Issuu

Geological Society of Australia

ABSTRACTS Number

Orogenesis in the Outback 12 - 1 6 th July, 1999, Alice Springs, Australia Specialist Group in Geochemistry, Mineralogy and Petrology

Orogenesis in the Outback A look at cyclicity and reactivation in orogenic belts

Alice Springs, central Australia, July 1999

Specialist Group in Geochemistry, Mineralogy and Petrology (GSA) In Association With

Tectonic Studies Group, UK IGCP368 Northern Territory Geological Survey Northern Territory Division of the Geological Society of Australia

Abstract Volume 54 Edited by J. A. Miller and I.S. Buick

ISSN 0729-01IX National Library of Australia Cataloguing-in-Publication Data

© Geological Society of Australia Incorporated, Specialist Group in Geochemistry, Mineralogy and Petrology, 1999 Copies of this abstract volume may be purchased from The Business Manager, Geological Society of Australia, Suite 706, 301 George St., Sydney, N.S.W. 2000 Australia. Example citation for papers in this volume: L.E.Beacom, T.B.Anderson & R.E.Holdsworth, 1999. Brittle fracture patterns and reactivation in basement and cover rock sequences. Geological Society of Australia, Abstracts No. 54, pp 6.

MAJOR SPONSORSHIP FOR OROGENESIS IN THE OUTBACK HAS BEEN PROVIDED BY

NORTHERN TERRITORY GEOLOGICAL SURVEY

SPONSORING COMPANIES AND ORGANISATIONS

Australian Geodynamics Cooperative Research Centre Normandy NFIVI Monash University Bookshop Board IGCP 368 Pasminco Acacia Victorian Division of the Geological Society of Australia Northern Territory Convention Bureau

STUDENT SUPPORT FOR OROGENESIS IN THE OUTBACK HAS BEEN PROVIDED BY

Specialist Group in Tectonics and Structural Geology GSA Specialist Group in Solid Earth Geophysics GSA Victoria Division of the Geological Society of Australia South Australia Division of the Geological Society of Australia Western Australia Division of the Geological Society of Australia Tasmania Division of the Geological Society of Australia

ABSTRACT LIST (presenting Author is first unless underlined)

L E . B e a c o m , T.B.Anderson & R.E.Holdsworth Brittle fracture patterns and reactivation in basement and cover rock sequences.

C . B i e r m e i e r & K. S t u w e Strain rates from inclusion trails in garnet: A coupled microstructural and metamorphic approach

S . B o d o r k o s , P.A. C a w o o d a n d N . H . S . O l i v e r Chronology of rapid orogenesis in the Palaeoproterozoic central Halls Creek Orogen, northern Australia.

S . B o d o r k o s , N . H . S . Oliver, P.A. C a w o o d , M . S a n d i f o r d , A . O r d a n d B . H o b b s Thermal modelling of temperature-time paths in the palaeoproterozoic Halls Creek Orogen, WA: constraints from the Tickalara Metamorphics.

J . B r a u n a n d R. S h a w A continental-scale tectonic model for the reactivation of Proterozoic sutures in the mid- to late Paleozoic

I. S. B u i c k , R. Frei, a n d I. C a r t w r i g h t The timing of multiple(?) high-temperature retrograde events in the Reynolds Range, central Australia

I. S. B u i c k , I. S . W i l l i a m s , R . L G i b s o n a n d J . A . Miller Polymetamorphism in the central zone of the Limpopo Metamorphic Belt, South Africa: Constraints from a SHRIMP zircon, monazite and titanite study.

I. S. B u i c k a n d M . H a n d Polyphase metamorphism and reactivation in the Reynolds-Anmatjira Range area, northern Arunta Inlier.

I. C a r t w r i g h t , I. S. B u i c k . J. K. V r y Episodic, channelled fluid flow from internal and external sources: Reynolds Range, central Australia

M . C h o , S. R. L e e , K. Y i a n d R. S t e r n Metamorphic and tectonic evolution of the Paleoproterozoic granulites in the northeastern Gyeonggi massif, South Korea

D . J. C l a r k , B. J. H e n s e n & P. D. K i n n y Geochronological constraints for a two-stage history OF the ALBANY-Fraser orogen, W.A.

N . G . C u l s h a w , R.A. J a m i e s o n . J . W . F . K e t c h u m , N. W o d i c k a , D . C o r r i g a n , & P.H. R e y n o l d s Polyphase ductile thrusting and extension in the lower orogenic crust: western Grenville Province, Georgian Bay, Ontario

P. Ding Field evidence for five Palaeoproterozoic orogenies in the Northern and Eastern Arunta Orogenic Province, central Australia

P. Ding Cyclic orogenies in the Halls Creek region, WA - Field evidence

A . D o m b r o w s k i a n d S. H o e r n e s

Controls on fluid flow in scapolite-bearing shear zones of the Late-Proterozoic Kuiseb Formation, Damara Orogen, Central Namibia. 28

LA.D.Fernandes, C.C. Porcher, A.O.M. Silva, R.A. Cliff & G.T.R. Droop Styles of tectonic reworking of basement in the Dom Feliciano Belt of southern Brazil during the Neoproterozoic

T. Flottmann, M. Hand, D. Close, C. Edgoose and I. Scrimgeour Thrust tectonic styles at the margins of the Amadeus Basin, Central Australia

B.R. Goleby, BJ. Drummond and RJ. Korsch The crustal structure of central Australia from deep seismic profiling

G.H. Grantham, A.S.T.D. Manhica, R.A. Armstrong The nature of the boundary between the Zimbabwe Craton and the Mozambique Belt, central Mozambique and new SHRIMP zircon data.

M. Hand, K. Slater, S. MacLaren, M. Sandiford Heat production rates in Australian Proterozoic terrains

S. L Harley

UHT crustal metamorphism: Advances in constraining peak temperatures and P-T histories 38

S. L Harley Elucidating P-T-f-event histories in complex high grade terrains: Lessons from Antarctica

B. Holdsworth, M. Stewart & J. Imber The textural and rheological evolution of long-lived fault zones in orogenic belts

O. Holm The geology of the Arthur Lineament, Tasmania

G. Houseman Mechanisms of lithospheric renewal associated with continental orogeny

M. H. Huang, I. S. Buick, I. S. Williams and R. Maas Polymetamorphism and the tectonic evolution of the Danba Domal Terrain, Sichuan Province, Western China

Hunze S., Buysch A., Stiiwe K.. Ehlers K. The staurolite isograd in the nine mile region. Broken Hill Block, Australia. Evidence for multiple metamorphic events?

K.J. Hussey, J. Smith and N. Donnellan Some regional implications of new geochronological constraints from the Tennant Creek and Arunta Inliers, central Australia.

R.A. Jamieson, C. Beaumont, 0. Vanderhaeghe, and P. Fullsack Thermal-Mechanical Models of Large Convergent Orogens

R.A. Jamieson, N.G. Culshaw, S.M. Carr, R.M. Easton, & J.W.F Ketchum The Grenville Orogenic Cycle: Constraints from a geological transect across Ontario and New York

W.P. Johnstone and C. Harris Fluid-rock interaction at c. 1000 Ma and c. 500 Ma in dronning Maud Land, east Antarctica: A stable isotope study of the Sverdrupfjella Group.

S. Keay, G. Lister I. Buick

Cyclicity in orogensis recorded by multiple episodes of zircon growth: an ion microprobe (SHRIMP) study of the Alpine evolution of the Cyclades, Greece. 57

J. Kim and M. Cho Progressive low-pressure metamorphism in northeastern Yeongnam Massif, Korea

E. Koester, LA.D. Fernandes. A.R. Pawley & G.T.R. Droop High-grade gneisses as source-rocks for syntranscurrent peralunninous granites in southeastern Brazil: an experimental study

M. Krabbendam When the Wilson Cycle breaks down: inhibiting reworking and How some orogens can produce strong lithosphere

P. H. Macey and C. Harris Magmatism, metamorphism and fluids in the northern margin of the Damara Orogen, Namibia.

R. Maas Rb-Sr, Sm-Nd and U-Pb (Pb-Pb) dating of metamorphic assemblages - some remarks on applications and recent developments

S. McLaren and M. Sandiford Cyclicity in the Australian Proterozoic- Influences and Controls

J.A. Miller, J. Mawby, M. Hand, LS. Buick, I.S. Williams The Larapinta Event: Palaeozoic metamorphism in the eastern Arunta Inlier

J.A. Miller, I.S. Buick and I. Cartwright The role of fluids in the development and reworking of high-pressure assemblages and textures: examples from the Corsican Ophiolite, France

A. Moller, R.A. Armstrong, M. Ballevre, B.J. Hensen & K. Mezger Crustal growth, metamorphism and deformation in the Strangways metamorphic complex: A summary of recent U-Pb and Sm-Nd geochronology.

A. Moller, I.S. Williams, S. Jackson & B.J. Hensen Palaeozoic deformation and mineral growth in the Strangways metamorphic complex: In-situ dating of zircon and monazite in a staurolite-corundum bearing shear zone.

Emily A. Neil and Gregory A. Houseman Episodic Intraplate Orogeny in central Australia; is it mantle driven?

Emily A. Neil and Gregory A. Houseman Lithospheric strength and plate boundary stresses During the Intraplate Alice springs orogeny

0. Nzolang, J.P. Nzenti, J.P. Tchouankoue, A.A. Ganwa Crustal anatexis and in-situ granite formation in the Pan-African north-equatorial fold belt: The migmatites and associated granites of the Bantoum area, west Cameroon, central Africa.

H.-J. Paech Pan-African overprint of the Grenvillian basement in central Dronning Maud Land.

M. K. Pandit Geochemistry of coexisting Na rich and K granites: A Late Proterozoic thermal event in the North Delhi Fold Belt, western India

H. Park and M. Cho Petrogenesis of the scapolite-bearing calc-silicate rocks in the western Imjingang belt, Korea

D. Phillips, R.L. Gibson, C.R. Anhaeusserand G.B. Kiviets "^^Ar/^^Ar approaches to dating shear zones.

C.C. Porcher, G.T.R. Droop, L.A.D. Fernandes and K.H. Brodie Re-equilibration of granulites in a transcurrent shear zone in south-eastern Brazil.

H. M. Rajesh Aluminous A-Type granitoid scenario of southwestern India: Characterisation and regional implications.

C. M. Read and 1. Cartwright Meteoric fluid infiltration during Palaeozoic exhumation of the Proterozoic Yalbadjandi Shear Zone, central Australia.

C. M. Read and 1. Cartwright. The role of fluids in a major crustal thrust zone, Redbank High Strain Zone, central Australia.

D. Rubatto, I. S. Williams, I. Buick Zircon and monazite record of Proterozoic metamorphism in the Reynolds Range, central Australia.

D. Rubatto Migration of subduction in the Alps

A. Rudge, B. Goleby and T. Barr The crustal structure of the Arunta Block and northern Amadeus Basin: from AGSO Deep Reflection Seismic Data

P. D. Ryan, The role of deep basement in continental tectonics

P. D. Ryan & John F. Dewey Ordovician arc, continent collision in the Caledonides of western Ireland.

M. Sandiford, 8. McLaren, M. Hand Intraplate orogeny and the geochemical structure of the continental crust: an outback Australian perspective.

M. Satish-Kumar and H. Wada Does carbon isotope equilibrium exists between graphite and fluid inclusions in granulites? A case study from the Nilgiri granulites, southern India.

101

M. Satish-Kumar, H. Wada, M. Santosh and M. Yoshida The thermal structure of a southern Indian granulite terrain based on carbon isotope thermometry

103

I. Scrimgeour and J. G. Raith High grade mylonitic reworking of the Kanadra Granulite, central Australia: Implications for the evolution of the eastern Arunta Inlier.

105

L. Songnian, L. Huaikun, Y. Haifeng, Z. Fengqing and Y. Chunliang Neoproterozoic events in the Phanerozoic Orogenic Belts of central China

106

S.A. Uberall, K. Stuwe. K. Ehlers and A.P. Nutman Deformation history of Proterozoic granites in the north of the Broken Hill Block, NSW, Australia

107

J.K.Vry High-grade metamorphism of low-T"diagenetic-hydrothermar' alteration in a polymetamorphic belt, Reynolds Range, Australia.

109

I. S. Williams and I. S. Buick The behaviour of zircon, monazite and their U-Pb isotopic systems during the hgih-grade metamorphism of quartzo-feldspathic sedimentary rocks.

110

M. Yoshida Recurrent tectonothermal events in a high-grade terrain: A look into geochronology from Sri Lanka and surrounding areas.

111

M. Yoshida, M. Arima and A.T. Rao Recyclic tectonothermal events in the eastern Ghats: Continuation to SW Australia and Antarctic.

113

G. Zhao, P. A. Cawood & S. A. Wilde Polymetamorphism of the Archean mafic granulites from the Trans-North China Orogen: textural evidence and tectonic implications.

115

G. Zhao' S. A. Wilde, P. A. Cawood Two types of metamorphism of the basement rocks in the western zone of the North China Craton and their tectonic implications.

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

BRITTLE FRACTURE PATTERNS AND REACTIVATION IN BASEMENT AND COVER ROCK SEQUENCES. L.E.Beacom^ ^ T.B.Anderson^ & R.E.Holdsworth^ 7 Dept of Geology, School of Geosciences, Queen's University of Belfast, BT7 1NN, N. Ireland, UK. 2 Reactivation Research Group, Dept of Geological Sciences, University of Durham, DH1 3LE, UK.

This study presents a quantitative and qualitative assessment of the nature and distribution of fracture arrays in highgrade basement gneisses and in overlying sequences of sedimentary rocks. Examples are taken from the Late Archaean to Proterozoic Lewisian basement Complex in NW Scotland and the unconformably overlying Late Proterozoic and Early Palaeozoic sediments. Fracture populations are classified by orientation, relative age, movement sense and direction, fault rock infill, spacing, length, width, connectivity and fractal statistics. The textural evolution and deformation mechanisms define the rheological controls on fracture development and provide a means to access the potential for reactivation of basement fault zones into the overlying sediments. The fracture evolution pathways described for basement rocks illustrate the influence of pre-existing structures and fluid activity on fault zone development. In the deep brittle regime (>10km), rock-dominated processes produce pseudotachylyte and cataclasitic fault rocks which are controlled by pre-existing ductile foliation and lithology. Syntectonic fluid activity in these fault zones has led to reaction weakening and the development of phyllonitic fault rocks which overprint the products of earlier brittle deformation. This occurred in the vicinity of the main load bearing region in the crust. At shallower crustal depths, hydrofracture meshes localise around major faults and produce incohesive breccias, gouges and veining. These fractures locally reactivate basement anisotropies but largely ignore many of the pre-existing structures. These findings suggest that significant reactivation of brittle structures only occurs in fault zones which have experienced long-term weakening. The fracture population characteristics within basement fault zones are used to classify reactivated and non-reactivated structures. Some of the large scale fracture attributes associated with the basement fault arrays are transferred into the fault zones in the overlying sediments. This allows the fracture characteristics in the sediments to be used to determine if basement structures are reactivated. The limitations of this approach can be seen at small scales, generally less than tens of metres, when bedding and lithological heterogeneities in the sediments act as the main controls influencing the fracture development.

O R O G E N E S I S IN T H E O U T B A C K

STRAIN RATES FROM INCLUSION TRAILS IN GARNET. A COUPLED MICROSTRUCTURAL AND METAMORPHIC APPROACH C. Biermeier & K. Stiiwe

Department Austria

of Geology

and

Palaeontology, Karl-Franzens-University,

Heinrichstr.

26 A-8010

Graz,

The magnitude of strain rates during continental deformation is one of the most difficult parameters to extract from metamorphic rocks as it is only indirectly recorded in the thermal, baric and textural record of the assemblages. However, it is one of the more important parameters to know, as it is intimately related to the duration of tectonothermal events. It is also ultimately useful to discern between episodic and continuous events. In this study we present preliminary results from a serial sectioning study in which we use the orientation of spiral inclusion trails in rotated garnet porphyroblasts from the Gleinalm Complex, Eastern Alps, to infer the different stages of fabric development during garnet growth and ultimately strain rate. The study is based on the assumptions that: (a) garnets do rotate during non-coaxial deformation at a rate that is proportional to the shear strain rate of the rock (despite: Bell et al., 1992; see: Passchier et al., 1992) and (b) that we know the thermal evolution - and therefore the heating rate - of the area (e.g. Neubauer et al., 1995; Stuwe & Powell, 1995). The Gleinalm area is part of the high grade metamorphic Austroalpine nappe complex which was stacked during the late Cretaceous and exhumed in a sinistral wrench corridor during the Late Cretaceous and Tertiary. Pelitic gneisses reached upper amphibolite fades metamorphic conditions during the stacking of this pile. Characteristic gamet-muscovite-biotite-quartz assemblages grew on the heating path syn-kinematically. This is evidenced by a strong planar mica fabric and garnet porphyroblasts with extreme spiral patterns. There is a total of about 5 vol% of garnet in the rock. Individual crystals range from 1 to 10 mm in size. Spiral patterns appear continuous and garnets have an overall single-phase appearance. Thus, we infer that all garnet grew during the Cretaceous kinematic and metamophic event. Fig. 1. Schematic illustration of garnet growth during shearing. The white area is a garnet porphyroblast at time t5. The dashed area is the surrounding mica schist matrix, shown for the timesteps t4 (shaded) and t5 (square box). Circles in the garnet show garnet radius at times t2 - t4. The radius increases with R=t^0.3 so that volumetric growth is linear with time, a is the angle of inclusion trail rotation per time step. Here it is constant, because the shearing rate is constant. We make the simplifying assumption that the shear strain, y, is related direcdv to a. Thus we use Y=t2Cay For this project, the internal structure (quartz inclusion trails) in the garnet porphyroblasts was studied by series of thin sections (using the methods of Johnson & Bell, 1996). About 120 thin sections were cut of 3 blocks, each 2,5x4,5x5 cm in size. In each thin section we mapped out the volumetric proportions of parts of garnet crystals that preserve a given orientation of the inclusion trails. With some assumption on the rate of garnet rotation during shearing (based on model results using ELLE; Jessel et al., 1999), we are able to derive how much garnet grew during different stages of shear strain. Independently of this we calculated the volumetric rate of gamet growth during the heating path. This information was obtained by calculating the equilibrium volumetric proportions of gamet in a rock of the relevant bulk composition using, the intemally consistent thermodynamic data set of Holland and Powell (1990) and the computer program THERMOCALC. This shows, that - for the bulk composition of the Gleinalm metapelites about 5 vol% of gamet grow at a linear rate by divariant reaction in the temperature range between 550°C and 650^C.

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

Coupling of the thermodynamic information with the microstructural information allows us to infer the orientation of the fabric as a function of volumetric garnet growth and therefore of temperature. From this, we have calculated shear strain rates for assumed heating rates. For example, for a heating rate of 10 °C/my and 5 vol% of gamet growth between 550°C and 650°C each vol% of garnet grows in 2my. Thus, for a linear volumetric growth rate, the radial growth of gamet radius, R, must occur at a rate of (Figure 1). Comparing the orientation of the inclusion trails with knowledge of R as a function of temperature (or time) allows us to infer the shear strain rate, y/my. On figure 1, the change of inclusion trail orientation (angle a ) between each time step (2my) is about 35®. Thus, from the schematic illustration of fig. 1, we estimate a preliminary result of a shear strain rate of 0.35 my"^ This study is part of FWF project P-12846-GEO.

REFERENCES

Bell, T.H., Johnson, S.E., Davis, B., Forde, A., Hayward, N., and Wilkins C., 1992. Porphyroblast inclusion-trail orientation data: eppure non son girate*! Journal of Metamorphic Geology, 10, 295-307. Holland T.J.B. and Powell, R., 1990. An enlarged and updated internally consistent thermodynamic dataset with uncertainties and correlations: the system K20-Na20-Ca0-Mg0-Fe0-Fe203-A1203-Ti02-Si02-C-H2-02.. Journal of Metamorphic Geology, 8, 89-124. Jessel, M., Bons, P., Evans, L., Barr, T. & Stuwe, K., 1999. Elle: A system for the simulation of deforming and metamorphosing rocks and ist application to anisotropic grain growth. Abstracts for DRM conference. Johnson, S.E. and Bell, T.H., 1996. How useful are 'millipede' and other similar porphyroblast microstructures for determining synmetamorphic deformation histories? Journal of Metamorphic Geology, 14, 15-28. Neubauer, F., Dallmeyer, R.D., Dunkl, I., and Schimik, D., 1995. Late Cretaceous exhumation of the Gleinalm dome, Eastern Alps: kinematics, cooling history and sedimentary response in a sinistral wrench corridor. Tectonophysics, 242, 79-98. Passchier, C.W., Trouw, A.J., Zwart, H.J., Vissers R.L.M., 1992. Porphyroblast rotation: eppur si muove*? Journal of Metamorphic Geology, 10, 283-294. Stiiwe, K. and Powell, R., 1995. PT paths from modal proportions: application to the Koralm Complex, Eastern Alps. Contrib Mineral Petrol, 119, 83-93.

OROGENESIS IN THE OUTBACK

CHRONOLOGY OF RAPID OROGENESIS IN THE PALAEOPROTEROZOIC CENTRAL HALLS CREEK OROGEN, NORTHERN AUSTRALIA S. Bodorkos^^ P.A. Cawood^ a n d N.H.S. Oliver^

[TSRC, School of Applied Geology, Curtin University, GPO Box U1987, Perth \NA 6845. *Email: simonb@lithos. curtin. edu.au ^EGRU, School of Earth Sciences, James Cook University, Townsville QLD 4811.

This study integrates well-constrained U-Pb SHRIMP dating and CL imaging of zircons with field and metamorphic data to resolve discrete but very closely spaced events in the evolution of the Palaeoproterozoic Halls Creek Orogen, which represents the interface between the colliding Kimberley and North Australian cratons (Tyler et al 1995). One of the aims of this work is to construct an absolute time frame for the evolution of the Tickalara Metamorphics in the northern Central zone of the Halls Creek Orogen, in response to the -1850 Ma collisional event, which involved intrusion of voluminous mafic and felsic plutons broadly synchronous with upper greenschist- to lower granulite-facies, low-pressure metamorphism. The data presented here require rapid deposition, burial and metamorphism of the Tickalara Metamorphics within 10-20 million years, implying a rate of orogenic process (sedimentation, magmatism and metamorphism) not previously documented in Australian Proterozoic mobile belts. A greenschist-facies quartz-muscovite psammite was targeted to constrain the ages of detrital zircons, while avoiding the overprinting effects of high-grade metamorphism observed elsewhere in the unit. The zircons are characterised by well-developed and complex oscillatory zoning, commonly truncated by mechanically abraded edges and fracture boundaries. The zoning structures are inferred to be of igneous origin, with some grains exhibiting evidence for an extended crustal prehistory. Nine of the 27 SHRIMP analyses from this sample returned 207p|3/206p|3 aggg older than 1900 Ma. The other 18 analyses form a single, mostly concordant group with a 207py206p|3 age of 1864 ±4 Ma (chi-squared = 0.69), placing an upper limit on the age of the protolith to the Tickalara Metamorphics (Bodorkos et al. in press). Zircons were also extracted from an uppermost-amphibolite facies garnet-sillimanite metapelite containing -10 vol% leucosome material, in order to constrain the timing of metamorphism. The suite comprises a relatively high proportion of euhedral grains, and although most cores retain complex oscillatory zoning, many are overgrown by thin (-10-20 |im) structureless rims with constant (high or low) CL response. In addition, a small number of zircons analysed displayed little internal structure. Five of the 38 analyses from this sample yielded ^^'^Pb/^^^Pb ages older than 1900 Ma. The remaining 33 analyses form a large, mostly concordant group with ages in the range 1885-1830 Ma and a ^^"^Pb/^^^Pb age of 1857 ±5 Ma. However, chi-squared = 2.81 for this age group, indicating scatter in the ages well beyond that expected solely from analytical uncertainty. This suggests the presence of two or more temporally (and geologically) distinct age components in the data set. A relationship is evident between the internal structures sampled by the ion beam and the intragrain ^^'^Pb/^^^Pb ages in the post- 1900 Ma data set. Of the 33 analyses, 18 pre-1855 Ma ages fall within areas of zircon displaying well-developed oscillatory zoning, which are interpreted as detrital zircons which survived the HTLP metamorphic event. Treating these oscillatory-zoned zircons as a single group yields a ^^"^Pb/^^^Pb age of 1867 ±4 Ma (chisquared = 0.84), which is a statistically valid single population. In addition, 10 post-1850 Ma ages are from rims with no CL-visible structure, which overgrow and truncate oscillatory-zoned cores. These rims are interpreted as the expression of limited zircon dissolution and reprecipitation during uppermost-amphibolite facies metamorphism, and the 10 analyses yield an age of 1843 ±4 Ma (chi-squared = 0.77), also a single population within analytical error. The remaining five ages are in the range -1860-1850 Ma from zircons with little internal structure, and are of uncertain geological affinity. In an attempt to account for all the data, the mixture modelling program MIX (Sambridge and Compston 1994) was used to estimate the number of components and ages of those components in the dataset of 33 post-1900 Ma ages. The best fit model comprised two components: an older group of 19 analyses with a ^^'^Pb/^^^Pb age of 1867 ±4 Ma and a younger group of 14 analyses with a ^^'^Pb/^^^Pb age of 1845 ±4 Ma. The older group matches almost exactly with the 18 pre-1850 Ma ages in oscillatory-zoned zircon, and the ^^'^Pb/^^^Pb age is identical to the older group distinguished on the basis of CL structures. The age is also in excellent agreement with the 1864 ±4 Ma zircon suite

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

from the low-grade psammite, supporting the inferred detrital origin of the oscillatory-zoned grains. The best fit MIX model assigned four of the five analyses from weakly-zoned zircons into the younger group, which is dominated by structureless rims. Nevertheless, the 1845 ±4 Ma age does not differ significantly from the 1843 ±4 Ma age derived from the 10 rim analyses without the unzoned zircons. Furthermore, the 1845 ±4 Ma age is identical to a conventional U-Pb monazite age of 1845 ±3 Ma from a granulite-facies metapelite less than 10 km to the west (Oliver et al in press), constraining the timing of peak metamorphism.

I I

The size of the zircons and the prominence of oscillatory zoning suggest that the detrital zircons in metasedimentary units of the Tickalara Metamorphics crystallised in a plutonic environment. The depth of emplacement of this plutonic zircon source is unknown, but is unlikely to have been deeper than shallow-mid crustal levels in view of the subsequent rapid erosion and deposition history. Assuming an arbitrary emplacement depth of -5-10 km for the -1865 Ma zircon-bearing granitoid(s) and a relatively rapid average exhumation rate of 1 mm/year (cf. 1.2 ±0.6 mm/year for the High Himalaya; Hubbard et al 1991), the source rocks are exposed to erosion at the surface at -1860-1855 Ma. Equally rapid burial then occurred, although at least part of the sedimentary succession was still relatively cold at -1850 Ma, since intrusion of the 1850 ±2 Ma Fletcher Creek Granite (Page et al 1995) created a prominent metamorphic aureole. Sedimentation in the proto-Tickalara depocentre, followed by tectonic loading resulted in the metasedimentary rocks of the northern Central zone reaching peak metamorphic pressures of 350-500 MPa (Thornett 1986) and depths of -10-15 km at -1845 Ma. Repeated mafic magma injection and remelting of pelitic wallrocks suggest that this thermal event was protracted (Oliver and Barr 1997, Oliver et al in press), followed by ductile D3 deformation at -1830 Ma during slow cooling from peak metamorphic conditions. Macroscopic D4 folding occurred at -1820 Ma, with truncation of limbs by kilometre-scale, brittle retrograde (muscovite-chlorite and chlorite-epidote bearing) late-S4 faults constitutes the first evidence for substantial cooling from peak metamorphic conditions -25-30 million years earlier. Our results permit estimates of the rate of crustal processes in the Halls Creek Orogen, and have implications for the rate of Proterozoic orogenic processes elsewhere, since regional-scale geochronological studies in comparable Proterozoic domains commonly lack the detail required to resolve events recorded by zircon which are separated by less than a few tens of millions of years. We suggest that the broad synchronicity of magmatism (providing a source for detrital zircon), sedimentation and metamorphism in the Halls Creek Orogen indicates tectonic activity at a major plate boundary, and that similar geologic histories in comparable Proterozoic plate margins elsewhere may have gone undetected in many cases due to low precision of the available geochronological data, sparse age data at a regional scale and/or inconclusive field relationships. REFERENCES

I I

Bodorkos, S., Oliver, N.H.S. and Cawood, P.A., in press. Thermal evolution of the central Halls Creek Orogen, northern Australia. Australian Journal of Earth Sciences. Hubbard, M., Royden, L.H. and Hodges, K.V., 1991. Constraints on unroofing rates in the High Himalaya, eastern Nepal. Tectonics, 10, 287-298. Oliver, N.H.S., Bodorkos, S., Nemchin, A.A., Kinny, P.D. and Watt, G.R., in press. Relationship between zircon UPb SHRIMP ages and leucosome type in migmatites of the Halls Creek Orogen, Western Australia. Journal of Petrology. Page, R.W., Tyler, I.M. and Blake, D.H., 1995. Geochronology of magmatism and high-grade metamorphism, Kimberley region, W.A. Australian Conference on Geochronology and Isotope Geo science Abstracts, 3, 25. Curtin University of Technology, Perth. Sambridge, M.S. and Compston, W., 1994. Mixture modeling of multi-component data sets with application to ionprobe zircon ages. Earth and Planetary Science Letters, 128, 373-390. Thornett, J.R., 1986. Evolution of a high-grade metamorphic terrain in the Proterozoic Halls Creek Mobile Zone, Western Australia. PhD thesis. University of Western Australia, Perth (unpubl.). Tyler, I.M., Griffin, T.J., Page, R.W. and Shaw, R.D., 1995. Are there terranes within the Lamboo Complex of the Halls Creek Orogen? Geological Survey of Western Australia Annual Review 1993-94, 37-46.

OROGENESIS IN THE OUTBACK

THERMAL MODELLING OF TEMPERATURE-TIME PATHS IN THE PALAEOPROTEROZOIC HALLS CREEK OROGEN, WA: CONSTRAINTS FROM THE TICKALARA METAMORPHICS S. Bodorkos^', N.H.S. Oliver^, P.A. Cawood\ M. Sandiford^ A. Ord^ and B. Hobbs^

[TSRC, School of Applied Geology, Curtin University, GPO Box U1987, Perth WA 6845. *Email: simonb@lithos. curtin. edu. au ^EGRU, School of Earth Sciences, James Cook University, Townsville QLD 4811. ^Department of Geology and Geophysics, University of Adelaide, Adelaide SA 5005. ^CS/RO Exploration & Mining, PO Box 437, Nedlands WA 6009.

The role of modern-style plate tectonic processes in Australian Palaeoproterozoic orogens characterised by extensive high-temperature, low-pressure (HTLP) metamorphic rocks is poorly constrained (e.g. Etheridge et al 1987, Wyborn et al. 1992). In such terranes, elucidation of the thermal history via the construction of wellconstrained P-T-t paths is of critical importance in the interpretation of possible tectonic environments. Detailed field, petrological and geochronological data from the Tickalara Metamorphics, within the Central zone of the Palaeoproterozoic Halls Creek Orogen in northern Australia, indicates it represents a section of the middle crust that underwent a protracted thermal event, with temperatures remaining above -500"C in the interval -1850-1820 Ma, and was subjected to a sequence of three clearly overprinting deformation events. Voluminous intrusion of intermediate and mafic plutons during this period suggests that advective heat transfer contributed to the elevated crustal geotherm. The Tickalara Metamorphics, which represent the oldest rock unit in the Central Zone, are dominated by clastic metasedimentary rocks, with minor metacarbonates and mafic metavolcanic rocks. Metamorphic grade varies from greenschist facies in the south to granulite facies in the north. Field, petrological and geochemical data relevant to the resolving the tectonothermal evolution of the northern Central zone include the following. Detrital zircon populations provide a maximum depositional age for the proto-Tickalara sedimentary rocks. Youngest detrital suites have been dated at 1865 ±2 Ma (Page et al. 1995a), 1864 ±4 and 1867 ±4 Ma (Bodorkos et al. 1998) from metasedimentary rocks throughout the unit. The northern Tickalara Metamorphics were intruded by the garnetbearing leucocratic Fletcher Creek Granite sill (U-Pb SHRIMP age 1850 ±2 Ma, Page et al. 1995b) which, despite its chemical resemblance to the ternary minimum melt composition, has a prominent -100 metre-wide pyroxenehomfels facies metamorphic aureole. This is interpreted as evidence that the host rocks were still cold (<300-400°C) at the time of intrusion. The Fletcher Creek Granite also contains a well-developed solid-state foliation, suggesting emplacement prior to HTLP metamorphism and associated deformation. Peak regional metamorphism in the northern Central zone reached temperatures of -700-750'C at pressures of 300-400 MPa, resulting in widespread development of garnet-biotite and garnet-cordierite±spinel assemblages in metapelites. Timing of peak metamorphism is constrained by a U-Pb SHRIMP age of 1845 ±4 Ma from overgrowths on zircon from an upperamphibolite facies metapelite (Bodorkos et al. 1998) and an identical conventional U-Pb monazite age of 1845 ±3 Ma from a nearby granulite facies stromatic migmatite (Oliver et al. in press). The Sally Malay layered maficultramafic intrusion (U-Pb SHRIMP age 1841 ±3 Ma, Trudu and Hoatson 1996) post-dates peak regional metamorphism, cross-cutting the regional foliation and superimposing a diatexitic contact aureole on local metapelitic gneisses. Repeated mafic magma injection has resulted in multiple generations of contact/sheath migmatites with anatectic zircons as young as -1835 Ma (Oliver et al. in press). East of the Sally Malay intrusion, emplacement of the Mabel Downs Tonalite (U-Pb SHRIMP age 1832 ±3 Ma, Page et al. 1995b) was synchronous with ductile D3 deformation. Mutual cross-cutting relationships between folds, unretrogressed hornblende-bearing intrusive phases and ductile shear zones suggest that the pluton was emplaced under amphibolite-facies conditions. Younger plutons in the same suite (e.g. Sally Downs Tonalite, U-Pb SHRIMP age 1821 ±4 Ma, Sheppard et al 1995) post-date D3, but are folded by F4 folds and cut by brittle retrograde S4 faults, which in turn pre-date a "stitching" -1810 Ma granitoid in the southern Halls Creek Orogen (Sheppard et al 1995). These brittle faults are the first evidence for significant cooling from peak metamorphism -25-30 million years earlier. Any thermal model of crust-mantle interaction to produce the observed mid-crustal T-t path must account for the following: (i) rapid erosion, deposition and burial of the proto-Tickalara sediments; (ii) protracted HTLP metamorphism accompanied by layered mafic intrusions; and (iii) relatively slow cooling from peak metamorphic conditions (-300'C in -30 million years). Two simple models are compared here: anomalous mantle heat flux, and

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

instantaneous crustal thickening. Both of these models are capable of producing the protracted HTLP metamorphism in the middle crust observed in the Halls Creek Orogen, followed by relatively slow cooling. The first model uses a simple 2-D anomalous mantle heat flux (qanom ~ 100 mWm"^) with a horizontal decay length scale of 100 km. The anomalous heat flux q^^^ is generated by simulating the mantle heat flow at a mid-ocean ridge shortly after the cessation of lithospheric rifting, which is then applied to the base of a 40 km-thick crust and allowed to decay exponentially with time. This model has the advantage of invoking thinning and/or rifting of the mantle lithosphere even during crustal thickening, and the observed large volumes of layered mafic-ultramafic intrusions in the middle- and lower crust are a predictable consequence of this process. In addition, mantle involvement could account for the generation of the Sr-undepleted, Y-depleted Mabel Downs Tonalite and its postpeak metamorphic emplacement in the middle crust. This trace-element geochemical signature is very rare in granitoids of the Australian Proterozoic and Wybom et al (1992) suggest that such intrusives are mantle-derived, from a source in which garnet is stable. A possible drawback of the model is that many aspects of mechanisms providing anomalous mantle heat flux, especially with respect to the inferred magnitude of the anomalous pulse, remain poorly understood. The second model is essentially one-dimensional and assumes constant mantle heat flow (q^ = 30 mWm"^). It involves the generation of a thermally stable 30 km-thick crust, followed by instantaneous thickening of this crust to 40 km by thrusting, resulting in repetition of the uppermost 10 km section. This has the effect of significantly increasing the internal heat production throughout the crustal section, especially in the upper crust. One of the most obvious advantages of this model is a well-understood mechanism, which can be invoked on the basis of field evidence for large-scale thrusting in many metamorphic belts, including the Halls Creek Orogen. In addition, extremely high temperatures in the lower crust would be likely to trigger extensive partial melting, which may possibly (over a long period of time) produce voluminous granitoids exhibiting temporal geochemical variation. Disadvantages include difficulties in accounting for widespread mafic plutonism, and also granitoids with geochemical signatures indicating a possible mantle source (such as the Mabel Downs Tonalite). Further work is needed to integrate the essential characteristics of the mantle-related processes described in the first model and the crustal thickening simulated by the second model, in order to produce a model that incorporates aspects of both lithospheric detachment or delamination, and crustal thickening via continental collision. There is also a need to establish a more tightly-constrained set of thermal boundary conditions, and to integrate fully the observed geological characteristics of the Halls Creek Orogen (e.g. simulation of the emplacement of intrusives of varying compositions at different crustal levels in the 0-30 million years following the instantaneous large-scale event, to observe the interplay between transient and ambient thermal effects). Although aspects of crustal evolution in the Halls Creek Orogen remain unclear, the substantial presence of (i) HTLP metamorphic rocks; (ii) voluminous 1860-1830 Ma layered mafic bodies, and (iii) 1830-1820 Ma tonalitic and granodioritic plutons with geochemical signatures possibly derived from mantle sources, suggest extensive asthenospheric upwelling occurred, probably as a result of lithospheric removal during continental collision. REFERENCES

Bodorkos, S., Cawood, P.A. and Oliver, N.H.S., 1998. Thermal evolution of the Halls Creek Orogen, W.A.: field, petrological and geochronological constraints. Geological Society of Australia Abstracts, 49, 46. Etheridge, M.A., Rutland, R.W.R. and Wybom, L.A.I., 1987. Orogenesis and tectonic process in the Early to Middle Proterozoic of northern Australia. In: Proterozoic Lithospheric Evolution, ed Kroner, A., Geodynamics Series, 17, pp. 131-147. American Geophysical Union, Washington D.C. Oliver, N.H.S., Bodorkos, S., Nemchin, A.A., Kinny, P.D. and Watt, G.R., in press. Relationships between zircon U-Pb SHRIMP ages and leucosome type in migmatites of the Halls Creek Orogen, Western Australia. Journal of Petrology. Page, R.W., Hoatson, D.M, Sun, S. and Foudoulis, C., 1995a. High-precision geochronology of Palaeoproterozoic layered mafic-ultramafic intrusions in the East Kimberley. AGSO Research Newsletter, 22, 7-8. Page, R.W., Tyler, I.M. and Blake, D.H., 1995b. Geochronology of magmatism and high-grade metamorphism, Kimberley region, W.A. Australian Conference on Geochronology Abstracts, 3, 25. Sheppard, S., Griffin, T.J. and Tyler, I.M., 1995. Geochemistry of felsic igneous rocks from the southern Halls Creek Orogen. Geological Survey of Western Australia Record, 1995/4, 81 pp. Trudu, A. and Hoatson, D., 1996. Depths of emplacement of Precambrian layered intrusions in the East Kimberley, AGSO Research Newsletter, 25, 10-12. Wybom, L.A.L, Wyborn, D., Warren, R.G. and Drummond, B.J., 1992. Proterozoic granite types in Australia: implications for lower crustal composition, structure and evolution. Transactions of the Royal Society of Edinburgh: Earth Sciences, 83, 201-209.

OROGENESIS IN THE OUTBACK

A CONTINENTAL-SCALE TECTONIC MODEL FOR THE REACTIVATION OF PROTEROZOIC SUTURES IN THE MID- TO LATE PALEOZOIC J. Braun^ and R. Shaw^ ^Research School of Earth Sciences, The Australian National University, Canberra, ACT, Jean. Braun@anu. edu. au ^Australian Geological Survey Organisation, Canberra, ACT

The Australian continent has evolved from its Archaean nuclei to its present-day extent through a series of accretion events punctuated by periods of intense tectonic activity during which the sutures between the various accreted lithospheric blocks have been reactivated. Reactivation of the sutures may have been driven by subduction and/or rifting along the continent margins and/or by forces originating within the mantle beneath the continent such as the delamination of a gravitationally unstable continental root or the rise of a mantle plume. The most recent series of large-scale tectonic events to reactivate ancient sutures between the various Australian continent major lithospheric blocks took place in the mid- to Late Palaeozoic and culminated in the Alice Springs Orogeny of central Australia. In this paper, we attempt to analyse the large-scale deformation patterns observed within the Australian continent which formed structures in the Amadeus Transverse Zone and the Fitzroy Trough in Western Australia during the Alice Springs Orogeny and in the Tasman Fold Belt System during equivalent Late Ordovician to Carboniferous events. To this end, we first present the results of a two-dimensional thin-sheet finite element model of a large continent subjected to convergent and divergent velocity boundary conditions along its margins. Deformation within the continent interior is dictated by the presence of cratonic blocks of varying strength, thickness and thermal state. The model predictions can be compared with observations of strain, paleo-topography, basin subsidence, crustal thickness, denudation, geochronology and metamorphic grade of exposed rocks, thereby providing support for the viability of the assumed tectonic model. We also present results from a more local 2D plane-strain fully-coupled thermo-mechanical model of the continental lithosphere which allows for localized deformation. The results show that variations in style along the northern margin of the Amadeus Basin may be explained by variations in the assumed initial geothermal gradient. In the central part of the Amadeus Basin, the lithosphere was cold and responded by movement along a single thrust plane (the Redbank Thrust Zone), whereas towards the east, the lithosphere was characterized by a greater geothermal gradient allowing for the formation of intra-crustal weak layers the upper part of which were exhumed during the latest stages of deformation. In the Fitzroy Trough, a similar tectonic style is observed (thick-skinned re-activation) but, this time, in response to extensional stresses. The model results suggest that the main bounding fault of the rift basin dips in the opposite direction to the assumed lithospheric-scale re-activated structure. The other margin is characterised by an array of listric normal faults activated at different stages during the evolution of the rift and soling into a low-viscosity detachment surface. This asymmetry agrees with the structures imaged in a deep reflection seismic line across the Fitzroy Trough.

A L I C E SPRINGS, AUSTRALIA, 1 9 9 9

THE TIMING OF MULTIPLE(?) HIGH-TEMPERATURE RETROGRADE EVENTS IN THE REYNOLDS RANGE, CENTRAL AUSTRALIA I. S. Buick\ R. Frei^ and I. Cartwright^

^Department of Earth Sciences, La Trobe University, Bundoora, Vic 3083. ^Geologisk Institut, Kobenhavns Universitet, 0ster Voldgade 10, DK'1350 Kobenhavn, ^Department of Earth Sciences, Mon ash University, Clayton, Vic. 3168.

Denmark.

In the Reynolds Range (central Australia), metasediments of the Reynolds Range Group were intruded by voluminous sheet-like -1.78 Ga granites and subsequently regionally metamorphosed from greenschist (-400 " Q to granulite (-750-800 °C) grade at - 5 kbar during the Che wings Orogeny (-1590-1570 Ma). At the highest regional metamorphic grades, granulite-facies metacarbonates, metapelites and metapsammites were partially retrogressed to amphibolite grade in narrow, kilometre-scale, strike-parallel zones after the metamorphic peak. The zones of high-temperature retrogression were channelways for water-rich fluids at, or near, the granite solidus (-650-700 'C). Locally, the retrograde zones record episodes of lower temperature retrogression at -550-600'C. Based on previous SHRIMP U/Pb zircon and monazite studies throughout the Reynolds Range area it appears that peak metamorphic conditions were reached at or around 1594±6 Ma, and that high-temperature (-650-700 "C) retrogression occurred as late as 1568i4 Ma. Constraints from field mapping, stable isotope geochemistry, geochronology and phase equilibria studies are in general agreement that the high-temperature retrograde fluids were exsolved from segregated partial melts that were derived from partially melted metapelite and metagranite. However, it unclear whether the lower-temperature retrogression is related to the same source because the terrain is dissected by mid-amphibolite facies shear zones that may alternatively been the fluid source. These shear zones may have developed during either of two periods: a) during the 1500-1400 Ma Anmatjira uplift phase that involved initial shearing on major through-going crustal structures such as the Redbank Thrust Zone; or b) during the 300-400 Ma Alice Springs Orogeny, which involved south-directed thrusting of Mesoproterozoic metamorphic basement over Neoproterozoic to Palaeozoic intraplate basins in central Australia, and that reactivated some of the earlier Mesoproterozoic shear zones. At Conical Hill (Reynolds Range) Lower Calcsilicate Unit metasediments of the Reynolds Range Group and underlying migmatitic Napperby Gneiss metagranite underwent regional uppermost amphibolite facies- (-680°C to l l O ' C ) , low-pressure metamorphism during the Chewings Orogeny. The Lower Calcsilicate Unit is extensively quartz veined and epidotised, and discordant grossular-andradite (grandite) garnet + epidote quartz veins may be traced over tens of metres depth into pegmatites that pooled at the Napperby Gneiss/Lower Calcsilicate Unit contact. The quartz veins were probably precipitated by water-rich fluids that exsolved from partial melts derived from the Napperby Gneiss during cooling from the peak of regional metamorphism to the wet granite solidus. As noted above, this is a general feature of the Reynolds Range at high grade. Stable isotope data are generally consistent with the retrogression at Conical Hill involving fluids exsolved from melts in the underlying Napperby Gneiss. However, the veins also show evidence of isotopic disequilibrium. Early formed grandite garnet and quartz give oxygen isotopic fractionations that imply minimum temperatures that are far higher (>720-820T) than that of the metamorphic peak. In contrast, epidote-quartz fractionations imply temperature of 560-625'C, in good agreement with petrological constraints on the temperature of epidotisation. The quartz veins contain an early assemblage of grandite gamet-titanite-clinopyroxene±k-feldspar±plagioclase, and a texturally later assemblage characterised by epidote (that replaces garnet and feldspars), with or without minor tremolite (that replaces clinopyroxene). This suggests that the mineral assemblage in the quartz veins either continued to equilibrate with cooling, or that it records a polyphase evolution. Pb stepwise leaching (PbSL) on garnet from three discordant quartz veins yielded comparable single mineral isochrons of 1566 ± 32 Ma, 1576 ± 3 Ma and 1577 ± 5 Ma, which are interpreted as the age of garnet growth in the veins. These dates are in good agreement with previous SHRIMP ages of zircon and monazite formed during hightemperature retrogression elsewhere in the Reynolds Range. The relatively small age difference between peak metamorphism and retrograde veining suggests that partial melting and melt crystallisation controlled fluid recycling in the high-grade rocks. However, PbSL experiments on epidote intergrown with, and partially replacing, garnet in two of the veins yielded isochrons of 1454 ± 34 Ma and 1469 ± 26 Ma. The -100-120 Ma age difference

OROGENESIS IN THE OUTBACK

between intergrown garnet and late epidote from the same vein suggests that the vein systems may have experienced multiple episodes of fluid flow, with die younger ages being related to the Anmatjira Uplift phase.

