci. 31, 3-23. Fitzsimons I.C.W., Kinny P.D. & Harley S.L, 1997. Two stages of zircon and monazite growth in anatectic leucogneiss: SHRIMP constraints on the duration and intensity of PanAfrican metamorphism in Prydz Bay, East Antarctica. Terra Nova 9, 47-51. Eraser, G., McDougall, L, EUis, D.J. & Williams, I.S., 2000. Timing and rate of isothermal decompression in Pan-African granulites from Rundvagshetta, East Antarctica. J. Metamorphic Geol. 18,441-454. Harley S.L., 1998. On the occurrence and characterisation of ultrahigh-temperature crustal metamorphism. In: Treloar P.J and O'Brien P.J. (eds), What Drives Metamorphism and Metamorphic Reactions ? Geol. Soc. London, Spec. Publ. 138,81-107.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

WAS THE KALAHARI SHIELD ATTACHED TO LAURENTIABALTICA DURING THE GRENVILLEAN OROGENY? C. Hatton De Beers GeoScience Centre, P.O. Box 82232, Southdale 2135, South Africa

Natal terrane. In a further extensional event at 1.06 Ga the Kalahari shield rifted along its northwestem margin, and separated from Laurentia-Baltica. The Kalahari shield was thus attached to LaurentiaBaltica only during the early phases of the Grenvillean orogeny. The rapid alternation between rifting and compression that is observed in both the NamaquaNatal and Grenvillean orogenies is best explained by over-riding of mantle plumes during accretion. The high heatflow observed in the Namaqua terrane (Robb et al, 1999) provides the clearest evidence for the presence of unusually hot mantle during the Namaqua-Natal event.

The Ghanzi-Chobe terrane, lying on the northwestern margin of the Kaapvaal and Zimbabwe cratons, which together constitute the nucleus of the Kalahari shield, has been interpreted as a magmatic arc (Watters, 1997). Penetrative foliation within the 1343±6 Ma Zongwe orthogneiss of southern Zambia (Hanson et al, 1988) is considered to be related to the accretion of the Ghanzi-Chobe terrane to the cratonic nucleus. Extension within the GhanziChobe terrane is recorded by the basaltic andesite and rhyolite of the Haiber Flats formation of the Sinclair group, respectively dated at 1086±44 and 1038±74 Ma (Hoal et al., 1989). The northwestern boundary of the Ghanzi-Chobe terrane is now defined by a suture formed during the Damara orogeny. Syntectonic granites associated with this suture are dated at 554±16 Ma (Kroner, 1982) and 551±19 Ma (Hanson et al, 1993), and the inferred extension of this suture into northern Zimbabwe is associated with resetting at 543±33 Ma (Vinyu et al., 1997). The nature of the terrane that lay to the northwest of the Ghanzi-Chobe terrane in the Mesoproterozoic is therefore unknown. The Namaqua-Natal terrane lies on the southern margin of the Kalahari shield. This terrane probably accreted to the older nucleus during the emplacement of the Nababeep and Modderfontein gneisses at 1212±11 and 1199±12 Ma, respectively (Robb et al, 1999). In the Namaqua terrane these early gneisses are associated with Palaeoproterozoic basement, but in Natal Palaeoproterozoic basement is absent. Extensive magmatism in a later event is marked by the emplacement of granites of the Oribi Gorge suite in Natal and the Spektakel suite in Namaqualand (Bomela granite, 1068±2 Ma; Bloukop granite, 1065±2 Ma, Thomas et al., 1996). A late phase of compression may be recorded by zircon overgrowths at 1032±12 Ma (Robb et al., 1999). Palaeogeographic reconstructions allow the possibility that the Kalahari shield was attached to Laurentia-Baltica at 1.2 Ga, but not to the reconstituted Laurentia-Baltica of 1 Ga. A possible sequence of events is the following. The Kalahari shield accreted to Laurentia-Baltica at 1.34 Ga. In an early extensional phase within Laurentia-Baltica the southeastern margin of the Kalahari shield rifted off, and by 1.2 Ga was replaced by the juvenile

References Hanson R.E., Wilson TJ., Brueckner H.K., Onstott T.C., Wardlaw M.S., Johns C.C., Hardcastle K.C. 1988. Reconnaissance geochronology, tectonothermal evolution, and regional significance of the middle Proterozoic ChomaKalomo block, southern Zambia. Precambrian Research 42, 39-61. Hanson R.E., Wardlaw M.S., Wilson T.J., Mwale G. 1993. U-Pb zircon ages from the Hook granite massif and the Mwembeshi dislocation zone: constraints on Pan-African deformation, plutonism and transcurrent shearing in central Zambia. Precambrian Research 63, 189-209. Hoal B.G., Hoal K.E.O., Boyd F.R., Pearson D.G. 1995. Age constraints on crustal and mantle lithosphere beneath the Gibeon kimberlite field, Namibia. South African Journal of Geology 98, 112-118. Kroner A. 1982. Rb-Sr geochronology and tectonic evolution of the Pan-African belt of Namibia, southwestern Africa. American Journal of Science 282, 1471-1507. Robb L.J., Armstrong R.A., Waters D.J. 1999. The history of granulite-facies metamorphism and crustal growth from single zircon U-Pb geochronology: Namaqualand, South Africa. Journal of Petrology 40, 1747-1770. Thomas R.J., De Beer C.H., Bowring S.A. 1996. A comparative study of the Mesoproterozoic late orogenic porphyritic granitoids of southwest Namaqualand and Natal, South Africa. Journal of African Earth Sciences 23, 485-508. Vinyu M., Martin M., Bowring S., Hanson R., Jelsma H., Dirks P. 1997. Tectonothermal evolution of the polymetamorphic Zambezi belt in NE Zimbabwe: constraints from single grain zircon data. Abstracts, 17th Colloquium on African Geology and Conference on Intraplate Magmatism and Tectonics of Southern Africa, p. 52. Watters B.R. 1977. The Sinclair-group: Definition and regional correlation. Transactions of the Geological Society of South Africa 80, 9-16.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

THE NEOPROTEROZOIC SNOWBALL EARTH HYPOTHESIS: TRIALS AND TRIBULATIONS P.F. H o f f m a n Department of Earth & Planetary Sciences, Harvard University, Cambridge, MA 02138, USA.

overlain directly by unusually thick, deep-water sequences. 'Cap carbonates' are the transgressive systems tracts of such sequences, and are a predictable response to the alkalinity flux from carbonate and silicate weathering during the ultragreenhouse transient. Cap carbonates contain a peculiar ensemble of sedimentary structures. Reverse-graded peloids (deep-water ooids) and giant wave ripples (false 'tepees') in may be related to hypercanes, expected where sea-surface temperatures >28°C. Primary barite hard-grounds and reef-like masses of former aragonite crystal fans in cap carbonates imply unusual seawater chemistry. Transfer of water loads from ocean basins to continents during the more heavily-glaciated later stages of a snowball episode will cause spatially variable base-level falls at continental margins due to lithospheric flexure. Incised paleovalleys with local relief of <380 m are associated with the Chuos glaciation in Namibia. If the tropical ocean was ice covered, glacial debris will collect on the flanks of low-lying carbonate shelves and platforms lacking nucleation sites for mountain glaciers, as in the Ghaub glaciation in Namibia. A characteristic pattern of large-amplitude deviations is observed before and after glaciation in many regions, and constitutes a strong biogeochemical argument against diachronous, regionalized glaciation. The isotopic pattern can be explained in detail by the snowball hypothesis, and the final preglacial deviation is one possible 'smoking gun'. The greatest weakness of the hypothesis is that we do not know the true nature of a snowball Earth in sufficient detail. Claims for observations contradictory to the hypothesis may overestimate our present understanding (a statement that applies equally to the preceding paragraph). We do not know the extent of dynamic ground-based glaciers. Nor do we know if the enhanced temperature fluctuations characteristic of frozen planets (e.g. Mars and snowball Earth) could account for periglacial patterned ground at low paleolatitudes, without recourse to high orbital obliquity. Does the apparent lack of high-latitude glacial deposits reflect distance from moisture sources or an absence of polar continents? Would seawater ^^Sr/^^Sr and be buffered during a snowball episode by carbonate dissolution, in response to lower pH due to submarine hydrothermal activity? Does the

The snowball Earth hypothesis was first proposed by Joe Kirschvink in 1989 as a possible explanation for equatorial glaciation at sea level in the late Neoproterozoic. He invoked a fundamental instability in the climate system caused by icealbedo feedback. This instability emerged from early energy-balance climate models and is replicated in a variety of GCMs. If sea ice ever reached the tropics, runaway albedo feedback would drive ice lines rapidly to the equator and global mean temperature would drop catastrophically. Kirschvink suggested that plate tectonics would be the Earth's salvation, because CO2 levels would rise inexorably due to volcanism and metamorphism, assuming that silicate weathering (which consumes CO2) was seriously curtailed while the land surface was frozen. Although the onset and termination of a snowball episode would occur abruptly, particularly in the tropics (years to decades), the snowball stage itself could last for millions of years due to the huge hysteresis in CO2. Diurnal and seasonal temperature fluctuations would be enhanced at all latitudes because of the low heat capacity of the icy surface and weak meridional heat transport. In the longterm, mean annual temperatures would slowly rise (non-linearly) due to progressive increase in CO2. The aftermath of a snowball episode would be very warm because high levels of CO2 (>0.1 bar) are required to overcome a snowball albedo. Enhanced silicate weathering would eventually restore the atmospheric greenhouse to a new steady state. Kirschvink speculated that snowball events only occur when most continents are at low latitudes, consistent with their absence in the Phanerozoic. Kirschvink's hypothesis was originally conceived to account for glaciomarine deposits close to the paleo-equator and the co-occurrence of banded iron-formation in the late Neoproterozoic. Many additional observations can be explained by deduction from the original hypothesis. The albedo feedback instability predicts that glacial episodes begin and end abruptly. The long duration of a snowball episode is supported by magnetostratigraphic data for the Marinoan glaciation in South Australia. Long-lasting glacial episodes with low average sedimentation rates will cause angular unconformities (in tectonically active areas) to be localized at glacial horizons. In areas of ongoing tectonic subsidence, glacial intervals will be

Geological Society of Australia Abstracts 65

Rodinia Symposium. Perth October 2001

deposition of banded iron-formation during glaciation imply that equatorial sea ice was too thin (<20 m), discontinuous, or impersistent to prevent oxygenic photosynthesis? This would provide the

refugium needed for eukaryotic algae and possible stem-group metazoa.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

LANDSAT TM AND DIGITAL ELEVATION MODELS AS REGIONAL MAPPING AND EXPLORATION TOOLS: THE STEP FROM 2D TO 3D D.A. Hollingsworth and P.A. Cawood Tectonics Special Research Centre, Department of Applied Geology, Curtin University of Technology, GPO Box U1987, Perth WA 6845, E-nnail: Hollingd@lithos.curtin.edu.au

displayed in either "true colour" or colour infrared, mimicking the features observed in colour or colour infrared aerial photographs, respectively. The second feature of the data is that it extends beyond the visible and near-infrared (NIR) into the MIR region - making it possible to highlight some unseen (in the visible to NIR) lithological variations. The third feature of Landsat TM is that, when combined with a DEM it allows 3D "perspective" views of the data as opposed to aerial photography which has a fixed stereo viewpoint. The advantage of being able to control the viewpoint becomes obvious when dealing in areas where both topography and structure effect outcrop pattern. With the use of a DEM it is possible for the geologist to view topography from a low angle, as if looking from one hill to the next, thereby "removing" the effects of topography. The fourth, and most important feature, is that the data can be tailored to suit individual needs by looking at combinations or ratios of different wavelength bands. When generating an image using Landsat TM data, it is necessary to assign a band (or mathematical combination of bands) to a colour (red, green or blue). Using the ratio of two bands as opposed to a single band often enables features to be highlighted. By using band ratios, it is possible to highlight such things as variations in vegetation, water, drainage systems, rock type, and structure. Regional-scale mapping can be undertaken rapidly if the stratigraphy can be correlated with other mapping in the area, or if "ground truthing" is carried out within the area of interest so that lithology can be matched with the Landsat TM image. The combination of Landsat TM with other data sources such as gravity or aeromagnetic data is easily achieved using modem software packages. The ability to quickly overlay data from various sources allows better recognition of possible target sites.

INTRODUCTION Landsat Thematic Mapper (TM) imagery is a valuable tool for the exploration or mapping geologist. It allows a quick and relatively inexpensive plan view of any area (worldwide) prior to fieldwork or geophysical surveys. In the Hamersley Province of Western Australia, an area of low rainfall and sparse vegetation, it has been successfully used in regional mapping surveys. In addition to rock types, drainage, structure, and vegetation have been defined from the imagery. With the addition of Digital Elevation Models (DEM's), it's now possible to view the interaction between lithology, structural features and topography.

GENERAL PRINCIPLES Rapid advances in technology over the past 10 years have seen the use of digital remote sensing data move from the domain of the large workstation computers to the average desktop PC. Where once gravity and magnetic surveys were the most common digital data sets, Landsat TM data and DEM's are now used widely in exploration. Landsat TM offers many advantages over the aerial photographs that are widely used in the construction of interpretive maps during the early stages of exploration. The principal advantages are that Landsat TM is a digital data set and that it extends beyond the visible spectrum into the mid-infrared (MIR) region. In an area such as the Hamersley Province of Western Australia (WA), aerial photographs tend to be discoloured through a ubiquitous iron-rich red dust that colours the rocks. By imaging beyond the visible light region of the spectrum, Landsat TM is able to minimize the effects of the discolouration and reliably distinguish between rock types. As Landsat TM data is a digital data set, it can be quickly and easy combined with other data sets over the same area. Landsat TM is readily available for the entire area of WA, and gives geologists the ability to quickly see relationships between regional fold belts, shear zones, rock types, and local features. The versatility of digital Landsat TM data lies in four aspects of the data. Firstly, the data can be

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

the Landsat TM data allowed recognition of regionally significant structures. Predictably, there are differences between published data and Landsat TM interpretation in other areas of the Hamersley Province that will require fieldwork to resolve - particularly in the mapping of BIF. The Marra Mamba Iron Formation, at the base of the Hamersley Group, contains more chert and less iron oxides than the Brockman Iron Formation. On the imagery, the two BIF units appear significantly and consistently different. However, the geologic maps that were compiled from black and white aerial photographs, which rely heavily on outcrop pattern, have not made the distinction between the two units in some areas. Thus, the Landsat images have provided information, particularly in the MIR, that was not available to the initial mapping work.

APPLICATION TO THE HAMERSLEY PROVINCE, WESTERN AUSTRALIA The initial aim of the data processing was to test the capability of Landsat TM to differentiate rock types (major and minor) within the Hamersley Province of Western Australia. This area is ideal for using Landsat TM data because there are large lithological/mineralogical variations in the stratigraphy, which are continuous over the entire province. The stratigraphy includes the Fortescue (clastic sediments, mafic volcanics, and shales), Hamersley (banded iron formation, shale, minor dolomite, and volcanics), Turee Creek (shale, sandstone, volcanics), and Wyloo (sandstone, conglomerate, mafic volcanics) Groups. Vertical relief of up to 600 m in the Hamersley Province also facilitates exposure of the recessive rock units and provides a good opportunity for the use of a DEM to better interpret the structure of the area. To test the capabilities of Landsat TM, a number of three band combinations and ratios were visually tested using ER Mapper 5.5. The best combination involved Band 7 (MIR), Band 5 (MIR), and Band 1 (visible - blue). This outperformed even the "standard" 7/5, 3/1, and 5/4 ratio combination for lithology delineation. The most notable response occurs in the Hamersley Group, which comprises Banded Iron Formation (BIF), dolomite, shale, rhyolite and dolerite sills. Shale units as thin as 50 m can be recognised from the Landsat images, and traced for lO's of kilometers along and across strike. In order to determine that the colour variation seen in the Landsat TM data was directly related to the underlying lithology, images were compared with detailed surface maps. There were no major discrepancies noted between field surveys and interpretation of the generated images. In addition.

CONCLUSIONS Landsat TM imagery was successfully used to define the lithology and structure in the Hamersley Province of Western Australia. It is important to note that regional mapping using Landsat TM imagery may not always be as successful as in the Hamersley Province, which features a wide range of fundamentally different lithologies and bulk-rock chemical composition. In areas which contain rocks with less lithological contrast, for example an area comprising primarily different sandstone units, the delineation of rock types from Landsat TM imagery is often more difficult. However, in nearly all terrains, structural features such as faults, folds, and shear zones can be seen in the images. In addition, mapping of large areas can be completed far quicker than by either field mapping or aerial photograph interpretation, at a scale that is convenient for the user.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

RECONSTRUCTION OF THE ARCHAEAN JIMBLEBAR GREENSTONE BELT, SYLVANIA INLIER, WA. D.A. Hollingsworth, P.A. Cawood and B. Monek Tectonics Special Research Centre, Department of Applied Geology, Curtin University of Technology, GPO Box U1987, Perth WA 6845. E-mail: Hollingd@lithos.curtin.edu.au.

granite-greenstone terrane to be accurately reconstructed. The first three of these deformation events were ductile compressive events, with the first event (Dl) producing thrusts, sheath folds and a pervasive fabric. The remaining ductile events produced open to tight folding (D2 and D3). The remaining three events are fault- and shear-related events. The major faulting event (D4) in the greenstone belt pre-dates emplacement of the Sylvania granitoids, and the D5 and D6 shear zone events are confined to the granitoid rocks and pre-date the 2.8 Ga unconformity with the overlying Mount Bruce Supergroup.

INTRODUCTION The Sylvania Inlier is a part of the Pilbara granite-greenstone terrane that is exposed to the south of the Hamersley Province of Western Australia (Fig. 1). The Inlier represents a basement high that was structurally active at around 2200 Ma during the Ophthalmian Orogeny; however, until recently the structural history prior to this event was poorly understood. Recent work within the Jimblebar Greenstone Belt (Fig. 2) in the eastern part of the inlier, has revealed a long and complex deformation history prior to 2800 Ma, and the identification of sedimentary structures (Monek, 1999) and igneous layering in mafic plutons (Tyler, 1991) has enabled the recognition of a coherent stratigraphy. A pre-existing basement of diorite and granodiorite is inferred from its presence both as clasts within sedimentary rocks of the greenstone sequence and as xenoliths within the surrounding granitoids. The lowermost mappable unit comprises quartz-gamet-amphibole schist, for which an intermediate volcanic protolith is inferred. This unit is succeeded by a sequence of siliciclastic and chemical sedimentary rocks, interleaved with mafic and ultramafic volcanic flows. This supracrustal sequence was then intruded by large volumes of mafic to ultramafic magma, resulting in formation of layered intrusions. Relict igneous textures are preserved in isolated low-strain zones throughout the beh, allowing the recognition of both intrusive and extrusive mafic to ultramafic rocks. Sedimentary rocks present within the belt include conglomerate, quartzite, pelitic schist, chert and banded iron formation.

Figure 1. Regional map showing the location of the Sylvania Inlier in relation to the major exposure of the Pilbara granite-greenstone terrane.

STRUCTURAL FRAMEWORK Within the Jimblebar Greenstone Belt and adjacent granitoid rocks of the Sylvania Inlier, six deformation events are recognised that pre-date deposition of the -2115 - 2300 Ma Mount Bruce Supergroup. The relative deformation chronology was determined from consistent crosscutting and overprinting relations observed throughout the greenstone belt and adjacent granitoid rocks. These relationships permit the progressive restoration of successive deformation events, providing insight into the original depositional relationship of the rock units. This allows the initial assembly of the

RECONSTRUCTION Restoration of the Jimblebar Greenstone Belt to its pre-deformed state was not possible due to complex deformation and poor preservation of the earliest events. Restoration to the pre-D2 event was achieved in six discrete stages. Stage one was the restoration of young (<750 Ma) north-west trending oblique slip faults that cross-cut the greenstone belt

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

"Mount Bruce Supergroup' 1(2775-2300 Ma):

Post-Dg basins Mount Bruce Supergroup

Jimblebar Greenstone Belt (pre-D^) Layered Intrusions

Unconformity

II r ' "

Ultramafic Gabbro

Post-D4 intrusions Megacrystic Granitoid

Fault

Granitoid

Undifferentiated Sediments Basalt Intermediate volcanics

Figure 2. A simplified map showing the distribution of rock units within, and adjacent to, the Jimblebar Greenstone Belt. restoration of a major south-east plunging D3 anticline. The anticline and associated parasitic folds are responsible for the curved shape of the greenstone belt. Stage six was the restoration of the syncline that separates the western and eastern parts of the belt. Removal of the syncline shows a greenstone stratigraphy that is disrupted by both thrust faults and associated drag features as well as early extensional features. A lack of similar units across the major D2 structure has made it impossible to restore the earliest events recognised within the Jimblebar Greenstone Belt

and surrounding granite. Displacement on these structures was minor and restoration involved matching of units on either side of the faults. Stage two involved the restoration of structures related to the formation of the ^2200 Ma Ophthalmia Fold Belt in the Mount Bruce Supergroup to the north of the Sylvania Inlier. Structures related to the Ophthalmian Orogeny trend east-west and reactivate both earlier D4 structures and also the margins of the greenstone belt in the west. The extent of D4 reactivation is quantifiable using offset on the postD4 granitoids that pre-date the Ophthalmian Orogeny. Restoration of the Ophthalmian related structures is done using the post-D4 granitoids and some interpretation of their sub-surface extent is required. Stage three involved the removal of the granitoid rocks that surround the greenstone belt. Removal of these granitoids leads to speculation as to the lateral extent of the greenstone stratigraphy prior to granite emplacement. Stage four was the restoration of a large D4 fault that separated the western and eastern portions of the greenstone belt. Drag on units adjacent to the D4 structure suggest a significant component of lateral movement on the structure. Restoration was achieved by assuming no major change in unit distribution due to vertical movement. Stage five concentrated on the

CONCLUSION The age and complex deformation history preserved in many Archaean granite-greenstone terranes often renders structural restoration difficult or impossible, and the amenity of the Sylvania Inlier to such reconstruction has important implications for the evolution of poly-deformed greenstone sequences elsewhere. In addition, unraveling of such deformation histories provides insight into the nature of deformation styles and processes during the Archaean. The crustal architecture of the Sylvania Inlier contrasts with the older (pre-3000 Ma) dome-and-keel provinces of the northern Pilbara (e.g. Hickman, 1984), and the existence of early structures in the Jimblebar Greenstone Belt

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

References

which suggest compression in a ductile deformation regime possibly indicates a significant horizontal component in tectonic force. The Jimblebar Greenstone Beh is older than most Archaean provinces for which terrane accretion models have been proposed (e.g. eastern Yilgam; Campbell & Hill, 1992) and the possibility that the Jimblebar Greenstone Belt represents an intermediary between such provinces and Early Archaean dome-and-keel terranes has important implications for the evolution of tectonic processes through the Archaean.

Campbell, I.H. & Hill, R.L (1992) A two-stage model for the formation of the granite-greenstone terrains of the Kalgoorlie-Norseman area, Western Australia. Earth and Planitary Science Letters 90, 11-25. Hickman, A.H. (1984) Archaean Diapirism in the Pilbara Block, Western Australia. In: Kroner A. & Greiling R. eds. Precambrian Tectonics Illustrated, pp. 113-127. E. Schweizerbartische Verlagbuchhandlung, Stuttgart. Monek, B. (1999) Structural history of the Jimblebar Greenstone Belt, Pilbara Craton,Western Australia. B.Sc.(Hons) thesis, Curtin University of Technology. Tyler, I.M. (1991) The geology of the Sylvania Inlier and the Southeast Hamersley Basin. Geological Survey of Western Australia, Bulletin 138.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

TECTONIC FRAMEWORK OF THE ITREMO SHEET AT -800 MA: IMPLICATIONS FOR THE POSITION OF CENTRAL MADAGASCAR IN RODINIA B. Hulscher\ C.McA. Powell and LC.W. Fitzsimons^ ^ Tectonics Special Research Centre, Department of Geology and Geophysics, The University of Western Australia, 35 Stirling Hwy, Crawley, WA 6009, Australia ^Tectonics Special Research Centre, Department of Applied Geology, Curtin University of Technology, GPO Box U1987, Perth WA 6845

RODINIA

and location relative to Central Madagascar basement. At present, the depositional age of the Itremo Group is poorly constrained (1855-800 Ma, Cox et al., 1998; Handke et al., 1999). With respect to the age of different lithological units in the basement. Tucker and others (1999) dated a tonalite gneiss underneath the Itremo Group at 2511 Ma. However, most of the top of this Archaean basement is formed by thick amphibolites. These thick dolerite and homblende-gabbro sills may well be genetically related to the formation of the Itremo sediments and not to the basement gneisses. The layered amphibolites are up to 800 m thick and contain hornblende, plagioclase, quartz and microcline (+biotite +sphene). (Up) towards the Itremo Group, they become interlayered with thin arenites and pehtes. At the tectonized contact with the official first occurrence (Moine, 1966) of the (thicker) 'Itremo' arenites, a layer parallel -15 m metadolerite sill occurs. Above the contact, the amphibolite layers thin in favour of increasingly abundant pelites and psammites. At even higher structural levels, where metamorphism and deformation are less penetrative, layer parallel basalt flows (<20 m) with amygdales and flow vesicles are preserved between hornblende bearing, biotite-rich pelites. We postulate here that the Itremo Group and the underlying thin sediment slivers and thick mafic sills are genetically related, and that they represent the onset of extension-related magmatism in Central Madagascar and the formation and infilling of a basin. As the formation age of both the Itremo group and the basal amphibolites is poorly constrained, we can envisage three scenarios for the development of this sequence, i.e., during: 1) the formation of the Mozambique Ocean, 2) widespread crustal extension at the break-up of Rodinia just prior to -800 Ma, or 3) the formation of a retro-arc basin during westward subduction undemeath Central

Many reconstructions of the late Mesoproterozoic supercontinent, Rodinia, place Madagascar attached to India in its younger Gondwanaland fit on the eastern side of a wide Mozambique Ocean that separates IndiaMadagascar-East Antarctica from the Congo-Sao Francisco block of Africa (e.g., Dalziel, 1997). When Rodinia broke up during the midNeoproterozoic, the Mozambique Ocean is postulated to have contracted by subduction beneath either or both of the India-Madagascar and east African margins while the Palaeo-Pacific Ocean grew between Australia-Antarctica and Laurentia. Final closure of the Mozambique Ocean is thought to have occurred in the late Neoproterozoic during the end-Precambrian formation of Gondwanaland. Based on observations on the sedimentary sequence of the Itremo Group and on basal amphibolites of the Itremo Sheet, we propose a new model in which Central Madagascar formed its own microcontinent at ca. 800 Ma, with the Mozambique Ocean subducting westward, resulting in the formation of a retro-arc basin and associated plutons at this time.

ITREMO SHEET UNITS

The Itremo Sheet (Collins et al., 2000) of Central Madagascar consists partly of the metasediments of the Itremo Group - a succession of quartzite, pelite and carbonate rocks more than 2 km thick. The Itremo Sheet tectonically overlies a series of reworked high-grade Archaean gneisses along a shallowly west dipping contact. The Itremo Group sediments have been previously interpreted as a passive margin (Cox et al, 1998) or platform sequence. This setting, and whether they were deposited directly onto the basement of Central Madagascar or whether the sequence is allochthonous, is debated. Our ongoing work on the contact between the Itremo Group and the so-called 'Archaean basement' is providing new clues to the type of basinal setting, and to its timing

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Madagascar when the Mozambique Ocean began to close at -^800 Ma. We favour the latter model, because a suprasubduction setting can explain not only the facies associations in and between the upper ('proper') Itremo Group and lower Itremo Sheet amphibolites, but also the patterns of deformation, metamorphism and intrusion found in both units around 800 Ma.

and formed between 804 and 770 Ma (Handke et al.. Tucker et al., 1999). These plutons are part of a much longer belt that can be traced over 1000 km through the Seychelles Islands into the Malani Igneous Province of NW India. The magmatic belt in Madagascar has been interpreted by Handke and others (1999) to be the roots of a former ca. 800 Ma continental magmatic arc.

Itremo Group Sediments

> -800 Ma Deformation

If the Itremo Group represents a passive margin sequence as proposed by Cox and others (1998), at the edge of Rodinia, one would expect coeval continental slope and rise deposits. Such evidence is lacking however. There is a facies change of thick, well sorted arenites intercalated with thin, two-mica pelites in the west, to abundant rhythmic-layered sandstones and siltstones and thinner, well sorted arenites in the east. The Eastern Itremo Group also contains hematite rich shales, and thicker biotite-rich pelites. While arenites dominate the Western Itremo Group, carbonaceous black muds and stromatolitic limestones cap the Eastern Itremo Group. These facies patterns are consistent with: • a basin axis towards the (south) west • infilling with source material from Central Madagascar (Archaean) gneisses • eastward transgression and onlap onto this basement, at which time the shallow marine limestones were deposited.

The style of deformation documented in the Itremo Group by Hulscher and others (2001) and Abello and Johnson (this volume), could reflect tectonic processes operating during arc formation. A penetrative deformation with isoclinal recumbent folds and an early layer-parallel cleavage is overprinted by low-P-high-T minerals in the contact aureoles of the ca. 800 Ma plutons (at < 4 kbar, > 650 °C). This indicates that either there was a separate episode of deformation before the plutons were emplaced, or that progressive deformation had occurred during intrusion. The first possibility raises the question of whether there is any Grenvilleaged late Mesoproterozoic deformation, which has been reported from the Mozambique belt (Costa et al., 1994) but not recognized in Madagascar before. In the second case, the isoclinal folds with penetrative fabrics formed early during the same tectonic event that emplaced the ca. 800 Ma plutons. There then are two possibilities. The first option is to consider that the early deformation could be due to widespread crustal extension. Although metamorphic facies are strongly attenuated, as expected with this model, this is due to the rapid injection of voluminous plutons under vertical rather than horizontal stretching conditions. Sheared pluton and aureole margins are often steep, but they are also sub-parallel and possibly reoriented by later deformation, making it difficult to distinguish between intrusion as largely layer parallel, sub-horizontal sheets, or as stocks that intrude the flat-lying fabric and the recumbent folds at a high angle, with ductile stretching of the wall rock leading to transposition into a vertical and thinned contact aureole. However, while later upright folding is indeed present, it is possible to trace older, tight, synintrusion upright folds close to the pluton, with subhorizontal fold axes, into the actual contact aureole, where the same fold axes rotate into vertical, parallel to a strong high temperature vertical stretching lineation that developed syn-intrusion in a stretched, vertical contact zone of both granite and wall rock (skam). In addition, the described layer parallel amphibolites underneath the Itremo Group preserve many angular cross cutting relationships by the

The high level Itremo graphitic schists, which are derived from black shale protoliths, are consistent with a restricted, epicontinental marine environment. Geochemistry by Raoelison (1997) indicates source rocks for the meta-pelites were granitic rocks and mafic volcanics. In summary, the Itremo Group could represent the infilling of a shallow, epicontinental, retro-arc basin. It is possible that this basin formed along a central axis that ran NW-SE and was located between present day Southwest and Central Madagascar. The Ranotsara Shear Zone may well have developed on other, reactivated NW-SE trending pre-existing fractures.

