Skip to content | Change text size
Earth Modelling Institute

 

Geodynamics of the Australian Plate
Research Meeting

Geology, Geophysics, Hazards & Resources
ABSTRACTS
(current at 30/05/09)
Mineralisation at the Australian plate margin - ancient and modern examples

David R. Cooke and Jacqueline L. Blackwell

CODES, The Australian Research Council’s Centre for Excellence in Ore Deposits, University of Tasmania, Private Bag 126, Hobart, 7001, Tasmania, Australia

Convergent margin settings are favourable environments for magmatic-hydrothermal systems. Porphyry copper-gold, epithermal gold-silver and skarn deposits have formed at various sites on the eastern and northern margins of the Australian plate during orogenesis over the past 450 m.y. Ancient and modern examples highlight that similar processes have probably caused mineralisation during the Paleozoic to recent history of the Australian plate margin.

The Macquarie Arc was an oceanic island arc situated to the east of mainland Australia during the Middle Ordovician to early Silurian. At least two mineralising epochs occurred in the Macquarie Arc’s evolution. The first, around 450 Ma, produced calc-alkalic porphyry Cu-Au deposits (e.g., Copper Hill, Cargo), and low sulfidation-style mineralisation at Cowal. Mineralisation occurred during a hiatus in volcanism, at a time of limestone deposition and shoaling of the arc. The second, and far more economically significant mineralising event occurred during the Benambran Orogeny, around 440 Ma. Alkalic porphyry deposits and skarns formed at Cadia and North Parkes after volcanism ceased in the Macquarie Arc. The alkalic intrusions were the final magmatic suite emplaced in the Macquarie Arc, apparently during the collision of the arc with mainland Australia.

Recent mineralising events on the Australian plate margin can help to provide insights into tectonic processes that may have occurred during the evolution of the Macquarie Arc. A giant calc-alkalic porphyry copper gold deposit formed in the Banda Arc at Batu Hijau (Sumbawa) 3.7. m.y. ago. Subduction of the Roo Rise beneath Sumbawa led to a kink or tear in the subducting slab beneath the Banda arc. This structure propagated as a NE-trending fracture zone through the overriding plate, providing a locus for magmatic-hydrothermal mineralisation at Batu Hijau. Similar topographic and thermal anomalies on down-going slabs can be shown to be spatially and temporally associated with porphyry and epithermal deposits of Miocene to Pliocene age elsewhere around the circum-Pacific (e.g., Philippines, Mexico, Panama, Ecuador, Peru and Chile). Similar processes may have been important to the formation of the calc-alkalic porphyry deposits of the Macquarie Arc 450 m.y. ago, but any physical evidence for ridge or seamount subduction is ultimately consumed by the subduction process, preventing any definitive proof of this phenomenon in ancient settings.

Alkalic epithermal gold deposits have formed on the PNG mainland (Porgera) and offshore PNG (Lihir) during the past six million years. They provide insights into processes that may have characterised the alkalic mineralising event at the culmination of Macquarie Arc magmatism in the Paleozoic.  Lihir is part of the 250 km long, Tabar-Lihir-Tanga-Feni, alkalic volcanic island chain located in the New Ireland Basin. Collision of the Ontong-Java plateau with the Australian plate (~ 25 Ma) initially caused subduction reversal (10 Ma), shutting down arc magmatism at New Ireland and triggering arc magmatism at New Britain. In the past 3.5 million years, alkalic volcanism and related magmatic-hydrothermal mineralisation at Lihir has occurred in a back-arc setting relative to the New Britain arc. Arc-normal structures have localised alkalic volcanism and mineralisation in the back arc. These cross-arc structures appear to be dilating episodically, due to the ongoing oblique collision of the Ontong-Java plateau with the New Britain subduction system. It may be that collision of the Macquarie Arc with an oceanic plateau terminated subduction in the early Silurian. Alternatively, it may be the Macquarie Arc’s collision with, and amalgamation to, mainland Australia that activated the cross-arc structures which localised alkalic magmatism and mineralisation at Cadia and North Parkes.


Boninites Ancient and Modern: Tectonic Setting and Possible Roles as Source Rocks for Palaeozoic Turbidite-hosted Gold Deposits

Anthony J Crawford(1) and Reid R Keays (2)

1. ARC Centre of Excellence in Ore Deposits (CODES), University of Tasmania
2. School of Earth Sciences, Monash University