ALICE SPRINGS, AUSTRALIA, 1999

POLYMETAMORPHISM IN THE CENTRAL ZONE OF THE LIMPOPO METAMORPHIC BELT, SOUTH AFRICA: CONTRAINTS FROM A SHRIMP ZIRCON, MONAZITE AND TITANITE STUDY I. S. Buick\ I. S. Williams^ R . L Gibson^ and J.A. Miller"^

^Department of Earth Sciences, La Trobe University, Bundoora, Vic 3083 ^Researcti School of Earth Sciences, Australian National University, Canberra, ACT 0200 ^Department of Geology, University of the Witwatersrand, Private Bag 3, WITS 2050, South Africa "^Department of Earth Sciences, Monash University, Clayton, Vic. 3168

The Limpopo Metamorphic Belt is situated between the Archaean Kaapvaal and Zimbabwe Cratons in southern Africa. It is a 250 km-long, composite E-W trending metamorphic belt, comprising a Central Zone (CZ) flanked by Northern and Southern Marginal Zones (NMZ and SMZ, respectively). Major shear zones separate the CZ from the NMZ and SMZ (Triangle and Palala Shear Zones, respectively). Both the NMZ and SMZ contain dominant tonalitetrondhjemite-granodiorite intrusive suites and subordinate metasedimentary rocks that are thought to be the granulite-facies equivalents of granite-greenstone sequences of the Kaapvaal (SMZ) and Zimbabwe (NMZ) Cratons. Medium pressure granulite-grade metamorphism in the SMZ occurred at -2.67-2.63 Ga and followed a clockwise retrograde P-T-t path (Stevens and Van Reenen, 1992; Kroner et al, 1999). In the NMZ, granulite-facies metamorphism and synchonous intrusion of chamoenderbites occurred between -2.71 Ga and -2.57 Ga (Kamber et al, 1997). In contrast, the Central Zone is lithologically markedly different. It comprises a sequence of granulitegrade supracrustals (Beit Bridge Group of metapelites, metapsammites, marbles, quartzites and calcsilicate rocks) that are tectonically interleaved with, and intruded by, the - 3 . 0 Ga Messina Layered Mafic Complex and extensive suites of now migmatitic granitic gneisses with inferred instrusive ages of -3.33-3.19 Ga, -2.73-2.60 Ga and -2.582.51 Ga (Kroner et al, 1999). Previous age determinations using evaporation or SHRIMP U/Pb dating of metamorphic zircon, or Pb-Pb stepped leaching of single minerals in granulite-grade assemblages (for example garnet and titanite) have returned ages of -3.24 Ga, -2.57-2.52 Ga and -2.03-2.0 Ga (Holzer et al, 1998; Kroner et al, 1999). Metasediments of the Beit Bridge Group were inferred to have been deposited prior to -3.3 Ga (Kroner et al, 1999). Despite the large number of intrusive/metamorphic events' that have affected the Beit Bridge Group metasediments, they appear to show a relatively simple high-grade metamorphic history. Peak metamorphic conditions in a variety of rock types reached - 8 5 0 " C at - 1 0 kbar and were followed by near-isothermal decompression to - 7 0 0 - 8 0 0 ^ and - 3 - 5 kbar (Harris & Holland, 1984; Droop, 1989). This clockwise P-T-t path was originally ascribed to Archaean tectonics (Harris & Holland, 1984), but has more recently been wholly attributed to -2.03-2.0 Ga Proterozoic re-working of the terrain (Kroner et al, 1999). Several phases of low-temperature retrogression (generally T<600'C, and commonly - 3 0 0 ' C ) are semi-pervasive throughout the CZ, and their absolute timing and fluid sources are unknown. In this study, we present new SHRIMP U/Pb constraints from zircon, monazite and titanite for the timing of deposition of protoliths to the Beit Bridge Group, polyphase high-grade metamorphism and retrogression. SHRIMP U/Pb age determinations were obtained from: 1) zircon and monazite from three granulite-facies metapelites along the length of the CZ- all metapelites contain the peak-metamorphic assemblage Grt-Sill-BtQtz±Ksp±Pl and show evidence of decompression to form late cordierite±spinel; 2) monazite and zircon from a metapsammitic gneiss interlayered with one of these metapelites; 3) zircon occurring in a garnet-bearing discordant leucosome that occurs in concordant metabasic layers within the Beit Bridge Group (W of Messina); 4) titanite occurring in medium-temperature, retrograde calcite-diopside-tremolite-grossular garnet veins that cross cut metabasic gneisses (East of Messina); and 5) titanite in interlayered marbles and calcsilicate rocks that are deformed within amphibolite-facies shear zones in the easternmost exposures of the CZ. All three metapelites, the metapsammite and the discordant leucosome in the metabasic granulite, contain a population of low Th/U metamorphic zircon with ages of -2.02 Ga -2.00 Ga as overgrowths on detrital or inherited zircon grains, thus confirming the importance of the - 2 . 0 Ga event along the length of the CZ. In the case of two of the metapelitic granulites, these metamorphic overgrowths occur on detrital zircons with a range of ages similar to those already determined elsewhere in the CZ (that is, greater than - 3 . 5 Ga). However, one metapelitic granulite (west of Messina) with the -2.02 Ga metamorphic zircon population additionally contains a main population of high Th/U zircon with igneous zonation and chemistry that is concordant at - 2 . 6 9 Ga. This age falls within a major period of granitoid emplacement in the CZ. While this could conceivably be a sample of metamorphosed, extremely

OROGENESIS IN THE OUTBACK

aluminous granite, such a relationship appears unlikely based on field evidence. Therefore, we tentatively interpret the -2.69 Ga zircon population as being detrital, thus suggesting that the precursors to some metapelitic sediments were deposited on metamorphic basement post-2.69 Ga. Zircon cores overgrown by -2.2 Ga metamorphic zircon from the partially melted metatabasite retain igneous zoning and Th/U chemistry, and yield a concordant age at -2.65 Ga that probably dates emplacement of the igneous protolith. The -2.2 Ga overgrowths in the metabasite are themselves overgrown by low Th/U, discordant rims at -1.93 Ga. Monazite from all three metapelitic granulites and one metapsammitic granulite yield concordant ages of -2.02-2.03 Ga that are similar to metamorphic zircon from the same rocks. In addition, one metapelitic granulite (east of Messina) contains a second, older monazite generation dated at -2.64 Ga. Zircon from the same rock contains two generations of metamorphic zircon, at -2.66 Ga and 2.57 Ga, that are overgrown by -2.02 Ga metamorphic zircon, thus indicating that this sample experienced multiple high-grade metamorphic events, even though petrographically the sample shows the same simple textural evolution as for other samples that is ascribed wholly to -2.0 Ga reworking of the CZ. High-grade metamorphism at -2.64-2.66 Ga has not previously been documented in the CZ, but is similar to the age of granulite-facies metamorphism in the SMZ. These data suggests that the Cz and marginal zones at least partly share an Archaean high-grade metamorphic history. Titanite from a late retrograde tremolite vein that cross cuts partially melted mafic granulite and in an amphibolitefacies shear zone within layered marble and calcsilicate rock yield ages within error of those already obtained from zircon and monazite, suggesting that the CZ cooled relatively quickly after the peak of the -2.2-2.0 Ga event. REFERENCES

Droop, G.T.R., 1989. Reaction history of garnet-sapphirine granulites and conditions of Archaean high-pressure granulite-facies metamorphism in the Central Zone Limpopo Mobile Belt, Zimbabwe. Journal of Metamorphic Geology, 7, 383-403. Harris, N.B.W., and Holland, T.J.B., 1984. The significance of cordierite-hypersthene assemblages from the Beitbridge region of the central Limpopo Belt: evidence for rapid decompression in the Archaean? American Mineralogist, 69, 1036-1049. Holzer, L., Frei, R, Barton, J.M. and Kramers, J.D., 1998. Unravelling the record of successive high grade events in the Central Zone of the Limpopo Belt using Pb single phase dating of metamorphic minerals. Precambrian Research, 87, 87-115. Kamber, B.S., Bino, G.G., Wijbrans, J.R., Davies, G.R., and Villa, I.M., 1997. Archaean granulites of the Limpopo belt, Zimbabwe: one slow exhumation or two rapid events. Tectonics, 15, 1414-1430. Kroner, A., Jaekel, P., Brandl, G., Nemchin, A.A. and Pidgeon, R.T., 1999. Single zircon ages for granitoid gneisses in the Central zone of the Limpopo Belt, Southern Africa and geodynamic implications. Precambrian Research, 93, 299-337. Stevens, G. and Van Reenen, D.D., 1992. Partial melting and the origin of metapelitic granulites in the southern Marginal zone of the Limpopo belt. South Africa. Precambrian Research, 55, 303-319.

ALICE SPRINGS, AUSTRALIA, 1999

POLYPHASE METAMORPHISM AND REACTIVATION IN THE REYNOLDS-ANMATJIRA RANGE AREA, NORTHERN ARUNTAINLIER I.S. Buick\ & M. Hand^ ^Dept of Earth Sciences and VIEPS, La Trobe University, Bundoora, VIC, 3083, AUSTRALIA ^Dept of Geology and Geophysics, University of Adelaide, Adelaide, S.A. 5001, AUSTRALIA

The Reynolds and Anmatjira Ranges occur in the northern tectonic province of the Arunta Inlier, and show a complex history of polymetamorphism and reactivation over a period of 1500 Ma. The ranges mainly comprise two lithostratigraphic successions: the pelitic to psammitic Lander Rock Beds (LRB; deposited at or before -1840 Ma) and their equivalents, which form local basement; and the shallow water sediments (pelite, carbonate, quartzite) of the-1800-1780 Ma Reynolds Range Group (RRG), which form a local cover sequence. These two sequences are separated by a disconformity, and locally an angular unconformity. The earliest metamorphic history pre-dates deposition of the Reynolds Range Group, and involves greenschist-facies deformation in the LRB that was overprinted by contact metamorphism around a suite of megacrystic granites emplaced at -2.5 kbar during the -1820-1800 Ma Mt. Stafford Tectonic Event (Vernon et al, 1990). The timing of the early greenschist-facies event is not known, but is likely to be younger than -1840 Ma, based on constraints from U/Pb ages of detrital zircons in the LRB (Vry et al, 1996). P-T-t paths associated with this event appear to be counterclockwise and involved a component of near-isobaric cooling (Collins & Vernon, 1990). Soon after deposition of the RRG multiple granitoids intruded both basement and cover at -1780 Ma during the Strangways Orogeny (Collins & Williams, 1995). In the RRG, contact metamorphic blasts (andalusite, cordierite) overprint a greenschist-facies fabric that formed earlier during the Strangways Orogeny. In the Anmatjira Range, the -1784 Ma Possum Creek charnockite (Collins & Williams, 1995) truncates leucosomes in partially melted, low- to medium-pressure (-5.5 kbar) granulite-facies metapelites, suggesting that the Strangways Orogeny was characterised by rapid lateral thermal gradients. Metamorphic assemblages formed during the Strangways Orogeny were variably overprinted during the -15901570 Ma Chewings Orogeny, which resulted in a continuous greenschist- to granulite-facies transition (-400^0 to 800'C at -5 kbar) along each range (Dirks et al, 1991; Hand, 1995). Mineral assemblages developed during the Chewings Orogeny are axial planar to tight sub-vertical folds that trend NW to SE along the ranges, and which are locally refolded into reclined orientations by a series of conjugate crenulations with kink-like geometries. At the highest grades these crenulation zones were the locus of melt segregation in migmatitic metapelites and metagranites, and subsequent chaneled retrogression in overlying carbonates where these melts finally crystallised (Buick etal, 1998). Along the length of the Reynolds Range the Chewings-age overprint on Strangways-age assemblages resulted in a range of apparent P-T vectors, from cooling (at low Chewings grade) to heating (at high Chewings Grade). The lack of map-scale fold interference patterns between Strangways- and Chewings-age structures suggests that Chewingsage deformation may have been essentially coaxial with Strangways-age structures. In the Anmatjira Range, where the Chewings-age overprint reached granulite grade and overprinted slightly higher temperature assemblages formed during the Strangways Orogeny, an apparent P-T vector of cooling resulted (Hand et al, 1992). However, petrological evidence suggests that the Chewings overprint is a prograde feature (Hand, 1995) and that granulitefacies metamorphism in both events was followed by limited decompression (Hand et al, 1992). Limited decompression (1-1.5 kbar) followed by cooling (a clockwise P-T-t vector) has been documented in both the LRB and RRG in the adjacent Reynolds Range after the peak of Chewings-age metamorphism (Vry & Cartwright, 1994; Buick etal, 1998). Tectonic models for regional scale metamorphism in the Reynolds and Anmajira Ranges have been the subject of considerable debate. Based on early geochronology Collins & Vernon (1990) and Collins et al (1995) suggested that metamorphism was transient and synchronous with -1780 Ma granite emplacement ie during the Strangways Orogeny. However, the more recent SHRIMP studies of Williams et al (1996), Vry et al (1996) and Rubatto et al (1999, this volume) have shown that : a) metamorphism occurred at -1590-1570 Ma (the Chewings Orogeny) and therefore could not be related to emplacement of these granites; and b) the RRG stayed above the granite solidus for

OROGENESIS IN THE OUTBACK

-26 Ma and initially cooled slowly (Williams et ai, 1996). Hand et al (1995) and Sandiford& Hand (1998) have noted that granites in this terrain are anomalously enriched in heat producing elements and have suggested that regional Chewings-age metamorphism resulted from burial of the -1820-1800 Ma and -1780 Ma high heat producing granites. In this case the granites, while not being the direct cause of metamorphism, have provided a long-lived thermal contribution during subsequent orogenesis. The Reynolds and Anmatjira Rranges are dissected by a series of NW-SE and E-W trending shear zones whose timing has also been subject to debate, and which define a "pop-up" structure (Collins & Teyssier, 1989). Throughout the area, medium-temperature chronometers (Rb/Sr muscovite; "^^Ar-^^ArHbde) typically record ages of -1400-1500 Ma (Collins et al, 1995; Buick et al, 1999 & unpublished data) that may reflect initial exhumation of the terrain during the Anmatjira Uplift Phase, as is recorded further to the south e.g. along the Redbank Thrust Zone. The metamorphic grade of the shear zones shows a similar distribution to that for the Chewings Orogeny, and increases from greenschist to medium-P amphibolite-facies (4-6 kbar at -550-600'C; Dirks et al, 1991; Hand, 1995) from NW to SE along the length of the ranges. However, where dated the shear zones have so far returned ages of -330 Ma and apparently formed during the Alice Springs Orogeny (Cartwright et al, 1999). It remains unclear as to what extent these shear zones reactivated Mesoproterozoic structures, and why their grade distribution mimics that of the Chewings Orogeny. REFERENCES

Buick, I.S., Cartwright, 1. & Harley, S.L., 1998. The retrograde P-T-t path for low-pressure granulites from the Reynolds Range, central Australia: petrological constraints and implications for low-P/high-T metamorphism. Journal of Metamorphic Geology, 16, 511-529. Buick, I.S., Frei, R. & Cartwright, I. 1999. The timing of high-temperature retrogression in the Reynolds Range, central Australia: constraints from single mineral Pb-Pb dating. Contributions to Mineralogy and Petrology, 135, 244-254. Cartwright, I., Buick, I.S., Foster, D.A. & Lambert, D.D., 1999. Alice Springs age shear zones from the southeastern Reynolds Range, central Australia. Australian Journal of Earth Sciences, 46, (in press). Collins, W.J. & Vernon, R., 1991. Orogeny associated with anticlockwise P-T-t paths: Evidence from low-P, high-T metamorphic terranes in the Arunta inlier, central Australia. Geology, 19, 835-838. Collins, W.J. & Teyssier, C., 1989. Crustal-scale ductile fault systems in the Arunta Inlier, central Austraha. Tectonophysics, 158,49-66. Collins,WJ. & Williams, I.S., 1995. SHRIMP ionprobe dating of short-lived Proterozoic tectonic cycles in the northern Arunta Inlier, central Australia. Precambrian Research , 71, 69-89. Collins, W.J., Williams, I.S., Shaw, S.E., & McLauglin, N.A., 1995. The age of the Ormiston Pound Granite: implications for Mesoproterozoic evolution of the Arunta Inlier, central Australia. Precambrian Research, 71, 91-105. Dirks, P.H.G.M., Hand, M. & Powell, R. 1991. The P-T deformation path for a mid-Proterozoic, low-pressure terrane: the Reynolds Range, central Australia. Journal of Metamorphic Geology, 9, 641-661. Hand, M., 1995. Structural and metamorphic evolution: Anmatjira Range and Jetty Peninsula Ph.D. Thesis (University of Melbourne, unpublished). Hand, M., 1996. Does a transient view help or hinder the interpretation of reaction textures in LP/HT rocks? Geological Society of Australia Abstracts, 42, 27-28. Hand, M., Dirks, P.H.G.M., Powell, R. & Buick, I.S., 1992. How well established is isobaric cooling in Proterozoic orogenic belts? An example from the Arunta inlier, central Australia. Geology, 20, 649-652. Sandiford, M. & Hand, M., 1998. Australian Proterozoic high-temperature, low-pressure metamorphism in the conductive limit. In: Treloar, P.J. & O'Brien, P.J. (eds) What Drives Metamorphism and Metamorphic Reactions? Geological Society of London Special Publication, 138, 109-120. Vernon, R.H., Clarke, G.L. & Collins, W.J., 1990. Local, mid-crustal granulite facies metamorphism and melting: an example in the Mount Stafford area, central Australia. In: High Temperature Metamorphism and Crustal Anatexis (eds Ashworth, J.R. & Brown, M.), pp. 272-315. Unwin & Hyman, London. Vry, J.K. & Cartwright, I., 1994. Sapphirine-komerupine rocks from the Reynolds Range, central Australia: constraints on the uplift history of a Proterozoic low pressure terrain. Contributions to Mineralogy and Petrology ,116, 78-91. Vry, J., Compston, W., and Cartwright, I. (1996). SHRIMP II dating of zircons and monazites: reassessing the timing of high-grade metamorphism and fluid flow in the Reynolds Range, northern Arunta Block, Australia. Journal of Metamorphic Geology, 14, 335-350. Williams, I.S., Buick, I.S. & Cartwright, I., 1996. An extended episode of early Mesoproterozoic metamorphic fluid flow in the Reynolds Range, central Australia. Journal of Metamorphic Geology , 14, 29-48.

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

EPISODIC, CHANNELLED FLUID FLOW FROM INTERNAL AND EXTERNAL SOURCES: REYNOLDS RANGE, CENTRAL AUSTRALIA I. Cartwright\ I. S. Buick^ J. K. Vry^ ^Department of Earth Sciences, Monash University, Clayton Vic. 3168, Australia ^School of Earth Sciences, La Trobe University, Bundoora Vic. 3083, Australia ^School of Earth Sciences, PO Box 600, Victoria University, Wellington, New Zealand An electronic version of this poster may be found at: http://www.earth.nionash.edu.au/-icart/central/text.html

Constraining crustal fluid-rock interaction is important for many reasons: (1) infiltration of large volumes of fluid may cause significant metasomatism; (2) fluid flow may transfer heat; (3) fluids may promote partial melting; and (4) fluids affect the rheology of the crust and thus can control deformation. However, due to data requirements, individual episodes of fluid flow are often considered in isolation, and there have been few studies of the history of fluid-rock interaction within an individual terrain. This is contrast to structural or metamorphic studies that tend to emphasize evolution with time. Here, we present a time-integrated study of fluid-rock interaction in the Reynolds Range, central Australia, and link episodes of fluid flow with the tectonometamorphic evolution of the terrain. The Reynolds Range is an ideal terrain for such a study as it contains a wide variety of lithologies that may be traced from low to high metamorphic grade. Thus, many often contentious questions, such as whether fluid-rock interaction in uniformly high-grade metamorphic terrains occurred during or prior to regional metamorphism, may be addressed. TECTONOMETAMORPHIC HISTORY Granites emplaced into the Lander Rock Beds caused contact metamorphism at 1.82 Ga (Collins & Williams, 1995). Following the deposition of the Reynolds Range Group sediments there was a further period of granite intrusion and contact metamorphism at -1.78 Ga. The main regional metamorphism to affect the Reynolds Range, M2, was at --1594 Ma (Vry et al., 1996; Williams et al., 1996). M2 occurred at 400-500 MPa and temperatures varied from --400 °C in the NW to --740140 °C in the SE. Fluids were exsolved from crystallising partial melts until approximately 30 m.y. after the thermal peak (Williams et al., 1996). A regionally extensive S2 fabric was developed during M2. FLUID FLOW DURING 1.82 GA CONTACT METAMORPHISM Lander Rock Bed metapsammites, and metapelites underwent low-grade (<500 °C) contact metamorphism at -250 MPa around S-type granites at 1.82 Ga. values of Lander Rock Beds range from 13.4±0.8%o to as low as 6.7%o adjacent to some larger plutons (Vry & Cartwright, 1998). These data suggest that these rocks were affected by contact metamorohic fluid flow. Rare Mg-Al-rich pods of sapphirine-bearing rocks within the Lander Rock Beds also have low O values (4.0±0.7%o). These pods may represent low-temperature diagenetic-hydrothermal deposits of Mg-rich chlorite formed at that time. The fluids at this stage probably involve a combination of metamorphic and igneous fluids; however, the low values of the Mg- and Al- rocks suggest some involvement of meteoric water. FLUID FLOW DURING L78 GA CONTACT METAMORPHISM Grandite-rich layers in calcareous rocks of the Lower Calcsilicate Unit of the Reynolds Range Group were formed by the infiltration of water-rich (XCO2 <0.3) fluids 1.78 Ga associated with the emplacement of the Napperby Gneiss at 1.78 Ga. The stable isotope values of calcite = -4.2 to -0.8%o, = 10.5-14.0%o) and the silicate fraction = 6.1-10.8%o) of the grandite-rich layers are most consistent with the infiltrating fluid being from a magmatic source. Locally Lower Calcsilicate Unit metasediments and the Napperby Gneiss at their mutual contact values of the have values as low as 2%o that imply they were infiltrated by meteoric water at 1.78 Ga. The metapelites from the Reynolds Range Group vary between 4.4 and 15.3%o. There is heterogeneity in values at all metamorphic grades of some 7-8%o, with the lower values commonly, but not always, from rocks adjacent to 1.78 Ga granites. This association implies that fluid flow during contact metamorphism reset stable isotope ratios in the metapelites. The lowest metapelite values are lower than those typical of the granites in this terrain, implying that, as in the Lower Calcsilicate Unit, fluid flow involved magmatic-meteoric fluids. FLUID FLOW DURING REGIONAL METAMORPHISM The 1.6 Ga regional metamorphism was a period when little significant pervasive fluid flow occurred. There are a number of lines of evidence for low volumes of fluids at this time: (1) ranges of stable isotope ratios in all rock units are largely independent of metamorphic grade (Buick & Cartwright, 1996); (2) stable isotope ratios produced by fluid flow during contact metamorphism are preserved through the regional event; (3) marbles define a trend of

OROGENESIS IN THE OUTBACK

increasing XCO2 values from <0.05 at greenschist facies to 0.7-1.0 at granulite facies; (4) melting of the granulite facies gneisses occurred by fluid absent reactions. FLUID F L O W DURING COOLING F R O M R E G I O N A L M E T A M O R P H I S M

Marbles of the Upper Calcsilicate Unit of the Reynolds Range Group locally contain wollastonite-, grossular-, and humite-bearing mineral assemblages together with a variety of high-variance skams. Mineral assemblages yield XCO2 values of <0.1, implying that water-rich fluids infiltrated these rocks. The source of these fluids was decimetre- to metre-thick, coarse-grained aluminous pegmatites that were derived by melting of underlying pelites during regional metamorphism and quartz veins. These pegmatites and veins cut, and hence postdate, the regional S2 fabric, and yield SHRIMP ages of 1589-1568 Ma (Williams et al., 1996). Infiltrated marbles have and values of 8 to 24%o and -7 to 2%o, which are lower than their unretrogressed equivalents. The lower and values approach equilibrium with the pegmatites. The occurrence of skarns and zones of massive wollastonite imply that fluid volumes were locally sufficient to cause metasomatism. Metapsammites in the Upper Calcsilicate Unit also show evidence for undergoing fluid-rock interaction at this time. These lithologies are interlayered with, and pass along strike into, quartz-poor anthophyllite or cummingtonite-bearing gneisses. In these gneisses, the amphiboles are typically randomly oriented and grow in rosettes; however, a relic S2 foliation is locally preserved as inclusion trails of quartz and ilmenite in cordierite. Metapsammitic rocks have values of 7- 13 %o again consistent with fluids being derived from the aluminous pegmatites. Fluids exsolved from partial melts within the Napperby Gneiss formed discordant quartz + garnet + epidote veins within the Lower Calcsilicate Unit and also caused extensive epidotisation that dies out over a few metres away from the contact with the Napperby Gneiss. PbPb ages of 1576 Ma from garnets and epidote in the veins (Buick et al., in press) confirm that fluid flow occurred during the early stages of cooling. M E T E O R I C F L U I D F L O W IN A L I C E S P R I N G S S H E A R Z O N E S

Sheared metagranites deformed under greenschist to amphibolite-facies conditions in the Alice Springs Orogeny at 334 Ma (Cartwright et al., 1999) have values as low as 0%o that are much lower than those of their unsheared counterparts ( 7 . 0 - 9 . 3 % o ) . The calculated fluid values ( - 1 to - 2 % o ) lie within the range of meteoric fluids, implying ingress of fluids from the Earth's surface at that time. Meteoric fluid flow is also recorded in Alice Springs age shear zones from the Anmatjira Ranges and the Redbank area (Read & Cartwright, this volume). TIME-INTEGRATED FLUID FLUXES AND FLUID SOURCES

Silica addition within Alice Spring shear zones at Sandy Creek yields time-integrated fluid fluxes of -10^ m^/m^. High-temperature retrogression in the marbles reset mineralogy and the stable isotopes over 10-200 m, implying time-integrated fluid fluxes of up to 600 mVm^. Silica metasomatism in these rocks, which occurs on smaller (1-20 m) lengthscales requires time-integrated fluid fluxes of --100-1000 mVm^. Fluid flow in the Reynolds range involved both internal (igneous fluids during contact metamorphism, high-temperature retrogression adjacent to pegmatites) and external (influx of meteoric water) sources. Fluid flow generally occurred as a response to the numerous tectonic episodes that were recorded by this terrain. REFERENCES

Buick, I.S., Cartwright, L, & Williams, I.S. 1997. High-temperature retrogression of granulite-facies marbles from the Reynolds Range. Journal of Petrology, 38, 877-910. Buick, I.S., Frei, R. & Cartwright, I., 1999. The timing of high-temperature retrogression in the Reynolds Range, central Australia: constraints from single mineral Pb-Pb dating. Contributions to Mineralogy and Petrology, in press. Cartwright, I., Buick, I.S., Foster, D.A., & Lambert, D.D., 1999. Alice Springs age shear zones from the Reynolds Range, central Australia: implications for regional tectonics. Australian Journal of Earth Sciences, in press. Cartwright, I., Buick, I.S. & Vry, J.K., 1996. Polyphase metamorphic fluid flow in the Lower Calcsilicate Unit, Reynolds Range, central Australia. Precambrian Research, 77, 211-229. Collins, W.J. & Williams, I.S., 1995. SHRIMP ion probe dating of short-lived Proterozoic tectonic cycles in the northern Arunta Inlier, central Australia. Precambrian Research, 71, 69-89. Vry J.K. & Cartwright I., 1998. Stable isotopic evidence for early fluid infiltration in a multiply-metamorphosed terrane: the Reynolds Range, Arunta Block, central Australia. Journal of Metamorphic Geology, in press. Vry, J., Compston, W. & Cartwright, L, 1996. SHRIMP II dating of zircons and monazites: reassessing the timing of high-grade metamorphism and fluid flow in the Reynolds Range, northem Arunta Block, Australia. Journal of Metamorphic Geology, 14, 335-350. Williams, I.S., Buick, I.S. & Cartwright, I., 1996. An extended episode of early Mesoproterozoic metamorphic fluid flow in the Reynolds Range, central Australia. Joumal of Metamorphic Geology, 14, 29-47.

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

METAMORPHIC AND TECTONIC EVOLUTION OF THE PALEOPROTEROZOIC GRANULITES IN THE NORTHEASTERN GYEONGGI MASSIF, SOUTH KOREA M. Cho\ S. R. Lee' ', K. Yi' and R. Stern'

^Department of Geological Sciences, Seoul National University, Seoul, 151-742, Korea ^Geological Research Division, Korea Institute of Geology, Mining and Materials, Taejon, 305-350, Korea ^J.C. Roddick Ion Microprobe Laboratory, Geological Survey of Canada, 601 Booth Street, Ottawa, Canada K1A 0E8

A coherent granulite complex is newly found in the Hwacheon area, northeastern margin of the Gyeonggi massif, South Korea, formerly regarded as the eastern part of the Imjingang belt. This Hwacheon granulite complex (HGC) consists mainly of pelitic migmatitic granulites and leucogranites together with minor mafic granulites and amphibolites. Mineral assemblages and reaction textures indicate that the HGC has experienced five stages of metamorphic events: pre- (Ml) and peak (M2) granulite-facies metamorphism; retrogressive, high (M3) and low (M4) pressure metamorphism of the upper amphibolite facies; and local retrogression (M5) producing andalusitebearing assemblages. Rare inclusions of kyanite and staurolite in M2 garnet of pelitic granulites suggest a crustal thickening process operative prior to M2. The M2 granulite-facies metamorphism at 7.2-8.8 kbar and 785-840°C induced a widespread partial melting in pelitic rocks and produced syn- to post-tectonic, (para-)autochthonous leucogranites. At the waning stage of M2, cordierite together with secondary garnet and hercynite formed in some pelitic granulites, suggesting a decompression to 6.1-6.5 kbar at 755-765°C prior to further cooling. Partial hydration of the M2 minerals produced M3 assemblages, characterized by the growth of secondary kyanite and coronitic garnet in pelitic and mafic granulites, respectively. The P-T conditions of M3 are estimated to be 7.2-8.4 kbar and 690-700°C, and suggest a quasi-isobaric cooling between M2 and M3. The clockwise P-T path for Ml to M3 stages is accounted for by crustal thickening and associated partial melting. It is also likely that, after the peak metamorphism, there is some residence time at the lower crustal level prior to the exhumation of the granulite complex. The timing of peak metamorphism was dated from the unzoned overgrowth rims of zircon in a migmatitic granulite to be 1872-1-11/-9 Ma, using the SHRIMP ion microprobe. Detrital cores of zircon, however, give diachronous UPb ages of ca. 2300, 2450, 2650 and 2900 Ma, attesting to the presence of Archean protoliths that has been previously unrecognized in South Korea. Moreover, the apatite age identical within error range to the zircon age suggests a rapid cooling of the HGC after the granulite-facies melting. The decompressional exhumation of HGC can be inferred from the growth of M4 cordierite mantling both M2 and M3 minerals such as garnet and kyanite in pelitic granulites. P-T estimates of M4 (4.5-5.3 kbar and 635-700°C) suggest a quasi-isothermal decompression from M3 to M4. Such an isothermal decompression was also reported in the Chuncheon amphibolite approximately 30 km south of the HGC (Lee and Cho, 1995). The age of exhumation during M4 is not precisely determined yet, but it probably corresponds to the Permo-Triassic time, as revealed by "^^Ar /^^Ar ages of hornblende (226 ± 8 Ma) from the garnet amphibolite and muscovite (202 ± 4 Ma) from the gneissose pegmatite. The integration of our P-T estimates suggests that the evolution of HGC is characterized by two distinct tectonometamorphic cycles. Widespread occurrence of retrogressive kyanite in aluminous granulites suggests that the granulite complex has been deeply seated prior to granulite-facies metamorphism and subsequent anatexis, and resided at the mid-crustal level after the Paleoproterozoic orogeny. Another tectono-metamorphic event during the Permo-Triassic time may be responsible for the final exhumation of the granulite complex toward the surface. Therefore, the granulite complex in the east-central part of Korean Peninsula could be considered as a Paleoproterozoic crustal basement that was reactivated during the Permo-Triassic orogeny. The age of reactivation of the granulite complex is coeval with the continental collision between the North and South China Blocks in China. Thus, it is likely that northern margin of the Gyeonggi massif together with the Imjingang belt may form a suture zone extending from the Chinese collision belt. REFERENCES

Lee, S.R. and Cho, M., 1995. Tectonometamorphic evolution of the Chuncheon amphibolite, central Gyeonggi massif. South Korea. Journal of Metamorphic Geology, 13, 315-328.

O R O G E N E S I S IN T H E O U T B A C K

GEOCHRONOLOGICAL CONSTRAINTS FOR A TWO-STAGE HISTORY OF THE ALBANY-FRASER OROGEN, W A D. J. Clark^ B. J. Hensen^ & P. D. Kinny2 Department of Applied Geology, University of NSW, Sydney 2052. ^TSRC, School of Applied Geology, Curtin University of Tectinology, GPO Box U1987 Perth 6845.

The Albany-Fraser Orogen (Myers, 1990) is an arcuate orogenic belt situated along the southern and southeastern margins of the Archaean Yilgam Craton in Western Australia (Fig. 1). Previous U-Pb zircon geochronological studies have bracketed the major period of tectonothermal activity to between c.1300 and 1100 Ma (e.g. Pidgeon, 1990; Black et al., 1992; Nelson et al., 1995). We present new geochronological data (SHRIMP U-Pb zircon, monazite and rutile) from samples with well-defined relationships to structural and metamorphic events recognised in the eastern Albany-Fraser Orogen, which suggest that the Mesoproterozoic Albany-Fraser Orogeny comprised two discrete thermo-tectonic episodes: Stage 1 (c.1345-1280 Ma); and Stage II (c.1214-1140 Ma). Stage I orogenesis resulted in the upper-amphibolite to granulite facies metamorphism of the Malcolm Gneiss (Mi: T=750-800'C, P=4-7kbar), and pervasive deformation during three structural events (D1.3). Abundant c.1330-1314 Ma (Nelson et al., 1995) monzogranitic plutons (Recherche Granite) intruded after the Dj structural event, and were strongly foliated and folded during D2 and D3 respectively. D3 is associated with NW-SE horizontal shortening, and occurred prior to the 1313±16 Ma (zircon age) intrusion of linear aplite dykes. Mj metamorphism accompanied and outlasted Stage I deformation and plutonism. Subsequent to Stage I, the orogen underwent a period of exhumation and erosion, resulting in the deposition of sediments in shallow intracratonic basins near Mount Ragged. This interpretation is supported by the discovery of a 1321 ±24 Ma (n=12, chi-squared = 1.7) population of detrital zircons from Mount Ragged. The detrital zircons demonstrate that these rocks unconformably overly, and were derived from, Stage I basement. During this period, sedimentation was accompanied by the scattered intrusion of mafic dykes into the basement rocks. Stage II commenced deformation (D4a) and granulite facies metamorphism (M2a: T>800'C, P=5-6kbar) in the Salisbury Gneiss terrain. A zircon crystallisation age from a syn-D4a metasedimentary anatectic leucosome provides a lower age limit of 1214±8Ma for peak M2a conditions. A second generation of metamorphic zircon formed in these rocks during the exhumation of the terrain at 1182±13Ma. M2a metamorphism is not recognised in the Malcolm Gneiss or in the Mount Ragged metasedimentary rocks. A distinct lower amphibolite facies metamorphism (M2b: T=530-600"C, P=4kbar) overprints Mj assemblages in the Malcolm Gneiss, and resulted in the formation of metamorphic rutile in the Mount Ragged metasedimentary rocks. The 1154±15Ma age of these rutiles provides a minimum estimate for the time of burial and metamorphism of the Mount Ragged metasedimentary rocks. It is proposed that burial (D4b) and metamorphism (M2b) of the Mount Ragged metasedimentary rocks, and the overprinting of the Malcolm Gneiss, occurred as a consequence of overthrusting by high grade, early-Stage II rocks (including the Salisbury Gneiss terrain) along an inferred NE-SW trending structure (the Rodona Fault). Deformation relating to the overthrusting continued until at least 1140±6Ma (monazite age on syn-D4b pegmatite). The final event attributed to Stage II is the emplacement of c.ll40Ma (Nelson et al., 1995) post-kinematic monzogranite plutons (Esperance Granite) throughout the terrain. The two-stage thermo-tectonic history of the Albany-Fraser Orogen correlates with adjacent Mesoproterozoic orogenic belts in Australia, and the Windmill Islands to Hunger Hills region of East Antarctica, suggesting that Mesoproterozoic Australia assembled in two stages. Our work supports a model, previously proposed by Myers et al., (1996), in which the three Australian pre-Mesoproterozoic continents (the North, West and South Australian cratons) collided to form a single continent at c.1300 Ma, and subsequent reactivation affecting basement and cover between c.1200 and 1100 Ma was intracratonic in nature. Two comparable and broadly contemporaneous compressional orogenies controlled the formation of the Kibaran Belt in Africa (e.g. Kampunzu, 1997) and the Grenville Belt in Canada (e.g. McLelland et al, 1996), suggesting that tectonic events in Mesoproterozoic Australia follow a similar pattern to that recognised for Rodinia amalgamation worldwide.

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

REFERENCES Black, L. P., Harris, L. B. & Delor, C. P., 1992. Reworking of Archaean and Early Proterozoic components during a progressive, Middle Proterozoic tectono-thermal event in the Albany Mobile Belt, Western Australia. Precambrian Research, 59, 95-123. Kampunzu, A., B., 1997. The Kibaran Orogenic Cycle (ca. 1400-1000 Ma): review and new perspectives in the framework of Rodinia reconstruction. Proterozoic Geology of Madagascar, Program with 5, 34. McLelland, J., Daly, J. S. & McLelland, J. M., 1996. The Grenville Orogenic Cycle (ca. 1350-1000 Ma): an Adirondack perspective. Tectonophysics, 265, 1-28. Myers, J. S., 1990. Albany-Fraser Orogen. Geological Survey of Western Australia Memoir, 3, 255-264. Myers, J. S., 1995. Geology of the Esperance 1:1 000 000 sheet. Western Australian Geological Survey, 1:1 000 000 Geological Series Explanatory Notes, lOp. Myers, J. S., Shaw, R. D. & Tyler, I. M., 1996. Tectonic evolution of Proterozoic Australia. Tectonics, 15, 14311446. Nelson, D. R., Myers, J. S. & Nutman, A. P., 1995. Chronology and evolution of the middle Proterozoic AlbanyFraser Orogen, Western Australia. Australian Journal of Earth Science, 42, 481-495. Pidgeon, R. T., 1990. Timing of plutonism in the Proterozoic Albany Mobile Belt, southwestern Australia. Precambrian Research, 47, 157-167.

1 AMO O

Mt Rawed J ?^ MtRagged

Pcnii 4

34'3(r-

YILGARN CRATON

/ I \ | Malcolm Gneiss Salisbury Gneiss • mam ™ • I

Recherche Granite (c.l300Ma) Bspcrance Granite (c.ll40Ma)

J / /

^ Cape Pasicy 34'-

Caix- Arid

.Mount Ragged metased. rix:ks I Cainozoic cover Major structure (inferred)

^^y * Pi Malcoln

MiddlcU.

30km

—L-J— Fig. 1 Main geological features of the study area, within the Nornalup Complex of the eastern Albany-Fraser Orogen (modified after Myers, 1995). The middle inset shows the three major tectonic units recognised in the Albany-Fraser Orogen (Myers, 1995). Abbreviations are: Biranup Complex (BC); Fraser Complex (FC); Nornalup Complex (NC). 123'

124*

20 OROGENESIS IN THE OUTBACK

POLYSTAGE DUCTILE THRUSTING AND EXTENSION IN THE LOWER OROGENIC CRUST: WESTERN GRENVILLE PROVINCE, GEORGIAN BAY, ONTARIO N.G. Culshaw\ R.A. J a m i e s o n \ J.W.F. Ketchum^ N. Wodicka^ D. Corrigan^ & P.H. Reynolds^ ^Department of Earth Sciences, Dalhousie University, Halifax, NS, Canada, B3H 3J5 ^Department of Earth Sciences, Memorial University, St. John's, NF, Canada, A1B 3X5 ^Geological Sun/ey of Canada, 601 Booth Street, Ottawa, ON, Canada, K1A 0E8 The geology, structure, metamorphism, and geochronology of the western Central Gneiss Belt, Grenville orogen, have been documented on a transect along the well exposed shoreline of Georgian Bay, Ontario (Fig. 1; Culshaw et al. 1997). The results offer some insight into the tectonic evolution of Laurentian crust and margin in the interval 1160-980 Ma, as well as on the style of deformation and thermal evolution of the lower crust during the development of a large convergent orogen. A number of lithotectonic units with contrasting pre-Grenvillian and Grenvillian histories have been documented. Recognition of the pre-Grenvillian tectonic affinities and age contrasts of these units has proven essential for reconstructing the tectonic history of this part of the orogen. Penetrative ductile shear affected most rocks along the transect. Lithological boundaries are generally parallel to the dominant fabric, and discrete shear zones are typically difficult to distinguish from host tectonites. Most shear zones display both thrust-sense and normal-sense kinematic indicators. Important discrete shear zones include the thrust-sense Parry Sound shear zone, which formed at ca. 1160 Ma but was transported along with its host rocks over Laurentian crust sometime after 1080 Ma (Culshaw et al., 1997), and the ca. 1020 Ma normal-sense Shawanaga shear zone (Ketchum et al., 1998) that formed in the late stages of, or shortly after, the Ottawan orogeny (G4 of Jamieson et al., this volume). Metamorphic grade is high throughout the transect region, with Grenvillian upper amphibolite to granulite facies assemblages recording temperatures of 750-900''C and pressures of 8-12 kbar. Contrasting P-T-t paths have been obtained from different lithotectonic units. Those from allochthonous units (Composite Arc Belt and Laurentian margin; Table 1 of Jamieson et al., this volume; Carr et al., in press) probably reflect the combined effects of multiple Grenvillian thermal and tectonic events (Wodicka et al., in press). Grenvillian P-T-t paths from parautochthonous units appear to reflect the combined effects of post-1080 Ma Grenvillian thrusting and ca. 1020 Ma extension (Jamieson et al., 1995). Structural, metamorphic, and geochronological data (Fig. 1) indicate that Grenvillian convergence along the transect began with transport of the previously deformed and metamorphosed (ca. 1160 Ma; G2) Parry Sound domain over the Laurentian margin. This may have begun as early as 1120 Ma (G3) and was well underway by 1080 Ma (G4). This was followed by out-of-sequence thrusting and further convergence along successively deeper, NWpropagating ductile thrust zones during the Ottawan orogeny (G4). A major episode of ductile extension affected the transect at ca. 1020 Ma. A final stage of convergence affected the Grenville Front Tectonic Zone at ca. 1000 Ma (G5). Data from the transect are compatible with models that suggest that the rheology of the lower crust plays an important role in orogenic evolution. Pervasive ductile shear was important during both convergence and extension, with the dominantly sub-horizontal fabrics along the transect largely attributable to the effects of late, ductile extension (Culshaw et al. 1997). Strength contrasts between strong granulites, weak migmatites, and their host rocks, in part inherited from early Grenvillian and/or pre-Grenvillian tectonic epsiodes, strongly influenced both local- and crustal-scale patterns of deformation. REFERENCES

Carr, S.D., Easton, R.M., Jamieson, R.A., & Culshaw, N.G. (in press). Geologic cross-section across the Grenville orogen of Ontario and New York. Canadian Journal of Earth Sciences. Culshaw, N.G., Jamieson, R.A., Ketchum, J.W.F., Wodicka, N., Corrigan, D. and Reynolds, P.H. 1997. Transect across the northwestern Grenville orogen, Georgian Bay, Ontario. Tectonics, 16, 966-982.

ALICE SPRINGS. AUSTRALIA, 1999

Jamieson, R.A., Culshaw, N.G. and Corrigan, D. 1995. North-west propagation of the Grenville orogen: Grenvillian structure and metamorphism near Key Harbour, Georgian Bay, Ontario, Canada. Journal of Metamorphic Geology. 13, 185-207. Jamieson, R.A., Culshaw, N.G., Carr, S.D., Easton, R.M., and Ketchum, J.W.F. (this volume) The Grenvillian orogenic cycle: constraints from a geologic transect across Ontario and New York. Ketchum, J.W.F., Heaman, L.M., Krogh, T.E., Culshaw, N.G., and Jamieson, R.A. 1998. Timing and thermal influence of late orogenic extension in the lower crust: a U-Pb geochronological study from the southwest Grenville orogen, Canada. Precambrian Research, 89, 25-45. White, D.J., Easton, R.M., Culshaw, N.G., Milkereit, B., Forsyth, D.A., Carr, S.D., Green, A.G. and Davidson, A. 1994. Seismic images of the Grenville orogen in Ontario. Canadian Journal of Earth Sciences, 31, 293-307. Wodicka, N., Ketchum, J.W.F., and Jamieson, R.A. in press. Grenvillian metamorphism of monocyclic rocks, Georgian Bay, Ontario: implications for convergence history. Canadian Mineralogist. LAURENTIA + LAURENTIAN MARGIN

GFTZ

1200 1^1100 CD

1 1 1 1 1 1 1 1 1 1

BRITT

G H / MR

PSD

900

: 0 0 0 1 1 1 1 . 1 . 50

CMBBZ .

0 o

1 1 11 100

1 , 200

150

CMB

G2 1

•

• o;

••

< 1000 •

COMPOSITE ARC BELT

(OTTAWAN)

§° 0

1 . (

250

t . 1 300

• zircon 0 hornblende 350

400

b) .

pfo

foreland AGL-33

GLIMPCE-J V= H

Figure 1. a) U-Pb (zircon) and ^^Ar/^^Ai (hornblende) data indicating the limes of peak metamorphism and cooling through ca. 500°C respectively. Data sources are listed in Culshaw et al. (1997) and Carr et al. (in press), b) Crustal-scale cross-section extending from the Grenville Front to the Composite Arc Belt, based on geological and structural data from the Georgian Bay transect and seismic reflection profiles (e.g., White et al., 1994). Unit names and abbreviations as follows: GFTZ - Grenville Front Tectonic Zone; BRITT -Britt domain, SH Shawanaga domain; PSD - Parry Sound domain; uGH - upper Go Home domain; IGH-RA - lower Go Home, southern Rosseau, and Algonquin domains; MR-MS - Moon River, Muskoka, and Seguin domains; CMBBZ Central Metasedimentary Belt boundary thrust zone; CMB - Central Metasedimentary Belt. Nomenclature for other major units and orgenic events from Carr et al. (in press) and Jamieson et al. (this volume; Tabic 1).

O R O G E N E S I S IN THE O U T B A C K

FIELD EVIDENCE FOR FIVE PALAEOPROTEROZOIC OROGENIES IN THE NORTHERN AND EASTERN ARUNTAOROGENIC PROVINCE, CENTRAL AUSTRALIA p. Ding

North Flinders Exploration (A division of Normandy NFM Limited 24-25 Greenhill Road, Wayville, SA, 5034, Australia

More than ten sequential Palaeoproterozoic orogenic events, which are represented by angular unconformities, have been recognised in central and northern Australia. However, only the eight early ones of the Palaeoproteozoic orogenies have been discovered or studied by myself. The younger orogenies, which are well documented in the Mount Isa region and the Chewings Ragne area, will not be discussed here. Each unconformity separates a new cycle of stratigraphic sequence (cover) from the previously deformed older stratigraphic sequence (basement). The deformations associted with each of these orogenies have been temporarily labelled D1 and D8 in a younging order. Different orogenies are of variable intensities, and even the intensity of each orogeny varies from place to place. The Palaeoproterozoic orogenies and their associated strata in central and northern Australia are presented in Table 1. Up to date only five of the eight Palaeoproterozoic orogenies have been recognised in the Lander River area of the northwestern Arunta Inlier (Table 2). The Cycle 1 stratigraphic sequence is newly named the Yundurbulu Range Group by myself, which includes the Mt Stafford beds (Pys) and the newly named Yuendumu Metamorphics (Pyy). The Weldon Metamorphics (Pye) was Division One of B M R . this group comprises mainly granulite-facies meta-sediments and mafic rocks with different degree of migmatisation. The Cycle 2 stratigraphic sequence in the area is nelwy recognised during this study and named the Woodforde River Group (Pw). This group is well exposed in the middle and southeast part of the Reynolds Range, to the southeast of Algamba Bore. A clear angular unconformity relationship has been observed between the two Groups. The Woodforde River Group comprises mainly metamorphosed mature sediments (cyclic quartz sandstone, shale and carbonate rocks). Metamorphic grade varies from upper greenschist to lower granulite facies. The Cycle 3 stratigraphic sequence is named the Lander River Group (PI) and comprises mainly a low-grade metamorphosed turbidite sequence with few interbeds of quartz-rich meta-sandstone with occasional conglomerate, which is unconformably overlying the Woodforde River Group or sillimanite gneiss in several places. No Cycle 4 and Cycle 5 strata have been recognised in the area up to date. However there are evidence of deformation and magmatism of the two Cycles. The Cycle 6 stratigraphic sequence in the area is recognised and named the Coniston Group (Pc) by the author. This Group comprises mainly a greenschist facies upper sub-unit (Pcb) of intensely schistose meta-rhyollite and tuffaceous sandstone, and a basal sub-unit (Pea) of conglomerate and sandstone. This Group has been intruded by a microgranite (1785 ± 22 Ma), which was intensely deformed and retrogressed to a greenschist facies schist (Warimbi Schist of B M R ) . No Cycle 7 strata have been recognised in the area, but there are evidence of deformation and magmatism of Cycle 7. The Cycle 8 stratigraphic sequence in the area is newly named the Giles Range Group (Pi) by the author. A significant angular unconformity separated the Giles Range Group from the Coniston Group and older rocks. This Grouop is mainly very low-grade metasediments. An tectonic modal has beendeveloped to explain the observed cyclic orogenies.

