Subsurface Modelling of the Gilmore Fault Zone

SUBSURFACE MODELLING OF THE GILMORE FAULT ZONE:

IMPLICATIONS FOR LACHLAN TECTONIC RECONSTRUCTIONS

Deepika Venkataramani

Supervisors:

Dr David Boutelier,

Dr Robert Musgrave,

Dr Alistair Hack

& Prof Bill Collins

Thesis submitted in partial fulfilment of the requirements for the

Master of Philosophy (MPhil) degree, School of Environment and Life

Sciences,

University of Newcastle

March 2017 DECLARATION

The work presented in this thesis is entirely the result of original research conducted by the author unless otherwise acknowledged. This work has not been accepted for the award of any other degree or qualification at any other university or tertiary institution.

Deepika Venkataramani March 2017

i ABSTRACT This study considers the tectonic evolution of the Lachlan Orogen by modelling the subsurface morphology of the Gilmore Fault Zone (GFZ). The GFZ marks a distinct geophysical contrast between (high gravity, low magnetic intensity) high-grade metamorphic rocks found in the Wagga Metamorphic Belt (WMB) to the west, and the (low gravity, uniformly high magnetic intensity) low-grade metavolcanic rocks found in the Macquarie Arc and Silurian rift basins to the east. Subsurface structure around the GFZ in the vicinity of Barmedman (34°8'33.94"S 147°23'11.39"E) has been inverted by iterative 2.5D potential-field modelling of gravity and magnetics, constrained by pre-existing reflection seismic profiles, potential-field interpretations by previous workers, and physical properties data collected on representative lithologies. My findings show that the surface structure mapped as the Gilmore Fault is an east- dipping, shallow thrust fault, and does not correspond to the major crustal ‘suture’ envisaged in regional tectonic studies. This shallow east-dipping fault should be renamed the Barmedman fault. The Barmedman Fault flake (as opposed to the GFZ) is curved, and terminates abruptly to the north, indicating the Barmedman Fault flake is the base of a series of thrust flakes imposed on the pre-existing main fault in the GFZ. The GFZ is best described as a steep west-dipping fault zone constituting the eastern flank of the Silurian Tumut Trough. I conclude that the modelled structure of the GFZ is not consistent with the terrane accretion model since the GFZ does not mark a suture between significantly different geological units. The modelled structure of the GFZ shows evidence of multiple contractional and extensional events which are the main characteristics of the accretionary orogen tectonic model. However, steep faults are observed as well, indicating that significant lateral slip, a characteristic of the orocline tectonic model, is mechanically possible.

ii ACKNOWLEDGEMENTS I thank Su God for all that I am allowed to be and for all the blessings. I acknowledge the Aboriginal custodians of the land in the region that I studied, the Elders, past, present and future, and Country itself. I OFFER MY HUMBLE GRATITUDE TO ALL THOSE WHO HAVE HELPED ME ‘ACHIEVE’ UP TO THIS POINT.

I acknowledge the funding provided by the Geological Survey of NSW. I would especially like to thank Phil Gilmore, John Greenfield, Rosemary Hegarty, Jaime Robinson and Mark Eastlake for their encouragement and eagerness to help me solve problems and make discoveries. I acknowledge and thank Astrid Carlton for her multiple emails, phone calls to technical support, patience and help in processing the magnetic field data. I would also like to thank Linda Stenning for helping me process and correct the collected gravity data.

Thank you to my academic SUPERvisors for pointing me in the right direction, for their multidisciplinary guidance in helping me fine tune my research and sharpen my logic like a well-crafted Katana. I thank David Boutelier for his calm, precise guidance on understanding tectonic concepts and for helping me to organise myself as a researcher. I thank Bill Collins for always setting aside time to address key concepts and improve my understanding. Thank you for your time, patience and guidance. Thank you Alistair Hack for remembering that I wanted to get involved in Geophysics and for bringing this research topic to my attention. I would have missed out on this wonderful opportunity if it weren’t for you! Finally I thank Bob Musgrave for his altruistic desire to teach and help me learn. Thank you for always encouraging me to exhibit my work and for your support at conferences. Thank you for helping me grow my knowledge in the field of Geophysics. Every conversation with you is an opportunity to expand my knowledge. I appreciate you all.

No word is powerful enough to convey the debt of gratitude to my divine parents. Thank you for setting my feet upon the path of knowledge and teaching me that there is no end, that learning goes on throughout life but for also imbuing me with the strength and confidence to tackle any obstacle in my way. Venkat, thank you for your well-placed proverbs and the endless support only a dad like you can give. Thank you Amrita for having unstinting confidence in me, because of this I will always have confidence in myself. To Anu, I thank you for teaching me, “To strive, to seek, to find, and not to yield.”(Alfred Lord Tennyson)

iii Table of Contents DECLARATION...... i ABSTRACT ...... ii ACKNOWLEDGEMENTS ...... iii TABLE OF CONTENTS...... iv LIST OF FIGURES ...... vi LIST OF PLATES……………………………………………………………………………………………………………………..…vii LIST OF TABLES ...... vii CHAPTER 1: INTRODUCTION ...... 1 1.1 THE SOUTHERN TASMANIDES OF EASTERN AUSTRALIA ...... 1 1.2 AIMS ...... 3 1.3 OBJECTIVES ...... 5 CHAPTER 2: GEOLOGICAL SETTING...... 8 2.1.1 Orogenic Events...... 11 2.1.2 The Macquarie Arc...... 12 2.1.3 The Gilmore Fault Zone ...... 12 2.2 EXISTING TECTONIC MODELS ...... 16 2.2.1 The Lachlan Orogen Formed By Suspect Terrane Accretion ...... 16 2.2.2 The Lachlan Orogen Formed As An Extensional Accretionary Orogen ...... 16 2.2.3 The Lachlan Orogen Was Subjected To Subduction Flip/ Polarity Reversal ...... 18 2.2.4 The Lachlan Orogen Achieved Its Current Form By Oroclinal Bending...... 18 2.2.5 Extrusion of the WMB: ...... 20 CHAPTER 3: METHODS AND DATA ACQUISITION...... 21 3.1 PREVIOUS WORK ...... 22 3.2 REFLECTION SEISMIC DATA ...... 22 3.3 GRIDDED POTENTIAL FIELD DATA ...... 24 3.4 COLLECTED TMI AND MAGNETIC SUSCEPTIBILTY ...... 25 3.4.1 Magnetic/Solar Storms And CMEs...... 25 3.4.2 Buckshot Gravel...... 28 3.4.3 High Frequency Power Lines ...... 28 3.4.4 Problems With Processing/ Despiking Data...... 28 3.5 COLLECTED GRAVITY ...... 28 3.6 PETROPHYSICAL PROPERTY DATA ...... 31 3.6.1 Petrophysical Properties Analysis...... 34 3.6.2 Remanence ...... 37 iv 3.7 GEOLOGICALLY CONSTRAINED INVERSION BY ITERATIVE FORWARD MODELLING ...... 39 CHAPTER 4: GEOPHYSICAL MODELLING RESULTS AND GEOLOGICAL INTERPRETATION ...... 42 4.1 REFLECTION SEISMIC INTERPRETATION ...... 42 4.2 GEOLOGICAL INTERPRETATION ...... 43 4.3 RESULTS OF INVERSION ...... 44 4.3.1 Cross-section 1 ...... 44 4.3.2 Cross-section 2 ...... 50 4.3.3 Cross-section 3 ...... 52 CHAPTER 5: TECTONIC SYNTHESIS ...... 56 5.1 IMPROVED FIT...... 56 Box 1 ...... 56 Box 2 and Box 3 ...... 57 Box 4 and Box 5 ...... 57 Box 6 ...... 58 Box 7 ...... 58 5.2 GEOLOGICAL IMPLICATIONS...... 58 5.3 EVALUATION OF EXISTING TECTONIC MODELS ...... 60 5.3.1 Terrane Boundary ...... 61 5.3.2 Extensional Accretionary Orogen/ Accordion Tectonics ...... 61 5.3.3 Subduction flip/ polarity reversal ...... 62 5.3.4 Orocline model ...... 62 5.3.5 Extrusion of the WMB ...... 63 5.4 TECTONIC EVOLUTION ...... 64 5.4.1 490-440 Ma (Late Cambrian – Late Ordovician) ...... 65 5.4.2 440 Ma (Benambran Orogeny)...... 67 5.4.3 435- 425 Ma (Silurian Extension)...... 69 5.4.4 425- 420 Ma (Bindian Orogeny) ...... 70 5.4.5 Post Bindian ...... 71 5.5 SUMMARY ...... 74 CHAPTER 6: CONCLUSIONS AND RECOMMENDATIONS ...... 76 REFERENCES ...... 78 APPENDICES ...... 88 APPENDIX A1: ...... 88 APPENDIX A2: ...... 89 APPENDIX A3: ...... 90

v APPENDIX B1: ...... 91 APPENDIX B2: ...... 92

LIST OF FIGURES

Figure 1.1: Tasmanides of eastern Australia…………………………………………….2 Figure 1.2: Map of NSW, showing the GFZ, Siluro-Devonian basins, Macquarie Arc, turbidites and MORBs…………………………………………………………………...6 Figure 1.3: Simplified iterative method used to arrive at the final model…………….....7 Figure 2.1: Simplified structural map showing the subdivision of the LO into the western, central and eastern zones……………………………………………………...10 Figure 2.2: Location map showing the five major geological units considered in the study area……………………………………………………………………………….13 Figure 2.3: TMI grid……………………………………………………………………15 Figure 2.4: Bouguer gravity grid……………………………………………………….15 Figure 2.5: Model showing tectonic switching present in extensional accretionary orogen model …………………………………………………………………………..17 Figure 2.6: Image contrasting the accretionary orogen model with the subduction flip/ quantum tectonic model……………………………………………………...18 Figure 2.7: The Lachlan orocline model……………………………………………….19 Figure 2.8: Extrusion model……………………………………………………………20 Figure 3.1a,b : Composite image of reflection seismic lines 99agsL1-L3 and corresponding geological model of Direen et al. (2001)……………………………… 23 Figure 3.2: Multitrack profile of six lines of collected magnetic field data……………26 Figure 3.3: Graph comparing collected and gridded TMI data………………………...27 Figure 3.4: Map showing lines of rover magnetometer collected (25m spacing) along roads surrounding the town of Barmedman……………………………………………27 Figure 3.5: Map showing line of collected Gravity data……………………………….29 Figure 3.6 a: Stacked profile comparing the grid-interpolated gravity (blue) with the field collected gravity (orange). Note the goodness of fit on flanks of the anomalies…30 Figure 3.6 b: This graph shows the good fit and correspondence between grid- interpolated gravity (blue) and field collected gravity (orange)……………………..…30 Figure 3.7: Map showing location of samples collected in the field…………………...32 Figure 3.8: A general overview of 2.5D modelling ………………………...……….…39 Figure 4.1: Geology map of study area over cross-section 1 and 2…………………….46 Figure 4.2: Map interpretation by Bell (in prep) superimposed on 1VD of the TMI…..47 Figure 4.3: Composite image of modelled cross-section 1 ……………………………48 Figure 4.4: Composite image of modelled cross-section 2 ……………………………53 Figure 4.5: Composite image of modelled cross-section 3 ……………………………55 Figure 5.1: Evolution of the LO from 490 -440 Ma…………………………………....66 Figure 5.2: Tectonic model showing the onset of the Benambran orogeny …………...68 Figure 5.3: Tectonic evolution of the LO between 435-425 Ma Silurian extension…...69 Figure 5.4: Tectonic evolution of LO between 425 -420 Ma…………………………..70 Figure 5.5: Evolution of the LO during the Tabberabberan Orogeny………………….72 Figure 5.6: Tectonic evolution of the LO in the extensional phase following the Tabberabberan …………………………………………………………………………73 Figure 5.7: Tectonic evolution of the LO during the Kanimblan Orogeny…………….74

vi LIST OF PLATES Plate 1a: Ferruginous grains attached to magnetic base plate of a Leica DGPS……………….33 Plate 1b: Scintrex Autograv (CG5) gravimeter (levelled) and GPS at base station……….…...33 Plate 1c: Temora Volcanics sampled rock cut to size for petrophysical property measurements…………………………………………………………………………………...33 Plate 1d: Bronxhome Formation and 1 sample of foliated Barmedman granite- sampled rock cut to size for petrophysical property measurements…………………………………………...33 Plate 1e: Wagga turbidites sampled rock cut to size for petrophysical property measurements…………………………………………………………………………………...33 Plate 1f: Wagga turbidites sampled rock cut to size for petrophysical property measurements…………………………………………………………………………………...33 Plate 1g: Yiddah Formation sampled rock cut to size for petrophysical property measurements…………………………………………………………………………………...33 Plate 1h: Yiddah Formation sampled rock cut to size for petrophysical property measurements…………………………………………………………………………………...33 Plate 1i: Wagga turbidites sampled rock cut to size for petrophysical property measurements…………………………………………………………………………………...33

LIST OF TABLES Table 1a: Table of rock samples and corresponding petrophysical property data……..35 Table 1b: Table of Mean and Standard deviation values of collected petrophysical properties……………………………………...... 36 Table 2: Range of inverted petrophysical properties derived from cross-section 1……38 Table 3: Range of inverted petrophysical properties derived from cross-section 2……38 Table 4: Range of inverted petrophysical properties derived from cross-section 3……38 Table 5: Model sensitivity analysis showing RMS values for different dip directions...41

vii CHAPTER 1: INTRODUCTION

1.1 THE SOUTHERN TASMANIDES OF EASTERN AUSTRALIA

This study aims to elucidate the role of the Gilmore Fault Zone (GFZ) in the tectonic evolution of the Lachlan orogen (LO) within the Tasmanides of Eastern Australia. The

Tasmanides consist of a series of Neoproterozoic to Triassic orogenic belts which once marked the paleo-Pacific edge of the Gondwana continent (Fig 1.1). These recorded multiple cycles of convergent margin tectonics, arc volcanism, and back-arc extension characterising accretionary orogens (Royden & Burchfiel, 1989; Glen, 1992; Collins,

2002a, b; Collins & Richards, 2008). There are four major Orogens within the

Tasmanides: The Lachlan Orogen forming the bulk of New South Wales and Victoria, the Delamerian Orogen to the west, the Thomson Orogen to the north and the New

England Orogen to the east (Fig 1.1).

The geodynamic evolution of the Tasmanides, more specifically the LO, is controversial.

The present consensus is that the LO formed as a vast, early Paleozoic arc/back-arc system behind a long-lived west-dipping subduction zone that closed in the middle

Paleozoic, (Powell, 1984; Collins and Vernon, 1992; Fergusson & Coney, 1992b; Gray

& Foster, 1997; Scheibner & Basden, 1998; VandenBerg et al., 2000; Foster & Gray,

2000; Collins & Hobbs, 2001), This view has also been modified to include the possibility of multiple subduction zones (Gray & Foster 2004), or that stalling and subduction flip events occurred (Aitchison & Buckman, 2012). Alternatively, Scheibner (1985) suggests that the Tasmanides are composed of tectonostratigraphic terranes bound by crustal sutures.

Figure 1.1: Tasmanides of eastern Australia. The following are volcanic belts of the Macquarie Arc: jnvb= Junee-Narromine volcanic belt; mvb= Molong volcanic belt; rgvb= Rockley-Gulgong volcanic belt; kvb= Kiandra volcanic belt. The western boundary of the Tasmanides or ‘Tasman line’ as defined by Scheibner (1974) and/or Glen et al. (2012), separate the Tasmanides from Cratonic Australia. The Lachlan Orogen (LO) is divided into: ELO= Eastern LO; CLO= Central LO; WLO= Western LO. The study area, centred on the Junee-Narromine volcanic belt and showing the Gilmore Fault Zone (GFZ) as the boundary between the CLO and the ELO is boxed in a black rectangle. Image adapted from Glen et al., 2012 and Belica et al., 2017.

The controversy regarding the geodynamic evolution of the LO is further enhanced by the lack of consensus regarding the nature of the basement or substrate of the LO.

Evidence has been found for continental (Rutland, 1973; White et al., 1976; Christensen

& Mooney, 1995), oceanic (Crook, 1969, 1974a; Meffre et al., 2011; Spaggiari et al.,

2003a, 2004a; Forster et al., 2015) or mixed (Scheibner, 1974; Finlayson et al., 2002;

Glen et al., 2007), allowing a variety of geodynamic models to be proposed.

Recent identification of curvilinear geophysical features throughout the Lachlan

(Musgrave & Rawlinson, 2010; Cayley et al., 2011; Musgrave, 2015) has rejuvenated the proposition of a more complex tectonic evolution with diachronous ‘extrusion’ from

oblique collision resulting in both major thrust as well as strike-slip faults (Morand &

Gray, 1991).

1.2 AIMS

As already mentioned, several models have been formulated for the tectonic evolution of the Lachlan Orogen:

 Terrane boundary (Scheibner, 1985)

 Extrusion of the WMB (Morand & Gray, 1991)

 Extensional accretionary Orogen/Accordion Tectonics (Collins, 2002 a,b)

 Subduction flip/ polarity reversal (Aitchison & Buckman, 2012)

 Orocline (Musgrave & Rawlinson, 2010; Cayley, 2012; Musgrave, 2015)

Each model includes a number of predictions regarding the subsurface geometry of the

GFZ:

- Is the GFZ a west-dipping or east-dipping fault zone (suture model, Subduction

flip/ polarity reversal)?