TECTONIC SETTING AT - 800 MA A supra-subduction setting for Central Madagascar and the sediments and amphibolites of the Itremo Sheet could explain the features outlined below.

- 800 Ma Plutons and Arc formation? The tectonic units of Central Madagascar (the Antananarivo Block and the Tsaratanana Thrust Sheet) and the Itremo Group are intruded by a 450 km-long meridional belt of gabbroic and granitoid batholiths. The gabbros and granitoids are coeval

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

The opposite is in fact true. The eastern Itremo terrane boundary and other, major internal faults dip westwards. This is the case for most terrane boundaries and major faults throughout Central Madagascar (when away from later vertical shear zones). While faults could have been reoriented during Gondwanaland amalgamation, pressure gradients have preservation potential. Nevertheless, the eastern Itremo Group margin sediments provide no evidence for westward thrust loading as they are low greenschist facies. Moreover, the grade of the early penetrative fabric increases westwards. Another altemative, proposed by several authors (Kroner et al., 2000; Collins et al., 2000), involves a subduction zone that dipped east from a trench that separated Central Madagascar from Southern India, now incorporated in the Betsimisaraka suture between Central Madagascar and the Antongil Block, a fragment of the Dharwar craton. This scenario is not consistent however with the presence of -800 Ma mafic plutons and granodiorites in the Tsaratanana Sheet, a rift complex or ophiolite remnant (Nicollet, 1990; Berhe, 1990) that occurs west of the proposed suture zone in Madagascar, at highest structural levels. In addition, the Antongil Block lies east of the Betsimisaraka suture but lacks -800-750 Ma plutons. This could be explained instead by the initiation of a westward dipping subduction zone under a Central Madagascar microcontinent, with westwards obduction of the Tsaratanana Sheet across the Antananarivo Block, just prior to the generation of the 790-780 Ma plutons that intrude both tectonic units.

much less deformed ca. 800 Ma granitoids and gabbroic feeder dykes. The second option, and the preferred tectonic model, is therefore that the penetrative fabric and recumbent folds are related to upper-crustal imbrication of the postulated continental magmatic arc (Handke et al. 1999) during contractional deformation, analogous to retro-arc thrusting in the Patagonian Andes. Both the extensional and contractional deformation model could account for the observation that heat locally outlasted deformation (Abello & Johnson, this volume); in the latter case through voluminous pluton injections into the Itremo Group, once incorporated in the arc. In the arc scenario, the formation of the (upper) Itremo Group and lower dolerite and gabbro sills could be explained by extension in a shallow retroarc setting, while the observed early structures could reflect later deformational contractional components.

Arc formation: Westward Subduction? Thus, if the Central Madagascar Antananarivo Block formed the basement to an Itremo retro-arc basin that lay to the west of it, the subduction zone is likely to have been west-dipping along Madagascar's eastern boundary. However, Handke and others (1999) argued for an east dipping subduction zone, with Madagascar at the outboard margin of Rodinia next to India, facing the Mozambique Ocean to the west. Eastward subduction on Madagascar's western edge is not consistent with the lack of -800 Ma plutons in South Madagascar (SW of the Ranotsara Shear Zone), which could well have been a part of the Madagascar microcontinent at this time, as it shares a 2.5 Ga affinity with the Antananarivo Block. In the westward subduction scenario however, the Itremo Group would have developed in an epicontinental setting between South Madagascar and the Antananarivo Block, and South Madagascar would possibly be too far from the (eastern) trench for arc plutons to intrude it. Secondly, the stacking order and dip direction of tectonic units in Madagascar is not consistent with an east dipping subduction zone; not at -800 Ma nor during Gondwanaland amalgamation. Even if the Itremo Group would have been formed as a passive margin sequence on the eastern outboard margin of Rodinia, instead of in an epicontinental basin, and if it would be incorporated into a magmatic arc above an east dipping subduction zone around -800 Ma, one still would expect relicts of eastward-increasing pressure gradients within the resulting sediment nappes, and of east dipping thrusts with evidence of burial undemeath basement that was overthrust towards the west, e.g., akin to the Western Cordillera basement and sediments of the North Chile-Argentina Andes.

MADAGASCAR: A RODINIA MICROCONTINENT? A scenario of westward subduction and continental magmatic arc formation leaves the question of whether Madagascar lay outboard on the African margin of Rodinia, or whether it lay close to India. In the latter case, the westward subduction implies it formed its own microcontinent, separated from the Dharwar craton by an oceanic fragment. A firm attachment of Central Madagascar to the African collage of continental blocks in Rodinia at the time of Rodinia break-up, as required by the continental extension model, seems unlikely, considering that Central Madagascar is straddled on each side by possible sutures. Indeed, if the Vohibory belt in the SW tip of Madagascar represents an ophiolite remnant, incorporated during collision at the end of the Neoproterozoic, it is more likely to have been another strand of the existing Mozambique Ocean, on the western side of a Madagascar microcontinent, than a new ocean fragment that formed if Madagascar rifted away from African Rodinia during Rodinia break-up.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

HULSCHER, B., COLLINS, A.S., DAHL, K.L., FITZSIMONS, LC.W., JOHNSON, S.P., JONSSON, M.K., PASSMORE, A.R. & POWELL, C.MCA. Evidence of -800 Ma and possibly older deformation and plutonism in Madagascar. GSA-SGTSG Abstracts. 64: 91-92. KRONER, A., HEGNER, E., COLLINS, A.S., WINDLEY, B.F., BREWER, T.S., RAZAKAMANANA, T. & PIDGEON, R.T., 2000. Age and magmatic history of the Antananarivo Block, central Madagascar, as derived from zircon geochronology and Nd isotopic systematics. Amer. J. Sci. 300: 251-288. MOINE, B. 1966. Grand traits structuraux du massif SchistoQuartzo-Calcaire (Centre-Ouest de Madagascar). Comptes Rendue de la Semaine Geologiques. 93-97. NICOLLET, C. 1990. Crustal evolution of the granulites of Madagascar: In: Vielzeuf, D., Vidal, Ph. (Eds.) Granulites and Crustal Evolution. Kluwer, Amsterdam. 291-310. RAOELISON, I.L. 1997. Structure and metamorphism of the Itremo group, Central Madagascar. Unpubl. MSc. Thesis. Rand Afrikaans University. TUCKER, R.D., ASHWAL, L.D., HANDKE, M.J., HAMILTON, M.A., LE GRANGE, M. AND RAMBELOSON, R.A. 1999. U-Pb geochronology and isotope geochemistry of the Archaean and Proterozoic rocks of north-central Madagascar. Joum. Geol. 107: 135-153.

References ABELLO, J. & JOHNSON, S.P. 2001. Significance of Neoproterozoic deformation and magmatism in the Itremo Group, Central Madagascar: evidence for the break-up of Rodinia. Rodinia Symposium Abstracts (this volume). BERHE, S.M. 1990. Ophiolites in northeast and east Africa: implications for Proterozoic crustal growth. Joum. Geol. Soc. Lond. 147:41-57. COLLINS, A.S., KRONER, A., RAZAKAMANANA, T. & WINDLEY, B.F. 2000. The tectonic architecture of the East African Orogen in central Madagascar - a structural and geochronological perspective. J. African Earth Sc. 30 (4a): 21.

COSTA, M., CADOPPI, P., SACCHI, R. & FANNING, C.M. 1994. U-Pb SHRIMP dating of zircons from Mozambique gneiss. Boll. Soc. Geol. Italy. 113: 173-178. COX, R., ARMSTRONG, R.A. & ASHWAL, L.D. 1998. Sedimentology, geochronology and provenance of the Proterozoic Itremo Group, central Madagascar, and implications for pre-Gondwana paleogeography. Jour. Geol. Soc. London. 155: 1009-1024. DALZIEL, I.W.D. 1997. Neoproterozoic-Paleozoic geography and tectonics: Review, hypothesis, and environmental speculation. Bull. Geol. Soc. Amer. 108: 16-42. HANDKE, M.J., TUCKER, R.D. & ASHWAL, L.D. 1999. Neoproterozoic continental arc magmatism in west-central Madagascar. Geology. 27: 351-354.

Rodinia S y m p o s i u m , P e r t h October 2 0 0 1

G e o l o g i c a l Society of Australia A b s t r a c t s 65

PALEOPROTEROZOIC BASEMENT IN NW SONORA, MEXICO: A KEY PORTION OF THE SW MARGIN OF LAURENTIA TO RECONSTRUCT THE RODINIA SUPERCONTINENT A . Iriondo^ ^ W.R. Premo^ a n d M J . Kunk^ ^U.S. Geological Survey, MS 974 Box 25046, Denver, Colorado 80225, USA ^University of Colorado at Boulder, Campus Box 399, Boulder, Colorado 80309, USA

Laurentia is an amalgamation of juvenile volcanic arc complexes of Paleoproterozoic (Mojave, Yavapai, and Mazatzal) and Mesoproterozoic (Grenville) crustal terranes juxtaposed along distinct tectonic boundaries (Fig. 1).

The SW margin of Laurentia is one of the key Neoproterozoic to Early Paleozoic rifted margins used in continental reconstructions of the Rodinia supercontinent (e.g., SWEAT, AUSWUS, Siberianwest Laurentian connection). A common premise for all these reconstructions is that the SW margin of

111°W

119°W

'North America" block (1.71 -1.65 Ga) MSM Mojave-Sonora megashear

Caborca block (1.78-1.60 Ga)

Yavapai province (1.84-1.66 Ga)

Outcrops of Proterozoic rocks

MGSZ ' Moore Gulch Shear Zone

Mazatzal province (1.75-1.62 Ga)

Nd province boundary El]

Nd provinces

CUKJ Mojave-Yavapai geochemical boundary

Figure 1. Distribution of major Paleoproterozoic crustal provinces (Mojave, Yavapai, and Mazatzal) and basement blocks (Caborca and "North America" in northern Sonora, Mexico) of SW North America, as defined by U-Pb zircon geochronology, Nd and Pb isotopes, major and trace element geochemistry, and petrologic and structural analysis of Proterozoic basement rocks. The magmatic age range listed for the different Paleoproterozoic provinces and blocks reflects U-Pb zircon geochronology shown in Figure 2. The fields of basement blocks in Sonora, and the trace of the Mojave-Sonora megashear (MSM) are from T.H. Anderson and L.T. Silver in the early 80's. Abbreviations for geographic features: HER, Hermosillo; LA, Los Angeles; LV, Las Vegas; PHX, Phoenix; SD, San Diego; TUC, Tucson.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Caborca block rocks exhibit geochemical and isotopic affinities to rocks from either the Yavapai province or the Mojave-Yavapai Transition zone, whereas rocks of the "North America" block have signatures similar to those of the Mazatzal province of southern Arizona.

Detailed isotopic, geochemical, and temporal characterization of representative Proterozoic granitoids in the Quitovac region of NW Sonora, Mexico (Table 1), has identified two distinct Paleoproterozoic basement blocks that coincide spatially with the previously proposed, age-defmed, Caborca and "North America" blocks (Fig. 1).

TABLE 1. GENERAL CHARACTERISTICS OF PROTEROZOIC GRANITOIDS FROM THE CABORCA AND "NORTH AMERICA" BLOCKS IN THE QUITOVAC REGION, NW SONORA, MEXICO Feature Age of maqmatism (U-Pb zircon crystallization ages) • Paleoproterozoic • Mesoproterozoic 1.4 Ga granite

"North America" Block

Caborca Block

Caborca "Grenvillian"

1714-1657 Ma Away from Quitovac

1777-1693 Ma Away from Quitovac

1126-1112 Ma N.A.*

Plutonic rock type

Calcalkaline granite and guartzmonzonite

Calcalkaline granite and two-mica granite

Alkaline granite

Nd radiogenic isotopes • cNd at T = crystallization age • Nd model age (TDM) • Nd whole-rock isochron • Nd province • Major element Geochemical series K2O (%) CaO (%) Na20 (%) A/CNK Fe/Mg • Trace element and REE Nb-Y discrimination diagram for granites Y (ppm) Rb (ppm) Ba (ppm) Th (ppm)

Y.REE (ppm) (LaA^b)N

+3.4 to +3.9

+0.6 to +2.6

-4.0 to-1.4

1800-1740 Ma 1663 ± 3 6 Ma (sNd ~ + 2 . 5 ) Nd province 2 & 3

2070-1880 Ma 1715 ± 9 4 Ma (eNd ~ + 1 . 2 ) Nd province 2

1770-1720 Ma N.D.f

Calcalkaline 0.6-2.1 2.53-4.17 4.1-4.5 Metaluminous to peraluminous 0.72-0.82

High-K calcalkaline 4.1-5.9 0.99-1.53 3.1-3.8 Peraluminous

High-K calcalkaline 4.7 2.01-2.18 3.4-3.6 Metaluminous to peraluminous 0.83-0.88

"Volcanic arc" granites

35 137 1042 12 122-188 4.2-12.0

0.85-0.89

74 590 4 23-97 10.7-49.1

N.A.*

"Within plate" and "volcanic arc" granites 54

222

1279 23 200-262 (Aibo = 423) 6.4-15.3

* N.A. = not applicable. t N.D. = no data.

overlaps in magmatic ages among Paleoproterozoic igneous rocks (Fig. 2). We conclude that it is erroneous to try to differentiate these provinces or blocks in SW North America based solely on U-Pb crystallization ages. In addition to a continuum of Paleoproterozoic magmatism among these provinces from approximately 1.8 to 1.6 Ga, we also observe a ubiquitous ~1.4-Ga magmatic pulse followed by a poorly understood Grenvillian age (-1.1 Ga) magmatic event in all the provinces and blocks. This indicates that these crustal provinces may have shared a common magmatic evolutionary history during the Proterozoic. In addition, we propose that the concept of age-defmed blocks in Sonora cannot be justified because of the large magmatic age overlap between them. Additional geochronologic studies in Sonora should be complemented with multidisciplinary studies, as mentioned above, to characterize the basement and determine any possible correlations with the existing, well defined,

The Paleoproterozoic Mojave, Yavapai, and Mazatzal crustal provinces of SW North America are reasonably well defined based on multidisciplinary studies that combine U-Pb zircon geochronology, Pb and Nd whole rock radioisotopes, major and trace element geochemistry, and petrologic and structural analysis. In contrast, the origin of the Paleoproterozoic basement that extends from these provinces into Sonora, Mexico, has often been interpreted solely on U-Pb zircon geochronology of igneous rocks that propose the existence of two blocks with significant age differences; the older Caborca basement block (1.81.7 Ga) and the younger "North America" block (1.7-1.6 Ga). We have compiled a database of U-Pb zircon ages, including 17 new ages from the Sonoran blocks and 161 previously published ages from the Paleoproterozoic crustal provinces and blocks of SW North America, which indicates that there are large

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Yavapai crust; and (3) The NW part of the Caborca block is composed only of Yavapai crust and the remainder of the Caborca block is in fact Mazatzal crust. The resulting margin in this last case would contain parts of the Mojave, Transition Zone, Yavapai, and Mazatzal provinces. These competing scenarios can be tested by undertaking structural, geochemical, and isotopic studies in appropriate parts of the Sonoran Desert that expose portions of Precambrian basement.

Paleoproterozoic cmstal provinces in SW North America. Proterozoic basement rocks of NW Sonora are a critical piece of this puzzle. We present several alternative hypotheses (Fig. 3) for the distribution of Paleoproterozoic provinces in the SW margin of Laurentia: (1) Caborca basement in NW Sonora is an extension of Mojave-Yavapai Transition zone crust that is juxtaposed against the Mazatzal province. In this scenario the margin would be a triple-junction of Mojave, Transition Zone, and

1.71 1.65 t

E (0

n = 19

0) E 3

.iiii

''North America" block Mexico

iiilN

ill

n= 7

1.74

1.71 1.65

Caborca block Mexico

- M 1.74

1.66

granne

;

granite

^ ; n = 10

1.66

Mazatzal province USA

diabase

/i n = 31

Yavapai province USA

I '

undated diabase

n = 34

i . .

Transition zone Mojave-Yavapai USA

diabase n = 25

I, ^ » r I ^ M / ' diabase &

Mojave province USA

anorthosite n = 53

1.8

1.7

1.6

1.8

U-Pb Zircon Age (Ga)

1.7

1.6

1.5

1.4

1.3

1.2

1.1

1.0

U-Pb Zircon Age (Ga)

Figure 2. Time of Proterozoic magmatism in major cmstal provinces and blocks in SW North America. A: Histograms showing U-Pb zircon ages from Paleoproterozoic plutonic and volcanic rocks. The vertical scale, shown with tick marks, is the number of samples in each 10-m.y.-interval. B: Graph showing U-Pb zircon ages for Paleoproterozoic and Mesoproterozoic igneous rocks. Ages are offset vertically for clarity. Where available, error bars represent errors cited in references at the 95% confidence level. In both graphs, the vertical gray bands indicate the overlap in magmatic ages. The graphs for the Caborca and "North America" blocks in Sonora, Mexico, contain all existing U-Pb zircon data including our new ages from Quitovac.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Figure 3. Hypothetical distributions of Paleoproterozoic provinces in the SW margin of Laurentia after continental rifting in the Neoproterozoic-Early Paleozoic. These hypotheses are based on current geological knowledge for the margin, in addition to our new studies in the Quitovac region, and assume no significant strike-slip redistribution of the provinces after continental rifting.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

MAGNETIC ANOMALIES ALONG THE SOUTHERN MARGIN OF THE KALAHARI PLATE - GRENVILLEAN OR PAN-AFRICAN? J. Jacobs\ RJ. Thomas^ and A. Golynsky^ ' Universitat Bremen, FB Geowissenschaften, PF 330 440, 28334 Bremen, Germany e-mail: jojacobs@uni-bremen.de ^ Council for Geoscience, PO Box 572, Bellville 7535, South Africa ^ VNIIOkeangeologia, St. Petersburg, Russia

provided by K-Ar and Ar-Ar mica dating across the shear zone. --1.1 Ga dates are recorded to the west and -500 Ma dates to the east. The different aeromagnetic signatures on either side of the shear zone thus mirror the differences recorded in the agedata. We interpret the striped magnetic crust west of the shear zone to represent -1.1 Ga crust without a ductile Pan Afi-ican overprint, while to the east, the amorphous magnetic pattern is caused by -1.1 Ga crust with a Pan African overprint. Thus, the Heimefront Shear Zone probably represents the front of the major Pan-African orogen at mid-crustal levels in west East Antarctica. Major crustal reworking east of the Heimefront Shear Zone occurred between -580-510 Ma, synchronous with the collision between E- and W-Gondwana. In a wider sense, this region of westernmost Dronning Maud Land forms the western front of the southern extension of the East African Orogen, the high-grade core of which is exposed several hundred kilometres further to the east in Central Dronning Maud Land.

In western Dronning Maud Land, East Antarctica, the Grenville-age crust exposed forms the detached southeastern segment of the wider Namaqua-Natal-Maud-S. Mozambique metamorphic belt. The originally great curved swathe of the Namaqua-Natal-Maud-S. Mozambique belt probably formed during indentation of the Kaapvaal Craton into Laurentia at ^1.1 Ga, resulting in the greater Grenville orogen (i.e. Grenville + Namaqua-NatalMaud-S. Mozambique Belt). In southern and eastern South Africa the aeromagnetic signature of the -1.1 Ga crust is typified by a highly characteristic "striped" pattern of elongate, ESE-trending aeromagnetic anomalies, of which the Beattie Anomaly is the most prominent and well-known. This pattern continues through the Falkland Plateau into westernmost Dronning Maud Land, where it terminates abruptly at the Heimefront Shear Zone. To the east of this structure a more amorphous pattern of small aeromagnetic anomalies is apparent across much of Dronning Maud Land. The Heimefront Shear Zone is a major Mesoproterozoic terrane boundary which was reactivated in Pan African times. Evidence for this is

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

HIGH JO2 METASOMATISM DURING WHITESCHIST METAMORPHISM IN THE ZAMBEZI BELT, NORTHERN ZIMBABWE: AMALGAMATION OF THE AFRICAN CRATONS DURING THE FORMATION OF GONDWANALAND AND NOT RODINIA S.P. Johnson Tectonics Special Research Centre, Department of Geology and Geophysics, The University of Western Australia, 35 Stirling Highway, Crawley 6009, Western Australia.

RATIONALE

- 650°C under high j02 and pRiO conditions. Syntectonic mineral phases such as kyanite porphyroblasts have preferentially grown within the foliation plane and define a dominant lineation trend. These structural fabrics are identical to those within the underlying Chewore Ophiolite and surrounding Chewore Inliers, where they have been dated at 524 ± 16 Ma (Goscombe et al., 1998). SHRIMP dating of zoned metamorphic zircons from the whiteschists reveal a more complex history with concordant ages at 1100 - 1050 Ma, 890 - 880 Ma and discordant ages of less than 590 Ma (one near concordant analyses is 590 ± 20 Ma). The significance of these age clusters is unclear; however, it does indicate that the whiteschist protolith is older than 1100 Ma and possibly (due to the geochemical similarity) the same as the Mesoproterozoic OIB's of the Chewore Ophiolite.

The age of amalgamation of the various southern African Cratons, namely the Congo and Kalahari Cratons, into the present-day configuration has long been an issue of debate. Some authors infer collision during the amalgamation of Rodinia along the 1.0 - 1.3 Ga orogenic belts (Kibaran and Irumide Belts) and that the younger transcontinental orogenic belt, the Pan African Belt, formed as an intracratonic orogen or mobile belt during Africa's collision into East Gondwanaland at c. 550 Ma. However, a 150 km long, 40 km wide, arcuate zone of very high pressure eclogite and whiteschist fragments within the Zambezi Belt of Zambia and northern Zimbabwe, indicate that this belt could not have formed as an intracratonic orogen. More likely this belt represents the mid-crustal levels of a Himalayan-style orogenic belt with the Lufilian Arc representing the foreland basin to this orogenic system. This abstract presents metamorphic and geochronological data for the high-pressure Kadunguri Whiteschists of the Chewore Inliers, northern Zimbabwe in an attempt to constrain the timing of this high pressure event and subsequent final amalgamation of Africa.

REGIONAL SYNTHESIS

The Kadunguri Whiteschists record a very-high pressure tectonothermal event under extreme metasomatic conditions at or near the base of a thickened continental crust (i.e. 50km) with the peak of metamorphism requiring a subduction zone / suture zone geothermal gradient of 10°C k m \ The nature and origin of the extreme metasomatic fluid is enigmatic and unlikely to be related to subduction zone fluids. One possibility is that such a fluid is produced by the dehydration of subducted upper continental lithosphere. The similarity in style and orientation of structural fabrics of the whiteschists and those known to be Pan African, the growth of peak metamorphic porphyroblasts that define this fabric and the presence of <590 Ma metamorphic rims on zircons, indicates that this tectonothermal event was related to continental collision during the Pan African event. This suggests that the Zambezi Belt represents the final result of a Neoproterozoic Wilson cycle involving the closure of a significant ocean basin and the possible subduction of continental crust in a Himalayan-style orogen. The transcontinental Pan African Belt therefore

THE KADUNGURI WHITESCHISTS

The Kadunguri Whiteschists crop out on the southern margin of the Mesoproterozoic Chewore Ophiolite as a semi-continuous block of whiteschist some 5 X 1.5 km. Lithologies range from typical whiteschists (talc + kyanite) to orthoamphibolebearing whiteschists (gedrite + kyanite). Whole rock major element analysis indicate that all lithologies can be described in the simple MFASH system with all Fe occurring as Fe^^, indicating the highly oxidised nature of the lithologies. Geochemical variation diagrams and isocon plots indicate that these lithologies developed by extreme metasomatism of alkalic ocean island meta-basalts from the neighbouring Chewore Ophiolite. Mineral parageneses indicate peak PT conditions of between 1 3 - 2 1 kbar (more probably 1 3 - 1 5 kbar) and 550

Geological Society of Australia Abstracts 65

Rodinia Symposium, Perth October 2001

represents a significant suture in West Gondwanaland and indicates that the present-day configuration of Africa was not bom until the Neoproterozoic c. 550Ma. Considering the rarity of accordion-style ocean opening and closure it is unlikely that the Kalahari and Congo Cratons were together prior to the amalgamation of Gondwanaland, therefore in future reconstructions

of Rodinia these two cratons must be treated as separate, disparate cratons. References Goscombe, B., Armstrong, R., & Barton, J. M. 1998.

^^ PP

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

EAST AFRICA AND MADAGASCAR WERE NOT PART OF RODINIA, AND EAST GONDWANA WAS NOT A COHERENT BLOCK A. Kroner Institut fur Geowissenschaften, Universitat Mainz, 55099 Mainz, Germany.

The EAO is characterized by extensive structural reworking and metamorphic overprinting of Archaean rocks, particularly in Tanzania and Madagascar, and these rocks either constitute marginal parts of cratonic domains or represent crustal blocks (terranes or microcontinents?) of unknown derivation. Furthermore, there is evidence for extensive granitoid magmatism in the time period -840 to <600 Ma whose predominant calcalkaline chemistry suggests subduction-related active margin processes. If these were related to the breakup and dispersal of Rodinia, this must have begun prior to 840 Ma. Peak metamorphic events during the Pan-African accretion and amalgamation process leading to formation of the supercontinent Gondwana are diachronous across the EAO and occurred between -640 and -550 Ma ago (Fig. 1). It is not clear whether there were two distinct pulses, one at 640620 Ma and one at 570-550 Ma, or whether one major granulite-facies event was diachronous. PT trajectories show a counter-clockwise evolution in Tanzania and Malawi, suggesting that processes other than, or in addition to, continental collision, such as magmatic underplating, were responsible for the high-grade event(s). This is supported by Pb and Nd isotopic data, indicating involvement of juvenile material in the generation of the Malawi and Tanzania granitoids (see summary in Maboko and Nakamura, 2001). The variation in age of peak metamorphism across the EAO from Ethiopia to Antarctica suggests that this is not a uniform belt but was assembled from a number of terranes that accreted at different times over a period of more than 100 m.y. (Kroner et al., 2001). Simple accretion, from west to east, does not seem to hold, since high-grade metamorphism in the Malawi terrane is young, whereas it is significantly older farther east, in Tanzania and Mozambique, and then young again in Madagascar, Sri Lanka, southern India and Antarctica (Fig. 1). The available data, therefore, suggest that East Gondwana was not a coherent block colliding with West Gondwana (Africa) some 550-530 Ma ago but consisted of individual terranes whose dimensions and origins remain to be defined.

Rocks associated with the amalgamation of the supercontinent Rodinia should either belong to an active continental margin setting and be generated between -1000 and -1300 Ma ago, or should bear the imprint of continental collision, i.e. be tectonized and form part of linear mobile belts during the same time range. This is well documented by the Grenville belt of North America and the Eastern Ghats belt of India, but is a matter of controversy when it comes to the various crustal components of eastern and central Africa. The -1300—950 Ma Kibaran and Irumide belts, although linear in nature and probably related to collisional processes, are little understood geodynamically, and their relation to Rodinia amalgamation is not established. The terrain farther east in Africa, now generally referred to as the East African Orogen (EAO), experienced its main structural evolution during the Neoproterozoic and as part of the amalgamation of Gondwana. Contrary to previous assertions, there is no evidence of a major structural or metamorphic event at -1000 Ma in this region, but there are increasing signs for localized Kibaran-age granitoid intrusions in parts of the EAO, in particular in southern Tanzania, southern Malawi, northern Zimbabwe and parts of northern Mozambique (see summary in Kroner, in press). The relation of these localized occurrences to the later, more voluminous, Neoproterozoic magmatic rocks is difficult to establish because of pervasive Pan-African structural overprinting. Madagascar shows similar features in that there is no linear 1000 Ma belt, there are no major occurrences of Kibaran-age rocks, and there is no evidence of deformation and metamorphism at this time. In fact similarities in rock assemblages and ages, particularly in central Madagascar, suggest a close genetic link with East Africa (Kroner et al., 2000; Muhongo et al., 2001). The mainly Archaean easternmost part of Madagascar is separated from the remaining part by the Betsimisaraka belt, interpreted by Collins & Windley (2001) as a suture zone where part of the Mozambique ocean vanished during amalgamation of Gondwana. Most of Madagascar is therefore likely to have been part of East Africa in Neoproterozoic times and was therefore also part of West Gondwana.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Figure 1. Early Mesozoic Gondwana reconstruction (mostly after Lawver et al., 1998) showing main geologic units in East and Southeast Afi-ica, Madagascar, southern India, Sri Lanka and East Antarctica (after Shackleton, 1996, with modifications of Madagascar fi-om Collins et al., 2000; Collins and Windley, 2001 and Krsner et al., 2000). LHB = Lutzow-Holm Bay, NNB = Namaqua-Natal belt; NC = Napier Complex; RC = Rayner Complex. Normal numbers are ages for magmatic events, numbers in italics are ages for metamorphic events (from Krsner, in press).

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Rodinia and Gondwana supercontinent formation and dispersal. S. Afr. J. Geol., Kroner, A., Hegner, E., Collins, A.S., Windley, B.F., Brewer, T.S., Razakamanana, T. and Pidgeon, R.T., 2000. Age and magmatic history of the Antananarivo Block, central Madagascar, as derived from zircon geochronology and Nd isotopic systematics. Am. J. Sci., 300, 251-288. Kroner, A., Willner, A.P., Hegner, E., Jaeckel, P. and Nemchin, A., 2001. Single zircon ages, PT evolution and Nd isotopic systematics of high-grade gneisses in southern Malawi and their bearing on the extent of the Mozambique belt into southern Africa. Precambrian Res., 109, 257-291. Lawver, L.A., Gahagan, L.M. and Dalziel, I.W.D., 1998. A tight fit-early Mesozoic Gond-wana, a plate reconstruction perspective. In: Motoyoshi, Y. and Shiraishi, K. (Eds.) Origin and evolution of continents. Mem. National Inst. Polar Res., Tokyo, Japan, Special Issue, 53, 214-229. Maboko, M.A.H. and Nakamura, E., 2001. Isotopic dating of juvenile Neoproterozoic crust in the Usambara Mountains of northeastern Tanzania: A link between the Mozambique belt and the Arabian-Nubian shield? Precambrian Res., in press Muhongo, S., Kroner, A. and Nemchin, A.A., 2001. Zircon ages from granulite facies rocks in the Mozambique belt of Tanzania and implications for Gondwana assembly. J. Geol., 109, 171-189. Shackleton, R.M., 1996. The final collision zone between East and West Gondwana: where is it? J. afr. Earth Sci., 23, 271287.

