Issuu

Geological Society of Australia

ABSTRACTS Number

25-29 September 1995

SPECIALIST GROUP IN TECTONICS AND STRUCTURAL GEOLOGY

GEOLOGICAL SOCIETY OF AUSTRALIA SPECIALIST GROUP IN TECTONICS AND STRUCTURAL GEOLOGY

CLARE VALLEY CONFERENCE 25-29 September, 1995

ABSTRACTS No. 40

TABLE OF CONTENTS

List of Oral Presentations

ii-v

List of Poster Presentations

vi-viii

Abstracts (oral and poster) arranged alphabetically by first author

1-185

ORGANIZING COMMITEE Chairman Prof. Ron Vernon School of Earth Sciences, Macquarie University, Sydney, NSW 2109 Ph. (02) 850 8413 Fax (02) 850 8428 e-mail: ronald.vernon@laurel.ocs.mq.edu.au

Secretary Dr Nick Cook Pasminco Exploration, Broken Hill, NSW Ph. (080) 88 8293 Fax (080) 88 8312 e-mail: ncook@geko.com.au

Treasurer Dr Scott Johnson School of Earth Sciences, Macquarie University, Sydney, NSW 2109 Ph. (02) 850 7694 Fax (02) 850 8428 e-mail: scott.johnson@mq.edu.au

Editor Dr June Hill School of Earth Sciences, Macquarie University, Sydney, NSW 2109 Ph. (02) 850 7694 Fax (02) 850 8428 e-mail: june.hill@mq.edu.au

General Committee: Dr Paul Lennox Dept. Applied Geology, University of N. S. W., Kensington, 2052 Ph. (02) 385 4809 Fax (02) 385 5935 e-mail: P.Lennox@unsw.edu.au

Dr Dick Glen NSW Geological Survey, P.O.Box 536, St Leonards 2065 Ph. (02) 901 8888 Fax (02) 901 8256 e-mail: richardg@ozemail.com.au

Dr Colin Wilkins Dept. of Geology and Geophysics, University of Sydney, Sydney 2006 Ph. (02) 692 2924 Fax (02) 692 0184 e-mail: colin@es.su.oz.au

Dr Brian Marshall Dept. Applied Geology, University of Technology, Sydney, Broadway 2007 Ph. (02) 330 1759 Fax (02) 330 1755 e-mail: bmarshal@acacia.itd.uts.edu.au

LIST OF ORAL PRESENTATIONS AUTHOR & TITLE (Where the presenter of the talk is not the first author listed, the presenter is underlined.)

Adshead, N. S. Integration of metamorphic P-T-t paths and structural development using inclusion trails in porphyroblasts Ames, C. C. Three dimensional structural modelling of the Isa Valley Baldwin, S. L., Lister, G. S. & Vandenburg, L. C. Rapid operation of a major crustal shear zone in the Aegean metamorphic core complexes of Ios, Cyclades, Greece Bateman, R. Microstructural evolution of magmas Beaumont, C., Ellis, S. & Pfiffner, A. Geodynamical models of the transition from subduction to collision tectonics Bell, T. H. & Wang, J. Rotated staurolite porphyroblasts in the Littleton Schist at Bolton Connecticut: true or false? Braun, J. Numerical models of crustal deformation applied to the Alice Springs Orogeny of Central Australia Brown, M. Feedback mechanisms and channelized magma transfer Buick, L S., Williams, I. S. & Cartwright, L SHRIMP constraints on the timing of regional granulite-facies metamorphism and deformation in the Reynolds Range, Central Australia: implications for the tectonics of low-pressure/high-temperature metamorphism Cartwright, L, Harley, S. L., Buick, I. S. & Barr, T. Cooling and uplift rates of granulite-facies rocks from the Rauer Group, Antartica: constraints from oxygen isotopes in marbles and implications for tectonic processes Christie, J. M. Quartz paleopiezometers and the effects of annealing Cox, S. F., Streit, J. E., Kanagawa, K. & Fitz Gerald, J, D. Microstructural and mechanical evolution of fault rocks in the presence of reactive pore fluids Cox, S. J. D., Korsch, R. J., Nicoll, M. & Totterdell, J. 3D modelling of the Bowen Basin: A solution using a combination of software Crowhurst, P. V., Hill, K. C. & Foster, D. A. Thermochronological constraints on the tectonic evolution of northern Papua New Guinea Davis, B. K. Microstructural processes and development of macroscale fold geometries in the Robertson River Metamorphics, North Queensland Dunlap, W. J., Teyssier, C., Hirth, J. G. & McDougall, I. Thermal, mechanical and kinematic evolution of a ductile duplex, Central Australia

•• 11

PAGE NO.

2 5

6 8 10

18 22

26 30

32 33

Flottmann, T. & Sandiford, M. Tertiary compression^ tectonics of the Mt Lofty Ranges and Gulf of St Vincent region, South Australia Foden, J., Turner, S. & Sandiford, M. Granite composition in the Delamerian Orogen: a response to orogenic crustal thickening and thinning Gibson, H. J., Wilson, C. J. L. & Green, P. F. The roles of apatite fission track analysis (AFTA) and field structural mapping in tectonic studies: example from the Murchison Basin, New Zealand Gray, D. R., Gregory, R. T., Norris, R. J. & Cox, S. C. Regional scale sheath-folding and heterogeneously distributed shear strain in an evolving nappe pile, Otago Schist, New Zealand Grujic, D., Casey, M., Davidson, C., Hollister, L., Kundig, R., Pavlis, T. & Schmid, S. Ductile extrusion of Higher Himalayan Crystalline of Bhutan Himalayas. Quartz microfabric study Hand, M., Bendall. B. & Sandiford, M. Metamorphic evidence for Palaeoproterozoic oblique convergence in the eastern Gawler Craton Hand, M., Fanning, M. & Sandiford, M. Low-P high-T metamorphism and the role of high-heat producing granites in the northern Arunta Inlier Henderson, J., Henderson, M. & Wright, T. Kinematic analysis of asymmetrical fabrics in metaturbidites from the Archean Slave structural Province, northern Canada Hickey, K. A. Sequential foliation development and mineral growth in high-grade metasedimentary gneisses, Broken Hill: the production of composite foliations during folding of gneissic rocks Hobbs, B. E., Zhang, Y. & Ord, A. Crenulation cleavage or strain slip cleavage? Hoek, J. D. Mafic dykes as deformation structures Holcombe, R. J., Fielding, C. R. & Stephens, C. J. Tear fault termination of the fold-thrust belt in the northern New England Orogen Huang, W., Shan, W. L. & Fu, Z. R. Tectonic significance of bedding-parallel foliation: a case study in Beijing West-Hill, North China Hunt, G. W., Muhlhaus, H-B, Hobbs, B. E. & Ord. A. The role of elasticity in visco-elastic fold evolution Jamieson, R. A., Beaumont, C., Hamilton, J. & Fullsack, P. Tectonic assembly of inverted isograds in convergent orogens Little, T. A. An algebraic technique for measuring three-dimensional displacement gradients and strain in brittle shear zones MacCready, T. Quart c-axis fabrics not aligned with rock lineations: a technique for viewing fabrics in the XZ plane of oblique final strain increments

••• HI

67 70 72 74

76 79 92

Mancktelow, N. Deviatoric stress and the interplay between deformation and metamorphism Mason, R. A. Structure of the Alice Anticline, Papua New Guinea: serial balanced cross-sections and 3-D restoration Mawby, J., Foden, J,, Kelley, S. & McDougall, I. Chronological constraints on the thermal evolution of Harts range, Arunta Inlier Miller, J. A. & Cartwright, I. A clockwise P-T path for granulite facies metamorphism in the Mallee Bore area , northern Harts Range: implications for the Proterozoic tectonic evolution of the Arunta Inlier, Central Australia Miller, J. M. & Gray, D. R. Subduction and sediment accretion by underplating in the eastern LachlanFold Belt Morand, V. J. The Barrabool Hills metagabbro: a curious piece of the Cambrian jig-saw puzzle in Victoria Munroe, S. M. Tectonic and structural controls on hydrothermal fluid flow at the Porgera gold deposit, PNG Neil, E. A. & Houseman, G. A. Geodynamic models of central Asia: strong Tarim Basin or weak Tien Shan? Nguyen, P. T., Powell, C. McA., Harris, L. B. & Hein, K. A, A. Development of vein systems in shear zones at the Revenge mine, Kambalda, Western Australia: evidence of palaeoseismic events Norris, R. J., Youngsen, J. H. & Jackson, J. A. The structural evolution of active fold and fault systems in central Otago, New Zealand: evidence revealed by drainage patterns O'Halloran, G. J. & Cas, R. A, F. Sedimentary responses to deformation - examples from the Late Devonian Mansfield Basin Oliver, N. H. S. Kanmantoo pelite-hosted copper deposit: pervasive and channelized late-syntectonic fluid flow above a pluton Oussa, S. A. Structural evolution of the Kalinjala mylonite zone Park, Y. & Means. W. D. Direct observation of deformation processes in crystal mushes Passchier, C. W. Non-simple shear zones Power, W. L. & Cox, S. J. D. The great Australian lineament debate - what is important? Preiss, W. V. Tectonic evolution of the mid-north, South Australia Sandiford, M., Bingemer. A. & Hand, M. Basement involved deformation, heat production and low-pressure metamorphism in the Adelaide Fold Belt and some wider implications

95 99

102 107

109 Ill

113

115 117

119

122

124

132 133 136 139 141 146

Sandiford, M. & Coblentz, D. D. Continental collision, ridge torques and stress in the Indian-Australian plate Scott, R. J. & Lister, G. S. The origin of Cordilleran metamorphic core complexes Stiiwe, K. Uplift versus exhumation. A kinematic model and implications for the geometrical evolution of the Tauern Window, eastern Alps Teasdale, J. P. Preliminary investigations on the tectonic evolution of the western Gawler Craton Upton, P. Deformation as a driving force for fluid flow within an active orogenic belt, Southern Alps, New Zealand Van Kranendonk, M. J. & Collins, W. J. Lateral escape tectonics following ca. 3300 Ma extensional core complex formation in the eastern part of the Archean Pilbara Craton, Western Australia Vandenburg, L. C. & Lister, G. S. Structural analysis of basement tectonites from the Aegean metamorphic core complex of Ios, Cyclades, Greece Venn, C. & Lister, G. S. Tectonic evolution of the high pressure schists and gneisses of the Amoss-Mayavetch metamorphic core complex, New Caledonia Williams, M. L. & Karlstrom, K. E. Looping P-T paths, LPHT metamorphism, and the evolution of the middle crust: the Proterozoic Orogen of southwestern USA. Williams, P. F. & Johnston, D. H. Development of regional scale fold nappes: an example from the Monashee Complex, Canadian Cordillera Winsor, C.N. The prediction of rock mass discontinuities in the Clare Valley Syncline, based on regional geology interpretation Worley, B. A., Wilson, C. J. L. & Powell, R. Crenulation cleavage formation: evolving deformation mechanisms with increasing metamorphic grade

147 153 157

160 162

166

168

169

171

173

177

179

LIST OF POSTER PRESENTATIONS AUTHOR & TITLE

PAGE

(Where the presenter of the poster is not the first author listed, the presenter is underlined.)

Aslund, T., Cartwright, L & Oliver, N. H. S. Syntectonic granite emplacement and subsequent metasomatism: Revenue Granite, Mt Isa. Berry, R. F. Palaeozoic tectonics of western Tasmania Betts, P. & Valenta, R. The early tectonic history of the Fiery Creek Dome region in northern Mount Isa Inlier Bordokos, S. & Oliver, N. H. S. Emplacement, timing, and genesis of the Mabel Downs Granodiorite in the Halls Creek Orogen, East Kimberley Bucher, M., Foster, D. A. & Gleadow, A. J. W. The Ar/Ar thermochronology of granites from the western Lachlan Fold Belt, southeastern Australia: implications for tectonic development Coughlin, T. J. & Holcombe, R. J. The kinematic evolution and tectonic implications of major arc-oblique lineaments, in the northwest Argentine Andes Daly, S. J., Fairclough. M. Fanning, C. M. & Rankin, L.R. Tectonic evolution of the western Gawler Craton: a Palaeoproterozoic collision zone and likely plate margin Ehlers, K. & Lister. G. S. Transience vs ambience Flottmann, T. & James, P. Structural geometry and strain distribution in two superposed and reactivated basins: the Adelaide foreland fold-thrust belt Forster, M., Baldwin, S. L. & Lister, G. S. Structural and thermal evolution of the upper plate of the Aegean metamorphic core complex of Ios, Cyclades, Greece Foster, D. A. & Ehlers, K. The Mesoproterozoic and Neoproterozoic thermal and tectonic history of the southern Gawler Craton: constraints from Ar/Ar geochronolgy Glen, R. A. & Hancock, P. L. Folding as a mechanism of mild tectonic inversion: the southern margin of the Bristol Channel Basin, England Gray, D. R. & Miller, J. M. Implications of a subduction complex for the tectonic evolution of the Lachlan Fold Belt Harbort, T. A., Holcombe, R. J. & Stephens, C. J. Structural GIS and computer applications in structural geology Hill, E. J. Extensional deformation of continental crust prior to break-up and sea-floor spreading: a study of Misima Island, Papua New Guinea

NO,

11 13

41 42

62 gg

James, P. R., Flottmann, T., Barret, L., Buhrer, D., Macdonald, A. & Szmidel, R. The geometry and kinematics of thrusts and shear zones from the southern Adelaide fold-thrust belt on the southern Fleurieu Peninsula Jiang, Zh., Oliver, N. H. S., Barr, T. D., Power, W. L. & Ord, A. Modelling of fluid migration along faults in different structural environments Johnson, S. E. & Moore, R. R. 3-D computer movies of porphyroblast microstructures Korsch, R. J. & Totterdell, J. M. Eatern margin of the Bowen-Gunnedah-Sydney Basin: geometry of the BurungaLeichhardt-Moonie-Goondiwindi-Mooki-Hunter fault system Leeming, P. M., Harris, L. B., Nash, C. R., Rankin, L. R. & Crowe, W. Lithospherical domains and deformation of the Eastern Ghats Tectonic Province and southern Singhbhum Craton, northern Orissa, India Lennox, P. G. & Flood, P. G. Age and structural characterisation of the Texas Megafold, southern New England Orogen, eastern Australia Lister, G. S. Porphyroblasts - the inside story Marshall, B., Franklin, B. J. & Tomkins, A. G. Emplacement and deformation of the Wondalga Granodiorite, near Adelong, Lachlan Fold Belt, NSW Masuda, T. & Mizuno, N. Deflection of non-Newtonian simple shear flow around a rigid spherical body using a finite element method McLaren, A. C, & Fitz Gerald, J. D. Microstructural changes and deformation during the phase transformations in solid ammonium nitrate Miller, D. T. Delamerian fold thrust belt - a new tectonic model Offler, R., O'Hanley, D. & Lennox, P. Tectonic significance of veins in the Woodsreef asbestos deposit, southern New England Fold Belt, NSW Oliver, N. H. S. & Barr, T. D. Dehydration and anatexis caused by mafic melt migration through fractures, Halls Creek Orogen, east Kimberleys Oliver, R. L. Influence of tectonics on the occurrence of talc in the Mount Lofty Ranges, South Australia O'Sullivan, P. B., Kohn, B. P., Foster, D. A. & Gleadow, A. J. W. Fission track evidence for Late Permian-Early Triassic and Middle Cretaceous denudation in the eastern Lachlan Fold Belt, NSW, Australia Parker, A. J. & Hume, R. G. Geoscience data model: towards a standard for spatial data transfer Potma, W. A. Pre-Isan extension: effect on the geology of the Mitakoodi Quartzite/Overhang Jaspilite contact, Kuridala region, Mount Isa

•• Vll

81 83 85

91 97

100

104

106 121

126

128

130

134 137

Powell, R. Calculated mineral euqilibria: representation and relevance to tectonothermal evolution Pryer, Lu L. Influence of water on the character of mylonite produced by shearing during lowgrade metamorphism in feldspar rich rocks Ronaszeki. J.. Ascott. B. & Powell. C. McA. The three-dimensional structure of the Mount Whaleback Mine, Newman, WA Saruwatari, K., Michibayashi, K. & Shimizu, I. Shape preferred orientation of spinel within dunite: evidence of plastically deformed peridotite in the Yakuno Ophiolite, Japan Schaefer, B. F., Hand, M., Bendall, B., Foden, J. & Sandiford, M. Chronology of magmatism and deformation in the Lincoln Complex; constraints on the Kimban Orogeny Scott, M.M. Thrusting of the northern Molong volcanic belt, Wellington, New South Wales Scott, R. J. Geometric evolution of the Buckskin-Rawhide metamorphic core complex, SW USA Sims, J. P. & Wilson, C. J. L. Strain-related textural development in a granulite-facies shear zone - the Rauer Group, east Antartica Stuwe, K. & Fabel, D. Long versus short term denudation rates for the eastern Alps. Constraints on the uplift history? Suzuki, S., Morikawa, T., Nakayama, Y. & Masuda, T. Energetics of grain boundary migration of quartz Teasdale, J. P., Schaefer, B. F. & Sandiford, M. Post-Delamerian thrusting along the Paralana Fault, northern Flinders Ranges, South Australia Upton, P. Deformation-induced reaction in the Alpine Fault mylonites, Southern Alps, New Zealand Willman, C.E., Hendrickx, M., VandenBerg, A. H. M., Cayley, R. A. & Taylor, D. A down-cutting thrust in the Mount Useful Slate Belt of eastern Victoria Wilson, C. J. L. & Zhang, Y. Experimantal analogues, microstructure development in high temperature high strain zones Yamaguchi, H. & Masuda, T. Deformation of carbonate rocks in Modi Khola of the Annapurna area, central Nepal Yassaghi, A., James, P. R. & Flottmann, T. Ductile thrusts and shear zones of the Adelaide Hills; geometrical constraints Zakowski, S. Migmatization during low pressure-high temperature metamorphism in the Mt Lofty Ranges, South Australia Zhang, Y., Scheibner, E., Hobbs, B. E., Ord, A. & Drummond, B. Lithospheric structure in the eastern Australian passive margin as revealed by gravity modelling

••• Vlll

138

143

144 148

149

150 151 155

158

159 161

164

174

176

180 181 183

185

Abstracts

INTEGRATION OF METAMORPHIC P-T-t PATHS AND STRUCTURAL DEVELOPMENT USING INCLUSION TRAILS IN PORPHYROBLASTS N.S. Adshead Department of Earth Sciences, James Cook University, Townsville, Q. 4811, Australia Integrated microstructural, electron microprobe and bulk geochemical work has revealed a strong correlation between metamorphic reactions, the timing of growth of andalusite, garnet and staurolite porphyroblasts, the geometry of their inclusion trails and the macroscopic structural development of the Kanmantoo region of the Adelaide Fold Belt. Multiple phases of deformation and concurrent metamorphism reveal that the prograde metamorphic path of the rocks is closely linked to the deformation path through a staircase-like pattern of pressure and temperature jumps, synchronous with the development of vertical and horizontal foliations. The intimate relationship between successive growth of the different porphyroblastic phases (within a layer of the same bulk composition) and their structural position in developing folds are preserved in the Strathalbyn Anticline/Macclesfield Syncline fold couplet. This demonstrates that the growth of a porphyroblastic phase is controlled by its position within the structure, and hence pressure and temperature variation. It enables the apparent anomaly of synchronous growth of andalusite and staurolite, over a potentially wide range of PT space, to be resolved. Therefore, the rocks have not undergone a simple isobaric heating/cooling path, but a more complex P-T-t path which involves changes in both temperature and pressure. The growth of andalusite or staurolite is episodic and varies depending on movement of the rock back and forth across the reaction boundary as it follows an anticlockwise staircase-like P-T-t path. This enables a fully integrated structural/metamorphic path through the orogen to be determined.

THREE DIMENSIONAL STRUCTURAL MODELLING OF THE ISA VALLEY C.C. Ames Department of Earth Sciences, Monash University, Clayton, V7C 3168, Australia. Geophysical data can place important constraints on three dimensional structural interpretations, since any interpretation must place magnetic units at the correct position, depth and geometry to produce a match with observed geophysical measurements. Although the model produced may not be innately unique, it does provide a way in which a conceptual model can be examined and determined if it is viable in the light of geophysical and geological data. Both two dimensional and three dimensional geophysical techniques were used to determine the structure and structural history of the Isa Valley. The Isa Valley in the region surrounding Lake Moondarra, is an area of complex deformation. As a result many different interpretations have arisen attempting to explain the geometry and type of faulting under the non-magnetic Mt. Isa Group cover rocks (Bell, 1991; Winsor, 1986; Dunnet, 1976). To give further understanding to the nature and timing of faulting events to incorporate into the three dimensional model, an animated two dimensional reconstruction using a multimedia software package (Macromedia DIRECTOR) was attempted in the smaller scaled, well mapped region east of Lake Moondarra (Bain et al, 1992). Faulting within the Eastern Creek Volcanics (E.C.V.'s) accounts for about 20-25% of the shortening within the E.C.V.'s. The remainder of the apparent shortening was accomplished via open folding. Based on the two dimensional reconstruction and the viability of the three dimensional model, deformation within the E.C.V.'s occurred in the following order (most recent back to earliest):(1) South dipping normal faulting. (2) North-south striking strike/slip faulting. (3) Dominant NW trending sinistral strike/slip scissor faulting and N to NE striking dextral faulting. (4) Spoon shaped, south dipping normal faulting. (5) North dipping, open folding preceding major faulting events. (6) Syndepositional faulting. The resultant film produced demonstrates both the plausibility and reliability of the partial reconstruction to an open fold with a thinning in the stratigraphic thickness of the E.C.V.'s to the NE. The method of reconstruction provides a basis for ongoing attempts to reconstruct the difficult structure within the Mt. Isa Inlier. A three dimensional structural model of the Isa Valley was constructed on the interpretation of aeromagnetic data, the two dimensional partial reconstruction and previous geological studies of the region. VULCAN(ENVISAGE) three dimensional geological modelling was used to gain an understanding of the structure and to obtain a synthetic geophysical response using NODDY geophysical calculations of the block model of the magnetic E.C.V.'s. This calculated geophysical response was compared to the aeromagnetic data to deduce the viability of the model. Three dimensional modelling demonstrated variation in the magnetic susceptibility of the E.C.V's north of the major E-W repetition (Spillway Fault). Modelling implied the region is composed of a major north plunging folding event with less amplitude and longer wavelength to the east. This fold was then faulted by the major south dipping normal fault with greater displacement in the east. Spoon faults are bounded by the Mt. Isa Fault Zone and the Police Creek Fault. Subsequent NE block up, NW-SE trending sinistral faulting results in minimal displacement under the Mt. Isa Group near Lake Moondarra and greater displacement in the Eastern Creek Volcanics.

The match between the calculated and real data was reasonable, however resulting amplitude of the calculated magnetics implies the model was deeper than reality, even though the model provided a similar magnetic signature under Lake Moondarra. Overall the three dimensional model proved to be a viable structural model that fits the geological and geophysical datasets. REFERENCES Bain J. H. C., Heinrich C. A. & Henderson G.A.M. 1992. Stratigraphy, structure, and metasomatism of the Haslingden Group, east Moondarra area, Mount Isa: a deformed and mineralised Proterozoic multistage rift-sag sequence. In Stewart A. J., & Blake D. H. (Editors), Detailed Studies of the Mount Isa Inlier. Bureau of Mineral Resources, Australia, Bulletin 243., 125-136. Bell, T.H. (1991) The role of thrusting in the structural development of the Mount Isa Mine and its relevance to exploration in the surrounding region. Economic Geology, 86, 1602-1625. Dunnet, D. (1976) Mt Isa - reconstruction of a faulted ore body. Phil Trans. R. Soc. Lond, 283, 333-344. Winsor C. N. 1986. Intermittent folding and faulting in the Lake Moondarra area, Mt Isa, Queensland. Australian Journal of Earth Sciences 33, 27-42.

SYNTECTONIC GRANITE EMPLACEMENT AND SUBSEQUENT METASOMATISM: REVENUE GRANITE, MT ISA T. Aslund .1. Cartwright & N.H.S. Oliver Department of Earth Sciences, Monash University, Clayton, VIC 3168, Australia School of Applied Geology, Curtin University, GPO flax U 1987, Perth, WA 6001, Australia The Revenue Granite intruded lower Corella Formation metasediments at approximately 1740 Ma (Di extension). Located 10 km north of Duchess, the Revenue Granite was syntectonically emplaced and underwent substantial metasomatism shortly after intrusion. Deformation in and around the Revenue Granite was relatively complex due to the close coincidence of magmatism, deformation and metasomatism. Five sets of folds, and three sets of foliations and stretching lineations are recognised, the bulk of which are related to pre-D2 deformation. Scapolitisation (temperatures >500°C) was associated with the formation of L-tectonite and small scale shears. Hypotheses regarding the origin of the metasomatising fluids need to account for the stable isotopic composition of the fluid (8 0 8 to 10 %o, 8 C —2 %o) as well as its high salinity (at least 40 wt% NaCl). Two possible origins of the fluids are discussed, (1) late-stage evolution of granite magma, and (2) rapid heating of formation pore fluids. Di metasomatism in the Revenue Granite - aureole system was essentially a de-metalling event resulting in losses in all components with the exception of Na, Ca, CI, and Sr (Ti, Zr and A1 were predominantly conserved). Thus, the Revenue Granite may actually represent a now depleted source region for hydrothermal ores such as Cu and Au(?). Deformation played an important role as fluid flow was strongly structurally controlled. 1

RAPID OPERATION OF A MAJOR CRUSTAL SHEAR ZONE IN THE AEGEAN METAMORPHIC CORE COMPLEX OF IOS, CYCLADES, GREECE S. L. Baldwin1, G. S. Lister2 & L. C. Vandenburg2 1 2

Department of Geosciences, University of Arizona, Tucson, Arizona, U.S.A. V.I.E.P.S., Department of Earth Sciences, Monash University, Clayton, Victoria, 3168

Step-heating experiments have been performed on biotite, K-feldspar, and white mica from the lower structural levels of the major ductile shear zone that forms the carapace of the lower plate of the Aegean metamorphic core complex of Ios, Cyclades, Greece. These experiments have produced 40 Ar/ 39 Ar apparent age spectra with considerable complexity, consistent with the fact that the rocks in the South Cyclades shear zone have recorded the effects of many different episodes of deformation and metamorphism. Pronounced heterogeneity within and between samples can be explained as a result of recrystallization and grain growth during short-lived thermal pulses. The thermal events must be short-lived since otherwise there would be widespread resetting of apparent ages retained within grains that have not been recrystallized during the younger episodes of deformation and grain growth. Similarly, ambient temperatures between each thermal pulse must be relatively low to allow the preservation of relatively old apparent ages. The shapes of the 40 Ar/ 39 Ar apparent age spectra obtained from K-feldspar samples can be explained as due to differing rates of argon release from multiple diffusion domains with different plateau ages. Multidomain spectra have also been obtained from biotite (with up to three different components), and these correlate with observed microstructures showing up to three different episodes of recrystallization and grain growth within the one sample. Apparent age spectra obtained from white mica separated from different samples also vary greatly. There is again an exact correlation of apparent age with the microstructure, consistent with the relative timing of different episodes of recrystallization and grain growth that can be determined as the result of fabric and microstructural analysis. The implication is that radiogenic argon is released sequentially from different generations of recrystallized grains. 40 Ar/ 39 Ar thermochronology has thus proven able to provide constraints on the absolute age of different deformation and recrystallization events within a small well-studied area in which structural and microstructural control has been established.

MICROSTRUCTURAL EVOLUTION OF MAGMAS Roger Bateman Departamento de Geologia y Mineria, Universidad de Sevilla, Spain

Many reaction microstructures in igneous rocks are products of incomplete solid-state processes: however, progressive crystallization of the magma is also commonly reflected in disequilibrium microstructures, and the structural geologist must be able to recognize them. These microstructures represent incompleted reactions between crystals and melt, commonly because the products of reaction armourplate the residues of the unstable phase(s). Such reactions may take place through the evolution of a single magma, or through compositional change during mixing of different magmas. Reactions and relative timings of growth may be deduced using the same principles as used in interpretations of metamorphic-deformational microstructures. Two examples are discussed here. The Scottish Strontian pluton consists of several granitoid facies, zoned from a tonalite through porphyritic granodiorite to a biotite granite, and emplaced in an extensional fault (Hutton, 1988, Geological Society of America Bulletin, 100, 13921399). The Santa Olalla pluton (Bateman et al., 1992, Lithos, 28, 111-131.), north of Sevilla in Andalucia, is asymmetrically zoned from a pyroxene gabbro through dominant tonalites to a cordierite quartz monzonite. Tonalites of both have abundant enclaves. Complicated reaction series may be identified within a single suite of rocks. Reactions such as pyroxene-^biotite+quartz; pyroxene-*^ amphibole; biotite-^ hornblende; and hornblende+ titanite-^biotite+quartz can be identified. In these Spanish and Scottish plutons, 2 or 3 populations (respectively) of pyroxenes can be inferred. The first pyroxene (pxl), here found only in the Scottish rocks, occurs as resorbed inclusions surrounded by large, irregular biotite crystals. These pyroxenes are present in felsic rocks. In appinites, these biotites may themselves be corroded and resorbed, and enclosed in hornblende. Examples can be found where a resorbed pyroxene grain is surrounded by several separated resorbed, optically continuous grains of biotite (btl) that appear to have constituted a single grain before reaction with and replacement by later hornblende. This reaction pxl-•btl is stabilized by relatively high aSi02. The replacement of biotite by hornblende apparently took place within appinites. Biotite-pyroxene contacts are typically very irregular, and are lined by fine quartz grains. The pyroxene crystal faces away from biotite are typically idiomorphic. Irregular grains of titanite, with rims of fine quartz are associated with coarse biotite, suggesting Ti is supplied to biotite by the reaction pxl+titanite-H)tl+ quartz+Cameit. This btl may be equivalent to the biotite that has been resorbed and transformed to hornblende in other rocks (see below), because similar biotite-pyroxene clots occur in the appinites, where there is an overprinting reaction producing hornblende. The second pyroxene (px2) is represented by amphibole clots, although fresh pyroxene rarely remains within them. They consist of commonly square, polycrystalline clots of amphiboles with a polygonal-granoblastic texture. Many have hornblende rims enclosing zoned actinolite crystals with Al-poor cores. Grainsize in the Spanish clots varies widely, but the hornblende rim is invariably coarser than the actinolite in the clot core. These are interpreted as a result of a pyroxene-melt reaction, the rate of which is ultimately controlled by Al diffusion through armour-plating hornblende as hornblende progressively replaces first pyroxene then actinolite, with concomitant grain growth (Castro et al., 1992, Canadian Mineralogist, 30,1093-1112. These clots are commonly rimmed by biotite bt2. The third type (px3) is formed by small prisms that occur mainly within the appinite groundmass, included in plagioclase which in places has an intercumulus microstructure. Px2 and px3 cannot be equivalents: the amphibole clots are much larger, and show much more complete reaction. For the latter reason, they are probably also younger than the clots. Possibly, the clot pyroxenes grew in one melt, and were subsequently added to this appinite magma as xenocrysts where they formed clots. The small idiomorphic pyroxenes may have grown in the appinite melt. 6

Intermediate tonalites are characterized by the presence of amphibole clots, idiomorphic hornblendes, and complex biotite-hornblende microstructures. Idiomorphic hornblende crystals in the groundmass very typically contain resorbed residua of biotite (bt2), and rarely pyroxene. In some examples, some flakes have recrystallized, and now have well developed (001) crystal faces. Other biotite-hornblende contacts are crenellated and scalloped, clearly indicating resorption. These resorbed biotites were formerly phenocrysts up to 2 mm in size. In both appinites and the tonalites, a later generation of biotite (bt3) can be identified. This biotite has crystallized around the hornblendes containing residual biotite. A typical mode of occurrence is for later biotite to crosscut residual biotite and host hornblende prisms at an angle of near 90°. This may be simple recrystallization, but some hornblende grains are fully enclosed in biotite. In some cases the new biotite extends well into the groundmass, and may be included within a single crystal of groundmass plagioclase. This latter observation indicates that even the growth of the second generation of biotite predated much of the precipitation of plagioclase. The new biotites commonly are associated with fine quartz blebs, and this may indicate that this biotite originated by the reation hb-^bt3+quartz. In the most felsic granitoids, without intervening growth of hornblende, two generations of biotite cannot be identified. An observation common to all rocks is that there is a stage of hornblende growth that postdates biotite growth. The relative stability of biotite and hornblende can be related to intensive variables or composition in all rocks. The change from biotite to hornblende crystallization across all rock type reflects a change in these conditions. Further crystallization precipitated the younger biotite3 generation. The growth of px-btl clots reflects lower fH 2 0 and/or higher f 0 2 and fK 2 0, relative to the growth of amphibole clots. Thus, the renewed stability may indicate a return to more hydrous (and hence lower f 0 2 and fK 2 0) conditions in the associated melt. In bulk compositional terms, biotite is favoured by higher Si0 2 and K 2 0, and lower CaO. Total pressure also weakly favours hornblende over biotite. Together, the details of Strontian's microstructures reveal a complex history of crystallization, mixing and replacement. Plagioclase can also show great microstructural detail of a magma's life history. In the Santa Olalla tonalites, acid attach and Nomarski imagery (highlighting the microtopography produced by dissolution: Stamatopoulou-Seymour et al., Contributions to Mineralogy and Petrology, 104, 73-84, 1990). Very fine detail is revealed on plagioclase zoning, and the repeated truncation and dissolution of these microchemical stratigraphies. These multiple dissolution events may be attributed to magma composition change during magma mixing, or melting during decompression. In the Scottish rocks, there is a limited amount of solid-state deformation. Mineralogically, there is a great similarity with the igneous microstructures, but microstructurally very different, and very different in meaning. time late crystallization GRANODIORITE

pxl+M btl+M STciot px2+M -*-act-»- hb+M+bt2 Amph clot

TONALITE

APPINITE

plag+qz+K

M->- px3

xenocryst cumulates

plag+qz

px3+M

bt(Bt & Amph clots)+M* hb+ti+M bt3+qz)

HYBRIDIZATION

px: pyroxene hb: hornblende act: actinolite bt biotite ti: titanite M: melt M*:meit aftrer hybridization Bt biotite in clots Amph: amphibole in clots

Plag

GEODYNAMICAL MODELS OF THE TRANSITION FROM SUBDUCTION TO COLLISION TECTONICS Christopher Beaumont1, Susan Ellis1, & Adrian Pfiffner2 Oceanography Department, Dalhousie University, Halifax, Nova Scotia B3H4J1 Canada Geological Institute, University of Bern, Baltzerstr. 7, CH-3012 Bern, Switzerland

The subduction and collision phases of convergent margins may be investigated using geodynamical models of crustal deformation. Using finite element techniques, we explore the subduction/collision process for a case driven by two distinct, superimposed phases of relative convergence between the underlying mantle plates. We describe how the basally-driven (mantle subduction) model of crustal-scale collisional orogenesis (fig. l(i) boundary conditions, Willett et aL, 1993, Geology, 21, 371-374) can be expanded to include both a precursor phase of oceanic subduction (fig. l(ii) boundary conditions) and the collisional phase. The basic model (fig. 1) assumes a crustal layer with uniform properties, strongly coupled to the underlying mantle lithosphere in what are interpreted to be oceanic and continental regions. In modelling, the pro-lithosphere is initially flexed down under slab pull, L (fig. l(ii)), and the pro-crust subducts beneath the retro-crust. Later, L is progressively reduced to determine how crustal deformation changes when slab buoyancy increases as, i) L = 0 Configuration, Collision for example, during slab breakoff or attempted continental Pro(a r o 3 )/2 Retro - x = S _ -o subduction. Model results (fig. 2) show a three stage evolution: \ Wet Feldspar" \ 1 / Coulomb Plastic Stage 1 (creation of subduction \ Rheology N^ Z 0= 15" shear zone); Stage 2 (flexural rebound, pro-crust detachment, -- • i ¥ • " TT Moho -El. tectonic underplating of the shear vx = vp vx = o zone/suture, and initiation of x<s vz=o retro-thrusting (retro-charriage); x>S, z = h and Stage 3 (development of Implied Mantle doubly-vergent orogen, thrusting Subduction and exhumation (when eroded) of suture in hanging wall of the retoshear). The principal control on styles of deformation is the ii) L > 0 Configuration, Oceanic Subduction buoyancy of the pro-lithosphere.

x <S

Jt (Subduction Load) / { ^ p l i e d Mantle Subduction

Figure 3 shows formation of a composite tectonic wedge when a potential detachment horizon is included in the pro-crust. The wedge has two regions, an outer accretionary-style contractional pro-wedge, and an inner extensional wedge sandwiched between the suture with the retrocrust and the crustal detachment horizon.

Material above the detachment is first accreted, technically Figure 1 shortened, and stacked in the

outer wedge, then extended and thinned when funnelled into the subduction zone through the inner wedge. The vertical extension associated with the shortened outer wedge contributes to the creation of the topographic surface slope necessary for the formation of a critical taper geometry. We interpret this composite wedge to be fundamental to the tectonics of Alpine (Penninic-style) nappes. Nappes formed in this way may be subducted or retained in the wedge depending on the exit mass flux at its base. Material in the pro-lithosphere beneath the detachment moves steadily toward the subduction zone and is eventually subducted (fig. 3). This model thus also explains how in an orogen, crustal rocks originally almost adjacent to each other end up in very different structural positions. STAGE 1 : Subduction of Oceanic Lithosphere Retro-

STAGE 2 (cont.): End Partial Subduction of ^ _ Continental Margin Retro-thrusting

Increased Buoyancy

Negative Buoyancy

i i STAGE 2 : Partial Subduction of Continental Margin

S T A G E

(Obduction)

3 :C o U i s i o n a l

Increased Buoyancy

Retro-wedge

ac^ot

Suture Zone

r e

Pro-wedge

otv

Constant or Zero Buoyancy Force Figure 2 flattening of nappes

9 Figure 3

nappe stack formation

•I I \ \ \ \ V \ \ \ \ \ 7\ /\ /\ /N /\ /\ /N / \ / \ / \ xj>x\ /\x/\ x\• \x•/Nx\••••XXX /\ .\ /\ /N/ \/ \/ / /

ROTATED STAUROLITE PORPHYROBLASTS IN THE LITTLETON SCHIST AT BOLTON CONNECTICUT: TRUE OR FALSE? T.H. Belli and J. Wang* 1

Department of Earth Sciences, James Cook University, Townsville, Q. 4811, Australia department of Geology, Shanxi mining Institute, Taiyuan, Shanxi 0300242, China The axes of spiral, sigmoidal and other more complex inclusion trail geometries in garnet and staurolite porphyroblasts have different trends but do not change orientation across the Bolton Syncline. This indicates that neither mineral phase rotated during formation of this fold, or the main schistosity in these rocks, contrary to previous claims. Generally poorly developed inclusion trails in garnet porphyroblasts in these rocks show no relationship to the matrix, indicating that growth of this mineral occurred before the formation of the staurolite porphyroblasts. Three different orientations for the axes of spiral and other inclusion trail geometries are preserved within these garnet porphyroblasts. All samples contain one of these axes which is that associated with the youngest period of garnet growth. Three rocks contain evidence for a differently oriented older axis in the porphyroblast cores. The axes of sigmoidal, spiral and other complex inclusion trail geometries preserved in staurolite porphyroblasts are also constantly oriented across the Bolton Syncline with an orientation lying at some 500 to the youngest garnet axes. These porphyroblasts commonly show a reversal in the spiral asymmetry from core to rim about the same axis. The apparent rotation suggested previously for these porphyroblasts resulted from incorrect determination of the orientation of the spiral axes preserved in single staurolite crystals. Spiral axes cannot be determined from the distribution of poles to ellipsoidal faces of quartz inclusion trails when the latter are near cigar shaped in 3-D.

PALAEOZOIC TECTONICS OF WESTERN TASMANIA R.F. Berry Geology Dept., University of Tasmania, GPO Box 252C, Hobart, Tas 7001, Australia The structural history of western Tasmania is dominated by non-cylindrical folding with strong evidence of multiple reactivation of early fold and fault structures. This complexity has led to a plethora of local unconformities and apparently contradictory structural relationships. Many of the fold trends that have been considered as Devonian structures are open Cambrian folds which have been tightened in the Devonian. The passive margin of eastern Australia collided with an oceanic arc, and major slices of forearc lithologies were thrust over Tasmania in the early Middle Cambrian. Allochthonous blocks are the mafic/ultramafic complexes (Berry & Crawford 1988), the Forth Metamorphic Complex, the Badger Head M.C. and other small metamorphic inliers (Woodward et al 1993). Other possible allochthonous blocks are the high grade metasediments of the western part of the Tyennan Complex and the Bowry Formation. The early part of the ophiolite emplacement, as recorded in high temperature mylonites, involved thrusting towards the SW. Thrusting to the S, recorded in widespread cataclasites in western Tasmania (Findlay & Brown 1992), represents the later stages of obduction. The ophiolite emplacement coincides with the rapid sag phase which initiated the Kanmantoo cycle of deposition. A possible interpretation is that tectonic loading of the continental margin produced the new cycle of active deposition in a foreland basin setting (Turner et al 1994, Coney et al. 1990). The Delamerian Orogeny in Tasmania is a complex event with rapidly changing stress patterns. Deposition continued throughout this event but the locus and style of deposition change very rapidly. The geometry of the associated folding has more in common with the Delamerian folding in the Olary region than with the intense high temperature history in the Kanmantoo region. Part 1 Middle Middle Cambrian-extensional phase: The first event is a middle Cambrian extensional phase with rapid subsidence, active syn-orogenic deposition and major postcollisional felsic-dominated volcanism. Extension stresses are recorded in the hydrothermal vein geometry and the Henty dyke swarm. The extension was approximately east west. Part 2 Late Middle to early Late Cambrian: The extensional phase was closely followed, and may overlap with, a NS compressional event which has produced E-W trending folds in the Fossey Mountain Trough, at Adamsfield and near Bathurst Harbour. Part 3 Late Cambrian: The last phase of the Delamerian orogeny reactivated earlier extensional faults as reverse faults (Henty Fault). Major reverse faults and upright open N-trending folds were formed in western Tasmania. This phase also caused major uplift of the Tyennan block with syn-orogenic sediments (Owen Conglomerate) accumulating in synclinal cores. Active tectonism ceased in the latest Cambrian and this was followed by an Ordovician Silurian sag phase. The Silurian deformation phase recognized in eastern Victoria and further north was only represented as a depositional hiatus. During the Early Devonian shelf sedimentation continued. The Devonian deformation throughout Tasmania is characterized by the complexity of fold orientations (e.g. Williams 1989. In many areas, the fold geometry is controlled by the preexisting Cambrian fold trends which were tightened during the Devonian. This led to Devonian cleavage orientations which transect the folds with which they are associated. In the Fossey Mountain 'trough E-W Cambrian folds are tightened. In the Dundas Trough, Ntrending Cambrian folds are tightened with an associated NNW-striking Devonian cleavage. NNE-trending folds north of Tullah are controlled by the reactivation of the Henty Fault. Most western Tasmania granites were intruded late syn- to post- kinematic with respect to the NNW-trending cleavage and associated folding and faulting. 1

The subsequent N- to NNE-trending compression produced WNW-trending folds and thrusts in the south, and NW-striking thrusts and associated folds in the north. On the west coast, there is an associated phase of brittle wrench faults dominated by NNE-striking sinistral movement on the Henty Fault. A return to E-W compression is found at a few isolated localities. Regional metamorphism associated with the Devonian orogeny was prehnite -pumpellyite grade with local zones of greenschist facies in the vicinity of late syn- to post-orogenic granites. References Berry RJF. & Crawford AJ. 1988. The tectonic significance of Cambrian allochthonous mafic-ultramafic complexes in Tasmania. Australian Journal of Earth Sciences, 35: 523-533. Coney PJ., Edwards A., Hine R., Morrison F. & Windrim D. 1990. The regional tectonics of the Tasman orogenic system, eastern Australia. J. Struct. Geol. 12,519-543. Findlay R.H. & Brown A.V. 1992. The 10th Legion Thrust, Zeehan district. Distribution, interpretation, and regional and economic significance. Mineral Resources Tasmania unpublished report 1992/02,27pp. Turner N .J., Bottrill R.S., Crawford A J. & Villa 1.1992. Geology and prospectivity of the Arthur Mobile Belt Bull. geol. Surv. Tasm. 70,226-232 Turner S„ Sandiford M., Flottman T. & Foden J. 1994. Rb/Sr dating of differentiated cleavagefromthe upper Adelaidean metasediments at Hallett Cove, southern Adelaide fold belt J. Structural Geology 16,1233-1241. Williams E. 1989. Mid Palaeozoic deformation, in Burrett, CP. and Martin, E.L. (Eds.). Geology and Mineral Resources of Tasmania Special Publication of the Geological Society of Australia 15,239-253. Woodward N.B., Gray DR. & Elliott C.G. 1993. Repeated Palaeozoic thrusting and allochthoneity of Precambrian basement, northern Tasmania. Aust J. Earth Sci. 40,297-311.

THE EARLY TECTONIC HISTORY OF THE FIERY CREEK DOME REGION IN NORTHERN MOUNT ISA INLIER Peter Betts & Rick Valenta Department of Earth Sciences, Monash University, Clayton, Victoria 3168, Australia Structures and geometries of the Mount Isa Inlier are dominated by a series of Middle Proterozoic N-S and E-W shortening events termed the Isan Orogeny. In many parts of the Inlier these structures have almost completely obscured earlier rift related structures associated with the deposition of Cover Sequence 2 (1810-1740 Ma) and Cover Sequence 3 (1680-1590 Ma). Isan Orogeny structures are less intense in the Fiery Creek Dome region than many other portions of inlier providing a geological window where earlier structures can be assessed. The surface geology of the study area is comprised of dome and basin outcrop pattern. This pattern reflects a combination of basement geometry, laccolithic emplacement prior to the deposition of Cover Sequence 3 at approximately 1700 Ma (age of the Weberra granite) and shortening during the Isan Orogeny. The Surprise Creek unconformity defines the boundary between Cover Sequence 2 and Cover Sequence 3. Reconstruction of post-unconformity structures throughout the entire Western Succession generally indicates a low angular unconformity associated with either eastward or northward tilting of the Cover Sequence 2. Examination of the Mellish Park Syncline above and below the Surprise Creek Unconformity indicates tighter folding below the unconformity. A high bedding angle between Cover Sequence 2 and Cover Sequence 3 on the western limb of the Mellish Park Syncline and the variation in the fold geometry above and below the Surprise Creek Unconformity suggests: a) localised preIsan folding during the emplacement of a pluton which cores the dome or b) fold buttressing against the pluton coupled with slip along the unconformity during Isan Orogeny shortening. An inferred NW-SE direction of extension during the deposition of Cover Sequence 3 is based on detailed mapping along the Fiery Creek Fault. The syn-sedimentary Fiery Creek Fault is a NE oriented, steeply NW dipping regional scale structure that has a complex movement history. The Fiery Creek Fault was active during the deposition of the Surprise Creek Formation during NW-SE extension. Evidence includes thickening and coarsening of sediments in the hanging wall of the Fiery Creek Fault and absence of stratigraphic units across the fault. Structural inversion of the fault during later shortening has resulted in variable dip-slip movement along the fault. Relative offset across the fault changes from normal-sense in the northeast to reverse in the southwest. This geometry is due to tectonic inversion of the earlier normal fault during Isan Orogeny shortening. Enhancement of near surface magnetic anomalies sourced from the Fiery Creek Volcanics and magnetic modelling over the southern portion of the Fiery Creek Fault suggest that the fault has a normal dip-slip movement in the subsurface. This contradiction between the sense of movement at the surface and in the subsurface indicates an increase in the thickness of the Surprise Creek Formation in the hanging wall proximal to the Fiery Creek Fault.

EMPLACEMENT, TIMING, AND GENESIS OF THE MABEL DOWNS GRANODIORITE IN THE HALLS CREEK OROGEN, EAST KIMBERLEY S. Bodorkos & N.H.S. Oliver School of Applied Geology, Curtin University, GPO Box U1987, Perth, W.A. 6001 Igneous rocks of most northern Australian Proterozoic terrains are dominated by a bimodal distribution of granitic and basaltic compositions, lending weight to the model of Etheridge et al. (1987) that proposes that these belts were generated by intracratonic reworking of older crustal materials. In modern collisional settings (especially those involving subduction of oceanic crust) the abundance of andesites and their intrusive equivalents is commonly interpreted as evidence for the contamination of melt source materials with subducted, hydrated slab and wedge material. The Halls Creek Orogen thus provides considerable interest as a possible Proterozoic plate margin because it contains widespread intrusives of granodioritic to tonalitic composition, which could possibly be derived by melting of contaminated subducted slab material in a collisional setting analogous to modern plate tectonic margins (Ogasawara 1988). Recent work by GSWA also highlights cross-orogen disparities in timing and nature of sedimentation, deformation and metamorphism (Tyler et al 1994), and igneous activity (Page & Sun 1994) which could support a collisional, plate-margin origin for the orogen. The Mabel Downs Granodiorite is the focus of this study, as one of the largest plutons in the Halls Creek Orogen with an "unusual" composition. There is some debate over the relative timing of this body, and finalized geochronological results (by R.W. Page at AGSO) have yet to be released. The aim of this study is to examine the field relations, petrology, whole rock geochemistry and radiogenic isotope systematics of the pluton in order to determine its genesis, timing and relationships to metamorphism and deformation, and also to assess whether the pluton could have been generated in a collisional plate boundary setting. This part of the Kimberley also provides several very well exposed examples of magma mingling and also syntectonic magma emplacement. Mixing of mafic and felsic compositions to produce intermediate magmas is observed at local scales, and it is possible that this process on a larger scale at depth has contributed to some of the bulk compositional characteristics of the Mabel Downs intrusion. The pluton is sill-like with respect to external markers, but defining its margins is difficult due to interfingering with the wallrocks at a scale of 10-100 metres over distances of up to 500 metres. In many exposures of the body, particularly (but not exclusively) towards its margins, there is a wide range of compositions present ranging from adamellite to gabbro. Mutual crosscutting patterns and relationships with metamorphic assemblages and deformation fabrics suggest that emplacement of the different phases occurred over a relatively short time period. Key outcrops contain strong evidence suggesting that at least part of the body was emplaced synchronous with two major deformation phases, possibly closely spaced in time (D2 and D3 in regional terminology). This evidence takes the form of melts of similar composition displaying a range of relationships with respect to the established (relative) deformational chronology. For example, melts of granodioritic composition have been found as axial planar features synchronous with F4 folds, as flow-foliated magma lying in F3 hinges and shear zones, and as large enclaves containing a regional S2 fabric. Mingled mafic-felsic magmas occur pre-, syn- and post-D3. Recent AGSO geochronological results (AGSO newsletter) from leucocratic phases of nearby major mafic intrusions suggests two major phases of melt emplacement at 1850 and 1830 Ma to the south and west of the study area. Although confirmation of absolute timing of the different phases of the Mabel Downs body awaits detailed geochronology, it is thought that: 1) the pluton was emplaced synchronous with a major deformation event, and possibly spanned the development of two or three short-lived deformation phases during regional low pressure metamorphism; and 14

2) mafic, intermediate and felsic melts were emplaced simultaneously, and both mingling and mixing occurred during this time.

References Etheridge M.A., Rutland R.W.R. & Wyborn L.A.I. 1987. Orogenesis and tectonic processes in the Early to Middle Proterozoic of northern Australia. American Geophysical Union, Geodynamics Series 17, 131-147. Ogasawara M. 1988. Geochemistry of the Early Proterozoic granitoids in the Halls Creek orogenic subprovince, northern Australia. Precambrian Research 40/41,469-486. Page R.W. & Sun S. 1994. Evolution of the Kimberley region, W. A. and adjacent Proterozoic Inliers - new geochronological constraints. Geological Society of Australia Abstracts 37, 332-333. Tyler I.M., Griffin T.J., Page R.W. & Shaw R.D. 1994. The Halls Creek Fault System: repeated reactivation of a major tectonic boundary within the north Australian craton. Geological Society of Australia Abstracts 36, 167168.

NUMERICAL MODELS OF CRUSTAL DEFORMATION APPLIED TO THE ALICE SPRINGS OROGENY OF CENTRAL AUSTRALIA J. Braun Research School of Earth Sciences, Australian National University, Canberra, ACT 0200, Australia The Alice Springs Orogeny of Central Australia is a well-documented episode of continental shortening. This Devono-Carboniferous tectonic event led to the formation of crustal-scale structures that are still observable in the present-day gravity field (Figure 1). The magnitude of the gravity anomalies as well as reflection seismic images of the lithosphere indicate that a major thrust fault cuts through the Moho. The nature of the tectonic process that led to crustal shortening in Central Australia remains however a mistery. Two "classes" of mechanisms are possible: 1) those active along plate boundaries producing compressional stresses that are transmitted inside the plate interiors and re-activate pre-existing weaknesses, and 2) those arising in the mantle/asthenosphere directly beneath the continent.

Figure 1: Artificially illuminated map of the Bouger gravity anomaly in Central Australia

We present the results of a simple numerical model in which a mechanism of the second type is assumed to drive continental deformation: the mantle lithosphere underneath the Arunta Block to the north of the Redbank Deformed Zone (RDZ) is assumed to "subduct" beneath the mantle underlying the Amadeus Basin (Figure 2). In the model, the lithosphere is regarded as a viscoelasto-plastic layer floating on an inviscid asthenosphere. The model predicts the formation of a narrow, southward dipping shear zone/fault that accommodates thrusting of the northern Arunta Block over the southern crustal block underlying the Amadeus Basin (Figure 2a). The fault offsets the Moho to produce a structure very similar to the one observed in a deep reflection seismic survey across the RDZ (Figure 2b from Shaw et al., 1991, BMR Bull. 236, 385-408) and confirmed by a study of seismic travel time anomalies (Lambeck, 1991, BMR Bull. 236, 409-427). Following thusting and thickening, the lithosphere subsides on both sides of the orogen and a deep foreland basin develops, the geometry of which is very similar to the Devonian section of the Amadeus Basin. In the later stages of deformation, movement along the crustal fault is hindered by the increased pressure due to the finite crustal thickening. This leads to a widening of the zone of deformation by the development of secondary faults and a crustal-scale fold in the footwall beneath the foreland basin (Figure 2a). These structures are very similar to the Ormiston thrust zone and MacDonnell Homocline along the northern margin of die Amadeus basin (Figure 2b).

Sedimentary basin

Erosion surface

High strain shear zone

20km

Sub-continental mantle "subduction"

Southern Northern Amadeus i D A r u P t a Basin | Province Q MacDonnell Homocline

Central Arunta Province

Northern Arunta Province Ngalia Basin

Ormiston Thrust Zone Redbank Thrust Zone Devonian sediments Pre-ASO sediments Mylonite zone

Lower amphibolite Moho? |Amphibolite Upper amphibolite Upper amphibolite - granulite Granulite with amphibolite regression Granulite

20km

Figure 2: a) Numerical model results and b) baserved structures and distribution of metamorphic rocks

The model also predicts the "original depth" of rocks that have been brought to the surface by denudation following thrusting and erosion (Figure 2a). These predictions are compared to the observed metamorphic grade of rocks in the central and northern Arunta Province (Figure 2b, from Shaw, 1987, ANU PhD Thesis). The model predicts that the metamorphic grade of rocks should increase rapidly from the edge of the foreland basin towards the main thrust fault. The peak in metamorphic grade is reached at some distance to the north of where the fault outcrops at the surface. This pattern in the distribution of metamorphic rocks is very similar to that observed in central Australia across the RDZ.