Boninites are petrographically and compositionally distinctive rocks that form generally in forearc settings in intra-oceanic arcs. Key compositional features include simultaneously high SiO2 (>53%) and high MgO (>6%) and very low contents of TiO2 (<0.4%).  A spectrum of compositions from end-member low-Ca boninites with very low CaO and Al2O3 contents, to high-Ca boninites with relatively high CaO and Al2O3 exists, the latter passing into the depleted end of the arc tholeiite
compositional range. Key ingredients for boninite genesis are a supply of heat to elevate temperatures in the forearc mantle wedge from ambient temperatures probably around 600°C at ~50km depth, to temperatures in excess of 1200°C. For end-member low-Ca boninites, we suggest that the prime petrogenetic mechanism may be initiation of subduction at a recently active spreading centre in response to a major reorganisation of plate boundaries. Thus the ‘burst’ of low-Ca boninitic magmatism around the W Pacific at 55-50Ma is taken to reflect initiation of subduction at spreading ridges in W Pacific marginal basins as a response to plate reorganisations consequent upon India – Asia collision. Thus in ancient foldbelts, low-Ca boninites can be useful indicators of plate polarity (since they are apparently restricted to forearc regions of intra-oceanic arcs), and also of major (global?) plate reorganisations.

From this perspective, we examine the occurrence of boninites in foldbelts, and use them to interpret the tectonic development of the Lachlan Fold Belt in Victoria and Tasmania. Furthermore, we show that boninites contain 1-2 orders of magnitude more gold than mid-ocean ridge basalts (and even more relative to typical upper crustal rocks and granites). The occurrence of a thick, imbricated pile of autobrecciated, glassy, and thus eminently alterable boninites in the basement of the western part of the Lachlan Fold Belt may be a significant reason why the overlying Ordovician turbidite succession is so endowed with gold.
Subsidence and hydrocarbon maturation on Australian continental margins in the full lithosphere context

Alexey Goncharov(1), Ian Deighton(2), Sandra McLaren(3) and Christian Heine(4)

1. Geoscience Australia
2. TGS Nopec, London
3. School of Earth Sciences, University of Melbourne
4. StatoilHydro, Oslo
 
The extent to which basement and crustal controls on subsidence modelling should be quantified is equivocal. Uncertainty in parameters of subsidence modelling that can only be defined qualitatively may outweigh the uncertainty related to parameters that can be quantified, and as such finding the right balance between quantifiable and unquantifiable uncertainties is a challenge.

Conventional burial and thermal geo-history models use inferred values for heat flow or geothermal gradient and determine palaeo heat flow purely from stripped basement subsidence. In our approach, advanced burial and thermal geo-history modelling was carried out using Fobos Pro modelling software utilising six basement and crustal controls on hydrocarbon (HC) maturation: total sediment thickness, crustal thickness, lithosphere thickness, crustal heat production, sediment heat production, and dynamic topography. The transient thermal solution was determined each time stretching and sediment loading occur. Heat flow was determined from the calculated palaeo temperatures, and was not a direct input in the procedure. High quality refraction seismic data were needed to constrain basement depth and composition, and crustal thickness. Interpreted reflection seismic profiles, heat flow measurements, content of radioactive elements in rock samples taken from basement outcrops onshore or dredge samples, and high quality vitrinite reflectance and temperature data from wells were also used to constrain input parameters.

The position of HC maturation windows in vertical section is a result of complex interplay of all six basement and crustal controls. We determined the relative significance of these controls on model outputs, estimated possible ambiguities and discovered somewhat enigmatic limits to certain sensitivities (e.g., lithosphere thickness). We tried to establish ultimate end-member scenarios that would correspond to coldest and hottest maturation scenarios, and discovered that they fail to satisfy PMT (predicted Moho depth test, it compares crustal thickness predicted by subsidence modelling to that independently measured or calculated). We demonstrated that PMT is a powerful tool in discriminating multiple possible solutions. For example, for one of the wells tested, ~1 km shifts of top dry gas maturation windows result from upper crustal heat production variation between 2 and 4 µW/m3. This variation corresponds to models with substantially different responses to the PMT, thus allowing to prefer a lower heat production scenario. Finally, we married these results to interpreted source rock intervals in some study areas, and demonstrated how important even small shifts in maturation windows can be. On the basis of this we conclude that subsidence and HC maturation modelling will lead to ambiguous results if based on reflection seismic interpretation alone without considering the full lithosphere context.


Timor Collision: Stratigraphic Constraints

David W. Haig, Myra Keep, Eujay McCartain

School of Earth and Environment, The University of Western Australia, 35 Stirling Hwy, Nedlands, Perth, WA, 6009

Evidence is presented for the timing of four distinct phases in the development of the Timor orogen, as determined from stratigraphic analysis. These include initial collision (10.9-9.8 Ma GTS2004), emplacement of the early nappes creating loading and diapirism (9.8-5.5 Ma GTS 2004), a tectonic quiet interval (5.5 Ma-4.5 Ma GTS 2004) that extended for over a million years and may represent the time of locking of the subduction system, and a post 4.5 Ma phase of uplift, unroofing and further diapirism in response to isostatic rebound. This history significantly changes some previous interpretations for timing of the collision.
 