The available data suggest that East and central Africa were not involved in the formation of Rodinia, or at least do not favour a Rodinia connection. The world some 1000 Ma ago may not have consisted of a single supercontinent and a vast ocean, and other parts of West Gondwana, such as Brazil, may also have questionable links to Rodinia. Maybe, after all, Rodinia was a large continent but not a supercontinent. References Collins, A.S., Windley, B.F. and Razakamanana, T., 2000a. Neoproterozoic crustal-scale extensional detachment in central Madagascar: implications for extensional collapse of the East African Orogen. Geol. Mag., 137, 39-51. Collins, A.S. and Windley, B.F., 2001. The tectonic evolution of central and north Madagascar and its place in the East African orogen. J. Geol., in press. Jacobs, J., Fanning, C.M., Henjes-Kunst, F., Oelsch, M. and Paech, H.-J., 1998. Continuation of the Mozambique belt into East Antarctica: Grenville-age metamorphism and polyphase Pan-African high-grade events in central Dronning Maud Land: J. Geol., 106, 385-406. Kroner, A., in press. The Mozambique belt of East Africa and Madagascar: significance of zircon and Nd model ages for

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

DOES IT TAKE A SUPERPLUME TO BREAKUP A SUPERCONTINENT? — A CASE FOR RODINIA Z . X . L \ \ X . H . l P , P.D. K i n n y ^ a n d H. Zhou"^ ^ Tectonic Special Research Centre, Department of Geology and Geophysics, The University of Western Australia, Crawley, WA 6009, Australia ^ Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, P.O. Box 1131, Guangzhou 510640, China ^ Faculty of Earth Sciences, China University of Geosciences, Wuhan 430074, China "^Tectonics Special Research Centre, Department of Applied Geology, Curtin University of Technology, WA 6845, Australia

Australia (Crawford and Hilyard, 1990; Zhao et al., 1994; Wingate et al., 1998; Barovich and Foden, 2000). Our systematic geochronological and geochemical analyses of granitoids, gabbros, dyke swarms and rift volcanics in South China since then, in combination with existing results, have revealed that the Neoproterozoic bimodal magmatic activity there occurred in two major episodes (Fig. lb): one between 840 Ma and 800 Ma (with a peak at ca. 825 Ma), the other between 780 Ma and 740 Ma (with two peaks). There was also a minor episode at around 860 Ma. Whereas many pre-820 Ma intrusions are unconformably overlain by Neoproterozoic rift successions (e.g., Li et al., 1999), younger intrusions are often intruding the rift succession. In places ca. 780-750 Ma granites are unconformably overlain by Upper Sinian successions (< 750 Ma) which are interpreted by Wang and Li (2001) to form a rift cover. This demonstrates that crustal unroofing may have continued from prior to 850 Ma (Li et al., 1999) until after 780 Ma. Furthermore, some 760-750 Ma mafic dykes that are clearly co-magmatic with the granodiorites show REE and trace element patterns similar to those of mantle plume origin (e.g., Zhao et al, 1994). The ca. 780-740 Ma magmatic rocks in South China can thus also be interpreted as of mantle plume origin. This raises the question of whether we are seeing the record of two successive mantle plumes spread over 100 million years, or the results of a mantle superplume (Li et al., 2000, 2001). Similar observations are found in other parts of Rodinia.

INTRODUCTION Anderson (1982) first recognised the connections between the concentrations of hotspots, the AtlanticAfrican geoid high, and the former position of the supercontinent Pangaea. He proposed that the thermal isolation of the supercontinent led to the formation of a large mantle upwelling (or superswell), which eventually broke up the supercontinent. However, Anderson and others (e.g. Condie, 2000) do not believe that the mantle superswell developed beneath a supercontinent could be a mantle superplume arising from the coremantle boundary. Advances in recent years in global seismic tomography (e.g., van der Hilst et al., 1997) demonstrated the whole-mantle nature of mantle convection, thus reviving the idea of mantle plumes being the driving force for plate motions (Morgan, 1971) and supercontinent breakup. Could the breakup of Rodinia be the result of a mantle superplume, as argued by some for the breakup of Pangaea (e.g. Doblas et al, 1998; also see reviews by Story, 1995; Courtillot et al., 1999)? In this paper we examine the question using both new and existing data from South China, and published results from other continents, and speculate on the idea that mantle plumes (superplumes in particular) are the driving force of plate movement and supercontinent cycles.

NEOPROTEROZOIC BIMODAL MAGMATISM, RAPID UNROOFING, AND CONTINENTAL RIFTING IN SOUTH CHINA Li et al. (1999) interpreted 830-820 Ma bimodal igneous intrusions in South China as possible results of a mantle plume, based on the evidence of (1) their anorogenic origin, (2) a continental scale (-1000 km) rapid unroofing accompanying the magmatic intrusions, (3) broadly synchronous continental rifting, and (4) their identical age to the plumerelated mafic dyke swarms and basaltic eruptions in

Laurentia Both the ca. 780 Ma and ca. 720 Ma igneous activities in Laurentia have previously been interpreted as of mantle plume origin (Park et al., 1995; Heaman et al., 1992), but 850-800 Ma magmatic activities are rare. The 780-720 Ma bimodal igneous intrusions and extrusions are

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

± 7 Ma rift volcanics in the Adelaide Rift Complex (Preiss, 2000), 777 ± 7 Ma and 760 ± 12 Ma granites in westem Tasmania and King Island which were previously regarded as of orogenic origin with little supporting evidence (Turner et al., 1998), and the 755 ± 3 Ma Mundine Well dyke swarm (Wingate and Giddings, 2000). The 827 ± 6 Ma Gairdner Dyke Swarm is found to be erosionally truncated by the Sturtian glacial deposits (760-700 Ma), indicating uplifting after the dyke intrusion. However, no direct contact between the dykes and the basal Adelaidean units has yet been found. There is nonetheless a continent-wide hiatus for the interval 1000-840 Ma, which is consistent with the observation in South China.

abundant (Fig. la) along rift margins in both eastern and western Laurentia. Intriguingly, alkaline intrusions at the older end of the age spectrum are often found unconformably overlain by rift successions of shghtly younger age (e.g., Fetter and Goldberg, 1995), although little attention has so far been given to such intrusions along western Laurentia. We interpret these intrusions as the resuhs of anatectic melting above a mantle plume, and their erosional contact with the rift succession as the results of lithospheric doming and unroofmg above the plume head.

' (a)

^^^ J La iffi.flffl 720

720

740

760

740

760

780

800

820

3 Laurentia N

India, Seychelles and Madagascar Neoproterozoic bimodal, anorogenic magmatism in India spans from ca. 860 Ma to 730 Ma, with one broad peak at ca. 840-790 Ma, another broad peak at ca. 780-730 Ma (with a small sub-peak at ca. 780 Ma), and a possible minor peak at ca. 860 Ma (Fig. Id). The broad pattern of age distribution is very similar to that of South China and Australia. The rock types in India include alkaline granitoids (syenites), gabbros, mafic dykes, and bimodal volcanics (such as in the Malani igneous suite). As on the other continents, some intrusive bodies are seen to be unconformably overlain by Neoproterozoic rift successions (Das Gupta, 2000), indicating doming and unroofmg during and/or after the intrusions. Similar aged biomodal magmatic activities are also well developed in the Seychelles (Tucker et a l , 2001; Torsvik et al., 2001) and in Madagascar (Handke et al., 1999; Kroner et al., 1999; 2000; Tucker et al., 1999) (Fig. le). Handke et al. (1999) and Tucker et al. (2001) interpreted these igneous rocks, together with the Malani igneous suite in northwestern India, as of island arc origin because of their possible continental margin location in Rodinia. Others have argued that they were developed in an extensional (mantle plume?) environment (e.g.. Kroner et al., 2000). Similar rocks exist in southern and eastern India as well, and their petrological and geochemical characters, and the lack of coeval island arc volcaniclastic rocks in those regions, make a mantle plume origin a more plausible interpretation.

840

(b) South China

780

BOO

620

640

860

880

(C)

Australia

720

740

760

780

800

820

840

860

880

I" AdiJiM-'(e) Seychelles and Madagascar

720

740

780

840

860

880

Age (Ma)

Figure 1. Histograms of Neoproterozoic bimodal, anorogenic magmatic activities in central and westem Rodinia. Distribution curves were generated using ISOPLOT of Ludwig (2001).

Australia

Southern Africa

Neoproterozoic bimodal magmatic activities spread from ca. 850 Ma to ca. 750 Ma in Australia (Fig. Ic). Apart from the ca. 820 Ma rocks previously recognised as of mantle plume origin, there are also ca. 850 Ma lamprophyric dykes in the southern Yilgam (Robey et al., 1989), 810-800 Ma kimberlite pipes and lamproites in northern Kimberley (Pidgeon et al., 1989), 802 ± 10 and 777

Igneous rocks in South Afi-ica (e.g., Hanson et al., 1988; Frimmel at al., 2001) again share a similar age distribution (Fig. If). These rocks in places are clearly related to continental rifting, and some have been interpreted as of mantle plume origin (e.g., Frimmel et al., 2001). In southwestem Africa, ca.770 Ma syenites are unconformably overlain by rift successions no younger than 746 ± 2 Ma, again

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

(Zhao et al., 1994) or beneath South China (Li et al., 1999), and the ca. 780 Ma plume head west of Laurentia (Park et al., 1995). This superplume caused broad doming over Rodinia, and its heat led to widespread anatectic melting. Continental rifting over the superplume eventually led to the breakup of Rodinia by ca. 750 Ma.

indicating rapid unroofing at the time of the magmatism. Neoproterozoic magmatism of similar ages are also reported in the Congo and the Sao Francisco cratons.

DID A SUPERPLUME BREAK UP RODINIA? Both a ca. 825 Ma mantle plume (e.g. Li et al., 1999) and a ca. 780 Ma mantle plume (e.g., Park et al., 1995) have previously been proposed for the breakup of Rodinia, although the age difference between the plume-related dyke swarms in Laurentia and Australia (e.g., Wingate et al, 1998) has been used as evidence against the role of a mantle plume in the breakup of Rodinia. However, the following observations suggest that the breakup of Rodinia was probably caused by a mantle superplume which underplated much of Rodinia: • Geochemical and/or radiating patterns of mafic dyke swarms call for mantle plumes at ca. 825 Ma and ca. 780 Ma, coinciding with the timing of rifting and breakup of Rodinia; • A common early Neoproterozoic hiatus in South China, Australia, East Antarctica, India, Central and Southem Africa, and possibly Laurentia, plus evidence of syn-magmatic rapid unroofing in many of the continents, suggest trans-Rodinia lithospheric doming and unroofing from ca. 860 Ma to ca. 750 Ma or younger; • Widespread Neoproterozoic anorogenic, commonly biomodal, magmatism on most of these continents demands a prolonged (from ca. 860 Ma until 740-720 Ma), trans-Rodinia mantle heat source, which can be readily provided by a superplume; • The common occurrence of alkali or peralkali granitoids is consistent with them being of interplate, possibly plume-related origin; • The occurrence of kimberlite pipes and lamprophyres in Western Australian cratons (Pidgeon et a l , 1989) indicate a hot mantle beneath the Australian craton; • The common occurrence of anorogenic granitoids just prior to the rifting event is inconsistent with them being the results of passive mantle upwelling beneath stretched and thinned continental lithosphere (e.g.. White and McKenzie, 1989). We envisage that, like the mantle superswell (superplume?) developed underneath Panagea (Anderson, 1982), the Rodinia superplume rose beneath the central and western parts of Rodinia, with a diameter of well over 6000 km (Fig. 2). There could have been a number of normal plumes developed within this superplume, like the ca. 825 Ma plume head beneath just off South Australia

Figure 2. Schematic diagram showing the proposed Neoproterozoic superplume beneath Rodinia.

MANTLE SUPERPLUMES AS A DOMINANT DRIVING FORCE FOR PLATE TECTONICS? Condie (1998; 2000) demonstrated that the Earth's history is dominated by episodes of continental growth and supercontinental events, including the Neoproterozoic supercontinent Rodinia, but argued that mantle plumes (or superplumes) are results of slab avalanches during the formation of supercontinents rather than being associated with their breakups. However, we argue from evidence presented in this paper that the Rodinia superplume, probably as a result of both the thermal insulation of the supercontinent (e.g., Anderson, 1982) and the double push-ups of subducted slabs (e.g., Kellogg et al., 1999) surrounding it, was developed during (and caused) Rodinia breakup rather than its assembly. This is in line with the interpretation that a late Palaeozoic to Mesozoic superplume caused the breakup of Pangaea (e.g., Doblas et a l , 1998), and led to a pulse in oceanic crust production during the midCretaceous (e.g., Larson, 1991). We therefore suggest that superplumes are the main driving force for plate tectonics, which is characterised by episodic assembly and dispersion of supercontinents.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

References

tasmania: clues from south china. Geol. Soc. Aust. Abstracts, V. 64, p. 110-111. Li, Z.X., Li, X.H., Kinny, P.D., and Wang, J., 1999, The breakup of Rodinia: did it started with a mantle plume beneath South China? Earth Planet. Sci. Lett., v. 173, p. 171-181. Ludwig, K.R., 2001, Users manual for Isoplot/Ex rev. 2.49. Berkeley Geochronology Centre Special Publication, No. la, p. 56 pp. Morgan, 1971, Convection plumes in the lower Mantle. Nature, v. 230, p. 42-43. Park, J.K., Buchan, K.L., and Harlan, S.S., 1995, A proposed giant radiating dyke swarm fragmented by the separation of Laurentia and Australia based on paleomagnetism of ca.780 Ma mafic intrusions in western North America. Earth Planet. Sci. Lett., V. 132, p. 129-139. Pidgeon, R.T., Smith, C.B., and Fanning, C.M., 1989, Kimberlite and lamproite emplacement ages in Western Australia, in Ross, J., Jaques, A.L., Ferguson, J., Green, D.H., O'Reilly, S.Y., Danchin, R.V., and Janse, A.J.A., eds., Kimberlites and Related Rocks Volume 1: Their Composition, Occurrence, Origin and Emplacement, Geol. Soc. Aust. Special Publication No. 14, p. 382-391. Preiss, W.V., 2000, The Adelaide Geosyncline of South Australia and its significance in Neoproterozoic continental reconstruction. Precamb. Res., v. 100, p. 21-63. Robey, J.V.A., Bristow, J.W., Marx, M.R., Joyce, J., Danchin, R.V., and Amott, F., 1989, Alkaline ultrabasic dikes near Norseman, Western Australia, in Ross, J., Jaques, A.L., Ferguson, J., Green, D.H., O'Reilly, S.Y., Danchin, R.V., and Janse, A.J.A., eds., Kimberlites and Related Rocks Volume 1: Their Composition, Occurrence, Origin and Emplacement, Geol. Soc. Aust. Special Pub. No. 14, p. 383-391. Storey, B.C., 1995, The role of mantle plumes in continental breakup: Case histories from Gondwanaland. Nature, v. 377, p. 301-308. Torsvik, T.H., Carter, L.M., Ashwal, L.D., Bhushan, S.K., Pandit, M.K., and Jamtveit, B., 2001, Rodinia refined or obscured; palaeomagnetism of the Malani Igneous Suite (NW India). Precamb. Res., v. 108, p. 319-333. Tucker, R.D., Ashwal, L.D., and Torsvik, T.H., 2001, U-Pb geochronology of Seychelles granitoids: a Neoproterozoic continental arc fragment. Earth Planet. Sci. Lett., v. 187, p. 27-38. Turner, N.J., Black, L.P., and Kamperman, M., 1998, Dating of Neoproterozoic and Cambrian orogenies in Tasmania. Aust. J. Earth Sci., v. 45, p. 789-806. van der Hilst, R.D., Widiyantoro, S., and Engdahl, E.R., 1997, Evidence for deep mantle circulation from global tomography. Nature, v. 386, p. 578-584. Wang, J., and Li, Z.X., 2001, Sequence stratigraphy and evolution of the Neoproterozoic marginal basins along southeastern Yangtze Craton, South China. Gond. Res., v. 4, p. 17-26. White, R., and McKenzie, D., 1989, Magmatism at Rift Zones: The Generation of Volcanic Continental Margins and Flood Basalts. J. Geophys. Res., v. 94, p. 7685-7729. Wingate, M.T.D., Campbell, I.H., Compston, W., and Gibson, G.M., 1998, Ion microprobe U-Pb ages for Neoproterozoicbasaltic magmatism in south-central Australia and implications for the breakup of Rodinia. Precamb.Res., v. 87, p. 135-159. Wingate, M.T.D., and Giddings, J.W., 2000, Age and palaeomagnetism of the Mundine Well dyke swarm. Western Australia: implications for an Australia-Laurentia connection at 755 Ma. Precamb. Res., v. 100, p. 335-357. Zhao, J.X., Malcolm, M.T., and Korsch, R.J., 1994, Characterisation of a plume-related ~ 800 Ma magmatic event and its implications for basin formation in centralsouthern Australia. Earth Planet. Sci. Lett., v. 121, p. 349367.

Anderson, 1982, Hotspots, polar wander, Mesozoic convection and the geoid. Nature, v. 297, p. 391-393. Barovich, K.M., and Foden, J., 2000, A Neoproterozoic flood basalt province in southern-central Australia: geochemical and Nd isotope evidence from basin fill. Precamb. Res., v. 100, p. 213-234. Condie, K.C., 1998, Episodic continental growth and supercontinents: a mantle avalanche connection? Earth Planet. Sci. Lett., v. 163, p. 97-108. Condie, K.C., 2000, Episodic continental growth models: afterthoughts and extensions. Tectonophysics, v. 322, p. 153162.

Courtillot, v., Jaupart, C., Manighetti, I., Tapponnier, P., and Besse, J., 1999, On causal links between flood basalts and continental breakup. Earth Planet. Sci. Lett., v. 166, p. 177195. Crawford, A.F., and Hilyard, D., 1990, Geochemistry of Late Proterozoic tholeiitic flood basalts, Adelaide Geosyncline, South Australia, in Jago, J.B., and Moore, P.S., eds., The Evolution of a Late Proterozoic — Early Palaeozoic Rift Complex: The Adelaide Geosyncline. Geol. Soc. Aust. Spec. Pub. No. 16, p. 49-67. Das Gupta, P.K., 2000, The Neoproterozoic supercrustal Kumool Group, India: a case history of sediment-starved basin. Compact CD abstract volume, 31st International Geological Congress, Rio de Janeiro, Brazil, 6-17 August 2000. Doblas, M., Oyarzun, R., Lopez, R.J., Cebria, J.M., Youbi, N., Mahecha, V., Lago, M., Pocovi, A., and Cabanis, B., 1998, Permo-Carboniferous volcanism in Europe and Northwest Africa: a superplume exhaust valve in the centre of Pangaea? in Kinnaird, ed.. Aspects of tensional magmatism., J. Afr. Earth Sci., v. 26, p. 89-99. Fetter, A.H., and Goldberg, S.A., 1995, Age and geochemical characteristics of bimodal magmatism in the Neoproterozoic Grandfather Mountain rift basin. J. Geol., v. 103, p. 313-326. Frimmel, H.E., Zartman, R.E., and Spath, A., 2001, The Richtersveld Igneous Complex, South Africa: U-Pb zircon and geochemical evidence for the beginning of Neoproterozoic continental breakup. J. Geol., v. 109, p. 493508. Handke, M.J., Tucker, R.D., and Ashwal, L.D., 1999, Neoproterozoic continental arc magmatism in west-central Madagascar. Geology, v. 27, p. 351-354. Hanson, R.E., Wilson, T.J., and Wardlaw, M.S., 1988, Deformed batholiths in the Pan-African Zambezi belt, Zambia: Age and implications for regional Proterozoic tectonics. Geology, v. 16, p. 1134-1137. Heaman, L.M., LeCheminant, A.N., and Rainbird, R.H., 1992, Nature and timing of Franklin igneous events, Canada: implications for a late Proterozoic mantle plume and the break-up of Laurentia. Earth and Planet. Sci. Lett., v. 109, p. 117-131. Kellogg, L.H., Hager, B.H., and van, d.H.R.D., 1999, Compositional stratification in the deep mantle. Science, v. 283, p. 1881-1884. Kroener, A., Hegner, E., Collins, A.S., Windley, B.F., Brewer, T.S., Razakamanana, T., and Pidgeon, R.T., 2000, Age and magmatic history of the Antananarivo Block, central Madagascar, as derived from zircon geochronology and Nd isotopic systematics. Am. J. Sci., v. 300, p. 251-288. Kroener, A., Windley, B.F., Jaeckel, P., Brewer, T.S., and Razakamanana, T., 1999, New zircon ages and geological significance for the evolution of the Pan-African orogen in Madacascar. J. Geol. Soc., London, v. 156, p. 1125-1135. Larson, R.L., 1991, Latest pulse of Earth; evidence for a MidCretaceous superplume. Geology, v. 19, p. 547-550. Li, Z.X., 2000, Neoproterozoic hiatus in central Rodinia: legacy of a mantle plume that started the Rodinia breakup?.: Compact CD abstract volume, 31 st International Geological Congress, Rio de Janeiro, Brazil, 6-17 August 2000. Li, Z.X., 2001, Understanding the precambrian tectonic events in

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

GRENVILLE-AGE CONTINENTAL COLLISION IN SOUTH CHINA: NEW SHRIMP AGE CONSTRAINTS AND IMPLICATIONS TO RODINIA CONFIGURATION Z X

L i \ X . H . L i ^ H. Z h o u ^ a n d P . D . Kinny"^

^Tectonic Special Research Centre, Department of Geology and Geophysics, The University of Western Australia, Crawley, WA 6009, Australia ^Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, P.O. Box 1131, Guangzhou 510640, China ^Faculty of Earth Sciences, China University of Geosciences, Wuhan 430074, China "^Tectonics Special Research Centre, Department of Applied Geology, Curtin University of Technology, WA 6845, Australia.

In this study we determined SHRIMP U-Pb ages of zircon grains from (1) foliated granite and orthogneisses in the basement of the Hainan Island at the southwestern end of the Cathaysia Block, (2) Mesoproterozoic metasediments underlaying Neoproterozoic rift and platform successions of the Yangtze Block in southern Sichuan Province, (3) a granitic gneiss within the Mesoproterozoic paragneisses and metasediments, and (4) inherited zircons in a mafic dyke that intruded Neoproterozoic granitic rocks in central Sichuan Province (Fig. 1). Amongst the two samples from the basement of Cathaysia, one gives a concordant age of ca. 1430 Ma, interpreted as the age of its crystallisation. The other has an identical magmatic crystallisation age, and in addition, has metamorphic zircon overgrowths dated at ca. 1300 - 1000 Ma, and an inherited zircon core of ca. 1790 Ma.

ABSTRACT Estimates for the age of collision between the Cathaysia and Yangtze Blocks in South China (Fig. 1) are generally in the range of ca. 1000 Ma to 800 Ma, but there are models for a collision as young as in the Mesozoic. The older estimates are based on conventional ages of deformed volcanic rocks (formed in an arc environment?), while the younger ones are based on the ca. 970 Ma age of the ophiolite in northeastern Jiangxi and southern Anhui Provinces, and the ca. 820 Ma granites along the orogenic belt (e.g., X.H. Li, 1999). However, the ca. 820 Ma granites have recently been interpreted to be due to cmstal melting above a mantle plume in an extensional environment (Z.X. Li et al., 1999). No reliable metamorphic age related to the orogeny (e.g., the Sibao Orogeny; see discussions in Z.X. Li, 1998) has yet been reported.

Pre-Neoproterozoic Outcrops

* X.H. 111997 * * Qiuetal., 2000

Figure 1. Precambrian tectonic framework of South China and sampling localities (e.g. 99KD33).

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Furthermore, the ca. 1430 Ma granitic intrusions in Hainan Island are consistent with the model that the Cathaysia Block was connected to southwestern Laurentia during the Mesoproterozoic (Li et al., 1995), where there is a transcontinental ca. 1400 Ma plutonic province (e.g. Nyman et al, 1994). If South China was indeed between Australia and Laurentia in Rodinia, as suggested by Li et al. (1995; 1999) (Fig. 2), the presence of the Sibao suture within South China would suggest that Rodinia did not exist until ca. 1000 Ma. This is supported by recent palaeomagnetic results (Wingate et al., this volume).

The granitic gneiss within the Mesoproterozoic paragneisses and metasediments in southern Sichuan Province gives, for the first time in South China, a reliable metamorphic(?)-zircon crystallisation age of 1007 ± 14 Ma. Detrital zircons from the silicic metasediments, and inherited zircons found in the mafic dyke, have an age spectrum very similar to those from the Cathaysia samples, including the ca. 1300 - 1000 Ma ages. We interpret these new results as the first reliable evidence for Grenville-age metamorphism and continental collision in South China. We take the ca. 1300 - 1000 Ma zircon overgrowths in the basement of Cathaysia, as well as the ca. 1000 Ma granitic gneiss in southern Sichuan, as evidence for a protracted orogenic process which let to the collision between the Cathaysia and Yangtze Blocks by ca. 1000 Ma. The presence of a Cathaysia signature among detrital zircons at the southern margin of the Yangtze Block suggests that those sediments were likely formed in a foreland basin on the Yangtze side of the late Mesoproterozoic orogen.

References Li, X.H., 1997, Timing of the Cathaysia Block formation: constraints from SHRIMP U-Pb zircon geochronology: Episodes, v. 20, p. 188-192. Li, X.H., 1999, U-Pb zircon ages of granites from the southern margin of Yangtze Block and the timing of Neoproterozoic Jinning Orogeny in SE China: termination of Rodinia assembly?: Precamb. Res., v. 97, p. 43-57. Li, Z.X., 1998, Tectonic evolution of the major East Asian lithospheric blocks since mid-Proterozoic - a synthesis, in Martin, F.J., Chung, S.-L., Lo, C.-H., and Lee, T.-Y., eds.. Mantle Dynamics and Plate Interactions in East Asia, Volume 27: AGU Geodynamics Series: Washington, D.C., American Geophysical Union, p. 221-243. Li, Z.X., Li, X.H., Kinny, P.D., and Wang, J., 1999, The breakup of Rodinia: did it started with a mantle plume beneath South China?: Earth Planet. Sci. Lett., v. 173, p. 171-181. Li, Z.X., Zhang, L., and Powell, C.M., 1995, South China in Rodinia: part of the missing link between Australia-East Antarctica and Laurentia?: Geology, v. 23, p. 407-410. Nyman, M.W., Karlstrom, K.E., Kirby, E., and Graubard, C.M., 1994, Mesoproterozoic contractional orogeny in western North America: Evidence from ca. 1.4 Ga plutons: Geology, p. 901-904. Qiu, Y.M., Gao, S., McNaughton, N.J., Groves, I>.L, and Ling, W., 2000, First evidence of >3.2 Ga continental crust in the Yangtze craton of South China and its implications for Archean crustal evolution and Phanerozoic tectonics: Geology, v. 28, p. 11-14.

B = Belt Basin C = Cathaysia T = W. Tasmania Y = Yangtze

Sibao Orogen

+—+

Hainar Island Mojavia YavapaiMazatzal

Ca. 1.4 Ga magmatic provinces Grenville-aged metamorphism Rivers from the Sibao Orogen Interpreted foreland basins

•—

2000 km

Figure 2. A schematic diagram showing the Sibao Orogen as a possible suture in central Rodinia

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

NEW U-PB SHRIMP AGES FROM THE TUGELA TERRANE, NATAL BELT, SOUTH AFRICA: INSIGHTS INTO THE ASSEMBLY OF RODINIA S. McCourt\ R.A. Armstrong^ and S.T. Johnson^ ' Department of Geology University of Durban-Westville, Durban, South Africa. ^ Research School of Earth Sciences, Australian National University, Canberra, ACT 0200, Australia ^ School of Earth and Ocean Sciences, University of Victoria, Victoria, Canada

development of north-verging folds and thrust faults, tectonic interleaving of the terrane with ophiolite, and voluminous granitic magmatism (the Mtungweni granitoids). U-Pb SHRIMP analyses on zircons from units identified as important time markers, indicate that oceanic arc development was underway by 1209 ± 5 Ma, the age of the Kotongweni tonalite. The depositional age of the metasedimentary Dulumbe gneiss is constrained to being post-1175 ± 9 Ma, the age of the youngest detrital component in the sequence. The oldest detrital grain has an age of 1289 ±14Ma which is not only significantly older than anything else reported from the Natal belt but also indicates that the Tugela terrane did not receive detritus from the Archaean Kaapvaal Craton. Subsequent crustal thickening, tectonic burial of the terrane and peak metamorphism occurred at 1182 ± 19 Ma, the age of metamorphic zircon, and inherited zircons that suffered syn-metamorphic Pb-loss in the Dulumbe paragneiss. Subsequent uplift and northward emplacement of the Tugela terrane onto the Kaapvaal Craton (D3), is constrained by the crystallisation age of the late syn-kinematic Mkondene diorite (1161 ± 9 Ma) and Mtungweni granitoids (1155 ± 6 Ma). The massif-type Mambulu anorthosite was emplaced at 1145 ± 6 Ma.

ABSTRACT We report on the results of a geochronological study of the Tugela terrane, the northernmost tectonic element of the Mesoproterozoic Natal belt of South Africa. The Tugela terrane consists of a west-plunging structural stack, with the structurally highest, most allochthonous and, in terms of published models, most oceanic thrust sheet exposed in the west of the terrane. Our study focused on the three highest tectonostratigraphic packages in the thrust stack. These are, from west to east, the Tugela, Mandleni, and Madidima, respectively. The Tugela tectonostratigraphic package consists of an oceanic arc sequence, including arc tholeiites and a tonalite intrusion (the Kotongweni tonalite). The Mandleni tectonostratigraphic package is characterized by a geochemically enriched bimodal magmatic succession (the Dondwana tectonite unit). Metasedimentary rocks including the feldspathic Dulumbe gneiss dominate the Madidima tectonostratigraphic package. All the thrust sheets are characterized by a penetrative schistosity / gneissosity, and by synkinematic upper amphibolite to granulite grade metamorphism. Voluminous mafic to ultramafic magmatism, including the emplacement of the Tugela Rand layered complex, the Mkondene diorite and the Mambulu massif-type anorthosite, occurred late during this tectonic event. Subsequent exhumation and lower amphibolite grade metamorphism was coeval with the

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

TECTONIC EVOLUTION OF THE MESOPROTEROZOIC ALBANY-FRASER OROGEN AND NEOPROTEROZOIC LEEUWIN GNEISS COMPLEX, SOUTHWESTERN AUSTRALIA: AUSTRALIAN PORTIONS OF ANTARCTIC - AUSTRALIAN OROGENS J.S. Myers^ and D.R. Nelson^ ' Department of Earth Sciences, Memorial University of Newfoundland, St. John's, Newfoundland A1B 3X5, Canada ^ Geological Survey of Western Australia, Mineral House, 100 Plain Street, East Perth, WA 6004, Australia and Department of Applied Physics, Curtin University of Technology, GPO Box U1987, Perth, WA 6001, Australia

hundreds of kilometres long, at deep crustal levels. Locally derived granitic melts formed in low strain zones during the peak of metamorphism. Between c.1300 and 1280 Ma, the resulting thrust pile was transported northwards and elevated onto the edge of the West Australian continent. Associated tectonic fabrics formed at decreasing metamorphic grades from amphibolite to greenschist facies, and ductile structures were increasingly superceded by brittle structures. After a tectonically dormant period of 100 million years, there was renewed intrusion of large volumes of granite in the southern part of the Australian portion of the orogen, accompanied by regional dextral transpression between c. 1180 and 1130 Ma. Most of the Leeuwin Complex was derived from c. 780 and 695 Ma granites intruded into Mesoproterozoic anorthosite and c. 1090 Ma granite. The rocks were strongly deformed and converted to granulite facies gneisses at c. 615 Ma, an event that may reflect the amalgamation of an Australian Antarctic continent with India to form an eastern component of Gondwanaland. These gneisses were intruded by granite at c. 535 Ma, and again deformed and recrystallized in granulite facies during an episode of extension.