- Is it a vertical structure accommodating horizontal slip (extrusion model, orocline

model)?

- Is there any evidence in subsurface geometry for multiple pulses of

extension/compression (Accordion model)?

The aim is to better constrain the subsurface geometry of the Gilmore Fault zone, thereby constraining which of the various tectonic models of the Lachlan are compatible with the

Gilmore fault Zone and which are not.

The subsurface geometry of the GFZ can be assessed using joint inversion of magnetic and gravity along profiles, aided by seismic reflection surveys and constrained by

petrophysical measurements of representative lithologies. A profile is constructed (Fig

1.2), made of a number of blocks with constrained petrophysical properties (Fig 1.3), generating two independent signals. Geometries of blocks and petrophysical properties are varied to match observed signals. The goodness-of-fit will be statistically calculated.

The modelled geometries would give a good indication of the nature of the subsurface allowing me to evaluate the geodynamic evolution (Fig 1.3).

This area has previously been modelled by Direen et al. (2001), who used the same software and potential field approach as I have, and interpreted by Glen et al. (2002).

These authors have synthesised an instructive model of the first order structure of the LO but the area surrounding the GFZ itself, is poorly modelled. The signal fit surrounding the GFZ is poor, particularly on the flanks of prominent anomalies which suggests that the dip on the edge of the anomaly source is not sufficiently modelled. This potentially leads to multiple permutations regarding the possible attitude of surrounding strata.

The work done by Direen et al. (2001) and Glen et al. (2002) also does not consider the presence of oroclines. Other Authors (Musgrave & Rawlinson, 2010; Cayley et al., 2011;

Musgrave, 2015) have published work on oroclines in the LO which is based on more

‘recent’ magnetic and gravity compilation grids and with new ways of enhancing and filtering (Appendix A) geophysical data (Cooper & Cowan, 2006). However none of these works fully considers the role of the GFZ. A re-examination of the GFZ and its role in the evolution of the Lachlan Orogen is further justified.

1.3 OBJECTIVES

Field-work:

 Collect magnetic susceptibility, magnetic field and gravity measurements along the location of the modelled cross-sections.  Collect core samples (where possible, representative, unweathered hand specimens) for laboratory analysis.  Compare collected potential field data (smaller line spacing, ‘higher resolution’) with existing compilation grids. Determine if these data sets differ significantly thus effect the validity of model.

Laboratory analysis:

 Extract petrophysical properties from samples (magnetic susceptibility, saturated density) and use to inform ranges of values used in the model.  Calculate NRM (Natural Remnant Magnetisation) and Koenigsberger ratios to determine degree of remanence possibly present (Chapter 3).  Use to inform modelling input parameters (magnetic susceptibility, saturated density). If there is a large remanence contribution, then correct for this in the models.

Modelling:

 Model three individual cross sections along the GFZ in the Cootamundra and Forbes 1:250 000 map sheets using magnetic, gravity and reflection seismic data (Fig 1.2).  Produce a geologically constrained model of magnetic and gravity data to yield a revised model of the GFZ.  Establish a new tectonic framework with the modelled cross-section, using cross- cutting relationships.  Verify or validate existing models for the tectonic evolution of the Lachlan Orogen.

Figure 1.2: Map of NSW, Australia showing the trace of the GFZ and other important faults as well as Siluro-Devonian basins, Ordovician volcanic belts (Macquarie Arc), turbidites and mid ocean ridge basalts (MORB). Note that most of these faults extend through the majority of the state. The location of the 1VD is represented by the blue box. Image adapted from Glen et al. (2002). The first vertical derivative (1VD) of Total magnetic intensity (TMI) of the study area, within the 250K Cootamundra and Forbes map sheets, shows the location of the modelled cross - sections (solid black lines numbered 1 to 3). Four major faults (dashed and solid lines) bisect these cross-sections. 6

Figure 1.3: Simplified iterative method used to arrive at the final model. Each cross- section typically took 4 different revisions. Modelled bodies were evaluated to determine what geological model they best represented. Root Mean Square (RMS) values were used to evaluate geologically sensible models.

CHAPTER 2: GEOLOGICAL SETTING

2.1 THE LACHLAN OROGEN (LO)

The LO is an 800 km wide portion of the >1500 km wide Tasmanides (Glen, 2013). It is generally classified as an accretionary orogen that records the retreat of the Paleo-pacific margin eastward from the Gondwanan continent (Scheibner, 1972, 1987; Cas, 1983;

Royden & Burchfiel, 1989; Glen, 1992, 2005; Zen, 1996; Scheibner & Basden, 1998;

Scheibner & Veevers, 2000; Collins, 2002a, b; Crawford et al., 2003a; Cawood, 2005;

Cawood et al., 2009).

Accretionary orogens may start on continental boundaries and grow by crustal thickening and addition of continental fragments, oceanic crust and island arcs through subduction- related processes (Gray & Foster, 2004).

Gray & Foster (2004) summarise some widely accepted facts about the LO:

 The LO started on the eastern edge of the Gondwanan margin, above a continent-

dipping subduction zone, and grew by accretion of submarine sedimentary

terranes, island arcs, oceanic crust and continental fragments (Fergusson &

Coney, 1992a; Gray, 1997; Foster et al., 1999; Cayley, 2012) from 520 Ma to

340 Ma.

 Peak deformation was late Ordovician to Silurian (Gray & Foster, 2004). The LO

can be divided into zones on the basis of structure and deformation patterns (Gray,

1997; Fergusson, 2003; Spaggiari et al., 2003b): the Western Lachlan zone,

Central Lachlan and the Eastern Lachlan zone (Fig 2.1).

o The western LO is dominated by an east-vergent thrust system with

alternating zones of northwest- and north-trending structures (Cox et al.,

1991a; Gray & Willman, 1991a, b). Faults in this zone show east-younging

(450-395 Ma).

o The central LO is characterised by northwest- trending structures and

consists of a southwest-vergent thrust-belt linked to a fault-bounded

metamorphic complex, the Wagga Metamorphic Belt (Fergusson, 1987a;

Morand & Gray, 1991). The age of deformation is poorly constrained.

o The eastern LO is dominated by a north–south structural grain and east-

directed thrusting which caused inversion of extensional basins in the west

(Glen, 1992). Deformation in this zone (400-380 Ma) is younger than the

other zones of the LO (Gray & Foster, 2004) with the exception of the

Narooma Complex (445-440 Ma).

o The Governor Fault represents the boundary between the western and

central LO (VandenBerg et al., 1995, 2000) whilst the GFZ represents the

boundary between the central zone and the eastern zone.

 Much of the western zone is under cover and evidence of basin inversion of

extensional basins and distributed shear strain (Silurian Tumut; Siluro-Devonian

Hill End and Cowra troughs and other Devonian basins etc.) is restricted to the

central and eastern LO (Glen, 1992; Gray, 1997; Gray & Gregory, 2003; Spaggiari

et al. 2003b).

Figure 2.1: Simplified structural map showing the subdivision of the LO into the western, central and eastern zones. Major faults (thick lines) have been identified (1-21) as well as aeromagnetic trend lines (thin lines). Note the location of the GFZ (no.12, circled in green) as the fault delineating the central from the eastern zone. The location of the WMB and Narooma accretionary complex have also been illustrated. Modified from Gray & Foster, 2004.

In order to engage in the debate surrounding the nature of the LO, some key aspects need to be introduced:

2.1.1 Orogenic Events

Orogenic events involve large-scale collisional processes but are diachronous and vary in intensity and importance throughout the LO (Gray et al., 1997; Foster et al., 1999). They are often named after towns where type locations of unconformities are identified. Only those events recorded in the Macquarie Arc, will be discussed in this thesis. These events will be explained more in Chapter 6. The resultant variable effects of shortening and localised extension are, in part, responsible for the formation of large-scale structures such as the GFZ which is examined in this study (Fig 1.1).

The LO (with focus on the Macquarie Arc) contains evidence of at least three thin-skinned shortening events (Gray et al., 1997; Foster et al., 1998; Foster et al., 1999) with intervening periods of extension: the Benambran (Ordovician), Tabberabberan (mid-

Devonian) and the Kanimblan (Early Carboniferous). The Benambran orogeny is often considered the main cratonisation event in the LO (Cayley, 2011). The Tabberabberan and Kanimblan (overlaps with the Hunter-Bowen orogeny) saw the dismemberment of the Macquarie Arc into four separate belts, inversion of intervening transtensional basins and the emplacement of I- and S-type granites (Glen, 2005). In addition to these four orogenies, the Silurian Bindian orogeny (400 Ma) is also considered by some authors to be an important event. Morand & Gray (1991) suggest that the GFZ formed during this time in response to NNE-SSW shortening event and Cayley & Musgrave (2016) suggest that south-east directed slab rollback post-Bindian initiated oroclinal folding of the

Macquarie Arc.

2.1.2 The Macquarie Arc

The Macquarie Arc consists of high-K calc-alkaline to shoshonitic evolved basalt, basaltic andesite and andesite (Glen et al., 2007) and is the volcanic island arc that accreted onto the Gondwanan margin. This collision event is associated with the

Benambran Orogeny (late Ordovician, 440 Ma). The Macquarie Arc was dismembered in a Siluro- Devonian extension event into four volcanic belts: Junee–Narromine

Volcanic Belt; Molong Volcanic Belt; Rockley–Gulgong Volcanic Belt; and, Kiandra

Volcanic Belt (Fig 1.1; Glen et al., 1998; Glen, 2005). The Junee-Narromine volcanic belt is considered the core of the arc and contains sixteen smaller igneous complexes

(Narromine, Cowal etc.) which extend over >200 km (Percival et al., 2011).

2.1.3 The Gilmore Fault Zone

The GFZ marks a very distinct lithological contrast (Fig 2.2) between Silurian S-type granites, Ordovician metasediments of the WMB (in the central zone) and the Ordovician- early Silurian calc-alkaline Macquarie Arc in the east zone (Scheibner, 1985; Wormald

& Price, 1990). The arc volcanics immediately east of the GFZ are part of the Junee -

Narromine Volcanic Belt, which is mostly under cover (Crawford et al., 2007; Percival

& Glen, 2001) and occupy a series of basement highs and troughs, such as the late

Silurian- early Devonian Tumut Trough, which is adjacent to the GFZ (Stuart-Smith,

1990).

The contrast between the high to moderate magnetic intensity signature of the Macquarie

Arc and Tumut Trough, and the consistently low magnetic intensity of the WMB (Fig

2.3) clearly define the GFZ along its entire strike length on the 1:250 000 Cootamundra

and Forbes map sheets (Warren et al., 1995; Stuart-Smith, 1991; Suppel et al., 1986;

Wyatt et al., 1980).

Figure 2.2: Location map showing the five major geological units considered in the study area (after Glen et al., 2007). The three horizontal lines represent the cross-sections modelled herein with corresponding reflection seismic lines (red). The Gilmore Fault Zone in this image (Gilmore Suture as defined by Scheibner & Basden, 1998) has been traced as the eastern boundary of the Wagga Metamorphic Belt. The exact boundary is what has contributed to some of the confusion surrounding tectonic models of the area. Note the splay of the fault between Gidginbung and West Wyalong, which I redefine in this study as the Barmedman Fault. The map only shows surface geology and the Tumut Trough is only shown to outcrop far south of the cross-sections. The GFZ is the western flank of the Tumut trough thus I have modelled and extrapolated it below shallower units. 13

Between Gidginbung and West Wyalong, the GFZ as defined by Scheibner & Basden

(1998) splits into two splays (Fig 2.2), between which are the Upper Ordovician

Gidginbung Volcanics, a unit of the Macquarie Arc (Glen et al., 2007). The western splay, which locally forms the boundary between the high- and low-intensity magnetic belts, was interpreted as the principal trace of the GFZ during geological mapping of the

Cootamundra 1:250 000 geological sheet (Warren et al., 1995). Glen et al. (2002) identified the western splay as the GFZ, and interpreted it as the east-dipping western limit of an imbricated zone. Short wavelength TMI anomalies between the two faults splays suggest that the wedge of Gidginbung Volcanics they surround is relatively shallow, when compared to other Macquarie Arc source rocks east of the eastern splay, which are characterised by broad TMI anomalies.

The GFZ is marked by a high, positive gravity anomaly on its western side (Fig 2.4). The steepest gravity gradient appears just east of the GFZ, and the anomaly decreases much more slowly to the west. The source of this gravity anomaly, presumably corresponding to relatively dense volcanic rocks of the Macquarie Arc, thus appears to dip west.

The GFZ is one prominent feature whose poorly constrained role as a major thrust and/or strike-slip fault has the potential to inform the new tectonic models. Modelling subsurface geometry, along with overprinting relationships of recognised deep crustal structures, will allow the kinematic history of the fault, and the relationships between the central and the eastern LO to be ascertained. The GFZ forms the boundary between the central and eastern Lachlan provinces, extending for many hundreds of kilometres through New South Wales, Victoria and Tasmania. For this reason I have narrowed the study area and interpretations to include only the extent of the GFZ within the 1: 250 000

Cootamundra and Forbes map sheets (Fig 1.2). 14

Figure 2.3 (upper): TMI and Figure 2.4 (lower): Bouguer gravity grids. Both images show the location of the numbered cross-sections and towns. The major tectonostratigraphic units, faults and trends in the study area have been superimposed on the images. Some of the sampled lithologies sampled are noted: WMB=Wagga metamorphic belt. Gravity and magnetic grids, profiles and images were supplied by the Geological Survey of New South Wales, and were extracted from their state wide compilations as of 1 May 2015.

2.2 EXISTING TECTONIC MODELS

The current views on the formation of the LO with special emphasis on the eastern LO and what the GFZ represents in these different scenarios are outline below:

2.2.1 The Lachlan Orogen Formed By Suspect Terrane Accretion

The GFZ is considered by Scheibner (1985) to be a terrane boundary between a micro - continent that collided with the passive Gondwana margin. Episodic terrane accretion and dispersion coupled with subduction zone tectonics would have been responsible for many of the extensional features observed such as volcanic arcs, rifts and basins as well as the resulting position of the Tasmanides.

If the GFZ is a terrane boundary/crustal suture it should juxtapose crustal blocks of distinct, compositionally different rock packages, and there should be contrasting density and magnetic susceptibility of the mid-lower crust on either side of the GFZ. Furthermore if the GFZ is a convergent suture it should be indicated by a crustal-penetrating thrust fault.

2.2.2 The Lachlan Orogen Formed As An Extensional Accretionary Orogen

More recently it is has also been suggested that at least the eastern LO formed as an extensional accretionary orogen that evolved from growth and destruction of volcanic arc and back arc systems by slab rollback coupled with rapid (>10 million years) orogenic cycles from the mid-Cambrian to the Silurian (Fig 2.5; Collins, 2002a, b). In this model, growth occurs via off-scrapping and addition of crustal blocks transferred from the subducting slab, onto the Gondwanan margin.

This alternating pattern of long-term slab retreat and short-term subduction advance

(Accordion-style tectonics) has been linked with a tripartite association including the formation of HTLP regional metamorphic zones and S-type granites (Collins & Richards,

2008). The WMB and the tripartite association is considered indicative of tectonic switching. In this model the generation of S-type granites is a by-product of the Silurian

(435-430Ma) extension, during which time rift basins such as the Tumut Trough formed

(Cas, 1983; Zen, 1996). The southern extent of the GFZ corresponds to the western flank of the Tumut Trough. Thus the GFZ could have been formed as an extensional fault allowing the development of a depo-centre for a deep Silurian sedimentary pile (Tumut

Trough) on its eastern side. However, if there is evidence of shortening after the Silurian extension then the fault might have been reactivated as a thrust and/or re-oriented such that it might be difficult to identify it as a normal fault.

Figure 2.5: Model showing tectonic switching present in extensional accretionary orogen model (modified from Collins, 2002a, b). A, B- Slab retreat induces regional extension and formation of backarc; C- incoming oceanic plateau induces flat slab subduction and local thickening and short lived orogenic belt; D- extension mode re- established. V.F.=volcanic front; D.F.=distal arc flank; SL=sea level.

2.2.3 The Lachlan Orogen Was Subjected To Subduction Flip/ Polarity Reversal

Aitchison & Buckman (2012) also suggest LO growth as an accretionary orogen but instead of ‘accordion-type’ growth specify ‘quantum addition’ of juvenile, island-arc material whereby collision with the continent margin initiated shortening of the back arc accompanied by subduction polarity flip (Fig 2.6). They use the distribution of turbidites as evidence for an east-dipping subduction zone, east dipping WMB (therefore east dipping GFZ) under the Macquarie Arc. They also attribute the formation of S-type granites to crustal thickening as opposed to during extension (Collins & Richards, 2008).

Figure 2.6: Image contrasting the accretionary orogen model (left hand side) with the subduction flip/ quantum tectonic model suggested by Aitchison & Buckman (2012; right hand side). On the left hand side. The subduction flip model suggests an arc continent collision in which the Macquarie Arc transfers to the Gondwanan margin via a subduction flip as a result of stalled subduction channel. Image modified from Aitchison & Buckman (2012) and not to scale.