ALICESPRINGS, AUSTRALIA, 1999

Table l . T Y P E PALEOPROTEROZOIC OROGENS AND THEIR ASSOCIATED STRATA IN CENTRAL AND NORTHERN AUSTRALIA CYCLE T Y P E OROGEN ORIENTATION (STRATIGRAPHIC SEQUENCES) Cycle8--1750-?1700 ma WNW Harts Range Orogen (Harts Range Group; Giles Range Group) Cycle7--1780-1750 ma Tomkinson Creek Orogen (Tomkinson Creek NE Group) Cycle 6--1810-1780 ma Hatches Creek Orogen (Hatches Creek Group; WNW Coniston Group) Cycle5--1830-1810ma Birthday Creek Orogen (Birthday Creek Group: NNW including Mount Winnecke Formation and Supplejack Down Sandstone) Cycle 4--1855-1830 ma Wilson Creek Orogen (Wilson Creek Group, E-W and ENE including Nanny Goat Creek beds, Helena beds and part of the Nongala beds); Ord River Orogen (Ord River Group including Whitewater Volcanics and Panton River Formation) Cycle 3--1900-1855 ma Lander River Orogen (Lander River Group); NE Tennant Creek Orogen (Warramunga Group); Halls Creek - Pine Creek Orogen (Halls Creek Group; Tanami Mine Succession; Mount Partridge Group + South Alligator Group + Finniss River Group). Cycle 2 - 7 1 9 8 0 - 1 9 0 0 ma West Kimberley Orogen (Bullaman Group); WNW Napperby Orogen (Woodforde River Group) Cycle 1 --2450-71980 ma Tanami Orogen (Tanami Group), East NE Kimberley Orogen (Sophie Downs Group); Arunta Orogen (Yundurbulu Range Group) >2500 Ma Archaean basement

Table 2. PALEOPROTEROZOl[C OROGENIES AND THEIR ASSOCIATED STRATIGRAPHIC SEQUENCES IN THE NORTH:ERN AND EASTERN ARUNTA OROGENIC PROVINCE CYCLE OROGENY AND THEIR ASSOCIATED ORIENTATION STRATIGRAPHIC SEQUENCES Harts Range Orogeny Cycle8--1750-1700 ma Harts range Group & Giles Range Group WNW Davenport Orogeny (?Early Strangways Orogeny) Cycle6--1810-1780 ma Coniston Group & Hatches creek Group WNW Barramundi Orogeny Cycle 3--1900.1855 ma Lander River Orogen (Lander River Group: NE including most of the low grade Lander Rock beds previously mapped in the area to the northwest of Algamba Bore) Reynolds Range Orogeny Cycle 2 - - ? 1980-1900 ma Napperby Orogen (Woodforde River Group: WNW including previous Wickstead beds, Woodforde River beds) Yuendumu Orogeny Cycle 1 >-71980 ma Arunta Orogen (Yundurbulu Range Group: NE including the Mount Stafford beds, Weldon Metamorphics and Yuendumu Metamorphics) >2500 Ma Archaean Basement

OROGENESIS IN THE OUTBACK

CYCLIC OROGENIES IN THE HALLS CREEK REGION, WA-FIELD EVIDENCE p. Ding North Flinders Exploration (A Division of Normandy NFM limited) 24-25 Greenhill Road, Wayville, SA 5093, Australia

The Halls Creek region (or East Kimberley region) in Western Australia was mainly studied and documented by B M R , GSWA and AGSO. The region was described by previous workers as a Proterozoic Orogenic Province, which was deformed and metamorphosed during Barramundi Orogeny and was cratonised shortly after a transitional period. Through my own field study, I recognised a quite different story for the region. It is concluded here that rather than a single and major orogeny, the Halls Creek region recorded five full-cycle orogenies, cycle 1 to cycle 5, which are comparable with that of the Tanami region. Cycle 1 Sophie Downs Group (Ps) is a medium to high-grade metamorphic sequence, which was intensely deformed and became recumbent before the intrusion of Sophie Downs Granite ( - 1 9 1 0 Ma). In its type area, around the Sophie Downs Granite the Sophie Downs Group was metamorphosed to amphibolite facies, which was previously mapped as Biscay Formation by B M R in 1967 and by AGSO in 1996. In the northern area the Sophie Downs Group is the amphibolite to granulite facies Tickalara Metamorphics (Pst), which was considered a high-grade metamorphic equivalent of the whole "Halls Creek Group" (Plumb et al., 1985). The precursor of this group comprises interbedded pelite, psammite, carbonate and calc-silicate rocks, which represent mature deposits in a shallow extensional basin. The cycle 1 depositional basin is named the East Kimberley Orogen here. The majority of the pre-orogenic concordant mafic bodies within the Sophie Downs Group were dolerite sills, although some concordant mafic layers are probably metamorphosed basalt lavas. Later mafic and ultramafic intrusive bodies are also present in this Group. No reliable direct age has been obtained for this stratigraphic unit. This Group is recognised and named by the author. The Sophie Downs Group shows medium to high-grade (upper amphibolite facies) burial metamorphism. The bedding parallel schistosity ( E S I ) and stretching lineation ( E L I ) represent pre-orogenic fabrics formed by preorogenic bedding parallel shearing (EDI), the extensional deformation phase of cycle 1. The E D I fabrics ( E S I and E L I ) in the Sophie Downs Group have been folded by the first orogenic event (Dl), the East Kimberley Orogeny, and formed regional NE trending F1 folds with regional schistosity of axial plane S I . During the burial phase of the next orogenic cycle (cycle 2), a progressive metamorphism (M2) in the northern area replaced staurolite by sillimanite in amphibolite zone and replaced biotite by cordierite and sillimanite in transition granulite zone. All fabrics became sub-horizontal under the heavy load during the deep burial. The recumbent D l structures were later reworked by D2 and D3 to form dome structures. Cycle 2 Bullaman Group (Pb) is proposed and named by the author to include mainly the Ding Dong Downs Volcanics (Pbd) in the east zone, and some of the high-grade metamorphic sequence in the West Kimberley. The Ding Dong Downs Volcanics includes meta-basalt, meta-felsic volcanic rocks and meta-sediments, in the Halls Creek area. The high-Th zircon populations in the felsic volcanic rocks indicate a volcanic crystallisation age of 1912 +/- 3Ma (Page and Sun 1994). The Ding Dong Downs Volcanics have generally been metamorphosed to biotite-grade of greenschist facies in the Halls Creek area, and developed bedding parallel schistosity and stretching lineation during subsiding period. The meta-sediments have been folded during the second orogenic event (D2), the West Kimberley Orogeny, and probably formed a NW trending upright fold belt. The Sophie Downs Granitoid ( - 1 9 1 0 Ma), which is an alkali-feldspar granite plug, intruded into the high-grade Sophie Downs Group. At the same time some micro granites ( - 1 9 1 0 Ma) intruded into the low-grade Ding Dong Downs Volcanics. The interpreted NW trending cycle 2 depositional basin is named the West Kimberley Orogen by the author. Cycle 3 Halls Creek Group (Ph) comprises three Formations: the lower Saunders Creek Formation (200 m), the middle Biscay Formation (2000-3000 m), and the upper Olympio Formation (> 4000m, up to 8000 m). The Saunders Creek Formation comprises mainly shallow water sandstone, whereas the Biscay Formation contains bimodal volcanic rocks and sub-volcanic intrusive rocks, which have zircon age of - 1 8 8 0 Ma and 1870 Ma. The Olympio Formation is a thick sequence of deep-water turbidite and contains alkaline volcanic or sub-volcanic rocks, which have been dated as young as -1857-h/-5 Ma. The Halls Creek Group was deposited in a NE trending extensional basin, which is named the Halls Creek Orogen (cycle 3) by the author to differentiate from the East Kimberley Orogen (cycle 1). The Halls Creek Orogen and the East Kimberley Orogen occupied the similar

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

geographic areas. The Pre-orogenic Woodward Dolerite, emplaced mainly into the Biscay Formation and the lower level of the Olympio Formation. The Halls Creek Orogeny may have occurred at about 1860 Ma (before the deposition of Whitewater Volcanics). The unconformity contact between the Sounders Creek Formation and its basement has been sheared in places and became a detachment boundary. Pre-orogenic granitoid (1865 Ma) became foliated after the Halls Creek Orogeny, whereas the post-orogenic Bow River Granitoid (1860 - 1850 Ma) remains massive. Cycle 4 stratigraphic units include the Panton River Formation of Ord River Group, the Whitewater Volcanics and the Koongie Park Formation. The Halls Creek Group, particularly the Biscay Formation, is unconformably overlying by a metamorphosed turbidite sequence in the east zone. The turbidite sequence above the unconformity was mapped as typical Olympio Formation by BMR and AGSO, but is now named the Panton River Formation (Pop), a member of the cycle 4 Ord River Group, by the author. Panton River Formation was recognised in the east zone only. The Whitewater Volcanics are widely exposed from the West Kimberley to the East Kimberley and was reported to overlie the Halls Creek Group unconformably. The U-Pb zircon ages obtained from the Whitewater Volcanics are inseparable at 1855 +/- 5 Ma and 1854 +/- 5 Ma (Page, R. & Sun, S. S. 1994). The Whitewater Volcanics have been strongly deformed and metamorphosed during cycle 4 orogeny and formed E-W trending folds, which were further interfered by D5 in the East Kimberley and by the Neoproterozoic Yampi Orogeny (1000 Ma) in the West Kimberley (the King Leopold Belt). In the central zone, a fault-bounded sequence, which includes deformed and metamorphosed sedimentary and felsic volcanic rocks, was named the Koongie Park Formation by AGSO and GSWA. A SHRIMP U-Pb zircon age of 1843 +/- 2 Ma has been obtained from these rocks (Page and Sun, 1994). No direct relationship has been observed between the Koongie Park Formation and the Halls Creek Group. Cycle 5 Caroline Pool Group (Moola Bulla Formation of AGSO) is a thinly bedded sandstone sequence with pebbly sandstone and fine conglomerate at its basal position in the east zone. This group overlies the Olympio Formation unconformably, and may be correlated to the Speewah Group in the Kimberley Basin and the Revolver Creek Formation in the central zone. Felsic volcanic interbeds in the Valentine Siltstone of the Speewah Group has been dated as 1834±3 Ma. The Caroline Pool Group is unconformably overlain by the Kimberley Group. The Kimberley Group is generally a post-orogenic cover sequence, which is the stratigraphic equivalent of the cycle 6 Hatches Creek Group in the Devon Port Province. The Kimberley Group was intruded by the Hart Dolerite (1790 Ma), the Fish Hole Dolerite and associated granophyre. The latter has a zircon U-Pb age of -1800 Ma. The postorogenic San Sou Granitoid (1790 Ma) intruded into the Halls Creek Group, the Ding Dong Downs Volcanics, and the cycle 2 granite in the southern area of the east zone. The Kimberley Group has been gently folded but substantially faulted in the Halls Creek area.

OROGENESIS IN THE OUTBACK

CONTROLS ON FLUID FLOW IN SCAPOLITE-BEARING SHEAR ZONES OF THE LATEPROTEROZOIC KUISEB FORMATION, DAMARA OROGEN, CENTRAL NAMIBIA A. Dombrowski\ and S. Hoernes^

^Mineralogisches Institut der Universitat Wurzburg, Am Hubland, 0-97074 Wurzburg, Germany ^Mineralog.-Petrolog, Institut, Universitat Bonn, Poppelsdorfer SchloU, D-53115 Bonn, Germany

Scapolite formation in amphibolite facies metasedimentary rocks may indicate the presence of evaporitic layers in the precursor sediments or interaction with chlorine and/or C02-rich fluids during metamorphism. These fluids can be derived from either an external reservoir or by successive devolatilization reactions during prograde metamorphism. Deciphering an evaporitic nature of the protolith poses much of a problem because clear evidence such as pseudomorphs after evaporite minerals are usually lacking. Here we present petrological and stable isotopic constraints on scapolite formation and fluid-rock interaction in amphibolite facies metaturbidites of the pan-African Kuiseb Formation in the northern Khomas Trough of central Namibia. REGIONAL GEOLOGY

The Late-Proterozoic Kuiseb Formation in the Khomas Trough, central Namibia, comprises a succession of multiply deformed metaturbidites which were deposited in an elongate submarine fan system at an convergent continental margin (Kukla, 1992). Proceeding subduction of the Kalahari Craton beneath the Kongo Craton led to incorporation of these sediments into an accretionary prism. The amphibolite facies metamorphic overprint is related to crustal thickening and subsequent thermal relaxation during the collision of the Congo and Kalahari cratons in late panAfrican times. The structural evolution of the Khomas Trough is characterised by five phases of ductile deformation (Kukla, 1992). The expression of a markedly heterogeneous regional structural regime associated with thrusting are kilometre-wide low-strain zones alternating with narrow high strain-zones which can be traced laterally for more than 100 km. The Kuiseb metaturbidites consist of metapsammitic and metapelitic rocks with intercalations of calcsilicate rocks, marble, tremolite schists, graphite schists as well as amphibolites of the Matchless Member. An outstanding feature of the northern IChomas Trough are two scapolite-bearing sequences situated in two of the major thrust slices. Mineral assemblages in the scapolite-bearing sequences reflect metamorphic peak conditions conditions of 620°C at 2.5-4 kbar. Metamorphism and deformation are post-dated by the intrusion of the Donkerhuk granite and related rocks. PROTOLITH CHARACTERISTICS

The results of geochemical and stable isotope investigations show that neither REE patterns nor carbon isotope signatures of scapolite ( 5 ^ ^ C p d b = - 8 to - l l % o ) and calcite to -10%o) indicate the existence of former evaporite sequences in the precursor sediments of the scapolite-bearing horizons of the northern Khomas Trough (Dombrowski et al., 1996). Moreover, mass balance calculations demonstrate that no significant metasomatism affected the Kuiseb metapelites and -psanmiites. Intensive fluid-rock interaction with externally derived fluids can thus be precluded. The original sediment presumably consisted of a mixture of carbonate, sand and mud deposited by large-volume high-density turbidity currents. The chlorine-rich fluids which triggered scapolite formation were clearly derived from pore fluids of the sediment pile rather than from an external fluid reservoir. Scapolite formation is related to decarbonation reactions in the carbonate-bearing portions of the rocks. Thus, the release of large quantities of CO2 enhanced the permeability of the rocks and gave rise to increased fluid flow through the calcsilicate layers. FLUID E V O L U T I O N A N D S T A B L E I S O T O P E G E O C H E M I S T R Y

The fluid evolution in the scapolite-bearing shear zones of the northern Khomas Trough can be reconstructed using stable isotope characteristics of different lithologies and chlorine contents of scapolite and hydrous minerals such as biotite, apatite and amphibole. Scapolites of the southern scapolite-bearing sequence display quite uniform compositions while in the northern horizon, CI activity gradients are preserved on a millimeter scale. It can thus be concluded that fluid flux, at least in the northern horizon, was rather low. CI contents in biotite and apatite in both scapolite-bearing sequences do not show any significant differences. However, at the outcrop scale, biotites of metapelitic rocks are considerably variable with respect to their CI contents ranging from Cl-free to Cl-enriched. Fluid migration in the metapelitic layers is therfore interpreted to have been strongly channelized along a network of microfractures. This way, parts of the biotites were able to exchange with the Cl-bearing fluid phase while others remained unaffected.

ALICE SPRINGS, AUSTRALIA, 1999

Further evidence for the fluid evolution in the northern Khomas Trough come from coexisting scapolite and epidote. The occurrence of epidote indicates infiltration of H20-rich fluids while the presence of scapolite requires elevated CO2 activities in the coexisting fluid. Infiltration of aqueous fluids would result in homogenization of CI activity gradients which, however, are preserved in scapolites of the northern Khomas Trough. A ready explanation is the local buffering of fluid composition by decarbonation reactions which lowered the H2O activity in the aqueous NaCl-bearing fluid until conditions for scapolite formation were achieved. The migrating fluids probably were derived from dehydration of adjacent metapelites and residual pore fluids. The whole rock values of scapolite-bearing (5^^0SMOW=10-5-12.8%o) and scapolite-free (5^^0SMOW=9.242.8%o) Kuiseb samples are essentially identical. Thus, scapolitization cannot be explained by equilibration of typical Kuiseb metapelites with an exotic externally derived fluid phase. Calculating whole rock fractionation factors for each sample using the refined increment method described by Hoffbauer et al. (1994) indicates the interaction of scapolite-bearing rocks with a homogeneous fluid phase while the metapelites either remained unaffected or did not reach equilibrium with the coexisting fluid. This has important implications for the fluid flow patterns of the different rock types. Although high fluid fluxes can be precluded for both, scapolite-free and scapolite-bearing rocks, the throughput of fluid in the calc-silicate layers has been distinctly higher. In metapelitic layers fluid migration was probably focused along short-lived fratures leading to disequilibrium patterns of the whole rock fractionation factors. DISCUSSION

Defining the conditions of scapolite formation in the Kuiseb metaturbidites has important implications on the fluid flow history of deep crustal shear zones in the northern Khomas Trough. Fluid flow patterns are controlled by smallscale processes rather than the large-scale geotectonic evolution of the Khomas Hochland accretionary prism. The scapolite-bearing sequences reflect fluid flow processes which are related to the main deformation phase (D3) and are dominantly controlled by the protolith composition. There is no need to infer the throughput of large amount of fluids along deep crustal shear zones during D3 to explain scapolitization of the Kuiseb metasedimentary rocks. During continent collision prograde metamorphic devolatilization reactions produced substantial amounts of fluid which interacted with the Kuiseb Formation rocks together with pore fluids from the sediment pile. Local variations in fluid composition are the result of a complex interplay between internal buffering and activity variations of H2O, CO2, HCl and NaCl that arose from initial variations in protolith composition and subsequent modification of permeability during devolatilization. Scapolite formation in the Kuiseb Formation metaturbidites is hence an expression of the original protolith heterogeneity in combination with the small-scale fluid evolution history in the Khomas Hochland accretionary prism. REFERENCES

Dombrowski, A., Hoernes, S., and Okrusch, M., 1996. Scapolitization in the Kuiseb Formation of the Damara Orogen: geochemical and stable isotope evidence for fluid infiltration along deep crustal shear zones. Communs. geol. Surv. Namibia, 11, 21-29. Hoffbauer, R, Hoernes, S., and Fiorentini, E., 1994. Oxygen isotope thermometry based on a refined increment method and its application to granulite-grade rocks from Sri Lanka. Precambrian Research, 66, 199-220. Kukla, P.A., 1992. Tectonics and sedimentation of a Late-Proterozoic Damaran convergent continental margin, Khomas Hochland, central Namibia. Geol. Surv. Namibia Mem., 12, 95 pp.

OROGENESIS IN THE OUTBACK

STYLES OF TECTONIC REWORKING OF BASEMENT IN THE DOM FELICIANO BELT OF SOUTHERN BRAZIL DURING THE NEOPROTEROZOIC LA.D.Fernandes ^ C.C. Porcher A.O.M. Silva; R.A. Cliff G.T.R. Droop ^ 1. Department of Geology, Universidade Federal do Rio Grande do Sul, P.O. Box 15065 - 91501/970, Porto Alegre, RS, Brazil. 2. School of Earth Sciences - University of Leeds, LS2 9JT- UK. 3. Department of Earth Sciences, University of Manchester, Oxford Road, Manchester, M13 9PL, UK.

Old continental crust is a major component of collisional orogenic belts of diverse ages and tectonic settings in several continents. However, to most geologists working on precambrian shields of Brazil, the recognition of basement rocks within orogenic belts brings up the debate about the 'reworking of older crust vs. new crustal addition' during orogeny. Despite the fact that orogenic studies of the last two decades have demonstrated that precambrian orogens, like their phanerozoic analogues, are unique amalgamations of geotectonic units of diverse nature and age, the general lack of geological constraints for isotopic data in the Southern Brazilian Shield has allowed the adoption of all sorts of geotectonic models for the evolution of this segment of the continental crust. Models based on concepts derived from Plate Tectonics as the main orogenic process stand side-by-side with others based on the activity of Mantle Plumes. However, in the last few years, the change of paradigm of orogenic studies resulted in the establishment of research programmes directed at the investigation of rock units and processes related to both, the old and the younger tectonic episodes within an orogenic belt. It became clear that correct understanding of the nature and ages of rocks and their fabrics are essential to unveil the evolution of any segment of the continental crust. In other words, to find out what exactly happened during this otherwise generic 'reworking of older crust', it is necessary to study geological processes, constrain as accurately as possible their physical conditions and sequential development. Rock sequences representing older continental basement that crop out within the Dom Feliciano Belt were affected by metamorphism and deformation during the Neoproterozoic. Abundant magmatism genetically related to collisional and transcurrent tectonics was the predominant style of 'reworking' in the eastern and western parts of the central geophysical domain of this belt. Along the western segment of this belt orthogneisses of the transamazonian (Palaeoproterozoic) basement and schists of the brasiliano (Neoproterozoic) cover were tectonically interleaved under low-grade metamorphic conditions. Following a synthesis of the principal rock units and structures related to tectonic processes in these two key-areas we discuss the implications of this evolution for the assembly of the Gondwana Supercontinent. Styles of reworking of palaeoproterozoic basement are remarkable distinct along the eastern and western segments of the central geophysical domain. While in the west neoproterozoic reworking took place under conditions of low metamorphic grade, NE-SW tectonic transport directions and abundant aqueous fluids, the older crust in the eastern segment was mainly subjected to superposed high-grade metamorphism and associated partial melting. In this region, deformation on flat-lying shear zones with E-W transport direction of the early stages was followed by nucleation of large transcurrent shear zones and emplacement of crustal-derived syntectonic granites during later stages (Femandes & Koester 1999). The increasing temperatures and more extensive reworking along the eastern part of the DFB might be interpreted in terms of polarity of the tectonic processes. Most geotectonic models proposed for this belt agree about the presence of 'reworked basement' along its eastern part. In this area, the presence of large volumes of neoproterozoic calc-alkaline granitic magmas has been recognised on the basis of petrologic and isotopic studies. While the more cautious referred to the tectonic environment of this magmatism as 'thick continental crust', others preferred the bold (and easier to test) designation of 'magmatic arc of continental margin type'. This interpretation not only had implications in terms of existence of older continental crust, but also predicts that it would be extensively reworked by magmatism, metamorphism and deformation with the progressive development of orogenic processes. One of the major questions related to the evolution of the DFB is the relative age of closure and polarity of subduction of former oceans (Charrua in the West and Adamastor in the East). Closure of these oceans has originated the magmatic arcs (eastern or western, respectively) and their collision with the Kalahari and Rio de La Plata cratons gave rise to the orogenic collage (DFB). While there is no sufficient reliable radiometric data to decide which subduction zone is older, the style of reworking reported for the basement rocks can be used as source of speculation about the polarity of subduction. The more intensive orogenic magmatism and high-grade metamorphism of the eastern segment suggests that this region was close to a major thermal anomaly. If this anomaly is related to consumption of oceanic lithosphere (whether Charrua or Adamastor), subduction towards the West is more compatible

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

with the nature, intensity and history of tectonic reworking of this segment of older crust. While the nature of the protoliths (orogenic magmatism) and tectonic transport directions of the eastern segment can be more readly related to early-orogenic E-W directed collisional episodes, the tectonic significance of the younger deformation observed in the western segment remains unclear. Because it shows kinematic and metamorphic compatibility with the mid-crustal transcurrent shear zones nearby, this deformation was attributed to the accommodation of differential strain produced by collision between the Kalahari and Congo cratons, during the late stages of assembly of the SW-Gondwana.

O R O G E N E S I S IN T H E O U T B A C K

THRUST TECTONIC STYLES AT THE MARGINS OF THE AMADEUS BASIN, CENTRAL AUSTRALIA T. Flottmann,'' M. Hand,'' D. Closed C. Edgoose^ and I. Scrimgeour^ 1 Department of Geology and Geophysics, The University of Adelaide, SA 5005 2 Northern Territory Geological Survey, P.O. Box 2655, Alice Springs, N.T 0871 tflottma@geology. adelaide. edu. au

Intracratonic deformation at the northern and southern margin of the Amadeus Basin during the Petermann 570 - 530 Ma) and the Alice Springs orogenies (400 - 300 Ma) respectively (Fig.l), is characterised by the formation of major crustal wedge systems which were inserted at the base of the Amadeus Basin. In both orogens the wedges were emplaced towards the basin, that is towards the north in the Petermann Orogen and towards the south in the Alice Springs Orogen. In both orogens, the evaporitic Bitter Springs Formation/Pinyinna beds form a detachment between the upper successions of the Amadeus Basin and lower units. The former were displaced along major backthrusts towards the south in the Petermann Orogen and towards the north in the Alice Springs Orogen. These foreland and hinterland propagating thrust systems form major tectonic wedges (or triangle zones) which are now exposed at different crustal levels in both orogens (Figs 2 amd 3). The Petermann Orogen exposes mid-crustal kyanitegrade cover sequences with pervasive deformational fabrics that formed during intense crustal stacking. Despite the considerable metamorphic overprint and pervasive deformation, a basic stratigraphic template is clearly preserved, allowing a schematic restoration of the major thrust stacks. Results suggest over tenfold thickening of a stratigraphic package c. 1.5 km thick that includes: (1) basement, (2) a rift succession (Bloods Range beds), (3) the Dean Quartzite and (4) the evaporitic Pinyinna beds, occurred during the Petermann Orogen. Along the northern edge of Amadeus Basin in the vicinity of Ormiston Gorge, balanced sections allow a full restoration and forward modelling of the wedge/backthrust system. The results suggest an overall shortening of about 19 km occurred during the Alice Springs Orogeny, which is almost an order of magnitude less than the shortening in the Petermann Orogen. Ongoing shortening after the emplacement of the wedge^ackthrust system in both orogens led to regional tilting of the wedge/backthrust system towards the Amadeus Basin. We suggest that although the wedge^ackthrust systems along the Amadeus basin margins developed at different times they have several key features in common. Firstly, both are developed in the footwall of major crustal discontinuities consisting of the Woodroffe Thrust in the south, and the Redbank Shear Zone in the north. Both fault systems acted as crustal strain guides that prompted displacement along footwall shortcut thrust systems leading to the formation of crustal wedges. The partitioning of displacement into the wedge/backthrust system appears largely due to the presence of the mechanically weak horizon of the Bitter Springs Formation/Pinyinna beds which forms a suitable stratigraphically controlled detachment zone. Along this decollement, the upper successions were backthrust above the foreland propagating wedge systems. In all likelyhood the style of deformation is common where the stratigraphy of fold thrust belts contains suitable detachment horizons and thrust propagation is controlled by major pre-existing fault zones.

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

THE CRUSTAL STRUCTURE OF CENTRAL AUSTRALIA FROM DEEP SEISMIC PROFILING. B.R. Goleby, BJ. Drummond and RJ. Korsch Australian Geodynamic Cooperative Research Centre, Australian Geological Sun/ey Organisation, P.O. Box 378, Canberra, ACT, 2601, Australia

Seismic reflection transects through Central Australia have shown the presence of a variety of deformation styles, including elements of both thick-skinned and thin skinned tectonics that coexist within the same province. The Central and Northern Provinces of the Arunta Block have a 'thick-skinned' style of crustal deformation with the major thrusts being south-directed. The Amadeus Basin and parts of the southern Arunta Block have a 'thinskinned' style of deformation, with a major detachment located within the Bitter Springs salt layer in the lower part of the sedimentary succession. This deformation of the northern margin of the Amadeus Basin was by southdirected thrusts, with thick sediments deposited in a foreland setting. These south-directed thrusts link with deeper penetrating northward-dipping thrusts. AMADEUS BASIN

GT 24°S " I

Gardiner Thrust

MH OTZ RTZ

ARUNTA BLOCK

NGALIA BASIN

22®S

MacDonnell Homocline

Figure 1: Schematic cross section through central Australia along the deep seismic traverses, from the central Amadeus Basin in the south to the Ngalia Basin in the north. The thick-skinned faulting within the Arunta Block contrasts with the thin-skinned styles within the Amadeus Basin.

Arunta Block

Within the central and northern Arunta Block, the major displacements occurred along planar, north dipping faults and shear zones that extend from the surface to the crust-mantle boundary (Figure 1), the most prominent being the Redbank Thrust Zone (RTZ, Figure 1). The overall character of seismic reflections changes several times along the traverse indicating fundamental differences in the crustal composition. One good example is across the Redbank Thrust Zone. Another is at the northernmost extent of the seismic traverse beneath the Ngalia Basin (Figure 2). This figure shows two distinct domains of reflectivity separated by a moderately north-dipping crustal penetrating shear zone (Domain Shear Zone, Figure 2). The northern domain consists of a series of short reflectors at differing orientations. The shear zone itself is imaged as a series of linear reflectors (Figure 2). Within the southern domain, the reflectivity can be divided into several predominant trends. The orientation of the reflectors defining the shear zone (Sl-Sl", Figure 2) is pervasive throughout the Arunta Block, with the Redbank Thrust Zone being the best example. We infer that this dip defines the youngest deformation. The two parallel dashed lines (S2-S2", Figure 2) are truncated by the main shear zone and other similar shears, inferring that the dashed trend represents an earlier 'deformation' or lithological trend. Between these two trends, there are several other trends, the next most common being an "S" pattern (S3-S3", Figure 2). In all, the seismic images several phases of 'deformation'.

Amadeus Basin The MacDonnell Homocline (MH, Figure 1) results from the interplay of a series of southward-directed faults south of the Redbank Thrust Zone and the northward-directed back-thrusting of the Amadeus Basin sediments on a

OROGENESIS IN THE OUTBACK

decollement near the base of the sediments (see also Flottman and Hand, 1999). New reprocessing and interpretation of the seismic data through this region supports this hypothesis (Rudge et al., this volume). Northern Arunta Block Seismic

Figure 2. Portion of migrated deep seismic data within the northern Arunta Block at the northern end of the seismic traverse showing differences in reflectivity and hence crustal composition across a major crustal-penetrating shear zone. In the Amadeus Basin, the Gardiner Thrust (GT, Figure 1) has a ramp-flat geometry with about 30 km of horizontal northward movement and up to 6 km of vertical uplift of the upper plate (Korsch et al., 1998) resulting in significant repetition of the Amadeus succession . This thin-skinned style of deformation is north-directed, but appears to be synchronous with the south-directed thick-skinned thrusting in the Arunta Block (Korsch et al, 1998). Comparison with Pyrenees A comparison with the deep seismic reflection data from the Pyrenees of Spain suggests that many of the features seen in central Australia may be typical of continent-continent orogenies. The Pyrenees are an Early Cretaceous to Eocene example of the interaction between foreland sedimentation and thrust kinematics (Coney et al., 1996). The crustal section through central Australia (Figure 1) shows a gross similarity with the crustal structure for the Pyrenees (Figure 3). In both sections, there is a major basal detachment surface with the sedimentary succession

Figure 3: Cross section of the Pyrenees along the ECORS profile, (from Cloetingh et al, 1998, p. 39)

ALICE SPRINGS, AUSTRALIA, 1999

thinning away from the axial zone. There is evidence of significant thrusting at the northern margin, and also within the central sections of both basins. The deep seismic data from both profiles show a major discontinuity within the lower crust, with a section of mantle material being thrust upwards. REFERENCES

Cloetingh, S., Stephenson, R., Marzo, M., Munoz, J.A. and Verges, J., 1998. Tectonic Geomorphology and Sedimentary Dynamics. 9th Workshop of the ILP Task Force, "Origin of Sedimentary Basins", EuroconferenceEuroprobe Oliana, Program and Field Trips, 64pp. Coney, P.J., Munoz, J.A., McClay, K.R. and Evenchick, C., 1996. Syntectonic burial and post-tectonic exhumation of the southern Pyrenees foreland fold-thrust belt. Journal of the Geological Society, London, 153, 9-16. Flottman, T. and Hand, M., 1999. Folded basement-cored tectonic wedges along the northern edge of the Amadeus Basin, Central Australia: evaluation of orogenic shortening. Journal of Structural Geology, 21, 399-412. Korsch, R.J., Goleby, B.R., Leven, J.H. and Drummond, B.J., 1998. Crustal architecture of central Australia based on deep seismic reflection profiling. Tectonophysics, 288, 57-69. Rudge. A., Goleby, B.R and Barr, T., 1999. The Crustal structure of the Arunta Block and northern Amadeus Basin: from AGSO Deep Reflection Seismic Data. This Volume.

OROGENESIS IN THE OUTBACK

THE NATURE OF THE BOUNDARY BETWEEN THE ZIMBABWE CRATON AND THE MOZAMBIQUE BELT, CENTRAL MOZAMBIQUE AND NEW SHRIMP ZIRCON DATA. G.H. Grantham'' , A.S.T.D. Manhica^ R.A. Armstrong^ ^Council for Geosciences, P/BagX112, Pretoria 0001, South Africa ^ Geological Survey of Mozambique, Maputo, Mozambique, ^ PRISE, Australian National University, Canberra, Australia

(grantham@geoscience.org.za)

Recent mapping indicates that the margin between the Zimbabwe Craton and the Mozambique Belt in central western Mozambique is defined by a shear zone which is characterised by a strong, near vertical N-S planar fabric which is defined by mineral fabrics as well as lenticular leucosome blebs. Lineations and kinematic indicators in this zone are sparse however most suggest a sinistral sense of movement. This planar fabric is the only fabric in a ~2300Ma granite to the west whereas the fabric postdates earlier fabrics in the Chimoio Granodiorite Gneiss and the Nhansipfe Megacrystic Granite gneiss. New SHRIMP zircon data from these intrusions give ages of 1107 +/- 14 Ma (MSWD = 0.39) and 1112 +/- 18 Ma (MSWD) = 0.52. These data confirm the Grenvillian age suggested by Rb/Sr data which suggested ages for these orthogneisses of ~1200Ma and 980Ma respectively. These data also indicate that the emplacement of these bodies may have been responsible for the ~1070Ma ages recorded by Ar/Ar studies on biotite from Archaean gneisses which underlie the Zimbabwe Craton in central western Mozambique. Ar/Ar data from micas across the craton/ mobile belt margin and coincident with the shear zone suggest cooling through ~300°C at ~460Ma whereas zones on either side of the shear zone yield Ar/Ar biotite ages of - 5 5 0 M Manhica et al. (1998). Rb/Sr mineral whole/rock studies currently being undertaken are aimed at confirming the Ar/Ar data, particularly those data which suggest a Grenvillian-age overprint on the Zimbabwe Craton particularly since this data could originate from "excess" Ar in the mica. No evidence for a Pan-African suture has been recognised in the area. Consequently no evidence to support the existence of the Mozambique Ocean prior to the amalgamation of East and West Gondwana is seen as has been suggested by some workers who proposed the fragmentation of Rodinia and subsequent amalgamation of East and West Gondwana during the Pan-African in this area. This data would suggest that the Kalahari Craton constituted part of East Gondwana and that the Pan African amalgamation occurred along the western margin of the Kalahari Craton. These data would also suggest that the Pan-African event in southern Mozambique was intra-cratonic. REFERENCES

Manhica A.D.S.T., Grantham G.H. & Guise P.D. 1998 An ^°Ar/^^Ar study of the Zimbabwe Craton-Mozambique Belt boundary in the Manica-Chimoio area, western central Mozambique. Gondwana X abstracts. Cape Town. Jnl. Afr. Earth Sci., 27,1 A, p 135.

A L I C E SPRINGS, AUSTRALIA, 1 9 9 9

HEAT PRODUCTION RATES IN AUSTRALIAN PROTEROZOIC TERRAINS Martin H a n d \ Kerry Slater^, S a n d r a M a c L a r e n \ M i k e Sandiford^

^ Geology and Geophysics, Adelaide University, Adelaide, S.A. 5005. ^ Northern Territory Geological Sun/ey, P.O. Box 2655, Alice Springs, NT, 0871.

In regions with substantial outcrop, calibrated airborne radiometric data offers a fast way to estimate regional heat production rates, and provides a general way to assess the homogeneity of heat producing bodies. In combination with TMI and gravity data, generalised volumetric estimates of rock units can be made that allow crustal heat production to be assessed. In the Arunta and Mount Isa Inliers in the Australian Protero-zoic, heat production rates that are largely concentrated in regional scale granites that are at least several km's thick, are up to 10 (Figures 1 and 2). These bodies therefore represent major reservoirs of thermal energy. For example, a 4-5 kilometer thick sheet of the Anmatjira Orthogneiss (Figure I) contributes more to the surface heat flow than that which is commonly attributed to the entire continental crust (e.g. MacLennan & I'aylor, 1996). Although the surface heat production rates in many parts of the Australian Proterozoic appear very high by global standards, they are consistent with the high surface heat flow that characterises much of the province, and suggest that some regions of crust may generate in excess of 70 mWm'^. Crustal heat production of this magnitude should exert a major, or even controlling influence on the long-term thermal and mechanical stability of the lithosphere (e.g. Sandiford & Hand, 1 9 9 8 ) , and is likely to have played a central role in shaping the geologic evolution of parts of the Australian lithosphere containing Proterozoic crust.

References

McLennan, S.M., Taylor, S.R., 1996. J. Geol. 104, 369-377. MacLaren, S., Sandiford, M. & Hand, M. 1999. Geology, in press. Sandiford, M. & Hand, M. 1998. EPSL., 162, 97-110.

OROGENESIS IN THE OUTBACK

UHT CRUSTAL METAMORPHISM: ADVANCES IN CONSTRAINING PEAK TEMPERATURES AND P^T HISTORIES S. L Harley Department of Geology and Geophysics, University of Edinburgh, Kings Buildings, West Mains Road, Edinburgh, Scotland EH9 3JW. email: sharley@glg.ed.ac.ul<

Ultra high temperature (UHT) regional metamorphism is defined as metamorphism in which crustal rocks are subjected to temperatures of 900-1100®C at moderate pressures (7-13 kbar). Classic examples include the Napier Complex of Antarctica, Wilson Lake and Sipiwesk Lake in Canada, the Eastern Ghats of India and Labwor Hills in Uganda. UHT metamorphism also has been recognised from Scotland, southern India, Sri Lanka, Madagascar, and localities in Antarctica. Many of these areas or localities include or involve polycyclic granulites that may have been through one or more metamorphic / tectonothermal event prior to the event that left its UHT imprint on them. For example, in the Napier Complex there is geochronological evidence for several events, magmatic and metamorphic, preceeding the late Archaean metamorphism that many workers consider responsible for the UHT assemblages. Similarly, in the Rauer Group of East Antarctica and the Highland Series of Sri Lanka there are complex prehistories of magmatism, sedimentation and probable metamorphism that are not related to the UHT event recorded in rare sapphirine-orthopyroxe-silimanite granulites in these terrains. The key UHT indicators in crustal granulites are mineral assemblages involving sapphirine, garnet, aluminous orthopyroxene, cordierite, sillimanite, spinel and quartz. Experimentally-constrained FMAS and KFMASH grids involving these phases and osumilite and melt show that sapphirine -H quartz is stable only at >1040°C in reduced rocks, that osumilite is restricted to >900®C and has a stability limit of 9 kbar in FMAS, and that orthopyroxene + sillimanite + quartz is restricted to pressures (P) greater than 8 kbar in KFMASH. These criteria, many of which have only been experimentally developed since 1995, and grids newly isoplethed for mineral compositions (Harley 1998a) allow peak P-T conditions to be defined and the post-peak P-T paths delineated. New Fe-Mg exchange thermometry using the gamet-orthopyroxene calibration of Ganguly et al (1996) yields temperatures of 900-1100°C for many UHT areas. However, the best UHT compositional indicator is high AI2O3 content (8-12 wt%) in orthopyroxene. Thermometry based on the AI2O3 content of orthopyroxene coexisting with garnet, in assemblages constrained to be >1000®C, often support these UHT conditions. For example, the Aranovich and Berman (1997) thermometer returns 960±50°C for UHT granulites from the Napier Complex (Harley 1998b), consistent with the notion that under most circumstances Al net-transfer and exchange reactions close to diffusion at temperatures that are much greater than those relevant to Fe-Mg exchange. Preserved UHT P-T records are varied in that both near isobaric cooling (IBC) and near-isothermal decompression (ITD) post-peak P-T paths are deduced from reaction textures. In contrast the prograde P-T histories of most UHT areas are poorly known. Kyanite inclusions (Motoyoshi and Ishikawa 1997) or pseudomorphs after kyanite (Raith et al 1997) imply clockwise P-T trajectories for specific UHT-ITD occurrences. At Mather Peninsula, Rauer Islands, mineral compositions (Harley 1998c) produce a peak P-T estimate, 11-12 kbar and 1033±30°C, that is consistent with petrogenetic grids. A UHT-ITD path is derived here from grid-based interpretation of reaction textures in which garnets are replaced by lower-pressure equivalents such as sapphirine -1- orthopyroxene + cordierite. Identical textures described from several other UHT occurrences are interpreted similarly, and ITD from 10-12 kbar to ca. 7-8 kbar at temperatures in excess of 900°C, and even 1000°C, is considered to be the post-peak P-T path typical of these UHT localities. Elemental mapping of Al zoning in orthopyroxene may provide insight into pre- and post-UHT P-T histories (Harley 1998b). At Mather Peninsula orthopyroxenes in garnet show rimward zoning from 8 to 10.5 wt % AI2O3. This reflects heating through 90±20°C in the interval 930-1030°C at 9-12 kbar, and corresponds with the temperature interval over which melts would be generated (e.g. Carrington and Harley 1995). In the Napier Complex, where IBC has previously been inferred, orthopyroxene that coexists with sapphirine and quartz zones rimwards from 12 to 9 wt% AI2O3. These features reflect the reaction 2Mg-tschermaks = sapphirine + quartz and lead to a calculated minimum cooling through 80°C within the stability field of sapphirine + quartz (Harley 1998b; Harley and Motoyoshi 1999). The new estimate of at least 1120±20®C for the peak temperature of regional metamorphism in the Napier Complex is the most extreme for any UHT terrain. Textures involving sapphrinie +

ALICE SPRINGS, AUSTRALIA, 1999

quartz in UHT granulites from from In Ouzzal, Algeria, may also be interpretable in terms of the tschermaks reaction and hence imply similarly extreme peak metamorphic temperatures. The question of how such high temperatures can be generated, even transiently, prior to decompression and/or cooling of the UHT granulites is still controversial. Local heat sources may explain individual cases, but do not provide a general explanation for the rare clockwise P-T paths and common ITD histories which imply that significant heat-transfer followed burial and accompanied exhumation. Advected heat has to be a major contributor to the thermal budget, and this heat must in many cases be delivered during deformation of crustal rocks that are significantly older than the metamorphism itself and have been through early metamorphic events or cycles. Convective thinning of the lithospheric thermal boundary layer (TBL) during or after a phase of crustal thickening provides a plausible model for large-scale UHT metamorphism, and would account for rapid exhumation and hence ITD P-T paths if TBL removal triggered crustal extension. However, specific evaluation of this and other models requires better constraints on the timescales of metamorphism and post-peak ITD or IBC in all UHT terrains or occurrences so far investigated. REFERENCES

Aranovich, L.Ya. and Berman, R.G., 1997. A new gamet-orthopyroxene thermometer based on reversed AI2O3 solubility in Fe0-Al203-Si02 orthopyroxene. American Mineralogist, 82, 345-353. Carrington, D.P. and Harley, S.L., 1995. Partial melting and phase relations in high-grade metapelites: an experimental petrogenetic grid in the KFMASH system. Contributions to Mineralogy and Petrology, 120, 270291. Ganguly, J., Cheng, W. and Tirone, M., 1996. Thermodynamics of aluminosilicate garnet solid solution: new experimental data, an optimized model, and thermodynamic applications. Contributions to Mineralogy and Petrology, 126, 137-151. Harley, S.L., 1998a. 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 ?" Geological Society London, Special Publications, 138, 81-107. Harley, S.L., 1998b. An appraisal of Peak temperatures and thermal histories in ultrahigh-temperature (UHT) crustal metamorphism: the significance of aluminous orthopyroxene. In: Motoyoshi, Y and Shiraishi, K. (eds) "Origin and Evolution of Continents". Memoirs National Institute of Polar Research, Special Issues, 53,49-73. Harley, S.L., 1998c. Ultrahigh temperature granulite metamorphism (1050^C, 12 kbar) metamorphism and decompression in garnet (Mg70)-orthopyroxene-sillimanite gneisses from the Rauer Group, Bast Antarctica. Journal of Metamorphic Geology, 16, 541-562. Harley, S.L. and Motoyoshi, Y., 1999. Al-zoning in orthopyroxene in a sapphirine quartzite: evidence for >1120°C UHT metamorphism in the Napier Complex, Antarctica and implications for the entropy of sapphirine. Contributions to Mineralogy and Petrology, in press. Motoyoshi, Y. and Ishikawa, M., 1997. Metamorphic and structural evolution of granulites from Rundvagshetta, Lutzow-Holm Bay, East Antarctica. In Ricci, C.A. (ed) "The Antarctic Region: Geological Evolution and Processes" Terra Antartica Publications, 65-72. Raith, M., Karmnakar, S. and Brown, M., 1997. Ultra-high temperature metamorphism and multistage decompressional evolution of sapphrinie granulites from the Palni Hill Ranges, southern India. Journal of Metamorphic Geology, 15, 379-400.

OROGENESIS IN THE OUTBACK

ELUCIDATING P-7-t-EVENT HISTORIES IN COMPLEX HIGH GRADE TERRAINS: LESSONS FROM ANTARCTICA S. L Harley Department of Geology & Geophysics, University of Edinburgh, Kings Buildings, West Mains Road, Edinburgh, Scotland EH9 3JW. email: sharley@glg.ed.ac.uk

Elucidation of the complex histories typical of many high-grade terrains is a major challenge to petrology, structural geology and geochronology. The complexity inherent in these terrains arises not only from the interplay between the many processes that may influence their evolution in any single tectonic event but also their propensity to be reworked and overprinted in subsequent tectonothermal episodes. Even where such overprinting is documented and characterised in terms of age, grade and structural style it is still difficult to extract unambiguous P-T-t information on the earlier events. Indeed, the attribution of particular assemblages and P-T conditions to specific events and ages may be uncertain even in terrains for which detailed geological and geochronological data are available. Two terrains in East Antarctica, the Rauer Islands and the Napier Complex, indicate the variety of geological and isotopic complexity possible in reworked high-grade terrains and illustrate some approaches that can be applied with varying degrees of success to unravel this complexity. These terrains, and others, demonstrate that there are key gaps in the present understanding of the processes affecting zircon U-Pb and other isotopic systems that may lead to considerable ambiguity in the interpretation of age data in terms of specific metamorphic events in such complexes. T H E R A U E R ISLANDS: M U L T I P H A S E R E W O R K I N G A N D A P R O B L E M A T I C U H T E V E N T

The Rauer Islands consists of both Archaean and Proterozoic crustal components. Its complex geological evolution involves deformation and magmatic events prior to 2800 Ma, polyphase reworking of Archaean crustal precursors during Proterozoic granulite facies events (Harley et al 1995; Kinny et al 1993) and further overprinting of Archaean and Proterozoic lithologies at 530-500 Ma by mylonite zones and potentially by amphibolite-granulite facies high strain zones (Sims et al 1994). Archaean relation can be unravelled in low-strain windows such as the Scherbinina Layered Complex (SLC), a poly deformed and metamorphosed Fe-tholeiite intrusive cut by several generations of deformed metabasite dykes. The SLC and three of these dyke generations are intruded by tonalitic sheets representative of Archaean felsic orthogneisses present throughout the NE Rauer Islands. U-Pb S H R I M P dating of magmatic zircons precisely define the magmatic age of the SLC to be 2844±6 Ma (Harley et al 1998). However, a meta-tonalite sheet that cuts the SLC preserves a zircon isotopic array dominated by oscillatory zoned grains with cores up to 3470±30 Ma old. These zircons are interpreted as inherited because of the persuasive field evidence for emplacement of the tonalite into the SLC, but in the absence of clear field relations it would be easy to incorrectly interpret this as the intrusive age of the tonalite itself and hence derive an entirely spurious early-mid Archaean geological history. A Proterozoic tectonothermal event has been inferred in the Rauer Islands from two types of syn- to latemetamorphic intrusives that yield 1030-1000 Ma zircon ages (Kinny et al 1993). However, as this event has not been detected in the zircon U-Pb data obtained on the Archaean gneisses the important question arises as to whether the Archaean and Proterozoic components shared a common late-Proterozoic history or were juxtaposed in an even younger event (Hensen and Zhou 1997). The observation that both the Archaean and Proterozoic gneisses show resetting of zircon at ca. 530 Ma may imply that this, rather than ca. 1000 Ma, is the age of the principal high-grade event. Such an interpretation requires that all the 1000 Ma zircons in each of the dated Proterozoic intrusives are inherited rather than magmatic, and is not preferred here as the ages are consistent with relative event sequence. Instead, the resetting seen the Archaean gneisses reflects Pb-loss within irregular compositional zones and healed cracks in older zircon, and is attributed to the effects of greenschist-amphibolite facies fluid infiltration. This overprint correlates with ca. 530-510 Ma higher-grade deformation and metamorphism evident in Proterozoic paragneisses exposed in areas to the south-west of the Rauer Islands, and records the amalgamation of the Rauer Islands with those areas during the East Gondwanian Pan-African orogeny. Given this complexity in overprinting events there is a high potential for ambiguity in the event context of specific P-T records obtained from gneisses in the Rauer Islands. This is exemplified by magnesian gneisses that preserve ultrahigh temperature (UHT) metamorphic assemblages including orthopyroxene + sillimanite ± quartz and garnet + orthopyroxene ± sillimanite. Mineral compositions (Harley 1998a) produce a peak P-T estimate, 11-12 kbar and 1033±30''C, that is far more extreme than obtained from most other granulites in the terrain. A UHT-ITD path is deduced from interpretation of reaction textures in which magnesian garnets are replaced by lower-pressure

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

equivalents such as sapphirine + orthopyroxene + cordierite. The pre-UHT P-T evolution in this case may be recorded by orthopyroxenes within garnet, which show rimward increases from 8 to 10.5 wt % AI2O3 that probably reflect heating over the interval 930-1030°C at 9-12 kbar. Although the final stages of ITD recorded in these rocks can be correlated with the 'normal' P - I history preserved in the Proterozoic components of the Rauer Islands, and hence with either 1000 Ma or 530 Ma events, presently available isotopic data (e.g. monazite; garnet Sm-Nd: Hensen and Zhou 1997) do not constrain whether the main UHT history is a vestige of Archaean tectonism or is simply an early stage of the youngest high-grade event to affect the terrain.

I I

B a

THE NAPIER COMPLEX: DO ZIRCON AGES CONSTRAIN THE UHT EVENT ? The Napier Complex, a very strongly deformed granulite terrain in which few cross-cutting relationships are preserved, is generally regarded as the best-documented example of UHT metamorphism. Although the occurrence of UHT metamorphism at 950-1050°C and 7-11 kbar, followed by near-isobaric cooling (IBC) is well established from mineral assemblages and thermobarometry (Harley and Black 1997), the age of the UHT metamorphism is controversial. This arises because complex Archaean evolution of the Napier Complex variably reset and affected the usually robust zircon U-Pb system. Initial felsic magmatism in the Napier Complex occurred at 3840-3770 Ma, and is recorded from felsic orthogneisses at Mount Sones and Gage Ridge. The 2980±9 Ma Proclamation Island chamockite in the northernmost part of the Complex may define the age of an early tectonothermal event that is of minor importance or entirely absent in the main area of UHT metamorphism, which has been interpreted to have occurred at ca. 2840 Ma on the basis of complex zircon populations in a granitic gneiss (2837±15 Ma) and a paragneiss (2822±22 Ma) (Harley and Black 1997). These authors recognise an event at 2480-2450 Ma, but attribute this to an upper-amphibolite to lower granulite facies deformation that post-dates and is unrelated to the IBC that followed the UHT event. In contrast, other workers interpret the 2480-2450 Ma U-Pb zircon ages, and ca. 2500 Ma Sm-Nd whole rock isochrons, to approximate the age of the UHT event. In this case, zircons older than ca. 2500 Ma must reflect pre-UHT material, including inherited grains in pre- or syn-metamorphic melts and detrital grains in paragneiss precursors. Interpretations of the timescale of IBC and deep-crustal residence, and hence models for the nature of the UHT metamorphism, critically depend upon clarification of the event significance of the zircon populations. New mineral geochemical approaches have to be adopted in order to resolve this issue. Imaging and SIMS analysis have been used to petrologically examine Napier Complex zircons. Zircons from a sapphirine-orthopyroxene quartzite show radiation-induced self-fracturing and fracture sealing on a sub-micron to 10 micron scale that leads to a "brain" texture. Cores (2480 Ma) define the minimum age of the "brain" texture: this is not likely to be the age of UHT metamorphism. Zircons in a migmatitic paragneiss exhibit (1) oscillatory zoned core domains, and (2) rim domains that are relatively unzoned. Type (1) domains show steep HREE enrichments and are interpreted to have formed in equilibrium with a broadly granitic partial melt, but not garnet. 2655 Ma is a minimum age for these domains as they too display sealed fractures. Type (2) domains preserve flat HREE consistent with equilibration with garnet. Zircon growth in these domains at ca. 2480 Ma could either have occurred with garnet under UHT conditions, or from garnet as it broke down during post-UHT retrogression. The latter interpretation is consistent with the enhanced development of the type (2) domains where biotite has formed from garnet. The recognition of 2655 Ma zircons in this paragneiss indicates that the late-Archaean evolution of the Napier Complex is more complicated than either of the present end-member interpretations suggest. Notwithstanding the problems and difficulties illustrated in this analysis of the geological records preserved in the Rauer Islands and Napier Complex, integrated petrological-isotopic approaches provide a far greater insight into the evolution of these and similarly complex terrains than was possible a decade ago. Further resolution of ambiguities in age attribution and clarification of P-T-f-event histories in such complex terrains will be greatly assisted by developments in process-oriented or petrological approach to zircon and other accessory phase geochronology, and ultimately by the direct dating of metamorphic minerals using isotopic systems with closure temperatures similar to peak conditions. REFERENCES Harley, S.L., 1998. Ultrahigh temperature granulite metamorphism (1050^C, 12 kbar) metamorphism and decompression in garnet (Mg70)-orthopyroxene-sillimanite gneisses from the Rauer Group, East Antarctica. Journal of Metamorphic Geology, 16, 541-562. Harley, S.L. and Black, L.P., 1997. A revised Archaean chronology for the Napier Complex, Enderby Land, from SHRIMP ion-microprobe studies. Antarctic Science, 9, 74-91. Harley, S.L., Snape, I. and Black, L.P., 1998. The early evolution of a layered metaigneous complex in the Rauer Group, East Antarctica: evidence for a distinct Archaean terrane. Precambrian Research, 89, 175-205.

OROGENESIS IN THE OUTBACK

Harley SL, Snape I, Fitzsimons ICW 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. and Zhou, B., 1997. East Gondwana amalgamation by Pan-African collision? Evidence from Prydz Bay, Eastern Antarctica. In: The Antarctic Region: Geological Evolution and Processes, Ricci, C.A. (ed.). Terra Publications 115-119. Kinny, P.D., Black, L.P. and Sheraton, J.W., 1993. Zircon ages and the distribution of Archaean and Proterozoic rocks in the Rauer Islands. Antarctic Science, 5,193-206. Sims, J.R., Dirks, P.H.G.M., Carson, C. and Wilson, C.J.L., 1994. The structural evolution of the Rauer Islands, East Antarctica: mafic dykes as passive markers in a composite Proterozoic terrain. Antarctic Science, 6, 379394.