FEEDBACK MECHANISMS AND CHANNELIZED MAGMA TRANSFER Michael Brown Department of Geology, University of Maryland at College Park, MD 20742, USA Migmatite (mixed rock composed of granite in a metamorphic host) occurs in the higher grade parts of orogenic belts which exhibit heterogeneous deformation at all scales (Brown, 1994, Earth Science Reviews, 36, 83-130). Upper-amphibolite facies (lower T ) melt will be "wetter" than granulite facies melt (higher T)\ undersaturated melt has the potential to migrate greater distances. However, the common view that migmatite largely represents high a(H 2 0) melting for which migration was limited by the solidus is not supported by the occurrence of migmatitic structure developed by low a(H 2 0) volatile phase-absent melting in rocks characterized by extreme metamorphic conditions (800-1000°C at 0.9-1. lGPa). Understanding melt-producing reactions, changes in melt fraction with T (and P), segregation mechanisms, and conditions for magma mobility in different protolith compositions has led to better comprehension of links between anatectic migmatite (metatexite and diatexite) and granite (e.g. Brown, Rushmer & Sawyer, 1995, Journal of Geophysical Research, 100, in press). Many granites are located in orogenic belts: associated with high-grade metamorphic rocks; emplaced synkinematically along shear zones during crustal contraction and extension; and, emplaced in lower-grade metamorphic rocks at higher structural levels. Q.E.D. the crust may be considered as migmatitic at all scales! The plumbing by which melt segregates and transfers through the crust is marked by magma that became stuck in the system. Outcrop-scale information in migmatites indicates the importance of deformation to drive melt segregation and provide sites for melt accumulation. The same features occur at map-scale in orogenic belts, where regional tectonic structures may control location of granite plutons in the brittle upper crust, and lower-grade units are separated from higher-grade migmatitic crust by shear zones that represent crustal-scale decollements located along Theologically weak units. Melting at high a(H20) involves volume reduction, segregation may be realized by convection and filter pressing, and melt retention is expected, since the system tends to "suck in" melt (Brown, Averkin, McLellan & Sawer, 1995, Journal of Geophysical Research, 100, in press). Melt fraction may increase until the contiguity of the solid framework breaks down and bulk magma mobility with entrainment of residual material takes place, unless deformation drives out melt at lower melt fractions. By contrast, major crustal melting involves hydrate breakdown under low a(H20) volatile phase-absent conditions, which leads to an increase in volume and segregation by convection and filter pressing. Increasing volume may lead to pore fluid pressures high enough to facilitate melt-enhanced embrittlement, cataclastic flow and fracture to allow melt escape; effectively the system "blows out" melt (Brown, Averkin, McLellan & Sawyer, 1995, Journal of Geophysical Research, 100, in press). There are three end-member models by which magma might ascend through the crust to its level of emplacement, vzz. as a diapir, as a dyke or along a shear zone. The water content of granite melt can vary from < 3 wt% (strongly water undersaturated) to > 10 wt% ("very wet"), which has a significant effect on the viscosity of granite magma. Because most melt transfer properties are inversely dependent on viscosity, whether granite magma ascends through the crust as bodies closer to the dyke or diapir end-member may resolve to a better knowledge of viscosities of the magmas concerned. The application of differential stress to an anisotropic body such as the crust, in particular when it contains melt, leads to a strongly 18

heterogeneous strain field. Melt segregation and transfer likely is scale-invariant. Melt may migrate through structurally-created pathways, such as fractures and shear zones, accumulate in a variety of dilatant sites, such as boudin necks, or immigrate as discordant bodies of leucosome, all of which demonstrate the important role played by deformation in the extraction of granitic melt. Melt will flow to low-pressure regions. Migration of magma along major crustal shear zones, particularly transcurrent shear zones, appears fundamental in some orogenic settings. The buoyancy force of the magma then will interact with regional stresses in the most convenient way to facilitate ascent. Thus, shear zones may be expected to localize the sites of magma ascent, although the mechanism of ascent still may be buoyancy-dominated (diapiric), and buoyancy forces may cause some deformation of the wall rocks. Zones dominated by buoyant magma flow may be characterized by a steep magmatic lineation, whereas zones characterized by strike-slip shearing may record sub-horizontal magmatic lineations. One consequence of an anisotropically layered crust is that melting in one zone may be capped by an unmolten layer above. If the partially molten zone develops instabilities due to gravity, then melt might be expected to migrate parallel to the layering and accumulate at dome culminations. If shear zones preferentially nucleate at these sites, then magma ascent within shear zones may be realized. Additionally, crustal-scale pinch and swell or boudinage of Theologically stronger units may take place, which may allow upward movement of weaker migmatitic crust that also may facilitate magma segregation, accumulation and ascent. Anatexis has a significant implication for crustal rheology and it is clear that there are feedback mechanisms between deformation and melting. Because the effective viscosity of non-linear materials is strain-rate dependent, strain concentration in anatectic domains enhances contrasts in viscosity, facilitates melt segregation and ascent, and partitions deformation preferentially back into the zones in which partial melting has occurred. This feedback loop intensifies the perturbation, induces complex kinematic patterns and promotes the formation of shear zone systems. During ductile transpression, partitioning of shear strain between coaxial and noncoaxial components takes place, and linked systems of transcurrent shear and subhorizontal decollement zones are developed. An example of middle crust that preserves evidence of ductile transpression synchronous with melting is the St. Malo migmatite belt, France. Here melt has frozen during transfer along an interconnected system of shallow-oriented shear zones within the migmatite belt, and marginal to the migmatite belt larger volumes of magma have been frozen during ascent within major strike-slip shear zones. In this case, magma ascent through the crust from the zone of generation to the level of emplacement is inferred to have been channelized, having occurred through a small number of major strike-slip shear zones. Furthermore, anatexis was important in enabling the transpressive deformation to propagate. The strike-slip shear zones themselves were distorting, and as bends in the shear zones amplified so the resultant geometry became increasingly inefficient for transcurrent displacement. This kinematic inefficiency likely was counterbalanced by volume loss due to deformation-enhanced melt segregation and transfer out of the system. Although the volume of magma that has migrated out of the migmatite belt is unknown, it was this significant volume loss that facilitated contractional deformation during transpression and propagation of the deformation front. Adjacent to the migmatite belt at a higher structural level the Mancellian granites are interpreted to represent emplacement into the upper crust of fugitive magma from migmatitic rocks analogous to the St. Malo migmatite belt at deeper structural levels. Ascent is inferred to have been channelized with emplacement into tensile bridges developed between left-stepping segments of a major transcurrent fault zone. Along individual granite pluton/country rock contacts, however, local space creation by stoping clearly was the method by which emplacement was completed. 19

THE AR/ AR THERMOCHRONOLOGY OF GRANITES FROM THE WESTERN LACHLAN FOLD BELT, SOUTHEASTERN AUSTRALIA: IMPLICATIONS FOR TECTONIC DEVELOPMENT M. Bucher. D.A. Foster & A.J.W. Gleadow School of Earth Sciences, La Trobe University, Bundoora, VIC 3083, Australia The general lack of medium to high grade regional metamorphic rocks within the southern part of the Tasman orogenic belt on mainland Australia provides an opportunity to apply thermochronologic methods on igneous rocks to investigate the state of the crust at the time of formation and their respective cooling paths thereafter. The Lachlan Fold Belt of southeastern Australia contains voluminous mainly post-tectonic granites and related volcanic rocks. The geology of the western Lachlan Fold Belt is dominated by a thick succession of Lower Palaeozoic submarine fan deposits which form the framework for these igneous rocks. The crystallization ages of these plutons, previously studied by Bt[±Ms±Hbl] K-Ar data (Richards, & Singleton, 1981), point to most of the granites being emplaced within two restricted periods of time. Additional Ar- Ar laser fusion and laser step-heating experiments on Bt and Ms confirmed this range in age across the basement terrains from Late Silurian/Early Devonian [-415-385 Ma] in the west of the state of Victoria to Middle/Late Devonian [-375-355 Ma] in the area around Melbourne. A small number of plutons in the western Lachlan Fold Belt defy the general trend and give CambroOrdovician ^Ar-^Ar biotite ages of -500 Ma. These ages are chronologically more similar to plutons within the Delamerian orogen to the west in western-most Victoria and southeastern South Australia. Sixteen alkali feldspar [Ksp] samples from granites covering an area from the Gariwerd Mts. [Grampians] in the west to Tynong in the east, and Mt. Jeffcott in the north to Mt. Eliza in the south, have been analyzed using the Ar- Ar incremental heating technique. The Ksp ages of these plutons, given by high-temperature plateaus in the age spectra, are always younger than their respective Bt and/or Ms ages and decrease from the Grampians-Stavely terrane (Chappell et al., 1988) in the west [MacKenzie River: 397±2 (±2a) Ma; Hopkins River: 393±1 Ma] to the central Stawell terrane [Stawell: 399±4 Ma; Mt. Cole: 388±1 Ma; Mt. Jeffcott: 388±3 Ma] and finally to the eastern Melbourne terrane [Lismore: 359±2 Ma; Wallinduc: 371±2 Ma; Harcourt: 364±1 Ma; You Yangs: 364±3 Ma; Crosbie: 369±3 Ma; Mt. Black: 358±2 Ma; Glenaroua: 377±1 Ma; King Parrot Creek: 373±3 Ma; Lysterfield: 371±2 Ma; Tynong: 369±3; Mt. Eliza: 350+1 Ma]. These data reflect the established K-Ar and fission track age patterns on micas (Richards & Singleton, 1981) and sphene and apatite (Gleadow & Lovering, 1978; Foster & Gleadow, 1992) of younging of the plutonic activity from west to east. The inspection of the individual argon release spectra reveals that [i] Ksp is less retentive with respect to radiogenic Ar than Bt and/or Ms, [ii] all samples cooled relatively quickly and with only minor radiogenic Ar loss after Ksp closed its Ar system. The youngest reliable ages in the individual age spectra defining an upper age limit where diffusive loss in Ksp became insignificant range across the basement terranes from -300 Ma in the west to -280-330 Ma and, in the east, to -300-335 Ma. These values indicate that cooling of the individual plutons from magmatic temperatures at the time of their formation to the minimum closure temperature of alkali feldspar [-170-200 °C for low-temperature domains] lasted -100 Ma in the west, -50-90 Ma in the central area, and only about 30-50 Ma in the east confirming the rise of the crustal level in which these granites intruded. This thermochronology indicates that the plutons were the final vestiges of orogenic activity in the terranes they intruded and that deformation migrated eastward through time within the western Lachlan Fold Belt. Apatite fission track data (Gleadow & Lovering, 1978; Foster & Gleadow, 1992) from Palaeozoic rocks along parts of the southeastern continental margin of Australia suggest that final cooling of the basement terranes below -110 °C occurred during Cretaceous time. This reflects reheating followed by denundation concentrated near the northern margin of the sed40

imentary basins during the rifting of Australia and Antarctica. The Ar- Ar systematics of alkali feldspar from several granites along the southeastern Australian continental margin do not reveal any radiogenic argon loss at this time. Prevailing temperatures during rifting were sufficient enough to disturb the apatite fission track systems, but did not affect alkali feldspar, indicating that temperatures remained below -130 °C. References 40

Richards, J.R. & Singleton, O.P., 1981. Palaeozoic Victoria, Australia:-igneous rocks, ages and their interpretation. Journal of the Geological Society of Australia 28, 395-421. Chappell, B.W., White, A.J.R., & Hine, R., 1988. Granite provinces and basement terranes in the Lachlan Fold Belt, southeastern Australia. Australian Journal of Earth Sciences 35, 505-521. Gleadow, A.J.W & Lovering, J.F., 1978. Thermal history of granitic rocks from western Victoria: a fission track dating study. Journal of the Geological Society of Australia 25, 323-340. Foster, D.A. & Gleadow, A.J.W., 1992. Reactivated tectonic boundaries and implications for the reconstruction of southeastern Australia and northern Victoria Land, Antarctica. Geology 20, 267-270.

SHRIMP CONSTRAINTS ON THE TIMING OF GRANULITE-FACIES METAMORPHISM AND DEFORMATION IN THE REYNOLDS RANGE, CENTRAL AUSTRALIA: IMPLICATIONS FOR THE TECTONICS OF LOWPRESSURE/HIGH-TEMPERATURE METAMORPHISM. Ian S. Buick1. Ian S. Williams2 & Ian Cartwright3 School of Earth Sciences, La Trobe University, Bundoora, Vic. 3083, Australia. Research School of Earth Sciences, Australian National University, Canberra, ACT 0200, Australia. 3 Department of Earth Sciences, Monash University, Clayton, Vic, 3168, Australia. 2

Studies of regionally-extensive low-pressure/high-temperature metamorphism of amphibolitefacies grade commonly indicate a link between local metamorphic highs and close proximity to intrusive rocks of broadly granitic composition. This may indicate that batholithic granites provide the heat source and therefore drive both regional metamorphism and deformation in these highly thermally perturbed terrains (e.g. Lux et al, 1986; Nature, 323, 794-797). The results of thermal modelling suggest that granite-driven low-pressure/high-temperature metamorphism occurs as events of typically short duration (<1-10 Ma; Rothstein & Hoisch, Journal of Metamorphic Geology, 12, 723-734; Sandiford et al., 1995, Australian Journal of Earth Sciences, 42, 233-240), and that the metamorphic peak is followed by rapid cooling at aproximately constant pressure (>10's to 100's °C/Ma; Rothstein & Hoisch, Journal of Metamorphic Geology, 12, 723-734). This is in accord with thermochronological studies of some (Foster et al., 1992, Geological Society of America bulletin, 104, 176-191), but not all such terrains (Bowring, & Karlstrom, 1990, Geology, 18, 1203-1206). Recently, a similar model has been invoked to account for temporal and spatial relationships between metamorphism, deformation and granite intrusion during low-pressure/high-temperature regional metamorphism of the Reynolds Range, Arunta Block, central Australia (Collins & Vernon, 1991, Geology, 19, 835-838). However, new SHRIMP age determinations presented here suggest that plutonism and regional metamorphism in the Reynolds Range were unrelated, that regional metamorphism was an extended event and that the terrain subsequently cooled slowly. The ages coupled with evidence for decompression after the peak of metamorphism imply that new models are required to account for the tectonic evolution of this terrain. Within the Reynolds Range the Lander Rock Beds and early granitic intrusions, which together form local basement, are unconfomably overlain by metamorphosed sediments of shallow marine origin (the Reynolds Range Group). The earliest metamorphism and deformation (MIDI) occured in the basement, was of low-pressure facies, and was post-dated by early granite emplacement at - 1.82 Ga (Collins & Williams, 1995, Precambrian Research, 71, 69-89). The Reynolds Range Group was deposited between - 1.82 and 1.78 Ga (Collins & Williams, 1995, Precambrian Research, 71, 69-89). A second generation of granites was emplaced into the Reynolds Range Group and the basement at -1.78 Ga (Collins & Williams, 1995, Precambrian Research, 71, 69-89) and was associated with extensive contact metamorphism. Subsequently, both the Reynolds Range Group and basement were regionally metamorphosed (M2-D2) during sub-horizontal convergent deformation. M2-D2 regional metamorphism occurred at - 4 to 5 kbar and increased in grade from greenschist- to granulite-facies (-400 to 750-800 C) along the length of the Reynolds Range (Dirks et al, 1991, Journal of Metamorphic Geology, 9, 641-661). M2 isograds cut the trend of large-scale, upright, NW-SE trending, tight D2 folds and rock units. SHRIMP age determinations on apparently metamorphic zircon from granites that intruded prior to M2 are also -1.78 Ga, which was interpreted by Collins & Williams (1995) as dating M2-D2 metamorphism and deformation, in support of a tectonic model in which regional metamorphism and deformation were driven by intrusion of numerous -1.78 Ga granite sills. In the high-grade portion of the Reynolds Range, peak-M2 granulite-facies assemblages in marbles, calcsilicate rocks and semipelites/psammites of the Reynolds Range Group were partially retrogressed in narrow (10's- 100fs x 1000fs metres) zones. These retrograde zones 22

represent channelways for high-temperature (650-700 °C) water-rich fluids derived from the crystallising M2 partial melts sourced from -1.78 Ga granites (Hand & Dirks, 1991, Journal of Structural Geology, 14, 591-604) and metapelites (Cartwright & Buick, 1995, Journal of Metamorphic Geology, 13,397-418) that segregated into conjugate crenulation sets that deform S2. Within the retrograde zones, new, commonly randomly oriented mineral assemblages overprint relic S2. The retrograde zones are characterized by resetting of mineralogy and oxygen isotope ratios, local major element metasomatism, and the emplacement of hightemperature pegmatites and quartz veins. We have dated newly formed zircon and monazite from quartz vein sets in the retrograde zones, in addition to those from an unretrogressed granulite-facies sub-aluminous metapelite that contains S2, in order to constrain the timing and duration of M2-D2 regional metamorphism. U-Th-Pb isotopic ages were measured by the SHRIMP on zircon and monazite from an unretrogressed granulite-facies sub-aluminous pelite, and both semi-concordant and highly discordant aluminous quartz-rich segregations from the retograde zones. Zircon cores from the sub-aluminous pelite show its likely protolith to be an igneous rock 1812 ± 11 Ma old, itself derived from a source containing components as old as 2.2 Ga. Low-Th/U overgrowths on the zircons grew during M2 granulite-facies metamorphism at 1594 ± 6 Ma, and are a minimum estimate for the age of S2. Monazite cooled to its blocking temperature at 1576 ± 8 Ma. Zircon cores from the semi-concordant segregation are dominantly >2.3 Ga old. Two generations of low-Th/U overgrowths on the zircons give indistinguishable ages for the older and younger of 1589 ± 8 Ma and 1582 ± 8 Ma respectively (combined age 1586 ± 5 Ma). The monazite age is the same, 1576 ± 1 2 Ma. Zircon intergrown with sillimanite from the late discordant segregation gave 1568 ± 4 Ma. The age of regional low-pressure/high-temperature metamorphism and the main fabric-forming event in the Reynolds Range Group is -1.6 Ga, not -1.78 Ga as previously thought. The age difference between granite emplacement and regional metamorphism (-180 Ma) is much greater than the typical timescale for single events in low-pressure/high-temperature orogenic belts. Therefore, regional metamorphism and deformation were not driven by the emplacement of numerous granites within the terrain. The chronology of zircon growth described above also allows us to place limits on the duration of the -1.6 Ga tectonometamorphic event. The event started at least as early as 1594 ± 6 Ma, with the formation of the zircon overgrowths in the granulite-facies sub-aluminous pelite, and continued at least until the crystallisation of the zircon in the post-S2 quartz-sillimanite segregations at 1568 ± 4 Ma. High temperatures (> 650-700 °C) were maintained for at least 26 ± 3 (a) Ma, consistent with a very slow time-integrated cooling rate of -2-6°C/Ma from peak temperatures of -800 °C. Slow cooling is also indicated by the significant age difference between zircon and monazite from the unretrogressed pelite. We interpret the -26 Ma time interval to be the minimum duration of a single high-temperature event, rather than of a series of unrelated events (e.g. thermal pulses) because the granulite-facies metamorphism and the hightemperature retrogression have a simple causual relationship. Had there been multiple thermal pulses, it seems unlikely that an early major thermal pulse would have caused peak metamorphism but no retrogression, but a later minor one released the fluids, unless the temperature had remained high in between. Both the extended duration of the M2 event and the subsequent slow high-temperature cooling history are inconsistent with the prediction of thermal models for granite-driven low-pressure metamorphism, and suggest a more long-lived heat source. Therefore the tectonic setting for M2-D2 in the Reynolds Range is unlikely to be that envisaged by Collins and Vernon (1991). Initial slow cooling from peak temperatures may also explain controversies as to the shape of the retrograde Pressure-Temperature-time (P-T-t) path in the Reynolds Range, where P-T-t segments of near-isobaric cooling (Clarke et al., 1990, Journal of Metamorphic Geology, 8, 65-88) and near-isothermal decompression (1-2 kbar; Vry & Cartwright, 1994, Contributions to Mineralogy and Petrology, 116, 78-91) have been deduced from mineral reaction textures. While retrograde P-T-t paths with a substantial cooling component are generally to be expected in low-pressure/high-temperature terrains, slow high-temperature cooling may allow decompression textures to be recorded in some rocks during minor erosionally-controlled uplift (uplift rates of 0.1-0.25 mm/yr) towards the end of convergent deformation. 23

COOLING AND UPLIFT RATES OF GRANULITE-FACIES ROCKS FROM THE RAUER GROUP, ANTARCTICA: CONSTRAINTS FROM OXYGEN ISOTOPES IN MARBLES AND IMPLICATIONS FOR TECTONIC PROCESSES I. Cartwrightl. S.L. Harley2, IS. Buick3 & T. Barr1 1

Department of Earth Sciences & VIEPS, Monash University, Clayton, Vic. 3168 Australia Grant Institute of Geology, University of Edinburgh, West Mains Road, Edinburgh EH9 3JW, UK 3 School of Earth Sciences & VIEPS, LaTrobe University, Bundoora, Vic. 3083, Australia

One of the primary requirements in understanding tectonometamorphic processes is documenting the timescales over which metamorphism occurs. Slow rates of uplift and cooling following the peak of metamorphism may indicate that the processes that caused metamorphism occurred over long time periods and affected significant volumes of the crust. Crustal thickening and subsequent exhumation is one such long-lived process. By contrast, rapid cooling from the peak of metamorphism may imply that metamorphism was the result of processes that operate over shorter timescales (e.g. advective heat transfer), which often affect smaller crustal areas. Methods of estimating the timescales of metamorphism include: 1) cooling rates calculated from cation diffusion between coexisting minerals; 2) direct radiometric dating of discrete tectonometamorphic events; 3) dating of cores and rims of metamorphic minerals that show evidence for growth along a well-defined sector of a metamorphic P-T path; 4) ages yielded by minerals with differential closure temperatures; or 5) calculation of the relative rates of deformation and heat transfer. Because oxygen isotope diffusion and fractionation coefficients are known for a range of minerals, oxygen isotopes may be used as both geothermometers and geospeedometers. Where a slower diffusing mineral is contained within a much larger volume of a second mineral which has more rapid diffusion coefficients (as is often the case in marbles), the closure-temperature approach of Dodson (1973) may be used as a first approximation to calculating cooling rates. During cooling, diffusion should cause isotopic resetting over a given width at the margins of the grains; hence, smaller grains should yield lower closure temperatures. Here we use such a method to estimate the cooling rate of Rauer Group marbles. The Rauer Group, east Antarctica, is a polymetamorphic and polydeformational granulite-facies orthogneiss dominated terrain with four distinct lithological associations: 1) felsic to intermediate orthogneisses that locally preserve relict igneous textures; 2) layered paragneisses (including Ferich pelites, semipelites, quartzites, leucocratic gneisses, and calcsilicates); 3) homogeneous felsic to intermediate orthogneisses with strong fabrics that contain boudins and rafts of marble and Mgrich metapelites; and 4) layered gneisses comprising mafic granulite and minor marble that are often preserved as rafts within the orthogneisses of association 1. (Harley & Fitzsimmons, 1991; Harley & Buick, 1992; Buick et al.91993). Associations 1 and 2 were only metamorphosed by the last major metamoiphism (M3) at -1030-1000 Ma (Kinney et al., 1993) at 700-900 MPa, and 84G±40 °C. Following M3, the rocks underwent near isothermal decompression to 200-400 MPa while temperatures remained in excess of 750 °C. In areas of low D3 strain, Associations 3 and 4 preserve evidence of older tectonic fabrics and mineral assemblages of a pre-M3 metamorphic event (P = 1000-1200 MPa, T = 1000-1050°C. Overall M3 probably represents extensive Proterozoic reworking of rocks that had undergone an earlier high-temperature, high-pressure metamorphism (possibly in the Archaean). The samples from this study come from a 1.5 m wide layer of diopside + scapolite-bearing marble contained within mafic orthogneisses on Filla Island. Individual and groups of round diopside grains were separated from the marbles for oxygen isotope analysis by acid dissolution, sieving, and hand picking. There is a broad correlation between grain size and 8 18 0(Di) values. In any single sample, diopside grains with grain radii (r) > 0.045mm have relatively uniform 8 1 8 0 values. In individual samples, temperatures calculated from A180(Cc-Di) fractionations using the data of Matthews (1994) for the average 8 1 8 0 values of diopside grains of r > 0.045 mm range from 861-826°C with an overall average of 842±46°C. This is in excellent agreement with the estimated temperature of M3 metamorphism in this area (800-880°C), indicating that the coarser diopside grains at least preserve close to peak M3 metamorphic 8 1 8 0 values. The smaller diopside 24

grains yield temperatures of as low as 640-650°C, suggesting that their 8 0 values have been reset during cooling. As a first approximation closure temperatures for calcite + diopside isotopic diffusion were calculated using the method of Dodson (1973). The results suggest that cooling rates were between 1 and 10°C / Ma. These estimates are not significantly altered by adopting numerical diffusion models that take into account variable cooling rates and avoid the drawback that Dodson's equations assume: 1) an infinite reservoir; and 2) isotopic fractionations that are proportional to 1/T (whereas oxygen-isotope fractionations generally vary as a function of 1/T ). The 8 0 data from the Rauer Group marbles imply that cooling rates from M3 metamorphism were relatively slow. Thermal modelling of metamorphism due to the emplacement of plutonic rocks predicts cooling rates of >100°C/Ma (Rosthstein & Hoisch, 1994). Unless the region was subjected to multiple intrusions over a protracted period of time, the slow cooling rates combined with textural evidence for post-M3 decompression are probably more consistent with metamorphism occurring due to tectonic thickening. The stable isotope data may also help to constrain the geochronological data (Kinney et al, 1993). Mafic granulites in the Rauer Group have folded fabrics that are cut by -1.0 Ga tonalite veins. Leucogneisses that probably represent partial melts of metapelites have been dated at -1.0 Ga. However, metapelites and leucogneisses yield -0.5 Ga Sm-Nd ages (Hensen & Zhao, 1995), and many gneisses contain zircons with -0.5 Ga rims (Kinney et al, 1993). Biotite K-Ar and Ar-Ar ages of 0.49 Ga for gneisses south of the Rauer Group (Zhao et al, 1992) are interpreted as marking the stage at which the terrain cooled through 500°C. If the granulite event was at 0.5 Ga, cooling rates would have been ~30-40°C/Ma. However, the oxygen isotope data are probably more consistent with slower cooling rates. We suggest that the granulite facies event is represented by the 1.0 Ga ages. The 0.5 Ga event may represent a period of retrogression fluxed by fluids associated with 0.5 Ga pegmatites that cut the granulites or a later thermal event at ~600°C which reset the Sm-Nd isotope system. References 18

Buick, I.S., Harley, S.H. & Cartwright, I. 1993. Contributions to Mineralogy and Petrology, 113, 557-571. Dodson, M.H. 1973. Contributions to Mineralogy and Petrology, 40, 259-274. Harley, S.H. & Buick, I.S. 1992. Journal of Petrology, 92, 693-728. Harley, S.H. & Fitzsimmons, I.C.W. 1991. Journal of Metamorphic Geology, 9, 231-244. Hensen, B.J. & Zhao, B.1995. Australian Journal of Earth Sciences, in press. Kinny, PX>., Black, L.P. & Sheraton, J.W. 1993. Antarctic Science, 5,193-206. Matthews, A. 1994. Journal of Metamorphic Geology, 12, 211-220. Rosthstein, D.A. & Hoisch, T.D. 1994. Journal of Metamorphic Geology, 12, 723-734. Zhao, Y., Song, B., Wang, Y., Ren, L., Li, J., & Chen, T. 1991. In: Yoshida, Y. Kaminuma, K. & Shiraishi, K. (eds), Recent progress in Antarctic Earth Sciences. Terrapub, Tokyo, 155-161.

QUARTZ PALEOPIEZOMETERS AND THE EFFECTS OF ANNEALING John M. Christie Department of Earth and Space Sciences, UCLA, Los Angeles, CA, USA 90024 Experimental Data. Annealing experiments were carried out on two quartzites, flint and finegrained novaculite at temperatures of 700, 750 and 800C at 1.0 GPa pressure and 850C at 1.5 GPa — all within the a-quartz stability field. One percent of water by weight was added in each experiment. 2

Grain growth kinetics for novaculite are well fitted by the relation D - D = ct, where D is the initial grain size, D is the grain size at time t, and c is temperature dependent according to the relation c=c exp (-Eg/RT); Eg is the activation enthalpy and R is the gas constant. Eg= (Q+PV*) where Q and V* are activation energy and volume, respectively. Experimental values for novaculite are c = 7.5 x 10" m Y , Q = 281 kJmol" and V* = -1.86 x 10" n^mol" . Grain sizes increased by factors of 4 to 5 in the longest experiments (= 800h) at 800 and 850C and grain growth was faster in flint. (Note that grain sizes indicated are mean linear intercepts measured on polished and etched surfaces.) Dislocation microstructures were studied by TEM in each of the four aggregates from the same experiments. Dislocation densities (r) were estimated in the quartzites only; the flint and novaculite were undergoing rapid grain growth and were unsuitable for this purpose. The initial densities (r ) were high: 1.64 x 10 cm" in the metamorphic quartzite (KH-1) and 1.76 x 10 cm" in the sedimentary quartzite (NWH). (More than 80 micrographs were analyzed from each rock to obtain these values). Both quartzites showed similar dislocation annihilation rates and the data are presented together. Densities were reduced by an order of magnitude in experiments of 790h at 800C and 205h at 850C. 0

Dislocation recovery is best approximated by second order kinetics: dr/dt = -Kr , where r is the dislocation density at time t. The data were fitted to the derived relation (1/r - 1/ro) = kt, where k = ko exp [-(Q + PV *)/RT]; k is the rate constant and Qr and V * are respectively the activation energy and volume for recovery. Data from the 1.0 GPa experiments give values of k = 1.2 x 10 m V and E = 146 (± 12) kJmol" ; the latter value is similar to that for steady-state flow of quartzite by dislocation creep and dynamic recrystallization. r

There is a very significant pressure dependence: annihilation rates at 850C and 1.5 GPa are 3-4 times faster than predicted by the 1.0 GPa data. Analysis of the data for 1.0 and 1.5 GPa yields values of Q = ^SkJmol" and V * = -1.96 x 10" m mol" . This large and negative activation volume indicates that recovery rates will be appreciably slower at shallower crustal levels (P < 1.0 GPa). 1

Consequences. Assuming that peak temperatures in deformed metamorphic rock may persist for millions to several tens of millions of years after deformation, as is suggested by conductive cooling models, these kinetic data place significant limitations on applications of experimentally derived piezometric relationships, such as recrystallized grain size and dislocation density. Grain size: Initial grain sizes of 100mm (corresponding to flow stress =10 MPa) will remain stable for 10 - 10 years at 400C, but only 1 My at 500C and 10 years at 600C. Unusually small grain sizes of 10mm (flow stress = 1 GPa) would be stable for = 10 My at 7

400C, but only 10 at 500C. Thus grain size in pure quartz rocks that are water saturated is only reliable as a piezometer in rocks deformed at low temperatures (< 500C) if peak temperatures persist for more than 1-10 My after deformation ceases. 3

Dislocation density: An appropriate and conservative measure of the stability of this parameter is the halving time t(l/2) — the time for the dislocation density to decrease to half of r at any temperature and pressure. (Practical experience indicates that a factor of two is a maximum estimate of error at high densities (r>10 cm" ), though errors are much smaller at lower values). Halving times for typical natural densities of 10 and 10 cm" are given in Table I, which also shows the large pressure effect. TABLET Halving times in years 0.0 P(GPa) 1.5 1.0 300 400 T(C) 300 400 300 400 lxlO r = 10 cm" 7x10^ 2xl0 3xl0 2xl0 7xl0 2xl0 1x10 10 cm" 7xl0 3xl0 2xl0 7xl0 0

At low pressures t(l/2) is high (10 - 10 years) at 300C and significantly shorter (10 - 10 years) at 400C. At temperatures above 400C, annealing is rapid enough for recovery to alter the initial dislocation density, so that simple application of the piezometric relations is precluded. 6

Conclusion. It is evident that recovery and grain growth are sufficiently rapid at temperatures much above 400C (greenschist facies) to prevent simple application of experimental piezometric relations. However, the kinetic laws have potential usefulness in determining details of the thermo-mechanical history of rocks. Examples will be discussed.

THE KINEMATIC EVOLUTION AND TECTONIC IMPLICATIONS OF MAJOR ARC-OBLIQUE LINEAMENTS, IN THE NORTHWEST ARGENTINE ANDES. T.J.Coughlin & R.J. Holcombe Department of Earth Sciences, The University of Queensland, Brisbane, Qld4072, Australia This paper presents the initial findings of a satellite lineation interpretation and time-space analysis of the northwestern Argentine Andes. The following commentary is supported by reconnaissance field observations and constitutes a prelude to a three year study aimed at resolving some of the fundamental structural and tectonic problems of the Argentine Andes. Several major lineaments oriented at a high angle to the north-south trending Andean Orogen have previously been identified in northwestern Argentina (Mon, 1976; Ramos, 1977). New digitally mosaiced LANDSAT Thematic Mapper and Multispectral Scanner data, from the Argentine and Chilean andes, allows for the accurate plotting of major lineament traces and an interpretation of their significance in the context of a more regional tectonic framework. Analysis of the data to date has confirmed the findings of previous workers and identified an extensive pattern of northeast and northwest trending lineaments (Figure 1).

The lineaments correspond to both narrow and broad zones of (often extreme) topographic variation. Their structural manifestation varies from local faulting and strike changes to broad zones of anomalous internal structure and focussed igneous activity, defining the structural boundaries of offset tectono-stratigraphic terranes (Mon, 1976; Ramos, 1977; Allmendinger et al., 1983).

A literature-based time-space analysis of the major morphotectonic elements between the latitudes 24° and 32° has identified multiple, complex cycles of superposed arc-orogenic episodes since at least the midPalaeozoic. The distribution of the major Palaeozoic structural stages and the Jurassic to Recent Andean Cycle are presented in Figure 1.

Figure 1. Satellite lineation interpretation (arc oblique lineations in thick-line, arc parallel and normal lineations in thin-line).

The spatial distribution of the major arc-oblique lineaments relative to the identified structural stages and tectonic cycles,

demonstrates their influence on the geological development of the region (Figure 1). The most significant lineament, in terms of tectonic implications, is that previously identified in the northeast of the study area as the Hualfin Lineament (Ramos, 1977). The trace of the (extended) Hualfin Lineament (Figure 1), corresponds to a fundamental tectonic boundary, separating the Sierras Pampeanas, a thick-skinned structural terrane in the South, from the Cordillera Oriental thin-skinned structural terrane in the North (Mon, 1976). The lineament also defines the southern boundary of the late Cenozoic volcanic Puna plateau and corresponds to the southern extent of the transition zone from steep to flat subduction of the Nazca Plate (Allmendinger et al., 1983). The north-south trending fold-thrust belt of the Argentine Precordillera terminates against the southern margin of the Hualfin Lineament and proximal to its confluence with the northwest trending Valle Fertfl fault. Previous workers have identified a consistent pattern of right-lateral and left-lateral strike-slip displacement on regional northeast and northwest trending fault zones respectively (Mon, 1976; Ramos, 1977; Allmendinger et al., 1983). Northeasterly rotated Neogene faults, enechelon fold patterns and offset stratigraphy have been interpreted as indicative of late Cenozoic right-lateral wrenching on the Hualfin Lineament (Ramos, 1977, Allmendinger et al., 1983, Urreiztieta et al., 1993). Cretaceous basalt vents and facies changes in sedimentary rocks occurring along and proximal to the Hualfin Lineament (Mon, 1976; Allmendinger et al., 1983), indicate a minimum Cretaceous age for the associated structures. The relative timing of displacement and absolute age of the structures however, remains contentious. This lineament analysis, constitutes the prelude to a continued study of the kinematics and tectonic influence of major cross-orogen structures and in particular, the Hualfin lineament and Valle Fertfl Fault (Figure 1). Initial field observations have shown that at the southern extent of the Puna Plateau, the Hualfin lineament corresponds to a broad topographic depression bounded on its south by a significant fault zone. The fault zone overprints a pre-Carboniferous deformation and may be significant in controlling obvious variations in sedimentary facies and structural style across the topographic depression. References Allmendinger, R.W., Ramos, V.A., Jordan, T.E., Palma, M & Isacks, B.L. 1983. Palaeogeography and Andean Structural Geometry, Northwest Argentina. Tectonics, Vol.2, No.l, pp 1-16. Mon, R., 1976. The structure of the eastern border of the Andes in north-western Argentina. Geol. Rundsch, 65, pp 211-222. Ramos, V.A., 1977. Basement tectonics from LANDSAT imagery in mining exploration. Geol. Mijnbouw, Vol. 56, pp 243-252. Urreiztieta, M., Rosello, E.A., Gapais, D., LeCorve, C & Cobbold, P.R. Neogene dextral transpression at the southern edge of the Altiplano Puna (NW Argentina). Ext. Abstracts, II Intl. Sym. on Andean Tectonics, Oxford, pp 267-269.

MICROSTRUCTURAL AND MECHANICAL EVOLUTION OF FAULT ROCKS IN THE PRESENCE OF REACTIVE PORE FLUIDS Cox. S.FA Streit, J.E.2, Kanagawa, K.3 and FitzGerald, J.D.4 1

Department of Geology, University of Newcastle, NSW 2308, Australia Institut fur Geowissenschaften und Lithospharenforschung der Justus-Liebig Universitat, 35390 Giessen, Germany ^Department of Geology, Chiba University, Chiba, Japan 4 Research School of Earth Sciences, The Australian National University, Canberra, ACT 0200, Australia 2

A variety of geological and geochemical evidence indicates that pore fluids are commonly present in crustal faults at seismogenic depths. Fluid migration through active faults is promoted by permeability enhancement associated with both aseismic creep and episodic seismic failure, and potentiually has a major impact on the mechanical behavior of faults. We discuss examples of natural fault rocks and simulated, experimental fault rocks to illustrate the role of reactive pore fluids in producing time-dependent effects on the microstructural evolution, fluid transport properties and mechanical behavior of fault rocks. The first of these examples involves fine-grained cataclasites in faults that localised migration of externally-derived fluids through Late Proterozoic granite on King Island (Tasmania, Australia). Banded cataclasites and injection cataclasites in small faults that form part of the Currie Shear Zone (CSZ) record a history of repeated slip and production of wear products at temperatures around 250-300°C. Individual slip events were each separated by intervals of compaction and hydrothermal alteration of wear products during post-slip fluid-rock interaction. Marked contrasts between the K/Na ratios of successive cataclasite layers, and between the composition of cataclasites and the granitic host rock, record a complex history of fluid-rock interaction due to episodic fluid migration within the CSZ. Variations in the chemistry of successively formed cataclasites are interpreted in terms of fluid migration paths, fluid/rock ratios and the evolving chemistry of fluids circulating through the CSZ. The formation of banded cataclasites in the CSZ is interpreted in terms of repeated seismic slip and periods of interseismic fault healing being associated with time-varying fault permeability, episodic fluid flow and fluid-rock interaction. Time-dependent variations in shear stress, fluid pressure and fault cohesion were key factors controlling rupture nucleation and recurrence, and the evolution of shear strength in the CSZ. Insights about time-dependent variations in the mechanical behavior and microstructures of fault rocks at hydrothermal conditions are also provided by experimental studies of simulated fault gouges. We describe a suite of experiments on fine-grained quartz aggregates in simulated gouge layers 0.5mm to 1mm thick, and inclined at 30° to the shortening direction between forcing blocks made of porous quartz sandstone. Tests have been conducted at a temperature of 1200K, a confining pressure of 300 MPa and a pore water pressure of 200 MPa. Two types of gouge have been investigated: a finegrained, initially angular quartz aggregate with a maximum grainsize of 5|nm, and a coarser quartz aggregate with a maximum grainsize of 250|im. Time-dependent strength changes, potentially similar to those associated with interseismic fault healing, were explored using compaction and sintering treatments prior to sliding tests. After hot isostatic pressing (HIP) treatments for intervals of 15 minutes to two hours, gouge layers attain strengths comparable to that of the forcing blocks. They have shear

strengths that are approximately double those associated with sliding on bare interfaces at the same conditions, with apparent coefficients of friction between 0.7 and 0.75 near their peak strength. Microstructural studies indicate that dissolution-precipitation processes assist rapid compaction of the simulated gouge during the HIP process. Rapid stress drop at axial displacements greater than 1mm for the fine-grained gouge are associated with localised shear failure along one of the interfaces between the gouge layer and the forcing blocks. At higher displacements, stable sliding occurs along this interface with friction coefficients between 0.4 and 0.5. The coarse quartz gouge tends to be weaker than the finer gouge for comparable HIP treatments, apparently because it does not compact as rapidly as the finer gouge and has attained a lower cohesive strength when sliding is commenced. The studies of natural and simulated fault rocks highlight the importance of reactive pore fluids in promoting changes in the strength and permeability of wear products between slip events. We argue that rupture nucleation and recurrence, and the time-dependence of fault shear strength during the seismic cycle, are controlled by a complex interplay between shear stress recovery, fluid pressure evolution and evolution of the cohesive strength of wear products.

3D MODELLING OF THE BOWEN BASIN: A SOLUTION USING A COMBINATION OF SOFTWARE S J D Cox , R J Korsch , M Nicoll & J Totterdell Australian Geodynamics Cooperative Research Centre CSIRO Exploration & Mining, PO Box 437, Nedlands, WA 6009 AGSO, GPO BOX 378, Canberra, ACT 2601 We are building a structural and stratigraphic model of Queensland's Bowen Basin in order to address two principal goals: (i) a 3D geological understanding of a large and important resource bearing area; (ii) an assessment and development of the technology required for 3D geology modelling. The resulting model also will form the basic input for a time coded reconstruction of the formation of the basin, which has significant implications for our understanding of the tectonics of adjacent areas, particularly thrusting in the New England Fold Belt The study builds on a major compilation of seismic reflection data acquired from the last 30 years, as part of an NGMA project Sedimentary Basins of Eastern Australia (SBEA). That project also includes the Surat and Bowen-Gunnedah Basins in NSW, and our modelling will be extended into these areas. The outputs from the SBEA project are a series of gridded representations of sequence-boundaries and interpretations of faults intersecting each of these. Lacking sufficient control data, seismic two-way travel times are used in place of depths. The interpretations were captured, checked and gridded using Petroseis software. Petroseis produces gridded surfaces which are generally smooth, but can include discontinuities corresponding to fault offsets. We used the GIS package GRASS to verify and adjust the gridding, in particular where deeper surfaces appeared to intersect shallower surfaces as an artefact of the gridding methods. This occurs in areas that are poorly constrained where minimum curvature interpolation causes overshoots. Rules expressed as boolean and arithmetic operations allowed corrections in GRASS, though the resulting interpretation is not unique. We used the mining system Vulcan to verify and label polylines representing the seismic interpretations. The 3D visualisation capability was particularly useful in lining-up fault traces to determine which ones sampled the same faults. When incompatibilities and misinterpretations were discovered they were also corrected in the Petroseis database. The now-clean database was imported into EarthVision. The Geologic Structure Builder module allows the fault sequence and geometry to be specified. The tops of stratigraphic zones may be tagged as depositional, channel erosion, or unconformity. This overcomes the problems found earlier with Petroseis gridding scheme, but clear sequence boundary properties must be defined in advance. EarthVision also allows a single logical surface to comprise a set of partial grids, which enables it to represent surfaces with repeated values (eg due to reverse faulting). The final model was achieved by using a sequence of unrelated software packages. No one of the packages was capable of completing the task alone. These operations required information structure and format conversions as it was passed from one package to the next, with the consequent dangers of information loss or corruption. However, this general approach, of using key features from a variety of software, seems to be currently the most realistic method of 3D modelling for geology. 1

THERMOCHRONOLOGICAL CONSTRAINTS ON THE TECTONIC EVOLUTION OF NORTHERN PAPUA NEW GUINEA P.V. Crowhurst. K.C. Hill & D.A. Foster Victorian Institute of Earth and Planetary Sciences, School of Earth Sciences La Trobe University, Melbourne, Victoria, 3083, Australia. The Bewani-Torricelli-Prince Alexander Mountains probably formed as a tholeiitic island arc in the Late Eocene - Early Oligocene (fig. 2A). The arc is interpreted to have accreted to the New Guinea margin by the Late Oligocene (fig. 2B), but may have formed on a ribbon of extended continental crust along the margin. Inferred rollback of the subducting slab beneath New Guinea in the Early Miocene placed the margin and arc into extension, creating starved graben in northern New Guinea and causing regional subsidence(fig. 2C). Extension caused the development of metamorphic core complexes adjacent to the graben. This caused middle crustal rocks to cool rapidly from temperatures >500°C in the New Guinea Mobile Belt. Two inferred core complexes that have been dated show rapid cooling from 27-23 Ma and 20-18 Ma (fig. 1). Continued subduction beneath New Guinea resulted in formation of the Maramuni Arc in the Middle Miocene and the end of extension (fig. 2D). In the Late Miocene, collision of the Melanesian Arc caused regional uplift of all basement of northern Papua New Guinea (fig. 2E), mainly from 8-5 Ma, causing at least 3-4 km of denudation. The compressional deformation propagated south causing uplift, denudation and cooling in the Papuan Fold Belt at -4 Ma, but continuing to the present (Crowhurst et al, in press; Spec. Publ. Geol. Soc. Lon.). 35

100

Time (Ma) 20

Bewani-Torricelli Mountains New Guinea Mobile Belt

0U 200 (Zircon Fission Track Age) 20Ma 1} 3

Rapid cooling possibly

to extensional <D 300 due CU tectonics and denudation of the metamorphic core H1

(Apatite Fission Trade Age) 5Ma iRapid cooling due to erosional denudation caused by arc collision 18Ma (K-Ar age - biotite) 23Ma (K-Ar age - biotite)

400

.Cooling history of metamorphic rocks from the Prince Alexander Mountains Mid-crustal rocks |>15km depth • 20Ma (K-Ar age - hornblende) 37Ma (K-Ar age - hornblende) 27Ma(K-Arage- hornblende) (Crystalisation Age ?) Figure 1. Cooling histories of the mid-crustal rocks now exposed in the Mobile Belt and of the intrusive and metamorphic rocks from the Bewani- Torricelli -Prince Alexander Mountains constrained by fission track and K-Ar dating. (K-Ar ages are from Rogerson et al., 1987; GSPNG, Memoir 12; Hutchison & Norvick, 1980; BMR Record 1980/24 and Page, 1976; BMR Bull. 126). Closure temperatures used to generate the curves are:- K-Ar hornblende (~500°C); K-Ar biotite (~300°C); zircon fission track ages (~200-250°C) and apatite fission track ages (~120°C).

Early Oligocene - Late Eocene Papua New Guinea

Pacific Plate

Island arc (future Bewani-Torricelli Mtns)

^Tonalites, diorites granodiorite Late Oligocene

Accreted arc

Palaeogene marginal basin

i New subduction zone forming Early Miocene

Sepik Graben

27-20 Ma

Incipient low-angle y detached normal fault Metamorphic core complex

Middle Miocene Maramuni Arc

Incipient obduction

Prince Alexander /metamorphic rocks

Rollback

Volcaniclastics

Pliocene New Guinea Papuan Mobile Belt Fold Belt Obducted ophiolite

Pacific Plate,

Sepik Bewani-Torricelli Basin Mountains

E Indo-Australiair Plate

Figure 2. 34

New Guinea Bismark Plate

Solomon Sea Plate Melanesian Arc

TECTONIC EVOLUTION OF THE WESTERN GAWLER CRATON: A PALAEOPROTEROZOIC COLLISION ZONE AND LIKELY PLATE MARGIN S J.Daly , M.C. Fairclough . C.M. Fanning and L.R.Rankin 1

Department of Mines and Energy, South Australia. P.O. Box 151, Eastwood\ SA 5063 Research School of Earth Sciences, ANU. GPO Box 4 Canberra ACT 2601 World Geoscience Corporation Ltd\ 65 Brockway Rd, Flore at, WA 6014. 1

High resolution aeromagnetic images of the western Gawler Craton show that it is dominated by a broad curvilinear belt of anastomosing shear zones broadly subparallel to the Karari Fault Zone (KFZ). Within and adjacent to these major shears are a number of crustal blocks which have different origins and metamorphic histories. Development of Proterozoic shears, and some reactivation of interpreted Archaean precursors, has resulted in recycling and reequilibration of Archaean and Proterozoic paragneisses of the Gawler Craton. The shear zones have also acted as foci for multiphase acid and basic intrusives and extrusives. Sedimentary sequences, deposited during the Palaeoproterozoic and metamorphosed from greenschist to amphibolite facies, now occur as isolated fault bound remnants in the centre of the Gawler Craton overlying Archaean Mulgathing Complex. To the west and north, flanking these central low to medium grade metasediments, are thicker more persistent Palaeoproterozoic paragneisses near Ooldea and Mt Woods. These constitute the Coober Pedy and Mabel Creek magnetic and gravity ridges. Lithologies include BIF, aluminous sediments, calcsilicate and carbonate, now with granulite facies assemblages. Hypersthene-sillimanite and sapphirine-quartz assemblages from MESA DDH Ooldea 2 indicate very high grade conditions of 8-10kb and 950-1000°C. Exsolution of hematite-magnetite symplectites in garnet indicates subsequent rapid exhumation probably at 700°C. SHRIMP U-Pb analyses of metamorphic zircon give ages of 1742+27 Ma for the Mt Woods area and 1657±7 Ma at Ooldea respectively. The high grade rocks at Ooldea strike east-west and are truncated to the south-east by the curvilinear KFZ. Similar sillimanite-hypersthene paragneisses in the Coober Pedy Ridge occur immediately to the north of the gravity-delineated KFZ. We propose that the KFZ is one of the main fractures along which these sediments were buried and rapidly exhumed. Immediately southeast of the KFZ (on TALLARINGA) the aeromagnetic images show lateral rotation of complexly folded Archaean gneisses, with both dextral and later major sinistral movements. Further south, Archaean gneisses are intruded by Proterozoic igneous rocks, the latter having foliations parallel to the Karari Fault Zone. Adjacent and sub-parallel to this region, prominent aeromagnetic and gravity highs separated by anastomosing shears, comprise the Fowler Suture Zone (FSZ; Fairclough et al. 1994, MESA digital dataset). The complex aeromagnetic signature of the FSZ presents a dramatic contrast to the magnetically quiet Archaean rocks. East of the FSZ, east-west trending shears within Archaean rocks show progressive anticlockwise rotation into the FSZ indicating major sinistral strike-slip movement. The east-west shears are interpreted as Archaean structures reactivated during the development of the FSZ, which is considered to be genetically related to the KFZ. The Fowler Suture Zone consists of Palaeoproterozoic meta-igneous calc-alkaline cumulus gabbro-diorite and tonalite. Ultramafic rocks have also been intersected in drillcore. Pelitic schist and BIF from drill core southeast of Lake Tallacootra appear concordant with the regional magnetic fabric. The aeromagnetic geometries of the polyphase mafic and ultramafic rocks indicate that they were emplaced during the development of the FSZ. It is possible that

the oldest mafic bodies may have been seafloor sills intruded into the original Proterozoic basin. Zircons from a mildly deformed cumulus gabbro in MESA DDH Colona 43 give a magmatic U-Pb age of 1730±10 Ma. Zircons interpreted as having formed during amphibolite facies metamorphism of a meta-gabbro/basalt in MESA DDH Nundroo 2 have an U-Pb age of 1537±10 Ma, though we note that the latter material has undergone granulite facies deformation. We conclude. that formation and metamorphism associated with the FSZ is constrained between 1730 Ma and 1540 Ma, significantly younger than previously thought. The younger age of 1540 Ma, implies a post Hiltaba Suite high grade event in the west Gawler Craton. Large multiphase granitoid plutons occur to the southeast of the FSZ and are possibly of similar age and lithology to the mildly deformed 1630 Ma St Peter Suite. Undeformed plutons have been interpreted as car. 1585 Ma Hiltaba Suite, however it is likely that some faulted and/or foliated granitoids may also be related to this Suite, given our new geochronological constraints on fabric development. Whilst largely strike-slip movements are evident in the aeromagnetic images of the FSZ and KFZ this is not consistent with outcrop scale observations. Stretching lineations in the Ifould Lake area and elsewhere generally pitch greater than 60-70 degrees to the north within subvertical foliation planes. Rare S-C fabrics are consistent with an east block up dip-slip movement component. These lineations could be interpreted to represent the most recently developed fabric only, however strike-slip has been interpreted elsewhere on the Craton to be the final deformation phase, and no evidence of a discrete dip-slip overprint is locally evident as structural repetition or superimposed fabrics. Such observations indicate that sinistral strikeslip movement is locally and spatially only a minor component of the deformation, and that to achieve the enormous displacement apparent in aeromagnetic images, an extreme degree of associated dip-slip movement is required within the FSZ. Bleeker (1990, Ph.D Thesis, New Brunswick University) describes a similar contradiction between geophysical and geological evidence in the Thompson Nickel Belt, Canada. Detailed structural analysis in that region has revealed that transpressional deformation, with strain partitioning into broad dip-slip and intense narrow strike slip domains, can account for the apparent discrepancy. Geophysical images show many geometric similarities to the Thompson region. It is suggested that transpressional collision can also account for regional structural features apparent in the FSZ with broad basement zones deformed by dip-slip strain accounting for most of the rare outcrop, separated by volumetrically minor (non-outcropping) zones of concentrated strike-slip deformation. It is proposed here that the FSZ represents an oblique Palaeoproterozoic (ca. 1730-1540 Ma) collisional zone between two Archaean-Palaeoproterozoic microplates, one being the Mawson Continent comprising the Gawler Craton and parts of the East Antarctic Shield, the other, perhaps the Yilgarn Craton. Associated with this collision are the development of a duplexstyle shear system, syn-orogenic intrusions, and intrusion of late stage zoned plutons during a subsequent period of crustal relaxation. It is further proposed that the KFZ is the leading edge of the over-riding plate and the high grade rocks at Ooldea and along the Coober Pedy ridge have resulted during collision. Rapid exhumation at c.1660 Ma indicates beginning of crustal relaxation and extensive igneous intrusion reflected in the intensive gravity low southeast of the FSZ. The FSZ may thus have played a significant role in the amalgamation of the Mawson Continent (including the Gawler Craton and parts of the East Antarctic Shield) with other Archaean and Proterozoic Cratons to form the Australian Continent during the PalaeoMesoproterozoic.

MICROSTRUCTURAL PROCESSES AND DEVELOPMENT OF MACROSCALE FOLD GEOMETRIES IN THE ROBERTSON RIVER MET AMORPHICS, NORTH QUEENSLAND B.K. Davis James Cook University of North Queensland\ Townsville, Qld 4811, Australia Rocks defining a macroscale antiform (25 km2) in the Proterozoic Robertson River Metamorphics have been affected by four deformations, Dxto D4, of distinctive style. The fourth deformation, D4, was responsible for formation of the macroscale geometry of the fold. Geometric analysis, combined with microstructural study of over 700 spatially oriented thin sections (e.g. Johnson, 1992), has revealed micro-scale and macro-scale geometries consistent with an interpretation of D4 folding via accommodation of progressive D4 shearing strain along pre-existing cleavages. Progressive shearing strain during D4 is interpreted to have been redistributed due to strain partitioning during progressive D4 deformation, and was accommodated by the phyllosilicate-rich zones defining both the S2 differentiated crenulation cleavage and the Sj foliation. D2 matrix crenulations are continuous with inclusion trails defining S2 in porphyroblasts. The matrix crenulations often have larger interlimb angles than those in the porphyroblasts (i.e. looking 'less developed'). In some thin sections the matrix crenulations also have narrower hinge regions than those in the porphyroblasts. These geometries are interpreted to have formed via 'unfolding1 of the D2 crenulations (e.g. Bell., 1986; Phillips and Key, 1992) due to the accommodation of D4 shearing strain along pre-D4 cleavage surfaces, with narrowing of the limb regions caused by strain-enhanced dissolution of material adjacent to the zones accommodating progressive shearing strain. On the fold limbs this shearing strain during D4 is interpreted to have operated in both a synthetic and an antithetic sense, relative to that operating at the scale of the macroscopic fold limb, with synchronous progressive rotation of the S2 cleavage. Synthetic progressive shearing strain (relative to the macroscopic fold limb) is interpreted to have been accommodated by Sj and S2 foliations that were oriented at a low angle to the axial plane of the macroscale antiform. Decrenulated Sj in the hinges of D2 differentiated crenulations was particularly important in this respect and commonly lies axial planar to microscale D4 crenulations. The processes of antithetic and synthetic shear are interpreted to have been operating synchronusly during D4 folding, with the relative importance of either process being largely dependent on the orientation of Sl and S2 cleavages in different domains. Despite the intensity of D4 deformation, a separate cross-cutting S4 cleavage has rarely been produced at either the meso- or macroscale. Instead, accommodation of progressive D4 shearing strain by S{ and S2 has resulted in the formation of extensive zones of composite S r S 4 and S2-S4 cleavages, which trend approximately parallel to the axial plane of the macroscale fold. These zones separate relatively wider domains where S! and S2 have not been modified but have simply been folded around the antiform. The 'reworking' of Sx and S2 foliations during D4 has resulted in the formation of apparent D] and D2 intersection lineations that are subparallel. This is because zones of composite S r S 4 that intersect folded S2 produce a lineation parallel to composite S2-S4 that has intersected folded S,. Intersection of zones of composite cleavage with folded bedding produces similar effects.

References Bell, T.H., 1986. Foliation development and reactivation in metamorphic rocks: the reactivation of earlier foliations and decrenulation due to shifting patterns of deformation partitioning. Journal of Metamorphic Geology, 4,421-444 Davis, B.K., 1995. Regional-scale foliation reactivation and re-use during formation of a macroscopic fold in the Robertson RiverMetamorphics, north Queensland, Australia. Tectonophysics, 242, 292-311 Johnson, S.E., 1^92. Sequential porphyroblast growth during progressive deformation and low-P high-T (LHPT) metamorphism, Cooma Complex, Australia: The use of microstructural analysis to better understand deformation and metamorphic histories. Tectonophysics, 214, 311-339. Phillips, E.R. & Key, R.M., 1992. Porphyroblast-fabric relationships: an example from the Appin Group in the Glen Roy area. Scottish Journal of Geology, 28, 89-101.

THERMAL, MECHANICAL AND KINEMATIC EVOLUTION OF A DUCTILE DUPLEX: RUBY GAP DUPLEX, CENTRAL AUSTRALIA W J Dunlap1. C. Teyssier2, J.G. Hirth3 & I. McDougall1 1 2 3

Research School of Earth Sciences, The Australian National University, ACT0200, Australia Department of Geology and Geophysics, University of Minnesota, Minneapolis, MN 55455, USA McLean Laboratory, Woods Hole Oceanographic Institute, Woods Hole, MA 02543, USA

Over the past decade the evolution of the Ruby Gap duplex of central Australia has been examined using structural and microstructural analysis, constraints from rock deformation experiments, results of 40Ar/39Ar isotopic analysis and thermal modelling, and field mapping. These results allow us to reconstruct the growth of the duplex, which accounts for up to 65 km of Palaeozoic crustal shortening in the southeastern Arunta Block. The analysis reveals that the duplex formed over an extended period, from about 350 to 270 Ma. Thrust sheets went through a cycle of imbrication and ductile deformation while partitioned into either the downgoing lower plate or the uplifting upper plate of a megathrust. As the imbrication process scavenged the lower plate, thrust sheets were transferred to the upper plate, experiencing relatively rapid exhumation, and eventually passing into the brittle regime. Late in the development of the thrust system, a gneiss complex forming the highest structural level of the upper plate, immediately above the duplex, became unstable and cascaded into a foreland of thick carbonates containing intraformational salt detachments. The Ruby Gap duplex is an antiformal stack of five thrust sheets that were deformed and superposed under greenschist fades conditions during the 400 to 300 Ma Alice Springs Orogeny (Dunlap et al 1991, Geology, 19, 1213-1216; Dunlap and Teyssier, 1995, Precambrian Research, 71, 229-250; Dunlap et al., in press, Tectonics). Duplex formation ensued as a consequence of overthrusting of the Amadeus Basin sediments by a basement-cored nappe 6-10 km thick forming the upper plate of the megathrust system. Mylonitized Heavitree Quartzite (undeformed prior to thrusting) caps each of the thrust sheets in the duplex. Deformation late in the duplexing process occurred near the brittle ductile transition for quartzite, and was localized along the fault zones bounding the thrust sheets, whereas the interiors of the thrust sheets remained relatively unaffected by the latest deformation. The microstructure of the quartzite varies from nearly undeformed in the lowest thrust sheets (toward the foreland) to coarsely recrystallized in the upper thrust sheets (toward the hinterland). Recrystallized grain size of both quartz and cleavage-forming white mica increases systematically toward the upper sheets. Consequently, estimates of the flow stress decrease smoothly toward the upper thrust sheets. In addition, a transition in the regime of dislocation creep in quartz occurs toward the upper sheets, from a regime dominated by subgrain rotation to one dominated by grain boundary migration. The change in dislocation creep regime across the duplex probably reflects the thermal structure of the system during deformation, with the hinterland portion of the duplex being the hottest. However, the transitions in both recrystallized grain size and dislocation creep regime are consistent with either an increase in temperature or a decrease in strain rate, or both, during deformation. Over fifty mineral separates from around the duplex have been dated by the K/Ar and 40 Ar/ 39 Ar methods. The overall pattern that has emerged is one where the rocks of the upper plate have apparently cooled below the closure temperatures of amphiboles and micas between about 400 and 350 Ma. In contrast, cleavage-forming white micas from the mylonitized quartzite within the duplex yield apparent ages falling mostly in the range of 340 to 310 Ma. The apparent ages of many of these cleavage-forming micas are believed to represent "deformation ages" (that is, ages of crystallization instead of cooling). White micas from the hottest portion of the duplex, however, probably retain cooling information. These interpretations are strongly dependent on the thermal history during deformation, a problem which we have addressed through multidomain analysis of K-feldspar 40 Ar/ 39 Ar diffusion experiments (cf. Lovera et al, 1989, JGR, 94,17,917-17,935). Seven diffusion experiments have been performed on K-feldspars, enabling the thermal history of the various structural levels of the duplex to be assessed. The characteristic pattern of the ^Ar/^Ar release spectra is one of staircaselike increases in apparent age, from miniumum ages of 220 to 250 Ma to maximum ages of about 280 to 400 Ma. These age gradients suggest that the K-feldspars have tracked the thermal history of the duplex during progressive closure to argon diffusion. The thermal models derived from 39

inverting the diffusion information all indicate that the duplex formed in the approximate temperature range of 250 to 350°C. In addition, after ductile deformation in the duplex had ceased and the system reached its final geometry (-post 310 Ma), relatively rapid cooling of the entire thrust system ensued, between 300 Ma and 270 Ma. This rapid cooling appears to have resulted from the combined effects of 1) detachment of the upper plate gneisses and transport into the foreland, 2) uplift and erosion of the upper plate and 3) relaxation of isotherms perturbed by thrust sheet advection. Preliminary analysis of the entire data set suggests the following evolution for the duplex (Fig. 1). By about 340 Ma duplex formation was underway, with the uppermost sheets experiencing imbrication and deformation at temperatures in excess of about 330°C. By about 320 Ma the imbrication process appears to have been nearly complete, with ductile deformation occurring at about 250°C in the cooler foreland portion of the duplex and at about 300°C in the hotter hinterland portion of the duplex. Subsequently, the entire thrust stack cooled as it was thrust upward and exhumed through the brittle-ductile transition in quartzite. As a result, deformation partitioned onto narrow fault zones, forming late cataclastic textures on fault surfaces. Ductile deformation then ceased in all but the most hinterland portion of the duplex. When the thrust sheets reached their final geometry, the thermal perturbation brought about by thrusting relaxed, passing white micas in the uppermost (hinterland) thrust sheets through their closure temperatures for argon by about 310 Ma. Final rapid cooling of the thrust system (at about 5-10°C/Ma) occurred at 300 to 270 Ma as extensional structures dissected the upper plate. Slow cooling through the temperature range 200 to 150°C continued to at least 220 Ma, probably as a result of erosion and isostatic compensation alone.