2.2.4 The Lachlan Orogen Achieved Its Current Form By Oroclinal Bending

The idea of tectonic escape from subduction rollback through formation of oroclines

(Musgrave & Rawlinson, 2010; Cayley, 2012; Musgrave, 2015) or curvature of orogens

(Moresi et al., 2014) provides a much simpler view of the Tasmanides evolution and

accounts for many of the aeromagnetic and petrographic trends visible. The orocline model suggests the Macquarie Arc was the only arc and associated subduction zone active during Lachlan Orogen formation during the Ordovician. According to the orocline model the WMB is in the south-moving core of the orocline (Fig 2.7). If so, deformation in the

WMB should be extensional or transtensional and the high-temperature metamorphism should be synchronous with development of the GFZ. The GFZ should have a major component of syn-kinematic high-metamorphic grade rocks on its western margin. This increased metamorphic grade would be reflected in increased density and should be revealed during gravity modelling.

Transtensional motion would be recorded in steeply dipping, crust penetrating faults. The

2.5D cross-section models cannot account for lateral displacement but in order for this large extension to occur there would need to be steep, vertical structures and vertically plunging bodies, visible in cross-section.

Figure 2.7: The Lachlan orocline hypothesis (after Moresi et al., 2014; and Cayley and Musgrave, 2016). Clockwise oroclinal folding brought on by collision of the Selwyn block.

2.2.5 Extrusion of the WMB:

Alternatively, the GFZ is thought to be a steeply west-dipping strike-slip fault (Stuart-

Smith, 1990a, et al., 1992) that formed in response to the compression and south-easterly movement of the WMB, as a tectonic wedge (Fig 2.8), over the Macquarie Arc in the late

Silurian to early Devonian (~ 420–400 Ma; Morand & Gray, 1991; Glen et al., 1992;

Foster & Gray, 2000).

Figure 2.8: Extrusion model (Morand & Gray, 1991). Model showing Late Silurian-Early Devonian movement of WMB in a south- southeast direction relative to adjacent crustal blocks. The southeast moving wedge is recorded in bounding faults: sinistral motion on the Gilmore Fault (GF), dextral motion on

the Kiewa Fault (KF) and dip-slip reverse motion on the Indi Fault (IF). MWFZ = Mt Wellington Fault Zone.

CHAPTER 3: METHODS AND DATA ACQUISITION

The following methods and interpretation of cross-sections 1 and 2 have been published in Exploration Geophysics as a case study (Venkataramani et al., 2017). The third cross- section is also discussed herein.

The lack of outcrop and initial uncertainty between competing tectonic and structural interpretations warrants the construction of a geologically constrained inversion of magnetic and gravity data to yield a revised model of the GFZ. Three reference 2.5D cross-section models were constructed on the basis of mapped surface geology, Total

Magnetic Intensity (TMI) and gravity extracted from compilation grids. The first two profiles were also constrained by pre-existing reflection seismic sections (Fig 2.2). Where possible, petrophysical properties (magnetic susceptibility, density and Koenigsberger ratios) were extracted from fresh rock samples and used to constrain model parameters.

The first two profiles correspond to reflection seismic lines 99AGSL1, 2, and 3 and extend for more than 80 km. They were simultaneously inverted by iterative forward modelling of body geometry and petrophysical property contrasts. I modelled a third profile further north in an attempt to identify the continuation of the GFZ, but this model did not have the same degree of constraint on the mid-crust as the two more southerly profiles as there were no available seismic lines. I followed the Type III (lithologic) inversion approach of Oldenburg & Pratt (2007). The inverted models form the basis for further interpretation of the evolution of the Lachlan Orogen (Chapter 4).

Two attempts were made to collect rover TMI and gravity measurements in the field.

Unfortunately the magnetometer data were severely affected by noise, and the gravity data traverse did not extend far enough to define the major structures. Therefore, I used grid derived data as the basis of the potential field models. The ground gravity data did 21

confirm that the grid interpolation of the gravity was a reliable representation (Fig 3.6 a,b).

3.1 PREVIOUS WORK

Direen et al. (2001) produced joint gravity and magnetic models along the same reflection seismic profiles examined in the present study. While these models provided useful insights into the geological interpretation of the first-order architecture of the LO (Glen et al., 2002), the correspondence between the model responses and the magnetic and gravity data is poor in places, particularly on the flanks of the prominent magnetic anomalies that mark the GFZ (Fig 3.1a, b). The gradient at the flanks of anomalies is strongly controlled by the dip of the geological source (Foss, 2002). I have produced revised potential field models that significantly reduce these misfits and therefore better constrain the dip of bodies, while at the same time incorporating new geological constraints from the East Riverina mapping program of the Geological Survey of New

South Wales.

3.2 REFLECTION SEISMIC DATA

Data from seismic reflection surveys conducted by the Australian Geodynamics

Cooperative Research Centre (AGCRC) and the New South Wales Department of

Mineral Resources (NSWDMR) in 1999 (lines 99AGSL1, 99AGSL2, 99AGSL3) are used in the deep crustal interpretation. However, my modelled cross-sections are based on longer profile lines (>80 km) that are slightly offset from the seismic profiles in order to best target observed magnetic and gravity anomalies. The geological structure imaged by the three seismic lines has been re-interpreted by Glen et al. (2002), but without additional work to address the misfits in the existing potential field models around the

Gilmore Fault Zone, the precise geometry and tectonic role of this particular structure remains ambiguous (Fig 3.1a, b).

Figure 3.1 a, b: Composite image of reflection seismic line 99AGSL1, L2, L3 and corresponding geological model of Direen et al. (2001). Figure 3.1 (a) corresponds to my cross-section 1 whilst Figure 3.1 (b) corresponds to my cross section 2. The red boxes (1-7) highlight mismatches between geology, observed and calculated signals and are compared with the responses to the revised model shown as boxes (1-7) in Figure 4.3 – 4.4. Major features visible in the seismic line (A-E) were used as a basis for modelling. CDP = common depth point. For this study, two-way travel times were converted to depth based on an assumed average crustal velocity of 6000 m.s-1 (Direen et al., 2001).

3.3 GRIDDED POTENTIAL FIELD DATA

Given that all three profiles extend for more than 80 km and model features extending more than 20 km into the subsurface, I considered directly fitting aeromagnetic line data to represent an unnecessary oversampling, made more complex by the continuation of the profiles over two separate surveys with differing ground clearance. Any loss of resolution or introduction of artefacts into the grid produced from these surveys should not exceed the line spacing of the data (mostly  250m) and so have little influence at the scale of the modelling. I have therefore followed Direen et al. (2001) in fitting aeromagnetic data derived from grid compilations. Magnetic and gravity grids, profiles and images were supplied by GSNSW and were extracted from state wide compilations as of 1 May 2015

(Fig 2.3-2.4).

TMI data were sampled at 50m spacing from the GSNSW state-wide TMI grid locally merged from the Cootamundra and Forbes 1:250 000 sheet aeromagnetic acquisition programs conducted by Geoscience Australia (GA) and NSWDMR in 1993. These programs were conducted using 200-400 m spaced east-west flight lines and a terrain clearance of 80-100 m. The same aeromagnetic data produced the grids that were sampled for the model profiles in Direen et al. (2001). Gravity was sampled from the GA

540-m compilation grid of GSNSW gravity data. The GA grid locally incorporates a mix of data: 300-m spaced line data along the three seismic lines; the 2013 Riverina survey with a 1 km station spacing which extended over all but the westernmost 10 km of cross- section 1; and a mix of 3 km to 15 km spaced point data over cross-section 2.

There are many narrow, near surface bodies in all three cross-sections. Some of these bodies have a horizontal extent less than 1 km, which is less than the smallest grid spacing in the above mentioned grids. The geometry of these bodies (smaller than the

gravity spacing) were controlled only by magnetic data. Again I state that the resolution of these grids is more than sufficient for modelling both the long and shorter wavelength features along the selected profiles.

Data was sampled in ModelVision™ at 50 m which represents grid spacing for TMI but is an oversampling of the gravity, acquired over the area with a station spacing ranging between 1 km and 15 km. I modelling both long and short wavelength anomalies but did not attempt to model detailed geometries at depths or spacings in the upper 250 m of the profiles recognising the potential loss of fidelity in modelling lengths shorter than the line spacing.

I recognise the potential dangers of using interpolated data. To verify that no significant short wavelength features were processed out, I collected ground data (magnetics and gravity) at shorter intervals (25m and 100m respectively) than the sampled grid data, over profiles representing the central 25 km of cross-section 1, and compared these with the sampled grid profiles (Figs 3.3 and 3.6a,b).

3.4 COLLECTED TMI AND MAGNETIC SUSCEPTIBILTY

A Geometrics proton precession (G856AX) rover magnetometer was used to collect magnetic data at 25m spacing (Fig 3.2, 3.3). A Geometrics Cesium vapour (G858) magnetometer was set up at a permanent base station, with measurements recorded every

10 seconds, and used in the diurnal correction of the rover data. Rover data was collected on roads east and west of the town of Barmedman and thus corresponded to a portion of the line modelled as cross-section 1 (Fig 3.4).

3.4.1 Magnetic/Solar Storms And CMEs

Two separate attempts were made to collect magnetic data in the field. The first trip occurred from 18-20 March 2015. This corresponded with the 2015 St Patricks Day Solar 25

storm. This is considered the worst magnetic storm in the last decade and in this current solar cycle. The storm reached a G4 (severe) level on NOAA’s geomagnetic storm scale and registered a Kp index (indicator of global geomagnetic storm activity) between 6 to

8 (out of 9 which is extreme geomagnetic storming). The rover and base magnetometers recorded off the scale spikes and fluctuations. The data could not be processed effectively and thus had to be discarded.

The second trip (6-11 October 2015) coincided with a CME (Coronal mass ejection) and a CIR (co-rotating interaction region) that hit the Earth’s magnetic field. The highest recorded Kp index for this geomagnetic storm was 7. Both rover and base station magnetometers recorded large scale fluctuations and spikes but an attempt was made to salvage and process the data. Prediction of solar weather is not an exact science and warnings are often only issued within a day or two of the actual event.

Figure 3.2: Multitrack profile of six lines of collected magnetic field data (1;1_;1_2;1_3;1_4;1_5). The ‘curves’ are displayed perpendicular to the corresponding line from which they were measured. Many of these lines exhibit data spikes even after despiking and diurnal corrections and were discarded.

Figure 3.3: shows two lines (1_1 and 1_3) that did not appear to be largely affected by solar storm activity. These are superimposed on the corresponding portion of the line modelled as cross-section 1(red) which is extracted from the TMI grid. Note that the collected data is plotted on a secondary axis. The offset in the two scales results from a shift in the regional field made during compilation and stitching of the grid. Amplitudes of the grid anomalies are smaller, and their wavelengths wider, than the corresponding ground collected anomalies, a result of upward continuation resulting from the 80–100 m ground clearance of the airborne data. Some inconsistency in the position of peaks is to be expected, given the irregular path of the ground profiles, which do not perfectly match the position of the grid profile. Note the good correspondence on the flanks of most of the peaks.

Figure 3.4: Map showing lines of rover magnetometer collected (25m spacing) along roads surrounding the town of Barmedman. Lines 1_1, 1_2 and 1_3 overlap a portion of modelled cross-section 1 and seismic line 99AGSL3. Lines 1_1 and 1_3 are compared to the grid-extracted TMI signal over the same interval (Figure 3.3).

3.4.2 Buckshot Gravel

A study by Clarke & Chenoweth (1995) suggests that ferruginous grains and fecal pellets of biogenic origin contribute to the high susceptibility in buckshot gravel found on road surfaces. These grains are a result of weathering, bush fires and bacterial break down of rock. I do not correct for the effect of magnetic buckshot gravel but cannot rule out its effect on the collected magnetic data (Plate 1a).

3.4.3 High Frequency Power Lines

I could not avoid passing under cables and in some instances walking parallel to high frequency powerlines. In almost all these cases the rover magnetometer recorded off the scale fluctuations. These were later edited out but as a result a few kilometres of data had to be discarded.

3.4.4 Problems With Processing/ Despiking Data

I experienced multiple problems in processing the collected magnetic data. The large scale errors were easily removed however there was great uncertainty regarding the validity of the data due to the October 2015 solar storms. I used high and low pass filters to view the data in Oasis montaj™ software however it is possible I may have ‘over- corrected’ the data by removing some actual anomalies as well as artefact data spikes (Fig

3.2, 3.3).

3.5 COLLECTED GRAVITY

From 8-10 October 2015 I collected ~17 km of gravity data along a single line at 100m station spacing using a Scintrex Autograv (CG5) gravimeter (Plate 1b). The line corresponded with parts of the same roads traversed in the above mentioned magnetic survey (Fig 3.5). 28

I also recorded position and elevation using a Leica RTK GPS (Plate 1a) which was used in the processing of data. Though utmost care was taken in levelling the gravimeter before every reading I cannot rule out the effect of wind and distant road traffic. I made multiple trips between the allocated base station and the AFGN (Australian fundamental gravity network) base station at Young Airport. The line of collected data closely corresponds to a portion of the profile modelled in cross-section 1(Fig 3.5, 3.6 a,b) between approximately 525000E – 545000E and along 6220000N.

Figure 3.5: Map showing line of collected Gravity data. Gravity stations were spaced 100 m apart and measured along roads east and west of the Barmedman.

This data was collected at a station spacing (100m) less than the station spacing for the gridded data (varies from 1- 15 km) in an attempt to assess the level of interpolation possibly present in the models. If modelled sections fall between the gridded data then some of the modelled anomalies might be a product of the ‘interpolation’ of the data and thus not real. 29

Figure 3.6 a (left): Stacked profile comparing the grid-interpolated gravity (blue) with the field collected gravity (orange). Note the goodness of fit on flanks of the anomalies. Figure 3.6 b (right): This graph shows the good fit and correspondence between grid- interpolated gravity (blue) and field collected gravity (orange). Note the good fit on the flanks of anomalies, between the gridded and collected data, particularly between 541000 – 543000E. The fact that the interpolated grid data is a good representation of the observed (collected) data adds confidence to the 2.5 D forward modelled cross sections.

The data was processed and corrected for instrument calibration, atmospheric effects, drift and the free-air and Bouguer corrections, assuming a Bouguer density of 2670 kg/m3 , the same density used in Bouguer correction of the grid data. Tidal corrections were made by gravity meter using an inbuilt algorithm. The corrected gravity data were tied to the

AFGN via the base station to produce absolute gravity for comparison to the grid gravity.

The collected gravity did not appear to vary significantly from the grid data. The data recorded a few shorter wavelength features, which is to be expected as more data is captured over a shorter station spacing, but exhibited the same overall long wavelength trend. This verifies that there are virtually no ‘large artefact anomalies’ modelled as part of the study.

3.6 PETROPHYSICAL PROPERTY DATA

The petrophysical properties of magnetic susceptibility, saturated density and

Koenigsberger ratios (Koenigsberger, 1938) were measured for selected rocks in the laboratory. As outcrop was limited (Fig 3.7), hand samples collected in the field were supplemented by samples taken from two diamond drill cores from the core library of the

Geological Survey of NSW. Specimens were prepared as 2.5 cm diameter cylinders with a water-cooled diamond coring bit mounted in a drill press (Plate 1c- i). Magnetic susceptibility of specimens was obtained using a Bartington MS-2 meter with a well sensor, operating at the low-frequency setting (465 Hz). Specimens were vacuum- saturated with water, and the saturated density calculated using Archimedes’ principle.

Lastly, specimens were analysed in a Molspin Minispin spinner magnetometer for natural remanent magnetisation (NRM). Koenigsberger ratios were then calculated following

Equation 1.

−7 푄 = 푁푅푀×4휋 ×10 (1) 퐵푅푒푔×퐾푉표푙 where Q = Koenigsberger ratio

NRM = Natural Remanent Magnetisation intensity, A/m

BReg = Regional total magnetic intensity, T

ΚVol = SI magnetic susceptibility, normalised for volume

-5 Note: NRM is usually recorded in mA/m, BReg in nT, and KVol with a multiplier of 10 . These factors are allowed for in the calculation.

Figure 3.7: Map showing location of samples collected in the field. Only 7 locations are listed but more than one sample specimen was taken at each location (refer to Table 1a). A total of 38 samples were taken but none were ‘oriented’. WAG= Wagga Group, YID= Yiddah Formation, TEM= Temora Volcanics, BXH= Bronxhome Formation.

Plate a: Ferruginous grains attached to magnetic base plate of a Leica DGPS. The GPS was operated in RTK (real time kinematic) mode, using the Leica Geosystems SmartNet network service linking via a mobile phone. Plate b: Scintrex Autograv (CG5) gravimeter (levelled) and GPS at base station. GPS measurements used in Gravity processing. Plate c-i: selection of sampled rocks cut to size and used in petrophysical property measurements. c= Temora Volcanics; d= Bronxhome Formation and 1 sample of foliated Barmedman granite; e, f, i= Wagga turbidites; g & h= Yiddah Formation.

