Issuu

GEOLOGICAL SOCIETY OF AUSTRALIA - SPECIALIST GROUP: TECTONICS & STRUCTURAL GEOLOGY I

2 2 - 2 6 SEPTEMBER, 2 0 0 3

Kalbarri, Western Australia

PROGRAM & ABSTRACTS

FIRST PUBLISHED 2003 BY PROMACO C O N V E N T I O N S PTY LTD FORTHE

Geological Society of Australia Specialist Group: Tectonics & Structural Geology Field Meeting 2003 Edited by Reddy, S.M., Fitzsimons, I.C.W. and Collins, A.S.

This volume is a pre-Conference publication of the presentations to be given at the Specialist Group Field Meeting, to be held in Kalbarri, Western Australia, September 2003. The contents and any opinions expressed within the abstracts represent the views of the authors only. This publication is copyright. Apart from fair dealing for the purpose of private study, research, criticism or review as permitted under the Copyright Act, no part may be produced without prior written permission from the publisher. Additional copies may be obtained from: Promaco Conventions Pty Ltd PO Box 890 Canning Bridge WESTERN AUSTRALIA 6153 Tel: (08) 9332 2900 Fax: (08) 9332 291 I Email: promaco@promaco.com.au

ISBN: I 86308 107 0

ADDENDA Abstracts by Sheldon and Wheeler and Rolland et al were unfortunately omitted from the printed abstract volume. The 2003 SGTSG Committee apologise to the authors concerned for this oversight, and these two abstracts are included here as loose sheets. Both papers were presented as posters in the Faults and Fluid session at the 2003 Kalbarri SGTSG Field Meeting.

SGTSG Committee Perth 2003.

SGTSC; I'ieUl Meeting 2003

INFLUENCE OF PORE FLUID CHEMISTRY ON THE STATE OF STRESS IN SEDIMENTARY BASINS Heather A. Sheldon^'^ and John Wheeler^ ^Department of Earth Sciences, University of Liverpool, 4 Brownlow Street, Liverpool, L69 3GP, UK ^Now at CSIRO Exploration & Mining, ARRC, 26 Dick Perry Avenue, Kensington, Perth, WA 6151

INTRODUCTION T h e r m o d y n a m i c principles predict a relationship between stress and chemistry, such that solubility increases with the normal stress acting on a solid-fluid interface. We investigate this relationship between stress and chemistry in the context of a sandstone undergoing pressure solution, focusing on the implications for horizontal stress within basins. Pressure solution is a 3-stage process, involving (1) Grain contact dissolution; (2) Transport of solute to the adjacent pore space by diffusion; (3) Removal of solute from the pore fluid by precipitation and/or transport processes. It is driven by local gradients in chemical potential that arise from the difference in normal stress between grain contacts and free faces of grains. Pressure solution generally causes adjacent grains to move towards one another; however, if the pore fluid is oversaturated, there may be a driving force for solute to diffuse into the grain contacts and precipitate there, thus causing the grains to move apart. If the boundaries of the system are fixed, this driving force must be counterbalanced by a compressive stress. Consequently, there is a relationship between pore fluid saturation and stress in compacting sediments. METHOD A numerical model is used to simulate the evolution of porosity, fluid pressure, and stress, in a vertical, 1-D column of quartz sandstone undergoing pressure solution and cementation. The column forms part of a larger body of sandstone that is constrained laterally, i.e. it is undergoing uniaxial strain, with no horizontal movement of solid or fluid. The strain rate due to pressure solution depends on: (1) the "driving force", which in turn depends on the effective stress acting on grain contacts and the concentration of solute in the pore fluid (relative to the equilibrium concentration); and (2) various kinetic parameters, in particular the grain boundary diffusivity and the rate constants for dissolution and precipitation. These parameters are varied within the ranges Ahstrui I Siihniission

suggested in the literature, in order to illustrate the interaction between pore fluid saturation and stress. RESULTS The model predicts a smooth decrease in porosity with depth, until a minimum porosity (<1%) is reached. At this point the pore fluid becomes overpressured, thus inhibiting further compaction. The concentration of solute in the pore fluid in this overpressured region is close to the equilibrium value, implying a balance between addition of solute by grain contact dissolution, and removal of solute by precipitation on the free faces of grains. The horizontal stress in this region must be equal to the fluid pressure, which is approximately lithostatic, in order to prevent lateral expansion of the sandstone. Above the overpressured region, the concentration of solute, and hence the horizontal stress, depends on the rate constant for precipitation. If precipitation is slow relative to grain contact dissolution, e.g. due to grain coatings, solute builds up in the pore fluid and hence the horizontal stress is relatively large. This enhancement in horizontal stress can be as large as the enhancement associated with overpressure. Conversely, if the rate constant for precipitation is greater than or equal to that for grain contact dissolution, the level of supersaturation in the pore fluid is much less and the horizontal stress is approximately equal to the fluid pressure throughout the column. CONCLUSIONS In a laterally c o n s t r a i n e d sediment undergoing pressure solution, the horizontal stress must be equal to the fluid pressure, plus an extra term dependent on the concentration of solute in the pore fluid. This extra term can be significant, of the same order of magnitude as stress enhancement due to overpressure. We have focused on the evolution of pore fluid composition due to pressure solution, but any process that influences pore fluid chemistry could impact on the stress state within sedimentary basins.

SGTSa hicld Meeting 2003

FLUID FLOW AND ELEMENT MOBILITY IN MID-CRUSTAL SHEAR ZONES IN COLLISIONAL OROGENS: INSIGHTS FROM THE MONT BLANC MASSIF SHEAR ZONE NETWORK Y, Rolland ^ ^ S.F.Cox \ A.-M. Boullier ^ G. Pennachioni ^ N. Mancktelow ^ M. RossiM.Martelat ^ O. Vidal ^ ^ Geology Dept., ANU, Canberra, ACT 0200, Australia ^ Geosciences Azur, Universite de Nice-Sophia Antipolis, Pare Valrose, 06108 Nice ^LGCA-LGIT, BP53, Univ. J. Fourier, 38041 Grenoble, France ^ Dip. di Geologia, Via Giotto 1, 35137 Padova, Italy ^ Department Erdwissenschaften, ETH-Zentrum, CH-8092 Zurich, Switzerland

Networks of kinematically-related Alpine shear

transfer due to fluid-driven alteration. Fluid-

zones in the Mont Blanc nnassif granite have

driven alteration has occurred while the shear

localised fluid flow, hydrothermal alteration

zones were active, and is compatible with

and associated vein formation at mid-crustal

generally small, but locally important volume

depths during Alpine collision. The high relief

changes of the sheared granite. Very variable

of the Mont Blanc massif, and the tunnel

elemental depletions and enrichments (in

which cuts through it, provide an opportunity

particular for the Rare Earth elements) are

to investigate the coupling between shear

interpreted to reflect major differences in (1)

zone development,

fluid-rock

fluid chemistry, (2) fluid flow directions relative

fluid flow,

interaction, and mass transport during Alpine

to PIT gradients, and (3) connectivity between

deformation.

elements of the shear network and fluid

The shear zones have developed in response

reservoirs, in different parts of the shear

to NW-SE shortening and vertical extension at

network. Time-integrated fluid fluxes are

c. 20-18

fluid

estimated to be of the order of 10® m^m"^ in

inclusions and mineral thermo-barometry on

the zones of highest fluid flux. In addition, C,

shear-hosted veins indicate

zone

O, H stable isotopic composition of calcite-

formation at around 400°C and relatively high

chlorite-quartz assemblages in veins and

Ma. Calcite-quartz shear

pressures (500 MPa). Four main hydrothermal

shear zones indicate that fluid chemistries are

alteration styles have been recognised in the

the result of the mixing of fluids derived from

massif.

These assemblages are dominated

two reservoirs. In the core of the massif, the

by: (1) in the NW part of the massif, epidote-

shear networks have facilitated escape of

muscovite; (2) in the central part, chlorite-

reduced fluids from deeper crustal or mantle

phlogopite ± muscovite, followed by (3)

fluid reservoirs, whereas near the SE margin

actinolite-muscovite ± chlorite to the E; and (4)

of the massif there is evidence for mixing of

in the SE part of the massif, muscovite-biotite

deeply-sourced fluids with fluids originating

± calcile ± chlorite. Geochemical changes are

from the overlying

controlled by metamorphic minerals and mass

nappes.

Ahstrac / Submission

or adjacent

Helvetic

SGTSG Field Meeting

2 2 - 2 6 SEPTEMBER, 2 0 0 3

Kalbarri, Western Australia PROGRAM AND ABSTRACTS

Welcome I It is w i t h pleasure that w e welcome you t o the SGTSG Meeting

for 2003.

T h e C o n f e r e n c e w i l l p r o v i d e a challenging m i x of 'M

papers, w o r k s h o p s and field trips t h r o u g h stunning geological locations.

# ii

W i t h t h e a d d i t i o n of several social f u n c t i o n s , t h e Conference w e e k in this outstanding location promises t o be an exciting time. We look f o r w a r d t o meeting w i t h you during the Conference week. SGTSG C o m m i t t e e 2003 GEOLOGICAL SOCIETY OF AUSTRALIA Specialist Group : Tectonics and Structural Geology KALBARRI, 22-26 September 2003 http://www.sgtsg.gsa.org.au Conference Committee Steve Reddy - Chair Peter Cawood - Vice Chair Alan Collins - Secretary Ian Fitzsimons - Treasurer Ian lyler Anthony Gartrell Sandi Occhipinti Bregje Hulscher Peter Schaubs Myra Keep Roberto Weinberg

Sponsors The Conference Committee would like to thank the following sponsors:

Suggested Citation Reddy, S.M., Fitzsimons, I.C.W. & Collins, A.S. (Editors) 2003. SGTSG Field Meeting, Kalbarri, 22-26 September 2003, Geological Society of Australia Abstracts, 72, 173 p.

CSiiv

PETROLEUM EXPLORATION & MINING

Conference Organiser Promaco Conventions Pty Ltd ABN: 68 008 784 585 PO Box 890, Canning Bridge 6153 Western Australia Email: promaco@promaco.com.au Web: www.promaco.com.au Tel. (08) 9332 2900

Fax. (08) 9332 291 I

WESTERN AUSTRALIAN DIVISION

Maxwell Geosciences

Joumal ''^Virtual explorer

General Information

Pre Conference Workshop

Location of Sessions

Microstructural Analysis Using Electron Backscatter Diffraction (EBSD) David J. Prior (Liverpool University) & Steven M. Reddy (Curtin University)

All conference sessions from Monday 22 September to Friday 26 September will be held at the Zuytdorp Restaurant in Kalbarri.

Messages

A noticeboard will be in the registration area for messages to delegates. Please check this daily. Urgent messages may be telephoned through to the office of Promaco Conventions on (08) 9332 2900 for transfer to Kalbarri.

Mobile Telephones

As a courtesy to speakers, mobile telephones and pagers are to be turned off within the lecture rooms during all sessions.

Name Badges

Each attendee of the conference will be issued a name badge at registration. The badge is the official pass and must be worn at all times.

Program and Book of Abstracts

Additional copies of the Book of Abstracts may be purchased from the registration desk during the conference for $33.00 (incl.GST).

Registration Desk

The registration desk will be serviced from 8.00am to 4.00pm each day of the conference and until the conference close on Friday. Major credit cards will be accepted at the registration desk for payment of registration fees and functions.

Speaker Preparation

Speakers are reminded it is important to check their AV presentations prior to the scheduled time. The venue will be available during the breaks at tea, lunch or prior to sessions each day. A technician will be on hand to assist you.

Keynote Speakers Prof Ian Dalziel (University of Texas at Austin) Prof Mark Harrison (Australian National University) Prof Mike Sandiford (University of Melbourne) Dr Dave Prior (University of Liverpool) Prof Bruce Hobbs (CSIRO) Prof Brendan Murphy (St Francis Xavier University, Canada) Prof Chris Morley (University of Brunei) Prof Stephen Cox (Australian National University) Prof Tim Bell (James Cook University)

A short course providing an introduction to microstructural analysis using Electron Backscatter Diffraction (EBSD) on the Scanning Electron Microscope was held at the Cottesloe Beach Hotel, Cottesloe on Saturday 20 September 2003.

Social Program WELCOME RECEPTION Sunday 21 September 2003 After arrival in Kalbarri, registration followed by a light meal will be held at the Kalbarri Golf Club at 7:00pm. Drinks will be available for purchase at the bar. Cost: included for delegates and students additional tickets $20 Monday 22 September 2003 A poster session will be held at the Zuytdorp Restaurant at 6:00pm. Authors will be available for discussion. Drinks will be available for purchase at the bar. Tuesday 23 September 2003 It is said that you haven't been to Kalbarri if you haven't been to Findlays for a seafood barbecue. Tonight is the night for that indulgence. (At your own expense) Cost: $20 approx (food only) Wednesday 24 September 2003 At the return of the field trips, a poster session will be held at the Conference venue from 5:00pm to 6:30pm. Authors will be available for discussion. Drinks will be available for purchase at the bar. At the conclusion, delegates may transfer to the Kalbarri Golf Club for a barbecue dinner at 7:00pm. At 8:15pm, Dr Phil Playford will give a presentation entitled 'Carpet of Silver: the Wreck of the Zuytdorp in 1712\ Dinner cost: $20 (food only) CONFERENCE DINNER Thursday 25 September 2003 Enjoy a relaxing time at the Conference Dinner at the Kalbarri Golf Club from 7:30pm. Cost: $44 per person

Breakfasts / Lunches While the accommodation apartments have cooking facilities and there are cafes downtown, the Zuytdorp Restaurant will provide a full hot breakfast each day for $ 14.50. Lunch will be provided at the Zuytdorp each day and is Included in the registration fee.

Field Trips In keeping with the aims of the SGTSG, field trips are a fundamental part of the meeting. Field trips will be linked to IGCP project 453 - Modern and Ancient Orogens. Being located on the central west coast of WA, Kalbarri is well placed for exploring the Pinjarra and Capricorn Orogens. The pre- and post- conference field trips investigate the controversial tectonic evolution of these orogens. The mid-meeting field excursions will examine the spectacular geology of the Kalbarri region.

Pre-Conference Field Trip

FT I - Pinjarra Orogen: Leeuwin Complex and the Darling Mobile Belt). 17 to 21 September 2003 Four days were spent examining the best exposed outcrop of the Pinjarra Orogen, the Leeuwin Complex, in the far southwest of the state. The emphasis of the trip was on the processes associated with mesoscopic ductile deformation in the mid crust (folds, shear zones etc.) and the regional Proterozoic history of Western Australia. Leaders: Damien Janssen, Ian Fitzsimons and Alan Collins (Curtin University)

Mid-Conference Field Trips

Two mid-conference field trips are on offer on Wednesday 24 September to make the most of the spectacular Kalbarri region. You will not regret bringing your camera for these trips. FT2 - A Ramble Through the Red Beds: Unstructured Ordovician, Kalbarri Departs 9:00am. This trip offers stunning scenery and exposures of Palaeozoic sedimentary rocks of the Tumblagooda Formation, at the spectacular Z Bend in the Murchison River gorge and at Red Bluff on the coastal gorges south of Kalbarri. You will see fluvial and coastal depositional environments from a time when colonisation of land was just beginning, and a variety of trace fossils including large eurypterid trackways. Classic examples of several types of soft sediment deformation are also present in these superb sections. (Maximum participants = 35). Leader: Roger Hocking (GSWA). Departure 9.00am Cost: $40 FT3 - The Hardabut Fault: a Neogene or Pleistocene thrust localised by an older normal fault Departs 8:30am. The Hardabut Fault has moved as a thrust fault (dip ca 20°) along the line of a pre-existing major normal fault. The phase of compressive faulting and folding represented by this thrust is thought to have been responsible for uplift of the plateau into which the Murchison River gorge has been incised, probably during Neogene or Pleistocene. The excursion will visit the entrance to the gorge, where the Ordovician Tumblagooda Sandstone has been folded and thrust over the Proterozoic Northampton Complex. The evidence for other Neotectonic activity in the Shark Bay to Kalbarri area will also be discussed during the excursion. Leader: Phil Playford (GSWA). Departure 8.30am Cost: $40

Post-Conference Field Trip (NOW Fully Booked)

FT4 - Capricorn Orogen: From the Yilgarn to the Pilbara (26 September to 3 October).

The post-conference fieldtrip to the Proterozoic Capricorn Orogen will journey from the margin of the Archaean Yilgarn Craton 400 km to the north, through a spectacularly changing landscape, to the Archaean Pilbara Craton. The Capricorn Orogen is a Palaeoproterozoic to Neoproterozoic belt between the Pilbara and Yilgarn Cratons, which records several periods of collision, transpression, and extension between 2000-750 Ma. The field trip begins by studying one of the best exposed examples of crustal-scale Palaeoproterozoic Shear Zones in the world — The Errabiddy Shear Zone, a 200 km long shear zone that developed during the 2000-1950 Ma Glenburgh Orogeny, but has been re-activated through to the Neoproterozoic. From here, the northward trek will include looking at the best exposed outcrops in the Glenburgh Terrane, including heterogeneously deformed medium- to high-grade granitic gneisses. Here, impressive igneous relationships can be seen adjacent to felsic gneiss derived from the same material. In the central part of the Capricorn Orogen we will look at the enigmatic relationship between basement gneisses and overlying metasedimentary rocks. Travelling north, we will examine the tectonic setting and significance of the Palaeoproterozoic and Mesoproterozoic Ashburton and Edmund Basins. Finally, the trip will spend the last day in the unique banded iron formations of the Hamersley Group, deformed into spectacular folds, faults and shear zones by Palaeoproterozoic deformation. The dual emphases of the trip will be, firstly on processes of mesoscopic to macroscopic deformation through time, including ductile deformation in the mid-crust and brittle deformation in the upper crust, and secondly, on the tectonic evolution of the Yilgarn and Pilbara collision. Leaders: Sandra Occhipinti, Steve Reddy, Ian Tyler, Keith Sircombe, Dave Martin, Alan Thome and Steve Sheppard (Curtin University, GSWA and UWA) Trip details: Depart - early afternoon Friday 26 September Kalbarri conference venue. Field trip runs for 8 days from 26 September to 3 October, ending in the iron-ore mining town of Paraburdoo. All meals Friday 26 September to lunchtime Friday 3 October provided. Accommodation is camping due to the extreme remoteness of area (camping equipment provided, excluding sleeping bags). Participants will return to Perth via coach or fly at additional cost. Coach travel to Perth includes an extra 2 days. The coach will travel through the central part of Western Australia, taking a route different to that taken during the field trip stopping at Nallan Homestead in the remote region of Cue on Saturday 4 October. Participants will arrive in Perth on Sunday 5 October. Breakfasts will be provided on Saturday 4 October, and Sunday 5 October. Dinner will be provided on Saturday 4 October. Lunches can be purchased en route. Flights to Perth domestic airport will be with Qantas, departing Paraburdoo on the evening of Friday 3 October.

Accompanying Persons Kalbarri offers plenty of attractions for the visitor. The following tours are available which include spectacular gorges, breathtaking scenery, the mighty Murchison River and historic Murchison House. Bookings can be made at the Kalbarri Tourist Bureau on 9937 I 104.

T O U R I : LOOP and Z BEND GORGE TOUR 8.00am - 12.00 noon Tuesday, Thursday, Saturday & Sunday ADULTS $44 CHILDREN $26 ( 6 to 14 Years) CHILDREN Under 6 years , FREE T O U R 2: HERITAGE TOUR I.OOpm-5.00 pm Monday and Friday ADULTS $42 T O U R 3: MURCHISON RIVER CRUISE Morning Cruise includes Lunch, Morning Tea. DEPARTS daily at 10.00 am - ( 3.5 Hours). ADULTS $34 CHILDREN $13 ( 3 - 1 4 Years)

Location Mops

T O U R 4: WHALE WATCHING 3 Hours - Departs Wednesday at 9.00 am ADULTS $38 CHILDREN $25 (under 14 Years) T O U R 5: MURCHISON HOUSE STATION Tour the Station in a 4WD Charabanc (includes Morning Tea) Day tours 8.00 am- 12.00 noon Monday, Wednesday, Friday ADULTS $44 CHILDREN $27.50 (under 14 Years) Other alternatives are available through the Kalbarri Tourist Bureau - http://www.kalbamwaJnfo

Conference Program

N B : The presenting author is shown with the paper title. All authors are listed on the abstract. Abstracts are included in program order from page 17. Poster abstracts are listed in alphabetical order. An author index can be found at the back of the book.

Conference Program

1 KEYNOTE SPEAKER Liverpool University, David Prior UK 1 Steven Reddy Curtin University, Australia 1 Craig Buchan Curtin University, Australia 1 Aaron Stallard Univ. Canterbury, New Zealand iTom Blenkinsop James Cook University, Australia

Watching microstructural changes inside an SEM Application of Electron Backscatter Diffraction to natural high-strain zone deformation Characterizing complex deformation paths during transpression Metamorphic foliation revealed: grain-shape data from the Otago Schist, New Zealand Fractal characterization of grain boundaries: a new microstructural technique

[DISCUSSION I MORNING TEA 1 KEYNOTE SPEAKER James Cook University, Tim Bell ;QLD Allen Kennedy Curtin University, Australia Sandra McLaren Australian National University [Michael Cottam Royal Holloway, University London, UK Peter Crowhurst CSIRO Petroleum, Australia

The preservation of extensive datable deformation histories in rock and their radical implications for structural and metamorphic processes SHRIMP 5'®0: critical analytical parameters The Cooper Basin argon laboratory: nature's test of the K-feldspar MDD model From Gondwana to orogeny: integrated thermochronological constraint of AUS-PAC boundary evolution in New Zealand Late Miocene - Recent tectonic development of the northern Cordillera Real, Ecuador: new insights from (U-Th)/He thermochronology

•DISCUSSION CONFERENCE CLOSE

Poster Sessions Monday 22 September: POSTERS Hark Barley

University of Western Australia

Jurassic to Miocene magmatism in Myanmar and the tectonic evolution of SE Asia

Alan Collins

Curtin University, Australia

Detrital footprint of the Mozambique Ocean: U/Pb SHRIMP and Pb evaporation zircon geochronology of metasedimentary gneisses in eastern Madagascar

Michael Cottam

Royal Holloway, University London, U K

More than just the Alpine Fault: reconstructing the initial geometry of the AUS-PAC boundary through New Zealand

|or\athon Dale

Regional eclogite facies metamorphism of the Adula Nappe: implications for the tectonic evolution of the European central Alps | British Antarctic Survey, Defining tectono-stratigraphic terranes in the Antarctic Peninsula using lithospheric heterogeneities UK Paleozoic accretionary orogenesis in western Mongolia Curtin University, Australia Curtin University, Is Neoproterozoic tectonism in Dronning Maud Land, Mozambique and Sri Lanka a southern Australia extension of the East African Orogen or an eastern extension of the Zambezi Orogen?

^Irson Dean Arjan Dijkstra Ian Fttzslmom

University of Melbourne, Australia

Ian Fitzsinnons

Curtin University, Australia

Ben Goscombe

University of Adelaide, Australia University of Adelaide, Australia

Ben Goscombe

Does the late Neoproterozoic Darling Fault Zone of Western Australia extend all the way to the Transantarctic Mountains? Variation in metamorphic style along the northern margin of the Damara Orogen, Namibia The metamorphic response of transpressional orogenesis: the Kaoko Belt, Namibia

Ben Goscombe

University of Adelaide, Australia

Granulites of the Malawi mosaic

Ben Goscombe

University of Adelaide, Australia

Structural and metamorphic architecture of the east Nepal Himalayas

iDavici Gray

University of Melbourne, Australia University of Western Australia

Mechanism of Late Cretaceous fold-nappe emplacement, Oman and tectonic considerations Back to the source: SHRIMP U/Pb ages of zircons from the Itremo Group and Molo sequence of central Madagascar

University of Rennes, France Ohio University, USA

Structural and AMS evidences for the oblique opening of a Miocene dykes swarm associated with the rapid acceleration of the Pacific plate subduction under southern Patagonia (Chile) The Acatlan Complex, southern Mexico: record of the closure of the Rheic Ocean?

University of Western Australia Monash University, Australia

Global palaeogeography in latest Precambrian and Cambrian

Monash University, Australia Monash University, Australia

Modelling insights into the dynamics of subducting slabs in the upper mantle

Sregje Hulscher Jean-Pierre Lefort Damian Nance Sergei Pisarevsky Gideon Rosenbaum Wouter Schellart Wouter Schellart

Formation of arcuate orogenic belts in the western Mediterranean region

Episodic opening of back-arc basins: a case study from the southwest Pacific

NB: The presenting author is shown with the paper title. All authors are listed on the abstract. Abstracts are included in program order from page 17. Poster abstracts are listed in alphabetical order. An author index can be found at the back of the book.

Fracture systems in granite pavements of the eastern Pilbara Craton, Western Australia: indicators of neotectonic activity? The potential for paleoliquefaction studies to contribute to Australia's earthquake hazard map Multiple chronostratigraphic units in the Strangways Metamorphic Complex, central Australia Structure and age of the northern Leeuwin Complex, Western Australia: constraints from field mapping and U-Th-Pb isotopic and chemical analysis The Nackara Arc of South Australia, a fault-fold system of significant economic potential Ductile to brittle shear zones in the Wyangala Fault, Cowra, New South Wales A new activity-composition model for amphiboles in NCKFMASO: application to greenschist facies metamorphism in Archaean greenstones Gold bearing H2O-CO2 fluids: insights from mineral equilibria modelling in Archaean greenstones An integrated process model of the Mt Isa copper deposits, Queensland Modes of crustal extension determined by rheological layering Kinematics of syn-intrusive extension associated with the Sybella Batholith End-member boudin classification and modified boudin structures Electron microprobe dating of monazites from the western Gawler Craton A kinematic history of the Selwyn Fault, Mornington Peninsula, Victoria, Australia Tectonic cycles in the Strangways Metamorphic Complex, a history of Palaeoproterozoic convergence and extension A hot-plate tectonic model for Proterozoic crustal evolution in Australia Fluid-driven deformation processes and controls on gold deposition, Argo shear zone, St Ives Goldfield, Western Australia Structural and deformational controls on the development of the Bendigo gold-quartz vein system Tectonic evolution of the Weekeroo Inliers, Curnamona Province, South Australia Petrology of the Elagiri alkaline body of the Southern Granulite Terrane of the Indian Shield Anorthosite bodies of the Chhotanagpur gneissic terrane of the Eastern Indian Shield deciphering the style of early crustal evolution Reactivation history of the Great Boundary Fault, northwestern India Middle Ordovician to Early Silurian tectonic controls on genesis of world-class gold deposits at 440 Ma in the Lachlan Orogen Tectonic history from 3D geological modelling: Kanowna district. Western Australia

SGTSG Field Meeting 2003

MONDAY 22 SEPTEMBER 2003

SGTSG Field Meeting 2003

MOUNTAINS, MANTLE PLUMES, LITHOPHERIC GENERATION IN PANTHALASSA-TYPE OCEANS AND MODELS FOR THE ORIGIN OF SUPERCONTINENTS J, Brendan Murphv\ R. Damian Nance^ Andrew J. Hynes^ Stephen T. Johnston"*, and J. Duncan Keppie^ ^Dept. of Earth Sciences, St. Francis Xavier University. Antigonish, N.S. B2G 2W5, Canada, ^Dept. of Geological Sciences, Ohio University, Athens, Ohio 45701, U.S.A; ^Dept. of Earth and Planetary Sciences, McGill University, Montreal, P.O. H3A 2A7, Canada, "^School of Earth and Ocean Sciences, University of Victoria, B.C., Canada, V8W 3P6, ^institute de Geologia, Universidad Nacional Autonoma de Mexico, 04510 Mexico D.F., Mexico.

There are more than 40 hotspots beneath the modern oceanic crust and no modern ocean plate can be consumed without hotspots being overridden by continental crust. This implies that hotspots, and their underlying plumes, must play an important part in orogenic processes in the geologic past. The overriding of a hotspot and its related buoyant swell may profoundly change the tectonothermal expression of ongoing subduction at a continental margin. Recent geodynamic analyses have emphasized the relationship between modern flat slab subduction zones and the overriding of buoyant oceanic crust. Although most models for the evolution of the Late Mesozoic-Cenozoic Laramide orogeny in the western United States involve flat-slab subduction, the mechanisms proposed are controversial. The progressive overriding of the Yellowstone hot spot and its buoyant swell may have aided in the progressive eastward-directed shallowing of the subduction zone, that is commonly invoked to explain the time-transgressive eastward migration of deformation up to 1200 km into the continental interior and the cessation of magmatism and account for calculations which show that the Yellowstone plume would require between 42-89 million years to account for the mass deficit beneath the western Cordillera. According to plate reconstructions, the ancestral Yellowstone plume would have been beneath northern Nevada at ca. 35 Ma (the probable time of the Carlin gold deposition), and beneath the Kula plate prior to 55-50 Ma, at which time it would have collided with the continental margin. Evidence for the existence of the plume in the oceanic realm (i.e. prior to 55 Ma) is derived from Late Cretaceous basaltic terranes of the Coast Ranges of British Columbia and Washington, and from the Yukon territory. Some basaltic provinces of the Coast Ranges (such as

the 60-50 Ma Crescent terrane) were seamounts generated by the Yellowstone plume that accreted to western North America in the Eocene. In the Yukon territory, the ca. 70 Ma Carmacks basaltic volcanics have plume-type geochemistry. Paleomagnetic data indicate that the Crescent and Carmacks basalts were both erupted at paleolatitudes similar to that of the Yellowstone hotspot. The Carmacks basalts translated 17.2 + 6.5° to the north since its eruption but the Crescent volcanics translation northward is only 4.72° +/- 8.9°. An examination of the geological evolution of the Crescent terrane indicates that it was formed in a shallowing-upward Loihi-type oceanic setting culminating in the eruption of subaerial lavas. We combine estimates of the paleo-elevation of the Crescent volcanics and the age of the ocean floor onto which they were erupted to determine the excess elevation at the time of eruption. This excess elevation is then used to assess the characteristics of the plume thought to have fed these volcanics, and the characteristics are compared with those of the modern Yellowstone plume. Plate reconstructions indicate that the Crescent terrane was emplaced into ca. 20 Ma crust, and the presence of subaerial lavas implies an uplift due to the plume of ca. 4.2 km, which we use to calculate a minimum buoyancy flux of 1.1 Mg s \ similar to that of the modern Yellowstone plume. Published paleomagnetic data indicate that the Crescent terrane was formed at a paleolatitude similar to that of the Yellowstone hot spot. The Crescent seamount was accreted within 5 million years of the cessation of plume magmatism. Plate reconstructions indicate that it would have originated about 750 km to the west of the North American plate margin if it developed above a fixed Yellowstone plume, and are therefore consistent with the recorded very

SGTSG Field Meeting 2003

short interval between its formation and tectonic emplacement. We interpret the Crescent terrane as due to the ancestral Yellowstone plume. Such a plume would have generated an elongate swell and related plateau that would have been overridden by the North American margin. Taken together, the relationship between the plume, oceanic plateau and flat-slab subduction in Laramide times would have been analogous to the relationship between modern Andean flat-slab subduction zones and the Juan Fernandez and Nazca oceanic plateaus and their related hotspots. How can we detect the role of mantle plumes in the geologic record? The Laramide orogeny of the western United States is proposed as a modern analogue for the Silurian-Devonian Acadian orogeny and subsequent diachronous, voluminous, short-lived magmatism and basin

formation in the northern Appalachians. Shallowing of the Benioff zone accounts for several enigmatic features associated with plate convergence in the northern Appalachians including (1) Wenlockian-Ludlovian termination of arc-related magmatism in the Avalon terrane followed by a period of relative magmatic quiescence from 395 to 380 Ma and (2) diachronous migration of the Acadian deformation front from ca. 415 Ma in the southeast to ca. 370 Ma in the northwest, extending more than 600 km into the continental interior. The flattening of the subduction zone is attributed to overriding of a plume by the convergent margin, which may explain (1) the abrupt termination of magmatic quiescence by 380-370 Ma voluminous felsic magmatism and plume-related lamprophyres in the southeast (Meguma terrane) as the plume thermally eroded the oceanic lithosphere, causing melting of the lower crust; (2) Late Devonian regional high-7, low-P metamorphism in the Meguma terrane related to the thermal anomalies above a plume; (3) synchronous Devonian emplacement of Meguma gold deposits and associated siderophile elements possibly derived from fluid circulation above an ascending plume; (4) rapid Late Devonian uplift and erosion of as much as 10 km due to dynamic uplift over a plume; (5) migration of magmatism to the north (Avalon terrane: Cobequid Highlands) so that plumerelated Carboniferous magmatism occurred in and around the Carboniferous-Permian Maritimes basin; (6) the high-density lens at the base of the crust beneath the Maritimes basin as plume-derived underplated mafic rocks; and (7) a subsidence mechanism for formation of the Maritimes basin by cooling of a decapitated plume head.

In the Proterozoic, identification of the imprint of plume activity in mountain building is less clear. There is growing acceptance that global scale collisional orogenesis from ca. 1.2 to 1.0 Ga is related to the assembly of the supercontinent Rodinia. One of the corollaries of the existence of a supercontinent is the coeval presence of a Panthalassa-like "peri-Rodinian" ocean. By comparison with Panthalassa and the modern Pacific, such an ocean would have contained plume-related hotspots and underplated buoyant lithosphere in addition to ocean ridges, and ensimatic subduction zones, each generating juvenile crust broadly coeval with the life-span of Rodinia. Although most of this crust was subsequently destroyed by subduction, vestiges have been preserved in terranes that accreted to the leading edges of Rodinia following its breakup beginning at about 0.75 Ma. These terranes are characterized by having U-Pb crystallization ages and calculated Sm-Nd depleted mantle model ages that are coeval with the 1.2 to 0.75 Ga lifespan of Rodinia and are indicative of an origin as peri-Rodinian oceanic lithosphere. Subduction has consumed much of this lithosphere, but vestiges are preserved in Neoproterozoic collisional orogenic belts in West Africa (e.g., the Trans Saharan orogenic belt) and in Brazil (e.g., the Tocantins and Boborema provinces). Peri-Rodinian crust is also preserved in accretionary orogens such as the Arabian Shield and as recycled crust if subduction zones become re-established along continental margin after accretion. Such recycling is exemplified by the Avalonian belt of Atlantic Canada, which was located along the northern Gondwanan margin in the late Neoproterozoic. Avalonia is dominated by 0.630.57 Ga arc-related igneous activity, but has ca. 1.0 Ga Sm-Nd depleted mantle model ages, consistent with formation of proto-crust within the peri-Rodinian ocean followed by recycling during Neoproterozoic subduction (Murphy et al., 2000). Identification of lithosphere generation in Panthalassa type oceans has fundamental implications for geodynamic models for the origin of supercontinents. In recent years, two endmember models for the formation of supercontinents have emerged. In the classical Wilson cycle, oceanic crust generated during supercontinent breakup (the "interior ocean") is consumed during amalgamation so that the supercontinent turns "inside in" (intraversion). Alternatively, following supercontinent breakup, the exterior margins of the dispersing continental fragments collide during assembly so that the supercontinent turns "outside in" (extraversion). These end-member models can be distinguished by comparing the Sm-Nd crustal formation ages

SGTSG Field Meeting 2003

of accreted mafic complexes in the collisional orogens of supercontinent assembly with the breakup age of the previous supercontinent. For supercontinents generated by intraversion, these crustal formation ages postdate rifting of the previous supercontinent. For supercontinents generated by extraversion, the oceanic lithosphere consumed during assembly pre-dates breakup of the previous supercontinent so that crustal formation ages of accreted mafic complexes are older than that of rifting. In the Paleozoic Appalachian-Caledonide-Variscan orogen, a key collisional orogen in the assembly of Pangea, crust formation ages of accretionary mafic complexes post-date the formation of the lapetus Ocean (i.e. are < ca. 0.6 Ga) suggesting

supercontinent assembly by intraversion. By contrast, the Neoproterozoic East African and Brasiliano orogens, which resulted in the amalgamation of Gondwana, are characterized by mafic complexes with crustal formation ages (ca. 0.75 to 1.2 Ga) that predate the ca. 750 Ma breakup of Rodinia. Hence, these complexes must have formed from lithosphere in the ocean that surrounded Rodinia, implying that this ocean was consumed during the amalgamation of Gondwana. These data indicate that Pangea and Gondwana were formed by intraversion and extraversion, respectively, implying that supercontinents can be assembled by fundamentally distinct geodynamic processes.

SGTSG Field Meeting 2003

OCEANIC EXTENSIONAL DETACHMENT SYSTEMS AT MODERN RIDGES AND WITHIN OPHIOLITES Arian H. Diikstra^ and Peter A. Cawood^ ^Department of Applied Geology, Curtin University, GPO Box U1987, Perth, WA 6845 Australia, a.dijkstra@curtin.edu.au ^Tectonics Special Research Centre, Curtin University, GPO Box U1987, Perth, WA 6845, Australia

It is becoming increasingly clear that detachment faults play an important role in sea floor spreading at modern mid-ocean ridges, in particular in settings in which magma supply to the ridge system is low or episodic in character, and is unable to fully accommodate extension. Such settings include slow and intermediate spreading rate ridge systems and/or neartransform, inside-corner environments. In such settings, gabbros and peridotites are often exposed at the ocean floor and it has been suggested that oceanic detachments are in part responsible for the exhumation of such rocks from deep crustal or mantle levels. Examples of well-documented extensional detachments at active ridges include a striated gabbro/peridotite massif in the vicinity of the Fifteen-Twenty Fracture Zone at the Mid-Atlantic Ridge (MacLeod et al., 2002) and a gabbro massif at Atlantis Bank (Southwest Indian Ridge - Dick et al., 2000). Interestingly, in both cases the observed detachment systems seem to sole out at relatively shallow, mid-crustal levels. We present geological observations on oceanic detachment systems within oceanic lithosphere recently formed at a slow-spreading ridge that is exposed on Macquarie Island (Southern Ocean), and within ophiolites derived from slow-to-intermediate spreading rate ridge systems (e.g., the Bay of Islands Ophiolite, Newfoundland and the Othris Peridotite Massif, Central Greece). On Macquarie Island, extensional deformation is localised in gabbroic dykes and intrusions within ultramafic rocks immediately below the crust-mantle boundary (the Eagle Bay shear zone system). The shear zone system is cut by undeformed dykes, which are related to the development of a sheeted dyke complex within pre-existing gabbroic crust.

Deformation occurred under amphibolite facies conditions. Smaller, mainly greenschist facies shear zones are exposed at shallower levels within the crustal section on Macquarie Island. In the Bay of Islands Ophiolite, a flat-lying extensional detachment is exposed in the Lewis Hills (the Carol Mountain shear zone), within mantle rocks and lower crustal cumulates (Suhr & Cawood, 2001). This shear zone system was mainly active under granulite facies conditions. In the Othris Peridotite Massif, a fragment of oceanic mantle lithosphere formed at a slowspreading ridge, highly deformed fine-grained peridotite tectonites have recorded spreadingrelated deformation predating and coeval with abundant melt impregnation (Dijkstra et al., 2001). Our observations show that evidence exists for extensional detachment systems at deep crustal and mantle levels at slow and intermediate spreading rate oceanic ridge systems. We discuss whether the currently available observations support geometrical tectonic models of single detachment faults that cut through the entire oceanic lithosphere at spreading ridges (the so-called 'oceanic corecomplex' models), or whether extensional deformation at ocean ridge systems is accommodated by a series of low-angle detachments that sole at different, rheologically weak, levels within the oceanic lithosphere. References: Dick et al., EPSL 179, 31-51, 2000; Dijkstra et al., J. Petrology 42, 5-24; MacLeod et al.. Geology 30, 879-882, 2002; Suhr & Cawood, GSA Bulletin 113, 1025-1038, 2001.

SGTSG Field Meeting 2003

EXTENSION ALONG THE AUSTRALIAN-PACIFIC TRANSPRESSIONAL TRANSFORM PLATE BOUNDARY NEAR MACQUARIE ISLAND N a t h a n R, Daczko^'^ (ndaczko@mail.utexas.edu), K a r a h L. Wertz^'^ (karah@mail.utexas.edu), S h a r o n Mosher^ (mosher@mail.utexas.edu), M i l l a r d F. Coffin^'^ (mcoffin@ori.u-tokvo.ac.ip). T i p MeckeP'^ (tip@mail.utexas.edu) ^Department of Geological Sciences, University of Texas at Austin ^Institute for Geophysics, University of Texas at Austin ^Ocean Research Institute, University of Tokyo and Institute for Frontier Research on Earth Evolution, Japan Marine Science and Technology Centre

The Australian-Pacific transform plate boundary fault zone along the Macquarie and McDougall segments of the Macquarie Ridge Complex, south of New Zealand, is characterized by dominantly normal faults and pull-apart basins, in apparent conflict with the regional transpressional tectonic setting. We propose that present day curvature of the transform is inherited from a preexisting divergent plate boundary and that the overall extensional kinematics shown by faults along the main plate boundary trace and exposed on Macquarie Island result from local stresses related to right steps in the plate boundary faults and not to the current transpressional setting. Transpression along the Australian-Pacific transform plate boundary has resulted in uplift of the -1500 km long Macquarie Ridge Complex. Macquarie Island, the only exposure of the complex, sits atop a 5 km high, 50 km wide submarine ridge of oceanic crust and lies --4.5 km east of the major active plate boundary fault zone. Thus, Macquarie Island and the surrounding sea floor provide a unique opportunity to study an active oceanic transform fault using complimentary datasets that include marine geophysics and land based geology. Mapping of recent faults affecting the topography of Macquarie Island shows the island is extensively cut by high angle normal faults forming pull-apart basins. Furthermore, evidence for reverse motion is rare. Using marine geophysical data, including bathymetry, reflectivity and seismic reflection data, collected along the Australian-Pacific plate boundary to the north and south of the island, we have defined a 5-15 km wide plate boundary zone. Within this zone a series of en echelon faults step to the right along the main plate boundary trace. At the

right stepping fault terminations, elongate depressions (up to 10 km wide and 1.2 km deep) parallel the plate boundary. We interpret the depressions as responding to local stresses located at extensional relay zones or pull-apart basins. Conclusion The Macquarie and McDougall segments of the Australian-Pacific transform plate boundary is comprised of a series of major en echelon, right lateral faults that step to the right, producing extensional relay zones and pull-apart basins between fault tips where displacement is transferred. Subsidiary faults increase in number in the transfer zones and form smaller pull-apart basins accommodating local extensional stresses. The zone of associated deformation is 5-15 km wide. Recent faulting on Macquarie Island reflects dominantly extensional tectonics, apparently in conflict with the island's transpressional tectonic setting and history of uplift. However, we infer that the island is located within an extensional local stress field related to right steps in the plate boundary located -4.5 km to the west, and fault geometries and kinematics are not related to current transpression at the plate boundary. We propose that transpression at the plate boundary is partitioned into a strike-slip component parallel to the boundary, forming the en echelon strike slip faults and a convergent component causing the anomalous bathymetric ridge and trough morphology of the McDougall and Macquarie segments of the MRC.

SGTSG Field Meeting 2003

THE NORFOLK RIDGE AND D'ENTRECASTEAUX OROCLINE: MODERN ANALOGUE FOR RIBBON CONTINENTS AND TERRANE WRECKS? Stephen T. Johnston School of Earth & Ocean Sciences, University of Victoria, PO Box 3055 STN CSC, Victoria, British Columbia, Canada V8W 3P6 stj@uvic.ca

The origin of map-view bends of orogenic belts and arcs remains enigmatic. Here I summarize geological evidence indicating that a bend of the northern end of a ribbon continent extending north from the Northland Peninsula, New Zealand through New Caledonia and the Loyalty Islands and into the submarine d'Entrecasteaux ridge (the NNNCd'E ribbon continent) is an orocline that has formed as a result of oroclinal orogeny (buckling about vertical axes of rotation due to pinning of the leading edge of a migrating lithospheric beam). An analogue model is used to investigate the relationship between orocline development, the rotation of the Vanuatu-New Hebrides arc and the origin of the North Fiji basin. The NNNCd'E ribbon continent terminates to the northeast in the Vanuatu-New Hebrides arc. The asymmetric, triangular, northwest-tapering North Fiji lies northeast of (behind) the arc. Paleomagnetic, geological and GPS data imply that the Vanuatu-New Hebrides arc has rotated --60 clockwise since 12 Ma about a pole of rotation located at its northwest end, opening the North Fiji basin. The analogue model demonstrates that the arc was forced to rotate clockwise due to its southward advance being impeded by the NNNCd'E ribbon continent.

In this model the ribbon continent originated as a linear, north-trending feature that ended in the arc, just west of its mid-point. Southward advance of the arc buckled the ribbon continent, giving rise to the orocline. Buckling forced the arc to rotate clockwise, opening the asymmetric North Fiji basin. Orocline formation has involved tearing of the ribbon continent lithosphere from the oceanic lithosphere that bounds it on both sides, allowing sub-slab DUPAL anomaly mantle to seep into and contaminate the adjacent Vanuatu-New Hebrides magmatic arc. The lithospheric-scale of buckling is consistent with orocline formation controlling the clockwise rotation of the Vanuatu-New Hebrides arc and formation of the North Fiji basin. Buckling of the NNNCd'E ribbon continent against the migrating Vanuatu-New Hebrides arc explains the curved ribbon continent, the clockwise rotation of the arc and the origin of the North Fiji basin. This ongoing oroclinal orogeny provides us with an opportunity to further understand the processes responsible for and involved in the buckling of lithospheric beams, and to refine interpretations of ancient. orogens that are thought to be the products of oroclinal orogeny.

SGTSG Field Meeting 2003

UTILISING SPATIO-TEMPORAL CONSTRAINTS IN TECTONIC RECONSTRUCTIONS OF OROGENS: AN EXAMPLE FROM THE CENTRAL MEDITERRANEAN G. R o s e n b a u m a n d G.S. L i s t e r Australian Crustal Research Centre, School of Geosciences, Monash University, Melbourne (gideon@nnail.earth.monash.edu.au)

The central Mediterranean region is located between the converging plates of Africa and Europe. Its eastern part consists of an Early Mesozoic oceanic lithosphere (the Ionian Sea), whereas the western part, the Tyrrhenian Sea, is a young extensional basin that formed since the Late Miocene (ca. 10 Ma). The Tyrrhenian Sea consists of an attenuated continental crust, and locally, of Pliocene-Pleistocene oceanic crust. The eastern and the western margins of the Tyrrhenian Sea are marked by an arcuate orogenic belt, comprising of the Apennines in the Italian peninsula, the Calabrian arc in southern Italy and the Maghrebides in Sicily. The Apennine-Maghrebide belt has been subjected to orogenic processes associated with subduction of a west-dipping lithospheric slab, simultaneously with back-arc extension in the Tyrrhenian Sea. Deformation in the TyrrhenianApennine system followed an earlier deformational event that took place during the Late-Oligocene-Early Miocene and involved the opening of the Ligurian-Provengal Basin as a result of the rollback of a northwest-dipping subduction zone. The opening of the LigurianProvengal Basin was accompanied by a counterclockwise rotation of the Corsica-Sardinia microplate, which led to a progressive collision of Corsica-Sardinia with the former western margin of Adria, and gave rise to the formation of a north-south striking orogen in the Apennines. Based on palaeomagnetic results, it has been suggested that Corsica and Sardinia stopped rotating at ca. 16 Ma, which coincides with the time of termination of back-arc extension in the Ligurian-Provengal basin. We analysed spatial and temporal constraints that can link the geometry of the subducting slab with the history of subduction rollback and backarc extension in the Tyrrhenian-Apennine region. Two methods have been used in order to constrain crustal deformation processes: (1) analysis of spatio-temporal data related to thrusting as inferred from structural studies and from the depositional ages of foreland basins; and (2) spatio-temporal analysis of syn-rift deposits from the floor of the Tyrrhenian Sea and

from back-arc basins within the ApennineMaghrebide belt. These constraints were combined with constraints on the role of subduction rollback derived from the distribution of orogen ic-related (calc-alkaline) magmatic centres and from the present-day geometry of the subducting slab as Inferred from tomographic images. The constraints provided a preliminary reconstruction model, which was then refined by additional constraints based on palaeomagnetic and palaeogeographic data. Results of our analysis suggest that the major stages of opening in the Tyrrhenian Sea did not begin before the Late Miocene (ca. 13-10 Ma). Deformation in the back-arc region involved two episodes of back-arc extension induced by the rollback of a west-dipping subducting slab. The first period of extension (10-6 Ma) was prominent in the northern Tyrrhenian and in the western part of the southern Tyrrhenian. The second period of extension, which began in the latest Messinian (6-5 Ma), mainly affected the southern Tyrrhenian, and was accompanied by extreme rates of subduction rollback, in excess of 60-100 km/Myr. The transition between the two stages of opening at about 6-5 Ma, was triggered by docking of the Internal Carbonate Platform in the central Apennine subduction system, which we propose, led to the formation of a slab tear. Subsequently, the remaining Ionian slab beneath Calabria was narrower, and was consequently subjected to accelerated rates of subduction rollback. Our results show that spatio-temporal constraints can be used to provide geologically sound reconstructions of orogenic belts. In the example of the Tyrrhenian-Apennine system, it provided predication for the timing of tectonic events, such as the docking of the Internal Carbonate Platform onto the orogenic edifice. It also enabled us to better understand the tectonic responses caused by tearing of a retreating slab.

SGTSG Field Meeting 2003

ZIRCON AGE CONSTRAINTS ON ARC MAGMA SOURCES AND TECTONIC EVOLUTION IN JAVA, INDONESIA P J H a m i l t o n \ P K i n n y ^ H S m y t h ^ a n d R Hall^ ^ CSIRO Petroleum, PO Box 1130, Bentley, West Australia, 6102, Australia, joe.hamilton@csiro.au ^ Department of Applied Geology and Tectonics Special Research Centre, Curtin University of Technology, Hayman Road, Bentley, Western Australia, 6102, Australia. ^ SE Asia Research Group, Royal Holloway University of London, Egham, Surrey, TW20 OEX, UK

Radiometric dating methods are being applied to rock samples from East Java in order to define volcanic arc evolution and associated regional uplift and cooling histories in a precise temporal framework. We report here results from SHRIMP U-Pb dating of zircons, undertaken with the additional objectives of ascertaining the age and provenance of crustal contributions to arc magmatism and of the sediments in the East Java Basins. Two tuffaceous sediments have yielded rather scattered isotopic data. For a Sambipitu Formation sample, scatter results from there being two distinct populations defining ages at 12.8±0.2 and 18.6±0.2Ma. Two distinct eruptive events have been sampled in these sediments. One explanation could be the assimilation by the younger event of older igneous rocks with subsequent sedimentary redistribution as tuffaceous sediment. Alternatively, there could have been sedimentary mixing of the younger volcaniclstics with those derived from the older event. For a Semilir Formation sample, individual zircon dates range from 19Ma to 31 Ma (17 grains) reflecting a variable sedimentary and/or volcanic sampling of zircons crystallised at different times over a period of --12Ma of igneous activity in the same region. Zircon dates indicative of intrusion ages were obtained for three samples and range from ~25Ma (1 grain, two spot analyses) for a diorite from Nanggulan to 9.3±0.2Ma (20 grains) for a diorite from Ponorogo and to 5.7±0.5 Ma (22 grains) for a diorite body at Trenggelek. However, the Nanggulan diorite together with a granodiorite from Jember and volcanic breccia clasts from Turen also yielded ranges of very old zircon ages. For these three samples 16 of a total of 41 zircon ages were between 500 and 600 Ma. The remaining 25 ages are > 600 Ma and range up to 3200 Ma. Each sample exhibited a similar distribution of ages with a dominance of "Pan-African" ages but with a total range extending to greater than 2000Ma. The presence of these old inherited zircons has implications for

the sources of melts beneath the volcanic arc. There are three possible sources for this inheritance; • continental sediment on the plate being subducted beneath Java during the Paleogene, • continental crust included in the basement beneath the Paleogene arc, or • a continental fragment subducted beneath Java. If a fragment of crust was subducted below or accreted to Java it would imply the existence of a microcontinental fragment which originated from Gondwana, and was separated by Mesozoic rifting of the Gondwana margin. Magma rising from the subduction zone could intrude this continental sliver if accreted to Java and assimilate the zircons prior to eruption. Alternatively, if the subducting slab included a microcontinental fragment, its partial melting could release old zircons into the rising magma. Although there is no evidence for the existence of pre-Cretaceous crust in East and Central Java, tectonic reconstructions would allow for this possibility. In conclusion these data have provided; • The most precise radiometric dates yet available to contribute to the establishment of a temporal framework of arc development in Java. • The first indications of continental crust ranging in age from about 500Ma to 3200 Ma has contributed to arc magmas in East Java. • Precise constraints for establishing cooling curves for arc intrusives in combination with other mineral chronometers (hornblende Ar-Ar. AFTA and apatite UTh/He) with different blocking temperatures. This in turn has important implications for sediment supply rates, provenance, hydrocarbon reservoir quality and basin evolution and architecture in East Java. • Precise radiometric dates that may be used fro improve Tertiary plate reconstructions of SE Asia.

SGTSG Field Meeting 2003

EARTH'S ACCRETIONARY OROGENS: MECHANISMS OF CRUSTAL GROWTH AND OROGENESIS Peter Cawood Tectonics Special Research Centre, Curtin University, GPO Box U1987, Perth, Australia, p.cawood@info.curtin.edu.au, Fax: +61-8-9266-7972

Classic models of orogens involve a Wilson cycle of ocean opening and closing with orogenesis related to continent-continent collision. Such models fail to explain the geological history of a significant number of orogenic belts throughout the world in which deformation, metamorphism and crustal growth took place in an environment of on-going plate convergence. These belts are termed accretionary orogens but have also been refereed to as non-collisional orogens, Pacifictype orogens, Turkic-type and exterior orogens. Accretionary orogens evolve in generally curvilinear belts comprising dominantly mafic to silicic igneous rocks and their sedimentary products and accumulated largely in marine settings. They are variably deformed and metamorphosed by tectono-thermal events aligned parallel to, and punctuating, fades trends. Accretionary orogens form at sites of subduction of oceanic lithosphere and consist of magmatic arcs systems along with material accreted from the downgoing plate and eroded from the upper plate. Deformational features include structures formed in extension and compressive environments during steady-state convergence (arc/backarc vs. accretionary prism) that are overprinted by short regional

compressive orogenic events. Orogenesis takes place through coupling across the plate boundary with strain concentrated in zones of mechanical and thermal weakening such as the magmatic arc and back arc region. Potential driving mechanisms for coupling include accretion of buoyant lithosphere (terrane accretion), flat slab subduction, and rapid absolute upper plate motion over-riding the downgoing plate. The Circum-Pacific region provides outstanding examples of accretionary orogens. The Pacific formed during breakup of Rodinia in the Neoproterozoic and has never subsequently closed, resulting in a series of overall ocean-ward younging orogenic systems that have always faced an open ocean, yet have been the sites of repeated tectono-thermal events and continental growth. Accretionary orogens have been active throughout Earth history. They have been responsible for major growth of the continental lithosphere through the addition of juvenile magmatic products and include Archean greenstone belts, the Paleoproterozoic Birimian orogen (W. Africa), the Arabian-Nubian shield (Pan African) and Paleozoic orogens in Asia.

SGTSG Field Meeting 2003 OPHIOLITES AND ACCRETED TURBIDITE FANS: TECTONIC PROCESSES AND CONTINENTAL GROWTH THROUGH TIME IN CONVERGENT MARGIN SETTINGS 1

David Gray , David Foster , Robert Gregory , Catherine Spaggiari , Ben Goscombe

1 ^School of Earth Sciences, University of Melbourne, Melbourne, 3010, Victoria. ^School of Geological Sciences, University of Florida, Gainesville, FL, 32611-2120, USA. 4

Department of Geological Sciences, Southern Methodist University, Dallas, TX, 75275, USA.