The Albany-Fraser Orogen is exposed for 800 km along the southern coast of Western Australia. The orogen resulted from the collision of a West Australian continent with a South Australian - East Antarctic continent at c.1300 Ma. The orogen was split longitudinally during the rifting of Australia from Antarctica at c. 100 Ma, and the southern part of the orogen is exposed along the adjacent coast of Antarctica in Wilkes Land and the Hunger Hills. The Albany-Fraser Orogen is truncated to the west by the Darling Fault and juxtaposed against the Neoproterozoic Leeuwin gneiss complex. This complex strikes north-south and is exposed for 100 km along the southwestern coast of Australia. The Leeuwin Complex was truncated to the south by the opening of the Southern Ocean, and the southern continuation of this gneiss complex is exposed in the Prydz Bay - Denman Glacier region of Antarctica. This presentation will provide an overview of the tectonic and magmatic evolution of both the AlbanyFraser Orogen and Leeuwin Complex, and draw comparisons with their Antarctic equivalents. The Albany-Fraser Orogen mainly consists of orthogneiss and granite. The orthogneiss was mostly derived from c. 2630 and 1700 - 1600 Ma granitic protoliths that were intensely deformed at c.1300 Ma. These rocks, together with c.1300 Ma gabbro and granite, were stacked in thrust and duplex structures a few kilometres thick and

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

RECONSTRUCTIONS OF CONTINENTS DURING THE PROTEROZOIC - A WAY TOWARDS RODINIA L J . Pesonen\ S. Mertanen^ and S.-A. Elming^ ^ Department of Geophysics, University of Helsinki, PO Box 64, Helsinki, Finland (lauri.pesonen@helsinki.fi) ^ Geological Survey of Finland, PO Box 96, Espoo, Finland ^ Department of Applied Geophysics, University of Lulea, Lulea, Sweden

together to represent Laurentia. The same approach holds for the other continents, too. However, this assumption does not necessarily hold for the pre-2.0 Ga old cratons and their reconstructions. The used poles and their ages are listed in Pesonen et al. (2001). Some 50% of the data were taken from the Global Palaeomagnetic Data Base (GPDB). The rest were compiled from literature and personal data bases by workshop attendants. Before accepting a pole into the analysis it had to pass a certain reliability criteria, where the poles are classified into four classes (A, B, C and D), where A is very reliable and D is unreliable (Pesonen et al., 1989). An analysis of the global data after the filtering process (accepted poles were mainly of class A-B and only occasionally of C) revealed that at specific times there were several poles available ft-om many continents so that the existence of supercontinents could be studied by paleomagnetism. The time intervals for which we have reliable poles from more than two, continents were 2.0 Ga, 1.83 Ga, 1.77 Ga, 1.65 Ga, 1.5 Ga, 1. 25 Ga, 1.10 Ga and 1.05 Ga (Pesonen et al., 2001). In making the continental reconstructions at various times we used the Euler-techniques and the GMAP-programme (Torsvik et al, 1990). This technique allows the continents to be placed in correct paleolatitudes and azimuthal orientations (with respect to present North) but they do not allow the paleolongitudes to be defined. The errors in paleolatitudes vary but are generally less than 10 degrees. The geological and tectonic features (e.g., major orogenic belts, huge fracture zones) of each continent were outlined onto the plots. The continents were then slided together by moving them along paleolatitude-lines (which is allowed by paleomagnetism since the paleolongitude is undeterminable). Occasionally, if a better (geologically or paleomagnetically) matching was required, we plotted a continent into its antipodal hemisphere (with inverted orientation). This is allowed since we do not know the true polarity of the Precambrian poles. By this way we ended to eight plots of possible continental assemblies during Proterozoic times. The results and data are given in Pesonen et al. (2001). Here we give three examples

INTRODUCTION The importance of supercontinents to the crustal and mantle processes on Earth has recently been emphasized in several studies (e.g., Pesonen et al., 2001 and references therein). These processes include mantle superplume events, slab avalanches to the mantle, global glaciations, carbon isotope excursions, truncations of major dyke swarms, peaked distribution of the ages of black shales and stromatolites, eustatic sea level changes, etc. At least three supercontinent assemblies have existed during or since the Neoproterozoic (Rodinia, ca. 1200-780 Myrs, Gondwana 550-400 Myrs and Pangaea 350-250 Myrs). The geological importance of supercontinents for the Earth's evolution has lead some people to seek whether global wide continental assemblies existed prior to Rodinia (e.g., assemblies like "Vaalbara", "Ur", "Atlantica", etc., see Pesonen et al. 2001). Although paleomagnetism is the only method which provides direct knowledge of the positions of continents during the past, the use of paleomagnetism in many of these previous reconstructions has been obscure. In this paper we use good quality paleomagnetic data as a prime source for making continental reconstructions during the Proterozoic. The paper is an output of the Fourth Scandinavian Paleomagnetic Data Base Workshop, held in 1999 in Arhus, Denmark (Abrahamsen et al., 2001).

USED TECHNIQUES AND SOURCES OF DATA Paleomagnetic data were obtained from fifteen continents or cratonic blocks (Laurentia (L), Baltica (B), Siberia (S), North China (NC), South China (SC), India (I), Amazonia (Am), Sao Fransisco (SF), West Africa (WA), Congo (C), Kalahari (K), Australia (A), East Antarctica (EA), Dronning Maud Land (DL) and Coats Land (CL)). Although the present continents have grown by accretion of different cratons we show the continents in some cases as a unity. For example, most of the Laurentian poles come from the Superior craton and only a few from the Wyoming craton. Since there is no strong evidence that these cratons have drifted separately during 2.0-1.05 Ga, we treat the poles

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

The survey shows that the continents, except Australia, have been generally at shallow to moderate latitudes during the Proterozoic times. Figure 1 shows the assembly of five continents at 1.83 Ga ago.

Laurentia and the Rio Negro Juruena belt in Amazonia. Our paleomagnetic data suggest that Siberia was in contact with Laurentia already at 1.5 Ga ago, much earlier than previously thought (750 Ma; Hoffman, 1991) and that the orientation of Siberia is opposite to that shown in many previous reconstructions. A novel scenario for the birth of Rodinia during 1.2-1.05 Ga is presented in Figures 2 and 3.

1830 Ma

1100 Ma

Figure 1. Global reconstruction at 1.83 Ga showing five continents (Laurentia, Baltica, Amazonia, India and South China) in a possible new Mesoproterozoic supercontinent (?) assembly.

Figure 2. Global reconstruction of the continents at 1.10 Ga showing nine continents in an assembly leading to Rodinia (see Figure 3).

of the paleomagnetically based reconstructions (at 1.83 Ga, 1.10 Ga and 1.05 Ga) and a summary of the results.

SUMMARY

In this model Siberia separated from Laurentia at about 1.1 Ga and amalgamated to Baltica during 1.1-1.05 Ga. Laurentia and Baltica were still united at 1.25 Ga ago when Congo/Sao Fransisco continent began to approach Baltica. At ca. 1.2 Ga ago, Congo Sea separated Baltica from Laurentia. Slightly later, at about 1.08 Ga ago, Congo/SaoFransisco collided with Baltica causing the Irumidian and Sveconorwegian orogenies in these continents. In our model, which for Amazonia is not based on paleomagnetic data, Amazonia collided with Laurentia also at ca. 1. 1 Ga ago to form the Sunsas/Araguipe and Grenvillian orogenic belts. This model is, however, not consistent with new palaeomagnetic data from Amazonia (Tohver et al., 2001), which places Amazonia frirther to south and in different orientation than in our model (Fig. 3).

The five continents in the 1.83 Ga assembly are mostly in the northern hemisphere and occupy moderate to shallow latitudes. The data are consistent with a supercontinent that could have existed at ca. 1.9-1. 8 Ga ago and could be related to the proposed global superplume event at this time (Condie, 1998). This assembly probably initiated already ca. 1.9 Ga when Laurentia collided with Baltica causing the Nagssugtoqidian orogeny in Laurentia and the Lapland-Kola orogeny in Baltica. Paleomagnetic data show that during the final docking to Baltica, Laurentia rotated ca. 70 degrees clockwise. From 1.83 Ga to ca. 1.25 Ga the Laurentia-Baltica landmass rotated and drifted to nearly equatorial latitudes (Buchan et al. 2001). Meanwhile Amazonia drifted towards Baltica and became docked with it in a long-lasting (1.88-1.45 Ga) island-arc accretion which culminated with the formation of the Svecofennian/Gothian belts in Baltica, the coeval Hudsonian/Labradorian belts in

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

1050 Ma

The models presented in this work are based on palaeomagnetic data. To test the models we urgently need new palaeomagnetic poles from rocks of the age of ca. 1.1-1.05 Ga from most continents, especially from the Amazonia and Congo/Sao Fransisco cratons. References

Abrahamsen, N., Pesonen, LJ., Van der Voo, R., 2001. The 4th Nordic Paleomagnetism Workshop and Symposium. (Submitted to DGU). Buchan, K., Ernst, R., Hamilton, M., Mertanen, S., Pesonen, LJ., Elming, S.-A, 2001. Rodinia: the evidence from integrated paleomagnetism and U-Pb geochronology. Prec. Research (in press). Condie, K.C., 1998. Episodic continental growth and supercontinents: a mantle avalanhce connection? Earth Plan. SciLett., 163,97-108. Hoffman, P. 1991. Did the breakout of Laurentia turn Gondwana inside out? Science 252, 1409-1412 Pesonen, L.J. Elming, S.-A., Pisarevsky, S., Mertanen, S., D'Agrella-Filho, M., Meert, J., Schmidt, P., Bylund, G 2001. Assemblies of Continents During the Proterozoic - A Paleomagnetic Survey (in prep). Tohver, E., Van der Pluim, B., Van der Voo, R., Scandolara, J., Rizzotto, G., 2001. Reconstructing Rodinia: the view from Amazonia. In: Eos, Transacts., AGU, 82, No. 20, May 15, S129. Torsvik, T.H , Smethurst, M. and Pesonen, L.J., 1990. GMAPGeographic mapping and palaeoreconstruction package. NGU rapport Nr. 90.019, 62 pp.

Figure 3. Global reconstruction at 1.05 Ga ago showing the palaeomagnetically valid Rodinia supercontinent. Note that Amazonia (with a ?-mark) is based on its Gondwana position due to lack of paleomagnetic data from this continent at 1.05 Ga. The second position of Amazonia with broken lines is based on new paleomagnetic data from Amazonia of slightly younger intrusive rocks (K-Ar age of ca. 1.04-0.98 Ga; Tohver et al., 2001) and may relate to events younger than 1.05 Ga. At about 1.05 Ga, these landmasses (CongoSaoFransisco/Baltica and Laurentia/Amazonia?) amalgamated with other cratons to form the supercontinent Rodinia. In the last stage of this aggregation, Baltica collided with the northeastern margin of Greenland causing late-Sveconorwegian overprints (ca. 1.0-0.85 Ga) in NE Greenland and in Baltica. It is possible that Kalahari collided at this time with Laurentia to cause similar lateGrenvillian/Namagua-Natal overprints in Laurentia and Kalahari.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

AN ANIMATED HISTORY OF RODINIA C.McA. Powell, S.A. Pisarevsky and M.T.D, WIngate Tectonics Special Research Centre, Department of Geology and Geophysics, The University of Western Australia, 35 Stirling Hwy, Crawley, WA 6009, Australia

between 680 and 600 Ma. Powell and Pisarevsky (2001) combined this information with new analysis of the palaeomagnetic data and proposed a new model of the Neoproterozoic tectonic history of the East Gondwanaland. According to this model, India together with the Rayner block of Antarctica was not a part of Rodinia, but recombined with the rest of Eastem Gondwanaland between 680 and 610 Ma. Another striking result was obtained by Wingate et al. (2001) from the 1070 Ma Bangemall Sills (Westem Australia). This new palaeomagnetic pole strongly contradicts such popular models as SWEAT (Moores, 1991) and AUSWUS (e.g. Karlstrom et al., 1999; Burrett and Berry, 2000). New AustraliaLaurentia fit AUSMEX was proposed, in which the Grenville Province of Laurentia continues into northeastem Australia to join the Cape Range Musgrave - Albany-Frazer orogenic system. A recent palaeomagnetic study of the deep drillhole Empress lA in the Officer Basin (Pisarevsky et al., 2001) also supports this fit. A comparison of late Mesoproterozoic palaeomagnetic poles from the Kalahari craton and the correlative Gmnehogna fragment in East Antarctica (Powell et al., 2001) shows that the Kalahari-Gmnehogna craton shows, that in Rodinia, the Kalahari craton could have lain to the southwest of Laurentia with the Namaqua-Natal orogenic belt facing outboard and away from the Laurentian craton. Kalahari could have lain off the westem margin of Australia until 800-750 Ma, when breakup associated with the 755 Ma Mundine Well mafic dyke swarm caused Kalahari to rotate anticlockwise away from the westem margin of Australia. Predictions from this model can be tested against the known geology of the Kalahari craton. The position of Baltica constrained by palaeomagnetic data (e.g. Hyodo and Dunlop, 1993; Pisarevsky and Bylund, 1998; Weil et al, 1998 and references therein). Baltica appears to have had a long-lived passive margin along the Urals from the late Mesoproterozoic to the Vendian, when it became a transpressive margin. Similarly, Siberia appears to have had a passive southem margin until ca. 800 Ma, when ophiolites were emplaced and subduction commenced beneath its southwestem edge. East Greenland north of ca. 73° N also appears to have had a passive margin from the late Mesoproterozoic through the Neoproterozoic. These features place the passive margin facing a late

There is general agreement that the Earth's continental crust could have been assembled in a supercontinent, Rodinia, in the late Mesoproterozoic and early Neoproterozoic. There are several versions of its composition and configuration (e.g. Hoffman, 1991; Dalziel, 1997; Weil et al., 1998). Laurentia is thought to lie at the core of Rodinia, because it is surrounded by late Neoproterozoic passive margins formed during the breakup of the supposed supercontinent. Australia, Antarctica and possibly South China are considered to have lain along Laurentia's westem margin, and Baltica, Amazonia and Rio de la Plata cratons along its eastem margin. The position of Siberia is disputed, but it is generally shown as lying along the northern margin of Laurentia. The position of the Congo and Kalahari cratons is uncertain, with at least four positions having been shown for Kalahari in the last few years. We have used palaeomagnetism to determine permissible fits for Rodinia, and geology to refine permissible fits into plausible reconstructions. Some of the geological constraints we have used are continuity of tectonic belts, and the presence of passive or active continental margins. There is also the global balance of rifted margins that needs to be accounted for in any acceptable reconstruction. The palaeomagnetic data provide quantitative constraints for the Precambrian palaeoreconstructions. Unfortunately, these data are distributed very non-uniformly in time and space. The majority of palaeomagnetic results for Rodinia times (roughly between 1000 and 750 Ma) came from Laurentia. McElhinny and McFadden (2000) did what seems to be the most reasonable analysis of these Laurentian data, and we have used their APWP as a framework for Rodinia reconstructions. Two recent discoveries forced us to review the shape and history of Rodinia. Fitzsimons (2000) demonstrated three separate "Grenvillian" provinces in East Antarctica with statistically different ages, which are separated by two younger belts of the Pan-African age. There is also evidence for the Neoproterozoic sinistral displacement along the Darling Mobile Belt (Pinjarra Orogen) on the present-day westem margin of Australia (Harris, 1994). Harris (1994) also suggested that the Darling Mobile Belt continued into Antarctica and found an equivalent structure in India, roughly estimating the time of the event

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Karlstrom, K.E., Harlan, S.S., Williams, M.L., McLelland, J., Geissman, J.W and Ahall, K.-L, 1999. Refining Rodinia: geologic evidence for the Australia-Western U.S. connection in the Proterozoic. GSA Today, 9, #10: 1-7. Li., Z.X., Li., X.H., Kinny, P.D. and Wang, J., 1999. The breakup of Rodinia: did it start with a mantle plume beneath South China? Earth Planet. Sci. Lett., 173: 171-181. McElhinny, M.W. and McFadden, P.L., 2000. Paleomagnetism: Continents and Oceans. Academic Press, San Diego: 386 pp. Moores, E.M., 1991. Southwest US - East Antarctic (SWEAT) connection: a hypothesis. Geology, 19: 425-428. Pisarevsky, S.A. and Bylund, G., 1998. Palaeomagnetism of a key section of the Protogine Zone, southern Sweden. Geophys.J.Int, 133: 185-200. Pisarevsky, S.A., Li, Z.X., Grey, K. and Stevens, M.K., 2001. A palaeomagnetic study of Empress 1 A, a stratigraphic drillhole in the Officer Basin: new evidence for the low-latitude position of Australia in the Neoproterozoic. Precambrian Res., 110: 93-108. Powell, C. McA. and Pisarevsky, S.A., 2001. Late Neoproterozoic assembly of East Gondwanaland. Geology (in press). Powell,C.McA, Jones,D.L, Pisarevsky, S.A. and Wingate, M.T.D., 2001. Paleomagnetic constraints on the position of the Kalahari craton in Rodinia. Precambrian Res., 110: 33-46. Trompette, R., 1997. Neoproterozoic (-600 Ma) aggregation of Western Gondwana: a tentative scenario. Precambrian Res., 82: 101-112. Weil, A.B., Van der Voo, R., Mac Niocaill, C and Meert, J.G., 1998. The Proterozoic supercontinent Rodinia: paleomagnetically derived reconstruction for 1100 to 800 Ma. Earth Planet. Sci. Lett., 154: 13-24. Wingate, M.T.D and Giddings, J.W., 2000. Age and palaeomagnetism of the Mundine Well dyke swarm. Western Australia: implications for an Australia-Laurentia connection at 755 Ma. Precambrian Res., 100: 335-357. Wingate, M.D.T., Pisarevsky, S.A. and Evans, D.A.D., 2001. Paleomagnetic constraints on Rodinia reconstructions at 1070 Ma: no SWEAT, no Auswus. Terra Nova (in press).

Mesoproterozoic-early Neoproterozoic ocean at the edge of any late Mesoproterozoic supercontinent. Proposed positions of Western Africa, Amazonia and Rio de La Plata cratons generally agree with the hypothesis of WA-A-RP mega-continent (e.g. Trompette, 1997). It is commonly assumed that the breakup of Rodinia started on the western (in present coordinates) boundary of Laurentia (e.g. Hoffman, 1991; Dalziel, 1997; Li et al., 1999). Palaeomagnetic data from Mundane Well dyke swarm (Wingate and Giddings, 2000) suggest that this breakup started before 755 Ma. We present an animated series of maps showing the permissible reconstruction of Rodinia at the end of the Mesoproterozoic, and its possible breakup after 810 Ma (Fig. 1). References Burrett, C and Berry, R., 2000. Proterozoic Australia-Western United States (AUSWUS) fit between Laurentia and Australia. Geology, 28: 103-106. Dalziel, I.W.D., 1997. Neoproterozoic-Paleozoic geography and tectonics: review, hypothesis, environmental speculation. GSA Bulletin, 109: 16-42. Fitzsimons, I.C.W., 2000. Grenville-age basement provinces in East Antarctica: evidence for three separate collisional orogens: Geology, 28: 879-882. Harris, L.B., 1994. Neoproterozoic sinistral displacement along the Darling Mobile Belt, Western Australia, during Gondwanaland assembly. Journal of the Geological Society, London, 151:901-904. Hoffman, P.F., 1991. Did the breakout of Laurentia turn Gondwana inside out? Science, 252: 1409-1412. Hyodo, H. and Dunlop, D.J., 1993. Effect of anisotropy on the paleomagnetic contact test for a Grenvillian dike. J. Geophys. Res., 98: 7997-8017.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Figure 1. (a) Global palaeogeography at ca. 810 Ma, at the beginning of Rodinia breakup. C is Congo, K is Kalahari, T is Tarim, Au is Australia, SC is South China, M is the Mawson craton. La is Laurentia, S is Siberia, Am is Amazonia, W is West Africa, and B is Bahica. India is at the North Pole, and Rio de la Plata is not specifically labelled. Continental blocks are positioned according to palaeomagnetic data and rotation poles in Powell and Pisarevsky (2001). (b) Global palaeogeography at ca. 750 Ma, after Rodinia had begun to break up. Australia, India (In), Congo, South China, and Laurentia are position according to palaeomagnetic data and rotation poles in Powell and Pisarevsky (2001). Mollweide projection, (c) Global palaeogeography at ca. 610 Ma, after India was assembled in its East Gondwanaland position, and the Brazilide ocean had closed merging Congo-Sao Francisco, and possibly Kalahari, with Amazonia and Rio de la Plata cratons.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

NEOPROTEROZOIC EXTENSIONAL EVENTS IN SOUTH AUSTRALIA IN THE PRELUDE TO RODINIA BREAKUP W.V. Preiss Office of Mineral and Energy Resources, PIRSA, GPO Box 1671 Adelaide SA 5001.

clear; such terranes may include Cambrian volcanic arcs and parts of Tasmanian Proterozoic crust.

INTRODUCTION The Neoproterozoic to Middle Cambrian Adelaide Geosyncline (Preiss, 1987, 2000) of eastern South Australia borders and onlaps older Precambrian cratons. It records the rift events that led to eventual continental separation, and the first effects of plate convergence in the Middle to Late Cambrian Delamerian Orogeny. The approximately north-south trending Torrens Hinge Zone (THZ) separates the Archaean to Mesoproterozoic Gawler Craton (Daly et aL, 1998) to the west, with its thin platform cover of Neoproterozoic and Early Cambrian sediments of the Stuart Shelf, from the rift complex of the Adelaide Geosyncline to the east, with its extremely thick, deformed sedimentary pile. The Cumamona Province (Robertson et al, 1998; Conor, 2000) is a block of Proterozoic crust northeast of the Adelaide Geosyncline that has a central cratonic region surrounded by rifted and deformed basement at its periphery. The Adelaide Geosyncline and the superimposed Delamerian Orogen (-0.5 Ga) have less certain tectonic relationships with the Palaeozoic Tasman orogenic belts of eastern Australia, the original boundaries being largely obscured by later Palaeozoic tectonic structures, e.g. Moyston Fault of western Victoria (Vandenberg et al, 2000) as well as extensive younger cover. The basement beneath the Adelaide Geosyncline underwent repeated extension and subsidence to provide accommodation space for up to 15 km of sedimentary fill comprising the Neoproterozoic Callanna, Burra, Umberatana and Wilpena Groups, as well as Early to Middle Cambrian deposits of the Hawker, Normanville, Kanmantoo and Lake Frome Groups. The stratigraphy and structure of the basin record at least four separate rift events, culminating in the break-up of Rodinia which is interpreted to have taken place at --0.7 Ga. In addition, renewed rifting in the Early Cambrian formed the Kanmantoo Trough (Belperio et aL, 1998), orientated obliquely to the earlier rifts. This trough formed an incipient but failed rift between Australia and Antarctica, long after the eastern continental margin of AustraliaAntarctica had been established. Whereas the effects of the -0.5 Ga Delamerian Orogeny on the sedimentary fill, extending south into Antarctica as the Ross Orogen, are well documented, the role of potential colliding or accreting cmstal blocks is less

BASEMENT GEOLOGY The Gawler Craton is of a collage of late Archaean, Palaeoproterozoic and Mesoproterozoic orogenic belts involving numerous sedimentary and volcanic successions and plutonic intrusions, variably deformed and metamorphosed. Near the eastern margin, on eastern Eyre Peninsula, an -1.74 Ga proximal clastic and felsic volcanic succession (Moonabie Formation, Macgregor Volcanics) overlies older Palaeoproterozoic metamorphic and plutonic rocks and passes eastward on Yorke Peninsula into more distal sediments and volcanics of similar age (Wallaroo Group). The Wallaroo Group is moderately deformed and metamorphosed; the age of this deformation is uncertain, but at least its later phases are coeval with intrusion of-1.59 Ga granites of the Hiltaba Suite. The preNeoproterozoic history of the THZ is also uncertain, but Palaeoproterozoic to Mesoproterozoic rocks of the eastem Gawler Craton extend beneath the Neoproterozoic sediments in the THZ and form its basement. Basement within the Delamerian Orogen is known from the Willyama and Mount Painter Inliers and five inliers in the Mount Lofty Ranges near Adelaide. Deformation and metamorphism in the inliers took place in the -1.6 Ga Olarian Orogeny, best known from the Willyama Inliers of the Cumamona Province, where it affects a distinctive 1.71-1.65 Ga metasedimentary succession (Willyama Supergroup) with lesser felsic and mafic volcanic intercalations. The lower part of this stratigraphy (Cumamona Group) is affected by pervasive albitisation (partly diagenetic, partly syntectonic) making recognition of the precursor lithotypes uncertain. While only a relatively thin succession (-4 km) is involved in repeated recumbent to inclined complex folding, the total stratigraphic content of the sedimentary basin, which includes the giant Pb-Zn-Ag Broken Hill orebody, is uncertain, and hypothetically could include older sediments overlapping with the age range of the rocks on the eastem Gawler Craton. In the central, cratonic portion of the Cumamona Province, the deformed Willyama Supergroup is overlain by essentially undeformed Mesoproterozoic mafic and

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

felsic volcanics and clastic sediments, and by thin remnants of flat-lying Neoproterozoic and Cambrian platform cover.

mobilisation of the evaporitic rift sediments during subsequent deposition.

SYNDEPOSITIONAL TECTONICS OF THE ADELAIDE GEOSYNCLINE

Rifting at this time resulted fi-om generally E-W extension, bounded to the west by the THZ. Major grabens and horsts formed adjacent to the THZ in the Mid-North and Willouran Ranges, with thick, coarse-grained, basal fluvial facies (Rhynie Sandstone) and local mafic volcanism. The bimodal Boucaut Volcanics (rhyolite component - 7 8 0 Ma), south of the Cumamona Province, probably resulted ft-om the same extension event. The remainder of the fill includes lacustrine to paralic carbonates (e.g. Skillogalee Dolomite), but even these have strong westerly clastic input adjacent to the Gawler Craton. These were followed by clastic sediments deposited during marine transgression into the rift basins. The troughs were probably still intracontinental, with eastern and western sediment provenances; sedimentation extended onto the periphery of the Cumamona Province, but not to its central cratonic region, nor onto the Gawler Craton in the west.

-780 Ma. Burra Group

The distribution and orientation of mafic dykes, shifting depocentres, and disposition of synsedimentary extensional faults help to define the structure of distinct rifted troughs with different orientations and timing. These suggest both changes in the extensional stress regime through the basin's nearly 150 million year pre-breakup history, and control by pre-existing basement structures, in particular the NNW-trending G2 corridor (O'Driscoll, 1983). The THZ marks the eastern limit of unattenuated Precambrian crust; it is expressed as a zone of extensional faulting during deposition of the Burra Group, flexuring during the later sag phases of deposition, and a zone of gentle, open folding in the Delamerian Orogeny, situated between the Gawler Craton and the main thrust front of the Delamerian Orogen.

-700 Ma. Yudnamutana Subgroup of the Umberatana Group

NEOPROTEROZOIC EXTENSIONAL EVENTS AND THE TIMING OF CONTINENTAL BREAKUP The following recognised:

four major rift events

New rifts trending NW to E-W were established around the margins of the Cumamona Province, where very thick Sturtian glacial successions were deposited. Iron-rich facies, comparable to those of the Rapitan in the NW Canadian Cordillera, were deposited in the Baratta Trough south and west of the Cumamona Province, but not in the equivalent Yudnamutana Trough to the northwest and Torrowangee Trough east of Broken Hill. Postglacial marine transgression onto the Gawler Craton is recorded by the carbonaceous siltstone of the Tapley Hill Formation, marking the onset of the first major sag-phase sedimentation and probably the start of continental separation. Younger strata of the Umberatana and Wilpena Groups (including deposits of the Marinoan glaciation) represent shallow to deep shelf environments, with further minor rifting likely to have continued. However, there is little evidence from sediment thickness and facies pattems for major growth faulting, except along the THZ during deposition of the Wilpena Group.

are

-830 Ma. Arkaroola Subgroup of the Callanna Group Early NE-SW extension manifested by the 827 Ma Gairdner Dyke Swarm within the Gawler Craton and coeval dykes near Broken Hill (Wingate et ai, 1998) followed a brief period of stable, epicontinental sedimentation in a shallow epeiric sea and did not lead to the formation of major depocentres. There is no evidence of significant uplift prior to this magmatism, as might be expected from major mantle plume activity. Coeval mafic lavas, originally widespread, were deeply buried by later rift deposits, and are either locally preserved near basin margins (e.g. Wooltana Volcanics) or have been brought to the surface as xenoclasts in diapiric breccia.

-800 Ma. Curdimurka Subgroup of the Callanna Group

^ 9 0 Ma. The Koonenberry Belt of western NSW

The first major rift valleys, formed by mostly NW-trending extensional faults, were restricted basins with mixed immature clastics and carbonates deposited under evaporitic conditions. Associated felsic and mafic volcanism was minor and localised. These troughs underlie and are mostly obscured by younger, more extensive Neoproterozoic strata, but to some extent their distribution can be inferred from the distribution of diapirs which resulted fi-om the

This belt contains thick basalts, interlayered in mostly fine-grained clastics, that are provisionally correlated with volcanics at Mount Wright dated at - 5 9 0 Ma (Crawford et aL, 1997). No similar late Neoproterozoic mafic volcanism, reflecting major cmstal extension, is recorded in equivalents within the Adelaide Geosyncline in South Australia, but may be represented on King Island. The

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

trending orogenic beh as a resuh of the 1.6 Ga Olarian Orogeny and, perhaps. Broken Hill-style base metal mineralisation, should be sought in the conjugate continental margin. In addition, Conor (1998) has suggested a link between the 1.6 Ga regional alteration and mineralising systems of the eastern Gawler Craton, portions of the Cumamona Province and the Cloncurry district of Queensland, and the Wemecke Breccias of Canada. Such a linkage would be consistent with the original juxtaposition of the Canadian Cordilleran margin with eastern Australia as suggested by the stratigraphic similarities in the rift and immediately post-rift sedimentary successions on the two passive margins.

autochthoneity of the Koonenberry Belt is open to question, as no basement is known by which to Hnk it to the Austrahan cratons. The mafic magmatism is therefore of uncertain significance in considering the timing of continental separation.