Stratigraphic relationships suggest that the collision was between the Banda Arc and an ancient Timor Plateau — a continental terrace/plateau that was contiguous with the Australian mainland and similar in morphology and bathymetry to present-day Exmouth Plateau. From the Early Cretaceous to the Middle Miocene, Timor Plateau lay in the middle bathyal zone. The eastern slope of Timor Trough may be partly the result of Early Cretaceous subsidence of the Timor Plateau. A palaeoenvironmental re-evaluation is made of the basal section cored in DSDP Leg 27 Hole 262 located low on the eastern flank of the Timor Trough.
 
Key stratigraphic criteria used in previous interpretations of the age of collision are critically assessed. These include the ages of the youngest pre-collisional strata and the oldest synorogenic unit; the presence of a supposed mid-Pliocene unconformity spanning planktonic foraminiferal zone N20; and the age of the youngest unit attributed to the Banda Terrane.
 
The central role of biostratigraphy in reconstructing the pre-collisional stratigraphic succession and in determining stratigraphic breaks indicative of particular tectonic events, and in evaluating facies, is emphasised.  Constraints on the age of collision based on non-stratigraphic criteria are used to support the proposed history of collision outlined above.  
 
Timor Collision: onshore and offshore deformation

Myra Keep, David Haig, Eujay McCartain

School of Earth and Environment, The University of Western Australia, 35 Stirling Hwy, Nedlands, Perth, WA, 6009.

New biostratigraphic dating places the collision between the Australian Plate and the Banda Arc at 10.9-9.8Ma.  Colliision produced a complex intercalation of thrust slices from both the Australian Plate and Banda Arc sides of the plate boundary. Initial thrust emplacement occurred between 9.8-5.5 Ma, producing, amongst other things, a series of south-directed thrust slices of material from the Gondwana and Australian Margin megasequences. These thrusts produced complex antiformal thrust stacks that generally include Triassic limestones from the Gondwana Megasequence as the highest structural unit.  Thrust stacks preserve elements of the Gondwana and Australian Margin megasequences, including the remnants of the collided “Timor plateau”, analogous to the offshore Exmouth Plateau of the present day. Emplacement of these thrust slices and subsequent loading of the crust caused remobilisation of underlying Triassic mudstones, which were emplaced in the Viqueque region prior to 5.5 Ma. These remobilised muds form diapirs and melange deposits, containing blocks of Australian-derived materials. After a tectonic quiet zone from 5.5-4.5 Ma, further thrust emplacement occurred. 

Intercalation of Australian-derived material with material from the Banda Terrane has been complicated by probable over-folding of Banda Terrane thrust slices, resulting in unpredictable outcrop locations, probable inverted stratigraphy at some locations, and complex structural interactions. Late to Recent high-angle faults control much of the present-day topographic expression of the island.

Seismic structural interpretation of offshore sequences from re-processed 2D seismic data along the southern margin of East Timor mimic onshore deformation styles, with the structural style dominated by shortening, preserved as multiple south-directed thrusts, cut by late normal faults, both listric and high-angle planar faults. The only offshore well, Mola 1, encountered the Pliocene—Pleistocene boundary close to it’s maximum depth at ~3km, indicating significant shortening and thickening of Pleistocene units.
The Geodynamics of Great Earthquakes

Gordon Lister and Marnie Forster

Research School of Earth Sciences, The Australian National University

The crust and mantle above major subduction zones is mechanically weakened by the flux of heat and water that can be associated with subduction zone processes.  In consequence the lithosphere of the over-riding orogens might seem to act more like a fluid than a rigid plate.  Such fluid-like behaviour has been noted for the Himalaya and for the crust of the uplifted adjacent Tibetan Plateau, which appear to be collapsing.  Similar conclusions as to the fluid-like behaviour of an orogen can also be reached for the crust and mantle of Myanmar and Indonesia, since here there is evidence for arc-normal extension adjacent to rolling-back subduction zones.   Both of these regions were recently affected by devastating earthquakes.  This raises the question as to  how stress build up due to gravity-driven fluid-like behaviour of an orogen interacts with stress build-up due to relative motion of the tectonic plates, a topic that can be addressed utilizing modelling and simulation software infrastructure being developed by the AuScope National Collaborative Infrastructure Strategy (NCRIS).

The 2004 Great Sumatran Earthquake was the result of motion on a gently dipping (~5-15°) giant megathrust that decoupled the Indonesian crust from the Indian slab. The rupture nucleated just to the north of the Simeulue cusp, and propagated over two entire structural domains of the arc.  However, failure occurred only to the north of the boundary between the Indian and Australian plates, i.e. sensu stricto affecting only the interface between the Indian plate and SE Asia.  This rupture therefore would have had the effect of increasing the loading on the adjacent (also shallow dipping) interface between the Indonesian crust and the Australian slab.  In consequence this interface subsequently failed, in the 2005 Great Earthquake at Nias.  This observation suggests that structural analysis can play a role in understanding the factors that trigger motion on the giant megathrusts responsible for these disasters, so we have begun to analyze the structure of the region in more detail.  Several datasets have been considered, but principally: a) Centroid Moment Tensor (CMT) data provided by the Global CMT project; and b) Shuttle Radar Topography Mission (SRTM) digital elevation data that can be portrayed so as to highlight fault traces that evident at the surface.  In addition we included specific observations as to fault geometry that have been made in the field, accurately located hypocentres in a dataset provided by Robert Engdahl, and tomographic imagery derived from the Widimantoro dataset.