^Department of Applied Geology, Curtin University of Applied Technology, Perth, 6845, W.A. School of Earth and Environmental Sciences, Adelaide University, South Australia, 5005, Australia. The architecture, structural character and structural relationships within the crust of ancient orogenic belts, as well as the relative juxtaposition of tectonic elements record tectonic processes related to growth of continental crust. Examples from the Neoproterozoic Pan African Damaran Orogen of SW Africa, the Palaeozoic Tasman Orogen of E Australia, and the Mesozoic Otago Schist Belt of New Zealand provide a disconnected history of tectonic process through time and illustrate continental growth from marine successions involving subduction-accretion. Lithologies include basalt, chert and/or turbidite in the low-grade belts (e.g. Lachlan Orogen) or monotonous quartzo-feldspathic schist, with varying amounts of micaceous schist, greenschist and metachert (e.g. Southern Zone, Damara Orogen and Otago Schist belt, NZ). The preservation of, as well as the nature of the preserved oceanic lithosphere (Tethyan or Oman type Cordilleran or Lachlan Orogen type) to a first order approximation depends on tectonic setting (forearc, backarc, continental margin), the age of oceanic lithosphere (young or old), and the thickness of the turbidite fan. Tethyan ophiolites form in sediment-starved settings with pelagic sequences, and involve underthrusting of formerly passive margin continental crust (e.g. Oman). Whereas the Makran, Otago, Lachlan and Franciscan Complex (western USA) have large volumes of turbidite sediments and involve subsequent structural interleaving of slices of oceanic lithosphere. Such large sediment volumes tend to clog up the associated subduction zone resulting in marked deformation within the thrust-wedge (accretionary prism).

versus

The turbidite package shows stratal disruption and classic melange/ broken formation of accretionary prisms (e.g. Shimanto, Japan and Kodiak complex, Alaska), chevron folding (e.g. Lachlan) and/or fold nappes and schistosity (e.g. Shimanto complex, Japan; Southern Zone, Damara Orogen; Otago Schist belt, NZ). Deformation of the turbidite package depends on the tectonic setting (forearc, backarc, continental margin), the thickness of the turbidite fan, the residence time of the turbidites on the seafloor prior to deformation, and the preserved level within the subduction accretion system. Boundary conditions for the orogens studied are: • thick sediment fill, old lithosphere, back arc setting: LACHLAN OROGEN, EASTERN AUSTRALIA • thick sediment fill, old lithosphere, forearc setting: OTAGO SCHIST, NEW ZEALAND • thick sediment fill, forearc setting: DAMARA OROGEN, NAMIBIA • thin pelagic fill, young lithosphere and a passive continental margin transition to a subduction zone setting: OMAN Conclusions: Convergent margin tectonic settings involving large accumulations of turbidites show a similarity in style and nature of orogen development at least since the NeoProterozoic. These settings represent important sites of growth of new continental crust, with the newly accreted crust consisting of a supra-crustal recycled layer (turbidites) underlain by mafic oceanic crust, the real addition to the continental crust.

SGTSG Field Meeting 2003

ANTARCTIC PENINSULA TERRANE ORIGINS AND TRANSPORT KINEMATICS: IMPLICATIONS FOR THE LATE PALAEOZOIC AND MESOZOIC EVOLUTION OF THE PACIFIC MARGIN OF GONDWANA. Alan P.M. Vaughan\ Ian L. Millar^, John D. Bradshaw^ Rudolph A.J. Trouw"^ ^British Antarctic Survey, High Cross, Madingley Rd, Cambridge, CB3 OET, U. K. ^NERC Isotope Geosciences Laboratory, Keyworth, Nottingham, NG12 5GG. ^Department of Geological Sciences, University of Canterbury, Christchurch, N.Z. "^Departamento de Geologia, I. Geo, UFRJ, 21910-900, Rio de Janeiro, Brazil, e-mail: a.vaughan@bas.ac.uk

INTRODUCTION Recent structural and geochronological data from the Antarctic Peninsula suggest that it consists of several exotic blocks; one of these has affinities with south-east Australia; another contains conglomerates derived from an as yet unidentified Palaeozoic magmatic province. Overall terrane transport kinematics has implications for the Palaeozoic-Mesozoic evolution of the Palaeo-Pacific margin of Gondwana.

EXOTIC TERRANES The large-scale kinematics of the midCretaceous, Palmer Land orogenic event and palaeontological affinities of an invertebrate fauna at Mount King on Alexander Island suggest that a Permo-Carboniferous component of the eastern part of the LeMay Group accretionary complex may have originated in south-east Australia. The View Point submarine fan conglomerate in the northern Antarctic Peninsula, is dominated by Ordovician to Carboniferous granites and volcanic clasts, and has zircon provenance ages no younger than Carboniferous, suggesting that this may represent glaciomarine deposits or reworked glacial tillite with a currently unidentified Palaeozoic source. A 3.1 Ga granite clast indicates reworking of ancient cratonic material within this province. Cretaceous rocks of the LeMay Group on Charcot Island have Hf-isotope signatures that are unlike anything in adjacent Triassic LeMay Group rocks or in the nearby

magmatic arc rocks of the Antarctic Peninsula Central Domain.

KINEMATIC DATA Kinematic data from the southern Antarctic Peninsula and plate reconstructions for the southern Pacific suggest that dextral shear was dominant during the early Cretaceous bringing terranes from modern-day south-west. This is consistent with evidence for south-east Australian invertebrate faunas on Alexander Island. Hf-data from the Charcot Island LeMay Group rocks are also consistent with a source to modern-day south-west. In the northern Antarctic Peninsula, however, kinematic data from accretionary complex rocks suggest that shear was sinistral in the late Triassic and early Jurassic (also seen in western Chile at this time). This would have brought accretionary complex rocks bearing granite clasts from modern-day north, suggesting a possible African or South American origin.

CONCLUSIONS Overall, taking into consideration data from Australia, New Zealand and both South and North America, terrane movement appears to have been from the southern to the northern hemisphere throughout the late Palaeozoic and Mesozoic, but with perhaps a brief reversal of terrane movement in the late Triassic and early Jurassic.

SGTSG Field Meeting 2003

STRUCTURAL GEOLOGY AND GEOCHRONOLOGY OF THE YIDUN ARC, EASTERN TIBET: A RECORD OF TRIASSIC CONVERGENCE AND EXHUMATION Reid, A,J„ Wilson, C J . L and Phillips, D. ^School of Earth Sciences, University of Melbourne, Victoria, 3010, Australia.

The Yidun Arc is located between the Songpan Garze Fold Belt and Qiangtang (North Tibetan) Block in the eastern Tibetan Plateau. Despite its location in the modern day plateau, the tectonic history of the Yidun Arc centres around the Permo-Triassic closure of the Palaeotethys Ocean. This pre-Cenozoic structural and thermal history of the Yidun Arc has been investigated through field structural observations and "^^Ar/^^Ar dating. The western Yidun Arc exposes Palaeozoic metasediments upon which the Triassic volcanic arc was built. Deformation intensity within these metasediments reach a maximum in a major high-strain zone along the western margin of the Yidun Arc, interpreted to be a major westdirected thrust and termed the Jinsha Thrust Zone. The Jinsha Thrust is also the location of a series of ophiolite outcrops that define the Jinsha River Suture. The emplacement of both foliated and unfoliated granitic bodies into the metasediments of the Jinsha Thrust in the Early Triassic, places a tight constraint on the age of ductile deformation and therefore the timing of suturing of the Yidun Arc with the Qiangtang (North Tibetan) Block. Middle to Upper Triassic flysch and intercalated arc volcanics unconformably overlie and are in faulted contact with the Palaeozoic sequences of the western Yidun Arc. Towards the base of this sequence, conglomerate horizons containing abundant mafic and ultramafic clasts, suggesting the melange of the Jinsha River Suture was exhumed and eroded during the Middle Triassic. During the late Triassic these overlying sequences were deformed and intruded by large granitic batholiths.

"^^Ar/^^Ar dating of hornblende, muscovite and biotite separates from the high-grade zone of the Jinsha Thrust was undertaken to constrain the timing of cooling within the thrust. These preliminary results show hornblende and muscovite ages cluster around -223 Ma, while biotite yields a younger age '-204Ma. A muscovite sample from structurally overlying mica-schists yields a concordant age of --220 Ma. These data suggest rapid cooling from 550°C to below 400°C at -223-220 Ma, residence at temperatures in excess of biotite closure for -1520 million years followed by a renewed phase of cooling. A muscovite sample from west of the high-grade zone shows an older age -240Ma, while another sample to the east gives a younger age -215Ma. The cooling history of the Jinsha Thrust shows stepwise cooling related to Early-Middle Triassic progressive suturing of the Yidun Arc and the Qiangtang Block, followed by a subsequent phase of Late Triassic exhumation. The reactivation of this major structure during the Late Triassic is interpreted to have occurred in response to regional stresses due to the closure of the greater Palaeotethys Ocean at this time and the final suturing of the Qiangtang Block to the southern margin of Eurasia.

SGTSG Field Meeting 2003

FROM A NEOPROTEROZOIC ACTIVE MARGIN SETTING TO A PALEOZOIC SHELF SEQUENCE: A GEODYNAMIC HISTORY OF THE CADOMIAN-VARISCAN BASEMENT SLIVERS IN CENTRAL EUROPE (GERMANY, BOHEMIAN MASSIF) Ulf L i n n e m a n n \ Neal J . M c N a u g h t o n ^ , R o l f L. R o m e r ^ M i c h a e l G e h m l i c h ^ K e r s t i n D r o s t ^ C h r i s t i a n Tonk® ^Staatliche Naturhistorische Sammlungen Dresden, Museum fur Mineralogie und Geologie, Konigsbrucker Landstr. 159, D-01109 Dresden, Germany, uif.nnnemann@snsd.smwk.sachsen.de ^Centre for Global Metallogeny, School of Earth and Geographic Sciences, University of Western Australia, Nedlands, Western Australia 6009, Australia, nmcnauqh@aeol.uwa.edu.au ^GeoForschungsZentrum Potsdam, Telegrafenberg, D-14473, romer@gfz-potsdam.de "^Staatliche Naturhistorische Sammlungen Dresden, Museum fur Mineralogie und Geologie, Konigsbrucker Landstr. 159, D-01109 Dresden, Germany, kerstin.drost@snsd.smwk.sachsen.de ^Staatliche Naturhistorische Sammlungen Dresden, Museum fur Mineralogie und Geologie, Konigsbrucker Landstr. 159, D-01109 Dresden, Germany, mmg@snsd.de ^Staatliche Naturhistorische Sammlungen Dresden, Museum fur Mineralogie und Geologie, Konigsbrucker Landstr. 159, D-01109 Dresden, Germany, mmg@snsd.de

Neoproterozoic rocks in the Saxo-Thuringian part of Armorica were overprinted during Cadomian orogenic processes at the northern margin of Gondwana. The Cambro-Ordovician overstep sequence in Saxo-Thuringia most probably was deposited in a failed-rift setting that reflects the separation of Avalonia and other terranes from Gondwana mainland. Silurian to Early Carboniferous shelf sediments of Saxo-Thuringia were deposited at the southern passive margin of the Rheic Ocean. SHRIMP U/Pb geochronology on detrital and inherited zircon grains from preVariscan basement rocks of the northern part of the Bohemian Massif (Saxo-Thuringia, Germany) demonstrate a distinct West African provenance for sediments and magmatic rocks in this part of the Armorican Microplate. Nd-isotope data of

Late Neoproterozoic to Early Carboniferous sedimentary rocks show no change in sediment provenance from the Neoproterozoic to the Lower Carboniferous, which implies that SaxoThuringia never left its West African source until the Variscan Orogeny leading to the Lower Carboniferous configuration of Pangea. Hence, large parts of the Armorican Microplate may have never left Africa in pre-Pangean times, and Armorica simply represents a remnant of a "Greater Africa" in Gondwanan Europe. The separation of Armorica from the Gondwana mainland and a long drift during the Paleozoic is not supported by the presented data.

SGTSG Field Meeting 2003

CHANGING TECTONIC STYLE THROUGH TIME IN THE SOUTHERN APENNINES FOLD AND THRUST BELT (ITALY): GEOLOGICAL AND GEOPHYSICAL CONSTRAINTS FROM THE LUCANIA OILFIELDS AREA Stefano Mazzoli\ Peter Shiner^ Vacuity of Environmental Sciences, University of Urbino, Campus Scientifico Sogesta, 61029 Urbino (PU), Italy; e-mail: s.mazzoli@geo.uniurb.it ^Enterprise Oil Italiana, Via Due Macelli 66, 00185 Rome, Italy

INTRODUCTION For many years, fold-thrust belts have been commonly interpreted as 'thin-skinned' structures, developed above a detachment with the underlying basement remaining undeformed. However, in the last few years, models of basement fault reactivation have been increasingly applied. Both 'thin-skinned' and deep-rooted, inversion-dominated deformation appear to be involved in building fold-thrust complexes in the Apennine belt of southern Italy. DATA The southern Apennine belt consists at outcrop of folded and E-transported pelagic basin and carbonate platform successions of MesozoicPalaeogene age, and unconformably overlying Miocene-Quaternary foredeep and wedge-top basin deposits. Well data in the Lucania region demonstrate large-scale thin-skinned thrusting of the units above, with 57 km of horizontal displacement since earliest Pliocene times. This includes the shortening that ramps up through the buried Apulian platform carbonates, which form the reservoir unit in several major oil fields in the Lucania area. The detachment between the shallow allochthon and the buried carbonate platform unit is marked by a 'melange zone' several hundreds of metres thick, which has been penetrated by numerous oil wells. As a result of

overpressuring, gas shows are common throughout this interval. Traps are formed in the buried Apulian carbonates by fold structures developed in the hanging wall of major reverse faults. The density of data associated with oilfield development in the Lucania area provides an opportunity to constrain structural models at a variety of scales. Good quality seismic data image the front of the main (Monte Alpi) field and show involvement of the Base Apulian seismic reflection in compressional structures. Locally well-defined fault plane reflectors image the moderate-steeply dipping, convex-up reverse faults that bound the structures. Integration of well, VSR and seismic data confirm the dominance within the reservoir of moderate-high angle reverse and normal faulting. RESULTS AND CONCLUSION Taken together the elements above suggest that a thick-skinned structural model is most appropriate for the deep Apulian structures. The overlying overpressured 'melange zone' most probably played a primary role in the construction of the orogen, allowing a relatively fast (12.5 mm/y) emplacement of the detached thin-skinned units onto a yet essentially undeformed Apulian foreland succession. The latter was later cut by relatively lowdisplacement, thick-skinned structures of Middle Pliocene-Early Pleistocene age.

SGTSG Field Meeting 2003

DID LOWER CRUSTAL CHANNEL FLOW CREATE THE HIMALAYA? Mark Harrison Research School of Earth Sciences, The Australian National University

It has long been suspected that ductile flow of molten crust plays a significant role in orogenyV Bird^ proposed that a fluid lower crust could be laterally extruded from underneath high topography via planar channel (Poiseuille) flow. More recently, Beaumont et al.^ investigated the coupling of channel flow to focused surface denudation and proposed that the crystalline core of the Himalaya was created by this process. This view - that the Greater Himalayan Crystallines (GHC) represent extruded Tibetan middle crust has been widely adopted"^'^ ®. Despite the growing popularity of this idea, I am unaware of any evidence that actually supports such a model and note several facts that preclude it outright. The heightened interest in the channel flow model stems from bright spot anomalies beneath the Yadong-Gulu rift, S Tibet, being interpreted as evidence that anatexis is actively occurring at >15 km depth, and that the region between the Main Central Thrust and South Tibetan Detachment is an earlier extruded equivalent^. This interpretation is problematic as: 1) the bright spots appear to have the seismic properties of an aqueous fluid^ 2) the seismic velocity structure beneath S Tibet from broadband studies indicates generally cold crust^; 3) geologic and geochemical observations of the uplifted Yadong-Gulu rift flank reveal no evidence of shallow anatexis but instead record a flux of juvenile heat and mass throughout the Tertiary^®; and 4) ^He/'^He ratios in hydrothermal fluids sampled directly above the bright spots contain a mantle-derived component^ ^ suggesting their origin from mafic magmas in the mid-crust. Evidence advanced to support the extrusion model in the form of magneto-telluric and geomagnetic data are equally consistent with the above four facts. Putting aside objections to the underlying conjecture (i.e., widespread shallow anatexis), we note that conditions sufficient for dehydration melting (>750°C at 15 km) are required by this interpretation. As the seismogenic Moho is <750X beneath S Tibet^^, this constraint implies that the region between 15 to 80 km depth is essentially isothermal. Numerical modeling has shown that a thermally mature continental collision zone can evolve a near-isothermal structure where the mid-crust has been extruded under rapid denudation along a narrow range front^. If the model were to apply to the shallow anatexis hypothesis^ then the GHC should show evidence of being: /) a ductiley extruded melange, 11) extruded through the Gangdese arc, ill)

geochemically equivalent to Lhasa Block crust, and iv) extruded through a narrow (<50 km) Himalayan aperture. However, the GHC is welldocumented to be comprised of the following stratigraphy that can be traced the length of the Himalaya^^: Fm. I, metapsammitic gneisses form the basal unit; Fm. II is directly above and dominated by calc-silicates; Fm. Ill is a ca. 490 Ma augen gneiss sheet. The persistence of this coherent layering across the Himalaya is inconsistent with the particle paths predicted by the modeP (i.e., requirement / not met). Hundreds of detrital zircon U-Pb dates from the GHC^"^-^' reveal a continuum of ages from 550 Ma to >3 Ga, but none characteristic of Gangdese arc magmatism (140-40 Ma) (i.e., requirement 11 not met). Isotopic characteristics of the GHC strongly contrast with that inferred for the Lhasa block (e.g., 8Nd= -17 vs. -7)^° (i.e., requirement ill not met). Lastly, the GHC exhibits a 100-150-km-long nappe phase along >70% of its length^^ (i.e., requirement iv not met). I conclude that there is as yet no evidence that specifically supports the shallow anatexis model. Rather, it appears that local high heat flow and hydrothermal activity in S Tibet are largely driven by episodic emplacement of mantle-derived magmas. Thus the underlying assumption for appealing to the coupled channel flow/rapid denudation model is without foundation and several facts appear to rule out its relevance to the formation of the Himalaya. 1. Judges 5:5, King James Version of the Bible; 2. Bird, 1991, JGR, 96, 10275; 3. Beaumont et al., 2001, Nature 414, 738; 4. Searle et al., 1997, J. Geo/. 105, 295; 5. Hodges et al., 2001, Tectonics, 20, 799; 6. Grujic et al., 2002, EPSL 198, 177; 7. Nelson et al., 1996, Science 274, 1684; 8. Makovsky and Klemperer, 1999, JGR 104, 10,795; 9. Owens and Zandt, 1997, Nature 387, 37; 10. D'Andrea et al., 1999, EOS 80, 1785; 11. Hoke et al., 2000, EPSL 180, 297; 12. Ruppel and McNamara, 1997, EOS 78, 535; 13. Le Fort, 1996, The Tectonics of Asia (A. Yin and T.M. Harrison, eds.), CUP, 95; 14. DeCelles et al., 2000, Science 288, 497; 15. Myrow et al., 2003, EPSL 212, 433.

SGTSG Field Meeting 2003

INTEGRATED LOW TEMPERATURE THERMOCHRONOLOGY: A TALE OF TWO TERRANES Kohn, B. P.J Gleadow, A. J. W.,^ Carter, T. J., ^ Lorencak, M.,^ Belton, D. X.^ and Foster, D.A ^ ^School of Earth Sciences, The University of Melbourne, Parkville, Victoria 3010 ^Dept. of Geological Sciences, University of Florida, Gainesville, FL 32611, USA b.kohn@unimelb.edu.au

The recent development of apatite (U-Th)/He (AHe) thermochronometry offers an exciting and unparalleled opportunity to provide thermal history information (between MO-SOX) characteristic of the shallower depths of the crust. AHe thermochronometry therefore complements the apatite fission track (AFT) method (most sensitive between -70-110°C) in that it is most responsive over the temperature range where the AFT system is least sensitive. We report the results of studies from a rapidly-cooled and a slowly-cooled terrane, to illustrate how integrated modeling of low temperature thermal histories using the two methods can lead to new insights into upper crustal and surface processes. Metamorphic core complexes within the highly extended Colorado River extensional corridor in the southwestern USA are the result of mid-Tertiary extension achieved along low-angle normal faults. Previous AFT analyses on these complexes were to a degree hindered by relatively low uranium concentrations and young ages, giving rise to relatively large uncertainties in age determinations. Despite these limitations, first-order estimates of fault slip rate, dip of fault while active, timing of extension and palaeogeothermal gradients were reported. The potentially higher resolution of the AHe data provides a means to test these different aspects of late-stage cooling in core complexes. The Harcuvar Mts, Arizona, display continuously over -60 km a broadly domed footwall exposure of a low-angle fault, which has been exhumed with a top-to-the-northeast sense of shear. Rocks at the northeast end of the range resided at depths of 1020 km before extension, whereas the southwestern end was structurally shallower. Hence, AFT and AHe ages record cooling due to exhumation along the low-angle fault, because the pre-extension temperatures would have been well above closure temperatures for both these systems. All AHe analyses yield ages within error of the corresponding AFT ages, but they reveal a more detailed picture of the exhumation history. Previous AFT data from the Harcuvar Mts yield an average detachment fault slip rate estimate of -7.7 km/Myr. However, rather than a single slip rate for

movement along the detachment, the AHe data suggest a more complex two stage cooling history. This involves an initial slip rate of -3.7 km/Myr (-2015 Ma), followed by apparent accelerated slip of -17 km/Myr (-15-13 Ma). The latter cooling phase is related to rapid slip on the detachment combined with thermal relaxation of hot footwall rocks that were at nearly uniform shallow depths beneath the distended hanging wall due to isostatic rebound. In cratonic settings, where Phanerozoic stratigraphic or structural information is commonly scarce or missing, the dual analytical approach can potentially elucidate upper crustal t-T paths. Such paths are more tightly constrained when, in addition to outcrop sampling, deep borehole samples are also available. AFT results from the southern Canadian Shield, including a 3.44 km deep borehole in Sudbury, reveal the spatial distribution of burial under foreland sediments during late Ordovician to Silurian and late Permian to early Triassic time following the Taconic and Alleghanian Orogenies respectively to the east. However, the difficulty in replicating AHe ages from rocks which have experienced slow cooling and prolonged residence in the He partial retention zone, presents a potential problem. Such problems may arise from factors such as variable grain size, chemical variations of grain composition and the presence of U and Th-rich micro-inclusions. Where AHe ages from the Shield can be replicated their interpretation involves a modelling strategy, which takes into account the thermal history experienced by the samples. Our results show that parts of southeastern Ontario had cooled to temperatures <-40°C by late Jurassic time, whereas samples deep in the Sudbury borehole also record a 'weak' late Cretaceous heating event not recorded by AHe data from surface samples. In the Shield environment, a combined AFT and AHe approach is essential for revealing an integrated low temperature Phanerozoic thermal history, which would not otherwise have been possible by using either method alone.

SGTSG Field Meeting 2003

THE FINAL AMALGAMATION OF GONDWANALAND: CONSTRAINTS FROM CENTRAL MADAGASCAR Breqje Hulscher ^ and Ian C.W. Fitzsimons ^ 1 TSRC (Tectonics Special Research Centre), Dep. of Geology and Geophysics, The University of Western Australia, 35 Stirling Hwy, Crawley, WA 6009, Australia (e-mail: bhulscher@tsrc.uwa.edu.au) 2 TSRC, Dep. of Applied Geology, Curtin University of Technology, GPO Box U1987, Perth, WA 6845

From as early as the 1960's, geologists have recognized two major thermal events in the late Neoproterozoic to Cambrian history of East Africa. The periods from -750 to 620 and -580 to 510 Ma were both characterized by extensive magmatism, metamorphism and deformation and are thought to relate to the closure of the Mozambique Ocean, resulting in the formation of the East African Orogen and the amalgamation of Gondwanaland. There are several schools of thought as to which period actually represents the final closure of the Mozambique Ocean. One model envisages an early closure by 620 Ma, involving collision of the 'African' Tanzania and Congo Cratons with various other fragments. In this early closure model, tectonism between 580 and 510 Ma is considered post-collisional, and is generally interpreted in terms of orogenic collapse. In the second model, the Mozambique Ocean did not close until -550 Ma, and tectonism before 620 Ma is explained in different ways, including the early accretion of continental fragments before final ocean closure, or magmatic underplating in an active continental margin setting. In order to resolve these ambiguities, we need to locate all suture zones and establish their age of closure. However, lack both of outcrop and geochronological data makes this problematic, and another approach is to determine the style of tectonism at 550 Ma. In the early closure model, magmatism and deformation between 580 and 510 Ma should be driven largely by extension. In contrast, this time period should be dominated by shortening related to convergence in the model for final amalgamation at 550 Ma. These two end-member models also have different explanations for the widespread exhumation of the East African orogen during the 580 to 510 Ma time frame. In the early closure model, post-orogenic extensional collapse is the main driving mechanism for this exhumation, whilst in the late ocean closure model the exhumation resulted from convergence related reverse movements leading to uplift and erosion. The geology of Madagascar has been interpreted to support either model for

Gondwanaland amalgamation. In an attempt to provide further constraints, we have built a tectonic framework for Central Madagascar by integrating structural field and satellite imagery studies with a targeted geochronology and geochemistry program. Most of Central Madagascar comprises the Mesoproterozoic supracrustal Itremo Group and the granitic -2.5 Ga basement of the Antananarivo Block. Both are intruded by two main pulses of granitoids at 820 to 720 Ma and 550 to 510 Ma. The older pulse is regarded as arc magmatism unrelated to the final amalgamation of Gondwanaland, and we focus here on the younger events. We find no evidence of magmatic activity related to the East African Orogeny before -580 Ma. The main phase of east-west shortening in Central Madagascar had started by 550 Ma and involved mostly thin-skinned folding and thrusting of autochthonous cover and basement, leading to tectonic interslicing at cover-basement contacts. Around 540 Ma, possibly signifying a lock-up stage, strain partitioning increases and deformation in the cover rocks is dominated by sinistral transpression. This is associated with vertical movements between blocks bound by NW-SE and N-S trending strike-slip zones, possibly similar to the style of deformation responsible for exhumation in southern Madagascar. This sinistral transpression continued until -530 Ma, when lateral escape could no longer be accommodated. Thus, Central Madagascar provides evidence for continued convergence between -550 and -530 Ma, rather than for a post-collisional extensional setting at this time. Along the length of the suture zone(s) of Gondwanaland, oblique convergence was accommodated during this period by horizontal shortening coupled with significant, dominantly sinistral, strike-slip. Therefore, we favour the model for late stage closure of the Mozambique Ocean but do not exclude a multiphase assembly, possibly involving the accretion of continental fragments other than Madagascar to East Africa by 620 Ma.

SGTSG Field Meeting 2003

DOES THE SOUTHERN CONTINUATION OF THE EAST AFRICAN OROGEN INTO EAST ANTARCTICA REPRESENT AN AREA OF TECTONIC ESCAPE? Joachim Jacobs FB Geowissenschaften, Universitat Bremen, PF 330440, 28334 Bremen, Germany jojacobs@uni-bremen.de

The collision of parts of E- and W-Gondwana during Late Neoproterozoic/Early Paleozoic times led to one of Earth's largest known orogens, the East African-Antarctic Orogen (EAAO) and to the amalgamation of major parts of Gondwana. The EAAO has a combined strike length of more than 8000 km and a width of up to 2000 km. In Dronning Maud Land, East Antarctica, the EAAO has overprinted 1.1 Ga metamorphic rocks at high-grade conditions. The western orogenic front of the EAAO is a large dextral transpression zone, represented by the Heimefront Shear Zone and can be traced by aerogeophysical means under the ice. Along this shear zone, characteristic high-amplitude and elongate Grenville-age magnetic anomalies of the Maud Belt terminate sharply immediately to the W. The Heimefront shear zone also coincides with Late Mesoproterozoic Ar-Ar and K-Ar mineral ages to the W and Paleozoic Ar-Ar and K-Ar mineral ages to the E of the shear zone. The central part of the EAAO is characterised by polyphase Late Neoproterozoic/Early Paleozoic deformation and metamorphism and large volumes of late to postorogenic A2-type magmatism. The suture between parts of E- and W-Gondwana has not been unequivocally identified in this southern part of the EAAO, however, large structures under the ice point to a number of potential sites for the main suture. Large parts of the EAAO seem to have evolved within an overall sinistral transpressional setting. However, the southern extension of the EAAO is characterised by both sinistral and dextral tectonism as well as orogenic collapse with associated late-tectonic magmatism in a pull-apart-type setting. This observation points to a possible S-directed escape tectonic

regime of the EAAO. Further arguments for such a S-directed break-out are: •

Southward fanning-out of major, very elongate shear zones, similar to the situation in SE Asia today

•

Intense reactivation of the western foreland, with large amounts of differential denudational unroofing and possibly crustal thinning

•

Arrested blocks of not-overprinted crust in the southern-central portion of the EAAO (the Coats Land block), that escaped overprint by southern break-out.

•

During the break-up of Gondwana, the southern portion of the EAAO is characterised by a collage of microplates. These microplates could indicate shear zone bound blocks that were shaped during the Sdirected escape tectonic regime.

This interpretation has important implications for Gondwana reconstructions. Some reconstructions use Mesoproterozoic structures as starting points to e.g. reconstruct southern Africa and East Antarctica. If this area however was affected by large scale transcurrent shearing during Late Neoproterozoic/Early Paleozoic times, continental reconstructions based on Mesoproterozoic structures would be obsolete. This model could be tested by a careful reconstruction of this part of Gondwana solely based on the step-by-step subtraction of sea floor magnetic anomalies.

SGTSG Field Meeting 2003

NEOPROTEROZOIC SUPERCONTINENT, SUPERPLUME, TRUE POLAR WANDER, AND THE SNOWBALL EARTH: WERE THEY INTERRELATED? Li, Z . X P \ E v a n s ,

Z h a n g , S.

(3)

^^^ Tectonics Special Research Centre, The University of Western Australia, Crawley, WA 1090, Australia (zli@tsrc.uwa.edu.au: Fax: +61-8-93801037) ^^^ Department of Geology & Geophysics, Yale University, Yale University, P.O. Box 208109, New Haven CT 06520-8109, USA (dai.evans@yale.edu) ^^^ School of Earth and Land resources, China University of Geosciences, Beijing 10083, China (shzhang@cuqb.edu.cn)

The Neoproterozoic time (1000-545 Ma) recorded a number of major, and often enigmatic, events in the Earth's history: the assembly of the supercontinent Rodinia at around 1000 Ma, possibly superplume events, and the development of equatorial glaciations at sea level which led to the speculation of snowball Earth events. Were there any causal relationships between these events? We have recently obtained new palaeomagnetic results from the Xiaofeng dyke swarm in central Yangtze craton, South China, which we dated at 802 ± 10 Ma using the SHRIMP U-Pb zircon method. Together with new understanding regarding the timing of the final Rodinia assembly, the timing and location of superplume events, and existing palaeomagnetic data, this enabled us to re-examine the configurations of Rodinia and possible causal relationships between the various Neoproterozoic global events. Our new palaeomagnetic pole places South China at between ca. 55° and 70'' palaeolatitude, in par with high palaeolatitude indicated by a similar-aged pole for India. Although no reliable palaeomagnetic data of that age are available for Australia, a reported pole from the ca. 820 Ma (?) Wooltana Volcanics would place Australia in a high-latitude position too. However, coeval poles from Laurentia indicate a low-latitude position that would separate Laurentia significantly from the high-latitude continents. If we ignore the ca. 820-800 Ma poles from Laurentia which are not regarded as of high quality, and use the palaeopoles of that age from the other continents, we can reconstruct a Rodinia similar to that of Li et al. (1999, EPSL

173, 171-181), with South China between western Laurentia and eastern Australia, and one end of Rodinia occupying a polar position at that time. However, such configuration would require Rodinia to break up by ca. 750 Ma. Alternatively, by matching the paired ca. 820-800 Ma and ca. 750 Ma poles from South China, India, Australia (ca. 820 Ma pole unreliable) and Congo, as well as the ca. 725 Ma pole from Laurentia, we are able to reconstruct a Rodinia with South China adjacent to northern India and western Australia. Interestingly, this reconstruction would require a rapid 90° rotation of Rodinia around an equatorial axis between ca. 800 Ma and ca. 750 Ma, a hallmark of true polar wander. We suggest that the initiation of 825-800 Ma mantle plumes under the polar end of Rodinia (South China, India, Australia and southern Africa), as part of the Rodinian superplume, triggered the TPW event, which brought the entire supercontinent to an equatorial position. This sudden increase of emerged land area at the equator increased both the CO2 drawdown (through enhanced weathering, particularly plume magmatism erupted during 830-745 Ma) and the amount of solar energy reflected away by the supercontinent, which together led to low-latitude glaciation in Sturtian time. Following recovery from the Sturtian ice age(s), high sea-levels accompanied by thermal substance of postRodinia passive margins led to greenhouse conditions which persisted until the Marinoan ice age.

SGTSG Field Meeting 2003

RECONSTRUCTING RODINIA: IS THERE ANY LIGHT AT THE END OF THE TUNNEL? Ian W. D. Dalziel Tectonics Special Research Centre, Institute for Geophysics, Jackson School of Geosciences, University of Texas at Austin, USA.

The Tectonics Special Research Centre was funded to determine the supercontinents of which Australia has been a part over the past 3.0 Ga. The Centre came into being in the aftermath of proposals that Laurentia broke out of the Rodinia supercontinent, defined as the supercontinent that amalgamated with a global 'Grenvillian' orogenic event at the end of the Mesoproterozoic and in fragmenting 'begat' all subsequent continents. The early 1990's conception of Rodinia was that the present cordilleran margin of Laurentia was adjacent to East Gondwana (ie Antarctic and Australian) cratons, and that the present AppalachianCaledonian margin was adjacent to Baltica and West Gondwana (ie South American and African) cratons . Opening of the Pacific Ocean basin in the Neoproterozoic was suggested to have been accompanied by closure of a major Mozambique ocean basin, with East and West Gondwana cratons suturing along the East African orogen. Subsequent work has emphasized that neither East nor West Gondwana was amalgamated until the end of 'Pan African' orogenesis into the Cambrian, and many alternative Rodinia reconstructions have been proposed in the past decade on both geologic and paleomagnetic grounds. In my view these fail to consider either the budget of late Precambrian rifted margins or compatability with the timing and geometry of well established Cambrian Gondwanaland amalgamation and end-Paleozoic Pangea assembly. Nonetheless no 'smoking gun' has emerged to point towards any particular

reconstruction as preferrable. This is true even as other developments , most notably the emergence of the 'snowball Earth' hypothesis , have made Neoproterozoic-early Paleozoic paleogeography especially important for the understanding of the evolution of the Earth system. I will review three lines of evidence that I regard as the most promising. First there is the use of lead isotopic 'signatures,' which appear to rule out the possibility that the Arequipa massif of proto-Andean South America is a former part of Laurentia while pointing to collision of the protoAppalachian margin of Laurentia with Amazonia in Grenvillian orogenesis. Second, the identical timing of the Keeweenawan magmatic event in Laurentia and the Umkondo of the Kalahari craton is best explained by those two cratons being continuous, or colliding, at ^1.1 Ga. Third, the presence of 'Pan African' detrital zircons in sedimentary strata at the base of the lower Paleozoic section of the Argentine Precordillera emphasize the importance of that terrane as the best clue to elucidating pre-Pangea paleogeography. It appears to have been derived from the Gondwanaland margin at the end of the Precambrian, faunally isolated as part of Laurentia during the Cambrian and Early Ordovician, and reunited with Gondwanaland in the mid- to Late Ordovician.

SGTSG Field Meeting 2003

CONTRASTING FEATURES OF THE TRANS-HUDSON (1.91.8 GA) AND GRENVILLE (1.2-1.0 GA) OROGENS IN CANADA: DO THEY REFLECT AN EVOLUTION OF THE LITHOSPHERE THROUGH TIME? David Corrigan Geological Survey of Canada, 615 Booth Street, Ottawa, Ontario, Canada, K1A 0E9

There has been much debate on the thermal and mechanical evolution of the lithosphere from the Archean to the Proterozoic. The most commonly held view suggests that cooling, thickening and compositional differentiaton of the lithosphere from the earliest Archean to the MesoProterozoic occurred gradually, with modern-type Wilson-cycle tectonic styles not appearing before ca. 2.0 b.y. ago. Within the North American Craton, the Trans-Hudson (1.90-1.78 Ga) and Grenville (1.20-1.00 Ga) orogens provide useful snapshots and insights on orogenic styles and evolving parameters spanning over ca. 900 million years. Although both orogens share similar characteristics such as overall length and breadth, they are marked by fundamental differences. In general, the Trans-Hudson Orogen is characterized by medium- to low-pressure and medium-temperature conditions, abundance of crustal-derived melts and absence of significant extensional shear zones. In contrast, the Grenville Orogen is characterized by widespread medium- to high-pressures and temperatures, numerous large anorthositic and associated anhydrous plutonic suites, as well as an

abundance of major, syn-collisional extensional faults. Although some of the noted differences in overall metamorphic grades can be attributed to variable physical parameters such as duration of continent-continent collision and absolute amount of convergence, most of the features can be better explained by a temporal variance of parameters including crustal and lithospheric thickness, geothermal gradient, and radiogenic heat production. The extreme width coupled with moderate metamorphic grades observed throughout the Trans-Hudson Orogen is compatible with an orogenic system with relatively thinner and warmer lithosphere. In contrast, metamorphic features observed in the Grenville Orogen are compatible with a cooler, more rheologically competent lithosphere. The postulated presence of a cooler, denser, and hence more gravitationally unstable lithospheric mantle during the Grenvillian Orogeny may also have played a role in the generation of 'plateaulike' regions of uplift and the generation and emplacement of anorthositic suites.

SGTSG Field Meeting 2003

THE 2.0 GA USAGARAN ECLOGUES, TANZANIA: AGE AND KINEMATIC HISTORY OF THE OLDEST HP/LT TERRANE ON EARTH Alan S, Collin$\ Steven M. Reddy\ Craig Buchan^ and Abdul Mruma^ ^ Tectonics Special Research Centre, Department of Applied Geology, Curtin University, GPO Box U1987, Perth, Australia. ^Department of Geology, University of Dar es Salaam, Tanzania. Corresponding author: a.s.collins@curtin.edu.au. Fax: +61-8-9266-3153

Eclogite facies rocks metamorphosed at low/medium temperatures and high pressures require an anomalously low geothermal gradient to form. Such rocks are usually interpreted to have formed by the subduction of relatively cold continental or oceanic crust followed by rapid exhumation. The paucity of eclogite facies rocks in the Archaean and Palaeoproterozoic suggests that either eclogite-facies rocks did not form at subduction zones in the way they do today, or simply that they aren't preserved either because of different exhumation rates and processes in the early Earth, or because of subsequent orogenic reworking. In this study we present new structural and U/Pb SHRIMP data from the world's oldest subduction related eclogite, the 2.0 Ga Usagaran eclogites of Tanzania. The Usagaran Belt of Tanzania lies to the east, and is thrust over, the Archaean Tanzania Craton in central Tanzania. The orogen is made up of a greenschist-facies series of Proterozoic metasedimentary and metavolcanic rocks (the Konse Group; that have detrital zircons indicative of a Tanzanian craton source concordant grains with ages of 25502750 Ma) and a structurally higher, amphibolitegrade series of orthogneisses, kyanite+garnet paragneisses and amphibolites that preserve relicts of medium temperature eclogite-facies metamorphism (Isimani Suite). The eclogite facies mafic relics have published chemical signatures suggestive of a MORB protolith and are imbricated within a >10km thick package of felsic gneisses that are extensively deformed by non-coaxial flow. U/Pb SHRIMP dating of zircons from two of these felsic gneisses yield concordant zircons with a range of 207p^/206p^ ages from 2489±16 Ma to 2908±5 Ma. The zircon ages match the complexity of the rocks, with locally preserved migmatitic textures completely transposed in higher-strain regions by non-coaxial flow. Sub-spherical, low-U, near concordant zircons from a mafic eclogite yielded a weighted mean 207p^/206p^ age of 1982±28 Ma, which is interpreted as the age of eclogite-facies metamorphism. This is confirmed by more precise

age of 1998±2 Ma from low Th/U rims that mantle zircons from a leucocratic gneiss. Deformation in the Isimani Suite can be broadly subdivided into five events. The first of these (D^), is strongly overprinted by a pervasive deformation (D2) at amphibolite facies conditions, which resulted in the accumulation of high strains throughout all of the exposed Isimani rocks. The geometry of foliations and lineations developed during D2 deformation are variable and have different shear directions that enable five D2 domains to be identified. Analysis of these domains indicates a geometrical and kinematic pattern that is interpreted to have formed by strain and kinematic partitioning during sinistral transpression. Euhedral prismatic zircons and high Th/U rims from a pegmatite that crosscuts the D2 foliation yield a age of 1991 ±2 Ma that is interpreted as the time of pegmatite crystallisation. The pervasive high strain deformation is therefore dated to between 2001 (age of eclogite-facies metamorphism) and 1989 Ma at maximum 1a error. Subsequent greenschist facies deformation, localised as shear zones on boundaries separating D2 domains have both contractional and extensional geometries that indicate reactivation of the Isimani Suite. isotopic ratios from muscovites suggest that this reactivation and associated greenschist-facies metamorphism was a result of the Neoproterozoic East African Orogen. The Usagaran eclogite-facies mafic and pelitic rocks are interlayered with felsic orthogneisses that, based on recent published U/Pb age data, extend throughout much of eastern Tanzania as the protoliths to the Neoproterozoic East African orogen. We propose that the Usagaran eclogites mark the suture between this terrane and the Tanzania craton and are, to date, the earliest known representation of subduction-zone metamorphism.

SGTSG Field Meeting 2003

LEWISIAN TERRANES Pete Kinnv\ Gary Love^ and Clark Friend^ ^Tectonics Special Research Centre, Curtin University of Technology, Perth, W.A. pk@lithos.curtin.edu.au h h e now defunct Department of Geology, Oxford Brookes University, Headington, 0X3 OBP, U.K.

The Lewisian Gneiss Complex of northwest Scotland traditionally has been viewed as a contiguous, exhumed block of lower crust which experienced two major 'orogenic cycles': the late Archaean Scourian and the Palaeoproterozoic Laxfordian episodes. However, detailed SHRIMP U-Pb geochronology has now shown that the complex is composed of at least ten distinct tectonostratigraphic terranes, each with its own unique initial accretionary and subsequent metamorphic history. These terranes were assembled into their present configuration during the Palaeoproterozoic, via a series of collisional events and lateral movements upon shear zones. Exposures of the complex on the Scottish mainland, west of the Moine Thrust, are now subdivided from north to south as follows: The Rhiconich Terrane, occupying the region north of Loch Laxford, consists largely of 2840 to 2800 Ma TTG suite gneisses, cut by basaltic dykes and c. 1855 Ma granite-pegmatite sheets. Amphibolite fades metamorphism at c. 1740 Ma marks docking with the Assynt Terrane to the south along the Laxford Shear Zone (LSZ). The Assynt Terrane, occupying the region south of the LSZ as far as Loch Inver, consists of 3030 to 2960 Ma TTG suite gneisses metamorphosed to granulite fades at c. 2490 Ma (the Badcallianl prior to the intrusion of the Scourie dyke swarm. Amphibolite fades metamorphism, in common with the Rhiconich Terrane, occurred at 1740 Ma. The Gruinard Terrane consists of 2860 to 2825 Ma TTG gneisses south of the Strathan Line shear zone south of Loch Inver, through to the Gruinard Belt shear zone where Proterozoic units of the Gairloch Terrane occur, and probably south beyond the Tollie Antiform. Experienced granulite facies metamorphism at c. 2730 Ma, later docking with the Assynt Terrane post-2490 Ma, with associated 'Inverian' retrogression. The Gairloch Terrane comprises juvenile Palaeoproterozoic volcanic, metasedimentary and intrusive rocks in the region between the Gruinard Belt and Loch Shieldaig. Indudes the Loch Maree Group and c. 1905 Ma Ard Gneiss. Later metamorphosed to amphibolite facies. The lalltaig Terrane, a small, shear-bounded block in the vicinity of Loch Shieldaig, contains the c.

2000 Ma lalltaig Gneiss. Experienced granulite facies metamorphism at c. 1875 Ma. The Rona Terrane, extending south from the Gairloch Shear Zone through Loch Torridon and including the Inner Hebridean islands of Raasay, Rona, Coll and Tiree, includes TTG gneisses up to c. 3135 Ma old cross-cut by c. 2955 Ma granite sheets and neosomes, together with younger, c. 2880 Ma TTG gneisses. UndenA^ent amphibolite facies metamorphism at c. 1670 Ma. On the Outer Hebridean Islands, the complex is subdivided, again from north to south, as follows: The Niss (or 'Butt') Terrane, a thin slice of Palaeoproterozoic rocks exposed at the Butt of Lewis, comprises c. 1870 Ma dioritic gneisses and the c. 1860 Ma Ness Anorthosite, deformed and metamorphosed under amphibolite facies. The Tarbert Terrane, extending south from the Butt of Lewis to the Langavat-Finsbay shear zone on South Harris, is characterised by Archaean TTG gneisses, some of which have protolith ages of c. 2830 Ma and, at one locality, an older component dated at c. 3125 Ma. Intruded by basaltic dykes and later by the c. 1675 Ma Harris granite sheets, with associated amphibolite facies metamorphism. The Rolneabhal Terrane, occupying the region of South Harris between the Langavat-Finsbay shear zone to the north and the Ensay shear zone along the south coast, consists of the c. 1880 Ma Langavat and Leverburgh metasedimentary belts and the enclosed South Harris Igneous Complex, including the c. 1880 Ma Roineval Anorthosite and Scarista Tonalite. Underwent UHT granulite fades metamorphism at c. 1870 Ma. Docked with the Tarbert Terrane post-1675 Ma. The Uist/Barra Block consists of the remaining, as yet largely undifferentiated Archaean gneisses of the southern Outer Hebridean islands. Divided east-west by The Outer Hebrides Fault Zone that places granulite facies and ex-granulite fades rocks on top of amphibolite facies gneisses. Includes TTG gneisses with protolith ages of 2825 to 2750 Ma and the gabbroic-anorthositic Corodale Gneiss. Granulite facies metamorphism on Barra has been dated at c. 2730 Ma, whereas granitic neosomes on South Uist have been dated at c. 1740 Ma. Assembly of the Lewisian Gneiss Complex appears to have been completed by c. 1670 Ma.

SGTSG Field Meeting 2003

ON THE DEEP CRUSTAL STRUCTURE OF THE LATE ARCHAEAN EASTERN YILGARN CRATON: A COMPARISON TO PALAEOZOIC AND MODERN ANALOGUES Richard Blewett\ Paul Henson\ Bruce Goleby\ David Champion\ Kevin Cassidy\ and Bruce Groenewald^ pmd*CRC do Geoscience Australia, GPO Box 378 Canberra ACT 2601 (Richard.Blewett@ga.gov.au) Geological Survey of Western Australia, PO Box 1664 Kalgoorlie WA 6433 (bruce.groenewald@doir.wa.gov.au)

Seismic reflection data provide a unique view of the structure of the Earth's lithosphere as it provides constraints on the subsurface geology that are not obtainable by any other method. The technique successfully images the geometry of rock packages and their bounding structures, so that inferences can be made about tectonic processes. The late Archaean East Yilgarn Craton is an economically important (7% of Global Exploration expenditure) part of Western Australia. In order to understand better the mineral systems of this area, an appreciation of the 3D and 4D geology is necessary. New seismic reflection data (2001) complement data acquired in 1991 and 1999, and together they provide a deep crustal image of an across-strike transect of over 400 km. The Northeastern Yilgarn deep seismic reflection data is characterised by: • a subdivision of the crust into three broad layers; • a prominent E-dip to most reflections, and; • four E-dipping crustal-penetrating shear zones. The crust is approximated 40 km thick in the Northeastern Yilgarn region, and thickens eastwards. This deepening of the Moho is achieved by a series of short steeply dipping ramps and long sub-horizontal flats. The crust is subdivided into three subhorizontal layers. The lowest layer is considered to be a ductile lower crustal unit. There are only two instances where dipping reflections are seen in this layer and they both relate to the deep penetrating shears. The upper surface of this sub-horizontal layer is defined by a sudden change in reflection character to a zone where large packages of dipping reflectivity are the norm. The middle layer is characterised by numerous prominent east-dipping large-scale lozenges. The boundary between the middle and upper layers is diffuse and irregular in geometry.

The upper layer is far more complex and variable and is thin-skinned in character. There is a very pronounced E-dip to the majority of the reflectivity seen along the seismic traverse, both in the middle and upper levels of the crust. West-dipping features are imaged to the west (Kalgoorlie). Dips are generally shallow (--30°). Four prominent E-dipping zones that transect the crust, and divide Eastern Yilgarn into five distinct domains. The shears can be traced from the surface to the lower crust and in two cases to the Moho. All deep penetrating shear zones have a complex geometry indicative of polyphase deformation of a thick-skinned nature. In the upper crust, there is no single 'detachment' surface, but in many places it appears that there could be two or more subhorizontal shear zones. Between the Leonora and the Laverton region, there is an undulating contact between the granitoid-greenstone succession and the inferred felsic basement material. This contact is interpreted to be a low angle shear zone. A key feature of the Kalgoorlie area is the 'Y-front' structure formed by the intersection of E- and W-dipping structures. There is also no clear evidence for a similar 'Yfront' further to the east.; At eastern end of transect, inversion of Proterozoic and younger sedimentary basins show that Archaean structures have been reactivated. It is uncertain as to how far this deformational event extends westwards into the Yilgarn Craton. The thin- and thick-skinned nature of the Eastern Yilgarn compares favourably with other Archaean terranes (e.g., Abitibi), as well as Palaeozoic and Modern accretionary and extensional orogenic belts. This suggests that tectonic processes in the late Archaean involved plate forces, and that younger analogues are valid for a comparative understanding.

SGTSG Field Meeting 2003

THERMAL AND MECHANICAL CONTROLS ON THE EVOLUTION OF ARCHAEAN CRUSTAL DEFORMATION: EXAMPLES FROM WESTERN AUSTRALIA Simon Bodorkos\ Mike Sandiford\ Roberto Weinberg^ and David Hansen^ ^School of Earth Sciences, University of Melbourne VIC 3010 (bodorkos@unimelb.edu.au) ^School of Geosciences, Monash University, Clayton VIC 3800 Most Archaean granite-greenstone terrains are characterised by an early first-order tectonostratigraphy that comprises a relatively dense, mafic volcanic-dominated greenstone sequence overlying a less dense tonalite-trondhjemitegranodiorite (TTG) assemblage. Regional deformation of this configuration has resulted in crustal-scale "dome-and-keel" structure, with supracrustal rocks defining synformal keels surrounding and/or flanking antiformal domes cored by granitoid or felsic gneiss. However, important differences exist between Middle and Late Archaean terrains in terms of crustal growth patterns and geometry of deformation. Many pre-3.0 Ga terrains (e.g. the eastern Pilbara Craton) comprise pseudocircular, domal granitoids separated by curviplanar greenstone keels with a wide range of strike orientations between belts. Dome-and-keel formation was invariably accompanied by felsic magmatism, linear zones of deformation are rare at terrain-scale, and crustal ages display no consistent spatial pattern within cratons. In contrast, post-3.0 Ga terrains (e.g. Eastern Goldfields Province, Yilgarn Craton) comprise elliptical, antiformal granitoids that are separated by elongate, parallel greenstone belts. Deformation largely post-dated major felsic magmatism, and first-order crustal age zoning is consistent with craton growth via lateral accretion (e.g. west-east younging in the Yilgarn Craton). We contend that these contrasts reflect fundamental changes in the thermal and mechanical behaviour of the Archaean upper lithosphere (and its response to tectonic forces) through time. The crustal thermal regime is sensitive to the abundance and depth distribution of heat-producing elements (HPEs). However, the role of "conductive incubation" (long-term steepening of the geotherm due to burial of HPEs in the crust) in the Archaean (when HPEs were more abundant and greenstone-over-granitoid stratification was common) remains underappreciated. Assuming (1) a uniform thermal conductivity k = 3 Wm'^X"^ for the Archaean lithosphere, (2) a reduced heat flow c/^ = 40 mWm"^ at 30 km

depth, and (3) that all crustal HPEs are hosted by (and uniformly distributed within) a felsic TTG layer 10 km thick, simple one-dimensional thermal models show that long-term lower crustal temperatures (Taokm) depend on the heat production rate (H) of the TTG layer and the thickness (z) of the overlying greenstones. Stratigraphic, geochemical and geochronologic data imply that in the East Pilbara, H = 4.6 ±0.6 |jWm"^ and z = 14 ±2 km at the onset of c. 3.30 Ga dome-and-keel development, yielding Taokm = 700 tlO^C. In contrast, H = 3.5 ±0.6 |jWm"^ and z = 8 ±2 km in the Eastern Goldfields, yielding Taokm = 550 ±50°C when c. 2.65 Ga dome-andkeel formation was taking place. The fact that the effective viscosity of the lower crust is strongly controlled by the temperature field means that the disparity in Tsokm values has important implications for both the style of deformation and the structural record preserved by the rocks. In the Eastern Goldfields, relatively low Taokm values imply a cold, strong crust, with subsolidus granitoid doming driven by a combination of horizontal shortening and buoyancy. Large-scale structures formed by tectonic stresses were "locked into" the crust, in similar fashion to modern orogenic belts. In contrast, conductive incubation in the East Pilbara increased Taokm by 200-300°C, permitting high-a(H20) melting of the TTG layer at c. 3.30 Ga to produce the felsic magmas that accompanied doming. The combined effects of thermal softening and partial melting in the midcrust decreased its effective viscosity by several orders of magnitude, and even a relatively small initial perturbation may have triggered significant gravity-driven deformation and magmatic diapirism. Consequently, large-scale structures in the rocks may dominantly reflect thermal and mechanical re-equilibration of the crust, and it is possible that only a fragmentary record of the tectonic stresses prevalent during early crustal assembly is preserved.

SGTSG Field Meeting 2003

THE CRUSTAL GEODYNAMICS OF THE 2.75-2.65 GA GLOBAL CRISIS Patrice F. Rev \ Pascal Phllippot ^ and Nicolas Thebaud ' ' ^ School of Geosciences, Building F05, The University of Sydney, NSW 2006, Australia, ^ IPGP, Laboratoire de Geosciences Marines, UMR 7097, 4 Place Jussieu, 75252 Paris, France. Corresponding author prev@es.usvd.edu.au.

Assuming that the period 2.75-2.65 Ga corresponds to a single, but global geodynamic event, we investigate through numerical experiments the mechanisms that could have led to the profound continental reworking that occurred at that time. Although the extent of the crisis at the Earth's surface pledges in favour of

Our favoured model involves a global rearrangement of convection cells in the deep mantle and formation of multiple mantle plumes. The greenstones em placed at the surface and

the involvement of mantle plumes, our numerical

the plumes that spread in the thermal boundary

experiments suggest that the thermal impact of mantle plumes alone is unlikely to explain both the amplitude and timing of the thermal anomaly, as observed in the Superior Province (Canada) and the Yilgarn Craton (Australia). Similarly, moderate crustal thickening cannot lead to significant reworking of the continental crust within the observed time constraint. Crustal thickening with a factor > 1.5 is also unlikely because it is not consistent with the moderate metamorphic grade observed at the surface of many Archaean cratons. Burial of a radiogenic crust under a 10 km thick greenstone cover also falls short of explaining, not so much the amplitude and the extent, but the timing of the thermal anomaly. In contrast, the combination of the thermal anomaly related to the greenstone blanketing effect with the heat transfer from a plume head spreading at the top of the thermal boundary layer can adequately explain the

layer contributed to heat the crust from both above and below. This produced massive crustal partial-melting that reached its climax ca. 40 Myr after the emplacement of the plumes and associated greenstone cover rocks. In turn, this led to gravitational instabilities in the crust, as dense greenstone cover rocks progressively sink into the thermally softened crust and, granite domes rose in response. The extraction of heatproducing elements toward the upper part of the crust has contributed to the cooling and stabilization of the cratons. This succession of events, which is not incompatible with platetectonic processes, may have profoundly changed the nature of the crust exposed at the surface and could explain the contrasting geochemical signatures of Archaean and postArchaean shales.

amplitude, the timing and the extent of the 2.752.65 Ga crisis.