3.6.1 Petrophysical Properties Analysis

A total of 38 oriented hand specimens (Fig 3.7) were taken from the Wagga Group turbidites (Ordovician turbidites), Temora Volcanics (late Ordovician to early Silurian intermediate volcanics of the Macquarie Arc) and Bronxhome Formation (late Ordovician to early Silurian sandstones, siltstones and shales) and Yiddah Formation (Silurian to early Devonian quartz-rich sandstones, siltstones and conglomerates). Unweathered outcrop was rare and where possible samples were taken from diamond drill core found at the GSNSW core library. Fourteen unoriented samples were taken from two available cores drilled in the mineralised diorite, andesite and porphyry of the Gidginbung

Volcanics (early Silurian Macquarie Arc rocks) in the Barmedman and Yiddah districts

(Table 1a, b).

The calculated Koenigsberger ratios fall in the range 0.01 ≤ Q ≤ 9.41, Table 1a and b, excluding 5 samples from the quartz-rich Wagga Group turbidites and Yiddah Formation which have negative magnetic susceptibility, a consequence of the diamagnetic (and hence negative) contribution of the quartz in these rocks exceeding the positive susceptibility contributed by the small proportions of magnetite and other ferromagnetic and antiferromagnetic minerals present. Only 16 of the samples have Q ≥ 1, a condition where remanent magnetisation would dominate the resulting anomalies. The lower Q values of the remaining 31 samples imply that the resultant magnetisation of sources of anomalies is dominated by the induced magnetisation, and hence is close to parallel to the present field. Only one of the sedimentary or granitic samples has a magnetic susceptibility > 10  10-5 SI or NRM intensity > 10 mA/m, so none of these units significantly impact the magnetic model.

Table 1a: Table of rock samples and corresponding petrophysical property data.

Saturated Density (kg/m Density Saturated

Geographic coordinates: coordinates: Geographic coordinates: Geographic

Magnetic Susceptibility Susceptibility Magnetic (Q) ratio Koenigsberger

Sample/ Drill hole code hole Drill Sample/

Sample Volume (cm Volume Sample

Declination ( ° ) Declination

Inclination ( ° ) Inclination

NRM (mA/m) NRM

(10

Package

) SI

)

Barmedman granite: Devonian FG 16.93 0.25 0.03 11.22 2650 YID 5.82 0.86 0.32 11.16 2560 YID 5.77 3.24 1.23 11.27 2550 YID 2.62 0.28 0.23 11.45 2580 YID 4.15 1.24 0.65 10.84 2610 YID 0.47 0.37 1.75 10.7 2530 Yiddah Formation: YID 9.34 2.27 0.53 10.71 2530 Silurian- early negative Devonian YID -1.41 0.56 value 10.66 2450 sediments and YID 0.94 4.04 9.41 10.69 2490 volcanics YID 0.49 0.33 1.48 10.15 2440 YID 9.68 2.11 0.47 10.33 2430 YID 4.73 1.5 0.69 10.57 2430 YID 3.63 2.72 1.63 9.65 2370 YID 5.37 2.2 0.89 10.24 2450 112.0 TEM 7 4.96 0.1 R 12.18 2950 055.5 085.1 Temora 931.2 Volcanics: TEM 5 238.9 0.56 N 11.2 2930 329.5 -003.5 late 425.3 216.8 Ordovician- TEM 7 2 1.11 R 12.06 2890 194.7 065.8 early Silurian 283.0 TEM 2 178.5 1.37 R 11.66 2910 259.1 041.9 BXH 5.15 0.11 0.05 11.64 2410 Bronxhome Formation: BXH 6.16 2.63 0.93 11.36 2430 Ordovician- BXH 6.36 3.55 1.22 10.22 1800 early Silurian BXH 24.01 18.91 1.72 7.08 3920 MMRH1001 11.9 0.14 0.03 N 10.08 2610 MMRH1001 21.7 0.9 0.09 N 11.06 2610 MMRH1001 17.33 0.24 0.03 R 12.12 2580 Gidginbung MMRH1001 41.77 16.05 0.83 N 11.97 2640 Volcanics: 5141. 975.3 Ordovician- MMRH1001 19 2 0.41 R 11.97 2810 early Silurian 1431. MMRH1001 27 5.9 0.01 R 13.24 2860 MMRH1001 39.11 0.6 0.03 N 8.95 2670 MMRH1001 31.75 0.73 0.05 R 10.71 2670

MMRH1001 12.31 0.36 0.06 R 11.37 2770 2834. YDH09 5927 22 1.04 R 8.63 2810 1999. YDH09 07 125.6 0.14 R 10.74 3080 1123. 106.9 YDH09 12 9 0.21 R 10.64 2800 YDH09 43.84 0.77 0.04 R 10.72 2780 133.9 YDH09 9 75.82 1.23 R 11.12 2760 WAG 0 5.48 undefined 10.82 2530 WAG 0 2.18 undefined 11.06 2550 WAG 3.7 1.72 1.01 10.82 2540 WAG 0.45 1.26 6.13 11.14 2540 WAG 5.52 1.92 0.76 10.87 2540 WAG 8.1 0.71 0.19 10.49 2660 Wagga WAG 3.46 0.25 0.16 10.11 2560 Group: WAG 3.54 0.59 0.36 9.9 2620 Ordovician WAG 4.94 0.59 0.26 8.1 2880 metasediment s/ turbidites WAG 0.9 0.36 0.88 11.13 2570 negative WAG -0.93 2.18 value 10.77 2470 WAG 2.43 2.71 2.43 6.18 3050 WAG 4.03 2.1 1.13 11.17 2480 WAG 0 1.25 undefined 11.06 2500 WAG 1.76 1.92 2.37 5.68 3910 WAG 1.02 0.12 0.26 9.76 2500

Notes: * YDH09 and MMRH1001 are diamond drill core samples (GSNSW Londonderry core library). All calculations are volume corrected. Polarity of volcanic samples indicated in Koenigsberger ratio column: N=Normal, R=Reversed.

Table 1b: Table of Mean and Standard deviation values of collected petrophysical properties (from Table 1a).

Sample/ Drill hole hole Drill Sample/

Saturated Density Density Saturated

Koenigsberger Koenigsberger

Susceptibility Susceptibility (mA/m) NRM

Magnetic Magnetic

ratio (Q)ratio

(10

(kg/m

code

) SI

)

YID 3.969 3.355 1.671 1.209 1.607 2.511 2493.846 71.477 TEM 437.928 352.942 159.795 106.197 0.785 0.568 2920.000 25.820 BXH 10.420 9.075 6.300 8.532 0.980 0.701 2640.000 902.035 MMRH1001 749.814 1711.203 111.138 324.110 0.171 0.276 2691.111 98.672 YDH09 1845.404 2417.607 628.680 1233.854 0.532 0.558 2846.000 132.212 WAG 2.433 2.479 1.584 1.318 1.328 1.703 2681.250 362.158

3.6.2 Remanence

Of the sampled units, the dominant influences on the magnetic model are the two representatives of Macquarie Arc volcanic rocks, the Gidginbung Volcanics and the

Temora Volcanics. The Temora Volcanics outcrop along strike from the Gidginbung

Volcanics about 35 km south-southeast of cross-section 1. The polarity of NRM was determined for samples from these two more strongly magnetised units. The sign of the inclination of the NRM was used to classify the remanence as normal or reversed polarity.

Lack of orientation of the core precluded precise determination of remanence declination and inclination for the Gidginbung Volcanics. Samples from drill hole MMRH1001 show a mix of normal and reversed polarity. Samples from drill hole YDH09 are all reversed.

Samples from the Temora Volcanics were oriented, allowing a full vectorial definition of

NRM. NRM is reversed polarity in three of these samples and normal polarity in one.

The magnetisation direction of the normal sample was inverted to allow combination with the reversed polarity samples, and the resulting Fisher (1953) mean has a declination =

190.8°, inclination = 65.5°. The wide scatter of directions results in a very large α95 of

66°. In total, only 4 of the 18 samples of volcanic rocks combine reversed polarity with a

Q > 0.5. Given this observation, and the paucity of well-constrained oriented remanence data, I did not generally include remanence in the models. However, the anomaly due to one body (corresponding to the Temora Volcanics) was impossible to fit without applying reverse polarity remanence, and was well modelled with a Q ratio of 1.37, declination of

54.4° and inclination of 35.6°, corresponding to the measured polarity and Q of one of the four Temora Volcanics samples.

The measured saturated density and magnetic susceptibility values informed the modelled values (Table 2, Table 3, Table 4). Slightly different values were used to model the

Ordovician basement but the range of values still overlap. This is evidence of the complexity of the substrate to the Macquarie arc.

Table 2: Range of inverted petrophysical properties derived from cross-section 1. Magnetic Density susceptibility (SI) ( kg/m3) Late Devonian sediments 0 2670 - 2800 Late Devonian granites/intrusions 0.1 - 0.22 2750 - 3100 Late Silurian- Early Devonian sediments and volcaniclastics 0 - 0.09 2550 - 3000 Silurian granites/intrusions 0 2780 - 2820 Ordovician turbidites 0 - 0.03 2670 - 2870 Ordovician volcanics 0.005 - 0.07 2760 - 3000

Table 3: Range of inverted petrophysical properties derived from cross-section 2. Magnetic Density susceptibility (SI) (kg/m3) Late Devonian sediments 0.000001 - 0.04 2730 Late Devonian granites/intrusions 0.01 - 0.03 2750 - 2770 Late Silurian - Early Devonian sediments and volcanics 0 - 0.12 2640 - 2820 Silurian granites/intrusions 0.0001 2670 - 2700 Ordovician turbidites 0 - 0.009 2690 - 2770 Ordovician volcanics 0.0015 -0.11 2600- 2980 Late Cambrian - Early Ordovician MORB 0.00001 2700

Table 4: Range of inverted petrophysical properties derived from cross-section 3 .

Magnetic Density susceptibility (SI) (kg/m3) Late Devonian sediments 0.006 - 0.04 2670 - 2730 Late Silurian - Early Devonian sediments and volcanics 0.009 - 0.05 2660 - 2800 Silurian granites/intrusions 0.018 2760 Ordovician turbidites -0.0018 - 0.02 2760 - 2950 Ordovician volcanics 0.021 - 0.05 2680 - 2770 Late Cambrian - Early Ordovician basement 0.08 - 0.0804 2900 - 3020

3.7 Geologically Constrained Inversion By Iterative Forward Modelling

The GFZ is a geologically complex area for which the seismic lines provide the only independent control on the deep structure (up to 25 km) and where petrophysical constraints were limited to two drill holes and a few unweathered, unoriented samples.

As a consequence, direct inversions of the geometry, including Bayesian modification of a reference model (Bosch et al., 2006; Lane et al., 2007), were unlikely to converge efficiently.

Three profiles representing geological models of the crust surrounding the GFZ were inverted by parametric iterative forward potential field modelling of 2.5D cross sections

(Oldenburg & Pratt, 2007; Musgrave & Dick, 2017), using ModelVision™.

2.5D modelling entails modelling a 2D cross-section but also extending modelled bodies along strike in map view to fit observed potential field trends and geology (Fig 3.8). The geometry of bodies were constrained by the seismic reflection data.

Figure 3.8: A general overview of 2.5D modelling. A single line of data can be extracted (eg. from the TMI) and modelled along strike to match regional features but also in cross-sectional view (green body).

While my methodology was similar to that previously followed by Direen et al. (2001), I sought to improve the closeness of the fit of the model response. Strong emphasis was placed on modelling the gradient on the flanks of anomalies because these are most sensitive to the dip direction of the source of the anomalies. Modelled bodies also had to be geologically justified. A background susceptibility of 0 SI and a background density of 2670 kg/m3 (the average crustal density used in the Bouguer correction of the data) were employed. Certain structures, like the GFZ, have different orientations in different tectonic conceptual models. I modelled these different possibilities and used a sensitivity test (RMS) to pick the most likely models (Appendix B). Root mean square error (RMS) is the difference between the model response and data, summed over all interpolated data points (Table 5).

Direen et al. (2001) assigned a density of 2850 kg/m3 and a susceptibility of 0 SI to the middle crust. I instead model deeply penetrating bodies but found very little model response below 25 km. Therefore I do not explicitly model the middle crust.

I modelled the cross-sections across three separate areas of interest along the GFZ (Figs

2.2 – 2.4). TMI and gravity profiles were extracted from gridded base maps and a specific regional field (TMI at 57656 nT, and gravity at -300 µm.s-2 relative to the GA grid) was chosen. For the purpose of modelling, the region fields were chosen on long intervals of low, relatively constant field, and in the case of gravity, away from the high gravity anomalies of the Macquarie Arc. Applying a constant value to the regional field, as opposed to a quadratic or higher order function, better enables modelling of deeper bodies and avoids aliasing caused by artificial regional field curvature.

Unlike the previous two cross-sections, cross-section 3 had no corresponding reflection seismic line and thus no constraint on structure of the mid crust. No previous studies

modelling this area have been found either. The same gravity and magnetic grids were used to model all three cross-sections. I decided on modelling this line further north of the previous two profiles as the northward continuation of the GFZ in map view was unclear. The trace of the GFZ could have included one of many inferred faults that varied according to author and different scale map sheets.

Table 5: Model sensitivity analysis. Dip direction RMS

Barmedman Magnetics Gravity GFZ fault (nT) (µms-2) west east 1.438 4.32

west west 1.811 4.42 Cross-section 1 east west 1.647 4.90 east east 1.737 4.90 west east 2.021 3.49 west west 2.125 3.78 Cross-section 2 east west 2.200 3.91 east east 2.146 3.61 Cross-section 3 west 1.712 4.10

Table 5: Model sensitivity analysis showing RMS values for different dip directions. The lowest RMS values correspond to preferred models (in bold). Different tectonic models have been discussed in Chapter 2- for example, I modelled an east-dipping GFZ (with the Barmedman fault dipping in both directions) to try to replicate the Aitchison & Buckman (2012) model. In this model, the accretion of the Macquarie Arc plugs the subduction channel causing a subduction flip to east-dipping subduction which would be reflected as an east-dipping GFZ. However from the results, the best fit achieved is with a west-dipping GFZ and an east-dipping Barmedman fault which would suggest a different tectonic model for this particular area. The Barmedman fault was also taken into consideration during the modelling process as this flake has often been interpreted as part of the GFZ thus contributing to the confusion in regional tectonic studies. In all models the Barmedman fault was best modelled as a shallow fault flake (up to 3.5 km deep). See Appendix B for figures relating to these RMS values.

CHAPTER 4: GEOPHYSICAL MODELLING

RESULTS AND GEOLOGICAL INTERPRETATION

4.1 REFLECTION SEISMIC INTERPRETATION The reflection seismic lines 99AGSL1, L2, L3 have been fully interpreted and modelled by Direen et al. (2001) and Glen et al. (2002). Figures 3.1a,b show a compilation of reflection seismic lines 99AGSL1, L2, L3 and the corresponding Direen et al. (2001) models. Many of the body boundaries in my starting models were based on features recognisable in the seismic images (Figure 3.1a, b).

The most dominant feature (in lines 99AGSL2, 99AGSL3) is the west-dipping belt of strong seismic reflectors (labelled A in Figure 3.1a, b). Strong reflector packages like this are commonly interpreted as volcanic or volcano-sedimentary packages (Direen,

1998) thus are likely to represent the Macquarie Arc, and the westwards dip of the seismic reflectors is consistent with this package being the source of the gravity anomaly that decreases gradually westward.

I modelled shallow weakly reflective packages with highly reflective bases as granites

(labelled B). 99AGSL2 shows distinct east-dipping reflectors within the Wagga

Metamorphic Belt (labelled C). These were interpreted as fault bounded deformed granites visible in the TMI image (Fig 2.2) as elongated ovals with very low magnetic relief.

99AGSL1 (near CDP 3800) and 99AGSL3 (near CDP 6000) show a shallow west dipping body with repeated parallel reflectors suggesting bedding within a shallow sedimentary basin (labelled D).

4.2 GEOLOGICAL INTERPRETATION

I created a simplified geology map of the area overlying cross-sections 1 and 2 (Fig 4.1).

This was difficult as a most of the Forbes and Cootamundra maps show extensive

Tertiary/Cainozoic and fluvial cover sequences. I drew my interpretation from a range of geology and metallogenic map sheets such as: Macquarie 1: 500 000 (Brunker &

Offenburg., 1970); Cootamundra 1:250 000 (Warren et al., 1996); Narromine 1:250 000

(Sherwin, 1997); Forbes 1:250 000 (Raymond et al., 2000); Nymagee 1:250 000

(Brunker, 1968); Cargelligo 1:250 000 (Meakin et al., 2006); Cootamundra 1:100 000

(Basden et al., 1975), Wyalong 1:100 000 (Duggan & Lyons, 2000), Tumut 1:100 000

(Basden, 1990); Nymagee 1:100 000 (MacRae, 1988); and looking at trends in 1VD TMI,

TMI and Bouguer gravity grids. I also used the interpretation of Glen et al. (2002) as a starting point.

Geochronological data (U/Pb, Ar/Ar, K/Ar, Rb/Sr, Sm/Nd), available in the GSNSW

Geoscientific data warehouse online database, were used as a guide to check that the mapped units were within the expected age ranges.