I B

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

THE TEXTURAL AND RHEOLOGICAL EVOLUTION OF LONG-LIVED FAULT ZONES IN OROGENIC BELTS R. Holdsworth\ M. Stewart & J. Imber 1 Reactivation Research Grp, Dept of Geological Sciences, University of Durham, Durham DH1 3LE, UK 2 Geology Dept, University of Southampton, Southampton, S04 3ZH, UK 3 Fault Analysis Group, Dept of Earth Sciences, University of Liverpool, Liverpool L69 3BX, UK

Reactivated faults and shear zones exposed in the deeply exhumed parts of ancient orogenic belts present opportunities to study fault zone processes that operate at different depths in the continental lithosphere. They also provide important insights into fundamental, deep-seated weakening mechanisms that may account for their longlived movement histories and the apparent weakness of many modem fault zones such as the San Andreas system. Field- and laboratory-based studies of a series of reactivated structures in the North Atlantic region and elsewhere suggests that they exhibit two fundamental structural architectures: Sub-vertical anastomosing: broadly symmetrical fault rock distributions, with central localisation with time/shallowing depth; originate as strike-slip faults; narrow (<5km); fragmentary preservation; multiple overprinting; slivers. Example: Great Glen Fault, Scotland. Inclined asymmetrical: asymmetric fault rock distributions migrating towards footwall (reverse) or hangingwall (normal) with time/shallowing depth; originate as dip-slip faults; broad (>5km); continuous preservation possible. Example: Outer Hebrides Fault Zone, Scotland. All crustal-scale fault zones exposed in basement preserve up to three key fault rock assemblages: (a) Early mylonites-cataclasites-psuedotachylites - preserve a record of deformation across the initial, primary frictional-viscous transition (10-15 km); (b) 'Low-strain' cataclastic crush belts - preserve a record of brittle deformation and intial fluid influx in the upper part of the primary frictional-viscous transition (8-12 km); (c) Reactivated, 'high-strain' hydrated cataclasites-phyllonites - preserve a record of fluid-influenced deformation and retrogression across a reaction-modified frictional-viscous transition (5-10 km). In general, fault rock fabric and rheology are determined by the interplay of six main controlling factors: (1) depth of deformation; (2) protolith composition; (3) structure and scale of fracture networks; (4) fluid influx (composition, volume); (5) intensity of strain and associated metamorphism; (6) strain history (strain rates, reactivation). Our studies suggest that the majority of long-lived faults and shear zones are weak due to the development of shallow, reaction-modified frictional-viscous transitions. This finding emphasises the fundamental role played by fluids in determining the long term rheology and architecture of the crust and, ultimately, the lithosphere.

OROGENESIS IN THE OUTBACK

THE GEOLOGY OF THE ARTHUR LINEAMENT, TASMANIA 0. Holm I CODES-SRC, University of Tasmania, GPO Box 252-79, Hobart, Tas. 7001 Australia.

The Arthur Lineament is a NE-trending belt of medium to high strain Neoproterozoic rocks in northwestern Tasmania, covering an area 110 km long and up to 10 km wide (Gee, 1967). Several orogenies affected Tasmania during the Late Proterozoic to Middle Cambrian. The Wickham Orogeny is the oldest recorded event in northwestern Tasmania. The syn-kinematic intrusions of a 777 ± 7 Ma granitoid in the Bo wry Formation at the Rocky River in northwestern Tasmania and the King Island Granitoid with an age of 760 ± 1 2 Ma provide the probable age of the Wickham Orogeny (Turner et al, 1995, Turner et al, 1998). The subsequent 510 to 490 Ma Tyennan Orogeny, commencing in the early part of the Middle Cambrian featured crustal collision, and was associated with emplacement of ultramafic-bearing allochthons and high P, low T metamorphism (Turner et ai, 1998). The timing of the main deformation that affected the Arthur Lineament remains contentious. Berry (1994) regarded the lineament to be a feature of the 760-770 Ma Wickham Orogeny which represented a major detachment between the weakly folded upper thrust sheet (Rocky Cape Group) and the strongly deformed footwall (Burnie/Oonah Formation). The Bowry Formation is interpreted by Berry (1994) as an allochthonous block emplaced around 500 Ma. Alternatively, Turner et ai (1998) argued that based on K-Ar ages from the Bowry Formation of the Timbs Group in the Arthur Lineament, the main episode of deformation in the lineament occurred at 510 ± 10 Ma, as part of the Tyennan Orogeny. Further contention exists over whether the metamorphic boundaries are gradational and lithological boundaries are conformable, or whether the lineament represents a series of faulted slices of varying composition and metamorphic grade. Lithologies within the Arthur Lineament include the medium- to high-strain part of the Bumie/Oonah Formation in the east and felsic to mafic schists of the Timbs Group in the west. The low strain Bumie/Oonah Formation to the east of the Arthur Lineament increases in strain approaching the lineament to the west. It undergoes a transition from a lower greenschist facies quartzwacke, interlayered with sparsely-chloritic pelitic siltstone and mudstone to a high strain greenschist facies quartz-mica schist with multiple generations of syn- to post-deformational quartz veins. The Timbs Group has been correlated with the turbidites and volcanogenic metasediments of the Ahrberg Group, which occur along the southwest margin of the Arthur Lineament. Lithologies of the Timbs Group vary in composition and metamorphic grade. Greenschist, relict blueschist and amphibolite assemblages are evident. Various authors (Spry, 1964, Spiller, 1974, Green and Spiller, 1977, Turner and Bottrill, 1993) have described blue amphiboles from within the Bowry Formation of the Timbs Group, but retrograde greenschist mineral assemblages are dominant in the metamorphic complex. Schists are commonly characterised by the presence of syn-kinematic albite porphyroblasts with inclusions of epidote, tourmaline, and glaucophane. Other minerals present are quartz, white mica, carbonate, garnet, green-brown prograde biotite and retrograde chlorite. Amphibolites are predominantly retrogressed and variably foliated. Prograde amphibole is typically pseudomorphed by actinolite. Primary titanomagnetite has been deformed and subsequently pseudomorphed by sphene followed by fme-grained epidote. The amphibolites are predominantly magnetite-bearing, and have been derived from tholeiitic magmas (Turner and Bottrill, 1993). The Ahrberg Group is unconformable on the Rocky Cape Group to the west and is faulted against the Timbs Group to the east. It has been subject to prograde greenschist facies metamorphism, reflected in the chlorite dominant mineralogy of the Bemafai Volcanics and the phyllitic textures of the metasediments. There is evidence of two episodes of ductile deformation, and at least one episode of brittle deformation. The Timbs Group and the high strain Bumie/Oonah Formation have undergone at least three episodes of ductile deformation. D1 produced N-S trending, isoclinal folding. D2 has resulted in moderately plunging, 020° trending, tight folds that are approximately coaxial with Fl. Late shear bands striking 340 overprint D1 and D2 structures. High strain zones in the lineament indicate a dextral wrench sense of movement, and are distinguishable from lower strain zones by strong schistose textures.

ALICE SPRINGS, AUSTRALIA, 1999 REFERENCES

Berry, R.F., 1994. Tectonics of western Tasmania: Late Precambrian - Devonian. Geological Society of Australia Abstracts, 39, 6-8. Gee, R.D., 1967. The Proterozoic rocks of the Rocky Cape Geanticline. In: The Geology of Western Tasmania-A Symposium. University of Tasmania, Hobart (unpubL). Green, T.H. and Spiller, A.R., 1977. Blue amphibole from Precambrian metabasalts. Savage River, Tasmania. American Mineralogist, 62, 164-166. Spiller, A.R., 1974. The petrology of the Savage River iron ore deposit, Tasmania. B.A. (Hons) thesis, Macquarie University. Spry, A.H., 1964. Precambrian rocks of Tasmania, Part VI, the Zeehan - Corinna area. Papers Proceedings Royal Society of Tasmania, 98, 23-48. Turner, N.J., 1989. The Precambrian. In: Eds Burrett, C.F. and Martin, E.L. Geology and Mineral Resources of Tasmania. Geological Society of Australia, Special Publication, 15, 5-46. Turner, N.J., Black, L.P., and Kamperman, M., 1995. Pre-Middle Cambrian stratigraphy, orogenesis and geochronology in western Tasmania. Geological Society of Australia Abstract Series, 39, 51-56. Turner, N.J. and Bottrill, R.S., 1993. Blue amphibole in the Proterozoic to Cambrian Arthur Metamorphic Complex, northwest Tasmania. Report Mineral Resources Tasmania. 1993/26. Turner, N.J., Black, L.P., and Kamperman, M., 1998. Dating of Neoproterozoic and Cambrian orogenies in Tasmania. Australian Journal of Earth Sciences, 45, 789-806.

O R O G E N E S I S IN THE O U T B A C K

MECHANISMS OF LITHOSPHERIC RENEWAL ASSOCIATED WITH CONTINENTAL OROGENY G. A. Houseman Department of Earth Sciences, Monash University, Clayton VIC, 3168 (greg@earth. monash. edu. au)

Mountain building is generally driven by an episode of crustal thickening caused by convergence and often associated with continental collision. Structures in the crust record the deformation of the uppermost part of the lithosphere, but we must infer what has happened to the mantle lithosphere on the basis of physical reasoning and geophysical measurement. Presently active orogenic zones provide an important guide to the types of process that may occur in the subcontinental lithospheric mantle: the Tibetan Plateau, the Tien Shan, the Alboran Sea, Southern California, and New Zealand serve to illustrate different facets of a process which can generally be described as gravitational instability of the mantle lithosphere. Various mechanisms which could cause the removal of the continental mantle lithosphere have been described, including convective thinning of the lithosphere (Houseman et al., 1981), mantle delamination (Bird, 1979), and subduction of the mantle lithosphere (Beaumont et al., 1994). I briefly review these concepts and discuss the similarities and differences in their geological consequences. The convective thinning concept has recently been put on a firm dynamical basis (Houseman and Molnar, 1997, Molnar et al, 1998, Neil and Houseman, 1999), and can be used to provide quantitative predictions, whereas the application of the other two concepts has till now depended essentially on an imposed kinematic framework. Unlike the oceanic lithosphere, the mantle layer of the continental lithosphere in cratonic areas appears quite stable for long periods of time (Jordan, 1975). The stability may be due in part to the continental mande lithosphere being depleted and intrinsically buoyant (Griffin et al., 1998). Under the action of convergence driven by plate boundary stresses, however, there is clear evidence of lithospheric involvement in the mantle circulation. The episodic renewal of at least part of the subcontinental lithosphere thus occurs, despite the stabilising influence of a thick buoyant crust. Where it has occurred, it seems that the instability is triggered by localised thickening of the crust and lithosphere. Lithospheric renewal can occur, however, only if it is driven by gravity acting on an unstable stratification of density. The excess density of the lithsospheric mantle relative to the asthenosphere is caused by thermal contraction. The implied threshold for stability (small perturbations decay, whereas large perturbations grow catastrophically) strongly suggests that the deformation of the lithosphere is governed by a non-Newtonian viscosity. For nonNewtonian viscosity the growth of the mechanical instability is likely to be defeated by thermal diffusion until a significant departure from horizontal stratification is induced by localised convergence. Once a sufficiently large thickness perturbation has been induced, however, the growth follows a strongly non-linear path, which ultimately concludes in the catastrophic removal of a large section of the mantle lithosphere beneath the convergent zone. The hypothesis that dislocation creep in olivine determines the mechanical stability of the continental mantle lithosphere thus explains the localisation in space, and episodicity in time, of those regions affected by lithospheric instability. Of the three conceptual mechanisms listed above for deformation of the continental lithosphere in orogenic zones, only subduction of the mantle lithosphere does not necessarily lead to local lithospheric renewal. The consequences of convective thinning and delamination mechanisms may be contrasted, however. In the case of convective thinning the downwelling of the mantle is expected to occur directly beneath the region of thickest crust. As the instability grows, the mantle lithosphere from adjoining regions flows into the downwelling and there is eventually replacement of the mantle lithosphere as it is thinned and stretched, even though crustal convergence may be continuing. In contrast, delamination describes a process in which the mantle lithosphere is peeled away from the base of the crust as a coherent sheet, without necessarily undergoing major internal deformation. Hot asthenosphere moves into the gap between mantle and crustal layers as it opens. The locus of deformation migrates away from its starting point and the mantle lithosphere is progressively replaced as the delamination proceeds. The two mechanisms are not necessarily mutually exclusive, however. The interaction of crustal buoyancy with externally imposed convergence can cause the convective thinning instability to mimic delamination (Houseman et al., 1999).

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

The injection of heat into the lithosphere consequent on convective thinning or delamination mechanisms may be marked in the geological record by a high temperature - low pressure regional metamorphic event (e.g. Piatt et al., 1998). Magmatism produced by melting of the upper mantle and or lower crust is also likely to affect the area, and the heating of the lithosphere will cause an increase in heat flow and surface elevation. With the increase in elevation caused by isostasy comes an increase in gravitational potential energy which changes the local stress balance, pushing the lithosphere into horizontal extension (England and Houseman, 1989). Even though the region may continue to be affected by plate-boundary stresses that drive convergence on the perimeter, its interior may undergo extension. The techniques of teleseismic traveltime tomography have been applied to the problem of imaging the lithosphere and upper mantle in a number of actively deforming orogens. The images thus derived begin to constrain how the lithospheric renewal mechanism occurs. At best they show the distribution, in three dimensions, of regions with a relatively fast seismic velocity. These regions are intrepreted as segments of cold lithospheric mantle descending through the surrounding warmer asthenosphere. Tomographic cross-sections through the lithosphere and upper mantle beneath the Transverse Ranges of Southern California (Kohler, 1999) show a narrow sheet of cold lithosphere descending directly beneath the zone of crustal convergence caused by the big bend of the San Andreas Fault. Preliminary results from the South Island of New Zealand (Stern et al., 1998) also suggest downwelling of cold material beneath the Alpine Fault convergent zone. For both of these examples, the width of the convergent zone is small (of order 300 km) and convergence has been relatively short lived (of order 5 Myr), so the geometry of downwelling appears simple. For the Tibetan Plateau the lengthscale is an order of magnitude greater, and it is likely that the convective thinning event occurred in the Miocene (Molnar et al, 1993). Yet the distribution of P-wave velocity in the mantle beneath the plateau (McNamara et al, 1997) could be interpreted to show a pattern of lithospheric downwelling which persists today. Numerical experiments on the Rayleigh-Taylor instability and results from seismic tomography have thus combined in unexpected ways to help us understand the enigmatic process of lithospheric renewal in orogenic zones. REFERENCES

Beaumont, C., P. Fullsack, and J. Hamilton, Styles of crustal deformation caused by subduction of the underlying lithosphere, Tectonophys., 232, 119-132, 1994. Bird,P., Continental delamination and the Colorado Plateau, J. Geophys. Res., 84, 7561-7571, 1979. England, P.C., and G.A. Houseman, Extension during continental convergence with application to the Tibetan Plateau, J. Geophys. Res., 94, 17561-17579, 1989. Griffin, W.L., S.Y. O'Reilly, C.G. Ryan, O. Gaul, and D.A. lonov. Secular variation in the composition of subcontinental lithospheric mantle: geophysical and geodynamic implications, in Structure and Evolution of the Australian Continent, Eds. J. Braun, J. Dooley, B. Goleby, R. van der Hilst, and C. Klootwijk, Geodynamics Series, 26, 1-26,1998. Houseman, G., McKenzie, D.P. and Molnar, P., Convective instability of a thickened boundary layer and its relevance for the thermal evolution of continental convergent belts, J. Geophys. Res., 86, 6115-6132, 1981. Houseman, G., and P. Molnar, Gravitational (Rayleigh-Taylor) instability of a layer with non-linear viscosity and convective thinning of continental lithosphere, Geophys. J. Int., 128, 125-150, 1997. Houseman, G., E. A. Neil, and M.D. Kohler, Lithospheric instability beneath the Transverse Ranges of California, J. Geophys. Res., in review, 1999. Jordan, T., The continental tectosphere. Rev. Geophys. Space Phys., 13 1-12, 1975. Kohler, M.D, Coupled crust-mantle deformation beneath the San Gabriel Mountains in the Southern California Transverse Ranges, J. Geophys. Res., (in press) 1999. McNamara, D.E., W.R. Walter, T.J. Owens, and C.J. Ammon, Upper mantle velocity structure beneath the Tibetan Plateau from Pn travel-time tomography, J. Geophys. Res., 102,493-505, 1997. Molnar, P., P.C. England, and J. Martinod, Mantle dynamics, the uplift of the Tibetan Plateau, and the Indian monsoon, Rev. Geophys., 31, 357-396, 1993. Molnar, P., G. Houseman and C. Conrad, Rayleigh-Taylor instability and convective thinning of mechanically thickened lithosphere: effects of non-linear viscosity decreasing exponentially with depth and of horizontal shortening of the layer, Geophys. J. Int., 133, 568-584, 1998. Neil, E.A., and G.A. Houseman, Rayleigh-Taylor instability of the upper mantle and its role in intraplate orogeny, Geophys. J. Int., in press, 1999. Piatt, J.P., J.-I. Soto, M.J. Whitehouse, A.J. Hurford, and S.P. Kelly, Thermal evolution, rate of exhumation and tectonic significance of metamorphic rocks from the floor of the Alboran extensional basin. Western Mediterranean, Tectonics, 17, 671-689, 1998. Stem, T., P. Molnar, D. Okaya, and D. Eberhart-Phillips, P-wave delay data and the missing mantle beneath the

OROGENESIS IN THE OUTBACK

POLYMETAMORPHISM AND THE TECTONIC EVOLUTION OF THE DANBA DOMAL METAMORPHIC TERRAIN, SICHUAN PROVINCE, WESTERN CHINA 'M. H. Huang, 'l. S. Buick, 'l. S. Williams and 'R. Maas

^Department of Earth Sciences, La Trobe University, Bundoora, Vic. 3083, Australia ^Research School of Earth Sciences, The Australian National University, Canberra, ACT 0200, Australia

The Danba Domal Metamorphic Terrain (DDMT), within the eastern portion of the Songpan-Garze Orogenic Belt (western China; SGOB), is a major belt of polymetamorphosed and deformed Mesoproterozoic to Mesozoic sedimentary and volcanic rocks between the North China Craton, the Yangtze Block and the Tibetan Block. Although the SGOB is generally characterised by regionally extensive, low-grade Triassic flysch sediments, the DDMT consists of: Mesoproterozoic migmatized orthogneiss basement; Neoproterozoic-Triassic covering metasediments; and syntectonic granites, and reached much higher metamorphic grades. Isograds have been defined within the DDMT based on the distribution of index minerals in metapelites, namely biotite, garnet, staurolite, kyanite, and sillimanite. At the highest grades a zone of migmatization can also be delineated in the central northern portion of the DDMT. These isograds are locally concentric around the basement orthogneiss domes and grade generally increases towards the north. The higher-grade sillimanite zone locally truncates the trend of lower grade zones, suggesting that it might reflect a separate metamorphic event.

Three metamorphic events have been distinguished in the metapelites: (1) an earliest (Ml) Barrovian-style pro grade metamorphism up to mid-amphibolite facies grade (Grt+St+ Ilm+Ky+Mu+Qtz) that probably developed during Indosinian-Yanshanian (Triassic-Jurassic) southwards-directed shortening (Dl); (2) a subsequent (M2) hightemperature Sill±Kfs overprint associated with E-W compression, and possibly the emplacement of syntectonic (-110 Ma) dioritic-syenitic intrusions and the migmatization of the gneiss domes (D2), and (3) late stage, retrograde (M3) growth of Qtz+Ser+Chl in variably developed SE-NW oriented crenulation cleavages, and in Miocene NW trending fault and shear zones that developed during the Himalayan Orogeny (D3-D4). Garnet porphyroblasts from all grades have been examined by electron microprobe to determine their element partitioning. Normal growth zoning has been found in the garnet through kyanite zones, indicating prograde growth during the Ml. In contrast, garnet grains in the sillimanite zone are generally compositionally homogeneous, or record complex zoning patterns that most probably reflect a metamorphic overprint during M2. P-T conditions estimated from metapelites and amphibolites from the different zones using conventional thermobarometry and the THERMOCALC computer program are as follows: biotite zone (-475°C, -4.2 kbar), garnet zone (-530-580' C, -5-7 kbar); staurolite zone (-570-590' C, -6-7.7 kbar ); kyanite zone (-580-610" C, -6.6-8 kbar); and sillimanite zone (600-690" C, 5-6 kbar for non-Kfs assemblages). Temperatures within the sillimanite zone increase northwards within the DDMT. Pressure calculations on metapelites containing kyanite and sillimanite suggest a slight decrease in pressure from Ml (8-6 kbar) to M2 (-6 to 5 kbar), however it is unclear whether this difference is significant statistically. No garnet was found associated with the muscovite-poor, migmatitic Sil+Kfs assemblages, rendering them less useful for thermobarometry. However, it is inferred that these assemblages record temperatures in excess of ~700°C and formed through muscovite dehydration partial melting. This partial melting might have been responsible for the generation of voluminous pegmatites that occur throughout the high-grade portion of the terrain. The age of high-grade metamorphism in the DDMT has been investigated by U/Pb dating of monazite from metapelites using SHRIMP II. Monazite intergrown with foliation-defining muscovite and biotite from a kyanite zone sample yielded a main population age of -180 Ma. This compares well with K/Av ages of 154-180 Ma determined from illite for regionally extensive low-grade metamorphism in the SGOB outside of the DDMT. In contrast, monazite included in foliation-forming biotite from two samples in the supposedly younger sillimanite zone yielded slightly older main population ages of -185 and -195 Ma respectively. These ages are generally consistent with the emplacement ages of voluminous Indosinian S-type granitoids throughout the SGOB (210-190 Ma), but the oldest age (-195 Ma) is significantly older than monazite from the kyanite zone sample. These data suggest that Ml occurred at or before -180 Ma. However, the significance of -185-195 Ma monazite in the sillimanite-zone samples remains unclear. Either the M2 event that formed the sillimanite zone was progressive with Ml Barrovian metamorphism, or M2 occurred later, but was not associated with new monazite growth. In the northern DDMT, structures in sillimanite zone rocks appear to be synchronous with those developed in -110 Ma

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

syenite intrusions (Rb/Sr WR isochron). If this is the case, then the -185-195 Ma monazite ages have been preserved through a younger M2 overprint that reached at least -TOO^C. Unlike most metamorphic monazite, in which Pb/U is relatively uniform, the Pb/U apparent ages of monazite from one sillimanite zone sample and the kyanite zone sample (taken close to the sillimanite-in isograd) range down to 170 and 140 Ma respectively. Backscattered electron images of selected monazite grains show no evidence of growth zoning, but lower Pb/U tends to occur close to the grain margins. This might reflect partial (diffusional?) radiogenic Pb loss from the monazite during a post-crystallization, high-temperature thermal event, possibly the inferred younger sillimanite zone overprint. Post-Ml, coarse grained pegmatites emplaced throughout the high grade portions of the DDMT yield Rb-Sr mineral isochrons at -180 Ma and -100-120 Ma, consistent with the Ml and M2 events reflecting two separate events rather than a continuous evolution. Further investigations are under way to constrain the timing of M2. Throughout the DDMT, Rb-Sr studies of biotite and muscovite in metapelites record significantly younger ages. WR-biotite model Rb-Sr ages throughout the DDMT consistently yield ages of -26-33 Ma regardless of regional grade, whereas WR-muscovite model Rb-Sr ages show a considerably larger spread (33-130 Ma). These data may be consistent with either: a) a thermal event that was hot enough to reset biotite (T(. -300 °C) but not muscovite RbSr (Tc -500'C) isotope systems; or b) initially slow differential cooling during the Mesozoic, followed by rapid cooling and exhumation in the Oligo-Miocene. The isotopic resetting does not appear to be related to pervasive deformation. However, major strike slip fault zones associated with granite emplacement to the south west of the DDMT developed in the interval 10-20 Ma during the later stages of the Himalayan Orogeny. The DDMT is regarded as a product of poly metamorphism combined with early crustal thickening and top-to-thesouth decollement caused by the subduction of the South China block under Laurasia during the Indosinian Orogeny (Ml in the DDMT), followed by uplift induced by granitoid intrusion due to the E-W collision between the Tibet and South China Blocks, possibly during the Yanshanian Orogeny (M2). Further uplift or exhumation and cooling might have occurred during the Himalayan movement when convergence between the India Plate and Eurasia took place.

OROGENESIS IN THE OUTBACK

THE STAUROLITE ISOGRAD IN THE NINE MILE REGION, BROKEN HILL BLOCK, AUSTRALIA. EVIDENCE FOR MULTIPLE METAMORPHIC EVENTS ? S. Hunze\ A. B u y s c h \ K. Stuwe^ K. Ehlers^

1 Department of Earth Science Monash University Clayton, Vic, 3168, Australia 2 Institut fur Geologie und Palaontologie; Universitat Graz; Heinrichstr. 26; A-8010 Graz, Austria

In the Broken Hill Block, New South Wales, Australia, early studies interpreted observations on the metamorphic geology as evidence for a single metamorphic event that changes in grade across the terrain (Binns, 1964; Phillips and Wall, 1981). It has also long been recognised that lower grade metamorphic assemblages overprint the high grade peak assemblages in large parts of the Block (e.g. Edwards, 1958). However, only few studies (e.g. Hobbs et al., 1984) have attempted to correlate different parageneses with different thermal events and the lower grade overprint has often been interpreted as part of the cooling path from the high grade events. Aside from recent geochronological studies (which indicate thermal events in the Broken Hill Block around 1690 Ma, 1640 Ma, 1600 Ma, 1200 and 500 Ma; Page and Laing, 1992; Ehlers and Nutman, 1997; Nutman and Ehlers, 1997; Harrison and McDougall, 1981) three independent arguments indicate to us that the lower grade overprint is not related to the retrograde path from the high grade event but is related to an independent and possibly much later thermal event: 1.) The regional distribution of M3 isograds does not coincide with the regional distribtuion of M2 isograds. The isograds for the two events cross on a map scale. 2.) The lower grade metamorphic overprint, M3, occurred as a static mineral growth and is seperated from the previous high grade event, M2, by at least one deformation episode, D3 (Stiiwe and Ehlers, 1997). 3.) Previous workers have suggested, that the M3 assemblage is related to shear zone development during D4, which has long been recognised as a later and possibly much younger event. In order to clarify the first point, we have mapped out - on a 1:10000 scale - the distribution of M2 and M3 parageneses in a region where: (a) M2 and M3 isograds are at angles, according to the maps of Phillips and Wall (1981) and Stevens et al., (1988) and (b) where pelitic schists record thermodynamically low-variant assemblages of both events. This region is the Nine Mile region north of the Broken Hill township. There, we have focused on mapping the orientation of the staurolite isograd but also have mapped out other mineral isograds that can be traced in the field. In this contribution we present the results from this mapping study. While the high resolution of the mapping for this project prevented us from extending the study to a regionally relevant part of the Broken Hill Block, we were able to show that different aluminous phases in the pelitic schists have different spatial distributions. Correlation of bulk compositional variations with mineralogical variations shows that this cannot be attributed to variations in bulk composition alone. We therefore speculate that our results are related to different thermal focuses of independent thermal events.

This contribution is part of FWF project P-12846-GEO. REFERENCES

Binns, R.A., 1964. Zones of progressive regional metamorphism in the Willyama Complex, Broken Hill district, New South Wales. Journal of the Geological Society of Australia, 11, 283-330. Edwards, A.B., 1958. Amphibolites from the Broken Hill district. Journal of the Geological Society of Australia, 5, 1-32. Ehlers, K. and Nutman, A.P., 1997. Thermochronological evolution of the Willyama Complex. Geodynamics and ore deposits conference abstract volume, Australian Geodynamics Cooperative Research Centre, Ballarat, 2628.

Harrison, M.T. and McDougall, L, 1981. Excess 40Ar in metamorphic rocks from Broken Hill, New South Wales:implications for 40Ar/39Ar age spectra and the thermal history of the region. Earth and Planetary Science Letters, 55,123-149 Hobbs, B.E., Archibald, N.J., Etheridge, M.A. and Wall, V.J., 1984. Tectonic history of the Broken Hill Block, Australia. In: Precambrian Tectonics Illustrated. Eds Kroner, A. and Greiling, R. Schweizerbart'sche Veriagsbuchhandlung, Stuttgart, 353-368. Nutman, A.P. and Ehlers, K., 1997. Evidence for multiple Paleoproterozoic thermal events and magmatism adjacent

A L I C E SPRINGS, AUSTRALIA, 1999

to the Broken Hill Pb-Zn-Ag orebody, Australia. Precambrian Research. Page, R.W. and Laing, W.P., 1992. Felsic Metavocanic Rocks related to the Broken Hill Pb-Zn-Ag Orebody, Australia. Geology, depositional age and timing of high-grade metamorphism. Economic Geology, 87, 21382168. Phillips, G.N. and Wall, V.J., 1981. Evaluation of prograde regional metamorphic conditions: their implications for the heat source and water activity during metamorphism in the Willyama Complex, Broken Hill, Australia. Bulletin de Mineralia, 104, 801-810. Stevens, B.P.J., Barnes, R.G., Brown, R.E., Stroud W.J., and Willis, I.L., 1988. The Willyama Supergroup in the Broken Hill and Euriowie Blocks, New South Wales. Precambrian Research, 40-41. 297-327. Stiiwe, K. and Ehlers, K., 1997. Multiple metamorphic events at Broken Hill, Australia. Evidence from chloritoidbearing parageneses in the Nine-Mile region. Journal of Petrology, 38, 1167-1186.

OROGENESIS IN THE OUTBACK

SOME REGIONAL IMPLICATIONS OF NEW GEOCHRONOLOGICAL CONSTRAINTS FROM THE TENNANT CREEK AND ARUNTAINLIERS, CENTRAL AUSTRALIA. K.J. Hussey\ J. Smith^ and N. Donnellan^ 1. Northern Territory Geological Sur/ey, PO Box 2655, Alice Springs, NT 0871. 2. Australian Geological Sun/ey Organisation, GPO Box 378, Canberra, ACT 2601.

NTGS and AGSO recently commenced a geochronology project to further investigate structural and stratigraphic relationships and correlations within and between the Arunta and Tennant Creek Inliers. The first samples dated document a newly recognised period of igneous activity at -1805 Ma in the northern-most Arunta Inlier; suggest that folding in the Davenport province is older than previously thought; broadly confirm correlations between units in the Davenport province; and preclude previously suggested correlations between the southern and northern Tennant Creek Inlier. In central-northern Barrow Creek map sheet, a volcano-sedimentary succession crops out as the Osborne and Taylor Ranges. These rocks have kilometer-scale, tight to isoclinal, upright to inclined folds with northwest-trending axes. These, and more recessive rocks immediately surrounding the Ranges, were correlated with the Wauchope and Ooradidgee Subgroups of the Davenport province by Haines et al (1991). Notwithstanding lithological similarities, and geophysical interpretations which link this region with the Davenport province, Haines et al (1991) erected a local stratigraphy and defined new units. The Strzeleckie Volcanics were selected to test the stratigraphic link with the Wauchope Subgroup in the main part of the Davenport province. In the Osborne and Taylor Ranges the dacitic to rhyolitic Strzeleckie Volcanics interfingers with, and is disconformably overlain in the northeastern part of the Ranges by, the epiclastic Tinfish Sandstone. The Strzeleckie Volcanics and the Tinfish Sandstone were mapped as conformably overlying the Gwynne Sandstone, the local basal unit of the Wauchope Subgroup, and are conformably to disconformably overlain by the Illoquara Sandstone. Disconformable relationships have also been recognised within the Wauchope Subgroup of the main outcrop area of the Davenport province. However, these disconformities are probably localised and of limited regional significance because Blake et al (1987) report an overall conformity within the Subgroup in the main part of the Davenport province. SHRIMP U-Pb dating of zircons from an ignimbrite near the base of the Strzeleckie Volcanics indicates it erupted at 1819 ± 9 Ma. This confirms it is broadly contemporaneous with other dated extrusive igneous rocks within the Wauchope Subgroup, and high-level felsic sills in the Davenport province, although the Treasure Volcanics, upper Ooradidgee Subgroup, is also similar in age (Ozchron database; RW Page, pers. comm.). In the extreme north of the Davenport province only volcaniclastic sedimentary rocks are present at this stratigraphic interval. On the basis of this it appears that subaerial felsic igneous activity at around 1820 Ma was concentrated in the southern parts of the Davenport province. Igneous rocks of this age also occur much further to the south, in the northern Arunta Inlier (Collins and Willams, 1995). Two granite samples were selected from the northern Arunta Inlier for dating purposes because they appear to constrain regional deformation in Barrow Creek. The medium to very coarse grained porphyritic and enclavebearing Ooralingie Granite shows a well developed subvertical foliation and is clearly intruded by the relatively undeformed Bean Tree Granite. These granites are part of the Barrow Creek Granitic Complex which intrudes greenschist to lower amphibolite facies Bullion Schist. SHRIMP U-Pb dating of zircons from the Bean Tree and Ooralingie Granites indicates magmatic ages of 1803 ± 6 Ma and 1809 ± 5 Ma respectively. The closeness of these ages suggests that the Ooralingie Granite was probably syn-tectonic and the Bean Tree Granite immediately posttectonic with respect to the same tectono-magmatic event. The northwest-trending foliation in the Ooralingie Granite is pervasive, and of similar metamorphic grade to the country rocks. Further, the foliation appears to parallel the trend of large-scale folds in central-northern Barrow Creek. It therefore seems plausible that the Ooralingie Granite, Bullion Schist and the Ooradidgee and Wauchope Subgroups were deformed immediately prior to emplacement of the Bean Tree Granite. Thus, large-scale upright folding in the Davenport province appears to have occurred around 1805 Ma, between the 1820 Ma Stafford tectonic event and the 1770 Ma Strangways orogeny of Collins and Shaw (1995) in the Arunta Inlier.

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

Another implication of recent dating is that the lithostratigraphic correlations between the Davenport province and the northern-most Tennant Creek Inlier suggested by Blake (1984) are not correct. Compston (1995) indicated a maximum age of 1784 ± 7 Ma for the Hayward Creek Formation, basal unit of the Tomkinson Creek Subgroup in the northern Tennant Creek Inlier. This shows that the Tomkinson Creek Subgroup is much younger than the 1820 Ma Wauchope Subgroup, and also, presumably, younger than the Hanlon Subgroup which conformably overlies the Wauchope Subgroup in the Davenport province (Blake et aL, 1987). A maximum age of approximately 1785 Ma has been reported for conglomerate at the base of the Reynolds Range Group in the Arunta Inlier (unpublished data referred to by Collins and Williams, 1995). This implies that this group is younger than the Wauchope Subgroup, and might correlate with the basal units of the Tomkinson Creek Subgroup. These constraints indicate that largescale northwest-trending folds in the Tomkinson Creek Subgroup may be related to a younger event than that which produced the northwest-trending folds in the Wauchope Subgroup, and should overprint the later. We would like to thank David Young, David Blake and Rod Page for their constructive reviews and comments. Published with the support and approval of the Director of NTGS and Executive Director of AGSO. REFERENCES Blake, D.H., 1984. Stratigraphic correlations in the Tennant Creek region, central Australia: Warramunga Group, Tomkinson Creek beds, Hatches Creek Group, and Rising Sun Conglomerate. Bureau of Mineral Resources Journal of Australian Geology and Geophysics, 9, 41-47. Blake, D.H., Stewart, A.J., Sweet, LP., and Hone, I.G., 1987. Geology of the Proterozoic Davenport province, central Australia. Bureau of Mineral Resources, Geology and Geophysics, Bulletin 226. Collins, W.J. and Shaw, R.D., 1995. Geochronological constraints on orogenic events in the Arunta Inlier. Precambrian Research, 71, 315-346. Collins, W.J. and Williams, I.S., 1995. SHRIMP ionprobe dating of short-lived Proterozoic tectonic cycles in the northern Arunta Inlier, central Australia. Precambrian Research, 71, 69-89. Compston, D., 1995. Timing constraints on the evolution of the Tennant Creek Block, northern Australia. Precambrian Research, 71, 107-129. Haines, P.W., Bagas, L., Wyche, S., Simons, B., and Morris, D.G., 1991. Barrow Creek, Northern Territory 1:250 000 Geological Series, Northern Territory Geological Survey, Explanatory Notes, SF53-6.

OROGENESIS IN THE OUTBACK

THERMAL-MECHANICAL MODELS OF LARGE CONVERGENT OROGENS R.A. Jamieson\ C. Beaumont^, O. Vanderhaeghe^'^, and P. Fullsack^ ^Department of Earth Sciences, Dalhousie University, IHalifax, N.S., Canada, B3H 3J5 ^Department of Oceanography, Dalhousie University, IHaiifax, N.S., Canada, B3H 4J1

Coupled thermal-mechanical models of orogens are designed to investigate interactions between tectonic processes including deformation, erosion, heat transport, and metamorphism. A number of recent numerical experiments indicate that the thermal evolution of orogens is closely linked to accretion of heat-producing material in the orogenic crust (e.g., Huerta et al. 1998, Jamieson et al. 1998). Both the volume and spatial distribution of heatproducing material play importnat roles. The maximum temperatures are limited by the competition between heat production, heat loss by diffusion, advection of cold material into or beneath the orogen, and redistribution of hot material within the orogen. These trade-offs can be quantified using dimensionless Peclet and Damkohler numbers, which represent ratios between advection, diffusion, and heat production. Model orogens including a substantial volume of tectonically accreted radioactive material ("tarm") at crustal levels reach P-T conditions compatible with Harrovian regional metamorphism. In large orogens, where a significant proportion of the crust consists of "tarm", conditions appropriate for partial melting can be achieved. Field observations, laboratory experiments, and conceptual models suggest that the presence of melt should substantially weaken the deforming crust. This in turn should affect the tectonic style, for example leading to plateau formation and/or syn-orogenic extension. Some preliminary results are presented from coupled thermal-mechanical models designed to investigate the role of partial melting in large convergent orogens. In the models, upper crustal "tarm" is incorporated into the orogen whereas the lower crust is subducted along with the upper mantle. P-T conditions necessary for partial melting are achieved in the lower orogenic crust after about 30 My of convergence. High temperatures in the lower model crust are accompanied by the development of an orogenic plateau, but no extension or crustal thinning is observed. The tectonic style of the hot, weak model orogen appears to be strongly influenced by the adjacent cool, strong, nonorogenic crust. In the models investigated so far, melting and accompanying rheological weakening are not sufficient to drive in-plane extension; a rapid change in boundary conditions, such as a reduction in convergence rate, is also required. Work is in progress to extend the results of these preliminary coupled thermal-mechanical models to conditions more appropriate to large natural orogens. REFERENCES

Huerta, A.D., Royden, L.H., and Hodges, K.V. 1998. The thermal structure of collisional orogens as a response to accretion, erosion, and radiogenic heating. Journal of Geophysical Research, 103, 15287-15302. Jamieson, R.A., Beaumont, C., Fullsack, P., and Lee, B. 1998. Barrovian regional metamorphism: where's the heat? In: Treloar, P.J., & O'Brien, P.J. (edsj "What drives metamorphism and metamorphic reactions?" Geological Society of London Special Publication 138, 23-51.

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

THE GRENVILLIAN OROGENIC CYCLE: CONSTRAINTS FROM A GEOLOGICAL TRANSECT ACROSS ONTARIO AND NEW YORK R.A. Jamieson\ N.G. Culshaw\ S.M. Carr^, R.M. Easton^ & J.W.F Ketchum^ ^Department of Earth Sciences, Dalhousie University, Halifax, NS, Canada, B3H 3J5; ^Department of Earth Sciences, Carleton University, Ottawa, ON, Canada, K1S 5B6; ^Ontario Geological Survey, 933 Ramsay Lake Road, Sudbury, ON, Canada, P3E 6B5; "^Department of Earth Sciences, Memorial University, St. John's, NF, Canada, A1B 3X5

The Grenvillian orogenic cycle (Wynne-Edwards, 1972; Rivers et al. 1989) spans the range ca. 1190-960 Ma. The scale and generally high metamorphic grade of Grenvillian rocks now at the surface suggests that at its peak it was a mountain belt of Himalayan proportions. Three orogen-wide tectonic episodes are recognized - ca. 1190-1140 Ma, ca. 1090-1020 Ma, and ca. 1000 Ma. It is generally agreed that the orogeny involved collision between Laurentia and one or more continental and/or magmatic arc terranes that originally lay to the southeast. There is much less agreement on the nature, timing, and geologic significance of the collisional event(s) and whether the orogeny was continuous or episodic. Variations in style along the length of the orogen may in part reflect variations in the nature of the Laurentian crust dating from earlier tectonism (ca. 1900-1400 Ma) associated with long-lived Mesoproterozoic subduction beneath the Laurentian margin (e.g.. Rivers, 1997). Recent syntheses of geological and geophysical data from Ontario and New York (Fig. 1 of Culshaw et al. this volume; White et al. 1994; McLelland et al. 1996; Culshaw et al. 1997; Carr et al. in press) provide some constraints on the nature of the Grenvillian orogenic cycle at the western end of the orogen. The geology of the region is described in terms of three contrasting tectonic divisions (Table 1): Laurentian crust and margin, a Composite Arc Belt (CAB) formed and assembled southeast of Laurentia, and the enigmatic Frontenac- Adirondack Belt (FAB) that may represent a distinctive part of the CAB or an offshore continental terrane. In an attempt to avoid some unresolved problems of nomenclature, correlation, and interpretation, the tectonic evolution of the orogen is described in terms of five stages, G1 through G5. Gl(1290-1220 Ma): This event, restricted to the CAB, involved the construction and amalgamation of volcanic arcs. A progression from tholeiitic to calc-alkaline to bimodal volcanism may represent a change from primitive to mature arcs followed by back-arc rifting. Amalgamation of arcs at 1245-1220 Ma (Elzevirian orogeny) was accompanied by widespread granitoid magmatism and variable but relatively subdued regional metamorphism and deformation. Evidence for links between Laurentia, CAB, and FAB at this time is tenuous. G2 (1190-1140 Ma): This is the earliest episode of tectonism normally included in the Grenville orogenic cycle. It has been equated with the waning stages of the Elzevirian orogeny (Gl) but significant differences in character, age, and distribution suggest that it should be considered separately (e.g.. Rivers, 1997). Geological features of this age do not fit easily into a coherent tectonic framework. Although this event is widely attributed to continental collision involving Laurentia (e.g.. Rivers et al. 1989; Rivers, 1997), there is little evidence that Laurentian crust in the study area was affected by tectonism at this time (Culshaw et al. 1997). In contrast, parts of the CAB and FAB were affected by granulite facies metamorphism and intense deformation, and it is likely that these two belts were contiguous by ca. 1160 Ma (Carr et al. in press). Anorthosite was emplaced in some of the affected regions during this interval. We suggest that this event involved convergence within and/or between CAB and FAB, offshore or at the distal edge of Laurentia. G3 (1140-1090 Ma): This was a period of relative quiescence within the western part of the orogen. However, an episode of ca. 1120 Ma tectonism that affected both allochthonous CAB and underlying Laurentian margin rocks may represent the intial encounter between Laurentia and offshore terranes. The Midcontinent Rift system (MCR) formed at 1109-1094 Ma. The main phase of rifting coincided with a gap in metamorphic ages within the orogen, and the end of rifting coincided with the onset of the Ottawan orogeny (G4), suggesting that the MCR either influenced, or was influenced by, the Grenville orogeny. Thermal effects related to the MCR probably affected the style of deformation during the subsequent Ottawan orogeny (G4). G4 (1090-1020 Ma): Widespread deformation and metamorphism, generally termed the Ottawan orogeny, affected Laurentia and parts of the CAB and FAB beginning at ca. 1090 Ma. This is interpreted as the major collisional event in the Grenvillian orogenic cycle. Northwest-directed thrusting affected the entire orogen during this time. In the transect area, high pressure metamorphism at ca. 1090-1085 Ma was followed by deformation and

OROGENESIS IN THE OUTBACK

metamorphism that propagated towards the Laurentian foreland from ca. 1080-1035 Ma, with regionally significant extension at ca. 1020 Ma (Culshaw et al., 1997; this volume). G5 (1010-980 Ma): Thrusting and metamorphism affected the Grenville Front Tectonic Zone (GFTZ ) at ca. 1000 Ma, followed by exhumation and rapid cooling. This has been attributed to a final stage of convergence and propagation of the orogen into its foreland. The location of the GFTZ in the transect area may be related to the presence of pre-Grenvillian tectonic boundaries, the limit of MCR-related thermal effects, or both. REFERENCES Carr, S.D., Easton, R.M., Jamieson, R.A., & Culshaw, N.G. (in press). Geologic cross-section across the Grenville orogen of Ontario and New York. Canadian Journal of Earth Sciences. Culshaw, N.G., Jamieson, R.A., Ketchum, J.W.F., Wodicka, N., Corrigan, D. and Reynolds, P.H. 1997. Transect across the northwestern Grenville orogen, Georgian Bay, Ontario. Tectonics, 16, 966-982. Culshaw, N.G., Jamieson, R.A., Ketchum, J.W.F., Wodicka, N., Corrigan, D. and Reynolds, P.H. this volume. Polystage ductile thrusting and extension in the lower orogenic crust: western Grenville Province, Georgian Bay, Ontario. Easton, R.M. 1992. The Grenville Province and the Proterozoic history of central and southern Ontario. In Geology of Ontario, edited by P.C. Thurston et al., Ontario Geological Survey, Special Volume 4, 715-904. McLelland, J.M., Daly, J.S., and McLelland, J.M. 1996. The Grenville orogenic cycle (ca. 1350-1000 Ma): an Adirondack perspective. Tectonophysics, 265, 1-28. Rivers, T. 1997. Lithotectonic elements of the Grenville Province: review and tectonic implications. Precambrian Research, 86, 117-154. Rivers, T., Martignole, J., Gower, C.F., and Davidson, A. 1989. New tectonic subdivisions of the Grenville Province, southeast Canadian Shield, Tectonics, 8, 63-84. White, D.J., Easton, R.M., Culshaw, N.G., Milkereit, B., Forsyth, D.A., Carr, S.D., Green, A.G. and Davidson, A. 1994. Seismic images of the Grenville orogen in Ontario. Canadian Journal of Earth Sciences, 31, 293-307. Wynne-Edwards, H.R. 1972. The Grenville Province. In Variations in Tectonic Styles in Canada, edited by R.A. Price and R.J.W. Douglas, Geological Association of Canada, Special Paper 11, 263-334. Table 1. Tectonic divisions, western Grenville orogen (see also Fig. 1, Culshaw et al., this volume) Poly/mpno-cyclic Belts Other Widely Used Names Tectonic Divisions (Rivers et al. 1989) (Carr et al. in press) (Wynne-Edwards 1972; Easton 1992) autochthonous polycyclic Laurentian foreland Southern Province, Superior Province Neoarchean to Mesoproterozoic Laurentian crust parautochthonous polycyclic Grenville Front Tectonic Zone (GFTZ) reworked ca. 1060-1000 Ma northern Central Gneiss Belt (CGB) distal Laurentian crust and margin southern Central Gneiss Belt (CGB) allochthonous polycyclic + with CAB allochthons (various domains recognized) and allochthonous monocyclic reworked ca. 1090-1020 Ma CMB boundary thrust zone (CMBBZ) Composite Arc Belt (CAB) Central Metasedimentary Belt (CMB) allochthonous monocyclic formed and assembled (various terranes recognized) ca. 1290-1220 Ma allochthonous monocyclic Frontenac-Adirondack Belt (FAB) Frontenac terrane (part of CMB) Central Granulite Terrane amalgamated with CAB by (Adirondack Lowlands and Highlands) ca. 1160 Ma

A L I C E SPRINGS, AUSTRALIA, 1 9 9 9

FLUID-ROCK INTERACTION AT c. 1000 Ma AND c. 500 Ma IN DRONNING MAUD LAND, EAST ANTARCTICA: A STABLE ISOTOPE STUDY OF THE SVERDRUPFJELLA GROUP. W.P. J o h n s t o n e and C. Harris Department of Geological Sciences, University of Cape Town, Rondebosch 7700, South Africa

Much of the high-grade East Antarctic coastal mobile belt appears to have developed at 550-500 Ma (e.g. Fitzsimons, 1998). In the Kiwanveggen and H.U. Sverdrupfjella mountain ranges of western Dronning Maud Land, a high-grade metamorphic event at between 1200-900 Ma with a thermal overprint at -500 Ma has been recognized (Groenewald et al., 1995). A series of altered basalts, the Sistenup lavas, which post date the high grade metamorphic basement rocks, exhibit ^^0-depletion as a result of interaction with a high temperature fluid. We have determined the oxygen isotope composition of whole-rock gneisses, quartz and quartz-calcite veins, as well as calcsilicate rocks from the Sverdrupfjella Group, in order to understand the role and relative timing of fluid-rock interaction. Two types of calcite were recognised in calc-silicate rocks in the field: (i) calcite associated with early scattered boudins, and (ii) calcite in metasomatic alteration bodies associated with late shear zones. The boudins range in and from 9.6 to 12.8%c and -11.1 to -6.0%o, respectively. The low values can best be explained as the result of decarbonation during early (1200-900 Ma) high-grade regional metamorphism. The calcite associated with metasomatic alteration bodies ranges in and from 10.6 to 14.9%o and -1.7 to -i-2.6%o, respectively. The large difference in values between the two types of calcite shows that the earlier boudins cannot be the source of carbon for the later metasomatic bodies. The earlier calc-silicate boudins record an older, more complete metamorphic history in contrast to the late metasomatic calc-silicates which probably record a younger, lower temperature and pressure metamorphic event. Similar carbonates from the Rauer Group of East Antarctica (Buick et al, 1994) show a correlation between and which can only be explained by substantial exchange between rock and water during decarbonation. These trends are absent in the Kirwanveggen boudins. The rocks of the Sverdrupfjella Group are dominated by a variety of ortho- and para-gneisses 7.0 to 12.0%o) with rare calc-silicate rocks 10.7 to 13.9%o) and late amphibolite dykes 6.3 to 7.6%o). The average value of the dominant gneisses in the Kirwanveggen is 8.6%o, whereas in the H.U.Sverdrupfjella it is 7.3%o. Quartz veins of a variety of ages in the Kirwanveggen range in value from 5.2 to I I M c (average 8.5%o). Under closed system "rock-buffered" conditions, the values of quartz veins should be greater than the bulk value of the rock. The lower than expected values of selected quartz veins indicates non-equilibrium, fluiddominated conditions. A notable feature of the host rocks is the generally low values of the felsic gneisses, and in particular the low average value (7.3%c) of the granite gneiss in the H.U.Sverdrupfjella. This may indicate low original values of the protolith (i.e. essentially mantle-derived), but an alternative explanation is that values were lowered by exchange with a fluid during metamorphism. Oxygen isotope data from samples collected across late reactivated shear zones in the Kirwanveggen indicates the presence of high temperature fluids. Regional aeromagnetic anomalies identified by Comer (1991) highlight the extent of these shear zones whereas the presence of magnetite attests to substantial fluid flow within these zones. High temperature fluid-rock interaction identified in the Sistenup lavas could be the result of the same late fluidizing event possibly at -500 Ma. The combination of aeromagnetic and stable isotope data used to interpret these shear zones is providing a better understanding of the role of fluids in the metamorphic evolution of the Sverdrupfjella Group. REFERENCES

Buick, I.S., Harley, S.L., Cartwright, I., Mattey, D., 1994. Stable isotopic signatures of superposed fluid events in granulite facies marbles of the Rauer Group, East Antarctica. Journal of Metamorphic Geology. 12, 285-299. Comer, B. and Groenewald, P.B., 1991. Gondwana reunited. South African Journal of Antarctic Research. 21, 172. Fitzsimons, I.C.W., 1998. Early Cambrian tectonism in East Antarctica: implications for Gondwana assembly and earlier supercontinents. Special abstracts issue Gondwana 10: Event stratigraphy of Gondwana. Journal of African Earth Sciences, 27, 74-75. Groenewald, P.B., Moyes, A.B., Grantham, G.H., Krynauw, J.R., 1995. East Antarctic cmstal evolution: geological constraints and modelling in westem Dronning Maud Land. Precambrian Research. 75, 231-250.