SGTSG Field Meeting 2003

AGE, EXTENT, AND RATE OF SEDIMENTATION OF AN ARCHEAN TURBIDITE BASIN, SLAVE CRATON, CANADA Wouter Bleeker^ ^Continental Geoscience Division, Geological Survey of Canada, 601 Booth Street, Ottawa, Ontario, Canada, K1A 0E8; e-mail address: wbleeker@nrcan.gc.ca

Introduction: Turbidites form the dominant type of sedimentary material in Archean granitegreenstone terrains, typically appearing towards the top of local stratigraphic sections. Despite the relative abundance of such turbiditic rocks in the Archean record, the nature of the sedimentary basins in which they were deposited remains controversial, with interpretations varying from fully oceanic settings (telescoped into putative "accretionary prisms"), to collisional foredeeps, and extensional ensialic settings. Here we present new observations on the ca. 2.66 Ga Burwash Formation and correlative turbidites, which cover as much as one-third of the moderate size Slave craton of the Canadian Shield and rank among the world's best preserved Archean turbidites. Our data and observations suggest a largely ensialic setting with more or less synchronous sedimentation peaking across the craton at 2661-2662 Ma. We define the basin in which these turbidites were deposited as the Burwash Basin. Age and Extent: Volcanic and volcaniclastic rhyolite complexes immediately below Burwash turbidites range in age from 2679±3 Ma to 2658±1 Ma. Hence, the total life-span of the Burwash Basin may have ranged from 2682 to ca. 2655 Ma, although the bulk of sedimentation appears to have occurred in a narrow interval from 2665 to 2658 Ma, with peak sedimentation rates of several meters per 1000 years. The extent of the Burwash Basin must have stretched well beyond the present rifted margins of the craton. The long axis of the basin (SW-NE) thus exceeded 800 km, whereas its width was at least 400 km. The long axis of the basin is now parallel to a belt of early chevron folds, which reflect the collapse of the Burwash Basin.

progressive off-stripping and reworking of underlying volcanic sequences and their plutonic centers, with only minor contributions from basement rocks. Coarse detrital titanites from the plutonic cores of the felsic volcanic centers indicate that plutonic centers were being uplifted and exhumed during turbidite sedimentation in adjacent depot centers. Rates of Sedimentation: Thin, yellowweathering, airfall tuff layers allow determination of precise ages of deposition and rates of sedimentation. Several tuff layers, distributed throughout as much as 10 km of vertical section, yield identical ages of 2662±2 Ma. Hence, at 2662 Ma, time-averaged rates of deposition were at least 2.5 m/1000 years and likely peaked at several times this value. This compares well with high sedimentation rates in tectonically active modern basins. Interestingly, tuff layers near the top of thick turbidite sections have the same age as nearby rhyolite complexes at the base, requiring significant lateral thickness variations and deposition of thick turbidite successions adjacent to still emergent felsic volcanic edifices. Conclusions: The 2.68-2.66 Ga Burwash Basin is one of the largest and best-preserved Archean turbidite basins. It formed in a largely ensialic setting, characterized by active volcanism and creation of accommodation space by contemporaneous extension. Peak rates of sedimentation (several meters/1000 years) were attained coeval with 2661-2662 Ma felsic volcanism, with detritus being derived from active volcanoes, and off-stripping and reworking of uplifted volcanic sections and their plutonic cores. These characteristics invite comparisons with turbidite basins in modern back-arc settings.

Setting and Provenance: The transition from underlying volcanic rocks to turbidites varies from conformable to unconformable. In several localities, high-angle unconformities cut down into older basement granitoids, thus tying turbidite sedimentation unequivocally to ensialic basement. Detrital zircon geochronology and geochemical mixing calculations suggest a

SGTSG Field Meeting 2003

TUESDAY 23 SEPTEMBER 2003

SGTSG Field Meeting 2003 46

SGTSG Field Meeting 2003

CONTINENTAL INTERIOR TECTONICS : AN AUSTRALIAN PERSPECTIVE Mike Sandiford School of Earth Sciences, University of Melbourne VIC 3010 (mikes@unimelb.edu.au)

Widespread distributed seismicity in continental interiors such as Australia provides a number of important insights into the tectonics of continental interiors. For example, it suggests that the normal state of stress is near failure, as might be expected for a self-organized critical state. But to what cause can we assign the state of stress, and what are the long-term ramifications of such tectonic activity, if any? These questions are perhaps best addressed in terms of the geological context of the seismicity, particularly the neotectonic framework. For example, neotectonic studies can potentially constrain the patterning of deformation through time, and provide the possibility of deducing the seismic efficiency and/or long term seismic recurrence rates. The aim of this talk is to provide such a perspective

for interpreting seismicity in Australia, and to explore the broader ramifications for continental interior tectonics. Aspects of the Australian record that will be covered, include the seismological record, the neotectonic record, and the origins of and changes in the stress field contributing to seismicity. The final part of the talk will speculate on the role of continental interior tectonics in continental evolution. Using the insights from the Australian neotectonic record as well as its ancient past, the argument will be posited that so-called "intraplate" tectonic leaves an important imprint in the mechanical and thermal structure of the continent, and the way in which it does so may well provide important clues as to the rheological constitution of the lithosphere at geological timescales.

SGTSG Field Meeting 2003

TOWARDS BUILDING A 3D GEOLOGICAL MAP OF AUSTRALIA RJ. Korsch^ ^ M.G. Nicoll^ and R. Lane^ Vredictive Mineral Discovery Cooperative Research Centre ^Geoscience Australia, GPO Box 378, Canberra ACT 2601 Australia Russell.Korsch@ga.gov.au. Malcolm.Nicoll@ga.gov.au, Richard.Lane@ga.gov.au

Within Australia, all the ore bodies at or close to the surface have probably been discovered. Thus future discoveries are likely to be at depths of 100m to 1000 m in areas where there is widespread, and often thick, regolith. There is also a need to understand the architecture of the mineral systems to much greater depths. Hence there is a need to understand the regional geology of Australia to significant depths. Currently, 3D interpretation of 2D geological maps, through construction of cross sections, is weak. This has been confirmed by several deep seismic profiles, which frequently show surprises, compared with the predicted geology. In an attempt to better understand the geology of Australia at depth, 3D geological models (that is, 3D maps) are being constructed at scales varying from individual mines, to districts and provinces, to the whole of Australia. The construction of a 3D geological model of the Australian continent is limited by the sparse nature of the available 3D-constraining data. To achieve the construction, we have adopted a twofold process. One is to erect an Australia-wide 'scaffolding' using available data sets such as geology, gravity, magnetics, potential field edge detection, deep seismic reflection, seismic tomography and geological cross sections. We then build regional-scale 3D models of specific areas to start to infill parts of the 'scaffolding'. In terms of methodology at the regional scale, we construct serial geological cross sections based on surface mapping and solid geology interpretations. These sections are refined in an iterative process involving integration with available seismic data, structural information inferred from edges detected in potential field data upwardly continued to various levels, and through comparison of observed potential field data with that calculated through 2.5D forward modelling of the sections. Key surfaces can be constructed from the sections, and these are used to infill the lithology volumes. We are starting to use 3D physical property models derived from geophysical data to enhance

our understanding of the geology at depth. AEM data has been used to image conductive regolith and basement features to depths of more than 100m. Fully 3D constrained inversion of potential field data can be used to provide insight and to test geological models to depths of many kilometres for their consistency with the observed potential field data. In many areas, potential field data provides the only regularly sampled information on the subsurface. The value of this approach is typified by work carried out around the Olympic Dam deposit in South Australia where there is no outcrop of the basement succession and few drill holes that penetrate very far into the basement. Sedimentary basins with a good coverage of seismic reflection data, usually collected by industry, are much more amenable to the building of a 3D geological model than an area under a thick cover of regolith with no seismic coverage. One such area for which we have built a 3D model is the Bowen, Gunnedah and Surat basins in eastern Australia. Here, the interpretations of >1000 seismic sections have been used to determine fault geometries and to produce several structure contour maps of mappable sequence boundaries. These have been integrated with petroleum well, stratigraphic borehole and field data to produce the 3D basin geometry. In conjunction with state and Northern Territory geological surveys, Geoscience Australia has been developing geological models for the Laverton-Leonora area of the Yilgarn craton, Tasmania, Eastern Gawler craton and the AruntaTanami region of the Northern Territory. The Predictive Mineral Discovery CRC is also involved in building models for Broken Hill, the whole of Tasmania, western Victoria, Mt Isa and the Eastern Goldfields region of the Yilgarn Craton. In summary, the building of 3D geological maps is providing a better integration of geological and geophysical data and a more complete understanding of the geology of Australia in three dimensions, leading to a more rigorous approach to predictive mineral discovery.

SGTSG Field Meeting 2003

COMPARISON OF LITHOSTRATIGRAPHY AND TECTONIC HISTORY ACROSS THE YILGARN CRATON She Fa Chen, Stephen Wyche, Angela Riganti Geological Survey of Western Australia, 100 Plain Street, East Perth, WA 6004

The Archaean Yilgarn Craton consists of the Narryer and Southwest Gneiss Terranes, and the Murchison, Southern Cross and Eastern Goldfields Granite-Greenstone Terranes (GGT). The central Southern Cross GGT contains two distinct greenstone successions. A 3 Ga lower greenstone succession is characterized by mafic volcanic rocks and banded iron-formation (BIF). It is subdivided into three lithostratigraphic associations and unconformably overlain by a 2.73 Ga upper succession of calc-alkaline volcanic and clastic sedimentary rocks. Recent geological mapping in the central Southern Cross GGT has recognized three major deformation events: Di N-S compression was restricted to the lower greenstone succession and produced thrust faults and tight to isoclinal folds; D2 E-W orogenic compression at 2730-2680 Ma formed N-trending folds with a weak axial planar foliation; D3 progressive and inhomogeneous E-W shortening at 2680-2655 Ma developed regionalscale, NW-trending sinistral and NE-trending dextral shear zones that are linked by N-trending contractional zones, forming large arcuate structures. Pidgeon and Hallberg (2000) subdivided greenstones in the northern Murchison GGT into five informal lithological assemblages. The older units (Assemblages 1-3; 3.0-2.8 Ga) consist mainly of mafic-ultramafic volcanic rocks and BIF. The younger units comprise rhyolite and dacite (Assemblage 4; 2.76-2.71 Ga), and clastic sedimentary rocks (Assemblage 5). Watkins and Hickman (1990) recognized four deformation events with poorly constrained ages in the Murchison GGT. Di produced layer-parallel fabrics and isoclinal to recumbent folds. D2 N-S compression formed E-trending folds. The combined effects of Di and D2 are broadly similar to the Di in the central Southern Cross GGT. D3 E-W compression was a regional folding event that produced upright folds with a strong axial planar foliation. D4 E-W compression developed regional-scale, NE- and NNE-trending dextral shear zones. Several unconformity-bounded stratigraphic sequences have been differentiated by Krapez et al. (2000) in the Eastern Goldfields GGT. The

Kambalda Sequence characterized by basalt and komatiite was deposited in a back-arc basin prior to 2700 Ma and from at least 2715 Ma. The Spargoville Sequence of rhyolitic and dacitic lavas with intercalated volcaniclastic and sedimentary rocks was deposited in an arcadjacent, volcano-bound basin between 2700 and 2683 Ma. The Kalgoorlie Sequence dominated by volcaniclastic and epiclastic rocks was deposited in an intra-arc rift basin at two stages (2681-2670 and 2661-2655 Ma). The younger Kurrawang, Merougil, and Jones Creek Sequences are composed mainly of clastic sedimentary rocks (Krapez et al., 2000). A complex structural history with both compressional and extensional events has been recognized in the southern Eastern Goldfields GGT. The earliest phase deformation (Di) is variously interpreted as both extension and shortening. A regional folding event (D2) at 26752657 Ma formed NNW-trending upright folds and axial planar foliation. D3 E-W compression at 2663-2635 Ma developed NNW-trending sinistral shear zones. However, various authors have argued for different timing of the events (e.g. D2 started after 2655 Ma, Weinberg et al., 2003). The greenstone lithostratigraphy and deformation sequence of the northern Murchison and central Southern Cross GGTs are broadly similar in age and character, but differ from, and are older than, those of the Eastern Goldfields GGT. This may suggest that the Murchison and Southern Cross GGTs have formed in similar tectonic settings that differ substantially from that of the Eastern Goldfields GGT. References Krapez, B., Brown, S. J. A., Hand, J., Barley, M. E., and Cas, R. A. F., 2000: Tectonophysics 322, p. 89-133. Pidgeon, R.T., and Hallberg, J.A., 2000: Australian Journal of Earth Sciences 47, 153165. Watkins, K.P., and Hickman, A.H., 1990: Geological Survey of Western Australia Bulletin 137, 267p. Weinberg, R.F., Moresi, L., and van der Borgh, P., 2003: Precambrian Research 120, 219-239.

SGTSG Field Meeting 2003

STRUCTURAL ANALYSIS AND TECTONIC SETTING OF THE JACK HILLS BELT, NORTHERN YILGARN MARGIN, WESTERN AUSTRALIA. Catherine V. Spaggiari and Dave Hollingsworth Tectonics Special Research Centre, Department of Applied Geology, Curtin University, Perth, W.A. 6102, Australia. c.spaggiari@curtin.edu.au

The Jack Hills belt is an approximately 100 km long major zone of deformation within the Narryer Terrain, which forms part of the northern margin of the Archaean Yilgarn Craton of Western Australia. The Jack Hills belt is significant not only in terms of craton margin development, but also because it contains metasedimentary rocks that host the world's oldest detrital zircons, dated at 4.4 Ga. Because of the deformation intensity, stratigraphic relations of the host metasedimentary rocks are problematic, as are potential source rocks for the old detrital grains. The metasedimentary rocks comprise Archaean - Palaeoproterozoic conglomerates, quartzites, and politic schists and are juxtaposed with slivers of mafic and ultramafic rocks, and banded iron formation. These are surrounded by predominantly granitic gneisses, intruded by Late Archaean granitoids. Within the Jack Hills belt metamorphic grade ranges from upper greenschist to amphibolite, and is up to granulite facies in the surrounding gneisses. Structurally, the Jack Hills belt is dominated by a major east-northeast-trending shear zone. Satellite imagery shows the shear zone as a major linear feature that cuts across the entire length of the belt, as well as the surrounding gneisses, and shows no evidence of crosscutting deformation. Any reworking of the shear zone therefore appears to be confined to the same structure. The satellite images also show that the main foliation is drawn into the middle of the shear zone on either side, indicating apparent

dextral, strike-slip displacement. This is supported by preliminary field data, such as S/C fabrics and augen structures. The preliminary field data show that the shear zone Is dominated by a moderately to steeply dipping, often mylonitic foliation, with tight to isoclinal intrafolial folds. A well-developed mineral lineation plunges predominantly between -^10 to 30° to the west-southwest, but there is some variation to this in different parts of the belt. In places the mineral lineation is accompanied by a strong rodding lineation, and/or stretched quartz veins or pebbles in conglomerate. The predominant west-southwest plunge of the lineations suggests a component of dip slip movement with the east-northeast side up. Fold hinges are predominantly subparallel to the stretching direction, and there is evidence of complex refolding away from the central part of the shear zone where the main foliation is drawn into the belt. Timing relationships of shearing are as yet. unclear, but the presence of detrital zircons as young as --1790 Ma in some of the metasedimentary rocks suggests that shearing may at least in part relate to the Capricorn Orogeny, i.e. during collision and suturing of the Yilgarn and Pilbara Cratons, with possible reworking in the Neoproterozoic Edmundian Orogeny.

SGTSG Field Meeting 2003

A STRUCTURAL ANALYSIS OF THE HAMERSLEY BASIN, WESTERN AUSTRALIA Andrew C. Duncan ADAW Pty Ltd, Post Office, Pickering Brook, W.A. 6076, Australia. E-Mail adaw@bigpond.com.

Structural data collected from much of the southern half of the Hamersley Basin during structural/stratigraphic mapping of the Marra Mamba and Brockman Iron Formations has demonstrated a need for significant modifications to the deformation models of previous workers. Significant areas of mapping of the major iron formations of the southern Hamersley Basin makes it apparent that there is no difference in the structural history from west to east across the Basin. The only variation across the Basin is in deformation intensity and to some extent, orientation. The field relationships indicate that there have been five deformation events. Of these, only two events have any major geometric effects (D2 and D3). Results: The first event (Di) is a north-west south-east extensional event and is characterised by layer-parallel rodding, trending north-east south-west and extensional shear like structures. This event is presumably associated with the development of the Hamersley Basin. The second deformation event (D2) resulted in the large-scale regional folds seen across the Hamersley Basin. These folds are predominantly east - west in orientation. Rare, outcrop-scale, parasitic folds occur associated with this event. Where they occur, they tend to be upright to south-verging folds suggesting a north over south tectonic movement. The occurrences of smallerscale D2 structures are more common towards the east of the Hamersley Basin and may have an associated axial surface cleavage. The third deformation event (D3) is the most obvious and dominant folding event at the outcrop-scale throughout the southern Hamersley Basin. The D3 folds are generally north verging

and trend west-north-west to possibly north-west in the west of the Basin and become nearly eastwest in the east of the Basin. The D3 folds obliquely crosscut the large-scale D2 regional folds throughout the Hamersley Basin, and in the east often produce enechelon fold interference patterns with the D2 structures. The D3 event increases in intensity across the Basin from west to east. The D3 event may produce an axial surface cleavage in the right lithologies and is associated with minor south over north thrust faults throughout the Hamersley Basin. The fourth deformation event (D4) is characterised by minor upright tight to open, chevron to kink like, east - west trending folds. This event appears slightly more developed in the eastern side of the Hamersley Basin. The fifth deformation event (D5) consists of two sets of very minor upright open warps or folds trending north-west and north-east. Also two sets of cross-cutting near vertical dolerite dyke trends occur with the same trend and may be related to the folding. However, no crosscutting or refolding evidence was observed. Conclusions: The D2 and D3 events pre-date the Ashburton Basin in age. Folds resulting from the •2 and D3 events are frequently crosscut by north-west - south-east trending, near-vertical, dolerite dykes that do not crosscut the Ashburton Basin. The Ophthalmia Fold Belt, recognised in the Hamersley Basin, either consists of two dominant deformation events (D2 and D3) or should be divided into two separate Fold Belts.

SGTSG Field Meeting 2003

REGIONAL DUCTILE SHEAR ZONES AND ASSOCIATED STRUCTURES IN THE SOUTHEAST HAMERSLEY PROVINCE, PILBARA CRATON, WA. D, A, Hollingsworth, L. B. Harris, & P. A. Cawood Tectonics Special Research Centre, Department of Applied Geology, Curtin University of Technology, G.P.O. Box U1987, Perth, Western Australia 6845. d.hollingsworth@curtin.edu.au

Introduction The Hamersley Province lies across the southern part of the Pilbara Craton and is defined as the current exposure area of the Mount Bruce Supergroup. The metasedimentary and metavolcanic rocks of the Mount Bruce Supergroup have been extensively deformed along the southern margin of the Hamersley Province, and less so to the north. Deformation took place during the Palaeoproterozoic Ophthalmian Orogenic Event, at around 2200 Ma, and is the most intense in the southeast part of the Hamersley Province, adjacent to the Sylvania Inlier. Although intense deformation has resulted in development of overturned and recumbent folds within the iron-formations of the Hamersley Group, very few regional detachments have been recognised. Previous detachments described were found either within the granitegreenstone basement of the Sylvania Inlier or within the sediments directly overlying basement. During recent mapping along the northern margin of the Sylvania Inlier, the we have recognised two major ductile shear zones, the first at the base of the Hamersley Group, and the second within the Hamersley Group. Scale and significance The ductile shear zones recognised within the Hamersley Group are sub-horizontal to gently south dipping in orientation, with the exception of one outcrop that has been overturned by later-stage progressive folding. The shear zones are interpreted to be major sub-horizontal detachment zones that facilitated development of the intense folding evident in the Hamersley Group rocks to the north of the areas of outcrop. The first shear zone, located approximately 35 km to the east of Newman is visible in three outcrops over an area of 6 km^, with the largest outcrop covering an area of nearly 2 km^. The southern most outcrop is separated from the other two exposures by a late east-west trending normal fault. The highly strained zone evident in these exposures is approximately 4 m thick and represents a major decoupling between the Hamersley Group and older units.

The second shear zone is located approximately 8 km to the north-north-west of Newman and crops out over an area of less than 500 m^. The total thickness of the high strain zone of this structure is not known as it is obstructed by colluvium, however; the zone has a minimum thickness of 2 m. While this structure is evident over a smaller area than that east of Newman, it represents a ductile detachment of similar scale and lies in an area of structural importance. The Homestead system of faults, predominantly normal or oblique slip, have been interpreted as late-stage extensional structures that have reactivated earlier detachments. The ductile shear zone identified lies just to the north of the interpreted location of the Homestead Fault System, and may be a preserved fragment of the earlier detachment surface. The high strain zones appear unremarkable from a distance, and the only feature that is clearly evident is the presence of abundant stretching lineations. Due to the high strain, the outcrop takes on a bedded appearance, which is a completely transposed fabric. While the transposed fabric is folded and cut by shear bands, it is only after close examination that early attenuated folds are recognised. Sheath folds, synthetic and antithetic shear planes, and inclined to recumbent folds are all seen to overprint transposed folds and shear bands that have been rotated during ongoing ductile deformation. Conclusions The recognition of these major ductile shear zones on the southeast margin of the Hamersley Province suggests that major decoupling within the Mount Bruce Supergroup was not limited to shale or dolomite-rich units during Ophthalmian deformation. The structures evident within iron-formations in the high strain zones, and their overprinting relations record a dynamic and ongoing deformation event in which earlier formed structures are overprinted and eventually obliterated as the detachment propagates northward.

SGTSG Field Meeting 2003

GEOCHRONOLOGY OF THE ASHBURTON PROVINCE Keith Sircombe Tectonics Special Research Centre, University of Western Australia, M004, 35 Stirling Highway, Crawley, WA 6009, ksircombe@uwa.edu.au

Introduction The Proterozoic Ashburton Province is an arcuate structure spanning some 450 km from near Pannawonica in the northwest to Turee Creek Station in the east and defines the northern margin of the Capricorn Orogen. The Ashburton Province records the evolution of the Capricorn orogeny on the southern margin of the Pilbara Craton and its interaction with the Gascoyne Complex during amalgamation with the Yilgarn Craton. Numerous models have been proposed but debate has been hampered by a lack of absolute ages for the succession, especially in the thick, regionally extensive and largely homogeneous turbiditic Ashburton Formation. The Ashburton Formation is a 5-12 km thick turbiditic succession of mudstone and immature sandstone with minor amounts of conglomerate and volcanics. Based on field, modal and palaeodirection data the succession has been interpreted as a longitudinal deep-marine basin with detritus fed from the southeast. As convergence continued submarine fan systems prograding northwards from the southern margin and ultimately culminated in the fluvial dominated sediments preserved in overlying basins (Mt. Minnie, Blair and Bresnahan). Two aspects of the Ashburton Province's geochronology have been examined. Firstly, zircons from rare volcanic sequences within the Ashburton Formation have been used to geochronologically constrain the succession. Secondly, detrital zircon geochronology has been used to examine the evolution of the basin. Dated volcanic constraints In the northwest, the June Hill Volcanics underlie the Ashburton Formation and have previously been used with an overlying tuff further south in the Capricorn Group to constrain deposition to a -40 Myr period in the Palaeoproterozoic. However, recent SHRIMP reexamination has indicated that the previous U-Pb age was actually an amalgam of crystallisation and inherited ages. A new SHRIMP U-Pb age is significantly younger and remarkably similar to

the overlying tuff. This suggests that the Ashburton Formation may be diachroneous across the province and is younging westward. The hypothesis of a diachroneity is further reinforced with recent geochronology on the Mt. Boggola volcanics. This volcanic sequence occurs in the southeastern end of the province and had previously been correlated with the June Hill volcanics. A new SHRIMP age indicates that the Mt. Boggola volcanics are significantly older than June Hill and the diachroneity across the Ashburton Formation may span -30 Myr. Detrital zircon geochronology To examine detrital zircon geochronology, seven samples of the Ashburton Formation were collected in a spatial and temporal transect across the central and eastern basin. Samples from lower in the succession tend to have polymodal and wide-ranging age components, whereas higher succession samples tend toward younger and unimodal components. While the polymodal samples contain individual grain ages that potentially match the younger range of known ages in the Pilbara and Yilgarn cratons, there are no prominent clusters in this age range and Archaean ages are uncommon. All samples have prominent late Palaeoproterozoic components that are interpreted as having a provenance in similarly aged granites known from the southern margin of the Capricorn Orogen. Conclusions Detrital zircons from the central Ashburton Formation basin indicate an evolution from a polymodal Archean-Palaeoproterozoic provenance to singular Palaeoproterozoic provenance. The ubiquitous presence of Palaeoproterozoic aged detritus in the Ashburton Formation suggests that the depocentre was linked to the Gascoyne Complex during amalgamation between the Pilbara and Yilgarn. New dates for volcanism related to the Ashburton Formation indicates that deposition within the depocentre was diachroneous, migrating from the southeast to northwest over -30 Myr.

SGTSG Field Meeting 2003

STRUCTURE, METAMORPHISM AND SHRIMP U-PB GEOCHRONOLOGY OF THE CENTRAL GASCOYNE COMPLEX, WESTERN AUSTRALIA C.A. V a r v e i r \ K.E. C u l v e r * ^ I.C.W. F i t z s i m o n s Tectonics SRC, Applied Geology, Curtin University, GPO Box U1987, Perth WA 6845, Australia. ''I Present address: Sons of Gwalia - Southern Cross Operations, PMB 10, Southern Cross WA 6426 *2 Present address: BHP Billiton Iron Ore Pty Ltd, Mt Whaleback Mine, Newman

The Gascoyne Complex is the metamorphic basement of the Capricorn Orogen, developed during Palaeoproterozoic collision of the Archaean Yilgarn and Pilbara cratons of Western Australia. Recent work has focused on the southern Gascoyne Complex, where 2550-2450 Ma granite basement of the Glenburgh Terrane was intruded by 2005-1970 Ma granitoid of the Dalgaringa Supersuite and deformed and metamorphosed by the 2000-1960 Ma Glenburgh Orogeny. This latter event is thought to mark accretion of the Glenburgh Terrane to the northern margin of the Yilgarn Craton. Later deformation and metamorphism in the southern Gascoyne Complex, associated with intrusion of the 1830-1780 Ma Moorarie Supersuite, is attributed to collision further north between the Glenburgh-Yilgarn terrane and the Pilbara Craton during the Capricorn Orogeny. The details of this collision, including the location of the suture zone remain unknown. CENTRAL GASCOYNE COMPLEX The central Gascoyne Complex is dominated by biotite granodiorite of the 1840-1800 Ma Minnie Creek Batholith, which extends for at least 300 km parallel to the NW-SE structural trend of the complex. Psammitic to pelitic schists of the Morrissey Metamorphic Suite occur south of this batholith, and are interleaved with granitic gneiss and intruded by weakly foliated muscovite-biotitetourmaline granite plutons. The contact between the Morrissey Metamorphic Suite and Minnie Creek Batholith is concealed by the Ti Tree Syncline, a narrow strip of Mesoproterozoic sedimentary rocks of the Bangemall Supergroup. The Morrissey Metamorphic Suite ranges from lower-greenschist fades muscovite-chlorite schist in the north, close to the TI Tree Syncline, to midamphibolite fades kyanite-staurolite-biotite schist in the south, where metamorphic conditions reached 8-9 kbar and 600-700°C. Structure is dominated by a pervasive E-W to NW-SE trending SI foliation, which is axial planar to early upright folding of bedding. SI is folded about NW-SE trending F2 folds, which are locally transposed by an axial

planar S2 cleavage. S2 is best developed close to the margins of the ovoid granite plutons, and is probably related to their intrusion. SHRIMP U-PB GEOCHRONOLOGY U-Pb analysis of detrital zircons from a biotitemuscovite psammite of the Morrissey Metamorphic Suite yielded a dominant age population at 1840 ±4 Ma and a minor population at 2050-1990 Ma. Other analyses were either discordant or gave scattered ages between 2700 and 2100 Ma. Concordant magmatic zircon from the Gurun Gutta Granite, one of the muscovite-biotite-tourmaline granite plutons, yielded a weighted mean 207Pb/206Pb age of 1652 ±5 Ma, interpreted as the igneous crystallization age. CONCLUSIONS Sedimentary precursors of the Morrissey Metamorphic Suite were probably eroded from southerly sources. 2700-2100 Ma detrital zircons match the age of granitic basement in the Yilgarn Craton and Glenburgh Terrane, whilst 2000 Ma grains match the Dalgaringa Supersuite. The 1840 Ma population is slightly older than reported ages for the Moorarie Supersuite, but could come from an unidentified early phase of this unit, or the Minnie Creek Batholith, whose age is less well defined. These sediments were deposited after 1840 Ma, and buried to depths of 30 km and metamorphosed during intense NNW-SSE shortening. A minimum age of metamorphism is provided by the 1650 Ma plutons that cut the SI foliation. These rocks provide evidence for substantial crustal shortening and burial under a Barrovian thermal regime sometime between 1840 and 1650 Ma. This is consistent with continental collision during the Capricorn Orogeny and probably correlates with 1830-1780 Ma events in the southern Gascoyne Complex. The Glenburgh provenance of the Morrissey Metamorphic Suite implies that any collisional suture must lie along or to the north of the Ti Tree Syncline.

SGTSG Field Meeting 2003

MESOPROTEROZOIC & NEOPROTEROZOIC REACTIVATION OF THE PALAEOPROTEROZOIC GLENBURGH & CAPRICORN OROGENS: EVIDENCE FROM DATING S. A. Occhipinti and S. M. Reddy Tectonics Special Research Centre, Curtin University of Technology, P.O. Box 1987 Perth 6004

Crustal-scale shear zones are fundamental discontinuities that often remain weak and are prone to complex deformation histories due to reactivation. However reactivation can be difficult to discriminate from the complex structural features associated with transpressional deformation. Detailed and regional-scale mapping around the northern margin of the Yilgarn craton within the Palaeoproterozoic Glenburgh and Capricorn Orogenies, combined with geochronological dating, has been employed to better investigate the tectonic amalgamation of the Yilgarn and Pilbara cratons. The age of Glenburgh and Capricorn deformation is largely constrained by overprinting relationships between pre- and post-tectonic granites. These age constraints indicate a complex Palaeoproterozoic history along the southern margin of the Capricorn Orogen. The Palaeoproterozoic Errabiddy Shear Zone records discrete pulses of magmatism and deformation dated at 2000-1950 Ma and 1830-1780Ma, that suggests collision of the Glenburgh Terrane and the Yilgarn Craton by c. 1950 Ma (the Glenburgh Orogeny) and subsequent reactivation during the greenschist facies Capricorn Orogeny (1830 1780 Ma). Further east, the Kerba Shear Zone, which marks the southern boundary of the Yarlarweelor Gneiss Complex and cuts the Capricorn age Kerba granite (c. 1808 Ma), has previously been thought to be of Capricorn age. These types of feaures have been used to suggest that southern Capricorn Orogen underwent little tectonic activity after the Palaeoproterozoic Infra-red laser probe dating of individual micas from samples from a range of tectonic units within the southern Capricorn Orogen show that this is not the case. The Errabiddy Shear Zone and rocks further north in the Glenburgh terrane record cooling through c.300°C at around 1000-900 Ma. Although these micas lie within

Capricorn or Glenburgh structures the ages are unlikely to represent the slow cooling of rocks from a Capricorn greenschist thermal event. More likely, they represent a thermal event associated with late Mesoproterozoic / early Neoproterozoic tectonism which could correlate with the 1070 - 750 Ma Edmundian Orogeny, which is regionally significant in the northern Capricorn Orogen. However, there is no evidence that the micas grew at this time. Therefore, although "^Âr/^Âr data reflect tectonism after the Capricorn Orogeny, there are few structural features that can be clearly related to Capricorn reactivation. Lower intercepts for U-Pb SHRIMP analyses of zircon to the south of the Capricorn Orogen indicate a possible thermal event at 980 Ma. East of the Errabiddy Shear Zone around the Kerba Fault, Infra-red laser probe "^Âr/^Âr dating yields Mesoproterozoic cooling and deformation ages much younger than previously envisaged. Early Mesoproterozoic ages may reflect either cooling or deformation ages. However, late Mesoproterozoic ages are clearly related to structural reactivation of Capricorn high-strain zones. This reactivation predates the Edmundian Orogeny but has a temporal equivalence to tectonism taking place In the Albany-Fraser belt on the southern side of the craton, or may be related to the formation of the Collier Basin In the central part of the Capricorn Orogen. Abundant discrete brittle-ductile structures within both the Errabiddy Shear Zone and Kerba Fault that cut through Capricorn-aged fold hinges and cut out fold limbs In both areas, and have previously been presumed to have formed during the waning stages of the Capricorn Orogeny. However, the ages presented here suggest that these structures may reflect MesoNeoproterozoic deformation.

SGTSG Field Meeting 2003

ASSEMBLING AND REACTIVATING THE CAPRICORN OROGEN: LITHOTECTONIC ELEMENTS, OROGENIES AND GLOBAL SIGNIFICANCE Ian M. Tyler^ and Peter A. Cawood^ ^Geological Survey of Western Australia, Mineral House, 100 Plain Street, East Perth WA 6004, Australia (ian.tvler@doir.wa.gov.au). ^Department of Applied Geology, Tectonics Special Research Centre, Curtin University, GPO Box U 1987, Perth WA 6845, Australia (p.cawood@info.curtin.edu.au).

The Capricorn Orogen was initiated during at least two Palaeoproterozoic suturing events that brought together the Archaean Yilgarn and Pilbara cratons to form the West Australian Craton. The orogen comprises Palaeoproterozoic Plutonic and medium- to high-grade metamorphic rocks of the 2550-1620 Ma Gascoyne Complex and a series of Palaeoproterozoic volcanosedimentary and sedimentary basins, including the 2200-1805 Ma Ashburton Basin, the ca. 1800 Ma Blair Basin, the 2150-1840 Ma Yerrida Basin, the 2020-1900 Ma Bryah and Padbury Basins and the 1840-1800 Ma Earaheedy Basin. The deformed margins of the Pilbara and Yilgarn cratons represent forelands to the orogen. Major pulses of deformation and metamorphism, which may represent the culmination of two separate Wilson cycles, affected the northern and southern margins of the orogen. The 2800 Ma to 2200 Ma rocks of the Hamersley Basin and the basal Ashburton Basin along the southern margin of the Pilbara Craton record the rifting and break-up of a Mesoarchaean continent, followed by accretion and collision from the south during the c. 2200 Ma Ophthalmian Orogeny. The Glenburgh Terrane of the Gascoyne Complex could represent a remnant of the colliding continent. Rifting and break-up of a Neoarchaean continent is represented by the 2150 Ma rocks of the basal Yerrida Basin along the northern margin of the Yilgarn Craton. The development of the Bryah and Padbury basins may reflect accretion of the Glenburgh Terrane, possibly already attached to the Pilbara Craton, onto the northwestern Yilgarn Craton during the 2005-1960 Ma Glenburgh Orogeny. Accretion was accompanied by extensive supra-subduction zone magmatism. The 1830-1780 Ma Capricorn Orogeny may represent an intracratonic event resulting from reactivation of the orogen during amalgamation

of the West Australian Craton with the North Australian Craton. The associated deformational, metamorphic and magmatic events extended across the entire orogen and provide a younger limit for the amalgamation of the Yilgarn and Pilbara cratons. The upper part of the Ashburton Basin and the Blair Basin developed as a foreland basin along the northern margin of the orogen at this time. Uplift of the Gascoyne Complex and the Yilgarn Craton, together with a presently unexposed early Palaeoproterozoic terrane, has provided the sediment deposited in the Ashburton and Blair basins, the upper part of the Yerrida Basin and the Earaheedy Basin. A more localized unnamed event took place within the orogen at the end of the Palaeoproterozoic ('-1670-1620 Ma). The orogen has been the site of further intracratonic reactivation with renewed basin formation, magmatism and orogeny during the Mesoproterozoic and the Neoproterozoic. The exhumed Palaeoproterozoic components of the orogen are unconformably overlain by the intracratonic Bresnahan Basin, and the Mesoproterozoic Edmund and Collier Basins (Bangemall Supergroup), which have been deformed during the Neoproterozoic Edmundian Orogeny. The Capricorn Orogen was the site of two late Archaean to Palaeoproterozoic Wilson Cycles that may have continued for 800 to 1000 m.y. The presence of similar collisional orogens in Palaeoproterozoic terranes elsewhere on Earth suggests that the early to mid-Palaeoproterozoic was a period of global supercontinent break-up and re-assembly. Mineralization within the orogen includes a wide variety of deposit types that can be related to collisional settings, including the world-class iron orebodies of the Hamersley Basin.

SGTSG Field Meeting 2003

LATE ARCHAEAN TECTONOTHERMAL HISTORY OF THE PROTO GAWLER CRATON; GEOCHEMICAL AND ISOTOPIC CONTRAINTS Greg Swain\ Ailsa Woodhouse^, Martin Hand\ Karin Barovich\ Michael Schwarz^ ^Continental Evolution Research Group, Geology and Geophysics, University of Adelaide. SA, 5005. greg.swain@adelaide.edu.au, martin.hand@adelaide.edu.au, karin.barovich@adelaide.edu.au ^Office of Minerals and Energy Resources, PIRSA, GPO Box 1671, Adelaide, SA, 5001. Woodhouse.Ailsa@saugov.sa.gov.au, Schwarz.MichaelP@saugov.sa.gov.au

The proto Gawler Craton in southern Australia is an extensive late-Archaean aged crystalline basement province of aluminous metasedimentary, felsic/mafic/ultramafic volcanic and granitic rocks. The Craton preserves a complex tectonothermal history of earliest Proterozoic (ca. 2440-2420 Ma Sleafordian Orogeny), Palaeoproterozoic (ca. 1730-1700 Kimban Orogeny), and Mesoproterozoic (ca. 1690-1540 episodic Kararan Orogeny) aged events followed by stabilization at ca. 1450 Ma. Globally, the late Archaean is marked by a period of super continental tectonic cycles dominantly between ca. 2.78-2.59 Ga which preserve prodigious metallogenic provinces. Accordingly there has been considerable effort expended in understanding late Archaean tectonic systems, both from a mineralisation and continental evolution perspective. However, within this context, there has been comparatively little attempt to develop a coherent understanding of the geodynamic development of the proto Gawler Craton. In this contribution we show that combined application of geochemical, U-Pb geochronological and Sm-Nd isotopic tools provides an avenue to explore the development of the proto Gawler Craton. The geochemical composition of felsic volcanic and granitic rocks from a wide distribution across the craton (ca. 2560 Ma Devil's Playground Volcanics; ca. 2520 Ma Hall Bay Volcanics; ca. 2517 Ma Coulta Granodiorite; ca. 2500 Ma Glenloth Granite) exhibit the hallmarks of typical late Archaean subduction related arc-type rocks including negative Nb and Ti anomalies and TTG compositions respectively. The age range of these inferred arc-related rocks coincides with

the eruption ca. 2510 Ma ultramafic komatlites of the Harris Greenstone Domain. These rocks show geochemical characteristics similar to typical plume related late Achaean komatiites (e.g. Superior Province) including Al-depleted and undepleted compositions, near chondritic A^Os/TiOa and flat REE patterns. Coincident with the development of apparent arc felsic rocks and temporally associated plume-related mafic/ultramafics was the deposition of a series of sedimentary successions derived from reworked bimodal crust with depleted mantle model ages (TDM) that range between ca. 32002800 Ma. The co-existence of arc-like felsic rocks, plume-related mafic/ultramafics and coeval sediments derived largely from ^ 2700 Ma crust suggests that the late Archaean development of the proto-Gawler Craton was shaped by a convergent margin situated on or adjacent to a late Archaean continental domain. Within this context the generation of plume-related mafic/ultramafic rocks and the accumulation of sedimentary packages in part derived from a bimodal source(s) may have been related to rifting of the arc, conceivably linked to a retreating subduction system. At around 2500 Ma, the convergent regime became collisional, leading to the formation of the 2480-2420 Ma Sleafordian orogenic system, which affected much of the proto-Gawler Craton and its equivalents in the East Antarctic Shield. This collision lead to temporary cratonisation up until the Palaeoproterozoic 1900 Ma), when rifting initiated a new cycle of continental reorganization involving the Proto-Gawler Craton.

SGTSG Field Meeting 2003 46

SGTSG Field Meeting 2003

CHEMICAL MONAZITE DATING OF THE YARLBRINDA SHEAR ZONE, WESTERN GAWLER CRATON, SA. Gary M Ferris^ & Ron F Berry^ ^Gawler Craton Team, Minerals, Petroleum and Energy, PIRSA, EmaihFerris.Gary@saugov.sa.gov.au ^School of Earth Science, University of Tasmania. Email: Ron.Berry@utas.edu.au

The Yarlbrinda Shear Zone is a poorly exposed high strain zone within the Nuyts Domain of the Gawler Craton. It is bounded in the north by the Koonibba Fault Zone and Is stitched in the south by the Koondoolka batholith (-1580 Ma). The shear zone is 150 km in length and about 4 km wide. It separates the western edge of the Gawler Range Volcanics from Tunkillia Suite and St Peter Suite granitoids. The shear zone strikes roughly north-south, but curves to the northwest where it is truncated by the Koonibba Fault and Yerda Shear Zones. The Yarlbrinda Shear Zone is a foliated and lineated zone of deformed granitoids (Tunkillia Suite: -1680 Ma). The dominant foliation is defined by the orientation of recrystallised quartz and fractured feldspar grains. It is subvertical and strikes N-S to NNW-SSE. S-C fabrics and porphyroclast shapes record a dextral sense of shear. Plagioclase is fractured, recrystallised and altered to sericite. K-feldspar grains are less deformed and form prominent augen up to 20 mm in length. The stretching lineation, defined by feldspar and quartz grains, is shallowly plunging, 10°-20°, to the north. Locally the stretching lineations are steep, plunging 70°-80° to the south. The latter orientations probably formed during reactivation in a D2 stress field associated with the activity of the Yerda Shear Zone (-1590 Ma). The Di deformation is bracketed by the age of the host rocks (1680 Ma) and intrusion of the Kondoolka Batholith at 1580 Ma. Granite within the Kondoolka Batholith is undeformed with the exception of minor brecciation and veining along the northern margin that is possibly related to late •2 movement. Alteration and gold mineralisation within the Yarlbrinda Shear Zone occurred late within the deformation cycle, possibly associated with the switch from N-S to WSW-ENE compression, and Hiltaba suite (-1580 Ma) plutons. Granitoids from the Yarlbrinda Shear Zone show a range of chemical monazite ages. Sample 387406 is weakly deformed. Original quartz grains have been deformed with strong

subgrain formation. Subgrain margins are lobate indicating active grain boundary migration. Feldspar grains are fractured and twinned. The plagioclase is strongly sericite altered. Monazite grains retain a rounded shape and 20 |Lim grain size typical of monazite in granitoids. Chemical dating of this monazite gives a peak at 1680 Ma consistent with retention of the magmatic age. There is a tail of younger ages down to about 1550 Ma. Sample 368554 is more deformed and recrystallised. Quartz grains are recrystallised to a fine grain size. The feldspar grains are strongly fractured and there are large patches of finegrained recrystallised feldspar. Plagioclase is extensively sericitised. Monazite grains are about 10 fim, and some grains have shapes suggesting they result from breaking of larger grains. One patch of small monazite grains is 500 |am long and is interpreted as the result of fracturing, stretching and recrystallisation of an original large monazite grain. The monazite from this sample has a chemical age of 1600 ±32 Ma which is interpreted as the age of the Di deformation. Sample 387420 is strongly deformed and sericite altered. Less than 50% of the feldspar survives. The quartz grains are small fragments in a flood of alteration. Pyrite is common throughout the slide. Monazite grains are rare but occur as small grains in veins and as large fractured grains. Sulphide is common in cracks in the larger grains. The preliminary data on monazite ages suggest two ages are present, one at about 1600 Ma, which we interpreted as due to the deformation and another population with ages near 1510 Ma which may be related to the mineralising event in the shear zone. Chemical monazite dating records a range of ages in the Yarlbrinda Shear Zone. These ages reflect recrystallisation driven by high strain and fluid flow. The least altered rocks preserve the magmatic age at 1680 Ma. High strain rocks have been reset at about 1600 Ma. The most altered rocks now record ages between 1500 and 1600 Ma.

SGTSG Field Meeting 2003

EARLY HIGH-TEMPERATURE SHEAR ZONES IN THE BROKEN HILL BLOCK, NSW Caroline Forbes, Pete Betts and Gordon Lister School of Geosciences, Australian Crustal Research Centre, pmd*CRC, Monash University, Clayton VIC 3800

Extensional tectonics is recognised as having a significant role in the evolution of many polydeformed terranes. High-temperature shear zones that form at mid-crustal levels are important in accommodating extension within these terranes, and have been recognised in metamorphic core complexes and rift environments. These structures influence hightemperature/low-pressure metamorphism and the processes of exhumation of mid- to lower-crustal levels. Recognising and understanding these high-temperature shear zones is a key to unraveling the history of extensional terranes. However, within polydeformed terranes with a complex history Involving early extension followed by multiple episodes of shortening, recognition of early high-temperature shear zones can be difficult. To understand hightemperature shear zones and their relation to the tectonic evolution of a complexly deformed terrane, any shortening and deformation overprinting effects subsequent to shear zone development needs to be accounted for. The Early Proterozoic Broken Hill Block, central western NSW, is an example of a polydeformed terrane and preserves evidence of a complex shortening history at mid-crustal levels. The presence of early high-temperature shear zones in this terrane has lead to a re-evaluation of the early history of the area. The Allendale Mine Area in the northern Broken Hill Block was chosen to study and characterise an early high-temperature shear zone. The shortening history of this area has been constrained through detailed structural mapping. Two episodes of shortening at upper greenschist to amphibolite fades metamorphic conditions overprint the shear zone. These episodes resulted in the development of meso-to macro-scale Type 2 fold interference patterns as the result of fold interference between F2

recumbent and F3 upright, --N-S trending folds. Mapping and metamorphic analysis of an early high-temperature shear zone in the area has lead to these zones being placed within the context of the pre-shortening crustal architecture of the area. Understanding the role of early hightemperature shear zones in the evolution of the Allendale Mine Area follows onto further understanding of the thermal regime during intense deformation of the area. Orogenesis during the Olarian Orogeny (1.6-1.58 Ga) and the heat source of this event to result in the development of a high-temperature/low-pressure peak metamorphic mineral assemblage during shortening has been an enigmatic subject in the evolution of the Broken Hill Block for many years. If high-temperature shear zones are related to extension, the timing of development of these shear zones becomes important in linking deformation and metamorphism. Previous suggestions of early extension at 1.69 Ga in the Broken Hill Block imply early hightemperature/low-pressure metamorphism, however this can not explain peak hightemperature/low-pressure metamorphism during the Olarian Orogeny since the lithosphere would have undergone thermal re-equilibration between periods of extension and orogenesis. Therefore, the extensional thermal regime could not be inherited into orogenesis, However, if the timing of extension was just prior to the Olarian Orogeny, ie. close to 1.6 Ga, the hightemperature/low-pressure thermal regime of extension could be inherited into orogenesis, thus accounting for peak high-temperature/lowpressure metamorphism during compression. This would imply tectonic switching between extension and orogenesis over a short time (10 to 20 million years).

SGTSG Field Meeting 2003

DETACHMENT FAULTING AND SHEAR-HOSTED MINERALISATION BETWEEN UPPER AND LOWER PLATES OF CONTRASTING TECTONOTHERMAL HISTORY IN THE PALEOPROTEROZOIC WILLYAMA SUPERGROUP, SOUTHCENTRAL AUSTRALIA George Gibson\ & Karol Czarnota^ ^ Broken Hill Exploration Initiative (BHEI) & Predictive Mineral Discovery CRC, Geoscience Australia, Canberra, ACT 2601, Australia. (George.Gibson@ga.gov.au) ^BHEI, Geoscience Australia, Canberra, ACT 2601, Australia

The multiply-deformed Paleoproterozoic Willyama Supergroup incorporates ca 1700 Ma upper and lower metasedimentary sequences subjected to different tectonothermal histories and separated by a detachment surface of inferred D1 age. Rocks units below the detachment surface consist mainly of sillimanite to granulite grade psammopelitic schists and quartzofeldspathic gneiss with subordinate amounts of intercalated quartzite, marble and calc-silicate rock, and have been extensively migmatised with the amount of partial melting increasing with structural depth and proximity to early magmatic intrusions. Early magmatism was bimodal, of pre- to syn-DI age and includes 1700 Ma A-type granites as well as former 1690-1670 Ma dolerite intrusions now represented by amphibolite and two-pyroxene granululite whose geochemistry is consistent with intrusion into a continental rift or extensional environment. Metamorphism accompanying D1 deformation was of the low pressure (P) - high temperature (T) type and peaked around 1690-1670 Ma. In contrast, the upper sequence is largely devoid of syn-DI migmatites and bimodal magmatic rocks and encompasses mainly psammopelitic rocks metamorphosed under andalusite, rather than sillimanite, grade conditions. It shares the same deformational history (D1-D3) as the lower sequence but was metamorphosed at lower temperatures and shallower crustal depths. Unlike the underlying sequence, it also incorporates a number of graphitic horizons and is chemically more reduced. Lower sequence rocks, on the other hand, are oxidized, contain appreciable magnetite and become increasingly albitised towards the contact with the overlying sequence, indicating little or no exchange of fluids between these two sequences during the course of D1 deformation. Rather, fluid flow appears to have been channeled towards, and along the detachment surface, promoting strain

partitioning (mylonitisation) as well as increased amounts of hydrothermal activity and mineralization along the boundary between the two sequences. Subsequent to initial low P - high T metamorphism, the detachment surface and its associated D1 mylonites underwent further deformation and higher pressure metamorphism attendant on D2 recumbent folding and related crustal thickening. D2 deformation commenced no earlier than 1640 Ma and probably culminated around 1600 Ma, leading to additional sillimanitegrade metamorphism and selective replacement of andalusite-bearing mineral assemblages in parts of the upper sequence subjected to greater amounts of tectonic burial. This, and the fact that compositional profiles through syn-D2 garnet porphyroblasts are flat, indicates that temperatures during D2 deformation were sufficiently high in many parts of the Willyama Supergroup to promote near complete thermal equilibration. Syn-D2 thermal re-equilibration has masked the D1 conditions and made recovery of thermobarometric data relating to the earlier D1 event difficult. It is nevertheless evident from their different magmatic histories and relict Ml mineral assemblages that the two metasedimentary sequences were originally metamorphosed at different crustal levels and have since been tectonically juxtaposed. In view of these differences, the style of metamorphism, and the restriction of bimodal magmatism to the lowermost sequence, we suggest that juxtaposition is the result of extensional processes and that the boundary between these two sequences formed at mid-crustal depths in a manner analogous to that which produced the metamorphic core complexes of the North American Cordillera.

SGTSG Field Meeting 2003

AGE CONSTRAINTS ON TECTONISM IN SOUTHWESTERN CURNAMONA PROVINCE Lachlan Rutherford\ Martin Hand^ Karin Barovich^ Continental Evolution Research Group, Geology and Geophysics, School of Earth and Environmental Sciences, University of Adelaide, South Australia 5005 ^lachlan.rutherford@adelaide.edu.au ^martin.hand@adelaide.edu.au \arin.barovich@adelaide.edu.au

Introduction: The tectonic evolution of the Curnamona Province in the southeastern Australian Proterozoic has been the source of on-going debate, principally centred around the timing of the major stages of the metamorphic and structural evolution. Two schools of thought have emerged. The first suggestion is that the major phase of highT metamorphism and associated deformation occurred in an extensional setting at around 1670 Ma, linked to the progressive development of the Willyama Basin. In this model the regional pervasive layer parallel foliation formed in system analogous to a core complex that modulated the sequence evolution of the Willyama Supergroup. The second view is that the bulk of the regional deformation and associated high-T, low-P metamorphism occurred at around 1600-1590 Ma, associated with crustal thickening. One of the reasons that this debate is not yet resolved is the relative paucity of geochronological data obtained from minerals that either define structural fabrics, or formed at identifiable points In the metamorphic evolution. We present garnet Sm/Nd and monazite U-Th-Pb chemical ages from metamorphic assemblages that formed during the pervasive regional high geothermal gradient metamorphism that characterises the Olary region of the southwestern Curnamona Province. Garnet Sm/Nd and monazite U-Th-Pb chemical age data is also presented from greenschist to amphibolite-grade shear zones that dissect the terrain. Proterozoic metamorphism: Sm-Nd isotopic analysis on two garnet- bearing assemblages developed within the ubiquitous bedding parallel fabric give ages of 1583 ± 6.1 Ma (2a) and 1583 ± 5 Ma (2a). A slightly older age of 1609 ± 8 Ma was obtained from a leachate of one of the garnet samples. The bedding parallel fabric is folded by the upright ENE-trending folds that predate the regional

S-type granites in the terrain. Reconnaissance electron microprobe chemical analysis of monazites from transitional granulite-grade metapelites yielded U-Th-Pb chemical ages of -1570 Ma. Palaeozoic metamorphism: Sm-Nd isotopic ages from east-west trending garnet +/- staurolite-bearing shear zones give ages of 536 ± 14 Ma (2a) and 505 ± 13 Ma (2a). Preliminary monazite U-Th-Pb chemical ages from the shear zone assemblages also yield ages around 500 Ma. Discussion: The garnet Sm/Nd age constraints obtained from the layer parallel foliation suggest that regional prograde metamorphism occurred at around 1585 Ma in the southwestern Curnamona Province. At present there is little isotopic age evidence for the hypothesized -1670 Ma tectonism. The suggestion that regional prograde metamorphism occurred at around 1585 Ma is consistent with the presence of regional-scale granites in the southwestern Curnamona Province inferred to be around this age, which were derived either largely or exclusively from melting of the Willyama Supergroup. These c. 1580 Ma granites obviously represent the highest-temperature event to have affected the Willyama Supergroup, and we interpret these to have been generated by compressionally triggered partial melting. This -1585 Ma age is identical to the timing of compressional regional high-T low-P tectonism in the Mt Isa Inlier and in central Australia, as well as the timing of compressional deformation associated with the Hiltaba Suite in the Gawler Craton. The causes of this high-T tectonism across Proterozoic Australia are unclear at this stage. However we speculate that it is a response to north-directed collision between elements of the southern and central Australian "cratons".

SGTSG Field Meeting 2003

INSIGHTS INTO THE EARLY ISAN OROGENY FROM THE SNAKE CREEK ANTICLINE - A REGIONAL SCALE RECUMBENT FOLD AT THE EASTERN MARGIN OF THE MOUNT ISAINLIER David Giles, Peter Betts and Laurent Aillieres Australian Crustal Research Centre, School of Geosciences, Monash University 3800, Victoria, Australia

The formation of large recumbent folds is a longstanding issue in structural geology. Such folds occur in many orogenic belts and imply largescale lateral translations - by their common juxtaposition of unrelated rocks in the hangingwall and footwall of basal shear zones. Suggested mechanisms of formation include gravitational collapse of previously overthickened crust, gravitational sliding along lowangle normal faults and push-from-rear style deformation associated with low-angle thrusts. These mechanisms each imply a different timing of fold formation with respect to the evolving orogen and each should produce a characteristic morphology and strain distribution. As such, large recumbent folds can provide important information on the processes by which orogens formed. The Snake Creek Anticline is a 10km-scale synformal anticline at the eastern margin of the exposed Mount Isa Block. The fold has an overturned western limb and an axial surface that dips steeply to the east. Strain intensity increases toward the western contact between the Soldiers Cap Group and the underlying Corella Formation. The axial surface foliation is a differentiated schistosity indicating that the fold is at least a second-generation structure. North of the Snake Creek Anticline this fabric curves from north-south trending to east-west trending and is axial planar to shallowly east-plunging upright folds. In the south toward the core of the Snake Creek Anticline, the fold axis (steeply plunging in the north) approaches parallelism with a moderately southeast-plunging mineral lineation defined by biotite and sillimanite. Removing the effects of later deformation reveals the original geometry of the structure - a large recumbent fold - highly non-cylindrical in its core with a crumpled frontal zone of east-west

trending upright folds. The distribution of strain is consistent with "push-from-the-rear" models of nappe formation, rather than "gravity sliding" or "gravitational collapse" which have been proposed for some fold-nappes in the European Alps. The Soldiers Cap Group rocks formed a northwest vergent thrust wedge facilitated by high pore fluid pressure in the underlying Corella Formation. Upright folds and cleavages developed at greenschist facies in the toe of the wedge - with shortening accommodated largely by pure-shear. At the same time recumbent sheath folds and flat lying schistosities developed at upper amphibolite facies in the body of the wedge - where there was a combination of topto-the-NW simple shear and vertical flattening. The regional distribution of metamorphic zones is consistent with microstructural observations showing that metamorphism was broadly synchronous with formation of the Snake Creek Anticline, but was outlasted by deformation. Dating of metamorphic monazite elsewhere in the eastern Mount Isa Block suggests that this metamorphic episode occurred during the early stages of the Isan Orogeny ca 1.60-1.58 Ga. The fold formed as ca 1.70-1.65 Ga volcanosedimentary basins, lying marginal to the presently exposed Mount Isa Inlier, were inverted and thrust to the north-west during the thinskinned early stages of the Isan Orogeny (ca 1.60-1.58 Ga). Presumably the body of the wedge was buried beneath another thrust-slice (or series of slices) toward the east. Formation of a thrust pile in this way would have resulted in burial, metamorphism and exhumation of successive thrust-slices as the pile built toward the northwest - thus exhuming the high-grade core of early Isan Orogeny at the eastern margin of the block.