310 Ma

Rapid Exhumation of

111111111 350°C

§ 11»111111111111111111111 <« 111111111

TRANSIENCE VS AMBIENCE K. Ehlers & G.S. Lister Australian Crustal Research Centre, VIEPS Department of Earth Sciences, Monash University, Victoria 3168, Australia We propose that short lived heating events may account for regional metamorphism in a number of classic terranes that have been previously interpreted as the result of slow conductive heating of an overthickened crust according to the model of England and Thompson (Fig. 1). Quantitative estimates for the duration of these metamorphic episodes suggest that they can take place in a time frame of < 1 m.y., and the rate of cooling after these events might exceed as much as 500°C/m.y. Although the source of heat for metamorphism has not been conclusively identified, the short time scales involved suggest that advective heat input may be important (e.g., as the result of magmatism or fluid transport), and/or that the role of dissipative heating as the result of penetrative ductile deformation of the flowing rock mass (e.g., in shear zones) should be re-evaluated. Temperature

Temperature conditions frozen in during declining stages ~ a second thermal pulse

isothermal decompression until retrograde metamorphism

apparent P-T path defined I by coexisting assemblages peak conditions frozen in during declining stages of a thermal pulse

peak conditions set at maximum entropy v Metamorphism under the conditions W of the ambient geothermal gradient

v Metamorphism under transiently maintained conditions during a thermal pulse

(

Figure 1. The different P-T-t trajectories involved in regional metamorphism; (a) with P-T conditions determined by the ambient geothermal gradient which changes very slowly as the result of conduction on the crustal and lithospheric length scales; and (b) transient metamorphism, where conditions are "frozen" into the mineral chemistry during the declining stages of a thermal pulse. In one case (a) isothermal decompression takes place, and the conditions of peak metamorphism change only slightly. Cooling rates are slow. In the other case (b) isobaric cooling takes place, at very rapid rates, and the conditions of peak metamorphism are only transiently maintained. 41

STRUCTURAL GEOMETRY AND STRAIN DISTRIBUTION IN TWO SUPERPOSED AND REACTIVATED BASINS: THE ADELAIDE FORELAND FOLD THRUST BELT

Thomas Flottmann & Pat James Department of Geology & Geophysics, University of Adelaide, South Australia, 5005 Restorable cross sections provide useful constraints on the configuration of sedimentary basins that subsequently undergo orogenic contraction. In the case of pervasive deformation, cross-sections should incorporate information on the strain taken up by the formation of pervasive fabrics such as cleavage. We present the results of a regional structural study of the southern Adelaide Fold-Thrust-Belt, South Australia (Fig. 1) in which we use restorable cross-sections to examine the structure of the Belt. The southern Adelaide Fold-Thrust-Belt records the transition from extension to orogenic contraction at the inboard margin of the protoPacific. Initial Neoproterozoic extension is manifested by the Adelaidean basin that is superposed and partly crosscut and by the complexly shaped Cambrian Kanmantoo basin. This basin truncates the Adelaidean basin along a deeply incised and in planview complexly shaped margin. During basin reactivation, which is related to the Cambro/Ordovician Delamerian Orogeny both basins reflect contrasting distribution of contractional strain. In the Adelaidean basin the strain distribution is largely controlled by different lithologies, and the fact that certain lithologies onlap onto the craton margin. In the Kanmantoo basin the distribution of strain and its geometry is strongly influenced by the strong anisotropics that are represented by reactivated growth faults, and which are also clearly seen on seismic sections. Furthermore strain variations are controlled by metamorphic grade, which is increasing towards the north and west in the Kanmantoo Group rocks. The geometry of the Kanmantoo basin, which was dominated by transtensional tectonics at its southern margin on Kangaroo Island and by margin normal extension on Fleurieu Peninsula, influences the style and amount of basement involved deformation in the foreland. Steep boundary faults at the southern margin are tranpressionally reactivated and here basement involvement due to upthrusting of the former margin is minimal. The central and northern part of the Kanmantoo basin is intercalated with several basement inliers, which are presumably upthrust along footwall shortcut thrusts and reactivated former growth faults. The results of this study which combines surface data, seismic interpretation and aeromagnetic data show that regional strain analysis can assist in depicting substantial fault zones which may remain elusive in regional mapping of clastic sequences.

t t t t X ' ^

j - X ' l Platform Rocks(€)

AUSTRALIA

L 1 JL/JL l i f) AUSTRALIA^ \

| Normanville Group (C)

^ffiyyvj Kanmantoo Group (C)

%¥ ¥ i

MURRAY

¥ f t f ^CRATON^

\ V

Adelaidean (NP) ^ f f| Basement

BASIN

\ v

Granites Shear zone

OUTBOARD TERRANES

— I

Metamorphic Complex

J£f

Anticline

/ /

Thrust

^ ^

Stretching lineation

Syncline

f f t

,wilson

rrK

Adelaid<

i TERRANBi V YORKE PENINSULA

CARBONATE SHELF ON CRATON MARGIN

^cjunm

'iafiSRSru gKANGAROa .ISLAND

SOUTHERN OCEAN L

Kangaroo Island Domain

TERTIARY COMPRESSIONAL TECTONICS OF THE MT LOFTY RANGES AND GULF OF ST VINCENT REGION, SOUTH AUSTRALIA Thomas Flottman and Mike Sandiford Department of Geology & Geophysics, University ofAdelaide, South Australia, 5005 In the Adelaide region E-W compression from Middle Eocene - Recent has resulted in the marked uplift of about 300 m that causes the present day topographic relief of Kangaroo Island, Fleurieu Peninsula and Mt Lofty Ranges which abuts against a depression formed by the current Gulf St Vincent. Seismic sections through Backstairs Passage (Flottman and Cockshell, 1995), and observations on the Fleurieu Peninsula show that the topographic contrast is associated with the development of steep reverse faults which in part reactivate the western boundary faults of the Early Palaeozoic Kanmantoo Trough and Southern Adelaide Fold Belt. There is no evidence for dow-to the Basin (Gulf) normal faulting. Therfore flexural subsidence due to topographic loading formed the shallow depression of the Gulf St Vincent, where up to 600 m of Tertiary sediments have accumulated since the Late - Middle Eocene. The shallow rise of Yorke Peninsula, which parallels the topographic relief on Kangaroo Island/Fleurieu Peninsula is interpreted as a flexural bulge. As such, the development of the Mount Lofty - Gulf of St. Vincent system can be seen as part of the current E - W compressional stress regime throughout SE- Australia, the origin and onset of which remains contentious. In part, E - W compression in SE-Australia can be related to farfield tectonic stresses associated with the development of the New Zealand collision, although modelling studies (Coblentz et al., 1995) suggest that the effect of this far-field component is considerably diminished in South Australia. We note that the current topography in the Mount Lofty Ranges appears insufficient to account for the flexural loading evidenced by the Tertiary sedimentary succession preserved in the Gulf of St Vincent. This suggests infracrustal loads associated with the transition between Gawler Craton and the Adelaide Fold which may locally augment the far-field E-W tectonic compression in the Adelaide region. References Coblentz, D., Sandiford, M., Richardson, R, Zhou, S., and Hillis, R., 1995, The origins of the Australian stress field, Earth and Planetary Science Lettersin press. Flottman, T., and Cockshell, C.D., 1995, Palaeozoic Basins of southern South Australia: New insights to their structural historyfromregional seismic data, Australian Joounral of Earth Sciences, in press.

GRANITE COMPOSITION IN THE DELAMERIAN OROGEN: A RESPONSE TO OROGENIC CRUSTAL THICKENING AND THINNING J.Foden1. S.Turner2, and M.Sandiford1 Geology and Geophysics, University of Adelaide, Adelaide, SA 5001 Earth Sciences, Open University, Milton Keynes, UK

The Cambro-Ordovician magmatic history of the South Australian Delamerian Orogen can be linked to the thermal and mechanical history of the fold belt during the ~40Ma interval over which deformation occurred. Pre-dating this convergent deformation, possibly during an extension episode in the Early Cambrian, mafic alkalic and tholeiitic magmas originated from the lithospheric and asthenospheric mantle and were emplaced at high crustal levels. When convergent deformation began however, magma compositions became more granitic, with I-type or sometimes transitional I-S-type compositions prevailing. When this deformation ceased, there was an abrupt transition to bimodal magmatism when the granitic rocks changed to A-types. We link granite generation to an influx of heat carried by mafic magmas, with granite chemistry reflecting the degree of interaction between crust and mantle-derived melts. The proportion of crustal contribution is maximal during maximum crustal thickening, but is minor in pre- and post-deformational regimes. The relative extent of fractionation and crustal assimilation (of Archean and Palaeoproterozoic lower crustal rocks) varied systematically though the ~40Ma deformation history (525Ma-485 Ma) of the orogen. The syn-deformational granites resulted from AFC-type processes in which the rate of assimilation was high compared to the rate of crystallisation. Ascent of these hybrid magmas transported heat to mid to upper crustal levels where its dissipation resulted in Buchan style metamorphism and local, S-type granite production. The abrupt termination of deformation at 490±5Ma is attributed to the rapid thinning of the mantle lithosphere with consequent changes in the lithospheric stress regime, leading to the transport of mafic magmas to upper crustal magma chambers with negligible crustal assimilation. As a result fractional crystallisation became the dominant mode of magmatic evolution resultant granites being hot, incompatible element-rich, A-types. Simple calculations based on the conductive cooling of initial sill-like mafic intrusions are used to model the time evolution of parameters controlling assimilation and fractionation, residual melt composition, density, viscosity, Stokes Law buoyancy, NdSr isotopic composition and zircon inheritance. Results imply that the general crustal cooling history of mafic intrusions falls into two general categories: 1 Those which heat significant volumes of theirroofrocks to temperatures > crustal minimum melting temperatures (700°C). 2 Those for which the crustal minimum melting isotherm always falls within the margins of the intrusion. Type 1 magma processes dominate when mafic magmas stall at the Moho or at deep crustal levels. TTiese will evolve to produce I-S type granite melts with Si02 contents > 67% with masses equivalent to > 70% of the initial mafic body in between 0.3 and 1.5 Ma (±0.5Ma). They have Pb-, Sr- and Nd isotopic compositions which are strongly influenced by the crust and may contain significant proportions of inherited zircons. Type 1 magma chamber evolution occurs when the crust is thickened during convergent deformation of the lithosphere. In convergent stress fields mafic magmas may only reach their positions of hydrostatic neutrality with difficulty and will stall at progressively greater depths relative to the surface as deformation proceeds. 45

Type 2 magma chambers arise when mafic magma canriseto shallow crustal levels as may be appropriate during extension when mantle-derived melts experience only lithostatic loads and travel via dilatant fractures to levels approaching their levels of neutral buoyancy. Intrusions experience only slight crustal contamination and may develop to highlyfractionatedmelts (A-type granite, rhyolite, dacite, latite, or trachyte). They experience rapid decline in melt mass reaching 70% silica with between 20 and 40% of their original mafic parent mass and have few inherited zircons. The model for granite genesis in the Delamerian Orogen involving direct heat and mass transfer from the mantle to the crust in combination with intracrustal recycling is in marked contrast to the models proposed for I, S and A-type granite genesis in the adjacent and younger Lachlan fold belt where granites are regarded as intracrustal melts.

References

Foden J., Turner, S. and Morrison, R.S. 1990. Tectonic implications of Delamerian magmatism in South Australia and western Victoria. Geological Society of Australia, Special Publication 13, The Adelaide Geosyncline (The Daily Volume ),p465-482 Sandiford,M., Foden, J., Shaohua Zhou and Turner,S. 1992. Granite genesis and the mechanics of convergent orogenic belts with application to the southern Adelaide Foldbelt. in Second Hutton Symposium vol., special publication Transactions of the Royal Society of Edinburgh: Earth Sciences, 83, 83-93.

STRUCTURAL AND THERMAL EVOLUTION OF THE UPPER PLATE OF THE AEGEAN METAMORPHIC CORE COMPLEX OF IOS, CYCLADES, GREECE M. Forster . S.L. Baldwin , & G.S. Lister ^Department of Earth Sciences, Monash University, Melbourne, Vic Australia Department of Geosciences, University ofArizona, USA The structural and tectonic evolution of blueschists in the upper-plate of Ios, Cyclades, Greece is examined. The apparent age spectra obtained by step heating experiments of white mica using Ar/ Ar geochronology can be exactly correlated with the fabrics and microstructures can be developed during deformation and metamorphism. The apparent age spectra can be interpreted according to simple models of diffusive loss of argon during the PT history subsequent to growth and/or recrystallization of white micas (Lister & Baldwin, 1995). MacArgon modelling based on the measured apparent age spectra suggests that the upper-plate cooled after the last episode of metamorphic recrystallization, and that it has not been subjected to temperatures in excess of mid to low greenschist facies conditions for any significant length of time since the Eocene (Van der Maar & Jansen, 1981). Of particular significance is the microstructural data which suggests that both the upper-plate and the lower-plate on Ios have been subjected to an upper greenschist facies thermal overprint, with garnet-biotite assemblages locally developed. MacArgon models suggest that the duration of thermal events during such a thermal overprint cannot be of duration greater than 1 MYr, in particular given the magnitude of the peak temperatures inferred using paragenetic data (Lister & Baldwin, 1995). Microstructures show that fine-grained biotite (usually <10 m|n) is developed in the upper-plate, in microdilation sites where the Alpine micas have pulled apart, or in cleavage splits in microcrenulations. In one area immediately adjacent to the normal fault which defines the lower boundary of the upper-plate, the retrogressed high pressure assemblages are overprinted by euhedral garnet, biotite and poikiloblastic albite. This overprinting of a medium pressure metamorphic event may be linked to the period of core complex formation and a short-lived thermal pulse. References 1

Lister, G.S., & Baldwin, S.L. 1995. Modelling the effect of arbitary P-T-t histories on argon diffusion in minerals using the MacArgon program for the Apple Macintosh. Tectonophysics in press. Van der Maar, P.A. & Jansen, J.B.H., 1983. The geology of the polymetamorphic complex on Ios, Cyclades, Greece, and its significance for the Cycladic massif. Geol. Rund. 72, 283-299.

THE MESOPROTEROZOIC AND NEOPROTEROZOIC THERMAL AND TECTONIC HISTORY OF THE SOUTHERN GAWLER CRATON: CONSTRAINTS FROM ^Ar/^Ar GEOCHRONOLOGY David A. Foster & Karin Ehlers VIEPS, School of Earth Sciences, La Trobe University, Bundoora, Victoria 3083 VIEPS, Department of Earth Sciences, Monash University, Clayton, Victoria 3168 Ar/ Ar analyses of hornblende, biotite, muscovite, and K-feldspar from Archaean and Paleoproterozic rocks of the southern Gawler Craton, exposed on the Eyre Peninsula, South Australia, reveal a protracted history of slow cooling punctuated by episodes of more rapid cooling and reactivation, starting during the late stages of the Kimban Orogeny. Seven hornblende samples collected from the Kalinjala mylonite near Port Neill yield ages between 1600 and 1611 Ma. These samples were collected from rocks exhibiting a spectrum of metamorphic grades and degrees of mylonitization. Therefore, there is no obvious correlation between hornblende apparent age and intensity of deformation in the shear zone at Port Neill. The mylonite zone appears to have cooled uniformly below -500°C by -1600 Ma after the D3 stage of the Kimban Orogeny. Fine grained amphibole from an amphibolite layer in the Hutchison Group, in the Cleve subdomain, gives a maximum age -1611 Ma, a plateau at 1550 Ma, and evidence partial argon loss as late as -450 Ma. The concordance between the maximum age of this sample and the hornblende ages from Port Neill suggest a similar cooling history below -500 °C. The plateau age is probably younger because of the finer grain size and lower closure temperature of the hornblende. The young initial ages in this sample are associated with slightly elevated K/Ca ratios, which suggests that they are related to argon trapped in less retentive phyllosilicate intergrowths. Hornblende from a metamorphosed mafic dyke, on the east side of a major north-trending shear zone at Drummond Point also yields a similar age of 1590-1600 Ma. These older hornblende ages contrast with results from Tumby Bay and Lipson Cove to the south of Port Neill, where hornblende ages from mafic dykes are 1428 ± 2 Ma and 1429 ± 2 Ma, respectively. Previous K-Ar biotite studies in the southern Gawler Craton showed a general trend of decreasing apparent ages from >1600 Ma in the west to <1500 Ma in the east. Our Ar/ Ar results are consistent with this pattern but reveal that local variations in biotite apparent age are related to the chemical composition and effective grain size of the biotite. Single grains of Mg-rich biotite from Bratten Cairn, north of Port Neill, yield ages as old as 1590 ± 5 Ma. Fe-rich biotite from the same area yields maximum ages of 1541 ± 3 Ma. Some grains of the Fe-biotite experienced partial argon loss at -1410 Ma and -1050 Ma, while some grains of Mg-biotite apparently lost argon at -1450-1400 Ma. By contrast biotite from Lipson Cove gives an age of 1420 ± 2 Ma, similar to the hornblende age from this area. K-feldspar separates from Mount Greenly, Port Neill, Lipson Cove, and Cape Tournefort (south of Port Lincoln) yield monotonically increasing age gradients from 8001540, 1360-1610, 800-1410, and 700-1450 Ma, respectively. The minimum ages from these spectra record cooling though about 280 °C, while the maximum ages probably record cooling below temperatures of >350°C. The thermochronologic results indicate most of the southern Gawler Craton cooled below -500°C by 1590 Ma, after the end of the Kimban event and during the initiation of magmatism associated with the Gawler Range volcanics and Hildaba Suite plutons. It is possible that significant post orogenic denudation was associated with this magmatism and coincident cooling of the basement rocks. The biotite and K-feldspar data indicate that cooling after this time was more gradual. The 1428-1420 Ma hornblende and biotite ages from the coastal areas north of Port Lincoln and south of Port Neill indicate rapid cooling from >500° to <300°C. Disturbance to the K-Ar system in biotite north of Port Neill suggests that cooling followed a significant thermal pulse, possibly associated with dolerite dyke intrusion. Further, low temperature (low greenschist facies) reactivation during the Musgrave Orogeny for some areas of the Gawler Craton is also suggested by the biotite single grain results. The minimum ages of the K-feldspar samples record cooling and denudation during Neoproterozoic rifting. 1

THE ROLES OF APATITE FISSION TRACK ANALYSIS (AFTA™) AND FIELD STRUCTURAL MAPPING IN TECTONIC STUDIES: EXAMPLE FROM THE MURCHISON BASIN, NEW ZEALAND HJ. Gibson . C.J.L Wilson* and Pi . Green* Geotrack International, P.O Box 4120, Melbourne University, Victoria 3052, Australia School of Earth Sciences, University ofMelbourne, Parkville, 3052, Australia. This paper presents results of apatite fission track analysis (AFTA™) and field structural data from the Murchison Basin, New Zealand and demonstrates how these two techniques can (and should) be combined to provide rigorous assessment for tectonic history reconstruction. Apatite fission track analysis (AFTA™) combines the measurement of fission track ages and confined track lengths in detrital apatite with an understanding of the kinetics of track annealing to provide a technique for the evaluation of low temperature thermal histories. Unlike other paleothermometry techniques (eg. vitrinite reflectance), AFTA has the ability to constrain the timing of maximum paleotemperatures. Where there is evidence to suggest cooling from maximum paleotemperatures can be attributed to denudation, AFTA can also provide constraints on the timing and magnitude of uplift and erosion. The Murchison Basin is located in the north western part of the South Island of New Zealand, 25 kilometres west of the Alpine Fault The region is characterised by Carboniferous to Late Cretaceous age crystalline basement rocks, which are overlain by up to 7000 metres of deep marine to terrigenous Tertiary sequences. A further 2000 to 3000 metres of overlying sequence is thought to have beenremovedduring the late Oligocene to Recent Kaikoura Orogeny. Through AFTA, maximum paleotemperatures were determined in twenty eight surface samples from throughout the Murchison Basin. Results indicate high paleotemperatures (>110°C) at the margins of the basin and adjacent to major faults, and lower paleotemperatures (<85°C) at the depositional centre of the basin. AFTA data indicate coolingfrommaximum paleotemperatures at the margins of the basin generally commenced between 35 and 25 Ma, during the Oligocene. The margins were subsequently re-heated prior to cooling which commenced throughout the basin between 12 and 9 Ma, during the middle to late Miocene. In addition, AFTA data in two outcrop samplesfromnear the Matiri Fault in the central Murchison Basin indicate cooling from subsequent peak paleotemperatures of 80 to 100°C commenced between 5 and 3 Ma, during the early Pliocene. Although we have no constraint on paleogeothermal gradients, key elements of the local stratigraphy provide sufficient evidence to suggest that cooling events identified by AFTA can be attributed to uplift and denudation in the Murchisonregion.The earliest paleo-thermal event (35-25 Ma) identified by AFTA coincides with an almost instantaneous change from deep marine to terrigenous clastic sedimentation in the Murchison Basin. Subsequent paleo-thermal events commencing between 12 and 9 Ma, and between 5 and 3 Ma, coincide with the top Longford Formation unconformity whichrepresentsa gap in the preserved stratigraphy from Late Miocene to Quaternary times (12 Ma to Ma). Assuming a constant geothermal gradient of ~30°G/km since the Oligocene and a constant paleo-surface temperature of 15°C (both similar to present-day values), we used maximum paleotemperatures to estimate denudation since the onset of cooling, for each AFTA sample. Field analysis of the orientation and sense of movement on major faults, in conjunction with the use of published structural maps has been essential in interpreting the spatial variation in the timing and magnitude of denudation throughout the Murchison Basin. Relative movement of blocks either side of major fault traces, was determinedfromestimated denudation (assumed equivalent to uplift) above individual AFTA samples which are now at surface (Figure 1). Note that AFTA data suggest blocks to the eastern side of both the White Creek Fault and the southern Matiri Fault have undergonerelativelygreater uplift and denudation compared to blocks to the west of these fault traces. From AFTA data alone this suggests reverse faulting 49 1

probably occurred on easterly dipping thrust planes (because a predominantly compressional environment has persisted since the late Oligocene). However, field structural observations show that thrusting occurs on NW dipping fault planes. Resultsfromthe two techniques may appear to be conflicting, but we propose that the near-surface expressions of the White Creek and southern Matiri Faults have rolled-overfromeasterly dipping thrust faults, becoming NW dipping thrust faults in recent times, probably since the Early Pliocene. In support of this proposal, field observations indicate the Murchison Basin is located in a classic compressional wrench system in which case the development of roll-over structures, as described above, is entirely plausible. In particular, the southern Matiri Fault reveal that NW dipping fault planes envelope asymmetric crenulation folds and sub-horizontal slickenside lineations. These indicate a dextral shear sense with a SW over NE transport direction. In conclusion we suggest it is critical to combine AFTA data andfieldobservations in order to test models of tectonic evolution. This is because AFTA data provide information about the total amount of uplift and denudation over an extended period of time (since the middle Miocene, in the case of the Murchison Basin), whilefieldstructural observations relate only to the latest fault movements. Integration of data from both techniques provides a framework from which to understand the dynamic nature of tectonically active regions.

LEGEND: Total denudation since maximum burial (m) <2400 2400-3000 >>3000 Time of maximum burial: LO Late Oligocene O Oligocene M Miocene

•

Murchison Basin

AUSTRALIAN PLATE

0°

PACIFIC PLATE

0L.

100km

Figure 1: Location Map of the Murchison Basin, New Zealand. Block Diagram (above) shows therelativemovement of fault bounded blocks in the Murchison Basin. This is based on total denudation since maximum burial (from AFTA™) and is assumed to approximate uplift 50

FOLDING AS A MECHANISM OF MILD TECTONIC INVERSION: THE SOUTHERN MARGIN OF THE BRISTOL CHANNEL BASIN, ENGLAND R.A. Glen1 & P.L. Hancock2 1 Geological Survey of New South Wales, Department of Mineral Resources, Box 536 St. Leonards, New South Wales, Australia 2 Department of Geology, University of Bristol, Wills Memorial Building, Queen's Road, Bristol BS8 1RJ, United Kingdom.

The Bristol Channel Basin (BCB) is one of several extensional basins that formed in and around the British Isles in the Mesozoic. Contractional inversion of some or parts of these basins started in the latest Cretaceous but was mainly a Palaeogene-early Neogene event. The margins of the eastern part of the BCB are unusually well exposed along the English and Welsh coasts of the Bristol Channel but most of the basin is still beneath sea level. The southern margin, exposed in Triassic and Early Jurassic (Liassic) sedimentary rocks for 21 km along the northern coast of Somerset, is more deformed than the northern edge exposed on the coast of Glamorgan in old south Wales. This southern edge is dominated by major WNW-trending normal faults accompanied by open regional-scale contractional folds. NE and NW-trending faults and WNW-trending thrusts are subordinate structures. Our remapping of the northern margin of the Somerset coast of SW England between Watchet and Hinkley Point has demonstrated that the history of faulting is much more complex than previously appreciated. It also indicates that the regional folds are of great importance and formed during inversion, rather than as rollovers during extension on listric normal faults as previously suggested by many workers. Most of the WNW-trending normal faults dominating the southern margin of the BCB were formed in indurated rocks and are thus post-depositional. This post-depositional extension was largely achieved on roughly planar steep (60-70°) normal faults, down throwing both towards and away from the basin, and cutting gently dipping (25-30°) normal faults formed before tilting of the beds. The axial traces of footwall anticlines and hangingwall synclines extending no more than 1-50 m from the faults are interpreted as tip-line folds. Listric normal faults are rare but show some of the largest displacements. Relatively tight anticlines in their hangingwalls were probably initiated as gentle extensional rollovers and tightened during reverse reactivation. Neither we nor others have yet established a well-ordered sequence of faulting: only one growth fault (Late Triassic) has been recognised, a surprising observation in the context of the widely held belief that syn-sedimentary faulting was important during the evolution of the British early Mesozoic basins. Some normal faults formed during or after regional contractional folding, perhaps as a result of outer arc extension, while at a few other localities they cut an axial-plane cleavage associated with neoformed thrusts. Although the ultimate cause of basin inversion has not been resolved, it is clear that inversion involved three important processes: (1) regional folding about WNW-trending axes as a result of N-S contraction, (2) reactivation in reverse mode of WNW-trending normal faults, and (3) propagation of neoformed thrusts. Regional folds possess axial-trace 51

lengths up to and exceeding 1 km, and wavelengths measured in hundreds of metres. The dimensions of these folds commonly exceed those of important normal faults implying that they did not originate as extensional rollovers. Furthermore, some of them are segmented across-strike by WNW-trending normal faults. Reactivation of old normal faults in reverse mode rarely resulted in net stratigraphic separations becoming reverse. Neoformed thrusts are relatively uncommon and mainly achieve displacements of only a few metres. Strikeslip faults and localised N-trending folds and pressure-solution surfaces possibly reflect an early phase of inversion related to E-W shortening, or they could be some of the secondary products of the N-S stretching that preceded regional N-S contraction. ACKNOWLEDGEMENTS. Stratigraphic data on the attached posters are taken from published and unpublished mapping kindly supplied by Dr. A. Whittaker, British Geological Survey. Supported by the Commonwealth Department of Industry, Technology and Regional Development, and the NSW Department of Mineral Resources, with maps carefully drawn by Li Li and Paul Buttigieg of the Cartographic Section of that department. Published with the permission of the Director-General, NSW Department of Mineral Resources.

REGIONAL SCALE SHEATH-FOLDING AND HETEROGENEOUSLY DISTRIBUTED SHEAR STRAIN IN AN EVOLVING NAPPE PILE, OTAGO SCHIST, NEW ZEALAND D.R. Gray1. R.T. Gregory 2 , R.J.Norris3 & S.C. Cox 3 1 2 3

Department of Earth Sciences, Monash University, Melbourne, Vic 3168, Australia Department of Geological Sciences, Southern Methodist University, Dallas, Tx, U.S.A. Department of Geology, The University of Otago, Dunedin, New Zealand

The Otago Schist belt in the South Island of New Zealand is a metamorphic welt with higher grade, schistose greenschist facies rocks flanked by low grade non-schistose rocks as part of a broad, regional antiformal warp. The bulk of the schist is metamorphosed under chlorite to biotite zone conditions with a peak metamorphic temperature of 390°C followed by thermal relaxation prior to the last fabric-forming deformation (Yardley, 1982, Contrib. Mineral. Petrol., 81, 317-321). Metamorphic zones (Hutton & Turner, 1936, Trans. R. Soc. N.Z., 65, 405-406), textural zones (Bishop, 1972, Bull. Geol. Soc. Am., 83, 3177-3198) and transposition zones (Mortimer, 1992, Tectonics, 12, 237-244) are grossly parallel to this structure. Characterised by flat-lying schistosity, it shows transitions from 'segregated' schistosity into transposition layering with a persistent rodding and stretching (a-) lineation (Mortimer, 1992). The quartz-rodding lineation is parallel to long axes of stretched conglomerate clasts, pulled apart-porphyroblasts, and locally sheath fold axes. Within the schist macroscopic recumbent folds (nappes) have been mapped by vergence changes in mesoscopic folds within the schistosity /transposition layering (e.g. Means, 1963, N.Z. Jour. Geol. & Geophys., 6, 801-816; 1966, N.Z. Jour. Geol. & Geophys., 9, 173-194; Turnbull, 1981, N.Z. Jour. Geol. & Geophys., 24, 65-86). Recently, Mortimer (1992) however, has questioned the importance of nappe-like structures in the schist and has argued that schist evolution is dominated by ductile overthrusting in a single, progressive, non-coaxial deformation. We maintain that nappe-like folds are an important part of stuctural modified from Mortimer (1992) Nappes thickening within the schist complex. 1. Brighton nappe Nappe development 2. Nenthorn nappe 3. Macraes nappe is initiated by buck4. Manorburn nappe ling within a 'sim5. Blackstone Hills ple shear' dominatnappe 6. Bendigo nappe ed regime with sub7. Cromwell nappe sequent modifica8. Horn Range nappe tion by shear 9. Roaring Meg nappe 10. Northburn nappe induced flattening 11. Remarkables nappe and stretching. 12. Niger nappe 13. Poseidon nappe These folds are attenuated with either lower hinges and lower limbs strongly overprint50km ed by transposition foliation. Several regional scale closures of nappe-like folds have been mapped in the schist (Fig.l): Manorburn nappe (Means, 1966, N.Z. Jour. Geol. & Geophys., 9, 173-194). Niger nappe (Torlesse over Aspiring) (Craw, 1985, N.Z. Jour. Geol. & Geophys., 28, 5575) Poseidon nappe (Caples over Aspiring) (Bishop et al., 1976, N.Z. Jour. Geol. & Geophys., 19, 827-848) Remarkables nappe (Caples over Aspiring) (Cox, 1991) Bendigo nappe (within Torlesse) (Paterson, 1971, Univ. Otago Hons. Thesis) Horn Range, Gentle Annie, Roaring Meg Folds (Turnbull, 1981)

Many of these nappes at the regional scale have sheath-like form with curved axial surface traces in map view. Where unaffected by younger structures the nappe axial traces have map lengths of at least 30 km. Axial plane separations are up to 2-6 km. Regional nappes display variations in mesoscale structures where the regional hinge is normal, oblique or subparallel to the regional stretch direction defined by the rodding lineation (Fig. 2). Individual nappe closures are bounded above and below by high strain zones with transposition foliation and sheath folds associated with shear strains (y) in excess of 100 (Cox, 1991, N.Z. Jour. Geol. Geophys., 34, 73-82). Strains are very high (XZ ) and vary within the schist pile; in areas where bedding is still recognisable XZ strains are low (Norris & Bishop, 1990, Tectonphysics, 174, 331-349). Hinge zones are characterised by steep enveloping surfaces, recognisable sedimentary layering and symmetric meso-folds in zones up to 2 km in map width. Upper limbs are characterised by strongly transposed layering, buckled and extended veins, asymmetric folds, and a crenulation lineation (L n) generally perpendicular to the transport direction (TD). Lower limbs are characterised by zones of high strain, have extremely flattened and extended buckled veins, and show low angular discordance between the dominant foliation (S ) and the envelopping surface (S ). Rodding is pronounced and folds are distinctly noncylindrical with marked fold axis variability due to sheath folding. Regional deformation kinematics requires the Caples terrane to be overthrust on the Torlesse terrane (Fig. 1), as shown in the Remarkables nappe near Queenstown (cf. Cox, 1991). A concentration of large scale, recumbent isoclinal folds near the Caples- Torlesse boundary indicates however, that nappe formation has been important in this process. How do these nappe-folds propagate and amplify in a simple shear dominated regime? Initiation may be like that propsed for the Helvetic part of the Alps (e.g. Morcles, Diablaret and Wildhorn nappes: Ramsay et al, 1983, Geology, 11, 439-442) Mechanisms of propagation may relate to (1) a simple 'plane strain stretch of the folded sequence where the hinges are fixed and progressively move apart (Fig. 2a), or via (2) hinge rolling as in a 'conveyor belt' style of fold propagation within the layering due to increasing shear strain (Fig. 2b, c). Mechanism (2) requires overprinting fabrics which are not necessarily observed as layering moves from upper limb to lower limb structural positions. Furthermore, many hinges commonly showing unmodified bedding (e.g. Remarkables and Alexandra nappes) suggesting that hinge rolling has not played a significant part in nappe development. cre

1 coaxial flattening

high y

hinge migration

Rodding within the schist originates from (1) localised sheath folding with rigid-body rotation of hinge segments into the principal stretch (X) direction accompanied by subsequent stretching and segmentation. Extension fracturing with quartz-vein growth, and dynamic recrystallisation to form a quartz/albite grain elongation lineation within the original veins, are integral to the segmentation process (e.g. Remarkables nappe); (2) superimposed strains during progressive deformation, where a superimposed crenulation event intensifies locally to produce a pronounced rodding fabric (e.g. Brighton nappe). Some minor folds may originate by axis parallel shearing (cf. Mortimer, 1992), but the sheath-like folds which dominate the Otago Schist at all scales indicate that the majority of folds originate initially at high angles to the bulk transport direction in the deforming schist pile.

IMPLICATIONS OF A SUBDUCTION COMPLEX FOR THE TECTONIC EVOLUTION OF THE LACHLAN FOLD BELT D.R. Gray & J.M. Miller Australian Geodynamics Cooperatice Research Centre, Department of Earth Sciences, Monash University, Melbourne, Vic, 3168 Australia. Recognition of a palaeo-subduction zone along the south coast of N.S.W. (see Miller & Gray, this volume) has major implications for the tectonic evolution of the Lachlan Fold Belt of southeastern Australia. We reintroduce a subduction-related magmatic arc model to explain continental accretion for eastern Australia (Fig. 1). Such a model can explain the magmatic activity, magmatic underplating of the lower crust, the high T-low P metamorphism, and in particular the tectonic vergence of the structural belts that make up the Central and Eastern Lachlan Fold belt (Gray, in press, Geol. Soc. London Spec. Pub. "Orogenic Belts"). It has major effects for the evolution of the Eastern Lachlan Fold Belt, but puts the Western belt into a back-arc setting for most of the Silurian to Early Devonian. Recently proposed rift-drift delamination models (Collins, 1994, Geology, 22, 143-146; Collins & Vernon, 1994, Tectonophysics, 235, 249-275) do not explain the causes of crustal thickening and the vergence relationships of the upper crustal structural belts which clearly remain enigmatic.

Western Lachlan Central Lachlan Eastern Lachlan _ Fold ,| Fold it Fold , Belt Belt ~ Belt

continental crust

Late Devonian ~360 Ma Late Silurian Early Devonian ~420-380 Ma

• -. - sediment I I I I oceanic crust

Central Lachlan Eastern Lachlan Fold Fold Belt Belt

rollback

Wagga-Omeo JMetamorphic, ®l© Belt H

Scenario 2 Early Silurian -440 -420 Ma

deformed sediment prism + oceanic crust 11 plutonic activity

subductionrelated accretionary complex

underplating

Scenario 1 Early Silurian -440 -420 Ma

0 0 transform J / / active faults J / / inactive faults

rear-arc thrust belt

Fig.l: Tectonic cartoons for the development of the Eastern and Central Lachlan Fold Belt. Models are based on tectonic vergence requirements (modified from Powell, 1983, fig. 10)

The speculative models (Fig. 1) require an ocean-continent collisional setting with local geological complexities due to: 1) changes from a convergent to a transform margin (normal to oblique convergence), 2) changes in subduction zone dip, and 3) variations in the distribution of continental and oceanic crust. Unlike previously published tectonic cartoons (e.g. Powell, 1983, J. Geol. Soc. Aust. 30, 353-373) the key to these models is that they are based on the requirements of structural vergence. The basic premise behind the tectonic cartoons is that thrust-belts develop on the leading edges of actively driving plates where they verge toward, and against the motion of these plates (Coney, 1973, Geology, 1, 131-134). The angle of subduction controls the development of magmatic activity, with roll-back controlling the migration of plutonism and deformation. The cartoons show a "piecemeal" build-up rather than the massive intraplate-collapse within quasi continental-oceanic lithosphere proposed by Fergusson and Coney (1992, Tectonophysics 214, 417-439). Key tectonic elements in the cartoons include: 1) Early Silurian arc (Wagga-Omeo Metamorphic belt) (Fig.l c,d) characterised by high T/ low P metamorphism (cf. Morand, 1990, J.Met. Geol. 8, 1-12 ) and bounded by linked strikeslip faults and thrust faults (Morand & Gray, 1991, Aust. J. Earth. Sci. 38, 203-221). 2) Accretionarv complex/forearc belt (Tig, lc) or a back-arc thrust beltfFig. Id)?: Tabberabbera zone which consists of a simply folded turbidite sequence transitional into a 2km wide in situ-derived tectonic melange/broken formation of the Wonnangatta Fault (Fergusson, 1987, Am. Geophys. Un. Geodyn. Ser. 19, 39-56). 3) Early Silurian-Mid Devonian thrust-belt (Fig. lb,c) (Delegate-Bungonia thrust-belt: Fergusson & VandenBerg, 1990, J. Struct. Geol., 12, 577-589) transitional into subduction complex (Wagonga Complex: Bischoff & Prendergast, 1989, Neues Jahrbuch fur Geologie und Palaontologie, 175, 39-64; Fergusson & VandenBerg, 1990) Tectonic History The Ordovician period was dominated by an massive influx of turbidite from the west and southwest (present coordinates). Initiation of subduction in the Early Silurian/mid-Silurian (Eastern belt) (Fig. 1 c,d) led to a continent-ocean collisional interaction (Chilean setting) of a micro-plate with development of the Wagga-Omeo Metamorphic Belt and rear-arc thrusting (Tabberabbera Zone) to form the Central Lachlan Fold Belt (Fig. lb). The Western Lachlan Fold Belt underwent east-directed thrusting in the mid-Silurian perhaps due to failed subduction (Gray, in press) and represents an along strike variation in a complex plate margin oceanic setting. Other interpretations have the Western belt in a back-arc position at this time. In the Late Silurian to Early Devonian east-directed thrusting was taking place in the Western and Eastern belts respectively (Fig. lb). An inferred change in plate motion led to oblique convergence with eventual docking and amalgamation of the Western and Central/Eastern Belts along the Mt. Wellington Fault Zone in the late Early Devonian (Fig. la). This collisional interaction led to a change in the angle of subduction and magmatic underplating inboard of the continental margin leading to a period of silicic volcanism and plutonism in the Western Belt and broad continental style sedimentation (Lambie facies) across the Eastern Belt (Fig. la).

DUCTILE EXTRUSION OF HIGHER HIMALAYAN CRYSTALLINE OF BHUTAN HIMALAYAS. QUARTZ MICROFABRIC STUDY Diordje Grujicl Martin Casey1, Cameron Davidson2, Lincoln Hollister2, Rainer Kiindig3, Terry Pavlis4, Stefan Schmid5 7

£77/ Zentrum, Geologisches Institut, 8092 Zurich, Switzerland; Princeton University, NJ 08544; Geotechnische Kommission, ETH-Z, 8092 Zurich, Switzerland; 4 University of New Orleans, LA 70148; 5 Geologisch-Palaontologisches Institut, 4056 Basel Switzerland.

Since the recognition of the fundamental tectonic framework of the Himalayas (Gansser, 1964) considerable advances have been made. Several questions remain, however, without a satisfactory answer: (1) The Main Central Thrust is a major intra-crustal ductile thrust zone that is associated with an inverted metamorphic field gradient both above - in the Higher Himalaya Crystalline and below - in the Lesser Himalaya. Several models have been put forward to explain this phenomenon. (2) The South Tibetan Detachment is a major north-dipping low-angle normal fault with up to 15-18 km of throw and 80 km of heave. This normal faulting appears to be incongruous with the south directed thrusting along the MCT and within the Higher Himalaya Crystalline. (3) Leucogranites of the Higher Himalayan Crystalline were emplaced contemporaneously with the high temperature, intermediate to low pressure metamorphism and were associated with both southward- and northward-directed shear zones. In 1993 an expedition was undertaken in central and eastern Bhutan. This expedition provided a North-South traverse across all these problematic structures. This paper presents a synthesis of the evolution of the quartz microfabric across the deformed rock-pile of Higher Himalaya Crystalline in the Bhutan Himalayas, correlating these with the associated geological framework, including lithological succession, thermal distribution and field structures. The integration of field mapping and microstructural observations within the Bhutan Himalaya leads to following conclusions: (1) Asymmetry of c and a axes fabrics indicates top-to the south shearing from the Lesser Himalaya to the leucogranite injection complex at the base of the Monlakarchung-Pasalum pluton. The obliquity of the texture and the inferred finite strain (plane strain to moderately constrictional), suggest the strain regime had a combination of rotational and irrotational strain path. No change in strain magnitude in the quartz tectonites is observed throughout Higher Himalaya Crystalline of the Bhutan Himalayas. (2) Notwithstanding minor differences, most of the specimens from Main Central Thrust Zone and Higher Himalaya Crystalline of the Bhutan Himalaya show very similar crystallographic fabrics. In most of the specimens from Bhutan Himalaya the inferred deformation mechanism suggest low to moderate temperature conditions of deformation that produced the observed crystallographic preferred orientation. Much higher temperature of deformation is indicated only in the quartz veins from the leucogranite. (3) The observed ductile deformation is pervasively developed in the rocks throughout the investigated area. The intensity of deformation increases only slightly in the vicinity of the Main Central Thrust. Such deformation might account for some observed phenomena in Himalayan orogen like inverted metamorphic field gradient and the emplacement of leucogranites.

(4) Our kinematic interpretation involves coaxial thinning and concomitant southward extension parallel to the thrust transport direction accompanied by synchronous shearing within the lower levels of the Lesser Himalaya. (5) To explain the simultaneous operation of southward thrusting along Main Central Thrust and normal fault extension on the South Tibetan Detachment a model of tectonically southward extruding wedge by ductile flow has been proposed for central Himalayas of Bhutan (Fig. 1). The process of extrusive flow suggested here can be approximated quantitatively by channel flow models that have been used to describe subduction zone processes. Due to the displacement distribution, governed by channel flow, the largest finite displacement is to be expected in the upper part of the "channel". This zone is underlain by the belt of leucogranite plutons. (6) Inverted metamorphic field gradient in the Bhutan Himalaya is explained as a result of deforming the metamorphic isograds by ductile shearing along pervasive planar fabric - a mechanism analogous to similar folding on a regional scale.

MCT

Fig. 1 Schematic cross section of the Himalayas displaying implied relations between the Main Central Thrust zone and South Tibetan Detachment zone. A shallow crustal wedge is tectonically extruded to the south relative to India and Tibet. It is bounded on the bottom by thrusts of the MCT zone and on the top by normal faults of the STD zone - a boundary that acts as hinterland dipping normal fault. The wedge is pervasively ductily deformed, although the deformation is more concentrated on the boundaries of the wedge. The penetrative deformation, and maximal finite displacement in its core result in passive folding of metamorphic isograds. Effects of shear heating end heating brought by granitic pluton is not taken in account. Due to the non parallel walls of the channel - wedge an inverse flow will initiate where the underthrusting rocks rich the critical low viscosity. Vertical scale ca. 2.5 times exaggerated.

METAMORPHIC EVIDENCE FOR PALAEOPROTEROZOIC OBLIQUE CONVERGENCE IN THE EASTERN GAWLER CRATON Martin Hand, Betina Bendall & Mike Sandiford Department of Geology and Geophysics, University of Adelaide

Kimban metamorphism in the Lincoln Complex on southern Eyre Peninsula (1850-1700 Ma) (Parker, 1993) is characterised by a substantial across strike variation in peak metamorphic pressures recorded by transitional granulite facies assemblages. Maximum pressures (8-10 kbar) occur in a narrow domain located within, and immediately to the east of the Kalinjala Mylonite Zone, a major NNE-trending sub-vertical structure (Parker, 1993). Orogen-parallel movement within the KMZ is recorded by shallow-plunging L-fabrics defined by the peak metamorphic assemblages (e.g. Oussa, 1993). Maximum metamorphic pressures decrease smoothly away from the KMZ across strike to the east, reaching a stable value of - 5 kbar approximately 25 km from the fault, defining an average apparent baric gradiant of 0.16 ± 0.03 kbar/knr 1 . This is accompanied by a general increase in the plunge of lineations away from the KMZ. Garnethornblend-whole rock Sm-Nd isotopic data suggests the high-P assemblages (1730 ± 20 Ma) are roughly coveal with the lower pressure assemblages to east (1714 ± 12 Ma). Lateral variations in extent of exhumation provide an important kinematic constraint on the Kimban Orogeny that to some extent is independent of structural interpretation. Since the inferred baric gradient cannot have been a direct function of topography, a dynamic environment in which large amounts of exhumation were localised into a narrow strike-slip zone is envisaged. The increase in metamorphic pressures toward this zone implies that net, strike normal movements predominated away from the KMZ. The presence of low-pressure rocks in the Hutchison Group immediately west of the KMZ further supports the notion that differential exhumation was localised along the KMZ during the Kimban Orogeny. It is also apparent that rather than being a retrograde feature as suggested previously (e.g. Parker, 1993), the KMZ played a central role in the evolution of the Kimban Orogeny. The metamorphic observations in the Lincoln Complex are suggestive of a large-scale transpressional environment where strain was partitioned between a region of broadly strike-normal movement and a narrow region of dominantly strike-parallel movement through which material was rapidly exhumed. Preliminary kinematic observations along the apparent baric gradient in the Lincoln Complex are consistent with differential exhumation toward the strike-slip boundary. Although pre-existing structures, particularly mafic dykes, complicate the local strain picture in the complex, the total deformation pattern within the Lincoln Complex may be of more value in identifying the kinematic environment of the Kimban Orogeny than structural studies in the KMZ itself, where preservation of material that passed through the orogen is minimal. References Oussa, S. (1993). Description of a granulite facies shear zone (Kalinjala Mylonite Zone), and inferred cooling rates following granulite facies metamorphism. Unpublished Hons. Thesis, The University of Melbourne. Parker, A.J. (1993). Palaeoproterozoic. in: Drexel, J.F., Preiss, W.V & Parker, A.J. (eds.), The Geology of South Australia: Volume i, The Precambrain. State Print, Adelaide.

LOW-P HIGH-T METAMORPHISM AND THE ROLE OF HIGH-HEAT PRODUCING GRANITES IN THE NORTHERN ARUNTA INLIER Martin Hand. (!>2), Mark Fanning3 & Mike Sandiford1 1 2 3

Department of Geology and Geophysics, University ofAdelaide, Adelaide SA 5005 School of Earth Sciences, Melbourne University, Parkville VIC 3052 Research School of Earth Sciences, Australian National University, Canberra, ACT 0200

SHRIMP U-Pb ages for zircons from low-P (4.7±0.4 kbar, 730±30°C) granulitic gneiss in the Anmatjira and Reynolds Ranges provide evidence for a significant time interval 200 Ma) between voluminous granite emplacement and regional (> 30002 km) high-T metamorphism. Highly deformed and migmatised orthogneiss in the southeastern Anmatjira Range contains structured zircons, the centres of which display complex multistage Pb-loss between 18201770 Ma. The rims have high U/Th ratios and an age of 1571 ± 19 Ma that is interpreted to reflect the time of new zircon growth during either melting of the orthogneiss, or crystallisation of garnet-bearing leucosomes that both parallel the gneissic layering, and occupy crosscutting conjugate shear bands. Felsic orthogneiss within a migmatitic high-strain zone on the southern edge of the Anmatjira Range also contains zircon with structurally discrete centres, that crystallised at 1767 ± 17 Ma. These are likely to have grown during the emplacement of the igneous protolith of the gneiss. Subsequently, high U/Th overgrowths and new, multifaceted grains grew at 1570 ± 14 Ma. In both these samples, syn-to late kinematic migmatitic structures represent the youngest high-temperature features. In the southeastern Reynolds Range 20 km to the south, microgranite/pegmatite that formed during partial melting of granitic orthogneiss contains zircons with well defined cores that have a mean age of 1773 ±19 Ma. These are enclosed by high U/Th rims (1570 ± 6 Ma) which are interpreted to have grown during crystallisation of the partial melt. The cores are interpreted to be the crystallisation age of the granitic precursor to the migmatite. Previous SHRIMP ion microprobe dating in the Anmatjira Range (Collins & Williams, 1995), identified two populations in migmatitic charnockitic gneiss. The majority of grains belong to a well defined population (1774 ± 6 Ma) inferred to represent the crystallisation age of the charnockite, while a subordinate group of euhedral high U/Th grains (1587 ± 6 Ma) were interpreted to have grown during hydrothermal fluid fluxing associated with pegmatite intrusion. However on the basis of the samples described above, we prefer an alternative interpretation and suggest the youngest zircons in the charnockitic gneiss grew during regional low-P granulite conditions in the southeastern Anmatjira Range. Low to intermediate-P high-T metamorphism must often reflect transient advection of heat due to magma ascent. However in the Anmatjira and Reynolds Ranges, there is no obvious advective heat source associated with c. 1575 Ma tectonism. The U-Pb data indicates that the major magmatism occurred around 1780 Ma (Collins & Williams, 1995, this study) and was therefore not associated with regional low-pressure granulite conditions at - 1575 Ma. This is consistent with the field observation that all the major granitic bodies are highly deformed and metamorphosed by the regional high-T event. The zircon data also places some constraints on the duration of high-T, low-P metamorphism in the northern Arunta Inlier. The oldest zircon growth in the Anmatjira Range occurred at 1587±6 Ma (Collins & Williams, 1995). Assuming the similarity in zircon ages from the Anmatjira Range and southeastern Reynolds Range is a valid basis for a metamorphic correlation between these two regions, the youngest zircon growth during this event occurred at 1570±6 Ma. This suggests the terrain stayed above 650°C for >17 ± 8.5 Ma. Together, the U-Pb and metamorphic data suggests relatively slow (5 ± 4 °CMa"1) regional cooling. This is consistent with the cooling rate of - 3°CMa"1 calculated by Williams et al., (1995) in c. 1575 Ma granulite contained entirely within the Reynolds Range and suggests that slow cooling rates were a feature of extensive low-P Mesoproterozoic granulite in the central northern Arunta Inlier. An obvious question that arises from these observations is the nature of a heat distribution that allowed high-T, low-P mineral growth to occur in the absence of a suitable advective heat 60

source. Intriguingly, although voluminous granitic magmatism pre-dated granulite facies metamorphism in the southeastern Anmatjira and Reynolds Ranges by ~ 200 Ma, there appears to be a general spatial relationship between high-T metamorphism and the distribution of granite (e.g. Stewart, 1981). These granites comprise ~ 60% of the exposure.(Stewart, 1981) and are characterised by high U-Th-K concentrations and a correspondingly high heat production (5-10 liWnr 3 ) that cannot be representative of heat production in the rest of the crust. We show that if heat production of this magnitude is concentrated at shallow-to mid crustal levels, (10-20 km) then for a range of thermal parameters, steady state geotherms may produce conditions required for low-P high-T metamorphism without significant melting of a refractory lower crust. Importantly, geotherms show a strong dependence on the depth of the heat-producing layer, with the implication that minor burial 5 km) may be enough to initiate low-P high-T metamorphism. Similarly, only small amounts of erosion may be required to terminate the event. The metamorphic evolution of the Anmatjira and Reynolds Range region provides some support for this style of crustal heating. The c. 1575 Ma high-T metamorphism was terminated by 1-1.5 kbar of high-T decompression rather than isobaric cooling from the peak pressure (e.g. Vry & Cartwright 1994). At a later time (possibly 1500-1400 Ma or 400-300 Ma (Collins & Shaw, 1995)), the terrain was reworked by a major system of mid-crustal (4-6 kbar) shear zones associated with crustal thickening (e.g. Collins & Teyssier, 1989). The distribution of isograds associated with these shear zones is essentially identical to that of the earlier higher-T metamorphism, implying a similarly distributed heat source.

References Collins, W.J. & Williams, I.S. (1995). SHRIMP ionprobe dating of short-lived Proterozoic tectonic cycles in the northern Arunta Inlier, central Australia. Precambrian Research, 71, 69-90. Collins, W.J. & Shaw, R.D. (1995). Geochronological constraints on orogenic events in the Arunta Inlier, a review. Precambrian Research, 71, 315-346. Stewart, A.J. 1981. Reynolds Range region. Australian Bureau of Mineral Resources, Geology and Geophysics, 1:100 000 Geological Map series, Canberra. Vry, J.K. & Cartwright, I. (1994). Sapphirine-kornerupine rocks from the Reynolds Range, central Australia: constraints on the uplift history of a Proterozoic low pressure terrane. Contributions to Mineralogy and Petrology, 116, 78-91. Williams, I.S., Buick, I.S. & Cartwright, I. (1995). An extended episode of early Mesoproterozoic metamorphic fluid flow in the Reynolds Range, central Australia. Journal of Metamorphic geology, in press.

STRUCTURAL GIS AND COMPUTER APPLICATIONS IN FIELD GEOLOGY T.A. Harbort. R.J. Holcombe & C.J. Stephens

Department of Earth Sciences, University of Queensland, Qld 4072, Australia The regional exploration geology and tectonics group at the University of Queensland use a combination of PC-based applications to capture, record, and analyse field data:

GPS/Location Excel Spreadsheet Utilities. Initial field locations are captured either using handheld GPS units and downloaded directly into computer text files, or are entered manually into spreadsheets as AMG (UTM) or lat/long. A number of spreadsheet templates are presented that parse various types of GPS output data lines, and enable translation to and from specific AMG and lat/long formats (eg. decimal degrees for GIS entry). • GeoData, a MS Access-based relational database template, is the main database used for field data entry, querying and archiving. It is based on nine tables linked in one-to-many relationships, with the field number as the unique key, and linked to an external database containing global (non-project-specific) data such as stratigraphic nomenclature and map sheet names. The philosophy of GeoData has been to reduce the number of required tables to a minimum to handle different types geological field data; and by using linked tables any number of observations of any sort can be entered for each location. Thus the Keywords table handles any sort of data that links text to a field location, e.g., fabric descriptions; mineralogic or petrologic observations; fossils; sedimentary structures; the Numeric table handles any sort of data that links numeric input and an associated text field with a field location, e.g., strain estimates, stratigraphic interval data, number of fossils, etc. Two structural tables are included, one that handles any sort of dip/plunge-plus-azimuth data (fabrics, faults, joints, etc), and one for any sort of azimuth-only data such as vergence or younging directions, trends;etc. All tables are linked to a single multi-table entry form:

• GEOrient & GEOCalculator, MS Windows-based structural plotting (stereographic

projections and rose diagrams) and geometric calculation packages (Holcombe, 1994, Geol.Soc. Aust. Abstracts, 36,73-74) are used for structural analysis of data using MS Access as the prime querying tool of the GeoData database.

• Maplnfo utilities. Maplnfo is used as the prime GIS spatial database for displaying and

analysing spatial relationships, thematic mapping, for digitizing, and for map output. A number of utilities have been developed for this environment. These include:

• GeoSymbol, a MapBasic routine for defining and plotting structural symbols from

either other MapInfoTables containing dip/plunge/azimuth and location data, or from text files containing the same data. Work is currently under way to link GeoSymbol and GeoData.

• MI_Cat, a Maplnfo automatic table cataloguing utility. Mi-Cat ferrets through any

nominated path on a computer, finds any Maplnfo tables, catalogues the name, type, date, size,, geographic location and extents, and full file and pathname of any table, and plots a lat/long-based rectangle map object of the location. Mi-Cat can then be loaded into any Maplnfo workspace as a graphical database of all available data sets on the system. Since Maplnfo is capable of producing many hundreds of tables in any wellused computer system, Mi-Cat is invaluable in housekeeping the system.

KINEMATIC ANALYSIS OF ASYMMETRICAL FABRICS IN METATURBIDITES FROM THE ARCHEAN SLAVE STRUCTURAL PROVINCE, NORTHERN CANADA Jack Henderson1. Mariette Henderson1 & Tom Wright2 1 Geological Survey of Canada 2National Science Foundation, USA

Structural fabric data from metasedimentary rocks of the northern Slave Province exhibit a regional clockwise transection of vertical N-S striking bedding (S 0 ) and bedding-parallel cleavage (Si or S 2 ) by the main cleavage (S 3 ). In thin sections of chlorite grade metapelites, S3 is axial-planar to microfolded, bedding-parallel mica cleavage. Microfolds commonly are asymmetrical S-folds, with steep limbs in which bedding-parallel fabrics are rotated anticlockwise relative to bedding enveloping surfaces. Similar geometrical relations are seen in metapsammites, where S3 comprises spaced selvedges separating rotated bedding segments in microlithons. In higher-grade metaturbidites, blocky porphyroblasts of biotite, garnet, andalusite, and cordierite have overgrown bedding and bedding-parallel cleavage, and contain inclusion trails of quartz and graphite (Si). Characteristically, blastesis antedated S3, and the S 3 cleavage refracts across relatively competent, porphyroblast-rich metapelitic beds at a larger angle than less competent, porphyroblast-poor metapsammitic beds. In metapelites, Si trails in densely-packed porphyroblasts parallel external bedding traces (S e ). In metapsammites, Si in dispersed porphyroblasts are oblique to external bedding traces, and are truncated at the margins of porphyroblasts, which are wrapped by S3 cleavage. Characteristically, Si are perpendicular to external S3. These data are interpreted to indicate that the metasedimentary rocks were rotated to the vertical, and cleaved parallel to bedding before the peak of regional metamorphism. Porphyroblasts overgrew bedding during static regional metamorphism, and included bedding laminations and bedding-parallel cleavage as Si. The rocks were subsequently sheared in a sinistral transpressive regime, and the S3 crenulation cleavage formed clockwise to the regional N-S bedding trend. Short limbs of asymmetrical micro-folds, microlithons, and porphyroblasts were rotated anti-clockwise relative to the regional N-S bedding trend.