O R O G E N E S I S IN T H E O U T B A C K

CYCLICITY IN OROGENSIS RECORDED BY MULTIPLE EPISODES OF ZIRCON GROWTH: AN ION MICROPROBE (SHRIMP) STUDY OF THE ALPINE EVOLUTION OF THE CYCLADES, GREECE. Sue Keay^*, Gordon Lister^ Ian Buick^ 1RSES, Australian National University, Canberra ACT 0200, Australia 2VIEPS, Dept. Earth Sciences, Monash University, Clayton, VIC 3168, Australia 3Dept Earth Sciences, LaTrobe University, Bundoora, VIC 3083, Australia *current address: Dept. Earth Sciences, University of Queensland, Brisbane QLD 4072,

Australia

The tectonic evolution of the Cyclades, Greece, during the Alpine orogeny consisted of polyphase metamorphism, deformation, fluid infiltration, anatexis and shearing. Part of this tectonic history is recorded by the development of new^ zircon rims on pre-existing grains in Cycladic rocks. These overgrowths (generally < 30 microns wide) can be identified by cathodoluminescence imaging and dated using a technique that offers high spatial resolution (SHRIMP). A complex pattern of U-Pb ages is revealed spanning the entire Alpine evolution of the Cyclades from the Cretaceous to the mid-Miocene. Significant age populations can be objectively defined using a mixturemodelling procedure in conjunction with age probability distribution diagrams. As zircon development occurs in response to external factors that are likely to broadly correlate with tectonic events, combining these ages allows some broad generalisations to be made about the relationship between the formation of metamorphic zircon overgrowths and tectonic events in the Cyclades. Clusters of ages in the ranges 140-120, 105-95, 80-75, 66-62, 56-52, 50-45, 44-39, 30-27 and 19-16 Ma are identified suggesting that these were discrete periods of active tectonism in the Cyclades during the Alpine orogeny. These periods of new zircon development are difficult to relate to P-T conditions as they occur in response to processes that have affected the host rock that are no longer visible in the main mineral assemblage due to overprinting by younger events. However, as most of the Cycladic rocks have not experienced temperatures in excess of - 600 "C, and/or occur in lithologies that did not melt (e.g., metacarbonate rocks), the formation of new zircon must be related to sub-solidus processes, such as fluid infiltration. Stable isotope evidence suggests that zircon growth in Cycladic rocks can be related to hydro thermal activity, which is facilitated by the geochemical nature and the degree of deformation of the host lithology. The strong cluster of ages at 80-75 Ma corresponds to the timing of ophiolite formation on Syros and also the timing of high-T metamorphism in the Cycladic Upper Unit. Ophiolite formation in the Cyclades was accompanied by sea-floor alteration prior to early Tertiary subduction/collision, producing a crustal sequence containing large quantities of water in the form of chlorite and clay minerals that would devolatilise during high-P metamorphism. The significant period of zircon growth between 50-45 Ma is interpreted as recording fluid infiltration during the transition from eclogite to blueschist facies metamorphism and is consistent with estimates of the timing of high-P metamorphism from other isotopic systems. Retrogression to greenschist facies conditions and further fluid influx through the breakdown of hydrous high-P minerals is recorded by zircon ages of 30-27 Ma. Localised migmatisation of metapelitic rocks and orthogneisses and the development of new zircon during partial melting occurred at 19-16 Ma. The youngest layers of zircon growth occur in metacarbonate rocks at 14-13 Ma and are interpreted to reflect fluid flow associated with the intrusion of post-peak metamorphism granitoids. These youngest zircon ages correlate with SHRIMP U-Pb ages obtained from monazite and titanite from related Cycladic samples, all of which are interpreted to have formed in response to shearing related to extension post-peak metamorphism. The reproducibility of ages from different samples from different areas of the Cyclades and their consistency with other geological evidence suggests that zircon can be used to constrain the tectonic history of complicated orogenic belts.

ALICE SPRINGS, AUSTRALIA. 1999

PROGRESSIVE LOW-PRESSURE METAMORPHISM IN NORTHEASTERN YEONGNAM MASSIF, KOREA J . K i m ^ ' ^ a n d M. Cho^

^Department of Geological Sciences, Seoul National University, Seoul, 151-742, Korea ^Isotope Research Team, Korea Basic Science Institute, Yusung-Gu, Taejon, 305-333, Korea

The Early Proterozoic terrane in the northeastern Yeongnam Massif, Korea, records prograde mineral parageneses and metamorphic conditions typical for the low-pressure-high-temperature metamorphism. Predominant rock types are pelitic to psammitic gneisses and anatectic migmatites, which are intruded at ca. 1.7 Ga by leucocratic, Stype granite. The metamorphic grade systematically increases over a distance of 10 to 15 km from lower amphibolite facies in the northern part to lower granulite facies in the southern part of the study area. Four metamorphic zones develop with increasing temperature: biotite-cordierite, sillimanite, garnet, and spinel zones. Quartz and oligoclase occur as excess phase. Primary muscovite is uncommon in the biotite-cordierite zone, while K-feldspar is ubiquitous in the sillimanite to spinel zones. Progressive sequences of mineral assemblages correspond to the regional metamorphism of facies series Type 2a of Pattison and Tracy (1991). The reaction, muscovite + quartz = sillimanite + K-feldspar + HjO (R1 in Fig. 1), defines the occurrence of sillimanite. In the sillimanite zone, the Fe/(Fe+Mg) values of biotite (0.55-0.65) and cordierite (0.41-0.51) gradually increase via the reaction, biotite + sillimanite + quartz = cordierite + K-feldspar + H2O (R2). Migmatitic gneiss first appears in the vicinity of the garnet isograd, and its occurrence is apparently governed by two-stage melting. The consumption of muscovite by R1 and the leucosome mineralogy free of mafic minerals suggest that the initial melting is induced by the reaction, quartz + plagioclase -i- K-feldspar + fluid = liquid (R3). The large volume of these leucosomes probably results from the influx of external HjO. Idioblastic porphyroblasts of garnet and cordierite in leucosomes, and biotite and sillimanite inclusions in garnet of migmatitic gneisses suggest the second-stage melting dominated by the fluid-absent, dehydration reactions of biotite (biotite + sillimanite + quartz = cordierite ± garnet -t- K-feldspar + liquid; R4). This dehydration melting accounts for the decrease in Fe/(Fe+Mg) of biotite, cordierite and garnet in the garnet zone.

1 6 5 u

CO CO o

2 KFASHorKMASH KFMASH

I 500

550

600

650

700

750

And^

800

Temperature (°C)

Fig. 1. Simplified petrogenetic grid of metapelites in the KFMASH system based on Pattison (1991). The arrow represents the metamorphic field gradient inferredfrommineral parageneses and textures.

OROGENESIS IN THE OUTBACK

Several geothermo-barometers and phase equilibria were used to estimate the prograde metamorphic conditions. The average P-T conditions are in the range of 600 - 625°C and 3.8 - 4.1 kbar near the garnet isograd, and 730 750°C and 4.8-5.8 kbar near the spinel isograd. The incompatibility of quartz and spinel as well as the absence of kyanite and orthopyroxene corroborate these P-T estimates. The occurrence of K-feldspar through R1 prior to melting reaction (R3) constrains the metamorphic field gradient located below the invariant point IPl. In addition, the common assemblage of gamet-cordierite-biotite-sillimanite-K-feldspar-quartz in migmatitic gneiss suggests the peak metamorphism has reached the P-T condition of invariant point IP2. Thus, phase equilibria together with the P-T estimates indicate the metamorphic field gradient of ca. 25°C/km. No evidence for significant overthickening prior to thermal peak suggests that the advective heat and/or fluid associated with the emplacement of leucocratic granite has played an important role for the low-pressure metamorphism in the northeastern Yeongnam Massif. REFERENCES

Pattison, D.R.M., 1991. P-T-a(H20) conditions in the thermal aureole. In: Equilibrium and kinetics in contact metamorphism, eds. Voll, G. et al., Springer-Verlag, Heidelberg, 327-350. Pattison, D.R.M., and Tracy, R.J., 1991. Phase equilibria and thermobarometry of metapelites. In: Contact metamorphism, ed. Kerrick, D.M., Reviews in Mieralogy, 26, Mineralogical Society of America, 105-206.

ALICE SPRINGS, AUSTRALIA, 1999

HIGH-GRADE GNEISSES AS SOURCE-ROCKS FOR SYNTRANSCURRENT PERALUMINOUS GRANITES IN SOUTH-EASTERN BRAZIL: AN EXPERIMENTAL STUDY

E. Koester ^ L.A.D. Fernandes^; A.R. Pawley^ & G.T.R. Droop ^ 1 Departamento de Geologia, Universidade Federal do Rio Grande do Sul, P.O. Box 15065 9150/970, Porto Alegre, RS, Brazil, e-mail: ferna@if.ufrgs.br 2. Department of Earth Sciences, University of Manchester, Oxford Road, Manchester, Ml3 9PL, UK.

The late-Proterozoic Dorsal de Cangucu Transcurrent Shear Zone (DCTSZ) of the Sul-rio-grandense Shield in Brazil is intruded by a suite of syntectonic, peraluminous two-mica granites. The DCTSZ transects an orthogneissic complex (the Arroio dos Ratos Gneiss Complex (ARGC)) which contains mega-enclaves of metasedimentary gneisses of the Varzea do Capivarita Metamorphic Suite (VCMS). Rocks of the VCMS may be much more voluminous at depth than their outcrop suggests, and are possible candidates for the source rocks of the peraluminous granitic magmas. The VCMS consists of high-grade forsterite marbles, calc-silicate-gneisses, pyroxene-biotite-gneisses and migmatitic garnet-cordierite-gneisses, which were metamorphosed to granulite-facies conditions during the Proterozoic. Gamet-biotite- and two-pyroxene thermometry yield peak-metamorphic temperatures of ca. 800°C, consistent with the formation of garnet- and cordierite-bearing leucosomes by the incongruent fluid-absent melting ofbiotite+ sillimanite + quartz. In order to understand the link between the peraluminous granites and the DCTSZ and the possible role of the highgrade VCMS metasediments as potential sources of the granitic magmas, a series of melting experiments has been performed on a natural metasediment from the area as the starting material. The metasediment is a semi-pelitic gneiss with the mineral assemblage: garnet (2 mode %), cordierite (9%), biotite (24%), K-feldspar (4%), plagioclase (23%), quartz (33%), retrograde phengitic muscovite (3%) plus accessory apatite, zircon, pyrite and rutile. The rock shows evidence of extensive retrogression in the form of fine-grained muscovite replacing cordierite. Experiments were performed on the powdered starting material, under fluid-absent conditions, at temperatures of 700°C to 900®C and at pressures of 5 kbar (using an internally heated gas pressure-vessel), 10 kbar and 15 kbar (using a piston-cylinder apparatus). Our results indicate that a wide range of granitic compositions can be generated by fluid-absent partial melting of retrogressed VCMS semi-pelitic gneiss. Preliminary results show that at 5, 10 and 15 kbar, melting begins at ca. 700°C, promoted by participation of the retrogressive muscovite, producing a very small amount of glass (quenched melt). Melt productivity greatly increases by breakdown of biotite and/or cordierite at temperatures >=850°C and at pressures of 10 and 15 kbar, and at >=800°C at 5 kbar. Orthopyroxene is a residual mineral at 5 kbar, garnet at 10 kbar, and hornblende at 15 kbar. The melts generated have a wide range of granitic compositions (68% < wt.% Si02 < 74%) and metaluminous to peraluminous affinities (0.8 < A/CNK < 3). At 900°C and 10 kbar the melts have similar compositions to the syntectonic peraluminous granites associated with theDCTSZ. These results demonstrate that the peraluminous granites associated with the DCTSZ could have been produced by fluid-absent melting of cordierite-biotite-gneisses of the VCMS during late-Proterozoic transcurrent shearing. Further experiments are currently in progress to assess whether or not such melt compositions can be generated by partial melting of other potential protoliths such as ARGC orthogneisses or mixed orthogneiss plus metasediment.

OROGENESIS IN THE OUTBACK

WHEN THE WILSON CYCLE BREAKS DOWN: INHIBITING REWORKING AND HOW SOME OROGENS CAN PRODUCE STRONG LITHOSPHERE M. Krabbendam Department of Earth Sciences, Monash University, Clayton VIC 3168, email: maarten@earth.monash.edu.au

Australia

To understand a problem like reactivation and cyclicity of orogenesis it is important not only to study the controlling factors that lead to poly-cyclic orogenesis, but also to the factors that inhibit orogens to reactivate and prevent poly-cyclicity. Although reactivation and poly-cyclicity is common, there are many orogens that show a remarkable lack of reactivation. Examples include the Limpopo belt that did not reactivate significantly since 2200 Ma, the Urals which did not rift during the break-up of Laurentia. All the Pan-African orogens that occur within Africa (Trans-Sahara Belt, Damara Belt, Oubanguide Belt) have not rifted during Gondwana break-up and are, therefore, unlikely to be reworked in the future. Do these orogens show a systematic difference with orogens that did rift or otherwise were reworked? As orogens display an enormous variety in their evolution and architecture it is useful to contemplate some mechanisms that may be responsible for reworking vs. stabilisation. I will briefly review some factors that may inhibit reactivation of orogens. Reactivation, in this contribution, is taken as a large-scale process, requiring a entire Wilson Cycle of orogenesis, rifting, ocean opening and closure and orogenesis. As all oceans will close at some point in time, the critical stage in the Wilson Cycle is the rifting event. Thus, the problem can be simplified to the question whether or not an orogen is likely to rift more or less along its length. This can in turn be reduced to the question whether orogenic lithosphere is stronger or weaker than adjacent lithosphere several 100 Ma after that orogeny. LUCK AND BAD LUCK

Much rifting is initiated by hot-spots. By heating the lithosphere, hot-spots lower the strength of the lithosphere, thus facilitating rifting. Hotspots, however, are commonly thought to create uplift and hence, generate extensional stresses themselves. The position of hot-spots is commonly regarded as independent of the position of pre-existing orogens. Thus, there is a component of luck (bad luck?) whether or not an orogen is hit by a hot-spot from below. A long time-gap, however, may exist between the plume-activity and continental separation. During this time, the lithosphere is not only weakened, but the overlying lithosphere also moves with respect to the hotspot, possibly until a pre-existing orogen is encountered. This means that that the coincidence of hot-spots and orogens is skewed away from a pure luck distribution. An example is provided by the Karoo Mantle plume that first hit the Archean Kaapvaal Craton at ~ 180 Ma, but only resulted in rifting and separation at -150 Ma (Cox 1992, Storey 1995) when the mantle plume was underneath the Meso-Proterozoic Natal Province. DEHYDRATION METAMORPHISM

During orogenesis, metamorphism and partial melting can result in large-scale dehydration of the lower crust, particularly so when temperatures increase above 600-800 °C and granulite-facies conditions are reached. The anhydrous ('under-saturated') nature of the resultant rocks also means that the free water pressure is near zero (Yardley and Valley 1989). The absence of free water and the scarcity of hydrous minerals (which are generally weaker phases) means that granulites have a higher shear strength than lower grade rocks of comparable composition. This suggestion is supported by observations in areas such as the Bergen Arcs in Norway were Proterozoic granulite underwent eclogite-facies metamorphism during the Caledonian Orogeny, The granulites are largely preserved and were only deformed and metamorphosed in zones of later hydration (Austrheim 1998). Large-scale re-hydration of the lower crust is an unlikely process because hydration tend to localise in discrete shearzones, rather than affect a terrane in a homogenous manner. This would suggest that dehydration due to partial melting and / or granulite-facies metamorphism may lead to a rock that has a higher strength than rocks of similar composition that did not experience these processes. If a large part of the crust underwent the processes, this may lead to a higher than average crustal strength. This mechanism would have a particular strong effect if the orogen was destroyed mainly by erosion rather than extensional collapse, because erosion predominantly removes the upper crust. It is interesting that many areas of continental crust that experienced UHDP metamorphism are composed of reworked hydrous (amphibolite-facies) gneisses, whereas reactivation of anhydrous, granulite-facies terranes is more rare.

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

PRESENCE OF THICKENED SUB-CONTINENTAL LITHOSPHERIC MANTLE.

Many models of Wilson Cycles argue that rifting along orogens is caused by weakening because the thickened crust supposedly displaces the strong sub-continental lithospheric mantle (SCLM). This ignores the fact that the SCLM will thicken too during orogenesis, so if the argument is correct, the SCLM needs to be thinned. Thinning of the SCLM can occur during convective removal of the lower part (the thermal boundary layer) or by conductive heating. The chance of convective removal increases with the amount of lithospheric thickening. This implies that a orogen that only experienced limited thickening may not experience convective removal; in such an orogen the amount of radiogenic heating is also limited. Numerical modelling suggests that such orogens may remain lithospheric strong zones 100's of Ma after orogenesis. It has been argued that the Urals did not experience orogenic collapse and still has a thickened SCLM (Berzin et al. 1996), possibly explaining why it did not rift during the break-up of Laurentia.

NATURE OF THE MATERIAL: PRE-COLLISIONAL EVOLUTION

The nature of the material involved in the orogen may also determine whether or not an orogen is likely to rift or not. Of particular importance is the ratio of mafic vs. felsic material. Mafic material is more dense and normally less radiogenic than felsic material. This implies that if an orogen contains much mafic material, for instance if a number of island arcs are incorporated, it will not heat up as much as a more felsic orogen. This may provide an alternative explanation for the stabilisation of the Urals, as this orogens contains large amounts of island arc material, has a very low heat flow and radiogenic heat production (Kukkonen et al. 1997).

THE EXCEPTION: OROGENIC BELTS ADJACENT TO ARCHEAN CRATONS

Archean lithosphere is colder, thicker and stronger than younger lithosphere: this configuration is stable because Archean lithosphere is depleted in Fe and has a lower density (eg. Durrheim & Mooney 1991). The strength of Archean lithosphere appears to be greater than the strongest lithospheric scenarios sketched above. Thus, a Proterozoic or younger orogen adjacent to an Archean Craton is susceptible to reworking regardless of the nature of the younger orogen.

SUSCEPTIBILITY TO REWORKING: SOME CONSTRAINTS

If the above arguments are turned around an orogen can be regarded as susceptible to large-scale reworking if: • it is hit by a hot spot; • it did not dehydrate entirely during the first orogenic cycle; • it is composed of material with a high radiogenic heat production capacity; • if its underlying SCLM is thinned by convective removal of the thermal boundary layer; • it is destroyed by extensional collapse, rather than by erosion; • it experienced large amount of crustal thickening; • it is adjacent to an Archean Craton.

References:

Austrheim, H., Erambert, M. & Engvik, A. K., 1997. Processing of crust in the root of the Caledonian continental collision zone: the role of eclogitisation. Tectonophysics, 273,129-153. Berzin, R., et al. 1996. Orogenic Evolution of the Ural Mountains: Results from an Integrated Seismic Experiment. Science, 274, 220-221. Cox, K. G., 1992. Karoo igneous activity, and the early stages of the break-up of Gondwanaland. Geological Society Special Publication, 68, 137-148. Durrheim, R. J. & Mooney, W. D., 1991. Archean and Proterozoic crustal evolution: Evidence from crustal seismology. Geology, 19, 606-609. Kukkonen, I. T. et al. 1997. Low geothermal heat flow of the Urals fold belt; implication of low heat production, fluid circulation or palaeoclimate? Tectonophysics, 276, 63-85. Storey, B. C., 1995. The role of mantle plumes in continental breakup: case histories from Gondwanaland. Nature, 377, 301-308. Yardley, B. W. D. & Valley, J. W., 1997. The petrologic case for a dry lower crust. Journal of Geophysical Research, 102, 12173-12185.

OROGENESIS IN THE OUTBACK

MAGMATISM, METAMORPHISM AND FLUIDS IN THE NORTHERN MARGIN OF THE DAMARA OROGEN, NAMIBIA P. H. Macey and C. Harris Department of Geological Sciences, University of Cape Town, Rondebosch 7700, South Africa

The Zerrissene Group turbidites of central western Namibia, consist of a succession of spectacularly folded, siliciclastic turbidites (88%) and carbonate rocks (12%) (see the front cover of J. Struct. Geol. 1999). These rocks were deformed during the Pan-African Damara Orogeny to produce tight, north-south trending folds with near horizontal fold axes, and a well-developed slaty cleavage (e.g. Miller, 1983). A second, weaker, deformation event (D2) produced generally east-west trending interference folds. This second deformation event was accompanied by the intrusion of numerous and voluminous syn-tectonic granitoid bodies, and was followed by the intrusion of posttectonic granitoids and syenites. The close temporal relationship between multiple episodes of deformation and magmatism, and the large volumes of magma, has led to a complex thermal regime in the area. A wide, up to 15km, thermal aureole has developed around the major intrusive bodies. Changes in the mineralogy of the siliciclastic rocks towards the contacts with granite progress from chlorite, to biotite, muscovite-out, cordierite/garnet and andalusite zones. Fluid inclusion studies of quartz veins together with the metamorphic mineral assemblages suggest an approximately 200°C temperature gradient across the Zerrissene Group rocks, from about 350 to 550°C. Temperatures of metamorphism decrease to the north, away from the main syn-D2 granite contacts which strike east-west. The syn-D2 plutons in the area are typical S-type granites with quartz values ranging from 10.8 to 13.4%o (mean 11.9%o). Biotite 5D values range from -57 to -76%o. These data would suggest magmatic waters with the approximate and 5D values of 10 and -45%o respectively. The values of quartz veins hosted by the siliciclastic rocks range from approximately I0%c to 18%o and show a positive correlation with distance from the main granite contact. The host siliciclastic rocks also have lower values close to the granite contact, though the correlation between host value and distance from the granite is not as strong as in the quartz veins. The values of carbonate rocks, and quartz veins hosted by carbonate rocks, show no correlation with distance from the granite contact. The values of the carbonates range from 13 to 27%o (vs. SMOW) and have somewhat lower values than equivalent undeformed carbonate units of the Otavi Mountain Land (19-29%o, Chadwick, 1993). There is a reasonable positive correlation between host rock and quartz vein value with the quartz veins being on average 2%c higher than the host rocks. This is indicative of a "rock-buffered" fluid-rock system. However, a simple "rock-buffered" system will not produce a correlation between quartz vein values and distance from the main granite contact (i.e. the heat source driving fluid circulation). Two possible explanations for the regional variation in quartz vein values are: (i) Quartz veins formed from fluids which are mixtures of magmatic fluid with meteoric and/or metamorphic waters having much higher with the relative proportion of the former decreasing away from the granite contact, (ii) Fluid-rock interaction involving magmatic fluids lowered the value of host-rock closest to the contact. There is evidence to support both explanations. Siliciclastic rocks from the same formations have lower S^^O values close to the contact with the syn-tectonic granite. The 5D values of water extracted from fluid inclusions in quartz veins by bulk decrepitation shows a positive correlation with quartz value. This correlation is consistent with fluids being mixtures of magmatic fluid and a fluid with 5D and S^^O values of about -20 and +ll%o ,respectively (metamorphic water ?). The carbonate horizons have generally maintained their sedimentary precursor stable isotope compositions. A subset of samples from outcrops either adjacent to siliciclastic rocks or the contact with the granite appear to have undergone lowering of their whole-rock values. The value of the CO2 component of mixed CO2-H2O fluid inclusions is -i-1.9%o which is close to the average value for the carbonate rocks in the area. This suggests that the carbonate rocks themselves are the source of the carbon. The Brandberg West area hosts a number of hydrothermal vein-type Sn-W deposits. Quartz-muscovite oxygen isotope geothermometry of mineralized veins suggests the temperature of mineralization to be approximately 550®C. Quartz veins from the Brandberg West Sn-W deposit have lower values than expected given their apparent distance from the nearest granite contact and are consistent with the presence of a hidden granite pluton close to the

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

surface. The association of economic mineral deposits with lower than normal quartz vein potential exploration tool.

values offers a

REFERENCES

Miller, R.McG., 1983. The Pan-African Damara Orogen of South West Africa/Namibia Geological Society of South Africa, Special Publication, 11,431-515. Chadwick, P.J., 1993. A study of the Berg Aukas-type Pb-Zn-V deposits in the Otavi Mountain Land, Namibia. PhD thesis (unpublished) University of Cape Town.

O R O G E N E S I S IN T H E O U T B A C K

Rb-Sr, Sm-Nd and U-Pb (Pb-Pb) DATING OF METAMORPHIC ASSEMBLAGES • SOME REMARKS ON APPLICATIONS AND RECENT DEVELOPMENTS R. Maas VIEPS Radiogenic

Isotope Laboratory,

Dept. of Earth Sciences, La Trobe University,

Melbourne

Metamorphic geochronology has long been the preserve of traditional methods such as Rb-Sr (whole rock and mica dating), Pb-Pb (whole rock) and K-Ar and ^Âr-'^Âr. In recent years, in situ U-Pb dating of U-rich accessories (chiefly zircon, monazite) using ion-microprobes has become as serious competitor, owing to its accuracy, spatial resolution and speed. While this trend will continue (e.g. laser-ablation ICP-MS-based accessory mineral U-Pb dating), the TIMS (thermal ionization mass spectrometry)-based methods (chiefly Rb-Sr, Sm-Nd and U-Pb, Pb-Pb) have responded by building on what is their real strength, their ability to date the formation and/or cooling of of rock-forming metamorphic minerals. It is these minerals from which PT and textural information is extracted (1,2), not the U-rich accessories which can be dated by micro-beam methods. The range of minerals that can be dated is expanding, and improvements have been made to age accuracy, precision and spatial resolution by employing chemical leaching procedures and by reducing the sample sizes required for analyis. This contribution describes aspects of modem metamorphic geochronology and outlines some current trends. Rock-forming metamorphic minerals amenable for dating by one or more of the methods discussed here include: garnet, micas, staurolite, hornblendes, pyroxenes, ilmenite-magnetite, sphene and andalusite-kyanite. Garnet, a useful metamorphic indicator mineral, combines high Sm/Nd, moderately high U/Pb, and low Rb/Sr (e.g 3,4), and can be dated by all three systems. It was the target of the very first Sm-Nd study (5) and has since become the main field of Sm-Nd dating in metamorphic rocks (e.g. 6-8). The results have generally been encouraging; problem areas are the influence of REE-rich inclusions (see below) and the closure temperature of the garnet Sm-Nd system (9). Eclogites are an obvious target for garnet Sm-Nd dating, because REE-rich inclusions are unimportant and Sm/Nd ratios are high (10,11), allowing even very young eclogites to be dated (12). However, data need to be interpreted with caution because of potential failure of garnet and omphacite to fully equilibrate on formation (13,14). U-Pb and Pb-Pb dating of garnet is possible despite low U contents (=1 ppm) because common Pb is also low. Early work focussed on Precambrian garnets (15,16) but high ^^Û-^^Pb ratios often observed in garnet also allows dating of younger assemblages (17,18), including some of late-Alpine (25-30 Ma) age (19). With favorable dispersion in Sm/Nd and U/Pb, useful age precisions (<±2 Ma for Tertiary ages, <10 Ma for all others) can be obtained. Garnet has low Rb/Sr and requires data for host rock and/or bulk matrix to construct Rb-Sr isochron diagrams; two-point ages derived from these and from Sm-Nd, U-Pb data have been used to constrain garnet growth periods, and metamorphic heating and burial rates (20,21). Staurohte has surprisingly high U/Pb ratios, despite modest (1-4 ppm) U content, and samples weighing just a few mg can therefore be dated with good precision (18). The same study reported U-Pb data for kyanite and anatase; the kyanite-whole rock age was similar to the U-Pb ages for coexisting staurolite, implying similar closure characteristics. Ilmenite and magnetite, important in determination of fo2, are also amenable to precise U-Pb and PbPb dating (2,18). It should be pointed out that none of these minerals is sufficiently enriched in radiogenic Pb to be analysed by microbeam techniques. In a particularly interesting development, leaching techniques are increasingly used to improve accuracy and precision of isochrons. One of the most fertile areas is the application of step-leaching to selectively release common and radiogenic Pb from U-bearing minerals such as sphene, hornblende, pyroxene, garnet, staurolite, biotite, sillimanite apatite and tourmaline (22-26). The progressive leaching unmixes otherwise undateable blends of common and radiogenic Pb to generate precise Pb-Pb and U-Pb isochrons, essentially producing mono-minerallic isochron systems. This removes the need to combine data for different minerals to obtain an isochron which can be source of error. It has been shown (27) that the two different Pb components can be derived from the same mineral (and not in part from foreign inclusions), but in many cases inclusions will play an important role (28). If they are cogenetic (same age, same initial isotope ratio), their presence may be beneficial, but unequilibrated inclusions can cause spurious ages. Selective leaching is often the only way to recognize and/or remove such inclusions (26). Leaching with strong acids also strongly increases the Sm/Nd ratio of garnet, presumably be removing LREE-rich monazite and apatite inclusions (29). If they are cogenetic, the leachate and corresponding residue will lie on the same isochron, creating, in effect, a "single mineral" Sm-Nd isochron system. Similar results have been obtained for biotite Rb-Sr systems, again creating "single mineral" isochrons (30).

ALICE SPRINGS, AUSTRALIA, 1999

Another interesting development is the push to analyse smaller samples. With improvements in mass-spectrometric sensitivity, analysis of single grains becomes possible, e.g. Rb-Sr dating of single Alpine white mica flakes taken directly from thick sections to a precision of ±1 Ma (31). Micro-coring or microdrilling can be used to sample areas <lmm in diameter, in situ and therefore with the benefit of preserved textural context. Like in laser-ablation work, traverses can be sampled on single crystals, avoiding obvious inclusions, sites of alteration, and creating singlemineral isochron systems (e.g. on crystals showing trace-element zoning). This method has been driven to new extremes by (32) who report Rb-Sr ages for mylonite formation obtained from tiny samples (weighing 3 micrograms) of andalusite and muscovite grown in pressure shadows around the andalusite, and from calcite-i-mica in pressure shadows around pyrite in carbonate mylonites. Ages obtained on this scale are used to obtain geological strain rates in large fault systems; obviously such techniques may also be used to date other heterogenous systems, such as pseudotachylites, skams and hydrothermally altered rocks where textural context and sampling volume are important.

1 Cliff (1985) Journal of the Geol. Soc. of London 142, 97-110 2 Burton and O'Nions (1990) Contributions to Mineralogy and Petrology 106, 66-89 3 Hickmott et al. (1987) Geology 15, 573-576 4 Hickmott and Spear (1992) Journal of Petrology 33, 965-1005 5 Notsu et al. (1973) Geochemical Journal 7, 51-54 6. van Breemen and Hawkesworth (1980) Trans Royal Soc. of Edin., Earth Sci. 71, 97-102 7 Vance and O'Nions (1990) Earth and Planetary Science Letters 97, 227-240 8 Getty et al. (1993) Contributions to Mineralogy and Petrology 115, 45-57 9 Mezger et al. (1992) Earth and Planetary Science Letters 113, 397-409 10 Griffin and Brueckner (1985) Chemical Geology 52, 249-271 11 Miller and Thoni (1995) Chemical Geology 122, 199-225 12 Oberhansli et al. (1985) Chemical Geology 52, 165-184 13 Thoni and Jagoutz (1992) Geochimica Cosmochimica Acta 56, 347-368 14 Jagoutz (1994) ICOG 8 Abstracts, USGS Circular 1107, 156 15 Mezger et al.(1989) Contributions to Mineralogy and Petrology 101, 136-148 16 Mezger et al.(1991) Journal of Geology 99,415-428 17 Vance and Holland (1993) Contributions to Mineralogy and Petrology 114, 101-118 18 Lanzirotti and Hanson (1995) Geochimica Cosmochimica Acta 59, 2513-2526 19 Vance and O'Nions (1992) Earth and Planetary Science Letters 114, 113-129 20 Christensen et al. (1994) Contributions to Mineralogy and Petrology 118, 1-12 21 Burton and O'Nions (1991) Earth and Planetary Science Letters 107, 649-671 22 Gariepy et al. (1990) Geology 18,1078-1081 23 Frei and Kamber (1995) Earth and Planetary Science Letters 129, 261-268 24 Frei et al. (1995) Geology 23, 1095-1098 25 Berger and Braun (1997) Chemical Geology 142, 23-40 26 Dahl and Frei (1998) Geology 26, 111-114 27 Frei et al. (1996)) Geochimica Cosmochimica Acta 61, 393-414 28 DeWolf et al. (1996)) Geochimica Cosmochimica Acta 60,121-134 29 Zhou and Hensen (1995) Chemical Geology 121, 317-326 30 Maas (1999) submitted 31 Cliff (1994) ICOG 8 Abstracts, USGS Circular 1107, 62 32 Miiller et al. (1998) EOS 79, F951

OROGENESIS IN THE OUTBACK

CYCLICITY IN THE AUSTRALIAN PROTEROZOIC - INFLUENCES AND CONTROLS

S. McLaren and M. Sandiford Department of Geology and Geophysics, University of Adelaide, SA5005

Many Australian Proterozoic terranes record a series of deformational, magmatic and metamorphic events which can be broadly correlated through both time and space. Despite similarities of evolution on many scales, we also recognize marked differences often in the duration of crustal activity and the extent of terrane cyclicity. Reconciling both the similarities and differences in the record of Australian Proterozoic crustal activity to some generalized model of crustal behaviour is an ongoing challenge, any solution to which must appeal to both physical and geological arguments. The concept of terrane reworking must be understood in terms of the factors that control long-term crustal weakness. As crustal strength is closely allied to thermal structure, insights into this problem can be taken from understanding the parameters which control crustal thermal regime. Although many models of the evolution of metamorphic belts have concentrated on the variations in orogenic boundary conditions (eg. mantle heat flux and magma input) there has been relatively little effort expended on evaluating the role of internal variation in thermal parameters (notable exceptions include Chamberlain & Sonder, 1992). In this contribution we discuss the thermomechanical consequences of variations in crustal heat production. Working from these first-principles we then suggest how thermo-mechanical models may provide insights into the causes of crustal reworking in Australian Proterozoic terranes. Crustal strength is dependent not only on the amount of heat production in the crust, but also on its distribution. The most stable thermo-mechanical state corresponds to the case where crustal heat production is concentrated into the upper crust. However, prior to major periods of crustal differentiation, it is likely that crustal heat production was distributed throughout the lower crust, which would result in significant crustal weakening, accompanied by elevated thermal regimes. In order for a terrane to become increasingly stable in a thermo-mechanical sense the high heat producing elements must be removed from the lower crust and concentrated into the upper crust (for example, by magmatism and fractionation and/or compressional deformation). The crust will remain weak, and therefore susceptible to re-activation, until any anomalous heat production it contains has been removed, or emplaced at higher crustal levels. The series of changes recorded in the geological system (ie. the cyclicity) will reflect this underlying first-order thermo-mechanical control. In the context of the Australian Proterozoic (which is characterized by extraordinarily high concentrations of the heat producing elements, often in granites and granitic gneisses) the application of thermo-mechanical models may be quite appropriate. Depending on the initial distribution of heat producing elements in the crust and the opportunity for vertical crustal re-organization, different patterns of crustal reworking will be recorded. It is no coincidence that terranes where a large degree of crustal reworking has occurred have their record of tectonic activity terminated by some large magma extraction event associated with the generation of highly radiogenic, dominantly granitic melts in the upper crust (eg. eastern Mount Isa Inlier, Mount Painter Inlier, Tennant Creek and Davenport Provinces). This final migration of the heat producing elements into the upper crust in the form of these granites, results in an increase in crustal strength and correspondingly increased thermo-mechanical stability. Cycles of magma generation throughout the Proterozoic (eg. Wybom et al., 1992) attest this trend. REFERENCES

Chamberlain C.P. and Sonder L.J. 1990. Heat-producing elements and the thermal and baric patterns of metamorphic belts. Science 250, 763-769. Wyborn L.A.I., Wyborn D., Warren R.G. and Drummond B.J. 1992. Proterozoic granite types in Australia: implications for lower crust composition, structure and evolution. Transactions of the Royal Society of Edinburgh: Earth Sciences 83, 201-209.

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

THE LARAPINTA EVENT: ORDOVICIAN METAMORPHISM IN THE EASTERN ARUNTA INLIER J.A. Miller\ J. Mawby^ M. Hand^ I.S. Buick^ I. S. Williams'^

1 Department of Earth Sciences, Monash University, Clayton, VIC, 3168, Australia 2. Department of Geology and Geophysics, Adelaide University, Adelaide, SA, 5005, Australia 3. Department of Earth Sciences, La Trobe University, Bundoora, VIC, 3083, Australia 4. RSES, Australian National University, Canberra, ACT, 0200, Australia

Metamorphism in the Aninta Inlier is demonstrably polymetamorphic, and occurred during three main periods. The first period involved emplacement of granitic rocks and multiple episodes of high-grade metamorphism and occurred during the Palaeo- to Mesoproterozoic (~1780 Ma to —1570 Ma: Strangways Orogeny, Chewings Orogeny). Metamorphism of this type is mostly of low-pressure/high-temperature facies (-2.5 to -5 kbar). The second period of metamorphism occurred during the -300-400 Ma Alice Springs Orogeny, when thick-skinned, south-directed intraplate deformation thrust Arunta Inlier basement over the northern margin of the Amadeus Basin. In the Arunta Inlier basement this orogeny resulted in medium-pressure, greenschist to mid-amphibolite-facies metamorphism that re-worked Proterozoic assemblages, primarily in discrete shear zones. The third period of major metamorphism in the Arunta Inlier has only recently been identified on the basis of SHRIMP and Sm-Nd dating. It is also Palaeozoic in age and developed prior to the Alice Springs Orogeny in the interval -480-460 Ma (early to mid Ordovician). In the eastern Arunta Inlier (Harts Range Complex, Harts Range) pervasive medium- to high-pressure granulite-facies metamorphism (-8-10 kbar, -800 °C) occurred at this time in the Irindina Supracrustal Assemblage (Miller et aL, 1997; Mawby et al., 1998; Hand et al, 1999). Peak metamorphism was associated with extensive fluid-absent partial melting of both metabasic and metapelitic lithologies, resulting in the formation of coarse-grained garnet-bearing assemblages. Peak metamorphism was followed by approximately 4 kbar of near-isothermal decompression to 6 kbar and 650-700 associated with the formation of a regional sub-horizontal foliation that is in places mylonitic. SHRIMP data suggests that the terrain was above the granite solidus (-650 °C) for 20-25 Ma implying that decompression was slow. The regional extent of the Ordovician metamorphism is currently poorly constrained. However, Scrimegeour et al. (1999) have suggested that mylonitic reworking of the Kanadra Granulite further NE may also be Ordovician in age. The granulite-facies metamorphism and high-T decompression was coeval with the development of the late Cambrian to early Ordovician marine Larapinta Group, which accumulated in a slowly subsiding sedimentary basin that appears to have had its depocenter along the axis of the now exhumed Arunta Inlier. The Brady Gneiss, which structurally overlies the Irindina Supracrustal Assemblage in the Harts Range Complex, underwent peak metamorphism at approximately 7 kbar and 700°C and followed a retrograde P-T path qualitatively similar to the Irindina Supracrustal Assemblage. It is suggested that metamorphism of the Irindina Supracrustal Assemblage and the Brady Gneiss should be referred to as the LARAPINTA EVENT to highlight the synchroncity between the timing of high-grade metamorphism, and deposition of the Larapinta Group in the Amadeus Basin to the south of the Harts Range. REFERENCES Mawby, J., Hand, M., Foden, J. & Kinny, P., 1998. Ordovician granulites in the southeastern Arunta Inlier: a new twist in the Palaeozoic history of central Australia. Geological Society of Australia Abstracts Volume, 49, p. 296. Miller, J.A., Cartwright, I., & Buick, I.S., 1997. High grade metamorphism in the Harts Range. Petrology and P-T constraints from Mallee Bore, northern Harts Range, central Australia. Journal of Metamorphic Geology, 15, 613-629. Miller, J.A., Buick, I.S, Williams, LS. & Cartwright, L, 1998. Re-evaluating the metamorphic and tectonic history of the eastern Arunta Inlier, central Australia. Geological Society of Australia Abstracts Volume, 49, p. 316. Hand etal., 1999. Scrimgeour, I. & Raith, J.G., 1999. High grade mylonitic reworking of the Kanandra granulite, central Australia: implications for the Palaeozoic evolution of the eastern Arunta Inlier. Geological Society of Australia Abstracts Volume, 54, p 101.

OROGENESIS IN THE OUTBACK

THE ROLE OF SEAFLOOR ALTERATION IN THE DEVELOPMENT OF HIGHPRESSURE ASSEMBLAGES IN THE CORSICAN OPHIOLITE, FRANCE J. A. Miller\ I.S. Buick^ & I.Cartwright^

1. Dept of Earth Sciences and VIEPS, Monash Univeristy, Clayton, VIC, 3168, AUSTRALIA 2. Dept of Earth Sciences and VIEPS, La Trobe University, Bundoora, VIC, 3083, AUSTRALIA

The heterogeneity of high-pressure mineral assemblages in the Corsican Ophiolite attests to the important role early seafloor alteration takes in controlling the mineralogical and textural recrystallisation of rocks during high-pressure metamorphism. In this study we examine a series of variably deformed pillow lavas from Defile de Lancome, in Alpine Corsica, that have been metamorphosed up to eclogite-facies. The development of eclogite-facies mineralogies in these pillows has been strongly controlled by the pattern of brecciation and fracturing developed during seafloor alteration. This has also controlled how the rocks have behaved during deformation and retrogression during exhumation of the high-pressure nappe sequences. The Corsican Ophiolite is one of the Western Alpine Ophiolites, which represent remnants of the Piemonte Ocean that was subducted prior to the collision of the Apulian micro-plate with the European plate. These ophiolites have generally been affected by high-pressure metamorphism during the late Cretaceous and Tertirary (Cliff et aL, 1997; Rubbato et al, 1998), producing blueschist- and eclogite-facies mineralogies. Subsequent greenschist facies overprinting of the high-pressure mineral assemblages related to exhumation of the high-pressure nappe sequences has been documented from several areas including the Corsican Ophiolite. Evidence for three episodes of metamorphic activity have been identified in the ophiolitic rocks of Alpine Corsica (Fournier et al, 1991; Lahondere, 1991). Early eclogite-facies metamorphism (1.2 GPa, 530''C) was substantially overprinted by a pervasive blueschist-facies metamorphism (0.7-0.9 GPa, 400 °C) that produced lawsonite-rich mineralogies. These high-pressure mineralogies were locally retrogressed to greenschist-facies assemblages associated with late SEdirected thrusting and the formation of albite veins. In the Corsican Ophiolite oxygen isotope enrichments and depletions in different rock units are consistent with early hydrothermal alteration whilst the rocks were still on the seafloor. This early seafloor alteration caused fracturing and brecciation in the pillow lavas, allowing the ingress of seawater along discrete fracture zones. This in turn led to the formation of narrow zones of intense alteration, whilst the basalt clasts themselves remained relatively unaltered. Such a pattern of fracturing is recorded in unmetamorphosed pillows from the Troodos Ophiolite on Cyprus and is identical to the style of fracturing recorded in the pillows from Defile Lancome. The composition of these narrow alteration zones was dominated by clay and smectite minerals, whilst the basalt clasts had essentially an igneous mineralogy. The resultant pillow thus had pronounced rheological and compositional variations. During high-pressure metamorphism these regions behaved very differently. The relatively unaltered basalt clasts have statically transformed to the assemblage gt + omph + laws whilst retaining an igneous texture. In comparison, the discrete fracture zones have essentially acted as micro-scale reactivation zones along which deformation and fluid flow have been channelled. Thus, while the basalt clasts record the highest grade metamorphic event experienced by the pillows, the discrete fracture zones record only the latest metamorphic event to have affected the rocks. REFERENCES

Cliff, R. A., Bamicoat, A. C., & Inger, S., 1998. Early Tertiary eclogite facies metamorphism in the Monviso Ophiolite. Journal of Metamorphic Geology, 16, 447-455. Fournier, M., Jolivet, L., Goffe, B., & Dubois, R., 1991. Alpine Corsica metamorphic core complex. Tectonics, 10, 1173-1186. Lahondere, D., 1991. Les schistes bleu et les eclogites a lawsonite des unites continentales et oceaniques de la Corse alpine: nouvellees donnees petrologiques et structurales. These de doctorat, MontpeUier, France. Rubatto, D., Gebauer, D., Compagnoni, R., 1997. Dating the UHP/HP metamorphism in the Western Alps (SesiaLanzo and Zermatt-Saas-Fee); evidences for subduction events at the Cretaceous-Tertiary boundary and in the middle Eocene. Fifth international eclogite conference. Terra Abstracts, 9, Suppl. 1, 30-31.

ALICE SPRINGS, AUSTRALIA, 1999

CRUSTAL GROWTH, METAMORPHISM & DEFORMATION IN THE STRANGWAYS METAMORPHIC COMPLEX: A SUMMARY OF RECENT U-PB AND SM-ND GEOCHRONOLOGY

A. Moller^^*, R.A. Armstrong2, M. Ballevre^, B J . Henseni & K. Mezger^

1* School of Geology, University of New South Wales, Sydney, NSW 2052 curr. address: Institutf Geowiss., Univ. Mainz, D-55099 Mainz, annoeller@mail.uni-mainz.de 2 RSES, The Australian National University, Canberra, ACT 0200 3 Geosciences Rennes, Universite de Rennes I, F'35042 Rennes Cedex 4 Max-Planck Institut, Abteilung Geochemie, D-55099 Mainz

CRUSTAL GROWTH Evidence for rock formation ages in the Strangways Metamorphic Complex (SMC) is given by inherited cores of zircons analysed by SHRIMP, Pb isotopes from leached feldspar and whole rock Nd model ages. The oldest zircon cores found have an age of 2.1 Ga, but most are between 1760 and 1860 Ma old and correspond to earlier recognised orogenic phases (c.f. Collins & Shaw, 1995). Common Pb compositions from leached feldspars suggest multistage Pb isotope growth, partly in a high U environment, and extraction from a mantle like source in the Early Proterozoic. Most Nd model ages calculated from metapelites as well as mafic rocks in the SMC fall within a time span of 30 Ma at about 2.55 Ga (mostly unpublished results by Hensen & Zhou). The age cluster is better constrained than expected for a model dependent calculation and suggests either a uniform source with little disturbance of the REE systematics post mantle extraction or extremely good mixing of material throughout the SMC. The Nd data do not require the complex crust evolution suggested by previous authors and are consistent with the common Pb data. HIGH-GRADE METAMORPHISM IN THE SMC Two Proterozoic phases of high grade metamorphism at 1770 & 1730 Ma have previously been reported for the SMC. To resolve the timing and duration of these metamorphic events, SHRIMP U-Pb dating of zircons which crystallised during metamorphism and deformation was carried out on two locations. Euhedral zircons grew in orthopyroxene-bearing, structurally defined, leucosomes in mafic rocks. The zircons record the crystallisation age of the high-grade leucosomes, which coincides with the peak of metamorphism. Two leucosomes, a layer parallel (Ml, Dl) vein and a second crosscutting (M2, D2) vein were analysed and provide unambiguous evidence for a single high grade metamorphic episode followed by cooling at ca. 1715 Ma. The two metamorphic (Ml, M2) events of earlier workers could not be temporally resolved by SHRIMP. They occurred within 10-15 Ma of each other and the second event apparently disturbed U-Pb systematics of zircons which grew during the first event at ca. 1730 Ma. We conclude that M1-M2 and the correlated periods of intense deformation (D1-D2) are part of a single tectonothermal event, and that the age of the main high-grade metamorphic event in the SMC has to be revised to 1715-1730 Ma. Previous, older estimates for granulite metamorphism in the SMC (1730-1770 Ma) based on U-Pb zircon dating of deformed granitic intrusive rocks may represent inherited zircons or pre-metamorphic intrusive ages and are not necessarily linked to high-grade metamorphism. Zircon from an Opx-leucosome from E of the Woolanga fault show some inheritance of pre-1730 Ma cores, some results at about 1730 Ma as well as strong recrystallisation of the equant multifaceted grains which is evident by very bright luminescent zones. This results in disturbance of the U-Pb system and an estimate of about 1570 Ma for the age of this event, 120 Ma after granulite facies metamorphism in the SMC, but contemporaneous with high grade metamorphism in the Reynolds and Anmatjira Ranges (e.g. Williams et al, 1996). Conventional U-Pb monazite ages of high grade metapelites from widely-spaced localities in the SMC gave ages of ca. 1715-1730 Ma. This age range corresponds to zircon ages from Opx-leucosomes in mafic rocks. It is interpreted to record prograde monazite growth and the peak of metamorphism. Discordant results of similar age due to Pbloss are recorded by monazite from shearzones, which apparently were generally not reset or recrystallised during extensive retrogression of the high grade assemblages. Crystallisation or Pb closure of monazite in a Opxleucosome occurred at 1704 Ma. Discordant and younger ages were determined in metapelites from the eastern SMC. Metapelites from Mt. Pfitzner and from east of the Woolanga fault show discordance or different age populations at ca 1700 and 1400 Ma, respectively. Backscatter images reveal recrystallisation fronts (possibly fluid-driven) in some grains. The concordant 1400 Ma result of a low U monazite fraction provides an estimate for the age of non-recent Pb-loss in other fractions.