SGTSG Field Meeting 2003

THE HILTABA E V E N T - A N EXAMPLE OF HOTSPOT RELATED FLAT SUBDUCTION? Pete Betts, David Giles and Gordon Lister School of Geosciences, Australian Crustal Research Centre, Monash University, Clayton VIC 3800

The ca. 1600-1580 Ma Hiltaba Event represents a continental-scale tectono-thermal event that is preserved throughout the Proterozoic terranes of eastern, southern, and central Australia. In the Gawler Craton this event marks the transition from arc-related magmatism (St. Peters Suite) to continental back-arc magmatism in the overriding plate of a north-dipping subduction zone. This magmatism was characterised by voluminous outpourings of ca. 1600-1590 Ma bimodal volcanics (Gawler Range Volcanics) and the emplacement of the variably deformed A-type granites and minor picritic dykes of the Hiltaba Suite (ca. 1590-1580 Ma). Syn-orogenic magmatism also occurred in the Curnamona Province (Mundi Mundi Granites). Coincident with magma emplacement in the Gawler Craton was an episode of ca. 1600-1575 Ma HT-LP metamorphism in the Mount Isa Inlier, Curnamona Province, Coen and Yambo Inliers and the Arunta Inlier. This period of HT metamorphism was coincident with a period of regional -north-south directed crustal shortening in the Arunta Inlier (Chewings Orogeny), Mount Isa Inlier (Isan Orogeny), Curnamona Province (Olarian Orogeny), and the Georgetown Inlier (Jana Orogeny). Tectonic models for the Hiltaba Event must account for the coincidence of event across the entire continent. Key observations that constrain a holistic plate tectonic interpretation include: (1) switching off the arc magmatism along the southern margin of the Gawler Craton at ca. 1620 Ma; (2) A-type magmatism and bimodal volcanism in the central Gawler, which have a strong plume signature; (3) Syn-deformation emplacement of the Hiltaba Granites throughout the Gawler Craton; (4) structural observations in the eastern Mount Isa Inlier that suggest that the HT thermal regime of the lithosphere pre-dates

the regional shortening; (5) regional shortening throughout a large area of the continent interior coincident with A-type magmatism. We suggest that these observations can be explained by the southward retreat of a northdipping subduction along the southern margin of the continent across a mantle hotspot or plume. Roll-back of this subduction zone resulted in lithospheric extension and attenuation, and elevated geothermal gradients in the overriding plate (Gawler Craton, Curnamona Province, and North Australia Craton), as well as arc-related magmatism along the southern margin of the Gawler Craton. As the subduction zone retreated across the hotspot the dip of the subduction zone became shallow, arc-related magmatism switched off and flat subduction ensued. Flat subduction resulted in a switch from crustal extension in the overriding plate to regional crustal shortening that extended as far north as the Mount Isa and Georgetown Inliers. Crustal shortening appears to have occurred slightly earlier in the Curnamona Province and Arunta Inlier (ca. 1600-1590 Ma), whereas peak metamorphic conditions in . the Georgetown and Mount Isa Inlier suggest that the onset of orogenesis occurred between ca. 1590-1575 Ma. The thermal erosion of the flat slab beneath the Gawler Craton may have allowed renewed rollback of the subduction hinge and interaction between the hotspot and the continental lithosphere resulting in voluminous crustal melting and the emplacement of the Hiltaba Granites and associated picritic melts. The distribution of ca. 1580-1500 Ma A-type granites along the eastern Proterozoic terranes of Australia shows that they become increasingly younger to the north. We speculate that this may represent a hotspot trace as Australia migrated southward during this period

SGTSG Field Meeting 2003

STRUCTURAL INTERPRETATION OF THE YARDILLA 1:100 000 SHEET: IMPLICATIONS FOR THE YILGARN - ALBANY FRASER SUTURE Sarah Jones Geological Survey of Western Australia, Kalgoorlie Office, P.O. Box 1664, WA 6433

The Yardilla 1:100 000 sheet is located about 100 km southeast of Kalgoorlie in the Eastern Goldfields Granite-Greenstone Terrane. It is dominated by Archaean granitoid and supracrustal rocks of the Yilgarn Craton, with Proterozoic gneisses of the Albany Fraser Orogeny in the southeast, and Proterozoic Woodline Formation sedimentary rocks in the northwest. Structural trends differ markedly from the typical NNW-trending structural grain of the Eastern Goldfields and is indicative of a complex tectonic history related to overprinting by the Mesoproterozoic Albany Fraser Orogeny. Five deformation events (Di to D5) are preserved in Archaean rocks. A fine penetrative foliation (Si), typically parallel to bedding, and rare tight to isoclinal recumbent Fi folds represent Di. A stretching lineation is developed on Si, parallel to Fi fold hinges. The D2 event resulted in NNW-trending open to tight upright folds with an axial planar foliation (S2) formed during E-W compression. Refolding of Di fabrics during D2 resulted in local type-1 and type-2 fold interference patterns. Although visible on aeromagnetic images, regional scale D3 and D4 brittle structures were not observed in the field. The D5 event, attributed to Albany Fraser-related deformation, is subdivided into Dgg, Dgb and Dgc. Dsa is characterised by moderately NE-plunging open folds formed during NW-SE compression. On Yardilla, these folds now plunge to the SE as a result of clockwise rotation during Dsb dextral displacement. This rotation also affected earlier fabrics, with a shift from the regional NW trend to a local NE trend, parallel to the suture zone. Continued NW-SE compression during D^c resulted in the development of a steep NE-striking cleavage (S5). The Ssc cleavage progresses from a spaced cleavage in the northwest, to a strong penetrative schistosity in the suture zone. This fabric is associated with an increase in metamorphic grade from greenschist to amphibolite fades, characterised by an increase in mica grainsize, followed by the incoming of garnet in the suture zone. In the Proterozoic rocks of the Fraser Range In southeastern Yardilla, three deformation events were recognised (Dpi to Dps). A NE-striking, steeply

dipping gneissic banding (Sfi) with a shallow NEplunging lineation and dextral shear sense indicators represents Dp^j Df2 resulted in steep shear bands with steeply plunging lineations; and, Df3 is represented by subvertical NW-striking dextral strike-slip faults. Two stages have been proposed for the Albany Fraser Orogeny, with Stage I representing the main continent-continent collision at c. 1300 Ma, and renewed NW-SE convergence and intracrustal reactivation in Stage II at c. 1200 Ma. Recently, three phases have been suggested for the orogeny; with an initial continent-continent collision representing Stage I, and an early Stage II cratonscale thermal anomaly caused by emplacement of mafic dykes and granitoids, followed by reactivation of the orogen as a result of renewed NW-SE shortening (late Stage II). Steeply dipping NE-striking Dpi fabrics in the gneisses of the Fraser Range are correlated with regional D2 fabrics in the Albany Fraser Orogeny, attributed to dextral transpression in a late phase of Stage I deformation. The steep overprinting Df2 shear bands with steep lineations are similar to structures in the eastern part of the orogen that are attributed to NW-SE convergence and reactivation during a late phase of Stage II. The NW-striking Dp3 faults in the Fraser Range are absent from eastern parts of the orogen, but are similar to D3 structures in the western part of the orogen attributed to late NW-SE convergence. In summary, the D5 event in Archaean rocks of the Yilgarn Craton is attributed to Albany Fraserrelated deformation, with rotation of Di-Dsa structures from a regional NNW trend to a local suture-parallel NE trend. The clockwise rotation suggests a large component of dextral shear during collision, consistent with dextral shear sense indicators in the gneisses of the Fraser Range. The NE-oriented S5 fabric in Archaean rocks in the suture zone is parallel to Dpi fabrics in the Fraser Range and most likely formed at the same time. Subsequent reactivation of the Albany Fraser Orogeny may be indicated by the steep Dp2 shear bands that overprint earlier fabrics in the Fraser Range gneisses.

SGTSG Field Meeting 2003

MIX AND MATCH: USING ZIRCON GEOCHRONOLOGY TO CORRELATE LATE MESOPROTEROZOIC METAMORPHIC BELTS IN ANTARCTICA AND WESTERN AUSTRALIA I.C.W. Fitzsimons Tectonics SRC, Applied Geology. Curtin University. GPO Box U1987. Perth WA 6845. Australia. i.fitzsimons@curtin.edu.au Mesoproterozoic Grenville-age metamorphic belts are widespread in East Gondwana and preserve evidence for multiple tectonic events between 1350 and 900 Ma. Previously they were regarded as a single collisional orogen marking the assembly of East Gondwana, but they actually comprise three distinct provinces juxtaposed by 550-500 Ma tectonism. This is best preserved in Antarctica, where all three Grenville-age domains are exposed, each with a different age for high-grade tectonism: the Maud (1090-1030 Ma), Rayner (990-900 Ma) and Wilkes (1330-1130 Ma) provinces. These correlate closely with rocks in southern Africa (Namaqua-Natal and Mozambique belts), India (eastern Ghats) and Australia (Albany-Fraser). Grenville-age rocks with evidence for high-grade tectonism at 1090-1030 Ma also occur as blocks in the Neoproterozoic Pinjarra Orogen of Western Australia. Their age is similar to that of the Maud Province, raising the possibility that they are related. PINJARRA OROGEN Grenville-age rocks in the Pinjarra Orogen comprise 1090 Ma granitic orthogneiss in the Leeuwin Complex, and psammitic to politic paragneiss in the Northampton and Mullingarra complexes deformed and metamorphosed to amphibolite or granulite fades at 1080-1030 Ma. This tectonism is traditionally believed to reflect collision of Australia with India at 1100 Ma, since these rocks lie between India and Australia in Gondwana reconstructions. This need not be the case, however, given that East Gondwana is now known to have assembled at 550 Ma and it is likely that India and Australia attained their Gondwana positions at this time. MAUD PROVINCE The Maud Province is interpreted as an 11501100 Ma magmatic arc, back-arc basin, and continental margin sedimentary sequence developed at the edge of an unexposed craton that collided with the southeastern margin

(present-day coordinates) of the KaapvaalZimbabwe Craton of southern Africa at 1100 Ma. This collision resulted in pervasive deformation, granulite-facies metamorphism, and magmatism at 1090-1030 Ma. The unidentified craton is widely assumed to be the East Antarctic Shield, but again this need not be the case given widespread evidence that East Antarctica did not assemble until 550 Ma. DETRITAL ZIRCON POPULATIONS Evidence for the possible identity of the colliding cratons in both cases is provided by detrital zircon. SHRIMP U-Pb zircon age data for three paragneiss samples from the Pinjarra Orogen and two samples from the Maud Province reveal a number of striking similarities. 2100-1110 Ma grains dominate all samples, with a marked lack of c. 1500 Ma grains. Although different samples are dominated by different populations within this range, significant detrital populations at 1100-1115, 1160-1220, 12801310, 1350-1390, and 1420-1440 Ma occur in samples from both regions. CONCLUSIONS Age spectra from the Pinjarra Orogen and Maud Province are indistinguishable within the uncertainties of the data, and imply that paragneisses in both areas were part of the same sedimentary sequence eroded from the same source rocks, and are thus fragments of the same collisional orogen. Pre-1130 Ma detrital zircon populations in both areas match the ages of basement rocks in the Albany-Fraser Orogen and Wilkes Province, but these rocks cannot provide the youngest detrital population at 11301100 Ma. These latter grains were probably eroded from the magmatic arc exposed in the Maud Province, consistent with deposition at an active margin of an Australian-Antarctic Craton. The sedimentary rocks were then deformed and metamorphosed as the Kaapvaal-Zimbabwe Craton collided with this convergent margin at 1100 Ma.

SGTSG Field Meeting 2003

STRUCTURAL HISTORY OF THE SOUTHERN LEEUWIN COMPLEX, SOUTHWEST AUSTRALIA Janssen, D, P., Fitzsimons, I. C. W., Collins, A. S. Tectonics S.R.C., Department of Applied Geology, Curtin University of Technology, GPO Box U1987, Perth WA 6845. Email: d.ianssen@curtin.edu.au; i.fitzsimons@curtin.edu.au: a.s.collins@curtin.edu.au

INTRODUCTION The Leeuwin Complex is one of three basement inliers outcropping within Phanerozoic basin sediments on Australia's western margin, and forms part of the Pinjarra Orogen. To the east, the Dunsborough Fault separates the crystalline Leeuwin basement from Phanerozoic Perth Basin sedimentary rocks, while to the west the complex is bounded by the Indian Ocean. The Leeuwin Complex is dominated by various felsic orthogneiss units, interleaved with occasional mafic layers, and preserves predominantly north-south trending structures. The southern end of the Leeuwin Complex between Cosy Corner and Augusta provides an extensive east-west section across those structures, and a structural history with relative timing can be constructed. Four deformation events have affected the region, D1 to D3 being folding events and D4 being brittle in nature. The lithologies that are deformed range from amphibolite bands, through to anorthositicleucogabrroic compositions, a range of felsic gneisses, and a small meta-sedimentary package. Deformation events The dominant D1 structure observed in the field is a gneissic foliation, which is axial planar to the F1 isoclines which fold an earlier lithological layering. The isoclines are north-south trending, upright to inclined, shallowly plunging, with axial planes dipping steeply to the east. Due to the upright nature of the folds, and the pavement outcrop, hinges are rarely observed in the field, and are generally only seen on rare vertical east-west faces, or inferred by repeated lithologies. An associated mineral stretching lineation, defined by hornblende or biotite, is sub-horizontal and parallel to F1 hinges. A slight variation in orientation of all D1 structures is due to the effects of D3 (discussed below). The scarcity of reliable kinematic indicators and the nature of the folding suggest that D1 was the result of pure shear east-west flattening. The second deformation event in the south, seems to have been a tight to isoclinal folding event

that deforms the D1 foliation, but produces structures coaxial to those developed in 01, suggesting that D2 is a continuation of the D1 event. Tight F2 folds are observed at localities such as Skippy Rock and Elephant Rock, where the exposure of hinges is enhanced by a shallow plunge to the north (although the plunge may be due to the effects of 03). The third deformation to affect the southern end of the Leeuwin Complex is a WNW-ESE trending regional open warping, with a wavelength of approximately ten kilometres. It is identified in the field through the undulation of the D1 mineral stretching lineation through the horizontal. The F3 folds plunge shallowly to the west, and this causes the D1 mineral lineations to be rotated towards the east noticeably. This deformation has effectively smeared the poles to the D1 foliation and they no longer trace a great circle. Evidence for a fourth deformation is not widespread, and consists of small near-vertical brittle faults, that trend east-west, and offset earlier fabrics by only a few centimetres. CONCLUSION Since the crystallisation of the felsic orthogneiss, there has been sub-horizontal east-west flattening, forming isoclinal and tight folds that have axial planes dipping steeply to the east. Extension has caused boudinage structures, and a strong mineral stretching lineation parallel to the hinges of F1 folds. Some stage later, these structures were overprinted by a sub-horizontal approximately north-south compression, which induced WNW-trending open folds, with a shallow westerly dip. The northern Leeuwin Complex preserves a similar structural history, with four deformation events having been identified, although axial planes tend to be more inclined and plunges slightly steeper. Between the northern and southern ends, structures tend to dip shallowly to the east, and display a pervasive foliation. The relationship between these structures will be crucial in reconstructing a history for the entire complex, and hopefully will be resolved shortly.

SGTSG Field Meeting 2003

HIGH-GRADE METAMORPHIC EQUIVALENTS OF THE AMADEUS AND GEORGINA BASINS IN THE HARTS RANGE REGION, CENTRAL AUSTRALIA David Maidment^ ^ Ian Williams^ & Martin Hand^ ^Geoscience Australia, GPO Box 378, Canberra, ACT 2601. ^Research School of Earth Sciences, ANU, Canberra ACT 0200. david.maidment@anu.edu.au, ian.williams@anu.edu.au ^Continental Evolution Research Group, Geology and Geophysics, University of Adelaide, Adelaide SA 2002. martin.hand@adelaide.edu.au

A detrital zircon SHRIMP study has been undertaken to assess the possible correlation between high-grade metasediments in the Harts Range region of the eastern Arunta Inlier, and surrounding low-grade sediments of the Amadeus and Georgina Basins. Recent work has shown that the upper amphibolite to granulite facies Irindina Supracrustal Assemblage (ISA) of the Harts Range is not Palaeoproterozoic, as had been assumed, but as young as Cambrian. Neoproterozoic to Early Ordovician stratigraphic units in the Georgina and Amadeus Basins have been sampled to test whether detrital zircon patterns are comparable to that of ISA metasediments, and whether units in the basins might be correlatives of the high-grade Aruntahosted metasediments. Four units from the Georgina and Amadeus Basins have been analysed. A sequence from the southern Georgina Basin was sampled in the Elua Range, consisting of the Late Neoproterozoic Grant Bluff Fm, the Early Cambrian Mt Baldwin Fm, the Late Cambrian Arrinthrunga Fm and the Cambro-Ordovician Tomahawk Beds. The northern Amadeus Basin sequence was sampled at Ross River Gorge and at Ruby Gap and includes the Neoproterozoic Heavitree Quartzite, the Early Cambrian Arumbera (IV) Fm, the Late Cambrian Goyder Fm and the latest Cambrian Pacoota Sandstone. Clusters of detrital zircon ages occur at --2.5 Ga, 1.9-1.7 Ga and 1.2-1.0 Ga, consistent with derivation from sources in the Arunta and Musgrave Inliers. These populations are similar to those found in ISA metasediments, suggesting that both sequences shared a similar source. There is a particularly close similarity between the detrital zircon spectra of the Tomahawk Beds and the upper amphibolite facies (700°C, 7 kbar) Brady Gneiss.

The Georgina and Amadeus basins, and ISA sequences also share a common change in provenance over time. The Neoproterozoic to Early Cambrian units of the Georgina and Amadeus Basins contain few zircons younger than --800 Ma, whilst the Late Cambrian to Cambro-Ordovician units show an increasing abundance of zircons younger than 800 Ma. Similarly, the structurally lowest units of the ISA contain few <800 Ma zircons, whilst the upper units show an increasing abundance of 550-800 Ma zircons. In contrast to the older zircon components, a lack of known sources in central Australia suggests that the 550-800 Ma zircon population is unlikely to be derived from intracontinental sources. Subsidence leading to the formation of the NW-trending Larapintine Seaway in the Early Ordovician resulted in a change in basin configuration, which probably led to an influx of sediment from the continental margin to the east. The axis of this seaway corresponds to the development of high-grade metamorphism at depth (the Larapinta Event). The similarities in depositional age and provenance changes indicate that protoliths to the high-grade metamorphics of the Harts Range region were deposited in a sub-basin (the Irindina sub-basin) initiated between the Georgina and Amadeus Basins at -'540 Ma. These sediments were rapidly buried to --20 km by 520 Ma, when extensive partial melting of the package occurred to produce localised granites. A second phase of rapid subsidence began at around 500 Ma, leading to a further -20 km of burial by 480 Ma. The currently available evidence indicates that the Larapinta Event occurred within an intracratonic rift setting, concurrent with the formation of the Larapintine Seaway at the surface.

SGTSG Field Meeting 2003

U-PB IN DETRITAL-ZIRCON LA-ICP-MS AGE DATING OF EARLY PALAEOZOIC SANDSTONES FROM THE LACHLAN OROGEN: SEDIMENT PROVENANCE AND IMPLICATIONS FOR TECTONIC EVOLUTION Richard J. Squire, School of Earth Sciences, University of Melbourne, Victoria, 3010, rsquire@unimelb.edu.au

Ian H. Campbell, Research School of Earth Sciences, Australian National University, Canberra, ACT, 0200, lan.Campbell@anu.edu.au

Charlotte Allen, Research School of Earth Sciences, Australian National University, Canberra, ACT, 0200, charlotte.aHen@anu.edu.au

Chris, J.L Wilson, School of Earth Sciences, University of Melbourne, Victoria, 3010, cilw@unimelb.edu.au

Detrital-zircon age spectra based on about 850 ICP-MS analyses for Early Palaeozoic sandstones from the western sub-province of the Lachlan Orogen indicate that major changes in zircon provenance occurred in the Early Ordovician (ca. 490 Ma), Late Ordovician (ca. 455 Ma) and probably the late Early Silurian (ca. 430 Ma). The timing of these changes correlate with regional tectonic events, although there is a surprising absence of zircons associated with Silurian or Early Devonian magmatism and volcanism. The oldest rocks examined included the quartz turbidites of the St Arnauds Group at Stawell and east of the Coongee Fault near Ararat. These sandstones are dominated by zircon ages that are Middle Cambrian or older and thus match more closely the Cambrian quartz turbidites of the Delamerian Orogen (e.g.. Glen Thompson Sandstones) than the Ordovician sandstones that occur elsewhere in the Lachlan Orogen. Interflow sediments between the MORB-like tholeiitic basalts at Stawell also display strikingly similar age spectra to the Cambrian quartz turbidites. Therefore, the basalts at Stawell were erupted into a sedimentary basin that formed part of the East Gondwana passive margin. In strong contrast, sediments intercalated with tholeiitic basalts at Mt Moornambool (south-southwest of Stawell) are dominated by Middle to Late Cambrian zircons. These zircon ages not only indicate different provenance at Mt Moornambool, but also confine the age of tholeiitic magmatism in that region to early Late Cambrian (about 500 Ma). The Monageeta Shale overlies tholeiitic basalts in the eastern Bendigo Zone, and

displays a broad zircon age peak from about 600 Ma to 500 Ma, but does not include the Grenville (900-1200 Ma) or the generally less common Palaeoproterozoic or Archaean age zircons. In contrast, the Glen Creek Sandstone that overlies tholeiitic basalts in the eastern Melbourne Zone contains no Archaean or Proterozoic age zircons, and may be distinguished by an abundance of Early to Middle Ordovician age zircons. Therefore, the Cambrian tholeiitic basalts and overlying quartz turbidites of the Lachlan Orogen were emplaced in strikingly different palaeogeographic settings and thus cannot be correlated as simply as previously thought. Upper Ordovician sandstones (Sunbury Group) in the western Melbourne Zone display strikingly similar age spectra to the lower and middle Ordovician quartz turbidites with broad peaks between about 490-650 Ma and 900-1200 Ma, although they contain a distinctive peak at about 465-470 Ma. The Middle Ordovician peak broadly coincides with the age peak in the Glen Creek Sandstone indicating that provenance for some of the detritus included rocks associated with an unrecognised magmatic event at ca. 470465 Ma. The youngest rocks examined were the Upper Silurian Kilmore Siltstone and late Early Devonian Norton Gully Sandstone from the Melbourne Zone. Both samples display similar zircon ages and may be distinguished from the Upper Ordovician sandstones by the absence of the Middle Ordovician peak. The provenance of these sandstones, therefore, matches more closely the Early to Middle Ordovician sandstones than the Late Ordovician sandstones in the Lachlan Orogen.

SGTSG Field Meeting 2003

DATING IN WESTERN VICTORIA: IMPLICATIONS FOR THE EVOLUTION OF THE LACHLAN AND DELAMERIAN OROGENS John Miller\ David Phillips\ Christopher Wilson^ and Jon Dugdale^ ^ pmd-CRC, School of Earth Sciences, University of Melbourne, Victoria, Australia, 3010. ^ MPI mines Pty Ltd, 10-16 Queen St, Victoria, Australia, 3000.

Two major components of the Tasman Fold Belt outcrop in western Victoria; the Cambrian Delamerian Fold Belt and the Ordovician to Devonian Lachlan Fold Belt. The boundary between the Lachlan and Delamerian Fold Belts has been placed at various positions ranging from the Avoca to Yarramyljup Faults in the last 25 years. The orogenic boundary is currently defined as the east-dipping Moyston Fault on the western margin of the Stawell Zone. step-heating experiments carried out on amphibole, muscovite, biotite and sericite indicate that the rocks in the hangingwall of the Moyston Fault cooled rapidly from - 500X to below 400°C at about 500 Ma at a rate of 15 to 20°C per million years. This places the entire western region of the Stawell Zone into the Cambrian Delamerian Orogen. The rocks in western Victoria are now interpreted to be a hybrid orogenic zone deformed by both Delamerian and Lachlan orogenesis. A single fault does not define the orogenic boundary. These age data also resolve the problem of the variable structural complexity between the Stawell and Bendigo 440 Ma gold deposits: the rocks at Bendigo were not deposited until after the cessation of the Delamerian deformation that affected the Stawell rocks. The location of two major Delamerian Faults (Yarramjyiup and Moyston) appears to be controlled by rifted fragments/blocks of underlying dense Proterozoic lithosphere. This block has previously been identified by chemical variations in Cainozoic volcanic rocks and P-wave velocity data. Unlike the rocks to the east and west, the Phanerozoic units overlying this block are of low metamorphic grade and were probably protected from deformation by the underlying rigid Proterozoic basement. Delamerian deformation in the Stawell Zone deformed young (<20 Ma) hot buoyant oceanic

crust producing garnet-bearing amphibolites. West of the Stawell Zone, mantle exhumation during Cambrian rifting resulted in marked thermal interactions between the crust and mantle highlighted by syn-tectonic plutonism. Localisation of strain along lithospheric boundaries during Delamerian orogenisis produced faults with substantial vertical offsets (i.e. the Moyston Fault) producing sharp changes in metamorphic grade. The deformation is characterised by craton-verging thrusting. The Lachlan Orogeny occurred about 40 million years after the Delamerian Orogeny and deformed broadly similar lithologies. The grade of metamorphism in the turbidites of the western Lachlan Orogen is markedly uniform, with evidence for transitional blueschist metamorphism previously documented in the volcanic substrate. West-dipping faults that verge away from the craton are dominant, none of the faults have large vertical offsets. Plutonism in the western Lachlan Fold Belt occurred at least 40 to 60 million years after deformation and the geometry of the western Lachlan Orogeny has previously been interpreted to represent an accretionary wedge. The marked change between the Delamerian and Lachlan styles of deformation probably reflect differences in the cooling of the underlying volcanic substrate and a reduction in the amount of thermal interaction between the crust and mantle. The Lachlan Orogen deformed older, colder and denser oceanic crust producing a distinct style of thin-skinned deformation compared to the hotter Delamerian Orogen. ACKNOWLEDGMENTS This project has been funded by the pmd-CRC and MPI mines Pty Ltd. This paper is published with the permission of MPI mines Pty Ltd and the CEO of the Predictive Mineral Discovery Cooperative Research Centre.

SGTSG Field Meeting 2003 STRUCTURAL RE-EVALUTION OF THE HILL END TROUGH, EASTERN LACHLAN OROGEN, NSW Jeffrey J Vassallo\ Richard A Glen \ Vladimir David ^ ^ The Geological Survey of New South Wales, Department of Mineral Resources, PO Box 536, St Leonards NSW 1590

Introduction: The Hill End Trough (HET) has iconic status in the Lachlan Orogen. It was the best example of a Silurian-Devonian trough identified by Gordon Packham in his classical studies of the evolution of the orogen in the 1960s. It was also the focus of considerable controversy between Chris Powell, Mike Rickard and Gordon Packham in the 1970s, regarding the age of deformation. This debate was rekindled in the 1990s by Packham and Glen & Watkins. Here we present a new model of the structural evolution of the HET, eastern Lachlan Orogen, NSW, that has considerable implications for the assembly of eastern Australia from the Silurian to the Carboniferous, as well as for controls on orogenic, sediment hosted gold and base-metal deposits. Our work is based on re-evaluation of previous geological mapping coupled with two detailed (1:12 500 scale) east-west transects across the trough. Results: Our re-evaluation of the Hill End Trough has revealed a more complex structural history than suggested by previous workers. We have identified a regional cleavage that lies at variable angles to SQ. This cleavage, labelled S I , is crenulated by S 2 , the main N - S trending cleavage that is axial planar to the regional fold set. Si is defined by aligned biotite, muscovite and chlorite grains indicating it was formed at upper greenschist facies metamorphic conditions. We infer that Si is part of a thrusting event, based on the presence of Middle-Devonian thrusts on the flanking shelves, the over-thickening of stratigraphic units (Cunningham Formation) in the trough and the presence of large domains with consistent asymmetric vergence. Computer modelling of gravity data suggests an increase in the current depth to the basement under those

asymmetric domains. One major Di thrust in the slate and metapsammite Cunningham Formation follows the approximate location of the Sawmill Hill Fault and is clearly truncated by the Carboniferous Nindethana Thrust. The Sawmill Hill Fault is coincident with a major facies change, thus suggesting that at least one Di thrust formed by the reactivation a synsedimentary growth fault. An indication of the three-dimensional shape of Di thrusts comes from the alternating flat-tosteep changes in the F2 envelope and their consistent vergence to the west in the vicinity of the Sawmill Hill Fault. This surface geometry in the F2 envelope suggests that the Di thrust has a ramp-flat geometry at depth. As S2 consistently crenulates Si, It also suggests that the Di thrusts are folded. The occurrence of deformed beddingparallel quartz tension vein arrays in many other units in the HET also suggests that many second and third order Di structures exist throughout the Hill End Trough, possibly in the Hill End and Lewis Ponds areas. Structural relationships suggest these Di shear zones control the location of prospective base metal-gold mineralisation. The main N-S fabric (now labelled S2) is . principally defined by sericite, biotite and chlorite and wraps around the syn-late Di porphyroblasts. F2 folds are upright chevron-sinusoidal, close to isoclinal folds. Carboniferous folds and faults during the Kanimblan Orogeny deformed Late Devonian sediments on both flanks of the HET and we tentatively ascribe D2 structures in the HET to this event. Acknowledgments. Published with the permission of the DirectorGeneral, NSW Department of Mineral Resources.

SGTSG Field Meeting 2003

VALIDATING SOUTH-SIDE UP MOVEMENT OF BLOCKS WITHIN THE MOLONG ZONE, EASTERN LACHLAN FOLD BELT Lennox, P.G.\ Trzebski, R.^ and Kohn, B.^ 1. School of BEES, UNSW, Sydney 2052, p.lennox@unsw.edu.au,r.trzebski@usnw.edu.au 2. School of Earth Sciences, University of Melbourne, Parkville, Victoria 3010, b.kohn@uninnelb.edu.au

The meridional Molong Zone was once part of the Ordovician Macquarie volcanic arc before it was subdivided in the Silurian. The zone contains commonly meridional faults sometimes with long and complex movement histories, is intruded by extensive I and S-type granites and the country rocks show evidence of repeated foliation development during the latest Ordovician to Eariy Carboniferous Lachlan Orogen. Blocks within the Molong Zone show evidence of south-side up movement. DISCUSSION The Carcoar & Barry Granodiorites and Sunset Hill Granite near, NSW form northern outliers of the Wyangala Batholith. These granites and their enclosing country rock display a progressive increase in foliation development and more ductile features from north to south over a 40 km distance. These eariiest Silurian granites were emplaced at 612 km into a pull-apart structure bounded by two meridional, en echelon regional faults with a slight dog-leg shape. Subsequent dominantly east-west shortening has uplifted the granites during mainly west-over-east thrusting and southward movement. Fission track studies are underway In an attempt to resolve the amount of vertical movement of the southern section of this part of the Molong Zone compared with - 40 km north. The Wyangala Batholith is approximately 160km long and 30-50km wide and consists of over 30 individual plutons within the Molong Zone. Two thirds of these plutons are S-type, about 30% l-type and the rest A-type, dykes or gabbro. Four eastwest, high resolution gravity surveys spaced about 40km apart were completed across the Wyangala Batholith. Some of the north-south elongated plutons show clear evidence of roof zones at their northern extremities suggesting; rotation about east-west axes, or differential erosion of the plutons, or original tilted emplacement. It is possible that many of the surface faults root into an upper crustal detachment at - 7km as proposed for the two southern transects. The steeply, west-dipping Copperhannia Thrust is the

major fault on the eastern side of the transects in the northern most transects and is similar in orientation to the major faults on the eastern side of the southern two transects. Pooriy constrained dating suggest sericite formed twice on the Copperhannia Thrust; once during the late Middle Devonian Tabberabberan event and again during the Eariy Carboniferous Kanimblan event. The Wyangala Fault forms a steep east-dipping, eastover-west fault on the western side of the northern most transect. A similar oriented fault occurs about 10km west of the western most end of the southern transect. These major west or east-dipping faults may root into the same or different detachments in the upper crust. Some of these major faults change dip along their length. Overall northeasteriy striking faults invariably show dextral strike-slip movement, whereas northwesteriy striking faults show sinistral strike-slip movement. This would be consistent with the overall dominantly east-west shortening observed throughout the Molong Zone. Complementary studies of the deformation observed in the various plutons and enclosing Ordovician to Silurian country rocks indicates that many of the plutons have more deformed eastern margins and show west-over-east movement on generally west-dipping faults. Some of the meridional faults have more complex histories reflecting eariy, normal dip-slip movement in some cases prior to latest Ordovician to latest Silurian granite intrusion, inversion sometimes with a change of dip and finally significant strike-slip movement. In some cases it appears there has been south-side up movement of the granites and enclosing country rocks. CONCLUSION The south-side up movement observed within the Molong Zone may have resulted from the southward tectonic transport advocated for the Benambra Terrane between the Eariy Silurian and Middle Devonian by the Geological Survey of Victoria.

SGTSG Field Meeting 2003

WEDNESDAY 24 SEPTEMBER 2003

SGTSG Field Meeting 2003

CARPET OF SILVER: THE WRECK OF THE ZUYTDORP IN 1712 Phillip Playford Geological Survey of Western Australia, 100 Plain St, East Perth WA 6004

In 1927, a stockman working on Murchison House Station, Tom Pepper, found wooden wreckage at the foot of a line of steep cliffs about 60 km north of Kalbarri. In 1954 I relocated this wreckage, following directions from Tom Pepper, and soon afterwards organized two expeditions to the site. Many silver coins were found there, including schellingen and double stuivers bearing the name Zeeland and the date 1711. Through correspondence with museums and archives in the Netherlands, Cape Town, and Jakarta, I was able to prove that this wreck was that of the Zuytdorp, a great ship of the Dutch East India Company that had disappeared after leaving the Cape of Good Hope, bound for Batavia (Jakarta), in April 1712. This was the first Dutch wreck to be found and identified on the coast of Western Australia. I formally gave the name Zuytdorp Cliffs to the line of precipitous cliffs that extend north from Kalbarri to Steep Point, a distance of some 200 km. These cliffs form the eroded scarp of the Zuytdorp Fault, perhaps the most prominent Quaternary fault scarp in Australia. Clear evidence was found at the wrecksite that many people survived the wreck. They climbed to the top of the cliff, lighting a huge fire and indulging in a drinking spree. They left many broken gin bottles and various other items. Three survivors' camp sites were identified inland from the wreck, and there is evidence that some people may have reached a large spring

frequented by Aboriginal people of the Malgana Tribe, 50 km north of the wreck. The Zuytdorp was carrying 250,000 guilders in cash, to be used for trade in Asia. This coinage was kept in chests stored in the captain's cabin, and it is clear from wreckage on the seafloor that the chests went straight to the bottom after the wreck came to rest in front of the cliff. When divers first examined the site, on one of the very few days each year that it is possible to dive there, they found that much of this coinage was preserved as a 'Carpet of Silver' on the seafloor. This consisted of hundreds of thousands of silver coins polished by sand and wave action. Several thousand coins have since been recovered by divers of the WA Maritime Museum, but the major part of the Carpet of Silver has been taken by looters. There are two major unsolved mysteries relating to the Zuytdorp: the fate of the survivors and the whereabouts of the looted coinage. It is certain that no survivors ever returned to civilization, and they must eventually have died in Western Australia. There is an intriguing possibility that some may have joined and interbred with Aborigines of the area, a question that may eventually be solved through DNA research. In relation to the looted coinage, there is no proof as to who was responsible or where the coinage has gone, although there are a number of strong indications.

SGTSG Field Meeting 2003

THURSDAY 25 SEPTEMBER 2003

SGTSG Field Meeting 2003

IMPACT OF STRUCTURE AT THE RESERVOIR SCALE , BARAM DELTA PROVINCE, NW BORNEO Chris Morlev\ Mark Tingay^, Pieter Van Rensbergen^ and John Warren^ ^ Department of Petroleum Geoscience, University of Brunei Darussalam, Brunei Darussalam ^ NCPGG, The University of Adelaide, South Australia 5005, Australia ^ Renard Centre of Marine Geology, Universiteit Gent, Krijgslaan, 281-s8, 9000, Gent Belgium

The structural studies at the reservoir rock-scale in Baram Delta Province of NW Borneo have addressed four main geological aspects 1) the effects of faults as seals to hydrocarbons, 2) reduction of porosity and permeability by deformation bands associated with faulting, 3) the type and distribution of overpressure, 4) the impact of shale intrusions on reservoir compartmentalisation and imaging and 5) the timing of fluid (including hydrocarbon) migration events. In NW Borneo the basic gravity deformation style of deltas (growth faults, shale diapirs, toe thrusts) has been modified by the presence of inversion anticlines within more proximal parts of the system. A range of studies using outcrop data, seismic reflection data, well log data and geochemistry, have enabled the distribution of different types of overpressure (disequilibrium compaction, inflation) to be linked with different structural provinces, and the structural evolution of the delta. Varying stress conditions both temporally and spatially appear to have resulted in the most likely fluid pathways being either perpendicular to depositional strike (NW-SE) parallel to it (NESW). Different types of fluid migration (within shale diapirs, shale pipes, hydraulic fracture complexes, gas chimneys) can sometimes be identified from seismic reflection data. Commonly reactive diapirs appear to later evolve lateral and vertical intrusive complexes within country rock that: 1) dim seismic reflections resulting in overestimation of the size of the real mobile shale mass, 2) intrude fault planes,

enhance their sealing potential, 3) compartmentalise reservoirs by shale-filled hydraulic fractures, and 4) pump fluids into higher reservoirs. Isotope analysis indicates associated fluid migration commonly has left carbonate cement layers and nodules within reservoirs. Growth faulting affects poorly lithified syn-kinematic strata and better lithified prekinematic strata, changing the deformation and sealing properties of the fault with depth. Deformation of poorly lithified sediments by even small faults with a few metres displacement can impact reservoir quality due to the development of broad zones (up to 20 m wide) of deformation bands that significantly reduce average porosity, and are likely to impose a directional permeability. Shale smear is a common feature of fault planes, sand smear also occurs in poorly lithified units. Inversion of normal faults tends to breach earlier fault seals. Fortunately reactivation is very selective, only a few faults in any region tend to be inverted. Hydrocarbon reservoirs sealed by normal faults lie adjacent to inversion faults with no associated trapped hydrocarbons. There is tremendous potential complexity to the impact of structural events on reservoir geology, not only due to a variety of faulting-related aspects, but to the way structural events have triggered and controlled overpressure, shale intrusions and the timing and directionality of fluid migration events

SGTSG Field Meeting 2003

PROTO-OROGENESIS: THE NORTH WEST SHELF LEGACY Mvra Keep\ Ian Longley^, Mat Harrowfield^ and Warwick Crowe^ ^Tectonics Special Research Centre, School of Earth and Geographical Sciences, M004, University of Western Australia, 35 Stirling Highway, Nedlands, 6009. mkeep@tsrc.uwa.edu.au. ^Woodside Energy, 1 Adelaide Terrace, Perth, 6000.

Detailed seismic interpretation and tectonic analyses of areas around the North West Shelf of Australia have sought to document the nature, location and intensity of Neogene deformation. Results have shown that distinct changes occur in deformation style around the margin, and that basement topology controls strain distribution. The most important conclusion of this work, however, is the documentation that at this active collisional margin, shortening strain is so small as to be negligible (1-2%). We use the term protoorogenesis to describe the relatively small strains during collision, and note that the intensity of this deformation varies widely around the margin. Pre-existing rift compartments strongly control the Neogene response of Australian cratonic material to collision. The basement topology partitions strain and controls relative uplift within compartments. In the adjacent Browse Basin, discrete zones of high strain occur, manifest as transpressional and inversion structures that occur in narrow channels between basement highs. These zones preserve some of the most

intense shortening strain yet recorded for the region. Further to the southwest, interaction between the Yilgarn and Pilbara blocks during the Neogene caused discrete deformation in the Canning and Carnarvon basins, including zones of relatively intense transpression. Overall, regional maps of Neogene collisional shortening structures show strain accommodated in narrow zones, outboard in the Timor Sea and inboard in the Carnarvon Basin. Pulses of Neogene deformation at 25 Ma, 8 Ma and 3 Ma correlated well with regional tectonic activity. The 8 Ma event appears to be more widespread than the others. This 8 Ma event is attributed to collision of an outer promontory of Australian continental crust with the Sunda Arc. Attenuation and uplift of this promontory facilitated the coincident uplift and underplating of Sumba Island, and probably absorbed a significant proportion of the collisional strain.

SGTSG Field Meeting 2003

SINGLE TECTONICS - A NEW PARADIGM FOR THE NEOCENE OF NORTHWEST AUSTRALIAN Warwick Crowe, Myra Keep, Matt Harrowfield, Nathan Palmer and Jane Cunneen Tectonics, Special Research Centre, School of Earth and Geographical Sciences, M004, University of Western Australia, 35 Stirling Highway, Crawley WA 6009; wcrowe@tsrc.uwa.edu.au

The significance of Neogene deformation in trap formation, modification and/or breach along the North West Shelf has long been recognised. However our understanding of the structure and tectonics in the Neogene is at best superficial, particularly within the Canning and Carnarvon basins, especially when compared to that in the late Paleozoic and Mesozoic. Early Miocene to present tectonism along the NW shelf of Australia, attributed to the collision between the northern margin of Australia and elements of the Pacific Plate, show a distinct change in character between the Timor Sea region at the collisional edge and the intraplate shelf margin of the Carnarvon Basin to the SW. Dispersed normal fault deformation along the collisional outer shelf margin in the Timor Sea region contrasts markedly in both style and spatial distribution with coeval deformation in the Roebuck (offshore Canning) and Carnarvon basins, including possible fault reactivation along the Fitzroy Trough in the Canning Basin. Prominent NE-SW-trending Neogene strikeslip fault zones transect the length of the offshore Canning Basin and step inboard across the Beagle Sub-basin to a sinistral tranpressional zone along the Enderby Terrace, the effective margin of the Pilbara Block. This transpressional deformation, typified by the development of anticlinal inversion structures, extends SE along the inboard margin of the Dam pier and Barrow sub-basins. Strain from this localised transpressional deformation appears to have dispersed through a series of faults and anticlines which splay from the Long Island Fault Zone in the NW Cape region.

transiently imposed resistance at the leading edge of the Australian plate during early Miocene plate reorganisation. In this model deformation was internally driven through minor intraplate convergence between the Pilbara and Kimberley blocks. Transpressional strain was localised around the NW inboard margin of the Pilbara Block, and strain was partitioned through the Roebuck Basin along localised strike-slip faults into the southern Browse Basin. Within the onshore Canning Basin strain was partitioned along reactivated major bounding faults of the Fitzroy Trough. NW-SE-trending strike-slip zones within the Roebuck Basin accommodated strain between the northwestern Carnarvon Basin and in the northern Canning Basin. At the SW corner of the Leveque Shelf (Kimberley Block) is a complex zone of deformation where reactivated faults along the Fitzroy Trough and in the Roebuck Basin have converged and dissipate through the Browse Basin along discrete structures within the Barcoo Sub-basin. Persistent Mid-Miocene to Pliocene deformation in the Carnarvon, Roebuck and possibly Canning basins may be attributed to a major compressional pulse between 12 Ma and 4 Ma associated with oblique convergence between the Caroline Plate and Papua New Guinea. A spatial relationship with the distribution of recorded earthquake foci and the known Neogene deformation, in particular to the SW of the Leveque Shelf, at the convergent point of deformation in the Fitzroy Trough and Roebuck Basin, supports a single deformation regime that began in the latest Oligocene and continues into the present.

We attribute strain localisation within the intraplate shelf margin of the NW shelf to a limited inertial response of the Pilbara Craton to a

SGTSG Field Meeting 2003

EVOLUTION OF THE NEOGENE FAULT POPULATION, LAMINARIA AREA, TIMOR SEA Jane Cunneen and Myra Keep Tectonics Special Research Centre, School of Earth and Geographical Sciences, M004, University of Western Australia, 35 Stirling Highway, Nedlands, 6009. icunneen@tsrc.uwa.edu.au.

The interaction between reservoir-level Jurassic faults and Neogene faults in the Timor Sea is poorly understood. A fault population analysis of Neogene faults in the Laminaria area uses high resolution 3D seismic data as a tool for determining fault linkage properties and the likelihood of Neogene trap breach. The evolution of the Laminaria fault population is described in terms of fault initiation, fault growth, and strain partitioning. Fault initiation occurred in two stages. Large EW-trending faults initiated in the Early Miocene, with smaller ENE-WSW-trending faults starting in the Early Pliocene. The mechanisms for fault growth changed with development of the system. The early stages of the fault system consisted of small fault segments, which grew by radial propagation of the fault tips. The main stage of fault growth was achieved with growth by segment linkage during the Miocene, and it was during this time that the present-day fault lengths were established. Following the linkage of fault segments, fault growth was achieved by through

an increase in cumulative displacement during the Pliocene and Early Pleistocene, without large increases in fault length although there was likely some growth by radial propagation of the fault tips, especially in the dip direction. Partitioning of strain onto the major faults occurred from the Late Pliocene and only the major faults remain active today. Bulk Neogene extension across the Laminaria area Is about 3%, which is considered a low-strain environment. Extension measured from cross sections is up to 50% greater when the major structures are intersected, suggesting strong partitioning of strain onto the faults bounding the Laminaria High and Vidalia Graben. Although Jurassic horst structures exert a geographic control on the location of Neogene faults, displacement-length relationships indicate that Neogene faults are unlikely to penetrate the Mesozoic sequence, decreasing the likelihood of Neogene trap breach.

SGTSG Field Meeting 2003

STRAINSPOTTING IN THE TIMOR SEA: NEOGENE COLLISION AND FLEXURAL MODIFICATION OF AUSTRALIA'S PASSIVE MARGIN Mat Harrowfield\ Myra Keep\ Jane Cunneen\ Warwick Crowe^ and Neil Thompson^ 1. Tectonics, Special Research Centre, School of Earth and Geographical Sciences, M004, University of Western Australia, 35 Stirling Highway, Crawley WA 6009; mharrowfield@tsrc.uwa.edu.au: 2. Woodside Energy, 1 Adelaide Terrace Perth, neil.thompson@woodside.com.au

Australia's North West Shelf is a MesozoicRecent passive margin constructed upon a failed Permo-Carboniferous rift. Since the early Neogene, the North West Shelf has been pervasively modified by ongoing collision between the Australian continent and the Eurasia/Pacific island arc. From the Timor Sea region, we describe two components to this modification: 1) localised brittle dissection of the upper crust and passive margin cover; and 2) long-wavelength continuous amplification of deep-seated basement topology. At shallow levels, bi-vergent normal faulting created horst and graben geometries, reminiscent of extensional deformation. At deeper levels, amplification of basement topology juxtaposed Palaeozoic/Mesozoic and Neogene depocentres and renewed apparent uplift of outboard platforms. This amplification is consistent with ductile shortening. Rudimentary restoration of structural cross-sections and pre-tectonic prograde anatomies identifies less than 1% strike-normal length change and suggests depocentre subsidence was an apparent artefact of net contraction. We interpret Neogene modification to reflect flexural collapse of detached sedimentary

cover above an irregular, ductilely-shortened basement. This basement is thought to be the buried relic of the Permo-Carboniferous rift. Local variations in the style and intensity of Neogene modification belie compartmentalisation of this inherited basement and its underlying lithosphere. Neogene collision was predominantly buttressed by the salient upper-plate compartment of the Bonaparte Basin, the hardlinked basement architecture of which induced a strong structural inheritance within the collapsed pile. This protruding lithospheric relic was the bulldozer blade against which the Timor accretionary prism accumulated. In the adjacent lower-plate embayment of the Browse Basin, reduced strain and an amorphous soft-linked basement resulted in weak flexural sag and independent deformation of the sedimentary cover. We draw significant parallels between Mesozoic and Neogene modification of the North West Shelf and consider its evolution as the juxtaposition of two generations of continuous long-wavelength topology.

SGTSG Field Meeting 2003

EVOLUTION OF FLUID PATHWAYS DURING DEFORMATION OF THE MURRUMBIDGEE GROUP, TAEMAS AREA, LACHLAN FOLD BELT, NSW Stephen F Cox Department of Geology and Research School of Earth Sciences The Australian National University Canberra, ACT 0200, Australia (e-mail: sfcox@geology.anu.edu.au)

Structural and stable isotope studies of a one kilometre thick, Lower Devonian carbonate sequence (Murrumbidgee Group) in the Taemas area of the Lachlan Fold Belt (south-eastern Australia) indicate that externally-derived fluids migrated through the sequence during upright folding and associated reverse faulting at depths of several kilometres and temperatures in the range 150°C to 200°C. The evolution of fluid pathways during crustal shortening was controlled by growth of fault-related and foldrelated fracture networks at transiently supralithostatic fluid pressures. Systematic changes in 0-isotope compositions of veins up through the carbonate sequence are related to buffering of externally-derived fluid compositions by progressive reaction with the host rocks along the structurally-controlled fluid pathways. At the base of the carbonate sequence, calcite veins are depleted by up to 23%o relative to unaltered host-rock limestones. ^®0-depleted alteration haloes up to 20 metres wide in wall-rock in areas of intense vein development are related to fluid discharge from faults and associated vein arrays. Higher in the carbonate sequence, fault- and fold-related veins typically exhibit progressively less depletion relative to the distal host rocks. In the upper parts of the sequence, vein is usually less than 1-2%o less than unaltered host rocks. Systematic depletion of vein by up to 23%o also occurs immediately adjacent to the high displacement Warroo Fault, which bounds the eastern side of the carbonate sequence. Decreasing depletion within faults upwards through the carbonate sequence indicates vein formation was associated with upwards infiltration of fluids having an initial of -8%o. These fluids are interpreted to be evolved meteoric fluids or formation waters which migrated through the Black Range Group, a volcanic sequence underlying the Murrumbidgee Group. Reactive transport modelling of this flow system indicates

time-integrated fluid fluxes of approximately 10^ moles H2O cm"^ At the eastern boundary of the area, fluids which percolated through parts of the Warroo Fault also had an initial of -8%o. This fault is interpreted to have tapped fluids from the same reservoir which supplied fluids that migrated through faults at the base of the Murrumbidgee Group during contractional deformation. Within-site variations in between veins are interpreted in terms of variations in (1) relative timing of formation of veins during progressive migration of the geochemical front through the sedimentary sequence, (2) changes in connectivity between the backbone part of the fracture-controlled flow network and developing fracture arrays, and (3) local variations in stable isotope compositions of host rocks along fluid pathways. Systematic variations in stable isotope compositions of veins indicate that most faults were well-connected to an external fluid reservoir. This suggests that growth of fault and fracture networks has been driven largely by invasion of high pore fluid factor fluids. A fault- and fracture-related flow regime involving a combination of (1) near-lithostatic fluid pressures, (2) upwards migrating fluids and (3) near-surface derivation of fluids, requires that flow was topographically-driven. For topographically-driven downflow at nearhydrostatic fluid pressures in the recharge part of the system, near-lithostatic fluid pressures can develop in the discharge part of the system provided low, time-averaged permeabilities throttle fluid outflow, and provided the elevation difference between recharge and discharge areas is at least 3km.

SGTSG Field Meeting 2003

QUATERNARY TECTONISM IN THE CARNARVON BASIN WESTERN AUSTRALIA Phillip E Playford Geological Survey of Western Australia, 100 Plain St, East Perth WA 6004

Quaternary tectonism has played a major role in shaping the coastal and near-coastal geomorphology in the Carnarvon Basin, between Barrow Island and Kalbarri. East-west compression has resulted in long broad anticlines in Tertiary and Cretaceous rocks, along an arcuate belt trending northeast at Barrow Island, north-northeast at Cape Range, north at Cape Cuvier, and north-northwest at Shark Bay. In contrast, the Zuytdorp Cliffs are thought to be the eroded scarp of a normal fault that moved during the latest Pleistocene to early Holocene. Evidence of Cainozoic tectonism continues into the Perth Basin, where the Darling Scarp is probably a result of Tertiary and perhaps early Pleistocene movement along part of the Darling Fault. The anticlines between Barrow Island and Shark Bay have formed over reverse faults that resulted from reversal of movement along older normal faults. Compression began in the Eocene and continued into the Quaternary. This phenomenon was first recognized in Rough Range South no. 5 well, which penetrated a fault with displacement that is normal in Jurassic and Permian rocks and reverse in the overlying Cretaceous and Tertiary. The west side of Cape Range Anticline shows a series of four shoreline terraces, the highest (and oldest) being up to 60 m above sea level, while the others fall step-by-step to about 5 m in the lowest, which has been dated as 123,000 years B.P. The other terraces are probably older Pleistocene in age. They show evidence of mild warping parallel to the anticlinal axis and are thought to have developed during successive sea-level highstands, while the anticline was rising. The west flank of Cape Cuvier Anticline, 250 km south of Cape Range, similarly shows folding of a coral reef thought to have grown during the last interglacial period. Edel Land, Peron, and Nanga Peninsulas in Shark Bay are developed in Pleistocene and Holocene eolianites. Drilling at Dirk Hartog Island, along the same trend as Edel Land Peninsula, has shown that the Quaternary section there overlies an anticline in Tertiary limestones with its axis parallel to the length of the island. This anticline is probably

associated with reverse faulting at depth, and the two peninsulas are expected to have had similar origins. On the west side of Hamelin Pool the crests of Holocene intertidal stromatolites show six levels, the highest being about 80 cm above sea level. The oldest (and highest) stromatolites are entirely dead and are being actively degraded, the youngest (and lowest) are fully living, and those in between are partly living and partly dead. The different stromatolite levels have probably formed as a result of periodic uplift due to folding of the Nanga Peninsula Anticline during about the past 5,000 years. The en-echelon series of cliffs between the Kalbarri area and Dirk Hartog Island are thought to be eroded fault scarps, defined by Late Quaternary (perhaps early Holocene) normal faulting. The most prominent of these scarps form the Zuytdorp Cliffs, a sharply defined line of precipitous cliffs, in Tamala Limestone, forming one of the most spectacular features of the Australian coast. The cliffs extend for some 170 km from just north of Kalbarri to Dulverton Bay. North of there the cliff line is offset some 5 km east, and continues to Steep Point and along the west coast of Dirk Hartog Island. South of Kalbarri the linear cliff line is offset to the west along more irregular cliffs of Tumblagooda Sandstone (Ordovician or earliest Silurian) that extend southwest for some 15 km to Bluff Point. From there the cliffs continue south-southeast for 30 km to Shoal Point, and are again sharply defined, suggesting relatively recent fault movement. The Hardabut Fault, which trends northeast some 40 km east of Kalbarri, operated as a normal fault during the early Palaeozoic, and as a reverse fault during the Quaternary. Reversal of movement along this fault has uplifted the plateau to the west, causing the Murchison River to cut a spectacular gorge into the Tumblagooda Sandstone. There is no clear relationship between these Quaternary structural features and modern seismicity in Western Australia. The reason for this apparent quiescence is not clear, but it could be explained partly by the lack of seismic recording stations in relevant areas.

SGTSG Field Meeting 2003

LINKING EARTHQUAKES TO GEOLOGY: CONTEMPORARY DEFORMATION CONTROLLED BY ANCIENT STRUCTURE Dan Clark\ Mike Dentith^ and Mark Leonard^ ^Geoscience Australia. GPO Box 378, Canberra ACT, 2601, dan.clark@ga.gov.au, ^Dept. of Geology & Geophysics, University of Western Australia, WA, 6009.