-520 Ma. Early Cambrian rifting and the nature of the Kanmantoo Trough Early Cambrian sedimentation (Hawker Group, Normanville Group) in the Adelaide Geosyncline was mainly in shelf to ramp settings, dominated by carbonates and reefs, but generally deepening upward. In the late Early Cambrian, the Kanmantoo Trough formed by rifting across the trend of older troughs, truncating the THZ in the south. Here the trough swings into E-W orientation on Kangaroo Island, and is characterised by rapid subsidence and infill with turbidites in two megasequences, each culminating in shallow-water sandstone. The sedimentary fill is ensialic, overlying Neoproterozoic sediments east of the THZ and probably directly overlying down-faulted basement where the trough forms a re-entrant into the Gawler Craton on Kangaroo Island. Local, dominantly mafic volcanism (-520 Ma), immediately pre-dating formation of the trough, extends under the Murray Basin into the Padthaway Ridge of southeast SA and into the Glenelg River region of Victoria, where Kanmantoo equivalents have long been inferred.

References Belperio, A.P., Preiss, W.V., Fairclough, M.C., Gatehouse, C.G., Gum, J., Hough, J. and Burtt, A., 1998. Tectonic and metallogenic framework of the Cambrian Stansbury Basin Kanmantoo Trough, South Australia. AGSO Journal of Australian Geology and Geophysics, 17:183-200. Conor, C.H.H., 1998. Alteration and mineralisation in the Moonta-Wallaroo district of the eastern Gawler Craton, a comparison with the southern Cumamona Province. Geological Society of Australia. Abstracts, 49:88. Conor, C.H.H., 2000. Definition of major sedimentary and igneous units of the Olary Domain, Cumamona Province. MESA Journal, 19:51-56. Crawford, A.J., Stevens, B.P.J, and Fanning, C.M., 1997. Geochemistry and tectonic setting of some Neoproterozoic and Early Cambrian volcanics in westem New South Wales. Australian Joumal of Earth Sciences, 44:831-852. Daly, S.J., Fanning, C.M. and Fairclough, M.C., 1998. Tectonic evolution and exploration potential of the Gawler Craton, South Australia. AGSO Joumal of Australian Geology and Geophysics, 17:145-168. Flottmann, T. and James, P., 1993. Influence of basin architecture on the style of inversion and fold-thmst belt tectonics - the southern Adelaide Fold-Thrust Belt, South Australia. Joumal of Stmctural Geology, 19:1093-1110. Flottmann, T. James, P., Rogers, J. and Johnson, T., 1994. Early Palaeozoic foreland thmsting and basin reactivation at the Palaeo-Pacific margin of the southeastem Australian Precambrian Craton: a reappraisal of the stmctural evolution of the Southem Adelaide Fold-Thmst Belt. Tectonophysics, 234:95-116. O'Driscoll, E.S.T., 1983. Deep tectonic foundations of the Eromanga. Basin. APEA Joumal, 23:5-17. Preiss, W.V. (compiler), 1987. The Adelaide Geosyncline - late Proterozoic stratigraphy, sedimentation, palaeontology and tectonics. South Australia. Geological Survey. Bulletin, 53. Preiss, W.V., 2000. The Adelaide Geosyncline of South Australia, and its significance in continental reconstmction. Precambrian Research, 100:21 -63. Robertson, R.S., Preiss, W.V., Crooks, A.F., Hill, P.W. and Sheard, M.J., 1998. Review of the Proterozoic geology and mineral potential of the Cumamona Province in South Australia. AGSO Joumal of Australian Geology and Geophysics, 17:169-182. Vandenberg, A.H.M., Willman, C.E., Maher, S., Simons, B.A., Cayley, R.A., Taylor, D.H., Morand, V.J., Moore, D.H. and Radojkovic, A., 2000. The Tasman Fold Belt System in Victoria. Geological Survey of Victoria, Special Publication. Wingate, M.T.D., Campbell, I.H., Compston, W. and Gibson, G.M., 1998. Ion-probe U-Pb ages for Neoproterozoic basaltic magmatism in south-central Australia and implications for the breakup of Rodinia. Precambrian Research, 87:135-159.

DELAMERIAN OROGENY In South Australia the first evidence of compressive deformation is the onset of the Delamerian Orogeny, around the early Middle Cambrian; all earlier deformation relates to diapirism and rotation of extensional fault blocks. It is likely that sedimentation of the mostly Middle Cambrian Lake Frome Group redbeds in the central Flinders Ranges region was still proceeding while north-west directed thrusting was under way in the Mount Lofty Ranges region, where extensional faults, especially those defining the Kanmantoo Trough, were inverted (Flottmann and James, 1993; Flottmann (3/., 1994).

CONTINENTAL RECONSTRUCTION The Adelaide Geosyncline records the formation of a Neoproterozoic passive continental margin during the break-up of Rodinia. In trying to identify a potential conjugate continental margin on another present-day continent it will be necessary to compare not only the sedimentary record preserved on the ancient passive margins, but also details of the basement geology. In particular, the Palaeoproterozoic to Mesoproterozoic rocks of the Cumamona Province are critical; extensions of the distinctive --1.7 Ga Na-metasomatised metasediments, with their magmatic, metamorphic and deformational history in a generally northeasterly

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

DENUDATION OF THE GRENVILLE OROGEN RECORDED BY DETRITAL ZIRCON PROVENANCE STUDIES OF EARLY NEOPROTEROZOIC SANDSTONES ON THE MARGINS OF LAURENTIA R.H. Rainbird Geological Survey of Canada (GSC), 601 Booth St., Ottawa, Ontario, Canada K1A 0E8; rrainbir@nrcan.gc.ca

REGIONAL STUDIES - NORTHERN CORDILLERA

INTRODUCTION The assembly of Rodinia was accompanied by global collisional orogenesis preserved today as extensive linear belts of moderate to high-grade metamorphic rocks, such as the Grenville orogen of Laurentia and Baltica. The presence of these metamorphic rocks at surface today indicates uplift and erosion of lO's of kilometres of crust. The bulk of this mountain building was completed by 1.0 Ga, but erosion and denudation has continued ever since (a remnant of this Grenvillian highland is the Laurentian "mountains" of southwestern Quebec). The record of this uplift and erosion should be preserved in the vestiges of a pan-Rodinian drainage system that would have delivered huge volumes of detritus to adjacent basins.

The big-river hypothesis was tested by comparing the detrital zircon geochronology of the Shaler Supergroup with that of potentially correlative strata from the Ogilvie and Mackenzie platforms in the northern Canadian Cordillera (Rainbird, et al. 1996, Rainbird, et al. 1997). Of 54 detrital zircons, separated from five regionally correlative samples from the Mackenzie Mountains supergroup and the Pinguicula group, 85% were of Mesoproterozoic age with a high proportion of these clustering in a range between 1.25-1.0 Ga. As well, Sm-Nd isotopic data from intercalated mudrocks indicated a source with relatively juvenile model ages ( T D M = 1.74-1.54 Ga) consistent with the provenance indicated by detrital zircon geochronology. These results strongly supported the big-river model and suggested that the fluvial system was laterally extensive and may have formed a broad cratonic sheet originating from multiple sources along the length of the Grenvillian mountain front.

INITIAL STUDIES - SHALER SUPERGROUP U-Pb dating of detrital zircons was applied to assess the provenance of fluvial sandstones of the Shaler Supergroup, an early Neoproterozoic (c. 1.00.75 Ga) intracratonic basin succession preserved on the northwestern (present coordinates) margin of Laurentia in the Canadian Arctic. These and correlative sandstones exhibit consistent northwesterly paleocurrents suggesting that they derived from erosion of the adjacent craton. Detrital zircon geochronology supported this assertion with a significant proportion of the ages matching that of the adjacent Archean Slave craton and its marginal Paleoproterozoic orogenic belts. However, the majority of the detrital zircons from the Shaler Supergroup yielded late Mesoproterozoic ages, unlike any known proximal source terrain, but quite similar to the ages of extensive synorogenic plutons in the Grenville province of southeastern Laurentia. This observation led to the development of a hypothesis whereby detritus was transported approximately 3000 km northwestward from the rising Grenvillian orogeny by a pan-continental fluvial system (Rainbird, et al. 1992).

REGIONAL STUDIES - CENTRAL CORDILLERA Detrital zircon geochronology of Neoproterozoic sedimentary rocks from sedimentary basins located along the western margin of Laurentia, and now exposed mainly in the central Cordillera, yielded detrital zircons with characteristic "Grenvillian" ages (e.g. Stewart, et al. 2001). Although they attribute some of these ages to provenance from local volcanic centres, much of the material is considered to have derived from the Grenville orogen, elements of which extend into west Texas and northern Mexico. These data provide further support for the big-river hypothesis and for the presence of a craton-scale fluvial blanket in Neoproterozoic times.

SIBERIA Affinities between Laurentia and Siberia were proposed by Sears and Price (1978) and Condie and

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

A CRYPTIC GRENVILLIAN FORELAND BASIN IN THE U.S. MIDCONTINENT

Rosen (1994) and the similarities between the northern Cordillera and foldbehs along the eastern margin of Siberia have long been recognised (Khudoley and Guriev 1990). Subsequent models have been proposed but the position of Siberia remains an enigmatic component of the Rodinia reconstruction. Neoproterozoic (upper Riphean) sandstones of the Siberian platform share first-order similarities with their counterparts in Laurentia and yielded indirect evidence for a "Laurentia-Siberia" connection in their detrital zircon age profiles. The detrital zircon ages of the Mayamkan Formation, a fluvial arkose from the southern Sette-Daban fold belt, is almost identical to that of early Neoproterozoic fluvial sandstones of the Shaler and Mackenzie Mountains supergroups in northwestern Canada and include a high proportion (-^85%) of Mesoproterozoic zircons (Rainbird, et al. 1998). There is no known source region for detritus of such age in Siberia and isopachs and facies relations suggest provenance from an exotic terrane to the east. These data indicated that provenance of the late Riphean sandstones was mainly from rocks of Grenvillian age that were perhaps adjacent to a unidentified continent (Laurentia or Baltica?).

Proximal parts of the big- river system may be present in a westward-tapering wedge of coarse, immature clastic red-beds of interpreted Neoproterozoic age from the subsurface of western Ohio (Shrake, et al. 1991). These rocks, known as the Middle Run Formation, have been cored and imaged in subsurface seismic profiles, which reveal a wide, well-defined zone of east-dipping reflectors inferred to represent thrust structures of the Grenville front tectonic zone (COCORP Line OH-1; Hauser 1993). In the subsurface to the west is a shallow, east-dipping sequence of sedimentary strata, similar to Middle Run Formation elsewhere. Together these rocks were interpreted as the molassic phase of a previously unrecognized foreland basin to the Grenville orogen (Hauser 1993). A potential correlative to the Middle Run Formation is the Jacobsville sandstone, a >900m thick succession of subarkosic to sublithic arenites, conglomerates and siltstones, which are exposed in the Keweenaw Peninsula-eastern Lake Superior region. Tectonic uplift and provenance from the south is indicated by petrology and paleocurrent analysis (Kalliokoski 1982). Paleocurrents from fluvial units in the main part of the basin suggest axial (NE and SW) transport, perhaps related to development of a fluvial trunk system, common in foreland basins.

SCOTLAND Provenance of the late Mesoproterozoic to early Neoproterozoic Torridonian succession in NW Scotland has considerable consequences for the nature and timing of the formation and break-up of Rodinia. Previous provenance studies based on the sedimentology, geochemistry and mineralogy of these relatively immature coarse clastic rocks argued for a possible influence from Laurentia, from which Scotland is now separated. Recent detrital zircon geochronology studies indicated that the lower part of the Torridonian (Stoer Group) is composed mainly of sediment weathered from proximal late Archean sources in the Hebridean block, a piece of the North Atlantic craton of Laurentia, orphaned with the opening of lapetus (Rainbird, et al. 2001). The upper part of the Torridonian (Torridon Group) exhibits more varied provenance with distinctive detrital zircon age modes at c. 1.80 Ga, 1.65 Ga and 1.1 Ga. The latter two modes are considered to be distinctive of the Grenville Province in eastern Laurentia. These data together with paleocurrents suggest that the Torridon Group could have been deposited by a late-post Grenvillian foreland trunk river system. This is significant because the foreland basin part of the big-river system should be preserved somewhere along the great length of the Grenville thrust front.

PANGEA ANALOGUE An excellent and better preserved analogue to the proposed Rodinian paleogeographic model comes from the supercontinent Pangea, which amalgamated during the Appalachian-Hercynian orogeny in the Carboniferous-Triassic. The paleogeography of the Laurasia block of Pangea was dominated by a series of coalescing, alluvial-deltaic wedges and axial braided rivers that filled foreland basins formed by flexural loading along the Alleghenian-Appalachian thrust front (Absaroka Sequence of (Sloss 1988). The Central Appalachian Basin is an example whose infill has been interpreted to represent an Amazonscale drainage system (Archer and Greb 1995). As the foreland basins filled, excess detritus was transported westward, across the craton, by fluvial and eolian processes. In a similar fashion, the foreland basin to the Grenville orogen was probably overfilled with excess detritus spilling westward across Laurentia to intracratonic basins on its opposite margin. The lack of preservation of the proximal parts of this system is in part due to Phanerozoic cover, but may also be related to subdued flexural loading, a consequence to climate-

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Rainbird, R. H. et al 1997. Pan-continental River System Draining Grenville Orogen Recorded by U-Pb and Sm-Nd Geochronology of Neoproterozoic Quartzarenites and Mudrocks, Northwestern Canada. Journal of Geology 105, 118. Rainbird, R. H. et al 1998. U-Pb geochronology of Riphean sandstone and gabbro from southeast Siberia and its bearing on the Laurentia-Siberia connection. Earth and Planetary Science Letters 164/3-4, 409-420. Rainbird, R. H. et al 2001. Detrital zircon geochronology and provenance of the Torridonian, NW Scotland. Journal of the Geological Society [London] 158, 15-27. Sears, J. W. & Price, R. A. 1978. The Siberian connection: a case for Precambrian separation of the North American and Siberian cratons. Geology 6, 267-270. Shrake, D. L. et al 1991. Pre-Mount Simon basin under the Cincinatti Arch. Geology 19, 139-142. Sloss, L. L. 1988. Tectonic evolution of the craton in Phanerozoic time. In Sloss, L. L. Sedimentary Cover-North American craton. Geological Society of America, Boulder, Colorado. The Geology of North America, D-2, pp. 25-51. Stewart, J. H. et al 2001. Detrtital zircon provenance of Meosproterozoic to Cambrian arenites in the western United States and northwestern Mexico: Geological Society of America Bulletin 113, (in press).

controlled erosional unroofing of the mountain front (Rainbird et al. 1997). References Archer, A. W. & Greb, S. F. 1995. An Amazon-scale drainage system in the early Pennsylvanian of central North America. Journal of Geology 103, 611-628. Condie, K. C. & Rosen, O. M. 1994. Laurentia-Siberia connection revisited. Geology, 22, 168-170. Hauser, E. C. 1993. Grenville foreland thrust belt hidden beneath the U.S. midcontinent. Geology, 21, 61-64. Kalliokoski, J. 1982. Jacobsville Sandstone. In Wold, R. J. and Hinze, W. J. Geology and Tectonics of the Lake superior Basin, Geological Society of America, Memoir 156, 147-155. Khudoley, A. K. & Guriev, G. A. 1990. South Verkhoyansk and Cordillera-a comparative analysis. Soviet Geology N6, 67-76 (in Russian). Rainbird, R. H. et al 1992. Sampling Laurentia: Detrital zircon geochronology offers evidence for an extensive Neoproterozoic river system originating from Grenville orogen. Geology 20, 351-354. Rainbird, R. H. et al 1996. The early Neoproterozoic sedimentary Succession B of northwest Laurentia: correlations and paleogeographic significance. Geological Society of America Bulletin 108, 454-470.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

THE HIGH PRESSURE BELT IN THE GRENVILLE PROVINCE TECTONIC SETTING, EVOLUTION, PRESERVATION AND POSSIBLE CORRELATIONS WITH THE SVECO-NORWEGIAN BELT IN SWEDEN T. Rivers Department of Earth Sciences, Memorial University, St. John's, Newfoundland, Canada A1B 3X5.

normal-sense displacement. Available geochronologic data are compatible with the interpretation that displacements on the detachments in the hangingwall of the HP beh were in part coeval with the uplift of the HP belt along a crustal-scale thrust-sense shear zone in its footwall, pointing to tectonic extrusion as the mechanism for exhumation of the HP beh. Despite important differences in their preGrenvillian evolution, including the development and closure of a major Mesoproterozoic backarc basin in the western Grenville Province but its absence farther east, comparisons of the Grenvillian tectonic evolution of the HP belt in the western and central Grenville Province show comparable overall histories of thrusting and normal fauhing. However, we infer an important episode of footwall collapse beneath the HP terranes in the western Grenville Province that is less well developed farther east and may be correlated with the pervasive retrogression of the HP units. In addition, the lowest major fauh in the detachment system transects the HP belt, being structurally below it in the western Grenville Province, but above it in the central Grenville Province, thereby placing the HP belt in its hangingwall in the west, but in its footwall farther east. Tectonic extrusion of the HP rocks in both areas was achieved by alternating and/or coeval episodes of compression and extension in a foreland-ward propagating orogen. Mid- to late-Ottawan thrusting and normal faulting resulted in the HP terranes being in the mid to upper crust by late Ottawan times, after which they were passively transported towards the foreland as orogenic float during the ensuing Rigolet orogeny (-1010-1000 Ma). High pressure, relict eclogite facies units have also been reported from the Sveco-Norwegian orogen in southern Sweden, and preliminary comparisons are drawn between the tectonic setting and timing of the HP rocks in the Grenvillian and Sveco-Norwegian orogens.

The allochthonous terranes in the Grenville Province can be grouped into High Pressure (HP) and Low Pressure (LP) belts on the basis of the time and character of their Grenvillian metamorphic signatures. Both are largely composed of Paleo- and Mesoproterozoic plutonic units that constituted parts of the former Laurentian margin. The HP belt, characterised by eclogites and co-facial HP granulites, is known in two locations, each of several hundred square km extent, about 800 km apart in the western and central Grenville Province. In both of these areas, the HP belt tectonically overlies the Parautochthonous beh, the lowest lithotectonic unit in the Grenville orogen. The HP assemblages formed during the Ottawan orogeny between -11201060 Ma, with maximum P-T estimates in the central Grenville Province, where prograde metamorphic textures and mineralogy are preserved, being between 1800-1400 MPa and 900-800 P-T estimates from the western Grenville Province, where the HP units are tectonically disrupted and the HP mineralogy is extensively retrogressed, yield post-peak P-T conditions in the range 1450-1350 MPa and 750-700 Metamorphism in the underlying Parautochthonous beh took place at -1010-1000 Ma (Rigolet orogeny), and occurred as a result of tectonic emplacement of the HP belt as an orogenic wedge on the foreland to the northwest, with metamorphic, foreland-directed fold-thrust belts developed in the underlying Parautochthonous terranes as a result of basal accretion to the advancing orogenic wedge. Metamorphism in the Parautochthonous beh is typically of Harrovian type, with maximum P-T in the range 1100 MPa and 650°C. Metamorphism in the LP beh, which tectonically overlies the HP belt, was of variable temperature during the Grenvillian orogenic cycle, in some terranes being recorded by the '^^Ar/^^Ar system in muscovite, but not hornblende, suggesting 500 > T > 350 The LP beh is separated from the HP beh by an important system of ductile detachments with

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

PALAEOMAGNETIC EVIDENCE FOR LOW-LATITUDE GLACIATION: THE GEOCENTRIC AXIAL DIPOLE HYPOTHESIS FOR THE PROTEROZOIC? P.W. Schmidt CSIRO Division of Exploration and Mining, PO Box 136, North Ryde, NSW 1670.

PALAEOMAGNETISM ELATINA FORMATION

suggesting that, for some of the time at least, the magnetization and the reversal processes shared a similar time constant that has resulted in many zones of mixed polarity. Since Harland's controversial suggestion of low palaeolatitude glaciations, the lingering doubts that the deposits were glacially derived have abated, only to be replaced with a multitude of arguments for and against their apparent disposition towards low palaeolatitudes. Of the mechanism that has been promoted, some are easily laid to rest. The ballistic ejecta theory would predict lateral thinning of the deposits. There is no evidence of this. Furthermore, the ejecta theory does not explain many other features that are unique to glaciations. The equatorial ice-ring theory fails in that by analogy with Saturn, equatorial insolation, although reduced, remains higher than at other latitudes. There is also debate as to whether an ice-ring could exist as close to the Sun as Earth. After giving short shrift to impact ejecta and equatorial planetary ring ideas, the three possible scenarios that seem most probable to explain the observations are 'Snowball Earth' [8], large obliquity [9] and the gross failure of the geocentric axial dipole (GAD) hypothesis to describe the Proterozoic geomagnetic field [10]. A comprehensive summary of Neoproterozoic glacial deposits and the current state of play are provided by Evans [11]. In this work Evans concludes that Phanerozoic 'archetypal' glaciation is unable to account for the low-latitude continental sea-level glaciations observed in the Proterozoic. The remainder of this article will focus on the GAD hypothesis for the Proterozoic.

THE

There is now irrefutable evidence that at least some glaciations in the Proterozoic occurred in low palaeolatitudes, confirming the early suggestion of W.B. Harland [1]. The best evidence for low latitudes is currently from palaeomagnetic studies of the Neoproterozoic Elatina Formation in the Flinders Ranges, South Australia [2,3]. In addition, there are recent data from the Neoproterozoic Rapitan Group of North America that also indicate low palaeolatitudes of formation [4] and from Palaeoproterozoic glacial successions of North America [5] and Africa [6] that are consistent with low palaeolatitudes of formation. The arguments originally articulated by Embleton & Williams [7] concerning the timing of remanence of the Elatina Formation at Pichi Richi Pass have been entirely vindicated. Mineragraphic identification of high temperature forms of titanohematites coupled with the low metamorphic grade (burial diagenesis <135°C to 160°C) and observed discrete unblocking temperatures just below the hematite Curie temperature, effectively ruled out any chemical or thermal mechanisms of remagnetisation. However, perhaps the strongest evidence for the Elatina Formation having acquired its characteristic magnetization early is a positive fold test on soft-sediment slumps. This test has now been performed on three separate occasions and each time it has proved positive. However, none of the above studies answered growing objections that the Elatina palaeopole was a virtual geomagnetic pole (VGP), having sampled the field for less than a hundred years. Since then there have been two independent regional palaeomagnetic studies that overcome these doubts. The earlier study involved 10 sites while the latter 58 sites. The combined result yields a palaeopole at 41.6°S, 359.6°E (dp = 3.0°, dm = 5.9°) and a palaeolatitude of 7.9°±3.0°. Neither study was able to provide a convincing magnetostratigraphy although Schmidt & Williams [2] state that "it is possible to define a rough magnetostratigraphy... where polarity changes from normal... to reversed... down-section". Sohl et al.'s Fig. 10b [3] confirms this. The earlier study identified magnetisation reversals within specimens

THE GEOCENTRIC AXIAL DIPOLE HYPOTHESIS The Geocentric Axial Dipole (GAD) hypothesis underpins the whole palaeomagnetic edifice although it is almost certainly only approximate. The dipole equation states that tan [/] = 2tan[A], where I is magnetic inclination and /I is geographic (palaeo)latitude. The elegant simplicity of the dipole equation has probably conferred upon it a degree of reverence that has deterred some from doubting its veracity, let alone seeking an improvement.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

careful to 'bin' the data temporally and geographically, and also examined separately data from sedimentary and crystalline rocks. Although they found the GAD hypothesis entirely adequate for Cenozoic and Mesozoic results, their analysis suggested the persistence of small but significant non-dipole fields for the Palaeozoic and the Precambrian. The distributions of |/| for these times was found to be consistent with a 10 percent quadrupole and, surprisingly, a 25 percent octupole. The size of the octupole is surprising because it is generally assumed that the higher the order, the more quickly a multipole decays with distance from its source. Certainly the present octupole component is smaller than the quadrupole and the hexadecapole is smaller than the octupole. It may be that the octupole could be larger than the quadrupole by virtue of it belonging to the dipole family (McFadden et al, [17]). The dipole and quadrupole families are intrinsically different, have opposite symmetries and display distinctively different secular variation patterns. Perhaps it should not be surprising that these analyses suggest small but persistent non-dipole components. However, octupole and quadrupole contributions will affect palaeolatitude determinations differently. A small (<-30%) axial octupole field will not affect the lowpalaeolatitude interpretation of Proterozoic glaciation. Extreme (>50%) octupole content of the same sign as the dipole field produces multi-valued inclination/latitude relationship with the field having zero inclination at ±25° latitude as well as at the equator.

However, the closer palaeomagnetists look at the ever-growing database, the more small anomalies and departures from expectations appear. While these may be second order effects they need to be calibrated if palaeomagnetism is to move forward and yield reliable palaeolatitudes and palaeoreconstructions. Research topics potentially affected by the implications of non-dipole geomagnetic fields include those of low-latitude glaciation and supercontinents (Pangaea, Rodinia or its replacement) evolution. A new era of palaeomagnetic research should aim to refine the GAD by adding the first two zonal harmonics. While it is true that the GAD hypothesis has stood palaeomagnetism in good stead over the decades, there has always been an underacknowledged awareness that the GAD is an approximation. After all, there is no reasonable doubt that the Late Tertiary geomagnetic field, a time for which we have an order of magnitude more data than for any other period, contained a persistent - 4 percent (zonal) quadrupole field. Pioneering work by Wilson and Ade-Hall [12] showed that poles from young rocks of Europe and Asia were 'far-sided'. That is, they overshot the present north geographic pole by a few degrees. Numerous studies have subsequently confirmed this effect and demonstrate that it is a worldwide phenomenon (see Merrill et al, [13] for a frill discussion). Merrill and McElhinny [14] suggested that regional average poles should be calculated two ways, using the standard dipole relationship and using a modified relationship to correct for the quadrupole. Thus, for quite a while the question has been, how good is the GAD approximation for early geomagnetic times? To extend the investigation of (zonal) non-dipole fields to earlier times, Evans [15] examined the distribution of (absolute) inclinations, |/|, from rocks up to 600 Ma that were available from the pole lists of the time (Irving's and McElhinny's). Making the assumption that the geographic locations where the rocks sampled had acquired their magnetisations were uniformly distributed on the globe, he compared the distribution of |/| with that expected from a dipole. Evans [15] concluded that there was no reason to doubt the validity of the GAD hypothesis for the past 600 Ma. Piper and Grant [16] repeated Evans' procedure for rocks up to 3000 Ma and came to the same conclusion, however, as noted by Merrill et al [13], the data used were not selected for quality and it is likely therefore that the data set contains results that are themselves biased. Kent and Smethurst [10] also point out that Piper and Grant [16] did not class ('bin') their data according to geographical region and their results may therefore be biased by regions for which more data has been acquired. Indeed, Kent and Smethurst [10] also analysed the distribution of |/| for Precambrian and Phanerozoic data. They were

A moderate quadrupole field will offset equatorial palaeolatitudes, but not enough to account for the observations, e.g. a 30% quadrupole corresponds to a 13° palaeolatitude error. However, it should be emphasised that analyses of the geomagnetic field configuration using |71 distributions can not resolve a quadrupole field. Because only magnitudes of inclinations are analysed, in effect the quadrupole field becomes asymmetric (the sign of negative inclinations is ignored) and cancels itself. Therefore any quadrupole content may be difficult to recognise in such gross statistical analyses. Kent and Smethurst [10] also discussed the possibilities of other causes for the flatter Palaeozoic and Precambrian distributions, including that of a biased distribution of continents. If continents were spread around the equator, as has been suggested [18], then it would be expected that the distribution of |/| would be skewed to low values, which in effect is the same as adding an octupole of the same sign to the dipole. They frirther speculate that supercontinents produce geoid highs, which migrate equatorwards to align the Earth's rotation and principal inertia axes. If this mechanism has acted in the past then distributions of |/1 for times that

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

supercontinents existed would be expected to be skewed to low values. Interestingly, Bloxham [19] has recently claimed that a significant octupole contribution can be generated from heat-flux variations at the coremantle boundary. Moreover, such variations are plausible for the Palaeozoic and Proterozoic although it is emphasised they would not be expected to be a permanent feature but wax and wane on -250 Ma cycle. Bloxham further speculates, somewhat ironically, the reason that a larger octupole is not manifest in the Tertiary field may be because the time frames examined are too short to sample the complete geomagnetic cycle. The possible bias of palaeolatitudes ascribed to measured palaeomagnetic inclinations is of growing concern, particularly in the face of robust evidence that such inclinations imply low-latitudes for continental sea-level glaciation. While Merrill et al [13] cast doubts on the viability of determining nondipole components beyond the quadrupole, it nevertheless has such important ramifications for palaeomagnetic studies, of palaeolatitudes, palaeoreconstructions and palaeo-intensities alike, that it is compelling, if not incumbent upon us, to examine different methods that might be applied. Moreover, as already noted, a different method than the distribution of |/1 is required to determine quadrupole content.

Asymmetry versus Age

2000

4000

Age (Ma)

Figure 1. Asymmetry of results from GPMDB versus age for times prior to 540 Ma. While it is clearly highly non-unique, for each asymmetric set it is also possible to make some simplifying assumptions to determine a compatible axial non-dipole configuration. A further way is to analyse palaeomagnetic data from large dyke swarms, providing the swarm is large enough and oriented such as to sample the field appropriately [22]. Of course, all the usual assumptions of averaging secular variation etc. are implicit, but probably not attainable, at least not with current data sets. While it may not be possible to characterise the morphology definitively, the dyke data are a useful adjunct to other approaches. The largest dyke swarm known from the Proterozoic is the 1250 Ma Mackenzie-Sudbury dyke swarm of Canada. This swarm has been palaeomagnetically studied along its entire length, and while the data are of variable quality by today's standards, they have been used to test the viability of this approach. Using a modification of the methods developed by Creer et al [23] and Georgi [24] palaeoinclinations of the dyke swarm have been analysed. The modification involves re-introducing the sign of the inclinations and iteration, since the palaeolatitudes are not known a priori and necessarily depend initially on assuming a GAD. These methods express the components of the geomagnetic field in terms of spherical harmonics and through either least squares or minimisation methods determine the Gauss coefficients. The ratio of the axial quadrupole to the axial dipole is known as G2, while the ratio of the axial octupole to the axial dipole is G3. Using either the steepest gradient method or the downhill simplex gave the same result for the best fitting G2 and G3 for the Mackenzie-Sudbury swarm. Errors have also been estimated using bootstrap methods. By random re-sampling of the data and re-analysing each new set an overall mean and a standard error have been estimated.