OROGENESIS IN THE OUTBACK

PALEOZOIC SHEARZONES

Palaeozoic activity in the SMC was until recently thought to be restricted to exhumation of the granulites by thrusting along low-grade shearzones and was placed into the lower Carboniferous to Devonian Alice Springs Orogeny (ASO, 300-400 Ma). In-situ dating of complex zircons in a staurolite bearing shearzone reveals a phase of new zircon growth at 443 ±6 Ma (prior to the generally accepted age of the ASO), followed by development of the St-Bt-Chl assemblage in which monazites have an age of about 385 Ma (Moller et al., this volume). The zircon growth event is roughly contemporaneous with high grade metamorphic events in the Harts Range further to the east (e.g. Hand et al., 1999) East of the Woolanga fault, foliated garnet-bearing amphibolites occur within the West Bore shear zone, cutting through granulite-facies gneisses of the eastern SMC. Large euhedral garnets (up to 3 cm) occur within finegrained recrystallised leucocratic diffusion haloes of plagioclase-quartz. Garnet grew late or post-kinematic by dehydration reactions during prograde metamorphism at about 600 °C and 4-5 kbar, indicating heating near the end of a period of intense deformation. A Sm-Nd isochron on a single arrested reaction texture gives an age of 381 ±7 Ma for peak metamorphism and associated deformation. Amphibolite-facies conditions prevailed during shear zone development within the Strangways Metamorphic Complex during the Alice Springs Orogeny. The temperature conditions require a higher than normal geothermal gradient, and indicate that the Alice Springs Orogeny was associated with significant crustal thickening in this part of the SMC. OPEN ISSUES A N D CONCLUSIONS

The new age determinations indicate that there are different units within the SMC which have to be investigated separately to reconstruct the complex metamorphic and exhumation history of the whole complex. The geological significance of the noted events of recrystallisation in zircon and monazite in the eastern SMC needs to be resolved (and how it is possible that other events such as the emplacement of the Mud Tank carbonatite do not seem to have had any effect). The results also indicate that good structural and textural control is essential to decipher discrete metamorphic or deformation events in complex metamorphic terranes. REFERENCES

Collins, W.J. & Shaw, R.D. 1995. Geochronological constraints on orogenic events in the Arunta Inlier: a review. Precambrian Research 71, 315-346. Hand, M., Mawby, J., Kinny, P. & Foden, J. 1999. SHRIMP U-Pb constraints on the timing of Palaeozoic intracratonic deformation in the southeastern Arunta Inlier, central Australia. Tectonics, In Press. Moller, A., Williams, I.S., Jackson, S. & Hensen, B.J., 1999. Palaeozoic Deformation and Mineral Growth in the Strangways Metamorphic Complex: In-situ Dating of Zircon and Monazite in a Staurolite-Corundum Bearing Shearzone - Orogenesis in the Outback, Alice Springs, this volume. Williams, I.S., Buick, I.S. & Cartwright, 1. 1996. An extended episode of early Mesoproterozoic metamorphic fluid flow in the Reynolds Range, central Australia. Journal of metamorphic Geology 14, 29-47.

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

PALAEOZOIC DEFORMATION AND MINERAL GROWTH IN THE STRANGWAYS METAMORPHIC COMPLEX: IN-SITU DATING OF ZIRCON AND MONAZITE IN A STAUROLITE-CORUNDUM BEARING SHEARZONE A. Moller^'*, I.S. Williams^, S. Jackson^ & BJ. Hensen^ 1* School of Geology, University of New South Wales, Sydney, NSW 2052 curr. address: Institutf Geowiss., Univ. Mainz, 0-55099 Mainz, amoeller@maiLuni-mainz.de 2 RSES, The Australian National University, Canberra, ACT 0200 3 GEMOC, School of Earth Sciences, Macquarie University, Sydney, NSW 2109

The correlation of metamorphic and/or deformation history with the age of accessory phases is a problem which has won more attention recently. Dating minerals with high closure temperatures in-situ (newly grown zircon by SHRIMP and monazite by LA-ICP-MS) enables the direct correlation of deformation and metamorphic history as deduced from mineral assemblages and textures with the isotopic results. Exhumation of the granulites in the Strangways Metamorphic Complex (SMC) of the Arunta Inlier in central Australia has been explained by thrusting along lower-grade shearzones. Final exhumation during the Alice Springs Orogeny has previously been dated using minerals with low closure temperatures (biotite, muscovite) at 300-400 Ma. It has been debated whether the metamorphic grade of the shearzones is correlated with their age during periodic (stepwise) exhumation, i.e. whether the amphibolite facies shears originated in the Proterozoic and greenschist facies shears in the Palaeozoic.

STAUROLITE-CORUPWUM BEARING SHEARZONE

A staurolite bearing shearzone has been found within high-grade metapelites in the north-central SMC at the Edwards Creek copper prospect. Textural relationships of minerals in this shearzone and in-situ dating of complex zircons (Fig.l) reveal a phase of new zircon growth at 443 ±6 Ma (prior to the generally accepted age of the ASO) and followed by development of the St-Bt-Chl assemblage in which monazites have an age of about 385 Ma. The metapelitic "protolith" was probably affected by desilification through shearzone-fluids which left corundum clusters, possibly at the site of previous garnets. Proterozoic zircon cores with polycrystalline Palaeozoic overgrowths occur within the chlorite-biotite matrix, but also completely included in staurolite. The overgrowths therefore predate or are contemporaneous with staurolite growth. Some zircons show cracks which are partially filled by xenotime. Xenotime also occurs as inclusions in the monazite aligned along biotite and chlorite in the matrix. Monazite and xenotime growth thus postdates the deformation which cracked the zircons. Other phases in the biotite-chlorite matrix include rutile, apatite and a high-Th-phosphate. These accessory minerals are the subject of ongoing investigations to decipher the timeframe of mineral growth activity in this shear zone.

IMPLICATIONS FOR THE EVOLUTION OF THE SMC AND THE ARUNTA BLOCK

It can be speculated that deformation and growth of zircon as well as the assemblage staurolite-biotite-chlorite in the shearzone occurred at the same time as growth of Opx-And symplectites under reduced water activities in metapelitic granulites (Ballevre et al., 1997), where lack of water inhibited recrystallisation and mineral growth. Together with evidence from other areas (see below) the Palaeozoic growth of staurolite in the shearzone suggests that it can no longer be assumed that higher grade shearzones are older than lower grade ones and the Palaeozoic influence on the SMC was merely a retrograde one. The zircon growth event is contemporaneous with or somewhat younger than high grade metamorphic events in the Harts Range (Miller et al., 1998, Hand et al., 1999) further to the east. There is a marked contrast between the SMC and the Harts Range (HR). Upper amphibolite facies metamorphism during the Palaeozoic either caused widespread resetting or recrystallisation of monazite in the HR or those high grade sequences have to be re-interpreted as Palaeozoic formations, as suggested by the discovery of Neoproterozoic detrital zircons in metapelites from the Mallee Bore granulites (Miller et al., 1998). Evidence for Palaeozoic activity in the SMC has so far been sparse. Evidence for prograde metamorphism to amphibolite grade has recently been found to occur late to post-kinematically in the eastern SMC at 381 ±7 Ma (Ballevre et al., submitted), contemporaneous with the age of the monazites in the staurolite shear at Edwards Creek.

FOOD FOR THOUGHT

The Palaeozoic evolution in the Arunta Block, which involves not only 'retrograde shearing' and exhumation along thrusts, but deformation on a crustal scale and high-grade metamorphism, has to be put in the wider scenario of the whole Australian plate. The Ordovician-Silurian plate-tectonic and magmatic activity at the eastern continental margin may have provided some of the driving forces for the Palaeozoic events recently identified in the central

OROGENESIS IN THE OUTBACK

Australian region, and future work on the high grade rocks of the Arunta Block has to take into account the progress in interpretation of the young fold belts of Victoria and NSW. REFERENCES Ballevre, M., Hensen, B J . & Reynard, B., 1997. Orthopyroxene-andalusite symplectites replacing cordierite in granulites from the Strangways Range (Arunta block, central Australia): A new twist to the pressure-temperature history. Geology 25, 215-218. Ballevre, M., Moller, A. & Hensen, BJ., submitted. An Alice Springs (380 Ma) age for a prograde amphibolitefacies shear zone in the Strangways Metamorphic Complex, Arunta Block, submitted to Precambrian Research. Miller, J.A., Buick, I.S., Williams, I.S. & Cartwright, I., 1998. Re-evaluating the metamorphic and tectonic history of the Eastern Arunta Block, central Australia, 14th Australian Geological Convention , Townsville, Geological Society of Australia, abstracts 49, 318

Fig. 1: Backscatter electron image of complex zircon with metamict core, broad Proterozoic metamorphic rim and polycrystalline heterogeneous Palaeozoic overgrowth. Bright inclusions are xenotime. Size of ion-microprobe spots is about 20 |im.

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

EPISODIC INTRAPLATE OROGENY IN CENTRAL AUSTRALIA; IS IT MANTLE DRIVEN? E. A. Neir and G. A. Houseman^ ^AGCRC, VIEPS Department of Earth Sciences, Monash University, Clayton, 3168. ^VIEPS Department of Earth Sciences, Monash University, Clayton, 3168. eneil@earth. monash. edu. au

Central Australia has experienced multiple events of deformation and tectonic activity since the Proterozoic. The last tv^o events to have effected the region were the Petermann Ranges Orogeny (-600-500 Ma) and the Alice Springs Orogeny (-400-300 Ma). Both these events have been recognized as examples of intraplate orogeny. There are two possible sources for the stresses required to produce intraplate orogeny. These stresses could originate at the plate boundary and be transferred through the plate to the site of the deformation. Alternatively, the stresses could be mantle derived and acting on the base of the crust to produce thickening. The intraplate deformation would then be an expression of deeper processes occurring in the mantle. Plate boundary driven orogeny may involve extensive deformation of both the plate boundary region and the intraplate region simultaneously, as seen in the modem India-Asia collision producing both the Himalayan orogen at the boundary and the intraplate Tien Shan range 2000 km further north. Mantle driven intraplate orogeny, however, provides a possible mechanism for crustal thickening and deformation within the interior of a plate without requiring deformation of the plate boundary. In this study we investigate the possibility of producing intraplate orogeny from forces produced by gravitational instability of the mantle lithosphere. The Rayleigh-Taylor instability describes the gravitational instability of a layer of dense fluid on top of a less dense fluid. Under these unstable conditions, small deflections of the boundary between the two layers grow, at first exponentially, until the system overturns and the high-density material drops into, and is replaced by, hot asthenosphere. Here we analyse the Rayleigh-Taylor instability of lithosphere that consists of a buoyant crustal layer above a high-density mantle layer, which overlies a lower density asthenospheric half space. Incorporating a buoyant crustal layer slows the growth of the instability in the mantle. The downwelling lithospheric mantle transmits shear stresses to the base of the crust, causing crustal thickening. This crustal thickening mechanism provides a model for mantle-driven intraplate orogeny.

f^O.Ov f^O.87

asthenosphere

Numerical simulations of the mantle instability were calculated for crust and mantle layers with a constant density and constant Newtonian viscosity. These calculations show that the buoyant crustal layer may be thickened, and the surface uplifted, over the lithospheric downwelling, to produce a maximum crustal thickening factor of -1.4 (for typical lithospheric parameters) (Figure 1). This is enough to thicken a 35 km crust to 50 km and produce a significant intraplate mountain range. We find that thick, buoyant continental crust causes the instability to occur at a horizontal wavelength of between 200 and 400 km. Such an instability occurring in the mantle lithosphere beneath central Australia could account for the shortening associated with the intraplate orogenies which have occurred there. Figure 1. The development in time of the lithospheric downwelling caused by gravitational instability of the mantle lithosphere. Solid lines show the base of the lithosphere at sequential dimensionless times. Labels refer to the dimensionless time (O at each stage of the downwelling. Dashed lines represent the position of the Moho initially (black), and at the end of the experiment (grey). S is the maximum crustal thickening obtained in the experiment.

OROGENESIS IN THE OUTBACK

For Rayleigh-Taylor type instability to occur within the sub-continental lithospheric mantle it must occur on a time scale that is faster than the timescale for thermal diffusion in the lithosphere. Otherwise the downwelling lithosphere will have time to thermally equilibrate with the surrounding asthenosphere. It would then loose its thermally induced density difference and the gravitational instability would be unable to grow. For Rayleigh-Taylor type instability to occur in continental lithosphere, before thermal equilibration can occur, requires a lithospheric viscosity of Pa s (Neil and Houseman, 1999). A viscosity of order Pa s implies that the lithosphere must be relatively weak in order for Rayleigh-Taylor type instability to occur. If, therefore, gravitational instability is responsible for the intraplate deformation in central Australia, it implies that the central Australian lithosphere must have been weak or at least weakened prior to each intraplate event. Heating of the lithosphere reduces its strength and therefore its ability to resist gravitational instability. Sandiford and Hand (1998) have proposed that heating and weakening of the lithosphere prior to both the Petermann Ranges Orogeny and the Alice Springs Orogeny occurred as a result of the deposition of thick sediment sequences in the Amadeus Basin. These sediments insulated the high heat producing basement of central Australia, heating the crust and elevating the Moho temperature. Since the strength of the lithosphere is sensitive to the temperatures at the base of the crust (England, 1983) such a mechanism may have reduced the viscosity of the lithosphere allowing gravitational instability of the lithosphere to occur. Sandiford and Hand (1998) proposed that prior to the Petermann Ranges Orogeny the 4 km thick sedimentary cover in the southern Amadeus Basin could insulate the high heat producing basement and increase the Moho temperature by ~30> reducing the lithospheric strength in that region. Once the lithosphere is sufficiently weak for gravitational instability to occur, the high-density lithosphere could sink into the asthenosphere, thickening and shortening the overlying crust to produce the Petermann Ranges Orogeny. After detachment of the downwelling mass has occurred, the lithosphere enters a phase of thermal re-equilibration. During this time the topographic elevation produced by the crustal thickening is eroded. The hot asthenosphere which has replaced the lower layers of the lithosphere slowly cools over a period of order 100 Myr (determined by the thermal time constant of the lithosphere). The heat lost from the cooling asthenosphere propagates up through the crust during which time the crust is heated. During this period the lithosphere gradually cools and strengthens, the elevated topography over the convergent zone is removed by erosion, and a period of sedimentation occurs. Such sedimentation could contribute to the 7-8 km of sediment deposited in the northern and southern regions of the Amadeus Basin proposed by Sandiford and Hand (1998) to have insulated the lower crust and caused an increase in Moho temperature of -'110>C prior to the Alice Springs Orogeny. Once the lithosphere has had time to thermally re-equilibrate and thicken again, a second downwelling beneath the northern Amadeus Basin may have occurred to produce the Alice Springs Orogeny. For this second instability to occur the increase in Moho temperature caused by sedimentation (Sandiford and Hand, 1998) must have been sufficient to reduce the lithospheric viscosity to allow gravitational instability to grow. Preexisting topography on the Moho could also be a factor in determining where a second downwelling would occur. The time required for the lithosphere to thermally re-equilibrate could explain the -200 Myr separation of the Petermann Ranges and Alice Springs orogenies. We propose that crustal thickening driven by gravitational instability in the mantle is a possible source of stress required to produce intraplate orogeny. Such a mechanism could provide a source for the stresses that produced the Petermann Ranges orogeny and then, 200 Myr later, the Alice Springs Orogeny approximately 400 km further north. REFERENCES

England, P.C., 1983. Constraints on extension of continental lithosphere, J. Geophys. Res., 88,1145-1152. Neil E.A., and Houseman, G.A., 1999. Rayleigh-Taylor instability of the upper mantle and its role in intraplate orogeny, Geophys. J. Int., 138, (in press). Sandiford, M., and Hand, M., 1998. Controls on the locus of intraplate deformation in central Australia, Earth Planet. Sci. Lett., 162, 97"110. Work presented here was conducted as part of the Australian Geodynamics Cooperative Research Centre and this abstract is published with the permission of the Director, AGCRC.

ALICE SPRINGS, AUSTRALIA, 1999

LITHOSPHERIC STRENGTH AND PLATE BOUNDARY STRESSES DURING THE INTRAPLATE ALICE SPRINGS OROGENY E. A. Neil^ and G. A. Houseman^ MGCRC, VIEPS Department of Earth Sciences, Monash University, Clayton, 3168. ^VIEPS Department of Earth Sciences, Monash University, Clayton, 3168. eneil@earth. monash. edu. au

Between 400 Ma and 300 Ma central Australia experienced a period of widespread compression and uplift known as the Alice Springs Orogeny. This event caused the uplift of the Arunta Block, probably by reactivation of the Redbank Thrust Zone, and produced extensive folding, thrusting and low-grade metamorphism along the northern margins of the Amadeus and Ngalia Basins. During the earlier part of the Alice Springs Orogeny, when central Australia was undergoing shortening and compression, the region of the Canning Basin in north-western Australia experienced a period of extension driven subsidence. The Alice Springs Orogeny has been recognized as an example of intraplate orogeny and we investigate the lithospheric properties, plate boundary stresses and the distribution of lithospheric strength, which allow intraplate extension and compression simultaneously, in the regions of the Canning Basin and central Australia respectively. The thin viscous sheet model with non-Newtonian rheology and a laterally varying strength parameter is used in this study to attempt to constrain the tectonic environment required to produce compression in central Australia and extension to the north-west in the Canning Basin. We investigate a class of model for this intraplate event, in which deformation is driven by stresses acting on remote plate boundaries, in particular the northern boundary of the continent. On the northern boundary we impose a traction condition that causes the northern boundary to rotate in a clockwise direction. The rotation of this boundary could occur in response to two kinds of traction. An eastdirected traction z^ (initially equivalent to tangential stress CTxy), causes the upper boundary to move eastwards. It is also possible that the observed deformation has been produced by a plate boundary torque or bending moment which is implemented by a north-directed traction imposed on the upper boundary, such that it varies linearly from +Ty to -Ty across the northern boundary. These tractions cause dextral shearing between the northern boundary and the southern boundary, assumed to be embedded rigidly in the Gondwana supercontinent.

Figure 1. The evolution in time of the thin viscous sheet with rotational traction on the northern boundary (r^ = 1; Ty = 1; n = 3). To the left is a plot of the initially dimensionless strength distribution within the sheet, 5.0 at the north and south boundaries decreasing to 1.0 in the center of the sheet. This central weak zone is deformed as shown by the crustal thickness distribution (grey scale) with contours at 5 km and reference (initial) crustal thickness at 35 km. The region of thickening and compression corresponds to central Australia, and the region of thinning and extension corresponds to the Canning Basin.

Whatever the source of the boundary stresses that produced the Alice Springs Orogeny, the fact that deformation was focused in the central Australian and Canning Basin regions must reflect on the lithospheric properties in those regions. It appears from the concentration and intensity of deformation in the Arunta and Musgrave Blocks throughout the Proterozoic and Paleozoic that the lithosphere beneath the central Australian region has been relatively weak. A weak lithosphere would enable the central Australian lithosphere to deform more easily than surrounding terrains and absorb the strain associated with plate boundary stresses. If tectonic events occurred preferentially in central Australia because of a weak lithosphere, it appears that the lithosphere would have remained weak after each event or at least have been weakened again prior to each further event. We therefore include a central zone of weakness within the sheet representing the lithosphere of central Australia and the Canning Basin (Figure 1). The results show that either of these boundary stress configurations may produce clockwise rotation of the northern boundary of the plate and simultaneously, the regions of compression and extension across the center of the continent, with deformation concentrated in the weak zone (Figure 1).

O R O G E N E S I S IN THE OUTBACK

The maximum crustal thickening factor (Anax) in the weak zone depends almost linearly on the angle of rotation of the northern boundary (0). It appears that the actual stress regime that produces the rotation does not influence the dependence of on 6 since for all combinations of z^ and Ty tested the results follow the same path (Figure 2). To produce a maximum crustal thickness factor of 1.66 (corresponding to 100 km of shortening in a deforming zone of width 400 km) requires a rotation of the northern boundary of -45. Our results show that in the model a clockwise rotation of the northern boundary may be induced by an eastward shear traction or by an anticlockwise bending moment. In either case a rotation of approximately 45* produces crustal thickening in the eastern part of the weak region (corresponding to central Australia) and crustal thinning in the western part corresponding to the Canning Basin. If such a rotation occurred between northern and southern Australia during the Alice Springs Orogeny it could, in principle, be detected by palaeomagnetic data. By assuming that the deformation associated with the Alice Springs Orogeny occurred over a period of 100 Myr, and that the central Australian lithosphere was shortened 100 km in a deforming zone of width 400 km = 1.66) we can investigate the strength of the central Australian lithosphere implied by this orogenic event. The actual value of lithospheric strength prior to the Alice Springs Orogeny will depend on the thermal structure of the lithosphere. Since the magnitude of the driving force is relatively unconstrained, we calculate the relationship between applied boundary force and lithospheric strength required to produce 100 km of shortening in 100 Myr. England (1983) showed that, for lithosphere for which the integrated strength resides primarily in the mantle, the force per unit length acting on a vertical section may be expressed approximately as a function of Moho temperature. Therefore, the strength of the lithosphere and the boundary force required to deform it may be expressed in terms of temperature at the Moho (Figure 3).

A n g l e of rotation

(0)

Figure 2. The dependence of rotation angle (0) on the ln[/?,„yj. Each line represents a different combination of values of r^ and Ty, acting on the sheet shown in Figure 1 with a non-Newtonian viscosity law in which strain-rate is pro{2ortional to stress cubed (n = 3).

550

600

650

M o h o Temperature

700

fC)

Figure 3. Boundary force vs Moho temperature for different orogenic time spans (labeled on each curve in Myr) required to produce a maximum crustal thickening factor of |3„,ax = 1-66. Graph based on the England model for lithospheric strength. All parameters used are based on the rheology of wet olivine (Karato er ai 1986) and the geothermal gradient below the lithosphere is assumed constant at 8' C per km. The boundary stress is assumed to be a east directed shear (t^^ = 1, Ty = 0).

In this model there is a trade off (Figure 3) between the magnitude of the boundary force, the time interval during which the deformation occurs, and the strength of the lithosphere (described by Moho temperature). If the deformation occurred over 100 Myr then the boundary force required to deform the weak lithosphere is of order 10^^ Nm*' for a Moho temperature of --485 "C. Shorter periods of deformation would require higher Moho temperatures or greater boundary forces. If Moho temperatures prior to the Alice Springs orogeny were higher than those predicted here, shorter periods of deformation are permitted and it may be that deformation in central Australia was episodic over a 100 Myr period rather than continuous. REFERENCES

England, P.C., 1983. Constraints on extension of continental lithosphere, J. Geopliys. Res., 88, 1145-1152. Karato, S., Paterson, M.S., and Fitzgerald, J.D., 1986. Rheology of synthetic olivine aggregates: Influence of grainsize and water, J. Geophys. Res., 91, 8151-8176. Work presented here was conducted as part of the Australian Geodynamics Cooperative Research Centre and this abstract is published with the permission of the Director, AGCRC.

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

CRUSTAL ANATEXIS AND IN SITU GRANITE FORMATION IN THE PAN-AFRICAN NORTH-EQUATORIAL FOLD BELT: THE MIGMATITES AND ASSOCIATED GRANITES OF THE BANTOUM AREA, WEST-CAMEROON, CENTRAL AFRICA C. Nzolang, J.P. Nzenti, J.P. Tchouankoue, A.A. Ganwa University of Yaounde /, Department of Earth Sciences, B. P. 812, Yaounde,

Cameroon.

The pan-African north-equatorial fold belt (700-450 Ma) is an important link of the pan-African orogeny in so far as it represents the transition between the Trans-saharian fold belt in West Africa and the East African chains. It covers a broad region in Central Africa comprising Cameroon, Central African Republic, Chad, Nigeria, south Niger and west Sudan. The Cameroon part of that chain consists of three main structural and lithological domains: (1) the northern domain includes multiphase metamorphic formations and calc-alkaline intrusions (630 Ma) that are characteristic of an active margin environment; (2) the central Cameroon domain is affected by important shearing (N30°E, N70°E) marked out with numerous syntectonic calc-alkaline, alkaline to aluminous plutons, and it includes polycyclic metamorphic formations of medium- to low-pressure which contain remnants of reworked Archaean nuclei (2100 Ma); (3) the southern domain is marked by a prograde metamorphism reaching the granulite facies within the Yaounde Series and it overlaps the Archaean formations of the Congo Craton (3000-2800 Ma). Nzenti et al (1994) interpret the pan-African north-equatorial fold belt in Cameroon to be a typical intracontinental chain, which is characterized by a major intracrustal thrusting prepared by a rifting. The Bantoum area in west Cameroon is a part of the central domain; it is situated nearby the formations whose age is uncertain and assumed to be 2100 Ma ("Nde orthogneisses"; Lasserre and Soba, 1979). The regional structural evolution comprises two main phases of deformation: the phase D1 is dominated by the schistosity and foliation planes SI with a mean direction N30°E, associated with a stretching lineation LI and isochnal folds PI; the phase D2 which is essentially shearing is characterized by shearing planes C2 often filled with leucosomes, and folds P2 associated with an axial pl||ie schistosity S2. The studied area is made up of migmatitic gneisses to which are closely associated: (1) amphibolites interlayered or as centimetre- to metre-sized enclaves in these gneisses; (2) a set of porphyroid biotite granites, leucogranites and granodiorites occurring as concordant strips of variable widths (5 to 200 m) roughly elongated following the N70°E direction; (3) small veins of cordierite granites cut across the above formations. The leucogranites and cordierite granites are peraluminous and show the characteristics of S-types granites; they are interpreted as derived from a crustal melting in a collisional context. The mineral assemblages in gneisses and amphibolites indicate a high grade amphibolite facies metamorphism. The migmatization is evidenced by the development of: (1) quartzo-feldspathic leucosomes forming centimetric layers surrounded by thinner biotite-rich melanosome layers and parallel to the structural-regional trends; (2) granitic to pegmatitic leucosomes in veins or injected along the tectonic foliation or into shear planes. Therefore, the Bantoum region displays a highly anatectic zone showing an example of in situ granite formation during the Neoproterozoic event in Cameroon, possibly related to a collisional environment. These results seem to us of greatest interest in the search for a geodynamic model consistent with the evolution of the pan-African northequatorial fold belt in Cameroon. REFERENCES

Lasserre, M. and Soba, D., 1979. Migmatisation d'age panafricain au sein des formations camerounaises appartenant a la zone mobile de 1'Afrique centrale. C.R. Somm. Soc. Geol. Fr., 2, 64-68. Nzenti, J.P., Barbey, P., Bertrand, J.M. and Macaudiere, J., 1994. La chame panafricaine au Cameroon: cherchons suture et modele! In Soc. Geol. Fr., ed., 15e Reunion des Sciences de la Terre, Nancy, Abst. p 99.

OROGENESIS IN THE OUTBACK

PAN-AFRICAN OVERPRINT OF THE GRENVILLIAN BASEMENT IN CENTRAL DRONNING MAUD LAND. H.-J. PAECH Bundesanstalt fur Geowissenschaften und Rohstoffe (BGR), Postfach 510153, D-30631 Germany, e-mail: hans.paech@bgr.de

Hannover,

The geological history of central Dronning Maud Land (cDML) will be summarized on the basis of observations recently obtained during the international GeoMaud expedition and from the literature. The high-grade metamorphic basement in cDML is composed of supracrustal sequences (bimodal metavolcanic and metasedimentary rocks) and acid orthogneisses intruded prior to the metamorphism and deformation. The metamorphic basement is intruded by an anorthosite suite and a later voluminous post-kinematic granitoid suite (syenite, partly retrograded charnockite, monzodiorite and granite). These intrusive suites form the characteristic igneous province of cDML, and are locally accompanied by patchy chamockitization features in the metamorphic basement. The timing of the tectonic and metamorphic processes in cDML is now constrained by geochronological data (Jacobs et al. 1998). SHRIMP data has substantiated the accumulation age of the supracrustal protoliths as about 1080 Ma.The metamorphic basement of cDML was formed during the Grenvillian that was accompanied by acid intrusions that are now orthogneiss. Thus, the metamorphic basement in cDML has a Grenvillian origin. The Grenville structures were overprinted and almost completely obliterated by pervasive Pan-African metamorphism and tectonism. Furthermore, a particular orthogneiss also yield a Pan-African intrusion age. During the Pan-African events the Gruber anorthosite intruded (600 Ma) which was subsequently affected by high-grade metamorphism and tectonism (570 Ma). The final event in the history of the basement was the intrusion of large amounts of granitoid magma (about 500 Ma ago) and a small anorthosite, which do not show pervasive Pan-African foliation. Thus, the dominant structural and metamorphic features in cDML are Pan-African in age. Foliation and fold structures in the western part of cDML (Orvinfjella) trend uniformly ENE-WSW. In contrast, the tectonic trend in the eastern part (Wohlthat Massiv) varies dramatically. Around and in the outer part of the Gruber anorthosite, the foliation wraps around the rigid anorthosite core. This can be explained as the result of ductile behaviour of the metamorphic basement in an oblique sinistral regime, which is also evidenced in the Orvinfjella by sinistral shear zones (Bauer et al. 1999). In general, the tectonic position of cDML has much in common with the Mozambique Belt of eastern Africa, and other Gondwana regions of Antarctica. REFERENCES

Jacobs, J., Fanning, C.M.; Henjes-Kunst, F., Olesch, M. & 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. - T h e Journal of Geology, 106-406, Chicago. Bauer, W.; Jacobs, J. & Paech, H.-J. (1999): Indications for Pan-African collisional tectonics in central Dronning Maud Land (Antarctica).- Abstracts 8th Intem.Symp. Antarctic Earth Sciences Wellington.

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

GEOCHEMISTRY OF COEXISTING NA RICH AND K GRANITES: A LATE PROTEROZOIC THERMAL EVENT IN THE NORTH DELHI FOLD BELT, WESTERN INDIA M. K. Pandit

Department of Geology, University of Rajasthan, Jaipur 302004 India E-mail: panditmanoJ@hotmaiLcom

The Middle to Late Proterozoic Delhi Fold Belt in the western India is exposed in two (temporally distinct) domains. The diachronous sedimentation history of Delhi rocks, identifying Middle Proterozoic, North Delhi Fold belt (NDFB) and Late Proterozoic, South Delhi Fold Belt (SDFB) components has been established on the basis of distinct ages for intrusive granitoids. The 1700 - 1500 Ma granitoids accord Middle Proterozoic age to the NDFB while the SDFB, intruded by the 850 Ma granitoids (Erinpura granite) represents the younger component. Recently, much younger (possibly Late Proterozoic) granitic rocks in the NDFB have been reported from the Ajitgarh (27o26' N: 75o50' E) from the NDFB. The Ajitgarh pluton, comprising predominant K-rich alkali granite (pink) and subordinate Na-rich trondhjemitic granite (leucocratic) components, intrudes the metasedimentary rocks of North Delhi Fold Belt in the northwestern Indian shield. The alkali granite is characterized by predominant K-feldspar, quartz, sodic plagioclase and minor biotite and hornblende while 'trondhjemitic granite' encompasses predominant quartz and sodic plagioclase, minor K-feldspar and accessory homblende-sphene and zircon. Both the granites are undeformed and do not show any evidence of metamorphism or post-crystallization alteration/chemical substitution. Sodic affinity, high silica abundance, depletion in Ba, Rb, Sr and enrichment in the HFSE are the characteristic features of the trondhjemitic granite which differs from typical trondhjemites on account of low CaO (0.54 1.51%), low MgO (0.01 - 0.54%), low Sr (21 - 56 ppm) and high FeO/MgO ratio. These geochemical characters and the absence of intermediate rocks do not favour their evolution by crystal - liquid fractionation alone or through a simple two stage melting of a low-K tholeiitic source. The K-rich alkali granite has lower silica abundance and higher CaO, MgO and A1203 as compared to trondhjemitic granite. Lack of any systematic geochemical trends in case of alkali granite can be attributed to possible mixing with basic melts. Mafic enclaves, exclusively confined to the alkali granite, represent the remnant basic magmatism injected into the granitic melt chamber. No such enclaves have been observed in trondhjemitic granite. The close temporal and spatial association of both granite types and remarkable similarity in their REE patterns indicate a possible cogenesis between them. It is suggested that the trondhjemitic granite is the outcome of selective ion exchange of alkalis between the normal (potassic) granite melt and basic magma. Alkali exchange along the acid-base interface has been considered to be the most rapid process.

OROGENESIS IN THE OUTBACK

PETROGENESIS OF THE SCAPOLITE-BEARING CALC-SILICATE ROCKS IN THE WESTERN IMJINGANG BELT, KOREA H. Park and M. Cho

Department of Geological Sciences, Seoul National University, Seoul, 151-742, Korea

The Qinling-Dabie-Sulu belt, produced by Triassic continental collision between Sino-Korean and Yangtze cratons, may continue eastward to the Imjingang Belt in central Korean Peninsula (Ree et al., 1996). Part of this belt is exposed in South Korea, probably underlying the Devonian Imjin Group of North Korea, and is divided into northern Jingok unit and southern Samgot unit. While the Jingok unit comprises the Barrovian-type metapelites of garnet to kyanite zones, the Samgot unit consists mainly of calc-silicates, semipelitic schists, marbles, and amphibolites of the upper amphibolite facies. We present our priliminary results on the latter, in particular, scapolite-bearing calc-silicate rocks. Various iithologic layers of the Samgot unit are commonly intercalated with each other at outcrop as well as thinsection scales. These intercalations forming straight layers are the product of ductile deformation at high temperatures of impure carbonate-rich sequences. The calc-silicate rocks are granoblastic and 0.5-3 mm in average grain size. Prograde assemblages consist of various combinations of calcite, quartz, clinopyroxene, scapolite, clinozoisite, Ca-amphibole, calcic plagioclase, almadine-grossular garnet, biotite and titanite. No mappable variation in mineralogy was found. Absence of wollastonite and no evidence of partial melting, indicative of the upper amphibolite facies, are consistent with the P-T condition estimated from the Samgot amphibolite (8-11 kbar and 630-740 Ree et al., 1996). In the model system Ca0-Al203-Si02-H20-C02 (CASV), low-variance assemblages including garnet (Grs4o), Ca-plagioclase (Ango), meionitic scapolite (Eq.Ango), clinozoisite (Al/(Al+Fe)=0.9), quartz and calcite define various reactions intersecting approximately at 0.1-0.4 X^^q • ^^^ uncommon coexistence of clinozoisite and scapolite that are incompatible in the CASV system can be accounted for by the impurities in solid phases or the local variation in fluid compositions. In order to understand the behavior of fluid, we used the method of Oliver et al. (1992). When the model CASV system is extended to include FeO and MgO, aC02 / aH20 can be estimated from biotite-clinopyroxene-fluid phase relations. Our preliminary result suggests that fluid was not exchanged to a significant degree between individual layers. In addition, the occurrence of Cl-bearing scapolite (Xci=0.3) in close association with Cl-free one indicates the presence of Cl-bearing fluid at least on a local scale, and consequently of small salinity gradient. All of these observations suggest that metamorphic fluid is governed primarily by internal buffering in calc-silicate rocks. REFERENCES

Oliver, N., Wall, V., and Cartwright, I., 1992. Internal control of fluid compositions in amphibolite-facies scapolitic calc-silicates, Mary-Kathleen, Austraha. Contribution to Mineralogy and Petrology, 111, 94-112 Ree, J.-H., Cho, M., Kwon, S.-T., and Nakamura, E., 1996. Possible eastward extension of Chinese collision belt in south Korea: The Imjingang belt. Geology, 24, 1071-1074

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

40 a - / 3 9

A r r A r APPROACHES TO DATING SHEAR ZONES.

D. Phillips\ R.L Gibson^ C.R. Anhaeusser^ and G.B. Kiviets^ 1. Research School of Earth Sciences, Australian National University, Canberra, ACT, 0200, Australia. 2. Department of geology, University of the Witwatersrand, Private Bag 3, Wits, 2050, South Africa. 3. De Beers GeoScience Centre, P.O. Box, 82232, Southdale, 2135, South Africa.

An understanding of the timing and thermal evolution of crustal shear zones is crucial to the evaluation of tectonic models for metamorphic and structural terrains. Shear zones provide a focus for regional deformation, fluid flow and the subsequent recrystallisation and/or growth of new minerals such as micas, which are candidates for K-Ar and "Âr/^Âr dating techniques. While these minerals are relatively straightforward to analyse for argon isotopes, the interpretation of the results may be extremely complex. The principal goal of this presentation are to critically review "Âr/^Âr dating methods and interpretations as applied to shear zones, using selected examples from the literature as well as recent ÂrPÂr laser probe data from a shear zone related to the Vredefort impact event in South Africa. Of fundamental importance to the interpretation of "^Âr/^Âr data from shear zones is the relationship between the timing of mineral crystallisation and the thermal history of the shear zone in relation to relevant blocking temperatures for argon diffusion (e.g. Dunlap et al., 1991; West and Lux, 1993; Dunlap, 1997). In general, elucidation of this relationship requires some independent constraints on the thermal history of the deformation zone (e.g. Dunlap, 1997). If shearing and mineral growth occurred above the applicable closure temperatures, then the minerals will record cooling ages. However, if the minerals grew or recrystallised below these blocking temperature, then the time of mineral formation and shearing may be obtained. While desirable for constraining the age of shearing, the latter case is often compromised by the presence of detrital or inherited (unrecrystallised) grains from the host rock, which are not isotopically reset during shearing. Attempts to circumvent this problem include the analysis of fine grained mineral separates to exclude coarser detrital (or inherited) grains (e.g. West and Lux, 1993) and the in situ analysis of appropriate grains using the "^ÂTPÂT laser probe technique (e.g. Scaillet et al., 1990; present study). While the preparation of progressively finer grain sizes can improve the results, the complete removal of unrecrystallised material from bulk samples remains a very difficult task (cf. West and Lux, 1993). Analyses of single grains would seem to be an obvious solution - however, the fine grain size of micas in many brittle-ductile shear zones restricts the application of this approach. An alternative method involves the in situ laser probe analysis of clusters of appropriate grains - the main disadvantages of this method are that the ages are effectively K-Ar ages and the laser beam may cause degassing of adjacent or buried unrecrystallised material. The latter concerns can be addressed by undertaking repeat spot analyses of relevant areas, as well as analyses of surrounding material. An alternative, although difficult option is to excise clusters of appropriate material for laser step-heating analyses. "ÂrPÂv data may also be compromised by the incorporation of excess argon into minerals during recrystallisation/growth. In general an evaluation of the presence or absence of excess argon must be assessed by relation to geological and other isotopic constraints. A first order evaluation can be made from the reproducibility of the argon results (e.g. plateau age spectra), although this is not a guarantee of an excess-free system. The controversy surrounding the age of high pressure metamorphism and deformation in the western Alps is a good example of the problems associated with excess argon contamination (cf. Arnaud and Kelley, 1995). Some of the above-mentioned problems are manifest in a current study of a shear zone suspected to be related to the Vredefort impact structure in the Witwatersrand Basin, South Africa. The shear zone outcrops on the northwest edge of the Johannesburg Dome, where strongly cleaved granite basement rocks are overlain by Black Reef quartzites. The basement cleavages have been interpreted by some workers as predating the Black Reef quartzites (e.g. Hilliard and McCourt, 1995). However, detailed studies by McCarthy et al., (1986) indicated that the basement cleavages extend into the Black Reef quartzites and also cut Bushveld Complex dykes, suggesting a post-Bushveld tectonic event, possibly related to the Vredefort impact event. ^ÂTPÂT laser probe analyses of coarse muscovite porphyroblasts yielded ages ranging up to 3.1 Ga, in agreement with recent U-Pb zircon ages for granodiorites of the Johannesburg Dome. Locally, smaller disaggregated porphyroblasts yielded ages as young as 2047 ± 1 2 Ma, indicating variable loss of argon within highly sheared areas. Analyses of sericite clusters yielded apparent ages ranging from 1995 ± 18 Ma to 2337 ± 10 Ma, although most are younger than 2.1 Ga. The youngest five ages average 2027 ± 11 Ma, within error of recent U-Pb zircon determinations of the timing of the Vredefort impact event (Moser, 1997). Therefore, the current data are consistent with the geological model of McCarthy et al. (1986), relating the last shearing event in the Johannesburg Dome to the Vredefort impact. The discordance in the sericite results could result from incomplete resetting of inherited grains. However, geological and textural constraints and the results from disaggregated muscovite

O R O G E N E S I S IN THE OUTBACK

porphyroblasts suggest that the older apparent ages are more likely due to the incorporation of small amounts of excess argon, derived from the potassium-rich host rock. In conclusion, ^ATP^AT dating techniques, particularly laser probe methods, have enormous potential for constraining the timing of crustal deformation. However, data interpretation requires a thorough understanding of local and regional thermal and fluid flow regimes.

REFERENCES

Arnaud, N.O. and Kelley, S.P., 1995. Contributions to Mineralogy and Petrology., 121,1-11. Dunlap, W.J., 1997. Chemical Geology, 143, 181-203. Dunlap, W.J., Teyssier, C., MacDougall, 1. and Baldwin, S., 1991. Geology, 19, 1213-1216. Hilliard, P. and McCourt, S., 1995. South African Journal of Geology, 98, 349-355. McCarthy, T.S., Charlesworth, E.G. and Stanistreet, I.G., 1986. Transactions of the Geological Society of South Africa, 89,311-324. Moser, D.E., 1997. Geology, 25, 7-10. Scaillet, S., Feraud, G., Lagabrielle, Y., Ballevre and Ruffet, G., 1990. Geology, 18, 741-744. West, D.P. and Lux, D.R., 1993. Earth and Planetary Science Letters, 120, 221-237.

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

RE-EQUILIBRATION OF GRANULITES IN A TRANSCURRENT SHEAR ZONE IN SOUTH-EASTERN BRAZIL. C.C. Porcher\ G.T.R. D r o o p ^ L A . D . Fernandes'''K.H. Brodie^

1. CPGq, Universidade Federal do Rio Grande do Sul, P.O. Box 15065-9150/970, Porto Alegre, RS, Brazil, e-mail: ferna@if.ufrgs.br. 2. Department of Earth Sciences, University of Manchester, Oxford Road, Manchester, M13 9PL, UK.

The Paraiba do Sul Shear Zone (PSSZ) is one of the main structures of the Rio Paraiba do Sul Shear Belt. We have undertaken a petrological investigation of the metamorphic evolution of rocks from two areas of the PSSZ in Rio de Janeiro State, using petrographic and geothermobarometric methods integrated with structural analysis. The rocks of these regions were affected by two main high-grade metamorphic episodes, which probably occurred during a single Neoproterozoic orogenic cycle. The first metamorphic episode ( M l ) was syntectonic with respect to the main deformational event; this produced flat-lying fabrics and shallow NW/SE-trending lineations and is attributed to continental collision between the Congo and Sao Fancisco cratons. Thermobarometric results obtained for well equilibrated M l mineral assemblages give temperature conditions of c.850°C and pressures between 6 and 6.7 kbar in the southern area (Tres Rios city) and of 807°C to 877°C in the north (Santo Antonio de Padua region). These P-T conditions are assumed to reflect the thermal peak of metamorphism. The younger metamorphic episode (M2) was syntectonic with respect to the deformation that occurred during the development of the PSSZ. Lineations developed at this stage have shallow NE/SW plunges. M2 assemblages were only developed in the PSSZ, and the deformation clearly promoted re-equilibration of the previously metamorphosed rocks. The temperature and pressure conditions determined by the application of thermobarometric methods to M2 equilibrium assemblages are 715°C to 747°C and 4.5 kbar, respectively, for the southern area, and 734°C to 743°C and 5.2 kbar, respectively, for the northern area. In both areas, M2 pressures and temperatures are significantly lower than those of M l . There is no significant difference between the M l P-T conditions deduced for these two areas; the M2 P-T conditions of the two areas are also similar. In both areas, granulites affected by the transcurrent event show reaction microstructures (e.g. reaction rims of garnet around pyroxene) that indicate retrograde P-T paths dominated by cooling, consistent with relaxation of elevated geotherms in crust that had returned to 'normal' thickness. In the Tres Rios region, the isobaric cooling path seems to have been preceded by an isothermal decompression stage, recorded by orthopyroxene corona around clinopyroxene core; this decompression may reflect erosion and/or tectonic unroofing of crust that had been over-thickened tectonically during collision. The results of this study emphasise the importance of deformation in promoting metamorphic re-equilibration, thus making it possible for points on the retrograde part of a P-T path to be constrained with confidence.

OROGENESIS IN THE OUTBACK

ALUMINOUS A-TYPE GRANITOID SCENARIO OF SOUTHWESTERN INDIA: CHARACTERIZATION AND REGIONAL IMPLICATIONS H. M . R a j e s h Department of Geosciences, Faculty of Science, Osaka City University, Sugimoto 3-3-138, SumiyoshiKu, Osaka 558-8585, Japan. E-mail: rajesh@sci.osaka-cu.ac.jp

A suite of alkali granite and syenite plutons preserving evidence for a prominent Pan-African felsic magmatic event intrudes the Proterozoic granulite facies terrains of southwestern part of the Indian Peninsula. These plutonic masses are E-W or WNW-ESE to NW-SE elongated elliptical bodies showing sharp contacts with the country rocks, viz., migmatized felsic gneisses (garnet-biotite gneiss, biotite gneiss, homblende-biotite gneiss and biotite-hornblende gneiss) and/or chamockites (orthopyroxene-bearing quartzo-feldspathic rock of either igneous or metamorphic origin). Significantly these intrusives show spatial association with regional lineaments, implying that the orientation of these intrusives is directly controlled by the local fault pattern, and indirectly controlled by the rate of magma injection and heat loss into the country rocks. The proximity of the plutons to the edges or junctions of these megafaults of continental scale suggests that the magma generation and emplacement could be related to the formation and/or reactivation of these faults. Importantly, the plutons themselves are unsheared and largely unaffected by later deformational events, indicating the post-tectonic (anorogenic) nature of the magmatism. The field relations and undeformed igneous textures of most of these intrusions suggest that the melts migrated during the deformation in the associated shear zone/fault lineament, but crystallized after the deformation had ceased, leading to the general lack of deformational features. Hence the ages of these granitoid intrusions should have correspond closely to the conditions of deformation of the shear zone/fault lineament. Petrographically these granitoids are characterized by the presence of iron-rich hydrous mafic minerals, and/or sodic pyroxene, primary magnetite with or without subordinate ilmenite, and a collection of mineral wealth including molybdenite, zircon, phlogopite, fluorite, chevkinite and magnetite. According to oxide mineralogy, these granitoids belong to the magnetite series. Estimation of pressure, temperature and oxygen fugacity conditions using available geothermobarometers and mineral equilibria indicates that the parental magma of these granitoids was waterundersaturated, formed under high temperatures (~1000°C), at medium- to high-pressures (-6-8 kbar), under oxidizing conditions {JO2 above the NNO buffer), and possibly intruded at lower crustal levels. Geochemically these granitoids: 3. are alkaline (Na+K -^5-15 wt%), 4. have high K2O with typical high-K calc-alkaline to shoshonitic trend (an ultrapotassic variety also occurs), 5. are mildly aluminous to aluminous (AI2O3 --12-20 wt%), 6. are mostly metaluminous, with slightly peralkaline and peraluminous varieties, 7. have high total Fe (as Fe203), low CaO and MgO, medium Ti02/Mg0 and low /wg-number typical of A-type granitoids, 8. have abundant high-charge cations such as Ga, Zr, Zn, Y and Nb, owing to the incompatible nature of these elements in A-type magmas, 9. have low Rb contents for some granitoids, in comparison with average A-type granitoids worldwide, and 10. have low LILE, moderate to high transition element contents, and enrichment of LREE. Their major and trace element characteristics are typical of A-type granitoids. Although some of the recent studies have questioned grouping peralkaline and metaluminous-peraluminous rocks together as A-type granitoids, it is considered useful in the case of granitoid plutons where these two facies co-exists. Considering the aluminous nature of these granitoid plutons, they are best denoted as aluminous A-type granitoids. The Sr-isotope data suggest that these A-type granitoids were derived from a lower crustal source. Petrogenetic characterization of these granitoids, including trace element modeling, indicates that no unique process can account for their compositional diversity. Processes involving simple fractional crystallization, or partial melting, to complex ones involving combined partial melting-fractional crystallization was proposed for these granitoids. Within the constraints imposed by the high temperature, relatively anhydrous, K-rich nature of the magma, trace element modeling, and comparison with recent experimental studies on various granitoid source compositions, the distinctive features of the southwestern Indian aluminous A-type granitoids are suggested to reflect a characteristic parent magma composition and/or source rock of charnockitic (C-type) nature. The C-type signature of massif chamockites adjacent to these granitoids is also established in this study. Thus it follows that alkaline, aluminous A-

ALICE SPRINGS, AUSTRALIA, 1999

type granitoids from southwestern India can be generated from a dehydrated, charnockitic, mafic- to intermediate lower crust. The mechanism involved can be either partial melting or fractional crystallization or a combination of both with each taking lead at a particular stage of magmatic evolution. On a regional scale the occurrence of some of these A-type granitoids and, significantly, alkaline pegmatite's adjacent to exposures of arrested stages of charnockite formation (incipient charnockite) warrants further comment. Fluid inclusion studies on various granulite minerals (quartz, fluorapatite and cordierite) and quartz from some of the A-type granitoids suggests that they were enriched in CO2, CO2-H2O, F and other volatiles. It is apparent that a key factor for the resilience of the alkaline pegmatites adjacent to the incipient charnockites is the channeling of C02-rich fluids, either at grain scale or at outcrop scale. The transport of CO2 by felsic melts through the middle crust is considered to be part of a crustal-scale fluid system that linked mantle heat and CO2 input with upward migration of crustally derived felsic melts and incipient charnockite formation.

O R O G E N E S I S IN T H E O U T B A C K

METEORIC FLUID INFILTRATION DURING PALAEOZOIC EXHUMATION OF THE PROTEROZOIC YALBADJANDI SHEAR ZONE, CENTRAL AUSTRALIA. C. M. R e a d a n d I. Cartwright Dept. Earth Sciences, Monash University

and VIEPS, Clayton Campus,

VICTORIA

3168.