SEQUENTIAL FOLIATION DEVELOPMENT AND MINERAL GROWTH IN HIGH-GRADE METASEDIMENTARY GNEISSES, BROKEN HILL: THE PRODUCTION OF COMPOSITE FOLIATIONS DURING FOLDING OF GNEISSIC ROCKS K.A. Hickey Earth Science Department, James Cook University of North Queensland, Townsville, QLD 4811, Australia ABSTRACT A combination of mesoscopic mapping and microstructural analysis of over 1000 geographically oriented thin-sections suggests that metasedimentary gneisses of the Early Proterozoic Broken Hill Block in New South Wales, Australia, have undergone five stages of progressive high-grade deformation, D1-D5. These events are associated with an anticlockwise prograde P-T-f path and incorporate an apparent change in bulk shortening direction prior to D4. The first deformation is preserved as quartz and biotite inclusion trails in post-Di to syn-D2 Kfeldspar and cordierite porphyroblasts. Si is also locally preserved in the matrix where it is overprinted by the prominent gneissic foliation, S2. The first stage of deformation, Di, produced upright folds. Few Di fold hinges are exposed in the Broken Hill block, but they appear to be upright in nature. Metasedimentary gneisses in the Broken Hill region commonly have one well developed sillimanite-biotite gneissic foliation produced during the second deformation. S2 had an originally moderate to shallow dip prior to later D3 and D4 folding. A well developed southwest-northeast plunging sillimanite-biotite lineation, L l i e s within the S2 foliation. D2 produced a prominent set of inclined to recumbent macro- and mesoscopic folds with S2 as an axial plane fabric. During the development of this fold phase, initially northwest trending folds were variably rotated toward L|. This has resulted in D2 folds having a range of plunge directions. The third stage of deformation, D3, produced localised, upright to inclined, generally northwest-trending folds and crenulations of the S2 foliation and was not associated with development of a new penetrative foliation. Local growth of garnet porphyroblasts may have initiated during D3. A prominent coarsening of the matrix occurred over a period of time encompassing D2 and D3. The fourth deformation event produced widespread folding of So and S2. The folds have steep north-northeast striking axial planes and what appears to be a steep stretching direction. The D4 deformation was not accompanied by the development of a new penetrative foliation. Crenulations of So and S2 are developed in the hinges of D4 folds and on the margins of pre- or syn-D4 garnet porphyroblasts on fold limbs. In the latter case such crenulations typically do not extend far into the matrix. Rather than involve extensive axial-plane-parallel shear and the development of a new differentiated foliation, D4 folding proceeded by the reactivation of the pre-existing S2 foliation. D4 was associated with the major stage of garnet porphyroblast growth and a second stage of K-feldspar growth in the Broken Hill metasedimentary gneisses. The fifth deformation at Broken Hill, D5, is evident as subhorizontal crenulations overprinting D4 structures. It has not produced any major folds and is only locally well developed. The prominent role of reactivation during D4 folding at Broken Hill is considered typical of gneissic rocks, where well developed folds of an existing gneissic foliation often occur without any strong contemporaneous development of a penetrative axial-plane-parallel fabric in their 65

limbs, or in many cases their hinge (e.g. Hopgood and Bowes, 1990; Passchier et al., 1990). This pattern contrasts with that in many schist terrains where penetrative, often differentiated, axial-plane crenulations are typically well developed outside of fold hinges. A predominance of reactivation over crenulation may well arise in gneissic rocks because of difficulty in partitioning extensive shearing strain parallel to the axial plane across an existing gneissic foliation and intervening coarse-grained, phyllosilicate poor matrix. In the case of Broken Hill rocks this difficulty may have arisen in three ways. Firstly, the coarse grain size of the gneisses means that there are fewer grain boundaries per unit volume than in finer grained rocks such as schists. At temperatures as high as the amphibolite to granulite facies, intracrystalline deformation and recrystallisation is concentrated along grain boundaries rather than grain centres (eg. Bell and Johnson, 1989; Wenk and Christie, 1991). Consequently, a coarser grain size per unit volume may increase the competence of a polycrystalline aggregate. Secondly, the pelitic gneisses contain a high percentage of plagioclase, K-feldspar and cordierite. They are distributed through the matrix and locally overgrow and interlock with other minerals, particularly quartz. Their presence would have made it more difficult to propagate axial-planeparallel shear through the gneiss. Thirdly, D -D are associated with the generation of Kfeldspar, cordierite and garnet which resulted in a reduction in the volume of biotite and sillimanite available for foliation development. This would also make it difficult to propagate new penetrative foliations across the matrix between folia of the S2 fabric. Instead, any favourably oriented pre-existing foliation would have preferentially reactivated, with axialplane-parallel crenulations being preserved in fold hinges and on the margins of poiphyroblasts coarser than the average spacing of the reactivating foliation. References 2

Bell, T. H. and Johnson, S. E., 1989. The role of deformation partitioning in the deformation and recrystallisation of plagioclase and K-feldspar in the woodroffe Thrust mylonite zone, central Australia Journal of Metamorphic Geology, 7: 151-168. Hopgood, A. M. and Bowes, D. R., 1990. Contrasting structural features in the granulite-gneiss-charnockitegranite complex, Lake Baikal, U.S.S.R.: evidence for diverse geotectonic regimes in early Proterozoic times. Tectonophysics, 174: 279-299. Passchier, C. W., Myers, J. S. and Kroner, A., 1990. Field geology of high-grade gneiss terrains. SpringerVerlag, Berlin, 150 pp. Wenk, H.-R. and Christie, J. M., 1991. Comments on the interpretation of deformation textures in rocks. Journal of Structural Geology, 13: 1091-1110.

CRENULATION CLEAVAGE OR STRAIN-SLIP CLEAVAGE? B.E. Hobbs. Y. Zhang and A. Ord CSIRO, Exploration and Mining, Underwood Avenue, Floreat Park WA 6014 This paper summarises the results of finite difference modelling of the evolution of spaced cleavages during the folding of multi-layer "sequences with elastic-plastic and elastic-viscous rheologies. The constitutive behaviour also allows brittle failure parallel to the cleavage planes and subsequent slip if the stress conditions permit in elastic-plastic materials. In the early stages of deformation, strain is homogeneously distributed in the competent and incompetent layers. After a critical bulk shortening, however, one of two classes of strain localization begins within the incorporated layers. If the anisotropy of mechanical properties within the incompetent layers is low then a spaced cleavage develops which originates as shear banding at a high angle to the principal axes of strain. This type of spaced cleavage is what would have been called strain-slip cleavage in the classical literature and is commonly associated with layer parallel shear zones. If the anisotropy is high within the incompetent layers then a micro-folding instability nucleates and the spaced cleavage which develops is initially parallel to a principal plane of strain but rotates away from this plane as the folding accentuates. This second type of spaced cleavage is what is called crenulation cleavage in the classical literature. At high fold amplitudes neither type of spaced cleavage is parallel to a principal plane of strain although the departure from parallelism can be low. The paper describes the morphology of these two types of spaced cleavage and the controls on their development exerted by rheology and fluid flow.

EXTENSIONAL DEFORMATION OF CONTINENTAL CRUST PRIOR TO BREAK-UP AND SEA-FLOOR SPREADING: A STUDY OF MISIMA ISLAND, PAPUA NEW GUINEA E.J. Hill School of Earth Sciences, Macquarie University, Sydney, NSW 2109, Australia The youngest known metamorphic core complexes in the world occur as a number of small but topographically rugged islands known as the D'Entrecasteaux Islands, in eastern Papua New Guinea (Fig. 1). These complexes lie in a zone where sea-floor spreading is actively propagating westward into thickened continental crust (Taylor et al., 1995, Nature, 374, 534537). Sea-floor spreading about an east-west axis has been occurring over the last 5 Ma creating the Woodlark Basin, which is flanked to the north and south by continental ridges (Fig.l). Seismic activity on the ridges indicates that they are undergoing active continental extension (Abers, 1991, Geology, 19, 1205-1208). THE D'ENTRECASTEAUX ISLAND METAMORPHIC CORE COMPLEXES

A detailed study by Hill, Baldwin & Lister (1992, Geology, 20, 907-910) revealed that the D'Entrecasteaux Islands consist of high-grade metamorphic rocks (eclogite to amphibolite facies), which are separated from overlying mafic and ultramafic rocks (believed to be remnants of a mid-Tertiary obducted ophiolite, the Papuan Ultramafic Belt) by ductile shear zones and faults. Kinematic indicators show a normal sense of movement on these shear zones and faults. The high-grade metamorphic rocks and the mafic-ultramafic rocks are therefore considered to represent the lower and upper plate plates of the metamorphic core complex, respectively. In addition, there are some low-grade metamorphic rocks (greenschist facies) on the easternmost island of the group (Normanby Island). The relationship between these lowgrade metamorphic rocks and the metamorphic core complexes was not resolved during this earlier study. Because the metamorphic core complexes of the D'Entrecasteaux Islands lie just beyond the extreme western extent of the sea-floor spreading centre (Fig. 1), we would expect that continued propagation would rift the group of islands in two. Although the metamorphic core complexes are 2-4 million years older than the nearest sea-floor, their common east-west alignment and close spatial association suggests a common origin (Hill, Baldwin & Lister, 1995, J. Geophys. Res., 100(B6), 10441-10451). If the formation of these complexes is related to the development of the sea-floor spreading system, then we might expect to find older metamorphic core complexes along the flanks of the older sea-floor, to the east. To test this hypothesis a recent trip was made to Misima Island which lies along the southern continental flank of the Woodlark Basin (Fig. 1) and is known to consist largely of metamorphic rocks. MISIMA-WOODLARK ISLAND METAMORPHIC CORE COMPLEX

This reconnaissance investigation demonstrated that Misima Island is very structurally similar to the D'Entrecasteaux Islands. On Misima Island a high-grade metamorphic basement was also found to be separated from overlying rocks by a mylonitic ductile shear zone and sub-parallel faulting. However, the hanging wall consists of low-grade (greenschist facies) metamorphic rocks. Kinematic indicators suggest oblique-normal slip on the ductile shear zone. The upper plate of mafic-ultramafic rocks seen on the D'Entrecasteaux Islands do not crop out on Misima Island. However, if the Woodlark Basin is reconstructed to its pre-rifting configuration, Woodlark Island (which currently lies on the northern flank of the Basin, Fig. 1) would be connected to Misima Island. Basement outcrop on Woodlark Island consists of mafic rocks, which are reported to be very similar to those of the Papuan Ultramafic Belt (Ashley & Flood, 1981, J. Geol. Soc. Aust., 28, 227-240) and are therefore likely to be the same as the

upper plate rocks on the D'Entrecasteaux Islands. It can be inferred that Woodlark Island is the remnant of the upper plate of the Misima metamorphic core complex, and that sea-floor spreading subsequent to the formation of this complex rifted the upper plate from the rest of the complex (Fig. 2). The predominant structural trend on Misii$a Island is WNW/NW-ESE/SE, parallel to the shear zone which separates the high-grade from the low-grade metamorphic rocks. Parallel coastlines bound the S W side of Woodlark Island and NE side of Misima Island, and these may reflect the original bounding structure between the low-grade metamorphic rocks on Misima and the mafic upper plate rocks on Woodlark Island. The low-grade metamorphic rocks appear to be sandwiched between the high-grade metamorphic rocks of the lower plate and the mafic rocks of the upper plate, thus forming a previously undocumented middle plate in the metamorphic core complex (Fig. 2). These observations point to the possibility that the low-grade metamorphic rocks on Normanby Island also represent middle plate rocks.

Solomon Sea

1000 m bathymetric contours

Woodlark

D'Entrecasteaux \ lj

n /

> Woodlark Island

l \1 Islands

Rises s

Basin XMisima Island.

Pocklihgton

200 km scale

Coral Sea

Figure 1 Location of the Woodlark Basin and continental islands mentioned in the text. (Geometry of Woodlark Basin after Taylor et al.f 1995)

Continental crust above sea-level

Continental crust below sea-level

Woodlark Basin oceanic crust

pre-rift stage post-rift stage

lower plate

middle Plate

PPer

Plate

Figure 2:

Woodlark Basin

Hypothesized evolution of the Misima-Woodlark metamorphic core complex

Misima Island

Woodlark Island

MAFIC DYKES AS DEFORMATION S T R U C T U R E S J.D. Hoek School of Earth Sciences, The University of Melbourne, Parkville, Vic 3052, Australia Mafic dykes are deformation structures that have long been neglected by most structural geologists. In this talk I will argue that dykes deserve our attention by drawing attention to some fundamental observations and their implications, for the role of mafic dyke swarms in crustal extensional geodynamics, and by briefly outlining some aspects of dyke emplacement. Most mafic dykes in the continental crust form part of a mafic dyke swarm, consisting of thousands of near-vertical, parallel dykes of rather constant composition, commonly accommodating 5-30 % of crustal extension. Detailed structural analyses, e.g. of dyke swarms in the Vestfold Hills (Passchieret al., 1991; Hoek, 1994, 1995; Hoek & Seitz, in press) have shown that dyke swarm emplacement is not accompanied by ductile or brittle deformation of the host-rock. Field observations and theoretical arguments (e.g. Hoek, 1994) indicate that, commonly, dykes propagate in a plane normal to the smallest compressive principal stress a3, creating their own fracture rather than being injected in pre-existing joints. Dilation is caused by a positive driving pressure applied from within the fracture, defined as AP = P m - a3, where P m is the magma pressure. Many continental dyke swarms consist of tholeiitic basalt. Magmas of these compositions are commonly denser than typical crustal rock types such as granite or diorite (Fyfe, 1992). Assuming that dykes are fed by magma chambers at the base of the crust, with a magma chamber pressure equal to the lithostatic pressure at the base of the crust, dyke emplacement of such dense magmas is only possible when a3 is smaller than the lithostatic stress, i.e. when there is a horizontal extensional tectonic stress. This argument is in agreement with the observation that dyke swarms accommodate horizontal crustal extension. Since the elastic strain limit of rock under crustal conditions is in the order of 0.1-1%, the observation that dyke swarms accommodate up to 30% extension implies that the tectonic stress must be constantly re-applied after a dyke has been emplaced, and the fact that no non-elastic deformation occurs implies that the level of deviatoric stress required for dyke emplacement is lower than for brittle or ductile deformation (Parsons & Thompson, 1991; Hoek, 1994; Hoek & Seitz, in press). From these general observations, some questions emerge. [1] What is the mechanism of dyke swarm emplacement, how can it operate at an apparently low level of deviatoric stress, and is there a difference between dyke emplacement under high-grade and low-grade conditions? [2] What is the source of the extensional tectonic stress accommodated by dyke swarm emplacement? [3] What are the factors that determine whether (ultra)mafic magma, collecting at the base of the crust will lead to extensive mafic underplating or to dyke swarm emplacement and perhaps attendant flood basalt volcanism? It may be evident that such questions have a significant bearing on the geodynamics of the continental lithosphere, especially when the role of dyke emplacement as a mechanism for transport of thermal energy is taken into account. At present, world-wide, very few geologists are looking at dykes from a structural point of view. Yet, there is enormous potential for fruitful research on mafic dyke swarms. Two reasons for this high potential are the paucity of observational structural data on the one hand and, on the other hand, the relatively advanced state-of-the-art of theoretical analysis. The theory of dyke emplacement essentially applies concepts from existing, well established disciplines such as fracture mechanics and fluid mechanics. At present, these theories involve rather simplified geometrical description of individual dykes. Detailed observations of dykes in the field (e.g. Hoek, 1994, 1995) reveal the importance of small-scale irregularities that are

superimposed on the simplified large-scale view of dykes. Significantly, the existing theoretical framework is capable of incorporating such detail. Emplacement of an individual dyke involves the dilation of an isolated fracture of finite dimensions, and the elastic deformation of the host-rock. This problem can be adequately analysed using Linear Elastic Fracture Mechanics, leading to models for dyke emplacement that provide a direct relationship between fracture geometry, dilation geometry, and driving pressure distribution (Pollard & Segall, 1987). Existing applications involve the analytical solution of the driving pressure distribution responsible for the dilation geometry of a simplified dyke fracture geometry (e.g. Hoek, 1994, 1995). With increased computing power that is now widely available, it is equally possible, for example using finite elements or finite difference techniques, to analyse more realistic dyke geometries. Work on dykes in the Vestfold Hills (Hoek, 1994) has identified a variety of structural features including fracture process zones at arrested dyke tips, deformed host-rock bridges at offsets, and crosscut geometries between dykes, the formation of which is a function of the magnitudes of the lithostatic pressure and the deviatoric stress during emplacement. The fact that dyke emplacement involves elastic deformation allows their analysis in terms of the driving mechanism that caused the deformation: the state of stress. This is in marked contrast to the analysis of the more familiar deformation structures such as folds, foliations, or ductile shear zones. Due to the permanent, inelastic deformations that characterize those structures, there is no longer any straightforward relationship between the driving mechanism of the deformation and the final result. The main reason why dykes deserve more attention from the structural geological community lies in the potential of gaining considerable insight in the state of stress that leads to deformation. When we recognize that mafic dyke swarms do accommodate significant crustal deformation and start asking the obvious questions that follow from this recognition, the insights that will be gained will also benefit our understanding of other crustal deformation mechanisms that are, after all, also ultimately the result of an applied state of stress. References Fyfe, W.S., 1992. Magma underplating of continental crust. Journal of Volcanology and Geothermal Research, 50, 33-40. Hoek, J.D., 1994. Mafic dykes of the Vestfold Hills, East Antarctica An analysis of the emplacement mechanism of mafic dykes and of the role of dyke emplacement during crustal extension. Unpublished Ph.D. thesis, University of Utrecht. Hoek, J.D., 1995. Dyke propagation and arrest in Proterozoic tholeiitic dyke swarms, Vestfold Hills, East Antarctica, in: Baer, G. & Heimann, A. (ed.) Proceedings of the Third International Dyke Conference, Israel, September 1995. Hoek, J.D & Seitz, H.-M. (in press). Continental dyke swarms as tectonic indicators: An example from the Vestfold Hills, East Antarctica, in: Dirks, P.H.G.M., Hoek, J.D. & Passchier, C.W (eds.), Tectonics of East Antarctica. Precambrian Research. Parsons, T. & Thompson, G.A., 1991. The role of magma overpressuring in suppressing earthquake and topography: worldwide examples. Science, 253, 1399-402. Passchier, C.W., Bekendam, R.F., Hoek, J.D., Dirks, P.G.H.M. & Boorder, H. de, 1991. Proterozoic geological evolution of the northern Vestfold Hills, Antarctica. Geological Magazine, 128, 307-318. Pollard, D.D. & Segall, P., 1987. Theoretical displacements and stresses near fractures in rock: with applications to faults, joints, veins dikes, and solution surfaces, in: Atkinson, B.K. (ed.), Fracture Mechanics of Rock. Academic Press, London, 277-349.

TEAR FAULT TERMINATION OF THE FOLD-THRUST BELT IN THE NORTHERN NEW ENGLAND OROGEN R J. Holcombe, C.R. Fielding & C J. Stephens Department of Earth Sciences, University of Queensland, Qld4072, Australia The geometry of the northern New England Fold Belt in the Rockhampton region is dominated by Permo-Triassic thrusting of the Hunter-Bowen event. The Stanage Fault Zone is a complex dextral system of strike slip, oblique slip, and thrust faults that terminates a well-defined foldthrust belt in a tear fault system, displacing it offshore. Henderson et al. (1993) define and mapped the Stanage Fault Zone where it crosses the coast;, this paper examines the interaction between the extension to this fault zone, and the fold-thrust-belt around, and through, the Marlborough Block, some 50-100 km inland.

"Marlbomugh

Rockhampton

Mount

Be wen Foreland Basin Mainly Permian-Triassic structures \

thrust faults/ XW Marlborough % thrust tear faults

C-Tr plutons

Mainly Cretaceous-Tertiary structures

^nantifnrm ntlfrtrm open aantiform

Tertiary J fault basin

open synform

normal faults

Cretaceous fault basin GOZ = Gogango Overfolded Zone

multiple folds

Pervasive contractional structures (thrusts, folds, cleavage) occur for a distance of at least 100km inland from the coast in central Queensland (and local thrusts and minor folds persist

much further inland). Regional deformation is strongly heterogeneous, with considerable contraction focussed in the Gogango Oveifolded Zone (GOZ; see fig.), a narrow (~20km), west verging thrust-fold belt along the eastern margin of the Bowen Basin. The zone of strongly focussed deformation curves northeastward into a diffuse system of faults in the Marlborough area (the Stanage Fault Zone). Structures associated with this termination are complicated by the Marlborough Block, containing multiply deformed ophiolitic rocks, syntectonic metagranitic schists, and remnants of the Carboniferous? accretionary metasediments. Deformation is strongly partitioned both across and along the fold belt. Along the fold belt, tear faults separate compartments of fold-thrust packages. Section balancing is difficult because both folds and cleavage are common within thrust packages. Broad areas of steep to overturned dips, suggest that thrust propagation folds are common. Although several major thrusts can be defined, most individual thrusts have no more than a few tens or hundreds of metres of stratigraphic separation. The Marlborough Block is a thin (<2km), composite terrane riding on a very low angle, and even undulating, out-of-sequence thrust that overprints the earlier thrust packages. The basal thrust is a very brittle structure with little ductile behaviour, even within metres of the contact. In contrast, the Marlborough Block internally is an amalgam of numerous thrust packages with kilometre-scale ductile response in the rocks adjacent to each thrust. Kinematics of these thrusts are broadly similar to the kinematics of the footwall fold-thrust belt, and presumably the Marlborough thrusts derive from a deeper, more eastward, part of the same belt. The Stanage Fault Zone is a complex zone of linked faults that truncate the GOZ and the northern margin of the Marlborough block, and separate these allochthonous and paraallochthonous elements from a gently folded autochthonous terrane to the north. Faults include pure thrusts, strike-slip and oblique-slip vertical faults, and oblique-slip thrusts. The zone appears to have been a major tear fault system both for the nappe emplacement of the Marlborough Block thrust sheet, and for the in-sequence thrust belt in the footwall. Fergusson (1991) estimated shortenings of the order of 60% in the GOZ (an estimate we agree with on the basis of cleavage intensity) and an overall upper crustal shortening of 50-90km across the GOZ and Bowen Basin folded zone (see fig.). By matching equivalent terranes north and south of the Stanage Fault Zone we estimate thrust-tear fault contraction of about 30km across the GOZ itself. By matching deformed syntectonic terranes in the Marlborough block with the Broome Head Metamorphics described by Morand (1993) we would estimate about 50km translation on the Marlborough Block nappe system (in broad agreement with early estimates by Murray, 1974, and a 30 km translation cited by Leitch et al., 1994). References Fergusson C.L. 1991. Thin-skinned thrusting in the northern New England Orogen, central Queensland, Australia. Tectonics 10, 797-806. Henderson R.A., Fergusson C.L., Leitch E.C., Morand V.J., Rheinhardt J.J. & Carr P.F. 1993. Tectonics of the northern New England Fold Belt. In Flood P.G. and Aitchison J.C., eds New England Orogen, eastern Australia. Department of Geology and Geophysics, University of New England, Armidale, 505-515. Leitch E.C., Fergusson C.L. & Henderson R.A. 1994. Ophiolitic and metamorphic rocks in the Percy Isles and the Shoal water Bay region, New England Fold Belt, central Queensland. Australian Journal of Earth Sciences, 41, 571-579. Morand V.J. 1993. The Broome Head Metamorphics: high grade metamorphism in the northern New England Fold Orogen. In Flood P.G. and Aitchison J.C., eds. New England Orogen, eastern Australia. Department of Geology and Geophysics, University of New England, Armidale, 591-598. Murray C.G. 1974. Alpine-type ultramafics in the northern part of the Tasman Geosyncline - Possible remnants of Palaeozoic ocean floor. In Denmead A.K., Tweedale G.W. & Wilson A.F. eds The Tasman Geosyncline - A Symposium. 161-181. Geological Society of Australia, Queensland Division, Brisbane.

TECTONIC SIGNIFICANCE OF BEDDING-PARALLEL FOLIATION: A CASE STUDY IN BEIJING WEST-HILL, NORTH CHINA W. Huang*. W.L. Shan & Z.R. Fu, Department of Geology, China University of Geosciences, Beijing * now at AGCRC, VTEPS Department of Earth Sciences, Monash University, Vic 3168 Bedding-parallel foliation is one of the most common structural elements in many polydeformed terrains, particularly in the Proterozoic terrains. In most polydeformed terrains, the beddingparallel foliation is considered to be part of the Di structures, but its tectonic significance has not been fully understood (e.g. Black 1987). Revelation of the tectonic significance of the beddingparallel foliation may be the key to correctly reconstructing regional tectonic framework of a polydeformed terrain (e.g. the Mount Isa Inlier, Halls Creek et al.). However, in many cases, attempts on identifying the tectonic scenario of the bedding-parallel foliation are hampered by the fact that the bedding-parallel foliation is overprinted and/or reworked during later deformation events. In Beijing West-Hill, bedding-parallel foliation is little affected by later deformation. Thus the structural story in this area is considered to be an excellent example illustrating a coherent scenario of the development of bedding-parallel foliation The Beijing West-Hill southwest of Beijing is tectonically part of a young mobilised zone within the North China Craton (Yang & Yang, 1985). Three cover sequences (Neoproterozoic, Cambrian-Triassic and Jurassic) of the craton are separated by paraconformities. The Neoproterozoic mainly comprises limestone, dolomite and a little phyllite; the Cambrian-Triassic consists of limestone, phyllite and coal formation; the Jurassic sequence is composed of basicacid volcanic and volcaniclasts. The structural framework is mainly defined by a series of upright open folds and brittle fault zones, which developed during Indosinian-Yanshan Orogeny (195-137 Ma). Bedding-parallel foliation has differentially developed within all lithological units, and are in space associated with interlay shear zones and interlay recumbent folds. The bedding-parallel foliation regionally parallels or subparallels to the boundary of lithological units, although on local scale it cuts the bedding around the hinge of interlay recumbent fold. On the surface of bedding-parallel foliation, there is always a strong stretching lineation defined by elongated mineral grains. A simple shearing during foliation development is indicated by finite stain analysis in various lithological units (Huang, 1989; Shan et al., 1991). The style of deformation changes across different lithological units and different deformation zones. For instance, the competent layers such as the dolomite were boudinaged while the incompetent layers such as marl show ductile sheared and/or folded (Shan et al., 1991). The ductile shear zone laterally transfers into interlay recumbent fold. The variation of deformation style causes two interesting phenomena. The first phenomenon is that there is a dramatic change of stratum's thickness laterally. For instance, the thickness of the Qinbaikou Group (1000-900 Ma) around Huangyuan (where ductile shear zones dominate) is about one tenth of that west of Xianyonglin (where numerous interlay recumbent folds are identified), even similar lithological layers have been identified in both areas. The second phenomenon is that at many localities local upside-down strata are observed at the lower limbs of interlay recumbent folds, but overall the stratigraphic sequence is upright. The mineral assemblages that define the bedding-parallel foliation probably resulted from the earliest metamorphic event. Differential development of the bedding-parallel foliation corresponds to the metamorphic isograds. For instance, within the upper greenschist facies and amphibolite facies metamorphic zones, the bedding-parallel foliation developed intensively in every lithological unit; whereas in low greenschist facies metamorphic zone, bedding-parallel foliation has only developed within incompetent layers (Shan et al. 1991).

In summary, the structural features of bedding-parallel foliation and the relation between deformation and metamorphism suggest that the bedding-parallel foliation in Beijing West-Hill have probably resulted from intensive lateral flowing of rocks in mid-lower crust. The heat intrusion caused by tectonic-thermal mobilisation is probably the key factor igniting lateral flowing of rocks in mid-lower crust.

References Blake, D.H., 1987. Geology of Mount Isa Inlier and environs, Queensland and Northern Territory. BMR Bulletin 225. Huang, W., 1989. Microstructures and fabrics in the progressive deformed carbonates in Beijing West-Hill. GEOSCIENCE, 3:223-234. Shan, W.L., Song H.L., Fu, Z.R. and Ren J.Y., 1991. Principles, methods and practices of structural analysis. China University of Geosciences Press, 159pp. Yang, W.L. and Yang, S.N., 1985. Tectonics of China. Geological Press, Beijing, 341pp.

THE ROLE OF ELASTICITY IN VISCO-ELASTIC FOLD EVOLUTION G.W. Hunt1, H-B Muhlhaus2, B.E. Hobbs2 and A. Ord2 department of Civil Engineering, Imperial College London, U.K. Division of Exploration and Mining, CSIRO, Perth, Australia. It is commonly held amongst structural geologists that the folding of stratified rock is a slow process, and therefore is primarily a controlled viscous phenomenon. There is however a growing realisation, succinctly put by Price & Cosgrove (1990) for example at the end of the chapter entitied "Introduction to Folding", that much of what is observed may be due to inherent elasticity. In referring to the viscous studies of Biot (1965) and others, they comment "... their predicted results regarding L/a ratios have not fitted observed natural data. Only by treating rocks as highly nonlinear viscous materials has it proved possible to bring theory and field observations into line. This problem has arisen because it has been implicidy assumed, that elastic behaviour of rocks during the initiation of folding can be ignored" (after Price & Cosgrove (1990)). The talk will demonstrate one effect of introducing nonlinearity into the bedding relations, for a thin strip of elastic material supported within a visco-elastic (Maxwell) medium. We find that elasticity in the embedding medium provides the opportunity for "instantaneous" elastic behaviour to affect the form of deformation before viscosity and consequent dissipation have had time to play their part. Not only are wavelengths (L/a ratios) affected, but new structural forms can emerge; the way is open for some of the rich and varied bifurcation phenomena of elastic buckling (Hunt & Wadee 1991) to appear at the start of the evolutionary process, with continuing influence as time progresses. The response of the system, in a timescale associated with the rate of loading, mirrors that of pure elasticity, yet in the longer timescale is effectively governed by the viscous part of the embedding medium. What is unstable in the sense of the latter may be stable in the sense of the former, and both elastic and viscous parts are germane to what may be seen. In particular, with a softening nonlinearity in the bedding relation, the system setdes naturally at the start of evolution into a pattern of localised buckling, contrasting sharply with the strongly-periodic trend found in purely viscous formulations (Biot 1965; Muhlhaus, Hobbs & Ord 1994). Localised buckling has a wide range of spectacular variations, with a chaotic tendency, that is only now being understood in the purely elastic case; it will be demonstrated that the infinite set of allowable elastic solutions is interwoven with a set of unbounded (physically unrealisable) solutions, such that the boundary between them is fractal. Although the Maxwell representation of the bed is fundamentally that of a fluid, governed in the long term by viscosity, it is seen that for the limiting case of infinite length, the route to the fully relaxed state from an initial localisation remains inherently aperiodic; periodicity, so dominant for purely viscous embedding, never makes an appearance. Thus localised forms are seen to be an integral part of the geological timeframe.

REFERENCES Price, N.J. & Cosgrove, J.W. (1990), Analysis of geological structures, Cambridge: Cambridge University Press. Biot, M.A. (1965), Mechanics of incremental deformation, New York: Wiley. Hunt, G.W. & Wadee, M.K. (1991), Comparative lagrangian formulations for localised buckling, Proc. R. Soc. Lond. A 434,485-502. Muhlhaus, H-B., Hobbs, B.E. and Ord, A. (1994), The role of axial constraints on the evolution of folds in single layers, pages 223-231 of "Computer methods and advances in geomechanics, Vol 1" Siriwardane, H.J. and Zaman, M.M. (eds) Rotterdam: A.A. Balkema.

THE GEOMETRY AND KINEMATICS OF THRUSTS AND SHEAR ZONES FROM THE SOUTHERN ADELAIDE FOLD-THRUST BELT ON THE SOUTHERN FLEURIEU PENINSULA P.R. James, T. Flottmann, L. Barrett, D. Buhrer, A. Macdonald & R. Szmidel Department of Geology & Geophysics, University of Adelaide, South Australia, 5005 The study attempts to characterise contrasting structural responses to the Delamerian orogeny between a Cambrian paltform which forms the footwall to a sequence of major thrusts and a deformed Cambrian basin sequence in the hangingwall. Detailed geological and structural mapping was carried out in conjunction with Mines and Energy South Australia (MESA) and integrated within a Geographic Information System (GIS), also making use of recently acquired airborne geophysics of the South Australia Exploration Initiative. The aims of the mapping were to document the detailed geometry of shear zones and thrusts invoked by previous workers eg. Steinhardt (1991) & Flottmann et al. (1994). The area can be subdivided into three distinct zones. In the north, between Sellicks Beach and Normanville, upper Adelaidean quartzites and siltstones, and Cambrian carbonates form a steeply dipping to overturned homoclinal wedge above a major basal thrust. In the hangingwall, a leading imbricate fan displaces marker carbonates. Major chevron style propagation folds are similar to small-scale structures in the hangingwall of imbricate faults, and to the basal detachment. Displacement estimates calculated from balanced-section constructions and internal finite strains measured from deformed intraclastic carbonate conglomerates decrease higher in the thrust pile. Concomitantly the range of minor structures developed in association with the thrust complex, eg deformation stylolites, composite fabrics, minor folds, sygmoidal vein arrays etc. decrease in profusion and intensity. At the top of the thrust pile a major overturned synform in lowest Adelaidian conglomerates marks the transition to the intensely deformed zone, which consists of several thrusts that overall truncate the structures of the northern zone. The basal Normanville Thrust is superseded by the Normanville-Rapid Bay shear zone, where disrupted sequences of basement gneisses and retrograde phyllonites evidence largescale NW-directed lateral displacement. Major and minor structures include similarstyle folds, foliation boudinage, shear bands and extensional crenulation cleavages and a pervasive transposed LS tectonite fabric. This shear zone can be traced to the north and north east into the splayed shear and thrust zones of the Adelaide Hills, and to the south onto Kangaroo Island (Flottmann et al, 1995). In the southern zone, overlying the sheared Normanville Group carbonates of Rapid Headland, major units of the Adelaidian and Normanville Group are repeated in largescale recumbent isoclinal folds. These folds are discordant with low strain uniformly dipping Kanmantoo group metasandstones and metasiltstones. Minor thrusts and shear zones along with asymmetric folds indicate a single uniform north west vergence. References Flottmann T., James, P.R., Rogers, J. & Johnson, T., 1994. Early Palaeozoic foreland thrusting and basin reactivation at the south eastern Palaeo-Pacific margin of the Australian Precambrian Craton: a reappraisal of the structural evolution of the Southern Adelaide Fold-Thrust Belt. Tectonophysics, 234, 95-116. Flottmann, T., James, P.R.,Menpes, R.,Cesare, P., Twining, M., Fairclough, M., Randabel, J. & Marshak, S., 1995. The Structure of Kangaroo Island (South Australia): strain and kinematic partitioning during Delamerian basin and platform reactivation. Australian Journal of Earth Sciences 42, 35-49. Steinhardt, C., 1991. The microstructural anatomy of a major thrust zone on Fleurieu Peninsula, South Australia. Australian Journal of Earth Sciences 38, 139-150.

TECTONIC ASSEMBLY OF INVERTED ISOGRADS IN CONVERGENT OROGENS R.A. Jamieson1, C. Beaumont2, J. Hamilton2, & P. Fiillsack2 1

Department of Earth Sciences, Dalhousie University, Halifax, JVava Scotia, Canada, 5/5 Department of Oceanography, Dalhousie University, Halifax, Nova Scotia, Canada, B3H 4J1

"Inverted" metamorphic sequences are characterised by metamorphic temperatures that increase structurally upwards, with isograds typically parallel to associated thrust faults. Well known examples include high-temperature metamorphic soles beneath obducted ophiolites (e.g., Jamieson, 1986, J. Met. Geol., 4, 3-22) and metamorphic inversions associated with large thrust zones in orogenic belts (e.g., Main Central Thrust, Himalayas; e.g., Hubbard, 1989, J. Met. Geol., 7, 19-30). Processes commonly invoked to account for the formation of inverted sequences include shear heating along thrust faults (e.g., England & Molnar, 1993, Tectonics, 12, 145-157), transfer of heat from a "hot" hangingwall to a "cold" footwall during thrusting (e.g., Peacock, 1987, J. Geophys. Res., 92, 12761-12781; Royden, 1993, J. Geophys. Res., 98, 4487-4507) and tectonic juxtaposition of higher and lower grade rocks during or soon after metamorphism (e.g., Jamieson, 1986; Reddy etal., 1993, Geol. Soc. Special Publication, 74, 375-389). Rapid cooling and/or exhumation are generally considered to be necessary to preserve inverted sequences from subsequent thermal overprinting. These sequences are often referred to as "inverted metamorphic gradients" with the implicit assumption that they record an inverted distribution of isotherms at the time of metamorphism. However, these sequences should only be described in terms of inverted thermal gradients where it can be demonstrated that isograds correspond to isotherms, that pressure is constant or decreases upwards, that the isograds are synchronous, that the sequence is continuous, and that the present distribution of isograds represents the distribution at the time of peak metamorphism. In many well known examples these conditions are either not met or cannot be demonstrated, and use of the term "gradient" is therefore inappropriate. We present results from fully coupled thermal-mechanical models for a convergent orogen that is controlled by a combination of subduction of sub-orogenic lithosphere and surface denudation (e.g., Willett et al., 1993, Geology, 21, 371-374; Fullsack et al., 1994, AGU Fall Meeting, Abstracts, p.700). All the models use a common set of thermal parameters corresponding to an initial temperature at the base of the crust of 633°C. The "basic" model has a relatively weak crust, a convergence rate of 1 cm/yr, and total erosion. At amounts of convergence corresponding to a few crustal thicknesses this model produces an abrupt increase in maximum temperature (interpreted in terms of peak metamorphic grade) at the model surface in the vicinity of the retroshear. The model pattern closely resembles data from typical inverted metamorphic sequences including the Main Central Thrust (Fig. 1). Similar results were produced from a range of models with relatively high convergence rates (1-10 cm/yr), weak to moderately strong crust, and high to moderately high erosion rates. Models with low rates of convergence, low rates of erosion, and rheologically weak crust produce broad temperature distributions that do not resemble typical inverted sequences.

Based on these models we suggest that inverted metamorphic sequences are a general consequence of syn-orogenic exhumation associated with crustal-scale retro-shear zones, at moderate to high rates of convergence and erosion. The model inverted sequences are defined by points that reach peak temperatures at different times and in different places within the model orogen, and thus reflect tectonic juxtaposition rather than inverted thermal gradients. In fact, only the model with the highest convergence rates (10 cm/yr) produced inverted isotherms within the model crust. Although we do not argue that all inverted metamorphic sequences must result from tectonic assembly, we suggest that our models represent a reasonable way of accounting for many of their characteristics, particularly where continuity cannot be demonstrated.

800 O 7 0 0 C.600

x 500 <S 400 £ 300 - Model C6B 200

100

-20

distance (km)

Figure 1. Results of the basic model compared with metamorphic temperature data from the Main Central Thrust south of Mount Everest (Hubbard, 1989). Model temperatures are normalized to initial temperature at the base of the crust. Model parameters: crustal thickness = 35 km; temperature at base of crust = 633°C; convergence rate = 1 cm/yr; erosion = 100%; pyroxenebased rheology; results shown for 17.5 My, corresponding to convergence of 5 crustal thicknesses; solid circle (S) corresponds to model singularity (crust-mantle detachment point).

MODELLING OF FLUID MIGRATION ALONG FAULTS IN DIFFERENT STRUCTURAL ENVIRONMENTS Jiang, Zh. 1, Oliver, N.H.S.2, Barr, T.D. 1, Power, W.L.3 and Ord, A. 3 1

Monash University, VIEPS, Department of Earth Science, Clayton, Victoria, 3168 Curtin University School of Applied Geology, WA 600i. 3 CSIRO, Division of Exploration and Mining, Nedlands, WA 6009

For fault-controlled fluid migration during hydrothermal ore genesis, the formation of fracture networks is a primary structural condition. Not only does the progressive linking of fracture zones form appropriate long-distance channels for fluid migration, but it also provides the appropriate physical condition for ore deposition at particular structural sites. Why do the deposits usually occur in areas of relatively complex structure? Our modelling based on actual ore field data1 indicates that there are different fluid migration characteristics which vary with the local structural environment and may affect ore deposition. We used UDEC2 to model fluid flow in the Malage ore field of Gejiu, China (Fig.l). Flow velocities are increased gradually in time and space on the relatively connected faults, such as the NW-SE and E-W(western half) trending faults in the centre of the ore field (see Figs.l, 2 and 3). These are the main pathways for the ore-forming fluid. In the smaller secondary faults, flow velocities fluctuate with time and space, reflecting the variation of physical conditions such as fault aperture, fluid pressure etc., some of which are favourable to mineral deposition. These faults include the main ore deposition sites, such as the NW-SE branch fault (b-b') and the east section of the E-W fault (c-c")(see Figs 1, 2, 4 and 5). The detailed variation of fluid and aperture conditions modelled in this complex fault array can be related to the formation and location of mineral deposits, both predictively and conceptually.

Maximum flow rate: 2.0X10' Faults with fluid flow: flow rate is propertional to thickness of shaded line Faults with negligible fluid flow

y %

Fluid flow direction

\//A

Crest of granite stock Differential stress contours, MPa

geologic contacts Localities where fluid flow are measured Fig.l

Points of fluid injection

Flow rates and flow directions in Malage orefield

0 1

0.5

1 km

£ 20 E

0.6

"(U

itwI

5 10

-JV 1

0 )

2000 4000

'u _o

hO.1

-3.5 -

Q- 0.0035 -

-2.5

0.003 -

0.0025

0.0 1 1 6000 8000 10000

-100

100

200

300

400

500

600

Fault segment a to a" (m)

Calculation Steps (model time) Velocity(a')(m/s) —

"E 0.0045 a) 3 0.004

0.2 >a

0.005 -

/—"s

0.3

0.0055 -i

N(/>

0.4 E

/ /V» ////^ V!

^ 15

1-0.5

/;/."

VM/

Aperture(m)

Velocity(c')(m/s)

Flow Velocity(m/s)

Velocity(b)(m/s)

Fig.3 Variations of fault aperture and flow velocity along ore-feeding fault a to a" during the main metallogenic stage (at calculation step 2800)

Fig.2 History of flow velocities at the point a', c' and b respectively

r-0.070

0.0013 -i 0.0012

-0.065

E? 0.0012 £

0.0011

-0.055

0.0010 0.0010

-0.060 'uo CD

0.0012

% 0.0011 <

\V)

100

200

300

400

500

o£ UL

•0.050 600

0.006 0.005 E, <D 0.004 D 0.003 k.a <Q. 0.002 0.001 0.000 -

3.5 3.0 -2.5

-2.0

4->

-500

-1.5 1.0 0.5

1 1 1 1 1— 0.0 500 1000 1500 2000 2500 3000 Fault segment c to c" (m)

Fault segment b to b' (m) Aperture(m)

Aperture(m)

Flow Velocity(m/s)

Flow Velocity(m/s) Fig. 5 Variations of fault aperture and flow velocity along the ore-housing cross fault c to c" during main metallogenic stage (at calculation step 2800)

Fig.4 Variations of fault aperture and flow velocity along the ore-housing branch fault b to b1 during main metallogenic stage (at calculation step 2800)

References 1. Sun, J., Jiang, Zh., and Lei, Y., 1988, Tecto-geochemial features of Malage orefield, Gejiu mining area. Chinese Journal of Geochemisuy,Vol.7, No.4, p. 311-320. 2. Itasca consulting group, Inc., 1994, Universal Distinct Element Code, Version 2.0 82

3-D COMPUTER MOVIES OF PORPHYROBLAST MICROSTRUCTURES Scott E. Johnson1 & Ross R. Moore2 1

School of Earth Sciences, Macquarie University, New South Wales, 2109y Australia School of Mathematics, Physics, Computing and Electronics, Macquarie University, New South Wales, 2109, Australia 2

The 3-D shapes of geological structures provide an important constraint on the processes by which-they can form. With few exceptions, exposure of structures is practically 2-D, particularly microstructures such as those seen in and around porphyroblasts in thin sections. The interpretation of porphyroblast microstructures is presently controversial, and new information is required to move the debate forward. In an attempt to better understand two well-known porphyroblast microstructures, we have made serial thin sections of two different rocks that contain, respectively: 1) garnet porphyroblasts with spiral-shaped inclusion trails (Fig. 1), and 2) plagioclase porphyroblasts with inclusion trails that form outwardly opening, oppositely concave microfolds, commonly referred to as "millipede" microstructure (Fig. 2). The serial sections were spaced at 1.25-1.5 mm intervals on average, and we commonly obtained 6 to 12 sections through individual porphyroblasts.

Figure 2. Sketch showing outwardly opening, oppositely concave microfolds of S \ in and around a plagioclase porphyroblast, defining a "millipede" microstructure.

Figure 1. Sketch of a garnet porphyroblast with smoothly-curving spiral-shaped inclusion trails.

A "notebook" - a commented package that contains source code to perform various routines designed to reconstruct 3-D surfaces from 2-D curves collected in parallel serial sections through these surfaces was written for the program Mathematica (Wolfram Research Inc., USA). This notebook can be applied to many continuous surfaces for which data is collected as curves from parallel serial sections, and the resulting reconstructions provide an excellent resource for both teaching and research. Mathematica allows reconstructed 3-D surfaces to be viewed in any orientation, and sectioned in any orientation to any depth. It also has an animation capability that allows "movies" to be generated, which greatly helps to clarify the 3-D shapes of complex surfaces or groups of complexly interconnected surfaces. For the garnet porphyroblasts with spiral-shaped inclusion trails, we reconstructed the central spiral-shaped inclusion surface in a single porphyroblast (Fig. 3; see Johnson, 1993, and Moore & Johnson, 1993, for details). We then generated a movie that shows the surface rotating through 360° about its spiral axis, fully revealing its shape. For the plagioclase porphyroblasts with "millipede" microstructure, we reconstructed a single porphyroblast and 83

five inclusion surfaces that pass continuously from the porphyroblast into the surrounding matrix (see Johnson & Moore, 1996, for details). We then generated three movies that show progressive serial sectioning of these surfaces in three mutually orthogonal directions, revealing the shapes of the oppositely concave microfolds that define the microstructure. Figure 4 shows a reconstruction of the right-end S\ foliation surface that passes through the plagioclase porphyroblast in Fig. 2, illustrating the non-cylindrical nature of the microfolding. The left-end S\ foliation surface looks similar, but bulges in the opposite direction. The reconstruction of this "millipede" microstructure, particularly when viewed as a movie, has helped to demonstrate that it formed by heterogeneous extension parallel to S2 around the rigid porphyroblast.

Figure 3. 3-D representation of the central inclusion surface from a garnet porphyroblast with spiralshaped inclusion trails similar to that in Fig. 1. Numbers along the margins are true dimensions (mm).

Figure 4. Reconstruction of the right-end S1 foliation surface that passes through the plagioclase porphyroblast in Fig. 2. The bulge in the middle points towards the reader, the dark region to the left being a "valley".

Creating movies has significantly increased our understanding of the 3-D geometry of the two microstructures discussed above. Applying these techniques to a wider range of porphyroblast microstructures will enable us to better understand their formation, which hopefully will increase our general understanding of a number of relevant problems, including: 1) relative timing between deformation and metamorphism, 2) mechanisms of folding and crenulation-cleavage development, 3) shear senses in deformed metamorphic rocks, and 4) rates of tectonometamorphic processes.

References

Johnson, S.E., 1993. Unravelling the spirals: A serial thin section study and three-dimensional computer-aided reconstruction of spiral-shaped inclusion trails in garnet porphyroblasts. Journal of Metamorphic Geology, 11, 621-634. Johnson, S.E. & Moore, R.R., 1996. De-bugging the "millipede" porphyroblast microstructure. A serial thinsection study, 3-D computer animation and strain analysis. In press, Journal of Metamorphic Geology, 14, No 1. Moore, R.R. & Johnson, S.E., 1993. Reconstructing inclusion surfaces within metamorphic garnet crystals. The Mathematica Journal, 3, 70-75.

EASTERN MARGIN OF THE BOWEN-GUNNEDAH-SYDNEY BASIN: GEOMETRY OF THE BURUNGA-LEICHHARDT-MOONIE-GOONDIWINDIMOOKI-HUNTER FAULT SYSTEM R J. Korsch7 & J.M. Totterdell2 Australian Geodynamics Cooperative Research Centre, Australian Geological Survey Organisation, GPO Box 378, Canberra, ACT 2601, Australia Marine, Petroleum & Sedimentary Resources Division, Australian Geological Survey Organisation, GPO Box 378, Canberra, ACT 2601, Australia

The eastern margin of the Bowen-Gunnedah-Sydney basin system generally has been regarded as a faulted margin and several names applied along its length namely, from north to south, the Burunga, Leichhardt, Moonie, Goondiwindi, Mooki and Hunter faults. In the past, the faults have been interpreted as having extensional, strike-slip or thrust displacements. Recent work by the National Geoscience Mapping Accord project 'Sedimentary basins of eastern Australia' has involved the interpretation of over 1000 seismic sections in the Bowen, Gunnedah and Surat basins in both Queensland and New South Wales. This has provided the opportunity to study the geometry of the eastern margin of the Bowen and Gunnedah basins in several seismic sections, even where it is concealed beneath younger sediments of the Surat Basin. The northernmost segment, the meridional Burunga Fault, is a west-dipping thrust, interpreted to be a backthrust associated with a basement duplex to the east (Totterdell & Korsch, 1992; Elliott, 1993). The Burunga Anticline, above the thrust, is a fault propagation fold, with shortening being accommodated by thrusting at depth, and by a combination of thrusting and folding in the upper levels of the sedimentary pile as the thrust tip propagated up the ramp. Timing of the thrusting is after deposition of the Middle Triassic Moolayember Formation but prior to the deposition of the Early Jurassic Precipice Sandstone of the Surat Basin, that is, in the Middle-Late Triassic. Near the southern end of the Burunga Fault, a displacement transfer zone separates it, by en echelon overlap, from an east-dipping thrust, the Leichhardt Fault, which is also meridional. In this case, the Bowen Basin succcession in the hanging wall block dips gently to the west, and eventually several kilometres east of the fault the succession is absent. Farther south, the Moonie Fault is not linked to the Leichhardt Fault, but strikes northeast and is a low-angle thrust fault with a flat-ramp geometry. The major movement on this thrust also occurred after deposition ceased in the Bowen Basin, but prior to commencement of deposition in the Surat Basin, that is, in the Middle-Late Triassic. The succession in the anticline above the thrust fault shows a typical fault-bend fold style, but has been mostly eroded. Hence the Surat Basin succession sits directly on a basement of Tamworth Belt immediately southeast of the fault, but with a remnant of the Bowen Basin occurring farther to the southeast. Frontal or short cut thrusts with limited displacement commonly occur. In the southern Bowen Basin the meridional Goondiwindi Fault occurs in southernmost Queensland and northernmost New South Wales. Its northern tip is separated from the Moonie Fault by en echelon overlap at a displacement transfer zone, and its geometry is very similar to that of the Moonie Fault. A frontal thrust system has propagated further to the west into the sedint pile than at the Moonie Fault, but only has a very limited amount of displacement.

Farther south, the eastern limit of the Gunnedah Basin is defined at the surface by the Mooki Fault, which thrusts Late Carboniferous rocks of the Tamworth Belt on top of Early Permian rocks of the basin. The relationship between the Goondiwindi and Mooki faults is not clear because of the lack of seismic data and concealment of the faults below the Miocene flows and plugs of the Nandewar Volcano. In the Werrie Syncline, Carey (1934) mapped the Mooki Fault dipping 40-50° to the east. Based on a ground-based magnetometer survey, Ramsay & Stanley (1976) calculated a dip of 25° to the east for the fault in the vicinity of the BMR deep seismic line (BMR91.G01). In this seismic profile, it appears to have a shallow dip of less than 30° to the east. Reflectors in the Gunnedah Basin can be traced for at least 15 km beneath the Tamworth Belt, which was thrust over the eastern margin of the basin for at least this amount (Korsch et a/., 1993). At the Kelvin Fault, to the east of the Mooki Fault, Devonian rocks are thrust on top of Carboniferous ones and there is a greater displacement than on the Mooki Fault. To the south, the two thrusts converge, suggesting that the Mooki Fault is a diverging frontal splay fault in front of the main Kelvin Fault. In a series of cross sections across the Tamworth Belt, Woodward (1995) shows 45-58 km of displacement on the combined Mooki-Kelvin thrust system. In the Sydney Basin, Late Carboniferous rocks of the Tamworth Belt have been thrust over Early to Late Permian ones of the basin at the Hunter Thrust. Here the thrust front has propagated at least 30 km into the basin beyond the western limit of rocks of the New England Orogen (Glen & Beckett, 1989), causing considerable deformation in the basin. In conclusion, a series of thrust faults occur close to the eastern margin of the Bowen-GunnedahSydney basin system, but are not physically linked, usually being separated from the each other by en echelon overlap at a displacement transfer zone. In most cases, Permian rocks of the basins occur to the east of the faults, which have propagated into the basin system from the New England Orogen. Thus, the present eastern margin of the basin system is usually an erosional remnant due to uplift of the hanging wall block. References Carey, S.W., 1934. The geological structure of the Werrie Basin. Proceedings of the Linnean Society of New South Wales, 59, 351-374. Elliott, L.G., 1993. Post-Carboniferous tectonic evolution of eastern Australia. APEA Journal, 33, 215-236. Glen, R.A. & Beckett, J., 1989. Thin-skinned tectonics in the Hunter Coalfield of New South Wales. Australian Journal of Earth Sciences, 36, 589-593. Korsch, R.J., Wake-Dyster, K.D. & Johnstone, D.W., 1993. The Gunnedah Basin-New England Orogen deep seismic reflection profile: implications for New England tectonics. In Flood, P.G. &Aitchison, J.A. (Editors), New England Orogen, eastern Australia, University of New England, Armidale, 85-100. Ramsay, W.R.H. & Stanley, J.M., 1976. Magnetic anomalies over the western margin of the New England foldbelt, northeast New South Wales. Geological Society of America Bulletin, 87, 1421-1428. Totterdell, J.M. & Korsch, R.J., 1992. Structural configuration and tectonic development of the Bowen and Surat basins in the Taroom region. Australian Bureau of Mineral Resources, Record, 1991/102, 35-52. Woodward, N.B., 1995 . Thrust systems in the Tamworth Zone, southern New England Orogen, New South Wales. Australian Journal of Earth Sciences, 42, 107-117.

LITHOSTRUCTURAL DOMAINS AND DEFORMATION OF THE EASTERN GHATS TECTONIC PROVINCE AND SOUTHERN SINGHBHUM CRATON, NORTHERN ORISSA, INDIA P.M. Leeming . L.B. Hams , C.R. Nash \ L.R. Rankin \ and W. Crowe 1

1 2

World Geoscience Corporation, 65 Brockway Rd Floreat 6014, Australia Geology and Geophysics, The University of Western Australia, Nedlands 6907, Australia y

Geological mapping at 1:100,000 scale using Landsat TM data has delineated a collage of Precambrian lithostructural domains within the Eastern Ghats Tectonic Province and adjacent Singhbhum and Bhandara cratons in northern Orissa, India (Nash et al. submitted, Tectonophysics). Granulite-facies para- and ortho-gneisses dominate in the Baliguda, Sadingia, Phulbani and Angul domains in the northern Eastern Ghats, whilst migmatitic amphibolite-quartzite sequences constitute the Rengali and Balangir domains. Metamorphosed supracrustal pelitic sequences have been recognised in the Rampur and Tickarpara domains. A supracrustal volcano-sedimentary sequence (including BIF) on Archaean granite-gneiss basement dominates the Singhbhum Craton. Bounding shear zones have undergone a complex history characterised by multiple reactivation. Although precise ages are not yet available, the relative sequence of events has been established (although it is possible that different domains may have different early structural histories). }

In the northern Eastern Ghats, the earliest structures deform a tectonic foliation S\ parallel to lithological/compositional layering. High grade gneisses (eg core of Phulbani Domain) comprise a series of gneiss domes deforming an early foliation formed during vertical loading/sub-horizontal (ENE-WSW?) extension. Locally, the early gneissic foliation has been folded about tight to isoclinal folds with sub-horizontal axial surfaces, either as part of a progressive deformation sequence, or during two separate periods of deformation (in the absence of geochronology, these structures are tentatively grouped as Di). Folding of Di structures in mafic/felsic granulites about upright to SW verging, doubly plunging folds (F2) with axial surfaces oriented NW-SE to N-S and N- striking, high grade, reverse ductile shear zones (such as the Tumidhibandh shear zone separating the Rampur and Phulbani domains) indicates E-W shortening (terrane accretion?) during a second event (D2) resulting in dextral transpression in the southern Eastern Ghats. In a subsequent event (D3), Fi & F2 folds were tight to isoclinally folded with NE-striking axial planes. Broad, dextral (± reverse) ductile shear zones striking 080° to 100° (eg. Ranipathna shear zone) and minor, approximately N-S sinistral ductile shear zones indicate NW-SE shortening, with indentation of the Phulbani domain resulting in flattening fabrics and sub-horizontal extension along its NW boundary. D3 is characterised by retrograde metamorphism, charnocitisation (especially along shear zones) and granitoid intrusion (the -1150 Ma age of Aftalion et. al, J. Geol. 96, 663-676, may date D3). Close to isoclinal folds with E-W axial surfaces and WNW dextral, NE sinistral and ENEreverse/thrust brittle-ductile to brittle shear zones (in part reactivating previous structures) indicate approximately N-S shortening/E-W extension in a fourth (?Neoproterozoic) event (D4) in which the Singhbhum Craton has acted as a rigid indentor into the Eastern Ghats, forming its main architecture. D4 shear zones are developed on the margins of gneiss domes

within the Phulbani domain which have also acted as indentors. The Kerajang Fault, traced for over 250 km, was initiated in this event, with probable reactivation in the PaleozoicMesozoic (associated with coal basin formation). Dextral displacement of approximately 100 km along the Kerajang Fault is suggested by correlation of Gangpur Group carbonate sequences (N Singhbhum) and the Sausar Group (Bhandara Craton). Undoing this displacement, the Eastern Ghats Boundary Fault separating the Eastern Ghats from the Bhandara craton may correlate with the Sukinda Thrust along the SE margin between the Singhbhum Craton and Eastern Ghats. Minor faults (SE sinistral and ENE dextral, ESE extension fractures and NNE reverse) define a fifth event (D5). NNE-striking dextral brittle shear zones, NE-striking normal faults and NW-striking reverse faults indicate approximately NE-SW maximum compressive stress with sub-horizontal NW-SE minimum principal stress during a more pervasive brittle event (Dg) in which there has been reactivation suitably oriented ductile and brittle-ductile shear zones. Events in Orissa appear to correlate in terms of timing, deformation regime and tectonic transport directions with events recognised in the Central Indian Tectonic Zone, Albany Fraser Province of Western Australia, and parts of East Antarctica (Harris, 1994,in: Findlay, R.H, et.aL, Gondwana 8, 165-180; Harris,1995, Mem.Geol.Soc. India, 33, 47-71), suggesting coeval deformation throughout this part of proto-East Gondwanaland.

GHATS TECTONIC

P R O V W C ^ ^ ^ ^ v .

* I Tateara Domain

1 9 I Angii Doman

2 j Rafeangpur Domain

| 10 | Tkarpara Dorr»n

3 | Purunapart Domain /pnaba

| 11 \ Rampcr D o n w

1 4 j Bonal Qranfcaftnaaa I

| 12 ) Ptmfcart Domain

5 1 Sngtttxjm Grwvta 1 13 | Sormopu- Gonfia

1 6 1 Iron Or* Group

I ' frO Boiangr Anorffrata

1 7 j Koftan Basin ( v

I Rangal Domain » (hign yada gnecaas)

("abl Domain lI fauoracnjsats

I Gonchwana Basin f

[ Ovirtsgam Basri

DEi Bhandara Granfta /gnats

AGE AND STRUCTURAL CHARACTERISATION OF THE TEXAS MEGAFOLD, SOUTHERN NEW ENGLAND OROGEN, EASTERN AUSTRALIA P.G. Lennox &P.G. Flood Department ofApplied Geology, University of New South Wales, Sydney, NSW 2052, Australia. - Department of Geology & Geophysics, University of New England, Armidale, NSW 2351, Australia. Domain analysis of the bedding and cleavage data around the Texas megafold demonstrates it is a tight to isoclinal, steeply inclined synform, plunging steeply to the southsoutheast (Fig. 1). The trace of the steeply inclined axial surface is composite consisting of north-south and northwest-southeast oriented segments. This is consistent with independent palaeomagnetic and lineament studies (Aubourg et al, 1994; Vinayan et al., 1993). 1

n Cover a) / ^ T j y New England vv Permian 1 Orogen ^ slope basins V — y ^ ^ A 7 ill Texas m beds

/Terrica

Silver Bpur

40km

Beck trends" Megafol| outline

Tasman Sea ' \ 151 °30' x

Coffs Harbour M Fig. 1 a) Locality map of the New England Orogen (NEO) in eastern Australia and b) generalised map of the Texas megafold with the relevant Permian basins identified. The Texas beds defining the Texas megafold consist of four tectonostratigraphic units formed from monotonous turbidites and minor chert, greenstone and limestone (Fergusson & Flood 1984). Regional mapping and recent detailed mapping has shown that the Texas beds consist of many steeply dipping fault slices (Olgers et al, 1974; Forster 1991). Unconformably overlying the Texas beds are several spatially isolated Permian slope basin deposits of different ages (Briggs, in press). The Alum Rock and Pikedale deposits are entrained parallel to the structural grain on the northeastern limb of the megafold, whilst the Terrica and Silver Spur deposits near the megafold hinge zone contain folds probably developed during megafolding. At Silver Spur macroscopic folds probably related to megafolding fold the underlying Texas beds, unconformity and overlying Artinskian-Sakmarian (Early Permian) sequence. This provides an important constraint on the timing of megafolding.