Current models assume that earthquakes are driven by elastic energy stored as the result of relative motion of the tectonic plates.  Here we show that gravity-driven fluid-like flow of crust and mantle during afterslip also affects the outcome. Near Sumatra the aftershock sequence was dominated by faults produced by NW-SE compression, with slip close to parallel to the relative motion vector of the tectonic plates.  In the northern part of the rupture zone, however, things were quite different.  Aftershocks produced motion implying ~E-W arc normal stretching: a movement pattern that could only be driven by a westward surge of extending continental crust above a now weak basal detachment, perhaps defined by the initial rupture plane.  These competing movement patterns can be distinguished because the edge of the Indian plate is foundering, with slab roll-back in a direction orthogonal to its motion vector.
The Cretaceous/Tertiary geodynamic history of eastern Australia

R. Dietmar Müller(1), Lydia DiCaprio(1), Mike Gurnis(2), Kara Matthews(1) and Alina Hale(1)

1. EarthByte Group, School of Geosciences, The University of Sydney
2. California Institute of Technology, Pasadena, CA, USA

It is well known that the eastern half of the Australian Plate has been profoundly affected by Cretaceous/Tertiary geodynamic processes, leading to enormous changes in paleogeography, but early geodynamic models to investigate these processes were limited due to various simplifying assumptions involved and a lack of software infrastructures to efficiently link plate tectonics to mantle convection. We revisit this subject, using the finite element package CitcomS 2.2 to couple a regional high resolution model to the global mantle flow field and to GPlates-derived plate motions since the Early Cretaceous, thereby linking AuScope and Computational Infrastructure for Geodynamics (CIG) softwares. Our models have a Newtonian viscosity with dependencies on temperature, depth, composition and position. The viscosity of each model layer is varied to maximise model output match with observations, particularly mantle tomography and the tectonic subsidence history of the Eromanga and Surat basins derived from well data.  The early Cretaceous eastward drift of Australia over a subducting slab led to Cretaceous flooding in the Eromanga Basin area about 120 Ma, leading to deposition of fluvial and shallow marine sediments.  This episode was followed by uplift of eastern Australia, resulting in erosion and the establishment of the Ceduna Delta in the Great Australian Bight, where eroded Cretaceous sediments were re-deposited. Preserved Cretaceous Eromanga Basin sediments were subjected to intense sub-tropical weathering in a Late Cretaceous hothouse climate.  Australia's northward motion started in the Eocene, during a period of gradual global cooling and sea level fall, leading not only to a change in weathering regimes but also to major Miocene collisions in Tibet, New Zealand and Papua New Guinea, gripping Australia like a giant vice.  The associated mountain building episodes led to increases and re-orientations of regional Australian intraplate stress fields. We use published palaeo-stress models to quantify these Miocene stress changes in the Eromanga Basin, and suggest that they correlate with a well-documented period of gentle folding in the Eromanga Basin. Our conclusion is that a range of surface processes including major coastline change, sedimentation, erosion, folding, and faulting were ultimately driven by the interaction between mantle convection and plate tectonics.
Seismically Imaging the Australian Plate: ANU's Current Efforts and Future Directions

Sara Pozgay

Research School of Earth Sciences, The Australian National University

Both local and regional passive seismic array deployments have provided a suite of data for detailed seismic imaging of the Australian lithosphere.  Teleseismic tomography shows a clear transition between Proterozoic and Palaeozoic mantle lithosphere and ambient noise tomography details several deep sedimentary basins.  Current and future seismic deployments in the Gawler Craton, all of NSW, and surrounding the central Australian intercratonic suture zones will provide a wealth of data to further investigate continental dynamics.  Plate boundary imaging efforts include: a recent deployment in North Sumatra with UK & German colleagues to investigate different aspects of subduction zone segmentation; an upcoming deployment on a volcano in central Java designed to understand its structural, fluid, and magmatic components and to provide a suite of data for imaging the subduction system; and future efforts for imaging the subduction system near Rabaul in PNG.
The role of mantle and crustal detachments during lithospheric extension: implications for the evolution of the Australian plate

Gideon Rosenbaum(1), Klaus Regenauer-Lieb(2) and Roberto F. Weinberg(3)

1. School of Earth Sciences, The University of Queensland, Australia;
2. Department of Earth Sciences, The University of Western Australia, and CSIRO Exploration & Mining, Australia;
3. School of Geoscience, Monash University, Australia