In most intraplate regions, such as Australia, there is a general lack of understanding of why earthquakes occur where they do, or even why they occur at all. The lack of a definitive model, at any scale, describing Australian seismicity imposes significant limitations upon the methods that can be employed to estimate earthquake hazard. Perhaps the most unambiguous link between an earthquake event and geology occurs where the earthquake is sufficiently large to produce a surface rupture. Five such surface ruptures occur within 250 km of Perth, Western Australia (WA), in an area of ongoing seismicity known as the Southwest Seismic Zone (SWSZ). Importantly, these scarps occur in an area of low, rolling topography where there is no evidence in the landscape for regular or repeated displacement. The largest historical event to produce a known surface rupture in the SWSZ was the 1968 Ms 6.8 Meckering earthquake. A comparison of the 37 km long fault scarp and new high-resolution aeromagnetic data shows exceptional agreement. Virtually every feature in the scarp correlates with a basement feature mapped in the magnetics. It is clear that two basement trends exerted particular control on the faulting: NE-SW trending dykes, and NW-SE trending lithological contacts. Both trends accommodated reverse displacement. Two important observations with relevance to SWSZ seismicity models can be made from this data: 1. The surface rupture exploited pre-existing lines of weakness ~ no new faults were formed. It is clear that these structures were suitably oriented for reactivation in the prevailing E-W compressive stress field. 2. Movement on highly oblique intersecting faults, as occurred during the Meckering rupture, is self-limiting - the intersections between oblique faults tend to lock up after one or two large events, and seismicity must migrate elsewhere. A hypothesis of self-limiting seismicity, where movement along a major fault trend triggers movement on highly oblique intersecting faults, which subsequently lock up the intersections, is

proposed to account for the Meckering surface rupture and the lack of significant tectonic geomorphology in the area. This hypothesis has a potentially wide application to seismically active areas where relief is not developed (e.g. Tennant Ck). It could also explain presently aseismic areas bearing Quaternary fault scarps. The applicability of the model to this last circumstance has been tested over the prehistoric Hyden reverse fault scarp, ^250 km east of Perth. The linear scarp is about 30 km long and corresponds exactly to a N-S trending basement geological feature revealed in highresolution aeromagnetic data, suggesting that the rupture which formed the scarp exploited a preexisting structure. The N-S magnetic lineation forms part of a set that intersects a second NWSE trending set of linear basement trends. Both sets show evidence for having accommodated displacement. There is no present seismicity associated with the Hyden scarp - perhaps it is a more evolved system than Meckering which has already locked up! The hypothesis is also being tested in an area of current seismicity to the north of the 1979 Cadoux surface rupture, near Burakin. Over 18,000 events have been recorded in this area in the last two years. It is expected that an ongoing program of precise location of events will allow the failure surfaces to be mapped in 3-D, such that the orientation of the host faults can be quantified, and compared to features apparent on aeromagnetic images. Equally important, the temporal pattern of seismicity (i.e. fault failure) is being recorded. A dense GPS network established in the region will allow for an independent measure of strain distribution. The model proposed as a result of this work, if proven to be valid by ongoing research and future studies, has fundamental implications for how future seismic hazard assessments will be made in WA. Similar research is being conducted in other areas of Australia where Quaternary deformation is apparent to develop appropriate regional seismicity models for use in the next generation of seismic hazard maps.

SGTSG Field Meeting 2003

UTILISATION OF 3D STRUCTURAL RESTORATION TECHNOLOGY TO ADDRESS FAULT SEAL ISSUES IN THE TIMOR SEA Anthony Gartrell and Mark Lisk CSIRO Division of Petroleum Resources, ARRC, 26 Dick Perry Ave, Technology Park, Kensington, Perth, WA 6151, Australia

Introduction Leakage of hydrocarbons due to post-rift fault reactivation has been identified as the principle risk for hydrocarbon exploration in the Timor Sea, northwest Australia. Consequently, considerable effort has been applied in order to develop methodologies for assessing the risk of trap failure due to fault reactivation on a given prospect and to develop regional models in order to guide explorationists to the best prospects in the region. Here we look at some new approaches being used to provide better constrained trap integrity models by utilising 3D restoration technology. Stress History Estimation Evaluation of the contemporary stress field is commonly used as a tool to assess the risk of hydrocarbon losses due to fault reactivation in the Timor Sea. This approach considers that in a population of faults and fractures, those that are critically stressed with respect to the in situ stress tensor are more likely to act as conduits for fluid transmission. However, if the orientation or magnitude of the local stress tensor has changed significantly since the onset of hydrocarbon charge, then a trap assessed as secure in the present day may have actually failed during an earlier period. In order to obtain a more continuous record of the regions stress history, palaeostress conditions have been obtained by combining 3D structural restoration and fault slip inversion techniques. Restoration of subsurface faults, imaged in 3D seismic data, is used to obtain fault slip information. A computerised inversion procedure (TENSOR) is then used to

derive an optimised (reduced) stress tensor consistent with the measured fault slips. Early results using this technique indicate that the stress field in the Timor Sea has changed significantly since the onset of oil charge. Therefore, studies based on contemporary stress systems alone may lead to incorrect trap integrity assessments. Integrated Structural and Fluid Histories An important part of establishing a useful trap integrity model is to test it against the available fluid flow data. Integrated structural and fluid flow histories have been achieved by combining fluid inclusion techniques for mapping palaeohydrocarbon contacts with 3D structural analyisis and restoration at trap scale. Palaeo-oil-water contacts determined from GOI ™ (Grains containing Oil-bearing fluid Inclusions) data were used as a physical datum that is time specific to hydrocarbon charge. Progressive restoration of the trap until an originally flat palaeo-oil-water contact is obtained provides a 3D snapshot of the field at the time of initial oil charge. In this way, the charge history is linked directly with structural history, so that it is possible to identify the trap configuration (e.g. volume, areal extent, spill points) and likely migration pathways at the time of charge, as well as isolate subsequent modifications (e.g. tilting, fault reactivation) to the trap. Initial results using this approach provide evidence that challenges regional paradigms and highlight the critical role that fault intersection can play in providing highly effective fluid conduits.

SGTSG Field Meeting 2003

STRESS TRANSFER, FLUID FLOW AND TARGET SELECTION OF FAULT-HOSTED MINERAL DEPOSITS S, Micklethwaite\ S.F. Cox\ B. Davis^ & M. Coutts^ ^ Research School of Earth Science, Australian National University, Mills F^oad, Canberra, ACT 0200. Australia, email: Steven.Micklethwaite@anu.edu.au, SFCox@geology.anu.edu ^ RSG Global, PO Box 1671, West Perth, WA 6872. Australia, email: Brett.Davis@rsgglobal.com ^ Placer Dome Asia Pacific - Kalgoorlie West Operations, P.O. Box 1161, Kalgoorlie, WA 6433. Australia, email: Michele.Coutts@auriongold.com.au

During an earthquake on a seismogenic fault, the volumes of rock around the rupture undergo changes in static stress. These changes are due to shear stress relief on the mainshock rupture surface. Those volumes that are brought nearer to failure are closely associated with aftershocks. In this way the distribution of aftershocks around seismogenic fault ruptures have previously been successfully modelled on the Anatolian and San Andreas fault systems. We argue here that contemporary seismogenic fault systems can be used as an analogue to understanding paleo-fault systems. Thus the fault-hosted goldfields of the Kalgoorlie terrane, Western Australia, are best understood in this framework. In the Mt Pleasant goldfield, deposits are located on small displacement faults around the much larger displacement Black Flag fault (BFF). The goldfield is clustered over a short section (<10km) of the >30km long BFF and deposits are found in a range of lithologies. Field and map relationships show that the BFF is a dextral strike-slip structure, with two segments dying and linking at a dilatant jog where the goldfield is

located. Deformation and mineralisation were coeval in lower greenschist facies conditions ('-300°C, upper mid-crustal depths). Stress transfer modelling of segment-sized ruptures on the BFF at --lOkm depth, shows close spatial association of small faults and gold deposits with zones of increased proximity to failure (predicted aftershock location). Thus flow of mineralising fluids around the Black Flag fault was not uniform, but localised in specific sites where small-displacement structures were activated by aftershocks. Two main conclusions can be drawn from this research. Firstly, aftershock zones are expected to produce transiently permeable regions through which fluids can flow, or in which fluid mixing can occur. Secondly, stress transfer modelling, previously used for earthquake hazard prediction, has potential as a target prediction tool for the mineral industry.

SGTSG Field Meeting 2003

MODELLING DEFORMATION & FLUID FLOW ASSOCIATED WITH THE WALLABY GOLD DEPOSIT MINERALISING EVENT. Warren Potma \ Matthew Rovardi ^ Roberto Weinberg I ^ CSIRO Exploration & Mining - Computational Geoscience Group, pmd*CRC, warren.potma@csiro.au ^Camborne School of Mines, mattyrovardi@hotmail.com ^Monash University, weinberq@mail.earth.monash.edu.au

INTRODUCTION The Wallaby Gold Deposit (Laverton WA), first defined in 1998 and developed under the Placer Granny Smith Joint Venture, now contains a total open pit and underground resource of >7Mozs at 3.3g/t Au. This work presents the results of numerical modeling aimed at defining the links between fracturing and fluid flow responsible for the formation of the Wallaby lode gold system. BACKGROUND The Wallaby ore body is a structurally distinctive ore system. Unlike other gold deposits in the Yilgarn, mineralization is confined to a pipe-shaped alteration zone within the Wallaby Conglomerate, a lithologically and structurally homogeneous unit. The main mineralized lodes form a series of stacked shallowly NE dipping tensile features which are highly altered, veined and locally brecciated, and are, in turn, linked by an array of steep brittle mineralized structures. The lodes are exclusively developed within a pre-existing annular actinolite-magnetite alteration pipe, which plunges moderately to the south and is centered around a syenite dyke suite. This pre-existing alteration is thought to be the catalyst for localizing deformation and mineralization, by creating an anomalously brittle rock mass. PROCESS The numerical models to be presented aim to understand what geodynamic conditions were required to reproduce the observed distribution of tensile failure structures. Calculation meshes were developed depicting the geometry of the Wallaby deposit for use in Flac finite difference code modeling. These meshes comprised a moderately dipping pipe structure (representing the act-mag alteration halo) with moderate tensile strength and high permeability. This pipe was flanked by low permeability, high tensile strength rocks and overlain by an impermeable seal (representing

the shale sequence overlying the mineralization at Wallaby). All model runs incorporated horizontally directed shortening, in keeping with the geologic interpretation for the region. Initially, dry models were run to test the base line strain state of the models. The models were then run with various static pore fluid pressure conditions (from hydrostatic to lithostatic). Finally, several different fluid influx rates were applied at the base of the Wallaby pipe generating an over-pressured system. The modeled materials had an elasto-plastic rheology, where pore pressure has the ability to trigger yielding by decreasing the mean effective pressure. RESULTS Models run under static pore fluid pressure conditions failed to generate any tensile failure in the Wallaby Pipe. They did, however, generate shear structures in similar orientations to those observed in the field. Under supra-lithostatic pore pressure conditions, induced by fluid influx at the base of the pipe, a stacked series of shallow dipping tensile failures developed within the models. At even higher fluid influxes, these shallow dipping tensile failure zones were linked by steep fracture zones, indicative of those observed in the deposit. CONCLUSIONS Coupled deformation and fluid flow modeling of the Wallaby deposit indicates that the observed tensile structures which host the gold lodes require supra-lithostatic pore fluid pressures, induced by a combination of fluid influx at the base of the pipe and the effect of the shale unit seal above the pipe to propagate. The horizontal and vertical structures correspond to what is observed at Wallaby. The effect of these structures is to enhance permeability and focus fluid flow in the system.

SGTSG Field Meeting 2003

EFFECTS OF FLUID FLOW RATE VARIATION ON FAULTRELATED MINERALISATION STYLE E v a n s , K . A , \ O r d , A.^ ^ CSIRO Exploration & Mining, Box 312, Bayview Avenue, Clayton, VIC 3169, Australia ^ Predictive Mineral Discovery Cooperative Research Centre, CSIRO Exploration & Mining, PO Box 1130, Bentley, WA 6102, Australia

Fluids flowing through rocks during mineralising events profoundly affect rheological, mineralogical and chemical characteristics of those rocks. A predictive understanding of the mineralisation requires an ability to relate the details of the flow to its mineralising effects. Current analytical methods provide, in most cases, only a time-integrated picture of the fluid flow, and do not allow pertinent details, such as fluid flow rates, flow event episodicity and the temporal nature of chemical concentration gradients to be elucidated. Yet these are vital pieces of Information. Fluid flow rates and the local stress regime are interdependent, related via complex feedbacks through the rheological properties of the rock which, ultimately, determine the volume of rock that a fluid interacts with. This volume, in turn, fixes the fluid:rock ratio and the degree to which the system is fluid or rock buffered, with consequences for the isotopic and trace element signature of the mineralisation. We propose that flow geometry, mapped by alteration, could act as a broad indicator of the product of flow and strain rates. Low flow.strain values would be expected to correlate with pervasive flow and disseminated mineralisation, while higher values would be associated with fracture flow and space filling or extensive replacement mineralisation.

This study examines the consequences of current conceptual models for predicted geometries and chemical characteristics of mineralisation. Numerical modelling is used to place upper and lower limits on rheological parameters, deformation and flow rates consistent with endmember styles of fluid flow in a variety of rock types. Particular attention is paid to the response of faulted rocks to variations in fluid flow rates and the way in which they can act as a flow rate filter, transforming a slow steady flow input into an episodic input. The possibility of using mineralisation styles as a diagnostic to draw conclusions on variations in flow rates is investigated, as is the prospect that variations in flow rate in a host rock may reflect characteristics of the fluid source. For example, a slowly crystallising magma at depth might be expected to produce a slow steady supply of fluid, whereas the tendency of metamorphic devolatilisation to proceed in a number of discrete steps would give a contrasting style of fluid infiltration. The results of this work are illustrated by comparison with Carlin and Archaean Greenstone type deposits.

SGTSG Field Meeting 2003

SCENARIO TESTING OF FLUID-FLOW AND DEFORMATION DURING MINERALISATION: FROM SIMPLE TO COMPLEX GEOMETRIES Peter Schaubs\ Alison Ord\ Gordon German\ ^CSIRO Exploration and Mining, Australian Resources Research Centre, P.O. Box 1130, Bentley, W.A. 6102 We present the use of FLAC3D in conjunction with GoCad and 3DMACS to model deformation -fluid processes during mineralisation. Geometrically simple models are aimed at determining the effects of a number of deformation scenarios on volume strain, pore pressure and resultant fluid flow patterns. The initial model is made up of a simple fault region bounded by steeply dipping hanging- and footwall rocks. These rocks are truncated by an unconformity and flat-lying sandstone unit. Variables which are modified in the models include the style of deformation (e.g. compression, extension, strike slip, etc.), the dip of the fault, the strike of the fault with respect to the far-field stresses, and the mechanical properties and permeability of the footwall versus the hanging wall rocks. In this way we are able to assess which parameters are most important for causing fluid to flow either up or down the fault. The results of the models show that a low angle fault with permeability similar to the surrounding host rocks causes the fault to dilate and fluid to flow down from the sandstones into the fault. Steeply oriented faults, strike-slip deformation and high permeability faults cause fluid to flow up the fault. This has implications for the location of fluid mixing and mineralisation if it is assumed that the unconformity represents a boundary between two distinct fluid reservoirs. In the complex model, contacts between different units are smooth curved surfaces and are non planar. The objective of this model is to determine what affect the shape of a basalt dome has on fluid flow patterns and the position of regions of dilation in relationship to the formation of gold deposits. The model is made up of rigid

doubly plunging basalt dome which is blanketed by a thin weak altered metasedimentary unit and surrounded by moderately stiff metamorphic rocks. Deformation is applied so as to simulate horizontal compression perpendicular to the long axis of the dome. The altered metasedimentary unit contains regions of negative volume strain (contraction) on the flanks of the basalt dome where the dip is steep and at a high angle to the compression direction. Towards the top of the dome (but not at the crest) the weak altered metasedimentary unit contains regions of high positive volume strain (dilation) above the areas of contraction. This causes fluid flow rates to be highest close to the top of the dome where areas of contraction and maximum dilation are in close proximity. Contraction occurs within the matrix above the highest point of the dome. Regions of high positive volume strain are also regions that have failed in tension. These areas are more likely to have formed quartz veins which commonly host gold. Multi-regional, geometrically complex meshes are constructed using GoCad. The CSIROdeveloped software 3DMACS is then called via a web-browser. It is primarily used to set model parameters and properties, and translate these, along with the GoCad model, into FLAC3D. It also provides results for visualisation in GoCad and the creation of VRML files. 3DMACS can couple FLAC3D to other software, to additionally simulate thermal driven fluid flow and chemical reactions. In this way we are able to create models which simulate deformation, fluid flow, thermal and chemical processes all of which may be important for mineralisation.

SGTSG Field Meeting 2003

IMPERMEABLE LAYERS, ROCK FRACTURING AND FLUID FLOW LOCALIZATION Roberto Weinberg School of Geosciences, Monash University, Clayton, Melbourne VIC 3800

Permeability variation controls the flow of mineralizing fluids across the crust, and significant metal enrichment develops along focused fluid pathways. This work explores crustal-scale permeability variation during fluid flow and deformation by means of twodimensional numerical models using FLAC, a well-established commercial package. A body of low permeability rock (a seal, such as a granitic body, or a layer of politic rocks) lying across upward-directed fluid flow paths leads to increased pore pressure below it and along its lateral contacts. This high pore pressure causes increased fluid velocity along the lateral contacts, and a doubling or trebling of the velocity and volume of fluids travelling along these contacts compared to elsewhere. Another effect of the increased pore pressure along the seal's contacts is that it causes preferential fracturing of the contact region during deformation. Modelled permeability increase related to fracturing, resulted in a twenty-fold increase in velocity and fluid volumes through the contact area. The increased fluid velocity is not, however, linearly dependent on increased permeability. This is because the high permeability path provides an

escape valve for the pressurized fluids under the seal, leading to a rapid decrease in the pressure gradient and consequently a decrease in fluid velocity. The total volume of fluid travelling through the high-permeability, fractured contact area, depends not only on the value of permeability but also on the availability of fluid feeding into that area through low permeability rocks, which in turn is a function of the geometry of the permeable pathways and seal, and fluid volumes in the crust. In detail, the fracture pattern, fluid flow velocities and pathways through contact zones are extremely dependent on the elasto-plastic parameters of the rocks on either side of the contact, as well as their permeability contrast. The main conclusion of this research is that the presence of a lowpermeability body will focus fluids towards its lateral contacts making this area the most important fluid pathways through the crust. The results could help explain deposits formed at granite contacts (eg, deposits in the Leonora district, Yilgarn), as well as the distribution of gold along the Boulder-Lefroy Shear Zone in the Yilgarn Craton.

SGTSG Field Meeting 2003

FAULT (AND SHEAR ZONE) VALVE PUMPING REVISITED. A. Ord, N. Walters and B E. Hobbs. Predictive Mineral Discovery Cooperative Research Centre, CSIRO Exploration & Mining, PO Box 1130, Bentley, WA 6102, Australia

Following the classical work by Sibson, pumping of fluids up faults has been widely explored as a process for transporting hydrothermal fluids from deep (over-pressured) parts of the crust to shallower parts. However, Matthai and Roberts (1997) have drawn attention to constraints on the supply of fluid, to the fault from the host rock In the vicinity of the fault, in order to maintain significantly high fluid fluxes through the fault. The discussion rests, in part, on the time scale, x, that it takes for a pressure disturbance to diffuse over a distance, d, in a rock mass with hydraulic diffusivity, K. This time scale is given by: t = 5'/2K ."See Phillips, (1991)" As Matthai and Roberts point out, for highly impermeable rocks the time scale for a pressure transient to propagate over a distance of 100m is of the order of tens of thousands of years. Another important constraint is the storativity of both the fault and the country rock This parameter measures the volume of fluid that can enter a given rock mass for a unit change in pore pressure and is also involved in the definition of the hydraulic diffusivity. The main conclusion from the work of Matthai and Roberts (1997) is that large volumes of fluid are likely to be transported along faults only if the faults are highly permeable zones for large periods of time and/or the storativity of the rock mass adjacent to the fault is very large. In this paper we explore these issues in greater detail and extend the discussion to examine fluid transport in dilatant (aseismic) shear zones as well as in discrete, seismic faults. Using numerical models, we explore the flow patterns that develop and the volumes of fluid that can be transported through dilatant faults and shear zones that link overpressured and normally pressured regions in the crust.

Complete coupling between deformation induced dilatancy, pore pressure change and permeability change is incorporated in these models. We map out, in parameter space, the conditions where fault valve pumping is likely to be important for the transport of significant volumes of hydrothermal fluid. In particular we explore the development of damage zones at the ends of faults and the development of dilatant supply regions near faults, together with the influence of gas mixing, on the storativity and time scale for diffusion of fluid. We also couple fluid transport to the advection of heat and examine the thermal gradients that develop because of the transfer of fluid from lithostatically to hydrostatically pressured systems. These thermal gradients are responsible for mineralisation and alteration patterns in such systems and we briefly outline the systematics behind the development of these patterns. References. Matthai, S.K. and Roberts, S. G. (1997). Transient versus continuous fluid flow in seismically active faults: an investigation by electric analogue and numerical modelling. Chapter 16 in Fluid Flow and Transport in Rocks. Eds Jamtveit, B. and Yardley, B. W D. Chapman and Hall, 319pp. Phillips, O. M. (1991). Flow and Reactions in Permeable Rocks. Cambridge University Press, U.K. Sibson, R. H., Robert, F. and Poulsen, K. H. (1975). Seismic Pumping - a hydrothermal fluid transport mechanism. Jour. Geol Soc London. 131,653-9.

SGTSG Field Meeting 2003

SPHERICAL, DEFORMABLE, SHELL TECTONICS. Bruce Hobbs Office of Science and Innovation, Department of Premier and Cabinet, Governor Stirling Tower, 197 St Georges Terrace, Perth WA 6000, Australia.

There are still a number of quite enigmatic questions concerning Plate Tectonics. To date, for instance, there is no adequate explanation of why island arcs, and many mountain chains, have arcuate structures. There is no adequate explanation of why transform faults exist and why they have the spacing that is observed. There is still much discussion concerning the mechanism of slab roll-back, of the origin of the dip of subduction slabs (fluid dynamics predicts the dip should be vertical) and why some are very shallow, of the origin of back-arc basins and even the gross mechanisms involved in generating island arc and back arc igneous activity. Until quite recently (see Regenauer-Lieb et al, 2001; Regenauer-Lieb and Yuen, 2003), there has been no, ab initio, mechanism for initiating subduction; all fluid dynamic modelling, no matter how sophisticated the rheology, leads to a "stagnant lid" which consists of an outer thermal boundary layer (the lithosphere) which remains Intact with no subduction or rifting whilst the convective motions proceed in the mantle below. These issues are reviewed in Schubert et al (2001). The purpose of this paper is to examine many of these issues by approaching the problem from the outside inwards, like a geologist would, rather than from the inside outwards, like most geophysicists have to date. In doing this it is important that we operate on a sphere rather than in "flat Earth" mode, because, as will become apparent, many geometrical aspects of Plate Tectonics only become apparent on a sphere and are completely absent in the "flat Earth" environment. This was recognised by Yamaoka et al (1986) who first coined the term Spherical Shell Tectonics. The plates are also treated as deformable rather than as rigid materials as in classical Plate Tectonics. To this end, numerical calculations have been carried out on a sphere with the outer skin of the Earth represented as an elastic-plastic material, capable of failing in a localised shear mode, underlain by elastic-viscous (power-law) material. Elastic-viscous material can only localise in a shear mode if strain weakening occurs arising from processes such as chemical-, microstructural- or thermal-mechanical coupling.

Internal (radioactive) heat sources are included together with nominated thermal fluxes from the mantle. Coupling with fluid flow is also included so that, locally, where the problem is of interest, thermal advection due to fluid flow can be included together with coupling to mechanical behaviour. Convection within the Earth is not modelled specifically and velocity fields are imposed on various parts of the Earth in order to simulate observed Plate Tectonic motions or to replicate mantle velocity distributions observed in convection simulations. Even though mantle convection is not modelled specifically, the sublithospheric mantle is represented as an elasticpower-law viscous material with representative mantle density distribution; gravity is included so that isostatic response is realistically modelled. We examine the following issues: 1. Since plates are not rigid in these models and can deform in elastic, plastic or viscous modes, we can map out the distribution of both stress and strain within a plate as well as at plate boundaries. This gives some insight into the origins of back arc basins and the regional extent of localisation associated with rifting and subduction. 2. In keeping with other recent work (see Regenauer-Lieb and Yuen, 2003 and Regenauer-Lieb et al, 2001), through-going failure of the entire lithosphere to initiate a subduction zone or a zone of lithospheric extension only occurs if a strain weakening deformation process is included in the constitutive behaviour. We discuss the implications of this for the geometry (on a sphere) of plate boundaries and for internal "degassing" processes within the Earth. 3. In regions such as mid ocean ridges where advection of heat due to hydrothermal systems is well documented, the temperature distributions that result from hydrothermal circulation, coupled to constitutive behaviour, define the geometry of fracture systems that develop at the ridge. Overall, the detailed geometry of failure and deformation of non-rigid plate systems when viewed on a sphere enables us to place constraints on the rheology of the lithosphere and to understand the answers to many of the issues raised in the first paragraph. These constraints

SGTSG Field Meeting 2003

are absent if one operates in "flat Earth" mode because the geometrical constraints of incorporating the processes and geometry into a spherical geometry are quite restrictive and absent in Euclidian flat Earth geometries.

coupling of equation of state, rheology and thermal-mechanics, (in press), Schubert,G., Turcotte, D.L., and Olson, P. (2001). Mantle Convection in the Earth and Planets. Cambridge University Press, 940 pp.

References Regenauer-Lieb, K., Yuen D., and Brunland, J. (2001). The initiation of subduction: Criticality by addition of water? Science, 294, 578-580.

Yamaoka, K., Fukao, Y., and Kumazawa, M. (1986). Spherical Shell Tectonics: Effects of sphericity and inextensibility on the geometry of the descending lithosphere. Rev. Geophys. 24, 27-53.

Regenauer-Lieb, K. and Yuen, D. (2003). Positive feedback of interacting ductile faults from

SGTSG Field Meeting 2003

As an example, we present the following simulation: The four figures opposite are all threedimensional images of the shear strain rate plotted on the outside of the Earth in (a), on the inside of the lithosphere in (b) and (c), and as a composite image on both the inside and outside of the lithosphere in (d). The image in (c) is a zoom looking approximately along the diameter of the sphere. The model consists of two spherical plates representing the lithosphere; one plate has been held fixed relative to the centre of the Earth whilst the other rotates with a velocity field imposed on the base of the lithosphere about a Euler pole that is on the boundary between the two plates. The collision of these two plates here has produced a mountain range, best seen in (c), approximately the same size as the Himalayas. The images show that the locus of highest shear strain rate is behind the mountain range on the base of the lithosphere but well in front of the range on the top of the lithosphere. Figure (d) makes it clear that this distribution of shear strain rate comprises a lithospheric scale shear zone that, overall, dips at a quite shallow angle below the mountain range. Two or three dimensional "flat Earth" simulations lead to a shear zone dipping at about 45° under the mountain range. We compare this distribution of strain rate with that observed in the present day Himalayas.

100

SGTSG Field Meeting 2003

In this paper we are concerned with presenting a number of examples relevant to Plate Tectonics that address the issues presented in the first paragraph. In particular it is important to emphasise that the geometry of deformation of a spherical shell (representing the lithosphere) is quite different to what one would expect from a "flat Earth" situation and that much can be learnt about the rheological properties of the lithosphere and about the "outgassing" of the Earth by considering the deformation of the lithosphere from a spherical rather than flat Euclidian perspective.

(d)

101

SGTSG Field Meeting 2003

MODELING SHEAR ZONES IN GEOLOGY Klaus Reaenauer-Lieb^ and Dave Yuen^ ^Institute of Geophysics, ETH Zurich, Honggerberg HPP015, 8093 Zurich, CH; Fax: ++41-1-6332058, E-mail: klaus@tomo.ig.erdw.ethz.ch (now at CSIRO Exploration and Mining, Perth) ^Supercomputer Institute and Dept. of Geology and Geophysics, University of Minnesota, Minneapolis, USA, Fax: ++ 1-6126253819, E-mail: davey@krissy.msi.umn.edu

Shear zones are the most ubiquitous features observed in planetary surfaces. They appear as a jagged network of faults at the observable brittle surface of planets, and in geological exposures of deeper rocks they turn into smoothly braided networks of localized shear displacement leaving centimeter wide bands of "mylonitized", reduced grain-sizes behind. The overall size of the entire shear network rarely exceeds kilometer scale at depth. Although mylonitic shear zones are only visible to the observer, when uplifted and exposed at the surface, they govern the mechanical behavior of the strongest part of the lithosphere below 10-15 km depth. Mylonitic shear zones dissect plates, thus allowing plate tectonics to develop on the Earth.

crystallographic- /shape-preferred orientation and dynamic recrystallization is only a maximum of 10% of energy dissipated in the shear zones but it creates structural anisotropy. Shear zones become long-living features with a long-term memory. Water

MODELLING APPROACHES

A special thermo-mechanical role is attributed to the presence of water in nominally anhydrous minerals. We show that water directly affects the mechanical equation of state and has the potential to synchronize viscous and plastic flow processes at geological time scale. Using mechanical data from the laboratory, we can reproduce the basic mode of deformation of an entire mylonitic shear zone in a generic model of subduction initiation.

Physics

APPLICATIONS

The physics of ductile shear zones relies on feedback processes that turn a macroscopically homogenously deforming body into a heterogeneously slipping solid medium. Positive feedback can amplify strength heterogeneities by cascading through different scales. We identify basic, intrinsic length scales of strength heterogeneity such as those associated with plasticity, grain-size-, fluid-inclusion-, and thermal diffusion length scale. Energetics While brittle shear zones can be modeled without specific calculation of the energetics of deformation (Rudnicki and Rice Theory), the consideration of local changes in entropy is a key to understand ductile faulting. Shear heating introduces a jerky flow phenomenon potentially accompanied by ductile earthquakes. For the long time scale, deformational energy stored inside the shear zone through plastic dilation I

102

The approach has so far been applied to a wide range of subduction scenarios using forward 2 and 3-D modeling techniques. These shed light on the dynamic formation of oceanic arcs (feedback of mantle flow and lithosphere vs. lateral density contrasts), the effect of water breakdown reactions in subduction zone (its role for earthquake generation), and interaction of the overriding plate (flat Andean style tectonics). Future models will consider continent-continent collision. These models provide large-scale geodynamic constraints (material and volatile flux rates, stress and strain rates) for more detailed geological field studies. The next step of modeling lies in benchmarking basic feedback mechanism in applied field studies and zooming into the braided network of shear zone structure, without losing the large-scale constraint.

SGTSG Field Meeting 2003

A FULLY DYNAMIC MODEL OF EVOLVING EXTENSIONAL SHEAR ZONE SYSTEMS AND METAMORPHIC CORE COMPLEXES K l a u s G e s s n e r \ C h r i s Wijns^'^, L o u i s M o r e s i ^ ^ Computational Geoscience, CSIRO Exploration and Mining, PO Box 1130, Bentley WA 6102, Australia ^ Centre for Global Metallogeny, University of Western Australia, Crawley WA 6009 ^Computational Mathematics & Geophysics, Monash University, Clayton VIC 3800

Extension of a model of three rheological layers simulates the interaction of lower crustal flow and detachment faulting when metamorphic core complexes form in continental crust. Our numerical models demonstrate that stretching of a laterally homogeneous model lithosphere including a weak lower crustal layer leads to pronounced strain partitioning, where distributed flow of the lower crust drives brittle failure of the upper crust. THE PROBLEM The deformation behaviour of continental lithosphere in extension depends on a number of compositional and structural variables, which determine its rheological variation vertically and laterally. In the case of metamorphic core complexes, it has been suggested that the lower crust is sufficiently hot, enabling rapid lateral flow to accommodate horizontal pressure gradients across large normal-faults. It has been argued that in the case of metamorphic core complexes deformation is 'top driven', i.e. dominated by lateral movements of the strong upper crust due to gravitational instability of thickened crust. According to this concept, the displacement of strong upper crust exerts a shear stress on the lower crust, allowing plastic weakening to dominate deformation. This mechanism has been used to explain how strain localises to form initially shallow dipping shear zones on a lithospheric scale. Previous numerical models of continental extension have not been able to resolve the interaction between detachment faulting and lower crustal flow.

METHOD

allows the use of fast solution methods, and is not restricted by strain limits due to a deforming mesh. RESULTS Initial strain perturbations in the upper crust localise narrow, brittle-plastic detachment faults, which dissect the entire upper crust and exhume ductile lower crustal material in their footwall, while being rotated from a ca. 60° dip to subhorizontal. Once the upper crust is dissected, strain is largely accommodated by current-like lateral flow of lower crustal material relative to both upper crustal segments. While exhumation along the initial detachment continues, a second shear zone forms below the footwall of the initial detachment. For this region, our model predicts a high flow velocity gradient in the uppermost part of the lower crust, defining a ductile shear zone which operates continuously from the earth's surface to mid-crustal levels while the upper crustal segments remain in a kinematically stable position. DISCUSSION AND CONCLUSIONS Our model provides a dynamic explanation for the structural and metamorphic patterns of metamorphic terrains bounded by extensional shear zones. A weak lower crust underlying a rigid upper crust appears to be important, whereas lateral variations of crustal thickness may not be a requirement for the formation of metamorphic core complexes. We argue that distinct crustal architectures such as near-surface metamorphic core complexes and metamorphic domes can form at different depths of evolving shear zone systems during the same geodynamic process.

To simulate continental extension in rheologically stratified lithosphere we have used the numerical modelling software ELLIPSIS, which is capable of modelling high strain in geological materials with history dependent properties. This technique

103

SGTSG Field Meeting 2003

GEODYNAMICAL MODELING OF TECTONIC AND THERMAL THINNING IN THE NORTH CHINA BLOCK G e Lin^' ^ Y u e j u n W a n g \ G u a n g h a o C h e n \ Y a n h u a Z h a n g ^ ' \ Y a n g h u a W a n g ^ ' \ F e n g G u o ^ a n d W e i m i n g Fan^ ^ Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou, 510640, P.R. China ^ CSIRO Exploration & mining, PO Box 1130, Bentley WA 6102, Australia ^ Robertson Research International, Horizon House, Azalea Drive, Swanley, Kent BR8 8JR, UK.

The North China Block (NCB) is one of the areas, which experienced significant lithosphere thickness changes. Evidences from the studies on the petrology, mineralogy and geochemistry of deep upper-mantle derived xenoliths suggest that the thickness of the lithosphere of NCB was greater than 200 km in the Precambrian, about 150-200 km in the Palaeozoic and less than 100 km since the Mesozoic. While this long period of lithosphere thinning has been broadly attributed to thermotectonic changes in the NCB since the Palaeozoic, questions still remain as to the quantitative relationship of such huge thinning with thermal evolution and/or tectonic extension in the lithosphere. We have constructed a series of 2-D numerical models to explore the geodynamics of the lithospheric thinning in the NCB. Firstly, we employed a mountain-basin evolution system to model tectonic thinning processes. The NCB has probably experienced at least two stages of crustal deformation throughout the Mesozoic, as revealed by compressional structural styles before the late Jurassic and extensional tectonics in the late Mesozoic. The EW structural trending of the late Triassic QinlingDabie and Yanshanian orogenic belts, which represent the southern and northern boundaries of NCB respectively, indicates that N-S shortening occurred in NCB pre-Jurassic, while widely distributed NE-NNE-trending extensional basins suggest that crustal stretching during the late Mesozoic was in the E-W direction. The results of our model for the N-S shortening event show that at a dimensionless time of t = 0.9, the whole weak region plus part of the strong region of the model was thickened significantly. In particular, the eastern portion was thickened by about 30%. The compression model was then subjected to E-W crustal stretching, simulating the late Mesozoic extensional event. The results show that at a dimensionless time of t = 0.3, the whole area thickened in the previous compressive period became thinned. The strongest region in the model (the Ordos) and its north portion now are thicker

104

than the surround regions. These results suggest that the horizontal extension of a previously thickened, unstable lithosphere mainly contributed to the formation of extensional basins as a consequence of strike-slip collapse, which was possibly triggered by the descending subduction slab rolling-back easterly from the ancient Pacific Plate. We then used a finite difference code, FLAC, to simulate thermal thinning processes. Our preliminary results show that under given thermal conductivity and radioactive heat production in the lithosphere, increasing mantle thermal flux along the bottom of the lithosphere can lead to changes in the thermal state and lithospheric thickness of the NCB since the Palaeozoic. Lithospheric thickness can be thinned to less than 100 km when the mantle thermal flux increases from 20 to 35-40 mWm"^. The effects of tectonic extension on lithosphere thinning were explored by the models incorporating a constant extensional displacement rate at the lateral edges. The results show that the thinning of the lithosphere is still quite limited after a 22% extension (no increase in thermal flux). The inhomogeneously-thinned lithosphere in this case has a thickness ranging from about 175 to 150 km, much thicker than the final lithospheric thickness for the thermal thinning situation with a small increase in thermal flux. However, the effect of tectonic extension on lithosphere thinning is much greater if assuming extension starts after lithosphere is already inhomogeneously thinned by the increase of mantle thermal flux. In this case, bulk extension is dominantly localized at the segment of the lithosphere with the maximum thermal thinning, and a few percent bulk extension can generate much greater thinning at such locations. In summary, thermal thinning seems to be the dominant mechanism responsible for the huge lithosphere thinning in NCB. The change of the overall thermal state of the region from a stable low mantle thermal flux regime to an unstable high

SGTSG Field Meeting 2003

thermal flux one probably began from the Palaeozoic. This marked the start of a long-term thermal activation stage accompanied by active intrusive activities and volcanisms. Tectonic extension probably had more profound effects on lithosphere thinning at a later stage of this period when the lithosphere had already been inhomogeneously thinned. Such tectonic extension has probably played a dominant role in the formation of the sedimentary basins in the region because of its capability to strongly localize extension and thinning. Acknowledgements This work is support by the CSIRO/CAS exchange program, the Natural Science Foundation of China ( No: 49972045) and the CAS program ( No: KZCX2-113).

105

SGTSG Field Meeting 2003

MODELLING OF ROCK DEFORMATION BEHAVIOURS IN CRUST: BRITTLE FAILURE VERSUS PLASTIC FLOW Y. Zhang\ A. Ord\ B.E. Hobbs\ P.A. Roberts\ Ge Lin^ Yuejun Wang^and J.G. McLellan^ ^ Predictive Mineral Discovery Cooperative Research Centre, CSIRO Exploration & Mining, PO Box 1130, Bentley, WA 6102, Australia ^ Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou, China ^ Economic Geology Research Unit, James Cook University, Queensland, Australia Simulation of the initiation and propagation of discrete fractures and faults has been a challenging task in the field of numerical modelling. In this study, we aim to simulate various deformation scenarios, particularly folding and faulting, and show here the resultant patterns of fracture development. Our models explore several structural kinematic scenarios including folding, deposit-scale fault movement and crustal scale thrusting. We have also constructed an application model for an exploration cross section in the Kang-Jia-Wan gold deposit in the ShuiKou-Shan mineralisation district, China, where mineralisation is closely associated with severe rock brecciation. FLAC (a finite difference code) and Elfen (a finite element code) have been employed to simulate plastic or irreversible deformation and fracture development. Plastic deformation models assume a Mohr-Coulomb elastic-plastic rheology where the material deforms initially elastically but continues plastically to large strain once the maximum shear stress reaches a critical value (yield stress). Brittle models adopt the Rankine yield criterion, which states that the material will fail in tension and fracture if the maximum tensile stress reaches a critical value. The results show that the patterns of plastic strain localisation and

106

brittle fracture development consistently reflect the nature of the stress regime. For example, in the situation where far-field stress leads to shearing fault movement, a model assuming that plastic deformation dominates, predicts the development of strain localisation and dilation zones near fault tips, while the model which assumes that brittle failure dominates also demonstrates the initiation of fracture arrays at similar locations. Such fracture zones in the brittle model also represent zones of significant dilation. In the thrusting scenario, a plastic deformation model demonstrates that shear zones develop as thrusting proceeds, which corresponds to the formation of back-thrusts and new detachment thrusts. In contrast, the development of extensive discrete fracture zones dominates the model simulating brittle crustal slabs, but this still reflects the strongly compressive nature of the stress regime. Our test model based on the Kang-Jia-Wan gold deposit also predicted extensive fracture development at locations consistent with the observation of rock brecciation and mineralisation. The development of these fractures and brecciation is jointly controlled by structural and lithological features in the area, and this highlights the significance of structural and lithological criteria for mineral exploration in the region.

SGTSG Field Meeting 2003

TOPOGRAPHY OF THE MOHO UNDULATIONS IN FRANCE FROM GRAVITY DATA: THEIR AGE AND ORIGIN J.P, L e f o r t \ B.N.P. Agarwal^

^ University of Rennesi, Geosciences-Rennes, Institute of Geology, Tectonophysics Laboratory, Beaulieu Campus, 35042 RENNES Cedex, FRANCE ^ Department of Applied Geophysics, Indian School of Mines, DHANBAD 826004, INDIA

The complete gravity data set from France and its surrounding countries has been analyzed to compute the topography of the Moho undulations. This work is based on an improved filtering technique and an appropriate assumed density contrast between the crust and the upper mantle. Comparison with deep seismic refraction data reveals that this relief map expresses the continuity and geometry of the Moho undulations better than seismic refraction data. This gravity Moho map provides a far better correlation with surface geology than other geophysical techniques. Four domains have been recognized: (a) The Alpine domain where all the Moho undulations are concentric with the Alps; (b) The Armorican domain in which all the undulations are northwest-southeast oriented; (c) The Pyrenean domain, in which the undulations are parallel with the Mountain chain and (d) The Massif Central Domain which does not show clear structural orientation. Study of the topography and of the superficial structures associated with these undulations reveals that the undulations delineated in the Alpine Domain result from the Tertiary compression, which

shaped the Alps. The Armorican Domain was first created during the Lower to Middle Cretaceous opening of the Bay of Biscay. It is now slightly affected by the Tertiary to Quaternary closure of this Bay. The Pyrenean Domain was shaped by the Lower Cretaceous oblique opening of the Bay of Biscay. Comparison between the Moho undulations map and the stress map of France reveals that most of the undulations are perpendicular to the actual shortening directions. This observation suggests that the Mesozoic, Cenozoic and Quaternary stress directions were roughly the same. Massif Central is characterized by the convergence of these three sets of undulations. Its Post-01 igocene uplift was probably the result of the converging stresses recognized in the three surrounding domains. When the Moho undulations and the topography are compared, two types of periodic crustal instabilities can be recognized. One corresponds to the buckling of the crust developed under compression, the other to boudinage which was associated with extension'. Both phenomena show a typical wavelength of 200 to 250 kilometer.

107

SGTSG Field Meeting 2003

TWO MUCH RED WINE AT TOO MANY MINES John L, Baxter Principal Geologist, Continental Resource Management Pty Ltd., Belmont, Western Australia 6984 hermitage@iinet.net.au

Quality red wine is the result of complex interaction between a large numbers of well-defined factors such as: • Soil type • Climate and weather • Geomorphic aspect • Grape variety • Sugar content on picking • Processing • Maturing vats • Time of bottling. Low quality wine results from failing to create the conditions for the interaction of the component parts at appropriate times in the process. Mine geology is a complex interaction between a similarly large group of fairly well-defined factors such as: • Host rocks • Alteration and weathering • Ore controls o Physical o Chemical o Geometric • Exploration integrity • Resource & reserve estimations • Rock mechanics • Extraction methodology • Processing requirements • Mine management Profitable mines integrate all of these aspects in a complex adaptive system embracing all factors impacting on the extraction of the ore. The impact of structural geology on this system cannot be, but often is, underestimated in many mines. Understanding the 3-D geometry of the host rock and mineralization, the kinematics of the mineralizing event and the 3-D geometry of subsequent events can nearly always identify the geological components in these complex adaptive systems. Examples of deposits formed in a variety of tectonic settings will be presented from Western Australia, Queensland and Victoria. Simple techniques to resolve the structural geology and then the geometry of mineralization will be presented. Like seagulls, consultants arrive by air; squawk a lot occasionally shit on a few and depart by air. Red wine is often an accompaniment to an evenings discussions. The time frame can be anything from a

108

few days to a few months. Mine managers rightly are concerned with cost of having the seagull coming frequently. As a result the structural geology consultant in the mine environment is often a last resort when something fails. The most common failures are: • Stope grades do not reconcile with reserve estimations • Grade distribution is more variable than predicted • Lodes do not connect • Hanging walls fall in • Backs fall in on flatmakes Structural geology of an ore deposit has direct inputs into reserve calculations, mine design, extraction methods, ore reconciliation and mine stability. Like wines, poor quality mines are often the product of ignoring the impact of a single factor, structural geology, in the complex system. All sorts of excuses for omitting structural geology emerge retrospectively, but most can be related to time management. The demands on mine geologists time is high. Consultants can contribute to identification and focusing many of both geological and non-geological issues. The solutions are usually as simple as completing a 3-D geological map of the exposures and drill core, then creating a mine model using those reliable tools; controlled base map, hammer, hand lens, compass, stereographic net and computer modeling. Understanding the process of remediation has certainly led to copious quantities of red wine being consumed at mine sites all over Australia and elsewhere. Mining is an economic pastime; there is very little room for benevolence to research and creativity. Usually I have been introduced to these problems with a phase something like "We seem to be having difficulty with". Resolution is nearly always achieved by identifying the various contributors to the problem (grade control, face mapping, geostatistics, operational commitments, etc) and bringing them together to identify a course of action to overcome the difficulty. Revising the structural skills needed and identifying their importance to the problems with or without the bottle of red wine usually leads to a happy departure.

SGTSG Field Meeting 2003

FRIDAY 26 SEPTEMBER 2003

109

SGTSG Field Meeting 2003

110

SGTSG Field Meeting 2003

WATCHING MICROSTRUCTURAL CHANGES INSIDE AN SEM Dave Prior\ Gareth Seward\ Steven Celotto^ Michel Bestmann\ Sandra Piazolo\ Chris Spiers^ and John Wheeler^ ^Department of Earth Sciences, Liverpool University, L693GP, UK ^Department of Materials Science and Engineering, Liverpool University, L693GP, UK ^Department of Earth Sciences, Utrecht University, Budapestlaan 4, 3584 TA Utrecht, Netherlands

We wish to understand the processes by which microstructures, preferred orientations and resultant physical properties of rocks evolve and to use this information to help interpret, quantitatively, the history of a rock. However, we are limited to looking at the final frozen-in microstructures and and viable explanations of what has happened in the material are those that can explain the frozen-in microstructures and textures. Being able to watch microstructures and textures as they evolve provides a much better set of constraints upon the processes that operate in materials. The CamScan X500 crystal probe was built with the aim of conducting high temperature experiments in-situ in an SEM and quantifying microstructural and textural changes during those experiments using EBSD and imaging. In this talk we will show some of the results of experiments in metals and in rock forming minerals. We will focus on the results of phase transformation experiments in titanium and recrystallization experiments in rock salt. During heating, at '-882°C, the crystal structure of Ti transforms from hexagonal close packed (HCP - a) to body centred cubic (BCC - p), the opposite is observed when cooling from above 882°C. Historically, direct observation of the BCC phase has been limited due to the high transformation temperature, and the fact that the BCC phase cannot be 'frozen in' metastably by quenching to lower temperatures. During an in-situ experiment we are able to observe the 'transient' microstructure and crystallography. We can measure the relationship between a and p phases, and confirm that the first formed p grains have the Burgers orientation relationship (OR) {<0001>hcp//<110>bcc and <11-20>hcp//<111>bcc) during the up temperature transformation. The p grains are observed to have a lath shape

morphology within the parent a grains, this together with the specific OR relationship implies a shear dominated (martensitic) transformation mechanism. There is a change in morphology of p phase from lath shape to equiaxed grains and an accompanying order of magnitude increase in grainsize. Since the OR can only exist between a parent and daughter grain (i.e. the OR cannot be maintained with a randomly oriented neighbour) the observations suggest that grain growth processes may play a significant role in modifying the final p texture and that rapid grain growth is triggered by the phase transformation. Recrystallization is an important process in metals and minerals. Dry deformed rock salt samples were heated to temperatures between 350 and 410°C. Inter- and intracrystalline processes were observed using both secondary electron imaging and electron backscatter diffraction mapping techniques. Grain boundary migration (GBM) was observed between substructured grains. During GBM some unexpected features were observed. Often the growing grain exhibits new low angle boundaries in the swept area and in some cases the existing substructure of the growing grain is continued in the swept area. More rarely substructures within the swept area develop into a low angle boundary within the growing grain. These data are not explicable with existing grain boundary migration models. The observations imply that general grain boundaries have a structure that allows structural features of the 'dying' grain to be transferred to the growing grain. New experiments including high temperature deformation experiments will also be outlined.

Ill

SGTSG Field Meeting 2003

APPLICATION OF ELECTRON BACKSCATTER DIFFRACTION TO NATURAL HIGH-STRAIN ZONE DEFORMATION Steven Reddv & Craig Buchan Tectonics Special Research Centre, Department of Applied Geology, Curtin University of Technology, Perth, WA6845, Australia.

Zones of localized deformation (high strain zones) are commonly formed in the Earth's crust as a response to tectonic activity between adjacent lithospheric plates. Microstructural analysis provides important information regarding naturally occurring deformation processes within high-strain zones. However, high-strain zones commonly have complex deformation histories because of the spatial and temporal localisation of deformation during their development. Linking microstructural development to particular stages of this progressive deformation history may provide a significant advance in our understanding of how high-strain zones develop. However, such studies are difficult unless the temporal framework of deformation within the high-strain zone can be constrained. Here we present data from a high-strain zone in which a well-constrained temporal history of deformation has been established and we outline how Electron Backscatter Diffraction (EBSD) is being used to document the microstructural evolution of this shear zone over time. In the Western Alps, the Piemonte Ophiolite consists of rocks metamorphosed at eclogite facies conditions (r« 550-600 and P« 18-20 kbar) structurally beneath greenschist facies rocks metamorphosed at r« 400 X and P« 9 kbar. Mapping shows that the latter form a kilometre-wide calcite-dominated shear zone (the Gressoney Shear Zone - GSZ) dominated by topSE movement related to crustal extension. Radiometric dating (using the Rb-Sr decay

112

system) of micas within different GSZ fabrics, which dynamically recrystallized below their blocking temperature, indicate the time of deformation. Ages from different samples within the same fabric are reproducible and are consistent with the relative chronology derived from mapping. They show that the GSZ has a deformation history over a period of c. 9 million years (Ma) between c. 45 - 36 Ma ago. Four samples, used for Rb-Sr dating, have been studied using EBSD. These indicate a complex microstructural history that, in part, reflects the timing of deformation within different parts of the shear zone. Detailed analyses of orientation data also indicate a difference in crystallographic symmetry and preferred orientations for different grain size fractions. This reflects the operation of different deformation mechanisms under different conditions of pure and simple shear. Misorientation analysis and the dispersion of crystallographic axes around intracrystalline rotation axes indicate a link between intra sample slip systems and macroscopic vorticity axes inferred from the orientations of macroscopic structural elements with monoclinic symmetry. The apparent kinematic link between intragrain slip systems and external vorticity may provide an independent means of constraining slip directions in transpressional high strain zones. This may be particularly useful where mineral elongation lineations, recording the stretching direction, lie oblique to the kinematic shear direction.

SGTSG Field Meeting 2003

CHARACTERISING COMPLEX DEFORMATION PATHS DURING TRANSPRESSION Craig Buchan\ Steven M, Reddy\ Alan S. Collins^ ^ Tectonics Special Research Centre, Department of Applied Geology, Curtin University, GPO Box U1987, Perth, WA 6845 tel: +61 (0)8 9266 2446 ennail: c.buchan@curtin.edu.au

The Palaeoproterozoic Usagaran orogenic belt of Tanzania contains the Earth's oldest reported examples of subduction-related eclogite facies rocks. Detailed field mapping of gneisses exposed in the high-grade, eclogite-bearing part of the orogen (Isimani Suite) indicates a complex deformation and thermal history. Deformation in the Isimani Suite can be broadly subdivided into five events. The first of these (Di), associated with formation of eclogite facies metamorphism, is strongly overprinted by a pervasive high strain deformation (D2) and localised D3 folding at amphibolite facies conditions. The geometry of D2 foliations, lineations and shear sense indicators suggest variable deformation paths across the Isimani Suite that enable five D2 domains to be identified. Structural analysis of these domains indicates formation during strain and kinematically partitioned transpression. U-Pb SHRIMP zircon ages from the Usagaran eclogites and a post-Da pegmatite indicate that D2 deformation took place between 2001-1990 Ma (at maximum error). Subsequent greenschist facies deformation (D4 & D5) localised as thrust and extensional shear zones. These separate D2 domains and indicate post-D2 reactivation of the Isimani during the Neoproterozoic East African Orogen.

Electron Backscatter Diffraction (EBSD) studies of quartz domains within polymineralic samples from the five D2 structural domains has been undertaken to assess the relationship between macro-, meso- and microscopic structural features and better understand the structural evolution of transpressional high strain zones. By combining field geometry and kinematics with intra- and inter-grain orientation mapping, crystal orientation data and misorientation axes and angle distributions we: 1) characterise the deformation mechanisms operating within the different domains and hence assess variability of deformation conditions; 2) assess the effects of strain partitioning on an intra- and inter-domain scale; 3) quantify the effect of modal and textural distribution of quartz within polymineralic samples on the formation and preservation of crystallographic preferred orientation (CPO); and 4) assess whether preserved microscale kinematics reflect regional scale D2 kinematics or those of subsequent low strain deformation. By combining these observations we examine the relationship of finite strain to microstructural preservation, and address the problem of preservation of high strain textures during later low strain deformation.

113

SGTSG Field Meeting 2003

METAMORPHIC FOLIATION REVEALED: GRAIN-SHAPE DATA FROM THE OTAGO SCHIST, NEW ZEALAND Aaron Stallard, David Shelley Department of Geological Sciences, University of Canterbury, Private bag 4800, Christchurch, New Zealand aaron.stallard@canterbury.ac.nz david.shelley@canterbury.ac.nz

Introduction

The Otago Schist is a 150 km wide NW-SE trending belt of prehnite-pumpellyite to greenschist fades schist comprised of metasedimentary rocks that can be traced laterally into non-metamorphosed equivalents. Deformation within the schist was progressive, resulting in complex patterns of destruction and preservation of early-formed structures and textures.

sections normal to lineation. White mica commonly displays a bimodal distribution in sections perpendicular to lineation. An initial foliation developed subparallel to the limbs of isoclinal folds of bedding, and is composed of modified detrital quartz grains. Detrital and recrystallised white mica, quartz veins, strain shadow minerals and mica beards are structures that developed parallel to the grain-shape foliation. Progressive pressure solution resulted in increased aspect ratios and grain lengths, and enhanced alignment of grains parallel to the foliation plane. The development of a crenulation cleavage destroyed bedding, enhanced segregation layering, and provided a catalyst for change in dominant deformation mechanism from pressure solution to recrystallisation and minor grain boundary migration at higher metamorphic grade. Textural development of the foliation is linked to the dominant deformation mechanism, with the aspect ratio of quartz reduced once recrystallization became the dominant quartz deformation mechanism, while white mica continued to increase in size.

Results

Conclusion

A mineral elongation lineation is parallel to fold axes and defined largely by elongate quartz and white mica. Mean aspect ratios parallel to lineation range from 2/1 to 4/1 for quartz and 8/1 to 12/1 for white mica, while aspect ratios perpendicular to lineation are and 4/1 to 8/1 respectively. The 3D shapes of grains are that of stretched discs rather than pencil-shaped. Both quartz and white mica are aligned within the plane of foliation in sections cut parallel to lineation, but have varied orientations within

Foliation development in the Otago Schist is initiated by pressure solution of quartz and growth of white mica parallel to the mineral elongation direction. At higher metamorphic grade, and following increased segregation resulting form crenulation, quartz deforms by recrystallisation and grain boundary migration. This results in a decrease of quartz size and aspect ratio, while white mica continues to increase in size and becomes the dominant component of the grain-shape foliation.

The shape and orientation of quartz and white mica grains are described from a transect across metamorphosed sedimentary rocks of increasing textural development and metamorphic grade. The aim is, for the first time, to quantify the essential nature of the foliation (which is the shape-preferred orientation), and to document textural changes with increasing structural development. From this, we identify processes and deformation mechanisms important in the development of metamorphic foliation. The Otago Schist

114

SGTSG Field Meeting 2003

FRACTAL CHARACTERIZATION OF GRAIN BOUNDARIES: A NEW MICROSTRUCTURAL TECHNIQUE Tom Blenkinsop and Travis Murphy School of Earth Sciences, James Cook University, Townsville, QLD4811

Grain boundaries have scale invariant and therefore fractal geometries. One of the most widely used techniques to characterize their fractal geometries is the structured walk or coastline method, in which the length of the perimeter (P) of a boundary is measured by counting the number of steps of size L around the boundary. For a fractal shape, the relation between P and L gives the Coastline fractal dimension Dc from: P Dc characterizes the complexity of the grain boundary with great sensitivity. The more complex the grain boundary, the larger the value of Dc within the range 1 < Dc < 2. Manual determination of Dc is simple but very time consuming, which severely limits the number of increments of L that can be used to determine Dc. An automated method has been developed which allows several inherent problems to be solved in alternative ways. The best alternative has been evaluated by comparing results with other methods, and with fractal shapes that have known Dc values. GEOTHERMOMETRY An inverse relationship between Dc and temperature of deformation has been calibrated for quartz grain boundaries by Kruhl. This relationship is particularly useful in identifying intrusion-related deformation, as shown by studies on the Archaean Murehwa batholith in Zimbabwe. Quartz grain boundaries from granitic rocks in the batholith have relatively low fractal dimensions, suggesting that they were deformed at high temperatures that can be associated with intrusion of the batholith since no other high temperature events can be recognized. This result is confirmed by the presence of chessboard subgrains in the quartz, implying temperatures above the a-p quartz transition. By comparison, quartz grain boundaries from a greenschist facies shear zone have much higher fractal dimensions. The technique could potentially be applied to distinguish the

morphology of twin boundaries in calcite, whicfi has been proposed as a sensitive geothermometer. DEFORMATION MECHANISMS The high fractal dimensions of the greenschist facies shear zone quartz reflect the operation of recrystallization by grain boundary migration and bulging, in regime 1 of the Hirth and Tullis classification. This mechanism increases the complexity of the grain boundary and therefore Dc. Dc is therefore potentially a sensitive indicator of deformation mechanisms, although a detailed calibration has not been produced for any mineral. This approach has been used in the metallurgical literature to investigate the increase in complexity of grain boundaries produced by dislocation glide. DEFORMATION HISTORY A new development of the technique is the investigation of deformation history. A data set of experimentally deformed or annealed sphalerite grains as well as sphalerites from ore deposits has been compiled. Dc values from undeformed sphalerites have values very close to 1 (1 < Dc < 1.02), while Dc values in grains exhibiting obvious grain boundary migration recrystallisation have a range of 1.07 < Dc < 1.17. The recrystallised grains themselves have lower Dc values from 1.02 to 1.07 and a smaller grain size. A plot of fractal dimension against grain size clearly discriminates between undeformed or annealed grains, recrystallising grains and recrystallised grains. Transitional textures plot in fields between the end members. A major application of this result is the determination of deformation histories: undeformed, post-tectonic ore can be sensitively distinguished from pre to syn-tectonic ore. CONCLUSION Fractal analysis of grain boundaries is a new technique with several potential applications in microstructural studies, some of which also have economic implications.