A study conducted by M. Bell (in prep) at the University of Newcastle aims to unravel the kinematic history of the GFZ with emphasis on collecting new samples to analyse for age dates and to produce an updated geology map of the Riverina area. My study focuses on understanding the subsurface structure of the GFZ and using it to interpret the tectonic evolution of the LO. These two projects were part of a greater collaborative study thus it was decided that any further mapping might impinge on my colleague’s work. My geological interpretation (Fig 4.1) was based on geological maps (mentioned above) and used to constrain cross-sections 1 and 2. I have elected to include the portion of the map by Bell (in prep) that was used to constrain cross-section 3 (Fig 4.2).

In my interpretation I exclude cover sequences and units younger than late Devonian.

Understanding the geology in the area helped inform the possible modelled geometries such as whether there were folds, faults, unconformities, basins (of varying depth) and extrapolation of units to depth.

4.3 RESULTS OF INVERSION

Multiple iterations and refinements were made in an attempt to produce a geologically reasonable model, consistent with both mapped geology and tectonically admissible structures, which minimised misfit in both the gravity and magnetic responses. For those units where data were available, petrophysical measurements from outcrop or drill core within the area (Table 1a) were used to constrain the model (Tables 2, 3, 4).

Figures 4.3, 4.4 and 4.5 show the best-fit models and their resulting magnetic and gravity responses compared to data. Each model is repeated to show the density, magnetic susceptibility, and geological interpretation of each body. Residuals of gravity and TMI are shown with the same scale as the data, to illustrate the closeness of fit.

4.3.1 Cross-section 1

Cross section 1 (Fig 4.3) is 103 km long and roughly in the same location as reflection seismic line 99AGSL3. The combination of magnetic low and gravity high between 20

– 40 km is best modelled using dense west-dipping bodies at depth (labelled A on Figure

3b), with smaller contributions from two shallow bodies underlain by an eastward tapering wedge of a low magnetic susceptibility unit. The west-dipping bodies have a density and magnetic susceptibility range (Table 2) characteristic of the Macquarie Arc, whilst the petrophysical properties of the eastward tapering wedge are akin to those of turbidites.

A distinct boundary (labelled E on Figure 3a,b) can be drawn above these Macquarie Arc bodies (west to east) but this terminates at the base of a series of east-dipping tooth-shaped bodies at ± 5 km depth. These tooth-shaped wedges have petrophysical properties which fall into two sets, a higher susceptibility set which match Macquarie Arc volcanic rocks, and a low susceptibility set which matches the Siluro-Devonian sedimentary rocks of the

Yiddah Formation. The western boundary of this series of wedges (at about 40 km along the profile) corresponds to the location of the western Gidginbung–West Wyalong fault splay, which is linked to an inflection point in the magnetics. The shallow extent of the base of this series of wedges verifies the interpretation of the TMI anomalies that the western splay fault is shallower than the rest of the GFZ. Given this observation, which reduced the significance of this fault from defining the GFZ to being a secondary, antithetic fault that can only be observed around the latitude of Barmedman, I propose renaming this local feature as the Barmedman Fault. This distinction should help reduce the confusion surrounding the dip direction of the GFZ.

Figure 4.1: Geology map of study area over cross-section 1 and 2. Reference colours used in this map are also used in Figures 4.3 and 4.4.

Figure 4.2: Map interpretation by Bell (in prep) superimposed on 1VD of the TMI. Note that this interpretation further classifies geological ages into early, middle and late Silurian etc. This map shows the location of cross-section 3, north of cross-section 2, and has been used to inform cross- section 3. The reference colours for this image have thus been used in Figure 4.5.

Figure 4.3: Composite image of modelled cross-section 1 showing modelled and observed TMI, gravity and residual tracks and corresponding cross-sections coloured according to density (top, kg/m3), magnetic susceptibility (middle, SI) and simplified geological interpretation (bottom). Note that the unmodelled, white spaces correspond to areas without petrophysical contrast equalling 2670 kg/m3(density) and 0SI(magnetic susceptibility).This roughly corresponds to the location of reflection seismic line 99AGSL3. In the Geological interpretation, the GFZ is shown as a solid black line, the Barmedman fault is a thin dashed line near the surface and the Benambran Orogeny boundary is the thick dashed black. Note that the Barmedman Fault is a shallow east dipping fault whilst the GFZ is a steep west dipping fault. Boxes1-3 show improved fit between observed and calculated signals. V=H no vertical exaggeration. Reference found in Figure 4.1.

The Barmedman Fault is truncated by a distinct west-dipping contact that separates this part of the Macquarie Arc from what I interpret to be a basin of late Devonian sedimentary rocks and granites overlying late Silurian- early Devonian volcanics and volcaniclastics of the Tumut Trough occupying the deep half-graben shaped wedge extending between about 50 and 77 km profile distance. This is the local expression of the GFZ. From west to east this contact corresponds to an inflection point and drop in magnitude of the gravity signal and the eastern edge of a sharp peak in the TMI (Fig 4.3). The Macquarie Arc volcanics at the surface seem to be the main contributor to the increased TMI.

There was no geometric constraint on the half-graben wedge from the seismic sections, thus the Barmedman fault, GFZ and surrounding bodies were modelled dipping in alternating directions as a test of the sensitivity of the model to the dip direction (Table

5). The lowest RMS values and best fit to the observed magnetic and gravity curves were achieved through a shallow, steeply east dipping fault (Barmedman fault) terminating on a deeper west more gently dipping fault (GFZ). My results indicate that the Tumut half- graben extends to a depth of approximately 20 km and is the major contributor to the long wavelength low gravity anomaly between 50 and 60 km profile distance.

The low in the TMI curve between 44-50 km profile distance and the inflection in the gravity curve at 52 km profile distance (Fig 4.3) were difficult to model without a corresponding steeply west-dipping wedge. Furthermore, without the west-dipping wedge, unreasonably high magnetic susceptibilities and densities had to be assigned to the late Silurian- early Devonian bodies.

At 25 km depth, bodies have been assigned a horizontal base for a reason that derives from observations in cross-section 2 (see section 4.3.2). Modelling below this depth, particularly of the base of bodies, had little effect on the calculated gravity and TMI

despite an approximate Curie depth of between 55-60 km (Chopping & Kennett, 2015).

The final RMS error for cross-section 1 is 1.438 nT (TMI) and 4.32 μms-2 (gravity) with the GFZ dipping from 75°E at the surface to 55°W at depth, which indicates a very good fit (see fig. 4.3).

4.3.2 Cross-section 2

Cross-section 2 (Fig 4.4, Table 3) is approximately 50 km north of cross-section 1, 90 km long and close to seismic lines 99AGSL1 and 99AGSL2 in the 1:250K Forbes map sheet

(Fig. 2.2. The geology in the area is more complex than in the area around Cross-section

1 with many faults and folded strata which require the use of numerous, steeply plunging modelled bodies. The magnetic lows associated with 3 of the major peaks in the TMI

(between ± 32-39 km profile distance, Fig 4.4) correspond to steep west-dipping contacts or faults. Steeply plunging bodies were used to model the short-wavelength, high- amplitude observed TMI between these faults. There is little geometric constraint from the seismic sections on the middle portion of 99AGSL1 and 99AGSL2 below 10km depth. Modelled bodies were differentiated according to their correlation to surface geology and range of expected petrophysical properties.

Cross-section 2 (Fig 4.4) is best modelled with the same shallow, steeply east-dipping

Barmedman fault terminating against the west-dipping GFZ (Table 5) as the final RMS of 2.021 nT (TMI) and 3.49 μms-2 (gravity) was the lowest of all modelled scenarios.

The model also includes the west-dipping boundary between turbidites and underlying

Macquarie Arc (labelled E on Fig 3.1 a). The GFZ dips 86°W at the surface and 55°W at depth. Furthermore the west dipping wedge on the eastern edge of the profile (3.5 km to 20 km depth) is visible as a body of 2700 kg/cm3 underlying bodies of a much higher density range (2733-2783 kg/cm3), (Fig 4.4). This west dipping wedge was used to model

the decrease in gravity and TMI from 81 – 90 km profile distance. Given the low sensitivity of the models to structures below 25 km and the west verging trend of the neighbouring bodies I have speculatively modelled this wedge as soling into a putative west dipping basal detachment fault at 20 km depth.

The west dipping wedge at the eastern end of Cross-section 2 (3.5 km to 20 km depth), if extended to the surface, would correspond to Cambro-Ordovician MORB (mid ocean ridge basalts), of the Coolac Serpentinite Belt similar to the middle crust interpretation of

Glen et al. (2002). This wedge was required to model a long wavelength decrease in gravity between 81- 90 km profile distance and was modelled using a low magnetic susceptibility (0.00001SI), and an even lower density (2700 kg/cm3) than Direen et al.

(2850 kg/cm3, 2001). Both my assigned physical properties and those given in Direen et al. (2001) and Glen et al. (2002) are lower than would be expected for both MORB and serpentinite (Jindalee Group equivalent in Direen et al. (2001), rather they are better matched by a crust of continental affinity. In assigning a basaltic composition to the middle crust below the Macquarie Arc Glen et al. (2002) fail to take into account this apparent inconsistency in petrophysical properties.

My work thus suggests that the Macquarie Arc does not sit on a simple, continuous oceanic basement. Instead, the basement in areas consists of continental crust, as indicated by low densities, (Fig 4.4) whilst in other areas there are slices of serpentinite and MORB. I suggest that the slices of serpentinites and MORBs are part of the old

Coolac Serpentinite basement to the Tumut Trough, rather than basement to the whole arc.

4.3.3 Cross-section 3

Cross-section 3 (Fig 4.5, Table 4) is ~95 km long, approximately 60 km north of cross- section 2 and found within the 1:250k Narromine map sheet. RMS values for the final model was 1.712 nT and 4.104 μms-2. As previously mentioned, there were no existing reflection seismic lines to provide constraint on the mid-crust. Considering this important lack of constraint, I decided to apply similar geometries to what was observed in the previous two cross-sections.

The Barmedman fault does not continue into cross-section 3 rather it terminates a few kilometres north of cross-section 2 though it is not possible to precisely determine the location, on the map without modelling the intervening area.

The Macquarie Arc contributes to the long wavelength gravity signal over the entire profile whilst the source of most of the high frequency, short wavelength magnetic anomalies are attributed to Siluro-Devonian rock units (Fig 4.5).

Like in cross-section 1 and 2, the Benambran boundary can be modelled as a section of

Macquarie Arc wedged under the WMB. The contact between the WMB and the underthrust Macquarie Arc is however much steeper than in the previous two models.

The high amplitude, short wavelength TMI anomalies between 9 and 16 km profile distance were best modelled as slices and wedges verging towards a central point (~ 13 km). When compared to surface geology, these correspond to WMB and Ordovician turbidites. These Ordovician turbidites dip towards a central point which corresponds to an outcropping, deep-seated slice of Macquarie Arc (Fig. 4.2). This slice of Macquarie

Arc (0.02 SI, 2770 kg/cm3) can be easily distinguished from the turbidites (0.004-0.017

SI, 2750 -2920 kg/cm3) based on expected magnetic susceptibilities (Table 4).

Figure 4.4: Composite image of modelled cross-section 2 showing modelled and observed TMI, gravity and residual tracks and corresponding cross-sections coloured according to density (top, kg/m3), magnetic susceptibility (middle, SI) and simplified geological interpretation (bottom). Note that the unmodelled, white spaces correspond to areas without petrophysical contrast equalling 2670 kg/m3(density) and 0SI(magnetic susceptibility). The position of this line roughly corresponds to the position and combined length of reflection seismic lines 99AGSL2 & 99AGSL1. As in cross-section 1 the GFZ is shown as a solid black line, the Barmedman fault is a thin dashed line near the surface and the Benambran Orogeny boundary is the thick dashed black. Note that the Barmedman fault is still a shallow east dipping fault and the GFZ is a steep west dipping fault. A decollement is clearly visible in this cross-section at 20 km depth. A single Cambro-ordovician body has been modelled as the similar aged Coolac-serpentinite belt appears very close to the end of cross-section 2 in map view. V=H,no vertical exaggeration. Reference found in Figure 4.1. 53

The best fit to the GFZ is still achieved with a west dipping wedge however it is much steeper than in the previous two cross-sections (dips ~89°W to 35°W at depth). The GFZ still corresponds to the west-dipping western bounding fault of the Tumut trough however between 0 -5 km depth, the GFZ is also the western boundary of an intervening wedge of

Ordovician turbidites as mapped in figure 4.2. The Ordovian turbidites are thus overlaying the Siluro-Devonian volcaniclastic sediments filling the Tumut trough, which indicates that the Ordovician turbidites must have be thrusted over the Siluro-Devonian volcaniclastic during a shortening event that followed the deposition of those Siluro -

Devonian volcaniclastic sediments.

The most notable feature in this cross-section is the Devonian syncline (Hervey group overlying Trundle group which overlies Ootha group) between 60- 85 km profile distance, including and extending all the way to the purported base of the Macquarie Arc

(~23 km depth, Fig 4.5).

As with cross-section 2 there are no modelled bodies below 25 km depth as modelling beyond this depth had no appreciable effect on TMI and gravity. However, two bodies

(between 40-60 km and 80-90 km profile distance; and both at 15 – 20 km depth) of high density (2900 -3020 kg/cm3) were used to model the corresponding high gravity anomalies. No similar bodies have been modelled in the previous two cross-sections however these bodies do have the expected petrophysical properties for MORB and serpentinites. I interpret these as dense basement of oceanic affinity.

As previously mentioned the Macquarie Arc does not appear to have a basement of completely oceanic affinity. Cross-section 2 shows evidence of continental crustal slices beneath the Macquarie Arc whilst cross-section 3 gives evidence of some oceanic crust below the arc.

Figure 4.5: Composite image of modelled cross-section 3 showing modelled and observed TMI, gravity and residual tracks and corresponding cross-sections coloured according to density (top, kg/m3), magnetic susceptibility (middle, SI) and simplified geological interpretation (bottom). The residual is the difference between calculated and observed and has been scaled to show the goodness-of-fit. Note that the unmodelled, white spaces correspond to areas without petrophysical contrast equalling 2670 kg/m3(density) and 0SI(magnetic susceptibility). As in cross-section 1 the GFZ is shown as a solid black line and Benambran Orogeny boundary is the thick dashed black. Note the presence of 2 dense bodies(white with small circles) of oceanic affinity. Reference colours found in Figure 4.2, note that this colour scale and interpretation of again brackets is varies slightly from the age brackets in Figures 4.1, 4.3 and 4.4. V=H,no vertical exaggeration. 55

CHAPTER 5: TECTONIC SYNTHESIS

5.1 IMPROVED FIT

The reflection seismic lines (Fig 3.1 a, b) have previously been modelled by Direen et al.

(2001) and interpreted by Glen et al. (2002). Certain features of tectonic significance such as the GFZ and Barmedman Fault did not fit the data acceptably in these models. The modelling described in this Thesis shows significant differences from the earlier work by

Direen et al. (2001; Fig 3.1 a, b) and these are highlighted by seven boxes in Figures 4.3-

4.4. There are differences in the number and size of separately modelled bodies, mismatches between the gradients of the peaks at the flanks of modelled bodies and the observed magnetics and gravity, and in the resolution of the observed signals.

Box 1

In this part of Cross-section 1, the calculated TMI response to the model of Direen et al.

(2001) displays a mismatch in both the position of the peak and the gradients of its sides relative to the observed TMI (Fig 3.1 a). The calculated signal also does not exhibit the twin peaks present in the observed signal. Direen et al. (2001) suggested that this peak represents the boundary created by the GFZ, and they modelled the fault as shallow east- dipping, overlain by a thin slice of Macquarie Arc rocks. I agree with the east-dipping trend but have improved the fit using more sub-vertical wedges instead of shallow dipping slices (Fig 4.3). The revised Barmedman Fault has an overall less gentle dip and a listric form, and matches trend lines visible in the seismic reflection profile at depths down to about 1.5 km between CDPs 3800 and 4200. I interpret the western edge of this TMI peak to correspond to the Barmedman Fault, not the GFZ.

Box 2 and Box 3

In box 2 the calculated TMI response of the Direen et al. (2001) model is consistently higher than the data by about 200 nT (Fig 3.1 a), implying that the combined magnetic susceptibility contribution in this part of the model is too high. The gradient of the flanks of the calculated anomaly is less than that of the data, suggesting that the source bodies dip more steeply than shown in the Direen et al. (2001) model. I have significantly improved the fit by modelling a thicker and more steeply dipping wedge of Siluro -

Devonian sedimentary rocks (Yiddah Formation) with zero susceptibility (ie. no magnetic contrast with the background susceptibility).

The gravity response in box 3 on Cross-section 1 is poorly matched by the Direen et al.

(2001) model (Fig 3.1 a), which produces a response with a gentle gradient than is seen in the data. I improved this fit through two measures: increasing the density of the inferred

Ordovician Macquarie Arc volcanics west of the GFZ from 2800 to 2900 kg/m3, and changing the dip of the GFZ bounding the Tumut Trough to dip to the west. This westward dip proved robust during sensitivity analysis.