SOME POSSIBLE METHODS TO TEST THE GAD IN EARLY TIMES There are a number of tests of the GAD hypothesis, or ways of tracking the non-dipole content of the geomagnetic field during the Palaeozoic and Proterozoic. If a supercontinent can be defined with sufficient accuracy then the morphology of the field can be determined directly. This approach has led to varying opinions but most recently it has been shown that an axial octupole field eliminates the need for non-Bullard et al [20, 21] type Pangaeas. That is, a small relaxation of the GAD hypothesis obviates the degeneracy of manifold supercontinents. 'Ockham's razor' would dictate a small, realistic modification to the GAD rather than the insurmountable geological problems posed by latter-day Pangaeas. The examination of the evolution of asymmetric reversals through time may provide another approach to test the GAD. This may be achieved most easily by interrogating the GPMDB for studies where normal and reverse inclinations are reported separately. By simply recording any asymmetries that are present, and their magnitudes, against time any trends present should emerge. Fig. 1 shows an example for times prior to 540 Ma. However, the only discernible trend in this naive analysis of the distribution of asymmetry is a predominance of large asymmetries between 1000 Ma and 540Ma.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

of similar age that occur in Greenland and Baltica should be studied for comparison. Perhaps the main conclusion of the present study, initiated as it was by suggestions that the GAD hypothesis is only approximate, is that a small to moderate amount of axial octupole does not dramatically affect the reliablility of low (and high) palaeolatitude estimates. Only mid-latitude estimates are affected. Consequently, low-latitudes alleged for some Proterozoic glaciations remain a first-order geological and geophysical paradox.

This analysis yielded G2 = -0.12±0.02 and G3 = 0.29±0.015. The morphology of this field is shown in Fig. 2. Mackenzie-Sudbury Data and Model 90

pol e

Quad +0.29 Oct

30 Inc

9f k j N ^ o

150

-30 -60

-90

ACKNOWLEDGEMENTS

Co-latitude

The Australian Research Council and the International Geological Correlation Program are thanked for their support.

Figure 2. Inclination versus co-latitude for a dipole compared to the best-fitting non-dipole model (inset, upper and lower data bounds of Mackenzie-Sudbury data).

References

1. W.B. Harland, Geol. Rundsch., 54, 45-61 (1964). 2. P.W. Schmidt & G.E. Williams, Earth Planet. Sci. Lett., 134, 107-124(1995). 3. L.E. Sohl, N. Christie-Blick & D.V. Kent, GSA Bull., I l l , 1120-1139(1999). 4. J.K. Park, Can. J. Earth Sci., 34, 34-49 (1997). 5. G.E. Williams & P.W. Schmidt, Earth Planet. Sci. Lett., 153, 157-169(1997). 6. D.A. Evans, N.J. Beukes & J.L Kirschvink, Nature, 386, 262266(1997). 7. B.J.J. Embleton & G.E. Wilhams, Earth Planet Sci. Lett., 79, 419-430(1986). 8. J.L. Kirschvink, in J.W. Schopf & C.C. Klein (Eds.) The Proterozoic Biosphere: A multidisciplinary Study, Cambridge University Press, p. 51-52 (1992). 9. G.E. Williams, Earth Sci. Rev., 34, 1-45 (1993). 10. D.V. Kent & M.A. Smethurst, Earth Planet. Sci. Lett., 160, 391-402(1998). 11. D.A.D. Evans, Am. J. Sci., 300, 347-433 (2000). 12. R.L. Wilson & J.M. Ade-Hall, in S.K. Runcorn (Ed.) Palaeogeophysics, Academic Press, New York, pp. 307-312 (1970). 13. R.T. Merrill, M.W. McElhinny & P.L. McFadden, P.L. The Magnetic Field of the Earth: Paleomagnetism, the Core, and the Deep Mantle. Academic Press, Sydney, pp.527 (1996). 14. R.T. Merrill & M.W. McElhinny, M.W., Rev. Geophys. Space Phys., 15, 309-323 (1997). 15. M.E. Evans, Nature, 262, 676-677 (1975). 16. J.D.A. Piper & S. Grant, Phys. Earth Planet. Int., 55, 37-53, (1989). 17. P.L. McFadden, R.T. Merrill & M.W. McElhinny, J. Geophys. Res., 93 ,11583-11588 (1988). 18. N.M. Chumakov and D.P. Elston, Episodes, 12, 115-120 (1989). 19. J. Bloxham, Nature, 405, 63-65 (2000). 20. E.C Bullard, J.E. Everitt & A.G. Smith, Phil. Trans. R. Soc. London A, 258,41-51 (1964). 21. R. Van der Voo & T.H. Torsvik, Earth Planet. Sci. Lett., 187, 71-81. 22. P.W. Schmidt, in Niels Abrahamsen (Ed.), Proceddings of the Nordic Palaeomagnetic Symposium, Aarhus University, pp. 109-113 (1999). 23. K.M. Creer, D.T. Georgi & W. Lowrie, W., Geophys. J. R. astr. Soc., 33, 323-345 (1973). 24. D.T. Georgi, Geophys. J. R. astr. Soc., 39, 71-86 (1974).

Polarity notwithstanding, the quadrupole and the octupole components present in the Mackenzie dyke data are very similar to those determined by Kent and Smethurst [10]. The best fitting geomagnetic field at 1250 Ma comprised a negative 10 percent zonal quadrupole and a positive 30 percent zonal octupole. The octupole has the same polarity as the dipole while the quadrupole is of opposite polarity. The Mackenzie-Sudbury data set is shown in the inset of Fig. 2. Clearly the data leave much to be desired. Most of the individual palaeomagnetic studies were carried out prior to principal component analysis and many used AF demagnetisation only.

CONCLUSIONS The similarity of the zonal non-dipole field found by Kent and Smethurst [10] and the nondipole found here, coupled with Bloxham's [19] theoretical speculations and more recently Van der Voo and Torsvik's [21] findings, is encouraging although clearly very tentative. However, it may be that the problem with the GAD is not as great as it first seems. Much as early 20^^ century geologists were daunted by the prospects of mapping the geology of the sea floor, within a few decades the problem largely evaporated. The preliminary study presented here suggests that the Mackenzie-Sudbury dyke swarm should be the target for a new comprehensive palaeomagnetic (and palaeo-intensity) study, perhaps an international collaborative effort. Transects should be defined to average secular variation for several diagnostic palaeomagnetic isoclines. The samples should be analysed for both palaemagnetic directions and palaeo-intensities. They should be subjected to low-temperature, alternating field and thermal demagnetisation. Additionally, other dykes

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

A RE-EXAMINATION OF THE "YANBIAN OPHIOLITE SUITE": EVIDENCE FOR WESTERN EXTENSION OF THE MESOPROTEROZOIC SIBAO OROGEN IN SOUTH CHINA J.A. Sinclair Tectonics Special Research Centre, Department of Geology and Geophysics, The University of Western Australia, 35 Stirling Highway, Crawley WA 6009, Australia

FIELD RELATIONSHIPS: EVIDENCE FOR A LATER INTRUSION

INTRODUCTION The South China Block is divided into two major cratons, the Yangtze and the Cathaysia, which were joined together by a Late Mesoproterozoic suture (Z.X. Li at al, 1996). Ophiolite suites from the eastern end of the suture have given Sm-Nd ages of 1034±24 Ma, and 1020±30 Ma (Chen et al., 1991), and a SHRIMP U-Pb zircon age of 968 ± 23 Ma (X.H. Li et al., 1994). Where this suture goes further west is unknown. The "Yanbian ophiolite suite" is believed to be a Proterozoic ophiolite in southwest Sichuan, on the western edge of the Yangtze craton. Dating of the volcanic rocks has given Rb-Sr ages of 1006±58 Ma (J. Li, 1983, cited in Sun and Vuagnat, 1992). The traditional tectonic interpretation of the complex is that it represents a north/south trending suture, joining up with the northern margin of the Yangtze craton (Sun and Vuagnat, 1992). However recent work contradicts this view, with structures observed pointing to an east/west trend. This has led to the possibility that it may be an extension of the Yangtze/Cathaysia craton suture. Associated with this suture at the eastern end are 850-800 Ma granitic intrusions, traditionally interpreted as syn- to late-orogenic (Z.X. Li et al, 1999). However their age is 150 Ma younger than the Sibao Orogen, and is similar to the age of a continental rifting event (Z.X. Li et a l , 1999). Ultramafic intrusions dated at 828 ± 7 are indistinguishable in age from the granites, indicating co-magmatism (Z.X. Li et al., 1999). This does not support their orogenic origin. Li et al. (1999) thus argued that this magmatism is related to a mantle plume that signalled the beginning of the breakup of Rodinia. Granitic intrusions are also associated with the "Yanbian ophiolite suite". Ultramafic and mafic intrusions present in the area were thought to be part of the ophiolite complex itself, but recent geological evidence has caused this idea to be questioned.

The lower part of the "Yanbian ophiolite suite" comprises volcaniclastic siltstone and sandstone, grading into intercalated massive basalt and volcaniclastic siltstone. Thick pillow basalts make up the top part of the section. Sun and Vuagnat (1992) interpreted this as a typical volcanic type sequence from an N-type MORE ophiolite. However the large proportion of sediment and its volcaniclastic nature indicates that the "ophiolite" is likely to be of back-arc or island-arc origin. At the base of the "Yanbian ophiolite" are large masses of olivine gabbro and gabbronorite. This section has been interpreted as being the lower part of the ophiolite sequence (Sun and Vuagnat, 1992). However field studies have observed intrusive relationships between the plutonic rocks and the volcanic sequence. This indicates the gabbro intrusion is a later event.

GEOCHEMISTRY: EVIDENCE FOR BACK-ARC BASIN BASALTS AND CONTINENTAL INTRUSIONS Sun and Vuagnat (1992) interpreted the geochemistry of the "Yanbian ophiohte" to have Ntype MORB characteristics. However normalised rare earth element diagrams reveal an overall REE enrichment relative to N-type MORB, and strong LREE enrichment for the volcanic rocks. This pattern indicates a tholeiitic island-arc origin (X.H. Li, 1997). The plutonic rocks show a vastly different REE pattern, with an extremely high degree of LREE enrichment, and a relatively low level of HREE's compared to the volcanic rocks. This indicates a large crustal influence on the intrusion, supporting a plume type origin (Z.X. Li, 1999). Further work currently being undertaken in the areas of isotope dating and geochemistry will better constrain the age relationships and tectonic setting of the "Yanbian ophiolite" and associated plutonic rocks.

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Li, Z.X., Zhang, L. and Powell, C.McA., 1996. Positions of the East Asian cratons in the Neoproterozoic supercontinent Rodinia. Australian Journal of Earth Sciences, 43, pp. 593604 Li, Z.X., Li, X.H, Kinny, P.D. and Wang, J., 1999. The breakup of Rodinia: did it start with a mantle plume beneath South China? Earth and Planetary Science Letters, 173, pp. 171-181 Sun, M., and Vuagnat, M., 1992. Proterozoic Ophiolites from Yanbian and Shimian. Schweiz. Mineral. Petrogr. Mitt., 72, pp. 389-413

References Chen, J., Poland, K.A., Xing, F., Xu, X., and Zhou, T., 1991, Magmatism along the southeast margin of the Yangtze block: Precambrian collision of the Yangtze and Cathysia block of China: Geology, 19, p. 815-818. Li, X.-H., Zhou, G., Zhao, J., Panning, C.M., and W.Compston, 1994, SHRIMP iron microprobe zircon U-Pb age of the NE Jiangxi ophiolite and its tectonic implications: Geochimica, 23(2), pp. 125-131. Li, X.-H., 1997. Geochemical and Sm-Nd isotopic study of Neoproterozoic ophiolites from southeastern China: petrogenesis and tectonic implications. Precambrian Research, 81, pp. 129-144

100

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

PROMISE AND PERIL: MICROANALYSIS, PROVENANCE AND RECONSTRUCTION IN RODINIA AND BEYOND K.N. Sircombe Tectonics Special Research Centre, Department of Geology and Geography, University of Western Australia, 35 Stirling Highway, Crawley WA 6009.

enigmatic. For example, the Itremo Group in Madagascar, (Cox et al., 1998) and the "classic" search for the western source of the Belt supergroup (Ross et al., 1992). These studies typically involve only one or two sedimentary samples chosen to be roughly representative of the unit/succession/basin being examined. Correlation: Where the geological history is reasonably constrained and there is an attempt to match detrital zircon ages with potential source rocks either within a region or to long separated crustal blocks. For example, the Laurentia-Siberia link examined by Rainbird et al. (1998) and the Belt supergroup-North Queensland link examined by Blewett et al. (1998). Samples are also chosen to broadly represent of units/successions and typically three or more samples within a basin may be examined along with detailed geochronology of poorly constrained, but potentially important, protosources. Basin evolution: Helps constrain the geological history (tectonic evolution and paleogeography) of a margin involved in continental formation or breakup. For example, the development of the Paleozoic Delamerian Orogen in South Australia (Ireland et al, 1998) and the Perth Basin in Western Australia (Cawood and Nemchin, 2000). This approach is also applied to examine broadly uniform or monotonous sedimentary successions to confirm lithostratigraphic correlations and to search for finer provenance details indicative of tectonic evolution (e.g. Sircombe et al. 2001). Several samples are typically carefully chosen to represent the regional extent and variation in a succession. A vertical selection of samples through a sedimentary succession is a vital tool in examining the change in provenance through time.

INTRODUCTION

The acquisition of large, statistically valid sets of U-Pb isotopic data from single detrital zircon grains has become routine in a variety of methods (e.g. Dodson et al., 1988; Davis et al., 1994; Machado et al, 1996; Rainbird et al, 1998; DeCelles et al., 2000). In particular, the numbers of projects using microanalysis of single grains (e.g. SHRIMP or LAICPMS) have grown enormously in the last 5-10 years. The aim of such studies is to use the age of the detrital grains as a proxy for the age of the protosource where the mineral formed and thus construct a provenance for the sedimentary unit being examined. The application of provenance studies to tectonic issues involving continental reconstruction has been increasing. Recent Rodinian research has started in earnest to test, constrain and refme continental reconfigurations suggested by paleomagnetic and stratigraphic studies. One method applied to these projects is provenance analysis by microanalytical detrital zircon geochronology. This presentation is intended as a review of the method and a critique of potential applications to continental reconstruction.

APPLYING PROVENANCE ANALYSIS

Detrital zircon geochronology extracts zircon from a sedimentary unit and analyses the U-Pb isotopes to calculate an age. A relatively large number of analyses are required to ensure the statistical adequacy of the sample. This number (typically 59-60) is based on the calculation in Dodson et al. (1988) to ensure that there is only a 5% chance of missing any component that comprises more than 1 in 20 in the total population. Provenance interpretations are made on the presence (or absence) and the relative proportions of particular age modes seen in the zircon age distribution. The youngest age often comes under intense scrutiny as being the proxy for the maximum age of deposition - often in situations where other evidence about the age of a sedimentary rock is absent (e.g. Sircombe et al, 2001; Nelson, 2001). The application of detrital zircon geochronology provenance analysis in tectonic studies and continental reconstruction takes three forms. Reconnaissance: In situations where the geological history of a sedimentary unit is poorly constrained or

EXPECTING THE UNEXPECTED

Detrital zircon geochronology is prone to unexpected results and challenged assumptions. Examples include the pan-continental scale transportation systems suggested by Grenvillean aged zircons in Neoproterozoic sediments over 3000 km from the Grenville orogen (Rainbird et al, 1997) and the distinct lack of assumed Archean-aged zircons in Western Australian beach sands despite

101

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

reflected in the age spectrum seen in subsequent detrital zircons. Ideally detrital zircon geochronology should be applied in conjunction with other isotopic and geochemical methods (e.g. Sm-Nd) to ensure extra dimensions to the provenance interpretation.

close proximity to the Yilgam craton (Sircombe and Freeman, 1999). The last two "unexpected" examples illustrate an important issue in provenance analysis: sedimentary pathway is more important than proximity. It is very easy to fall into a temporal-centric trap where relatively local units or terranes exposed today dominate interpretation of past deposition. It is clear that mechanisms for transporting sediment thousands of kilometres without significant contamination from interposed terranes occur. Subsequent rifting and dispersal can orphan such sedimentary basins from a direct link to their source/s and erosion destroys correlative proximal successions. Thus it is important that a provenance study casts a wide net in the search for potential protosources. However, any proposed protosource must be supported by a reasonable and demonstrable connection to the area of deposition. Such provenance analysis attempting to link crustal blocks requires carefril consideration of the tectonic control of transportation systems. For instance, in the Perth Basin it appears that although volumetrically smaller Proterozoic sources dominated over Archean sources during the deposition of Mesozoic sediments because of their preferential uplift during rifting (Sircombe & Freeman, 1999). The pan-continental systems proposed by Rainbird et al. (1997) require large, stable cratonic areas - without vegetation that allow the dispersal of sediment after the foreland basin associated with the source orogen is frill. In other cases the foreland basin may act as a drain transporting sediment long distances (for instance the proposed Queensland protosource for the Torlesse terranes in New Zealand; Adams et al., 1998).

ARTIFICIAL BIASING

More insidious biasing occurs with sample selection at all levels from outcrop to hand-picking. Recent work with turbidites and fluvial units in Washington and British Columbia has demonstrated that individual fluvial sandstones in a succession although closely related in space and time - may have significantly different detrital zircon provenances (DeGraaff-Surpless et al., 2000). Conversely the related turbidites had generally consistent age distributions and reflected all components seen individually in the coarser sediments. Unfortunately such division between provenance localization and mixing is difficult to define apriori. Some finer sediments can reflect localized provenance whereas coarser sediments have a well-mixed provenance. For instance beach sands in eastern Australia represent both localized and well-mixed provenances even though there may not be any significant lithostratigraphic difference between samples (Sircombe, 1999). These cases stress the importance of a systematic and wideranging approach to sampling for detrital zircon geochronology for testing continental reconstruction. A simple "smash-and-grab" approach to sampling at a reconnaissance level will not ensure adequate provenance representation for testing correlation between crustal blocks. Once within the lab sampling biasing can be compounded by sample processing. Many of the methods used for sample preparation are so routine as to be almost dogmatic. Some methods routinely used for preparing igneous samples where a single age is expected and the goal is high precision are inappropriate for sedimentary samples. As discussed above sedimentary samples will often have a diverse range of age modes and this will be reflected in the nature of the zircon grains. It is vital that zircon grains are randomly chosen to ensure that the statistical adequacy of the sample as specified by Dodson et al. (1988). Conventional methods, particularly those involved in ID-TIMS, aim to select the "besf looking, low-U zircon in order to ensure concordant and thus more "accurate" results. In particular the Frantz magnetic barrier separator is routinely used to extract the least-magnetic fraction. Recent work with Archean metasedimentary rocks from the Slave craton in Canada has demonstrated that such rigourous selection runs the risk of inducing a bias in the sample (Sircombe and Stem, 2000; in review). Although in this case the principal age modes are

NATURAL BIASING

Like all methods, detrital zircon geochronology has a set of assumptions and limitations. There are also some subtleties in sample selection that may be easily overlooked although they can render data unrepresentative. Sedimentary processes at weathering, erosion, transportation and deposition cause biases that mean the sediment is not, and can never be, exactly representative of the source material it was derived from (Morton and Hallsworth, 1999). Some minerals are destroyed more easily in weathering and transportation and the process of deposition, particularly in placer deposits, can significantly alter the proportions of various heavy minerals. There is also the consideration that such heavy minerals, although key indicators of some protosources, may not be entirely representative. For instance, zircon is not an abundant constituent of mafic volcanism, thus although a mafic event may be geologically and volumetrically significant this may not be accurately

102

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

present in all magnetic fractions, it demonstrates that this cannot be assumed a priori and it certainly precludes any further provenance interpretation based on relative proportions of the age modes. Microanalysis allows a broader - and presumably more representative - range of zircon to be selected for analysis because although a grain may be high-U and "grungy" in appearance the probe-spot provide means to target optimal material and obtain some information from that grain. A striking example of the difficulties faced by conventional zircon analysis in ensuring a random and valid sample selection is given in a SHRIMP microanalytical study by Hallsworth et al. (2000). Detrital zircons from Rough Rock in the Carboniferous Pennine Basin had been previously examined by conventional means. Care had been taken to follow standard procedures and select grains according to physical attributes such as colour and morphology. The result found Archean and Caledonian (-400-450 Ma) aged grains only. The subsequent SHRIMP study found that nearly 50% of the detrital zircon were actually mid-Proterozoic in age and these ages had been completely missed by the conventional selection methods. In short, conventional analysis for detrital zircon geochronology is problematic and microanalysis is recommended along with sample selection processes to ensure a broadly representative sample.

Although a more technically correct depiction of the data, they can be difficult to interpret because it is the area beneath the curve rather than the height of the peaks that represents proportions. A compromise solution of using both probability density distributions and histograms is recommended. Finally, the status of single outlying ages in large sets of data requires careful consideration. Although statistically speaking such ages are of little consequence they can be of great geological significance (e.g. Hadean aged grains). The youngest age also takes on great value as the maximum age of deposition, but reliance on such a singular age should be weighted by supporting evidence such as whether the young grain is part of prominent age mode (cf Sircombe et al., 2001). If other data are lacking, there is also great temptation to have the maximum age of deposition become the proxy for the actual age of deposition. While this may be the case in some particular circumstances (e.g. Nelson, 2001) it can never be assumed, even in cases with wide provenance ranges. A simple case from eastern Australia illustrates the dangers of over-interpreting the maximum age of deposition (Sircombe, 1999). Here the age of the youngest detrital zircon from modern beach sand is ~ 100 Ma and often as high as -300 Ma.

INTERPRETATIONS

Detrital zircon geochronology for provenance analysis is a powerfiil tool in testing and refining continental reconstructions such as the various Rodinia configurations. However, like all methods there are assumptions and limitations that must be accounted for to ensure maximum benefit from analysis. • Is microanalytical detrital zircon geochronology justified? In many cases involving testing correlation for continental reconstructions microanalysis may be the only viable approach. However, consideration should also be given to conventional provenance analysis tools such as modal counting, heavy mineral analysis and wholerock geochemistry. At the very least these methods will provide extra dimensions to the provenance interpretation. • All sedimentary samples are naturally biased. • A clear aim of a provenance analysis project is required. Is it just reconnaissance, testing correlation or examining basin evolution? These aims will determine how systematically sampling needs to be done and how thoroughly knowledge of the lithostratigraphy is required (e.g. paleodirections, sequence stratigraphy). Reconnaissance level studies examining only a couple of samples in a poorly constrained succession may be inadequate for testing inter-craton correlation required by continental reconstruction.

RECOMMENDATIONS

Provenance interpretations are often a speculative exercise limited only by the researcher's Imowledge of potential protosources and pathways. Although there are methods for quantifying and objectively comparing large sets of data (Sircombe, 2000a) many interpretations rely on simple eye-ball comparison of histograms and probability destiny distributions (also known as "summed gaussian plots"). A lot of work remains to develop methods for easily and objectively comparing age distributions. Because conventional concordia plots usually become cluttered and unreadable as the number of plotted points increases, univariate diagrams are often chosen (Sircombe, 2000b). Unfortunately, there is a lot of scope for misuse (accidental or otherwise) in the presentation of data, and researchers did to be aware of these limitations. Histograms are particularly problematic because they discard the inherent error in the age measurement. Because bin-width is not fixed or standardized (recent detrital zircon literature has binwidths ranging from 5 to 100 Ma) there is the potential for altering the relative emphasis of particular modes during interpretation. The alternative is the "summed gaussian" or probability density distribution where the gaussian distribution represented by each age and its error is summed.

103

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

DeCelles, P.G., Gehrels, G.E., Quade, J., LaReau, B. & Spurlin, M., 2000. Tectonic implications of U-Pb zircon ages of Himalayan Orogenic Belt in Nepal. Science, 288: 497-499. DeGraaff-Surpless, K., McWilliams, M.O., Wooden, J.L. & Ireland T.R., 2000. Limitations of detrital zircon data for provenance analysis; an example from the Methow Basin, Washington and British Columbia. Abstracts with Programs Geological Society of America. 32 (6): 9. Dodson, M.H., Compston, W., Williams, LS. & Wilson, J.F., 1988. A search for ancient detrital zircons in Zimbabwean sediments. Journal of the Geological Society of London, 145: 977-983. Hallsworth, C.R., Morton, A.C., Claoue-Long J. & Fanning C.M., 2000. Carboniferous sand provenance in the Pennine Basin, UK; constraints from heavy mineral and detrital zircon age data. Sedimentary Geology, 137: 147-185. Ireland, T.R., Flottmann, T., Fanning, C.M. Gibson, G.M. & Preiss, W.V., 1998. Development of the early Paleozoic Pacific margin of Gondwana from detrital-zircon ages across the Delamerian Orogen. Geology, 26: 243-246. Machado, N., Schrank, A., Noce C.M. & Gauthier, G., 1996. Ages of detrital zircon from Archean-Paleoproterozoic sequences; implications for greenstone belt setting and evolution of a Transamazonian foreland basin in Quadrilatero Ferrifero, Southeast Brazil. Earth and Planetary Science Letters. 141: 259-276. Morton, A.C. & Hallsworth, C.R., 1999. Processes controlling the composition of heavy mineral assemblages in sandstones. Sedimentary Geology, 124: 3-30. Nelson, D.R., 2001. An assessment of the determination of depositional ages for precambrian clastic sedimentary rocks by U-Pb dating of detrital zircons. Sedimentary Geology, 141-142:37-60. Rainbird, R.H., McNicoll, V.J., Theriault, R.J., Heaman, L.M., Abbott, J.G., Long, D.G.F. & Thorkelson, D.J., 1997. Pancontinental river system draining Grenville Orogen recorded by U-Pb and Sm-Nd geochronology of Neoproterozoic quartzarenites and mudrocks, northwestern Canada. Journal ofGeology, 105: 1-17. Rainbird R.H. Stem R.A„ Khudoley A.K., Kropachev A.P., Heaman L.M.; & Sukhorukov V.L, 1998. U-Pb geochronology of Riphean sandstone and gabbro from Southeast Siberia and its bearing on the Laurentia-Siberia connection. Earth Planetary Science Letters. 164:409-420. Ross G.M., Parrish R.R. & Winston D., 1992. Provenance and UPb geochronology of the Mesoproterozoic Belt Supergroup (northwestern United States); implications for age of deposition and pre-Panthalassa plate reconstructions. Earth and Planetary Science Letters. 113: 57-76. Sircombe, K.N., 1999. Tracing provenance through the isotope ages of littoral and sedimentary detrital zircon, eastern Australia. Sedimentary Geology, 124:47-67. Sircombe, K.N., 2000a. Quantitative comparison of large sets of geochronological data using multivariate analysis: a provenance study example from Australia. Geochimica et Cosmochimica Acta, 64: 1593-1616. Sircombe K.N., 2000b. The utility and limitations of binned frequency histograms and probability density distributions for displaying absolute age data. Radiogenic Age and Isotopic Studies, Geological Survey of Canada, Current Research 2000-F2, 11 p.(http://www.nrcan.gc.ca/gsc/bookstore) Sircombe, K.N. & Freeman, M.J., 1999. Provenance of detrital zircons on the Western Australia coastline; implications for the geologic history of the Perth Basin and denudation of the Yilgam Craton. Geology, 27: 879-882. Sircombe, K.N. & Stem, R.A. 2000. Potential provenance biasing of detrital zircon geochronology due to magnetic susceptibility during sample preparation. Geological Society of America Abstracts with Programs, 32 (7). Sircombe, K.N., Bleeker, W. & Stem, R.A., 2001. Detrital zircon geochronology and grain-size analysis of a - 2 8 0 0 Ma Mesoarchean proto-cratonic cover succession, Slave Province, Canada. Earth and Planetary Science Letters, 189: 207-220.

• Careful consideration should go into the selection of samples. Larger and systematic studies have a better chance of detecting spatial and temporal variations to ensure accurate representation for correlation. Turbiditic samples have bettermixed and broader range of age modes than fluvial samples. The lithostratigraphic parameters of a samples united also need to be well constrained to aid later interpretation. • Detrital zircon may not accurately represent all the protosources that have contributed to a sediment. Complementary geochemical and isotopic methods should be pursued to provide further dimensions to provenance interpretations. • Randomness of selection and relatively large number of analyses are vital. Both are far easier to achieve using microanalytical methods. Frantz magnetic barrier separation should be avoided or the non-magnetic limit should be set reasonably high such as 1.8 amp and 10 degree side-slope. Handpicking should minimise handling of individual grains and subdivide material rather than select on the basis of colour and morphology. • Provenance interpretations must account for sedimentary pathway having greater importance than present proximity. Interpretations should look widely for potential protosources. Conversely, although long-distance transportation is clearly demonstrated, all interpretations relying on such pathways need to be able to demonstrate a reasonable link between source and basin. • Interpretations based on visual comparison are subjective. Limitations inherent in particular diagram formats need to be acknowledged. Over interpretation of outlying single-grains should be avoided. A clear differentiation between maximum possible age of deposition and actual age of deposition should be maintained. References Adams, C.J., Barley, M.E., Fletcher LR. & Pickard A.L., 1998. Evidence from U-Pb zircon and (super 40) Ar/ (super 39) Ar muscovite detrital mineral ages in metasandstones for movement of the Torlesse suspect terrane around the eastern margin of Gondwanaland. Terra Nova, 10: 183-189. Blewett, R.S., Black, L.P., Sun, S.S., Knutson, J., Hutton, L.J. & Bain, J.H.C., 1998. U-Pb zircon and Sm-Nd geochronology of the Mesoproterozoic of North Queensland; implications for a Rodinian coimection with the Belt Supergroup of North America. Precambrian Research. 89: 101-127. Cawood, P.A. & Nemchin, A.A., 2000. Provenance record of a rift basin; U/ Pb ages of detrital zircons from the Perth Basin, Western Australia. Sedimentary Geology, 134: 209-234. Cox, R., Armstrong, R.A. & Ashwal, L.D., 1998. Sedimentology, geochronology and provenance of the Proterozoic Itremo Group, central Madagascar, and implications for preGondwana palaeoceanography. Journal of the Geological Society of London. 155: 1009-1024. Davis, D.W., Hirdes, W., Schaltagger, U. & Nunoo, E.A., 1994. U-Pb age constraints on deposition and provenance of Birmian and gold-bearing Tarkwaian sediments in Ghana, West Africa. Precambrian Research, 67: 89-107.

104

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

VOROGOVIAN FAILED RIFT - EVOLUTION OF TYPICAL SEDIMENTARY BASIN DURING OF THE RODINIA BREAK UP: LATE NEOPROTEROZOIC OF SIBERIAN CRATON, RUSSIA J.K. Sovetov Institute Geology of United Institute Geology, Geophysics and Mineralogy of Siberian Branch of Russian Academy of Sciences, 630090, Novosibirsk, Russia.

carbonate platform, 3) deep water basin and accumulation of the gravitites in ramp fan, 4) continental embackment, shallow water shelf and carbonate formation related to tides (Fig.2). The basin evolution was reconstructed with lithofacies analysis and their associations that consider as depositional systems.