We use numerical modelling to investigate the development of crustal and mantle detachment faults during lithospheric extension. Our models simulate a wide range of rift systems with varying values of crustal thickness and heat flow, showing how strain localization in the mantle interacts with localization in the upper crust and controls the evolution of extensional systems. Model results reveal a richness of structures and deformation styles, which grow in response to a self-organized mechanism that minimizes the internal stored energy of the system by localizing deformation at different levels of the lithosphere. Crustal detachment faults are well developed during extension of overthickened (60 km) continental crust, even when the initial heat flow is relatively low (50 mW/m2). In contrast, localized mantle deformation is most pronounced when the extended lithosphere has a normal crustal thickness (30-40 km) and an intermediate (60-70 mW/m2) heat flow. Results show a non-linear response to subtle changes in crustal thickness or heat flow, characterized by abrupt and sometime unexpected switches in extension modes (e.g. from diffuse rifting to effective lithospheric-scale rupturing) or from mantle- to crust-dominated strain localization. We interpret this non-linearity to result from the interference of doming wavelengths. Disharmony of crust and mantle doming wavelengths results in efficient communication between shear zones at different lithospheric levels, leading to rupturing of the whole lithosphere. In contrast, harmonious crust and mantle doming inhibits interaction of shear zones across the lithosphere and results in a prolonged rifting history prior to continental breakup. Our model results are discussed in the context of ancient and recent extensional processes in the Australian plate. For example, post-orogenic extensional processes in eastern Papua New Guinea resulted in the development of metamorphic core complexes in the D'Entrecasteaux Islands and the exhumation of Pliocene eclogites below extensional detachments. Unlike typical Basin-and-Range core complexes, the D'Entrecasteaux islands core complexes are underlain by mantle doming, indicating that the process of strain localization was not restricted to the crust.


Combining plate reconstructions, geodynamic models and seismic tomography to constrain the tectonic evolution of the Southwest Pacific

W. P. Schellart (1), B. L. N. Kennett (2), W. Spakman (3) and M. Amaru (3)

1. School of Geosciences, Monash University, Melbourne, VIC 3800, Australia
2. Research School of Earth Sciences, The Australian National University, Canberra, ACT 0200, Australia
3. Department of Earth Sciences, Utrecht University, Utrecht, The Netherlands

The Southwest Pacific region east of the Australian continent is tectonically complex and is home to numerous fossil and active subduction zones. There remain numerous geological controversies in the region regarding the polarity and continuity of fossil subduction zones in New Zealand and New Caledonia, origin of obducted ophiolites, presence of high-pressure metamorphism, occurrence of widespread Cenozoic magmatism, geochemistry of the magmatic and obducted rocks, and the potential disappearance of one or more ocean basins. Traditional models for the Cenozoic evolution of the Southwest Pacific region include two distinct subduction zones, a northwest-dipping subduction zone in New Caledonia and a southwest-dipping subduction zone in the Northland region. These models, however, fail to meet numerous spatio-temporal, geophysical and geological constraints. In a new tectonic model, the Northland and New Caledonia regions are interpreted to bear the hallmarks of a ~2500 km wide continuous northeast-dipping subduction zone along which the South Loyalty Basin was subducted in the Eocene-Early Miocene. In this new model the Eocene-Early Miocene volcanic activity along the Loyalty-Three Kings-Northland plateau seamount chain is interpreted as arc volcanism and its timing constrains subduction activity. Post-subduction and syn- to post-obduction Late Oligocene to Early Miocene volcanism in New Caledonia, Northland and the Norfolk Basin is interpreted as slab-detachment induced volcanism, and therefore constrains the timing of slab detachment. Such data thus constrain the longevity of subduction and the timing of slab detachment.

The geological and geochronological data have been incorporated into regional tectonic reconstructions and coupled to different “absolute” global reference frames. With these reconstructions, and insight obtained from geodynamic models of progressive subduction to constrain slab sinking velocities and sinking directions, we were able to predict the geographical location and depth of the detached South Loyalty slab in the mantle. Global S-wave and P-wave mantle tomography models have been used to see if the predictions fit with the regional tomography. Both tomography models identify a previously unrecognized isolated lower-mantle high-velocity anomaly located below the Tasman Sea at ~1100 km depth, striking NW-SE and with lateral dimensions ~2200 km by 600-900 km. It is found that high-velocity anomaly in the tomography models shows a good agreement with the predictions from the reconstructions in terms of geographical location, size, geometry and depth. The agreement between the plate reconstructions and mantle tomography provides strong support for the existence of a continuous 2500 km wide subduction zone and negates earlier models involving two distinct subduction zones. The agreement also provides constraints on the viscosity of the lower mantle (~10exp22 Pa s) and lower mantle sinking velocities (~1.5 cm/yr).
The motion and position of the Australian plate since the break-up of Pangea: implications for Australian plate margins

Maria Seton and R. Dietmar Müller

EarthByte Group, School of Geosciences, The University of Sydney

The past latitude and longitude positions of the Australian plate can be reconstructed using absolute plate motion models.  This is easily done for time after 100 Ma using hotspot tracks.  However, for times prior to 100 Ma we can either use palaeomag data but we don’t have constraints on longitudes or hotspots back to 130 Ma but not good constraints.