Box 4 and Box 5

Boxes 4-7 (Fig 3.1 b) highlight the mismatches in modelled lines 99AGSL1, L2 which corresponds to my modelled cross-section 2 (Fig 4.4). The TMI response to the Direen et al. (2001) model highlighted in Box 4 does not show the three small peaks that are present in the data I modelled. My cross-section (Fig 4.4) shows that the magnetic data can be modelled with several shallow east-dipping tabular bodies, the edge of which corresponds to the Barmedman Fault. Note how Box 5 (Fig 3.1 b) also shows small east dipping bodies but these correspond to the peak labelled GFZ. However, although the

magnetic and gravity gradients appear to be to the east, a significantly better fit of both signals is obtained when the GFZ dips to the west (Figures 4.3, 4.4 and 4.5).

I disagree with this shallow east-dipping GFZ interpretation.

Box 6

Direen et al. (2001) modelled several Ordovician volcanic bodies with clearly defined contacts (Fig 3.1 b). However, the calculated TMI is a poor fit to the observed peaks. I instead have modelled this section with multiple steeply plunging bodies of alternating

Ordovician volcanics and Siluro-Devonian volcaniclastics (Fig 4.4) with a closer fit to the observed TMI. I do not agree with the position of the GFZ in Figure 3.1 b instead, I suggest that the first peak in Box 6 (Fig 4.4) corresponds to what I interpret as the GFZ, contact between Macquarie Arc and West dipping Tumut Trough.

Box 7

The TMI response to the Direen et al. (2001) model has a negative trough in the TMI at the start of Box 7 which is not present in the observed TMI (Figure 3.1 b). Seismic data

(labelled D on Figure 3.1 b) shows a tick-shaped shallow basin of strong, layered reflectors indicative of sedimentary basins. I have improved this fit by modelling an asymmetric tick-shaped, west-dipping basin (Fig 4.4) rather than a rounded basin (Figure

3.1 b). Mapped surface geology (Fig 4.1) shows this corresponds to a late Devonian basin.

5.2 GEOLOGICAL IMPLICATIONS

I suggest the west dipping boundary labelled E on Figure 3.1 a, b represents the fault developed during the Benambran orogeny as it separates the Wagga belt turbidites from the underthrusting Macquarie Arc. The boundary between the Macquarie Arc and the 58

Tumut Trough, to the east, is the GFZ. Stuart-Smith (1991) refers to the GFZ as the western bounding fault of the Tumut Trough and thus the late Devonian basin, granite and underlying late Silurian-early Devonian half graben modelled here represent the

Tumut Trough.

The west dipping wedge at the eastern end of cross-section 2 (3.5 km to 20 km depth) corresponds to Cambro-Ordovician MORB, of the Coolac Serpentinite Belt similar to the middle crust interpretation of Glen et al. (2002). I emphasise the differences in approach, of modeling the long wavelength decrease in gravity between 81- 90 km profile distance:

I modelled the wedge with 0.00001SI magnetic susceptibility instead of 0 SI as in Direen

3 3 et al. (2001) and assigned a lower density of 2700 kg/cm instead of 2850 kg/cm as in

Direen et al. (2001). Both these ranges of petrophysical properties are lower than would be expected for both MORB (Glen et al., 2002) and serpentinite (Jindalee Group equivalent in Direen et al. (2001)), rather they are better matched by a crust of continental affinity.

The findings are suggestive of a complex tectonic environment with the presence of both continental (Fig 4.4) and oceanic basement (Fig 4.5) below the Macquarie Arc. Such along strike variations are possible in most of tectonic models for the Macquarie Arc

(Collins, 2002a, b; Aitchison & Buckman, 2012; Cayley, 2012).

Although constraints on the deeper parts of the model are weak, all three models achieve the best fit to the data if the Tumut Trough is extended to >15 km depth. In Figure 2.3 the GFZ is marked by a high gravity anomaly on its western side. I have already suggested that this corresponds to dense Macquarie Arc rocks dipping west under the WMB. The model of the Macquarie Arc, particularly between 9-50 km profile distance on Cross-

section 1 (Fig 4.3) and 0-30 km profile distance on cross-section 2 (Fig 4.4), shows good correspondence between observed and modelled response. Given these findings I suggest that the base of the Macquarie Arc is at least 20 km deep.

The slice of MORB was used in modelling the gravity decrease in cross section 2 but its effect on gravity decreased with depth with increasing distance westward. Given the west dipping geometric trend of modelled bodies in cross-section 2 (from 60 – 90 km profile distance) I have modelled the eastern edge of the MORB as a decollement and have also interpreted that the Macquarie Arc roots into this decollement at ±20 km depth.

5.3 EVALUATION OF EXISTING TECTONIC MODELS

The noteworthy findings from cross-sections 1, 2 and 3 are summarised below:

 The GFZ appears as a west-dipping, crustal penetrating thrust fault.

 Cross-cutting relationships suggest the Barmedman Fault is a shallow back thrust

accommodating movement during the Benambran orogeny which pre-dates the

GFZ.

 The GFZ is the western bounding fault of the Tumut Trough.

 The GFZ is the boundary between the Tumut Trough and the Macquarie Arc

which are both wedged under the WMB.

 The GFZ does not juxtapose crustal segments of fundamentally different age and

geological character based on the modelled range of petrophysical properties of

observed rock types.

 The mid crustal rock immediately to the west of the GFZ (in cross-section 1)

appears to be denser than the rocks immediately to the east, but this difference

becomes less visible in cross-section 2 and less-so in cross-section 3.

 Given the method of modelling (i.e. multiple parallel 2.5D cross-sections), and

the fact that the cross-sections are striking roughly perpendicular to the main

geological structures, we cannot visualise horizontal slip/ transcurrent motion (i.e.

strike slip motion is out of plane in the cross-section) but we can see shortening

and extension visible as steeply plunging and near vertical faults/structures. The

cross-sections are roughly perpendicular to strike. This is the best approach for

modelling geometry, and dip-slip motion. However, although the cross-sections

are long (>80 km) regional-scale profiles, the employed technique is not

appropriate to ‘capture’ and identify transcurrent (along-strike) motion and along

strike variation.

5.3.1 Terrane Boundary

Many tectonic models include the newly named Barmedman Fault as part of the GFZ. I emphasise that these faults are different in character and should be separated. The evaluation of the following tectonic models is based on the GFZ as the fault in question.

Given these findings, I suggest that the GFZ is not the crustal suture proposed in the

Scheibner (1985) suspect terrane model. The GFZ is a crustal penetrating fault but it does not ‘suture together’ fundamentally different terranes as I interpret rocks of the Macquarie

Arc and Ordovician turbidites on both sides of the fault.

5.3.2 Extensional Accretionary Orogen/ Accordion Tectonics

The results suggest that the LO could have formed as an extensional accretionary orogen.

The extensional accretionary orogen model (Collins, 2002a, b) suggests the pattern of long term slab retreat and short term subduction advance is linked to the formation of high temperature, low pressure (HTLP) zones , such as the WMB, and S-type granites with a tripartite association (inboard S-type granite; outboard oceanic arc; intervening

turbidite-filled backarc basin). This model also suggests that the Tumut Trough was formed during one of these slab retreat phases (extension in the Silurian). Since the GFZ bounds the Tumut basin to the east, it must have been formed as an east dipping extensional fault. However, the current orientation of the GFZ demonstrates that it must have been rotated and reactivated as a west-dipping thrust during one or more subsequent shortening event. In addition to the GFZ, all three cross-sections show multiple west- dipping, crustal penetrating faults that are interpreted to be evidence of successive extension and basin inversion events punctuating LO formation. In addition, the GFZ might have accommodated some strike-slip movement (Stuart-Smith, 1991) but this is not directly visible in the cross-sections, as previously discussed.

5.3.3 Subduction flip/ polarity reversal

The cross-sections presented here show no evidence of east-dipping subduction.

Furthermore, attempts were made to model this possibility, but sensitivity testing, my means of calculating RMS error, in which the direction of dip was varied always achieved the minimum error with west-dipping subduction system. Therefore the results suggest that there is no evidence for a subduction flip or east-dipping subduction (Aitchison &

Buckman, 2012, Fergusson et al. 2013) associated with the GFZ and Macquarie Arc in this part of the LO.

5.3.4 Orocline model

The GFZ is only a small cog in the clock-wise, Z-shaped mega-fold that constitutes the giant Lachlan orocline (Musgrave & Rawlinson, 2010; Cayley, 2012; Musgrave, 2015).

According to the orocline model the WMB is in the south-moving core of the orocline

and deformation in the WMB should be extensional or transtensional and the high-grade metamorphism should be synchronous with development of the GFZ as an oblique/normal fault. This increased metamorphic grade would be reflected in increased density on the western side of the fault. As mentioned before, I cannot account for lateral

(along-strike) displacement in the geophysical models but in order for this movement to occur there would need to be steep, crustal-penetrating faults and vertically plunging bodies. The GFZ and surround bodies certainly fit this description (feature prominently in cross-section 2). Furthermore we do see the increased density of the western side of the GFZ, prominently in cross-section 1 but less so with the other two cross-sections possibly suggestive of the increasing level of deformation as the WMB moves further south.

In the Cayley (2012) model (Fig 2.6) Tumut Trough formation occurs between 430 - 400

Ma, however this is a ‘Victorian-centric’ view and the timing constraint and exact kinematics for the NSW portion of the model is not well constrained.

5.3.5 Extrusion of the WMB

The extrusion-type model by Morand & Gray (1991) suggests the GFZ formed as a strike- slip fault in response to compression and south-easterly movement of the WMB as a tectonic wedge over the Macquarie Arc. As previously mentioned, we cannot view transcurrent motion (in the cross-sections) therefore I cannot verify, using the modelled cross-sections, if the GFZ was a strike-slip fault at some stage in its formation. However the models presented here suggest that the GFZ formed as an extensional fault since it bounds the Tumut basin to the east, but currently appears structurally as a thrust further displacing the WMB over the Tumut Trough and Macquarie Arc. Regional tectonic studies suggest that the shortening event producing the underthrusting of the Macquarie

arc under the WMB corresponds to the Benambran Orogeny (~440Ma; Gray et al., 1997;

Foster et al., 1998; Foster et al., 1999; Cayley, 2011; Aitchison & Buckman, 2012). The shortening of the Silurian basins corresponds to the Bindian (~420 Ma) Tabberabberan

(~390 Ma) and Kanimblan (~350 Ma) orogenies.

5.4 TECTONIC EVOLUTION

Herein I present an interpreted tectonic evolution of the Macquarie Arc and Eastern LO from late Cambrian (490Ma) to late Devonian (350Ma) based on the findings of cross- sections 1- 3.

Interpretation of gravity, TMI and deep seismic-reflection data suggest that the Macquarie

Arc forms the base to the Tumut Trough (Basden, 1990; Dadd, 1998; Meffre et al.,2007),

Cowra Trough (Glen et al., 2002), Hill End Trough and the Mumbil Shelf (Glen et al.

2002; David et al. 2003). The following sketches also show the formation of Siluro-

Devonian basins and though the Tumut Trough is specifically mentioned, it is suggested that the post-Benambran orogeny extension events can generally be applied to all similar basins throughout the Macquarie Arc.

Figures 5.1- 5.7 have been constructed from the findings present in cross-section 1-3 (Fig

4.3- 4.5). I have placed my work in the greater context of the evolution of the LO by applying the position of faults and geometries of modelled bodies to the tectonic diagrams

(Fig 5.1 - 5.7). Reasonable estimates and constraints were used in developing the following tectonic sketches:

A mean ocean depth of 3.88 km (Kennett, 1982) and an average distance between the trench and magmatic front of arcs is 166±60 km (Gill, 1981). Although the average crustal

thickness for juvenile arc crust is ~20 km (Suyehiro et al., 1996) I have used a reasonable maximum crustal thickness of 30 km for the Macquarie Arc as shown in cross-section 1 and ≥1 km thickness for Narooma accretionary complex (Gray & Foster, 2004).

It is difficult to determine what mechanism or geodynamic events triggered the compression/ shortening. Furthermore, it is beyond the scope of this study. For the purpose of the illustration, a buoyant continental fragment is often drawn entering the trench and causing shortening.

5.4.1 490-440 Ma (Late Cambrian – Late Ordovician)

Throughout the Cambrian and Ordovician, the NSW portion of the Tasmanides (Percival et al., 2011) evolved from a few volcanic seamounts on the edge of the Delamerian continental margin to a series of depositional or back-arc basins and a volcanic island arc

(Glen, 2005; Glen et al., 2009).

Subduction related magmatism in the Macquarie Arc could have started as early as 490

Ma (Glen et al., 2007) but this date is poorly constrained. It is unclear whether the

Macquarie arc initiated as a continental arc and became later more oceanic when a backarc basin opened separating the arc from the continent, or if the arc formed outboard of the continent directly. The polarity of the subduction is also controversial, but the > 460

Ma Narooma accretionary complex (Cawood 1976, 1983; Percival et al., 2011) to the east suggest a west-dipping subduction forming the edge of the continent. The Delamerian highlands (Cambrian age Gondwana continental margin) supplied the detritus in the backarc throughout the early Ordovician.

Figure 5.1 illustrates the tectonic setting of the LO at 490 Ma. The down-going plate is the oceanic paleo-Pacific plate and subducts westward under the Gondwanan continent

at a steep angle. There is little evidence of the forearc and accretionary prism for the LO, apart for the Narooma complex (Fig 2.1). The pictured magmatic arc is the Macquarie arc and is constructed on a sliver of continental crust. The WMB is the backarc basin.

Figure 5.1: Evolution of the LO from 490 -440 Ma. No vertical exaggeration however the Wagga marginal basin is assumed to be >1000km wide thus cannot be drawn to scale.

The nature of the basement and substrate to the Macquarie arc and LO has been suggested as continental (Rutland, 1973; White et al., 1976; Christensen and Mooney, 1995), oceanic (Crook, 1969, 1974a; Direen et al., 2001; Glen et al., 2002; Spaggiari et al.,

2003a, 2004a; Meffre et al., 2011 ; Forster et al., 2015), or a mixture of both (Scheibner,

1974; Finlayson et al., 2002; Glen et al., 2007 ). The work presented here has shown that the mid crust should not be classified as simply oceanic or continental but that it contains evidence of both.

Recent work suggests that many island arcs in the western Pacific, such as Vanuatu (Buys et al., 2014), have zircon signatures of continental Australia. This suggests that the

Vanuatu arc was built on a fragment of continental Australia before it was rifted off during retreat of the southwestern pacific subduction margin. Similar implications are suggested for Tonga, Fiji and the Solomon Islands. A crustal velocity model (Finlayson et al., 2002) for the Macquarie Arc shows p-wave velocities within the range expected for continental crust (Christensen & Mooney, 1995). It is entirely possible, in an extensional accretionary orogen setting, that a continent-like ribbon rifted off continental Australia during a subduction retreat phase, and re-accreted onto the margin as the reworked Macquarie Arc, during renewed subduction advance (Schellart et al., 2006; Smyth et al., 2007; Buys et al., 2014).

5.4.2 440 Ma (Benambran Orogeny)

The late Ordovician – early Silurian deformation has been related to the Benambran

Orogeny (Browne 1947; David & Browne 1950). Multiple phases within the ‘Benambran orogeny’ particularly in the backarc have been identified (Offler et al., 1998; Glen et al.,

2007) but they do not leave any evidence in the geophysically modelled cross-sections therefore I cannot verify them.

By the start of the Benambran Orogeny the Macquarie Arc had accreted to the Gondwana

Plate and had been thrust under the WMB. This period also shows crustal thickening throughout the LO (Gray & foster, 2004). Some authors suggest that this large scale compression event was brought on by the collision of a Tasmania microcontinent

(Fergusson & Coney 1992a; Gray, 1997; Foster et al., 1999; Cayley, 2012), others suggest the northwards strike-slip transport of the allochthonous Bega Terrane along the eastern margin of Gondwana, into a forearc position outboard of the Macquarie Arc. (Glen et al.,

2007). Squire and Miller (2003) suggested that collision of the Macquarie Arc occurred at 455 Ma and was driven by the collision of a seamount (Glen et al., 1998).

Figure 5.2 shows a combination of thick and thin-skinned deformation. The thick-skinned thrusts illustrated in the backarc are speculative, and largely informed by analogue models

(Boutelier & Chemenda, 2011). The folded turbidites are a product of some thin-skinned deformation but this requires large shortening of the backarc and therefore some form of thick-skinned deformation deeper in the crust and lithosphere. Analogue models of backarc shortening have shown that multiple thrusts may be created (Boutelier &

Chemenda, 2011). The Barmedman fault forms as a back thrust accommodating movement on the (BB) contact between the WMB and underthrust Macquarie arc.

Figure 5.2: Tectonic model showing the onset of the Benambran orogeny. Benambran orogeny initiated by subduction of buoyant fragment (for the purpose of illustration, it is suspected to be continental). This fragment would be <15km thick as subduction continues and doesn’t fail (Cloos, 1993). This buoyancy anomaly produces compression in the back arc and folding of turbidites (thin-skinned deformation). This event also causes formation of large scale crustal faults and a decollement (thick-skinned deformation). Macquarie Arc magmatism ceases. No vertical exaggeration.