Three post-bedding foliations have been observed both in the field and thin section; the first is well developed and oriented subparallel to bedding, the second patchily developed foliation is usually at a high angle to bedding and probably developed during megafold formation and the third weakly developed foliation is oriented east-west. Only rarely are two of these three foliations detected in the same exposure. Often adjacent outcrops will contain a foliation at a low angle to bedding and another foliation. The first foliation is often a slaty cleavage, is axial surface to isoclinal microfolds within some thin sections and appears to be related to transposition in other thin sections. The second foliation outcrops as a spaced crenulation cleavage in rare exposures or as kinks in other exposures and in thin section is often an opaque seam cleavage. The New England region was part of a east-facing, almost meridional subduction system with the Texas beds, considered to represent part of an accretion-subduction complex in the Late Devonian to early Carboniferous. Changes in plate vectors at the end of the Carboniferous led to oblique subduction and back-stepping of the system, so that the New England area was affected by a dextral transpressional system for a short interval. This was resolved in the Texas area by the formation of an post-Early Permian megafold with poorly developed axial surface foliation. Continuing tightening of this structure led to the fanning of this foliation about the megafold. Megafolding is post Early Permian (post-Artinskian) since it folds Permian rocks of at youngest mid-Artinskian age at Silver Spur near Texas. This is contrary to existing tectonic models, which propose Late Carboniferous (Murray etal., 1987), late-Early Permian (Korsch & Harrington 1987) or Late Permian (Collins et al., 1993) for megafold formation. Therefore any tectonic model for the SNEFB must include Early Permian subduction. References Aubourg, C., Klootwijk, C. & Korsch, R.J., 1994. Magnetic Fabric and Palaeomagnetic studies in the Texas and Coffs-Harbour Blocks, New England Orogen. Australian Geological Survey Organisation Record, 1994/58. Briggs, D.J.C. (in press) Permian Productidina and Strophalosiidina from the Sydney-Bowen Basin and the New England Orogen: systematics and biostratigraphic significance. Association of Australasian Palaeontologists Memoir. Collins, W.J., Offler, R., Farrell, T.R. & Landenberger, B., 1993. A revised Late Palaeozoic-Early Mesozoic tectonic history for the southern New England Fold Belt. In Flood, P.G. & Aitchison, J.C. eds. New England Orogen, eastern Australi, pp. 69-84. Department of Geology & Geophysics, University of New England, Armidale, NSW. Fergusson, C.L. & Flood, P.G., 1984. A late Palaeozoic subduction complex in the Border Rivers area of southeast Queensland. Proceedings of the Royal Society of Queensland, 95, 47-55. Forster, P.,1991. Geology of a Transect Through the Pikedale Area, Border Rivers Area, Queensland. B.Sc. (Honours), University of New South Wales, Kensington. Korsch, R.J. & Harrington, H.J., 1987. Oroclinal Bending, Fragmentation and Deformation of Terranes in the New England Orogen, Eastern Australia. In Leitch, E.C. & Scheibner, E. eds. Terrane Accretion and Orogenic Belt, pp 129-139. American Geophysical Union, Boulder, Colorado, USA. Murray, C.G., Fergusson, C.L., Flood, P.G., Whitaker, W.G. & Korsch, R.J., 1987. Plate tectonic model for the Carboniferous evolution of the New England Fold Belt. Australian Journal of Earth Sciences, 34,213-236. Olgers, F., Flood, P.G., & Robertson, A.D., 1974. Palaeozoic geology of the Warwick and Goondiwindi 1: 250,000 sheet areas, Queensland and New South Wales. Australian Bureau of Mineral Resources Report, 164 109pp. Vinayan, P. K., Taylor, G. R., Balia, L. M., & Lennox, P. G. 1993. The Integration of Vector and Raster-Based Remotely Sensed Data for Geological Exploration. In E. G. Masters & J. R. Pollard (Ed.), Advanced Remote Sensing Conference, 1 (pp. 349-358). UNSW, Sydney: School of Geography, UNSW.

PORPHYROBLASTS - THE INSIDE STORY G. S. Lister Department of Earth Sciences, Monash University, Clayton, Victoria, 3168

There has been a decade of controversy in relation to the question as to whether or not porphyroblasts actually rotate. The debate has been marked by heated debate and poorly constrained argument. T.H. Bell suggested that "porphyroblasts do not rotate" but made no mention of the actual frame of reference. The earth rotates and thus so must porphyroblasts! Now "porphyroblasts do not rotate with respect to geographic coordinates". But even this framework is difficult to define. We cannot examine the problem in respect to improperly defined frames of reference. This becomes the starting point for a discussion as to the physical significance of the obervations that have been reported. Fabric and microstructural analysis offers a tool that can be carefully and objectively used to constrain the geological history of metamorphic tectonites. In this contribution a deliberate asttempt will be made to focus on the controversial elements of the Bell model with which specific issue can be taken. These issues are as follows: • foliations in general are not "shear foliations" as asserted; • the presence of additional crenulations adjacent to stress concentrators is not evidence for periods of pervasive deformation for which evidence in general has been erased by the process of reactivation; • reactivation of a foliation is a process that does take place, but not in the way described by Bell and his co-workers; • the sense-of-shear determined using porphyroblasts in general gives the "right answer" but not if we use the theory espoused by Bell; • folding is the result of buckling of competent layers, and the 'passively' superimposed effects of pervasive strain, not shear folding; • continuum principles can be usefully applied to the flow of rock masses without disadvantage. There are now several examples of case studies in which different lines of evidence have been used to constrain the movement picture, and these have produced unequivocal data in support of more classical interpretations of the nature and significance of microstructures observed in porphyroblasts.

AN ALGEBRAIC TECHNIQUE FOR MEASURING THREE-DIMENSIONAL DISPLACEMENT GRADIENTS AND STRAIN IN BRITTLE SHEAR ZONES T.A. Little Geology Department, Victoria University of Wellington, P.O. Box 600, Wellington, New Zealand Many methods for analysis of natural fault-slip data invert for a deviatoric stress tensor that is best-fit with respect to the slip pattern observed on a sample of faults. For each sample, stress is assumed to have acted homogeneously. Combining data from multiple fault sets thus invalidates the procedure. Two important field observations are commonly ignored in such stress-based treatments: 1) the magnitude of offset or slip on a fault, and 2) the spatial position of the faults relative to one other. Where exposures of brittle deformation zones are excellent, for example along coastal cliffs adjacent to major faults, fault-slip data can be analysed using a more comprehensive, kinematic approach—an adaptation of Wojtal's (1989, Journal of Structural Geology, 11, 669-678) technique for measuring faulting-related finite strain. As a modification to his method, displacements are allowed to be three-dimensional, and computations are algebraic, rather than graphic. Coordinate axes, x and z, are affixed to the deformed rock parallel the strike and dip directions of the brittle shear zone (Fig. 1). The y axis is perpendicular to the boundaries of the brittle deformation zone. For each fault, a computer spreadsheet program uses an input of basic fault-slip data and the apparent stratal offset on the outcrop face to solve algebraically for a net-slip vector (Fig. 2). These are resolved into three reciprocal displacement components: Au, Av, and Az, that restore slip on the fault and return the rock to an undeformed state. Fault-slip vectors observed in a transect across an array of faults are summed to obtain curves of cumulative displacement (ZAu, SAv, XAw) vs. the traverse distance, y, perpendicular to the boundaries of the brittle deformation zone. The curves provide a visual record of the displacement field. Their linearity is a test of the statistical homogeneity of the faulting-related deformation within a given domain. Where the curves are approximately linear, their slopes correspond to the mean reciprocal displacement gradients, 3u/3y, dv/dy, and 3z/dy of the bulk deformation. If the deformation is assumed to occur in a constant-volume transpressive shear zone, these gradients can be used to construct the reciprocal deformation matrix, from which all the finite strain parameters may be calculated (e.g., Sanderson & Marchini, 1984, Journal of Structural Geology, 6, 449-458). Ratios of the displacement gradients provide data on the bulk kinematics of the zone (angles, a and p in Fig. 1). I illustrate the technique by applying it to fault-slip data measured in seacliffs adjacent to the active Awatere fault in New Zealand. There, offset marker horizons are abundant and wellexposed, but fault-plane striations were not observed on every fault. One solution to incomplete data is to use the subset of faults for which complete slip information is available to derive: 1) an empirical curve relating fault-plane strike to slip-vector pitch/slip-sense, and 2) an empirical curve relating gouge thickness to displacement magnitude. These relationships can be used to assign missing values to the fault-slip data set. By running the spreadsheet over a range of uncertainty in the assigned values, robustness of the solution may be assessed. Within 300 m of the dextral-reverse Awatere fault zone, slip on distributed subsidiary faults causes a large magnitude of finite transpression. Outside of this damage zone, faults in the seacliffs accommodate very small magnitudes of bulk oblique-extension. In conjunction with field observations of cross-cutting relationships, these data reveal a distinct change in kinematic style along this part of the Marlborough Fault system in the last 5 Ma. The method provides useful and quantitative kinematic information while avoiding some of the assumptions used by stress inversion techniques. Unlike geometric moment summation techniques for measuring faulting-related strain from transect data (e.g., Peacock & Sanderson, 1993, Journal of Structural Geology, 15, 1513-1516), strain is not assumed to be infinitesimal. Knowledge of the distribution of faulting-related finite strain adjacent to major crustal faults could provide new insight into mechanical processes of fault inception and growth.

Geographic Coordinates

Cumulative Displacement Components

Tectonic Coordinates

Fig. 1. Diagram illustrating concept of displacement gradients in faulted rocks and meaning of symbols used in text. strike and dip of outcrop face

length, sense of offset of bedding on outcrop face strike and dip of bedding

Fig. 2. Diagram showing outcrop observations used as input for calculation of net-slip vector. Net-slip vector is calculated trigonometrically from fault-plane triangle defined slip-lineation, bedding trace, and outcrop trace.

QUARTZ C-AXIS FABRICS NOT ALIGNED WITH ROCK LINEATIONS: A TECHNIQUE FOR VIEWING FABRICS IN THE XZ PLANE OF OBLIQUE FINAL STRAIN INCREMENTS Tyler MacCreadv Department of Geology and Geophysics, University of Wyoming, Laramie, WY 82071-3006, U.S.A. Current address: Department of Earth Sciences, Monash University, Clayton, Victoria 3168, Australia. To interpret c-axis stereonet patterns of deformed quartz the patterns must be viewed in the XZ strain plane. This plane is usually assumed to contain the rock lineation. This study describes several c-axis patterns from a region of complex, amphibolite facies deformation in the Ruby Mountains core complex, Nevada, USA, where final strain increments were oblique to finite strain. The quartz recrystallized during these final strain increments while minerals defining the rock lineation (sillimanite and feldspar) were unaffected. An empirically justified technique has been devised for rotating the quartz c-axis patterns away from the rock lineation to view them in the appropriate plane. The c-axis orientation data for this study was originally collected from thin sections cut parallel to the rock lineation. Stereonet plots of this original data showed atypical, and hence uninterpretable, fabric patterns. The proper orientation for each pattern was determined by rotating the c-axis data around the pole to the foliation until the c-axes within the foliation plane lay at the center of the stereonet. The rotation produces an internal symmetry in the patterns. In addition, the rotated patterns across the area show a more consistent geographic orientation than the rock lineations. This orientation is inferred to be the XZ plane for the final increments of strain on the basis of this geometric consistency. The rotated view plane is confirmed to be the XZ plane of the final strain increments by its alignment with the strain in a mylontic shear zone that represents the latest stage of plastic strain along the edge of the study area. The complete development of the quartz textures during the final strain increments implies that they record very little of the finite strain history. Quartz is the only mineral recrystallized during the late strain thereby providing insight into the geometry of strain increments that cannot be determined from other minerals. Example of technique for rotating c-axis data for viewing in the proper XZ strain plane:

original data

ciS rotation axis

(section || rock lineation)

rotated data ('typical' c-axis

DEVIATORIC STRESS AND THE INTERPLAY BETWEEN DEFORMATION AND METAMORPfflSM Neil Mancktelow Geologisches Institute ETH-Zentrum CH-8092 Zurich, Switzerland The structures studies by structural geologists (folds, boudins, faults, shear zones etc.) reflect a heterogeneity that is characteristic of rock deformation on all scales; This heterogeneity in strain reflects both the variation in material properties and a variation in deviatoric stress in both time and place during the deformation history. For non-hydrostatic stress, the "pressure" is taken as the isotropic component of the stress field producing volume change, related to this volume change through the compressibility, and is therefore equivalent to the mean stress (oi + G2 + c>3)/3. Variations in deviatoric stress must therefore be reflected in effective pressure gradients over the same length scale - the fundamental question has been and still is if the magnitude of these tectonically produced pressure gradients is large enough to be observable in the structural and metamorphic record. Firstly, it can be very readily established that even without the introduction of external tectonic forces, isostatic equilibrium in the lithosphere does not imply a lithostatic pressure distribution and that strictly lithostatic pressure will be the exception rather than the rule (e.g. Turcotte & Schubert 1982). The difference in vertical normal stress, and thus in pressure, between a point under high mountains and a point at the same depth (relative to the geoid) under the adjacent lowlands is fundamental to explaining the development of extensional structures in the internal mountain belt while thrusting is still active in the external foreland (Molnar & Lyon-Caen 1988). Considering the effect of tectonic stresses for the simplest case, where principal stress directions are either vertical or horizontal, the rock has a viscous rheology and z is taken as vertical, the pressure difference from lithostatic is given by: where the values of 8 range between 0 for oblate stress and 1 for prolate stress (for plane stress 8 = 0.5). Values of AP are clearly positive for compressive deformation (o^ > <J ) and negative for extensional deformation (cr < Cj). This is the basic reason why brittle normal faults may extend further into the lithosphere than brittle thrusts, since the brittle/ductile transition will be lowered by the "tectonic underpressure" inherent in an extensional regime and raised by the "tectonic overpressure" in a compressional regime (e.g. Sibson 1974). The magnitude of this divergence from lithostatic pressure depends directly on the strength of the lithosphere - a topic that is still hotly debated! Estimated values range from only 10's of MPa up to 150-300 MPa, depending on your particular prejudice. Perhaps more important, however, are local geometric effects due to the concentration of stress within specific loadbearing horizons or stress risers due to geometric confinement of highly incompressible material. The generally accepted proposition that the lithosphere is rheologically laminated, with the strongest, relatively thin layers occurring in the middle crust at the brittle/ductile transition and in the uppermost mantle, implies that these zones will act as load-bearing layers and that the effective pressure at these levels may therefore differ significantly from lithostatic during tectonic loading. It can readily be shown that, on any scale, strong layers imbedded in a weaker matrix result in a strong refraction of the principal stress axes into orientations within the layer near parallel or perpendicular to the layer walls. Again shortening along the layer will lead to an overpressure, the suppression of brittle behaviour, and the expulsion of fluids or diffusion of material away from the layer along the tectonically established pressure gradient. The opposite will occur during extension of the layer. This explains the commonly observed brittle or semi-brittle boudinage of layers on the limbs of folds during continued plastic deformation of those parts of the layers still being progressively shortened. It may also help explain retrogression/hydration of layers during extension, and the eventual cessation of this retrogression and preservation of unstable assemblages in the cores of boudins, since once separation has occurred the boudins are no longer internally extended and the pressure gradient disappears. It is also an explanation for Z

the common observation of vein material or, at higher temperatures, melt segregations within the necks of boudins. In the case of melt, both migration and formation, by decompression melting, is aided by the tectonic underpressure associated with boudin formation. Important non-lithostatic pressure gradients may also occur as the result of confined flow, the most important examples involving extrusion of material in a long but narrow zone between two stronger or "rigid" plates. This geometry is characteristic of all models for rigid indentation, sideways and vertical extrusion, "flower structures" and "orange-pip" extrusion of material from subduction zones, often suggested as the mechanism for rapid exhumation of deeply buried rocks. A simple consideration of the necessary force balance along such zones demonstrates that there must be a gradient in the normal stress parallel to the zone to counteract the integrated effect of resistance on the boundaries. This implies a pressure gradient in the same direction, pressure increasing in the opposite direction to the direction of flow. In fact, this necessary pressure gradient can be considered as the driving force for the extrusion. For lateral extrusion, such as proposed in the Eastern Alps and the Himalayas, the lateral pressure gradient will also result in an associated vertical component, strongest in the highest pressure region near the neutral surface, to produce tectonic windows (e.g. the Tauern Window and the syntaxes in the Himalayas). Strong non-lithostatic pressure gradients may also be established during subduction if the subducting material is confined between strong upper and lower plates. The principle is that of a journal bearing, where a thin fluid film is drawn into a narrowing channel and very large pressures can be developed in the lubricating film. The limitation will be the strength of the confining plates, and in a subduction zone this limitation is most likely given by the upper plate. At shallow subduction depths, the overlying accretionary prism will provide little constraint, but once the "backstop" is reached, and particularly below the base of the crust, where the upper constraint is provided by relatively "cold" lithospheric mantle, the strength of the adjacent plates may be sufficient to allow considerable overpressure development in the subduction channel. These effects may be locally intensified by the subduction of topography on the down-going plate (particularly seamounts). Overpressures would be promoted in compressive margins (shallow subduction angle, younger/less dense downgoing plate, no back-arc basin e.g. the Andes) compared to extensional margins (e.g. the Marianas). Changes in subduction geometry could result in the rapid elimination of the geometrically determined overpressures, and thus to near isothermal decompression that would be interpreted from the metamorphic record as a major exhumation event. In all these cases it is easy to demonstrate that tectonically developed pressure variation should be the rule rather than the exception - the fundamental question remains if the magnitude of these variations is geologically significant. This can only be answered when the rheology of rocks under geological conditions is better understood. However, as noted by Hobbs (1972) more than 20 years ago, the geometry of current rock mechanics experiments (i.e. simple uniaxial compression) precludes the determination of the full rheology, and in particular the non-linear components of the rheology that are critical to determining the pressure variation expected in, for example, heterogeneous shear zones such as subduction zones.

References

Hobbs, B.E., 1972. Deformation of non-Newtonian materials in simple shear. In: Flow and Fracture of Rocks (edited by H. C. Heard et al.). Am. Geophys. Union, Geophys. Monogr. 16, 243-258. Molnar, P. & Lyon-Caen, H. 1988. Some simple physical aspects of the support, structure, and evolution of mountain belts. Geol. Soc. Am. Spec. Pap. 218, 179-207. Sibson, R.H. 1974. Frictional constraints on thrust, wrench, and normal faults. Nature 249, 542-544. Turcotte, D.L. & Schubert, G. 1982. Geodynamics - Applications of continuum physics to geological problems. John Wiley & Sons, New York.

EMPLACEMENT AND DEFORMATION OF THE WONDALGA GRANODIORTTE, NEAR ADELONG, LACHLAN FOLD BELT, NSW B. Marshall1, B.J. Franklin1 & A.G. Tomkins2 department of Applied Geology, University of Technology - Sydney, PO Box 123 Broadway, NSW2007. 2 CRA Exploration Pty Ltd, PO Box 1559, Mt Isa, Queensland 4825. The NNW-trending Late Silurian I-type Wondalga Granodiorite (WG) abuts the Gilmore Fault Zone (GFZ) on the east and intrudes or is locally faulted against the Ordovician tholeiitic Nacka Nacka Metabasic Igneous (NNMIC) on the west (Basden, 1990). Between Adelong and Wondalga, the western contact zone comprises a tract of intensely foliated and mylonitized WG, aplitic dykes and later mafic dykes, together with foliated intimately associated NNMIC (Veness, 1973; Basden, 1990). The tract of mylonitised rocks is termed the Wondalga Shear Zone (WSZ) (Basden, 1990); it is the locus of much of the data here presented. The contact zone (up to 500m wide) is defined as the tract within which neither the WG nor the NNMIC exceed 90% by area, and the contact is marked where the two units are in approximately equal proportions. To the east of the contact zone, NNMIC quickly reduces to sparse, small (<20cm) rounded to ellipsoidal enclaves that are found throughout the WG. To the west, irregular fingers and patches of amphibole-rich WG in NNMIC occur up to 2km (map distance) from the contact. Field relationships are consistent with selective assimilation of solid NNMIC by WG melt, and concurrent metasomatism of the NNMIC. From east to west across the WSZ and the largely coincident contact zone, the geometry of the minor component changes. NNMIC enclaves marginal to the WSZ are ovoid and have rounded terminations; further into the WSZ they are increasingly common and tend to be elongate-lenticular with more angular terminations; and towards the middle of the WSZ they exist as planar dyke-like bodies (from 3mm-2m thick) interlayered with approximately equal amounts of leuco- and mesocratic WG. Further westward, the NNMIC layers become increasingly dominant, whilst the WG becomes discontinuous, and then progressively more discordant and irregular. Aplitic dykes and NNMIC dyke-like layers within the WSZ are foliated, boudinaged and folded. The changing geometry of the two main components partly reflects deformation in the WSZ, but, based on some layers being the form-surface for shearzone-related folds and also on strain estimates, much is inherited. The mineralogy of the NNMIC and Wagga-Omeo Zone (Scheibner, 1993) metasediments (WOZM) away from the WSZ requires low-pressure metamorphism at 650-700°C. Because the hornblende in mesocratic WG, and at reactionfrontsbetween WG and NNMIC enclaves, is optically identical with that in the NNMIC, the WG was emplaced under conditions approximating peak metamorphism. Where influenced by the WSZ, brown hornblende is rimmed by blue-green hornblende in both the NNMIC and the WG. This retrogression, together with the mylonitic microfabric of the WG in the WSZ, is in accordance with lowpressure metamorphism at 500-550°C. Vergence relationships for the principal foliation in the WOZM indicate that the WG occupies a shallow-plunging (5°-*SSE) regional antiform. Much of the WG has an ill-defined foliation, but the more intense fabrics are confined to the WSZ on the west of the intrusion and the GFZ 97

on its east. Within the WSZ, the WG and aplitic dykes are proto- to orthomylonitic (Wise et al, 1984) and exhibit vertical to steeply W-dipping S- or S-C foliations (Berthe et al, 1979), an intersection lineation, and elongation and streaking lineations in S and C. In the less metasomatised equigranular NNMIC, S is a hornblende-defined schistosity and C is generally absent; with increasing metasomatism, a more gneissic S overprints heterogeneous distributions of biotite and now-porphyroclastic plagioclase. The folded and boudinaged aplitic dykes and NNMIC layers have S as their axial plane foliation. Dyke-hingelines plunge either shallowly (10-40°) SSE or very steeply, depending on the dyke's original orientation; hingelines in the NNMIC layers mainly plunge 10-30° towards SSE. The elongation and streaking lineations approximate the plunge of the shallow hingelines. Thin (<0.5m) conjugate tracts of subvertically dipping ultramylonite trend approximately NNE and WNW, overprint S and C, and respectively exhibit shallow S- or E-plunging mineral lineations. S-C relationships in the WSZ indicate an oblique-slip, right-lateral, reverse displacement. The NNMIC therefore moved north and up with respect to the WG. A similar movement-sense exists on the GFZ to the east, but a more complex picture has been reported (Stuart-Smith, 1991). Dextral and sinistral displacements, respectively in the NNE- and WNW-trending seams of ultramylonite, are consistent with ongoing extension in the main shear zone. Unfolding the NNMIC layers and many of the dykes suggests a shallow SSE dip before WSZ development. This could have resulted in localised underplating of the NNMIC by the WG and thereby have promoted assimilation and metasomatism The deformed and complex contact between the NNMIC and the WG results from: (a) emplacement (by local underplating ?) of the WG into folded NNMIC under peak or near peak metamorphic conditions (low pressure at 650-700°C); (b) selective assimilation of suitable layers and parts of the NNMIC by sheet-like invasions of WG melt, together with metasomatism of more resistent NNMIC layers and bodies, still at near peak conditions; (c) activation (or reactivation) of the WSZ under retrograde conditions (low pressure at 500550°C), thereby resulting in shear-zone fabrics and mesoscale folding of aplitic dykes and NNMIC layering; (d) under very similar metamorphic conditions, partitioning of strain into conjugate extensional shear zones within the WSZ, followed by emplacement of mafic dykes. The proto-WSZ and -GFZ and the pre-Silurian (?) structure of the WOZM and NNMIC have seemingly controlled emplacement of the WG. References Basden, H., 1990. Geology of the Tumut 1:100 000 Geological Sheet 8527. New South Wales Geological Survey, Sydney. Berthe, D., Choukroune, P. & Jegouzo, P., 1979. Othogneiss, mylonite and non-coaxial deformation of granites: the example of the South Armorican Shear Zone. J. Struct. Geol., 1, 31-42. Scheibner, E., 1993. Structuralframeworkof New South Wales. Geol. Surv. NSW, Quarterly Notes, 93, 1-36. Stuart-Smith, P.G., 1991. The Gilmore Fault Zone - the deformational history of a possible terrane boundary within the Lachlan Fold Belt, New South Wales. BMR J. Austral. Geol. & Geophys., 12, 35-50. Veness, V.R., 1973. Metamorphic and plutonic geology of the Wondalga area, NSW. B Sc (Hons) thesis, Sydney University. Wise, D.U. et al (8 authors), 1984. Fault-related rocks: suggestions for terminology. Geology, 12,145-164.

STRUCTURE OF THE ALICE ANTICLINE, PAPUA NEW GUINEA: SERIAL BALANCED CROSS-SECTIONS AND 3-D RESTORATION R.A. Mason CSIRO Division ofExploration and Mining, Private Mail Bag Wembley, WA 6014 The structure of thefrontal,western part of the Papuan Fold Belt is characterised by basement involved structures attributed to inversion of earlier formed extensional faults. The Alice anticline formed due to inversion of an extensional fault system active during the Tertiary. Complex forelimb geometry of the Alice anticline changes markedly along strike and suggests different amounts of shortening in this part of the structure. Balanced cross-sections quantify differential shortening for different profiles of the anticline which range from approximately 25% to 5%. Palinspastic restoration of early normal faults and later developed thrusts demonstrates that such differential shortening is compatible between cross-sections. Three-dimensional restoration of the Alice anticline makes use of a series of balanced crosssections and is based on a line-length method. Paradoxically, this restoration reveals non-plane strain in the balanced cross-sections upon which it relies. However, the restoration also reveals and quantifies a component of rotation about vertical axes, which would not be obvious by application of conventional methods of structural analysis. Rotations about vertical axes are attributed to pinning of progressive, foreland directed deformation. The distribution of rotations about vertical axes suggest that zones of pinning are coincident with, and therefore due to, transfer zones or off-steps in the early extensional fault geometry. Two transfer zones associated with the original extensional geometry acted as obstructions to deformation and have effectively pinned contractional structures during their formation causing the rotations about vertical axes. Such rotations are not manifest in changing fold axis orientations and have only been derived from a consideration of material balance during deformation. A general fracture system is developed in rocks in the Alice Anticline area. This fracture system is often symmetrical about fold axis orientations and typically comprises a set of conjugate shear fractures and another set, interpreted as extensional, which is sub-parallel to the obtuse bisector of the conjugate set. Unfolding of bedding using the three-dimensional restoration results in more consistently oriented areal fracture sets indicating that fracture orientations are strongly influenced by rotations about vertical axes.

DEFLECTION OF NON-NEWTONIAN SIMPLE SHEAR FLOW AROUND A RIGID SPHERICAL BODY USING A FINITE ELEMENT METHOD Toshiaki Masuda1 & Naoya Mizuno Institute of Geosciences, Shizuoka University, Shizuoka 422, Japan 1 Present address: Department of Earth Sciences, James Cook University of North Queensland, Townsville, Q.4811, Australia We present examples of deflection of simple shear flow of non-Newtonian, power-law viscous materials around a rigid spherical body, and examine how it is influenced by the mangitude of the stress exponent (n). The analysed range of n was between 2 and 5 and the angular velocity of the sphere was half of the far-field shear strain rate, which is exactly the same as that for a Newtonian fluid. All the kinematic and dynamic parameters show characteristic distributions around the sphere. Contrasts between the maximum and the minimum values of pressure, differential stress, vorticity and kinematical vorticity number decrease with increasing n , whereas those of velocity vector and strain rate increase with increasing n . The change in these contrasts are as much as a few tens of percent. Thus, the deflection of simple shear flow of nonNewtonian viscous materials around the sphere is, as a first approximation, similar to that of a Newtonian one.

Particle paths.

X Shape of separatrix.

7= 0 n=3

n-5

7= 1

7=2

7=3

7=4

7=5 Strain ellipses, y: simple shear strain.

101

CHRONOLOGICAL CONSTRAINTS ON THE THERMAL EVOLUTION OF THE HARTS RANGE, ARUNTA INLIER J. Mawbv1. J. Foden1, S. Kelley2, & I. McDougall3 1

Department of Geology and Geophysics, University of Adelaide, Adelaide, SA 5005 Earth Sciences, Open University, Milton Keynes, UK RSES, Australian National University, Canberra, ACT 2600

The Palaeozoic Alice Springs Orogeny (400-300 Ma) in the southern Arunta Inlier is a good example of coupled basement-cover deformation involving Neoproterozoic cover and Palaeoand Mesoproterozoic basement (e.g. Collins & Teyssier, 1989). It is becoming increasingly recognised that the effects of the Alice Springs Orogeny in the basement involved significant thermal and structural reworking (Collins & Teyssier, 1989; Foden et al., 1995; Dunlap & Teyssier, 1995). In this contribution, we provide further evidence of Palaeozoic isotopic reworking of basement rocks in the Harts Range region of the eastern Arunta Inlier. Basement rocks in the Harts Range are divided into two domains juxtaposed along a major structural boundary. The Harts Range Group consists of amphibolite and sillimanite-bearing gneiss that record maximum P-T conditions ~700°C, 7-8 kbar and structurally overlies kyanite-bearing rocks of the Entia Dome Complex (700°C, > 8 kbar) (Arnold et al., 1995). Existing isotopic data suggests these probably represent Proterozoic P-T conditions (e.g. Foden et al., 1995). In the Harts Range Group, pervasive mylonitic reworking of coarsegrained peak assemblages was associated with overall south-directed transport with respect to the Entia Dome, and the development of an intense sub-horizontal stretching lineation (Collins & Teyssier, 1989). Mineral assemblages within the mylonites in the Harts Range Group suggest that reworking occurred at - 570-630°C and ~5-6 kbar. In the Entia Dome, spatially restricted sillimanite-bearing shear zones also show south-directed basement transport. These preserve essentially identical P-T conditions (570-590°C, 4.5-5 kbar) to the mylonitic fabrics in the Harts Range Group. High resolution 40Ar/39Ar ages obtained from hornblende in the Harts Range Group range in age from 360 to 370 Ma, and are slightly older than corresponding hornblende ages in the underlying Entia Dome Complex (330-340 Ma). Whether this age difference indicates that cooling through - 500°C occurred earlier in the structurally higher Harts Range Group, or reflects the presence of excess Ar (e.g. Dunlap & Teyssier, 1995) is not clear. What is apparent though is that mylonitic-reworking of the Harts Range Group and the Entia Dome at crustal depths of -15 km (Collins & Teyssier, 1989) is older than - 350 Ma. This is consistent with Foden et al., (1995) who suggested that cooling through 550-600°C in the Harts Range region occurred at around 450 Ma. On this basis, it seems likely that mylonitic reworking of the higher grade (Proterozoic?) assemblages occurred at this time. The kinematic similarity between these mylonite zones and greenschist facies deformation that interleaved basement and cover further south at —325-311 Ma (Dunlap & Teyssier, 1995) suggests that these zones may also be components of the Alice Springs Orogeny. In the Entia Dome Complex, muscovite 40Ar/39Ar ages cluster at 315 Ma, suggesting that ~150°C cooling occurred over an interval of -20 Ma. Subsequent cooling through K-Ar closure in K-feldspar occurred at 300 Ma. From these data, it seems apparent that the Entia Dome and Harts Range region remained hotter than the rest of the southeastern Arunta Block (e.g. Dunlap & Teyssier, 1995) till about 315 Ma. However until the timing and duration of higher temperature (550-600°C) structural features in the basement, with kinematics consistent with the Alice Springs Orogeny is constrained, many of the important details of basement reworking during this event remain obscure. 102

References Collins, W.J. & Teyssier, C. (1989) Crustal ductile fault systems in the Arunta Inlier, central Australia. Tectonophysics, 158, 49-66. Arnold, J., Sandiford, M. & Wetherley, S. (1995). Metamorphic events in the eastern Arunta Inlier, Part 1. Metamorphic Petrology. Precambrian Research, 71, 183-205. Dunlap, W.J. & Teyssier, C. (1995). Palaeozoic deformation and isotopic disturbance in the southeastern Arunta Block. Precambrian Research, 71, 229-250. Foden, J., Mawby, J., Kelly, S., Turner, S. & Bruce, D. (1995). Metamorphic events in the eastern Arunta Inlier, Part 2. Nd-Sr-Ar isotopic constraints. Precambrian Research, 71, 207-227.

103

MICROSTRUCTURAL CHANGES AND DEFORMATION DURING THE PHASE TRANSFORMATIONS IN SOLID AMMONIUM NITRATE. A.C. McLaren - and J.D. Fitz Gerald Research School of Earth Sciences, The Australian National University, Canberra, A.C.T. 0200, Australia Department of Engineering, The Australian National University Between room temperature and its melting point (169°C), ammonium nitrate exhibits four stable forms, designated IV, III, II, and I, respectively. The temperature ranges over which each these forms are stable at atmospheric pressure are shown in the figure below where the specific volume is plotted as a function of temperature. 1 2

"EO) E (E1)

0.64

£ 0.60

o oo c0QL

0.57

Temperature (°C) Form IV (distorted CsCl structure) and form HI (NiAs structure type?) are both orthorhombic and the transformations between these forms are reconstructive. On heating, HI transforms to II which is tetragonal and structurally similar to IV: consequently the EH to II transformation is also reconstructive. However, the II to I transformation is displacive; I is cubic with the CsCl structure. Forms II and IV are simply slight distortions of form I, brought about essentially by increasing rotational and/or translational order on cooling. The reconstructive transformations into EI from either II (on cooling) or from IV (on heating) occur only in the presence of some solvent for NH4NO3 (such as water), even in very small amounts. In perfectly dry specimens displacive transformations between IV, n, and I only occur. Although NH4NO3 is not a structural analogue of any important rock-forming mineral, it is, however, a very suitable material for studying the microstructural changes that take place during different types of solid state phase transformations. The microstructural changes that have been observed in thin films in the light microscope will be presented in a short video. Being reconstructive, the transformations involving form HI usually lead to a significant reduction in grain size. However, the "massive" transformations between II and I lead to significant grain growth. The II —> IV transformation is "martensitic" - small lenses of IV propagate rapidly across large grains of n, and then thicken steadily until all of the host grain is consumed. The II -1 transformations are extremely sensitive to a shear stress. The application of such a stress always induces the phase boundary between these phases (initially held stationary by a stable temperature gradient) to move so as to increase the volume fraction of the lowtemperature phase. Under certain conditions, stress can induce spectacular instability of the phase boundary. However, the motion of the phase boundary is apparently not sensitive to the orientation of the applied shear stress. These types of behaviour have all been observed in thin films of NH4NO3 mounted between two glass slides on a FUTRON™ in-situ deformation apparatus. In this case the plane of shear is parallel to the specimen film plane. The apparatus 104

has been modified firstly by replacement of the PID heater controller by a simple Variac to produce smooth ramps when changing temperatures. Secondly, the motor drive has been disconnected to allow for constant shear stress experiments to be conducted. For such experiments, the stress is applied through the action of a small mass suspended from the end of one of the glass slides via a wire that rides around a small pulley. Mass can be added or removed so that shear stress can be changed quickly and reasonably smoothly. In addition to microstructural changes, permanent change of shape (ie plastic deformation) of a polycrystalline specimen may also occur during a phase transformation if a shear stress that is less than the yield stress of the weaker phase is simultaneously applied. Repeated cycling through a phase transformation in the presence of such an applied stress can lead to very large strains, similar to those associated with superplasticity. This effect, commonly termed transformation plasticity, has been observed in many metallic systems. Recently, similar behaviour has been observed in ammonium nitrate and other ionic crystals in experiments using a simple dead-load creep apparatus (McLaren and Meike, in preparation). This apparent weakness of a polycrystalline ammonium nitrate specimen during a phase transformation is independent of the nature of the transformation (ie displacive versus reconstructive) and appears to be the result of the small external stress biasing the direction in which the relatively large internal stresses (associated with the volume change at the transformation) operate. It is these internal stresses that drive the deformation: there is no intrinsic weakness of the specimen during the transformation. Creep experiments on two-component nitrate systems in which only one component undergoes a phase transformation indicate that the dispersed, non-transforming phase reduces the transformation strain very much more significantly than would be expected if the effect were due simply to the relative volume ratios of the two components (McLaren, unpublished work). These observations on ionic crystals may provide important geophysical information about deformation in the neighbourhood of the major discontinuities associated with phase transformations in particular in the transition zone of the Earth's mantle.

105

DELAMERIAN FOLD THRUST BELT - A NEW TECTONIC MODEL D.T. Miller

School ofEarth Sciences, Flinders University ofSA, GPO Box 2100, Adelaide SA 5001 The Delamerian Fold Thrust Belt is a major geological province within South Australia, extending from Kangaroo Island in the south to the Olary District in the north east and beyond the Flinders Ranges in the north. The geology of this region has excited the interest of industrialists, governments and academic geologists since the historically and economically important discoveries of metal deposits and coal deposits within the Belt. Although the geology of this region has been studied for more than a century and many major advances of geological knowledge have resulted from those studies, the development of a true understanding of the structure and tectonic history of the Belt has been lacking. Previous tectonic models for the Delamerian Fold Belt, typically incorporating basement wrenching, have not adequately explained the structural pattern between the Stuart Shelf and the Murray Basin Province. The Delamerian Fold Thrust Belt can now be explained in terms of thin-skinned folding and thrusting. The combination of geophysical techniques together with relatively recently developed structural geology methods never before applied within the Flinders Ranges have been used to develop a completely new interpretation of the vertical and lateral distribution of rock masses along a 250km transect across the Belt. Several types of geophysical data from very mixed ages and sources have been successfully reduced to a common useable database. The previously available geological and geophysical data have been improved with detailed geological field work carried out along the transect. The new database was used to apply the techniques of structural modelling and geological section balancing. The result is a completely new model of the geological structure along the 250km transect. The new model compares favourably with the structure of arcuate fold-thrust belts. The balanced geological cross section depicts a fold-thrust belt propagating towards the Gawler Craton and Stuart Shelf along a basal decollement at 8 to 12km depth. The results of this study represent a major advance in the understanding of the structural and general geological development of the Delamerian Fold Thrust Belt.

106

A CLOCKWISE P-T PATH FOR GRANULITE-FACIES METAMORPHISM IN THE MALLEE BORE AREA, NORTHERN HARTS RANGE: IMPLICATIONS FOR THE PROTEROZOIC TECTONIC EVOLUTION OF THE ARUNTAINLIER, CENTRAL AUSTRALIA Jodie A. Miller, Ian Cartwright Department of Earth Sciences and VIEPS, Monash University, Clayton, Australia, 3150 P-T-t paths have been one of the main tools used to infer tectonic settings for metamorphism. Clockwise P-T-t paths have been used to infer thinning of overthickened continental crust during waning convergent deformation, while anticlockwise P-T-t path have generally been explained as a consequence of rapid cooling following the intrusion of magmas, with or without convergent deformation (Harley, 1989). However, interpretations of metamorphic PT-t paths of some terrains have remained controversial despite a long history of investigation. Studies in the Harts, Anmatjira, Reynolds and Strangways Ranges of the Arunta Inlier, central Australia have resulted in conflicting interpretations regarding the P-T-t history of these terrains. The ambiguity associated with P-T-t paths derived from studies on the Arunta Inlier highlights many of the problems associated with the construction of qualitative P-T-t paths in older, complex, multiply metamorphosed terrains. Among these are: (1) the reliance upon interpretation of sometimes ambiguous mineral assemblages and reaction textures complicated by the possible superposition of multiple metamorphic events; and (2) the extraction of geologically meaningful P-T's from rocks that have experienced temperatures in excess of 700°C, is complicated by the fact that both intra- and intergranular diffusion rates are sufficiently rapid that mineral compositions may re-equilibrate during the initial stages of cooling (Harley, 1989). Investigation of new areas within the Arunta Inlier, such as the Mallee Bore area on the northern margin of the Harts Range, may provide additional insights into the P-T-t evolution of different areas within the Arunta Inlier. 134*30'

135*00'

134*50'

134*40'

135*10.'

135*20' Q

Regional Geology of the Harts Range

22*50'

Alcoota Station

Basement - Early Proterozoic Entia Gneiss Complex Y / / A Qonaaalabi Gneiss

100k

TC INLIER

MALLEEBORE AREA

Complex |.

.'IStrangways Metamorphic

Harts Range

23*00'

Complex Cover - Early-Mid Proterozoic

iMt. Riddoch Station

Harts Range Group

HIM®

ENTIADOME:

ly-'.y.'-l Irindina Gneiss R^S^i Mt. Riddoch Amphibolite

Huckitta r>

Bruna Gneiss

Dome^v

E555 Granitic Gneiss

•Yambla v. '.".'Window.;

/ / / Shear Zone

23*10'

[Florence Shear •Y.y.vZone;.Y.Y

] Queenie Flat Granite Huckitta Granodiorite

:Artnarpav ^Window-

I Tertiary Sediments

15km

Figure 1. Regional Geology of the Arunta Inlier and the Harts Range showing the location of the Mallee Bore area. Abbreviations in inset: AB - Amadeus Basin; NB - Naglia Basin; WB - Wiso Basin; TC - Tennant Creek; GB - Georgina Basin; NP - Northern Province; CP - Central Province; SP - Southern Province. 107

The Mallee Bore area is located on the northern side of the Plenty Highway, approximately 30 km from the Entia Dome (Fig. 1), and is dominated by Proterozoic garnetiferous orthogneisses and paragneisses. Three episodes of metamorphic activity have been identified in the Mallee Bore area on the basis of textural and mineralogical evidence. Geothermobarometry using independent Fe-Mg exchange thermometers and net transfer equilibria, as well as average P-T calculations (Powell and Holland, 1994) using the computer program THERMOCALC, indicates that the peak of regional metamorphism occurred at around 8-1 lkbar and 720-770°C. Large sillimanite-facies shear zones were developed subsequent to regional metamorphism under conditions of around 700-730°C and 4.4-6.4kbar. Garnet porphyroblasts within the mylonitic fabric were retrogressed after the cessation of shearing, probably when the rocks had cooled sufficiently for fluids to start exsolving from crystallizing partial melts (ca. 600-650°C). Other granulite terranes (e.g. Broken Hill; Corbett and Phillips, 1981) record retrogression at similar temperatures suggesting that retrogression as a result of the influx of fluids derived from the crystallization of partial melts may be a common phenomenon in partially-melted highgrade rocks. Decompression of the Mallee Bore rocks between regional metamorphism and the onset of shearing is consistent with a clockwise P-T path. Comparison of the findings for the Mallee Bore area with the Harts Range shows that while both areas apparently record clockwise P-T paths, the form of the P-T path is dissimilar (single cycle as opposed to multi-cyclic, e.g. Arnold et al., 1995) and the timing of regional metamorphism appears to be ambiguous (e.g. Foden et al., 1995; Collins and Shaw, 1995). Moreover, shear zones in the Mallee Bore area are sillimanite-bearing as opposed to kyanite-bearing in the Harts Range, suggesting that there was a higher geothermal gradient in the Mallee Bore area than the Harts Range. Nevertheless, the P-T conditions of metamorphism in the Mallee Bore area are consistent with HP-HT metamorphism recorded in the Harts and Strangways Ranges and contrasts with the LP-HT style of metamorphism seen in the northern tectonic province (e.g. the Reynolds Range). These variations in metamorphic style within the Arunta Inlier suggest that in Proterozoic polymetamorphic terrains the metamorphic and tectonic setting may vary on quite small scales, rendering inappropriate large-scale tectonic models.

References Arnold, J., Sandiford, M. & Wetherly, S., 1995. Metamorphism in the Entia Dome, Harts Range, eastern Arunta Block. Precambrian Research, 71, 183-205. Collins, W.J. & Shaw, R.D., 1995. Geochronological constraints on orogenic events in the Arunta Inlier: a review. Precambrian Research, 71, 315-346. Corbett, G.J. & Phillips, R.N., 1981. Regional retrograde metamorphism of a high grade terrain: the Willyama complex, Broken Hill, Australia. Lithos, 14, 59-73. Foden, J.D., Mawby, J., Kelley, S., Turner, J.P., and Bruce, D., 1995. Metamorphic events in the eastern Arunta Inlier, Part2. Nd-Sr-Ar isotopic constraints. Precambrian Research, 71, 207-227. Harley, S. L., 1989. The origin of granulites: a metamorphic perspective. Geological Magazine, 126, 215-331. Powell, R. & Holland, T. J. B., 1994. Optimal geothermometry and geobarometry. American Mineralogist 79 120-133. ' '

108

SUBDUCTION AND SEDIMENT ACCRETION BY UNDERPLATING IN THE EASTERN LACHLAN FOLD BELT J.M. Miller & D.R. Gray Department of Earth Sciences, Monash University, Melbourne, Vic, 3168, Australia. An imbricate stack of turbidite, chert, mafic volcanics and chaotic block-in-matrix melange (Fig. 1) along the south coast of N.S.W. defines a palaeo-subduction zone within the Palaeozoic Lachlan Fold Belt of eastern Australia. The sequence has the structural signature of a subduction zone comprised of lithologies in situ prior to subduction. The key arguments for a subduction zone are the presence of "classic" chaotic block-in-matrix melange, broken formation along high strain zones associated with large scale imbrication and underplating, scaly cleavages, initial "soft sediment" deformation, and structural complexity associated with a marked deformation front. Like other subduction accretionary complexes (e.g. Kodiak Complex, Aleutians and Shimanto Complex, Japan) biostratigraphic data (Stewart & Glen, 1992, Quart. Notes N.S.W. Geol. Survey, 85, 1-6) indicate a complex imbricated sequence with a generalised younging seawards towards the former trench/subduction zone.

Beramgui-Iike F t folds High strain zone (broken formation) at some loca there is an age reversal across this contact. Chert and mafic volcanics, Mid Late Ordovician (Wagonga Formation)

Decollement

Chaotic block-in-matrix melange comprised of chert, turbidite and rare pillow lava (Bogolo Formation).

Fig. 1: Generalised imbricate stacking within the Narooma accretionary complex. The stacking order is clearly established within the Narooma anticlinorium and at Batemans Bay.

Coastal and inland structural belts define a gross zonation towards the subduction complex (see Fig. 1); these correspond approximately to the "inland" and "coastal facies" of Powell (1983a, Geol. Soc. Aust. Specialist Group in Tectonics and Structural Geology Field Guide 1, 1-118). Later Middle Silurian to Middle Devonian east-west shortening has resulted in large scale folding and thrusting within both the inland and coastal structural belts. This deformation is also associated with a marked deformation front that intensifies towards the south coast (see Fig. 6, Fergusson and Coney, 1992, Tectonophysics 214, 417-439). The inland structural belt contains broken formation associated with a scaly cleavage (Fergusson and Vandenberg, 1990, Jour. Struct. Geol., 12, 577-589). Zones of oblique dextral shear are also found within the chert on the south coast. Within the coastal structural belt a "stacking order" is clearly established at Narooma and Batemans Bay. Uppermost in the imbricate stack is an Early Ordovician to Late Ordovician turbidite sequence (Fig.l). At Murruna Point and Narooma this turbidite is separated from the underlying chert and volcanic sequence by a high strain zone containing broken formation. Pressure solution, dilational veining &nd boudinage is extensive within the high strain zones. Much of the rock within these zones has the appearance of a low grade mylonite. At Narooma there is an age inversion across this tectonic contact - Early Ordovician turbidite overlying Late Cambrian to-Late Ordovician chert. The well documented slaty cleavage at a low angle to bedding within the coastal belt turbidite (Williams, 1972, Am. Jour. Sci., 272, 1-47; Powell

109

and Rickard, 1985, Jour. Struct. Geol., 7, 385-400; Wilson and de Hedouville, 1985, Jour. Struct. Geol., 7, 401-408) is a direct result of this imbrication. At Batemans Bay the contact between the two units is conformable and Late Cambrian to Early Ordovician in age (Etheridge et a/., 1973, Jour. Geol. Soc. Aust., 19, 465-470; Bischoff and Prendergast, 1987, Neues Jahrbuch fur Geologie und Palaontologie, 175, 39-64). The lowest structural unit in the subduction complex is a "classic" block-in-matrix melange (Fig. 1). Blocks are dominantly comprised of turbidite as well as chert and rare pillow lava. The melange has blocks of random sizes that are not stratiform but randomly oriented and poorly sorted within the matrix. The necked regions of the blocks often do not even contain undulose extinction indicating prelithifrcation deformation. A strong tectonic overprint has further disrupted the blocks and resulted in a strike parallel alignment of blocks that were initially aligned at random. A pervasive slaty cleavage has formed synchronously with this later deformation and is defined by quartz rich lenses, chlorite and white mica. This foliation contains a stretching lineation that strikes 330° and is defined by "pull apart" chlorite, pressure shadows on framboidal pyrites and the long axes of the quartz rich lenses. The chaotic blockin-matrix melange is dominantly comprised of turbidite that has a similar provenence to the overlying turbidite sequence. It is believed to have been sourced from lithologies similar to the overlying chert and turbidite sequences. If it is not sourced from these lithologies a source older than Middle Cambrian is required that has turbidite of a very similar provenence to the Early Ordovician to Late Ordovician turbidite. The melange has formed prior to the slaty cleavages and is either an olistostrome or melange generated by upwelling within an accretionary complex (cf. Cloos and Shreve, 1988, Pure & Applied Geophysics, 128, 455500). In both scenarios the melange has been underplated or accreted beneath the chert and turbidite implying the contact between the melange and chert is a detachment of a similar style to the detachment between the chert and turbidite. In a classic accretionary complex scenario underplated packets are separated by low angle detachments (Sample and Fisher, 1986, Geology 14, 160-163) - this is exactly what is observed on the south coast. Accretion can not have been via offscraping because the metamorphic grade of lower greenschist facies is too high (offscraped sediments will not be metamorphosed above zeolite facies - Sample and Moore, 1987, Bull. Geol. Soc. Am., 99, 7-20). All sediment beneath the high strain zone between the turbidite and chert is considered to have been accreted. However, the boundary between the inland and coastal structural belts may represent the true extent of accretion implying the presence of other, as yet undetected, low angle detachments. One problem is that we cannot constrain which way the detachments dip. The early deformation is associated with an extension lineation oblique to the later structural trends. This implies oblique deformation whose timing is probably constrained to the Early Silurian by the presence of Late Ordovician sediments within the complex. The oblique nature of the deformation is enforced by the oblique style of the Early Silurian deformation affecting the Lachlan Fold Belt ("Benambran Orogeny") (cf. Powell, 1983b, Jour. Geol. Soc. Aust., 30, 353-373). At some stage the deformation has changed from oblique imbrication at a low angle to bedding to massive shortening via folding and thrusting during the Middle Silurian to Middle Devonian. This may correspond to a change in convergence vector or the oblique component of deformation being partitioned further back in the complex. Powell's (1983a) accretionary prism model for the south coast was discredited because the lithologies of the inland and coastal facies are now believed to be sedimentologically indistinguishable. Many geologists still perceive an accretionary prism should contain "exotic" offscraped and underplated pelagic and hemipelagic lithologies. If sediment subduction occurs then a "classic" accretionary prism will not form and the lithologies at the toe of the complex will have been insitu prior to subduction (eg. Scholl et al.y 1980, Geology 8, 564-568). Even at margins where significant accretion occurs much of the accretionary prism may be comprised of lithologies in situ prior to subduction (Hilde, 1983, Tectonophysics, 99, 381-397). The south coast may represent the toe of a subduction zone at which sediment subduction has occurred resulting in in situ lithologies inheriting structural features "normally" seen within the accretionary prism. Alternatively we may be looking at what "classic"' accretionary prisms really look like at a deeper structural level.

110

THE BARRABOOL HILLS METAGABBRO: A CURIOUS PIECE OF THE CAMBRIAN JIG-SAW PUZZLE IN VICTORIA. V.J. Morand Department of Geology, University ofBallarat, P.O. Box 663, Ballarat, Vic. 3353. Situated a few kilometres west of Geelong in central southern Victoria, the Barrabool Hills Metagabbro is probably part of the Heathcote "greenstone" belt, a belt of Cambrian basic to intermediate volcanics metamorphosed to low regional grade. It is separated from the main metabasite belt by Devonian granites and Cainozoic cover. Unlike the rest of the Heathcote Belt, the Barrabool Hills occurrence is entirely intrusive, consisting of several small outcrop areas of metagabbro, intruded by very small granite bodies. Initially described as epidiorite, the gabbro ranges from coarse to fine-grained and has a sporadically occurring weak igneous layering. No volcanic rocks or equivalents of the Cambrian chert and shale which overlie the metabasalts in the Heathcote Belt have been found, and as a consequence the metagabbro is undated, although very likely is Cambrian. A further problem is that the only other Paleozoic rocks in the area are the intrusive granites, which are correlated with the Late Devonian You Yangs Granite. This metabasic rock (whole rock analyses indicate it is gabbro) is interpreted as layer 3 of oceanic crust, as it forms one intrusion several kilometres in length and breadth. It is too thick to be a sill intruding the volcanic sequence, such as those found in parts of the Heathcote and Mount Wellington metabasite belts. Therefore the Barrabool Hills Metagabbro is the lower part of Cambrian oceanic crust which forms an ophiolite sequence marking the eastern edge of the Bendigo-Ballarat structural zone. The igneous layering is vertical and strikes NW - this may be the general trend of the belt of basic rocks in this area, if the layering was initially horizontal. Metamorphic assemblages are anorthite-amphibole (actinolite to hornblende), with minor chlorite. This indicates amphibolite facies metamorphism, much higher grade than the prehnitepumpellyite to lower greenschist facies that characterises the main Heathcote belt. Despite the small granite intrusions, and the nearby larger granite occurrence at Dog Rocks, the metamorphism of the gabbro appears to be regional. Although most rocks are massive with finely granoblastic recrystallised plagioclase and both large grains (replacing pyroxene) and needles of amphibole, foliated samples are common. These latter rocks range from weakly deformed to mylonitic. In addition, one sample of ultramylonite and one of cataclasite have been recovered. Unfortunately most samples are not in situ, but the few shear zones located in outcrop appear to form a sub vertical conjugate set indicating NS compression. On the microscale, mylonitic rocks are strongly foliated, some with S-C fabrics, and they typically contain rounded porphyroclasts of deformed hornblende and plagioclase surrounded by well aligned fine-grained hornblende and plagioclase. Thus the ductile shearing occurred slightly after the main metamorphism, but at about the same grade, since amphibolite facies assemblages characterise the mylonites. The shear zones are interpreted 111

as resulting from a regional deformation, with the gabbro deforming heterogeneously by strong strain partitioning into discrete shear zones, as is typical of plutonic rocks. This synmetamorphic deformation is quite different to the contact metamorphism observed in aureoles around Late Devonian granites in central Victoria. These aureoles have hornfelsic textures with very little evidence of deformation. Also, the grade of metamorphism of the metagabbro is the same throughout the entire outcrop area, with no increase in the vicinity of the granite intrusions. The Barrabool Hills Metagabbro records a regional metamorphic event of amphibolite facies grade, accompanying a strong regional deformation. The age of this orogenic event is pre-Late Devonian, and post-Cambrian, but otherwise unconstrained. The most likely event is the mid-Devonian Tabberabberan Orogeny, responsible for the E-W shortening of the Bendigo-Ballarat zone and thrust emplacement of the Heathcote belt. However, the NS compression suggested by the shear zone orientations and the possible NW trend of the belt here are at odds with this suggestion.

112

TECTONIC AND STRUCTURAL CONTROLS ON HYDROTHERMAL FLUID FLOW AND MINERALISATION AT THE PORGERA GOLD DEPOSIT, PNG S. M. Munroe Research School of Earth Sciences, Australian National University, Canberra, ACT 0200 Vein and fault-hosted mineralisation at the Porgera gold deposit is spatially and temporally associated with a suite of Late Miocene (6.0 Ma) mafic alkaline stocks and dykes (the Porgera intrusive complex). The stocks were emplaced after a period of intense folding and thrusting to form the Papuan Fold Belt. Both deformation and magmatism occurred in response to arccontinent collision in the Middle - Late Miocene (-10 Ma). Approximately 3600 m of uplift occurred at Porgera after mineralisation, and during the formation of the highlands of Papua New Guinea. Two stages of mineralisation at Porgera are distinct in their paragenesis and structural association. Stage I mineralisation is hosted by extension fractures which have various orientations and are spatially associated with the intrusives. The veins form two NNE striking sets which dip steeply and gently. The angle between the two dominant veins sets in any one location is approximately 90°. Mineralisation in these veins consists of pyrite, spalerite, galena, quartz and calcite with lesser amounts of native gold, chalcopyrite, marcasite, arsenopyrite, friebergite, tetrahedrite and siderite. Native gold occurs as sub-microscopic inclusions within pyrite resulting in gold grades up to 20 g/t within the vein. Stage II mineralisation is hosted by faults and fractures of the Roamane Fault Zone which strike ENE and dip at 75° south. Stage II mineralisation consists of quartz, pyrite and roscoelite (vanadium rich white mica) with lesser native gold, Au-Ag-Hg-Pb tellurides, apatite and barite. Gold grades within these veins are up to hundreds of grams per tonne. The Roamane Fault Zone experienced dip-slip normal movement during stage II mineralisation and dextral strikeslip movement after stage II mineralisation. The fault zone consists of a principal displacement zone (which strikes ENE) and numerous E-W striking, steeply dipping subsidiary faults. In addition, a number of ENE striking vertical extension fractures splay from the principal displacement zone. The Roamane Fault Zone and the stage II mineralisation always over-print stage I sulphide veins. Movement on the Roamane Fault Zone resulted in extensive brecciation which generated a low permeability pathway for hydrothermal fluid movement. Injection cataclasites associated with the Roamane Fault Zone and stage II mineralisation indicate episodic rapid fault slip on the fault occurred during mineralisation. The stage I (extension fracture) veins and stage II (fault-hosted) veins formed in a magmatichydrothermal system after intrusion at 6.0 Ma and before approximately 5.8 Ma. Therefore, in a relatively short period of time, two contrasting fracture mechanisms and two contrasting mineralisation types have developed in one area. For a vertical stage I vein with an assumed vertical intermediate compression (g2) direction and stress difference ( a r a 3 ) of 10-20 MPa (where a 2 = (a>+ a3)/2) a fluid pressure approaching lithostatic (48-53 MPa at 2000 m depth) is required for extension fracturing, according to the hydrofracture criteria (Pf > g 3 + T ) (Table 1). The presence of sub-horizontal stage I sulphide veins in extension fractures also indicates that fluid pressures during this stage of mineralisation were near lithostatic. The formation of sub-vertical and sub-horizontal stage I veins (with an average angle between them of 90°) suggests rapid changes in the principal stress directions which may be caused by intrusive complex and caldera subsidence in response to cooling or contraction of magma at depth. The timing of formation of the fault-related ENE vertical extension fractures with respect to fault loading and the fault-slip cycle is uncertain. Here, I assume that the vertical extension fractures formed as the fault ruptured, under peak conditions of fault loading. At this stage in the fault cycle, the effective minimum principal stress (o»3') was in the tensile field. It is assumed that the maximum principal stress (Gj) direction is vertical and the stress difference (a 1 - a 3 ) is 20 - 30 MPa. If c 3 ' = -5 MPa (T0 = 5 MPa), the effective maximum principal stress (o/) would be in the range 15 to 25 MPa, indicating fluid pressures prior to failure of 28 to 38 MPa. This 113

estimated fluid pressure is less than that estimated for the stage I veins and indicates that fluid pressures increased only slightly above hydrostatic prior to fault failure = 0.5 to 0.7) (Table 1).

Table 1. Estimated stress andfluidpressure conditions during mineralisation at Porgera

Stage I - Vertical extension fractures immediately prior to formation and mineralisation q, MPa q, MPa q, MPa P MPa K 1

58 to 63

48 to 53

43 to 48

q, - q, MPa

0.9 to 1.0

10 to 20

Stageq IIMPa - Roamane Faultq immediately priorq toMPa failure and vertical extension fracture MPa P MPa ky formationq - q MPa 53 38 to 43 33 to 23 28 to 38 0.5 to 0.7 20 to 30 ^vertical = Pgz> Pr = 2650 kg.m , g = 10 m.s , z = 2000 m, T = 5 MPa, al>a2> q P > q + T for extension fracture 4

CJ = a f

vertical

5,6

, q = (Gi - o )/2 2

probably also applicable to sub-horizontal veins where locally q = a ai = sub-lithostatic during collapse of intrusive complex or withdrawal of magma values assumed, G = a , ^ -> 0.4 post fault failure during mineralisation, = P /a

vertical

vcrtiC

vcrtical

The change in style of deformation from stage I to stage II mineralisation is a function of a change in the stress field in the area at that time. The estimates shown in Table 1 indicate a reduction (by approximately 35%) in the horizontal principal stresses between stage I and stage II mineralisation. The formation of the Roamane Fault Zone in an extensional tectonic environment is a result of relaxation of compression across the Papuan Fold Belt in response to cessation of subduction and continent - arc collision at approximately 10 Ma. Presumably convergence across the Australia - Pacific plate has been transferred to another margin by 6.0 Ma. Such changes in the tectonic setting result in changes in the deformation mechanisms which may be favourable to hydrothermal ore mineralisation. At Porgera, the introduction of the Roamane Fault Zone has been critical in the formation of a large high-grade gold resource. Active deformation on the Roamane Fault Zone during hydrothermal fluid flow may have aided mineralisation by subjecting fluids to rapid changes in pressure, temperature, fluid-fluid and fluid-rock interaction (Figure 1).