We use a “hybrid” reference in which a reliable moving hotspot frame for the last 100 Ma is linked to a True Polar Wander Corrected reference frame for earlier times. The stark difference between these two approaches is clear from Fig. 1c, which illustrates the location of the alternative set of subduction plate boundaries in relation to present coastlines at 130 Ma.

In addition, we test three alternative geomagnetic polarity timescales to examine the differences that timescales can play in reconstructing positions.  We use a new timescale of Gee and Kent (2007) which is an update of the Cande and Kent (1994) and Gradstein et . al. (1994) timescale used previously.

Our development of a hybrid absolute reference frame used GPlates AuScope Infrastructure and enables the comparison between alternative models and seamless integration with related data sets.

The implications for the location of the Australian plate particularly its eastern position.  Also for the direction and speed of convergence along the subduction boundary to the east.
Fluid movement in plate boundary deformation - lessons from New Zealand, Japan, and Cascadia

Rick Sibson

Department of Geology, University of Otago, P.O. Box 56, Dunedin 9054, New Zealand
rick.sibson@otago.ac.nz

Fault systems along divergent and convergent plate boundaries are loci for the most vigorous systems of fluid redistribution and fluid-rock interaction on Earth.  In turn, aqueous flow through fault systems, through a combination of physical and chemical weakening, contributes to the localisation of deformation along plate boundaries.  The stress state appears to play a critical role dictating whether flow systems in the upper crust are predominantly hot/cold hydrostatically pressured and driven by topography or convection, or involve episodic discharge from regions of fluid overpressuring.  These points are considered for several areas where high-resolution seismological and related geophysical data are yielding insights into active flow systems.

Massive convective circulation of hydrothermal fluids in the back-arc Taupo rift system, hosting the most active rhyolitic province on Earth, appears to extend throughout the 7-8 km deep seismogenic zone, suggesting that hot/cold hydrostatic pressure prevails throughout the rapidly (c. 1 cm/yr) extending brittle upper crust. In contrast, NE Honshu, Japan, is a magmatic arc under compression.  Tomography reveals flow paths of magmatic and hydrothermal fluids in the mantle wedge and within the crust.  Continued reactivation of steep reverse faults in the upper crust during ongoing compressional inversion is consistent with a range of seismological and/or electrical evidence for heterogeneous overpressuring of the mid-crust and lower seismogenic zone, locally to near-lithostatic levels.  This creates an environment conducive to episodic 'fault-valve discharge' with accompanying hydrothermal deposition.  Note that the margin of the Australian plate in the NW South Island of New Zealand is likewise an active inversion province with evidence for deep input of subcrustal fluids.   In the Cascadia subduction system, the association of episodic tremor (possibly within the mantle wedge) with episodes of aseismic slip down-dip from the locked portion of the subduction thrust interface seems likely to be associated with the transfer of slab-derived fluids into the overlying crust where bright-spot reflectors and high electrical conductivity also suggest the presence of trapped overpressured fluids in interconnected pore-space.

Our understanding of fluid redistribution is compromised by our practice of representing orogenic belts by 2D cross-sections combined with 2D modelling of fluid flow in the plane of the sections.  Future mineral discovery will require full comprehension of the 3D characteristics of structural permeability and hydrothermal flow systems that give rise to high-flux flow, and also their time-variance.


Epeirogenic uplift at the eastern margin of the Australian plate

T.A. Stern(1), W.R. Stratford(1,2) and M.L. Salmon(1,3)

1. SGEES, Victoria University of Wellington, New Zealand
2. Institute for Geography and Geology, University of Copenhagen, Denmark
3. RSES, ANU, Canberra, Australia

Vertical motions of the Earth’s surface profoundly affect our living environment yet remain poorly understood within the realm of plate tectonics. It is widely thought that at, or behind, subduction plate-boundaries, vertical motions are primarily due to the interaction between the two tectonic plates. An alternate view is that Rayleigh-Taylor flow-like instabilities in the viscous mantle of the overriding plate, can have an important effect, resulting in the sudden (~ 1 my) release of gravitational potential energy that accumulated over periods of 30-40 my. Late Cenozoic vertical movements, coupled to seismically determined lithospheric structure, in New Zealand provide a test of these contrasting mechanisms.