115

SGTSG Field Meeting 2003

THE PRESERVATION OF EXTENSIVE DATABLE DEFORMATION HISTORIES IN ROCK AND THEIR RADICAL IMPLICATIONS FOR S TRUCTURAL AND METAMORPHIC PROCESSES T. H, Bell, A.P. Ham and P.W. Welch School of Earth Sciences, James Cook University, Townsville, Qld 4811, Australia

Bedding is reactivated on one or both limbs of pre-existing folds during younger deformations. This destroys (via decrenulation) newly developing foliations and tends to rotate relics of earlier oblique foliations into parallelism with SQ. Consequently, the 2 or 3 foliations (apart from bedding) that are generally preserved in the matrix of multiply deformed rocks preserve only a very small and recent portion of the extended deformation and metamorphic history that most of the rocks within an orogen have undergone. Porphyroblasts preserve evidence of this earlier history, but have been regarded as showing the remains of a foliation that was formed only 1 or 2 deformations prior to that seen in the matrix. We have found that the memory of successive deformations and metamorphic reactions recorded by porphyroblasts is very, very much more extensive than this. Foliation intersection axes preserved in porphyroblasts (FIAs) reveal a consistent succession of 5 FIAs with constant trends across a large tract of Acadian Vermont. FIA sets represent periods of time over which the direction of bulk shortening remained constant during orogenesis. This extensive history has been confirmed by absolute dating of monazite grains within the foliations defining the FIAs. Dating the youngest three FIA sets using monazite grains within successive foliations has revealed a progression in foliation ages from 431 ±2 to 349±3 Ma within porphyroblasts and from 366±3 to

116

327±5 Ma in pervasive matrix foliations. This history will probably extend by another 30 Ma when we find monazite grains defining the first two FIA sets. Significantly, these ages reveal over 100 million years of regionally continuous deformation, which gels well with the relentless continuity of relative plate motion. However, on the scale of very large outcrops the deformation is very discontinuous. What controls whether syntectonic porphyroblasts grow at a particular location, once the T and P and bulk composition are appropriate, is whether the deformation partitions through that location at the scale of a porphyroblast. This control is so intimate that porphyroblasts stop and start growing episodically at different times in the same compositional horizon on opposite limbs of regional folds, and even locally on mesoscopic folds. Furthermore, large-scale competency contrasts can affect regional partitioning of deformation and thus the localization of porphyroblast growth. Regional folds and gneiss domes, which commonly appear to be quite young structures, in all cases that we have studied except one, formed during the earliest deformation and were modified, but the only structures preserved throughout the history of orogenesis.

SGTSG Field Meeting 2003

SHRIMP 8

CRITICAL ANALYTICAL PARAMETERS

Allen Kennedy^ and Richard Stern^ 1. Dept. of Applied Physics, Curtin University of Technology, GPO Box U1987 Perth WA 6000 3. J.C. Roddick Lab., Geological Survey of Canada, 615 Booth St, Ottawa, ONT, Canada K1A 0E8

SHRIMP measurement of 8 ^Ô within individual zones of highly complex minerals will allow us to track the 6 of fluids and minerals associated with metamorphism, plutonism, diagenesis, ore deposition, hydrothermal alteration and supergene weathering. Obtaining U-Pb ages for the same zones will constrain the rates of geological processes, and this will be a major advance in SHRIMP applications. In addition, the role of low temperature fluids in the growth and/or resetting of U-Pb geochronometer minerals can be studied with in situ 6 measurements. Conventional bulk techniques achieve 0.1 per mil precision for 6 The SHRIMP cannot achieve the same precision, but a 6 ^Ô precision and reproducibility of 1.0 per mil is possible. Typical uncertainties for U/Pb ages are 1.0-2%, which is an order of magnitude larger uncertainty than that needed for O isotopes. This abstract documents our progress towards 5 ^^0 analysis, by examining our measurements of 8 in magnetite and zircon using both a Faraday cup and electron multiplier, but without charge neutralisation. As with any analytical technique it is essential that the analyst adopts a procedure that optimises all critical parameters. Count rates, count times, delay intervals, and data reduction methods must be chosen by the analyst. Additional complexity is added by critical instrumental and sample parameters that are partially under the control of the analyst or intrinsic to the instrument, oxygen isotope systematics, and the sample. The abundance of ^Ô, and respectively 0.98, 0.037, and 0.204, places major constraints on the count intervals, count rates, delay intervals and total count interval for data collected on an electron multiplier. Since Poisson statistics apply to counting processes it is possible to calculate minimum times to achieve 2 sigma precision and reproducibility of 1 per mil. One million counts per second (cps) on equates to approx. 370 cps on ^Ô and 2040 cps on and the count time to achieve 1 per mil data on 8 is 33 minutes. For 8 ^Ô this time would be 3 hrs. We used short data collection intervals, the "snapshot" technique, with 1 s. on 10 s. on background, and 15 s. on with the electron multiplier, and respective count intervals of 0.1 s.,10 s. and 10-15 s. for the Faraday cup. Delay times were matched to the settling time of the electrometer or magnet. The long total data collection interval necessitated by oxygen isotope abundances means

it is essential that secondary ion emission is stable for long periods. This depends upon the Cs ion source stability, sample charging/neutralisation, changing pit geometry. Cs implantation effects, and secondary ion emission characteristics. Our measurements show that secondary ion emission can change dramatically depending on the primary ion beam intensity used. In some instances rapid growth of the secondary signal during the first hour prevented reproducible measurement, while under other conditions long periods of stability occurred. Mass resolution, abundance sensitivity, magnet stability, collector discrimination, and, standard homogeneity are not critical issues for SHRIMP analysis of magnetite and zircon. With such high count rates the effects of electron multiplier dead-time and aging on measured isotope ratios must be understood. A series of measurements made with different assumed dead-time values, and electron multiplier voltages (to simulate aging of the multiplier), show that these parameters must be carefully assessed and controlled if reliable data is to be obtained. In addition, any non-linearity of the electron multiplier must also be accounted for. A 10 volt change in the inflection point on the electron multiplier operating plateau or a 5% error in the dead-time correction equates to a 1.5 per mil change in 8 SHRIMP 8 analysis of magnetite, an electrical conductor, is straightforward and 10 sets of 6 cycles produced 0.5 to 1.2 per mil precision and reproducibility with both the Faraday cup and electron multiplier. Instrumental Mass Fractionation (IMF) was approx. - 2 per mil for the Faraday cup and +4 per mil for the electron multiplier. 8 ^^0 analysis of zircon, an insulating silicate, does not appear to be possible without electron charge neutralisation. Although 1 per mil precision can be achieved the measured 8 changed continually with time. Consecutive data sets differed by up to 10 per mil and the range of 8 ^^O was 35 per mil. Our data shows that the primary beam Cs current (sputter rate) controls the 8 value of the emitted secondary ions.

7/7

SGTSG Field Meeting 2003

THE COOPER BASIN ARGON LABORATORY: NATURES TEST OF THE K-FELDSPAR MDD MODEL Sandra McLaren, W. Jim Dunlap Research School of Earth Sciences, Australian National University, Acton, ACT 0200, email, sandra.mclaren@anu.edu.au, jim.dunlap@anu.edu.au.

Argon thermochronology has proven to be a very useful tool in reconstructing thermal histories of deformed magmatic and metamorphic rocks that have been exhumed and cooled to surface conditions. In the vast majority of such cases, argon data for micas and amphiboles are interpreted in terms of cooling through some estimated closure temperature. In contrast, the age gradients exhibited by slowly cooled Kfeldspars may arise due to a range of closure temperatures from about 350°C to 150°C.

age variations are not representative of the apparent cooling history but reflect only later recrystallization processes. These workers refute the basis of the multi-domain method and suggest that the age gradients in K-feldspars are a direct reflection of late-stage recrystallization through processes including dissolutionreprecipitation and, in particular, perthitization. These processes lead to the development of a variety of microstructural subregions with mixed ages.

K-feldspars are one of the most widely used geochronometers in earth science research, largely because they are common in many kinds of continental crustal rocks and also because they retain argon quantitatively in a variety of situations. The large range of potential closure temperatures exhibited by K-feldspar samples, together with mathematical analysis of their argon release behaviour, has lead to the development of the 'Multiple diffusion domain' (MDD) model. The MDD theory suggests that the strong "^^Ar/^^Ar age gradients seen in K-feldspars are the result of variable argon retention by diffusion domains. The spectrum of diffusion domain sizes correlates with physical microstructure, and these domains are thought to close progressively to diffusive argon loss as the rocks are exhumed and cooled. If the domains have closed to diffusion the domain information can be inverted into continuous cooling histories that track the temperature-time, and hence exhumation history.

The Cooper-Eromanga Basin, in north-east South Australia provides an ideal natural laboratory in which to test the validity of the MDD model. The basin system contains a thick accumulation of Permian-Triassic and Cretaceous sediments, which is underlain by Carboniferous granites of the Big Lake Suite (and equivalents). These granites contain high concentrations of the heat producing elements, U, Th and K, and as a consequence the region is characterized by an extremely high geothermal gradient regime. Contemporary bottom hole temperatures in wells floored by granite (at 3-4 km depth) are between 160 and 230°C. These temperatures are likely to be within the closure temperature window for argon loss from Kfeldspars. We sampled K-feldspar bearing granites from a number of exploration drill-holes for "^^Ar/^^Ar and microstructural analysis. As the ages of the granites, the sedimentary history of the basins (including sedimentation rates), and the modern day thermal regime are all well known, the Cooper Basin K-feldspars provide an ideal parametric test of the competing MDD and recrystallization models.

However, despite the apparent success of the method in elucidating crustal thermal histories, it has been suggested that the effects of recrystallization on K-feldspar "^^Ar/^^Ar spectra may be significant, to the extent that the apparent

118

SGTSG Field Meeting 2003

FROM GONDWANATO OROGENY: INTEGRATED THERMOCHRONOLOGICAL CONSTRAINT OF AUS-PAC BOUNDARY EVOLUTION IN NEW ZEALAND Michael A. Cottam\ Geoffrey E. Batt^ Suzanne L. Baldwin^ ^Department of Geology Royal Holloway, University of London, Egham, Surrey, UK, TW20 OEX. E-mail: m.cottam@gl.rhul.ac.uk ^Department of Geology Royal Holloway, University of London, Egham, Surrey, UK, TW20 OEX. E-mail: g.batt@gl.rhul.ac.uk ^Department of Earth Sciences, 204 Heroy Geology Laboratory, Syracuse University Syracuse, NY 13244-1070, USA. E-mail: sbaldwin@syr.edu Thermochronological analysis of Australian Plate rocks from South Island, New Zealand reveal a multi-phase thermal history for the region constraining the development of the AUS-PAC boundary. Multichronometer cooling histories highlight two discrete cooling episodes separated by a well-constrained cooling hiatus. Initial cooling at -25-23 Ma is attributed to transpression associated with distributed deformation during early AUS-PAC development; later cooling at --8 Ma is linked to the onset of oblique convergence across the boundary zone. Thermochronology is presented for 14 granitoid and gneissic samples from a transect northwest of, and parallel to, the Alpine Fault the modern day surface expression of the AUSPAC boundary. Due to lower rates of exhumation Australian plate rocks record much fuller histories of AUS-PAC boundary evolution than their Pacific Plate neighbours. K-feldspar Ar/Ar analysis of samples, modelled using Multi Diffusion Domain principles, are supported, where possible, by envelopes of high temperature Ar/Ar Mica analysis and low temperature (U/Th)-He and Apatite fission track analysis. The resultant data range from probable Jurassic intrusion signals through to the exhumation of the last 5 Ma.

zone of distributed deformation which represented the first through going AUS-PAC structures in South Island: the 'proto Alpine Fault Zone'. This early transpressive exhumation lifted the granitic and gneissic bodies to mid-crustal depths, within the zones of partial retention for several chronometers, and thus beautifully poised to record the second cooling event. This initial cooling episode gives way to a wellconstrained cooling hiatus suggesting little or no exhumation. The latter, more rapid, and apparently ubiquitous, cooling is attributed to the initiation of oblique convergence across the wide AUS-PAC boundary zone following the movement of the Pacific Plate's pole of rotation, and the corresponding change in plate motion. This event is assigned an age of - 8 Ma, several million years earlier than previous estimates of the commencement of oblique convergence. The spatial patterns of these two discrete cooling phases have important implications for our understanding of the AUS-PAC boundary zone across New Zealand. These results also offer new insight into the style and processes of fault propagation at such plate boundaries and the way in which these structures evolve in response to changing boundary conditions.

The initial cooling episode at '-25-23 Ma, not seen in all samples, is attributed to transpression associated with the development of the wide

119

SGTSG Field Meeting 2003

LATE MIOCENE - RECENT TECTONIC DEVELOPMENT OF THE NORTHERN CORDILLERA REAL, ECUADOR: NEW INSIGHTS FROM (U-TH)/HE THERMOCHRONOLOGY P-V. Crowhurst^ and R.A. Spikings^ 1. CSIRO Petroleum, PO Box 136, North Ryde, New South Wales 1670, Australia. E-mail: peter.crowhurst@csiro.au 2. Section des Sciences de la Terre, Universite de Geneve, Rue des Maraichers 13, CH-1211 Geneve 4, Switzerland. E-mail: Richard.Spikinqs@terre.uniqe.ch

Numerous apatite fission-track (AFT) thermal history reconstructions along the Andean Cordillera have identified cooling commencing at 9 Ma and continuing until the present day along the entire orogen. The onset of cooling is coincident with the well-established Quechuan phases of Andean tectonism and is generally attributed to erosional and tectonic exhumation. However, the lack of sensitivity of the AFT method at temperatures <'-60°C, coupled with ambiguity regarding the initial conditions of track annealing (e.g. the length of an unannealed fission track) suggests the AFT method may have frequently failed to identify specific periods after 9 Ma when cooling rates were high. Forward modeling of (U-Th)/He apatite age data obtained from the juxtaposing Paleozoic Mesozoic Loja and Salado terranes in the northern Cordillera Real, has improved the resolution of the AFT thermal histories for the last 9 My. Each terrane yields a linear (U-Th)/He agealtitude profile, which suggests they acted as coherent blocks during periods of cooling driven by exhumation. The Loja terrane resided at

120

temperatures >70-80°C until --2 Ma, when it cooled rapidly to temperatures <40°C at rates >25°C/my. In contrast, the Salado terrane preserves an older, two stage history and cooled rapidly from >90°C to <40°C at 5-4 Ma and 9-8 Ma at rates as high as 40"C/my. Cooling and exhumation at 9 Ma was previously identified by AFT analysis and has been attributed to increased compressive stress driven by the flat subduction of the Carnegie Ridge. Cooling and exhumation at 5 Ma was also previously identified by AFT analysis in far northern Ecuador although the new models show that the basement terranes in central Ecuador were exhumed as coherent blocks during this period. Vertical reactivation of the Llanganates fault, which separates the Salado and Loja terranes at --5 and -'2Ma is coincident with the main stages of formation of the juxtaposed Interandean Valley structure (ramp valley), which provides further constraints on the growth phases of the valley and the cordillera.

SGTSG Field Meeting 2003

POSTERS MONDAY 22 SEPTEMBER 2003

121

SGTSG Field Meeting 2003

122

SGTSG Field Meeting 2003

JURASSIC TO MIOCENE MAGMATISM IN MYANMAR AND THE TECONIC EVOLUTION OF SE ASIA M-E- B a r l e v \ K h i n Z a w ^ A . L . P i c k a r d \ ^School of Earth and Geographical Sciences, The University of Western Australia, 35 Stirling Highway, Crawley, WA 6009. (email mbarley@geol.uwa.edu.au) ^Centre for Ore Deposit Research, University of Tasmania, Hobart, Tasmania, 7001.

The Mesozoic to Tertiary tectonic evolution of Southeast Asia is the result of the convergence and collision of fragments of Gondwanaland with Eurasia culminating in the collision of India. A rapidly growing geochronological database is placing tight constraints on the timing and duration of magmatic episodes and tectonic events in the Himalayas, Tibet and eastern Indochina. However, there is little comparable high-precision geochronology for Myanmar. Myanmar is made up of the west Myanmar and Shan-Thai (Sibumasu) terranes, separated by the right lateral Sagaing Fault. This SHRIMP U-Pb in zircon geochronology on granitoids from the Mogok Metamorphic Belt (MMB, a belt of high grade metamorphic rocks at the edge of the Shan-Thai Terrane), the Myeik Archipelago (Shan-Thai Terrane) and west Myanmar. Strongly deformed granitic orthogneisses in the Mogok Metamorphic Belt (MMB) near Mandalay contain Jurassic (-170 Ma) zircons that have partly recrystallised during -^43 Ma high-grade metamorphism. A hornblende syenite from Mandalay Hill also contains Jurassic zircons with evidence of Eocene metamorphism rimmed by thin zones of 30.9 ±0.7 Ma magmatic zircon. The relative abundance of Jurassic zircons in these rocks is consistent with suggestions that southern Eurasia had an Andean-type margin at that time. Mid-Cretaceous to earliest Eocene (120 to 50 Ma), l-type granitoids in the MMB, Myeik Archipelago and west Myanmar confirm that prior to the collision of India, an up to 200km wide magmatic belt extended along the Eurasian margin from Pakistan to Sumatra. The primitive l-type Khanza Chaung granodiorite in the Wuntho batholith in the west Myanmar terrane has a magmatic age of 94 ± 1 Ma. Triassic (-240 Ma), Jurassic (--170 Ma) and Early Cretaceous xenocryst zircons in this granitoid correspond with peaks of granitoid magmatism in the Shan-Thai terrane and establish that west Myanmar was part of the margin of Eurasia during the Mesozoic. In the Late Cretaceous west

Myanmar was 450 to 1100 km south of its present position outboard of the Shan Thai terrane of southern Myanmar. A suite of highly fractionated metaluminous to peraluminous l-type granitoids with associated Sn-W-Ta mineralisation em placed in the Myeik Archipelago of southern Myanmar (Shan-Thai terrane) have magmatic ages of 82 ± 1.4 Ma (Kawthoung), 62 ± 1.2 Ma (Hermyngi) and 50 ± 0.5 Ma (Auk Bok). Xenocryst zircons in these granitoids are either Proterozoic or derived from older members of the suite. This suite which extends into adjacent peninsular Thailand and was em placed into thickened continental crust well inboard of the subduction zone during rapid convergence and subduction of the IndiaAustralia plate. The primitive l-type Shangalon granodiorite in the Wuntho Batholith of west Myanmar has a magmatic age of 38.5 ± 0.6 Ma Ma indicating that subduction continued until about 40 Ma. Metamorphic overgrowths to zircons in the MMB orthogneiss near Mandalay date a period of Eocene (-43 Ma) high-grade metamorphism possibly during crustal thickening related to the initial collision between India and Eurasia (at 65 to 55 Ma). This was followed by emplacement of syn-tectonic hornblende syenites and leucogranites between 35 and 23 Ma. Similar syn-tectonic syenites and leucogranites intruded the Ailao Shan-Red River shear belt in southern China and Vietnam, and the Wang Chao and Three Pagodas faults in northern Thailand (that most likely link with the MMB) were also active at this time. The complex history of Eocene to early Miocene metamorphism, deformation and magmatism in the MMB provides evidence that it may have played a key role in the network of deformation zones that accommodated strain during the northwards movement of India and resulting extrusion or rotation of Indochina.

123

SGTSG Field Meeting 2003

DETRITAL FOOTPRINT OF THE MOZAMBIQUE OCEAN: U/PB SHRIMP AND PB EVAPORATION ZIRCON GEOCHRONOLOGY OF METASEDIMENTARY GNEISSES IN EASTERN MADAGASCAR Alan S. Collins^ Alfred Kroner^, Ian C.W. Fitzsimons^ and Theodore Razakamanana^ ^ Tectonics Special Research Centre, Department of Applied Geology, Curtin University, GPO Box U1987. Perth, Australia. ^ Institut fur Geowisschenshaften, Universitat Mainz, 55099 Mainz, Germany ^ Departement des Sciences de la Terre, Universite de Toliara, Toliara, Madagascar. Corresponding author: a.s.collins@curtin.edu.au. Fax: +61-8-9266-3153

U-Pb Sensitive High-mass Resolution Microprobe (SHRIMP) and Pb evaporation analyses of detrital zircons from metasedimentary rocks in eastern Madagascar reveal that: 1) The protoliths of many of these rocks were deposited between --800 and 550 Ma; 2) these rocks are sourced from regions with rocks that date back to over 3400 Ma, with dominant age populations of 3200-3000 Ma, -2650 Ma, -2500 Ma, and 800-700 Ma. The Dharwar Craton of southern India is a potential source region for these sediments, as here rocks date back to over 3400 Ma and include abundant gneissic rocks with protoliths older than 3000 Ma, sedimentary rocks deposited at 3000-2600 Ma and granitoids that crystallised at 2513-2552 Ma. The 800-700 Ma zircons could potentially be sourced from elsewhere in India or from the Antananarivo Block of central Madagascar in the latter stages of closure of the Mozambique Ocean. The region

124

of East Africa adjacent to Madagascar in Gondwana reconstructions (the Tanzania craton) is rejected as a potential source as there are no known rocks here older than 3000 Ma, and no detrital grains in our samples sourced from Mesoproterozoic and early Neoproterozoic rocks that are common throughout central east Africa. In contrast, coeval sediments 200 km west, in the Itremo sheet of central Madagascar, have detrital zircon age profiles consistent with a central East African source, suggesting that two late Neoproterozoic provenance fronts crop out in east Madagascar at approximately the position of the Betsimisaraka suture. These observations support an interpretation that the Betsimisaraka suture separates rocks that were derived from different locations within, or at the margins of, the Mozambique Ocean basin and therefore, that the suture is the site of subduction of a strand of Mozambique Ocean crust.

SGTSG Field Meeting 2003

MORE THAN JUST THE ALPINE FAULT: RECONSTRUCTING THE INITIAL GEOMETRY OF THE AUS-PAC BOUNDARY THROUGH NEW ZEALAND Michael A, Cottam\ Geoffrey E. Batt^ ^Department of Geology, Royal Holloway, University of London, Egham, Surrey, UK, TW20 OEX. E-mail: m.cottam@gl.rhul.ac.uk ^Department of Geology, Royal Holloway, University of London, Egham, Surrey, UK, TW20 OEX. E-mail: g.batt@gl.rhul.ac.uk

Previously enigmatic rocks of the Eraser Complex immediately west of the Alpine Fault, the present AUS-PAC boundary in South Island, New Zealand, are remnants of Australian Plate rocks deformed by a former configuration of the AUS-PAC boundary zone. Correlation with several other enigmatic arcuate structures immediately west of the Alpine Fault and along strike from the Eraser Complex suggests this 'proto Alpine Fault zone' was a broad zone of tectonism deforming both Pacific and Australian Plate material, very different to the narrow, focused, zone of today. Much of this zone has been concealed or removed by subsequent oblique convergence across the AUS-PAC boundary. Geochemical analyses and lithological comparison show that rocks of the Eraser Complex, a fault bounded sliver of high-grade rocks directly west of the Alpine Fault, represent deformed equivalents of New Zealand's Western Province. The correlation of similar lithologies between Australia, New Zealand and Antarctica, for example - Devonian granitoid rocks intruded into Ordovician sediments (Karamea Suite and Greenland Group of New Zealand's western province = Lachlan Fold Belt granitoids of Australia = Admiralty Intrusives & Robertson's Bay group Antarctica), supports the widely held view that the Western Province of New Zealand

is part of the Australian plate and may be regarded as a "tectonically rafted slice" of Gondwana. Detailed thermochronometry of these rocks shows a multi-phase cooling history associated with the development and evolution of the AUSPAC boundary from the Mesozoic onwards. KFeldspar Ar/Ar and Apatite (U/Th)-He thermochronology, both from within the Eraser Complex and other areas of South Island's 'West Coast', show that rocks of the Australian Plate have been deformed and exhumed during AUSPAC boundary evolution, albeit to a lesser extent than their Pacific Plate neighbours. These data show that Australian plate deformation is an important factor in AUS-PAC boundary evolution. Although less spectacular than exhumation of the Pacific Plate, Australian Plate deformation needs to be factored into attempts to model AUS-PAC boundary evolution, rather than being downplayed by regarding the Australian Plate as a simple rigid indentor. The recognition of this 'proto AUS-PAC boundary zone' also has significant implications for understanding the evolution of other obliquely convergent plate margins.

125

SGTSG Field Meeting 2003

REGIONAL ECLOGUE FACIES METAMORPHISM OF THE ADULA NAPPE: IMPLICATIONS FOR THE TECTONIC EVOLUTION OF THE EUROPEAN CENTRAL ALPS Jonathon Dale^ and Tim Holland^ ^School of Earth Sciences, University of Melbourne, Victoria 3010, Australia, jdale@unimelb.edu.au ^Department of Earth Sciences, University of Cambridge, Downing Street, Cambridge, CB2 3EQ, UK

Significant recent progress has been made towards unravelling the Mesozoic-Tertiary evolution of the Central Alps. The current synthesis involves the collision of Europe with Apulia (Africa) following the subduction of the intervening crust in a long-lived south-facing subduction zone. The highest pressures are recorded by metamorphic rocks which were near the southern margin of Europe prior to subduction, in the pelitic schist-hosted, mafic pods of the Adula Nappe. These rocks (and those of the associated Cima-Lunga unit to the west) preserve the only record of Tertiary eclogite facies metamorphism in the Central Alps. Within the Adula Nappe, metabasic mineral assemblages indicate an increase in pressure from north to south, consistent with the proposed south-dipping subduction zone. However, the detailed geometry of this subduction zone remains poorly understood. This study seeks to impose constraints via integrated petrology and thermobarometry of the high-pressure amphibolite- and eclogite-facies rocks. In the northern Adula Nappe, metabasic rocks preserve high-pressure amphibolite- to eclogite facies assemblages dominated by garnet and either barroisite or glaucophane. The modal abundance of omphacite in these rocks increases from north to south until in the central Adula Nappe, eclogites (modal garnet + omphacite > 70%) are widespread. Within the central Adula Nappe, kyanite modes in eclogites increase southward while mica and amphibole modes decrease. In the southern Adula Nappe, eclogite facies assemblages are composed dominantly of garnet and omphacite, however retrogression of eclogites is pervasive and eclogites are rare. Metabasic assemblages were hydrated by reactions driven by decompression. Garnet, barroisite and omphacite are replaced by chlorite, hornblende and plagioclase in amphibolites. Eclogites show several stages of re-equilibration under high-pressure conditions involving the replacement of kyanite by white

126

micas and the growth of amphiboles. Within highpressure amphibolites and eclogites, the patterns of growth zoning in garnets, a decrease in the Sicontent of phengite and an increase in the pargasite component of amphiboles indicate that these phases grew during decompression and heating under high-pressure conditions. Thermobarometric calculations (average-PT (THERMOCALC), garnet-amphibole and garnetclinopyroxene thermometry) establish that at maximum pressures (Pmax), the metamorphic conditions in the Adula Nappe were 17 kbar/640°C (north), 22 kbar/750°C (central) and 25 kbar/750°C (south). Average-Pf calculations based on assemblages equilibrated before, during and after P^ax constrain PT-paths for several different parts of the nappe, and these PT-paths show clockwise trajectories with heating accompanying the early stages of decompression from P^ax- These PT-paths are displaced to successively higher pressures and temperatures from north to south. Textural and mineral chemical observations and thermobarometric calculations establish that since peak pressures and during decompression, northern and central parts of the Adula Nappe behaved as a coherent tectonic unit with a common metamorphic history. Assuming (1) that Pmax was reached in different parts of the nappe at similar times, and (2) that since Pmax there has been no significant changes in N-S nappe length, the metamorphic conditions determined for Pmax define a regional metamorphic field gradient of 0.2 kbar/km and 9.6°C/km increasing from north to south. This field gradient constrains the depth of the Adula Nappe at peak pressures to between 60 km (north) and 80 km (central) depth, corresponding to a dip of 45° in a south dipping subduction zone. These results are in excellent agreement with current structural- and geophysical-based tectonic models for the Central Alps.

SGTSG Field Meeting 2003

DEFINING TECTONO-STRATIGRAPHIC TERRANES IN THE ANTARCTIC PENINSULA USING LITHOSPHERIC HETEROGENEITIES A,A. Dean, P.T. Leat, and A.P.M. Vaughan Geosciences Division, British Antarctic Survey High Cross, Madingley Road, Cambridge CB3 OET United Kingdom Alison.Dean@bas.ac.uk

The Antarctic Peninsula has conventionally been described as a Late Triassic to Miocene continental magmatic arc that formed by eastward-directed subduction of proto-Pacific and Pacific oceanic plate beneath a part of the prebreak-up margin of the former Gondwana supercontinent. Current theory suggests that while this basic premise holds, the tectonic history is complicated by recognition of several faulted crustal blocks (tectono-stratigraphic terranes) that could have allochthonous or exotic origins. Mafic dykes are common, albeit minor, components of continental magmatic arc terrains throughout the world. Most have undergone varying degrees of fractionation and assimilation prior to emplacement in the crust, obstructing determination of their provenance. However, some preserve a primitive chemical character that reflects their origins. Such dykes, derived from partial melts in the lithospheric, or rarely asthenospheric mantle, provide palpable evidence as to the nature of their enigmatic source regions. Mafic dykes with primitive signatures (MgO > 8 wt%), indicating a relatively unmodified composition, were intruded as swarms in the

Antarctic Peninsula during a restricted interval between 110 and 90 Ma. Preliminary work identifies two distinct suites. Oscar II Coast, Graham Land dykes with low LREE/HREE ratios, are interpreted as derived from partial melting of asthenospheric mantle wedge, and Black Coast, Palmer Land dykes with high LREE/HREE ratios, are interpreted as derived from partial melting of sub-arc lithospheric mantle. The later dykes of supposed lithospheric derivation, plus crustal and mantle xenoliths and xenocrysts they carry, are the focus of this study, providing a geochemical profile of crust/mantle architecture beneath the Antarctic Peninsula, refining the characterisation of terranes previously defined by structural, stratigraphic and geophysical means. Initial results support a predominantly lithospheric source for Palmer Land basaltic dykes. Within this group at least three sub-groups have been identified indicating fundamental differences in lithospheric character across proposed terrane boundaries, which distinguishes distinctive differences in the subcontinental lithosphere in this region, and may indicate an allochthonous origins for some terranes.

127

SGTSG Field Meeting 2003

PALEOZOIC ACCRETIONARY OROGENESIS IN WESTERN MONGOLIA Arian H. piikstra\ Craig Buchan^, W. Dickson Cunningham^ Fraukje M. Brouwer"^ and Gombosuren Badarch^ ^Department of Applied Geology, Curtin University, GPO Box U1987, Perth, WA 6845, Australia ^Tectonics Special Research Centre, Curtin University, GPO Box U1987, Perth, WA 6845, Australia ^Department of Geology, University of Leicester, University Road, LE1 7RH, Leicester, UK "^institut fur Geologie, Universitat Bern, Baltzerstrasse 1, 3012 Bern, Switzerland ^Institute of Geology and Mineral Resources, Ulaan Baatar, Mongolia

The margins of the Precambrian Siberian craton exposed in Central Asia consist of a tectonic collage of "terranes", which amalgamated during the Late Proterozoic-Early Palaeozoic (Sengor et al., 1993). This collage, referred to as the Altaids Orogen, was the site of extensive creation of juvenile continental crust during the Paleozoic. The Altaids Orogen is also the proposed type-area of so-called "Turkicstyle" or accretionary orogenesis. Accretionary orogens form by continuous accretion of subduction complexes, turbidite basins, island arcs, and oceanic fragments above a long-lived, oceanward migrating subduction zone system. The Altaids Orogen is one of the largest accretionary orogens in the world, yet is also one of the least well studied. Accretionary orogenesis in the Altaids ended with the docking of the North China Craton during Permo-Triassic times. The Mongolian Altai Mountains is one of the key areas in which the development of the Altaids Orogen can be studied. The Altai is essentially a late Cenozoic intracontinental transpressional mountain belt (Cunningham et al., 1996). However, Cenozoic structures are localised and Late Proterozoic-Early Palaeozoic plutonic and polydeformed metamorphosed basement rocks are exposed in uplifted fault blocks. Recent work in Central and Western Mongolia by the team of the University of Leicester has highlighted the importance of microcontinental fragments in the Altaids tectonic collage (Buchan et al., 2001). The docking of these microcontinental fragments has lead to punctuated 'mini-collisional' episodes during the development of the Altaids. Several of such possible microcontinental fragments have been recognised in outcrop within Central and Western Mongolia by our team, mainly on the basis of their high metamorphic grade and complex polyphase deformation history. In this presentation we present detailed structural observations and cross-sections from selected key areas in Western Mongolia. In general, we recognise three main lithotectonic units related to the development of the Altaids Orogen: (1)

128

amphibolite to granulite grade basement orthogneisses and paragneisses. (2) regionally widespread greenschist to amphibolite facies, flysch-like metasediments (Cambrian to Silurian?), often with a volcaniclastic component, (3) largely unmetamorphosed and weakly deformed shallowmarine to continental clastic deposits (Devonian?) interpreted as a post-orogenic, molasse-like sequence. In addition, numerous granitoid intrusions of various Paleozoic ages intrude the sequence. An important Palaeozoic structural relationship is exposed in the Dariv Range. In this area, a NeoProterozoic shelf-type sequence and high-grade Precambrian metamorphic basement were thrust towards the north over a (-570 Ma) ophiolitic sequence with clear supra-subduction zone characteristics (in fact, we deem this highly unusual Dariv Ophiolite, which has a dacitic sheeted dyke complex, a fore-arc ophiolite type locality). The Dariv Range area is interpreted as a suture zone formed by the collision of continental fragments during the Cambrian. A similar suture zone occurs further east in Mongolia, marked by the Bayankhongor Ophiolite (Buchan et al., 2001). Recognition of this suture zone in the Altai region has important implications for the amalgamation history of basement terranes in the region and the polarity of Paleozoic subduction. Our results confirm that the Altaids Orogen mainly grew by accretion of oceanic and turbidite basins, but we argue that pre-existing continental blocks (possibly fragments of Gondwana?) played a greater role in the development of the CAOB than previously appreciated. [References: Buchan et al., J. Geol. Soc. London 158, 445-460, 2001; Cunningham et al.. Tectonics 15, 142-156, 1996; Sengor et al.. Nature, 364, 299-307, 1993.

SGTSG Field Meeting 2003

IS NEOPROTEROZOIC TECTONISM IN DRONNING MAUD LAND, MOZAMBIQUE AND SRI LANKA A SOUTHERN EXTENSION OF THE EAST AFRICAN OROGEN OR AN EASTERN EXTENSION OF THE ZAMBEZI OROGEN? I.C.W. Fitzsimons Tectonics SRC, Applied Geology, Curtin University, GPO Box U1987, Perth WA 6845, Australia. i.fitzsimons@curtin.edu.au

The East African Orogen has long been interpreted as a north-south trending collision zone resulting from late Neoproterozoic eastwest closure of the Mozambique ocean during the Gondwana assembly, but the continuation of this suture into Dronning Maud Land of Antarctica remains enigmatic. The location of a collisional suture is commonly identified by rock types typical of the suture zone itself, such as ophiolites or eclogites, but such features may be poorly preserved in deeply eroded Precambrian orogens. Given that a major suture zone is unlikely to juxtapose crustal blocks with similar histories, another constraint on the location of a suture is that it should not pass through a region of apparently consistent geology. One such region within the likely southern extension of the Mozambique suture, is the Maud Province of Antarctica, and correlative rocks in Mozambique and Sri Lanka. THE MAUD PROVINCE AND ITS GONDWANA NEIGHBOURS The Maud Province of western Dronning Maud Land is characterized by high-grade gneiss with metamorphic ages of 1090-1030 Ma. Further east, in central Dronning Maud Land, magmatism and granulite-facies metamorphism at 650-500 Ma is attributed to closure of the Mozambique ocean, but this region still preserves older 10901030 Ma metamorphic ages. Similar ages are obtained from gneisses along the eastern edge of the Zimbabwe craton in western Mozambique, the Lurio Foreland of northeastern Mozambique, and Sri Lanka, which lie adjacent to Dronning Maud Land in Gondwana reconstructions. It is likely that these represent dispersed fragments of an originally contiguous late Mesoproterozoic terrane, and that any Neoproterozoic suture zone developed during the assembly of Gondwana must pass to one side of this terrane. It has been suggested that these rocks formed a microplate in the middle of the Mozambique ocean. However, lack of a pervasive Neoproterozoic

overprint in western Dronning Maud Land and western Mozambique implies that late Mesoproterozoic rocks in these regions developed in situ adjacent to the eastern margin of the Archaean Kaapvaal-Zimbabwe craton. Indeed these rocks have been interpreted as a Mesoproterozoic collision zone between the Kaapvaal-Zimbabwe craton and another unidentified craton. It follows that the Maud terrane is likely to lie on the southern African side of any Mozambique suture zone. DOES THE EAST AFRICAN OROGEN PASS INTO ANTARCTICA AT ALL? In fact there is some doubt whether the Mozambique suture and East African orogen extend into Antarctica at all. Broadly north-south trending structures (present-day coordinates) consistent with the orientation of the East African Orogen are developed close to the margin of the Kaapvaal-Zimbabwe craton in Mozambique and Antarctica, but are associated with strike-slip movement and not convergence. Late Neoproterozoic structural trends in the Lurio Belt, Sri Lanka, and in central Dronning Maud Land and Sor Rondane Mountains of Antarctica are locally associated with thrusting, but are subparallel to the Antarctic coastline in a Gondwana fit and at a high angle to the expected trend of the East African Orogen. These structures are, however, consistent with the trend of the Neoproterozoic Zambezi Orogen of south-central Africa, where there is structural, petrological and isotopic evidence for Neoproterozoic ocean closure and continental collision. This suggests that Neoproterozoic tectonism in Dronning Maud Land, the Lurio Belt, and Sri Lanka developed along the southern margin of an extension of a "Zambezi" suture, rather than along a western margin of a "Mozambique" suture. If this suture passes into Antarctica it must do so in the Lutzow Holm Bay region of eastern Dronning Maud Land.

129

SGTSG Field Meeting 2003 DOES THE LATE NEOPROTEROZOIC DARLING FAULT ZONE OF WESTERN AUSTRALIA EXTEND ALL THE WAY TO THE TRANSANTARCTIC MOUNTAINS? I.C.W. Fitzsimons Tectonics SRC, Applied Geology, Curtin University, GPO Box U1987, Perth WA 6845, Australia. i.fitzsimons@curtin.edu.au

The Darling Fault extends north-south for 1000 km along the western margin of Australia and its curvilinear trace dominates magnetic and gravity Images of the region. Its morphology reflects Mesozoic rifting, but it exploited an older structure called the Darling Fault Zone developed during Neoproterozoic transcurrent movement. This zone juxtaposed the Archaean Yilgarn craton and Mesoproterozoic Albany-Fraser orogen with late Mesoproterozoic to Neoproterozoic gneissic blocks exposed along the western edge of Australia, collectively called the Pinjarra Orogen. East of the fault zone, there was no pervasive tectonism or magmatism after 1130 Ma and rocks have TDM Nd model ages older than 1.8 Ga, but rocks west of the fault zone underwent magmatism and deformation at 1100-1000, 750-700 and 550-500 Ma and have TDM Nd model ages of 2.2-1.1 Ga. The continuation of this boundary into Antarctica is poorly constrained, but there is evidence that it extends across Antarctica to the Transantarctic Mountains. QUEEN MARY LAND COAST The extension of the Darling Fault Zone into Antarctica is most likely marked by the Denman and Scott glaciers of the Queen Mary Land coast. Mesoproterozoic gneiss in the Bunger Hills, immediately east of these glaciers, has TDM Nd model ages of 2.3-1.8 Ga, no evidence of pervasive tectonism or magmatism after 1130 Ma, and an identical history to the Albany-Fraser orogen, whereas rocks to the west include granitoid protoliths emplaced at 3000 and 550 Ma, evidence of metamorphism at 1050 and 500 Ma, and TDM Nd model ages of 3.2-1.6 Ga. Limited corridors of aeromagnetic data in Queen Maud Land are too widely spaced to identify major crustal boundaries beneath the ice. There is some evidence of a "Bunger Anomaly" near the coast but its continuity is impossible to assess from the data available and its interpretation is uncertain.

130

LAKE VOSTOK Given the consistent orientation of the Darling Fault Zone in Australia for 1000 km, it might be expected to continue across Antarctica with the same north-south trend. This would take it through the Lake Vostok region of the plateau, where a recent detailed geophysical survey has Identified a north-south trending boundary between two regions of quite distinct gravity and magnetic characteristics. The location and orientation of this boundary is consistent with it being a continuation of the Darling Fault Zone. TRANSANTARCTIC MOUNTAINS A continued north-south trend across Antarctica leads to the central Transantarctic Mountains, where there is also evidence of a boundary in the Precambrian basement. Isotope data for the Granite Harbour Intrusive suite, emplaced during the Cambro-Ordovician Ross orogeny, show a distinct jump in TDM Nd model age at the Shackleton Glacier, from 2.2-1.6 Ga in the northern and central Transantarctic Mountains, to 1.5-1.1 Ga in the Queen Maud and Horlick Mountains. These ages mirror those on either side of the Darling Fault Zone in Australia, and the presence of 1100-1000 Ma zircon xenocrysts in granites from the Pensacola and Queen Maud Mountains allow further correlations with the Pinjarra Orogen. Traditionally this region of younger basement in the central Transantarctic Mountains has been interpreted as an allochthonous terrane accreted to the Ross margin, and older basement is inferred to lie inboard under the ice. It is also possible, however, that it is the continuation of the Pinjarra Orogen from Queen Maud Land, via Lake Vostok, to the Pacific margin of the Antarctic craton. More geophysical data are needed from the plateau to demonstrate whether these features reflect a fundamental boundary or a coincidental alignment of unrelated structures.

SGTSG Field Meeting 2003

VARIATION IN METAMORPHIC STYLE ALONG THE NORTHERN MARGIN OF THE DAMARA OROGEN, NAMIBIA. 1

Ben Goscombe , David Gray , Martin Hand ^School of Earth and Environmental Sciences, University of Adelaide, 5005, S.A. ^School of Earth Sciences, University of Melbourne, Parkville, 3010, Victoria.

The northern margin of the Inland Branch of the Pan-African Damara Orogen in Namibia, shows dramatic along strike variation in metamorphic character during convergence between the Congo and Kalahari Cratons (M3). Low-P contact metamorphism with anticlockwise P-T paths dominate in the west and high-P Barrovian metamorphism with a clockwise P-T path occurs in the east. The sequence of M3 mineral growth in contact aureoles show early growth of cordierite porphyroblasts that were pseudomorphed to biotitechlorite-muscovite at the same time that a andalusite-biotite-muscovite transposed foliation is developed in the matrix. The peak metamorphic assemblages were overprinted by crenulations and retrograde chlorite-muscovite. The KFMASH PT pseudosection for metapelites in the Ugab Zone and western Northern Zone contact aureoles, indicate tight anticlockwise P-T loops through peak metamorphic conditions of 540-570 and 2.5-3.2 kb. These semi-quantitative P-T loops are consistent with average PT calculations using THERMOCALC, which give a pooled mean of 556±26 and 3.2±0.6 kb, indicating a high average thermal gradient of 50 °C/km. In contrast, the eastern Northern Zone experienced deep burial. high-P/moderate-T Barrovian metamorphism with an average thermal gradient of 21 °C/km and peak metamorphic conditions of approximately 635 and 8.7 kb. The calculated PT pseudosection and garnet compositional isopleths in KFMASH, appropriate for the metapelite sample from this region, document a clockwise P-T path. Early plagioclase^-kyanitebiotite parageneses evolve by plagioclase consumption and the growth of garnet to increasing Xpe. XMg and Xca and decreasing XMn compositions, indicating steep burial with heating. The developed kyanite-garnet-biotite peak metamorphic parageneses were followed by the 2 resorption of garnet and formation of plagioclase moats, indicating decompression, which was

followed by retrogressive cooling and chloritemuscovite growth. The clockwise P-T loop is consistent with the foreland vergent fold-thrust belt geometry in this part of the northern margin. The Ugab Zone and western Northern Zone show evidence for an earlier M2 metamorphism that produced regionally pervasive foliations with low-P biotite-muscovite-chloritetandalusite parageneses formed on high average thermal gradients, possibly due to a period of high heat flow and associated granite emplacement at 580-570 Ma. Pervasive matrix foliations (M2) were overprinted by contact metamorphic parageneses (M3) in the aureoles of 530±3 Ma granite in the Ugab Zone and 553-514 Ma granite in the western Northern Zone. Available geochronological data suggests that convergence between the Congo and Kalahari Cratons was essentially coeval in all parts of the northern margin, with similar ages of 535-530 Ma for the main phase of deformation in the eastern Northern Zone and Northern Platform. NNE-SSW directed convergence across the Inland Branch of the Damara Orogen was also broadly contemporaneous with the 538-505 Ma deformation and high-grade metamorphism of the Central Zone immediately to the south. It is often assumed that there is no change in metamorphic style along the length of linear to arcuate convergent orogens. The northern margin of the Inland Branch clearly illustrates the potential for contrasting styles of metamororphism developing at the same time in different sectors along the length of a single orogenic margin. In this case, two processes were important in this disparate thermal histories; (1) initial and evolving differences in crustal architecture, possibly including thickness of the passive margin lithosphere and thickness of the overlying sedimentary succession, and (2) variation in input of heat into different sectors by granite emplacement. Variation in metamorphic style in apparently simple linear/arcuate margins of high-angle convergent orogens, may be more wide spread than is currently acknowledged.

131

SGTSG Field Meeting 2003

THE METAMORPHIC RESPONSE OF TRANSPRESSIONAL OROGENESIS: THE KAOKO BELT, NAMIBIA. 1

Ben Goscombe , Martin Hand , David Gray , Joanna Mawby ''school of Earth and Environmental Sciences, Adelaide University, 5005, South Australia. 2 School of Earth Sciences, University of Melbourne, Parkville, 3010, Victoria.

The Kaoko Belt of the Damara Orogen, Namibia, is the deeply eroded core of a sinistral transpressional orogen that has half-flower structure geometry centred on the major, 4 -5 km wide Purros Mylonite Zone. Formed between the Congo Craton in the east and Rio De La Plata Craton in Brazil, the Kaoko Belt represents the northern coastal arm of a triple junction within the Pan-African Orogenic System. Consisting of reworked basement and a cover of Neoproterozoic Damara Sequence, the Kaoko Belt can be sub-divided structurally into three parallel NNW-trending zones. The East Kaoko Zone comprises sub-greenschist fades shelf carbonates that have been uprightly folded. The Central Kaoko Zone contains a slope and deep basin fades succession that has experienced intense deformation, including pervasive reworking of basement into large-scale east-vergent nappes. The Western Kaoko Zone is predominantly deep basin fades of high metamorphic grade intruded by numerous Damara granitoids. It has experienced intense wrench-style deformation with formation of upright isoclines and steep, crustal-scale shear zones. The Kaoko Belt evolved through three distinct phases of a protracted Pan-African Orogeny in the late Neoproterozoic to Cambrian. (1) An apparently long-lived early Thermal Phase in the west was responsible for rare granitoids at 660 Ma (M1) and later pervasive partial melting, granites and relic porphyroclasts between 580-570 Ma (early-M2). (2) The architecture of the belt formed in the main deformation Transpressional Phase (570-550 Ma) that reworked earlier parageneses forming the pervasive L-S fabric and associated matrix assemblages (M2). Deformation involved progressive sinistral transpression, evolving both temporally and spatially towards the margin from wrench-style to high-angle convergence accompanying foreland vergent thrusts and nappes. (3) Transpression terminated prior to minor reworking in the Shortening Phase (M3), which generated upright, open folds during N-S shortening (530-510 Ma) and is correlated with high-angle convergence in the Inland Branch of the Damara Orogen.

132

The Western Kaoko Zone is comprised of orogen-parallel panels of distinctly different metamorphic grade separated by crustal-scale, strike-slip ductile shear zones, the isograd pattern being indicative of low-angle extrusional tectonics in this, the orogen core. Peak metamorphic conditions and matrix assemblages were attained during earlyM2 at high average thermal gradients (29-40 °C/km) and were intensely reworked by lower-grade pervassive fabrics during M2. In the north, immediately adjacent to the Purros Mylonite Zone, the Khumib Terrane of garnet-grade experienced peak metamorphism at 566 ®C and 5.7 kb. Along strike to the south is the granulite-grade Hoarusib Terrane that experienced peak early-M2 conditions at 845 °C and 8.1 kb and M2 reworking at approximately 560-580 °C and 4.8 kb. To the west the Coastal Terrane experienced early-M2 metamorphism at sillimanite-K-feldspar-melt grades and was reworked during M2 at muscovite-biotite grades. In the Central Kaoko Zone, metamorphic grade increases towards the west to higher structural levels and so constitutes an inverted Barrovian metamorphic sequence. Peak metamorphic matrix assemblages formed during pervassive deformation in the Transpressional Phase (M2) at conditions ranging 530 to 690 °C and 8.5- 9.0 kb across the Central Kaoko Zone with low average thermal gradients (17-23 °C/km). Clockwise P-T paths were experienced in both the Central Kaoko Zone and Western Kaoko Zone. The developed mineral parageneses preserve different portions of the P-T loops in different parts of the orogen, the burial phase is preserved in the Central Kaoko Zone and exhumation phase in the Western Kaoko Zone. Garnet Sm-Nd geochronology indicates that matrix parageneses, early-M2 in the WKZ and M2 in the CKZ, formed at the same time within uncertainties (576±15 Ma). This indicates that the thermal peak was contemporaneous across the belt, even though deformational phases of equivalent structural style were diachronous across the Kaoko Belt.

SGTSG Field Meeting 2003

GRANULITES OF THE MALAWI MOSAIC 1

Ben Goscombe , David Gray , Martin Hand ^School of Earth and Environmental Sciences, Adelaide University, 5005, South Australia. ^School of Earth Sciences, University of Melbourne, Parkville, 3010, Victoria.

Malawi is centred in a triangle region, here called the Malawi Mosaic, at the junction of 4 major orogenic belts (Zambezi, Ubendian, Mozambique and Irumide orogenic belts) and must surely be a contender for one of the most complex sector of crust preserved. The Malawi Mosaic is composed of a number of juxtaposed terranes, all of granulite grade. P-T loops and metamorphic conditions in all terranes have been constrained by petrologic interpretation in published petrogenetic grids. Peak metamorphic conditions have been calculated using THERMOCALC v3.1 (Powell & Holland, 1998) for homogenized core compositions and assuming aH20=0.25 and presented as pooled means for each terrane. The peak parageneses are interpretively linked with published metamorphic zircon and monazite U-Pb ages. These terranes are briefly summarized, from north to south through Malawai. Lulomo and Nyika regions are components of the Ubendian belt and correspondingly preserve the longest history. Main phase orogenesis is two successive high-grade events on an overall clockwise P-T path, through a high-P peak of 867±81 and 10.1 kb at 2002 Ma in the Lulomo region and at 1990 Ma a low-P peak at 673±77 and 4.5±1.1 Ma preserved in the Nyika region. The Kasunga region preserves petrologic and geochronologic evidence for three metamorphic cycles. Similar to the Nyika region, early low-P metamorphism was at 1986 Ma but with an anticlockwise P-T path through peak conditions at approximately 850 and 4.0-6.0 kb. A later high-P clockwise P-T path granulite overprint at 860±67 and 10.5±1.9 kb possibly occurred at 1100 Ma and was followed by amphibolite fades metamorphism anywhere between 990 Ma and 630 Ma. The Chipata region has not been well constrained. Nevertheless, this region has strong similarities to the Zambezi belt and is a granulite terrane of between 1100 and 850 Ma age, and has been reworked by Barrovian metamorphism at 671±51 and 7.0±1.0 kb, anywhere between 800 and 524 Ma. The Lilongwe region experienced an overall clockwise P-T path through two successive granulite metamorphic peaks. Initially a high-P peak at 785±49 and 9.8±1.5 kb at 635 Ma is preserved in most parageneses, followed by a lowP peak at 775±83 and 6.6±1.2 kb at 580 Ma. Further south in the Dedza and Balaka regions the 580 Ma metamorphism is also recorded and similarly record an overall clockwise P-T path to a

second low-P granulite peak at 535 Ma. The first peak is at 763±41 °C and 9.2±1.0 kb and the second peak is at 777±51 and 7.6±1.4 kb. The Blantyre region in the far south of Malawi preserves a clockwise P-T loop through near eclogitic conditions of 930±57 and 14.3±2.1 kb, followed by decompression to high-P granulite conditions of 892±43 and 12.6±1.8 kb at approximately 590540 Ma. The terranes comprising the Malawi Mosaic show the following features; (1) High-grade metamorphic events spanning 2100 to 520 Ma, encompassing some 35% of earths history. (2) Evidence of reworking by multiple granulite fades events in individual terranes. (3) Geochronological evidence for at least 9 highgrade metamorphic events, with pooled means of age determinations in the wider region being; 2058±14, 1992±10, 1140-1080, 987±18, 938±11, 840±19, 646±13, 580±9 and 535±6 Ma. (4) An extraordinary range of peak metamorphic conditions from 673±77 and 4.5±1.1 kb to 930±57°C and 14.3±2.1 kb. (5) Average thermal gradients range 20 to 50 °C/km. (6) Various P-T paths are documented, most are clockwise with multiple thermal peaks. (7) Each terrane can be uniquely characterized by peak conditions, P-T paths, average thermal gradient and age of metamorphic events. This data indicates juxtaposed terranes of contrasting tectono-metamorphic style, which infers that a diversity of tectonic environments, over time, caused high-grade metamorphism across the Malawi Mosaic. Progressive continental growth was largely towards the south and out-board towards the east. Nevertheless the youngest granulite metamorphic events (535±6 Ma) are in the core of the complex, are of high-P type (9.0-14.0 kb) and these occur in-board of the 646±13 Ma Mozambique belt to the east. The latest 535±6 Ma period is correlated with main phase deformation during N-S convergence in the Pan-African Orogenic System, which spans the continent from Malawi through the Zambezi Belt to the Inland Branch of the Damara Orogen.

133

SGTSG Field Meeting 2003

STRUCTURAL AND METAMORPHIC ARCHITECTURE OF THE EAST NEPAL HIMALAYAS. 1

Ben Goscombe , Martin Hand , David Gray 1 School of Earth and Environmental Sciences, Adelaide University, 5005, S.A. ^School of Earth Sciences, Melbourne University, Parkville 3010, Victoria.