Late Neoproterozoic Vorogovian series is max thickness about 5400 m and fulfil of the same name depression in north-west of the Yenisei Ridge into Siberian Craton. Vorogovian depression make up a system of failed rifts-aulacogenes have been originated during break up of the Rodinia (Sovetov, 1993; 1997; Sovetov, Blagovidov, 2000). Riftogenic character of the depression was well founded by sequences of depositional systems, bimodal paleotransport of clastic material, initial and recurrent rifting and join with ocean (Sovetov, 1997, 2000). The rifting and Vorogovian deposition are the first half of Late Neoproterozoic Baikalian geodynamic cycle and these were compared with stages of Paleoasian ocean spreading (Sovetov, 1993). Baikalian series with molasse of Ushakovskaya formation together united into Peribaikalian region, after author opinion, depositional events of the single geodynamic cycle. This cycle contain the Late Riphean transtensional regime and rifting of the Siberian craton and stage its transpression and collision in Vendian but the rifting stage considerably was less Ml displaid there than Vorogovian. Vorogovian series was bounded with regional sincollisional disconformities and riftogenic basin was deformed slighty before deposition of Vendian molasse. In this report author suggest to consider the Vorogovian depression as tectonotype of Late Neoproterozoic failed rift and Vorogovian basin as lithotypical one for Rodinia break up stages. Vorogovian aulacogene (150x50 km) together with Teisko-Chapsky one form the system of riftogenic depressions (Fig.l). These aulacogenes were developed autonomously on the different fundament namely upon Riphean Sukhopitskaya and Tungusikskaya sedimentary serieses of passive margin of the Siberian craton and Isakovsky ophiolity complex has been obducted on the craton at Late Riphean. The time of ophiolity formation around the Siberian craton and simultaneous Neoproterozoic rifting is 850-740 Ma (Vemikovsky et al, 2000), 775 Ma (Watanabe et al, 1999). Vorogovian aulacogene and basin have a four stages of development, namely: 1) graben and fluvial-deltaic deposition, 2) proto-gulf, the first great sea transgression and formation of the

Vorogovsky aulacogene

Teisko^hapsky aulaeogmcs

^ Y Southern outline of Craton • • • •Boundary of Vendian-Lower Paleozoic cover Major faults of Yenisei Ridge Fig.l Late Neoproterozoic (Late Riphean) riftogenic depressions at the Siberian Craton

At the stage 1 the graben has been originating with alluvial fans in sides and alluvial-delta plane in the central part of the rift. The cyclic terrigenous complex (about 500-1500 m) with deltaic type sedimentation offered by lower part of the Severorechenskaya formation. Lithofacies successions of prodelta, delta platform, mouth bars and fluvial channels make up coarsening upwards cycles from 10 to 20 m thick. The cyclic sedimentation was related to distributary chanels migration and basin level fluctuations. Accommodation basin was the lake with deep of some tens meters to 100 m. At Late Severorechenskaya time the sea entered the rift and prodelta black muds with organic matter was expanded. Thinning of clastics texture with time

105

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

distinguished: 1) terrigene about 330 m, 2) terrigenecarbonate about 300 m, 3) terrigene about 60 m, 4) terrigene-carbonate about 220 m. Supply the sediments was controlled by faults activity and it was determined on measuring of turbidity and massflow currents. The terrigenous sand was carried across from N to S, terrigene-carbonate one from SW to NE. Carbonate platform on paleotransport data continue to exist in south-westem part of Vorogovian aulacogene and was the main source of lime micritic and clastic material in basin. It was distinguished two stage of infilling of deep sea basin which has been beginning by sudden and swift carrying over of sand mass to deep and was finished by deposition of hybridic and carbonate turbidites. Two depositional events must account for by falling sea level and forming the I type of turbidite systems (Mutti, Normark, 1987) and should have the intercontinental importance. The recurrent rifting as a whole may to relate of start the Rodinia pieces drifting and lithosphere cooling. The stages of basin infilling may be correlate with growth of Middle Ocean Ridges and dynamics of border faults. The stage 4 is shallow marine shelf sedimentation on the continental embackment. Sukhorechenskaya formation about 2400 m thickness lay onto the Mutninskaya formation with gradual transition. It was established the zone of shelf break channels where between the zone of turbidites on the slope and zone of shallow marine terrigene-carbonate sedimentation on the shelf the preparation and accumulation sediments for massflow and turbidity currents took place. "Channel in channel" association (90-370 m) consist of massive sandstones and tidal terrigene-carbonate laminites. The deposits of single channel (1-8 m) possess often by slumps structure or they were turned into chaotic mass and breccia. The channel unloading took place after accumulation of critical mass and was initiated by seismic shocks and storms. The progradation of shelf break channels followed the progradation of subtidal limestones, sandstones and mixed deposits. The stromatolithic biostroms and bioherms are among the tidal currents deposits particularly inside of aulacogene. Stromatolithic buildups and mats produced the large part of lime mud. The sand was moved in forms of ripples, dunes or more often by quick tidal currents in shelf channels. The channel deposits have the clinoform structure with signs of turbulent currents, slumps and breaks. The faults activity in the both sides of aulacogene caused the syndepositional intrabasin small faults, periodic forming the channel net and contrary directions of their axes dip to NE and SW-W (Sovetov, Blagovidov, 1996). The modal thickness of channel cycles in south-westem slope is 4-8 m while in north-eastem slope it is 2-4 m and correlate with grains size. Periodicity of channels peak in 50-80 m should be compare with sequences and global

indicate to sea level uplift too. Clastic material was transported from the aulacogene apex and sides to along its axis. In lithofacies of ephemeral streams the gravel and sand grains are crude often as quartz breccia but the central river has carried subrounded gravel with pebbles. Directions of the currents in distributaries and mouth bars were at SW-W-NW and graben dip at W. The stage 2 it was high sea level and ingression into all Late Neoproterozoic rifts of Yenisei Ridge. This event at Late Severorechenskaya age possibly related to the onset the spreading of ocean floor and have the interregional importance. The carbonate ramp appeared in foremost external zone of Vorogovian aulacogene where it was forming rhythmic member of mud limestone tempestites (70 m) and calcarenitic microphytolithic barrier complex above (75 m); the internal zone was filling in by deposits of carbonate platform with patch dolomite stromatolithic bioherms, microphytolithic bars and interbiohermal lagunas (70 m). Transgressive carbonate complex link with black shale delta by gradual transition and contain angular gravel and sand grains of quartz. The external microphytolithic barrier consist of large bars (8-10 m) with foresets dip to S. The stage 3 namely of recurrent rifting and deep water basin manifested itself by deposits of Mutninskaya formation about 900 m thickness. This stage was related to sudden activation of border faults firstly of north-eastern than south-westem ones. The ramp deep water fan was appeared. Three peculiarities are distinguished by this turbiditic basin namely the broad expanding of sandy mass-flow deposits - megaturbidites with layers to 15 m thickness which similar with megaturbidites of Jurassic rift of North Sea (Shanmugam et al, 1995); more complicated to the known models structure of terrigenous and hybridic terrigene-carbonate turbidites; alternation the terrigene, terrigenecarbonate and carbonate turbidities associations. Sedimentation of different on composition members was the symbol of activity of the two faults zones was bordering the Vorogovian rift. The central part of the basin was filling by distal identical but thin layered turbidites (450 m). Massive sandy megaturbidites with indistinct grading have not often any structures. Hybridic turbidites in model succession contain 11 structural intervals: 1) gravelsandy graded, 2) sandy thick laminated, 3) sandy thin laminated, 4) sandy with antidune structure, 5) sandy large cross-bedded, 6) sandy small crossbedded, 7) sandy-micritic with rolls, 8) sandymicritic graded, 9) silty-micritic, 10) micritic massive, 11) micritic laminated. The carbonate turbidites do not contain the lower structural intervals of model succession. In Mutninsky deep sea fan the four stratigraphical turbidity associations was

106

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Sovetov J.K., Blagovidov V.V., Chigvintseva L.A. Baikalian cycle of geodynamic development of the south-western margin of Siberian Platform (Project N 95-05-08795). In: Russian Fund of Fundamental Research in Siberian Region (Earth Crust and Mantle). T. I, lEC RAS, Irkutsk, 1995, p.7677 Sovetov J.K., Blagovidov V.V., Shelf sedimentation at late stage of Vorogovian depression development (Yenisei Ridge). Geology and Geophysics, 1996, T.37, N 4, p.45-51 Sovetov J.K., Blagovidov V.V. Neoproterozoic riftogenic basins in the south-west of Siberian Craton and break-up of Rodinia. In: 31st IGC, Abstracts vol., General Symposia. 9.4. Precambrian Supercontinents, Rio-de-Janeiro, Brazil, 2000. Vail P.R., Audemard F., Bowman S.A., Eisner P.N., Perez-Cruz C. The Stratigraphic signatures of tectonics, eustasy and sedimentology - an overview. In: Cycles and Events in Stratigraphy. Springer-Verlag, Berlin Heidelberg New York: 1991, p. 617-659. Vemikovsky V.A., Vemikovskaya A.E., Chemykh A.I. In: Neoproterozoic orogenic belts of the western margin Siberian Craton: petrology and tectonic evolution. In: 31st IGC, Abstracts vol.. General Symposia. 9.3. Tectonic Evolution of Proterozoic Orogenic Belts, Rio-de-Janeiro, Brazil, 2000 Watanabe T, Postnikov A.A., Ota T., Maehara K., Agashev A., Morita T., Roser B.T. Kuvai Group volcanic rocks (Riphean) in the Beret area, Sayan Mountains, SW Siberia, A study of the Paleo-Asian Ocean. In: Special Reports on the Regional Studies of North-East Eurasia and Pacific in Hokkaido University, 1999, p.103-112.

eustatic sea level fluctuations (Vail et al, 1991). The more deep central depression where it was calklutite area was remained at the stage 4 of evolution of Vorogovian aulacogene. The stage 4 at a whole correspond to passive continental margin of mature ocean.

References

Mutti E., Normark W.R. Comparing examples of modem and ancient turbidite systems. In: Marine clastic sedimentology. Graham and Trotman, 1987, P. 1-38. Shanmugam G., Bloch R.B., Mitchell S.M., Beamish G.W.J., Hodgkinson R.J., Damuth J.E., Straume T., Syvertsen S.E., Shields K.E., Basin-floor fans in the North Sea: sequence stratigraphic models vs. sedimentary facies. Americ. Petrol. Geol. Bull., 1995, 79, p. 477-512. Sovetov J.K. Sedimentologic sequences in Late Riphean aulacogenes at south-western margin of the Siberian Platform related to sea-floor spreading. In: Report N 4 of the IGCP Project 283. Fourth International Symposium on Geodynamic Evolution of Paleoasian Ocean. Abstracts, UIGGM SB RAS, Novosibirsk: 1993, p. 164-167. Sovetov J.K. Late Riphean riftogenesis and Baikalian cycle of geodynamic development of the Siberian Platform. In: Riphean of Northern Eurasia. Geology. General Problems of Stratigraphy. Institute Geology and Geochemistry UrB RAS, Ekaterinburg, 1997, p.223-230 Sovetov J.K., Recurrent rifting and formation of deep water basins in Late Riphean aulacogenes in Yenisei Ridge. In: Sedimentary basins: Appropriateness of Structure and Evolution, Miragenesis. Institute Geology and Geochemistry UrB RAS, Ekaterinburg, 2000, p. 118-119

The research was financed by Russian Fund of Fundamental Research, project N 00-05-65447 and Universities of Russia Fund, project N ZN-328-98.

stromatolithic bioherms

channels Stage 4 Stage 3 Stage 2 microph bars and planed Stage 1 delta deposlte

flu^lde Fig. 2. Neoproterozoic of Yenisei Ridge. Vorogovian series. Reconsruction of Vorogovian riftogenic basin at stage 4, Sukhorechensky age. Shelf sedimentation on continental embakment with tidal influenced terrigene and carbonate deposits. It is displaied too the evolution of sedimentation fromfluvialand delta plane (Early Severorechensky age, stage 1) to regional trangression and carbonate platform (Late Severorechensky age, stage 2) to deep water basin (Mutninsky age, stage 3) and shallow-water shelf with strong tides influence (Sukhorechensky age, stage 4).

107

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

THE PARADOX OF PROTEROZOIC GLACIATION AT SEA LEVEL AND STRONG SEASONALITY NEAR THE PALAEOEQUATOR: EVIDENCE AND IMPLICATIONS G.E. Williams Department of Geology and Geophysics, Adelaide University, SA 5005, Australia.

PROTEROZOIC GLACIAL CONTROVERSY

Neoproterozoic world ocean frozen to an average depth of > 1 km, a mean global temperature of 50°C with surface temperatures as low as -110°C [13], and an atmospheric hydrological cycle that virtually shut down for up to tens of millions of years. According to this scenario, soluble ferrous iron became concentrated in anoxic seawater beneath the ice cover, and interaction with the atmosphere upon melting of the ice caused the precipitation of iron-formations at the top of glaciogenic deposits. Volcanic outgassing during glaciation raised atmospheric C02 to 350 times the modem level in the absence of carbon sinks, with the resulting extreme greenhouse conditions ending each snowball state and causing carbonate deposition. This scenario of a virtually frozen-over world conflicts with much geological evidence: (1) Numerous geologists have emphasised the need for a vigorous and long-lived hydrological cycle and unfrozen seas during Proterozoic glaciations [14-16]. Continuous deposition of Marinoan rippled tidal deposits shows that the sea remained unfrozen for many years [8,15]. (2) Geological and geochemical data indicate that Neoproterozoic iron-formations in several glaciogenic successions of Sturtian-Rapitan age are hydrothermal deposits that formed with rifting under extensional regimes where fumarolic activity produced metal-charged brines [15-17]. Significantly, deposition of those successions is coeval with Rodinian breakup [18,19]. Moreover, the iron-formations usually occur within [16,17] or near the base [20] of glaciogenic successions, contrary to the snowball Earth scenario. (3) Carbon isotope values for 'cap' and other carbonates are variable and inconsistent with the snowball Earth hypothesis [21,22]. Cap carbonates may mark the destabilisation of gas hydrate in permafrost upon flooding of exposed continental shelves following rapid post-glacial warming [23]. Upwelling of oceanic waters in low latitudes, bringing cold waters onto continental shelves where carbonates were precipitated in warm shallow seas and transported basinward, also may have occurred. (4) Claims that a snowball Earth would lack carbon sinks are unjustified. Evidence for unfrozen seas [8,14-16] permits carbonate deposition, which

Proterozoic glaciogenic deposits occur on most continents and are like those of the Phanerozoic, containing thick glaciomarine tillites, laminated and varve-like argillites with dropstones, glaciofluvial outwash recording abundant meltwaters, striated pavements, and permafrost regoliths. However, three features of Proterozoic glaciogenic successions are the basis for vigorous debate concerning the nature of the Proterozoic global environment: (1) Early suggestions that Neoproterozoic glaciomarine deposition occurred in low palaeolatitudes [1] were confirmed by work in the CSIRO Rock Magnetism Laboratory at North Ryde, Sydney, which showed that late Neoproterozoic (Marinoan) glaciation at '-^600 Ma in South Australia [2,3] and Palaeoproterozoic (Huronian) glaciation at -2300 Ma in Canada [4] occurred near the palaeoequator. The -750-Ma Rapitan glaciation in Canada also occurred in low palaeolatitudes [5]. A low palaeolatitude for Palaeoproterozoic glaciation in South Africa was inferred from data for overlying volcanics [6]. (2) Several Neoproterozoic glaciogenic deposits are associated with ironstones or iron-formations. (3) Many Neoproterozoic glaciogenic successions are capped by a post-glacial carbonate unit. Combined data [2,7] give a palaeolatitude of 7.9° ± 3° for Marinoan glaciation. Positive fold tests executed on soft-sediment cuspate folds in Marinoan ebb-tidal deposits [2,3] confirm the early timing of magnetisation. The folds display lee-side avalanche deposits and rill marks, are draped by symmetrical and interference ripples, and have crests that are scoured and locally truncated by overlying beds showing climbing ripples [8]. The folds indicate gravity slides on a tidal delta that were caused by instability generated by storm waves.

A SNOWBALL EARTH?

Scenarios for Proterozoic glaciation include global glaciation [1], a non-dipole geomagnetic field, and a large obliquity of the ecliptic (tilt of spin axis) [9,10]. Recently, the global glaciation scenario, now dubbed 'snowball Earth' [11,12], has attracted attention with its dramatic claims of a

108

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

wide and several metres deep that form in the upper part of permafrost with rapid drops of temperature during repeated severe winters. The permafrost may fracture with sharp explosive-like noises [30]. Ice wedges occur in relatively humid periglacial areas where water freezes in the cracks, and sand wedges mark drier periglacial areas where the cracks are filled by drifting sand. Summer warming causes expansion of the permafrost and characteristic upturning of material next to wedges. Measurements across sand wedges in Antarctica over two decades indicated mean growth rates of up to 1 mm/year [32], and estimated ages of ice wedges in Alaska based on measured growth rates of 1-3 mm/year were verified by radiocarbon dating [32,33]. These observations confirm that periglacial wedges are actively forming under strongly seasonal climates. Fossil periglacial sand wedges are reliable indicators of past climate, including seasonal temperature range, because they formed through mechanical processes and so their interpretation avoids uncertainties in the nature of the former atmosphere and hydrosphere and later diagenetic alteration. In contrast, such uncertainties cause problems for carbonate chemostratigraphy, which underpins the snowball Earth hypothesis. Varve-like argillites with dropstones and till pellets occur with other glaciogenic deposits in the Huronian Gowganda Formation in Ontario [15,16]. Most investigators regard these argillites as varves, suggesting a seasonal climate near the palaeoequator also during Palaeoproterozoic glaciation. Importantly, climate modelling shows there would be little seasonal variation with global refrigeration [34], hence marked seasonal changes of temperature could not occur near the equator, nor probably at any latitude, for a snowball Earth. Other explanations of Proterozoic glaciation in low palaeomagnetic latitudes that also may account for the climatic paradox must be considered.

also could have occurred where the ice was thin and patchy — after all, carbonates are forming below a permanent ice cover in saline lakes in Antarctica [24], Nor were all continental areas ice-covered [25], so chemical weathering is not precluded. Moreover, CO2 sublimes at -78.5°C at atmospheric pressure and hence it would be precipitated as the solid at high latitudes (P. Gammon, pers. comm., 2000). The Earth would not emerge from a snowball state. (5) Marked seasonal changes of temperature in low palaeolatitudes during Proterozoic glaciations is incompatible with a frozen world (see below). To avoid refutation, snowballers have modified their 'theory' — there is now a 'near-snowball' Earth scenario [26] — and introduced auxiliary and ad hoc hypotheses, some of which are untestable or fanciftil (see below). Such a ploy was condemned by Karl Popper, who stated that 'it rescues the theory from recitation only at the price of destroying, or at least lowering, its scientific status' [27]. The snowball Earth scenario cannot adequately explain the nature of Proterozoic glaciations, and we must look elsewhere for the resolution of the Proterozoic glacial controversy.

THE SEASONALITY PARADOX

A climatic paradox — that of a strongly seasonal climate near the palaeoequator — cannot be ignored and provides a key to resolving the controversy. Spectacular Marinoan periglacial sand wedges occur on the Stuart Shelf in South Australia, with wedges 3+ m deep marking polygons up to 30 m across and occurring with other periglacial structures in a fossil permafrost regolith of quartzite breccia and in conformably overlying periglacial aeolian sandstone [25,28]. The quartzose parent material would not contract with desiccation and the Marinoan age of the wedges is well estabhshed [20]. Hoffman [29] has suggested, in defence of the snowball Earth scenario, that glacial surge cycles cause the temperature changes required to produce the Marinoan sand wedges and comparable sand wedges in Antarctica. This idea is refuted by more than a century of observation and research on periglacial geomorphology and processes [30] and the lack of evidence for any Marinoan glaciation of the Stuart Shelf [20], let alone the thousands of glacial advances and retreats the idea implies. Sand wedges, ice wedges and composite sand-ice wedges occur in Antarctica and ice wedges are widespread in the Arctic. Wedges show vertical lamination and in plan they define polygons ~ 10-30 m across. Such wedges are confined to periglacial regions that have a strongly seasonal climate, with a mean monthly temperature range of about 40°C from midwinter to midsummer [31]. Wedges are best developed in areas that have been free of glaciers for up to thousands of years. It is widely agreed that wedges develop from thermal contraction cracks ~l-5 mm

A NON-DIPOLE GEOMAGNETIC FIELD?

A Proterozoic non-dipole geomagnetic field may invalidate the assumed concordance of palaeomagnetic and true latitudes based on the geocentric axial dipole hypothesis. Pleistocene periglacial sand wedges occur in North America as far south as 42.5° latitude [35] and Marinoan sand wedges in South Australia formed at a palaeolatitude of 7.9° ± 3°, so a discrepancy of -30-35° between palaeomagnetic and true latitudes may resolve the paradox of a strongly seasonal periglacial climate near the Marinoan palaeoequator. Modelling has shown that extremely large axial non-dipole components of the geomagnetic field are required to produce discrepancies of the order of 30° between true and palaeomagnetic latitudes (P.W. Schmidt, pers. comm., 2001). For instance, an axial quadrupole component of 50% produces a 20° shift in true

109

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

equatorial values. However, there seems to be more evidence for axial octupole fields in the Palaeozoic and Proterozoic [36,37]. Axial octupole fields have no effect on equatorial latitudes but a 50% octupole component can make true mid-latitudes appear to be equatorial. Such extreme non-dipole components have not been identified in the palaeomagnetic record, however, either by gross statistical analysis of Proterozoic data [37] or in data for the Mackenzie dyke swarm in Canada [38]. Those studies suggest an octupole component no greater than 30%.

(2) The amplitude of the global seasonal cycle would be increased. High latitudes would endure greatly contrasting seasons, and large seasonal changes of temperature would reach low latitudes. The monotonic temperature gradient directed from the summer to the winter pole [40] would cause atmospheric circulation across the equator around solstices, when frigid air from the anticyclonic province in the winter hemisphere would flow toward the deep thermal depression in the summer hemisphere. Around equinoxes the global climate would display a normal day-night cycle. (3) The directions of zonal surface winds such as the tropical easterlies and mid-latitude westerlies would be reversed for 8 > 54° as the circulation in 'Hadley cells' reversed direction [40]. (4) Reduction of the equator-to-pole temperature gradient would weaken climatic zonation. The stability of latitude-dependent climates thus would be lessened, and any Milankovitch-band fluctuations in insolation due to orbital variations could cause large or abrupt changes of climate over wide areas that might be recorded by the stratigraphic proximity of cold- and warm-climate indicators. Important features of Proterozoic glaciogenic successions are consistent with a large obliquity: (1) deposition of glaciomarine successions preferentially in low palaeolatitudes; (2) the evidence for marked seasonal changes of temperature near the palaeoequator; (3) the presence of palaeo-northwesterly winds relative to contemporary low palaeolatitudes moving toward or across the palaeoequator during Marinoan glaciation in South Australia [25]; (4) the large changes of temperature on a 103-year time scale suggested by the observed stratigraphic proximity of cold- and warm-climate indicators, the formation of several generations of Marinoan periglacial sand-wedge structures [15,25], and the numerous advances and meltings of Neoproterozoic ice sheets [15].

A PROTEROZOIC LARGE OBLIQUITY?

A large (not 'high') obliquity of the ecliptic (e) would produce a strange global climate [9,10]: (1) The ratio of solar radiation received annually at either pole to that received at the equator would be increased (Fig. 1). For c = 54° all latitudes would receive equal radiation annually and the climatic zones would disappear. For 8 > 54°, moderate to equatorial latitudes (< 40°) would receive less radiation annually than high latitudes. If an Earth with c > 54° were to enter a glacial interval through some independent cause — a large obliquity per se is not a cause of glaciation — low latitudes would be glaciated preferentially. In high latitudes the cold, arid winter atmosphere would allow only limited snowfall which would melt entirely during the very hot summer. Snow and ice could, however, form in low latitudes [39], which would experience the additional cooling effect of frigid winds during each solstice and no extreme summer temperatures. The increased albedo resulting from a snow cover in low latitudes would allow the accumulation of permanent ice. Glaciations could be diachronous if continents moved through low latitudes.

MECHANISMS FOR OBLIQUITY CHANGE

The hypothesis of a large obliquity has come a long way since I formed the idea 30 years ago as an explanation of Proterozoic low-palaeolatitude glaciation and accompanying strongly seasonal climate. At that time there were no known mechanisms either for inducing a large Proterozoic obliquity or for reducing such an obliquity to a value < 54° prior to Late Ordovician polar glaciation. However, the last three decades have seen major developments in our understanding of the Earth's early history and dynamical evolution: (1) The 'single giant impact' hypothesis for the origin of the Moon has been developed [41]. This widely accepted mechanism for lunar origin requires a target-body/impactor mass ratio of -2.0 [42],

2 3 4 Relative m e a n annual insolation

Figure 1. Latitudinal variation of relative mean annual insolation of a planet for various values of obliquity (s). From [10].

110

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

that should be explored and tested but not pushed with evangelical zeal. Both hypotheses assume the Proterozoic geomagnetic field approximated a geocentric axial dipole, and while available evidence does not contradict that assumption there remains much to be done in this area. Further palaeomagnetic, geochronological and sedimentological studies of Proterozoic glaciogenic successions are of course required. Demonstration of a reliable polar palaeolatitude for a pre-Ediacarian glaciomarine succession may rule out the largeobliquity hypothesis but would not confirm a snowball Earth because the paradox of a strongly seasonal climate near the palaeoequator would remain unresolved. Nor would the demonstration of synchronous glaciation in low palaeolatitudes on several continents necessarily favour a snowball Earth because synchronous glaciation is possible also with a large obliquity if glacial intervals were relatively brief or saw little plate motion. In my view, on present evidence the solution to the Proterozoic climatic paradox of glaciation near sea level in low palaeomagnetic latitudes under a strongly seasonal climate most likely will come via geomagnetism or celestial mechanics.

which would likely result in c > 70° for the primordial Earth (Fig. 2). Proponents of a snowball Earth have not challenged the concept that the early Earth had a large obliquity. This concept is supported by modelling the Archaean climate [44], and Trendall [45] implied the Archaean obliquity was not necessarily similar to the present value.

70 CO CD CD CT CD

60 - APPROACH VELOCITY • 30 km/s O 20 km/s _ • 10 km/s X 5 km/s •o

JQ o c o

EE 20 o

o CO 10 o

•

"2 9• X

•

• o

•

o • 6 • • XX

o X

O _

o o•

•

1 o

•

_ X

? -x

• X«O X X 1

0.05 0.10 Planetesimal mass / target planet mass

Figure 2. Impact-induced obliquity of Earth-like target planet for a range of approach velocities and impactor/target-planet mass ratios. Target planet has pre-impact obliquity of 0°. Each point is the most frequent value in a run of 500 impacts at specified impactor mass. From [43].

ACKNOWLEDGMENTS I thank the Australian Research Council for supporting my collaborative studies with Phil Schmidt (CSIRO, North Ryde) on Proterozoic palaeomagnetism and environments in Australia and other continents. Grant Young (University of Western Ontario) is thanked for helpful correspondence. Support from the 2000/01 IGCP Grant-in-Aid also is acknowledged.

(2) The mechanism of 'climate friction' for secular change of obliquity, which involves changes in the mass distribution of the Earth and the rate of spin-axis precession due to the waxing and waning of ice sheets [46-49], has been recognised in the past decade. This mechanism may cause either secular increase or decrease in a planet's obliquity depending on the length and severity of the ice ages and the planet's internal properties, and has been investigated as the cause of the postulated obliquitydecrease in late Neoproterozoic-early Palaeozoic time [48]. While results so far are inconclusive, the mechanism is still in the earliest stage of exploration; for example, the effects of fluctuating equatorial ice sheets are unknown. As stated by the celestial mechanicists at NASA's Goddard Space Flight Center who first recognised the mechanism [49]: 'Unlike tidal friction, a topic two centuries old, climate friction is such a new field that we can't yet determine its importance for changing our planet's tih. Only time will tell.'

References 1. W.B. Harland, Geol. Rundsch. 54, 45 (1964). 2. P.W. Schmidt, G.E. Williams & B J J . Embleton, Earth Planet. Sci. Lett. 105, 355 (1991). 3. P.W. Schmidt & G.E. Williams, Earth Planet. Sci. Lett. 134, 107(1995). 4. G.E. Williams & P.W. Schmidt, Earth Planet. Sci. Lett. 153, 157(1997). 5. J.K. Park, Can. J. Earth Sci. 34, 34 (1997). 6. D.A. Evans, N.J. Beukes & J.L. Kirschvink, Nature 386, 262 (1997). 7. L.E. Sohl, N. Christie-Blick & D.V. Kent, Geol. Soc. Am. Bull. 111,1120(1999). 8. G.E. Williams, Sed. Geol. 106, 165 (1996). 9. G.E. Williams, Geol. Mag. 112, 441 (1975). 10. G.E. Williams, Earth-Sci. Rev. 34, 1 (1993). 11. J.L. Kirschvink, in The Proterozoic Biosphere: A Multidisciplinary Study, J.W. Schopf & C. Klein, Eds. Cambridge Univ. Press, Cambridge, 51 (1992). 12. P.P. Hoffman, A.J. Kaufman, G.P. Halverson & D.P. Schrag, Science 281, 1342(1998). 13. S.K. Baum & T.J. Crowley, Geophys. Res. Lett. 28, 583

DISCUSSION AND CONCLUSIONS

(2001). 14. M.E. McMechan, Bull. Can. Petrol. Geol. 48, 246 (2000). 15. G. Williams & P. Schmidt, The Australian Geologist 117, 21

I can state categorically that I do not believe the Proterozoic Earth had a large obliquity — nor do I disbelieve it! The ideas of a Proterozoic large obliquity and a snowball Earth are just hypotheses

(2000). 16. G.M. Young, The Australian Geologist 118, 6 (2001). 17. G.M. Young, Sed. Geol. 58, 127 (1988).

Ill

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

35. W.J. Wayne, Geol. Soc. Am. Abs. Prog. 22 (7), A85 (1990). 36. R. Van der Voo & T.H. Torsvik, Earth Planet Sci. Lett. 187, 71 (2001). 37. D.V. Kent & M.A. Smethurst, Earth Planet. Sci. Lett. 160, 391 (1998). 38. P.W. Schmidt, Proc. Nordic Palaeomagnetic Symp., Aarhus Univ., Aarhus Geosci. 8, 109 (1999). 39. R.J. Oglesby & J.G. Ogg, Paleoclimates 2, 293 (1998). 40. B.G. Hunt, J. Meteorol. Soc. Japan 60, 309 (1982). 41. S.R. Taylor, Am. Scientist 75, 469 (1987). 42. S. Ida, R.M. Canup & G.R. Stewart, Nature 389, 353 (1997). 43. W.K. Hartmann & S.M. Vail, in Origin of the Moon, W.K. Hartmann, R.J. Phillips & G.J. Taylor, Eds. Lunar & Planetary Institute, Houston, 551 (1986). 44. G.S. Jenkins, J. Geophys. Res. 105, 7357 (2000). 45. A.F. Trendall, J. Geol. Soc. Aust. 19, 13 (1972). 46. B.G. Bills, Geophys. Res. Lett. 21, 177 (1994), 47. D.P. Rubincam, Paleoceanography 10, 365 (1995). 48. D.M. Williams, J.F. Kasting & L.A. Frakes, Nature 396, 453 (1998). 49. D.P. Rubincam, B.F. Chao & B.G. Bills, Skye & Telescope 95 (6), 37 (1998).