Teleseismic P-wave advances for earthquakes recorded on a dense array of seismographs in central South Island require high seismic velocities in the upper mantle beneath the Southern Alps. These advances are interpreted to represent an excess of cold, thickened, mantle lithosphere forming directly below the crustal root of the alps. In the last ~ 10 my. About 1400 m of suppressed elevation, or “negative dynamic topography”, for central South Island is calculated from the modelled excess mass of lithospheric thickening.

In contrast, the central and western North Island displays positive dynamic topography.   Here the land surface of the Australian plate stands at a relatively high elevation ( up to 1 km asl)  given its measured crustal thickness (~ 25 km), and seismic evidence points to the mantle lid of western and central North Island being attenuated or absent. Geological evidence points to a rapid uplift of this region ~ 5 Ma.  These contrasting, yet roughly coeval, vertical movements for two parts of New Zealand are difficult to explain within the context of plate tectonics. They can, nevertheless, be attributed to different stages of a common process — i.e. uniform thickening of the crust and mantle lithosphere (central South Island), then rapid, possibly convective, removal of the mantle lithosphere after ~15–30 my of shortening (western-central North Island).

We suggest that the eastern edge of the Australian plate contains a useful geological record, and active geophysical processes, from which we can make general advances in the understanding of epeirogenic uplift for continental regions.


Uplift of the Lord Howe Rise and formation of the New Caledonia Trough by detachment of lower crust during Eocene and Oligocene subduction initiation in the western Pacific

Rupert Sutherland (1), Julien Collot (2,3), Yves Lafoy(3), Graham A. Logan(4), Ron Hackney(4), Vaughan Stagpoole (1), Chris Uruski (1), Takehiko Hashimoto (4), Karen Higgins(4), Richard H. Herzer(1), Ray Wood(1), Nick Mortimer(5)

1. GNS Science, PO Box 30368, Lower Hutt 5040, New Zealand
2. Dep. of Geodynamics and Geophysics, IFREMER, Centre de Brest, B.P. 70, 29280 Plouzané, France
3. Service de la Géologie de Nouvelle Calédonie, Direction de l’Industrie, des Mines et de l’Energie de Nouvelle Calédonie, B.P. 465, 98845 Nouméa, New Caledonia
4. Geoscience Australia, GPO Box 378, Canberra, ACT 2601, Australia
5. GNS Science, Private Bag 1930, Dunedin 9016, New Zealand

We use seismic-reflection and rock-sample data to propose that the first-order physiography of the New Caledonia Trough and Norfolk Ridge formed in Eocene to Miocene time, and was associated with the onset of subduction and back-arc spreading at the Australia-Pacific plate boundary. Our tectonic model involves an initial Cretaceous rift that is strongly modified by Cenozoic subduction initiation and hence we are able to explain: complex sedimentary basins of inferred Mesozoic age; a prominent unconformity and onlap surface of Middle Eocene to Early Miocene age at the base of flat-lying sediments beneath the axis of New Caledonia Trough; gently-dipping, variable thickness, and locally deformed Late Cretaceous strata along the margins of the trough;  platform morphology and unconformities on either side of the trough that indicate a  phase of Late Eocene to Early Miocene uplift to near sea level, followed by rapid Oligocene and Miocene subsidence of c. 1100-1800 m; and seismic-reflection facies tied to boreholes that suggest absolute tectonic subsidence of coal beneath southern New Caledonia Trough by 2500-3100 m since Late Cretaceous time, substantially after the  phase of inferred Cretaceous rifting. The Cenozoic part of the model involves subduction initiation followed by rapid foundering of the subducted slab. This created a deep (>2  km) enclosed oceanic trough c. 2000 km long and 200 km across in Eocene and Oligocene time as the lower crust detached, with simultaneous uplift and local land  development along basin flanks. Disruption of Late Cretaceous and Paleogene strata was minimal during this Cenozoic phase and involved only subtle tilting and local reverse  faulting or folding. Basin formation was possible through the action of at least one  detachment fault that allowed the lower crust to either be subducted into the mantle or exhumed eastward into Norfolk Basin.  
Deep Fault Drilling Project — Alpine Fault, New Zealand: Active Deformation, Seismogenesis, and Mineralization in a Transpressive Plate Boundary Fault

John Townend

Institute of Geophysics, School of Geography, Environment, and Earth Sciences, Victoria University of Wellington, PO Box 600, Wellington 6005, New Zealand

The mid-crust is the locus of several fundamental geological and geophysical phenomena.  These include the transitions from brittle to ductile behavior, from unstable to stable frictional sliding, and from cataclastic to mylonitic fault rocks; earthquake nucleation and predominant moment release; maximum crustal stresses; and mineralization associated with fracture permeability.  Current understanding of deformation, seismogenesis, and mineralization in the mid-crust is based largely on remote geophysical observations of active faults and direct geological observations of fossil faults.