Retrograde shear zones are a common feature in high-grade metamorphic terrains, and are important during exhumation. The assemblages in shear zones are commonly more hydrated than their host lithologies (especially in granulite terrains), this requires the introduction of fluids from external sources. The introduction of meteoric fluids into Alice Springs age (300-350 Ma) amphibolite facies grade shear zones in the northern Arunta Inlier suggests that there was significant crustal-scale fluid movement at that time. This has significant implications for the largescale plumbing systems operating during the exhumation of the Arunta Block. The Yalbadjandi shear zone runs N W - S E along the southern edge of the Anmatjira Ranges, in the northern Province of the Arunta Inlier. At Mt Weldon it constitutes a 2 0 0 m wide, steeply north dipping shear zone juxtaposing the Weldon Metamorphics (granulite facies meta-sediments) and the Mt Airy Orthogneiss (granulite facies metagranite). Rb-Sr data from this shear zone indicates that it was active during the Alice Springs Orogeny (Read et al., in prep). Shear zones of similar age are present throughout the Reynolds Range to the south of this area (Cartwright et aL, 1999, Cartwright and Buick, 1999). Thrust movement on the shear zone is indicated by microshears, C-S fabrics and deformation styles of the hanging wall Weldon Metamorphics and the footwall Mt Airy Orthogneiss. The Yalbadjandi Shear Zone forms part of a major "pop-up" structure that was responsible for the exhumation of the Northern Province of the Arunta Inlier during the Alice Springs Orogeny (Collins and Teyssier, 1989). Within the shear zones both the Mt Airy Orthogneiss and Weldon Metamorphics show major grainsize reduction and a change in the mineral assemblages. The sheared Mt Airy Orthogneiss, previously a coarse grained granitic gneiss, forms a mylonite with relic K-feldspar augens, recrystallised quartz, epidote, with biotite and minor muscovite forming the shear fabric. Variation in the abundances of epidote and biotite occur locally throughout the shear zone. The Weldon Metamorphics are layered pelitic and psammitic granulite, comprising cordierite, Kfeldspar, garnet, biotite, sillimanite and quartz. The sheared granulites are mica schists which comprise quartz, muscovite and biotite, with variable relic feldspar depending on the intensity of shearing. Evidence for substantial fluid influxes into shear zones in the Anmatjira and Reynolds Ranges including the Yalbadjandi Shear Zone is indicated by the formation of hydrous mineral assemblages and the existence of quartz veins. The initial values of the Mt Airy Orthogneiss range from 6-9%o. The Weldon Metamorphics range from 9-15%o. Oxygen isotope ratios within the shear zone are as low as -6%c indicating that fluids with very low infiltrated the Yalbadjandi Shear Zone during amphibolite facies shearing. The only known source for such low values is meteoric fluids. Across the shear zone the low values are clustered in several narrow bands, suggesting the meteoric fluids were highly channelled within the shear zone. This fluid focussing might have resulted from heterogenous strain across the shear zone with the more highly strained rocks having higher permeabilities. The introduction of fluids into these zones could have driven metamorphic reactions causing strain softening, thereby increasing the transient permeabilities. Fluid flow, shearing and metamorphic reactions were probably mutually reinforcing. Temperature estimates from oxygen isotope ratios of quartz and biotite from both sheared Mt Airy Orthogneiss and the Weldon Metamorphics rocks indicate shearing occurred between 4 5 0 - 5 5 0 ° C . This is consistent with the mineralogy developed in the shear zones. Several sheared orthogneisses with low values contain abundant epidote and biotite. The increase in epidote content of these sheared rocks suggests that a significant amount of F e O and CaO was introduced during fluid infiltration and precipitated as epidote during shearing. This observation is supported by whole rock compositions which show a marked increase in CaO and FeO in the sheared rocks. Major element changes such as this require large fluid volumes, indicating that these shear zones were major fluid conduits. Shearing within the Redbank High Strain Zone in the south of the Arunta Inlier also involved meteoric fluids. Specific dating hasn't yet been carried out on the zones of meteoric fluid infiltration, however, Alice Springs ages have been well documented from this region (Shaw and Black, 1991). Therefore, on a larger scale meteoric fluids

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

played an important role during the Alice Springs shearing throughout the Arunta Inlier. A significant portion of the fluid used to rehydrate the granulites of the northern province were provided from meteoric fluids being channelled down large thrust structures into the ductile crust. Therefore, the fluid flow systems operating within the Arunta Inlier during exhumation were connected to the Earth's surface. The deformation systems operating at this time provided a pathway for meteoric fluids to penetrate into the ductile crust. The commonality of meteoric fluid signatures throughout the Arunta Inlier suggests infiltration of surface derived fluids was widespread during exhumation. It is possible then to make a broad assumption that meteoric fluid played a major role in the rehydration and shearing of Proterozoic Arunta rocks during the Palaeozoic Alice Springs Orogeny. This indicates that the surface of the Earth may in fact be an important fluid source during the exhumation and rehydration of metamorphic terrains.

REFERENCES

Cartwright, I. Buick, I. S., Foster, D. A. and Lambert, D. D. 1999. Alice Springs age shear zones from the southeastern Reynolds Range, central Australia. Australian Journal of Earth Sciences 46. Cartwright, I. and Buick, I. S. 1999. The flow of surface-derived fluids through Alice Springs age middle-crustal ductile shear zones, Reynolds Range, central Australia. Journal of Metamorphic Geology 17. Collins, W. J. and Teyssier, C. 1989. Crustal scale ductile faults systems in the Arunta Inlier, central Australia. Tectonophysics 158, 49-66. Read, C. M, Cartwright, I. and Maas, R. (in prep) Meteoric fluid infiltration into shear zones during Palaeozoic exhumation of a Proterozoic terrain, examples from central Australia. Shaw, R. D. and Black, L. P. 1991. The History and tectonic Implications of the Redbank Thrust Zone, central Australia, based on structural, metamorphic and Rb-Sr isotopic evidence. Australian Journal of Earth Sciences 38, 307-332.

O R O G E N E S I S IN T H E O U T B A C K

THE ROLE OF FLUIDS IN A MAJOR CRUSTAL THRUST ZONE, REDBANK HIGH STRAIN ZONE, CENTRAL AUSTRALIA. C. M. R e a d a n d I. C a r t w r i g h t .

Department of Earth Sciences, Monash University and VIEPS, Clayton Campus, VICTORIA

3168

Crustal-scale fault zones world wide, such as the Moine thrust in Scotland, the Alpine thrust in New Zealand are associated with the juxtapose rocks of different crustal levels and may form terrain boundaries. During exhumation events the infiltration of fluids is a common phenomenon as shown by the hydrous assemblages found within the shear zones that cut high-grade anhydrous metamorphic rocks. The sources of these fluids is often not apparent from the surrounding rocks. Substantial devolatilisation of high-grade rocks is unlikely to accompany later low grade shearing and igneous activity does not always occur during shearing. The Redbank High Strain Zone (RHSZ) forms a major crustal boundary separating the Central and Southern Provinces of the Arunta Inlier in central Australia. The RHSZ is a 400km long north-dipping zone of faults and shears that juxtaposes amphibolite facies Southern Province gneisses with Central Province granulites. Deep seismic, teleseismic and gravity work indicates that the RHSZ offsets the moho by some 25km (Korsch et al, 1998), implying that it is a crustal-scale thrust structure. Geophysical and structural studies on the Redbank have been extensive owing to the size of this major lineament. Exhumation along the Redbank High Strain Zone is thought to have occurred in two major episodes. Two generations of shearing have been established on the basis of differing metamorphic facies assemblages within mylonite zones and the intrusion of the Stuart Dyke Swarm into previously deformed rocks that were later sheared by Ahce Springs ages structures (Shaw and Black, 1991; Zhao et al, 1992). According to Shaw and Black (1991), Rb-Sr ages indicate that shearing occurred initially around 1500-1400 Ma forming amphibolite facies (Type 1) mylonites. This was followed llOOm.y. later by the formation of low-grade (greenschist facies) mylonites (Type 2) during the Alice Springs Orogeny (at 300-350 Ma).

The sheared rocks of the RHSZ are dominated by protomylonites and have locally-developed mylonite to ultramylonite fabrics. RHSZ rock compositions vary, but are dominated by granitic and mafic amphibole-bearing assemblages. The unsheared rocks to the south of the RHSZ are coarse-grained granitic gneisses that have assemblages of orthoclase, plagioclase, biotite and quartz. These rocks are associated with pegmatites, which are boudinaged and stretched into the shear zones. To the north of the RHSZ the rocks are dominantly mafic granulites comprising garnet cordierite plagioclase assemblages, which contain sheared amphiboles, biotite and quartz. There are also patches of sediments within the RHSZ that have upper amphibolite grade quartzofeldspathic and quartzite assemblages. The rocks show varying degrees of rehydration from the precursor gneisses surrounding them. Sheared mafic granulites along the northern margin of the RHSZ, which are likely to be part of the Mt Hay Granulite Massif, show replacement of high-grade garnet by cordierite, implying decompression following the peak of metamorphism (which occurred at 1768Ma, Black et aU 1983). Coarse grained amphibole defining a shear fabric may have formed during Proterozoic shearing on the same P-T time path. These rocks were subsequently cut by small oblique shears. Other rocks within a few metres to tens of metres have experienced extensive rehydration and metasomatism, locally causing silicification and epidotisation. The quartz-epidote rocks are 5 to 15m wide and occur parallel to the main shear fabric. They preserve macroscopic textural features which indicate that they are metasomatic equivalents to the gneisses that host them. These altered sheared mylonites suggest substantial amounts of fluid infiltrated the RHSZ, channelled along narrow zones of greater permeability. The variation in the degree of strain, evident in thin section, along these zones is possibly a feature of the different rates of recovery within bi- and polymineralic lithologies, where quartz and epidote recovery is rapid. However the differing apparent amounts of fluid infiltration across the shear zone indicates fluid flow was highly focussed and not solely dependant on the degree of strain. Oxygen isotope values across the RHSZ in several across strike traverses along the RHSZ indicate a complex history of fluid-rock interaction. values range from 3.3 to \l%o with most values between 6-9%o. Lowering of the values locally within larger shears occurs in conjunction with major silicification and epidotisation. Owing to the large volumes of fluids that have infiltrated these areas, the isotope signature of the fluids is likely to be preserved over large distances, and therefore be recorded in the rock values. However there are zones of

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

silicification and epidotisation which have values in the 6-9%o range. The lower values are indicative of the infiltration of meteoric fluids into the shear zones. Within the other metasomatic zones it is possible that fluids that effected these rocks were also meteoric but their isotopic signatures were changed by fluid-rock interaction prior to the fluids reaching them. Alternatively other crustal sources may also have contributed to the fluid budget. There is a lack of evidence of magmatic activity during the Alice Springs Orogeny implying hydrothermal circulation of meteoric fluids did not occur at this time. The anhydrous nature of the basement granulites precludes them from being a source of fluids during shearing. Therefore it is likely that much of the fluids came directly from the surface or from within the Amadeus Basin which covered the Arunta Inlier prior to exhumation. The lack of Alice Spring ages from the host lithologies surrounding the shear zones implies that isotopic resetting is generally not occurring out of the shear zones, with the exception of some 300-350 Ma ages in gneisses with no textural evidence of later shearing from Collins and Shaw (1995, unpub. data). This implies that fluid transfer throughout the unsheared rocks of the southern province was limited during the Alice Springs Orogeny. The evidence of fluid infiltration within the shear zones of the RHSZ is indicative of channelling of meteoric fluids through narrow pathways during the active exhumation of the terrain. Re-equilibration of the meteoric fluids increases their S^^O values, which suggests the fluids have enjoyed major fluid-rock interaction and are possibly further along the flow path than meteoric fluids seen elsewhere in the Arunta Inlier. This isotopic and petrologic evidence indicates the Redbank High Strain Zone has acted as a major fluid conduit during exhumation of the Arunta Inlier. Shearing is complex and has occurred during different uplift events, however the oxygen isotope analyses suggest there was infiltration of meteoric fluids into the zone during at least one of the shearing events. Elsewhere in the Arunta Inlier meteoric fluids signatures have been found in Alice Springs age shear zones. Although dating of the low sheared rocks along the Redbank hasn't been undertaken it is possible these shears are also Alice Springs age as they are not cross cut by younger structures. As with many major crustal structures world wide the juxtaposing of different tectonic and metamorphic terrains is coupled with fluid infiltration, and we suggest the earths surface could represent a major reservoir for the fluids present during exhumation. REFERENCES

Black, L. P., Shaw, R. D. and Stewart, A. J. 1983. Rb-Sr geochronology of Proterozoic events in the Arunta Inlier, central Australia. BMR Journal ofAustralian Geology and Geophysics 8 129-137. Collins W.J. and Shaw R.D. 1995. Geochronological constraints on orogenic events in Arunta Inlier: a review. Precambrian Research 71, 315-346. Korsch, R. J., Goleby, B. R., Leven, J. H. and Drummond, B. R. 1998. Crustal Architecture of central Australia based on deep seismic reflection profiling. Tectonophysics 288, 57-69. Shaw, R. D. and Black, L. P. 1991. The history and tectonic implications of the Redbank Thrust Zone, central Australia, based on structural, metamorphic and Rb-Sr isotopic evidence. Australian Journal of Earth Sciences 38, 307-332. Zhao, J. X., McCulloch, M. T. and Bennett, V. C. 1992. Sm-Nd and U-Pb zircon isotopic constraints on the provenance of sediments from the Amadeus Basin, central Australia: Evidence fro REE fractionation. Geochimica et Cosmochimica Acta 56, 921-940.

OROGENESIS IN THE OUTBACK

ZIRCON AND MONAZITE RECORD OF PROTEROZOIC METAMORPHISM IN THE REYNOLDS RANGE, CENTRAL AUSTRALIA D. Rubatto\ I. S. Williams\ I. Buick^ 2. Research School of Earth Sciences, Australian National University, Canberra, ACT 0200, Australia 3. School of Earth Sciences and VIEPS, La Trobe University, Bundora, Vic. 3038, Australia

The Reynolds Range (northern Arunta block, central Australia) consists of a poly-metamorphic basement (Lander Rock Beds) on which an Early Proterozoic (-1800 Ma) sedimentary sequence (Reynolds Range Group) was deposited. Basement and cover were intruded by numerous granitic bodies and then underwent metamorphism during the Mesoproterozoic. The metasedimentary rocks of the Reynolds Range Group record a spectacular prograde metamorphic field gradient from greenschist (<400 chlorite-muscovite grade) to granulite facies (>750 cordierite-garnet-sillimanite grade; Dirks et al. 1991). U-Pb age determinations have been carried out by ion microprobe (SHRIMP) on zircon and monazite from metapelites and metapsammites with variable metamorphic grade along a profile cutting across the isograds. The first occurrence of metamorphic monazite is at upper amphibolite facies (biotite-sillimanite grade). BSE imaging of the metamorphic monazites shows zoning in trace elements, which does not correspond to zoning in age. In contrast, metamorphic zircon rims around detrital cores are found only in granulite-grade rocks. The metamorphic rims are characterised by absence of zoning (cathodoluminescence imaging) and low ThAJ ratios. Their width generally increases with increasing metamorphic grade and is likely to be related to the chemical composition and the degree of partial melting of the host rock. In fact, zircons from a granulite-grade metapelite have abundant metamorphic overgrowths when compared to zircons from a metapsammitic rock of the same metamorphic grade, in which melting reactions related to mica break-down are likely to be less. SHRIMP analyses of metamorphic zircon overgrowths and monazites from eight samples yielded ages in the range 1569-1586 Ma. Both monazite and zircon ages are interpreted as formation ages because Pb diffusion in zircon is insignificant at such temperatures (Lee et al., 1997) and the monazite Pb-U system has been proved to be resistant at temperatures in excess to 700 °C (e.g. Copeland, 1988). In the Reynolds Range Group, no difference in age has been detected between monazite and zircon of the same sample. However, a variation in age is present in rocks of different metamorphic grades with a trend toward older ages at lower grade Figure 1). Even though the difference in ages between the samples is in the order of 1%, which generally is the analytical uncertainty of SHRIMP ages, the possibility that this age trend is geologically significant must be considered. These age results suggest the formation of monazite and zircon during prograde metamorphism lasting over 10-15 Ma. The indistinguishable ages of zircon and monazite indicate that no lead diffusion occurred in monazite during cooling, which would have resulted in monazite consistently giving younger ages than zircon. The present ages are significantly younger than the 1583-1594 Ma ages obtained by previous SHRIMP geochronology on zircons and monazites (Williams et al. 1996; Vry et al. 1996) from granulitic metapelites of the Reynolds and Anmatjira Ranges. The reason for the difference in ages from the three studies remains unclear, but might be the result of either a high-grade metamorphic event lasting over 20 Ma (Williams et al., 1996), or two distinct metamorphic pulses. In the case of prolonged metamorphism, the rocks investigated in this study (low Fe/Mg ratio, absence of garnet) and in the previous studies (high Fe/Mg ratio, presence of garnet) may have crystallised zircon and monazite during different metamorphic reactions as a function of their bulk composition. This hypothesis and the possibility of an age trend with increasing metamorphic grade require further investigation and may lead to a completely new way of interpreting zircon and monazite U-Pb ages opening the possibility to establish the duration of high-grade metamorphic events by U-Pb geochronology.

ALICE SPRINGS, AUSTRALIA, 1999

1595 1590 1585 1580 Ma 1575 1570 1565 1560

g ran u lite

amphib.

iU ^

• zircon m monazite

distance

(km)

Figure 1. Zircon (circle) and monazite (square) ages of the Reynolds Range Group plotted against distance. The metamorphic grade decreases from left to right. Error bars are at 95% confidence level. REFERENCES

Copeland, P., Parrish, R. R. and Harrison, T. M., 1988. Identification of inherited radiogenic Pb in monazite and its implications for U-Pb systematics, Nature, 333, 760-763. Dirks, P.H.G.M., Hand, M. & Powell, R., 1991. The P-T-deformation path of a mid-Proterozoic, low pressure terrane: the Reynolds Range, central Australia. J. Metamorphic Geol, 9, 641-661. Lee, J. K. W., Williams, I. S. and Ellis, D. J., 1997. Pb, U and Th diffusion in natural zircon. Nature, 390, 159-162. Vry, J., Compston, W. & Cartwright, I., 1996. SHRIMP II dating of zircons and monazites: reassessing the timing of high-grade metamorphism and fluid flow in the Reynolds Range, northern Arunta Block, Australia. J. Metamorphic Geol, 14, 335-350. Williams, LS., Buick, I.S. & Cartwright, I., 1996. An extended episode of early Mesoproterozoic metamorphic fluid flow in the Reynolds Range, central Australia. J. Metamorphic Geol, 14, 29-47.

OROGENESIS IN THE OUTBACK

MIGRATION OF SUBDUCTION IN THE ALPS D. Rubatto Research School of Earth Sciences, Australian National University, Canberra, ACT 0200, Australia.

The Alpine chain resulted from the convergence and the collision between the Adriatic (African) plate and the European plate during Cretaceous and Tertiary. This orogeny produced a large amount of eclogite-facies rocks varying from ultramafic to granitic in chemical composition. These eclogites occur in different tectonic position within the Alpine nappe pile. SHRIMP U-Pb dating of zircons from several high-pressure rock types has been conducted to define the timing of high-pressure (HP) metamorphism in the different portion of the chain and to better understand Alpine subduction process. Samples from the Adriatic continental margin (Sesia-Lanzo Zone) and the European continental margin (Monte Rosa nappe) have been investigated, as well as samples from the ophiolites of the Tethys (Zermatt - Saas-Fee), the oceanic basin separating the two continental blocks before collision. Cathodoluminescence imaging of the zircons showed that several crystals have oscillatory-zoned cores surrounded by rims. The cores are inherited from the pre-Alpine magmatic or sedimentary protolith of the eclogite-facies rocks, whereas the rims yield Alpine ages. The rims are characterised by absence of zoning and cathodoluminescence emission generally higher than the cores, indicative of relatively lower U-contents. SHRIMP analysis revealed that the zircon rims have lower Th/U ratio (<0.1) and generally lower U-contents than the cores. These zircon rims are interpreted to have formed by recrystallization of a previous detrital or magmatic crystal, which is generally preserved in the core, during Alpine eclogite-facies metamorphism. An eclogite and a Mn-rich metasediment from the ophiolites contain zircons that do not preserve inherited cores and have inclusions of rutile. In this case, zircons are thought to have grown during HP metamorphism in the presence of fluids. U-Pb dating of metamorphic zircon and zircon rims indicated a variation in age of HP metamorphism (subduction) within the Alpine chain. 4. Metamorphic zircon rims from an eclogite and an eclogitic micaschist of the Sesia-Lanzo Zone (Adriatic continental margin unit) yield an age of 65.2±4.6 Ma and 65.2±3.1 Ma, respectively. This unit records HP metamorphism at 15-18 kbar and 550-600 (Pognante 1989). The age of HP metamorphism in the SesiaLanzo Zone is significantly younger than the age of eclogites found in the Adriatic continental units of the Eastern Alps where Sm-Nd age determinations support and age of 100-90 Ma (Thoni and Jagoutz, 1992). 5. Zircons from an eclogite and two Mn-rich metasediments of the Zermatt Saas-Fee ophiolites (Tethys) yield an age of 44.1 ±0.7 Ma. The presence of coesite in this unit constrained the peak pressure at 28-30 kbar at temperatures of 600 (Reinecke, 1991). 6. Metamorphic zircon rims from a phe-rich metaquartzite of the metasedimentary cover of the Monte Rosa nappe (European margin) yield an age of 34.6±1.4 Ma. The peak metamorphic conditions for this unit are 10-15 kbar and 550 (Chopin and Monie, 1984). This age is in agreement with the age of HP metamorphism in two other units of the European margin (Gebauer, 1996; Gebauer et al., 1997). The new geochronological data show that: 11. the HP metamorphism was not synchronous across the Alpine chain; 12. the age of subduction decreases from SE to NW, consistently with SE dipping subduction; 13. this youngening of the HP metamorphism from the Adriatic margin (65 Ma) through the ophiolites (44 Ma) to the European continental margin (35 Ma) indicates migration of subduction episodes during the Alpine orogeny (figure 1); 14. convergence in this sector of the Alps was characterised by poly-episodic subduction of continental and oceanic crust during which the more external units (European margin) were subducted while the internal units (Adriatic margin and Tethys ocean) were ah-eady on the exhumation path.

I I

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

Europe

Sesia-Lanzo Tethvs ocean •

35 Ma •

continental crust

44 Ma

Adria

Meliata-Hallstatt ocean

65 Ma

90-100 Ma oceanic crust not to scale

Figure 1. Schematic paleogeographic reconstruction of the Alps for the early Cretaceous with the location of the future subduction zones.

REFERENCES Chopin, C. and Monie, P., 1984, A unique magnesiochloritoid-bearing high-pressure assemblage from the Monte Rosa, Western Alps: petrological and 40 Ar-39Ar radiometric study. Contributions to Mineralogy and Petrology, 87, 388-398. Gebauer, D., 1996, A P-T-t-Path for an (Ultra?-) High-Pressure Ultramafic/Mafic Rock-Association and its Felsic Country-Rocks Based on SHRIMP-Dating of Magmatic and Metamorphic Zircon Domains. Example: Alpe Arami (Central Swiss Alps). In Earth Processes: Reading the Isotopic Code, American Geophysical Union, 309-328. Gebauer, D., Schertl, H.-P., Brix M. and Schreyer, W., 1997, 35 Ma old ultrahigh-pressure metamorphism and evidence for very rapid exhumation in the Dora Maira Massif, Western Alps. Lithos (HP-metamorphism in nature and experiment) 41, 5-24. Pognante, U., 1989, Tectonic implications of lawsonite formation in the Sesia zone (Western Alps). Tectonophysics, 162, 219-227. Reinecke, T., 1991, Very-high-pressure metamorphism and uplift of coesite-bearing metasediments from the Zermatt-Saas zone. Western Alps. European Journal of Mineralogyy 3, 7-17. Thoni, M. and Jagoutz, E., 1992, Some new aspects of dating eclogites in orogenic belts: Sm-Nd, Rb-Sr, and Pb-Pb isotopic results from the Austroalpine Saualpe and Koralpe type-locality (Carinthia/Styria, southeastern Austria). Geochimica and Cosmochimica Acta, 56, 347-368.

OROGENESIS IN THE OUTBACK

THE CRUSTAL STRUCTURE OF THE ARUNTA BLOCK AND NORTHERN AMADEUS BASIN: FROM AGSO DEEP REFLECTION SEISMIC DATA A. Rudge\ B. Goleby^ and T. Barr^ 1. Monash University, Clayton Vic 3168 2. AGSO, GPO Box 378, Canberra ACT 2601

The Arunta Block is a Mesoproterozoic basement block, which has experienced reactivation of crustal-scale faults during the latest Proterozoic and the Middle to late Paleozoic. The southern margin of the Arunta shares its boundary with the Neoproterozoic to late Devonian Amadeus Basin.

The two lines which show the Arunta Block, BMR85-1A and BMR85-1B, show that this portion of the crust is dominated by a series of planar reflections dipping moderately to the north, which have been interpreted as a series of parallel, planar thick skinned thrust faults (Goleby et al 1989). Many of these faults appear to cut deep into the crust, and one, the Redbank Thrust Zone appears to extend from the surface to the base of the crust, cross-cutting and displacing the Moho. At the surface, a major change in metamorphic grade occurs across the thrust, with greenschist and amphibolite facies rocks exposed on the southern side and granulite facies rocks exposed on the northern side (Korsch et al 1998). This indicates that there was a major amount of displacement associated with this fault, and no other faults in the region appear to have the same amount as the Redbank Thrust Zone. Mylonites that have developed in the suture have been dated at about 1500-1400 Ma, with reactivation and further mylonite development occurring at about 400-350 Ma (Shaw and Black 1991). It is these two deformation events that led to the big crustal scale faults seen on the seismic sections. The two lines BMR85-1A and BMR85-1B run from the Ngalia Basin in the north, through the Arunta Block, to the northern province of the Amadeus Basin. These surveys image many crustal scale features, including the Redbank, Ormiston and Gardiner Thrust faults. Though the Gardiner thrust fault is considered to be due to 'thin skinned' tectonics, because it can only be traced through the Amadeus Basin succession. The Amadeus Basin succession is over 10 km thick, mainly being of a sub-horizontal nature. Very few folds and thrusts are present within this section, apart from the strongly upturned northern margin, which forms the MacDonnell Homocline. Reprocessing of the seismic data using detailed velocity analysis, static adjustments and migration has improved the quality of the section resulting in the clearer imaging of faults and the abilty to trace sedimentary layers continuosly. The crust of the southern Arunta Block appears to extend beneath the Amadeus Basin, however there is no evidence for the crustal scale faults that occur further north in the Arunta Block. The surface geology shows steeply, southerly dipping units, indicating strongly upturned beds, and one theory for its formation is rotation and limited backthrusting of the succession in front of a south directed thrust wedge (Korsch et al 1998). A unit called the Brewer Conglomerate presents a problem to interpretation, as a 10 km section of line BMR85-1B is poorly imaged. This is due to the poor coupling of the source and the rock, and has resulted in very poor energy penetration. The reflection seismic data indicates considerable amounts of crustal deformation within the region, particularly across the Redbank thrust zone, a structure which cuts the entire crust and offsets the Moho. How did this occur? Was it due to an intr-continental collision, or collision on the present continental margin? Reprocessing the AGSO seismic lines, combined with interpretation of gravity and magnetic surveys through this region, will be used to better constrain the structure and tectonic evolution of the region. REFERENCES

Korsch, R.J., Goleby, B.R., Leven, J.H. & Drummond, B.J., 1998, Crustal Architecture of central Australia based on deep seismic profiling, Tectonophysics, 288, 57-69 Wright, C., Goleby, B.R., Shaw, R.D., Collins C.D.N., Kennett B.L.N. & Lambeck, K., 1993, BMR Journal of Australian Geology and Geophysics, 13, 359-368 Shaw, R.D. & Black. L.P., 1991, Australian Journal of Earth Sciences, 38, 307-332 Goleby, B.R., Shaw, R.D., Wright, C., Kennett, B.L.N., Lambeck, K., 1989, Geophysical evidence for Hhick skinned' crustal deformation in central Australia,_Nature 337, 325-330

I I

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

THE ROLE OF DEEP BASEMENT IN CONTINENTAL TECTONICS p. D. Ryan, Geology Department, National University of Ireland, Galway, Ireland. ryan@alisanos. nuigalway. ie

Cyclicity in continental tectonics sometimes requires a memory of past processes to exist well beyond the thermal relaxation time of the lithosphere. Ihis contribution will argue that, if this 'memory' is stored in the continental lithosphere, it must be in the lower crust or upper mantle. A possible mechanism is the formation of granulite and eclogite facies rocks during crustal shortening involving a significant increase in rock density. These dense rocks can be in isostatic equilibrium with either the lower crust or the upper mantle and may not be exhumed during orogenic collapse although they are of crustal composition. It is shown that a hthospheric column that contains such high grade crustal rocks either above or particularly below the 'seismic' Moho will remain weaker than lithosphere that does not. Also such a lithospheric column will be sensitive to fluid influx related to subsequent, possibly much later, tectonic processes. In addition it will be shown that the development and retrogression of such dense crustal lithologies can significantly affect topography and heat flow during continent-continent or arccontinent collisions and during continental extension.

O R O G E N E S I S IN THE OUTBACK

ORDOVICIAN ARC, CONTINENT COLLISION IN THE CALEDONIDES OF WESTERN IRELAND. p. D. Ryan^ & J. F. Dewey^ 1. Geology Department, National University of Ireland, Galway, Galway, Ireland. 2. Department of Earth Sciences, Parks Road, Oxford, 0 X 1 3PR, U.K.

The Caledonian orogeny of NW Europe records a short lived (< 5 Ma) orogenic event, the Grampian Orogeny. This contribution reviews evidence from the west of Ireland that this was caused by an arc-continent collision followed by a subduction flip between 472 and 468 Ma. Grampian nappe formation and Harrovian metamorphism occurred a rate that is almost an order of magnitude faster than that associated with continent-continent collision. A mechanism allowing crustal thickening and heating to take place at these accelerated rates is proposed involving rapid emplacement of a 200km wide fore-arc and accretionary complex onto the Laurentian foreland followed by magmatic advection associated with subduction delamination. Also, possible explanations for the remarkable preservation of pre- and post-flip arc complexes in the upper plate are discussed and comparisons are made with Cenozoic arc-continent collision zones.

W •

ALICE SPRINGS, AUSTRALIA, 1999

INTRAPLATE OROGENY AND THE GEOCHEMICAL STRUCTURE CONTINENTAL CRUST: AN OUTBACK AUSTRALIAN PERSPECTIVE.

OF THE

M. Sandiford, S. M c L a r e n , M. Hand Department of Geology and Geophysics, University of Adelaide, Adelaide, 5005, Australia msandifo@geology. adelaide. edu. au

It is increasingly recognised that intraplate deformation has resulted in a significant structuring of the continental interiors. This is particularly evident in outback (central) Australia where intraplate deformation during the Petermann and Alice Springs Orogenies has resulted in the development of some of the largest gravity anomalies known from the continental interiors. In central Australia, intraplate deformation has resulted in a profound geochemical reorganisation of the crust that must have impacted on its long-term thermal evolution. This talk focuses on the coupling between geochemical and thermal structuring of the crust during (and following) intraplate deformation, which is of fundamental significance to the long-term mechanical behaviour of continental interiors and, ultimately, to the nature of preserved heat production distributions in the crust. Examples pertinent to the Alice Springs Orogeny will be used where appropriate. Deformation, erosion and sedimentation change both the length-scales and absolute amount of heat production in the crust, and thus impact on its long-term thermal evolution. For example, the burial of a radiogenic basement beneath a thick sedimentary succession can lead to significant deep crustal heating. Significant spatial variations in the thickness of a sedimentary succession can therefore lead to variations in lithospheric strength which are potentially capable of localising intraplate deformation. Indeed, the preserved sedimentary record in Central Australia suggests the spatial pattern of intraplate deformation closely mimicked the distribution of depocentres in the Centralian Superbasin (Sandiford & Hand, 1998; Hand & Sandiford, 1999). Likewise, erosion of radiogenic basement will lead to significant deep crustal cooling and associated strengthening, which should inhibit the future possibility of localising deformation. The evolution of the heat production field within continental interiors during intraplate deformation is illustrated with reference to the characteristic length scale for heat production distribution, h,, and the vertically integrated crustal compliment of heat production, q^. This parameterisation leads to an image of the continental interiors as entities that evolve (in a thermo-mechanical sense) through successive increments of deformation, erosion and sedimentation. The end-point in this evolution is a state (low hr and/or q^) sufficiently "strong" to withstand, without appreciable deformation, the natural fluctuations in stress levels experienced by the continental interiors (ie, thermo-mechanically stabilised cratons). From the point of view of thermo-mechanical coupling, the rates of deformation are crucial, with characteristic deformation rates in intraplate orogens possibly very different from plate margin orogens. For example, the Alice Springs Orogeny appears to have been a long-lived event 90 Ma) involving less than - 1 0 0 km of distributed shortening. A significant component (-25%) of the shortening was accommodated on the Redbank Shear Zone; a thrust that dips at -40° and can be traced on seismic reflection profiles to depths of about 45 km where it clearly displaces the Moho by about 20 km. A minimum bound on the displacement rate on the Redbank Shear Zone is therefore - 0.3 mm/yr, and allows that deformation rates may have been as much as two orders of magnitude slower than typical of plate margin orogenies. Such differences in crustal scale deformation rates have profound implications for the thermal and mechanical evolution of the deforming system. In particular, Alice Springs-style deformation at rates of less than 1 mm/yr will result in crustal cooling during ongoing deformation. This syn-deformational cooling in response to removal of heat production from the hanging wall block, is estimated to amount to 20-30°C at Moho depths, and will be followed by further cooling of -40-50°C once deformation stops. Associated increases in crustal strength may provide a mechanism for terminating intraplate deformation and "locking-in" gravity anomalies formed during the constructional phase of orogeny, in ways that are fundamentally different from plate margin orogenies. References Sandiford, M. & Hand, M, 1998, Controls on the locus of intraplate deformation in central Australia, Earth and Planetary Science Letters, 162, 97-110. Hand, M., <& Sandiford, M., 1999, Intraplate deformation in central Australia, the link between subsidence and fault reactivation, Tectonophysics, 395, 121-140.

OROGENESIS IN THE OUTBACK

Illustration of the impact of h^ and q^ on the thermal and mechanical state of the lithosphere. The grey boxes show the estimated effect of a factor of 2 change in q^ and hj..

I q

mWrn"'

ALICESPRINGS, AUSTRALIA, 1 9 9 9

DOES CARBON ISOTOPE EQUILIBRIUM EXISTS BETWEEN GRAPHITE AND FLUID INCLUSIONS IN GRANULITES? A CASE STUDY FROM THE NILGIRI GRANULITES, SOUTHERN INDIA. M. Satish-Kumar and H. Wada

Department of Biology and Geosciences,

Shizuoka University, Shizuoka 422-8529,

Japan

The most important reservoirs of carbon in Earth's crust are carbonate minerals, graphite and CO2. Apart from these it also occurs in minor amounts in other carbon bearing mineral phases such as cordierite or scapolite, and as other volatile species like methane. The origin of carbon is deep crust can be either from the recycling of crustal carbon or from the upwelling of mantle volatiles carried through magmas. Granulites act often as a major temporary reservoir of carbon, by the presence of graphite and carbonic fluid inclusions. Carbon isotopes are important tools in deducing the origin and evolution of crustal carbon. Here, we try to constrain the isotopic relation between graphite and carbonic fluid inclusions in a well-documented fluid inclusion enriched orthopyroxene bearing enderbitic chamockite from the Nilgiri Hills, southern India. The southern Indian granulite terrain is an early to late Proterozoic mobile belt that encompasses the Dharwar craton. This mobile belt comprises of amphibolite to granulite facies rocks of sedimentary and igneous origin, and has been conventionally divided into number of blocks by a series of E-W to NE-SW trending shear zones. The tectonic, metamorphic and lithologic variations between the blocks indicate that accretion might have happened during early to late Proterozoic age accompanied by the tectonothermal events. Nilgiri massif exposes medium- to high-pressure granulite facies rocks which are bounded by the Moyar shear zone in the north and Bhavani shear zone in the south. The prominent rock types are intermediate to acid granulites (called enderbites) and chamockites (Raith et al., 1990). During the time span of 2.53 and 2.48 b.y. granulite facies metamorphism occurred in the Nilgiri block (Peucat et al., 1989). The peak metamorphic conditions inferred are around 750±40°C and 7 to 9.5 kb. Several earlier studies have documented the abundance of carbonic fluids in Doddabetta enderbites (e.g. Touret and Hansteen, 1988; Srikantappa et al., 1992). Enderbite samples used in the present study comprises of a mineral assemblage of plagioclase + quartz + orthopyroxene + garnet + biotite, with minor amounts of pyrrhotite, chalcopyrite, ilmenite and graphite. Apart from the typical massive type enderbite, samples that contain centimeter scale pods and veins of quartz segregation's (more than 90% quartz with minor garnet and graphite) was also used for the present study. Touret and Hansteen (1988) reported very high abundance of high-density primary carbonic inclusions amounting to as much as 20 vol. percent in garnet cores. Fluid inclusion and thin section petrography were carried out which revealed the abundance of carbonic inclusions in quartz and garnet. Graphite occurs as interstitial as well as included in other minerals. Garnet and quartz were separated under binocular microscope to a purity of more than 99 percent. Still very thin needle shaped, dark mineral inclusions were present and are supposed to be rutile, ilmenite or graphite. Micro-thermometric studies were carried out on fluid inclusions in garnet and quartz that revealed the compositional and density characteristics. Primary high-density (about Ig/cm^) CO2 inclusions occur in garnet and quartz. Similar to the observations made by Touret and Hansteen (1988) the inclusion abundance in garnet is far higher than in quartz. Mineral separates were used for the extraction of CO2 using stepped heating with an interval of 100°C under high vacuum. Extracted gases were analyzed for carbon isotopes. Typical profiles of stepped heating results of garnet and quartz are presented in Fig. 1. Graphite was separated from the samples by digesting the silicate phases in HF. Separated graphite was then combusted with V2O5 to produce CO2 gas carbon isotope ratios were measured. The release of fluid inclusion CO2 from mineral separates suggest that the most of the inclusions decrepitiate at around 700°C. This is consistent with the peak metamorphic temperature estimates of 750-i-/-40°C. Optical observation of decrepitation using high temperature heating stage also shows similar results. The carbon isotope of fluid inclusions in enderbitic quartz is around -4%c and in garnet is around -6%o, indicating a mantle signature corroborating the noble gas studies of Dunai and Touret (1993). The carbon isotope ratio of graphite in the same sample is around -12%o. Fluid inclusions in garnet are considered to be the most primary ones and considering the synmetamorphic nature of such inclusions isotopic equilibrium might have existed between graphite and carbonic fluids. A fractionation of about 6permil at 750°C, consistent with the reported theoretical and experimental results of fractionation of carbon isotopes between CO2 and graphite (Chacko et al., 1991), is suggesting to an isotopic equilibrium during peak metamorphism.

100 O r o g e n e s i s IN THE O u t b a c k

Nigiri enderbite garnet

Nilgiri enderbite quartz 250

-- -4 ^ ^

4- -8 400

500

600

700

2 150 0)

800 1000

Temperature(°C)

Fig, I. Typical stepped heating profiles of CO 2 yield and corresponding from the Doddabetta enderbitic granulite.

400 500 600 700 800 1000

Temperature{®C)

S'^C of fluid inclusions in quartz and garnet

We also measured the carbon isotope ratios of fluid inclusions in matrix quartz in enderbite and quartz segregation in centimeter scale pods and veins. The of fluid inclusions in matrix quartz is around -A%o and those of quartz segregation is around -3%o. The of graphite in quartz segregation is around -10%o. Peak release of CO2 in quartz segregation is less (about one tenth) than the matrix quartz, coinciding with the less abundance of fluid inclusions. Textural and micro-structural evidence points to a recrystallization of matrix quartz during cooling related with the reactivation of the adjoining shear zones of Nilgiri massif. Fluid inclusions were also reequilibrated during this event as evidenced by the lower density than the inclusions in garnet core. We believe that isotopic resetting of fluid inclusions occurred during this recrystallization whereas the inclusions in garnet remained inert. Alternatively, the fluids in garnet cores were equilibrated with graphite at a higher temperature than the matrix and hence lower fractionation value. In quartz segregates carbon isotope fractionation between carbonic fluids and graphite is around l%c. The isotopic heterogeneity of graphite in enderbite and quartz segregation indicates a possible later stage graphite (re-?) crystallization in quartz pods and corresponding isotopic equilibriation. Altogether it can be assumed that carbon isotopic equilibrium exists between graphite and synmetamorphic carbonic fluids in granulites. Our results provide new evidence for C02-graphite fractionation in natural samples and have wider implications on carbon isotope systematics in granulites.

REFERENCES Chacko, T., Mayeda, T.K., Clayton, R.N. and Goldsmith, J.R., 1991. Oxygen and carbon isotope fractionations between CO2 and calcite. Geochimica et Cosmochimica Acta, 55, 2867-2282. Dunai, T.J. and Touret, J.L.R., 1993. A noble gas study of a granulite sample from the Nilgiri Hills, southern India: implications for granulite formation. Earth and Planetary Science Letters, 119,271-281. Peucat, J.J., Vidal, P., Bernard-Griffiths, J., and Condie K.C. 1989. Sr, Nd, and Pb isotopic systematics in the Archaean low-to high-grade transition zone of southern India: syn-accretion vs. post-accretion granulites. Journal of Geology, 97,537-550. Raith, M., Srikantappa, C., Ashamanjeri, K.G. a n d Spiering, B. 1990. The Granulite terrain of the Nilgiri Hills (southern India): characterization of high-grade metamorphism. In: Vielzeuf, D. and Vidal, Ph. (eds.) Granulites and crustal evolution, NATO ASI Ser. C311. Dordrecht: Kluwer, 339-365. Srikantappa, C., Raith, M., and Touret, J.L.R. 1992. Synmetamorphic high-density carbonic fluids in the lower crust: Evidence from the Nilgiri granulites, southern India. Journal of Petrology, 33, 733-760. Touret, J.L.R. and Hansteen, T.H. 1988. Geothermometry and fluid inclusions in a rock from the Doddabetta chamockite complex, southwestern India. Rendoconti della Societa Italiana di Mineralogia Petrologia, 43, 6582.

ALICE SPRINGS, AUSTRALIA, 1999

101

THERMAL STRUCTURE OF SOUTHERN INDIAN GRANULITE TERRAIN BASED ON CARBON ISOTOPE THERMOMETRY M. S a t i s h - K u m a r \ H. W a d a \ M. Santosh^ and M. Yoshida^ ^Department of Biology and Geosciences, Shizuoka University, Shizuoka 4 2 2 - 8 5 2 9 , Japan ^Centre for Earth Science Studies, Akkulam, Trivandrum 6 9 5 0 3 1 , India ^Department of Geosciences, Osaka City University, Osaka 5 5 8 - 8 5 8 5 , Japan

Carbon isotope thermometry, in the past two decades, has been successfully used as a tool in deducing the peak metamorphic temperature condition in amphibolite to granulite facies marbles, where most of the existing geothermometers could not be applied. Although there exists discrepancies between theoretical, experimental and empirical calibrations a general consensus has been achieved for granulite facies rocks with the help of extensive data base from Adirondack highlands in north America (Kitchen and Valley, 1995 and references therein). Carbon isotope thermometer is based on the exchange of carbon isotopes between calcite and graphite. During prograde metamorphism re-crystallization of calcite and graphite occurs and studies have proved that above 400°C equilibrium equilibrium isotope fractionation persists (Wada and Suzuki, 1983). Marbles are best suited for carbon isotope thermometers mainly due to the following reasons, (1) refractory nature of graphite, (2) abundant reservoir of calcite that is having a rapid isotope diffusion while graphite is slow in diffusion, (3) absence of other carbon bearing phase. The southern Indian granulite terrain is an Archaean to Proterozoic terrain that encompasses the Dharwar craton and forms a mobile belt. This mobile belt consists of amphibolite to granulite facies rocks of sedimentary and igneous origin. The granulite terrain has been conventionally divided into a number of blocks by a series of E-W to NE-SW trending Shear Zones. The tectonic, metamorphic and lithologic variations between the terrains indicate that accretion might have happened during early to late Proterozoic age accompanied by the tectonothermal events. Off the shear zones the Palghat-Cuvery Shear Zone (PCSZ) is recently identified as a tectonic divide separating the Archaean terrain in the north and the Proterozoic terrain in the south. In the present study we will examine the calcite-graphite thermometry in the granulite grade marbles from the terrains south of the PCSZ. In the Madurai Block the highlands are occupied by mainly massive charnockites while the lowlands are intercalated with metasedimentary and meta-igneous rocks. Kodaikkanal massif and Cardomom hill massif forms the main massive bodies that are mid- to late-Proterozoic in age, and are mainly made up of enderbitic granulites affected by multiple deformational and metamorphic events. Eastern margins of these massifs are in contact with a thick sequence of metasedimentary sequence of gametiferous gneisses, sillimanite bearing pelitic gneisses, graphitic schists, calc-silicate rocks, marbles and quartzites. Nearer to the Kodaikkanal massif recently there has been several detailed studies regarding the peculiar occurrence of sapphirine-bearing granulites (Raith et al., 1997 and references cited therein). The Trivandrum Block comprises dominantly of the meta-sedimentary rocks that are metamorphosed to an upper amphibolite to granulite facies conditions. Recent studies have proved a high temperature metamorphism for the granulites reaching more than 850°C (Chacko et al., 1996). Marble bands occur in the Achankovil Shear Zone area, in the boundary with the Madurai Block. Medium to coarse grained calcite rich marbles were used to determine the fractionations between calcite and graphite from different localities in the two terrains. Fractionation range from 4%o to 2%c (Fig. 1) and the corresponding temperature estimates based on the Kitchen and Valley (1995) calibration yields a range of about 700 to 1000°C. The corresponding regional temperature distribudon suggests that the metamorphic temperatures increase towards the highland massifs from the plain lands. This is consistent with the model of igneous charnockites acting as a dominant heat source of metamorphism. This study also serves as a test for the calcite-graphite thermometry during high temperature metamorphism. Till date several studies have pointed out the reliability of the thermometry in the granulite grade. Almost of the extensive database of Kitchen and Valley (1995) in Adirondack mountains fall above 3 %c. We here for the first time report fractionations less than 2.5%c indicating the preservation of ultra high temperature signature in the calcite-graphite system. Significant is that our data is perfectly coinciding with the petrologically observed temperature estimates of 950-1000°C by Raith et al (1997) for the sapphirine bearing granulites in the Madurai Block. This similarity in temperature estimate suggest that calcite-graphite system can be utilized for identifying the metamorphic temperature conditions in terrains undergone by ultra-high temperature metamorphism.

1 0 2 OROGENESIS IN THE OUTBACK

In order to identify the effect of retrograde isotope exchange, micro-sampling in calcite and graphite were carried out. Three graphite crystals were delaminated perpendicular to the c-axis and analyzed. The results show that graphite has in average l%c variation within a single crystal, the heavier values in the core with respect to the rim. The J^C values of Calcite do not show any isotopic heterogeneity. Fractionation between the core of the graphite crystal and calcite gave the lowest value of 2%c. This corresponds to a temperature condition of more than 1000°C. Zonation preserved in graphite can be either due to polymetamorphic effects or due to the retrograde isotope exchange. A two-stage thermotectonic evolution has been advocated to UHT granulites of Madurai Block (Raith et al., 1997). We consider that the zonation profile preserved in graphite is an artifact of the slow cooling from temperatures of around 1000°C. In summary, the present study gives key constraints regarding the ultra high temperature conditions in marbles from the southern Indian granulite terrain and also the carbon isotope exchange between calcite and graphite during cooling.

Sampling localities •

Valliyoor

X Pandalgudi O Ambasamudram + Parappadi •

Kannlsserl

^ Thalayutlu -1

S^^Ccaicilc F i g . 1 C a r b o n i s o t o p e f r a c t i o n a t i o n s in c a l c i t c - g r a p h i t e p a i r s f r o m s o u t h I n d i a n g r a n u l i t e t e r r a i n . Isothermal lines are after Kitchen and Valley (1995)

REFERENCES

Chacko, T., Lamb, M. and Farquhar, J. 1996. Ultra high temperature metamorphism in the Kerala Khondalite Belt. In "The Archaean and Proterozoic terrains in Southern India within East Gondwana" M. Santosh and M. Yoshida eds. Gondwana Research Group Memoir-3, 157-165. Kitchen, N., and Valley, J.W., 1995. Carbon isotope thermometry in marbles of the Adirondack Mountains, New York. Journal of Metamorphic Geology, 13, 577-594. Raith, M., Karmakar, S. and Brown, M. 1997. Ultra-high-temperature metamorphism and multistage decompressional evolution of sapphirine granulites from the Palni Hill ranges, southern India. Journal of Metamorphic Geology, 15, 379-399. Wada, H. and Suzuki, K. 1983. Carbon isotope thermometry calibrated by dolomite-calcite solvus temperatures. Geochimica et Cosmochimica Acta 47, 697-706.

ALICE SPRINGS, AUSTRALIA, 1999

103

HIGH GRADE MYLONITIC REWORKING OF THE KANANDRA GRANULITE, CENTRAL AUSTRALIA: IMPLICATIONS FOR THE PALAEOZOIC EVOLUTION OF THE EASTERN ARUNTAINLIER 1. Scrimgeour^'^ and J. G. Raith'' 11nstitute for Geological Sciences, University of Leoben, Peter-Tunner Strafle 5, 8700 Leoben, E-mail: scrimgeo@unileoben.ac.at; raith@unileoben.ac.at 2 Northern Territory Geological Survey, P.O. Box 2655, Alice Springs, N.T., 0871, Australia.

Austria.

Our understanding of the geological evolution of the Arunta Inlier, central Australia, has been revolutionised over the past ten years by numerous metamorphic and geochronological studies which have established that it has a complex, polymetamorphic history. Most recently, the identification of granulite facies metamorphism of Ordovician age in the Harts Range region (Mawby et al, 1998; Miller et al, 1998) has led to a radical reinterpretation of the evolution of the eastern Arunta Inlier. Metamorphism of the Harts Range Group (HRG), previously believed to be Palaeoproterozoic, is now considered to be entirely Palaeozoic in age. However, the regional extent of this Palaeozoic high grade event remains unclear. In the Huckitta region (Fig. 1), the HRG is bounded to the north by high grade mylonite zones, the Entire Point Fault (EPF) and Delny-Mt Sainthill Fault Zone (DMFZ), which variably rework the Palaeoproterozoic Kanandra Granulite. An understanding of the structural and and metamorphic evolution of these mylonites is important in determining the degree and extent of high grade Palaeozoic reworking north of the HRG. E V O L U T I O N OF T H E K A N A N D R A G R A N U L I T E

In the Huckitta region, the Kanandra Granulite is dominated by mafic granulites and semi-pelitic migmatites. Within pelitic and semi-pelitic units of the Kanandra Granulite, peak metamorphism (M^ is defined by a migmatitic Si layering in which the leucosomes contain coarse garnet porphyroblasts. Mineral assemblages associated with Si in metapelites vary according to differing bulk compositions, but locally comprise garnet-sillimanite-biotite or gamet-cordierite-sillimanite bearing assemblages. Localised silica undersaturated bulk compositions contain spinelgarnet (± corundum) assemblages. P-T estimates suggest that peak Mi metamorphism reached 5-6 kbar and Si in the Kanandra Granulite is locally overprinted by a non-coaxial strain fabric (S2) with a strong SSE-SSW plunging quartz stretching lineation (L2; Fig. 1). S2 is defined by biotite and sillimanite in metapelitic rocks, and by two-pyroxene -1- hornblende bearing assemblages in mafic lithologies. Rare leucosomes occur axial planar to F2 folds and locally truncate S2. Mi garnet and sillimanite contain fractures perpendicular to L2, which are infilled by biotite ± sillimanite. In quartz-poor domains, fractures within sillimanite are infilled by biotite-spinel-ilmenite ± corundum. Although Mi garnet is typically highly corroded by M2 biotite and sillimanite, euhedral M2 garnets locally overgrow Mi garnet, and are interpreted to reflect (prograde?) stabilisation of garnet during Mj. M2 reaction textures and preliminary P-T estimates (-5-6 kbar, 700-750°C) are generally consistent with isobaric cooling following Ml. However, due to a lack of geochronological constraints, we cannot exclude the possibility that M2 represents a second independent metamorphic event. Intense upper amphibolite facies mylonites (S3) with a shallowly east-west trending sillimanite lineation (L3) separate the Kanandra Granulite from the Harts Range Group to the south. These mylonites form the EPF, and resulted in substantial reworking of the southern part of the Kanandra Granulite. Kinematic indicators consistently suggest a sinistral sense of strike-slip movement. Within metapelites, the mineral assemblage associated with S3 is gamet-biotite-sillimanite-K-feldspar-quartz-ilmenite ± muscovite. Garnets are often porphyroclastic, but not corroded, and small euhedral garnets locally occur within the mylonitic fabric. Thermobarometery on S3 pelitic mylonites gives 6.8-7.5 kbar and 680-730°C. Two texturally distinct garnet-hornblende-plagioclase-quartz assemblages can be distinguished in mafic lithologies within the EPF. A granoblastic assemblage, formed at -5-6 kbar and 650-700®C, is overprinted by a finer grained mylonitic M3 assemblage, which recrystallised at higher pressures, ie. 7-8 kbar and 700°C. This suggest that either (1) the Kanandra Granulite within the Entire Point followed an up-pressure evolution during mylonisation, or (2) the coarse-grained assemblage is related to D2, and hence is temporally unrelated to S3 mylonisation. Following D3, the Kanandra Granulite was reworked by mylonites (DJ, which form a 2-3 km wide zone varying in grade from mid-amphibolite facies (biotite-muscovite ± sillimanite) in the south to greenschist facies (chlorite-

1 0 4

O r o g e n e s i s

IN T H E

O u t b a c k

muscovite) in the north. D4 mylonites have a southwest plunging mineral lineation and consistent south-up (reverse) kinematic indicators. These mylonites form the DMFZ, which is interpreted to be a major back-thrust that accommodated significant exhumation during the Devonian-Carboniferous Alice Springs Orogeny.