5.4.3 435- 425 Ma (Silurian Extension)

The purported incoming continental fragment (oceanic plateau or other buoyant fragments) is fully subducted by this time and initiates a subduction rollback event. This extensional period sees the formation of sedimentary basins such as the Tumut trough

(Jackalass slate sediments and Frampton volcanics-428Ma) as well as the intrusion of the post-Benambran granites in the wagga zone, such as Wantabadgery granite (435-425 Ma) and Ulanda granite (425Ma). The GFZ started forming as an extensional fault on the western flank of the Tumut trough.

Figure 5.3: Tectonic evolution of the LO between 435-425 Ma Silurian extension. Thickness of mantle under the arc extracted from estimates in Hyndman et al., 2005. The mantle is still thin as it has only been ±15 my since the last compressional episode and is still hot. Formation of Tumut trough and Barmedman fault at this time. Note that for the sake of illustration, the size of granites has been slightly exaggerated and are shown reaching the surface. No vertical exaggeration.

5.4.4 425- 420 Ma (Bindian Orogeny)

During this time there is subduction of another purported continental sliver (or other buoyant object carried by the lower plate) which induces a short-lived compressional episode. This causes inversion of the Silurian basins (such as the TT), steepening and rotation of faults and formation of the west-dipping GFZ as seen in the cross-sections.

The rotation of the GFZ may be explained by a vertical variation in the amount of underthrusting of the Macquarie arc west of the Tumut basin, under the WMB. If

Macquarie arc block west of the Tumut is dragged from below into a west dipping channel, then a dextral shear is generated causing est-dipping GFZ normal fault to rotate

(clockwise in these north-facing sketches) and become a west-dipping thrust.

Figure 5.4: Tectonic evolution of LO between 425 -420 Ma. Inversion of Tumut trough and formation of the GFZ. Age data based on work done by GSNSW (Forster et al., 2015). No vertical exaggeration

The Bindian is marked by NW-trending folds and cleavage (~420 Ma). I can constrain the time of inversion of the Tumut trough based on the intrusion of post-tectonic granites

(Mishurley granites aged ca. 420 Ma; latest data released by GSNSW).

5.4.5 Post Bindian

The cross-sections record further extension and basin inversion events shown in Figures

(5.5-5.7). It is assumed that the extension and compression events between 390 – 350 Ma are an expression of the extensional accretionary orogen. The Tumut trough experiences a second extension and inversion event (~390 Ma Tabberabberan Orogeny) which splits it into 2 basins with intervening Macquarie arc (Figure 5.5). The eastern most basin is unconformably overlain by late Devonian Hervey group (354-365 Ma age)-Figure 5.6 and is inverted at ~350 Ma as a result of the Kanimblan Orogeny (Gray et al., 1997; Foster et al., 1998; Foster et al., 1999).

Figure 5.5: Evolution of the LO during the Tabberabberan Orogeny. Due to lateral variation cross-sections 2-3 show the Tumut trough bisected into two basins, with intervening Macquarie Arc, unlike the single wedge basin visible in cross-section 1. This figure places the modelled geometries of cross-sections 2 and 3 into the bigger tectonic context.

Figure 5.6: Tectonic evolution of the LO in the extensional phase following the Tabberabberan. This figure shows cross-section 2-3 within the bigger tectonic context.

Figure 5.7: Tectonic evolution of the LO during the Kanimblan Orogeny. Figure places cross-section 2-3 within the bigger tectonic context.

5.5 SUMMARY

The GFZ appears as a west-dipping thrust fault in the forward modelled potential field sections. It has been shown here as the bounding fault of the Tumut Trough thus it may have been an extensional fault earlier in its development, but has since steepened and rotated. It is speculatively inferred that the GFZ may also have been a strike-slip fault but as previously discussed, identifying transcurrent motion and modelling along strike variation is not within the scope of this study. The observations further suggest that all the major west-dipping faults sole into a decollement and that the GFZ is not a basal detachment fault or terrane boundary. 74

The newly reclassified Barmedman fault is shown to terminate much shallower than the rest of the GFZ. It is an east-dipping listric, back thrust fault that appears to predate the

GFZ and inversion of the Tumut Trough.

The findings presented in this thesis show evidence of multiple episodes of shortening and extension in the overriding plate to long-lived west dipping subduction. It is thus inferred that the ±110 km length of the GFZ modelled here, places the LO within an extensional accretionary orogen/or accordion model.

The observations fit well within the proposed accordion model but they do not rule out the orocline if it can generate or be associated with multiple extension/contraction events.

The expected high grade metamorphism associated with the movement of the WMB (core of orocline) is visible in the models but more work is required on the timing, kinematic analysis and along strike variation (in NSW) of the faults to comment further on this model.

CHAPTER 6: CONCLUSIONS AND

RECOMMENDATIONS

This study has produced new insights into the tectonic evolution of the eastern LO by incorporating potential field modelling of the crust surrounding the GFZ with expected petrophysical property from sampled rock types as well as constraint from reflection seismic lines. All models were evaluated with sensitivity tests and further refined by using geological constraints and feasible geological concepts. I have improved on previous models of the area and shed light on some of the purported tectonic models by modelling the GFZ.

In this study I have shown the following:

 The GFZ is a west-dipping crustal penetrating thrust fault (425- 420Ma) that is

the western bounding fault of the Tumut trough.

 The GFZ is not a terrane boundary and is distinct from the shallowly terminating,

east dipping fault that should be separately classified as the Barmedman Fault.

 The Tumut trough extends to ≥15km depth whilst the Macquarie Arc extends to

≥25 km depth.

 The root of the Macquarie Arc appears to sole into a decollement at ±25km depth.

 The basement and substrate of the Macquarie Arc contains both continental and

oceanic crustal signatures.

 This work on the major thrust and/or strike-slip fault system is used to refine the

development of tectonic models of the evolution of the LO

 Evidence, presented here, supports formation as an extensional accretionary

orogen but does not exclude a giant orocline.

Further south the GFZ continues into Victoria but the trace of the GFZ appears more complex further north into NSW. Investigating the behaviour of this fault at depth would greatly contribute to our further understanding of the evolution of the LO in NSW.

However there is limited data to constrain any further modelling. Reflection seismic lines were of great use in constraining the morphology of the mid crust. If new reflection seismic lines, perpendicular to strike are conducted this would greatly support improve the prospects for continued potential field modelling. This study is limited in its ability to model transcurrent motion but this may be partly overcome by using seismic lines parallel to strike however none are currently available.

The ± 110 km modelled length of the GFZ provides an insightful view of the crust (up to

30 km depth) as it incorporates interpretations of structure, reflection seismics, potential fields, petrophysical properties and even consideration for the level of remanence.

This study produces a multidisciplinary regional scale model even with limited ‘fresh’ rock samples on which to obtain petrophysical properties. With access to more rock samples and seismic lines, this study paves the way for continued research on the geophysics, kinematics and timing constraints surrounding the GFZ.

REFERENCES

Aitchison, J.C., and Buckman, S., 2012, Accordion vs. quantum tectonics: Insights into continental growth processes from the Paleozoic of eastern Gondwana: Gondwana Research, 22, 674–680. doi: 10.1016/j.gr.2012.05.013

Basden, H., 1990, Tumut 1:100 000 Geological Sheet 8527, 1st edition: Geological Survey of New South Wales, Sydney

Basden, H., Adrian J., Clift, D.S.L., and Winchester R.E., 1975, Cootamundra 1:100 000 Geological Sheet 8528, 1st edition: Geological Survey of New South Wales, Sydney

Bell, M., in prep, The geodynamic significance of the Gilmore fault zone, Lachlan Orogen: structural characteristics, kinematic history and timing. Unpublished Masters Thesis, University of Newcastle

Belica, M.E., Tohver, E., Pisarevsky, S.A., Jourdan, F., Denyszyn, S., and George A,D., 2017, Middle Permian paleomagnetism off the Sydney Basin, eastern Gndwana: Testing Pangae models and the timing of the end of Kiaman reverse Superchron: Tectonophysics. doi: 10.1016/j.tecto.2016.12.029

Bosch, M., Meza, R., Jiménez, R., and Hönig, A., 2006, Joint gravity and magnetic inversion in 3D using Monte Carlo methods: Geophysics, 71, G153–G156. doi:10.1190/1.2209952

Boutelier, D., and Chemenda, A., 2011, Physical Modeling of Arc–Continent Collision: A Review of 2D, 3D, Purely Mechanical and Thermo‐Mechanical Experimental Models,Arc-continent collision, Springer Berlin Heidelberg, 445-473

Browne, W. R., 1947, A short history of the Tasman Geosyncline of eastern Australia: Science Progress, 140, 623–637

Brunker, R.L., and Offenburg, A.C., 1970, Macquarie 1:500 000 Geological Map, 1st edition: Geological Survey of New South Wales, Sydney

Brunker, R.L., 1968, Nymagee 1:250 000 Geological Sheet SI/55-02, 1st edition: Geological Survey of New South Wales, Sydney

Buys, J., Carl Spandler, C., Holm, R.J., and Richards, S.W.,2014, Remnants of ancient Australia in Vanuatu: Implications for crustal evolution in island arcs and tectonic development of the southwest Pacific: Geology, 42, 939–942

Cas, R.A.F., 1983, Palaeogeographic and tectonic development of the Lachlan Fold Belt of southeastern Australia. Geological Society of Australia, Special Publication, 10, 104 pp

Cawood, P.A., 1976, Cambro-Ordovician strata in northern New South Wales: Search, 7, 317–318 78

Cawood, P.A., 1983, Modal composition and detrital clinopyroxene geochemistry of lithic sandstones from the New England Fold Belt (east Australia): A Paleozoic forearc terrane: Geological Society of America Bulletin, 94, 1199–1214

Cawood, P.A., 2005, Terra Australis Orogen: Rodinia breakup and development of the Pacific and Iapetus margins of Gondwana during the Neoproterozoic and Paleozoic: Earth-Science Reviews, 69, 249–279

Cawood, P.A., Kroner, A., Collins, W.J., Kusky, T.M., Mooney, W.D. and Windley, B.F., 2009, Accretionary orogens through earth history. In: Cawood, P.A. & Kroner, A. (eds) Earth Accretionary Systems in Space and Time. Geological Society, London, Special Publications, 318, 1–36

Cayley, R. A., 2011, Exotic crustal block accretion to the eastern Gondwanaland margin in the Late Cambrian–Tasmania, the Selwyn Block, and implications for the Cambrian– Silurian evolution of the Ross, Delamerian, and Lachlan orogens: Gondwana Research, 19, 628–649. doi: S1342937X10002054

Cayley, R.A., 2012, Oroclinal folding in the Lachlan Fold Belt: consequence of Southeast-directed Siluro-Devonian subduction rollback superimposed on an accreted Ordovician arc assemblage in Eastern Australia, Abstract: Selwyn Symposium, GSA Victoria division Geological society of Australia

Cayley, R.A., Korsch, R.J., Moore, D.H., Costelloe, R.D., Nakamura, A., Willman, C.E., Rawling, T.J., Morand, V.J., Skladzien, P.B., and O'shea, P.J., 2011,Crustal architecture of central Victoria: results from the 2006 deep crustal reflection seismic survey: Australian Journal of Earth Sciences, 58,113-156

Cayley, R., and Musgrave, R., 2016, The giant Lachlan orocline-A new geodynamic model for the Ordovician-Devonian evolution of Australia: Earth Science Reviews

Chopping, R., and Kennett, B., 2015, Maximum depth of magnetisation of Australia, its uncertainty, and implications for Curie depth: GeoResJ, 7, 70–77. doi:10.1016/j.grj.2015.06.003

Christensen, N. I., and Mooney, W.D., 1995, Seismic velocity structure and composition of the continental crust: a global view: Journal of Geophysical Research, 100, 9761–9788

Clarke, J.D.A. and Chenoweth, L., 1995, Classification, genesis and evolution of ferruginous surface grains: AGSO Journal of Australian Geology and Geophysics, 16, 213–221. Available at: https://www.scopus.com/inward/record.uri?eid=2-s2.0- 0029433274&partnerID=40&md5=6fbcd020a367212a629d633de3d43150

Cloos, M., 1993, Lithospheric buoyancy and collisional orogenesis: Subduction of oceanic plateaus, continental margins, island arcs, spreading ridges, and seamounts: Geological Society of America Bulletin, 105, 715–737

Collins, W.J., 2002a, Hot orogens, tectonic switching, and creation of continental crust: Geology, 30, 535-538 79

Collins, W.J., 2002b. Nature of extensional accretionary orogens: Tectonics, 21(4), 1-6. doi: 10.1029/2000TC001272

Collins, W.J., and Hobbs, B.E., 2001, What caused the early silurian change from mafic to silicic (S-type) magmatism in the Eastern Lachlan Fold Belt?: Australian Journal of Earth Sciences, 48,25–41

Collins, W.J. and Richards, S.W., 2008, Geodynamic significance of S-type granites in circum-Pacific orogens. Geology, 36,559–562

Collins, W.J. and Vernon, R.H., 1992, Palaeozoic arc growth, deformation and migration across the Lachlan Fold Belt, southeastern Australia: Tectonophysics, 214, 381–400

Cooper, G.R.J. and Cowan, D.R., 2006, Enhancing potential field data using filters based on the local phase: Computers & Geosciences, 32,1585–1591

Cox, S. F., Etheridge, M. A., Cas, R. A. F., and Clifford. B. A., 1991a, Deformational style of the Castlemaine area, Bendigo–Ballarat zone: Implications for evolution of crustal structure in central Victoria: Australian Journal of Earth Sciences, 38, 151–170

Crawford, A. J., Meffre, S., and Symonds, P. A., 2003a, 130 to 0 Ma tectonic evolution of the southwest Pacific and analogous geological evolution of the 600 to 220 Ma Tasman Fold Belt System. In: Hillis R. R. & Mueller R. D. eds. Evolution and Dynamics of the Australian Plate: Geological Society of Australia Special Publication, 22, 383–403

Crawford, A., Glen, R., Cooke, D., and Percival, I., 2007, Geological evolution and metallogenesis of the Ordovician Macquarie Arc, Lachlan Orogen, New South Wales: Australian Journal of Earth Sciences, 54, 137–141. doi:10.1080/08120090701221615

Crook, K. A. W., 1969, Contrasts between Atlantic and Pacific geosynclines: Earth and Planetary Science Letters, 2, 289–303

Crook, K. A. W., 1974a, Cratonisation of west Pacific-type geosynclines: Journal of Geology, 87, 24–36

Dadd, K.A., 1998, Incipient Backarc Magmatism In the Silurian Tumut Trough, New South Wales - an Ancient Analogue Of the Early Lau Basin: Australian Journal of Earth Sciences, 45,109–121

David, T. W. E., and Browne, W. R., 1950, The Geology of the Commonwealth of Australia. Edward Arnold and Co. London

David, V., Glen, R. A., and Spencer, R., 2003, Three-dimension extension of the Hill End Trough based on the modelling of the regional gravity data. In: Vassallo J. J. & Glen R. A. eds. Evolution of the Hill End Trough, Geological Survey of New South Wales Report GS2003/291, 43 – 48

Direen, N., 1998, The palaeozoic Koonenberry Fold and Thrust Belt, Western NSW: a case study in applied gravity and magnetic modelling: Exploration Geophysics, 29, 330–339. doi:10.1071/EG998330

Direen, N., Lyons, P., Korsch, R., and Glen, R., 2001, Integrated geophysical appraisal of crustal architecture in the eastern Lachlan Orogen: Exploration Geophysics, 32, 252– 262. doi:10.1071/EG01252

Duggan, M., and Lyons, P., 2000, Wyalong 1:100 000 Geological Sheet 8330, 1st edition: Geological Survey of New South Wales, Sydney & Geoscience Australia, Canberra

Fergusson, C. L., 1987a, Early Palaeozoic back-arc deformation in the Lachlan Fold Belt, southeastern Australia: implications for terrane translations in eastern Gondwanaland. In: Leitch E. C. & Scheibner E. eds. Terrane Accretion and Orogenic Belts: American Geophysical Union Geodynamics Series, 19, 39–56

Fergusson, C. L., 2003, Ordovician–Silurian accretion tectonics of the Lachlan Fold Belt, southeastern Australia: Australian Journal of Earth Sciences, 50,475–490

Fergusson, C.L., and Coney, P.J., 1992a, Convergence and intraplate deformation in the Lachlan Fold Belt of southeastern Australia: Tectonophysics, 214, 417–439

Fergusson, C.L., and Coney, P.J., 1992b, Implications of a Bengal Fan-type deposit in the Paleozoic Lachlan fold belt of southeastern Australia: Geology, 20, 1047–1049

Finlayson, D.M., Korsch, R.J., Glen, R.A., Leven, J.H., and Johnstone, D.W., 2002, Seismic imaging and crustal architecture across the Lachlan Transverse Zone, a possible early cross cutting feature of eastern Australia: Australian Journal of Earth Sciences, 49, 311–321

Fisher, R. A., 1953, Dispersion on a sphere: Proceedings of the Royal Society of London. Series A, 217, 295–305. doi:10.1098/rspa.1953.0064