Figure 1. Projected section looking south-west at Porgera showing schematic fluid flow paths during mineralisation. Stage I: Flow concentrated around intrusive stocks and dykes. Stage II: Vertical extension fractures splay from the main fault and have allowed fluids to rapidly ascend into the fault during failure. Geometry of the intrusives after magnetic modeling of Logan (1993, Placer Ex. Report 93/08, unpublished). 114

GEODYNAMIC MODELS OF CENTRAL ASIA: STRONG TARIM BASIN OR WEAK TIEN SHAN? E. A. Neil1 & G. A. Houseman2 1

Department of Earth Sciences, Monash University, Clayton, Vic 3168, Australia Department of Mathematics and A.G.C.R.C, Monash University, Clayton, Vic 3168, Australia

The India-Asia collision has caused crustal thickening in Central Asia to occur to such an extent that the Tibetan crust is now twice the normal crustal thickness. In addition, this collision caused the reactivation of the Tien Shan mountain range causing the crust in this region to thicken considerably to 50 km. It is unusual then that the Tarim Basin, the region between Tibet and the Tien Shan, shows little internal deformation and does not seem to have thickened appreciably as a result of the collision. In addition, Central Asia is a region of high seismicity and tectonic activity, yet the Tarim Basin is unusually aseismic. To the north, however, in the Tien Shan, seismicity is high with vertical strain-rates as high as 2 % per m.y. (Molnar and Deng, 1984), suggesting that crustal thickening in the Tien Shan is rapidly continuing. Actual crustal thicknesses in Central Asia have been determined using seismic, gravity and topographic data. Surface wave studies for paths across Central Asia have constrained the crustal thickness of the Tibetan Plateau to between 65 and 75 km (Chen and Molnar, 1981; Romanowicz, 1982), whereas P-wave arrival time studies show the crustal thickness of the Tien Shan to be around 50 km (Roecker et al, 1993). Crustal thickness estimates based on analysis of gravity data, indicate that the crustal thickness decreases to about 45 km in the middle of the Tarim basin, increasing to about 52 km in the Tien Shan and 72 km in the Tibetan Plateau (Ma, 1986). The present distribution of crustal thickness in Central Asia gives two possible hypotheses as to how the Tien Shan was reactivated by the India-Asia collision even though it is quite some distance north of the collision front and still leave the Tarim Basin, which is closer to the collision front, relatively undeformed. These hypotheses are: that the Tarim Basin is unusually strong and has resisted deformation and thickening itself but has acted as a secondary indenter into the Tien Shan causing the reactivation of this range (Villotte et al, 1984; England and Houseman, 1985); or that the Tien Shan is an unusually weak region that thickened during the collision, drawing strain from the Tarim Basin and leaving it relatively undeformed (Roecker et al, 1993). In this study we test these two hypotheses using a finite element representaion of the thin viscous sheet model to approximate the deformation field in the India-Asia collision (England and McKenzie, 1982; England and Houseman, 1986). We use the model to investigate how variations in the strength of the lithosphere affect the distribution of crustal thickness and the distribution of vertical strain-rates in the collision zone. Numerical calculations obtained from the thin viscous sheet model, give more support to the theory that the Tien Shan has formed as the result of a strong Tarim Basin. If, however, the Tarim Basin did form as a result of a weak Tien Shan, then the weak regions must have a strength parameter that is 70-85% of that for the surrounding region depending on the rheology of the lithosphere. Our calculations after England (1983) show that such a weakness could be explained by a local increase in Moho temperature of approximately 12-15°C. The experiments show that within the model a strong Tarim Basin is able to transfer strain to the Tien Shan producing significant crustal thickness there, while resisting significant deformation itself. The results suggest that the Tarim Basin should have a strength parameter no less than 125% that of surrounding regions in order to produce the observed distribution of crustal thickening. Our calculations after England (1983) show that this increase in strength could be explaned a decrease in Moho temperature of approximately 24-30°C.

115

References Chen, W., and P. Molnar, Constraints on the seismic wave velocity structure beneath the Tibetan Plateau and their tectonic implications, J. Geophys. Res., 86, 5937-5962, 1981. England, P.C., Constraints on Extension of Continental Lithosphere, J. Geophys. Res., 88, 1145-1152, 1983. England, P.C. and D.P. McKenzie, A thin viscous sheet model for continental deformation, Geophys. J. R. Astron. Soc., 70, 295-321, 1982 (Correction to above, Geophys. J. R. Astron. Soc., 73, 523-532, 1983). England, P., and G. Houseman, Role of lithospheric strength heterogeneities in the tectonics of Tibet and neighbouring regions, Nature, 315, 297-301, 1985. England, P., and G. Houseman, Finite strain calculations of continental deformation 2. Comparison with the India-Asia collision zone, J. Geophys. Res., 91, 3664-3676, 1986. Ma, Xingyuan (ed), Map of lithospheric dynamics for China and adjoining ocean regions, Geology Publishing House. State Seismological Bureau. Beijing, 1986. (In Chinese) Molnar, P. and Q. Deng, Faulting associated with large earthquakes and the average rate of deformation in central and eastern Asia, J. Geophys. Res., 89, 6203-6228, 1984. Roecker, S. W., T.M. Sabitova, L.P. Vinnik, Y.A. Burmakov, M.I. Golvanov, R. Mamatkanova and L. Munirova, Three-Dimensional Elastic Wave Velocity Structure of the Western and Central Tien Shan, J. Geophys. Res., 98, 15,779-15,795, 1993. Romanowicz, B. A., Constraints on the structure of the Tibet Plateau from pure path phase velocities of Love and Rayleigh waves. J. Geophys. Res., 87, 6865-6883, 1982. Vilotte, J.P., M. Daignieres, R. Madariaga & O. Zienkiewicz, The role of a heterogeneous inclusion during continental collision, Phys. Earth Planet. Int., 36, 236-259, 1984.

116

DEVELOPMENT OF VEIN SYSTEMS IN SHEAR ZONES AT THE REVENGE MINE, KAMBALDA, WESTERN AUSTRALIA: EVIDENCE OF PALAEOSEISMIC EVENTS P.T. Nguyen . C. McA. Powell , L.B. Harris & K.A.A. Hein 1

Department of Geology & Geophysics, The University of Western Australia, WA 6907 Western Mining Corporation Limited\ St Ives Gold Mines, Kambalda, WA 6442

1 2

The Revenge gold mine, 13 km southeast of Kambalda, in the Lake Lefroy gold field, Western Australia, is located in Archaean metavolcanic and sedimentary rocks of the Norseman-Wiluna greenstone belt. The area has undergone four main periods of deformation. The first two events were ductile, the first producing regional south-overnorth thrusts and the second NNW-trending folds and shear zones. The third event was brittle-ductile wrench faulting associated with the development of shear zones, and the fourth event involved dextral reactivation of early structures. The Revenge gold deposit contains two main types of reverse shear zone: (i) N-trending, moderately southeast dipping (45°); and (ii) NNE-trending, shallowly northwest dipping (15°). These shear zones were formed as a conjugate set during D3 WNW-ESE sub-horizontal compression and sub-vertical extension (Figure 1)

(e) (d) (C) Figure 1: Revenge schematic east-west cross section of (a) north-trending, moderately east-dipping shear zone, (b) NNW-trending, shallowly west-dipping shear zone and their structural data: (c) moderately east dipping shear veins and lineations, (d) shallowly west- dipping shear veins and lineations (e) interpretation of principal stresses responsible for the shear zones.

117

There are three main types of veins consisting predominantly of quartz and carbonate within the shear zones: shear veins, extensional veins and hybrid veins. The moderately southeast-dipping shear zone has narrow (10-50 cm), laminated, moderately dipping shear veins and multiform, large (10-50 cm), sub-horizontal extensional veins (Figure la). The shallowly northwest-dipping shear zone has large (0.5-5.0 m), sub-horizontal shear veins, narrow (5-15 cm), east-dipping extensional veins and narrow (5-10 cm) hybrid veins (Figure lb). These vein systems represent different local stress fields within each shear zone where the vein development accommodated a sub-vertical far-field stress ((J3). Towards the margin of the shear veins, several thin slivers of wall rocks taper into thin layers (0.1-3.0 mm) of sulphide composed largely of pyrite, subordinate vein-wall materials and free gold. Pyrite and gold deposition is thought to have occurred by reaction of a sulphide-rich hydrothermal fluid and the iron-rich wall rock at the margin of the vein by the process of desulphidation. Shear failure could have been triggered by fluid-pressure fluctuation, which is an important factor in the development of the shearzone/quartz-vein system. Reopening of quartz veins occurred when pressure built up during a slip event and repeated growth of quartz trapped thin slivers of wall rock and pyrite as inclusion layers by the crack-seal mechanism. Other internal structures of the veins such as crack-seal textures, open-space filling and quartz fibres also indicate that the veins are formed by hydraulic fracturing in an episodic stress regime by crack-seal processes. Sub-vertical shortening is also recognised during development of the shear zones indicating episodic permutation of principal stresses. This alternation is indicated by stylolitic textures in the veins, sub-vertical quartz veins, superposed normal faults, fluid inclusion planes and a sub-horizontal crenulation cleavage. Such stress permutation has been described by Boullier & Robert (1992) and Cox (1995) as shear stress reversal and interpreted as an episode of fluid pressure drop after rupture along the shear vein corresponding to collapse of the veins during palaeoseismic events. Cross-cutting relationships of the veins with different vein textures and alteration styles may indicate changes in fluid chemistry and fluid pressure during the formation of the shear zone. References Boullier A.M. & Robert F. 1992, Palaeoseismic events recorded in the Archaean gold-quartz vein networks, Val d'Or, Abitibi, Quebec, Canada, J. Struct. Geol. 14, No. 2, 161-179. Cox S.F., 1995. Faulting processes at high fluid pressures: an example of fault-valve behaviour from the Wattle Gully Fault, Victoria, Australia, J. Geophysical Research, in press.

118

THE STRUCTURAL EVOLUTION OF ACTIVE FAULT AND FOLD SYSTEMS IN CENTRAL OTAGO, NEW ZEALAND: EVIDENCE REVEALED BY DRAINAGE PATTERNS Richard J. Norris1, John H. Youngson1, & James A. Jackson2 1

Dept. of Geology, University of Otago, Dunedin, New Zealand

Dept. of Earth Sciences, University of Cambridge, U. K.

While the Alpine Fault is the site of most of the interplate dislacement in New Zealand's South Island, associated deformation is distributed over at least 100 km to the east throughout South Island. In Otago, Quaternary faults and young uplifted ranges extend over 200 km as far as the coast at Dunedin. Within the eastern 100 km or so, the ranges and basins reflect a northeasterly trending fold-thrust system which may be similar to the early stages of more fully developed, older systems elsewhere in the world, such as the Laramide fold-thrust belt of the western USA. In this study, we use drainage patterns to gain an insight to the spatial and temporal evolution of both single structures and the fold belt as a whole. Central and east Otago is underlain by a basement of fairly uniform schist lacking major variations in resistance to erosion. A widespread late Cretaceous to Miocene unconformity surface is cut across the schist and overlain by a series of easily erodable Cenozoic sediments. The schistosity in the basement is subparallel to the unconformity surface over much of the area. During the Quaternary, fold/fault structures on a wavelength of 5-20 km, probably related to deep-seated reverse faulting, have developed within the area and resulted in the growth of elongate schist ranges. The unconformity surface is deformed over the ranges and the Cenozoic sediments have been largely stripped from them. Consequent drainage patterns have developed on the growing ranges and have evolved in response to the structural evolution of the fault and fold systems. By examining the pattern of drainage evolution, important evidence may be obtained regarding the development of structures. In particular, we interpret a number of aspects of structural evolution: (1) How simple asymmetric folds can develop into box folds. Taieri Ridge is a simple asymmetric fold and has a consequent drainage pattern perpendicular to the fold crest which reflects the asymmetry of the underlying structure. The Rock and Pillar range is a more symmetric structure with a boxfold shape, but still retains a highly asymmetric drainage pattern which is deeply incised through the uplifted western limb. (2) How apparently continuous ridges were formed by the coalescing of separate propagating fault segments. Blackstone Hill and Raggedy Range now form a continous ridge with a low saddle between them. The Ida Burn flows south and around the end of Blackstone Hill in a deeply incised gorge. The Pool Burn on the other hand now flows north parallel to Raggedy Range, against the regional trend, to join the Ida Burn. Abandoned gorges high on Raggedy Range attest to the migration of the Pool Burn around the end of the growing range as it propagated northwards to coalesce with Blackstone Hill. (3) Evidence for the relative ages (or relative uplift rates) of adjacent structures. Rough Ridge splits northwards into several subparallel segments. Consequent drainage from early formed ridges is disrupted and reversed by uplift of more easterly structures, providing evidence for a general easterly stepping of fault imbrication. (4) Evidence for the propagation direction of fold/faults as they grow. The original drainage pattern across a growing ridge is disrupted and reversed, with streams being gathered by a small number of rivers flowing parallel to the structure before cutting across it in a gorge. These gorges are interpreted to represent previous locations of the nose of the fold, and the asymmetry of the catchments provides evidence for the direction of fold/fault propagation. By making some fairly rough and ready measurements, we are able to estimate that, for a 0.5-1 metre uplift of the crest such as might be expected during a characteristic earthquake, the folds propagate approximately 15-45m laterally. This figure is consistent both with field data from elsewhere and with theoretical studies on fault propagation. 119

Using the results of our analysis, we are able to reconstruct the development of the fold-thrust belt with time. The nucleation, growth and coalescence of structures may be compared with published analogue models of fold belt evolution.

(a): Map of east Otago (modified from Bishop, D. G., IGNS Sci. Rpt 94/1, 1994) showing peneplain.. 'Faults' are taken from Mortimer (IGNS Geological Map 7, 1993): for the areas studied, these appear to be mainly the steep limbs of asymmetric folds and there is limited evidence of faulting at the surface. D: Dunstan Range; R: Raggedy Range; RP: Rock & Pillar Range; T: Taieri Ridge; RR: Rough Ridge; M: Manuherikia Valley; IV: Ida Valley; Mo: Maniototo depression; H: Hawkdun Range; K: Kakanui Range (b): Generalised sketch looking north over east Otago (modified from Cotton 1917). Letters as for (a) plus: Tc: Thompson's Creek; B: Blackstone Hill; O: Oliverburn; NRR- North Roueh Ridge; SRR: South Rough Ridge; LRR: Little Rough Ridge. Figure from Jackson, J. A., Norris, R. J., Youngson, J.: J. Struct. Geol. (in Press) 120

TECTONIC SIGNIFICANCE OF VEINS IN THE WOODSREEF ASBESTOS DEPOSIT, SOUTHERN NEW ENGLAND FOLD BELT, NSW. R OSler1, D.O'Hanley2 & P.Lennox3 1 Department of Geology, University of Newcastle, Callaghan, NSW 2308, Australia 2 Department of Geological Sciences, University of Saskatchewan, Saskatchewan, Canada S7NOWO 3 Department of Applied Geology, University of NSW, Sydney, NSW 2033 Several types of veins are present in the Woodsreef serpentinite each having a distinct colour, texture and mineralogy. All are developed east of a major, N20E trending shear zone which is exposed in the open cuts. They are more numerous near shear zones and disappear to the east away from them; veins with chrysotile asbestos decrease in thickness away from these zones. In outcrop, they form en echelon arrays often with well developed bridges, asymmetric tips and overlap ratios of 0.4. Microscopically, evidence for crack- seal deformation is abundant. All these features suggest they are the result of extension which is accord with the interpretation of Glen and Butt (1981). Detailed examination of the different arrays did not reveal a consistent chronological relationship between them. The majority of the veins dip east to southeast at moderate to steep angles: slight variations in orientation occur in the different pits. A similar variation in orientation exists for each vein type suggesting they have experienced a similar tectonic history. Throughout the open cuts, shear zones in massive serpentinite are common. They are defined by a rough, disjunctive, anastomosing cleavage and commonly, rodingite pods or partly rodingitised dolerite occur within them; most trend ~ meridionally or east-west. One or more sets of cleavage may be present, some of which are more intensely developed than others. They show a considerable variation in orientation, however, most trend meridionally and northeast. Analysis of shear bands in these zones, indicates that sinistral, oblique slip movement is dominant. The close spatial relationship between the veins and the shear zones suggests that they are pinnate fractures formed during movement on these zones and are not related to the development of cleavage in schistose serpentinite as proposed by Glen and Butt (1981). Analysis of the angular relationship between individual shear zones and the dominant sets of veins in the open cuts, indicates that compressive stress Gi, was east-north-east during the formation of the south-easterly dipping veins and east-south-east for the east dipping veins. However, these may not be the true Oi directions as movement on the adjacent Peel Fault System may have reorientated the veins and shear zones. References Glen, R.A. & Butt, B.C., 1981. Chrysotile asbestos at Woodsreef, New South Wales. Economic Geology, 76, 1153-1169.

121

SEDIMENTARY RESPONSES TO DEFORMATION - EXAMPLES FROM THE LATE DEVONIAN MANSFIELD BASIN G.J. O'Halloran1 & R.A.F. Cas1 1

Department of Earth Sciences, Monash University, Clayton, Vic. 3168, Australia

A common surficial response to deformation involves the emergence of landforms (uplift) and associated syn-tectonic sedimentation (eg. Riba 1976). The Mansfield Basin of east-central Victoria (Figure la) is the northernmost structural sub-basin of the Mt Howitt Province (Marsden 1976), and contains a thick sequence of Late Devonian-Early Carboniferous continental molasse-like sedimentary rocks (along with minor acid volcanics). It overlies Cambrian to Early Devonian marine sequences of the Lachlan Fold Belt which were deformed during a phase of thin-skinned, east-vergent Tabberabberan (Mid Devonian) deformation. The deformation observed within the lower parts of the Mansfield Basin, often attributed to Carboniferous (Kanimblan) reactivation of extensional structures, is suggested here to be essentially syn-depositional and at least Late Devonian in age.

Figure 1

(a) Geology of the Mansfield Basin, (b) Cross sections across southwestern basin margins.

122

An episode of syn-depositional deformation during the early history of the Mansfield Basin is indicated by structural, stratigraphic and sedimentological relationships preserved along present-day western basin margins. Repeating episodes of folding, erosion and sedimentation are demonstrated in these areas, where fold structures within the lower sequences are truncated by intrabasinal syn-tectonic unconformities (Figure lb). The Mansfield Basin is suggested here to have been initiated as a compressional feature, with syn-sedimentary reverse faulting being responsible for source uplift and subsequent basin deformation. Palaeocurrents within conglomerate units indicate derivation from the west and are consistent with episodic thrusting along basin margin faults providing elevated source regions. Periods of tectonic quiescence are represented by finer grained meandering fluvial facies (indicative of lower regional topographic gradients) which display drainage patterns that appear not to have been influenced by bounding faults to the west. An upsequence increase in the textural and compositional maturity of basin sandstones and conglomerates (O'Halloran & Cas 1995) is proposed to be a result of the incorporation of basin fill into ongoing basin deformation, with unstable metapelitic lithologies being progressively winnowed from clast populations. While faulted contacts also exist along portions of eastern and northeastern basin margins, a less tectonically active setting is indicated in these areas with no evidence for syntectonic unconformities. A "convergent successor" (intermontane) basin setting is envisaged for the lower units of the Mansfield Basin, with active sedimentation and uplift focussed along southwestern basin margins. This phase of Late Devonian intra-orogenic tectonism is suggested to have been a response to distal collisional events along plate boundaries to the east/northeast (? in the New England Fold Belt) with the reactivation of major shear zones in Victoria resulting in the subsequent development of molasse facies.

References Marsden, M.A.H., 1976. Upper Devonian - Carboniferous. In Douglas J.G. and Ferguson J.A. eds. Geology of Victoria, pp. 77-124. Victorian Division, Geological Society of Australia, Melbourne. O'Halloran, G.J. and Cas, R.A.F., 1995. Evidence for syn-depositional deformation from the Late Devonian of the Mansfield Basin, east-central Victoria. Australian Journal of Earth Science (in press). Riba, O., 1976. Syn-tectonic unconformities in the Alto Cardener, Spanish Pyrenees: a genetic interpretation. Sedimentary Geology, 15, 213-233.

123

KANMANTOO PELITE-HOSTED COPPER DEPOSIT: PERVASIVE AND CHANNELIZED LATE-SYNTECTONIC FLUID FLOW ABOVE A PLUTON N. H. S. Oliver School of Applied Geology, Curtin University, GPO Box U1987, Perth, Western Australia The Kanmantoo copper mine (now abandoned) displays a beautiful example of late syntectonic pervasive and channelized fluid flow. The distribution and chemistry of metasomatic alteration products and ore suggests it was formed not by direct reworking of an earlier stratiform syngenetic deposit, but rather during a late thermal pulse related to the intrusion of Delamerian granites. Furthermore, the evolution of fluid flow from pervasive to channelized with time corresponds well with a traditional ore genetic view of regional leaching followed by structural focussing of ore-forming components. Amongst metamorphic petrologists, argument still continues about the extent and significance of pervasive and channelized fluid flow, especially over long distances, and this paper addresses some aspects of that argument. The Kanmantoo Group east of Adelaide is host to a large number of base metal prospects and deposits, including the Kanmantoo copper deposit. The orebody is the largest mined resource in a belt that is still considered to be base-metals prospective. Many of the other deposits are hosted in metamorphosed black shales (e.g. pyritic ore at Nairae). This observation, along with depleted sulphur isotope ratios (Seccombe et al 1985), is consistent with a synsedimentary anoxic, probably biogenic origin for the ores. However, the Kanmantoo deposit is distinct from the bulk of the base metal deposits, in that it is copper dominant (rather than Fe or Pb-Zn); it shows enriched 6 34 S values atypical of a syn-sedimentary, black-shale hosted source; it is hosted in graphite-poor pelitic rocks; and the entire orebody lies in the axial planar region of Delamerian folds, and clearly cuts stratigraphy. Seccombe et al. (1985) explained these observations by appealing to Delamerian reworking of a former basinal-brine derived syn-sedimentary deposit, but for a variety of additional reasons, we prefer the earlier hypothesis of Thomson (1975) that it formed late during the Delamerian Orogeny. The deposit forms a 100m wide lens sitting in the axial planar zone of a broad F3 fold. Bedding and an earlier schistosity are subparallel form-surfaces defining the fold, and are overprinted by a near-upright S3 axial planar crenulation cleavage. Smaller ore lenses and numerous quartz veins (containing minor mineralization, mainly chaclopyrite) are abundant in the periphery of the orebody and elsewhere in the open cut and broader surrounds. These quartz veins, typically a few cm wide and a few metres long, are axial planar to F3 minor folds. They display relatively coarse-grained cm-scale selvages that are rich in staurolite, biotite, garnet and second generation andalusite. These minerals overgrow and locally include remnants of the earlier schistosity, and are typically coarse and/or decussate. Oxygen isotope data for mineral pairs extracted from these selvages (mainly biotite-quartz) are consistent with 400 - 550°C equilibration temperatures. These relations indicate 1) the veins formed at P and T consistent with regional metamorphic conditions 2) the veins may be kinematically related to the waning stages of the upright folding event 3) at least some mineralization occurred at this time A regional appraisal of the oxygen isotope characteristics, whole rock geochemistry and petrology of the surrounding metapelitis and metapsammites has been undertaken in order to assess the broader patterns associated with the deposit Cartwright et al (1995) have undertaken a study 40-80 km to the north, in similar rocks, and found that the oxygen isotope ratios of pelites and marbles both decreased substantially with increasing metamorphic grade. Depleted values of 6 l s O at the high grade end were used to support a model of pervasive fluid flow, with fluid migrating towards the high grade portions of the belt. In this study, similar depleted 6 l s O values are obtained for most of the rocks within an 8 km radius of the deposit (9 to 11 %o) even in rocks only at biotite grade. Only 10 km south of the deposit do 6 l s O values

124

reach what would be considered "normal" for pelitic rocks (> 13 %6) that have not been pervasively infiltrated by fluid. These values are remarkably similar to worldwide values reported for granites, but can also be generated by long-distance up-temperature pervasive flow of initially externally derived fluid, with high total fluid fluxes (Dipple & Ferry, 1992; Cartwright etal. 1995). Geochemical patterns indicate, for various reference frames, that there has been lOOm-scale depletion of Ca, Na, Fe, and Sr, and similar scale enrichment of Cu and S, around and within the orebody relative to the broader surrounds. At metre-scales within the orebody (i.e. vein selvages), Si, Al, Zn and K have been locally redistributed.^ These patterns indicate either metasomatism due to interaction of rock with an exotic fluid, or equilibrium fluid flow downtemperature. The association of such changes with rocks that contain the F3 axial planar veins suggests that fluid travelled at least partly through veins prior to and during interaction with the wallrocks, unlike the regionally 018-depleted rocks which are relatively vein-poor and appear to have interacted with a pervasively infiltrating fluid. The model for ore genesis thus involves superposition of a focussed fluid flow system on a regional pervasive fluid flow system. The nature of the geochemical and isotopic changes suggests that broad fluid flow towards the higher temperature parts of the belt (from south to north) was overprinted by ore-producing fluid flow (upwards) above a granite pluton. The sulphur isotope values are more consistent with a granite-derived origin for the sulphur in the deposit than a syn-sedimentary origin. The sulphur cannot have been derived by direct reworking of regionally important syn-sedimentary sources (c.f. Seccombe et al. 1985), because the solubility of sulphur in metamorphic fluids is sufficiently low that the original and voluminous depleted signature should have been preserved, even if fluid fluxes were extremely high (Oliver et al. 1992). Pervasive fluid flow cannot readily occur in heterogeneous low permeability rocks undergoing significant deformation, because stress and strain partitioning couples intimately with fluid flow channelization. Therefore, the regional pervasive fluid flow system is inferred to have occurred after the main D2 deformation in the region had ceased, when permeabilities were high, perhaps due to stress relaxation immediately after D2. The onset of further shortening in D3, coupled with granite emplacement, provided the necessary conditions for subsequent focussing of regional and possibly granite-derived fluid into the aureole of an inferred late metamorphic granite, resulting in ore deposition. This study may thus have recognised the fluid flow paths associated with the "regional leaching" and "fluid focussing" components of a simple, generalized epigenetic ore formation model. References Cartwright, I.,Vry, J. K.& Sandiford, M., 1995. Changes in the stable isotope ratios of metapelites and marbles during regional metamorphism, Mount Lofty Ranges, South Australia: implications for crustal scale fluid flow. Contrib. Mineral. Petrol., in press. Dipple, G. M. & Ferry, J. M., 1992. Fluid flow and stable isotopic alteration in rocks at elevated temperatures with applications to metamorphism. Geochim. Cosmochim. Acta, 56, 3539-3550. Oliver, N. H. S., Hoering, T. C., Johnston, T. W., Rumble, D. Ill, & Shanks, W. C. III., 1992. Sulfur isotopic disequilibrium and fluid-rock interaction during metamorphism of sulfidic black shales from the Waterville-Augusta area, Maine, USA. Geochim. Cosmochim. Acta, 56,4257-4266. Seccombe, P. K., Spry, P. G., Both, R. A., Jones, M. T. & Schiller, J. C., 1985. Base metal mineralization in the Kanmantoo Group, South Australia: a regional sulfur isotope study. Econ. Geol., 80, 1824-1841. Thomson, B. P. (1975). Kanmantoo Trough - regional geology and comments on mineralisation. In (Knight, C. L. ed.) Economic Geology of Australia and Papua New Guinea, Vol. 1, Metals. Melbourne, Australian Institute of Mining and Metallurgy, 555-560.

125

DEHYDRATION AND ANATEXIS CAUSED BY MAFIC MELT MIGRATION THROUGH FRACTURES, HALLS CREEK OROGEN, EAST KIMBERLEYS N, H. S. Oliver1 & T. D. Barr2 1. School of Applied Geology, Curtin University, GPO Box U1987, Perth, W.A. 6001 2. Department-of Earth Sciences, Monash University, Clayton, Victoria, 3168 Models for low pressure, high temperature "regional-contact" metamorphism appeal to the role of magmatism as the primary cause for heat advection (Lux et al. 1986, Sandiford et al. 1991, 1995). Some migmatites may form essentially as a contact metamorphic product of interaction of high grade metamorphic rocks with such magmas. In this paper, an outcrop is described of low pressure granulites from north-western Australia in which the interplay between deformation, migmatization, dehydration and melt emplacement can be directly examined. Short-lived orogeny was responsible for the formation of the mid-Proterozoic Halls Creek Orogen. Voluminous granitic and gabbroic melts intruded over a relatively short time interval, several of these being synchronous with deformation phases and regional metamorphism. The belt now comprises > 80% intrusive igneous rocks, so it comes as no surprise that the regional metamorphism in the high grade part of the belt is of a low pressure, high temperature nature. Peak metamorphic conditions were 750 - 850 °C at 3 to 4.5 kbar (Thornett 1986), and the pelitic rocks are dominated by cordierite and sillimanite, with a mappable garnet isograd occurring in already migmatitic rocks. Preliminary SHRIMP U-Pb analyses of zircons from granitic leucosomes reveals complex inheritance problems, and, as yet, little temporal resolution between different deformations and leucosome-forming events. In key outcrops in the Fletcher Creek Rockholes, pelites, mafic orthogneisses and psammites are strongly deformed by a composite S1-S2 contact-parallel fabric, and then folded around broad E-W upright F3 folds. Temperatures remained high throughout these deformations, as indicated by the abundance of migmatites formed throughout the 3 deformations, and the presence of abundant post-S2 pyroxene porphyroblasts in mafic rocks. In the pelites and psammites, common stromatic quartzo-feldspathic migmatitic leucosomes are parallel to the SI fabric, but leucosomes cutting this fabric are abundant, particularly near the margins of some mafic intrusive dykes. These second generation melts occur in three modes: 1) garnet-bearing boudin-neck leucosomes. These may have developed during D1 or D2 deformation, but appear to crosscut the Sl-parallel stromatic types; 2) locally garnet- or cordierite-bearing shear-zone leucosomes. These appear to be geometrically related to the D3 folding event, and occur in smalldisplacement (up to 2 m) brittle-ductile shear zones; and

126

3) garnet-, cordierite- and biotite-bearing "remelts" around mafic intrusions. These occur as 10 to 50cm-wide zones irregularly following the boundary of mafic dykes intrusive into the sequence, locally back-veining into the dykes. The mafic intrusions occur predominantly as syn-D3 dykes, but locally have intruded as sills as early as Dl. The boundaries of the dykes are extremely lobate, this observation alone suggesting they were intruded into incompletely solidified wallrocks. Simple one-dimensional heat flow modelling supports this hypothesis, in that the wallrocks must have been hot at the time of emplacement of the dykes in order to allow remelting. Fractures and shear zones also cut mafic meta-igneous rocks, and are locally filled with pyroxene-bearing (possibly cumulate) gabbro. These shear zones lie in a similar orientation to the ones in the metasediments, and are inferred to have formed at the same time (during D3). Dehydration of 2-pyroxene-hornblendeplagioclase granulite to a hornblende-absent assemblage occurs immediately adjacent to the shear zones, and similar assemblages are present in vein-like arrays ("slugs") throughout adjacent outcrops, even though they are not cored by melt. These relations suggest localized dehydration occurred due to passage of the water-poor melts. Detailed mapping of several parts of the outcrop reveals a consistency in the orientation of the late melt networks, even in the alignment of decimetre-scale zones of internally chaotic schollen (raft) migmatites (Figure). This requires that, even though melt was present, the rocks could withstand sufficient load to form both an aligned fracture array, and regional-scale F3 folds. Although the fractures locally formed conduits for crosscutting mafic melt (that enhanced local anatexis in pelites and dehydration in meta-mafic rocks), some fractures appear to have acted primarily as sinks for still partially molten felsic material within the outcrop, drawing in melt at cm- to m-scales. The rocks are similar in this respect to lower grade rocks that have undergone shear- and tensile-failure at high fluid pressures. Detailed penological, geochemical and isotopic work is underway to assess the scale of melt migration within this fracture array. The rapid sequence of events implied by petrographic relations, one dimensional heat flow modelling and preliminary geochronology suggests that intrusion, metamorphism, and deformation were intimately linked. These rocks thus also provide a good location to test models that appeal to localization of deformation due to thermal weakening by melt emplacement (e.g. Sandiford, 1991) References Lux, D.R., DeYoreo, J.J., Guidotti, C.V., & Decker, E.R. 1986. Role of plutonism in low-pressure metamorphic belt formation. Nature, 323, 795-797. Sandiford, M., Martin, N., Zhou, S. and Fraser, G. 1991. Mechanical consequences of granite emplacement during high-T low-P metamorphism and the origin of "anticlockwise" PT paths. Earth and Planetary Science Letters, 107,164-172. Sandiford, M., Fraser, G., Arnold, J., Foden J., and Farrow, P. 1995. Some causes and consequences of

127

INFLUENCE OF TECTONICS ON THE OCCURRENCE OF TALC IN THE MOUNT LOFTY RANGES, SOUTH AUSTRALIA Robin L Oliver Deptartment of Geology and Geophysics, University of Adelaide, SA 5005 Talc deposits in the Mt Lofty Ranges outcrop most abundantly in a north-south trending, structurally controlled zone incorporating particularly Gumeracha and the Pewsey Vale and Mt Kitchener areas, east of the Barossa Valley (Fig 1). Typically the talc occurs as pods within the metamorphosed Woolshed Flat Shale which stratigraphically overlies the Montecute Dolomite (= Skillogalee Dolomite) in the Burra Group in the Adelaidean (Neoproterozoic) sequence. The metamorphosed Woolshed Flat Shale consists essentially of muscovite + biotite + quartz. Completely sericitised andalusite(?) porphyroblasts manifest retrogradation from the regional sillimanite grade to the east. Associated calc-silicate lithologies consist of hornblende-actinolite + epidote-clinozoizite + scapolite + albite. The retrogradation is likely to be a consequence of hydration and metasomatism concentrated along and affecting the upthrust hanging wall of the north-south fault between Gumeracha and the Barossa Valley. Close association of the talc with concentrations of metasomatic albite feldspar (albitite) is particularly noteworthy. Typically, the talc pods are surrounded by albite "aureoles". Elswhere, and in more detail, coursely crystalline albite is in the form of veins or diffusely replacing the micas and quartz of the country rock; in general, the degree of albitisation decreases with distance from the talc lodes and also the calc-silicate lenses. It is suggested that the talc is produced from dolomite (completely consumed) plus quartz during regional metamorphism: 3CaMg(C03)2+ 4Si0 2 + H 2 0 = MgaSiJO + 3CaC03 + 3C02 Tremolite-actinolite may be formed at a slightly higher temperature: 6CaCOs + 4Si0 2 + 5(MgFe)3Si4O10(OH)2 = 3Ca2(MgFe)5Si8022(0H)2 + 6C0 2 + 2H 2 0. Whether talc or actinolite are produced may be dependent also on the partial pressure of C0 2 which is influenced by the concentration of Na in the fluid. It has been suggested that magnesite (also completely consumed) is an even more likely source of the magnesium for talc formation: 3MgC0 3 + 4Si0 2 + H 2 0 = Mg3Si4O10(OH)2 + CCX,. Magnesite layers and lenses are not uncommon within the Montacute (= Skillogalee) Dolomite. Field and laboratory indications are that the production of albite is at the expense of mica: KAl3Si3Oio(OH)2 + 6Si0 2 + 3Na = 3NaAlSi3Og + 2H + 2K

and

K(MgFe)3(AlSi)3O10(OH)2 + Na = NaAlSi3Og + K + 3(MgFe) + 2H

128

The source of the sodium to produce the albite concentrations is considered to be halite-rich strata (completely consumed) within the Woolshed Flat Shale. Halite and magnesite are likely associates. It is considered that the development of the talc and associated albite is a consequence of agitation and retrogradation to an appropriate PT on the eastern, hanging-wall side of the northsouth trending Gumeracha-Pewsey Vale fault.

Reference Dymoke, P. and Sandiford, M., 1992. Phase relationships in Buchan fades pelitic assemblages: calculations with application to andalusite-staurolite parageneses in the Mt Lofty Ranges, South Australia. Contr. Min. Pet. 110, 121-132.

Biotite Zone Stauroliteandalusite Zone Fibrolitic sillimanite Zone Prismatic sillimanite Zone Migmatite Zone Basement inlier

Upper limit of stauroliteandalusite

Figure 1. Map of part of the Adelaide Fold Belt (from Dymoke and Sandiford, 1992) showing talc localities in relation to regional metamorphic zonation

129

FISSION TRACK EVIDENCE FOR LATE PERMIAN-EARLY TRIASSIC AND MIDDLE CRETACEOUS DENUDATION IN THE EASTERN LACHLAN FOLD BELT, NSW, AUSTRALIA P.B. O'Sullivan. B.P. Kohn, D.A. Foster, & A.J.W Gleadow Australian Geodynamics Cooperative Research Center School of Earth Sciences, La Trobe University, Bundoora, Victoria 3083, Australia The recent literature abounds with references describing the timing and the tectonic development of the Lachlan Fold Belt of southeastern Australia (e.g. Fergusson and Coney, 1992, and references therein). Most have proposed that deformation responsible for the regional shortening (estimated up to -60%) and formation of structures throughout the fold belt occurred along a convergent margin setting during Early Silurian to Middle Devonian and Early Carboniferous. The presence of "seemingly undeformed" Late Carboniferous granites, eg. the Bathurst Batholith within the eastern Lachlan Fold Belt in New South Wales, suggest that all deformation ceased prior to their intrusion. However, due to the lack of relevant geologic control and incomplete low temperature geochronology, the timing of any post Middle Carboniferous denudation within the fold belt is largely unconstrained. In an effort to constrain the post Middle Carboniferous thermal and tectonic history of the Lachlan Fold Belt, a growing data base of apatite fission track analyses from sedimentary and granitic rocks exposed within the fold belt is being generated. Fission track results from surface samples collected from the eastern part of the fold belt (Fig. 1) suggest that rocks experienced two distinct episodes of rapid cooling since the Carboniferous. The episodes are supported by relationships between the apatite fission track ages and mean confined track lengths, as well as between apatite ages and standard deviations of track lengths. The youngest apatite ages (-90-100 Ma) and the oldest apatite ages (-240-250 Ma) tend to have the longest mean confined track lengths (>13 |lm) and the narrowest length distributions (standard deviations <1.5 |im), whereas the intermediate apatite ages (between -100-240 Ma) tend to have shorter mean confined track lengths and broader distributions (Fig. 2). Such patterns are indicative of thermal annealing of fission tracks in apatite at elevated paleotemperatures (up to ~110°C), affecting the suite of samples to varying degrees within a common style of thermal history, followed by a period of rapid cooling (Green, 1986). The first cooling episode (Late Permian-Early Triassic) is probably related to the Late PermianEarly Triassic Hunter-Bowen Orogeny which is known to have affected the New England Fold Belt and adjoining Sydney-Bowen Basin. Data presented here suggest that denudation occurred in the eastern half of the Lachlan Fold Belt in response to this orogeny as well. The relationship of denudation to structural style is not clear as denudation could have occurred in response to compressional reactivation of earlier structures or to more regional rock uplift. However, data from the Bathurst Batholith suggest that it experienced an episode of compressional deformation resulting in the repetition of paleo-isothermal surfaces recorded by the fission track data. The second cooling episode (mid-Cretaceous) occurred as a result of the onset of continental extension in the Tasman Sea at -96 Ma. At this time underplating might have occurred inward of the rift along the eastern margin of Australia resulting in a significant change in base level and km-scale denudation over much of the southeastern highlands, from Tasmania to northern New South Wales, between -90-100 Ma.

130

Figure 1. Regional map showing eastern New South Wales and the major tectonic blocks mentioned. Localities include: LFB, the Lachlan Fold Belt; BB, the Bathurst Batholith; CMB, the Clarence Morton Basin; NEFB, the New England Fold Belt; and SBB, the Sydney-Bowen Basin.

100

200

300

400

AGE (MA)

100

200

300

400

AGE (MA)

Figure 2. Apatite fission track data from the eastern Lachlan Fold Belt in New South Wales. Relationships are shown between: (a) mean track length (2a) and fission track age ( 2a); and (b) confined track length standard deviation and fission track age ( 2a). The broad trends suggested by the stippled bands are discussed in the text.

References

Fergusson, C.L. & Coney, P.J., 1992. Convergence and intraplate deformation in the Lachlan Fold Belt of southeastern Australia. Tectonophysics, 214, 417-439. Green, P.F., 1986. On the thermo-tectonic evolution of northern England: evidence from fission track analysis. Geology, 5, 493-506.

131

STRUCTURAL EVOLUTION OF THE KALINJALA MYLONITE ZONE S. A. Oussa Department of Earth Sciences, Monash University, Clayton, Victoria 3163 As a major tectonic and lithological boundary within the Gawler Craton, the Kalinjala Mylonite Zone is an important structure pivotal to the understanding of the processes active during the formation of the Gawler Craton. The zone extends the length of the eastern Eyre Peninsula (some 300 km) and is up to four kilometres in width. The zone separates Archaean and Early Proterozoic metasediments to the west, from Middle Proterozoic orthogneiss to the east, and coincides with a major aeromagnetic high. Deformational structures from within the Kalinjala Mylonite Zone indicate that movement along the zone was protracted, and that the history of the zone was more complicated than previously recognised. Structural and metamorphic evidence indicates that the Kalinjala Mylonite Zone was originally a high grade structure active at deep crustal levels (to at least 30 kilometres) and that later during the deformation history of the Craton, movement along the zone took place at distinctively lower grades. The progressive and protracted nature of the deformation along the zone indicates that the Kalinjala Mylonite Zone played an integral and important part in the evolution of the Gawler Craton. Comprising a series of progressive and overprinting shear structures, the rock units contained within the zone display similar kinematics which provide evidence for dextral shearing. There are several generations of shear structures all of which are coplanar, but have developed during changed metamorphic conditions. The earliest shear structures comprise minerals that display high grade ductile deformation structures and contain mineral assemblages that give calculated pressures and temperatures of formation of at 840±95°C and 10.1±1.2 kbar. Later mylonites, shear bands and shear fabrics concentrated towards the centre of the zone comprise a complex array of overprinting and parallel planar fabrics that are of successively lower grade. These later overprinting mylonites and shear bands have mineral grains that display structures characteristic of development at Lower Amphibolite and Greenschist facies grade. These structural and metamorphic indicators suggest that the Kalinjala Mylonite Zone was probably active during both prograde and retrograde metamorphic history of the Gawler Craton.

132

DIRECT OBSERVATION OF DEFORMATION PROCESSES IN CRYSTAL MUSHES Y. Park1 and W.D. Means2 1 1

Department of Earth & Environmental Sciences, Korea University, Seoul 136-701, Korea Department of Geological Sciences, Univ. at Albany, Albany, NY 12222, USA

Deformation experiments on the stage of a petrographic microscope, on polyphase, nonsilicate crystal mushes, probably have a useful contribution to make to understanding meltpresent deformation of rocks in nature and in the laboratory. The microstructural evolution over time, in an optical field-of-view, is observed directly, and this provides unambiguous and spatially highly resolved information on the processes that are operating, on the grainscale and above. We have carried out isothermal deformation experiments on crystal mushes one grain thick containing crystals of three ammonium compounds (see Means and Park, 1994, Geology, 22, 323-326) and initial melt fractions of 5-50%, at strain rates of 10"4'5 sec*1 to 10"6 sec'1, using time-lapse video recording of microstructural evolution over periods of tens of minutes to tens of hours. Two strain rate-dependent behavior fields have been encountered. At higher strain rates, the crystal frameworks yield by crystal plasticity with attendant recrystallization, by grain boundary sliding, and by a melt-assisted parting process that converts lath-shaped grains into stubby segments. In pure shearing deformations, the melt is expelled as the crystal framework collapses, and the product of this "filter pressing" looks similar to material deformed plastically in the solid state without melt present. Abundant evidence of intracrystalline deformation remains, in the form of kink boundaries, subgrain boundaries, and recrystallized regions of reduced grain size. At lower strain rates, the framework again collapses in pure shearing and melt is expelled, but the grains are almost entirely "strain-free" throughout the deformation. We recognize two processes that permit framework collapse without crystal plasticity or fracture: "contact melting" at sites where double occupancy of space is threatened, and grain boundary migration-assisted grain boundary sliding. In the latter, a grain boundary migrates from a position blocking grain boundary sliding to a position suitable for grain boundary sliding. Samples that have been filter pressed in the lower strain rate regime look like high-grade metamorphic rocks. In a simple shearing experiment in the lower strain rate regime, we have observed a remarkable deformation-induced coarsening process, in which one phase coarsens by an order of magnitude during a bulk shear strain of about 10. All these processes will be illustrated with time-lapse video movies. We remain uncertain about the analogy between our experiments and deformation of natural crystal mushes. Our microstructures look familiar from rocks, and our processes seem possible in nature. But what are the rigorous scaling rules for experiments of this kind? If we accelerate a natural strain rate by say, 10 or 106, what is the correct scaling of properties like the diffusivity of material through melt, or the reactivity at interfaces between crystals and melt? These questions will be discussed. 133

GEOSCDENCE DATA MODEL: TOWARDS A STANDARD FOR SPATIAL DATA TRANSFER A. John Parker1 & Richard G. Hume2 1 2

Geosurveys Australia Pty Ltd, 18 HighfieldAve, St Georges, SA 5064, Australia Australasian Spatial Data Exchange Centre,515 Bridge RdyRichmondy Vic 3121, Australia

The GEOSCDENCE DATA MODEL project is an AMIRA project (P431) supported strongly by Government and industry organisations. It aims to develop an Australian standard data model and data dictionary to provide a common basis of understanding for geoscience data. This is important to the digital geoscientific computing industry as a standard data model is required for the transfer of data between different computing systems both within an organisation and between organisations. More importantly, it will provide a way to develop computer systems in the future with this common understanding built in. Furthermore, what is frequently overlooked is that this common basis for understanding applies not only to digital data but also to textual and graphical information and the way we collect and record it both in the field and in the office. Recently, Standards Australia adopted the SDTS spatial data transfer standard as a basis for the transfer of cartographic and geographic data and spatially-related data (AS/NZS 4270). SDTS originated from the US National Institute of Standards and Technology and is now being widely implemented thoughout North America particularly in spatial database systems (eg GIS software such as Arc/Info and Maplnfo). In Australia, it has already been adopted and implemented for cadastral data. The GEOSCIENCE DATA MODEL is to be seen as independant of SDTS but the standard forms the framework upon which the project is based. The SDTS framework can be described as: 1. a conceptual model, 2. a data structure model, and 3. a data transfer model. The conceptual model varies with use and scale. For example, an outcrop or mine can be modelled as a point on a map of Australia, a polygon on a large scale map, or a 3-D object. The GEOSCIENCE DATA MODEL will not specify generic spatial primitive objects but will leave them to users to model as required. The data structure model defines much of the information needed for a complete description of the data, in this case geoscience data, and includes spatial reference, quality, a dictionary of terms and relationships between terms. The transfer model encapsulates this data structure for transfer. The primary aim of the GEOSCIENCE DATA MODEL project is to define the data structure model and, as geoscientists, our objective is to ensure that the model is complete, concise, accurate and describes all the data entities, attributes and relationships as we would require.

134

The first stage will be to develop a data dictionary to define entity types, attributes and the sources of this information. This list of terms would have no hierarchy, attributes would not be assigned to entities, and included (alternate) terms would be listed separately and crossreferenced. The concept of included terms is important because it allows us to "distill" a large number of related features into a few entities. For example, the entity type "fabric" would have included terms: foliation, schistosity, cleavage, lineation, fold axis etc. The second stage of the project will be to develop a relational data model from the dictionary of terms, describe attributes for each entity and describe the relationship of entities and attributes to other entities. For example, the entity type fabric might include the attributes: fabric type, declination (or dip), azimuth, etc. The GEOSCEENCE DATA MODEL project is up and running but, due to the restricted resources available, it has been agreed that priority be given to four broad themes: 1. Geochemical Data, 2. Drilling Data, 3. Geological Data (including structural data), and 4. Mineral Resource/Reserve Data. The project remains open for additional sponsors and, in particular, input from all geoscientists with an interest in ensuring that standards adopted by Government, industry and the broader geological community are consistent with current and potential or future geoscientific practices.

135

NON-SIMPLE SHEAR ZONES C.W. Passchier Department of Geosciencesy University of Main, 55099 Germany

In studies of shear zones, it is commonly assumed that flow in the zone was simple shear. This is necessarily so if wall rocks did not deform and no volume change occurred in the shear zone, but many shear zones develop in foliated rocks where the assumption of simple shear flow may be invalid. In order to assess the consequences of non-coaxial flow types other than simple shear in shear zones, a model is presented in which fabric accumulation is a function of vorticity and three principal stretching rates. For general non-coaxial flows in shear zones it is not possible to use the concept of a 'flow plane1 as defined for simple shear. Instead, it is possible to define a planar or linear 'material line attractor' towards which all material lines rotate. This is a concept that allows direct comparison of fabrics developed in different flow regimes. Planar and linear mylonitic fabrics, sheath folds and the shape of mantled porphyroclasts are presented as results of different arrangements of this material line attractor. For progressive deformation, it is also possible to expand the idea of a material line attractor and to define a general 'strain axis attractor' towards which the long axes of the strain ellipsoid rotate as a function of stretching rates and vorticity. Similar attractors can be envisaged for crystallographic preferred orientation elements and steady state foliations, leading to the general concept of 'fabric attractors'. Comparison of different flow types and resulting progressive deformation sequences lead to the following results: (1) In plane strain, shear zones may be stretching or shortening, but stretching shear zones are expected to be more common. Particular microstructures such as shear band cleavage may help to distinguish simple shear shear zones form stretching ones. (2) For a specific strain ratio, shear strain measured over a stretching shear zone may exceed that for a simple shear zone. This means that if simple shear is assumed, strain in the zone is severely underestimated. Shear zones in the Rosas granodiorite in Spain are shown as an example of such stretching shear zones. (3) If a shear zone is stretching parallel to the vorticity vector, stretching lineations may develop parallel to this vector under some circumstances, i.e. normal to the 'movement direction' of the shear zone. This implies that shear sense indicators are found on planes normal to the stretching lineation. The strike-slip Cap de Creus structure in Spain is discussed as an example of this type of shear zone.

136

PRE-ISAN EXTENSION: EFFECT ON THE GEOLOGY OF THE MITAKOODI QUARTZITE/OVERHANG JASPILITE CONTACT, KURIDALA REGION, MT. ISA. Warren A. Potma Australian Geodynamics Cooperative Research Centre Dept. of Earth Sciences, Monash University, Melbourne, Vic 3168, Australia A newly discovered angular unconformity at the base of the Overhang Jaspilite has defined the relative timing of syn Cover Sequence 2 extension within the Mitakoodi Culmination. Half graben style normal faulting associated with this extension has been discovered within the Mitakoodi Quartzite on the east limb of the Culmination. If the effects of later folding are removed the original orientation of the normal fault (here named the Buka Rock Fault) appears to have been hanging wall to the SE. Similar normal faulting has previously been reported on the west limb of the Mitakoodi Culmination by Williams (1989). The Buka Rock Fault exhibits a normal displacement of nearly 900 m and terminates at an angular unconformity which locally forms the contact between the Mitakoodi Quartzite (Malbon Group) and the overlying Overhang Jaspilite (Mary Kathleen Group). Pre-existing U-Pb dates for the underlying Argylla Formation (1766± 23 Ma) and the Mary Kathleen Group (1720±7 Ma), deposited immediately above the unconformity (Blake 1987), enable constrait on timing of extension. Local fold morphologies resulting from shortening during the Isan Orogeny (15201620 Ma) have been directly affected by the presence of the early extension fault. Mechanisms similar to those which occur during structural inversion (de Graciansky 1989) have locally reoriented folds and increased their intensity as the footwall of the normal fault has acted as an extremely competent structural backstop. The current map pattern resulted from a complex interaction of early extensional structures and up to three later shortening phases during the Isan Orogeny. References Blake, D.H., 1987. Geology of the Mount Isa Inlier and environs, Queensland and Northern Territory. BMR Bulletin, 225, 1-83. De Graciansky. P.C., Dardeau, G., Lemoine, M. & Tricart, P., 1989. The inverted margin of the French Alps and foreland basin inversion. Geological Society Special Publication, 44, 87-104. Williams, P.R., 1989. Nature and timing of early extensional structures in the Mitakoodi Quartzite, Mount Isa Inlier, northwest Queensland. Australian Journal of Earth Sciences, 36, 283-296.

137

CALCULATED MINERAL EQUILIBRIA: REPRESENTATION, AND RELEVANCE TO TECTONOTHERMAL EVOLUTION. R. Powell Dept Geology, Uni Melbourne, Parkville, Victoria 3052, Australia Calculated metamorphic mineral equilibria, involving geothermometry/ barometry, petrogenetic grids and other arcanery, may seem a long way from the concerns of most SGTSG members. However, in dealing with rocks in most orogenic belts, there is much that can be learnt in the interface between mineral textures and theoretical petrology. During the fabric development of a rock, there is a concomitant change in the mineral assemblages and mineral compositions that occur as the rock follows a pressure-temperature path. Calculated mineral equilibria allow that development to be put into a tectonic framework by allowing the pressure-temperature path to be deduced. Through this, the evolution of the orogenic belt may be understood in terms of the tectonothermal evolution of the lithosphere. It is only in the last few years that such calculations have become possible, through such software as THERMOCALC (Powell and Holland, 1988, J. met. Geol., 6, 173-204; and recent upgrades). In the poster, some of the new developments in THERMOCALC will be outlined, particularly in relationship to producing 3-D phase diagrams, and, via Mathematica, representing them as QuickTime movies.

138

THE GREAT AUSTRALIAN LINEAMENT DEBATE - WHAT IS IMPORTANT? W, L. Power & S. J. D. Cox Australian Geodyrtamics Cooperative Research Centre CSIRO Exploration and Mining, FOB 437, Nedlands, W4 6009 Australia Study of linear features (lineaments) in the Australian continent has led to much debate over the past 40 to 50 years, with many studies discussing the importance of linear or quasi-linear features visible in various forms of geological and geophysical data, over a wide range of scales. As the Australian Geodynamics Cooperative Research Centre moves towards the compilation of one of its major goals, a 4D geodynamic model of Australia, it was considered desirable to place the lineament debate in context. Are lineaments an essential tectonic element of the Australian continent, or do lineaments only exist in the imagination of practitioners of an inexact, non-systematic art? How can we reconcile the involvement of multiple cratonic blocks of widely disparate ages with continent-scale linear structures which were active for significant periods in the geological history? In this study, we consider the terminology used to describe linear features. Small features (like those observed in air photos for example), as well as continental-scale features have been termed lineaments, and this has confused the discussion. Considering only the large scale features, a major division may be made between those features derived from primary, regional or local scale linear features such as joints, faults, and shear zones; and lineaments comprised of aligned point features such as ore deposits. Despite the fact that many proposed lineaments are defined as occupying corridors of substantial width, continent scale features that are straight in one map projection have noticeable curvature in other projections. We suggest that where there is 1) certainty about the origin of a linear feature, or 2) a feature is identified in one, homogeneous data set, or 3) where a clear genetic basis for the feature can be established, that terms such as fault, aligned joint system, photo-lineament, gravity-lineament, or similar, specific terms be used. Lineament should be reserved for cases where the origin of the feature(s) is uncertain or compound. As part of this study, we reviewed lineaments at three relatively broad scales as case studies. These are the G3 gravity lineament, the Darling River Lineament, and lineaments seen in remote-sensing imagery in the North Bowen basin (Figure 1). In the case of the G3 lineament we consider two sides of the problem. First, does the most recently compiled gravity data show features which are continuous with, or coincident with the G3 corridor as described by O'Driscoll (1986)? Second, what is the connection with geological information, in particular in structural and time relationships. We find that, although there is a rough alignment of features within the gravity, this is neither straight or continuous, and appears to coincide with features in the geology which are unrelated to each other, some of which are also at a high angle to the G3 trend. The existence of a continentscale structure in the G3 position is highly speculative. In contrast, the Darling River Lineament appears to be a more substantial feature. Scheibner (1974) suggested that the Darling River lineament is a now inactive transform fault, which has

139

been partially dismembered by later faulting. Here we have a genetic hypothesis which is potentially testable. It should be possible to constrain both the period of activity of the structure, and the amount and timing of major transform offsets. Although much of the terrane through which the lineament passes is covered by younger sedimentary rocks, so much of the critical geological data is difficult to assemble and evaluate, the available information appears to suggest that it marks the location of a major geological boundary. Recent geochronological work also suggests relative vertical offset in the SW segment which, however, has the opposite sense to that observed in the aligned offshore bathymetry. The connection between these onshore and offshore segments across the New England fold belt is obscure. One possibility is that a weakness has been exploited and reactivated as a passive-margin scissor or transfer fault during the opening of the Tasman.

Figure 1. Location map of major lineaments considered as case studies. G3=G3 gravity lineament proposed by O'Driscoll (1986). DRL=Darling River lineament, as discussed by Scheibner (1974). NBBL=Study area for the north Bowen Basin lineaments discussed by Leach et al., (1986, 1987).

We also reviewed a finer scale example of linear features, in the north Bowen basin. These structures were initially discovered using Landsat imagery, but then later found to be faults, at least onshore. Related observations include offsets of Pleistocene coral reefs; alignments of seismic events, both onshore and offshore; and the approximate collinearity of the onshore faults with transform faults in the Coral Sea. In this case, careful investigation of additional evidence revealed that features originally identified as just "lineaments" in imagery could be related to a substantial geological explanation.

References Leach, J H J, Mallet, C W, & Hobbs, B E, Structure of the Bowen Basin, Queensland, Australia, 13th CMMI Congress, Geology and Exploration, 2,73-78,1986. Leach, J H J, Rynn, J M W, Hammond, R H, & Mallet, C W, The possible correlation of major geological structures m the Bowen Basin, bathymetric contours in the Coral Sea, and regional seismic activity in north ^ east Australia, In: Proceedings. Pacific Rim Congress 87. Gold Coast, Australia, p. 741-743,1987. O'Driscoll, E S T , Observations of the lineament-ore relation, Phil Trans. R. Soc. LondA 317, 195-218, 1986. Scheibner, E, Fossil fracture zones (transform faults), segmentation, and correlation problems in the Tasman Fold Belt system, In: The Tasman Geosyncline - a Symposium, eds. A. K. Denmead, G. W. Tweedale & A F. Wilson, Brisbane, Geological Society of Australia, Queensland Division, p. 65-96,1974.