CORRELATION WITH THE HARTS RANGE GROUP AND IMPLICATIONS

Immediately south of the EPF in the Huckitta region, the Harts Range Group underwent peak metamorphism and migmatisation at -8-9 kbar and 750°C. Correlation with other parts of the Harts Range Group suggests that this occurred at 480-460 Ma (Miller et al., 1998; Mawby et al., 1998). The northern part of the HRG was then reworked by EPF mylonites at ~7kbars and 700°C, which are equivalent to D3 in the Kanandra Granulite. The down-pressure evolution of the HRG prior to EPF deformation is in contrast to that of the Kanandra Granulite, where the EPF reworked lower pressure assemblages. If peak metamorphism in the HRG is exclusively of Ordovician age, then upper amphibolite facies D3 reworking of the Kanandra Granulite is also constrained to be Palaeozoic in age. Such high-T metamorphism is not known from the Alice Springs Orogeny, and is more likely to have occurred during the late stages of the Ordovician high-grade event. The timing of D2 high strain in the Kanandra Granulite is less well constrained, and could be related to either Proterozoic or Ordovician processes. During the Alice Springs Orogeny, the terrain was exhumed along the DMFZ. The presence of sillimanite within southern DMFZ mylonites suggests that the thermal regime was still elevated at this time. The DMFZ forms the northern margin of Palaeozoic highgrade reworking, and juxtaposed the Kanandra Granulite against Proterozoic granites to the north, which are overlain by unmetamorphosed Neoproterozoic sediments. The present study indicates that mylonitic reworking of the Kanandra Granulite occurred in three structurally and metamorphically distinct phases ( D 2 . 4 ) , and is the result of a complex polymetamorphic evolution. We interpret most of this high-grade mylonitic reworking to be of Palaeozoic age. This study also confirms that the EPF and DMFZ represent structures of fundamental importance in the Palaeozoic evolution of the eastern Arunta Inlier, separating regions to the south which underwent granulite facies metamorphism and deformation in the Palaeozoic, from regions to the north that have no Palaeozoic metamorphic overprint. REFERENCES

Mawby, J., Hand, M., Foden, J. & Kinny, P., 1998. Ordovician granulites in the southeastern Arunta Inlier: a new twist in the Palaeozoic history of central Australia. Geological Society of Australia, Abstracts, 49,296. Miller, J.A., Buick, I.S., Williams, I.S. & Cartwright, I., 1998. Re-evaluating the metamorphic and tectonic history of the eastern Arunta Block, central Australia. Geological Society of Australia, Abstracts, 49, 316.

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

105

NEOPROTEROZOIC EVENTS IN THE PHANEROZOIC OROGENIC BELTS, CENTRAL CHINA L Songnian, L Huaikun, Y. Haifeng, Z. Fengqing and Y. Chunliang Tianjin Institute of geology and Mineral Resources, CAGS, No. 4, 8th Road, Dazhigu, 300170, Tianjin, CHINA

The giant young orogenic belts are developed in the central China. The western part of the belts is named the Caledonian Qilianshan Orogen; the central and eastern segments are called the Indosinian Qinling-Dabie and Sulu Belts, respectively. The young orogenic belts are extended in NWW to near E-W and NE direction for about 3000 km in length. Quite a few Neoproterozoic thermo-tectonic events are preserved in the young orogenic belts. The Sulu Belt is Located in the southeast Shandong and north Jiangsu Provinces, to the east of the Tanlu Fault. It is currently considered as the early Mesozoic orogenic belt between the North China and Yangzi Blocks. However, there are abundant Neoproterozoic granitoid gneisses preserved in this belt covering more than 5000 km2. The gneisses were thought to be the early Precambrian stratigraphic sequences. However, based on recent mapping and other special research, these gneisses should be metamorphic plutonic granites, and the ages of them are concentrated in the interval of from 731Ma to 897Ma. It is noteworthy that eclogites as enclaves are preserved in the granitoid gneisses. Besides the early Mesozoic time information, many Neproterozoic ages from the eclogites have been reported. Therefore there is no doubt that the eclogite-bearing granitoid gneisses were formed during the Neoproterozoic thermo-tectonic events. The Qinling-Dabie Mountains are about 2000 km long in near E-W direction. They exhibit poly-orogenic features including Neoproterozoic, Paleozoic and Mesozoic collisions and breakups. The most extensive orogeny took place during the early Mesozoic at about 240-220Ma. A famous superhigh pressure metamorphic belt is preserved in the Dabie Mountain, which is located in the eastern part of the Qinling-Dabie orogenic belts. A number of the early Paleozoic, especially Neoproterozoic ages of thermo-tectonic events have been obtained. For example, a complex magmatic zone about 800km long and 10km to 30km wide has been distinguished in the western section of the Qinling Mountain. A few of ages from lOOOMa to 700Ma are measured from the magmatic zone. In addition, a suite of ophiolite is discovered in Shangzou-Xaiguan Fault. And an age of 832±32Ma of Sm-Nd isochron has been reported. It is obvious that the earlier collision and later breakup events in the Qinling-Dabie Belt took place during the Neoproterozoic time. The western segment of the Central Orogen named the Qilianshan Mountain is commonly regarded as the early Paleozoic orogenic belt. A very complex evolution of this belt has been described in many Chinese literatures. The ophiolites of CambrianOrdovician ages and pre-Late Devonian unconformity were discovered. Recently an eclogite-granitoid gneiss zone, which is situated along the northern margin of the Qaidam Basin, to the south of the Qilianshan belt, was reported. According to some new isotopic data, the emplacement of the granite rocks took place from the late Mesoproterozoic to early Neoproterozoic. And an age of near 800Ma has been measured from an eclogite sample. It is reasonable to suppose that there is a Neoproterozoic suture between the Qaidam and Qilianshan terrains. Meanwhile, a U-Pb age of about 680Ma from volcanic rocks has been measured. The volcanic rocks are preserved in the lower part of the subsequent rifting sequence. So both collision and rifting events of the Neoproterozoic time have been revealed in the western section of the Central Orogen. The older thermo-tectonic events have been reported in many places of the young Central Orogen. The most important ones are a great magmatic zone and eclogites as enclaves in granitoid gneisses of the Neoproterozoic age. It is reasonable to believe that the young mountains in the Central Orogen underwent a complex Neoproterozoic orogenic history. After the collision at the early Neoproterozoic time, subsequent rifting deposits were formed. The mobile belts from the Sinian to the early Paleozoic are developed. And finally we think that we should attach importance to the follwing points and keep them in mind in future work: 1.The huge magmatic zones of Neoproterozoic are preserved in the young Central Orogen. But different plutonic rocks including subduction, collision and post-orogenic granite types are developed in the zones. Geochronology, geochemistry and tectonic origin of plutonic rocks need to be studied in detail. 2. As enclaves in granitoid gneisses, eclogites have been discovered in many places of the Central Orogen. Some of the eclogites are products of the Neoproterozoic collision between terrains. The eclogite-bearing granitoid gneisses show evidence of collision belts or plate magines of the Neoproterozoic age. 3. Besides the collision event, the subsequent rifting basins are developed during the Neoproterozoic time in the Orogen. A preferred age of the transitional time between collision and rifting is concentrated at about 800Ma. It is noteworthy that the collision among older terrains and rifting of a new united block overlap in time but not in space.

1 0 6 OROGENESIS IN THE OUTBACK

DEFORMATION HISTORY OF PROTEROZOIC GRANITES IN THE NORTH OF THE BROKEN HILL BLOCK, NSW, AUSTRALIA S.A. Uberair, K. Stuwe^ K. Ehlers^ and A.P. Nutman^ 1) Department of Earth Science, Monash University, Clayton Vic. 3168, Australia 2) Department of Geology, University ofGraz, Heinrichstr 26, A-8010 Graz, Austria 3) AGCRC, Department of Earth Science, Monash University, Clayton Vic. 3168, Australia In the north of the Proterozoic Broken Hill Block, NSW, Australia, outcrop a series of granitic bodies with enigmatic relations to host rock - the "Mundi Mundi type granites". The two largest bodies (several km diameter) are the Cusin Creek and the Brewery Well granites. On a map scale, these granites have a roundish outline and appear to crosscut earlier structures. Thus, they have been inferred by previous workers to be undeformed and possibly of Pan African age. On the other hand, preliminary radiometric work suggests ages between 1690 and 1500 Ma, which is even older than the supposed age of the surrounding metasediments (Cooper & Ludwig, 1985). However, dating is difficult and the position of these granites remains largely unknown. In this study, we present the first detailed map of the internal variations of these granites and present an interpretation which supports a preorogenic position of these granites. In both granites, we have documented several fabrics and at least one thermal event subsequent to intrusion, all of which can be correlated with observations in the surrounding metasediments. CUSIN C R E E K G R A N I T E

The Cusin Creek Granite contains four different fabrics. These are well-preserved in several fine-grained mylonites, where folding of older foliations and successive downbreaking of the feldspars to micas can be observed. These mylonites were used to establish the timing relationship of deformation phases. The eastern and southern contacts of the pluton are covered by alluvium. The western and northern contacts are tectonic (the whole northern part of the pluton has been folded). Signs for a thermal event subsequent to intrusion have been found in relics of the originally intruded rocks, which were preserved in small pockets at the western margin of the granite. SHRIMP dating of few extracted zircons revealed two ages for Cusin Creek Granite: 1690 and 1600 Ma. The two different ages could be the result of the post-intrusive thermal event. BREWERY WELL GRANITE

The Brewery Well Granite granite intruded the surrounding Paragon metasediments. It contains two foliations, of which one trends in the same direction as the oldest foliation in Cusin Creek Granite. Other fabrics are present, but not conclusive. The eastern margin of the pluton was eroded prior to the deposition of Adelaidian sediments and has locally been disrupted by faulting. The Paragon metasediments around Brewery Well Granite contain two post-intrusive deformations and unconclusive traces of a third. Many rafts of the intruded metasediments are included in the granite, mainly on the eastern side. The rafts are from centimetre up to many tens of metres in diameter. At the east side of the pluton an aureole of white (1-2 cm long), needle-like crystals occurs in the Paragon metasediments. These crystals are probaly retrogressed andalusite or sillimanite. Along the western margin of the pluton many aplite-veins follow a big shearzone. The largest of these is sebveral hundred metres long and several ten metres wide. Dating of few extracted zircons revealed inherited ages older than 2000 Ma. This contribution was supported by the AGCRC and FWF Project P12846-GEO. REFERENCES

Cooper, J.A. and Ludwig, K.R; 1985 Inherited zircons in the Mundi Mundi Granite, Broken Hill, New South Wales. Australian Journal of Earth Sciences, 32, 467-470.

ALICE SPRINGS, AUSTRALIA, 1999

107

HIGH-GRADE METAMORPHISM OF LOW-T"DIAGENETIC-HYDROTHERMAL" ALTERATION IN A POLYMETAMORPHIC BELT, REYNOLDS RANGE, AUSTRALIA. J.K. Vry School of Earth Sciences, Victoria University of Wellington, P.O. Box 600, Wellington, New Zealand, email: Julie.Vry@vuw.ac.nz

The Early Proterozoic Lander Rock Beds in the Reynolds Range, N.T., central Australia, have undergone a complex polymetamorphic history involving: early low-grade regional metamorphism and deformation; subsequent low-P (c. 2.5 kbar) contact metamorphism at c. 1810 Ma (Collins & Williams, 1995; Vry et al 1996), which produced andalusite porphyroblasts in the aureoles of megacrystic S-type granites; associated hydrothermal fluid circulation, which involved igneous + metamorphic fluids near smaller granites like the Harverson Granite in the central Reynolds Range, and drew in an externally-derived fluid, probably a low-latitude basin brine, near larger granites like the Yaningidjara Orthogneiss in the SE Reynolds Range; focusing of those fluids during retrogression through c. 200 °C produced local diagenetic-hydrothermal deposits of Mg-rich chlorite (Vry & Cartwright, 1998); later regional re-metamorphism to high-T and low-P granulite facies conditions in the SE Reynolds Range (800-700 with pressures dropping from c. 5.3 to c. 3.5 kbar between c. 1594 - 1586 Ma; Buick et al, 1998; Vry & Cartwright, 1994); amphibolite facies retrogression (c. 650 ''C by c. 1568 Ma;Buick et al, 1998); and later shear zone development and local hydration at amphibolite facies and lower metamorphic grades, much of which probably occurred during the 400-300 Ma Alice Springs Orogeny (Buick et al, 1998). The Lander Rock beds are an ideal setting in which to study polymetamorphism. The rocks are for the most part monotonous in composition, and they contain local contact aureoles that have been overprinted to various degrees by the subsequent regional metamorphism, which extends from very low grade in the NW, to granulite facies in the SE, along the structural-stratigraphic strike of the range. Many of the rocks in the NW show little if any evidence of the contact metamorphism and subsequent regional metamorphism. Elsewhere in the area, it is possible to investigate the effects of contact metamorphism and associated hydrothermal circulation around large and small granites, and to study the overprinting effects of the later high-grade regional metamorphism, which range from minor to dominant, from NW to SE along the length of the range. This background of information makes it possible to consider the processes that produced the unusual characteristics of certain high grade rocks in the SE Reynolds Range. Most of the orthopyroxene in the high-grade Lander Rock Beds probably formed not as a result of biotite breakdown, but as a result of prograde devolatilization reactions involving the breakdown of diagenetichydrothermal Mg-rich chlorite. This occurred at high temperatures (c. 700-800 °C) near the peak of the granulite facies metamorphism. Varying proportions of chlorite (chl) and quartz (qtz) in the altered rocks produced a number of different high-grade rocks whose bulk chemistries plot along a mixing trend. Chlorite-dominated alteration zones produced non-foliated rocks composed largely of: orthopyroxene (opx) + spinel (spl) ± clinohumite or olivine, via reactions such as: chl + qtz = spl + 2 opx + 4 H20 (c. 750 °C), & chl = opx 4- olivine + spl + 4 H20 (c. 750 °C). Fluorine-bearing minerals (phlogopite, clinohumite, and wagnerite) are common in the opx + spl rocks, and apatite is relict. Elsewhere in the region, for example near Mt. Boothby, the rocks are more iron rich, and contain abundant magnetite. Similar rocks (gedrite + spinel + clinohumite) also occur in the overlying Reynolds Range group, where they are likely to have a related origin. Precursor rocks with more quartz developed into nonfoliated opx + cordierite (crd) rocks, and produced large orthopyroxene porphyroblasts in foliated quartz-biotitecordierite country rocks via reactions such as: 2 chl + 7 qtz = 8 opx + crd + 8 H20(c 700-750 °C). The matrix probably formed via reactions such as chl + muscovite + 2 qtz = biotite + crd + 4 H20. Although temperatures for a number of prograde metamorphic reactions can be estimated, pressure constraints are lacking for the prograde regional metamorphic history. Hydrothermal circulation following emplacement of the Yaningidjara Orthogneiss and the accompanying hydrous retrogression in the contact aureole would have modified the textures and mineral assemblages of the country rocks prior to the later regional high-grade metamorphic episode. Mats of fine micas and chlorite produced by previous retrogression of cordierite are a reasonable precursor to the mats of fine early sillimanite ± rare grandidierite that occur in the cores of cordierite in the paragneiss country rocks. This texture is likely to result from remetamorphism of previously retrogressed rocks. Where the zones of chlorite alteration affected rocks that already contained large porphyroblasts of contact metamorphic andalusite, spectacular pseudomorph and partial pseudomorph textures developed during the later

108 OROGENESIS IN THE OUTBACK

regional metamorphism to high grade conditions. The pseudomorphs were produced by prograde devolatilization reactions involving the andalusite porphyroblasts (± retrograde alteration) and chlorite, for example: 2 chlorite + 19 aluminosilicate = 5 cordierite + 11 corundum + 8 vapour (= 550 Mg-rich rocks). In more iron-rich rocks the cordierite and spinel assemblages formed. The pseudomorphs reflect metamorphic reactions at temperatures (= 500 to 600 ®C) well below those of the regional metamorphic peak. There is no evidence that the quartz in the matrix was ever in chemical equilibrium with the corundum or spinel in the pseudomorphs. These assemblages should not be interpreted in terms of a quartz-bearing chemical system, and were not produced, as previously proposed, during retrogression from the regional metamorphic peak. The compositions of other silica-undersaturated high-grade rock types, notably rocks rich in sapphirine and kornerupine require de-silicification at high temperatures. Rocks composed largely of nearly boron-free komerupine preserve spectacular symplectitic breakdown textures (e.g. cordierite -H sapphirine ± opx ± gedrite) that are interpreted to result from decompression at high temperatures during early retrogression following the M2 metamorphic peak (Vry and Cartwright, 1994). Elsewhere, coarse komerupine (presumably boron-bearing) is rimmed by cordierite + tourmaline. Relict magnesite clots (to 1.5cm dia.) in some coarse sapphirine + kornerupine rocks probably predates the high-grade regional metamorphism, and may date from the diagenetic-hydrothermal history. Adjacent to these clots, the kornerupine and sapphirine are corroded, and breakdown assemblages of chlorite and corundum ± tourmaline have formed. Fluid inclusions are abundant in all of the minerals in and adjacent to these magnesite clots, but several types occur: C02-rich inclusions, inclusions containing abundant daughter crystals, and probably also vapour-dominated inclusions. These fluid inclusions are unlikely to prove useful for constraining metamorphic pressures, but they indicate that at least at some times or locally during the high-temperature regional metamorphic history, metamorphic fluids were involved in active dissolution and element transport. REFERENCES

Buick, I.S., Cartwright, I., and Harley, S.L., 1998. The retrograde P-T-t path for low-pressure granulite from the Reynolds Range, central Australia: petrological constraints and implications for low-P/4iigh-r metamorphism. Journal of Metamorphic Geology, 16, 511-529. Collins, WJ., and Williams, LS., 1995. SHRIMP ionprobe dating of short-lived Proterozoic tectonic cycles in the northern Arunta Inlier, central Australia. Precambrian Research, 71, 69-89. Vry, J.K. and Cartwright, I., 1994. Sapphirine-kornerupine rocks from the Reynolds Range, central Australia: constraints on the uplift history of a Proterozoic low pressure terrain. Contributions to Mineralogy and Petrology, 116, 78-91. Vry, J.K. and Cartwright, I., 1998. Stable isotopic evidence for fluid infiltration during contact metamorphism in a multiply-metamorphosed terrane: the Reynolds Range, Arunta Block, central Australia. Journal of Metamorphic Geology, 16, 749-765. Vry, J.K., Compston, W., and Cartwright, L, 1996. SHRIMP II dating of zircons and monazites: reassessing the timing of high-grade metamorphism and fluid flow in the Reynolds Range, northern Arunta Block, Australia. Journal of Metamorphic Geology, 14, 335-350.

ALICE SPRINGS, AUSTRALIA, 1999

109

THE BEHAVIOUR OF ZIRCON, MONAZITE AND THEIR U-Pb ISOTOPIC SYSTEMS DURING THE HIGH-GRADE METAMORPHISM OF QUARTZO-FELDSPATHIC SEDIMENTARY ROCKS I. S. Williams^ a n d I. S. Buick^ ''Research School of Earth Sciences, Australian National University, Canberra, ACT 0200 ^Department of Earth Sciences, La Trobe University, Bundoora, Vic 3083

Evidence of reactions between the rock-forming minerals can be used to define, with reasonable confidence, the sequence of P-T conditions a particular suite of rocks has experienced during a particular metamorphic event. Placing that P-T path accurately within a numerical time frame can be difficult however. Many of the minerals used for radioisotope geochronology, for example amphiboles, micas and feldspars, have closure temperatures below the peak temperatures reached at high metamorphic grades. Those minerals provide valuable information on the retrograde path, but they retain little or no record of the prograde path. Recently, particularly following the advent of SHRIMP, extensive use has been made of minerals with higher closure temperatures, such as zircon and monazite, to provide information on the timing of near-peak metamorphic conditions. Zircon has proved especially valuable, the closure temperature of its U-Pb isotopic system ( - 9 0 0 "C) being above even the temperature of many magmas. Much remains to be learned, however, about how these minerals behave during metamorphism and, for different rock types, what points on the P-T path they might record. In the Reynolds Range, northern Arunta Block, late Palaeoproterozoic metasediments of the Reynolds Range Group overlie a basement of earlier Palaeoproterozoic granites and metasediments. The Reynolds Range Group has been subject to low to medium pressure (4-5 kbar) regional metamorphism that in the SE reaches granulite facies ( - 7 5 0 'C), producing local partial melting and, within the marbles and metapelites of the Upper Calcsilicate Unit, extensive zones of retrogression resulting from infiltration of hot (650-700 'C), water-rich metamorphic fluids. In places these zones are intruded by pegmatites and coarse-grained, quartz-rich, cordierite and sillimanite-bearing segregations up to several metres wide. Zircon from a granulite grade semipelite away from zones of obvious retrogression yields a range of Palaeoproterozoic ages reflecting the provenance of the metasediment. Most grains have narrow overgrowths of new zircon, characteristically low in ThAJ, which yield an age of 1594 ± 6 Ma, presumably reflecting the granulite grade event. Monazite from the same rock appears to be wholly metamorphic, all analysed grains yielding the significantly younger age of 1576 ± 8 Ma. Zircon from one semi-concordant metamorphic segregation from a retrograde zone is, in contrast, dominantly metamorphic. Thick overgrowths of two contrasting chemical compositions yield indistinguishable ages of 1589 ± 8 Ma and 1582 ± 8 Ma—the small zircon cores range in age from Palaeoproterozoic to late Archean. The monazite once again appears wholly metamorphic, yielding 1576 ± 1 2 Ma. A second, structurally latest segregation contains sillimanite-perforated metamorphic zircon megacrysts which yield 1568 ± 4 Ma. Temperatures high enough for metamorphic zircon and monazite growth were sustained for - 2 5 Ma, but nevertheless the detrital zircon, and therefore presumably also the early-formed metamorphic zircon, were not isotopically reset. At Cooma, in SE Australia, a small area o f early Palaeozoic quartz-rich turbidites has been regionally metamorphosed at low pressure ( - 4 kb) to the point of partial melting. Over a distance of about 8 km, the grade rises progressively from chlorite to biotite to andalusite to sillimanite grade, thence through a zone of migmatites and partial melting to a small body of low-temperature peraluminous granite. The granite appears to be a consequence, not the cause, of the metamorphism. The individual detrital zircon and monazite grains in the biotitegrade metasediment yield a wide range of ages reflecting the ages of the various components in the sediment's source region. With rising grade the monazite grains become increasingly corroded, although still retaining their age range, until eventually they disappear. At sillimanite grade and above monazite again is present, but yields only the age of the Silurian metamorphism, - 4 3 0 Ma. In contrast, zircon appears to be unaffected by the metamorphism either morphologically or isotopically until zircon overgrowths begin to form at sillimanite grade. The amount of new growth increases with grade, peaking in the migmatite leucosomes. The age of the zircon overgrowths matches that of the metamorphic monazite—the ages of the zircon cores, even in the granite, match closely the ages of the detrital zircons. The granite has formed from the deeper equivalents of the exposed sedimentary rocks, and the inherited zircon cores in the granite have survived temperatures in excess of 750 "C, preserving an accurate, unbiased record of the ages of the detrital zircons in that sediment. In contrast, all the monazite in the migmatite and granite appears to be newly grown.

1 1 0 OROGENESIS IN THE OUTBACK

RECURRENT TECTONOTHERMAL EVENTS IN A HIGH-GRADE TERRAIN: A LOOK INTO GEOCHRONOLOGY FROM SRI LANKA AND SURROUNDING AREAS M. Yoshida

Department of Geosciences,

Osaka City University, Osaka 558-8585,

Japan

The high-grade metamorphic belt, including Sri Lanka, LYtzow-Holm Bay, and southern most India, represents many indications of recurrent tectonothermal events spanning over ca 500 million years during the Mesoproterozoic to early Palaeozoic, although there have been some explanations that it was affected only by the Pan-African orogeny centered around ca 600 Ma. Duplications of deformational structures of more than four times are preserved well in all the terrains. Different kinematics and depth of each deformation point to the apparent recycling of different tectonics. These deformations coin mostly penetrative and partly non-penetrative structures on rocks of the terrains, thus rendering criteria to differentiate mineral associations relating to each deformation. Mineral reaction textures and mineral assemblages of different microstructural domains points to superpositions of distinctly different P-T conditions and PTt paths. Some of these microstructural events are correlatable to the deformational events, pointing to dramatic changes in the crustal environment during each stage of deformations. Geochronological studies by various methods have shown several fringes of ages including major ca 500 Ma and minor assemblies of ca 650-800 Ma and older ages with apparent inherited zircon SHRIMP ages of ca 1200-2700 Ma (Yoshida et al., 1996a). Occurrence of intrusive plutonic masses of ca 1000 Ma or older provides evidence that the surrounding metasupracrustals are Mesoproterozoic or older. Nd TDM ages and Sm-Nd isotopic characteristics indicate a common geotectonic signature of all these terrains as being composed of three distinct geologic units with definite Sm-Nd isotopic characteristics and TDM ages of Mesoproterozoic, Paleoproterozoic, or Late Archaean, reflecting derivations from different protolith of the three geologic units (Yoshida et al., 1996b). Reference of ca 550 Ma ages to the last high-grade metamorphism preceded by least deformed granite-pegmatite of ca 700 Ma and followed by granite-pegmatite of ca 450-500 Ma is well constrained by various geochronologic data including the Sm-Nd mineral isochron ages and concordant and lower intercept zircon SRHIMP ages of metamorphites and of U-Pb ages of the granitic intrusives. Reference of older ages to earlier tectonothermal events is strongly suggestive from various evidence associated with careful considerations on isotopic systematics and microstructures of rocks. Ideas that these areas suffered only one tectonothermal event at ca 550 Ma or one PanAfrican event spanning ca 450-750 Ma are discussed with regard to oversimplified explanation of SHRIMP age data (e.g., Shiraishi et al., 1994; Shiraishi et al, 1995; Yoshida, 1995). REFERENCES

Shiraishi, K., Ellis, D.J., Hiroi, Y., Fanning, C.M., Motoyoshi, Y. and Nakai, Y., 1994, Cambrian orogenic belt in East Antarctica and Sri Lanka: Implications for Gondwana assembly. Jour. GeoL, 102, 47-65. Shiraishi, K., Ellis, D.J., Hiroi, Y., Fanning, C.M., Motoyoshi, Y. and Nakai, Y., 1995, Cambrian orogenic belt in East Antarctica and Sri Lanka: Implications for Gondwana assembly: A reply. J. GeoL, 103,469-471. Yoshida, M., 1995, Cambrian orogenic belt in East Antarctica and Sri Lanka: Implications for Gondwana assembly: A discussion. Jour. GeoL, 103, 467-468. Yoshida, M. Bindu, R.S., Kagami, H., Rajesham, T., Santosh, M. and Shirahata, H., 1996a, Geochronologic constraints of granulite terrains of South India and their implications for the Precambrian assembly of Gondwana. J. Southeast Asian Earth Sciences, 14, 137-147. Yoshida, M., Kagami, H. and Unnikrishnan-Warrier, C., 1996b, Neodymium model ages from India-Antarctic sector of East Gondwana: Differentiation of isotopic provinces. Abstracts, IGC, 2, 573.

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

111

RECYCLIC TECTONOTHERMAL EVENTS IN EASTERN GHATS: CONTINUATION TO SW AUSTRALIA AND ANTARCTICA M. Yoshida\ M. Arima'and A.T. Rao'

1 Department of Geosciences, Osaka City University, Osaka 558-8585, Japan 2. Geological Institute, Yokohama National University, Yokohama 240-8501, Japan 3. Department of Geology, Andhra University, Visakhapatnam 530 003, India

The Proterozoic Eastern Ghats Mobile Belt (EGMB) running along the east coast of Peninsular India shows polyorogenic signature, ranging from late Archaean to early Palaeozoic, in all accounts including protolith c metaigneous rocks, metamorphic PTt path and geochronology. A broad zonation of the belt into four zones has bee recognized, viz., the Western Charnockite Zone composed mostly of late Archaean enderbite-charnockites wit considerable amount of basic granulites and some metasediments, the Western Khondalite Zone dominated wit Proterozoic metasediments with orthogneisses and younger intrusives, the Central Migmatite Zone where dominar Proterozoic migmatization develops along with intrusive granites and charnockites, and the Eastern Khondalite Zon having similar characteristics as the Western Khondalite Zone (Ramakrishnan et al., 1998). Several ductile shear zone separate EGMB into number of different blocks/subterrains, having distinct structural and geohistoric characteristic (Chetty and Murthy, 1998, GSI Sp. Publ 44). The protolith of EGMB is composed of sedimentary rocks of mostly continental and passive continental margi sediments of possible Proterozoic ages with some magmatic rocks of diverse tectonic settings. There are cles indications of riftogenic magmatic rocks of ca 1450 Ma and intrusive alkaline rocks of ca 1240-1450 Ma (Sarkar an Paul, 1998, GSI Sp. Publ. 44; Takano and Arima, 1999). Granitic and charnockitic intrusions of ca 950-1100 Ma hav signatures of convergent tectonic signatures, and some charnockitic rock of ca 1900 Ma from central EGMB has bee identified also as the subduction-related activity (Takano and Arima, op cit). Granite activity of ca 800 Ma with posi tectonic and intracratonic characteristics is known. Circa 500-600 Ma rejuvenation is also dominant throughout. Several deformational events have been identified from various places of EGMB (e.g., Halden et al., 1982, Trans. Roy< Soc. Edinb, Earth Sci., 73; Biswal et al., 1998), generally composed of banding structure (Dl), recumbent-isoclinal fold (D2), overturned tight to isoclinal folds (D3), upright open folds (D4), vertical ductile shears (D5), and brittle sheai (D6). There are, however, some observations of other structures including intrafolial rootless isoclinal folds with hinge of disperse plunges, mesoscopic sheath folds, macroscopic sheath folds, and domal structures (e.g., Biswal, et al., op ci Tani et al., 1998, J. Geosci. Osaka City Univ., 41) some of which are not easily understood to belong the abov deformational sequence. Although a picture of early recumbent folds related with westerly thrusting and later due til shearing under a dextral transpression proposed by Chetty and Murthy (op cit) appears to be generally acceptable EGGB might have experienced a further complex superposed deformational events. Peak conditions of the granulite facies metamorphism with temperatures over 900_C under about 8-9 kb, preceded by a isothermal burdening, was followed by a combination of either isobaric cooling (IBC)-isothermal decompression (ITD; or ITD followed by IBC (e.g., Dasgupta et al., 1998, GSI Sp. Publ. 44). A magmatic underplating under an extention< crustal condition is suggested for the peak metamorphic events followed by early IBC path, and a collisional event j considered to reflect the later ITD path. A very high pressure conditions of over 13 kb with temperature over 1150_C a the earliest event before the IBC-ITD path has been pointed out by Shaw and Arima (1998), indicating a piling of crust over 50 km thickness suggesting a collisional event before the extensional event. Detections of contradictory PTt path among authors may reflect a complex metamorphic history, including polyorogenic events (cf. discussions in Dasgupt< op cit); further detailed researches are required. Geochronologic data include mostly from northern (e.g., Aftalion et al, 1988, J. GeoL, 96) and central areas (e.g., Sha^ et al., 1997, J. Geol., 105) with sporadic data from some areas which are reviewed by Sarkar and Paul (op cit). TDIN ages range from ca 2.2 Ga to 3.0 Ga (Sarkar and Paul., op cit; Yoshida et al., 1996, J. SE Asian Earth Sci., 14). A TD> age of ca 3.0 Ga of basic granulite from northern EGMB was explained by Sarkar and Paul (op cit) to be the age of th extrusion of the basic igneous rocks from mantle and thus to provide the age of the protolith of metasupracrustj sequence. However, younger TDM ages (ca 2.2-2.9 Ga) and an inherited zircon age (2.75 Ga) (Shaw et al., op cit) c metasediments surrounding metabasites from central area do not conform the above idea. A preferred model is th sedimentation sometime earlier than ca 2.0 Ga, and some major tectonothermal events must have taken place earlier tha the intrusion of alkaline plutons of ca 1240-1450 Ma (e.g., Sarkar and Paul, op cit). Major granulite metamorphism i considered either ca 1450 Ma, or ca 1450 Ma and ca 2000 Ma. Later high- to middle-grade metamorphisms in EGM]

1 1 2 OROGENESIS IN THE OUTBACK

were dated as ca 1000-1100 Ma, ca 800 Ma, and ca 550 ma by various methods including zircon SHRIMP and EPMA monazite ages (e.g., Shaw et al., op cit; Simmat and Raith, 1998, Gondwana Research Group Miscl. Pub. 8; Bindu et al., 1998, GRG Miscl. Pub. 8). The ca 1000 Ma events has a signature of convergent orogeny associated with charnockite intrusions, and other events are intracratonic events associated with granitic activity. The ca 1000 Ma events in EGMB is traceable northeastwards to Albany-Flaser Granulite Belt of SW Australia as well as southeastwards to the Rayner and Napier complexes of East Antarctica. The younger events than ca 1000 Ma of EGMB are comparable with those of the Antarctic rocks, while the older events are seen in those of SW Australia, appearing to reflect recyclic amalgamation-disruption of crustal fragments surrounding EGMB ranging from late Archaean to early Palaeozoic. A possible model of early crustal history of the East Gondwana surrounding EGMB will be discussed. SELECTED REFERENCES

Biswal, T.K, Sawjeevan, G. and Nayak, B.P., 1998, Deformational history of Eastern Ghats Mobile Belt around Lathore, Balangir District, Orissa. Jour. Geol Soc, India, 51, 219-225. Ramakrishnan, M., Nanda, J.K. and Augustine, P.P., 1998. Geological Evolution of the Proterozoic Eastern Ghats Mobile Belt. GSI Spec, PubL 44, 1-21. Shaw, R.K. and Arima, M. 1998, A corundum-quartz assemblage from the Eastern Ghats Granulite Belt, India: evidence for high-P-T metamorphism? Jour. Metamorphic GeoL, 16. Takano, N. and Arima, M., 1999, Abstracts, Japan Earth and Planet. Sci. Joint Meeting, Tokyo, June 1999.

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

113

POLYMETAMORPHISM OF ARCHEAN MAFIC GRANULITES FROM THE TRANSNORTH CHINA OROGEN: TEXTURAL EVIDENCE AND TECTONIC IMPLICATIONS G. Z h a o , P. A. C a w o o d & S. A. W i l d e

Tectonics Special Research Centre, School of Applied Geology, Curtin University of Technology, Box U1987, Perth, WA 6845

GPO

The North China craton consists of Early Archean to Paleoproterozoic basement overlain by Mesoproterozoic to Cenozoic cover. Lithological, structural, metamorphic and geochronological data for the North China craton enable the division of the craton into western and eastern blocks of Archean age, separated by a north-south trending Paleoproterozoic belt, the Trans-North China orogen (Fig. 1; Zhao et al., 1998, 1999). The eastern and western blocks are composed predominantly of Early to Late Archean tonalitc-trondhjemitic-granodioritic ( T T G ) grey gneisses and syntectonic granitoid rocks with rafts of supracrustal rocks consisting of ultramafic to felsic volcanics and sedimentary rocks metamorphosed from greenschist to granulite facies at -2.5 Ga. The Trans-North China orogen comprises a series of belts containing the reworked Archean materials, represented by T T G gneisses and mafic granulites, amphibolites and greenschists, and Paleoproterozoic juvenile crustal components, represented by metavolcanics, metasedimentary rocks and syntectonic granites. All these rocks underwent widespread greenschist to granulite facies metamorphism at -1.8 Ga.

110°E

120°E 200

400 km

H e n g s h ^ ^ ^ Beijing

/ iLuliang /v J

I I

/''^Zhongtiao Mesoproterozoic to Cenozoic cover

Archean Paleoproterozoic . ^ I Major fault

Fig. 1 Distribution of the basement rocks in the North China craton and the distribution of the Trans-North China orogen and eastern and western blocks. WB—Western block; EB— Eastern block; TNCO—Trans-North China orogen.

The polymetamorphism of Archean mafic granulites from the Trans-North China orogen is indicated by complex textural relations of garnet porphyroblasts and associated inclusions and symplectic coronas. Garnet porphyroblasts show two distinguishable phases: inclusion-rich core and inclusion-poor rim. The inclusion-rich garnet-core which is intergrown with quartz mantles a fine-grained inclusion assemblage of orthopyroxene + clinopyroxene -fplagioclase -i- quartz, whereas the inclusion-poor gamet-rim is surrounded by plagioclase + orthopyroxene and/or

1 1 4 OROGENESIS IN THE OUTBACK

plagioclase + hornblende symplectic coronas. Separating the garnet core and rim is plagioclase + biotite retrogressive rims. These textural relations indicate the presence of two discrete metamorphic events. The first metamorphic event (Mj) produced the mineral assemblages: (1) fine-grained orthopyroxene + clinopyroxene + plagioclase + quartz (Mj-i), mantled by the inclusion-rich garnet-core; (2) inclusion-rich garnet core + quartz (Mj. 2); and (3) plagioclase + biotite retrogressive rims around garnet (M1-3). These assemblages and their P-T estimates define an anticlockwise P-T path involving near-isobaric cooling followed by cooling with decreasing pressure. The mineral assemblages for the second metamorphic event (Mn) includes: (1) garnet rim (inclusion-poor) + orthopyroxene (matrix) 4- plagioclase (matrix) + clinopyroxene (matrix) + quartz (matrix) (Mji-l); (2) symplectic or coronitic orthopyroxene + plagioclase ± clinopyroxene (Mn-2)» 0 ) symplectic plagioclase + hornblende (Mji3). These assemblages and their P-T estimates define a clockwise P-T path involving near-isothermal decompression followed by cooling with decreasing pressure. The anticlockwise P-T paths of the first metamorphic event preserved by the mafic granulites from the Trans-North China orogen is very similar to those P-T paths established for the Archean mafic granulites from the eastern and western zones of the North China craton (Zhao et al., 1998; 1999, in press), and thus, they may have resulted from the same metamorphic event which occurred at -2.5 Ga. The metamorphism involving anticlockwise P-T paths generally is considered to be related to underplating and intrusion of mantle-derived magmas, which may occur in continental magmatic arc regions (Bohlen, 1991), above hot spots driven by mantle plumes (cf. Newton et al, 1980; Zhao & Wu, 1996), and in continental rift or extensional basin environments (Sandiford & Powell, 1986). On the basis of lithological, structural, geochronological and geochemical data, Zhao et al. (1998) proposed that mantle plumes played a significant contribution to the heat source of the Late Archean metamorphism of the eastern and western blocks. By the end of the Archean, two continental blocks, eastern and western North China, had developed through the interaction of mantle plumes with the lithosphere from the Early Archean to Late Archean. The clockwise P-T paths recorded in the mafic granulites from the Tans-North China orogen is similar to those P-T paths of the Paleoproterozoic khondalite series and other rock units from the western block and Trans-North China orogen, and reflect tectonothermal processes characterized by initial crustal thickening, subsequent nearly isothermal exhumation and final cooling. This sequence of tectonic processes is thought to be related to continentcontinent collisional environments (England & Thompson, 1984). Considering the spatial and temporal relationships between these rocks and the eastern and western blocks, this collisional environment may be related to the amalgamation of the eastern and western continental blocks in the Paleoproterozoic. The collision of the eastern and western continental blocks in the late Paleoproterozoic caused the crust of the Trans-North China orogen to be thickening at the peak of metamorphism (Mji-l). Following peak metamorphism, the crust underwent rapid exhumation (Mn-2)» forming widespread decompression textures in the rocks. Finally, cooling associated with retrogression (Mn-3) occurred in the metamorphosed crust after exhumation had ceased. Therefore, the mineral reaction textures and P-T paths of the Archean mafic granulites from the Trans-North China orogen record a polymetamorphic history of the craton from -2.5 Ga to 1.8 Ga. REFERENCES

Bohlen, S. R., 1991, On the formation of granulites. Journal of Metamorphic Geology, 9, 223-229. Brown, M., 1993, P-T-t evolution of orogenic belts and the causes of regional metamorphism. Journal of Geology Society, London, 150, 227-241. England, P. C., and Thompson, A. B., 1984, Pressure-temperature-time paths of regional metamorphism, 1. Heat transfer during the evolution of regions of thickened continental crust. Journal of Petrology, 25, 894-928. Newton, R. C., Smith, J. V., and Windley, B. F., 1980, Carbonic metamorphism, granulites and crustal growth. Nature, 288,45-50. Sandiford, M., and Powell, R., 1986, Deep crustal metamorphism during continental extension: ancient and modem examples. Earth and Planetary Science Letter, 79, 151-158. Zhao, G. C., and Wu, F. Y., 1996, Mantle plume tectonics: a possible tectonic mechanism of the formation and growth of the Early Archean to Middle Archean crust. 30th International Geological Congress, Abstract 2, p. 339. Zhao, G. C., Cawood, P. A., and Lu, L. Z., 1999, Petrology and P-T history of the Wutai amphibolites: implications for tectonic evolution of the Wutai Complex, China. Precambrian Research, 93, 181-199. Zhao, G. C., Wilde, S. A., Cawood, P. A., and Lu, L. Z., 1998, Thermal evolution of Archean basement rocks from the eastern part of the North China craton and its bearing on tectonic setting. International Geology Review, 40, 706-721. Zhao, G. C., Wilde, S. A., Cawood, P. A., and Lu, L. Z., in press. Thermal evolution of two types of mafic granulites from the North China craton: evidence for both mantle plume and collisional tectonics. Geological Magazine.

ALICE SPRINGS, AUSTRALIA, 1 9 9 9

115

TWO TYPES OF METAMORPHISM OF THE BASEMENT ROCKS IN THE WESTERN ZONE OF THE NORTH CHINA CRATON AND THEIR TECTONIC IMPLICATIONS G. Zhao, S. A. Wilde, P. A. Cawood

Tectonics Special Research Centre, School of Applied Geology, Curtin University of Technology, Box U1987, Perth, WA 6845, Australia

GPO

The basement of the North China Craton is divisible into the Archean eastern and western blocks, intervened by the Paleoproterozoic Trans-North China orogen (Zhao et al., 1999). The eastern block is composed predominantly of tonalitic-trondhjemitic-granodioritic (TTG) gneisses and syn-tectonic granitoids, with minor rafts and sheets of supracrustal rocks including ultramafic to felsic volcanic rocks and metasediments, metamorphosed from greenschist to granulite facies with anticlockwise P - T - t paths (Zhao et al., 1998). The TTG gneisses make up over 80% of the basement and the structural style is dominated by ovoid domes, separated by linear belts. Basement rocks have protolith ages of 3.85 to 2.50 Ga and underwent regional metamorphism at 2.6-2.5 Ga. The western block has a two-layered basement, with Archean TTG gneisses and mafic igneous rocks unconformably overlain by Paleoproterozoic khondalite series metasediments. The structural style is characterized by Archean TTG gneiss domes in the northwest, flanked to the southeast by linear khondalite belts. Separating these two blocks is the TransNorth China orogen that extends as a north-south trending belt through the provinces of Shanxi, Hebei and Inner Mongolia. The orogen consists of a series of high-pressure terranes containing reworked Archean components derived from the eastern and western blocks, together with Paleoproterozoic igneous and sedimentary rocks metamorphosed at greenschist to granulite facies at -1.8 Ga (Lu and Jin, 1993; Zhao et al., 1999, in press). On the basis of lithological, structural, metamorphic and isotopic data, and especially considering the spatial and temporal relationships between rocks in the Trans-North China orogen and those in the eastern and western blocks, Zhao et al. (1999, in press) suggest two different metamorphic events have affected the basement of the North China craton: the first event at 2.6-2.5 Ga, resulting in the development of the eastern and western blocks of the craton; and the second event of 1.8 Ga, amalgamating the Archean eastern and western blocks to form the North China craton. The two metamorphic events are clearly defined by textural relations and their P - T paths of the basement rocks from the western block.

Fig. 1 Spatial distribution of the eastern and western blocks separated by the Trans-North China orogen. WB—Western block; EB—Eastern block; TNCO—Trans-North China orogen. The Late Archean metamorphic event (Mj) is only recorded in the Archean TTG gneisses and mafic granulites, amphibolites and greenschists from the western block. In the high-grade areas, the mafic granulites and TTG gneisses display three metamorphic stages. The M j - i stage is represented by the assemblage plagioclase + hornblende -h quartz as mineral inclusions within garnet or pyroxene grains, and records the early metamorphic conditions with temperatures of 600-700°C and pressures of 5 - 6 kbar. The peak stage (Mi-2) is represented by the assemblage hypersthene + clinopyroxene + garnet + plagioclase -i- quartz and records conditions with temperatures of 800-850 °C and pressures of 8.0-10.0 kbar. The post-peak stage (M1.3) is represented by garnet + quartz ±

116 OROGENESIS IN THE OUTBACK

clinopyroxene symplectic coronas around the peak metamorphic minerals and records the P-T conditions with temperatures of 700-750°C and pressures of 8.0-10.0 kbar. These mineral assemblages and their P-T estimates suggest an isobaric cooling anticlockwise P-T path for the Archean mafic granulites and high-grade TTG gneisses in the western block. In the low-grade areas, the best M j information is obtained from gametiferous amphibolites, in which a similar sequence of three metamorphic stages has been recognized. The minerals of the M j - i stage are preserved as inclusions within garnet grains and are represented by actinolite, chlorite, epidote and Na-rich plagioclase. Their P - T conditions are semi-quantitatively estimated at 400°C and 2.5 kbar. The Mi.2 assemblage is hornblende + plagioclase (An30) + quartz -H garnet, representing the peak of metamorphism with P-T conditions of 5.0 kbar and 650°C. The final stage (M1.3) is indicated by retrogressive actinolite -f- epidote + plagioclase + quartz and metamorphic conditions are estimated at 450^^0 and 5.0 kbar. These assemblages and their P - T estimates also define an isobaric cooling-type anticlockwsie P-T path. The Paleoproterozoic metamorphic event (Mn) is recorded by both Paleoproterozoic khondalites and the reworked Archean mafic granulites, which are only exposed close to the boundary with the Trans-North China orogen. The Paleoproterozoic khondalites exhibit four metamorphic stages. The M n - l assemblage is preserved as mineral inclusions within garnet grains, with a typical assemblage of plagioclase + biotite + quartz + garnet + staurolite + kyanite + rutile ± ilmenite; the Mn-2 stage is represented by matrix mineral assemblages of plagioclase -h biotite quartz + garnet + sillimanite + spinel ± ilmenite; the M11.3 stage is represented by cordierite + sillimanite symplectic coronas around garnets or by cordierite coronas around spinel; and the M n - l stage is defined by andalusite and muscovite porphyroblasts which overprint the regional foliation defined by biotite and sillimanite. These textural relations define a clockwise P-T path. In the reworked mafic granulites, the polymetamorphic features are shown by two distinguishable phases of garnet: inclusion-rich core and inclusion-poor rim, separated by biotite + plagioclase retrogressive rims. The earlier garnet-core intergrown with quartz mantles the fine-grained inclusion assemblage orthopyroxene -1- clinopyroxene + plagioclase + quartz, whereas the later garnet-rim is surrounded by plagioclase 4- orthopyroxene and/or plagioclase + hornblende symplectic coronas. Based on the two phases of garnet and associated inclusion minerals and symplectic coronas, the mineral assemblages assigned to the first metamorphic event (Mj) include: fine-grained orthopyroxene 4- clinopyroxene + plagioclase + quartz ( M M ) , mantled by the inclusion-rich garnet-core; inclusion-rich garnet core quartz (M1-2); and plagioclase + biotite retrogressive rims around garnet (M1.3). These assemblages and their P - T estimates define an anticlockwise P - T path involving near-isobaric cooling followed by pressure-down cooling. The mineral assemblages assigned to the second metamorphic event (Mji) includes: garnet rim (inclusion-poor) -1- matrix-type orthopyroxene + plagioclase + clinopyroxene -i- quartz (Mji-l); symplectic or coronitic orthopyroxene plagioclase ± clinopyroxene (M11.2); and symplectic plagioclase + hornblende (Mn.3). These assemblages and their P - T estimates define a clockwise P - T path involving near-isothermal decompression, which is similar to the P - T paths estimated for the Paleoproterozoic khondalites. The Late Archean metamorphism involving anticlockwise P-T paths in the North China craton is interpreted as resulting from the underplating and intrusion of mantle-derived magmas which were associated with mantle plumes (Zhao et al., 1998). The clockwise P-T-t paths of the khondalite series and associated reworked Archean mafic granulites from the western block reflect tectonothermal processes characterized by initial crustal thickening, subsequent near-isothermal exhumation and final cooling. These tectonic processes are thought to be related to continent-continent collision of the eastern and western continental blocks. The collision caused the crust to be doubled at the peak of metamorphism. Following peak metamorphism, the thickened crust underwent rapid exhumation and formed widespread decompression textures in the rocks. Finally, retrogressive cooling occurred after exhumation ceased. The mineral reaction textural relations and P-T paths of the khondalite series and reworked Archean mafic granulites from the western zone record the tectonothermal history of the amalgamation of the eastern and western continental blocks which resulted in the final assembly of the North China craton. REFERENCES

Lu, L. Z., and Jin, S. Q., 1993. P - T - t paths and tectonic history of an early Precambrian granulite facies terrane, lining district, south-eastern Inner Mongolia, China. Journal of Metamorphic Geology 11, 483-498. Zhao, G. C., Cawood, P. A., and Lu, L. Z., 1999, Petrology and P - T history of the Wutai amphibolites: implications for tectonic evolution of the Wutai Complex, China. Precambrian Research, 93, 181-199. Zhao, G. C., Wilde, S. A., Cawood, P. A., and Lu, L. Z., 1998, Thermal evolution of Archean basement rocks from the eastern part of the North China craton and its bearing on tectonic setting. International Geology Review, 40, 706-721. Zhao, G. C., Wilde, S. A., Cawood, P. A., and Lu, L. Z., in press. Thermal evolution of two types of mafic granulites from the North China craton: evidence for both mantle plume and collisional tectonics. Geological Magazine.