The Alpine Fault is a major dextral-reverse fault that is thought to fail in large earthquakes (c. Mw 7.9) every 200–400 years and to have last ruptured in 1717 AD.  Ongoing uplift has rapidly exhumed a crustal section from 20–30 km depths, yielding a recently formed (<~1 Ma), well-preserved sample of mid-crustal structures currently active at depth.
At a recent workshop funded by the International Continental Scientific Drilling Program (ICDP), 61 researchers from Australia, Canada, France, Germany, New Zealand, the United Kingdom, and the United States met to discuss the significance and feasibility of a multi-national program of Alpine Fault drilling and allied science. 

During the course of the workshop, three main scientific themes and associated research goals emerged:  (1) evolution of a transpressive orogenic system; (2) ductile and brittle deformation mechanisms, and their interaction; and (3) seismogenesis and the habitat of earthquakes.  The remarkable along-strike homogeneity of the Alpine Fault’s hanging wall, the rapid rate of slip, and the dextral-reverse kinematics enable us to examine the progressive evolution of fault zone materials by linking rocks exposed at the surface to their in situ protoliths at depth along common exhumation trajectories.  In the central portion of the fault, where exhumation rates are highest (6–9 mm yr–1), several sites can be identified at which rocks encountered at depth in boreholes would correspond to well-studied outcrops, enabling progressive geological deformation and petrological changes to be studied as functions of space and time along the transport path.  Moreover, drilling may be able to target the critical phenomena highlighted above more readily than in locations where uplift rates are lower and the processes occur at greater depths.

This presentation reviews the opportunities the Alpine Fault provides to relate real fault rocks and in situ measurements to earthquake rupture models, and to provide analogues for mesothermal mineralization environments.  What will be crucial to the scientific success of a long-term drilling program is coordinated research that puts drilling results in context, such as detailed hydrogeological and thermal data collection and modeling, and more extensive analysis of thermochronological and fluid/gas chemistry data.
Correlation of ore-forming magmatism with geometry of continental lithosphere underneath New Guinea from seismic tomography
 
Michiel van Dongen(1), Wim Spakman(2), Roberto Weinberg(1)

1. Monash University, School of Geosciences, Clayton, VIC, Australia;
2. Utrecht University, Department of Geosciences, Utrecht, Netherlands

Email: m.vandongen@uq.edu.au

New Guinea forms part of the northern convergent margin of the Australian plate. We integrate 56 seismic tomography sections that characterise the deep seismic structure of New Guinea with the Late Miocene to present-day tectonic evolution to understand the nature and geographical distribution of magmatic rocks and their associated ore deposits. We correlate the seismic structure with the source components of post-collisional magmatism, as indicated by isotope data. In western New Guinea, where cold continental lithosphere underlies the island, intermediate magmatic rocks occur and they have a mixed mantle and crustal isotopic signature. Mantle-derived mafic intrusions and lavas occur in the eastern part of New Guinea where continental lithosphere is absent and hotter lithosphere underlies the island. This relationship confirms that magmatic interaction with cold lithosphere changes the geochemical signature of mantle-derived magma. Ore deposit distribution in New Guinea shows a broad correlation to the mantle structure as well, with giant porphyry Cu-Au deposits such as Ok Tedi and Grasberg above the edge of cold continental lithosphere and giant Au-only deposits such as Porgera and Lihir above hot lithosphere. The mechanisms that could explain this distribution require an analysis of yet unconstrained factors, one of which is asthenospheric mantle contribution to Au-enrichment.
Dynamic rheology and strength of the lithosphere

Roberto F. Weinberg(1), Klaus Regenauer-Lieb(2) and Gideon Rosenbaum(3)

(1) School of Geosciences, Monash University, Australia; (2) Department of Earth Sciences, The University of Western Australia, and CSIRO Exploration & Mining, Australia; (3)School of Earth Sciences, The University of Queensland, Australia;

The strength of the lithosphere is the integrated force required to cause deformation at a given rate. It has previously been assumed that continents subject to deformation are weaker when they are hotter. We argue that it is not the steady-state heat flow of continents that control their strength, but time-dependent feedback effects, triggered by shear heating and thermal expansion. These effects localize strain into weak shear zones which control the time-dependent strength of continents. Here we present numerical results showing that a cold and strong continent is substantially weakened by development of intensely localized shear zones. In contrast, weakening effects are less efficient in an initially warmer continent where shear zones are more diffuse. This leads to self-organization of the dissipative structures, i.e. the width, length, distribution and heat generation of shear zones. As a result, regardless of initial temperature profiles and crustal thicknesses, all modelled continents yield similar time-dependent strengths that follow similar temporal evolution defining a lithospheric strength attractor. An important implication is that even cold and deeply-rooted Archaean cratons may be vulnerable to deformation as evidenced by a number of cratons that were rifted apart during the break-up of Gondwana.









 
Earth Modelling Institute
Contact Details

info@geodynamics.monash.edu.au
t: (+61 3) 9905 4414
f: (+61 3) 9905 4403