140

TECTONIC EVOLUTION OF THE MID-NORTH, SOUTH AUSTRALIA W.V. Preiss Mines and Energy, South Australia. Box 151, Eastwood, South Australia, 5063. The Mid-North region occupies the southern, north-south trending arm of the Nackara Arc of the Delamerian fold belt, imposed on the Adelaidean (Neoproterozoic) to Cambrian mainly sedimentary rocks of the Adelaide Geosyncline during the Delamerian Orogeny (ca 500 Ma). The arcuate form of the major Delamerian structures is not an orocline, but is controlled partly by varying orientations of the structural grain of the basement, partly by primary basin configuration, and partly by the interference of different Delamerian fold phases. Deposition in the Adelaide Geosyncline began in Willouran time at about 850 Ma with initial thin platform sediments followed by crustal extension and widespread basalt extrusion. Northwest-trending rifts developed, possibly controlled by a major linear basement weakness, the G2 corridor, with thick deposits of fine to medium-grained shallow-marine to lacustrine clastics and carbonates and local volcanics (ca 800 Ma). Associated evaporites are best developed in the intracratonic rifts of the Flinders Ranges, where they formed the source beds of later diapirs, but are minor in the Mid-North, where this sequence is represented by the strongly folded River Broughton Beds of the Spalding Inlier. At the start of Torrensian time (possibly ca 780 Ma), a new set of rifts developed along the meridional Torrens Hinge Zone, causing subsidence of the Willouran western rift shoulder and the first inundation of the basement in the Adelaide region. In the Mid-North, thick fluvial sand with minor silt, dolomite and conglomerate (Rhynie Sandstone) accumulated adjacent to active fault scarps bounding the Gawler Craton to the west. There was local basaltic volcanism. Marine transgression then led to deposition of finer grained clastics and minor dolomites, prograding up into deltaic sandstone (Bungaree Quartzite). These early Torrensian deposits comprise the Emeroo Subgroup of the Burra Group. Further very shallow marine transgression deposited the Skillogalee Dolomite, with a lower, pale coloured member, and an organic-rich, dark, upper member containing chert and bedded and intraclastic magnesite precipitated in coastal lagoons under groundwater influence. During the early Torrensian, the Spalding Inlier is interpreted to have been an east-tilted horst, forming a barrier between a deeply subsident trough to the west in which the full sequence of Emeroo Subgroup accumulated and the remainder of the basin. A much reduced thickness of Emeroo Subgroup onlapped the Willouran sediments on the east flank of the tilted horst. The barrier still existed during deposition of the Skillogalee Dolomite, which received major sand influx from the west but is thinner and devoid of clastics on the east flank. The barrier had little effect on sedimentation of the upper Burra Group, comprising several transgressiveregressive cycles of carbonaceous siltstone and dolomite deposited below storm wave base shallowing up to coarse-grained deltaic sandstone. The Burra Group is separated by a disconformity from the overlying Umberatana Group, commencing with the Sturtian Appila Tillite (possibly ca 700 Ma). The western region occupied a shallow marine shelf with up to 500 m of glaciomarine diamictite. Renewed rifting further east produced north-west trending troughs in which much thicker glaciomarine sediments (Pualco Tillite, Benda Siltstone and associated Braemar ironstone facies) accumulated, and where the angular unconformity with the overlying dropstone-bearing Wilyerpa Formation is due to rotation of extensional fault blocks. The carbonaceous, thinly laminated silts of the Tapley Hill Formation represent the first transgression beyond the Adelaide Geosyncline onto the cratonic Stuart Shelf. The Torrens Hinge Zone now became a zone of gentle flexuring, the thickness of the Tapley Hill Formation increasing gradually eastward from the Stuart Shelf into the deeper basin. The upper part records shallowing, culminating in the oolitic and stromatolitic Brighton Limestone.

141

At the beginning of Marinoan time (ca 650 Ma), renewed tectonism generated local incision of the Sturtian sediments. The ensuing transgression deposited very shallow marine sandy silt in the west and deeper water laminated mud in the east. This succession contains several transgressive-regressive carbonate cycles in the shallower water regions, mainly in the Flinders Ranges. The Yerelina Subgroup records the second (Marinoan) glaciation, with several glacial advances and retreats. In the shallow shelf regions, these are disconformable on lower sediments of the Umberatana Group, but there are no known angular relationships; in the deeper basin, sedimentation may have been more continuous as the first glacial dropstones appear in laminated silt. The Wilpena Group is represented in the Mid-North mainly by the deeper marine Ulupa Siltstone, with the transgressive Nuccaleena Formation at the base overlying a regional disconformity. No Cambrian sediments are preseved in the Mid-North. The low metamorphic grade of the Neoproterozoic rocks makes it unlikely that there was ever a thick cover of Cambrian sediments such as those of the Kanmantoo Trough. During the Delamerian Orogeny, regional, relatively upright, commonly gently doubly plunging folds with average wavelengths of about 15 km and lengths of 100 km or more were imparted on the very thick sedimentary pile of the Adelaide Geosyncline in the Mid-North. Axial planes vary in dip from very steeply east through vertical and steeply west to gently west, and many folds display marked asymmetry toward the east. Axial plane slaty cleavage is commonly well developed, and is accompanied by mid-greenschist facies metamorphism with the growth of sericite, chlorite and green biotite. Locally in open structures cleavage is very weak, e.g. at Mintaro where flaggy bedding partings dominate, permitting the extraction of high-quality flagstone. This structural style contrasts with the thrust-dominated, northwest-vergent Delamerian deformation of the southern Mount Lofty Ranges. The difference can be properly understood only by considering both the basement-cover relationship and the timing of deformation. Near Adelaide, west of the basement inliers, Adelaidean rocks are of similar metamorphic grade to the Mid-North but much more intensely deformed. Competent units such as thick sandstones form rigid, fault-bounded, gently southeast-dipping "slabs", but pelitic units contain gently east-dipping cleavage and low-angle thrusts which then ramp up through the competent beds. Thrusts originate from retrograde shear zones within the basement, which is involved in the deformation. The presence of evaporitic Willouran sediments at depth within and northeast of the G2 corridor is inferred from the Spalding Inlier and the small diapir at Burra. As in the Flinders Ranges where these were the source beds of the major diapirs, the Willouran sediments provided a detachment between the basement and cover in the Mid-North. East of Adelaide, the main cleavage and thrusting are assigned to Dl. Locally, these are crenulated and folded on north-south axes by F2. In the high-grade Adelaidean and Cambrian metasediments of the eastern Mount Lofty Ranges, F1 folds are rare. SI is sub-parallel to bedding and is refolded by F2 which forms the major, upright, north-south folds. These latter folds can be traced directly into the dominant meridional folds of the Mid-North, which are therefore also F2. There is no evidence of Dl having affected the rocks of the Mid-North, except along the western margin of the ranges southwest of Clare, where north-south folds and crenulations affect highly foliated phyllitic metasiltstone and dolomite of the lower Burra Group. A series of poorly exposed granitoid plutons was intruded, possibly syn-D2, east of the ranges along the western margin of the Murray Basin. After substantial stripping of the Delamerian mountain chain in the early to mid-Palaeozoic, hydrothermal fluids associated with a small, subvolcanic intrusive at Burra introduced the copper orebody in the Skillogalee Dolomite. The Mid-North was peneplained and deeply weathered, probably largely in the Mesozoic. During the Cainozoic, the Flinders and Mount Lofty Ranges and Mid-North have undergone persistent uplift and incision as Delamerian reverse faults are being reactivated under an east-west compressive regime.

142

INFLUENCE OF WATER ON THE CHARACTER OF MYLONITE PRODUCED BY SHEARING DURING LOW GRADE METAMORPHISM IN FELDSPAR RICH ROCKS L. L. Pryer Research School of Earth Sciences, Australian National University, Canberra 0200. The character of mylonites produced by shearing in granites and other feldspar-rich rocks, during low grade metamorphism is strongly influenced by the amount of water available both before, during and after mylonitisation. In addition to the change in effective pressure and its control on 1) rock strength, 2) increased fracture propagation, and 3) possible increased quartz ductility, water has an important chemical control over the alteration of many of the minerals responsible for the strength of the rock. The strength of the feldspars, (as well as other high temperature phases), is directly related to the extent of hydration in these minerals at the time the stress is applied (reaction enhanced weakening) which, in turn, controls the character of the resulting mylonite. Therefore, mylonites produced during wet mylonitisation differ from those produced by dry mylonitisation. In the presence of water, medium and high temperature minerals tend to be unstable at low grade, breaking down to form fine grained masses of hydrated minerals such as 'sericite1, chlorite, epidote, muscovite, talc, etc. along with albite and quartz. These masses are extremely weak due to their fine grain size and inherently weak minerals. When subjected to a high differential stress, these masses are first to register strain and often localise it, such that shearing is initiated on the pseudomorphs of feldspars and hornblendes. At moderate to high total strain, any evidence of the existence of these phases is completely erased. Shearing results in the production of phylonites with strongly preferred orientation of the hydrous phases. During dry mylonitisation at or above the temperature necessary for quartz ductility unaltered feldspars in the host rock are generally the strongest phase. Failure during shearing produces angular to subrounded pieces of broken feldspar surrounded by quartz ribbons. The ultimate size and shape the feldspar fragments depends on the strain rate and total strain of the mylonite zone and on the composition of the rock itself. High matrix/feldspar ratio will result in larger fragments than one with a low matrix/feldspar ratio. At temperatures below that for quartz ductility, the quartz will be the strongest phase. Local concentrations of feldspars will fail brittly and tend to localise shear. The ultimate character of all of the fragments in the mylonite will be angular to rounded depending on the extent of abrasion the fragments have undergone. The ingress of water after shearing, will produce hydrated phases that are not preferentially oriented and that tend to occur in localised areas, pseudomorphing the fragments that they have replaced. The extent of feldspar alteration and the amount of preferred orientation in the matrix where the feldspars have been completely altered, can be used to determine the timing of fluid influx relative to shear deformation. An increase in the amount of fluid influx during progressive shearing produces a mylonite that displays characteristics of both wet and dry shearing, with localised zones of phylonite contained within wider zones of anhydrous mylonite.

143

THE THREE-DIMENSIONAL STRUCTURE OF THE MT. WHALEBACK MINE, NEWMAN, W.A. J. Ronaszeki1, B. Ascott2 & C.McA. Powell2 l

BHP Iron Ore Ltd., P.O. Box 655, Newman, W.A. 6753, Australia Department of Geology and Geophysics, University of Western Australia, Nedlands, W.A. 6907, Australia

The Mt. Whaleback Mine is situated in the southeastern Ophthalmia Fold Belt on the southern margin of the Pilbara Craton, Western Australia. Mt. Whaleback is the largest single-pit ironore mine in the world, with a projected remaining lifespan of about 30 years. The structure of the mine is complex and has a significant influence not only on ore genesis, but also on the stability of the open-pit itself. The final wall limits for the ultimate pit are currently under design, with particular importance being placed on design of the south wall in the west pit area, which will stand some 500 m high and be hosted in low-strength Mt. McRae Shale. The in-situ stress field is considered to have a strong influence on the stability of this wall. To model final wall open-pit designs accurately a 3-D structural model of the Whaleback Pit and its surroundings is currently under construction. This model is based on a good understanding of the complex polyphase deformation history of the mine, in particular the superposition of the three main fold phases. Four separate fold sets, Fi_4, have been identified in the Ophthalmia Fold Belt, of which F2-4 occur at Mt. Whaleback. Fi is a set of small-scale recumbent intrafolial folds which was recognised and described by Tyler (1991). F2 and F3 are the main regional fold sets developed in the fold belt at a range of scales from macroscopic to microscopic. F4 is a late NE/ENEtrending set of folds that produces mesoscopic dome-and-basin interference patterns where they coincide with minor F2 or F3 folds (Johnson, 1994). F2 folds are characterised by tight to close interlimb angles, moderate to shallowly-dipping axial planes, northward vergence, and overturned northern limbs. F2 folds are ductile in nature and commonly plunge shallowly to the west. However, the non-cylindrical style of the folds results in abundant plunge reversals. F3 folds are open and upright with a NW- to WNW- axial trend. Large-scale F3 folding is best recognised on shallow-dipping F2 upright limbs where they produce zones of bedding that dip shallowly to the north (Johnson, 1994). Johnson (1994) identified two major regional structural domains: a southern domain dominated by F2 (containing Mt. Whaleback) and a northern domain dominated by F 3 folds. F 2 is developed in the northern domain but in isolated zones and with less intensity compared with F2 in the southern domain. Regional-scale F2 folds are rare in the northern domain and hence F 3 dominates the fold pattern in this area. Johnson (pers. comm. 1995) considers that this spatial pattern of F2 folds relates to the distribution of blind thrusts at depth. F 4 folds are open and upright with NE-trending axial surfaces. These folds are characterised by a brittle deformation style, commonly with box-fold geometry (Ronaszeki, 1994). Mt. Whaleback Structure F 2 folds are the dominant fold set at Mt. Whaleback. The folds show a strong variation in interlimb angle from elasticas to open in profile and are commonly either overturned to the south or recumbent. Fold style is strongly dependent on lithology (ductility contrast) and the number and thickness of individual layers. Whereas the average regional trend of the F2 folds is around 270°, at Mt. Whaleback the folds show a wide variation in plunge direction from S to WNW (particularly with small-scale folds on the steep limb of the large F 2 folds). Two hypotheses are considered for this variation in F2 plunge direction: firstly, that the angular 144

variation relates to rotation of the hangingwall rock above two large-scale low-angle normal faults that truncate the mine's stratigraphy (if this is the case all the folds above the fault should have a common angular mismatch with the regional trend), or secondly, that the variation in the fold plunges is an artefact of the interference of large F3 and F4 folds that pass through the mine. Two large, low-angle, normal faults, the East Footwall Fault Zone (EFFZ) and Central Fault (CF), cut the F2 fold set. The CF is located 50 - 200 m above the EFFZ. Both faults have large offsets which increase east to west from about 100 m up to about 300 - 400 m, suggesting they are bounded to the west by tear-faults. The orientation of the CF and EFFZ is sub-parallel to the axial planes of the large-scale F2 folds but they truncate the Si axial plane cleavage and are not considered to be F2 axial plane faults (Ronaszeki, 1994). F3 folds at Mt. Whaleback are rarely developed at mesoscopic scale. Except for two occurrences where small-scale F3 folds (10 m wavelengths) can be observed to refold F2 folds, the occurrence of F3 folding can be inferred only from variation of F2 axial surface dips and fold plunges. The most common wavelength of F3 folds in the area is likely to be in the order of 1—2 km. F4 folds at Mt. Whaleback are more widespread than F3 folds. The folds tend to be smaller in wavelength commonly in the range of 10 - 200 m. The folds can be seen to refold the CF and EFFZ. In areas of poor exposure, F4 folds can be inferred from variation of F2 axial surface dips and fold plunges. A large high-angle fault zone, the Whaleback Fault Zone (WFZ) cuts the main F2 fold set and the low-angle normal faults (Hobbs et al., 1987). The WFZ is deformed in one area by F4 (Ronaszeki, 1994). Near-contemporaneity of the WFZ and F4 folds has been suggested, however, since the trend of the F4 folds parallels the trend of the South Whaleback fault. The final structural model has been produced by stacking several north-south cross-sections together. The sections have been constructed using existing drill-hole and face-mapping information and supplemented by collection of structural data from surface outcrop. Where factual information is limited, downdip projection has been applied and interpretation made based on the general nature of the fold styles. References Hobbs, B., Mason, R., Davidson, D. & Preston, K., 1987. Structural Geology of the North Wall of the Mt. Whaleback East Pit and Implications of Slope Stability. 1987. CSIRO Division of Geomechanics, 48pp. (unpublished). Johnson, T.M., 1994. The geometry of the SE Ophthalmia Fold Belt, Pilbara Region, Western Australia - A combination of two regional fold sets. Geol. Soc. Aust. Abstracts, 36, 85-86. Ronaszeki, J., 1994. Structural Geology of the Mt. Whaleback Mine, Newman, W.A. Geol. Soc. Aust. Abstracts 36, 136-137. Tyler, I.M. 1991. The geology of the Sylvania Inlier and southeastern Hamersley Basin . Bull. West Aust. Geol. Surv., 138, 108pp.

145

BASEMENT INVOLVED DEFORMATION, HEAT PRODUCTION AND LOWPRESSURE METAMORPHISM IN THE ADELAIDE FOLD BELT AND SOME WIDER IMPLICATIONS Mike Sandiford, Annette Bingemer & Martin Hand,

Department of Geology, University of Adelaide, Adelaide, 5005, Australia

Delamerian (520 - 450 Ma) deformation in much of the northern Adelaide fold belt is characterized by the coupling of Mesoproterozoic basement (e.g, the Mt. Painter and Willyama Inliers) with the Neoproterozoic cover despite excellent potential sequences for detachment (for example, the partly evaporitic Wywyana Formation in the Northern Flinders Ranges). Basement in this region is characterized by anomalous radiogenic heat production (regional values of at least 5 -10 ^Wnr ) and high heat flows (80-120 mWnr ), as well as by local lowpressure intermediate-temperature metamorphic aureoles developed above unconformities, suggesting that the thermal regime is the important factor controlling basement-cover behaviour. In the Arkaroola region, where the Mt. Painter Inlier is exposed in the core of a large amplitude, tight anticline truncated by reactivated strike-slip faults (the Paralana fault system), this unusual "unconformity related" metamorphism produced temperatures in excess of 500°C at depths of around 10 km. Mineral textural relations imply some cooling of the structure during fold amplification, suggesting that anomalous heat production rather than heat refraction (c.f., Mildren & Sandiford, 1995) was the important causative factor. Significant lateral thermal gradients persisted throughout structural development implying that lateral variations in basement heat production helped localize deformation. Changes in the depth of burial of an anomalously radiogenic basement by only a few kilometres can effect dramatic changes in the thermal structure in the mid-deep crust, potentially changing Moho temperatures by up to 20-40°C per km of burial (depending on choice of thermal parameters). Since lithospheric strength is very sensitive to Moho temperature (for the "BraceGoetz" lithosphere, a change in Moho temperature of 100°C produces an order of magnitude change in strength), relatively slight changes in depth of burial of radiogenic basement must have important implications for the dynamics of distributed deformation in continental interiors (both during extensional and compressional deformation), in a wide variety of settings (intracratonic basins, Adelaide fold belt, Alice Springs orogeny etc.) as well as the origin of the low-pressure, high-temperature, Mesoproterozoic metamorphism that characterizes the basement complexes. References 3

Mildren, S. and Sandiford, M., 1995, A heat refraction mechanism for Low-P metamorphism in the northern Flinders Ranges, South Australia, Australian Journal of Earth Sciences, 42, 241-247.

146

CONTINENTAL COLLISION, RIDGE TORQUES AND STRESS IN THE INDIAN-AUSTRALIAN PLATE Mike Sandiford & David D. Coblentz Department of Geology, University ofAdelaide, Adelaide, 5005, Australia Many of the long-wavelength features of the Indian-Australian plate stress field are explicable in terms of a balance between ridge torques and the "collisional" torques imposed along the Himalayan, New Guinea and New Zealand collisions (Coblentz et al., 1995), motivating a reanalysis of the mechanical evolution of the Indian-Australian plates over the last 50 Ma (Sandiford et al., 1995). The excess gravitational potential energy of a collisional orogen that may be supported by focusing distributed ridge torques in a plate depends on the ratio V of the effective lengths of ridge systems and collisional fronts. In the Indian-Australian plate, the net torque due to the asymmetric distribution of midocean ridges and young oceanic lithosphere along the southern boundary is -8.4 x 10 N.m and is capable of supporting an excess potential energy of the Himalayan collision zone of up to 7.5 x 10 J/m . The concept that collision in the IndianAustralian plate may be sustained by ridge torques suggests that collisional driving stresses may be sensitive to /'. In the Indian-Australian plate, V changed significantly during progressive collision from -55 to 45 m.y., during amalgamation of the Indian and Australian plates at -45 m.y., and during the development of the New Guinea orogen through the late Oligiocene and Miocene. Such changes in /' may help explain the termination of spreading between India and Australia following the onset of the Himalayan collision and the early Miocene normal faulting in the high Himalaya. References 25

Coblentz, D., Sandiford, M., Richardson, R, Zhou, S., and Hillis, R., 1995, The origins of the Australian stress field, Earth and Planetary Science Letters., in press. Sandiford, M., Coblentz, D., and Richardson, R.M., 1995, Focusing ridge-torques during continental collision in the Indo-Australian plate, Geology, in press.

147

SHAPE PREFERRED ORIENTATION OF SPINEL WITHIN DUNITE: EVIDENCE OF PLASTICALLY DEFORMED PERIDOTITE IN THE YAKUNO OPHIOLITE, JAPAN Kazuko Saruwatari1, Katsuyoshi Michibayashi1 and Ichiko Shimizu2 institute of Geoscience, Shizuoka University, Shizuoka 422, Japan 2

Geological Institute, University of Tokyo, Hongo 113, Japan

The study of shape preferred orientation (SPO) of spinel within dunite revealed that an ultramafic body of the Oshima Peninsula in the Yakuno Ophiolite, Maizuru Belt, SW Japan, had been pervasively deformed at the high-T conditions. The ultramafic rocks in the ophiolite consist of dunite and harzburgite members. The dunite member is separated into two parts: a layered wehrlite part and a dunite part (Ishiwatari, 1985). Dunites occur in the both the dunite and harzburgite members. The dunites in the dunite member are the principal lithology, whereas those in the harzburgite member are with harzburgites. We have measured four quantities on individual spinel grains to analyze the development of SPO: maximum length (L), minimum length (W), area of the grain (A) and angle (G) between a reference line and the maximum length. Two quantities, grain shape (R) and size of the grain (D) were calculated from these four quantities. The grain shape (R) is defined as R=L/W. The grain size (D) is calculated as D=2sqr(A/rc). Based on the measurements, we defined the average angle to be the main SPO of spinel grains for each sample, and the measured angles were normalized by the average angle to study the orientation distribution (<|>) with respect to the SPO. The results are as follows. The <|) distribution for each sample shows an SPO ofspinel grains that is oblique to the compositional layers. This indicates that the foliation and lineation of these rocks have been created tectonically. The development of SPO of spinel grains depends on thengrain size, becoming stronger as the grain size increases. These spinel grains show compositional zoning in their Al-Cr contents. Sincethis Al-Cr zoning pattern appears to be formed by a stress-induced diffusion creep under high temperature conditions (Ozawa, 1989), this zoning pattern must have resulted from plastic deformation at high temperature (more than 900°C). Therefore spinel grains can be used as strain markers in the ultramafic body. Rf/(|) method was employed for strain analysis. The values of Rs from the method are plotted in a Flinn diagram. The results show that the spinel grains in the dunite member are oblate in shape and in the harzburgite member are prolate. This suggests that the strain form in the ultramafic body changes from the upper part toward the lower part. References Ishiwatari, A. 1985. Granulite-facies metacumulates of the Yakuno ophiolite, Japan: Evidence for usually thick oceanic crust. Journal of Petrology, 26, 1-30. Ozawa, K. 1989. Stress-induced Al-Cr zoning of spinel in deformed peridotites. Nature, 338, 141-144.

148

CHRONOLOGY OF MAGMATISM AND DEFORMATION IN THE LINCOLN COMPLEX; CONSTRAINTS ON THE KIMBAN OROGENY Bruce F. Schaefer. Martin Hand, Betina Bendall, John Foden and Mike Sandiford Deptartment of Geology and Geophysics, University ofAdelaide The Kimban Orogen is characterised by an extended tectonothermal evolution punctuated by episodes of voluminous felsic magmatism (the Donington Suite) and multiple dyke generations (the Tournefort Dyke swarm) over the period of -1850 - -1700 Ma (Drexel et al., 1993 and refs. therein). It is the aim of the current study to discern the temporal and spatial components of magmatism, deformation and metamorphism during the Kimban Orogeny in order to provide a framework for an evolving geodynamic picture. The Donington Granitoid Suite (Mortimer et al, 1988), comprises voluminous high temperature felsic magmatism at 1846 ± 6 Ma (Pb-Pb zircon, this study, cf 1843 ± 2 Ma, Mortimer et al, 1988), and is intruded by multiple generations of mafic dykes of the Tournefort Dyke Swarm. These dykes preserve several features important for understanding the geodynamic evolution of the Kimban Orogeny: • Geochemical and structural evidence for multiple dyke emplacement events between crystallisation of the Donington suite and the cessation of the Kimban Orogeny at -1700 Ma (see Drexel et al, 1993). The oldest mafic dykes apparently predate Donington Suite magmatism, and observations in the more prevalent Tournefort Dyke Swarm preserve crosscutting relationships and multiple dyke array orientations. • Strongly heterogeneous deformation and high grade metamorphism resulted in recrystallisation of dykes in the Tournefort Dyke Swarm. This takes the form of either static recrystallisation with partial preservation of igneous textures or of dynamic recrystallisation and amphibolitisation. High grade metamorphic mineral assemblages in the Tournefort Dykes have yielded SmNd ages of 1716 ± 14 Ma and 1730 ± 20 Ma, constraining not only the timing of a high grade Kimban metamorphic event, but also providing a minimum emplacement age for the dykes. The Tournefort Dyke swarm indicate a number of dyke emplacement events prior to peak metamorphism during the Kimban, and an evolving extensional stress field after Donington Suite magmatism in the period between -1850 - 1714 Ma is necessary to emplace the Tournefort Dyke swarm. Peak Kimban regional metamorphism is characterised by ductile deformation along discrete shear zones associated with the Kimban Orogeny which are seperated by regions of heterogeneously distributed lower strain, and appears to have ceased by -1700 Ma. References: Drexel, J.F., Preiss, W.V. and Parker, A.J. (eds), 1993. The Geology of South Australia. Vol. 1. The Precambrian. South Aust. Geol. Surv. Bull 54. Mortimer, G.E., Cooper, J.A. and Oliver, R.L., 1988. The geochemical evolution of Proterozoic granitoids near Port Lincoln in the Gawler Orogenic Domain of South Australia. Prec. Res. 40/41:387406.

149

THRUSTING OF THE NORTHERN MOLONG VOLCANIC BELT, WELLINGTON, NEW SOUTH WALES M.M. Scott Geological Survey of New South Wales, P.O. Box 53, Orange, NSW 2800, Australia The Wellington 1:100 000 Geological Sheet has recently been mapped by the Geological Survey of NSW and Australian Geological Survey Organisation as part of the National Geoscience Mapping Accord. Seven major N-S trending and west-dipping thrusts interleave Ordovician to Devonian sequences across a 20km wide strip of the northeastern Wellington 1:100 000 Sheet. The Ordovician to Devonian sequences commonly dip and young to the W-WNW, and the stratigraphy is repeated by thrusts dipping parallel to bedding. Two narrow (<4km wide) belts of Early-Late Ordovician mafic volcaniclastics of the Cargo Volcanics and Oakdale Formation, are bound by the Black Rock, Neurea, Bakers Swamp and Narragal Thrusts, and are interleaved between Silurian and Devonian sequences. The Black Rock, Narragal and Bakers Swamp Thrusts have ramped along west-dipping bedding planes of the Narragal Limestone, Camelford Limestone, Barnby Hills Shale, Cuga Burga Volcanics and Garra Formation. The Narragal Thrust has displaced Oakdale Formation eastwards over younger Silurian and Devonian sequences. The southern Mount Arthur Syncline is tight, trends NNE, the axial plane dips WNW, cleavage is poorly developed, and and the eastern limb of the Catombal Group is upright and westdipping. The Mount Arthur Syncline is cut to the south by the west-dipping Curra Creek Thrust. The northern Mount Arthur Syncline is isoclinal, trends NNW, the axial plane dips ENE, cleavage is well developed and dips ENE, and the eastern limb of the Catombal Group is overturned and ENE dipping (Powell et al., 1976). The change in geometry of the northern Mount Arthur Syncline is due to the fold developing in the footwall of the ENE-dipping Macquarie Thrust, and at the northern termination of the Curra Creek Thrust. The Macquarie Thrust has Oakdale Formation in the hanging wall, and cuts the Black Rock, Neurea, Bakers Swamp and Narragal Thrusts east of the Mount Arthur Syncline. The Late Devonian Catombal Group overlies the Early Devonian Garra Formation with a low angular unconformity in the Mt Arthur and Catombal Synclines (Powell & Edgecombe, 1978). The basal eastern contact of the main belt of Catombal Group has angular discordant boundaries with Early Devonian Cuga Burga Volcanics, Siluro-Devonian Camelford Limestone, Late Ordovician Oakdale Formation and Early Ordovician Cargo Volcanics. If this basal contact of the Catombal Group is an unconformity, and not a bedding-parallel fault, significant thrusting and folding occurred to expose Early Ordovician Cargo Volcanics in the Middle Devonian. References Powell C.Mc.A. & Edgecombe, D.R., 1978. Mid-Devonian movements in the northeastern Lachlan Fold Belt. Journal of the Geological Society of Australia, 25, 165-184. Powell C.Mc.A., Edgecombe, D.R., Henry, N.M. & Jones, J.G., 1976. Timing of regional deformation of the Hill End Trough: A reassessment. Journal of the Geological Society of Australia, 23, 407-421.

150

GEOMETRIC EVOLUTION OF THE BUCKSKIN-RAWHIDE METAMORPHIC CORE COMPLEX, SW USA. Robert J. Scott1*2 1

VIEPS, Department of Earth Sciences, Monash University, Clayton, VIC 3168, Australia Now at RSES, Australian National University, Acton, ACT 2601, Australia

Many misconceptions about the development of metamorphic core complexes (MCCs) stem from a failure to appreciate the evolving geometry of the detachment systems that bound them. Typically models for MCCs depict a single master detachment fault that is present at the onset of extension and forms a permanent boundary between the upper and lower plates (Wernicke, 1981, Nature, 291, 645-648). This model is not correct, and has served to confuse issues related the age and original orientation of the detachment faults that now bound the MCCs. A more complete understanding of the geometric development of MCCs requires careful integration of geological data from both the upper and lower plates of the detachment system, and is only possible where both elements are exposed and well preserved. The Buckskin-Rawhide metamorphic core complex (BRMCC) is one of several MCCs exposed in the southern half of the -300 km long Colorado River extensional corridor (CREC), a belt of extreme Oligo-Miocene extension (locally >400%) between southern Nevada, and west-central Arizona (Howard & John, 1987, Geol. Soc. Sp. Publ., 28, 299-311). The detachment system that bounds the BRMCC is rooted beneath the Transition Zone along the SW margin Colorado Plateau, and accommodated >60 km top-to-the-NE displacement (Howard & John, ibid.; Spencer & Reynolds, 1991, Tectonics, 10, 1204-1221). A three or four-stage process for the formation of the BRMCC is suggested. The first stage began at -30 Ma and involved the emplacement of voluminous intrusions into the middle crust in the region that now forms the axis of the CREC. Numerous intensely deformed synextensional intrusions, some as young as 21 Ma, are exposed in the denuded lower plate of the BRMCC (Bryant & Wooden, 1989, Az. Geol. Surv. Bull., 198, 47-50). Strain gradients in and around the intrusions suggest they may have facilitated the development of crustal-scale shear zones. The second stage in the formation of the BRMCC overlapped with the first, but marked the first expression of extensional deformation at the earth's surface. It was characterised by (1) development of numerous moderately- to steeply-NE-dipping normal faults, bounding asymmetric half-grabens up to several kilometres across; (2) widespread tilting of the half grabens and synextensional strata deposited within them; and (3) the onset of moderate to rapid rates of cooling (locally >50°C/MYr) in mylonitic lower plate rocks now exposed at the southwestern end of the BRMCC. The second stage of extension lasted from -27 to <16 Ma, and represented >60% of the extensional history of the BRMCC. Up to 2000 m of sedimentary and polymodal volcanic rocks were deposited within the actively rotating half-grabens during this time (Spencer & Reynolds, 1989, Az. Geol. Surv. Bull., 198, 103167, Yarnold, 1994, Sed. Geol., 89, 43-63). Abundant coarse clastic sedimentary rocks, including debris flow and rock-avalanche deposits are characteristic of this interval. The third stage of extension was heralded by a dramatic acceleration in the cooling rates of lower plate rocks exposed at the southwestern end of the BRMCC. Rapid cooling (~300°C/MYr) between 17 and 16 Ma is inferred to reflect rapid slip on either an existing upper crustal fault or a newly formed splay of the developing detachment system. This event is seemingly reflected in the geological development of the upper plate by the widespread development of an angular unconformity at about this time (Fig. 1). Shortly thereafter an enormous rock avalanche deposit (Artillery megabreccia), sourced from a major new escarpment in the southwest of the BRMCC, was dispersed northeastwards across the land surface. Correlative megabreccia deposits up to 100 m thick occur across the Buckskin, Rawhide and Artillery mountains for a distance of over 30 km in the extension direction (Spencer & Reynolds, 1989, Az. Geol. Surv. Bull., 198, 103-167), although the original extent of the deposit in this direction was probably closer to 10 km. 151

Upper plate

Phase IV (13->9 Ma) Phase m (<16-13 Ma) Phase II (-27 - <16 Ma)

Rawhide fault Lower plate

Artillery megabreccia

microbreccia chloritic breccia

Figure 1 Structural relations in the Buckskin-Rawhide metamorphic core complex (schematic). The third stage of extension (<16-13 Ma) was marked by a significant, but gradual, reduction in the energy of the depositional systems, reflecting either widening of the sedimentary basins or an overall reduction in topographic relief. Mylonitic rocks in the lower plate of the detachment system were first exposed at the surface during this interval. The mode of upper crustal deformation appears to have changed dramatically after -16 Ma. While synextensional strata deposited during this interval are intensely faulted, significant tilting of upper plate strata was much more localised than it had been during the earlier period of extension. Gentlydipping strata in the Rawhide Mountains, deposited between 16 and 13 Ma are cut by the detachment fault (Rawhide fault) at a low-angle (Fig. 1). All of the faults in the upper plate that developed during or prior to the third stage of BRMCC formation are abruptly truncated where they intersect the detachment fault (Fig. 1). These relations indicate that the Rawhide fault formed at a late stage (<13 Ma), at a very low-angle and probably within a few hundred metres of the earth's surface. The fourth and final stage in the development of the BRMCC began at -13 Ma and continued until detachment faulting ceased at 10-9 Ma. Crudely stratified conglomerates containing up to 30-40% clasts derived from the lower plate (Spencer & Reynolds, 1989, Az. Geol. Surv. Bull., 198, 103-167) were deposited in the Rawhide Mountains after 13 Ma. The deposits are faulted and locally tilted by up to 90° where they abut the Rawhide fault. However, in contrast to the older upper plate faults, listric faults that cut the post-13 Ma deposits are not truncated, but merge with the detachment fault. This suggests that only the youngest syn-extensional faults in the upper plate were directly related to slip upon the Rawhide fault. As for the detachment splay inferred to have formed at -16 Ma, the formation of the Rawhide fault was marked by the development of a widespread angular unconformity between pre- and post-13 Ma deposits, and an abrupt increase in the energy of the sedimentary environment. The cataclasites below the Rawhide fault record a long history of fault movement, probably over a range of PT-conditions. This suggests that the younger detachment splays propagated from a master detachment zone and cut progressively into the upper plate of the system. Only locally does the Rawhide fault cut downwards across older fault breccias in the lower plate (Fig. 1). However the extreme variation in the thickness of the oldest cataclasites in the lower plate (i.e. >300 to <20 m) and their significant obliquity (>30°) to the underlying mylonites (Fig. 1) indicates downward cutting splays may have been significant earlier in the development of the MCC. Structural, stratigraphic and thermochronological relations in the BRMCC suggest that large displacement low-angle normal faults (detachment faults) formed progressively once extension was under way, but may not have been present during the earliest stages of the deformation. The marked change in deformational style in the upper plate of the BRMCC after -16 Ma suggests that a fundamental transition from distributed extension on steeply dipping, deeply-rooted rotational normal faults, to rapid localised extension on large displacement, low-angle detachment faults may have actually occurred at a relatively late-stage during the development of the core complex.

152

THE ORIGIN OF CORDILLERAN METAMORPHIC CORE COMPLEXES R.J. Scott1*2 and G.S. Lister1 1

VIEPS, Department of Earth Sciences, Monash University, Clayton, VIC 3168, Australia Now at RSES, Australian National University, Acton, ACT 2601, Australia

Debate over the origins of metamorphic core complexes (MCCs) has largely concerned the relative importance of three factors: (1) reduction in "far field11 compressive stresses reflecting plate interaction, (2) gravitational collapse of over thickened crust and (3) pre- to synextensional magmatism. Although there is widespread acceptance that all of these factors may have played some role in localising extensional deformation, there is no clear consensus on either the overall mode of deformation during MCC formation (i.e. crustal-scale pure or simple shear) or the ultimate cause(s) for crustal extension. Here we review a range of geological and geophysical data relevant to the development of MCCs in the U.S. Cordillera and in particular the southwestern U.S., as it is the area with which we are most familiar. We conclude that the MCCs were formed during roll back of the then underlying Farallon plate, which triggered melting in the upper mantle and ultimately extension of the overlying crust. Subduction of the Farallon plate began in the early Mesozoic and essentially continued along the entire western margin of the USA until at least the late Oligocene, when oblique collision with the Pacific-Farallon spreading ridge (East Pacific Rise) resulted in the progressive development of a transform system along southwestern margin of North America (Coney, 1978, Geol. Soc. Am. Mem., 152, 33-50). Development of the transform system probably began adjacent to Baja California between 20 and 30 Ma and has gradually lengthened (propagating both north and southward) to form the modern day San Andreas fault system (Atwater, 1970, Geol. Soc. Am. Bull., 81, 3513-3536). Development of the transform system was broadly coeval with MCC formation in southern and west-central Arizona and some workers have argued that the two processes were closely related (e.g. Glazner and Bartley, 1984, Tectonics, 3, 385-396). However this does not explain the older (early Eocene-Oligocene) MCCs in the northwestern and central-western U.S., which were developed inboard of a compressional or transpressional plate margin (Wernicke et al., 1987, Geol. Soc. Sp. Publ., 28, 203-221). The apparent variation in tectonic setting during the development of the Cordilleran MCCs has caused some workers (e.g. Wernicke et al., ibid.) to doubt how successfully plate boundary interactions could be used to explain the development of specific features on land. Although the MCCs are all developed in areas of significant prior crustal thickening (Wernicke et al., 1987, Geol. Soc. Sp. Publ., 28, 203-221), gravitational collapse of the crust does not appear to have been the driving force for extension. Palaeogeographic (Elston & Young, 1991, JGR, 96, 12389-12406; Yarnold, 1994, Sed. Geol., 89, 43-63), geobarometric (Anderson et al., 1988, Geology, 16, 366-369) and thermochronologic studies (Foster et al., 1990, JGR, 95, 20005-20024; Knapp & Heizler, 1990, JGR, 95, 20049-20073) in west-central Arizona and adjacent California suggest erosion induced uplift ± gravitational collapse of the previously thickened crust had slowed or ceased up to 10 MYr before development of the MCCs began. By the Late Oligocene the thickness of the crust in western Arizona is interpreted to have been no greater than that of the adjacent Colorado Plateau. Thus gravitational collapse could only have contributed to extensional deformation if the crust in western Arizona was significantly weaker than that in surrounding areas, or the underlying mantle was significantly more buoyant. Although both alternatives are plausible, and Mesozoic structures in some MCCs appear to have been instrumental in localising pre- to synextensional intrusions and/or extensional deformation, the abrupt onset of rapid extension suggests an 'external trigger1 led to the development of the MCCs. During the Tertiary there was a major resurgence in arc-like magmatism in the western U.S., following an interval of inferred low-angle subduction during which the locus of magmatism migrated up to 1500 km east of the plate margin or ceased altogether (Burchfiel & Davis, 1975, Am. J. Sci., 275-A, 363-395; Coney & Reynolds, 1977, Nature, 270, 403-406). The cumulative volume of predominantly felsic, calc-alkaline magma erupted in the western 153

Cordillera after 40 Ma may have exceeded 10 km (Elston & Bornhorst, 1979, Rio Grande rift; tectonics and magmatism, AGU, 416-438). Throughout the Cordillera the resurgence of magmatism coincided with or slightly predated the development of the MCCs (Armstrong & Ward, 1991, JGR, 96, 13201-13224). However while volcanism occurred over vast areas, extreme crustal extension was generally restricted to a belt <200 km wide, located within the largest magmatic fields. Structural relations in the Buckskin-Rawhide metamorphic core complex (BRMCC), westcentral Arizona, suggest mylonitic deformation was preferentially, although not exclusively, localised in and around numerous sheet-like synextensional intrusions in the lower plate. This suggests the ambient temperatures in the surrounding rocks may have been too low to facilitate ductile deformation (i.e. T<350-300°C for quartz-rich rocks). If this was the case, the BRMCC and perhaps other MCCs may have been quite shallowly rooted in the earth's crust. However thermochronological data from non-mylonitic rocks (preserved between shear zones) at the SW-end of the BRMCC indicates that even the shallowest exposed levels in the lower plate were uniformly above 400°C until at least the early Miocene (well after the onset of extension). The high 'ambient' temperature of the lower plate rocks could be due to the addition of large volumes of magma to the crust, prior to or during the initial stages of MCC formation. However existing thermochronological data provides no conclusive evidence to support this hypothesis. None the less, the facts that (1) ductile deformation was localised in and around synextensional intrusions, (2) pre- to synextensional intrusions account for -30% of the exposed lower plate, and (3) the oldest post-Cretaceous intrusions (e.g. Richard et al., 1990, JGR, 95, 19973-19987) predate the earliest surface expression of extensional deformation by 2-3 MYr suggests that plutonism played a fundamental role in the development of the BRMCC. The resurgence of arc-like magmatism and the development of an intraplate stress field favouring extension that was both coeval with and parallel to the direction of plate convergence, are interpreted to reflect roll back of the then underlying Farallon plate. Studies of syn- to postextensional magmatism in the SW U.S. indicate the earliest magmas were sourced from the subcontinental lithospheric mantle (SCLM), but gave way to asthenosphere-derived magmas following MCC formation (Daley & DePaolo, 1992, Geology, 20, 104-108; Bradshaw et al., Earth Plan. Sci. Lett., 116, 45-62). The distribution of late-stage asthenosphere-derived magmas in SW USA mimics regions of significant Oligo-Miocene crustal extension (some post-dating MCC formation) (Livaccari & Perry, 1993, Geology, 21, 719-722). Although the emplacement of mantle-derived magmas in the crust may have helped trigger MCC formation, the timing of the change from predominantly lithosphere- to predominantly asthenospherederived magmas suggests that the upper mantle subsequently responded passively to thinning of the overlying crust (Bradshaw et al., Earth Plan. Sci. Lett., 116, 45-62). Magmatic patterns in the SW U.S. are interpreted to reflect melt depletion, thinning and ultimate delamination of the gravitationally unstable remnants of the SCLM, during and following roll back of the Farallon plate. Dehydration of the Farallon plate during low-angle subduction is likely to have resulted in widespread metasomatism of the overlying SCLM beneath western North America. While this would have dramatically reduced mantle solidus temperatures in the affected areas, it did not immediately lead to melting due to the low temperatures of the SCLM and underlying slab. However, progressive replacement of the oceanic lithosphere by hot asthenosphere during slab roll back is interpreted to have triggered partial melting in the upper mantle, particularly within the volatile-enriched regions of the SCLM. Thermal weakening of the crust due to the emplacement of mantle-derived magmas and the favourable stress regime induced by slab roll back could have facilitated the development of crustal scale shear zones and the rapid onset of extensional deformation, particularly if the intrusions were localised along preexisting crustal weaknesses. Thinning of the crust ± SCLM would have led to decompression and further melting in the upper mantle. The feed-back between melting in the upper mantle and crustal thinning explains why extreme extensional deformation was localised within one or two relatively narrow belts rather than being distributed across the whole region that was affected by the resurgence of arc-magmatism. 6

154

STRAIN-RELATED TEXTURAL DEVELOPMENT IN A GRANULITE-FACIES SHEAR ZONE - THE RAUER GROUP, EAST ANTARCTICA I.P. Sims & C.J.L. Wilson School of Earth Sciences, University of Melbourne, Parkville, Victoria 3052 Australia Granulite fades metamorphic assemblages and subsequent reaction textures are normally considered to represent the peak and retrograde components of P-T-t paths. However, the affect of strain on the development of equilibrated assemblages and textures is rarely considered, even though there is much evidence to suggest that re-equilibration is encouraged by intracrystalline deformation. Hence, in mid- to lower-crustal rocks, where creep processes dominate, the equilibrated mineral assemblages should more closely reflect the strain history of the rock rather than the peak thermal or baric history. In the Rauer Group, non-coaxial deformation under granulite-facies conditions was partitioned into domains that record high and low strains. High-strain domains are characterised by the development of a gneissic foliation that overprints pre-existing foliations, while low strain domains are characterised by reorientation of the existing foliations. The foliation relationships are complicated, however, by the repartitioning of strains during progressive deformation. Nevertheless, bulk re-equilibration and development of stable mineral assemblages dominates in domains of high strain while very local grain scale re-equilibration and development of discrete (decompression) reaction textures dominates in low-strain domains. Hence, mineral assemblage paragenesis is strongly dependant on the local strain-history as well as the bulkrock composition. This is accentuated by the flow stress of the different rocktypes under granulite-facies conditions (Fig. 1). For example, discrete reaction textures are most common in the mafic rock-types, which have high flow-stress, while bulk re-equilibration is dominant in more felsic rock-types with low flowstress. Textures in the Theologically strong, mafic gneiss, are dominated by peak metamorphic assemblages. For example, coarse garnet and clinopyroxene atolls overprint a gneissic foliation defined by plagioclase, hornblende, orthopyroxene and clinopyroxene. P-T estimates on the peak assemblage give conditions of 10.8 ± 1.8kb & 972 ± 125°C. Discrete decompression reaction textures that are developed between garnet and clinopyroxene, involve symplectic orthopyroxene and anorthitic plagioclase. P-T estimates on the decompression assemblage give conditions of 6.9 ± 2.4kb & 827 ± 127°C. Microprobe analysis suggests that the development of decompression textures are due to grain-scale chemical disequilibrium, with the remnant reactants zoned with respect to the dominant components of the products (which are generally unzoned, or are zoned with respect to components that must have existed in a melt phase e.g. albite component in decompression plagioclase) Textures in pelitic gneiss are dominated by shape fabrics that are aligned in the dominant foliation and lineation. These mineral textures generally indicate that 155

the assemblages developed concurrently with the deformation. For example cordierite sigma-tails on garnet aligned with the regional elongation lineation suggest exhumation of the terrain was accompanied by the bulk non-coaxial flow. Likewise, the development of recrystallised cordierite and then aligned sapphirine both after an initial lineated assemblage of orthopyroxene, sillimanite and quartz is indicative of continued deformation with decompression. Compositional data suggests that the peak metapelitic assemblages developed at 10 - 12kb and approximately 1050°C (Harley & Fitzsimons, 1991). Microprobe analysis suggests, however, that in most metapelitic gneiss (particularly Fe-rich bulk-compositions), the mineral assemblages readily equilibrated within the progressively developed high-strain zones to lower pressure conditions. Granoblastic mineral assemblages in very late formed high-strain zones record conditions of 5.6 ± 1.8kb and 809 ± 130°C. It is difficult to resolve the diverse P-T estimations derived from equilibrated assemblages without considering the effects of the strain-history. Remnant extreme P-T domains reflect early high-strain histories while many of the rocks record later high-strain equilibrated assemblages. Hence, P-T calculations on a range of equilibrated assemblages that formed in progressively developed highstrain domains actually defines the metamorphic path of the terrain, while local reaction textures that formed concurrently in the low-strain domains indicate increments along that path. References Harley, S.L. & Fitzsimons, I.C.W. 1991. Pressure-temperature evolution of metapelitic granulites in a polymetamorphic terrane: the Rauer Group, East Antarctica. Journal of Metamorphic Geology, 9, 231-243.

156

UPLIFT VERSUS EXHUMATION. A KINEMATIC MODEL AND IMPLICATIONS FOR THE GEOMETRICAL EVOLUTION OF THE TAUERN WINDOW, EASTERN ALPS K. Stuwe Department of Earth Science, Monash University, Clayton, Vic 3168, Australia A simple one-dimensional model is used to clarify the vertical kinematics of rocks during denudation that occurs simultaneously with convergent deformation. It is shown that rocks may move upwards or downwards in the crust, depending on the relative rates of denudation and thickening and depending on the initial depth of rocks in the crust but not necessarily depending on mean surface elevation. Uplift and exhumation may follow different patterns in time, with this pattern being characteristic of the erosion process. In particular, there may be a hiatus between surface elevation increase (i.e. uplift) and exhumation. This is because initial convergent deformation will produce uplift but no exhumation as it displaces all rocks to larger distances from the surface (Fig. 1). In isostatic equilibrium, only rocks at levels shallower than about 3500 m from the surface can dome upwards, while all rocks at deeper levels than this will move downwards . Exhumation during thickening can only occur if rapid denudation accompanies the thickening process. With increasing uplift, denudation rates may increase so that exhumation of rocks from increasingly larger depths may commence only once a significant part of the uplift history has been completed. These conclusions emphasize the important differences between uplift and exhumation which are often not discriminated in the literature. It is shown here on hand of some examples how this non-discrimination may lead to misinterpretations. For example, a hiatus between uplift and exhumation may imply that many metamorphic domes represent inverted synforms that were produced by the process of deformation and inverted by the process of exhumation. Such a hiatus is difficult to document because there is no simple evidence that can be used to document uplift. However, some data indicate such a time gap between uplift and exhumation in the Tertiary evolution of the Tauern Window of the Eastern Alps. There it is suggested that care must be taken in using the dome shape of the structure for inferring shortening amounts in the Tertiary evolution of the Eastern Alps. 30

Fig.l: Uplift history (top diagram) and exhumation history (bottom diagram) calculated for a one dimensional column with a simple kinematic model in which convergent deformation is accompanied by erosion at the surface. Strain and erosion rates are chosen so that rocks from 30km depth reach the surface within about 20Ma. These are representative values for the evolution of the eastern Alps. It may be seen that surface uplift commences at the time of onset of thickening, but that exhumation of rocks commences not until about lOmy later

km A<*"p««

157

ROCK TRAJECTORIES (EXHUMATION HISTORY)

LONG VERSUS SHORT TERM DENUDATION RATES FOR THE EASTERN ALPS. CONSTRAINTS ON THE UPLIFT HISTORY ? K. Stuwe 1 & D. Fabel 2 1

Department of Earth Science, Monash University; Clayton, Vic 3168, Australia School of Earth Science, LaTrobe University; Bundoora, Vic 3083, Australia

Denudation rate is a good measure of relief (Summerfield & Hutton, 1995) and - for a narrow mountain range like the eastern Alps - probably elevation. Thus, detailed knowledge of the denudation rates through time may be a useful data set to derive information on the changes of mean surface elevation with time (i.e. the uplift history). We constrain such details of the denudation history for the eastern Alps by comparing four different data sets that bear information on the subject. These are: (a) The present day denudation rates from stream sediment transport (e.g. Schroder and Theune, 1984); (b) The sedimentological record in the Molasse basins; (c) fission track thermochronology and (d) isotope geochronology in association with geobarometry. In principle, these data sets may be used to estimate the denudation rates for three different periods in geological time. These are: 1.) data set (d) gives the average denudation rate between the time of peak metamorphism and the present. 2.) The difference between data set (d) and (c) gives an estimate of the early denudation rates following metamorphism. 3.) Data set (a) gives an estimate of the current denudation rate. In theory, very detailed knowledge of data set (b) provides an independent estimate on the denudation rate during all periods (e.g. England, 1981). While we are well aware of the enormous uncertainties that are associated with some of these data, they form some of the very few measurable parameters that may be used to obtain information on the uplift history (e.g. Forest et al., 1995). In particular for the elevation of surfaces long eroded into the Molasse basins, there is little data that are otherwise available. It is also emphasizes that each of these data sets alone does not bear any information on surface uplift and it is only comparison between them that may bear such information! Comparison of the different data sets shows that the denudation rates in the eastern Alps have decreased continuously through time since the Cretaceous. This is in contrast to landform studies which indicate that surface elevation has increased, at least in the last few million year (Sakaguchy, 1973). We are currently working on extending the data sets and estimating the errors involved. However, in the present study we discuss a number of possible interpretations of this preliminary result.

References England, P.C., 1981. Metamorphic pressure estimates and sediment volumes from the Alpine orogeny: an independent control on geobarometers?. Earth and Planetary Science Letters, 44, 273-294. Forest C.E., Molnar, P. & Emanuel K.A., 1995. Paleoaltimetry from energy conservation principles F Nature, 347, 347-350. Summerfield, M.A. & Hutton, N.J., 1994. Natural controls of fluvial denudation rates in major world drainage basins. Journal of Geophysical Research, 99,13871-13883. Sakaguchy, Y., 1973. Uber die geomorphologische Entwicklung der Ostalpen. Zeitschrift fur Geomorphologie, Suppl. 718, 144-155. Schroder, W. & Theune C., 1984. Feststoffabtrag und Stauraumverlandung in Mitteleuropa. Wasserwirtschaft, 74, 374-379.

158

ENERGETICS OF GRAIN BOUNDARY MIGRATION OF QUARTZ Satoru Suzuki1, Tomoyuki Morikawa2, Yoko Nakayama1 and Toshiaki Masuda1 1

Institute of Geosciences, Shizuoka University, Shizuoka, 422, Japan

Geological Institute, University of Tokyo, Hongo, 113, Japan

The relationship between velocity of grain boundary migration and driving force has been empirically derived in annealed quartz aggregates at high pressure and temperature. Three different quartzose rocks were used for the starting material: agate and two metacherts. Grain size of the agate is less than 1 micron whereas grain sizes of the metacherts are 0.2-2.5 mm and 1-6 mm, respectively. These rocks were cored and sliced into discs 10 mm in diameter and approximately 4 mm thick. These discs were put together so that the agates were in direct contact with two metacherts. The annealing experiments were conductedusing a solid confining medium apparatus at temperatures of 800° and 1000°C, confining pressure of 400 and 800 MPa and annealing time range from 6 x 101 to 3.6 x 105 seconds. Grain growth of agate occurred during the experiments. Mean grain size (D) at 800°C is expressed as a function of annealing time (t) as D-l.l t021 (1) Boundaries at contact surfaces of agate with two metacherts have migrated into agate at the expense of fine-grained quartz in agate, indicating clearly that the grain boundary migration occurred at the contact surface. The migrated distance varies from place to place even in one sample. The maximum migrated distance was used as a representative migrated distance of the sample, which increased from 4 to 96 microns with increasing annealing time. The relationshipbetween M and t at 800°C is expressed as M=2.9t °-20 (2) Driving force for the grain boundary migration is presumably the grain boundary energy of quartz in the agate. Since grain boundary area (S) is related to mean grain size (D) of quartz as S=3/D from the stereological analysis, we can express the grain boundary energy as P=3yT> (3) where y is the specific surface energy of quartz. By differentiating equation (2) and eliminating t and D using equations (1) and (3), we obtained the velocity of grain boundary migration at 800°C as V=dM/dt=0.87(P/3y)38 (4) Assuming the activation energy of grain boundary migration as Q=ll kcal/mol, we finally obtained a general expression as V= 1.5x102P/3y)3-8 exp(-Q/RT)

(5)

where R is the gas constant and T is the absolute temperature.

159

PRELIMINARY INVESTIGATIONS ON THE TECTONIC EVOLUTION OF THE WESTERN GAWLER CRATON J.P. Teasdale Department of Geology and Geophysics, University of Adelaide, Adelaide, 5005 The Gawler Craton remains one of the least understood Precambrian terrains in Australia, mainly due to a paucity of continuous outcrop. In the western Gawler Craton, sporadic outcrops and drill hole intersections yield a wide variety of rock types which were previously difficult to correlate. The recent availability of high resolution aeromagnetic data produced under the South Australian Exploration Initiative (SAEI) by Mines and Energy, South Australia (MESA) provides an excellent tool with which regional syntheses and correlations can be made. This dataset provides a detailed insight into the geology of the region at scales up to 1:100,000. The western Gawler Craton is dissected by numerous crustal scale shear zones which often juxtapose domains of significantly different magnetic character. The shear zones have previously been described as intracratonic, separating either different levels or lateral equivalents of the same continental crust (e.g. Rankin et al., 1989; Drexel et al., 1993). Integrated geological and geophysical evidence increasingly supports the notion that the fault-bounded domains may have undergone significantly different tectonothermal histories. This implies that lateral tectonic accretion of unrelated crustal segments may have occurred; a hypothesis at odds with an intracratonic setting. Given the sporadic nature of outcrops and available drill core in the western Gawler Craton, one of the most effective ways of testing this hypothesis is to investigate the metamorphic evolution of the domains. Regional metamorphism generally operates at scales larger than those of the domains in the western Gawler Craton. If the structural continuity of the metamorphically useful lithologies can be proven using the aeromagnetic data, then it is reasonable to assume that the metamorphic effects are regional, not local. By defining pressure-temperature trajectories for these lithologies, it may be possible to evaluate the tectonothermal history of each domain. If the tectonothermal histories prove to be different, it may then be possible to define the accretionary history of the western Gawler Craton. One such shear zone is the Karari Fault Zone, which forms a discrete, northwest trending shear zone over 300km long. Investigations into mid-Proterozoic (Drexel et al., 1993) paragneisses at Ooldea, immediately to the west of the Karari Fault Zone have yielded spectacular, very high grade ferruginous granulites in various drill-cores. Diagnostic mineral assemblages and reaction textures include sapphirine-quartz, spinel-quartz, corundum-quartz, sillimanite-orthopyroxene-quartz±(sapphirine-cordierite), and the probable former existence of osumilite. These assemblages indicate peak metamorphic conditions in excess of 900°C and 8kb, followed by near-isobaric cooling. The Ooldea granulites lie in a set of regionally continuous, east-west trending magnetic highs (probably caused by banded iron formations) which are truncated to the east by the Karari Fault Zone. About 30km east of the Karari Fault Zone, supposedly Archaean garnet-cordierite-sillimanite paragneisses near Lake Ifould (King, 1951) preserve evidence for peak metamorphism at significantly lower temperatures and pressures, suggesting that the two domains juxtaposed by the Karari Fault Zone have undergone significantly different tectonothermal histories. References Drexel, J.F., Preiss, W.V. & Parker, A.J. (eds.), 1993, The Geology of South Australia: Volume 1, The Precambrian. State Print, Adelaide. King, D., 1951, Geology of the Pidinga Area. Trans. Roy. Soc. 5. Aust., 74 (1), 25-44. Rankin, L.R., Martin, A.R. & Parker, A.J., 1989, Early Proterozoic History of the Karari Fault Zone, northwest Gawler Craton, South Australia. Australian Journal of Earth Sciences, 36,123-133.

160

POST-DELAMERIAN THRUSTING ALONG THE PARALANA FAULT, NORTHERN FLINDERS RANGES, SOUTH AUSTRALIA Jonathan Teasdale, Bruce F. Schaefer & Mike Sandiford Department of Geology and Geophysics, University of Adelaide, Adelaide, 5005 The Mount Painter Inlier in the northern Flinders Ranges, South Australia, is an exhumed antiformal section of pre-Adelaidean crystalline basement. It is unconformably overlain by Adelaidean sediments to the north, south and west, and in part by younger sediments (Mesozoic-recent) to the east, except where the Paralana Fault forms the basement-cover contact. One notable feature of the Mount Painter Inlier is the presence of several outliers of younger material within the basement, including slivers of presumed Adelaidean tillite and carbonate immediately west of Mount Painter, and the breccia/melange forming Hidden Valley in the central part of the inlier. The Hidden Valley breccia/melange consists of a poorly sorted, weathered, pale green carbonaceous matrix of grits and clays, containing numerous exotic clasts and/or rafts ranging from a few millimetres to over 500m in size. The clasts include lower crustal maficultramafic granulites, high-grade calc-silicates, and numerous low-grade sedimentary blocks, as well as gneisses, mylonites and metsediments characteristic of the surrounding basement rocks. Importantly, large blocks of supposedly Ordovician granites and pegmatites (the British Empire Granite) are contained within the melange. The bounding faults of the western side of Hidden Valley outlier dip shallowly to the west at -40°, whereas a shallow (-20°), east-dipping fault defines the eastern exposure of the outlier. Numerous basement segments overlie the valley fill on sub-horizontal faults. We interpret the faults to be part of the Paralana Fault system. Kinematic indicators on the Paralana Fault show consistent evidence for reverse sense, west-over-east movement; a fact which cannot be reconciled with the regional-scale kinematics in the northern Flinders Ranges during the Delamerian. The geomorphology of the Mount Painter Inlier is dominated by a series of CretaceousTertiary peneplains forming a series of high plateaux (e.g. the Freeling Heights, Mawson and Paralana Plateaux). These peneplains imply Tertiary exhumation in the Mount Painter Inlier involved vertical movement of at least 1km. Evidence for this includes steeply dipping, deformed Tertiary conglomerates on the eastern margin of the inlier, along the Paralana Fault zone. These observations imply that the Mount Painter Inlier is, at least in part, allochthonous, with major reverse-sense thrusting along the Paralana Fault emplacing dense basement rocks over softer Adelaidean (and possibly younger) sediments causing gravitational overturn of the thrust fault. This hypothesis is supported by geophysical evidence, since the Mount Painter Inlier forms a distinct gravity low. The timing of movement and the origin of the exotic clasts in the Hidden Valley melange (which show no equivalents within the PalaeoMesoproterozoic basement) remains poorly constrained. However, the thrusting clearly postdates the emplacement of the late to post-Delamerian British Empire Granite. The exotic clasts may be derived from the Curnamona Craton against which the thrusting occurred. Acknowledgments Conversations with the late Dr Reg Sprigg, Jim Sears and Graham Teale were invaluable for developing our ideas here.