The whole of east Nepal between Mount Everest and Kangchenjunga has been mapped and nine detailed structural and metamorphic profiles across the Himalayan Metamorphic Front documented. This data, for the first time, accurately constrains the structural and metamorphic architecture of the east Nepal Himalayas and these results contrast with the current paradigm for the architecture of the Himalayas in general. The Himalayan Metamorphic front is comprised of three discrete structuralmetamorphic-chronometric terranes: (1) At the base a low-grade Lower-Plate of >2080m thickness. (2) At the top a high-grade Upper-Plate of >40404740m thickness. (3) Wedged between is an inverted Barrovian series in what is called the Main Central Thrust Zone (MCTZ) that ranges 3504050m in thickness. The main, crustal-scale structure controlling metamorphism is not the Main Central Thrust at the base of the Main Central Thrust Zone, as asserted for elsewhere in the literature, but is the High Himal Thrust at the base of the over-riding Upper-Plate. The Upper-Plate is composed of polydeformed high-grade gneisses that constitute part of the Neoproterozoic Greater Himalayan Sequences. Upper-Plate rocks are of granulite grade and metapelites have matrix assemblages of garnet-cordierite-sillimanite-K-feldspar. Peak metamorphism occurred at 20-22 Ma, at conditions of 837±59 and 6.7±1.0 kb, defining an average thermal gradient of 36 °C/km. This metamorphic terrane shows a discrete metamorphic break from the upper part of the underlying Main Central Thrust Zone, amounting to AT=+187 °C and AP—2.1 kb. Peak metamorphic garnets are compositionally flat, which is typical of the homogenisation experienced at high metamorphic grades. The Upper-Plate P-T path has not been confidently established. Apparently contradictory petrological evidence is preserved in these rocks; indicating both decompression (rare cordierite growth after garnet) and low-P hercynite-sillimanite-quartz prograde parageneses, suggesting a complex prograde P-T history that was ultimately terminated by nearisobaric cooling. The base of the Upper-plate is a 200-400 m thick high strain mylonite zone called the High Himal Thrust. Deformation fabrics in the High Himal Thrust show a complex evolution, the latest

134

movement episode produced a pervasive foliation with sillimanite-biotite±garnet±gedrite assemblages that formed at 674±33 and 5.7±1.1 kb. These assemblages are identical to those developed in shearbands that over-print peak metamorphic parageneses in both the basal Upper-Plate and upper part of the underlying MCTZ, indicating that the latest movement along the High Himal Thrust, juxtaposed these metamorphic terranes during their post-peak evolution. The High Himal Thrust is the only discrete, crustal-scale, high-strain zone in the Himalayan Metamorphic Front, and is the main structure controlling the metamorphic architecture. The MCTZ is comprised of Neoproterozoic to Ordovician metasediments of the Greater Himalayan Sequences and consitutes the Inverted Barrovian metamorphic series of the Himalayan Metamorphic Front. Clockwise P-T paths were experienced, with peak metamorphic conditions ranging from 550 °C to 650 °C at 8.8±1.1 kb, defining average thermal gradients of 18-21 °C/km. Garnets from the MCTZ show two-stage growth, recording evidence for two metamorphic peaks, possibly correlating with metamorphic events of 2022 Ma in the overlying Upper-Plate and 6-13 Ma in the underlying Lower-Plate. There is no discrete, high-strain structural break between the MCTZ and the Palaeoproterozoic Lesser Himalayan Sequences of the Lower-Plate. Strain is equally and pervasively partitioned by schistosic fabric development throughout the MCTZ and LowerPlate. Furthermore, metamorphic grade varies continuously throughout the entire metamorphic front below the High Himalayan Thrust with no significant metamorphic discontinuity between the MCTZ and the Lower-Plate. Lower-Plate rocks experienced a single metamorphic cycle at 6-13 Ma and peak metamorphic conditions of 561 ±12 °C at 8.5±1.9 kb, with average thermal gradient of 19 °C/km. What has been called the Main Central Thrust in the past is an unconformity between Palaeoproterozoic and Neoproterozic sedimentary sequences and was not a discrete high-strain zone during metamorphism.

SGTSG Field Meeting 2003

MECHANISM OF LATE CRETACEOUS FOLD-NAPPE EMPLACEMENT, OMAN AND TECTONIC CONSIDERATIONS 1

David Gray , John Miller , Robert Gregory ^School of Earth Sciences, University of Melbourne, Melbourne, 3010, Victoria. ^Department of Geological Sciences, Southern Methodist University, Dallas, TX, 75275, USA.

Strain magnitude and kinematic considerations require development of the major Saih Hatat foldnappe beneath the Samail Ophiolite nappe to involve: 1) underthrusting with fold-nappe development in the hanging wall, 2) a pinned, essentially undeformed upper limb, and 3) a'rolling hinge' where the upper limb cycles through the hinge into a zone of intense shear strain on the fold lower limb (e.g. Morcles nappe development in the Helvetic Alps). The upper limb of the fold (southwest side of the present dome) consists of relatively undeformed Permian to Cretaceous carbonates of the autochthon, whereas the lower limb (northeast side of the dome) has inverted stratigraphy, is isoclinally folded and intensely deformed. This structure is responsible for a deepening structural level towards the NE, and has produced an apparent NE-increase in deformation across the structural dome (c.f. Le Metour et al. 1990). This deformation gradient is reflected by strong fabric (L-S tectonite) development and increasing pressure of metamorphism, culminating in a pervasive schistosity and regional fold-nappes that have isoclinal, sheath-like fold geometry and markedly attenuated stratigraphy along fold limbs. Deformation associated with these early recumbent closures has produced a consistent N- to NE-trending stretching lineation. This lineation is defined by pressure shadows on framboidal pyrites, the long axes of deformed clasts in conglomerate units, mineral lineations defined by white mica, and more rarely by the long axes of pencils that form in units adjacent to the Saiq 2v volcanics. Within the fold-nappe there is a significant increase in strain towards a major structural break where X/Z strain ratios change from -16:1 to over 100:1; X, Y and Z are the maximum, intermediate and minimum principal

stretches respectively. Pressure shadows on pyrite are very common in limestones, dolomites, and some quartzites and mafic schists. In the foliation (XY plane) pressure shadows are long, straight to slightly arcuate, with tapered form. When fibres are curved they show a 20° to 40° rotation (commonly clockwise sense) of X during pressure shadow development. In XZ sections pressure shadow fibre tails are also generally long and straight, although those with slight curvature indicate top to the north shear sense. Strain magnitudes clearly vary with lithology; limestone> dolomite» mafic schist >quartzite. Units below the fold-nappe (lower plate) have a pervasive schistosity and stretching lineation associated with the formation of regional isoclinal folds with sheath-like form and hinges that are sub-parallel to the regional stretching lineation. These regional isoclines fold the earlier highergrade assemblages. Deformed conglomerates and calc-schists within the lower plate reflect a strong component of flattening accompanied by marked stretch in X, producing flattened "cigarlike" forms (constrictional strain) and extensive shear bands that indicate top to the NE shear sense. Strains here are more uniform than in the fold-nappe ranging from 20:1 to 30:1. Conclusion: Oman ophiolite obduction clearly involves ocean-vergent thrusting and over-folding within the continental margin platform to slope facies sequences. Structural/tectonic scenarios that accommodate the development of such folds and shear zones include lateral escape from a rising buoyant crustal slice along an inferred oceanward-dipping subduction interface, backfolding (retrocharriage) associated with major oceanwards-directed underthrusting (subduction), or simple underthrusting of the margin by the oceanic realm.

135

SGTSG Field Meeting 2003

BACK TO THE SOURCE: SHRIMP U/PB AGES OF ZIRCONS FROM THE ITREMO GROUP AND MOLO SEQUENCE OF CENTRAL MADAGASCAR Bregje Hulscher ^ and Ian C.W. Fitzsimons ^ 1 TSRC (Tectonics Special Research Centre), Dep. of Geology and Geophysics, The University of Western Australia, 35 Stirling Hwy, Crawley, WA 6009, Australia (e-mail: bhulscher@tsrc.uwa.edu.au) 2 TSRC, Dep. of Applied Geology, Curtin University of Technology, GPO Box U1987, Perth, WA 6845

During the assembly of Gondwanaland Madagascar lay within a major zone of collision between plates of Indian and 'African' origin. East Madagascar still contains a fragment of India. The question is when that fragment accreted and where the rest of Madagascar's tectonic units came from. Recent provenance studies have shed some light on Madagascar's tectonic history. In Central Madagascar, we recognise at least three metasedimentary sequences which are distinct in terms of their inferred depositional age and environment, their provenance and their geographical distribution. From west to east they are the Molo sequence, the Itremo Group and the metasediments that occur in the proposed Betsimisaraka suture of Collins and co-authors, which juxtaposes Central Madagascar with the fragment of India (the DhanA/ar Craton). The Itremo Group comprises quartzites, pelites and marbles and most likely has a Mesoproterozoic depositional age, making it significantly older than the Late Neoproterozoic Molo and Betsimisaraka quartzites and pelites that occur on either side of the Itremo Group. The Molo and Betsimisaraka metasediments were probably deposited during the East African Orogeny. At -560-520 Ma they were caught up in pervasive deformation and metamorphism up to granulite fades grade. Although the depositional age of these synorogenic metasediments is similar, they are separated by the lower strain, lower grade and much older Itremo Group. The question is whether the two synorogenic metasedimentary sequences could have been sourced by different continents. We conducted U-Pb SHRIMP analyses on zircons from the Itremo Group and Molo sequence. For the Itremo Group, previous results by other authors have been confirmed, with main peaks occurring at -1895 Ma and -2550 Ma. These two peaks represent two significant, but separate orogenic events that each produced igneous protoliths. Subsequent overgrowths on igneous zircons occurred at 2509 ± 10 Ma and -1855 Ma respectively. The -2.5 Ga event is also present in the basement of Central Madagascar and occurs not only as detrital grains in the Itremo Group but

136

also as xenocrystic zircons in granitoids that intrude the Itremo Group. The oldest of the intrusions in the Itremo Group gives a minimum age of deposition of 795±21 Ma (Andohamaho granite). In our study, the maximum age of deposition, given by the youngest concordant detrital zircon, is 1790±25 Ma. In addition to the -1.85 and -2.5 Ga ages, minor peaks occur at -2.07-2.17 Ga, -2.72 Ga, -2.89 Ga, and -3.09-3.23 Ga. The oldest concordant detrital zircon is 3231 ±16 Ma. The Mesoarchaean age population has not been identified in earlier studies of the Itremo Group. This population is significant as it affects the assessment of provenance sources. Collins reports similar 3000-3200 Ma detrital ages for the metasediments in the proposed Betsimisaraka suture. He adds there is no potential source of that age in the 'African' cratons. Since then, work on the Muva Group in Northern Zambia (De Waele, pers. com.) and a 'cryptic terrane' by Rainaud and coworkers has identified a source region for Mesoarchaean zircons in East Africa. In fact, all mentioned peaks in the Itremo Group from -1.8 to -3.2 Ga also occur in the Muva Group. Our data thus show that the Itremo Group can definitely be solely derived from an 'African' source. Our data also imply that for the metasediments in the proposed suture, the source could have been either India or 'Africa' (plus Central Madagascar) as both continents contain the -2.5 to 3.2 Ga zircon populations Collins identified. In the Molo sequence, minor peaks occur at -1.98 Ga and -800 Ma. Most zircons yield ages between 560 and 540 Ma. However, the latter ages reflect a resetting of detrital protolith ages as these recrystallised zircons yield metamorphic Th/U ratios for both cores and rims. The responsible granulite fades metamorphic event is -545 Ma and is related to the final amalgamation of East Gondwanaland during oblique convergence, which, if solely based on detrital zircon patterns, does not necessarily involve ocean closure within East Madagascar at this time.

SGTSG Field Meeting 2003

STRUCTURAL AND AMS EVIDENCES FOR THE OLBIQUE OPENING OF A MIOCENE DYKES SWARM ASSOCIATED WITH THE RAPID ACCELERATION OF THE PACIFIC PLATE SUBDUCTION UNDER SOUTHERN PATAGONIA (CHILE). Jean-Pierre Lefort\ Tahar Aifa^ and Francisco Herve^ ^ Techtonophysics, Geosciences-Rennes, Institut de Geologie. Campus de Rennes-Beaulieu, 35042, Rennes cedex, France. E-mail; lefort@univ-rennes1 .fr ^ Geophysics. Geosciences-Rennes. Institut de Geologie. Campus de Rennes-Beaulieu, 35042, Rennes cedex, France. E-mail; alfa@univ-rennes1.fr ^ Departamento de Geologia. Facultad de Ciencias Fisicas y Matematicas, Universidad de Chile. Casilla 10465 Correo21, Santiago, Chile. E-mail; fherve @cec.uchele.cl

Study of the orientation of 53 dykes in a Miocene doleritic dyke swarm located above the subduction of the Pacific oceanic floor beneath Southern America (Patagonia, Chile), shows that the opening of the dykes cannot be directly related with the azimuth of subduction during the same period of time. However, if we consider at this time: a/ the main fault direction, b/ the preferential orientations of the doleritic dykes, c/ the existence of two large transcurrent dextral shear zones parallel with the shoreline, d/ the azimuth of subduction of the Pacific floor at the time of the dykes emplacement, el the shape of the Taitao peninsula, f/ the macroscopic evidence of dextral and sinistral transtensions, g/ the direction and age of the graben recognized in the Golfo de Penas and the thrusts observed in the Cosmelli basin, a general kinematics model can be proposed. This model, which is based on a physical indentation experiment, characterized by

a weak confinement on one of its edges, explains why all these structures developed at the same time. Study of the shallow AMS tensors shows that the magnetic minerals located at the margins of many dykes witness an oblique opening (transtension) and confirms the general stress pattern which prevailed during Miocene time. Because of the size of the dyke swarm and of the evolution of the Pacific plate during Miocene time, the indentation cannot be a local feature. We postulate that the strong acceleration of the subduction which occurred at the same time as the formation of all these structures, changed the dip of the subduction plane, increased the friction forces above the subducting slab and dramatically modified the superficial stress. In Patagonia, this abrupt change in the dip of the subduction would have acted like a large indenter.

137

SGTSG Field Meeting 2003

THE ACATLAN COMPLEX, SOUTHERN MEXICO: RECORD OF THE CLOSURE OF THE RHEIC OCEAN? R. Damian Nance^ and J. Duncan Keppie^ ^Department of Geological Sciences, Ohio University, Athens, Ohio 45701, U.S.A., nance@ohio.edu ^Institute de Geologia, Universidad Nacional Autonoma de Mexico, Mexico D.F., 04510 Mexico

INTRODUCTION The Palaeozoic Acatlan Complex of southern Mexico comprises an assemblage of repeatedly deformed metasedimentary sequences, locally eclogitic maficultramafic suites, and arc-related granitic bodies that are considered to represent a vestige of either the lapetus or Rheic oceans. Forming basement to Mexico's Mixteco terrane, the complex is tectonically juxtaposed to the east against Grenville-age granulites (Oaxacan Complex) of the Zapotecan terrane, the Palaeo-zoic platformal cover of which bears fossils of Gondwanan affinity. The tectonothermal record of the Acatlan Complex is therefore important, not only to the accretionary history of Mexico, but also to the wider issues of Palaeozoic continental reconstructions. GEOLOGICAL SETTING The Acatlan Complex is made up of two major thrust sequences, a lower Petlalcingo Group and an upper Piaxtia Group, both unconformably overlain by the Pennsylvanian-Lower Permian Tecomate Formation. These sequences are deformed into two, major, N-S, upright folds separated by a steeply dipping zone. They are bound to the east by a dextral flower structure (Caltepec fault zone), which forms the boundary with the Grenville-age Oaxacan Complex. The lower sequence (Petlalcingo Group) is made up of thick low-grade metasedimentary rocks interpreted as either a passive margin sequence or trench and forearc deposits. The Petlalcingo Group comprises the Chazumba and Cosoltepec formations, and is pervasively migmatized at its base (Magdalena Migmatite) as a result of a possible plume-related thermal event of Jurassic age (171±1 Ma; U-Pb zircon on neosome) linked to the opening of the Gulf of Mexico. The Chazumba Formation comprises thick, polydeformed, amphibolite fades metapsammites and metapelites that contain several ultramafic-mafic tectonic lenses, also of Jurassic (171±1 Ma; U-Pb zircon) age. The Chazumba Formation is structurally overlain by the Cosoltepec Formation comprising extensive phyllites and quartzites, and minor mafic metavolcanic units. The Cosoltepec Formation contains 375-355 Ma detrital zircons, which provide a maximum depo-sitional age. Other detrital zircons

138

with ages of 560 Ma and ca. 1 Ga suggest a source in either the Yucatan Peninsula or Brasiliano orogens, and the Oaxacan Complex or Amazonian craton, respectively. The formation is metamorphosed in the greenschist facies and has been deformed by three sets of structures of Carboniferous and Permian age. The Petlalcingo Group is structurally overlain by the Piaxtia Group comprising eclogitic mafic and ultramafic rocks, high-grade metasedimentary units, granitic rocks and migmatites of the upper thrust sequence. These units may encompass several thrust slices and are inferred to represent obducted oceanic and/or continental lithosphere. The chemical affinities of the igneous suites and peak metamorphic grade of the Piaxtia Group vary from east to west: (a) In the eastern part of the steep zone, 442±1 Ma (UPb zircon) bimodal continental tholeiitic rocks were metamorphosed to upper amphibolite facies, cooling through 500-550°C by 414±15 Ma (^°Ar/'^Ar hornblende), (b) In the western steep zone, MORB/OIB basaltic rocks were metamorphosed to 14-15 kb and 500-550X in the south, whereas arc mafic rocks were metamorphosed to14-15kb and 600-650X in the north. Sm-Nd data define an isochron of 386±22 Ma interpreted as a metamorphic age, and garnet geochemical data plot in the Group C eclogite field, (c) West of the steep zone, basaltic lenses were metamorphosed to 5-7 kb at 315-330°C to produce blueschist. This Silurian-Devonian metamorphism appears to have been followed by partial melting of the metasediments to produce the S-type Esperanza granitoid. This has yielded a lower intercept U-Pb zircon age of 440±14 Ma and cooled through 550500°C by 418±18 Ma (U-Pb monazite). The upper intercept age of 1161+30 Ma probably represents inheritance from the metasediments, which contain detrital zircon derived from the Oaxacan Complex. A leucocratic phase of the Esperanza granitoid yielded an age of 478±5 Ma (U-Pb zircon). Collectively, these data indicate a complex interleaving of subducted oceanic lithosphere, volcanic arc, and tectonically thickened continental crust suggesting that the Piaxtia Group represents an accretionary prism. If so, the prism developed from the Arenig to the Devonian.

SGTSG Field Meeting 2003

Following exhumation, the Tecomate Formation was deposited unconformably on the Petlalcingo and Piaxtia groups. The formation consists of conglomerate, sandstone, slate, within-plate mafic and felsic volcanic rocks, and limestones that contain Pennsylvanian-Early Permian cono-donts. All units were then polydeformed in N-S, dextral shear zones and associated S-vergent thrusts synchronous with emplacement of the Totoltepec pluton at 289±1 Ma (U-Pb zircon). This Early Permian Ouachitan orogenic event included at least two phases of deformation that took place under greenschist facies conditions: (1) isoclinal folding associated with N-S dextral shearing and S-vergent thrusting during which the Totoltepec pluton was thrust over the Tecomate and Cosoltepec formations; and (2) NWthrough N- to NE-trending upright, open folding with an axial planar crenulation cleavage. Muscovite from the Tecomate Formation has yielded a K-Ar age of 288±14 Ma. Juxtaposition of the Acatlan Complex against the Oaxacan Complex along a major zone of dextral trans-pression (Caltepec fault zone) is likewise of Early Permian age (276±1 Ma, U-Pb zircon on syn-tectonic granite). The Acatlan Complex is unconformably overlain by the Leonardian Matzitzi Formation, which also oversteps onto the Oaxacan Complex. Locally, a listric normal fault places the Acatlan Complex against unmetamorphosed Late Devonian rocks that are themselves overlain unconformably by Mississippian-Leonardian sedimentary rocks. The Matzitzi Formation lacks penetrative deform-ation but was tilted before deposition of Meso-zoic rocks. Undeformed granitic dykes dated at 175±3 Ma and 172±1 Ma (Rb/Sr) locally cut the Chazumba Formation and the mafic-ultramafic lenses it contains. Unconformably overlying Jurassic rocks vary from continental to shallow-marine shelf sedi-ments and are common to both the Mixteco and Zapotecan terranes. Most reconstructions place a magmatic arc east of the Mixteco terrane in the Jurassic. TECTONIC SIGNIFICANCE Available evidence for the timing of tectono-thermal activity in the Acatlan Complex shows striking parallels with that of the Appalachian-Ouachita Orogen. Granitoid U-Pb zircon ages of ca. 480 Ma (Early Ordovician), ca. 440 Ma (Late OrdovicianEarly Silurian) and ca. 290 Ma (Early Permian), for example, closely match those of the Taconic, Salinic, and Alleghanian-Ouachitan tectonothermal episodes of eastern North America. Based on these broad tectono-thermal similarities, the Acatlan Complex has been linked to Appalachian-Ouachitan tectonics and the closure of the lapetus Ocean. Hence, the complex has been interpreted to represent a vestige of the lapetus that formed during Late Ordovician-Early Silurian collision (Acatecan Orogeny) between eastern Laurentia and

a Grenville-age crustal block (Oaxaquia), which Is inferred to underlie much of present-day Mexico and considered to represent either a microcontinent or part of the Columbian margin of Gondwana. In this context, the recently acquired U-Pb ages of ca. 480 Ma and 440 Ma for the Esperanza granitoid are particularly significant because, prior to these findings, the first deformation of the Acatlan Complex was considered to be of Devonian age (Mixtecan Orogeny). Models for its tectonic history were consequently focused on potential linkages with the Acadian event of the Appalachians. Evidence of a tectonothermal event of broadly Taconic or Salinic age within the complex broadens its parallels with the Appalachian Orogen and, thereby, strengthens its interpretation as a segment of lapetus. Support for this interpretation also lies in the complex's proximity to the Oaxacan Complex, the age and metamorphic grade of which is similar to that of the Grenville basement of the Appalachian Orogen. But direct correlation of the Oaxacan Complex with the Grenville Belt of Laurentia is precluded by the fact that the former lies east of the Acatlan Complex and is overlain by Palaeozoic strata with Gondwanan faunal affinities, whereas the latter lies west of the Appalachian Orogen and is overlain by marine strata bearing Laurentian fauna. For this reason, tectonic models attributing development of the Acatlan Complex to closure of lapetus do not place it on the Appalachian (Laurentian) margin, but instead, assign it to the opposing (Gondwanan) margin. However, unlike the continental margins of lapetus, which achieved rift-drift transition at the base of the Cambrian and so preserve early Cambrian shallowmarine successions, the oldest platformal strata that unconformably overlie the Oaxacan Complex are of late Cambrian-early Ordovician age, by which time the lapetus Ocean had already begun to subduct. In addition, lead isotope data indicate that the gneisses of the Oaxacan Complex were derived from a mixture of juvenile 1 Ga and Palaeoproterozoic-Archean sources. We therefore interpret the Acatlan Complex, not as a remnant of lapetus, but as a vestige of its Immediate successor, the Rheic Ocean. The Silurian-Devonian and Carboniferous-Permian events are synchronous with those recorded in the European vestiges of the Rheic Ocean. DISCUSSION AND CONCLUSIONS As recorded in the Appalachian Orogen, the closure of lapetus did not occur as the result of continentcontinent collision between Laurentia and Gondwana. Instead, it records accretion of smaller continental arc terranes (Avalonia and Carolina)

139

SGTSG Field Meeting 2003

that had rifted away from their former positions along the margin of Gondwana. That part of lapetus ahead of these peri-Gondwanan terranes was closed, while a new Rheic Ocean opened behind them. Opening of the Rheic was therefore concurrent with the closure of lapetus and produced a new and younger continental margin along the edge of Gondwana. Available evidence suggests a Late Cambrian to Early Ordovician age for the Rheic rift-drift transition, which matches the age of the oldest platformal strata overlying the Oaxacan Complex. Closure of the Rheic Ocean to form the Alleghanian-Ouachitan Orogen was the result of collision in the Late Palaeozoic between Laurentia (and its accreted terranes) and Gondwana. As a vestige of the Rheic Ocean, the Acatlan Complex can be interpreted to record events on

140

either the Rheic margin of Gondwana or the trailing margin of the peri-Gondwanan terranes. In this view, the first deformation of the complex between ca. 440 and 364 Ma could document an accretionary event on one of these margins concurrent with the closing Rheic Ocean, while the second (Carboniferous-Permian) episode would record the terminal closure of the Rheic Ocean itself. If so, the complex may have lain adjacent to NW South America throughout the Palaeozoic and its stratigraphic and tectono-thermal records promise new insight into the history of the Rheic Ocean, the closure of which produced the climactic Ouachitan-Alleghanian-Variscan orogenic belt between Gondwana and Laurussia during the Carboniferous-Permian assembly of Pangaea.

SGTSG Field Meeting 2003

GLOBAL PALAEOGEOGRAPHY IN LATEST PRECAMBRIAN AND CAMBRIAN Sergei A. Pisarevsky^ ^Tectonics Special Research Centre, The University of WA, 35 Stirling Highway, Crawley, WA 6009, Australia; e-mail: spisarev@tsrc.uwa.edu.au

Latest Precambrian (Vendian) and Cambrian times appear to be among the most enigmatic in the Earth's history. Several alternative tectonic reconstructions for this time interval have been published in recent years. Probably the most interesting discussion is happened around the history of the opening of the lapetus ocean. Most scientists agree that the lapetus ocean and Tornquist sea emerged as a result of major breakup of a supercontinent combined by Laurentia, Baltica and Amazonia with possible involvement of other crustal blocks. However, the initial configuration of these three major plates is debated. Recent palaeomagnetic, geochronological and geological data support "low-latitude" positions of Laurentia and Baltica before and during the opening of western lapetus and Tornquist sea. The relative position of these

two continents is likely to be roughly similar to their recent configuration. The "Baltica upsidedown" model is less justified and contradicts some recently published data. The assemblage of Gondwanaland was another major tectonic event in the Latest Neoproterozoic - Early Cambrian. East Gondwanaland resulted from an oblique collision of India/Rayner and Australia/Mawson continents, which may not have been complete until the Early Cambrian. History of the assemblage of the West Gondwanaland Is less clear, but it was also a prolonged process, and the final collision probably happened in Cambrian. Five global palaeogeographic reconstructions for 640, 600, 560, 530 and 500 Ma are proposed.

141

SGTSG Field Meeting 2003

FORMATION OF ARCUATE OROGENIC BELTS IN THE WESTERN MEDITERRANEAN REGION G. Rosenbaum and G.S. Lister Australian Crustal Research Centre, School of Geosciences, Monash University, Melbourne (qideon@mail.earth.monash.edu.au)

The Alpine orogen in the western Mediterranean region consists of three classic examples of arcuate orogenic belts: the Rif-Betic belt, the Apennine-Calabrian-Maghrebides belt, and the western Alps. All these belts contain highpressure/low-temperature (HP/LT) rocks of Cretaceous to Oligocene ages, which were exhumed during extensional deformation. The HP/LT rocks are presently exposed in the sole of metamorphic core complexes. The arcuate belts in the western Mediterranean follow the margins of postOligocene extensional basins: the Ligurian, Provengal, Algerian, Alboran and Tyrrhenian basins. The floor of these basins consists of attenuated continental crust (in Valencia Trough, the Alboran Sea and the northern Tyrrhenian Sea) or Neogene to Recent oceanic crust (in the Algerian-Provengal basin and the southern Tyrrhenian Sea). The basins have been developed since the Oligocene in the overriding plate, in the proximity of a northwest-dipping subduction zone. The widespread extension was simultaneous with the overall convergence motion of Africa with respect to Europe, suggesting that the development of back-arc basins in the western Mediterranean was driven by rollback of a subduction hinge, which was combined with relatively slow convergence rates of Africa-Europe since the Oligocene. We have studied structural relationships within different segments of the western Mediterranean Alpine orogen in order to reconstruct the formation of the arcuate belts. Our reconstruction shows that the arcuate shape of the Rif-Betic and the Apennine-CalabrianMaghrebides belts were obtained during subduction rollback of the African slab in a direction oblique or orthogonal to the direction of convergence. During subduction rollback, allochthonous units of Alpine origin were emplaced in an arcuate fashion in the periphery of the extensional basins. The ages of HP/LT

142

rocks usually predate the formation of oroclinal bending, suggesting that the HP/LT rocks were inherited from earlier collisional processes. In a lithospheric view, the subducting lithosphere in the western Mediterranean region evolved from a relatively linear NE-SW-striking slab located south of the European margin, to several narrow slab segments presently found beneath the Alboran region. North Africa and Italy. These remnant slabs are fragments of oceanic lithosphere consumed since the Oligocene, and can account for 98% of the Oligocene oceanic domain. The slab segments are partly detached from the surface and are separated from each other by regions of low velocity anomalies where tearing has occurred. The deepest slab tears are indicated by the absence of high-velocity anomalies down to depths of --600 km. Such tears are found between the Alboran slab and the North African slab and between the North African slab and the Calabrian slab. A shallower tear, which is characterised by the absence of lithospheric slab down to depths of -250 km, is found in the central Apennines, between the Calabrian slab and the northern Apennine slab. The formation of arcuate belts is therefore the crustal manifestation of lithospheric processes associated with progressive rollback and tearing of subducting slabs. Deep tears occurred where the retreating subduction zone collided with the northern margin of Africa resulting in the incorporation of continental crust at the subduction system. In the Gibraltar and the Calabrian arcs, the existence of narrow oceanic passages enabled further rollback in a direction that was roughly perpendicular to the direction of convergence. These changes in rollback directions could only have occurred after tearing of the slabs. During subduction rollback, these tears accommodated strike-slip motions, leading to an overall 'extrusion' of the Alboran to the west and Calabria to the east.

SGTSG Field Meeting 2003

MODELLING INSIGHTS INTO THE DYNAMICS OF SUBDUCTING SLABS IN THE UPPER MANTLE W.P, Schellart* and G.S. Lister ACRC, School of Geosciences, Monash University, Melbourne, VIC 3800 (*wouter@maH.earth.monash.edu.au)

Wadati-Benioff zones and high seismic velocity zones in the Earth's interior indicate that oceanic lithosphere sinks into the mantle, attaining a wide variety of shapes. Geological and Geophysical data suggest that subducting slabs not only move in a downdip slab-parallel direction, but also in a slabperpendicular backward direction. Such a slabperpendicular backward component would result in regressive (oceanward) hinge-line migration of the subducting lithosphere (i.e. slab rollback). Regressive motion would then result in extension in the overriding plate (i.e. back-arc extension) due to collapse of the overriding plate towards the retreating hinge-line. Thus, back-arc extension seems to be a direct consequence of slab rollback. As evidenced by the episodic opening of back-arc basins and the wide variety in shapes (from roughly symmetrical to highly asymmetrical) of arc - backarc systems, it is clear that slab rollback is a transient three-dimensional process with significant trench-parallel variations in retreat velocity. This concept is supported by GPS-data and paleomagnetic data from a number of arcs across the globe. The process of subduction and slab retreat has been investigated previously with numerical and analogue models. In most of the numerical models, slab retreat was externally imposed and therefore the self-consistent dynamical behaviour could not be investigated. Furthermore, the experiments were two-dimensional, therefore excluding threedimensional complexities, which are an integral part of arc - back-arc systems. Previously conducted analogue experiments, however, were executed in three dimensions. The analogue models were designed to investigate the influence of a stratified mantle on the subduction process, to investigate subducting plate - overriding plate interaction, or to investigate hinge-retreat. In this investigation, fluid dynamic laboratory experiments have been conducted to study the kinematics and dynamics of subduction and slab induced mantle convection. In the experiments, subduction and slab rollback were not kinematically imposed but were allowed to evolve naturally and were driven by buoyancy forces only, reflecting natural conditions. The experiments modelled a dense high-viscosity plate (subducting oceanic lithosphere) overlying a less dense low-viscosity

layer (upper mantle). The overriding lithosphere was not incorporated. Several important features of slab behaviour were investigated including the temporal variability of hinge-line migration, the kinematic behaviour of the slab and the slab induced upper mantle flow. Both a fixed and free trailing edge boundary condition of the subducting plate were investigated. Results show that hinge-line retreat is a natural consequence of subduction of a negatively buoyant slab. The migration rate increases during the free sinking of the slab until it approaches the 670 km discontinuity, resulting in a decrease in migration rate followed by a renewed increase and finally approaching a steady state. Slab retreat results in sub-lithospheric mantle flow from underneath the slab towards the mantle wedge. Experimental results indicate that all flow occurs laterally around the slab edges, forcing the slab and hinge-line to attain a convex shape towards the direction of retreat, as also observed for arc - back-arc systems in nature. No signs of flow underneath the slab tip have been detected. This might explain why long subduction zones are relatively stable or retreat slowly (e.g. South American subduction zone, Aleutian arc, Sunda arc), while narrow subduction zones retreat relatively rapidly (e.g. Scotia arc, Calabrian arc, Hellenic arc, Tonga arc. New Hebrides arc, Mariana arc). This is supported by experimental findings, which show that slab retreat is maximum for a relatively narrow slab (250-750 km wide). Further increase in slab width progressively retards slab retreat. For a fixed trailing edge, the slab does not sink vertically downward, but sinks at an angle in a regressive manner, while the surface part of the subducting plate is slightly extended. When the slab tip hits the upper-lower mantle discontinuity, it is draped backward on top of the discontinuity. For a free trailing edge, slab rollback is less prominent compared to the fixed trailing edge experiments. Slab sinking is oriented more vertically while the surface part of the subducting plate is dragged into the subduction zone. Both fixed and free trailing edge experiments indicate that the slab can transmit tensional stresses across a non-stationary hinge towards the surface part of the subducting plate (e.g. effective slab pull force), indicating that the lithosphere acts as a stress guide.

143

SGTSG Field Meeting 2003

EPISODIC OPENING OF BACK-ARC BASINS: A CASE STUDY FROM THE SOUTHWEST PACIFIC W,P, Schellart*, G.S. Lister, G. Rosenbaum and C. Duboz ACRC, School of Geosciences, Monash University, Melbourne, VIC 3800 (*wouter@mail.earth.monash.edu.au)

Back-arc basins have an extensional origin and are characterised by their limited lateral extent and relatively short time-span of back-arc activity. Several back-arc regions are defined by the sequential opening of several individual back-arc basins during distinct episodes. For example, in the Western Mediterranean, two back-arc basins can be identified, which both developed due to the eastward retreat of the Ionian slab subducting underneath the Calabrian arc. Back-arc extension started with the opening of the Liguro-Provencal basin from 30 to 16 Ma and was followed by a period of tectonic quiescence In the back-arc region. Back-arc extension resumed during opening of the Tyrrhenian basin from 10 Ma to Present. Another example of episodic back-arc activity is located in the Western Pacific, where the formation of several back-arc basins resulted from eastward rollback of the Pacific plate. Opening of the Parece-Vela basin from 31 to 15 Ma was followed by a period of tectonic quiescence in the back-arc region. Overriding plate extension resumed during opening of the Mariana Trough from 10 Ma to Present and rifting in the Izu-Bonin arc from 2 Ma to Present. One of the most striking examples of episodic opening of back-arc basins is located in the Southwest Pacific region east of Australia. Here, a sequence of five individual back-arc basins can be recognised, which can all be thought of as to have resulted from progressive eastward retreat of the Pacific slab since the Cretaceous. From west to east, these basin and adjoining bounding ridges are: the East Australian margin, the Tasman Sea, the Lord Howe Rise, the New Caledonia Basin, the Norfolk Ridge, the Norfolk Basin, the Three Kings Ridge, the South Fiji Basin, the Colville Ridge, the Havre Trough and the Kermadec-Tonga Ridge. Back-arc extension probably initiated in the New Caledonia basin and the Lord Howe Rise region in the Early Cretaceous. This resulted in extensive extension in the New Caledonia basin and possible formation of oceanic crust. In the Lord Howe region, extension was not as extensive and more diffusive. During the Late Cretaceous, the New Caledonia basin stopped opening, possibly followed by a period of tectonic quiescence in the back-arc region. A new phase of eastward rollback was accommodated by formation of the Tasman Sea,

J 44

which opened by seafloor spreading from the Late Cretaceous to Early Eocene (80-52 Ma). This was followed by a relatively long period of tectonic quiescence in the back arc region. Tectonic activity resumed with formation of the Norfolk back-arc basin from Late Oligocene to Early Miocene 2616 Ma) associated with eastward retreat of the Pacific slab. This was followed by ~ Middle Miocene opening and spreading In the Southern South Fiji Basin located to the east of the Norfolk basin, and finally extension in the Havre trough from the latest Late Miocene (7-6 Ma) to Present. The sequential opening of the back-arc basins led to the formation of several continental ribbons, which originated from the Australian continent, including the Dampier Ridge, the Lord Howe Plateau and the Norfolk Ridge. Further eastward, the ridges attain progressively more characteristic features of extinct volcanic ridges (remnant volcanic arcs), including the Three Kings Ridge, the Colville Ridge and the Kermadec Ridge. Back-arc basin formation is a direct consequence of hinge-retreat of the subducting plate. Thus, to understand the episodic nature of back-arc basins, the episodic behaviour of hingemigration needs to be investigated. We define episodic behaviour of the hinge as periods in which hinge-retreat, hinge-advance and hinge-stability are . alternated. To investigate the episodic nature of the hinge of the subducting plate, fluid dynamical laboratory models have been built to simulate subduction and slab rollback. The models consist of a dense high viscous plate (e.g. oceanic lithosphere) subducting into a less dense low viscosity layer (e.g. upper mantle). Results indicate that the episodic nature of the hinge is mainly the result of interaction of the slab with the upper-lower mantle discontinuity. The modelling results indicate that the episodic behaviour of the hinge is dramatically amplified if the subducting plate has a relatively high applied absolute horizontal velocity at its trailing edge (e.g. ridge push). This episodic behaviour of the hinge could explain the sequential episodic opening of back-arc basins in the Southwestern Pacific, because the absolute velocity of the subducting Pacific plate has been relatively high (> 5 cm/yr) since the Late Cretaceous.

SGTSG Field Meeting 2003

POSTERS WEDNESDAY 24 SEPTEMBER 2003

145

SGTSG Field Meeting 2003

146

SGTSG Field Meeting 2003

FRACTURE SYSTEMS IN GRANITE PAVEMENTS OF THE EASTERN PILBARA CRATON, WESTERN AUSTRALIA: INDICATORS OF NEOTECTONIC ACTIVITY? Simon Bodorkos^ and Daniel Clark^ ^School of Earth Sciences, University of Melbourne VIC 3010 (bodorkos@unimelb.edu.au) ^Earthquake Hazard & Neotectonics Group, Geoscience Australia. GPO Box 378, Canberra ACT 2601

Records kept over the last 100 years indicate that continental Australia is seismically active in comparison with similar intraplate regions elsewhere in the world. The locus of deformation associated with this activity is defined by the distribution of earthquake hypocentres, which primarily occur along "weak" structures that are favourably oriented with respect to the prevailing stress field. If the movement history of such structures can be inferred from their geomorphic expression, they have the potential to constrain the orientation of the in situ stress field. However, the link between earthquake events and their impact on the Australian landscape is poorly understood, especially with respect to smaller events. In this context, the Pilbara Craton (which has experienced seven A/f > 5 earthquakes since 1960, and numerous smaller events) is ideal for studying the geomorphological consequences of such events, as earthquake hypocentres are consistently shallow and good bedrock exposure means that rapid obscuring of seismic geomorphology by weathering is unlikely. GSWA mapping has recognised several natural rock bursts i.e. fracture systems up to 100 m long and 10 m wide that expose fresh rock in granite pavements and offset pavement surfaces by up to 15 cm. We examined rock bursts at four key localities, in order to establish whether the fractures are tectonic in origin, and whether the fractures are related to larger fault structures or recorded seismicity. We observed: (1) two elongate, arcuate fracture systems characterised by vertical offsets of 10-15 cm in pavement surfaces, and the occurrence of granite blocks up to 1 m long that appear to have been ejected up to 3 m perpendicular to strike. At Gallery Hill, a 35 x 5 m fracture zone trends '-315765"NE, and causes reverse offset of the pavement. At North Shaw, fractures in an 80 x 15 m rock burst dominantly trend -030790°, with normal offset indicated by 50 cm-wide grabens in the centre of the fracture zone. Subordinate fractures trending '-'120780°N show reverse offset. (2) Simple vertical fractures

up to 100 m long with apertures up to 2 cm and negligible pavement offset, striking 045-060° at Gallery Hill and 025-030° at Mulgandinnah Hill. (3) Violent disintegration of large exfoliation sheets up to 30 cm thick on gently sloping pavements. Pervasive sheet shattering at Muccan produced dozens of ^50 cm blocks in drifts atop the bedrock. At Mulgandinnah Hill, sheet fragments 5-10 m in diameter are translocated downslope but some smaller blocks are overturned, and the axial strikes of three "Atent" structures are highly oblique to the slope direction. The relatively small size and lateral discontinuity of the fracture systems suggests that they do not form part of larger fault structures, and a tectonic origin cannot be proved for all of these features. Nevertheless, a relationship with recent seismicity cannot be ruled out, and plausible alternative mechanisms for fracture formation (e.g. gravity-driven slope failure, insolation weathering) are largely unsatisfactory. In particular, two key observations support a tectonic origin: (1) Fracture surfaces invariably appear fresher than patinated Aboriginal petroglyphs on pavement surfaces, and fractures directly cross-cut at least two of the three petroglyph generations on the Muccan pavement. It is thus likely that that the rock bursts were formed rapidly and recently. (2) The inferred horizontal compression and extension orientations are compatible with the formation of all of the observed structures within a single, NESW compressive tectonic stress field. This is consistent with regional in situ stress patterns (NE-SW and E-W compression in the nearby Canning and Carnarvon Basins respectively). Elsewhere in Australia, sheet shatter and fracture events have been linked to specific earth tremors, with the energy imparted resulting in failure of rock platforms already subject to the regional in situ stress field. In the seismically active Pilbara, it is conceivable that widely scattered rock bursts might be related to the spatial distribution of earth tremor events.

147

SGTSG Field Meeting 2003

THE POTENTIAL FOR PALEOLIQUEFACTION STUDIES TO CONTRIBUTE TO AUSTRALIA'S EARTHQUAKE HAZARD MAP Phil Cummins\ Dan Clark\ Clive Collins\ Martitia Turtle^ and Roy Van Arsdale^ ^Geoscience Australia. GPO Box 378, Canberra ACT, 2601, dan.clark@ga.gov.au ^Turtle & Associates, HC 33 Box 48, Bay Point Road,Georgetown, ME 04548 ^Center for Earthquake Research & Information, University of Memphis. 3890 Central Ave. Memphis, TN 38152.

Australia has a low rate of modern seismicity compared to plate margin regions, and a short historical record of earthquakes. These factors combined manifest as significant uncertainty in the Australian earthquake hazard map. The current hazard map is based exclusively on historical and instrumentally recorded seismicity. It is an open question as to whether this is a valid characterisation of seismicity over the long term. For example, recent neotectonic investigations in areas of low topographic relief indicate that seismicity must be transitory in both space and time over large parts of Australia. One approach to address the problem of sparse historic seismicity in intraplate regions like Australia is to use paleoliquefaction studies to constrain the occurrence of prehistoric earthquakes. Liquefaction has been observed following large historic earthquakes in South Australia and Victoria, and numerous 'sand blows' were observed following the 1968 Meckering earthquake (Ms 6.8) in Western Australia. It therefore seems likely that prehistoric earthquakes in these areas would

148

have also induced liquefaction. Liquefaction deposits might also be anticipated in other areas which are geologically prone to liquefaction, but have not experienced an historical earthquake. We have identified 4 target areas for paleoliquefaction studies: (1) the Perth region, a major urban centre situated in a large sedimentary basin near the South West Seismic Zone; (2) the Murchison river area about 500 km north of Perth, which lies near the epicentres of two large historical earthquakes; (3) southeastern South Australia, site of pronounced liquefaction associated with a large (Ms 6.5) 1897 earthquake; and (4) the Goulburn river near the Cadell fault in Victoria, whose banks consist of poorly consolidated fluvial sediments and lie near a fault known to have experienced slip in the Quaternary. In this presentation, we will discuss why we have chosen these sites as having high potential for paleoliquefaction studies, and hope to present some results from preliminary surveys at of the four target areas.

SGTSG Field Meeting 2003

MULTIPLE CHRONOSTRATIGRAPHIC UNITS IN THE STRANGWAYS METAMORPHIC COMPLEX, CENTRAL AUSTRALIA Matthew M. Cobb\ Peter D. Kinny^, Martin Hand^ 1. Teconics Special Research Centre, Department of Applied Geology, Curtin University of Technology, Bentley, Western Australia 6102. 2. Continental Evolution Research Group, Geosciences, University of Adelaide, Adelaide 5005. m.cobb@curtin.edu.au. pk@lithos.curtin.edu.au. martin.hand@adelaide.edu.au The Yambah and Erontonga granulite units within the Strangways Metamorphic Complex (SMC), Central Australia differ from the rest of the complex, with a predominance of granulite-grade psammitic and semipelitic metasediments. This differs from the remaining units, which are characterised by a dominance of either calcsilicates or granulite-grade felsic orthogneisses. This difference is observed clearly in radiometric data coverage of the Strangways complex, in which both the Yambah and Erontonga units demonstrate a much higher concentration of radioactive components compared to the remaining parts of the complex. These observations have led to the proposal that the SMC may not have evolved as a single tectonometamorphic package as assumed by other recent authors, and instead may be closer to initial theories which consider the complex to comprise several stratigraphic packages of differing age, geochemical character and tectonic provenance. While these earlier stratigraphic and tectonic divisions have been demonstrated to no longer be correct, the concept of multiple stratigraphic divisions of differing ages is still possible.

granulite unit and the adjacent Mt Fitzner region. Samples taken from the Mt Fitzner region show a clear protolith signature in SHRIMP results, at an age of 1806 ± 2 Ma. This is an age not recognised in either the samples analysed from the Yambah or Erontonga granulites. We will test the "multiple tectonostratigraphic unit" hypothesis, through the use of comprehensive geochemistry and detailed description of the metamorphic development of the Yambah granulites. Nd isotope measurements, already established as a viable means of geochemically qualifying the nature of parent sources to lower grade metasediments, will be used in demonstrating differences in protolith chemistries for the two units. Metamorphic analysis of the Yambah and Erontonga politic granulites will then be compared to previously published metamorphic studies of the SMC to determine how closely the tectonothermal development of the former resembles that of the remaining SMC. Combined with SHRIMP geochronology, this will enable a more accurate assessment of the tectonic activity undergone in the Paleoproterozoic assembly of the ProtoAustralian continent.

Current results from ongoing geochronology in the SMC have provided evidence of a chronologic distinction between the Yambah

149

SGTSG Field Meeting 2003

STRUCTURE AND AGE OF THE NORTHERN LEEUWIN COMPLEX, WESTERN AUSTRALIA: CONSTRAINTS FROM FIELD MAPPING AND U-TH-PB ISOTOPIC AND CHEMICAL ANALYSIS Alan S. Collins ^ Tectonics Special Research Centre, Department of Applied Geology, Curtin University, GPO Box U1987, Perth, Australia. a.s.collins@curtin.edu.au. Fax: +61-8-9266-3153

Rocks in the northern Leeuwin Complex of far southwestern Australia preserve rocks that formed during the break up of Rodinia and that provide evidence for the subsequent amalgamation of Gondwana. Detailed field mapping, structural investigation and U-Pb isotopic zircon analysis using the Sensitive Highmass Resolution Ion Microprobe (SHRIMP) has revealed that: 1) protoliths of pink granite gneiss and grey granodiorite gneiss intruded at --750 Ma, coeval with break up of western Rodinia. 2) Granulite/upper amphibolite facies metamorphism occurred at 522 ± 5 Ma, in the Early Cambrian, ---lOO Ma later than previous

150

estimates and of identical age as estimates of the final amalgamation of Gondwana. 3) Three major phases of ductile deformation occurred during or after this metamorphism and represent a progressive strain evolution from steeply inclined (with respect to the present Earth surface) shortening (D1) to sub-horizontal E-W (D2) then NNW-SSE (D3) contraction. The D3 shortening vector is consistent with sinistral movement on the Darling Fault Zone and suggests that movement on this structure may be late in the evolution of the Pinjarra Orogen.

SGTSG Field Meeting 2003

THE NACKARA ARC OF SOUTH AUSTRALIA, A FAULTFOLD SYSTEM OF SIGNIFICANT ECONOMIC POTENTIAL Colin H.H. Conor Geological Survey Branch, Minerals Petroleum and Energy. Primary Industries and Resources SA

The Nackara Arc is a regional structural apex containing a number of potentially economic commodities, which include gold, copper and diamonds. Carbonatite, kimberlite, diamond indicators and micro-diamonds continue to attract exploration, and the diamond-bearing Springfield Basin falls with the axis of the Arc. Copper and gold mineralisation is associated with iron oxide and sericite alteration. Felsic and intermediate volcanics and subvolcanic units have recently been discovered and may be genetically related to gold mineralisation. The Nackara Arc is located at a bifurcation of the Neoproterozoic Adelaide Geosyncline, which forms a critical triple point separating the ArchaeanMesoproterozoic Gawler Craton, PalaeoproterozoicMesoproterozoic Curnamona Province and Cambrian basement below the Murray Basin. The Adelaide Geosyncline is a rift filled by spatially restricted rift-phase (Callanna Group) and extensive sag-phase sediments (Burra, Umberatana and Wilpena Groups). The latter three groups crop out in the Nackara Arc, and there is potential for the Callanna Group to be present at depth. The Burra Group is dominated by siltstone, but with carbonate and quartzitic interbeds. The Umberatana Group is also siltstone dominated but characterised by tillites and the Wilpena Group comprises siltstone, carbonate, and quartzite. Salt or mud diapirs were extensively sourced from the Callanna Group in the northern Adelaide Geosyncline. Inversion of the Adelaide Geosyncline occurred during the Cambro-Ordovician Delamerian Orogeny with very long folds developed parallel to the rift axis. The axes of these folds curve around the Nackara Arc. Also trending parallel to the original rift and hence the structural axis are major faults, which in the south have been shown to thrust the Adelaidean cover westward on to the Gawler Craton. A similar situation exists in the Nackara Arc with strike-parallel discontinuities separating tectonic slices. Lateral and vertical displacement between slices is interpreted by a combination of flexural slip and thrusting. The Nackara Arc also forms a structural divide with dome and basin style deformation dominating to the north. To the southeast the innermost part of the Nackara Arc exposes the belt of intrusives that core the Delamerian Orogen; these extend southward to Kangaroo Island and eastward to the southeastern corner of the Curnamona Province.

The Nackara Arc is the northern flexure of a sigmoidal structure that deforms the Adelaide Geosyncline. The Nackara Arc is itself a northwesterly-trending fold of regional scale, which refolds the earlier above mentioned Delamerianaged long folds. Smaller interference structures are economically important and form anomalous Dshaped anticlines that are cored by zones of disruption. These zones have been referred to in the past as diapirs, although they possibly represent sites of tectonic or hydraulic brecciation. Two prominent examples are the Mount Grainger and Paratoo 'diapirs', both are intensely potassic alterated (feldspar and sericite) and gold mineralised. Early regional mapping showed that a major thrust broke across trend to enter the Mount Grainger 'Diapir'. Mapping also recorded mafic and intermediate rocks in the cores of the Mount Grainger and Paratoo structures. Gold mineralisation has been related both to these igneous rocks and to structurally focussed potassic alteration (sericite). Range River Gold are currently investigating Cu-Au mineralisation associated with newly discovered rhyolitic welded tuffs and porphyries, and alkaline and mafic intrusives in the vicinity of Oodlawirra. 'Diapiric' breccias and haematite bodies are common along the northwestern axial trend of the Nackara Arc fold and locally host copper mineralisation. Also of significance is the Walloway carbonatite and the diamond-bearing Springfield Basin. From an economic viewpoint the Nackara Arc may represent a 'Christmas tree' structure of regional scale; the northwesterly trending fold axis, representing the trunk, and strike parallel faults, representing branches, are the loci for iron oxidepotassic alteration and gold and copper mineralisation. Lamprophyres were intruded late during the Delamerian Orogeny and kimberlitic intrusives are known to have been introduced along northwesterly trending faults and strike parallel thrusts in the Jurassic. The timing of alteration and mineralisation is not known, but is an element of current investigations.

151

SGTSG Field Meeting 2003

DUCTILE TO BRITTLE SHEAR ZONES IN THE WYANGALA FAULT, COWRA, NEW SOUTH WALES Czarnota,

a n d L e n n o x , P.G.^

^Geoscience Australia, GPO Box 378, Canberra, ACT 2601. ^School of BEES, UNSW, Sydney 2052.

Past researchers have advocated a simple dipslip west-over-east movement scenario for the Wyangala Fault. This study clearly demonstrates that the movement is complex and implies that fault movement in the Eastern Lachlan Fold Belt is more complicated than previously though. The over 30km long, arcuate Wyangala Fault consists of at least two sub-parallel, north to northwest-trending shear zones within the Early (?) Silurian Wyangala Granite and Ordovician to Early Silurian metasediments and volcaniclastics on the western side of the Molong High. The Wyangala Granite is a megacrystic to porphyritic K-feldspar, biotite granite of S-type character. It contains a heterogeneously developed meridional foliation (S) which is cross-cut by a more moderately west dipping C-type shear band foliation and extensive mylonites. One north-northeast and six meridional trending shear zones typically 200m wide with a spacing of 200m to 1km have been defined in the Wyangala Granite or along granite-country rock boundaries in a 77km^ area around Wyangala Dam. The shear zones were delineated on the basis of grain size, the presence of an S-C fabric and the occurrence of mylonites. These shear zones have been subdivided into three low strain shear zones to the west and four high strain shear zones to the east of the eastern Wyangala Granite-country rock contact. This pattern of higher strains on the eastern margin of granites is consistent with the regionally observed dominance of west-over-east reverse faults on the eastern margin of some plutons within the Wyangala Batholith. The movement history along the shear zones is complex. Microstructural analysis indicates that all the shear zones in the area have been subject to west-over-east movement. The slip direction determined by S-C surface intersections

152

indicates there is a progression of movement sense from west-over-east along the low strain shear zones in the west to almost pure sinistral strike-slip movement along the high strain shear zones in the east. Ductile and brittle shear sense indicators clearly indicate sinistral strike-slip movement occurred along fault planes within the Wyangala Granite and within metasediments and volcanics 10km north of the study area. Hence it is considered that sinistral strike-slip motion occurred during the transition from ductile to brittle conditions. The localization and development of the shear zones is ascribed to rheological contrasts and steeply dipping meridional mineralogical layering within the Wyangala Granite and geochemical analyses indicate that areas of increased shearing correspond to zones which have higher silica content. The low strain shear zones were developed under low to medium grade metamorphic conditions and the high strain shear zones were developed under slightly highergrade metamorphic conditions. Ar-Ar dating of biotite from within the Wyangala Granite near the spillway gives cooling ages consistent with regional foliation development and possible west-over-east movement in the granites being late Middle Devonian (i.e. Tabberabberan). It is considered that sinistral strike-slip movement occurred either during the Tabberabberan or as part of the Early Carboniferous Kanimblan event. However since the change from west-over-east to strike-slip movement occurred during the transition from a ductile to brittle environment the former period is favored and thereby suggests that the orientation of the stress field rotated during the Tabberabberan event.

SGTSG Field Meeting 2003

A NEW ACTIVITY-COMPOSITION MODEL FOR AMPHIBOLES IN NCKFMASHO: APPLICATION TO GREENSCHIST FACIES METAMORPHISM IN ARCHAEAN GREENSTONES Jonathon Dale\ Roger Powell\ Tim Holland^ and Fiona Elmer^ ^School of Earth Sciences, University of Melbourne, Victoria 3010, Australia, jdale@unimelb.edu.au ^Department of Earth Sciences, University of Cambridge, Downing Street, Cambridge, CB2 3EQ, UK

Quantitative mineral equilibria modelling in metabasic rocks has long been compromised by poorly understood activity-composition (a-)Q relationships in amphiboles. We present a powerful new a-X model for clino-amphiboles in the system Na20-Ca0-K20-Fe0-Mg0-Al203Si02-H20-02 (using the end-members tremolite, tschermakite, pargasite, glaucophane, ferroactinolite, K-pargasite and ferri-tschermakite), based on two large natural-assemblage datasets collected from the literature. The first dataset consists of 73 pairs of coexisting amphiboles (actinolite-hornblende, winchite-hornblende, and barroisite-glaucophane) equilibrated over the range 250-580°C, 2-18 kbar. The second consists of 831 amphibole-bearing assemblages equilibrated over the range 250-950°C, 2-26 kbar. The model incorporates three key methodological advances. Firstly, we consider the possibility of asymmetric solvi in multicomponent systems by using the recently reformulated van Laar model. This permits the complete description of non-ideal mixing between the seven amphibole end-members in terms of 21 interaction energies and 7 asymmetry parameters. Secondly, the model is calibrated by simultaneously regressing data from several different mineral assemblages, thereby greatly increasing the quality of control available in comparison with calibrations based on a single assemblage (e.g. garnet-hornblende-plagioclasequartz). Thirdly it utilises an appropriately robust least median of squares (LMS) approach to the

non-linear regression of noisy data typical of natural assemblages, minimising the damaging effect of outliers on regression results. The ubiquity of amphiboles in metabasic rocks ensures widespread applicability of the model. Here, we demonstrate its versatility in two applications involving Archaean greenschist facies metabasic rocks. (1) The calcic-calcic amphibole solvus has long been the source of much debate as to its nature and even its existence. Firstly, we predict the existence of a calcic-calcic amphibole solvus below 600°C between actinolite and pargasitic-hornblende amphiboles. The model amphibole compositions are consistent with literature observations, but also provide insights into why coexisting amphiboles are rarely reported in greenstone compositions at low- to medium-grades. (2) The greenschist to amphibole facies transition in metabasic rocks has previously proven to be a difficult area for mineral equilibria calculations and investigations (in NCKFMASHO). In particular, investigation into the interactions between amphibole and epidote has been prohibited by the inability of the amphibole model to realistically incorporate ferric iron. We present P-T and r-X(C02) pseudosections across the greenschist to amphibole facies transition for greenstone bulk compositions. These predict mineral assemblages and compositions that are consistent with observations in amphibole and epidote bearing greenstones.