18. C.McA. Powell, W.V. Preiss, C.G. Gatehouse, B. Krapez & Z.X. Li, Tectonophys. 237, 113 (1994). 19 C.McA. Powell & S. Pisarevsky, European Un. Geosci. XI, Abs. 98 (2001). 20. J.F. Drexel, W.V. Preiss & A.J. Parker, Eds., Geol. Surv. S. Aust. Bull. 54, vol. 1 (1993). 21. T. Praveetal., European Un. Geosci. XI, Abs. 101 (2001). 22. A.C. Hill & M.R. Walter, Precambrian Res. 100, 181 (2000). 23. M.J. Kennedy, N. Christie-Blick & L.E. Sohl, Geology 29, 443 (2001). 24. M.R. Walter & J. Bauld, Precambrian Res. 21, 129 (1983). 25. G.E. Williams, Aust. J. Earth Sci. 45, 733 (1998). 26. T.J. Crowley, W.T. Hyde & W.R. Peltier, Geophys. Res. Lett. 28, 283 (2001). 27. K.R. Popper, Conjectures and Refutations, Routledge & Kegan Paul, London, 37 (1963). 28. G.E. Williams, Precambrian Res. 32, 233 (1986). 29. P.P. Hoffman, The Australian Geologist 118, 7 (2001). 30. A.L. Washburn, Geocryology: A Survey of Periglacial Processes and Environments. Wiley, New York (1980). 31. J. Karte, GeoJoumal 7, 329 (1983). 32. R.F. Black, Antarctic J. US 17, 53 (1982). 33. R.F. Black, Geol. Soc. Am. Bull. 63, 1235 (1952). 34. W.D. Sellers, Palaeogeog., Palaeoclimatol., Palaeoecol. 82, 217(1990).

112

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

AUSMEX: A NEW RODINIA RECONSTRUCTION AT 1070 Ma MT.D. Wingate, S.A. Pisarevsky and D.A.D. Evans Tectonics Special Research Centre, Department of Geology and Geophysics, The University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia.

baked sedimentary rocks produced ages between 1050 and 1075 Ma (Compston and Arriens, 1968; Gee et al, 1976; Goode and Hall, 1981). Following sill intrusion, the southern parts of the basin were compressed northwards, resulting in an arcuate region of elongate, tight to open folds known as the Edmund Fold Belt (Figure 1). Folding may have been related to 1090-1060 Ma tectonothermal events (Bruguier et al, 1999) in the adjacent Darling Mobile Belt (Figure 2c), and certainly occurred prior to intrusion of undeformed dolerite dykes of the 755 Ma Mundine Well swarm (Wingate and Giddings, 2000), that cut across all rocks of the basin.

INTRODUCTION Most reconstructions of the Rodinia supercontinent place East Gondwanaland (Australia, E. Antarctica, and India), as a coherent tectonic entity, adjacent to western Laurentia (ancestral North America). The SWEAT reconstruction places eastern Australia (± South China) against western Canada (Dalziel, 1991; Hoffman, 1991; Moores, 1991; Powell et al., 1993), whereas the AUSWUS model places eastern Australia adjacent to the western United States (Brookfield, 1993; Karlstrom et al, 1999; Burrett and Berry, 2000). Although these models are very different, each is based on matching geological and tectonic features and age provinces. The available paleomagnetic data are presently inadequate to discriminate between the SWEAT and AUSWUS alternatives. In this contribution, we summarise a U-Pb and paleomagnetic study of Mesoproterozoic dolerite sills in the western Bangemall Basin of Western Australia (Wingate et al., 2001). The results yield a precisely-dated paleopole, BBS, at 1070 ± 6 Ma, that is the most reliable of this age for Australia. The Bangemall paleopole indicates that neither the SWEAT nor AUSWUS fit is likely at 1070 Ma. The data suggest a new reconstruction, referred to as AUSMEX (Australia - Mexico), in which the Grenville orogenic belt of southernmost Laurentia extends into the Cape River - Musgrave - AlbanyFraser orogenic system of Australia.

GEOCHRONOLOGY Zircon and baddeleyite crystals were recovered from coarse-grained samples of five dolerite sills in the western Bangemall Basin and dated by SHRIMP ion microprobe. The results indicate that the sills were emplaced during two separate events. For three samples, all baddeleyite and zircon ^^^Pb/^^^Pb ratios agree to within analytical precision and yield statistically identical ages of 1071 ± 8, 1067 ± 14, and 1068 ± 22 Ma. The results are combined to yield a mean age of 1070 ± 6 Ma (95% confidence interval), which we regard as the time of crystallisation of the younger sill suite. Zircon and baddeleyite from two samples indicate an age for the older sills of 1465 ± 3 Ma.

PALEOMAGNETISM THE BANGEMALL BASIN

Samples for paleomagnetic study were collected from dolerite sills and sedimentary rocks at 25 sites distributed throughout the sedimentary succession in the western Bangemall Basin, including the five dated sites. Two main types of magnetic behaviour were observed. Most samples yielded an inconsistently-directed remanence of low thermal stability (referred to as type L), which we interpret as a chemical remanent magnetisation (CRM) carried mainly by maghemite. Only type L magnetisations were found in samples from the two older sills dated at -1465 Ma. A consistently-directed magnetisation (referred to as type A) was isolated in 79 samples from 15 sites, including the three dated at ca. 1070 Ma, and is the only remanence present at five sites. Unblocking temperatures close to 580°C show the remanence to

The ca. 1.6 to 1.0 Ga Bangemall Basin of Western Australia (Figure 1) is an intracratonic basin that contains more than 6 km of fme-grained carbonate and siliciclastic marine sedimentary rocks known as the Bangemall Supergroup (Muhling and Brakel, 1985; Martin et al, 1999). Preserved over 87,000 km^, the basin unconformably overlies deformed Paleoproterozoic sedimentary basins and metamorphic and igneous rocks of the Capricorn orogen (Tyler and Thome, 1990). Extensive quartz dolerite sills, classified geochemically as high-Ti continental tholeiites, occur throughout the Bangemall succession. They are typically >100 m thick, mainly concordant to bedding, and generally medium-grained, with locally exposed chilled margins and coarse-grained phases. Several K-Ar and Rb-Sr studies of dolerites and

113

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

200 km

INDIAN OCEAN

Neoproterozoic & Phanerozoic basins I

I Collier Group Edmund Group _

[

Bangemall Supergroup

I Paleoproterozoic basins

1 — P a l e o p r o t e r o z o i c igneous I — - I & metamorphic rocks [

I Archean to Paleoproterozoic basin

I 1 Archean graniteL — I greenstone terranes

Figure 1. Location and geological setting of the Bangemall Basin. The star shows the location of a dyke that yields primary paleomagnetic directions similar to those observed in 1070 Ma sills. from flat-lying sills in the Glenayle area (M. be single component and that relatively pure Wingate, work in progress), in the eastern magnetite is the dominant remanence carrier. Site Bangemall Basin (Figure 1), together with the lack mean directions converge after correction for of significant deformation of the Mundine Well dyke bedding tilt, and the concentration parameter, k, swarm, indicate that the Bangemall Basin has increases from 6 to 30. Corrected directions are undergone no vertical-axis rotation since 1070 Ma. NNW with moderate downwards inclination, except The mean direction, after tectonic correction (and at one site, where the direction is SSE and upward. inversion of data from site 25), is D, I = 340°, +47° Several observations indicate a primary origin (a95 = 8°, A'^ = 11 sites). The paleomagnetic pole, for the A magnetisation: BBS, calculated as the mean of site virtual 1. Low within-site dispersion is typical of primary geomagnetic poles (VGPs), lies at 34°N, 95°E. thermoremanent magnetisations (TRM) in rapidlyLate Mesoproterozoic paleopoles for Australia cooled intrusions. were obtained previously from the Stuart dykes and 2. Single-domain magnetite grains in some sill Kulgera sills in central Australia (Idnurm and samples require heating close to 580°C to unblock Giddings, 1988; Camacho et al, 1991). Although their magnetisation, and there is no evidence for a isotopic data suggest that the Stuart and Kulgera thermal event between 1070 and 755 Ma that could intrusions are roughly similar in age at -1050-1090 cause a remagnetisation. Ma (Zhao and McCulloch, 1993), their paleopoles 3. Polarity reversals between, but not within, are significantly different (Figure 2). Reliable intrusions is supportive of a primary remanence. constraints on paleohorizontal are not available for 4. Positive fold tests show that the A magnetisation the Kulgera or Stuart intrusions (no tectonic was acquired prior to folding (probably soon after corrections were applied), and both suites are 1070 Ma, certainly before 755 Ma). located in crustal blocks that were deformed and 5. Although baked sedimentary rocks in sill contacts probably re-oriented during late Neoproterozoic appear to be overprinted by the A magnetisation, no (Petermann Ranges) and/or Carboniferous (Alice stable remanence was isolated in unbaked rocks, Springs) tectonothermal events. The BBS paleopole therefore the criteria for a valid baked-contact test does not agree with the KDS or SDS poles, but is are not satisfied completely. However, an undated more reliable than either. The BBS pole is inferred NNE-trending dyke (star in Figure 1) carrying the to be primary, is dated precisely, and structural distinctive Bangemall A direction yields a complete control is well-defined in adjacent sedimentary positive baked-contact test with its host rock, the rocks. 2.45 Ga Woongarra Rhyolite, suggesting that the dyke is similar in age to the sills, and that the A magnetisation in the sills is also original. IMPLICATIONS FOR RODINIA We conclude that the A component is a primary The BBS paleopole permits a direct test of the TRM acquired during sill emplacement at 1070 Ma. SWEAT and AUSWUS reconstructions at 1070 Ma, Directions of opposite polarity at one site imply that prior to any plausible age for Rodinia's the intrusive event spanned at least one reversal of fragmentation (Hoffman, 1991). Although there is the Earth's magnetic field, and that, collectively, the no precisely dated Laurentian pole at 1070 Ma, a A magnetisations adequately average paleosecular variation. Similar paleomagnetic directions obtained

114

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Figure 2. Early to mid-Neoproterozoic reconstructions of East Gondwanaland (Australia + E. Antarctica + India) and Laurentia, according to the SWEAT (a) and AUSWUS (b) hypotheses. The Laurentian apparent polar wander (APW) path is based on four mean poles (McElhinny and McFadden, 2000) and a pole (AB) from the Abitibi dykes (Ernst and Buchan, 1993). Laurentia is situated at its correct paleolatitude and orientation at 1070 Ma, according to an interpolated pole position (see text). 'Grenville-age' mobile belts are shown in red. Australian poles include the Lakeview dolerite, lAR (Tanaka and Idnurm, 1994), Stuart dykes, SDS (Idnurm and Giddings, 1988), and Kulgera sills, KDS (Camacho et aL, 1991). A reasonable fit between the new 1070 ± 6 Ma BBS pole and the Laurentian APW path is not obtained in the SWEAT or AUSWUS models. A possible reconstruction between Australia and Laurentia at 1070 Ma (c) places the Grenville and Cape River provinces at similar paleolatitudes. M is the Mawson block of East Antarctica (Fitzsimons, 2000). Paleolongitudes are arbitrary. Rotation parameters are listed in Wingate et al (2001). position for Laurentia at 1070 Ma by assuming constant APW between the 1080 and 1040 Ma mean poles. Superimposing the BBS paleopole (within uncertainty) and this interpolated pole for Laurentia at the projection axis (Figure 2c) places the continents at their correct orientations and paleolatitudes at 1070 Ma. Australia is situated at lower paleolatitudes than is permitted by the SWEAT or AUSWUS models. In this position, the

cluster of results define two well constrained mean poles for 1080 and 1040 Ma (McElhinny and McFadden, 2000). The BBS pole is separated by --30° from the 1080 - 1040 Ma Laurentian APW path segment in the SWEAT fit and by at least 40° in the AUSWUS fit, indicating that neither SWEAT nor AUSWUS is viable at 1070 Ma (Figure 2). To explore possible reconstructions between Australia and Laurentia, we approximate a pole

115

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Dalziel, I.W.D., 1991. Pacific margins of Laurentia and East Antarctica-Australia as a conjugate rift pair: Evidence and implications for an Eocambrian supercontinent. Geology 19, 598-601. Ernst, R.E. and Buchan, K.L., 1993. Paleomagnetism of the Abitibi dike swarm, southern Superior Province, and implications for the Logan Loop. Canadian Journal of Earth Sciences 1886-1897. Fitzsimons, LC.W., 2000. Grenville-age basement provinces in East Antarctica: Evidence for three separate collisional orogens. Geology 28, 879-882. Gee, R.D., De Laeter, J.R. and Drake, J.R., 1976. Geology and geochronology of altered rhyolite from the lower part of the Bangemall Group near Tangadee, Western Australia. Western Australia Geological Survey, Annual Report 1975, 112-117. Goode, A.D.T. and Hall, W.D.M., 1981. The middle Proterozoic eastern Bangemall Basin, Western Australia. Precambrian Research U, 11-29. Hoffman, P.P., 1991. Did the breakout of Laurentia turn Gondwanaland inside-out? Science 252, 1409-1412. Idnurm, M. and Giddings, J.W., 1988. Australian Precambrian polar wander: a review. Precambrian Research 40, 61-88. Karlstrom, K.E., Harlan, S.S., Williams, M.L., McLelland, J., Geissman, J.W., Ahall, K.-L., 1999. Refining Rodinia: geologic evidence for the Australia-Western U.S. connection in the Proterozoic. GSA Today 9, 1-7. Martin, D.McB., Thome, A.M., and Copp, LA., 1999. A provisional revised stratigraphy for the Bangemall Group on the Edmund 1:250 000 sheet: Western Australia Geological Survey, Annual Review 1998 - 1999, 51 -55. McElhinny, M.W. and McFadden, P.L., 2000. Paleomagnetism Continents and Oceans. International Geophysics Series Vol. 73, Academic Press, USA. Moores, E.M., 1991. Southwest U.S. - East Antarctic (SWEAT) connection: a hypothesis. Geology 19, 425-428. Muhling, P.C. and Brakel, A.T., 1985. Geology of the Bangemall Group - the evolution of an intracratonic Proterozoic basin. Western Australia Geological Survey, Bulletin 128, 266p. Powell, C.McA., Li, Z.X., McElhinny, M.W. et al, 1993. Paleomagnetic constraints on timing of the Neoproterozoic breakup of Rodinia and the Cambrian formation of Gondwana. Geology 1\, 889-892. Rivers, T., 1997. Lithotectonic elements of the Grenville province: review and tectonic implications. Precambrian Research 86, 117-154. Tanaka, H. and Idnurm, M., 1994. Paleomagnetism of Proterozoic mafic intrusions and host rocks of the Mt. Isa Inlier, Australia. Precambrian Research 69, 241-258. Tyler, I.M. and Thome, A.M., 1990. The northem margin of the Capricom Orogen, Westem Australia - an example of an early Proterozoic collision zone. Journal of Structural Geologyll, 685-701. Wingate, M.T.D., Pisarevsky, S.A. and Evans, D.A.D., 2001. A revised Rodinia supercontinent: no SWEAT, no AUSWUS. Terra Nova, in press. Wingate, M.T.D. and Giddings, J.W., 2000. Age and paleomagnetism of the Mundine Well dyke swarm, Westem Australia: implications for an Australia - Laurentia connection at 755 Ma. Precambrian Research 100, 335-357. Zhao, J.-X. and McCulloch, M.T. 1993. Sm-Nd mineral isochron ages of Late Proterozoic dyke swarms in Australia: evidence for two distinct events of mafic magmatism and cmstal extension. Chemical Geology 109, 341-354.

Cape River Province of northeast Australia is aligned with the southwest end of the --1250-980 Ma Grenville Province of Laurentia (Rivers, 1997). The Cape River Province contains 'Grenville-age' metamorphic and magmatic rocks and may correlate to the west with the Musgrave and Albany-Fraser orogens (Blewett et al, 1998). The Grenville Province may therefore have continued through Australia, without requiring the sharp northward bend envisaged in the AUSWUS model.

CONCLUSIONS The BBS paleopole is clearly incompatible with Laurentian data in either a SWEAT or AUSWUS reconstruction. The provocative fit, AUSMEX, shown in Figure 2c requires further testing by comparing additional Proterozoic paleopoles of precisely the same age from Australia and Laurentia, with subsequent reconstructions to be elaborated using geological and other constraints. The most compelling geological arguments used to generate the SWEAT and AUSWUS hypotheses, including correlation of Mesoproterozoic orogenic belts, Paleo- and Mesoproterozoic isotopic age provinces, and Neoproterozoic rift - passive margin sedimentary successions, remain robust in our new reconstruction. The SWEAT and AUSWUS models imply that Neoproterozoic separation of AustraliaAntarctica from Laurentia led to opening of the Pacific Ocean. The results of this study, however, indicate that western Laurentia was not the conjugate margin to eastern Australia-Antarctica. The origin of the Pacific Ocean is therefore undetermined, and up to 10,000 km of late Neoproterozoic rifted margins in eastern Australia and western Laurentia need to be matched with other continental blocks within a revised Rodinia supercontinent. References Blewett R.S., Black L.P., Sun, S.s., Knutson, J., Hutton, LJ., Bain, J.H.C., 1998. U-Pb zircon and Sm-Nd geochronology of the Mesoproterozoic of North Queensland: implications for a Rodinian connection with the belt Supergroup of North America. Precambrian Research 89, 101-127. Brookfield, M.E., 1993. Neoproterozoic Laurentia-Australia fit. Geology 21, 683-686. Bruguier, O., Bosch, D., Pidgeon, R. T., Byrne, D.I., Harris, L.B., 1999. U-Pb chronology of the Northampton Complex, Western Australia - evidence for Grenvillian metamorphism, sedimentation, and deformation and geodynamic implications. Contributions to Mineralogy and Petrology 136, 258-272. Burrett, C. and Berry, R., 2000. Proterozoic Australia - Western United States (AUSWUS) fit between Laurentia and Australia. Geology 103-106. Camacho, A., Simons, B. and Schmidt, P.W., 1991. Geological and paleomagnetic significance of the Kulgera Dyke Swarm, N.T., Australia. Geophysical Journal International 107, 3745. Compston, W. and Aniens, P.A., 1968. The Precambrian geochronology of Australia. Canadian Journal of Earth Sciences 5, 561-583.

116

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

POSITIONS OF THE NORTH CHINA BLOCK IN NEOPROTEROZOIC RODINIA: A PALAEOMAGNETIC CONSTRAINT S. Zhang^ and Z X Lp ^ School of the Earth and Land Resources, China University of Geosciences, Beijing, 100083, China, Email: shzhang@cugb.edu.cn ^ Tectonics Special Research Centre, Department of Geology and Geophysics, The University of Western Australia, Crawley, WA 6009, Australia.

Meso- to Neoproterozoic successions are well developed in the NCB, consisting of siliciclastic rocks, red beds and volcanic rocks, and their structures are relatively simple. We carried out palaeomagnetic analysis on successions in western Henan and eastern Liaoning Provinces, aiming to test the relationships between the NCB, Siberia and Laurentia during the assembly and breakup of Rodinia. Based on our results and existing data, we present here palaeogeographic reconstructions for ca. 800 Ma and ca. 650 Ma respectively. We suggest that the NCB could have been close to Siberia during this time interval.

INTRODUCTION Paleogeographic positions of the North China Block (NCB) in Precambrian time, especially its connections with other paleocontinents, have long been debated. Some Chinese and Russian researchers favored a NCB-Siberia connection during the Archaean and Paleoproterozoic, and suggested that they may have broke apart during the Mesoproterozoic (e.g. C. Li et al, 1982, and references herein). Qian (1997) proposed a NCBBaltica connection based on a comparison of the two cratons' basements. However, such a connection was refuted by Zhao et al. (2001). Z.X. Li et al. (1996), based on tectonostratigrapic correlations, suggested that the NCB, Siberia and Laurentia may have been together from ca. 1.8 Ga until latest Precambrian, and they placed the present north margin of NCB against the western margin of Yenisei Ridge, Siberia. Wang et al. (1997) suggested that the NCB was adjacent to northwestern American during mid-Neoproterozoic, mainly basis on biogeographical analyses.

SELECTED NEOPROTEROZOIC PALEOMAGNETIC DATA FOR THE NCB Previous workers already noticed the problem of remagnetisation and conflicting results among some of the existing data (Lin, 1988; Z.X. Li et al., 1996). We thus applied the 7-point criteria (Van der Voo,

Table 1 Selected Neoproterozoic to Cambrian paleomagnetic poles for the NCB No.

Abbr.

Rock unit/Region/Reference

a95

Plat

Plon

Criteria

(A95)

(«N)

CE)

1 2

5 6 7 + + +

CAM3 Upper Cambrian/NCB/Huangetal., 1999

5.4

31.7

329.6

+ +

CAM2 Middle Cambrian/NCB/ Huang et al, 1999

5.5

326.7

+ +

CAMl Lower Cambrian/NCB/ Huang et al., 1999

6.5

18.5

342.0

Dongjia Fm/W Henan Province/Zhang et al., 2000

8.4

60.8

277.4

+ +

Zhangqu Fm/N Jiangsu Province/Fang et al, 1983

8.3

46.4

288.7

Niyuan Fm/Xu-Huai/Fang et al, 1983

8.8

44.0

294.3

Zhaowei Fm/Xu-Huai/Fang et al, 1983

23.0

41.2

281.7

Jiayuan Fm /Xu-Huai/ Fang et al, 1983

20.0

39.7

284.0

Average Poles ZQ, NY, ZW, JY

5.7

42.9

287.0 +

10 NF

Nanfen Fm/E Liaoning Province/Lin, 1984

9.5

16.5

301.1

+ +

11 SJT

Sanjiaotang Fm /W Henan / Zhang et al, 2000

41.3

227.0

+ + +

12 CZ

Cuizhauang Fm/ W Henan/ Zhang et al, 2000

4.2

46.6

217.8

N/(n), number of sites/Samples a95/(A95), radius of 95% confidence circle for a direction/pole. Plon/Plat, east longitude/north latitude of VGP; The 7 criteria Q value of Van der Voo (1990) has been used for quality assessment. Underlined poles only were used for reconstructions in Fig. 1.

117

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

40 N 20 N / ^ C Z

WCAM3

JwS3T

^Al

CAMl^y

Equator

20 S Bv

40 S NF Qingbaikouan

Sinian

Cambrian

800

540

Figure 1. Paleo-latitudes and orientations of the NCB implied by the selected poles as shown in Table 1.

Figure 2. Paleogeographic positions of the North China Block (NCB) in globe reconstruction: AT-East Antarctica; AU-Australia; AZ-Amazonia; BC-Baltica; CN-Congo; IN-Great India; LA- Laurentia; SBSiberia; SCB-South China Block (a) the NCB in Rodinia at ^780-800 Ma, Model 1 position for the NCB is based on polarity option-1 of pole NF in Table 1, and Model 2 based on polarity option-2. The East Gondwana-SCB connection of is based on Z.X. Li et al. (1999). Euler pole for rotating Siberia relative to Laurentia is ( I T N , 12°E, 31°counterclockwise), Other continents were rotated to Laurentia according to Torsvik et al. (1996). Position of the Laurentian pole is (9°N, 136°E) from Buchan et al. (2000). (b) Paleogeographic positions of the North China Block during Rodinia breakup at - 6 5 0 Ma, Model 1 position is based on polarity option-1 of pole DJ in Table 1, and Model 2 based on polarity option-2. Euler pole for rotating Siberia to Laurentia is (11°N, 12°E, 31° counterclockwise). Other continents were rotated to Laurentia according to Torsvik et al. (1996).

118

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

References

1990) to select the more reliable data out of the ca. 40 data sets available. The selected poles, including three Cambrian poles, are given in Table 1. The ages for the pole are assigned through integrated interpretation of available isotopic data and stratigraphic correlation (Zhang et al., 2000).

Buchan, K. L., Mertanen, S., Park, R. G., Pesonen, L. J., Elming, S. -A., Abrahamsen, N., Bylund, G., 2000. Comparing the drift of Laurentia and Baltica in the Proterozoic: the importance of key paleomagnetic poles. Tectonophysics, V.319, p.167-198. Fang, D. J., Zhu, Z. W., Guo, Y. B., 1983. Study on the paleomagnetism of Upper Precambrian in Northern Jiangsu and correlation between the Upper Precambrian strata in South and North China. Scientia Geologica Sinica, (4), p. 324-336. Hoffman, P. F., 1991. Did the breakout of Laurentia turn Gondwanaland inside out? Science, v.252, p.1409-1412 Huang, B., Yang, Z., Otofuji, Y., Zhu R., 1999. Early Paleozoic paleomagnetic poles from the western part of the North China Block and their implications, Tectonophysics, v.308, p.377-402. Li, C., Wang Q., Liu X., Tang Y., 1982. Explanatory notes to the tectonic map of Asia, Beijing: Cartographic Publishing House, p. 1-49. Li, Z. X., Zhang, L. H., Powell, C. McA., 1996. Position of the East Asian cratons in the Neoproterozoic supercontinent Rodinia , Aust. J. Earth Sci., v. 43, p.593-604. Li, Z. X., Li, X. H., Kinny, P. D., Wang J., 1999. The breakup of Rodinia: did it start with a mantle plume beneath South China? Earth and planetary science letters, v. 173, p. 171-

POSITIONS OF THE NCB IN GLOBE RECONSTRUCTIONS Figure 1 shows the paleolatitudes and orientations of the NCB implied by the selected paleopoles according to polarity option-1. If one adopt another polarity (polarity option-2), the positions of the NCB would be the mirror image of the given position on the opposing side of the equator. In Figure 2a, we plotted the possible positions of the NCB against other continents in a ca. 780-800 Ma Rodinia reconstruction. We take pole NF in Table 1 for the ~780-800Ma reconstruction of the NCB. In the polarity option-1 reconstruction, the NCB could be placed against the Greenville margin of Laurentia, or other positions at the same latitude (such as east of Amazonia). However, considering that there is no evidence suggesting a Grenville-age orogenic belt along the northern margin of the NCB, a position opposite to the Grenville Belt is regarded as unlikely. We thus prefer to use the polarity option-2 in the reconstructions, and place the NCB against the Lake Baikal margin of Siberia in a northern latitudinal position. During the break-up of Rodinia, Laurentia rotated largely clockwise (Torsvik et al., 1996). Model 1 (according to polarity option-1 of pole DJ in Table 1) (Fig. 2b) NCB was at the tropic belt of north hemisphere, but its model 2 position (according to polarity option-2 of pole DJ) still allows it to be next to Siberia. Given that the relative orientations between Laurentia-Siberia and the NCB had slightly changed from the 780-800 Ma reconstruction, the breakup between them may have already occurred by then. Qiao et al. (2000) reported stratigraphic record of paleo-earthquakes, and dykes of -720-600 Ma age, in eastern NCB. These activities could be related to the breakup events. However, more reliable data from both continents are required to further evaluate the timing of the breakup.

181.

Lin, J. L., 1988. Middle to Late Proterozoic paleomagnetic results from Jixian, Kexue Tongbao, v.33, p.207-210. Lin, J. L., 1984. The apparent polar wander paths for the North and South China Blocks, Ph.D. thesis. Univ. Of Calf. Santa Barbara, 248pp. Qian, X. L., 1997. Tectonic correlations of the Precambrian Evolution of the North China Craton with the Baltica Shield. In: Qian X. L., You Z. D., Halls H. C. (Eds), Precambrian Geology and Metamorphic Petrology. Utrcht, the Netherlands, p.43-58. Qiao, X. F., Gao, L. Z., 2000. Earthquake events in Neoproterozoic and Early Paleozoic and its relationship with supercontinental Rodinia in North China. Chinese Science Bulletin, V. 45(10), p.931-935. Torsvik, T. H., Smethurst, M. A., Meert, J. G., Van der Voo, R., McKerrow, W. S., Brasier, M. D., Sturt, B. A., Walderhaug, H. J., 1996. Continental break-up in the Neoproterozoic and Paleozoic-A tale of Baltica and Laurentia. Earth-Sci. Rev., v. 40, p.229-258. Van der Voo R., 1990. The reliability of paleomagnetic data, Tectonophysics, v. 184, p. 1-9. Wang H. Z., X. Li, S. Mei, S. Zhang, 1997. Pangaea cycles, earth's rhythms and possible earth expansion, Proc. 30th Intern. Geol. Congr. v.l, p.l 11-128. Zhang, S., Li, Z. X., Wu, H., Wang, H., 2000. New paleomagnetic results from the Neoproterozoic successions in southern North China block and paleogeographic implications, Science in China, Series D, v. 43(Supp.), p. 233-244. Zhao, G., Cawood, P. A., Wilde, S. A., Sun M., 2001. Review of global 2.1-1.8 Ga orogens and associated cratons: A preRodinia Supercontinent? Earth-Science Reviews (in press).

119

Rodinia Symposium, Perth October 2001

Geological Society of Australia Abstracts 65

Author Index ABELLO, J ADAMS, C J ARMSTRONG, R. A

1 3 80

KINNY, P.D KRONER, A KUNK, M J

74, 78 71 64

BARLEY, M BERRY, R BETTENCOURT, J.S BEYTH, M BOGDANOVA, S.V BRADSHAW, M . A BREWER, T.S BURRETT, C

3 10 27 7 8 3 20 10

LEITCH, E LI, X.H LL, Z.X

13 74, 78 74, 7 8 , 1 1 7

CARTER, G.D CAWOOD, P.A COLLINS, A.S

CORDANI, U . G CUTTEN, H

D'AGRELLA-FILHO, M.S DALZIEL, 1 DE BRITO NEVES, B.B

11 13, 43, 55, 57 16, 20 2 4

27, 31 35 31

ELMING, S.-A

27, 82 36, 113

GALLEN, M J GERALDES, M.C GOLYNSKY, A GREY, K HARLEY, S.L HATTON, C HOFFMAN, P.F HOLLINGSWORTH, D.A HULSCHER, B

16, 20, 39, 60 43 27 68 45 48 52 53 55, 57 20, 60

IRIONDO, A

JACOBS, J JOHNSON, S.P JOHNSON, S.T

68 1,69 80

82 57 81

NELSON, D.R

81 27,31 82 3 85, 113 60,85 88 64

RAINBIRD, R.H RAZAKAMANANA, T REDDY, S.M RIVERS, T

91 20 43 94

SCHMIDT, P.W SINCLAIR, J.A SLQUEIRA, R SIRCOMBE, K.N SOVETOV, J.K

95 99 27 101 105

TEIXEIRA, W THOMAS, R.J TRINDADE, R.I.F

27 68 31

WILLIAMS, G.E WINDLEY,B.F WINGATE, M.T.D

108 20 85,113

ZHANG, S

ZHOU, H

120

MERTANEN, S MONEK, B MYERS, J.S PACCA, I.G PESONEN, L.J PICKARD, A.L PISAREVSKY, S.A POWELL, C.MCA PREISS, W.V PREMO, W . R

EVANS, D.A.D FITZSIMONS, I.C.W

MCCOURT, S

117

74, 78