161

DEFORMATION AS A DRIVING FORCE FOR FLUID FLOW WITHIN AN ACTIVE OROGENIC BELT, SOUTHERN ALPS, NEW ZEALAND P. Upton Geology Department, University of Otago, Dunedin, New Zealand Fluid flow within the upper brittle crust can be described in terms of topographic and thermal driving forces (e.g. Donaldson, 1962; Koons & Craw, 1991). Such descriptions tend to assume the impermeable nature of the brittle-ductile transition and the lower crust as a basal boundary condition. However, geochemical data (e.g. Ferry, 1986), microstructural studies (Etheridge et a/., 1983; Cox & Etheridge 1989), and stable isotopes (e.g. Wickham & Taylor, 1990; McCaig et al, 1990; Upton et al., 1995) all suggest that in regions of active deformation the brittle-ductile transition and the lower crust are not as impermeable as has been previously assumed and that fluid, sometimes with a meteoric signature, is active in the deformation of these regions. The penetration of fluids with a near-surface signature through the brittle-ductile transition cannot occur in a system with a static permeability structure and only topographic and/or thermal driving forces (e.g. Koons & Craw, 1991). In a deforming material, volume changes lead to pore pressure gradients, transient permeabilities and porosities. These provide an additional deformation-induced driving force which enhances fluid flow in deforming rock. Volume change may result from poro-elasticity or dilatancy.

fluid penetration

Figure 1: The Southern Alps two-sided collisional orogen, modified from Koons and Craw (1991). Three regions of dilatant softening and fluid explosion are shown, the Alpine Fault, the Main Divide Fault Zone (Cox & Findlay 1995) and the Frontal Thrust (Koons 1990).

Numerical modelling in two-dimensions of fluid flow through a deforming, frictional, dilatant Mohr-Coulomb material has been carried out using a fully coupled mechanical and flow calculation. As a result of the dilatancy, the mechanical driving force for flow dominates. The dilatant flow law is strain-dependent and deformation creates regions of strain hardening where dilation allows fluid penetration and regions of strain softening from which fluid is expelled. The regions of softening correspond to large-scale fault zones such as the Alpine Fault and

162

Main Divide Fault Zone of the Australian/Pacific plate boundary (figure 1). The resulting deformation-induced flow is able to penetrate into the regions of low static permeability, a result of dilatancy and related transient increases in permeability. Stable isotope evidence for the incursion of meteoric water into ductile rocks of the Southern Alps has been found in biotites of the Alpine Fault mylonites and quartz veins of the Alpine schists and Alpine Fault mylonites (Upton et ai, 1995). Analyses of quartz veins formed beneath the brittle-ductile transition using a stable isotope laser probe showed variations of up to 2.5%o in individual veins. The quartz has precipitated from water calculated to have had an oxygen isotope signature ranging from 12 - 6%o over a temperature range of 600° - 400°C as the rock was transported to the surface as part of the the high uplift region of the orogen.

References

Cox, S.C., Findlay, R.H., 1995, The Main Divide Fault Zone and its role in formation of the Southern Alps, New Zealand, submitted to New Zealand Journal of Geology and Geophysics. Cox, S.F., Etheridge, M.A., 1989, Coupled grain-scale dilatancy and mass transfer during deformation at high fluid pressures: examples from Mount Lyell, Tasmania. Journal of Structural Geology, 11,147-162. Donaldson, I.G., 1962, Temperature gradients in the upper layers of the Earth's crust due to convective water flow. Journal of Geophysical Research, 67,3449-3459. Etheridge, M.A., Wall, V.J., Vernon, R.H., 1983, The role of the fluid phase during regional metamorphism and deformation. Journal of Metamorphic Geology, 1,205-225. Ferry, J.M., 1986, Infiltration of aqueous fluid and high fluidrrock ratios during greenschist facies metamorphism: A reply. Journal of Petrology, 27,695-714 Koons, P.O., 1990, The two-sided orogen: collision and erosion from the sand-box to the Southern Alps, New Zealand. Earth and Planetary Science Letters, 86,307-319. Koons, P.O., Craw, D., 1991, Evolution of fluid driving forces and composition within collisional orogens. Geophysical Research Letters, 18,679-682. McCaig, A.M., Wickham, S.M., Taylor, H.P., 1990, Deep fluid circulation in alpine shear zones, Pyrenees, France: field and oxygen isotope studies. Contributions to Mineralogy and Petrology, 106,41-60. Upton, P., Koons, P.O., Chamberlain, C.P., 1995, Penetration of deformation-driven meteoric water into ductile rocks; isotopic and model observationsfromthe Southern Alps, New Zealand, submitted to New Zealand Journal of Geology and Geophysics. Wickham, S.M., Taylor, HP., 1990, Hydrothermal systems associated with regional metamorphism and crustal anatexis: Example from the Pyrenees, France. In "The role of fluids in crustal processes." Washington D.C., National Academy Press.

163

DEFORMATION-INDUCED REACTION IN THE ALPINE FAULT MYLONITES, SOUTHERN ALPS, NEW ZEALAND P. Upton Geology Department, University of Otago, Dunedin, New Zealand The currently active continental collision between the Australia and Pacific plates provides a well-constrained area in which to study deformation-induced reaction. Convergence has been occurring on the plate boundary for >5 Ma (Sutherland, 1995) and approximately 20-25 kms of uplift over this period has brought garnet-oligoclase zone schist and mylonites to the surface. The Alpine schist is a belt of Mesozoic metagreywackes found to the east of the Alpine Fault, along the western side of the Southern Alps. The schists are dominated by quartzofeldspathic metasediments. The metamorphic grade of the schists decreases sharply from garnet-oligoclase at the Alpine Fault to pumpellyite-actinolite at the Main Divide, approximately 20 kms east of the Alpine Fault (Grindley, 1963; Grapes & Watanabe, 1992). Adjacent to the fault are the Alpine Fault mylonites, an approximately 1 km thick band of mylonites which are associated with the current phase of movement on the Alpine Fault (Sibson et al.y 1979; Prior, 1988). The mylonites grade into the schist over a distance of 1-2 kms (Sibson et al.y 1979; Prior, 1988; Berryman et al.9 1992). The metamorphic grade of the mylonites is garnet-oligoclase (Grapes & Watanabe, 1992). Electron microprobe analysis of the schist/mylonite transition shows that mineralogical variation within the biotite, muscovite and garnets of the mylonites is greater than that in the schists. All garnets in the Alpine schist are zoned with Fe and Mg contents increasing toward the rim while Mn and Ca contents decrease. In addition to this schist zoning pattern some mylonitic garnets exhibit an outer rim of Ca enrichment. Plagioclase varied from Ani5 to An4o over the oligoclase zone, and by up to 18% within individual rocks. The plagioclase in both schist and mylonite is complexly zoned, with some grains becoming more albitic toward the rim while other grains in the same rock are more anorthitic on the rim. The increase in mineralogical variation within the mylonites compared to the schists and the Ca-rich mylonitic garnet rims suggests reaction has occurred during mylonitisation. Deformation within the orogen has allowed the penetration of surface water beneath the brittle-ductile transition. The stable isotopic signature of this fluid is found in the mylonitic biotites (Upton, 1995) where recrystallisation has resulted in interaction between the biotites and the fluid. This fluid/biotite interaction has not occurred in the schist. As the distinctive mineralogy of the mylonites is not observed in the adjacent oligoclase zone schists, which have experienced very similar temperature and pressure conditions, it is suggested that the reaction is deformation-induced. Mylonite formation occurs under conditions of rapid strain, when the characteristic time for strain is small. The deformation-induced reaction within the mylonites suggests that the characteristic time for reaction is also small under these conditions and that the reaction rate is a function of strain rate. References Berryman, K.R., Beanland, S., Cooper, A.F., Cutten, H.N., Norris, R.J., Wood, P.R., 1992, The Alpine Fault, New Zealand, variation in the Quaternary structural style and geomorphic expression. Annales Tectonicae, Special IssueSupplement to volume 6,126-163. Grapes, R., Watanabe, T., 1992, Metamorphism and uplift of Alpine schist in the Franz Josef-Fox Glacier area of the Southern Alps, New Zealand. Journal of Metamorphic Geology, 10,171-180. Grindley, G.W., 1963, Structure of the alpine schists of South Westland, Southern Alps, New Zealand. New Zealand Journal of Geology and Geophysics, 6,872-930.

164

Koons, P.O., 1995, Metamorphism as chaotic mixing. AGU abstracts pp S290, Spring Meeting, Baltimore, Maryland. Prior, D„ 1988, Deformation processes in the Alpine Fault mylonites, South Island, New Zealand. Unpublished Ph.D. thesis, University of Leeds. Sibson, R.H, White, S.H., Atkinson, B.K., 1979, Fault rock distribution and structure within the Alpine Fault zone, a preliminary account. In Walcott, R.I., Cresswell, M.M., (eds) "The Origin of the Southern Alps" Royal Society of New Zealand Bulletin, 18,67-72. Sutherland, R.S, 1995, The Australia-Pacific boundary and Cenozoic plate motions in the southwest Pacific: Some constraints from GEOSAT data, in press Tectonics. Upton, P., Koons, P.O., Chamberlain, C.P., 1995, Penetration of deformation-driven meteoric water into ductile rocks; isotopic and model observations from the Southern Alps, New Zealand, submitted to New Zealand Journal of Geology and Geophysics.

165

LATERAL ESCAPE TECTONICS FOLLOWING ca.3300 Ma EXTENSIONAL CORE COMPLEX FORMATION IN THE EASTERN PART OF THE ARCHAEAN PILBARA CRATON, WESTERN AUSTRALIA M.J. Van Kranendonk & W.J. Collins Department of Geology, University ofNewcastle, Newcastle, NSW 2308, Australia Tectonic models to explain the origin of the domal granitoid batholiths and intervening greenstone belts of the Archaean Pilbara Craton have varied between an essentially passive mode of "vertical tectonics" involving the diapiric rise of granitoid batholiths into a little deformed cover of greenstones (cf. Hickman, 1984, Geol. Surv. W. Aust. Bull. 127; Collins, 1989, Precam. Res., 43,41-62), and an active mode of "horizontal tectonics" in which greenstones are envisaged to have been thrust over sialic crust during a period of Himalayan style collisional orogeny (Bickle et a/., 1985, Geol. Assoc. Can., Spec. Paper 28, 325-341). In the latter model, the authors suggest that large, overturned folds along the western margin of the Shaw batholith did not conform to a diapiric origin, and - on the basis of N-S re-folds - that diapiric re-activation of the batholith occurred subsequent to crustal thickening to produce the domes. However, a perusal of the Marble Bar map sheet (Hickman & Lipple, 1979, Geol. Surv. W. Aust., 1:250,000) and results from regional mapping show that the western Shaw folds are atypical of the other batholiths and sit in intriguing isolation, unresolved with respect to regional sets of structures. Therefore, neither a diapiric nor a Himalayan-style model can account structures in the Pilbara Craton. NOfoue HTM for all the we have mapped a linked system of pome Furthermore, sinistral transcurrent shear: zones, thrusts and folds that postdate formation of the granitoid domes, which supports a model of early extensional tectonics (dome formation at ca. 3300 Ma: Williams & Collins, 1989, Earth & Planet. Sci. Lett., 97, 41-53) followed by collisional orogeny and lateral escape tectonics.

During ca. 3300 Ma extension and crustal thinning, pre-existing sialic nuclei {ca. 3450 Ma syn-volcanic lopoliths ± >3450 Ma ?basement gneisses) were reactivated into domal structures (e.g. Mt. Edgar and W Shaw batholiths) following an analogy with the Recent core complexes of the D Entrecasteaux Islands of Papua New Guinea (Baldwin et al., 1993, Tectonics, 12,611-628). Doming was accompanied + + JJDIU + / H « by intrusion of voluminous granitic plutons derived from remelted sialic crust (Collins, 1993, Precam. -p- rovjESCue W (cover?) 60, 151-174), while contemporaneous felsic O (gg)zLAuAUoikH SYM-onoGErtic Res., FM. volcanics and coarse clastic sediments were qtiA*\rt>\V QockS 5 + (§)* deposited in synforms adjacent to the rising domes ST*£rli>/ mire fficitfa 'piCFcr/oA/ (e.g. Wyman & Kelly belts). Progressive expansion 8 ^ Tflfci/Or r/J^-r cove*. of the Mt. Edgar batholith resulted in evolution frfttaz-SLi? ffi^r "cd~ Bz rif. Ei g or Z2°S Corur/^cL'Vowns /+) from a core complex into an active diapir (Collins,

^I tr v S H A V

II11 r

166

1989). Extension led to rifting further west and the development of a back-arc sequence, as represented by the ca. 3260 Ma Strelly succession (Krapez, 1993, Precam. Res., 60, 1-45). Following extension and 3300 Ma dome formation, contraction resulted in collisional orogeny and lateral escape tectonics of the ca. 3260 Ma Strelly granite association. This event is represented by an interlocking, penetrative set of ductile shear zones, tight folds (including those of the western Shaw batholith margin), north-verging thrust faults and associated lateral ramps, and syn-orogenic sedimentary rocks that we interpret to have been caused by northward indentation of the Shaw batholith (? and attached rocks to the east) between ca. 3.2-2.95 Ga. Shaw batholith indentation set up an interlocking system of sinistral shear zones across a northsouth central corridor between the Yule and Carlindi batholiths in the west and the Shaw batholith and North Pole dome in the east as shown in Fig. 1. The eastern margin of the shear system formed along the Shaw batholith margin under amphibolite facies conditions, and formed a number of splays northward (Fig. 1). In between the splays lie the folds mapped by Bickle et al. {op. cit.\ which contain a component of dextral shear on steeply plunging lineations. The folds are bounded by, but not clearly truncated by, the amphibolite-facies sinistral shear splays, and are interpreted to have formed during the sinistral shear event. In this model, the area of the nappes first formed in a jog between shear splays across a sharp change in the curvature of the Shaw margin (Fig. 2a). With increasing shear strain, the jog became the short limb of a fold which, as a result of over-rotation relative to the main compressional vector, changed from sinistral to dextral displacement (Fig. 2b). The folds became tightly constricted between the two main shear splays such that material was translated up to the north and this caused crystallization of the relatively high pressure metamorphic mineral assemblages estimated by Bickle et al. {op. cit.). Northerly indentation of the Shaw batholith caused the westerly extrusion of the Strelly granite and associated volcano-sedimentary rocks, which encountered the concave, northeast-dipping margin of the Yule batholith and were ramped sharply up to the north to its present position where it is preserved in a 60° east-dipping cross-sectional view (Fig. 1). An analogy of this tilting process is that of a car being forced up on its two side wheels when it overrides a ramp with one side. Northward movement of the Strelly package was terminated as it encountered the Carlindi batholith. When it stopped, the rocks coming along behind - just like in a highway car crash crumpled up along its southern margin across a series of tight NE-SW striking folds (e.g. Soanseville syncline) and thrust faults. A fine example of the structures preserved along the lateral ramp lie within the most tightly curved portion of the Yule batholith margin (Fig. 1). Progressive southwards accretion exposed progressively deeper rocks to the south. The earlier formed folds of the western Shaw margin were cut by greenschist facies sinistral shears and associated tectonic melange, as seen by folded amphibolite-facies fabric elements. The western Shaw lineament, from the southern Shaw batholith to the Lalla Rookh formation, represents the break between transported material to the west, and unaffected rocks of the Warrawoona Group to the east and a tight fold hinge, but not neccessarily a terrane boundary. During collisional orogeny, uplift produced syn-orogenic coarse clastics (Fig. 1). For example, the Strelly succession (+ Gorge Ck. Gp.) is unconformably overlain by the Lalla Rookh Fm., which - although previously interpreted as a strike-slip basin - must have been deposited throughout the entire journey of the Strelly package, as its base is subparallel to the older rocks and was therefore initiated prior to tilting of the Strelly package.

167

S T R U C T U R A L A N A L Y S I S OF BASEMENT T E C T O N I T E S F R O M T H E A E G E A N M E T A M O R P H I C C O R E C O M P L E X OF IOS, C Y C L A D E S , G R E E C E Leon C.Vandenberg & Gordon S. Lister VIEPS Department of Earth Sciences, Monash University, Melbourne, 3168 Australia. The island of Ios in the Cyclades archipelago, Aegean Sea, Greece, is one of a group of domal exposures of high pressure metamorphic rocks, subject to Miocene continental extension. The lower-plate ('basement') is composed of a complex of granitic intrusions into Hercynian (Mo) amphibolite facies schists and gneisses, later deformed by the Alpine orogeny.

These rocks

were overthrust by the Cyclades blueschist nappe, and Alpine high pressure - low temperature metamorphism (Mi) has resulted in the formation of glaucophane, chloritoid, and jadeite (preserved only as rare inclusions). However, metamorphic parageneses in the 'basement' are dominated by the effect of a Miocene greenschist overprint (M2). Later tectonic exhumation during the Oligo-Miocene continental extension and plutonic event occurred as the result of the operation of a south-directed, crustal-scale shear zone (the D4 South Cyclades shear zone) located in the upper levels of the 'basement', and then later (brittle) low-angle normal faults. While there is no exposure of a detachment fault as such on Ios, there is a mid-crustal equivalent which has juxtaposed the Ios 'basement' against deformed and metamorphosed tectonites of the upper plate blueschist unit. Five generations of penetrative ductile structures (Di - D5) have been recognized in the Ios 'basement'. Di fabrics are rare and only preserved in the microlithon regions of the dominant 52 fabric or as inclusion trails within Mo garnet porphyroblasts. S2 is a gneissic fabric or a differentiated schistosity. Granitoids were deformed during D 2 to produce a strong S2 augen gneissic (mylonitic) foliation. L^ appears to have formed originally with a NW-SE trend. D 3 was of sufficient intensity to produce penetrative, approximately north-south oriented stretching lineations, and recumbent folds.

Foliated granitoids with S 2 fabric were (locally) further

deformed during D 4 to produce spectacular zones of S-C mylonites and north-south oriented L4 mineral stretching lineations. S 2 is normally subparallel to S 4 (in D 4 shear zones) except where it has been thrown into (l-100m scale) recumbent folds with S 3 axial plane. On Ios we can demonstrate that significant reorientation of older lineations has taken place during D 4 Miocene stretching of the Aegean continental crust. As a result, most folds now have the same orientation.

The present lineation pattern on Ios is therefore not entirely due to Eocene

subduction or to Eocene collision as suggested by earlier workers. The geomorphology of the island is dominated by the effects of a N-S to NNW-SSE elongate structural doming of the S 2 , 5 3 and S 4 fabrics during D 5 .

168

TECTONIC EVOLUTION OF THE HIGH PRESSURE SCHISTS AND GNEISSES OF THE AMOSS-MAYAVETCH METAMORPHIC CORE COMPLEX, NEW CALEDONIA

C.Venn and G.S Lister Department of Earth Sciences, Monash University, Clayton, Victoria The tectonic origin of the high pressure metamorphics in northeastern New Caledonia is uncertain. Previous workers have not recognised young faults and consequently the significance of these fundamental tectonic features have been misinterpreted. Tectonic models have thus been based on the concept of a thrust system across the schist belt and these supposedly older tectonic features have strongly influenced the literature since work began in New Caledonia in the 1860's. The confusion surrounding Tertiary plate movements in the Southwest Pacific appears to stem from a misunderstanding of the structural evolution and relative timing fabric formation and metamorphism. This study aims to correlate fabrics, microstructures and metamorphic events observed in the high grade zone of the high pressure schist belt, in an attempt to provide the basis for new models for the structural evolution of New Caledonia. The correlation between microscopic and mesoscopic information allows the timing of deformation and metamorphism to be established. The link between mesostructure and large scale structure (i.e., macrostructure) can be used to interpret macroscale relations across the remainder of the island. New tectonic models for the origin of the southwest Pacific will thus result. Detailed fabric and microstructural studies have been determined across selected traverses within the high pressure schist belt and consistent fabric and microstructural relations from all key localities have been determined. The relative timing of deformation and metamorphism has been ascertained and metamorphic conditions have been estimated for some examples. Collisional orogenesis occurred early during the deformation history. High pressure low temperature metamorphism took place prior to or synchronous with the development of the first penetrative foliation across the schist belt, Si. Peak metamorphism occurred soon after, involving growth of omphacite and garnet at pressures and temperatures on the order of 15 kbar (14.9±5.9) and 542°C (±31°C) respectively. Pervasive development of a differentiated S2 fabric also developed and this fabric can be identified in all samples. Ongoing crustal shortening (during late D2 or early D3) is suspected to have resulted in the steepening of the S2 fabric and pervasive recumbent folding. In the lower grade regions of the high pressure schist belt S2 remained a shallow dipping fabric. Thrusting during this time may have juxtaposed eclogites in a serpentinite sheet derived from the oceanic lithosphere, from hot, deep tectonic levels onto colder rocks from shallower structural levels. The serpentinite and preserved eclogite assemblages are believed to be the upper plate during D3. Eclogitic rocks in the footwall of the thrusts were not cooled as quickly and hence were retrogressed to blueschist facies assemblages. The period of recumbent folding of the steep S2 fabric took place during D3 followed by a third metamorphic event. M3 is characterised by static recrystallisation of quartz into foam microstructures and decussate grain growth of white mica. D3b continued with the formation of major ductile shear zones, the development of a pervasive S3 crenulation cleavage and dynamic recrystallisation of the foam textured grains. The progressive nature of the deformation during D3, suggests that the recrystallisation event M3 was a short lived thermal pulse. Pressure-temperature calculations reveal that rocks were uplifted during D3. Without substantial erosion and/or extension accompanying thrusting, thrusting along cannot adequately account for the exhumation of the high pressure rocks in New Caledonia. The major D3b ductile shear a

169

zones are suspected to be reactivated late D2/early D3 thrust faults which acted as normal sense shear zones which unroofed and exposed a zone of deep crustal rocks, the Amoss-Mayavetch metamorphic core complex. The D3b shear zones in the high pressure schist belt intensely transposed earlier formed fabrics in a major (1-3 km thick) retrograde D3b ductile shear zone exposed on the Amoss Ridge (the Hippolyte shear zone). Although this shear zone was formed under high pressure metamorphic conditions, only remnants of the former high pressure mineralogy and microstructure are now preserved. The rocks in this region are the highly deformed equivalents of the deeper rocks in the Amoss River. A compressional orogeny post-D3, folded the high pressure tectonites and the D3 shear zones into a series of upright folds which can be identified in outcrop scale and as major regional structures across the orogen. A series of complex interference structures were produced, which are most clearly identified on the Amoss Ridge. Late extension occurred across the entire schist belt forming, orogen scale, south dipping normal faults across the entire schist belt. These transected all previously formed structures both on the mesoscopic and microscopic scale. Associated with late extension a thermal pulse increased the temperature at shallow crustal levels and static growth of garnet and accompanying greenschist facies minerals grew. The source of the M4 thermal pulse is not constrained, but may be related to plutonism and/or fluids from retrograde reactions or decarbonation. The high pressure tectonites in the schist belt of New Caledonia represent rocks formerly at depths of at least thirty kilometres. Reconstructed crustal thickness across this zone can only account for 12-15 km of the 30-50 km needed to generate high pressure rocks. Erosional processes and two periods of extension in the high pressure schist belt can account for the absence of such large crustal sections today. The early fabric formation is attributed to have formed during the initial stages of orogenesis while the late extension marks the closing stages and final destruction of an orogenic belt. Northern New Caledonia today represents the folded remnants of a metamorphic core complex and the root zone of an ancient mountain belt which may have extended from New Guinea throughout much of the southwest Pacific.

170

LOOPING P-T PATHS, LPHT METAMORPHISM, AND THE EVOLUTION OF THE MIDDLE CRUST: THE PROTEROZOIC OROGEN OF SOUTHWESTERN U.S.A. Michael L. Williams1 and Karl E. Karlstrom2 1 2

Department of Geology and Geography, University of Massachusetts, Amherst, MA, 01002 Department of Earth and Planetary Sciences, Univ. of New Mexico, Albuquerque, NM 87131

Proterozoic rocks of southwestern North America provide a classic example of the interaction of deformation, plutonism, and LPHT metamorphism during more then 200 my of middle crustal orogenesis. The rocks are exposed in a 500-km wide orogenic belt that stretches from southern California to Minnesota. Supracrustal rocks can be divided into two packages: 1.75-1.70 Ga arcrelated mafic metavolcanics and immature metasediments, and 1.7-1.65 Ga felsic metavolcanics, thick clean quartzites, and aluminous schists. Plutonic rocks occur in three age/tectonic groups: 1.75-1.70 Ga arc-related plutons ranging from gabbros to granites, 1.7-1.65 Ga granites, synchronous with the dominant deformation and metamorphism (Yavapai and Mazatzal orogenies), and 1.4 Ga "anorogenic" granites. Peak metamorphic conditions varied regionally from 3-6 kbar and 400 to 700°C, but many uplifts contain 2- and 3-A1 silicate (ky-sill-and) assemblages indicative of near triple point metamorphic conditions. Microstructures generally suggest that metamorphism was synchronous with deformation. Isobars are nearly horizontal, and in many ranges are roughly parallel to the Proterozoic/Phanerozoic unconformity. Interpretations of P-T-t histories from the Proterozoic rocks have varied widely and two distinct models have been proposed. The first, based on mineral textures, porphyroblast relationships, and inclusion vs. matrix thermobarometry, involves counterclockwise P-T-t paths (Grambling, 1986, Geology, 14, 149-152; Grambling et al., 1989, GSA Spec. Pap. 235, 87-110). The prograde path was interpreted to pass close to the kyanite-andalusite phase boundary with nearly isobaric cooling accounting for the present subhorizontal isobars. The second model, based on paragenesis of Al-silicate minerals plus quantitative analysis using the "Gibbs method", postulates that some regions arrived at peak conditions along a steep decompression path (Daniel, 1992, M.S. Thesis, Univ. New Mexico; Grambling and Dallmeyer, 1993, Jour. Met. Geol, 11, 739755). The apparently homogeneous triple point conditions thus represent consistent retrograde, perhaps closure, conditions rather than an atypical tectonic history. However, the P-T path calculations involve critical, and probably erroneous, assumptions about plagioclase, (1) that plag. inclusions are in equilibrium with adjacent garnet, and/or (2) that plag. zoning can be correlated with garnet zoning. Although some decompression during peak metamorphism may be indicated, these assumptions lead to an overestimation of the magnitude of the pressure change. Ultimately, textural data support aspects of both the clockwise and the counterclockwise models. We propose a single composite clockwise/counterclockwise style of P-T path for most if not all of the Proterozoic rocks (Figure 1). The early parts of the path result from heating and thickening of thin juvenile crust. The later parts involve near isobaric cooling at 3 kbar. Because the rocks originated at the surface and the isobaric cooling phase occurred at 8-12 km depth, the cooling paths cross the prograde paths with a looping geometry similar to that proposed by Harley (1989, Geol. Mag. 126, 215-331) for granulite facies rocks. Paths may vary from essentially clockwise to counterclockwise depending on the size of the loop. Regions such as central Arizona, with little or no early over-thickening and decompression have essentially no loop (i.e. counterclockwise paths). Regions with significant over-thickening and decompression such as the northern New Mexico or the Grand Canyon, have a large clockwise loop, and yet early and late reactions suggest counterclockwise behavior. All of the paths display temperature spikes associated with the input of plutonic heat. The spikes occurred at two main times in the overall history (1) during the syn-

171

tectonic clockwise segment where the maximum preserved temperature is generally determined by the proximity to the plutons and (2) during the isobaric cooling history, particularly associated with the 1.4 Ga plutonic event. These spikes reset isotopic systems and locally control the geometry of the metamorphic isograds. The nearly ubiquitous microstructural evidence for synchronous deformation and metamorphism through the protracted history results from increased deformation due to thermal weakening near plutons and increased strain rates associated with prograde dehydration reactions. Proterozoic rocks of southwestern North America may have resided in the middle crust for more than 200 Ma after the Yavapai/Mazatzal orogeny. Thus, they may provide some insight into the nature of processes in the middle crust both during and between orogenic events. More than anything, these rocks demonstrate the fundamental role that plutonic rocks play in influencing and locally controlling deformation and metamorphism at this crustal level. Both at 1.7-1.65 Ga and at 1.5-1.4 Ga metamorphic isograds are related to felsic plutons, although this relationship is better defined in ranges with small P-T-t loops. In addition, fabric-forming deformational pulses are commonly associated with pluton emplacement suggesting that plutonic heat may allow the rocks to deform plastically at stress levels that otherwise may not produce a penetrative deformation. We picture the Proterozoic mid-crust as a dynamically flowing medium in which the geometry and particularly the rate of flow is enhanced during transient periods of heating associated with pluton emplacement. Looping P-T paths may be characteristic of such regions, particularly those that were uplifted at some time after the thermal event that produced the metamorphism.

Figure 1. Representative P-T-t paths for (a) central Arizona, (b) north-central New Mexico, and (c) northernmost New Mexico. Paths do not show the effects of 1.4Ga plutons.

172

DEVELOPMENT OF REGIONAL SCALE FOLD NAPPES: AN EXAMPLE FROM THE MONASHEE COMPLEX, CANADIAN CORDILLERA P.F. Williams & DHL Johnston Department of Geology, University of New Brunswick Fredericton, NB, Canada E3B 5A3 The large dimension of fold nappes measured parallel to their axial surface (commonly tens of kilometres) makes it impossible to explain these structures in terms of conventional fold mechanisms. The Joss Mountain synformal nappe is such a structure in the Thor-Odin culmination of the Monashee Dome of the Canadian Cordillera. The fold involves Monashee "core" and "cover" rocks. It can be traced in a sequence of thick marbles, quartzites, calcsilicates and sillimanite and kyanite bearing schists of the cover, unconformably overlying core para and orthogneisses. The lower limb, on a gross scale, is upward younging and east facing. The axial direction defined by mesoscopic folds is approximately northwesterly trending and the limb extends at least of the order of 10 kilometres perpendicular to this direction. The upper limb is downward younging and east facing. The axial direction has a more southwesterly plunge and the limb extends at least several kilometres perpendicular to this trend (it can be interpreted as much more extensive). Outcrop is continuous over large areas and mesoscopic folds are abundant. Fold symmetries are consistent with their position on the large structure. The main hinge coincides for the most part with the Shushwap valley where exposure is poor and practically unmappable. Where the hinge crosses a major ridge, outcrop is excellent, but the structure is obscured by a large concentration of pegmatite and late plutonic rocks. The rocks of Thor-Odin are thoroughly transposed throughout the culmination. In the next Monashee culmination to the north (Frenchman's Cap) deformation is strong in the southwesterly dipping lower limb in the southwest corner of the culmination, but dies out down section, towards the northeast, where the original stratigraphy is well preserved. Age dating indicates that the parasitic folds on the upper limb may be considerably older than those on the lower limb. We interpret this structure in terms of thick skinned tectonics involving a crustal scale low angle shear zone. The shear zone has thrust geometry and carried accreted rocks over the shelf deposits on the edge of the North American continent. It spread towards the northwest carrying progressively more deformed rocks over their undeformed protoliths. All parasitic folds developed with northwest vergence, but in the upper, older and most deformed part of the shear zone stratigraphy became inverted and the symmetry of the folds was reversed.

173

A DOWN-CUTTING THRUST IN THE MT USEFUL SLATE BELT OF EASTERN VICTORIA C.E. Willman M. Hendrickx, A.H.M. VandenBerg, R.A. Cayley & D. Taylor Geological Survey of Victoria The Mount Useful Slate Belt (MUSB) is a northwest trending belt of deformed Lower Palaeozoic volcanic and sedimentary rocks in the southern part of the Lachlan Fold Belt. Structural style is dominated by low angle thrust faults and tight folds which were produced in a single major Mid Devonian deformation (Tabberabberan). Accurate mapping of the 3dimensional thrust geometry is possible due to the mountainous terrain with a relief of up to 1200 m. This paper describes the major Fullarton Fault which has thrust the youngest sedimentary rocks over older volcanics. A well defined stratigraphy consists of a basal Cambrian volcanic sequence overlain by a sequence of Ordovician to Silurian black shales, siltstones and sandy turbidites. An unconformity may occur between the Cambrian and Ordovician but the sedimentary sequence is internally conformable. The Cambrian volcanics consist of andesitic lavas and breccias, rhyolites and fine grained volcaniclastics, all metamorphosed to lower greenschist facies. The oldest sedimentary rocks are Lower Ordovician turbidites and graptolitic black shales which occur in minor thin fault slivers (Howqua Shale). The next formation is the Mount Easton Shale (Upper Ordovician) which is composed of richly graptolitic black shales, probably several hundred metres thick. The Mount Easton Shale passes into the Lazarini Siltstone and Donnelly's Creek Siltstone, both Early Silurian and consisting of laminated siltstone and minor sandstone. Their total thickness is about 1100 m. A major lithological change occurs with the Serpentine Creek Sandstone (Early Silurian) which consists of up to 1250 m of thick bedded quartz-rich turbidites. Above this, the youngest unit is the Murderers Hill Siltstone (Mid to Late Silurian) which consists of up to 1500 m of laminated siltstone and includes a minor member of graptolitic lithic sandstone and shale (Snake-Edwards Sandstone Member). The Murderers Hill Siltstone was succeeded by the Lower Devonian Walhalla Group but this has not been preserved in the Mt. Useful Slate Belt. Structurally, the Mt. Useful Slate Belt can be divided into two major sectors. In the western zone (10 km wide), the Ordovician to Silurian sediments are folded about regional first order anticlinoria and synclinoria (half-wavelengths up to 3 km) and tight second order folds with half-wavelengths averaging about 300 m. In the eastern zone (4 km wide), the Cambrian volcanics form an antiformal thrust stack exposed as a series of thrust windows through the two youngest sedimentary units (Serpentine Creek Sandstone and Murderers Hill Siltstone). The structural history of the MUSB involves the early development of stratigraphically controlled detachment faults such as the Thomas and Fullarton faults. These form the boundaries between three major thrust sheets, each with a slightly different structural history. The Thomas Fault developed first, at or near the base of the sedimentary sequence but was subsequently truncated by the higher-level and younger Fullarton Fault, as evidenced by detailed mapping in the eastern zone. Here, the Fullarton Fault has juxtaposed the base of the Serpentine Creek Sandstone over the Cambrian volcanics and has truncated the Thomas Fault

174

and displaced the Mount Easton Shale, Lazarini Siltstone and Donnellys Creek Siltstone. This totals about 1500 m of missing stratigraphy. In the eastern zone, the geometry of the Fullarton Fault, adjacent folds and S-C fabrics is consistent with east-directed low angle thrusting. The fault is gently folded indicating fault lock-up occurred before regional fold-shortening was completed. Above the fault, the Serpentine Creek Sandstone is either folded into recumbent folds with an enveloping surface parallel to the fault, or it forms cleaved sections of homoclinally dipping sandstone beds that have remained parallel to the fault. It appears that the competency of the underlying volcanics protected the Serpentine Creek Sandstone from significant fold-shortening The folded Fullarton Fault can be traced westward along the base of the Serpentine Creek Sandstone. Stratigraphic displacement along the fault becomes less apparent away from the Cambrian thrust windows as the detachment becomes sub-parallel to stratigraphy, both in the hangingwall and footwall. It is inferred that major horizontal detachment first occurred along the Thomas Fault, and within the underlying volcanic sequence. This event was probably characterised by fault imbrication in the lower parts of the sedimentary sequence and the development of an antiformal thrust stack in the Cambrian volcanics along the eastern margin of the belt. Subsequent lock-up of the Thomas Fault was probably caused by steepening and uplift of the fault adjacent to and above the volcanics along the eastern margin, as continued duplexing within the volcanics increased the amplitude of the antiformal stack. This allowed the transfer of fault shortening to a higher stratigraphic and structural level, that is, along the Fullarton Fault. The Fullarton Fault appears to have propagated from west to east as a horizontal thrust, with its orientation apparently more strongly influenced by the competent, thick bedded Serpentine Creek Sandstone in the hangingwall than the underlying more pelitic siltstone. One result of this stratigraphic control is that the Fullarton Fault was able to cut down through much of the underlying stratigraphy where it was west-dipping along the western margin of the Cambrian antiformal thrust stack, while the fault and the stratigraphy above it remained essentially horizontal. The arrival of the Fullarton Fault at the more competent antiformal stack of Cambrian volcanics probably initiated the lockup of the fault, and heralded in the last phase of deformation, the progressive development of folding as the major means of shortening together with some accommodation faults (e.g. Frog Hollow Fault). Southwest

Northeast

Cross-section through the Mt. Useful Slate Belt.

175

EXPERIMENTAL ANALOGUES, MICROSTRUCTURAL DEVELOPMENT IN HIGH TEMPERATURE HIGH STRAIN ZONES C.J.L. Wilson1 & Y. Zhang2 1

School of Earth Sciences, Melbourne University, Parkville, Victoria 3052 AGCRC, CSIRO Exploration and Mining, Wembley, WA 6014

Microstructural changes in three sets of experiments involving crystallographic slip in anisotropic polycrystalline ice and naphthalene are described and interpreted with the aid of computer models. The development of microstructure was followed using time lapse photography and transmitted light observations with deformation undertaken in plane strain. The deformation within a grain aggregate of ice that accompanies axial shortening is always inhomogeneous on a grain-scale. The extent of inhomogenity varies depending on the preexisting grain structure and the way it can accommodate intragranular slip. Grain interactions are extremely important in determining the bulk deformation and the degree of grain boundary migration. The consequence of high strains between grains is the formation of high stresses between neighbouring grains and under the appropriate conditions there may be either grain boundary migration or melting at these sites. Where a sample undergoes translation and shear during deformation then anisotropic grains in the appropriate orientation undergo bending. This develops into a buckle instablity and much of the strain is accommodated by grains in easy glide orientations. Where naphthalene is deformed in a simple shear regime there is first the development of kink bands. In these sites deformation is localized and extensive dynamic recrystallization is focused. Asparities induced during the experiments produce a trail of recrystallization behind them. When the translation is sufficient, the different trails connect in a recrystallized band that records part of the imposed simple shear. Without knowing the origin of such recrystallized bands the interpretation of the localized strain history could be misinterpreted. As these bands could be incorrectly considered to be through-going fault zones associated with extensive grain size reduction.

176

THE PREDICTION OF ROCK MASS DISCONTINUITIES IN THE CLARE VALLEY SYNCLINE, BASED ON REGIONAL GEOLOGY INTERPRETATION C.N. Winsor Gartrell School Mining Engineering, Metallurgy and Applied Geology, University of South Australia, The Levels, SA, 5095, Australia Structural geologists traditionally use small scale field data to interpret the large scale. Whilst this is a useful exercise and helps understand the structural evolution of an area, additional applications can be gained by applying large scale knowledge and local conditions to make discontinuities predictions. This benefits geotechnical engineers seeking to determine the nature of discontinuities beyond a zone of observation (Wittke 1993). Predictions based on rock mechanics theory are often unrealistic due to rock mass inhomogenity and anisotropy. The application of structural geological to rock engineering projects is generally poorly appreciated, despite indications that it is useful (e.g. Hobbs 1993). The term discontinuity describes, without reference to geological age, any mechanical defect of low tensile strength along which separation can occur (Priest 1993). Rock discontinuities include bedding, joints, foliations, and faults, which exert a fundamental influence on rock mass strength. Geological information that assists a rock engineer understand the factors affecting discontinuities are under examination as part of a research project: Discontinuity Analysis and Prediction in Polydeformed Areas (DAPPA) The scenic Clare Valley renowned for it superb wines is also the site of a extensive N-S trending F1 macrosyncline plunging 20.350, which exhibits a strike length of forty kilometres. Locally the fold plunges shallowly to the N at northern locations and to the S in the south (Figure 1, after Winsor 1977), the fold's western limb is overturned. The Clare Valley Syncline is an excellent area to examine variations in rock discontinuities along strike and record their geometry and spacing. Fractures in the Clare Valley syncline display facf and 'be' extension, and shear joints geometries controlled by the orientation of bedding and the macroscopic folding (cf Hancock 1985). In addition local fractures are related to a palaeocurrent direction. As the macrostructure is consistent, lithology similar and the factors controlling fractures understood, this information can be used to predict discontinuities at a new quarry, SW of quarry 1. The rock type at the new quarry is a dolomite, however despite this lithological difference, discontinuities are influenced by the macrostructure. The spacing of

LOCATION

Bedding

Cleavage

Fold Axis

Figure 1 Discontinuities at quarries on the western limb of the Clare Valley Syncline.

177

bedding and joint sets for each quarry in Table 1, indicate an apparent correlation between bedding plane spacing and that of 'ac' joints. At nearly all locations the 'ac' joint spacing is about the same as the bedding spacing. The spacing of 'be' joints appears to increase from north to south, possibly due to a regional control. Successful discontinuity predictions require a firm understanding of the rock types present, their effect on discontinuity spacing, deformation history, macrostructural and local conditions. REFERENCES Hancock, P.L., 1985. Brittle microtectonics: principles & practice./, struct, geol. 7, 437-457. Hobbs, B.E., 1993. The significance of structural geology in rock mechanics. IN Comprehensive rock engineering principles, practice and projects. Volume 1 fundamentals. Brown, E.T., vol. editor, p 25-62. Priest, S.D., 1993. Discontinuity Analysis for Rock Engineering. Chapman and Hall, London. Winsor, C.N., 1977. A study of fractures in the mid-north of South Australia. Unpublished BSc (Hons) thesis, University of South Australia. Wittke, W., 1993. Remarks on the practical application of rock mechanics. Int. soc. Rock Mech. News journal 1, 21-24, 41-42. TABLE 1 SPACING DATA CLARE VALLEY QUARRIES (METRES)

Set

Bedding

Quarry 1 Spalding (195 discont.) 0.25

Quarry 2 Andrews (105 discont.) 0.17

Quarry 3 Barina (191 discont.) 0.29

Quarry 4 Sevenhills (342 discont.) 0.23

New Quarry (280 discont.) 0.28

'ac' joints

0.34

0.10

0.29

0.31

0.33

'be' joints

0.11

0.17

0.20

0.26

0.39

poorly represented absent

poorly represented poorly represented

0.76

poorly represented poorly represented

Quartz vein- 0.44 north dipping Shear joints poorly south dipping represented

absent

Overturned bedding surface, Quarry 3 (Barina), 'ac' joints are subvertical, quartz veins related to palaeocurrent direction dip to the north (right of photo) and shear fractures dip south.

178

CRENULATION CLEAVAGE FORMATION: EVOLVING DEFORMATION MECHANISMS WITH INCREASING METAMORPHIC GRADE. Brenton A. Worley. Christopher J.L. Wilson and Roger Powell School of Earth Sciences, Melbourne University, Parkville, Victoria 3052, Australia Localisation of strain into the Wenchuan-Maowen Shear Zone, central China, has resulted in the refolding of a pervasive slaty cleavage (Si) and the formation of a crenulation cleavage at metamorphic conditions ranging from chlorite to garnet zone (Worley and Wilson 1995). This offers a unique opportunity to investigate the way in which the deformation mechanisms responsible for crenulation cleavage formation evolve with increasing metamorphic grade, in particular temperature. In the chlorite zone (~380-420°C) the crenulation cleavage initially forms with a zonal morphology as a result of a combination of microfolding and removal of quartz and chlorite from cleavage P-domains via stress controlled diffusive mass transfer along the pre-existing crenulated fabric. The scale of transport is from the limbs to the hinge zones of the microfolds and initially there is no evidence of recrystallisation of muscovite in the cleavage P-domains. With increasing strain a discrete crenulation cleavage develops and the P-domains evolve into the preferred diffusion path, increasing the scale of diffusion and leading to the introduction of an external fluid. This results in disequilibrium between the P- and Q-domains and recrystallisation of P-domain muscovite. The disequilibrium between the domains persists to biotite zone conditions (~490-550°C) and indicates a domination of diffusion along rather than between the domains. Also important to the crenulation cleavage formation at these temperatures is crystallisation of biotite with it's orientation constrained by the stress/strain conditions and the crenulation cleavage domains which now form the dominant microstructural anisotropy. At garnet zone temperatures (~550-580°C), muscovite in the P- and increasingly in the Q-domains is recrystallised, however these domains are in chemical equilibrium and the diffusional transport of material appears to be controlled by intracrystalline diffusion related to lattice defect rearrangements in the muscovite and quartz grains. Evidence of intracrystalline deformation is preserved as kinks in biotite porphyroblasts and undulose extinction and subgrain formation in quartz grains, however the preservation of these microstructures has been minimised by annealing recrystallisation occurring during the post-crenulation cleavage metamorphic peak. References Worley, B.A. and Wilson, C.J.L., 1995. Deformation partitioning and foliation reactivation during transpressional orogenesis, an example from the Central Longmen Shan, China. Journal of Structural Geology, (in press).

179

DEFORMATION OF CARBONATE ROCKS IN MODI KHOLA OF THE ANNAPURNA AREA, CENTRAL NEPAL Haruka Yamaguchi & Toshiaki Masuda Institute of Geoscienses, Shizuoka University, Shizuoka422, Japan The Annapurna study area is located in the central Himalaya where metamorphic rocks are exposed without klippes. The Main Central Thrust (MCT) lies in the middle of this area, separating the northern Higher Himalayan Sequence (HHS) from the southern Lower Himalayan Sequence (LHS). HHS consists of pelitic, psammitic, and calc-silicate gneisses and marble. It is intruded by Miocene granites. LHS is overthrusted along MCT by HHS and is composed of pelitic, psammitic, and calcareous schists. Foliations are well developed in the most part but not in the far north of HHS. Foliation strikes sub-parallel to MCT and dips gently to the north. The foliation contains a mineral lineation trending NE-SW. The metamorphic grade increases from the chlorite to the garnet zone in LH; whereas the metamorphic grade in HHS increases from the kyanite to the sillimanite zone with increasing distance from MCT, and then decreases into the chlorite zone in the far north. In order to characterize the deformation of metamorphic rocks in this area, we have focussed on microstructural development of calcite in carbonate rocks (marbles in HHS and calcareous schists in LHS) because of their ubiquitous occurence. We collected samples of these rocks along Modi Khola (river). Their microstructures show a clear contrast between HHS and LHS. Calcite grains of the marbles in HHS are coarse, equigranular and polygonal in shape. Their mean grain size ranges from 390 to 460 microns and mean aspect ratios range from 1.8 to 2.2. Those of the calcareous schists in LHS are fine andflattened with mean grain size of 40 - 50 micron and aspect ratio of 2.5 - 3.

180

DUCTILE THRUSTS AND SHEAR ZONES OF THE ADELAIDE HILLS; GEOMETRICAL CONSTRAINTS A. Yassaghi, P. R. James & T. Flottmann Department of Geology & Geophysics, Adelaide University, SA., 5005 The Adelaide Hills form the northern section of the Southern Adelaide Fold-Thrust Belt, and consist of a sequence of late Precambrian Adelaidean metasedimentary rocks direcdy overlying a number of early Proterozoic basement inliers. The Adelaidean rocks mainly comprise multiple units of competent quartzites of the Aldgate Sandstone and Stonyfell Quartzite Formations interleaved with incompetent pelitic and phyllite rocks of the Woolshed Flat and Saddleworth Formations. Six major stacked parallel shear zones with generally NE-SW trends and shallow SE dips have been identified which are bounded by lower boundary thrusts. These shear zones caused right lateral offset to the rock units. The higher shear zones, i.e. Summertown shear zone and Mt. Bold shear zone, represent the largest displacement. This may due to their association with basement inliers. All the shear zones are linked with minor imbricate ductile thrusts which form trailing edge propagating fans verging to the NW. Detailed structural analysis of the area has delineated an array of minor folds and fabrics, many of which are restricted to the shear zones and to associated minor imbricate ductile thrusts. Minor folds, the most prominent features of the structural geometry of the area, are disharmonic structures forming open-to-tight buckle folds plunging generally SSE. Two types of minor fold geometry can be distinguished. (1) Gently to moderately curvilinear, close-to-isoclinal recumbent sheath-like fold types with fold axes subparallel to the local stretching lineation (Lx). These folds are well developed within the Summertown and Mt. Bold shear zones, where basement and the basement cored Onkaparinga Anticline are located, respectively. In the Mt. Bold shear zone, minor folds are persistently cut by small-scale shear zones. (2) Gentle-to-tight asymmetric fold types display little or no detectable hinge line curvature and the fold axes vary from orthogonal to oblique to Lx. ITie minor folds are developed in rocks which lie in high strain zones and also where they have a strongly layered anisotropy, due to the interbanding of psammitic and pelitic layers. Rocks have no such layering, such as the phyllite or carbonate units, zonal and discrete crenulation folds are more developed and their axes are almost always orthogonal to Lx. Besides these small-scale compressional structures, extensional crenulation cleavages also exist. These structures which are always seen in mica-rich phyllites dip gendy NNW and are best observed in regions where the minor folds are absent Other notable minor structures are two types of cleavage development. Type one, which is well developed as a penetrative cleavage within die incompetent units, is more distributed and is named as the regional cleavage or SI a. This probably lies in the axial plane of major folds, which formed during the early stage of regional deformation. This cleavage often lies at 45° or less to the bedding and the vergence direction is consistently to the ESE, irrespective of the dip of the bedding. Closer to the ductile thrusts and shear zones the other type cleavage named fault-related cleavage or Sib, which in turn crenulates the Sla, is also developed. Their concentrations increase toward the lower boundary thrust of the shear zones in which crenulation folding of layer silicates and transposition of more competent layers are also intensified. These fault-related cleavages are also axial planar cleavages to minor asymmetric folds. Sla, Sib and bedding structures become parallel to each other at the base of most shear zones. The orientation of Lx, which is gently SE plunging in most parts of the area is roughly constant This orientation as well as the observation that the small-scale shear zones cross-cut the upper limbs of sheath-like folds suggests that these sheath like folds formed during the initial stages of the deformation. At least some of the strain responsible for the formation of the shear zone fabrics and structures, however, must have occurred at a later stage, most likely during a single progressive phase. Furthermore, the restriction of folds and fabrics within the shear zones and their associated ductile thrusts and the parallelism of all planer fabric elements at the base of shear zones might indicate that the overall shortening in the Adelaide Hills area was largely accomplished by slip along the the major mapped high strain discrete shear zones. 181

Montyfute Rd/

ADELAIDE HILLS AREA

Mitcham Quartzite Saddleworth Fm,

including quartzite

Stonyfell Quartzite Woolshed Flat Fm. Montacute Dolomite Aldgate Sandstone Basement Tertiary faults Thrusts Inferred thrusts Anticline Syncline

is •V -

Geological boundary Inferred geological boundary <s

v - -

Roads

MIGMATIZATION DURING LOW PRESSURE - HIGH TEMPERATURE METAMORPHISM IN THE MT LOFTY RANGES, SOUTH AUSTRALIA. S. Zakowski. Department of Earth Sciences, Monash University, Melbourne, Vic 3168, Australia. The role of granitoid plutons in the structural and metamorphic evolution of low-pressure high-temperature (LPHT) metamorphic belts has long been debated. The granites within the southern Adelaide Fold Belt occupy a central region of highest metamorphic grade characterised by the progressive development of migmatite, sillimanite, staurolite-andalusite and biotite zones down temperature from the plutons (Offler & Fleming, 1968; Mancktelow, 1990; Dymoke & Sandiford, 1992). This arrangement of plutons and metamorphic zones is used by Sandiford et al. (1992) to suggest that the currently exposed (and unexposed) magmatic bodies represent the heat source for LPHT metamorphism in the southern Adelaide Fold Belt. However, the timing sequence of migmatite generation with respect to deformation events, and the timing of granitic intrusion with respect to both migmatization and deformation shows that the granitic bodies could not themselves be the cause of the LPHT regional metamorphism. The present study has shown that the structural history of the Mt Lofty Ranges involved three separate deformation events. The earliest deformation event, DL is only locally preserved in the more competent igneous lithologies and andalusite porphyroblasts as a spaced schistosity and/or inclusion trails. As this is the extent of the data available regarding Dl, the stress regime during this deformation event remains uncertain. The D2 deformation event resulted in the formation of a pervasive, bedding parallel S2 fabric and L2 mineral elongation lineation. While S2 was subsequently folded by D3 and is now upright to inclined, the S2 schistosity probably originated as a sub-horizontal schistosity. Locally, S^Jabrics are well developed sets of low-angle normal shear bands (without a consistent sense of shear). The movement direction of individual S2b shear bands is parallel to the L2 extension lineation developed in S2, suggesting that both fabrics developed as part of an ongoing, subhorizontal deformation event. With the present data set it is not possible to determine whether D2 involved any non-coaxial deformation. It is suggested however, that D2 was extensional (vertical flattening and N-S extension) in nature because there is no evidence for major stratigraphic repetition in the internal parts of the Southern Adelaide Fold Belt consistent with nappe- or nappe-and thrust style tectonism (Jenkins & Sandiford, 1992). The third deformation event D3_was a regional folding event that failed to produce a penetrative axial planar cleavage in the high to medium grade metamorphic zones. F3 folds are regionally upright with statistically horizontal hinges. The orientation of the S3 and L3 fabric elements suggests that the D3 co-axial refolding of D2 structures involved approximately east-west subhorizontal shortening and N-S sub-horizontal extension. It is suggested that the different strain orientations of structures during D2 and D3 reflect fundamental changes in the regional stress field during the Delamerian Orogeny in the south eastern Mt Lofty Ranges. Four types and three generations of migmatites are observed in the study area. The first generation of migmatites (Mi) are the stromatic migmatites that are parallel to the Si schistosity. These have been isoclinally folded during the D2 and D2b deformation events. The most common regionally developed migmatites are the second generation of stromatic migmatites (M2) that are found parallel to the S2 schistosity. The third generation of migmatites are the shear zone, nebulitic and shollen migmatites (M3) that are locally developed in a linear zone of migmatization surrounding the Rathjen Gneiss. The structural overprinting relationships imply that there are at least three distinct generation of migmatites in the study area. The present study has also established that the M2 stromatic migmatites are in situ partial melts of the compositionally banded quartzofeldspathic schists and gneisses in which they are found. The melting process was facilitated by the approximate minimum melt composition of the protolith, elevated temperatures and high water activity during regional metamorphism and deformation.

183

Peak regional metamorphism is indicated by the growth of fibrolitic sillimanite elongated within the S2 foliation, and the regional formation of stromatic migmatites (M2). The three generation of migmatites, correlated to the development of peak metamorphic assemblage, namely sillimanite and K-feldspar, give evidence of two thermal pulses during the Delamerian Orogeny in the south eastern Mt. Lofty Ranges. It is suggested that the first, which probably begun prior to the appearance of penetrative structures was a regional thermal event, while the last was a thermal pulse that was localised around the 'Rathjen Gneiss' linear migmatite zone. The oldest and largest syn-tectonic pluton in the "core" of the Belt is the Rathjen Gneiss emplaced at 516 Ma (Sandiford et al., 1992). Structural fabrics and migmatites preserved in the pluton, namely: the earliest foliation Si; and the S2 schistosity and L2 mineral elongation lineation correspond to the Si, S2 and L2 fabrics preserved in the surrounding metasediments. Two phases of remelting of the pluton have generated stromatic migmatites parallel to S2 (the M2 migmatites) crosscut by the later M3 nebulitic and shear zone migmatites. These relationships suggest that the Rathjen Gneiss was emplaced prior to the earliest deformation event and was remelted during peak metamorphism. Thus by demonstrating that the currently exposed granitic bodies had undergone partial melting during peak metamorphism, the present study shows that the granitic bodies could not themselves be the cause of the LPHT regional metamorphism. It is suggested that the heat source for the early granitic magmatism and the LPHT metamorphism in the Mt Lofty Ranges was provided by extension-associated mantle melting and intrusion of mafic magmas. References Dymoke P. & Sandiford M., (1992). Phase relationships in Buchan Facies Series pelitic assemblages: calculations with application to -andalusite-staurolite paragenesis in the Mt Lofty Ranges, South Australia. Contrib Mineral Petrol 110, 121-132. Jenkins R.J.F. & Sandiford M., (1992). Observations on the tectonic evolution of the southern Adelaide fold Belt. Tectonophysics 214,27-36. Mancktelow N.S., (1990). The structure of the southern Adelaide Fold Belt, South Australia. Geol Soc Aust Spec Publ 16,369-395. Offler R. & Fleming P.D., (1968). A synthesis of folding and metamorphism in the Mt Lofty Ranges, South Australia. J geol Soc Aust 15,245-266. Sandiford M., Foden J., Shaohua Z. & Turner S., (1992). Granite genesis and the mechanics of convergent orogenic belts with application to the southern Adelaide Fold Belt. Trans R Soc Edinburgh Earth Sci 83, 83-93..

184

LITHOSPHERIC STRUCTURE IN THE EASTERN AUSTRALIAN PASSIVE MARGIN AS REVEALED BY GRAVITY MODELLING Y. Zhang 1, E. Scheibnerl, B. E. Hobbsl, A. Ordl & B. Drammond2 1AGCRC, CSIRO Exploration & Mining, Wembley, WA 6074; 2North Ryde, NSW 2113. 2AGCRC, Australian Geological Survey Organisation, GPO /to* 378, Canberra, ACT 2065 The horizontal maximum compressive stresses in eastern Australia do not align with the absolute plate velocity trajectories and therefore cannot be explained simply by plate rotation related boundary forces. The results of Zhang et al.'s mechanical models (1995) suggest that the body forces associated with an upper-plate passive margin geometry and an upper mantle wedge contribute significantly to the stress field in the region. However, the results of such mechanical analyses are sensitive to the densities and geometries adopted for the deep structures in a model, about which there are some uncertainties. In this study, therefore, we aim to obtain further constraints on the lithospheric structure by modelling gravity for the eastern Australian passive margin. The forward modelling program developed by Roach (1993) calculates gravity anomalies based on the density and geometry of a cross section using the integral method of Talwani et al. (1959). Three cross sections in eastern Australia are modelled. The first is the eastern part of the Broken Hill-Sydney-Tasman Sea Transect (Scheibner et al. 1991). This is a 750 km long and 150 km deep profile comprising sea water, crustal, upper mantle and asthenospheric layers. Two other cross sections are about l°-latitude away, to the North and South. The sensitivity of gravity to densities/geometries is investigated by assuming variable parameters for the modelled layers. The calculated gravity anomalies are compared with the observed Bouguer gravity anomalies. The results show that the calculated gravity anomalies closely reflect the assumed lithospheric density-geometry structures in the models. For the model with the density-geometry structures identical to those used in Zhang at al.'s mechanical models, the calculated gravity anomalies generally compare well with the observed Bouguer gravity anomalies. The gravity curves are both characterised by overall lower values for the continental lithosphere and higher values for oceanic lithosphere with a gradual transition across the continental-oceanic crustal boundary (COB). This agreement provides support in particular for a denser mantle wedge below the COB (Zhang et al. 1995). These results also support the idea that the eastern Australian passive margin may depart from strict local isostatic equilibrium, since the cross section generating such gravity anomalies actually involves a mass difference in different lithospheric columns, giving rise to a maximum static stress difference of 79 MPa at a depth of 120 km below the COB; the static stresses at this level are close to 4 GPa. Sensitivity analyses of gravity anomalies with respect to densities show that an attempt to achieve approximate isostatic equilibrium by using a smaller density for the oceanic upper mantle results in gravity anomalies significantly different to the observations. Modifications in the structural geometries could further improve the agreement between the calculated and observed gravity curves, though a perfect match is not necessary for modelling on such a scale. Furthermore, comparison of the results for the cross sections at different latitudes suggests that the geometrical features of the lithospheric stuctures, such as the dip angle of the mantle wedge, could change along the direction of the COB. References Roach, M., 1993. MODEL2D Version 3.0, User Manual. The University of Tasmania. Scheibner, E., Powell, C.McA. & Spencer, R., 1991. Broken Hill-Sydney-Tasman Sea Transect, NSW, Eastern Australia (2 sheets). Expl. Notes 29p. Inter-Union Comm. Lithosph. & Am. Geophys. Union, BGGT-5. Talwani, M., Worzel, J. L. & Landisman, M., 1959. Rapid gravity computations for two-dimensional bodies with application to the Menodcino submarine fracture zone. J Geophys. Res., 64, 49-59. Zhang, Y., Scheibner, E., Ord, A. & Hobbs, B. E., 1995. Numerical modelling of crustal deformation of the eastern Australian passive margin. Aust. J. Earth Sciences (submitted).

185

Notes