153

SGTSG Field Meeting 2003

GOLD-BEARING H2O-CO2 FLUIDS: INSIGHTS FROM MINERAL EQUILIBRIA MODELLING IN ARCHAEAN GREENSTONES. Fiona L, Elmer\ Jonathon Dale^ and Roger Powelf ''School of Earth Sciences, University of Melbourne, Victoria 3010, Australia, idale@unimelb.edu.au

In postulating a genetic model for Archaean greenstone gold deposits, various possible sources of fluid for ore formation have been proposed. Many of these 'gold-only' deposits are surrounded by large alteration patterns, eg. Bronzwing, W.A. and Kalgoorlie, as a consequence of infiltrating CO2-H2O fluid. The aim of this study is to combine petrographic observations and mineral equilibria calculations, in order to make useful predictions about the composition and volume of fluids formed during the metamorphism of metabasic rocks at depth. We suggest that fluids produced during metamorphism provide a viable source of fluid for the formation of greenstone gold deposits. Metabasic rocks (including greenstones) show only small changes in mineral assemblage over greenschist-amphibolite facies pressures and temperatures, compared to mineral compositions and mineral modes, which may vary greatly. The assemblage amphibole + chlorite + epidote + plagioclase + quartz + carbonate ± Fe-Ti oxides ± K-mica ± garnet is common under these conditions, the devolatisation of this assemblage produces large volumes of CO2-H2O dominated fluids. This fluid, produced at depth will ultimately escape via structural pathways to infiltrate and consequently alter the shallower metabasic sequences above and is therefore a likely mechanism of gold transportation.

154

To understand the nature of the fluids produced, a precise understanding of phase equilibria in metabasic rocks is required. Thermodynamic modelling using the computer program THERMOCALC in the system Na20Ca0-Fe0-Mg0-Al203-Si02-C02-H20, allows the construction of P-T pseudosections and TX(C02) pseudosections. These pseudosections can be used to understand the evolution of metabasic rocks during greenschist-amphibolite facies metamorphism and provide a framework for the calculation of internal buffering paths. The evolution of rocks along these paths constrains the volume and composition of fluids produced by mineral reactions, demonstrating that fluid composition is extremely sensitive to changes in mineral compositions and modes. Calculated mineral assemblages and mineral compositions are similar to what are seen in Archaean greenstones. In these calculations the predicted fluids produced within metabasic rocks at depth during greenschist-amphibolite facies metamorphism have compositions and volumes favourable for gold transportation. Upon migration and infiltration of the overlying greenstone pile these fluids would be capable of producing the large alteration sequences we see associated with Archaean greenstone gold deposits today.

SGTSG Field Meeting 2003

AN INTEGRATED PROCESS MODEL OF THE MT ISA COPPER DEPOSITS, QUEENSLAND K l a u s G e s s n e r \ P e t e r A . J o n e s ^ A n d y R. W i l d e ^ P e t e r A l t - E p p i n g \ M e l i s s a G r e g o r y ^ G o r d o n W . H. G e r m a n \ P a u l G o w ^ R. D u g ! Wilson"^ ^ pmd*CRC, Computational Geoscience, CSIRO Exploration and Mining, PO Box 1130, Bentley WA 6102, Australia ^ pmd*CRC, Economic Geology Research Unit, School of Earth Sciences, James Cook University, Townsville 4811, Australia ^pmcTCRC, School of Geosciences, Monash University, Clayton, VIC 3800, Australia ^ Mount Isa Mines Exploration Pty Ltd, 102 Oban Road, Mount Isa, QLD 4825, Australia

The Mt Isa Cu and Pb-Zn deposits represent one of the largest accumulations of base-metals on the planet. A large body of research has accumulated in the eighty years since production commenced. Nevertheless, there are many aspects of ore deposition that remain controversial or uncertain. The nature of the hydrothermal fluids, the fluid flow system, the role of adjacent uranium-enriched granitoid batholiths and even the absolute age of metal deposition remain poorly understood. We present the integrated approach taken in an ongoing research project of the Predictive Mineral Discovery Cooperative Research Centre (pmcTCRC) to develop a truly predictive exploration model for the Mt Isa Copper deposit type. Key elements in this investigation are coupled and stand-alone numerical models, which are able to simulate mechanical, fluid flow, thermal and geochemical process models. These simulations are used to test conceptual models and scenarios of ore formation. In this contribution we focus on the integration of 3D structural geometries with numerical models to simulate deformation, fluid flow, and heat flow in two and three dimensions. 3D STRUCTURE A major challenge for the project has been to synthesise available data from Mount Isa Mines' Exploration and Mining branches and to generate a 3D geometric model of the metal grade shells, host-rock architecture and structures. The software used for this purpose is GOCAD. MECHANICAL, THERMAL & FLUID FLOW MODELLING A key part of the project is building and running partly-coupled numerical process models which lead to a quantitative understanding as to how

the mechanical and thermal processes interact with flow patterns of crustal and basinal fluids. The software used for this is the finite difference modelling package FLAC (2D and 3D versions). Two-dimensional simulations are typically used as a tool to explore mechanical, thermal and fluid-flow coupled models based on simplified geometries of actual cross-sections or generic geometries that represent critical parts of the conceptual model of the deposit. 3D GOCAD models have been used as direct input to FLAC. This is made possible by transferring the regions and surfaces of the 3D GOCAD model into FLAC3D through a purpose-built, XML-based automated mesh generation software (3D Mesh Analysis Coupling Software, 3DMACS) developed by CSIRO Exploration and Mining's Computational Geoscience Group. The FLACmodels can then be tested for sensitivity to a variety of parameter combinations, such as variable velocity boundary conditions, activity of structural elements or fluid source regions. More specifically, the 3D FLAC models are used to test scenarios, which are constrained by newly acquired geochemical data as well as by numerical models of phase speciation and reactive transport. CONCLUSIONS The integrated numerical modelling approach has already generated significant new insights into the Mt Isa mineralising system. The ability to visualize complex 3D data has proven to be a major benefit to the project, particularly in communicating between individual team members. Significant barriers to be overcome include the limitations on model complexity imposed by the present generation of computers and the large range of parameters that need to be explored.

155

SGTSG Field Meeting 2003

MODES OF CRUSTAL EXTENSION DETERMINED BY RHEOLOGICAL LAYERING C h r i s Wijns^'^, K l a u s G e s s n e r ^ , R o b e r t o W e i n b e r g ^ , L o u i s M o r e s i ^ 1 Centre for Global Metallogeny, University of Western Australia, Crawley, WA 6009, Australia 2 CSIRO Exploration and Mining, PO Box 1130, Bentley, WA 6102, Australia 3 School of Geosciences, Monash University, VIC 3800, Australia

Metamorphic core complexes form during continental extension when stretching is strongly localised onto relatively few normal faults. Each fault accomodates large displacements, eventually dissecting the upper crust and resulting in exhumation of the lower crust. In our numerical models, the extending crust is mechanically stratified through the temperature gradient, which dictates the transition from a strong, brittle, upper crust to a weaker, ductile, lower crust. We use a twodimensional Lagrangian Integration Point finite element code which allows simulations to develop very large strains comparable to those found in analogue models, while still tracking strain history accurately for the constitutive laws. The viscoplastic formulation, with strain and/or strainrate weakening behaviour, gives rise to the localisation of deformation and necking of layers. Results show two distinct extension modes which depend upon vertical rheological contrasts: the distributed faulting mode and the metamorphic core complex mode. In the absence of preexisting weaknesses, the ratio of the integrated

156

strength of the upper to lower crust is an indicator of the resulting mode of extension. When the strength of the lower crust approaches that of the upper crust (small ratio), the result is distributed, densely spaced faulting, with limited slip on each fault, and no exposure of lower crustal rocks. An example could be faulting in the North Sea. Metamorphic core complexes develop when the lower crust is much weaker (large strength ratio) and flows easily. Significant block rotation, and consequent large displacements along faults, leads to the complete dissection of the upper crust, as may be the case for core complexes of theWestern U.S.A. and the Aegean. Numerical temperature-time paths for our metamorphic core complex models are comparable to field data from apatite fission-track thermochronology for the Central Menderes in western Turkey. The actual critical strength ratio for the transition between modes will depend upon such factors as the relative thickness of the lower crust with respect to the upper crust, and the degree of fault weakening.

SGTSG Field Meeting 2003

KINEMATICS OF SYN-INTRUSIVE EXTENSION ASSOCIATED WITH THE SYBELLA BATHOLITH R. J, Gordon & R. J. Holcombe Department of Earth Sciences, University of Queensland, Brisbane, Queensland, Australia 4072

Several studies in the Mount Isa Inlier have suggested "-1660 Ma Sybella Batholith intruded synchronously with a basin-forming extensional event. This study presents evidence of a generation of strong syntectonic fabrics within and around the Sybella Batholith that supports this interpretation. The syntectonic fabrics and extensional structures are overprinted by the multiple phases of contractional deformation of the Isan Orogeny. Analysis of the Isan deformation, including progressive reconstructions of cross-sections across the Sybella Batholith, reveals the pre-lsan geometry of the Sybella Batholith and the surrounding basins (sub-basins of the --1800 Ma Leichhardt Superbasin and --1660 Ma Isa Superbasin). Isan contraction across the system varies north to south with maximum contraction in the south of -50%. The pre-lsan batholith geometry consists of a lower stratigraphy-parallel body {--6 km thick)

with strong L-tectonite fabrics at the margins and magmatic textures at the core. The country rocks immediately surrounding the sill share the syntectonic Fi fabrics. A second, largely undeformed, sill-like body --3 km thick is sited within brittle rocks km higher in the sequence but offset to the north. A third, smaller, more bulbous microgranite -2km thick lies immediately above the eastern edge of the upper sill. The Fi brittle-ductile transition is inferred to lie between the lowermost sill and the upper bodies. Models based on strain and kinematic analyses of the unfolded granite L-tectonites imply transtensional Fi deformation with east to northeast-directed opening. Transtensional Fi deformation is most consistent with previously proposed extensional models of basin development rather than the alternative strike-slip model.

157

SGTSG Field Meeting 2003

END-MEMBER BOUDIN CLASSIFICATION AND MODIFIED BOUDIN STRUCTURES. 1

Ben Goscombe , Cees Passchier , Martin Hand , David Gray School of Earth and Environmental Sciences, Adelaide University, South Australia, 5005, Australia. School of Earth Sciences, University of Mainz, Germany. ^School of Earth Sciences, University of Melbourne, Parkville, 3010, Victoria.

In monoclinic shear zones, there are only three ways a layer can be boudinaged, leading to three kinematic classes of boudinage. These are (1) symmetrically without slip on the inter-boudin surface (no-slip boudinage), and two classes with asymmetrical slip on the inter-boudin surface: slip being either (2) synthetic (S-slip boudinage) or (3) antithetic (A-slip boudinage) with respect to bulk shear sense. In S-slip boudinage, the boudins rotate antithetically, and in antithetic slip boudinage they rotate synthetically with respect to shear sense. We have investigated the geometry of 2100 natural boudins from a wide variety of geological contexts worldwide. Five end-member boudin block geometries that are easily distinguished in the field encompass the entire range of natural boudins. These five end-member boudin block geometries are characterized and named drawn, torn, domino, gash and shearband boudins. Groups of these are shown to operate almost exclusively by only one kinematic class; drawn and torn boudins extend by no-slip, domino and gash boudins form by A-slip and shearband boudins develop by S-slip boudinage. In addition to boudin block geometry, full classification must also consider boudin train obliquity with respect to the fabric attractor

158

Boudin trains lying at a low angle (<10'') to the fabric attractor are classified as foliation-parallel boudin trains and those exceeding 10°, are classified as foliation-oblique boudin trains. All foliation-oblique boudin trains operate by S-slip boudinage, regardless of end-member boudin block geometry. Material layeredness of the boudinaged rock must also be considered and has been classified into object boudinage, singlelayer boudinage, multiple-layer boudinage and foliation boudinage of a foliated rock devoid of, or irrespective of, layers of differing competence. Analysis of modified (or complex) boudin structures recognizes two types: (1) Sequential boudin structures experienced continued extension (i.e. progressive congruent structures) and show sequential development of two or more different end-member boudin block geometry components during progressive boudinage. (2) Reworked boudins were modified by subsequent deformational episodes and are polyphase noncongruent structures (folded, sheared and shortened types). Correct classification of boudins and recognition of their modification is the crucial first stage of interpretation of natural boudin structures, necessary to employing them as indicators of shear sense, flow regime and extension axes.

SGTSG Field Meeting 2003

ELECTRON MICROPROBE DATING OF MONAZITES FROM THE WESTERN GAWLER CRATON Oliver Holm\ Ron Beriy, Patrick Lyons\ and David Steele^ ^Geoscience Australia, GPO Box 378, Canberra, ACT 2601 ^School of Earth Sciences, University of Tasmania, GPO Box 252-79, Hobart, TAS 7001 ^Central Science Laboratory, University of Tasmania, GPO Box 252-74, Hobart, TAS 7001

The Archaean Mulgathing Complex, of the western Gawler Craton, Is largely unexposed, with less than one percent outcrop. Despite this, it is considered to have excellent mineral potential, particularly for Archaean lode gold, albeit metamorphosed. Understanding the tectonic history is crucial for future exploration success in this region. However, there have been few constraints to assist in our understanding of the tectonothermal evolution of the Mulgathing Complex. Previous geochronological studies (U-Pb SHRIMP dating) show that it underwent granulite facies metamorphism during the Sleafordian Orogeny, at about 2.44 Ga. Chemical dating, via electron microprobe analysis, is now established as a fast and reliable technique. We derived chemical ages for

monazites of the Mulgathing Complex. Results for samples taken from units with known SHRIMP ages (U-Pb zircon) show that the method is accurate and reproduced the latter ages within expected errors and allowance for the lower closure temperature of monazite. For example, our chemical age for the Christie Gneiss is 2416±12Ma (N=78) and the SHRIMP age is 2437±11 Ma. These data contain statistically valid age populations at about 2.2 Ga and 2.5-2.6 Ga, though their geological significance is yet to be determined. We also show that some gneisses previously thought to belong to the Sleafordian Orogeny gave ages suggesting they probably formed during the Kimban Orogeny, ca 1.72 Ga; and sheared biotite granite in the northwest of the Complex recorded an age of 2.43 Ga.

159

SGTSG Field Meeting 2003

A KINEMATIC HISTORY OF THE SELWYN FAULT, MORNINGTON PENINSULA, VICTORIA, AUSTRALIA. J a n s s e n , D . P . \ G l e a d o w , A . ^ W i l s o n , C J . L . ^ S a n d i f o r d , M.^ tectonics S.R.C., Department of Applied Geology, Curtin University of Technology, GPO Box U1987 Perth WA 6845. Email: d.ianssen@curtin.edu.au ^School of Earth Sciences, The University of Melbourne, Victoria 3010. Email: gleadow@unimelb.edu.au; cjlw@unimelb.edu.au; mikes@unimelb.edu.au

INTRODUCTION The Selwyn Fault on the Mornington Peninsula, Victoria, has a complex kinematic history, with episodes of strike-slip, normal and reverse movement. The deformation tends to be brittle to brittle-ductile in nature, and is mostly preserved within the Devonian granite outcrops on the Mt Martha coastline. The kinematics and timing are constrained from tectonic, stratigraphic, and palaeostress indicators in the field, such as crosscutting relationships, offsetting and deformation of stratigraphic units, striations, and pseudotachylite. The Mornington Peninsula is one of a series of broadly northeast-southwest elongated uplifts across southern Victoria, and is the only one consisting predominantly of Palaeozoic rocks. The others comprise mainly Early Cretaceous rocks, including the Strzelecki Ranges of the South Gippsland, and the Otway Ranges. Inversion structures in the Tertiary sediments of the Gippsland Basin in Bass Strait suggest a similar structural history may be responsible for all the structural highs in southern Victoria. Kinematic History From evidence preserved within the Devonian granites along the Port Phillip coastline, the evolution of the Selwyn Fault can be separated into three different stress regimes over time. In the Pre-Eocene, a NNE-SSW compressional field produced strike-slip movement that affected the peninsula, with sub-horizontal striations forming on conjugate fault sets within the Devonian granite. These structures may have originated from tectonic activity related to the initial formation of the Bass Strait basins in the Latest Jurassic-Earliest Cretaceous.

160

From Eocene to Latest Miocene-Early Pliocene a NW-SE extensional stress regime was in place, with NE trending normal faulting occurring within the Selwyn Fault system. The timing for this regime is constrained by a basalt dyke intruded with a NE trend, and steep westerly dip. Similar dykes at Flinders and Cape Schanck have been previously dated using K-Ar, and gave an Eocene age. Tertiary stratigraphy at Balcombe Bay, Mt Martha, allows some constraint for later movements on the NE trending normal faults. A thickening of strata across the fault is indicative of normal movements at the time of deposition. From Latest Miocene to Present Day, a WNWESE compressional field has been producing a reverse/thrust regime. This stress field has forced the inversion of many pre-existing NE trending structures within the Selwyn Fault system, and inverted similarly orientated features throughout southern Victoria. This inversion caused high erosion rates and buckled the stratigraphy, forming possible petroleum traps, although it is likely that they would have missed the maturation window. Lithological descriptions from boreholes on either side of the fault, and the heights of tectonically raised wave-cut platforms since the last interglacial have allowed an estimate of the rate of movement on the Selwyn Fault in the current stress regime. The two estimates closely agree, and suggest an uplift rate of 40-50 metres per million years. This rich neotectonic record is important in reconstructing the recent history of southeastern Australia, and also, in understanding the potential for seismic risk associated with such structures - especially within urban areas!

SGTSG Field Meeting 2003

TECTONIC CYCLES IN THE STRANGWAYS METAMORPhIC COMPLEX, A HISTORY OF PALAEOPROTEROZOIC CONVERGENCE AND EXTENSION David Maidment^'^ & Martin Hand^ ^ Geoscience Australia, GPO Box 378, Canberra, ACT 2601. ^ Research School of Earth Sciences, ANU, Canberra ACT 0200. david.maidment@anu.edu.au, ian.williams@anu.edu.au. ^ Continental Evolution Research Group, Geology and Geophysics, University of Adelaide, Adelaide SA 2002. martin.hand@adelaide.edu.au The Late Palaeoproterozoic collisional assembly of northern and southern Australia is partially recorded by the Strangways Orogeny, a period of tectonism and magmatism which affected the Arunta Inlier of central Australia between 1780-1710 Ma. Cordilleran-style magmatism early in this period suggests that plate-tectonic-style processes drove this amalgamation, however much is still unknown about the detailed tectonothermal history of this complex orogenic belt. The Strangways Orogeny is developed at highest grade along the southern margin of the North Australian Craton in the Strangways Metamorphic Complex. The Strangways Orogeny has commonly been subdivided into two phases, the Early Strangways Event (1780-1750 Ma) and the Late Strangways Event at (1730-1710 Ma), although some workers recognise high-grade metamorphism only during the later period. Two granitoids from the Entia Gneiss Complex in the eastern Strangways Metamorphic Complex have been dated by SHRIMP to determine the timing of cordilleran-style magmatism. The Huckitta Granodiorite has a crystallisation age of 1759 Ma, while the Inkamulla Granodiorite has an age of 1766 Ma. These ages are consistent with ages of similar granitoids in the SMC and constrain arc-related magmatism to between 1766-1751 Ma, ie the Early Strangways Event. High-grade rocks of the Strangways Metamorphic Complex are overlain by a supracrustal sequence, here informally termed the Ledan Package. This previously undated package consists of the Mendip Metamorphics, the Ledan Schist and the Utopia Quartzite, and is much lower-grade than the basement it overlies. SHRIMP zircon analysis of quartzite from the Mendip Metamorphics indicates a maximum depositional age of -1760 Ma with zircon overgrowths, interpreted as metamorphic, at 1726 Ma. The Ledan Schist has a maximum depositional

age of -1774 Ma, consistent with constraints obtained from the Mendip Metamorphics. Initially there was doubt whether the 1726 Ma zircon overgrowths in the Mendip Metamorphics were formed during in-situ growth or were inherited. Therefore, to further constrain the depositional age of the Mendip package, a pegmatite from the Mendip Metamorphics was dated. The pegmatite yielded a zircon age of 1730 Ma, consistent with the 1726 Ma overgrowths being metamorphic. This indicates that the Ledan Package formed as a cover sequence deposited between the Early and Late Strangways Events, on exhumed Early Strangways basement. These data are the first conclusive evidence that the Strangways Orogeny consists of two distinct tectonothermal cycles separated by a period of exhumation and basin development. The timing of basin formation approximately corresponds to the age of emplacement of A-type granite at 1740-1745 Ma (based on recent SHRIMP dating and previous zircon geochronology) in the eastern Strangways Metamorphic Complex. Thus we propose that the deposition of the Ledan Package records the extensional dissection of the -1760 Ma convergent margin on the southern edge of the northern Australian Craton. Conceivably, such an extensional regime may have been linked to rollback-style processes, and contributed to the exhumation of the 1780-1760 Ma Early Strangways rocks. Return to a collisionally convergent regime associated with bulk sinistral transpression by -1730 Ma resulted in burial and metamorphism of the Ledan Package, and medium pressure granulite-grade reworking of the Early Strangways basement complex. This -1730 Ma event was part of large convergent regime that affected much of the southern Australian Proterozoic.

161

SGTSG Field Meeting 2003

A HOT-PLATE TECTONIC MODEL FOR PROTEROZOIC CRUSTAL EVOLUTION IN AUSTRALIA Sandra McLaren\ Roger Powell^ ^Research School of Earth Sciences, Australian National University, Acton, ACT 0200, email, sandra.mclaren@anu.edu.au ^School of Earth Sciences, University of Melbourne, VIC, 3010, email, powell@unimelb.edu.au

Proterozoic terranes in Australia record complex tectonic histories in the interval 2000-1100 Ma, that have previously been interpreted using simple intracratonic or plate-tectonic models. However these models cannot fully account for: (1) repeated tectonic reactivation (both orogenesis and rifting); (2) large aspect-ratio orogenic belts; (3) mainly high temperature-low pressure metamorphism; (4) rifting and sag giving thick sedimentary basins; (5) the nature and timing of voluminous felsic magmatism, and (6) the general absence of plate-boundary features. Australian Proterozoic terranes are characterized by an extraordinary, but heterogeneous, enrichment of the heat-producing

162

elements. This enrichment must contribute to long-term lithospheric weakening and, thus, we advocate a hybrid lithospheric evolution model with two tectonic switches: (1) plate-boundaryderived stresses and, (2) heat-producingelement-related lithospheric weakening. The Australian Proterozoic geological record is therefore a function of both the magnitude of these stresses, the way in which the heat producing elements are distributed, and how both of these change with time. It has more in common with an early Earth hot-plate style of evolution, than the modern rigid-plate style.

SGTSG Field Meeting 2003

FLUID-DRIVEN DEFORMATION PROCESSES AND CONTROLS ON GOLD DEPOSITION, ARGO SHEAR ZONE, ST IVES GOLDFIELD, WESTERN AUSTRALIA R O'Learv^ a n d S F C o x ^ 1 2

Department of Geology, The Australian National University, now at: Geoscience Australia (GPO Box 378 Canberra Act 2601) Department of Geology and Research School of Earth Sciences, The Australian National University

The Argo deposit is a newly redeveloped, shearzone-hosted gold deposit within the St Ives Goldfield in the Norseman-Wiluna greenstone belt of the Yilgarn Craton, WA. The lode is hosted by a north-striking and steeply westdipping, low displacement (50m) reverse shear zone, the Argo (A1) shear, and many associated smaller stacked fault and shear structures. The Argo shear forms part of a regionally developed sinistral strike-slip shear system. The A1 shear zone crosses a 500 metre thick, differentiated dolerite sill (Condenser Dolerite) which which intruded conformably between volcaniclastic sedimentary rocks (Black Flag Beds) and a large basalt package (Paringa Basalt) that forms part of a thicker Archaean greenstone sequence. Gold mineralisation is most strongly localised where the A1 shear cuts an iron-rich zone of the dolerite. The dip of the Argo shear zone varies with depth and plays a crucial role in influencing deformation mechanisms and the style of the gold mineralisation. The more gently-dipping sections form dilational bends and jogs and are associated with well-developed dilational breccias and vein networks, whereas the more usual steeper segments of the Argo shear lack intense veining, but are associated with intensely-developed ductile shear foliation. Active deformation in the shear zones was essential for enhancing permeability and localising fluid migration during gold mineralisation. In particular, the development of multiple stacked shear zones in the Argo deposit has increased localised flow and fluid-rock reaction within the Condenser Dolerite. Failure processes in the Argo shear zone are interpreted

to be controlled by the relative rates of build up of pore fluid factors fluid pressure /vertical stress) and shear stress (T) during repeated seismic cycles. Rapid K build-up after slip events favours extension veining followed by brittle shear failure. Rapid shear stress recovery, relative to post-seismic Xy recovery, leads to interseismic ductile creep episodes which are terminated by brittle shear failure events. The infiltration of Fe/Mg-K-Na/COs/S- enriched fluids through the A1 shear zone, and their discharge into the surrounding dolerite wall-rock, led to the development of a zoned pattern of hydrothermal alteration around the structure. A distal chlorite-dominant alteration zone is associated with low fluid/rock ratios. Potassic, biotite-dominant alteration is associated with increasing fluid/rock ratios, and proximal albitecarbonate-quartz alteration occurs in the core of the shear zone. Potassic alteration was associated with reaction-weakening and assisted strain localisation during shear zone development. Highest gold grades are associated with the most intensely altered zones, especially in the dilatant parts of the shear zone. From all the potential mechanisms that could control precipitation of gold in the Argo deposit, two mechanisms seem to be most important: (1) desulfidation reactions between the gold-bearing hydrothermal fluids and the iron-rich zone of the Condenser Dolerite in the ductile, steeper segments of the shear, and (2) cyclic fluid pressure drops during fault dilation associated with brittle failure along the more gently dipping segments of the Argo shear.

163

SGTSG Field Meeting 2003

STRUCTURAL AND DEFORMATIONAL CONTROLS ON THE DEVELOPMENT OF THE BENDIGO GOLD-QUARTZ VEIN SYSTEM Karia Olvera\ Stephen Cox ^ Dean G. TurnbulL^ Paul Quigley ^ Gary F. Johansen ^ ^Geology Department, Australian National University, ACT 0200 Australia, e-nriail: karla@geology.anu.edu.au ^Geology Department and RSES, Australian National University, ACT 0200 Australia, email:sfcox@geology.anu.edu.au ^BendigoMining N.L., Box 2113 Bendigo Mail Centre, Vic 3554 e-mail: dJhurnbull@bmnl.com.au "^BendigoMining N.L., Box 2113 Bendigo Mail Centre, Vic 3554 e-mail: pquigley@bmnl.com.au ^Bendigo Mining N.L., Box 2113 Bendigo Mail Centre, Vic 3554 e-mail: gjohansen@bmnLcom.au

The Bendigo goldfield is located in central Victoria, within the Bendigo-Ballarat zone of the Palaeozoic Lachlan Fold Belt. The host rocks are Lower Ordovician marine metasediments comprised of a quartz-rich turbidite sequence with alternating packages of sandstones and slates. This sequence was shortened during the Late Ordovician to Early Silurian, producing north-south trending upright folds. The Deborah and Sheepshead anticlines form part of a group of eleven anticlines that host over to 540 tonnes of Au in quartz vein systems within a goldfield which extends north-south parallel to the fold trend for 12 km and across strike for 4 km. The folds are typically chevron folds, with interlimb angles less than 40°. Well-developed axial surface cleavage is present in pelitic units. Folds plunge gently north or south; their amplitude varies from 200 to 300 m. The Bendigo goldfield is localised in a regional culmination zone where the regional folds change from north plunging to south plunging. There are three main types of lode-hosting structures: 1) the predominant bedding-parallel faults and associated saddle reefs, 2) east- and west-dipping bedding-discordant faults, which sometimes offset fold hinges (and saddle reefs when present), and 3) fault-related extension veins around both types of faults. Commonly, beddingparallel faults contain laminated quartz veins that develop into the historically high-grade mineralised saddle reefs in the hinges of anticlines and in synclines. At fold scale, mineralised structures display a wide diversity of geometries. Young's reef gold lode in the Sheepshead anticline is a fault-related, attenuated saddle reef with associated beddingparallel veins on both limbs of the anticline. It lies at the contact between the Railway shale and the

164

Alexandria Sandstone. Shywolup's Eastern Spurs reef, in the Deborah anticline, consists of a set of extension veins related to the reverse, west-dipping Shywolup's Fault. Saint Anthony's reef, located also in the Deborah Anticline, is a complex structure comprised of a saddle reef dissected by a westdipping, bedding-discordant reverse fault and its associated extension veins. Although these structures are relatively small in cross section, they are continuous for large distances along strike. Gold-bearing structures repeat with depth at -200 metres intervals. Lithology exerts a strong control on the location of gold lodes. The different types of quartz veins present in the Deborah and Sheepshead anticlines show a broad variety of textures and mineral assemblages. Bedding-parallel veins range in thickness from a few centimetres up to 2 metres. They exhibit generally both, uniformly laminated, and stylolitic textures. Arsenopyrite is the dominant sulphide phase in bedding-parallel veins. Extension veins are usually massive and lack wall-rock inclusions. They are gently to steeply dipping and some have been folded. Some are displaced by bedding-parallel faults. Cleavage-parallel veins are also present. They are generally composed of massive quartz. Chlorite is an accessory component of these veins. Extension veins locally offset cleavage-parallel veins. Ongoing research is focusing on the role that bedding-parallel flexural slip has played during folding and in localisation of bedding-parallel goldquartz lodes in the Bendigo area. We are exploring how the stratigraphic architecture has localised lode structures and controlled repetition of structures at depth.

SGTSG Field Meeting 2003

TECTONIC EVOLUTION OF THE WEEKEROO INLIERS, CURNAMONA PROVINCE, SOUTH AUSTRALIA Wolfgang Preiss and Colin Conor Minerals, Petroleum and Energy Division, PIRSA, GPO Box 1671 Adelaide 5000.

The Weekeroo Inliers (Walparuta, Coppertop, Morialpa and Montstephen) are four anticlinal inliers, with western bounding faults, of preAdelaidean basement in the Adelaide Geosyncline in eastern South Australia. They form the southwesternmost outcrop of the Olary Domain of the Curnamona Province, an aeromagnetically defined ovoid area of Proterozoic metasedimentary (Willyama Supergroup), igneous and meta-igneous basement. The base and top of the Willyama Supergroup are unknown and recorded history of the Olary Domain begins with sedimentation of fine to mediumgrained clastic sediments of the Curnamona Group at 1.7 Ga in a basin of unknown extent, but with probable connection with coeval sediments of the Mt Isa Inlier of NW Queensland. Coeval felsic volcanism provides U-Pb SHRIMP zircon dates. The lower part (Wiperaminga Subgroup) largely comprises albitised sediments; timing and origin of albitisation are controversial. The upper part (Ethiudna Subgroup) is calcareous, with calcsilicate, calc-albitite, schist and thin quartzite beds. Metabasalt in the Weekeroo Inliers appears to be stratigraphically associated with the Ethiudna Subgroup. Subvolcanic, bimodal mafic and felsic intrusives are consistent with an extensional regime, but stratigraphic thicknesses (2-3 km) are insufficient to invoke a major continental rift. A persistent, thin marble, calc-silcate and pyritic nietasiltstone marker with basemetal anomalism (Bimba formation) marks the base of the Strathearn Group and is overlain by volcaniclastic metasiltstone (Plumbago Formation: U-Pb SHRIMP zircon age 1693 Ma). Overlying Saltbush and Mount Howden Subgroups (carbonaceous pelites, in part with psammltic event beds) equate with the Sundown and Paragon Groups of the Broken Hill Domain. Although little Broken Hill Group occurs in the Olary Domain, the extensional event forming the Broken Hill Group basin is represented by the mafic Lady Louise Suite (U-Pb SHRIMP age 1685 Ma) intruded into Curnamona Group sediments.

Recent geochronology and certain stratigraphic relationships suggest that significant breaks exist within the Willyama Supergroup, and current detailed mapping in the Weekeroo Inliers is aimed at identifying such discontinuities, as well as tectonic excisions that may have resulted from synsedimentary extension and/or compressive deformation of the Olarian Orogeny. The Olarian Orogeny (1.6 Ga) involved overall SE to NW tectonic transport but individual fold axes vary greatly in orientation, due partly to sheath-style folding and partly to fold interference, producing structures of great complexity. The origin of a persistent bedding-parallel foliation is controversial (related to either isoclinal folding or to extension). Early folds tend to be isoclinal, recumbent, but later folds are more upright and with more consistent SW-NE and E-W trends. Amphibolite-facies metamorphism, intrusion of intermediate-felsic magmas, generation of ana-tectic melts, hydraulic brecciation and hydrothermal alteration are functions of the Olarian Orogeny. Neoproterozoic sedimentation of the Adelaide Geosyncline commenced at -830 Ma in a central, NW-trending rift zone; the Curnamona Province was on the eastern shoulder of this rift. Only during a third rift phase (-780 Ma) was this region onlapped by marginal Adelaidean clastic and carbonate sediments. A fourth rift (-700 Ma) formed NW-trending half grabens during the Sturtian glaciation; the rift faults were precursors of the present western bounding faults of the Inliers. The Delamerian Orogeny (0.5 Ga), like the Olarian, involved NW-directed plate convergence. An early phase of deformation was controlled by the Sturtian rift structures, but the later ENE-trending major folds of the eastern Nackara Arc reflect the overall compression direction. The Weekeroo Inliers result from interference of these fold phases and postDelamerian erosion.

165

SGTSG Field Meeting 2003

PETROLOGY OF THE ELAGIRI ALKALINE BODY OF THE SOUTHERN GRANULITE TERRANE OF THE INDIAN SHIELD Subrata Mukhopadhyay, Nilaparna Kar, Sarmistha Mukhopadhyay and Jvotisankar Rav Department of Geology University of Calcutta 35, Ballygunge Circular Road Kolkata 700019 Email: israv65@hotmaH.com

The Precambrian alkaline magmatism (-1600 Ma to ~600 Ma) in Eastern and southern India is represented by over forty alkaline intrusives with variable rock associations. The tectonic setting of the Proterozoic mobile belt and granulitic terrane of the peninsular India has been described by some previous workers. The alkaline magmatism is post tectonic and later than the crustal thickening of the peninsular India caused by the westward overthrusting of the Coastal Granulite Terrane (CGT) and northward underthrusting of the Southern Granulite Terrane (SGT) with respect to the cratonic regions. The alkaline intrusives were em placed essentially along deep faults and fault intersections. Occurrences of about 46 alkaline plutons from Indian peninsula are known of which about 23 belong to the aerial extent of the Southern Granulite Terrane. Elagiri alkaline pluton of Tamilnadu (12° 31' : 78° 35') has been described to be a "pyroxene syenite" body by several previous workers. In the course of present study, the Elagiri body is documented to comprise grey syenite, pink syenite, porphyritic syenite leuco syenite and pegmatoidal syenites. The entire complex is intruded by lamprophyre and gabbro intrusives. The Elagiri body is being considered to represent late/post tectonic magmatic event corresponding to stabilization of this belt.

166

SGTSG Field Meeting 2003

ANORTHOSITE BODIES OF THE CHHOTANAGPUR GNEISSIC TERRANE OF THE EASTERN INDIAN SHIELD DECIPHERING THE STYLE OF EARLY CRUSTAL EVOLUTION. Sougata Pal and Jvotisankar Ray Department of Geology, University of Calcutta. 35, Ballygunge Circular Road .Calcutta -700 019 ,India. E - m a i l : israv65@hotmaH.com

Anorthosites in the Precambrian shields all over the world have spatially and temporally limited occurrences, but they constitute significant proportion of the Precambrian crust in several continents. The Chhotanagpur Gneissic l errane of the Eastern Indian Shield is dominantly a high grade terrane coupled with volcano -sedimentary belt (singhbhum Proterozoic basin) to its south and constitutes a cratonic paradox. Such high grade belts (as the Chhotanagpur Gneissic Terrane) with characteristic magmatic imprints are considered by many workers to have developed along plate convergence. However, the Chhotanagpur Gneissic Terrane could well be a potential supra subduction zone tectonic belt .The Chhotanagpur Gnissic Terrane has a very long history of tectonism spanning over 1.6 Ga .This terrane witnessed intermittent phases of cratonisation and rigidation of crustal sector and suffered polyphase deformation fracturing, rifting and thinning . Several anorthosite bodies have been reported to dot the Chhotanagpur Gnissic Terrane, of which following four bodies are distinctive and petrologically very important.These are—1 :-Anorthosite of Bengal(23 °28':86° 40"), 2:- Anorthosite of Bela (25 :85 °5'), 3:Anorthosite of Dumka (24 °10':87 °05'), 4:- Anorthosite of Daltonganj(24°10': 84^10') The Bengal Anorthosite is an east -west trending 35 km long elongated body having a maximum central width of 8.2 km. The different varieties of Bengal Anorthosite are grey anorthosite , white anorthosite and mottled anorthosite .The range of An content of plagioclase is reported to be Ansg -Ane? -It has been suggested by workers that the magma parental to Bengal Anorthosite body suffered fractional crystallization which imprinted progressive enrichment of alkali in the later fraction. The Bengal anorthosite body has been equilibrated at 630-750^ C with about 6+ 1 kb pressure corresponding to a depth o f - 2 0 km . The anorthosite of Bela comprises following subtypesanorthosite, gabbroic anorthosite and meta gabbro.The rocks occurring as isolated blocks and pods within surrounding high grade granite gnisses. Composition of plagioclase in the anorthositic rocks has range from

Anso -An73. low structural states and high K/Rb ratio . This body might have been detached from an labradorite massif. The Dumka Anorthosite body is related with the syntectonic basic intrusive occurring within the gneisses and older meta basic rock . Petrographically the Dumka body comprises variants like anorthositic norite and noritic anorthosite which corresponds to two differentiated fractions . The anorthite content of anorthosite bodies varies from Anss -An48. The Daltonganj anorthosite body is associated with a differentiated ultramafic body and these are typically intrusive into metaultramafics related to earliest phase of anorthosite magmatism . Geochemically, the Daltonganj anorthosite body is quite distinct from the Bengal anorthosite . The Dumka and Bela bodies are bodily representation of phase of magmatic differentiation . The Bengal anorthosite body on the other hand corresponds to global phenomenon of massif type anorthosite magmatism (1.2-1.7 Ga). In the absence of proper modern analogs of Archaean anorthosite , it is indeed an difficult task to delineate the exact tectonic history . The anorthosite bodies in the investigated Chhotanagpur Gnissic Terrane are rather remarkable in the sense that they are totally confined within the highly defonned and metamorphosed granulite grade host rock . There is a possibility that these bodies had originally formed within a cratonic and continental shelf environment and might have been tectonically interrelated with the presently associated supracaistal sequences. However, typical absence of textural evidences of crystal cumulation inhibits us to consider gravitational sinking of plagioclase. Therefore it is unlikely that anorthosite bodies under discussion of the present area formed in a mid oceanic spreading ridge or Island arc environment which would have been favourable tectonic situation for layered type of anorthosite bodies

167

SGTSG Field Meeting 2003

REACTIVATION HISTORY OF THE GREAT BOUNDARY FAULT, NORTHWESTERN INDIA Amit Sahay e-mail: amitsdes@iitr.ernet.in Department of Earth Sciences, Indian Institute of Technology Roorkee 247667, India

This paper aims at reconstructing paleostress history and deciphering the pore-fluid pressure conditions during the reactivation of a regionalscale fault. Paleostress analyses of the mesoscopic structures suggest that three successive events of reactivation on the Great Boundary Fault occurred in thrust-type, strike-slip type and thrust-type tectonic-settings, respectively. Whereas the pore-fluid

pressure

was supralithostatic during the first and the third events of reactivation, it was sublithostatic during the second event. Each event of reactivation induced a fluid pressure gradient, which resulted in the focused and enhanced flow of syntectonic fluids within the high permeability locales, primarily fractures and faults. Fluid inclusion study on strike-slip veins reveals that the syntectonic fluids were highly

168

dense, Na-Ca-CI brines of formational water origin. Stratigraphic evidence in favor of a two-km thick column of overburden above Kaimur sandstone beds implies that the strike-slip faulting occurred at 160-202°C temperature and 53 MPa pressure. Variation in homogenization temperature reflects fluctuation in pore-fluid pressure during entrapment of syntectonic fluids and points to seismic pumping as a possible mechanism of fluid flow during faulting. High paleogeothermal gradient, obtained by fluid inclusion data, is ascribed to the high heat flow due to crustal stretching during the Proterozoic rifting, the basal and intermittent volcanism in the basin, and occurrence of Berach granite as the basement.

SGTSG Field Meeting 2003

MIDDLE ORDOVICIAN TO EARLY SILURIAN TECTONIC CONTROLS ON GENESIS OF WORLD-CLASS GOLD DEPOSITS AT 440 MA IN THE LACHLAN OROGEN Richard J. Squire, School of Earth Sciences, University of Melbourne, Victoria, 3010, rsauire@unimelb.edu.au J o h n McL. IVIiller, School of Earth Sciences, University of Melbourne, Victoria, 3010, imm@unimelb.edu.au

At about 440 Ma, a major gold-rich metallogenic event occurred in the Lachlan Orogen that produced world-class orogenic-gold deposits hosted by quartz turbidites in the western subprovince (e.g., Bendigo and Stawell) and worldclass porphyry-copper deposits hosted by volcanic and intrusive rocks of the Macquarie Arc in the eastern sub-province. Despite the similar age and gold-rich style of mineralisation, the deposits are now separated by up to 700 km and record strikingly different structural and metamorphic histories prior to 440 Ma. We present here a correlation of the Middle Ordovician to Early Silurian stratigraphic, structural, magmatic and metallogenic histories across the entire Lachlan Orogen. These data are used to establish a new tectonic hypothesis for the orogen that provides a tectonic link between the different world-class gold deposits at about 440 Ma. The major Middle Ordovician to Early Silurian tectonic events occurred at about 465-470, 455, 440 and 430-425 Ma. We believe that genesis of the world-class porphyry-copper and orogenic-gold deposits of the Lachlan Orogen at about 440 Ma is related to upwelling of the asthenosphere during slab rollback near the margin of the Macquarie Arc, which occurred in response to collision of a seamount south of the arc at about 455 Ma. The shoshonitic magmas and associated porphyrycopper deposits of the Macquarie Arc formed during extension in an intra-oceanic back-arc setting during slab rollback. Upwelling of the

asthenosphere during slab rollback not only assisted in the generation of the shoshonitic magmas but also provided the heat engine to drive hydrothermal fluids up to 600 km inboard of the arc margin. As a result, elevated fluid pressures drove brittle deformation associated with orogenic-gold deposits in the Bendigo and Stawell Zones, rather than representing a major phase of compression. Differences in the structural and metamorphic histories of the Stawell and Bendigo deposits mean that structural thickening and associated metamorphism at about 440 Ma in the western sub-province could not have provided the heat source to produce the gold-bearing fluids. Also, there are no changes in sedimentation in the adjacent Melbourne Zone at about 440 Ma (cf., 455 Ma deformation). Therefore, the western sub-province was still within an east-west shortening regime and thus did not record the extension that occurred in the Macquarie Arc. Basement architecture may also have been important in controlling the extent of the hydrothermal fluids and thus the location of the orogenic-gold deposits. The fluids associated with the Bendigo and Stawell deposits were probably channelled between the partiallysubducted seamount and buried pre-Cambrian basement. The source of the gold associated with the orogenic-gold deposits is not known, thus upwelling of the asthenosphere provided only a heat engine and fluid source.

169

SGTSG Field Meeting 2003

TECTONIC HISTORY FROM 3D GEOLOGICAL MODELLING: KANOWNA DISTRICT, WESTERN AUSTRALIA G-l- T r i p p " , B . K . Davis^ , N J . A r c h i b a l d ^ a n d J . T r o f i m o v s ^ ^ gerard_tripp@placerdome.com; ^ brett.davis@rsgglobal.com; ^jessica@mail.earth.monash.edu.au

Three-dimensional geological modelling is applied to mature exploration districts for the purpose of ore discovery, as a result of improved geological understanding. Data sources including regional geophysics, geological mapping and exploration drilling are analysed in 3-D to assess geology, structure and the distribution of mineralization. However, modelling these data sets at regional scales (1:50,000 and smaller) combined with sedimentology, volcanology and geochronology, has direct implications for the tectonic history of an area. In the Kanowna District, Western Australia this approach has enhanced the understanding of how Archaean tectonics impacted on greenstone architecture, and created pathways for mineralization. Kanowna District Geology The Kanowna district is located on the eastern limb of the regional-scale Scotia-Kanowna Dome; a granite-cored D2 fold antiform that has disrupted Archaean greenstones of the Kalgoorlie Terrane. The greenstone structure is controlled by a series of folded stratigraphy-repeating D1 thrust faults that juxtapose ultramafic, mafic and felsic lithological packages. Ultramafic and mafic rocks include layered komatiite flows and pillowed high-magnesium basalt sequences. The ultramafic rocks were contemporaneously extruded with dacitic volcanic rocks, which formed the sediment source for coarse epiclastic sequences of mafic to dacitic composition, finely layered turbidites, and bedded grits and sandstones. The primary distribution of these rocks is a function of significant lateral facies variations, and is further complicated by deformation. Granitoid rocks have disrupted the greenstone sequences and also intruded along early pre-D2 shear zones. Mineralization in the district is spatially associated with these intrusions, including the world-class Kanowna Belle gold deposit. The early D1 and D2 shear

170

nicka@g8oinformex.com;

zones and syn/late-tectonic granitic rocks are cross-cut by a series of low-displacement faults. Tectonic history Physical volcanology and provenance studies indicate contemporaneous magmatism, volcanism and sedimentation in the Kanowna district with a bi-modal magmatic association In localised volcanic centres. Coeval komatiitic and dacitic magmatism with locally derived epi-clastic sedimentation occurred on a regional scale in the Boorara Domain indicating autochthonous emplacement in an arc-related setting. Later deformation from N-S directed shortening during basin-closure produced thrust stacking of the stratigraphy as revealed by geological mapping, and detailed microfabric analysis of the Fitzroy, Last Chance and Gordons shear zones. A major change in tectonic transport occurred with thrust faults and stratigraphy folded by an ENE-WSW directed shortening that produced much of the regional fabric. The Kanowna anticline is interpreted as a folded thrust duplex bounded by the Kanowna Shear, a D1 transfer zone, and the D1 Shamrock fault roof thrust, with the core of the thrust duplex now exposed by erosion of the Shamrock Fault. The 2656±5Ma Scotia Batholith intruded synchronously with regional D2 folding, emplacing swarms of apophyses into D1 thrust faults at Kanowna Belle (2656±6Ma), and throughout the Kanowna district. Gravity modelling indicates a shear controlled intrusive pathway for the Scotia batholith as part of a swarm of NNW trending shear zones. New seismic interpretations demonstrate high density fracturing below the Kanowna district with flat lying intra-greenstone decollements rising up under the highly fractured zone, which may have controlled magmatic and hydrothermal fluid flow into the greenstones

SGTSG Field Meeting 2003

ISAM DEFORMATION IN THE WESTERN FOLD BELT OF THE MOUNT ISA INLIER R. J. Gordon & R. J. Holcombe Department of Earth Sciences, University of Queensland. Brisbane, Queensland, Australia 4072

Basin processes, including half-graben fault systems dominated crustal processes at Mount Isa for at least 200 Ma prior to the Isan Orogeny. The Isan Orogeny is a protracted crustal event that ranges from -1585 Ma to -1500 Ma. Previous interpretations are that the Isan event was dominated by west-verging thin-skinned thrusting in the east of the Mount Isa Inlier with subsequent east-verging thin-skinned thrusts to the west. The Mount Isa Fault System is a prominent but diffuse boundary that separates low-grade rocks to the east from more structurally complex rocks to the west that expose numerous highgrade mid-crustal windows. Included in these windows is the -1655 Ma Sybella Batholith containing strong syntectonic fabrics related to the mid-crustal extensional basin process. This paper documents the detailed threedimensional structure of the rocks west of the Mount Isa Fault, and through a series of restored cross sections shows that the Mount Isa Fault System evolved progressively from west to east with increasing brittle and less intense deformation. Contractional deformation involved elements of thin-skinned deformation linked to both folding and thick-skinned reactivation of basin faults. In particular the proto-Mount Isa and May Downs Faults (between 20 and 30 km apart) have major influences on the geometry and strain distribution of the panel of rock between them. Because of the obliquity of the bounding, thick-skinned faults and the interaction with a northeasterly-verging

block to the west, the amount of east-west horizontal contraction varies from -20% in the north to greater than 50% (perhaps -80%) -70 km to the south. The strain distribution interpreted from reconstruction correlates reasonably well with regional strain measurements (fry plots and xenolith aspect ratios) although complicated by local strain partitioning considerations. The Reconstructed sequence in the evolution of these rocks is: a) Initial gentle folding between the May Downs and Mount Isa Faults generated by semi-rigid blocks to the west and east of the Sybella Batholith converging on each other over a shallowly east-dipping deepcrustal detachment fault. b) Initiation of a mid-crustal decollement and associated development of a major thrust. A major fault propagation antiform developed above this thrust. c) Continued activity on the decollement with the development of another major thrust and associated subsidiary thrusts and intense folding. d) Movement on the mid-crustal decollement ceased; and the major basin margin faults reactivated as thrusts with shortening at depth again accommodated along the deep-crustal detachment fault. e) Sinistral strike-slip reactivation of preexisting fault planes, either late in the Isan Orogeny or post-lsan.

171

SGTSG Field Meeting 2003

172

SGTSG Field Meeting 2003

AUTHOR INDEX Agarwal, B N P Aifa, T Aillieres, L Allen, C Alt-Epping, P Archibald, N J

107 137 65 72 155 170

B Badarch, G Baldwin, S L Barley, M E Barovich, K Batt, G E Baxter, J L Bell, T H Belton, D X Berry, R F Bestmann, M Betts, P Bleeker, W Blenkinsop, T Blewett, R Bodorkos, S Bradshaw, J D Brouwer, F M Buchan,C

128 119 123 59, 64 119, 125 108 116 34 61, 124, 159 111 62, 65, 67 45 115 42 43, 147 29 128 40, 112, 113, 128

C Campbell, 1 H Carter, T J Cassidy, K Cawood, P A Celotto, S Champion, D Chen, G Chen, S F Clark, D Cobb, M M Coffin, M F Collins, A S Collins, C Conor, C H H Corrigan, D Cottam, M A Coutts, M Cox, S F Crowe, W Crowhurst, P V Culver, K E Cummins, P

72 34 42 22, 27, 54, 58 111 42 104 51 90, 147, 148 149 23 40, 70, 113 124,150 148 151,165 39 119, 125 92 88, 92, 163, 164 84, 85, 87 120 56 148

Cunneen, J Cunningham, W D Czarnota, K

85, 86, 87 128 63, 152

D Daczko, N R Dale, J Dalziel, I W D David, V Davis, B K Dean, A A Dentith, M Dijkstra, A H Drost, K Duboz, C Dugdale, J Duncan, A C Dunlap, W J

23 126, 153, 154 38 74 92, 170 127 90 22, 128 31 144 73 53 118

E Elmer, F L Evans, D A D Evans, K A Fan, W Ferris, G M Fitzsimons, 1 C V^ 1 Forbes, C Foster, D A Friend, C

153, 154 37 94 104 61 35, 56, 69, 70 129, 130, 136 62 28, 34 41

G Gartrell, A Gehmlich, M German, G W H Gessner, K Gibson, G Giles, D Gleadow, A J W Glen, R A Goleby, B Gordon, R J Goscombe, B

91 31 95, 155 103, 155, 156 63 65, 67 34,160 74 42 157, 171 28, 131, 132 133, 134, 158 155 Gow, P 28, 131, 132, 133 Gray, D 134, 135, 158 155 Gregory, M 28, 135 Gregory, R 42 Groenewald, B 104 Guo, F

173

SGTSG Field Meeting 2003

H 26 Hall, R 116 Ham, A P 26 Hamilton, P J Hand, M 59, 64 ,71, 131, 132 133, 134, 149, 158, 161 43 Hansen, D 54 Harris, L B 33 Harrison, M Harrowfield, M 84, 85, 87 Henson, P 42 Herve, F 137 Hobbs, B E 97, 106 Hobbs, B 98 Holcombe, R J 157, 171 Holland, T 126, 153 Hollingsworth, D A 52, 54 Holm, O 159 Hulscher, B 35, 136 Hynes, A J 19 J Jacobs, J Janssen, D P Johansen,G F Johnston, S T Jones, P A Jones, S K Kar, N Keep, M Kennedy, A Keppie, J D Kinny, P D Kohn, B P Korsch, R J Kroner, A L Lane, R Leat, P T Lefort, J P Lennox, P G Leonard, M Li, ZX Lin, G Linnemann, U Lisk, M Lister, G S Longley, I Lorencak, M Love, G Lyons, P 174

36 70, 160 164 19, 24 155 68 166 84, 85, 86, 87 117 19, 138 26,41, 149 34, 75 50 124 50 127 107, 137 75, 152 90 37 104, 106 31 91 25, 62, 67 142, 143, 144 84 34 41 159

M Maidment, D 71, 161 132 Mawby, J 32 Mazzoli, S 118, 162 McLaren, S 106 McLellan, J G 31 McNaughton, N J 23 Meckel, T 92 Micklethwaite, S 29 Millar, I L Miller, J McL 73, 135, 169 Moresi, L 103, 156 Morley, C 83 Mosher, S 23 Mruma, A 40 Mukhopadhyay, Sar. 166 166 Mukhopadhyay, Sub. 19 Murphy, J B 115 Murphy, T N Nance, R D Nicoll, M G

19, 138 50

O O'Leary, R Occhipinti, S A Olvera, K Ord, A

163 57 164 94, 95, 97, 106

P Pal, S Palmer, N Passchier, C Philippot, P Phillips, D Piazolo, S Pickard, A L Pisarevsky, S A Playford, P E Potma, W Powell, R Preiss, W Prior, D

167 85 158 44 30, 73 111 123 141 79, 89 93 153, 154, 162 165 111

Q Quigley, P

164

R Ray, J 166, 167 Razakamanana, T 124 Reddy, SM 40,57,112,113 Regenauer-Lieb, K 102 Reid, A J 30 Rey, P F 44

SGTSG Field Meeting 2003

Riganti, A Roberts, P A Romer, R L Rosenbaum, G Rovardi, M Rutherford, L S Sahay, A Sandiford, M Schaubs, P Schellart, W P Schwa rz, M Seward, G Shelley, D Shiner, P Sircombe, K Smyth, H Spaggiari, C V Spiers, C Spikings, R A Squire, R J Stallard, A Steele, D Stern, R Swain, G

Thebaud, N Thompson, N Tingay, M Tonk, C Tripp, G 1 Trofimovs, J Trouw, R A J Trzebski, R Turnbull, D G Tuttle, M Tyler, I M Van Arsdale, R Van Rensbergen, P Varvell, C A Vassallo, J J Vaughan, A P M W Walters, N Wang, Yanghua Wang, Yuejun Warren, J Weinberg, R Welch, P H Wertz, K L

51 106 31 25, 142, 144 93 64 168 43, 49, 160 95 143, 144 59 111 114 32 55 26 28, 52

Wheeler, J Wijns, C Wilde, A R Williams, I Wilson, C J L Wilson, R D Woodhouse, A Wyche, S

111 103, 156 155 71 30, 72, 73, 160 155 59 51

Y Yuen, D Z Zaw, K Zhang,S Zhang,Y

102 123 37 104, 106

111

120 72, 169 114 159 117 59 44 87 83 31 170 170 29 75 164 148 58

148 83 56 74 29, 127 97 104 104, 106 83 43, 93, 96, 156 116 23

175