Forster, D. B., Mcinnes, P., Downes, P. M., Maas, R., Norman, M., and Blevin, P. L., 2015, New lead isotopic and geochronologic constraints on mineralisation in the Macquarie Arc - insights from the Lake Cowal district, Geological Survey of New South Wales: Quarterly Notes,144, 1-21

Foss, C., 2002, Interpretation of formation dip from modelling TMI data, Abstract: EAGE 64th Conference & Exhibition, Florence, Italy, 27–30 May 2002

Foster, D.A., and Gray, D.R., 2000, Discussion and Reply: Timing of orogenic events in the Lachlan Orogen: Australian Journal of Earth Sciences, 47, 813–822

Foster, D.A., Gray, D.R., and Bucher, M., 1999, Chronology of deformation within the turbidite-dominated, Lachlan orogen: Implications for the tectonic evolution of eastern Australia and Gondwana: Tectonics, 18, 452–485

Foster, D. A., Gray, D. R., Kwak, T. A. P., and Burcher, M., 1998, Chronology and tectonic framework of turbidite hosted gold deposits in the western Lachlan fold belt,Victoria: 40 Ar–39 Ar results: Ore Geology Reviews, 13, 229–250

Gill, J. B., 1981, Orogenic Andesites and Plate Tectonics:Geological Magazine, 119, 516-517. doi: 10.1017/S0016756800026911

Glen, R., 1992, Thrust, extensional and strike-slip tectonics in an evolving Palaeozoic orogen—a structural synthesis of the Lachlan Orogen of southeastern Australia: Tectonophysics, 214, 341–380

Glen, R. A., 1994, Ordovician-earliest Silurian sedimentary rocks. In: Lewis, P. C., Glen, R. A., Pratt, G. W., and Clarke, I., (compilers), Bega-Mallacoota 1:250 000 Geological Sheet SJ/55-4, SJ/55-8, Explanatory Notes. Geological Survey of New SouthWales, Sydney, http://www.resourcesandenergy.nsw.gov.au/miners-and- explorers/geoscience-information/products-and-data/maps/geological-maps

Glen, R. A., 2005, The Tasmanides of eastern Australia: Geological Society, London, Special Publications, 246, 23–96

Glen, R. A., 2013, Refining accretionary orogen models for the Tasmanides of eastern Australia: Australian Journal of Earth Sciences, 60, 315–370. doi: 10.1080/08120099.2013.772537

Glen, R. A., Dallmeyer, R. D., and Black, L. P., 1992, Isotope dating of basin inversion- The Palaeozoic Cobar Basin, Lachlan Orogen, Australia: Tectonophysics, 214, 249–268

Glen, R., Korsch, R., Direen, N., Jones, L., Johnstone, D., Lawrie, K., Finlayson, D., and Shaw, R., 2002, Crustal structure of the Ordovician Macquarie Arc, Eastern Lachlan Orogen, based on seismic-reflection profiling: Australian Journal of Earth Sciences, 49, 323–348. doi:10.1046/ j.1440-0952.2002.00925.x

Glen, R.A., Meffre, S., and Scott, R.J., 2007, Benambran Orogeny in the Eastern Lachlan Orogen, Australia: Australian Journal of Earth Sciences, 54, 385-415

Glen, R.A., Percival, I.G., and Quinn, C.D., 2009, Ordovician continental margin terranes in the Lachlan Orogen, Australia : Implications for tectonics in an accretionary orogen along the east Gondwana margin: Tectonics, 28,1–17 doi:10.1029/2009TC002446

Glen, R.A., Quinn, C.D., and Cooke, D.R., 2012, The Macquarie Arc, Lachlan Orogen, New South Wales: Its evolution, tectonic setting and mineral deposits: Episodes, 35,177–186

Glen, R. A., Walshe, J. L., Barron, L. M., and Watkins, J. J., 1998, Ordovician convergent-margin volcanism and tectonism in the Lachlan sector of east Gondwana: Geology, 26, 571–574

Gray, D. R., 1997, Tectonics of southeastern Australian Lachlan Fold Belt: structural and thermal aspects. In: Burg J-P. & Ford M. (eds.), Orogeny through Time: Geological Society Special Publication, 121, 149–177

Gray, D.R., and Foster, D.A.,1997, Orogenic concepts-application & definition: Lachlan fold belt, Eastern Australia: American Journal of Science, 297, 859-891. Gray, D.R., and Foster, D. A., 2004, Tectonic evolution of the Lachlan Orogen, southeast Australia: Historical review, data synthesis and modern perspectives: Australian Journal of Earth Sciences, 51, 773–817

Gray, D.R., Foster, D.A., and Bucher, M., 1997, Recognition and definition of orogenic events in the Lachlan Fold Belt: Australian Journal of Earth Sciences, 44, 489–501

Gray, D.R., and Gregory, R.T., 2003, Fault geometry as evidence for inversion of a former rift basin in the Eastern Lachlan Orogen: Australian Journal of Earth Sciences, 50, 513–523

Gray, D.R., and Willman, C.E., 1991a, Deformation in the Ballarat Slate belt, central Victoria and implications for the crustal structure across SE Australia: Australian Journal of Earth Sciences, 38, 171–201

Gray, D.R., and Willman, C.E., 1991b, Thrust-related strain gradients and thrusting mechanisms in a chevron-folded sequence, southeastern Australia: Journal of Structural Geology, 13, 691–710

Hyndman, R.D., Currie, C.A., and Mazzotti, S.P., 2005, Subduction zone backarcs, mobile belts, and orogenic heat: GSA Today, 15, 4–10

Kennett, J. P., 1982, Marine Geology, Prentice-Hall, Englewood Cliffs, N .J.

Koenigsberger, J., 1938, Natural residual magnetism of eruptive rocks: Journal of Geophysical Research, 43, 119–130. doi:10.1029/TE043i002p00119

Lane, R., FitzGerald, D., Guillen, A., Seikel, R., and McInerney, P., 2007, Lithologically constrained inversion of magnetic and gravity data sets: Preview, 129, 11–17

MacRae, G.P., 1988, Nymagee 1:100 000 Geological Sheet 8133, 1st edition: Geological Survey of New South Wales, Sydney

Meakin, N.S., Colquhoun, G.P., Cameron, R.G., Hendrickx, M.A., Vassallo, J.J., and Forster, D.B., 2006, Cargelligo 1:250 000 Geological Sheet SI/55-06, 3rd edition: Geological Survey of New South Wales, Maitland

Meffre, S., Norman, M., Crawford, A. J., and Squire, R., 2001, Detrital zircons and sandstone provenance, In: CODES–DMR–SPIRT Ordovician Project, Final report (April), 9–14

Meffre, S., Scott, R.J., Glen, R.A., and Squire, R.J., 2007, Re-evaluation of contact relationships between Ordovician volcanic belts and the quartz-rich turbidites of the Lachlan Orogen: Australian Journal of Earth Sciences, 54, 363–383

Morand, V.J., and Gray, D.R., 1991, Major fault zones related to the Omeo Metamorphic Complex, Northeastern Victoria: Australian Journal of Earth Sciences, 38, 203–221. doi:10.1080/08120099108727965

Moresi, L., Betts, P.G., Miller, M.S., and Cayley, R.A., 2014, Dynamics of continental accretion: Nature, 508, 245-248. doi: 10.1038/nature13033

Musgrave, R.J., 2015, Oroclines in the Tasmanides: Journal of Structural Geology, 80, 72–98. doi: 10.1016/j.jsg.2015.08.010

Musgrave, R. J., and Dick, S., 2017, A 3D model for the Koonenberry Belt from geologically constrained inversion of potential field data: Geological Survey of New South Wales, Quarterly Notes, 149, 1–13

Musgrave, R., and Rawlinson, N., 2010, Linking the upper crust to the upper mantle: comparison of teleseismic tomography with long-wavelength features of the gravity and magnetic fields of southeastern Australia: Exploration Geophysics, 41, 155-162

Offler, R., Miller, J. McL., Gray, D. R., Foster, D. A., and Bale, R., 1998, Crystallinity and b0 spacing of K-white micas in a Paleozoic accretionary complex, eastern Australia: metamorphism, paleo- geotherms, and structural style of an underplated sequence: Journal of Geology, 106, 495–509

Oldenburg, D., and Pratt, D., 2007, Geophysical inversion for mineral exploration : a decade of progress in theory and practice, in B. Milkereit, ed., Proceedings of Exploration 07: Fifth Decennial International Conference on Mineral Exploration, 61– 95

Packham, G. H., and Falvey, D. A., 1971, An hypothesis for the formation of marginal seas in the western Pacific: Tectonophysics, 11, 79 – 109

Percival, I. G., and Glen, R. A., 2001, Correlation of successions within the Ordovician Macquarie Arc, central New South Wales: Geological Survey of New South Wales, 47– 60

Percival, I.G., Quinn, C.D., and Glen, R.A., 2011, A review of Cambrian and Ordovician stratigraphy in New South Wales, Geological Survey of New South Wales: Quarterly Notes, 137, 1-39

Powell, C. McA., 1984, Ordovician to earliest Silurian: marginal sea and island arc, Silurian to mid-Devonian dextral transtensional margin; Late Devonian and Early Carboniferous: Continental magmatic arc along the east- em edge of the Lachlan Fold Belt. In: J.J. Veevers (Editor): Phanerozoic Earth History of Australia, Oxford Univ. Press, Oxford, 290-340

Raymond, O.L., Duggan, M.B., Lyons, P., Scott, M.M., Sherwin, L., Wallace, D.A., Krynen, J.P., Young, G.C., Wyborn, D., Glen R.A., Percival I.G., and Leys M., 2000, Forbes 1:250 000 Geological Sheet SI/55-07, 2nd edition: Geological Survey of New South Wales, Sydney, Geoscience Australia, Canberra

Royden, L., and Burchfiel, B. C., 1989, Are systematic variations in thrust belt style related to plate boundary processes? (the western Alps versus the Carpathians): Tectonics, 8, 51–61

Rutland, R.W.R., 1973, Tectonic evolution of the continental crust of Australia. In: D.H. Tarling and S.K. Runcorn (Editors), Continental Drift, Sea Floor Spreading and Plate Tectonics: Implications to the Earth Sciences. Academic Press, London, 1003- 1025

Scheibner, E., 1972, Actualistic models in tectonic mapping, Report of the 24th IGC 3, 405–422

Scheibner, E., 1974, Fossil fracture zones (transform faults), segmentation and correlation problems in the Tasman Fold Belt System. In: Denmead A. K., Tweedale G. W. & Wilson A. F. eds. The Tasman Geosyncline—a symposium in honour of Professor Dorothy Hill: Geological Society of Australia, Queensland Division, Brisbane, 65–98

Scheibner, E., 1985, Suspect terranes in the Tasman Fold Belt System, east Australia: Tectonostratigraphic Terranes of the Circum-Pacific Region, 1, 493–514

Scheibner, E., 1987, Paleozoic development of eastern Australia in rela- tion to the Pacific region. In: Monger J. W. H. and Franchetau J. J. (eds), Circum-Pacific orogenic belts and evolution of the Pacific ocean basin: American Geophysical Union Geodynamic Series, 18, 133–165

Scheibner, E., and Basden, H. (eds), 1998, Geology of New SouthWales-Synthesis. Volume 2: Geological Evolution, Geological Survey of New South Wales: Memoir, 13, 1- 645

Scheibner, E., and Veevers, J. J., 2000, Tasman Fold Belt System. In: Veevers J. J. (ed) Billion-year earth history of Australia & neighbours in Gondwanaland: GEMOC Press, Sydney, 164–234

Schellart, W.P., Lister, G.S., and Toy, V.G., 2006, A Late Cretaceous and Cenozoic reconstruction of the Southwest Pacific region: Tectonics controlled by subduction and slab rollback processes: Earth-Science Reviews, 76, 191–233. doi:10.1016/j.earscirev.2006.01.002

Sherwin, L., 1997, Narromine 1:250 000 Geological Sheet SI/55-03, 2nd edition: Geological Survey of New South Wales, Sydney

Smyth, H.R., Hamilton, P.J., Hall, R., and Kinny, P.D., 2007, The deep crust beneath island arcs: Inherited zircons reveal a Gondwana continental fragment beneath East 85

Java, Indonesia: Earth and Planetary Science Letters, 258, 269–282. doi:10.1016/j.epsl.2007.03.044

Spaggiari, C. V., Gray, D. R., and Foster, D. A., 2003a, Tethyan- and Cordilleran-type ophiolites of eastern Australia: implications for the evolution of the Tasmanides, In: Dilek Y. & Robinson P. T. (eds), Ophiolites in Earth history: Special Publication of the Geological Society London, 218, 517–539

Spaggiari, C. V., Gray, D. R., and Foster, D. A., 2003b, Formation and emplacement of the Dolodrook serpentinite body, Lachlan Orogen, Victoria: Australian Journal of Earth Sciences, 50, 709–723

Spaggiari, C. V., Gray, D.R., and Foster, D.A., 2004, Lachlan Orogen subduction- accretion systematics revisited: Australian Journal of Earth Sciences, 51, 549–553

Squire, R., and Miller, J. M., 2003, Synchronous compression and extension in East Gondwana: tectonic controls on world-class gold deposits at 440 Ma: Geology, 31, 1073–1076

Stuart-Smith, P.G., 1990a, Evidence for extensional tectonics in the Tumut Trough, Lachlan Fold Belt,NSW: Australian Journal of Earth Sciences, 37,147–167

Stuart-Smith, P.G., 1991a, The Gilmore Fault Zone—the deformational history of a possible terrane boundary within the Lachlan Fold Belt New South Wales: BMR Journal of Australian Geology & Geophysics, 12, 35 – 49

Stuart-smith, P.G., 1991b, Gilmore fault zone eastern Australia in tectonostratigraphic terranes in the circum-pacific region: Circum-pacific council for energy and mineral resources- Earth science series

Stuart-Smith, P.G., Hill, R.I., Rickard, M.J., and Etheridge, M.A., 1992, The stratigraphy and deformation history of the Tumut region: Implications for the development of the Lachlan Fold Belt. In: CL. Fergusson and R.A. Glen (Eds), The Palaeozoic Eastern Margin of Gondwanaland: Tectonics of the Lachlan Fold Belt, Southeastern Australia and Related Orogens: Tectonophysics, 214, 211- 237

Suppel, D.W., Warren, A.Y.E., Watkins, J.J., Chapman, J., Tenison Woods, K., and Barron, L., 1986, A reconnaissance study of the geology and gold deposits of the west Wyalong-Temora-Adelong District: New South Wales Geological Survey: Quarterly Notes, 64, 1-23

Suyehiro, K., Takahashi, N., Ariie, Y., Yokoi, Y., Hino, R., Shinohara, M., Kanazawa,T., Hirata, N., Tokuyama, H., and Taira, A., 1996, Continental crust, crustal underplating, and low-Q upper mantle beneath an ocean island arc: Science, 272, 390- 392

VandenBerg, A. H. M.,Willman, C. E., Hendrickx M. A., Bush, M. D., and Sands, B., 1995, The geology and prospectivity of the 1993 Mount Wellington Airborne Survey area. Department of Agriculture, Energy and Minerals: VIMP Report 2 86

VandenBerg, A. H. M., Willman, C. E., Maher, S., Simons, B. A., Cayley, R. A., Taylor, D. H., Morand, V. J., Moore, D. H., and Radojkovic, A., 2000, The Tasman Fold Belt System in Victoria. Geology and mineralisation of Proterozoic to Carboniferous rocks: Geological Survey of Victoria Special Publication

Venkataramani, D., Musgrave, R.J., Boutelier, D.A., Hack, A.C., and Collins, W. J., 2017, Revised potential field model of the Gilmore Fault Zone, Exploration Geophysics. doi:10.1071/EG16148

Warren, A.Y.E., Gilligan, L.B., and Raphael, N.M., 1995, Geology of the Cootamundra 1:250 000 map Sheet. viii + 160 pp. Geological Survey of New South Wales, Sydney

Warren, A.Y.F., Gilligan, L.B., and Raphael, N.M., 1996, Cootamundra 1:250 000 Geological Sheet SI/55-11, 2nd edition: Geological Survey of New South Wales, Sydney

White, A. J. R., Williams, I.S., and Chappell, B.W.,1976, The Jindabyne thrust and its tectonic, physiographic and petrogenetic significance: Journal of the Geological Society of Australia, 23, 105-112

Wormald, R. J., and Price, R. C., 1990, The gabbro-qtz monzodioritealkai granite association in southern New South Wales: implications for intrusive related gold mineralisation: Geological Society of Australia Abstracts, 25, 265

Wyatt, B. W., Yeates, A. N., and Tucker, D. H., 1980, A regional review of the geological sources of magnetic and gravity fields in the Lachlan Fold belt of NSW: BMR Journal of Australian Geology and Geophysics, 5, 289–300

Zen, E. A., 1996, Crustal magma generation and low-pressure high temperature metamorphism in an extensional environment; possible application to the Lachlan Fold Belt: American Journal of Science, 295, 851–874

APPENDICES

APPENDIX A1:

APPENDIX A2:

APPENDIX A3:

APPENDIX B1:

APPENDIX B2: