an integrated multi-scale 3d seismic model of the archaean...

(2006) 75–90www.elsevier.com/locate/tecto

Tectonophysics 420

An integrated multi-scale 3D seismic model of theArchaean Yilgarn Craton, Australia

B.R. Goleby a,b,⁎, R.S. Blewett a, T. Fomin a,b, S. Fishwick c, A.M. Reading c,P.A. Henson a, B.L.N. Kennett c, D.C. Champion a, L. Jones a,d,

B.J. Drummond d, M. Nicoll a

a Predictive Mineral Discovery CRC, Geoscience Australia, PO Box 378, Canberra, ACT, 2601, Australiab Australian National Seismic Imaging Resource (ANSIR), Geoscience Australia, PO Box 378, Canberra, ACT, 2601, Australia

c Research School of Earth Science, Australian National University, Canberra, ACT, 0200, Australiad Geoscience Australia, PO Box 378, Canberra ACT, 2601, Australia

Received 8 April 2005; received in revised form 30 September 2005; accepted 4 January 2006Available online 11 May 2006

Abstract

The collection of a range of different seismic data types has greatly improved our understanding of the crustal architecture ofAustralia's Archaean Yilgarn Craton over the last few years. These seismic data include broadband seismic studies, seismicreceiver functions, wide-angle recordings and mine-scale to deep seismic reflection transects. Each data set provides information onthe three-dimensional (3D) tectonic model of the Yilgarn Craton from the craton scale through to the mine scale. This paperdemonstrates that the integration and rationalisation of these different seismic data sets into a multi-scale 3D geological/seismicmodel, that can be visualised at once in a single software package, and incorporating all available data sets, significantly enhancesthis understanding. This enhanced understanding occurred because the integrated 3D model allowed easy and accurate comparisonof one result against another, and facilitated the integrated questioning and interrogation across scales and seismic method. As aresult, there are feedback questions regarding understanding of the individual seismic data sets themselves, as well as the YilgarnCraton as a whole.

The methodology used, including all the data sets in the model range, had to allow for the wide range of data sets, frequenciesand seismic modes. At the craton scale, P-wave, S-wave and surface wave variations constrained the 3D lithospheric velocitymodel, revealing noticeable large-scale velocity variations within and across the craton. An interesting feature of the data, easilyidentified in 3D, is the presence of a fast S-wave velocity anomaly (>4.8 km s−1) within the upper mantle. This velocity anomalydips east and has a series of step-down offsets that coincide approximately with province and terrane boundaries of the YilgarnCraton.

One-dimensional receiver function profiles show variations in their crustal velocity across the craton. These crustal velocityvariations are consistent with the larger-scale geological subdivision of the craton, and provide characteristic profiles for provincesand terranes. The receiver function results and the deep seismic reflection data both agree on the depth to the Moho, and bothindicate an increase in Moho depth to the east. The 2D seismic refraction results in the south-west of the craton provide crustalthickness information, an indication of middle and lower crustal compositions, and information regarding the broad-scalearchitectural framework.

At the province- and terrane-scale, the deep seismic reflection data and the mine-scale seismic data provide geometricconstraints on crustal architecture, in particular the orientation of the region's fault systems as well as variations in the thickness of

⁎ Corresponding author. Predictive Mineral Discovery CRC, Geoscience Australia, PO Box 378, Canberra, ACT, 2601, Australia.E-mail address: [email protected] (B.R. Goleby).

0040-1951/$ - see front matter. Crown Copyright © 2006 Published by Elsevier B.V. All rights reserved.doi:10.1016/j.tecto.2006.01.028

mailto:bruce.�[email protected]

http://dx.doi.org/10.1016/j.tecto.2006.01.028

76 B.R. Goleby et al. / Tectonophysics 420 (2006) 75–90

the granite–greenstone succession. Integration of the results from wide-angle seismic refraction data coincident with the deepseismic reflection data provided additional constraints on likely upper crustal lithologies.

The integrated 3D seismic model implies the dominant geodynamic process involved the development of an orogenic belt thatdeveloped with a series of contractional (folding and thrusting) events, separated by equally important extensional events. Theseismic reflection data in particular suggests that extensional movement on many shear zones was more common than previouslythought.

The seismic reflection data suggest that the dominant mineral systems involved deeply sourced fluid flowing up crustal-penetrating shear zones. These deeply sourced fluids were further focussed into sites located above fault-breached domal regions inthe upper crust.Crown Copyright © 2006 Published by Elsevier B.V. All rights reserved.

Keywords: 3D seismic model; Seismic reflection studies; Tomography data; Receiver function data; Yilgarn Craton; Crustal architecture; Archaean

1. Introduction

The integration (merging) and simultaneous visuali-sation of different seismic data sets, each with its ownscale, different frequencies, different wave-modes anddifferent dimensions is the easy part of constructing a3D seismic model of an area. The interpretation of thismulti-seismic data volume is complex. The process ofintegration of crustal-scale and high-resolution 2Dseismic reflection profiling with 2D crustal scale P-and S-wave velocity models with 1D receiver functionvelocity models and with low-resolution 3D tomogra-phically inverted velocity models becomes a function ofthe resampled resolution of each data set and theconstraints chosen. If these constraints are chosencorrectly then a self-consistent 3D velocity model willbe constructed. When integrated with other geologicalunderstanding (e.g., geological mapping, geochemistry,structural geology, numerical modelling, etc.), the resultis an improved 3D geological/tectonic model of theregion.

The seismic studies available and used for this modelof the Yilgarn Craton include seismic reflection data(Goleby et al., 1993, 2000, 2003), seismic refractiondata (Dentith et al., 2000), portable short-period wide-angle data (Fomin et al., 2003), receiver functionanalysis (Reading et al., 2003, in press) and broadbandseismic experiments (Kennett, 2003; Fishwick et al.,2005). In addition to the regional coverage, this studyfocused on data sets from within the Eastern GoldfieldsProvince, the more mineralised eastern third of theYilgarn Craton (Fig. 1).

Individually, each seismic data set has provided newinsights into the crustal architecture or the velocitystructure of the region, however, when all the seismicdata sets are integrated, the results provide an evenlarger increase in understanding of the structure and

likely geodynamics of the crust and lithosphere of theYilgarn Craton. Taken together, these data enhance theknowledge required for area-selection by the explora-tion industry and the prediction of new mineraloccurrences.

By combining both architectural information fromthe 2D seismic reflection data sets with the 2D and 3Dvelocity information from the 2D and 3D refraction andtomographic data sets, then extending all to 3D, morecan be inferred about the crustal geological variationsand hence tectonic processes that have operated withinthe craton. The three areas where the biggest impact hasoccurred is in improving our understanding andassessment of the crustal architecture, understandingstructural controls on crustal history and understandingrelationships between deformation processes, timingand mineral systems, which all impact on reducing riskin exploration.

In particular, the regional deep seismic reflectiontraverses within the Eastern Goldfields Province havesuccessfully imaged the prominent crustal-scale geo-metric features (Fig. 2; Goleby et al., 2003) and haveshown the potential of the method to assist in thedevelopment of a mineral systems model for the region.Coincident with the seismic reflection traverses, a wide-angle reflection profile, using the same energy source(Fomin et al., 2003), provided additional information onupper crustal velocity variations.

P- and S-wave data have provided information on the3D velocity structure within the lithosphere of theYilgarn Craton. These data, when integrated with bothnear-surface geological and geophysical data, allowedinvestigation of the tomographically determined veloc-ity structure of the granite–greenstone succession withinthe Yilgarn Craton. A craton-wide receiver functionanalysis, using the same S-wave data set, was used toinvestigate the crustal–uppermost mantle Vs velocity

Fig. 1. (a) Map of Australia showing location of the Yilgarn Craton. (b) Map of the Yilgarn Craton showing subdivision into different provinces,terranes and domains, with location of the deep seismic reflection traverses. (c) Map of the Yilgarn Craton showing location of different seismicexperiments undertaken during the last few years. Dots are legacy broadband recording sites (BRS) recorded by the Research School of EarthSciences (Australian National University). Small and large stars refer to the recent BRS sites. Thick dashed line is the location of the wide-angleseismic survey. Seismic reflection traverses (regional- and mine-scale) are shown as thin lines. Grey dots are refraction stations.

77B.R. Goleby et al. / Tectonophysics 420 (2006) 75–90


structure of the region. This data set showed significantvariation within the different crustal blocks of theEastern Goldfields Province.

When integrated and visualised together, thesevarious seismic data sets allowed the construction of awell-constrained 3D velocity/geological model of theregion. The seismic reflection data provided the 2Dcrustal architectural constraints along several keyregional seismic transects. The seismic tomographydata sets, at a range of scales, were used to infill themodel at both crustal and lithospheric scales. This infillprocess was important in understanding the region'sdistribution of mineral deposits within a 3D crustalarchitecture framework.

The basic philosophy employed in this 3D modelconstruction fell into two methods, dependent on theavailable data. The first and most reliable was at thecrustal scale where there were more depth constraints,the second was at the lithospheric scale where onlyteleseismic data sets were available. At the crustal scale,the architectural constraints, primarily obtained from theseismic reflection sections, were extrapolated outwardand linked with distant reflection sections to constructthe key structural elements of the model space. Thesurface location of these major structural elements,primarily crustal shears, and the orientation of thedeeper-seated geophysical anomalies was fundamentalin this extrapolation process. In general, we chosesmoothly varying and continuous model parameterswithin similar geological terranes, but allowed suddenmodel parameter changes across crustal shears.

The principal software package used in the visualisa-tion and construction of the Yilgarn Craton is thecommercially available software GOCAD (http://www.earthdecisionsciences.com/). In addition, several surfacebuilding and gridding routines were used to manipulateand reformat the incoming data sets. The software wasnot limited by large object sizes, so resolution (cell size)of the various data sets was kept as detailed as possible,given the limitations and resolution of that particularseismic method. This allowed zooming into the modelto check and inspect the finer details in a given area. Inthe process, we did not assign any weights to one dataset over another, rather we visualised all the data we had.

Fig. 2. (a) Interpretations of a composite traverse of the 1991 Eastern Goldfietraverse showing prominent crustal-scale features imaged by the deep seismicThis composite section shows four sub-horizontal layers; the lowest is the mathen a mid-crustal layer containing large scale lozenges and a more compapproximately 630 km in length by 45 km in depth. (b) Detailed view of thtraverse, across the Leonora Mineral Field, showing drill hole locations, seismmiddle part of the 2001 northeastern Yilgarn Craton seismic traverse, acrospossible fluid flow paths.

During interpretation, however, the accuracy andprecision of one data set over another was taken intoaccount. This allowed the comparison of 1D velocityfunctions or borehole logs, detailed 2D sections fromreflection profiling, and sparse but even 3D velocitymodels, both in making the generalised velocity modeland in its geological interpretation.

2. Geology

In developing the 3D geological model of the YilgarnCraton, an understanding of the surface geology, inparticular key structural elements, including shearzones, terrane boundaries, large granitic areas, basementgneissic areas etc, was fundamental. Without thisknowledge, extrapolation of features from one area toanother and across one boundary to another could not beconstrained. The 3D model of the Yilgarn Craton wasconstructed from a scale of 1 :250,000. This chosenscale allows the essential geological packages and keystructures to be defined, and importantly displayed andinterrogated without confusing the model with largerscale (detailed and less relevant) complications. Thescale needs to be set so all the key features involved inthe main tectonic processes are included while the‘noise’ is filtered.

At the scale of the model, the geology of the YilgarnCraton is simplified into a region that is dominated bylarge volumes of Archaean granites and greenstones(Fig. 1). These granites and greenstones define a north–south tectonic grain to the craton, made up a regionalnorth–south fault pattern and the location of north–south elongated granite batholiths into regional north–south antiforms. The north–south faults define tecto-nostratigraphic belts, and subdivide the craton intofault-bounded provinces, terranes and domains basedon similar and differing geological characteristics (Fig.1). The province-scale subdivisions include an eastern-most Eastern Goldfields Province, a central SouthernCross Province and a western Murchison Province(Myers, 1995; Barley et al., 2003).

The Eastern Goldfields Province, the craton's mostmineralised province, has a broad lithostratigraphy(Swager et al., 1997; Barley et al., 2003) that has been

lds seismic traverse and the 2001 northeastern Yilgarn Craton seismicreflection studies. The raw seismic sections are displayed underneath.

ntle, above which are a thin non to poorly reflective lower-crustal layer,lex upper crustal granite–greenstone layer. The composite section ise western most part of the 2001 northeastern Yilgarn Craton seismicic interpretation and location of the SOG mine. (c) Detailed view of thes the Laverton Mineral Field, showing the seismic interpretation and

http://www.earthdecisionsciences.com/

http://www.earthdecisionsciences.com/


used to further subdivide the province into a number ofterranes (e.g., Kalgoorlie, Kurnalpi, Laverton, Duketon;Fig. 1) and in several cases, further into discretedomains (Swager and Griffin, 1990; Swager et al.,1992; Swager, 1997). This broad greenstone successionlithostratigraphy consists of komatiites, basaltic rocks,felsic volcanic rocks and sedimentary rocks; however,this lithostratigraphy cannot be correlated betweenterranes (Swager et al., 1997).

The craton's granites have been have subdivided byChampion and Sheraton (1997) into five principalgranite types; namely: high-Ca, low-Ca, mafic, highhigh-field strength elements (Hi-HFSE) and syeniticgranites. The high-Ca and low-Ca granites dominate thecraton by area, with 60% high-Ca granites and 20% low-Ca granites. The high-Ca granites were intruded over aprolonged period terminating at ∼2660 Ma. Champion(1997) interpreted these high-Ca granites as lower-crustal to upper-mantle melts developed during subduc-tion. In contrast, the low-Ca granites were emplacedtowards the end of orogenesis (2655 Ma to 2630 Ma)across the entire craton (Cassidy et al., 2002). The low-Ca granites are high-temperature crustal melts, inter-preted to have developed during delamination of thelower crust (Smithies and Champion, 1999).

Structurally, the deformation history for the YilgarnCraton is complex and long-lived (e.g. Cassidy andChampion, 2004). Deformation of the Eastern Gold-fields Province has been generalised from that deter-mined for the well-studied, gold-rich, Kalgoorlie terrane(Swager, 1997; Blewett et al., 2004a,b). The north–south, structurally bounded, regional-scale belts, formdistinct structural terranes (Swager, 1997) that areconsidered to reflect a sequence of extensional andcontractional deformational events. This deformationhistory, as summarised by Blewett et al. (2004a),commenced with an early extensional event that wasfollowed by a series of episodic contractional andextensional episodes. The first of these contractionalevents was north–south orientated recumbent foldingand thrusting during a D1 shortening event. This wasfollowed by a period of large scale upright folding duringthe east northeast–west southwest widespread contrac-tional and extensional events that evolved episodicallyand rapidly, with a diachronous series of approximatelycoaxial switches in tectonic mode during a series of D2

folding and thrusting events (Blewett et al., 2004a). D3

deformation was minor and occurred as north–southorientated strike–slip faulting and associated faulting.The deformation history ended with localised transpres-sive oblique and reverse faulting during D4 deformation(Swager, 1997; Blewett et al., 2004a).

The boundaries to each province, terrane and domainare mapped as regional scale shear zones. The deepseismic reflection images cross most of these boundariesand in most cases has successfully imaged the dip, depthpenetration and relationship to nearby faults and shearzones. The seismic does indicate which faults or shearzones cut which others, however, it does not indicateany timing information about when the fault movedthrough time. This information is provided through thetransferring of the deformation history understanding toeach fault or shear zone during the construction andinterpretation of the 3D model.

The cumulative effect of this deformation history hasresulted in the Eastern Goldfields Province having apronounced north–northwest–south–southeast fabric orregional strike that is seen in both the orientation of thegranite and greenstone units as well as the orientation ofthe major shear zones (Fig. 1). This has led workers(e.g., Blewett et al., 2004a) to conclude that, in theeastern Yilgarn Craton, and perhaps for the wholecraton, the preferred tectonic environment is a subduc-tion setting, possibly a long-lived accretionary margin(Barley et al., 2003).

3. Seismic data sets input into the 3D model

3.1. 2D seismic reflection traverses

The network of regional deep seismic reflectiontraverses recorded within the Eastern Goldfields Prov-ince crossed a number of regionally significant terraneboundaries and internal shear zones (Figs. 1 and 2).These include the regional crustal-penetrating shearzones such as the Ida, Ockerburry, Hootanui andYamarna Shear Zones (Fig. 1) as well as shear zoneslimited to shallow depths such as the Bardoc, Keith-Kilkenny and Celia Shear Zones. In addition, theseismic imaged a series of crustal-scale featuresincluding a change in the thickness of the crust acrossthe Eastern Goldfields Province, a subdivision of thecrust into three broad layers, an east dip to the majorityof the reflections, and the presence of a series of east-dipping crustal-penetrating shear zones. Each of thesefeatures was interpreted and input into the 3D model.

The deep seismic reflection data reveal an EasternGoldfields Province crust with a fabric that has apronounced 30° easterly dip. Within this east-dippingfabric, there are a series of east-dipping reflective bandsthat can be traced from the surface to the lower crust andin several cases to the Moho. These represent theprovince's terrane boundaries (Ida, Ockerburry, Hoota-nui and Yamarna) as well as numerous internal shear


zones which all divide the province into a series ofdistinct domains. The Ida, Hootanui and Yamarna ShearZones are clearly imaged, whereas about 9 km of theOckerburry Shear Zone, from 3 km to 12 km depth,appears to have been intruded by later graniteemplacement and/or affected by subsequent deforma-tion. All shear zones have complex internal geometrythat suggests that these shear zones have been deformedseveral times.

There are several locations where west-dippingreflections have been imaged, some show linkage witha crustal penetrating shear zone. The most important ofthese is the Bardoc Shear Zone, a highly mineralisedzone containing numerous operating mines. The BardocShear Zone and the crustal-penetrating Ida Fault form asuitable fluid-focusing mechanism that directed fluidsinto the Bardoc Shear Zone (Goleby and Drummond,2000). Fluid flow modelling (Hobbs et al., 1997)supports this hypothesis.

The three distinctly different sub-horizontal layers(Fig. 2) have different seismic characteristics andprovide constraints on the evolution of the craton.There is a broad correspondence between these layersand the receiver function data. The lowest layer, thelower crust, is essentially devoid of reflections otherthan several bands of east-dipping reflections thatextend through to the base of the crust. These east-dipping reflections have been interpreted as representingthe location of deep penetrating shear zones. The uppersurface to this lowest crustal layer is marked by a suddenchange in reflection character to a zone where overlyinglarge packages of dipping reflections are typical (Fig. 2).Seismic reflectivity in the middle crust is characterisedby numerous prominent east-dipping reflections thatdefine the boundaries to large-scale, lozenge-like, mid-crustal bodies. Drummond et al. (2000) describedduplex structures in the mid crust of the 91EGF01seismic traverse related to D2 thrusting and imbrication.Similar features are evident in segments of the01AGSNY1 seismic traverse (Fig. 2). One exceptionto this pattern is within the middle crust of theKalgoorlie region, where an anomalous region of lowreflectivity has been interpreted to represent eitherextensive deformation or large scale alteration withinthe area (Drummond et al., 2000). The boundarybetween this middle layer and the upper layer is diffuseand irregular. The upper layer is complex and variable inits seismic character, reflecting complexities within thegranite–greenstone succession.

Mine-scale seismic traverses, recorded within theKalgoorlie, Leonora and Laverton mineralised regions,were undertaken to define mine scale architecture and

developing three-dimensional structural models of thetop 5–10 km of the mineralised sequences within thegranite–greenstone succession (Cassidy et al., 2003;Stolz, 2003; Beeson et al., 2004). Integration of theresultant seismic reflection sections with the mine scalegeological and drilling data has produced 2D crosssections and 3D models through the ore rich mineralisedzones, which the exploration company has used toidentify drilling targets.

These seismic data and its interpretation was inputinto the 3D model in two forms, images of the seismicsections and lines representing the various structures,including lines representing the shear zones, theboundaries to the various layers, lithological markerhorizons and outlines to the upper crustal granite bodies.These lines were then extrapolated outwards from thetraverse, using the surface geology to link with thesame/similar features identified on the other seismictraverses. This process formed surfaces that defined thevarious provinces, terranes and domains, as well as theinternal structure within each of these. There was noattempt made to parameterise these surfaces other thansome identifying colour scheme. As discussed above, in3D there has to be a stage during construction of eachsurface regarding which fault cuts what surface. Oncethis has been determined, the implications of thisdecision on the whole model need to be determinedand resolved.

3.2. 2D wide angle seismic profile

Wide-angle data, recorded in conjunction with theacquisition of the 2001 deep seismic reflection surveywithin the Eastern Goldfields Province providedadditional velocity information on the top 10 km ofthe granite–greenstone succession (Fig. 3). These datareveal the presence of a higher-velocity region beneathone of the regions geological domes (Fomin et al., 2003,2004; Fomin and Goleby, 2005). The higher velocities(6.5–6.7 km s−1, top at 3 km depth) are restricted to a∼2 km thick body located to the western part of thewide-angle experiment (Fig. 3). This higher-velocitybody is interpreted to represent mafic rocks, occurringas sills or flows, in a poorly mineralised area within thegreenstone succession to the west of Laverton MineralField, (Fomin et al., 2004). This body is underlain bylower velocity rocks (6.0–6.1 km s−1), most likelygranite–gneissic in composition (Fomin et al., 2004).

The higher-velocity body appears to be relativelytransparent on the deep seismic reflection image, asthere are far fewer reflective horizons interpreted in thisarea compared with the eastern and western flanks of the

Fig. 3. The wide-angle data, plotted with the deep seismic reflection interpretation to show similarities and differences between the two methods.Colour scheme shown in insert. From Fomin and Goleby (2005). Wide-angle survey was coincident with the regional reflection transect(01AGSNY1).


line (Goleby et al., 2004; Fomin et al., 2004). Such acorrelation is suggestive of a relatively smooth, longwavelength velocity effect, probably related to meta-morphism associated with the formation of the green-stone belt that significantly altered the velocitydistributions within the internal structure of the area(Fomin et al., 2004). This can also explain why steeplydipping reflectors, well imaged to the east and west ofthe line, are not imaged in this region.

The boundary separating this high-velocity layerfrom the low-velocity layer below it is possibly acompositional boundary between greenstones andunderlying felsic gneisses. There is no evidence forhigh-velocity material below this boundary. Assumingthe Moho is associated with the deepest reflectionsmodelled, total crustal thickness in the region can bespeculatively estimated to be in the range of 32–37 km,consistent with both the deep seismic reflection data andthe receiver function analysis.

3.3. 1D receiver function studies

Receiver functions results, calculated from theteleseismic arrivals, not only cover the entire cratonbut are also concentrated within the Eastern GoldfieldsProvince. This focus on the Eastern Goldfields Provincewas part of an ongoing experiment to investigate grossvelocity differences between mineralised and non-mineralised terranes that would be attributed to the

crustal scale changes associated with the presence of thegold mineralising system (Figs. 1 and 4). Reading et al.(2003, in press) and Reading and Kennett (2003) presentmuch of the recent receiver function data available onthe Yilgarn Craton. Their results indicate a pronouncedvariation in crustal and upper mantle velocity structureacross the craton (Fig. 4), with each province showingcharacteristic velocity structures that are internallyconsistent within each province (Reading et al., 2003,in press). This variation correlates with the mappedgeological provinces.

The receiver function results show the Mohodeepening eastward across the craton (Reading etal., 2003). Reading et al. (in press) proposedextensive reworking of crustal rocks, resulting inthe lower crust becoming more felsic and thereforelower in seismic velocity, as the mechanism for theformation of the observed sharp Moho discontinuityobserved beneath parts of the craton. Given this, therewould be a corresponding increase in the Vp/Vs ratioin the region of the lower crust. It is therefore likelythat the crustal architecture seen across the provincewas ‘frozen in’ for each province prior to theassembly of the craton.

The receiver function data was input into the 3Dmodel as single 1D velocity laths, with their thicknessand colour representing the changes in Vs velocity withdepth. Given that this data set is 1D and the results arevery dependent on the data used in the received function

Fig. 4. 3D image of receiver function results, plotted as velocity ‘laths’, with the thickness and colour variations representing S-wave velocities.Numbers refer to receiver function site. Lines refer to main faults and craton boundary. The S-wave velocity colour bar used in displaying the ‘laths’shown.


inversion, this type of visualisation is an appropriatemeans of representing the data. It allows a rapidexamination of the results themselves and variationsacross the model without being too detailed ordescriptive. In the current 3D model, no attempt hasbeen made to contour these data; rather the model isused to identify craton scaled variations. The onesurface constructed from the receiver functions data isthe depth to the Moho as this is the most precise part ofthe data set. There was no attempt adding parameters toeither the laths or the Moho. In the model, the receiverfunctions successfully locate the subsurface extent ofseveral of the crustal-penetrating shear zones. Thewestern boundary to the Eastern Goldfields Province,the Ida Fault (Fig. 1), is defined by the presence of aregion of higher-velocity lower crustal material withinthe footwall of this crustal penetrating fault.

3.4. 3D seismic tomography

Teleseismic S-wave data from distant earthquakesshow that the lithosphere beneath the Yilgarn Craton hasa low velocity crust underlain by mantle with a S-wavevelocity faster than the world average (Kennett, 2003;

Fig. 5). Fishwick et al. (2005) confirms the existence ofa region of fast S-wave material beneath the YilgarnCraton to at least 250 km depth. They also show that thisfeature, although not continuous, has slower velocitiesin the central Yilgarn and faster zones at the western andeastern edges of the craton.

The sparse low-resolution 3D tomographicallyinverted S-wave velocity data were input into the 3Dmodel as a 3D grid, represented by an array of velocityvalues within a regular 3D grid. This data set was thencapable of being sliced through each axis as well asbeing 3D contoured to produce a series of iso-surfaces.Both displays were equally important as both were usedduring interpretation.

The 3D model of the teleseismic S-wave data showsa higher-velocity, southeast dipping body approximately70 km beneath the Murchison Province, increasing toapproximately 100 km beneath the Ida Fault, then toapproximately 120 km (Fig. 5). Although the spacing ofthe sample points, and hence the gridding of theteleseismic S-wave volume is coarse compared to theother data sets, the ∼50×50×50 km3 grid size showsthat there is an indication that this body is not just asingle southeast dipping body but rather it is broken into

Fig. 5. (a) Image of the 3D S-wave speed structure of the lithosphere beneath Australia at a depth of 75 km. Red/brown colours indicate S-wavevelocities slower than the world average, blue colours indicate S-wave velocities faster than the world average. From Kennett (2003). (b) 3D Image ofthe lithosphere beneath the Yilgarn Craton, looking northwards from a depth of approx. 100 km. The velocity spectra changes colour from green toblue at the Moho. Two of the regional crustal penetrating shear zones, the Ida Fault and the Yamarna Fault, are shown as east dipping white-colouredsurfaces. Pink colours indicate S-wave velocities faster than the world average; blue colours indicate S-wave velocities slower than the world average.The yellow line is the surface outline of the Yilgarn Craton; purple lines are the main province subdivisions.


a series of segments. The 3D model shows a first orderspatial correlation between the subsurface location ofthese breaks and the location of main provinceboundaries mapped within the craton.

4. 3D seismic features of the Yilgarn Craton crust

The seismic techniques described above subdividethe Yilgarn crust into a series of similar regions, each

characterised according to the following seismicfeatures of the crust (Goleby et al., 2003; Blewett etal., 2004b). These features include:

• the presence of a higher-velocity, southeast dippingbody within the mantle lithosphere beneath theYilgarn Craton,

• this mantle lithospheric body is not a simple dippingbody, rather is complexly shaped,


• each province, and in several cases terranes, has itown characteristic crustal seismic velocity structure,that can be used to identify that terrane or province,

• the Moho dips to the east and implies a change in thethickness of the crust across the craton,

• the crust can be subdivided into three distinct broadlayers,

• a pronounced east-dip to the fabric within the crustand mantle, best imaged by the majority of theseismic reflections,

• the presence of a series of east-dipping crustal-penetrating shear zones that appear to sole into thebase of the crust; and

• the observation that the seismic reflection techniqueis imaging district scaled structures associated withmineralised structures.

All these features are visible in the 3D model andgiven most are ‘imaged’ by more than one technique;their reliability and our confidence in the interpretationof the craton are enhanced. Fig. 6 shows two ‘snapshots’ of the 3D model illustrating several of thesefeatures. The following observations are taken frominterrogation and inspection of the integrated 3D model.

4.1. Yilgarn Craton crustal structure

There is a pronounced easterly dip to the fabric ofthe Yilgarn crust. The majority of reflectivity observedalong the seismic traverse, both within the middle andupper levels of the crust typically dips eastwards at 30°(Figs. 2 and 6a). West-dipping features are imaged butthese are far less common. The crust is subdivided intothree sub-horizontal layers. Many of the receiverfunction sites support this simple subdivision of theEastern Goldfields Province crust into three distinctsub-horizontal seismic layers as well as supporting adivision of the craton into at least three differentgeological provinces; a difference that exists through-out the entire crust.

The upper crustal layer, with its granite–greenstonesuccession, is complex and variable in its seismiccharacter. The lozenge-like bodies imaged within themiddle crust suggest large-scale compressional defor-mation, that at one time was coupled to upper crustaldeformation but later became decoupled, though theevidence suggest that there was not large amounts ofrelative horizontal movement. The ductile lower crustalis devoid of reflectivity with the exception of threeinstances where dipping reflections are recorded withinthis layer and all relate to the location of the interpreteddeep-penetrating shears zones.

4.2. Yilgarn Craton Moho depth (implied crustalthickness)

The depth to the Moho has been determined fromthree different seismic data sets, deep seismic reflection,seismic refraction and receiver function analysis. In theEastern Goldfields Province, the deep seismic reflectiontraverses has imaged the Moho and along-traversevariations in this boundary. The deep seismic data showsthe Moho as a prominent thin, sub-horizontal band ofreflections. The Moho in this area is interpreted to be atabout 36 km depth beneath the Ida Fault on the91EGF01 traverse (Goleby et al., 2000), 40 km beneaththe town of Leonora and about 46 km at the eastern endof the 01AGSNY1 traverse (Goleby et al., 2004). Thisdeepening of the Moho is achieved through a series oframps and flats, with the Moho generally sub-horizontalfor long sections, and then ramping down over a shortdistance. This feature is observed on both the north-eastern Yilgarn traverse 01AGSNY1 and the EasternGoldfields Traverse 91EGF01.

In the south-western part of the Yilgarn Craton,crustal-scale refraction profiles indicates the crust is∼35 km thick (Dentith et al., 2000), whereas in thenorthern Yilgarn Craton, refraction profiles indicate thatthe crust is ∼38 km thick (Drummond, 1998). In bothareas, the respective authors identified significant lateralvelocity variation. Receiver function data across thecraton support these depths where they coincide with thereflection or refraction interpretations and provide infillpoint values elsewhere. Reading et al. (2003) reportedMoho depths, based on receiver function analysis, of∼40 km at the western end of traverse 01AGSNY1,deepening to ∼45 km at the eastern end near the craton.

When the various Moho estimates are contoured anddisplayed in 3D, a craton-wide Moho surface isproduced (Fig. 6b). This surface shows a good first-order correspondence to the estimates of the crustalthickness obtained from the teleseismic S-wave volume.

4.3. Lithospheric structure

Teleseismic results (Kennett, 2003; Reading andKennett, 2003) show the Yilgarn Craton has fast S-wave velocities and contains a southeastwards dippinghigh-velocity three-dimensional body that is approxi-mately 70 km deep beneath the western YilgarnCraton, 100 km beneath the Ida Fault, and approxi-mately 120 km near the eastern margin of the YilgarnCraton (Figs. 5 and 6b). Although not clear, thelithosphere beneath the craton is of the order of 220 kmdeep.


Unlike the previous data sets, this data set had noother constraints and thus could only be used on its own.The locations of recorded mantle nodules, kimberlites,lamprophyres and carbonitites were displayed in themodel; however, the dramatic age differences betweenthe latter's ages (Phanerozoic, mostly Mesozoic) and theage of the craton (Archaean) raised unanswerablequestion regarding the possibility of relative movementsbetween the crust, the mantle and the lithosphere. Evenso, arguments presented earlier indicate very littlemovement post the D2 deformation (∼2660 Ma) sowe have assumed only little relative horizontal move-ment between the crust and mantle. Thus, the recentmantle nodules have sampled a region of the mantle thatwas not too different from that of the Archaean. Giventhis, it is interesting to note that all kimberlite locationslie within 100 km of one of the major breaks (offsets) inthe 4.8 km s−1 S-wave iso-surface.

5. 3D integration and interpretation of the 3D model

Each of the seismic data sets presented within,provides valuable information on the crustal architectureof the Yilgarn Craton. The merging and integration ofthese data sets through the construction of a 3Dgeological model of the craton (Goleby et al., 2002;Blewett et al., 2003) provides significantly moreinformation on the crustal architecture and possiblegeodynamic evolution of the Yilgarn Craton. Throughthe incorporation of seismic and geological data sets, the3D model presents both detailed velocity distributions atvarious scales within a defined geological crustalarchitecture (Fig. 6). This combination of velocity,structural and geological data set displays velocity–geological variations across the Yilgarn Craton, as wellas differences between the velocity structure of miner-alised and less mineralised terranes of the EasternGoldfields Province and their underlying relatively low-density and low-velocity middle crust.

The integration process including the differentseismic data sets into the 3D model focused on thepresentation and visualisation of the various data setsrather than trying to adjust any particular data set tomatch the scale, frequencies, and wave-modes ofdifferent dimensions of any other data set. This

Fig. 6. (a) View looking southwest of 3D model of the Kalgoorlie region, showthe greenstone architecture and the granite shapes were derived. Faults (e.eastwards at approximately 30°. Significant gold deposits are shown as goldYilgarn craton, showing teleseismic S-wave velocity variations, 4.8 km s−1 Sand wide-angle results. The north–south white body is the Ida Fault, the terranCross Province.

methodology allowed the integration of 2D high-resolution reflection data with 2D crustal-scale seismicreflection profiling with 2D crustal scale P- and S-wavevelocity models with 1D receiver function velocitymodels and with low-resolution 3D tomographicvelocity models. The most important part of the crustalcomponent of the model was the construction of aregional crustal architecture framework, which thenallows for the development of a 3D geological/tectonicmodel of the area. The above process resulted in thedevelopment of a terrane-scaled geologically consistent3D model, within which, a craton-wide velocity modelis superimposed.

At the lithospheric scale, the 3D image of theteleseismic data set indicates the lithosphere beneath thecraton is far more complex than originally suggested.The main observation being the presence of a fast S-wave velocity layer at depths of 100–120 km. From the3D model's velocity structure, we can categorise themantle beneath the Yilgarn Craton as being fast,depleted, refractory, cold, less dense, dry, strong andbuoyant; compared with the mantle beneath the easternpart of Australia which is slow, undepleted, fertile,warm, dense, wet, weak and less buoyant. The fastvelocities beneath the Yilgarn Craton suggest that themantle beneath the Yilgarn consists of material similarto harzburgite, with the 100–120 km deep, high S-wavebody indicating a possible compositional change fromharzburgite to garnet lherzolite or the change fromgarnet lherzolite to eclogite. We do not see evidence fora seismic low-velocity zone beneath the Yilgarn Craton.

Three possibilities exist as to the nature of thishigher-velocity body; these being a fossil southeast-dipping subduction zone, a delaminated lower crustalrestite layer or the remains of a mantle plume (Blewett,2005). The seismic data, nor the 3D model, supports aplume model (Campbell and Hill, 1988; Hill et al.,1992) as modern mantle plumes are characterised bylow P- and S-wave velocities (not high velocities), andthere is no impact on the stratification of the mantlelithosphere (Blewett, 2005). Although there is someevidence for the high-velocity body being be a fossileast-directed subduction zone, primarily the observationthat modern subduction zones are characterised by fastS-wave velocities, we support the interpretation that this

ing network of seismic reflection sections from which the main faults,g. brown surface between Kalgoorlie and Coolgardie) generally dipprisms, gold occurrences as red crosses. (b) View looking NNWof the-wave velocity body, receiver function data, refraction interpretationse boundary between the Eastern Goldfields Province and the Southern


layer represents delaminated material from the base ofthe crust. Delamination is a geodynamic process with ascale large enough to account the temporal link betweenthe late-stage low-Ca granites and the late-stage goldevent occurring almost simultaneously across the entireYilgarn Craton (Cassidy et al., 2002; Blewett, 2005).Delamination is consistent with most of seismic andgeological observations, in particular the presence of aMoho with gentle easterly dip, a thin crust, a fasteasterly dipping S-wave velocity layer body at 100–120 km and an apparent simple layering within the crustand upper mantle. The delamination argument is furtherstrengthened by Champion and Sheraton (1997) andSmithies and Champion (1999) who suggested that thedelamination of a dense garnet-rich lower crust (restite),formed during extraction of earlier voluminous high-Cagranite magma, allowed heat to be introduced into thebase of the crust. This heat resulted in widespreadmelting and emplacement of low-Ca granites duringlate-orogenic extensional collapse (Blewett, 2005). Thepresence of the relatively flat Yilgarn Moho (Drum-mond et al., 2000; Goleby et al., 2004) indicates someform of lower crustal thermal erosion or modification bydelamination (Nelson, 1998).

Evidence that the layer could be subduction relatedcomes from several authors (e.g. Cassidy and Champi-on, 2004; Morris and Witt, 1997) who proposedsubduction to account for geological observations;however, both their models have opposing dips to thesubduction. The high-velocity layer cannot be thesignature of slabs from both southeast- and west-directed subduction events. In addition, van der Veldenet al. (2006) interpreted several short but continuousreflections that dip into the upper mantle beneath thenortheastern Yilgarn as possible remnants of convergentlithospheric boundaries between plates, and concludedthat these reflections represent relict subduction zones.

The low velocity structure of the Yilgarn crust today(Fig. 6b), implies an essentially felsic composition(Drummond et al., 1993; Fomin et al., 2003). This felsiccomposition questions the inference for a dense garnet-rich lower crust as a residue of high-Ca granitemagmatism (Champion and Sheraton, 1997; Smithiesand Champion, 1999). However, the integrated 3Dmodel does not preclude either the presence of a smallpercentage of dense material (up to 20%) within thelower crust, or the possibility that the high-velocity layer(dense) in the upper mantle represents the delaminatedlayer of previously thicker Archaean crust.

Many of the features seen on the regional seismicreflection data, in particular the more recent faultmovements, support late extension (Swager, 1997).

These arguments are all consistent with the velocity andgeological structures shown within the 3D model.

At the crustal scale, the reflection data and receiverfunction results show an eastward thickening of theYilgarn crust and a series of prominent east-dippingcrustal-penetrating shear zones, which correlate at thesurface with mapped terrane boundaries (Fig. 6b). The3D geometry of these crustal-penetrating shear zonesterrane boundaries within the Eastern GoldfieldsProvince is important, as the major gold deposits areall spatially associated with these structures. TheKalgoorlie region and Laverton Tectonic Zone, twomineral-rich regions within the Eastern GoldfieldsProvince, are controlled by the location of domicallow-angle shear zones linked to the major crustal-penetrating structures (Henson et al., 2005). Thisgeometry is critical in focusing upward moving fluidsand subsequent distribution of fluids into the overlyingcomplexly deformed greenstones. Elsewhere, faultsunrelated or unlinked to the main crustal-penetratingstructures appear poorer in economic endowment.

From the 3D model, we can show that the golddeposits in the Yilgarn were the product of the focussingof fluid fluxes from deeper in the crust and uppermostmantle into the upper crust via the deep-crustalpenetrating shear zones that acted as pre-definedpathways for fluids to be focussed into appropriatenearby structures or lithologically/rheologically suitableareas. Henson et al. (2005) integrated the seismicreflection data, forward and inverse modelled potentialfield data, together with structural geology of mappatterns to show a strong positive relationship betweendomes and world-class ore deposits. The surface trace,3D geometry, and relationship of these faults to othershears are key controls in gold distribution. This fluidmigration and trapping process operates at a range ofscales from crustal (craton) scale to deposit scale. At thecrustal scale, these signatures are mapped by seismicreflection methods and, to a lesser extent, seismicrefraction methods. However, at the lithospheric scale,seismic tomography methods were required to image themain ‘cracks’ (pathways).

The 3D crustal architecture suggests a foreland basinmodel with regional folding and thrusting occurringduring the main compressive episode (Goleby et al.,2002; Blewett et al., 2003, 2004a). This episode wasseparated by equally important extension events, withthe seismic reflection data suggesting that extensionalmovement on shear zones was more common thatpreviously thought and that the major domain-boundingshear zones are best interpreted as having late extensionthat followed an episode of earlier west-directed


thrusting. The seismic reflection data led to thedevelopment of a model involving rapid foreland-directed ‘orogenic surge’ of the granite–greenstonewedge above a gently dipping basal detachment duringregional compression (Blewett et al., 2004a) duringorogenesis.

Acknowledgements

In all these seismic data sets, ANSIR, the NationalResearch Facility in Earth Sounding, role is acknowl-edged for its part in the provision of equipment andexpertise and in the data collection phases of the work.We thank Chris Fitzgerald for producing the figures. Wethank Dave Snyder and two anonymous reviewers forvery helpful comments in improving this paper. Thispaper is published with the permission of the ChiefExecutive Officer, Geoscience Australia, and theDirector of the Predictive Mineral Discovery Coopera-tive Research Centre.

References

Barley, M.E., Brown, S.J.A., Cas, R.A.F., Cassidy, K.F., Champion, D.C., Gardoll, S.J., Krape, B., 2003. An integrated geological andmetallogenic framework for the eastern Yilgarn Craton: Develop-ing geodynamic models of highly mineralised Archaean granite–greenstone terranes. AMIRA Project P624, Final Report.

Beeson, J., Neilson, I.E., Standing, J.G., Willson, M.J., Coward, M.P.,2004. Towards a 3-dimensional common earth model for theLaverton Greenstone Belt, Western Australia. Continental Tecton-ics: Discussion meeting in memory of the life and work of MikeCoward, Conference proceedings. Special Publication of theGeological Society. Unpublished.

Blewett, R.S., 2005. The 5Q's synthesis and predictive mineraldiscovery. In: Blewett, R.S., Hitchman, A.P. (Eds.). Final report—3D Geological Models of the Eastern Yilgarn Craton. PredictiveMineral Discovery Cooperative Research Centre, UnpublishedReport.

Blewett, R.S., Champion, D.C., Cassidy, K.F., Goleby, B.R., Bell, B.,Groenewald, P.B., Nicoll, M., Whitaker, A., 2003. Implications ofthe northeastern Yilgarn Seismic to Leonora–Laverton 3Dmap. In:Goleby, B.R., Blewett, R.S., Groenewald, P.B., Cassidy, K.F.,Champion, D.C., Jones, L.E.A., Korsch, R.J., Shevchenko, S.,Apak, S.N. (Eds.), The 2001 Northeastern Yilgarn Deep SeismicReflection Survey, Geoscience. Australia, Record 2003/28,pp. 127–143.

Blewett, R.S., Cassidy, K.F., Champion, D.C., Henson, P.A., Goleby,B.R., Jones, L., Groenewald, P.B., 2004a. The WangkathaaOrogeny: an example of episodic regional ‘D2’ in the lateArchaean Eastern Goldfields Province, Western Australia. Pre-cambrian Research 130, 139–159.

Blewett, R.S., Cassidy, K.F., Champion, D.C., Henson, P.A.,Goleby, B.R., Kalinowski, A.A., 2004b. An orogenic surgemodel for the eastern Yilgarn Craton: implications for goldmineralising systems. In: Muhling, J., et al. (Ed.), SEG 2004,Predictive Mineral Discovery Under Cover, vol. 33. Centre for

Global Metallogeny, The University of Western Australia,Publication, pp. 321–324.

Campbell, I.H., Hill, R.I., 1988. A two-stage model for the formationof the granite–greenstone terrains of the Kalgoorlie–Norsemanarea, Western Australia. Earth and Planetary Science Letters 90,11–25.

Champion, D.C., 1997. Granitoids in the Eastern Goldfields. In:Cassidy, K.F., Whitaker, A.J., Lui, S.F. (Eds.), Kalgoorlie '97. AnInternational Conference on Crustal Evolution, Metallogeny andExploration of the Yilgarn Craton—An Update. AustralianGeological Survey Organisation, Record 1997/41, pp. 71–76.

Champion, D.C., Sheraton, J.W., 1997. Geochemistry and Nd isotopesystematics of Archaean granites of the Eastern Goldfields, YilgarnCraton, Australia; implications for crustal growth processes.Precambrian Research 83, 109–132.

Cassidy, K.F., Champion, D.C., 2004. Crustal evolution of the YilgarnCraton from Nd isotopes and granite geochronology: implicationsfor metallogeny. In: Muhling, J., et al. (Ed.), SEG 2004, PredictiveMineral Discovery Under Cover. Centre for Global Metallogeny,vol. 33. The University of Western Australia, Publication,pp. 317–320.

Cassidy, K.F., Champion, D.C., McNaughton, N.J., Fletcher, I.R.,Whitaker, A.J., Bastrakova, I.V., Budd, A.R., 2002. Characterisa-tion and metallogenic significance of Archaean granitoids of theYilgarn Craton, Western Australia. Minerals and Energy ResearchInstitute of Western Australia, Report, vol. 222. 514 pp.

Cassidy, K.F., Blewett, R.S., Champion, D.C., Goleby, B.R., 2003.Northeastern Yilgarn Seismic Reflection Survey: implications fororogenic Au systems. In: Goleby, B.R., Blewett, R.S., Groene-wald, P.B., Cassidy, K.F., Champion, D.C., Jones, L.E.A., Korsch,R.J., Shevchenko, S., Apak, S.N. (Eds.), The 2001 NortheasternYilgarn Deep Seismic Reflection Survey, Geoscience. Australia,Record 2003/28, pp. 127–143.

Dentith, M.C., Dent, V.F., Drummond, B.J., 2000. Deep crustalstructure in the southwestern Yilgarn Craton, Western Australia.Tectonophysics 325, 227–255.

Drummond, B.J., 1998. A review of crust/upper mantle structure in thePrecambrian areas of Australia and implications for Precambriancrustal evolution. Precambrian Research 40/41, 101–116.

Drummond, B.J., Goleby, B.R., Swager, C.P., Williams, P.R., 1993.Constraints on Archaean crustal composition and structureprovided by deep seismic sounding in the Yilgarn Block. OreGeology Reviews 8, 117–124.

Drummond, B.J., Goleby, B.R., Swager, C.P., 2000. Crustal signatureof Late Archaean tectonic episodes in the Yilgarn craton, WesternAustralia: evidence from deep seismic sounding. Tectonophysics329, 193–221.

Fishwick, S., Kennett, B.L.N., Reading, A.M., 2005. Contrasts inlithospheric structure within the Australian craton—insights fromsurface wave tomography. Earth and Planetary Science Letters231, 163–176.

Fomin, T., Goleby, B.R., 2005. Lessons from combined interpretationsof wide-angle and conventional reflection data in the northeasternYilgarn, Western Australia. Tectonophysics, this volume.

Fomin, T., Crawford, A., Johnstone, D., 2003. Awide-angle reflectionexperiment with Vibroseis sources as part of a multidisciplinaryseismic study of the Leonora–Laverton Tectonic Zone, Northeast-ern Yilgarn Craton. Exploration Geophysics 34, 147–150.

Fomin, T., Goleby, B.R., Nicoll, M., 2004. Some lessons fromcombined interpretations of wide-angle and conventional reflec-tion data in the northern Yilgarn, Western Australia. 11thInternational Symposium on Deep Seismic profiling of the


Continents and their Margins, Programme and Abstracts. Centredes congrès, Mount-Tremblant, Quebec.

Goleby, B.R., Drummond, B.J., 2000. The 1991 Deep seismic survey,Eastern Goldfields, W.A. In: Goleby, B.R., Bell, B., Korsch, R.J.,Sorjonen-Ward, P., Groenewald, P.B., Wyche, S., Bateman, R.,Fomin, T., Witt, W., Walshe, J., Drummond, B.J., Owen, A.J.(Eds.), Crustal Structure and Fluid Flow in the Eastern Goldfields,Western Australia. Australian Geological Survey Organisation,Record 2000/34, pp. 46–52.

Goleby, B.R., Rattenbury, M.S., Swager, C.P., Drummond, B.J.,Williams, P.R., Sheraton, J.W., Heinrich, C.A., 1993. Archaeancrustal structure from seismic reflection profiling, Eastern Gold-fields, Western Australia. Australian Geological Survey Organisa-tion, Record, 1993/15.

Goleby, B.R., Bell, B., Korsch, R.J., Sorjonen-Ward, P., Groene-wald, P.B., Wyche, S., Bateman, R., Fomin, T., Witt, W.,Walshe, J., Drummond, B.J., Owen, A.J., 2000. Crustalstructure and fluid flow in the Eastern Goldfields, WesternAustralia. Australian Geological Survey Organisation, Record2000/34. 109 pp.

Goleby, B.R., Korsch, R.J., Fomin, T., Bell, B., Nicoll, M.G.,Drummond, B.J., Owen, A.J., 2002. Preliminary 3-D geologicalmodel of the Kalgoorlie region, Yilgarn Craton, Western Australia,based on deep seismic-reflection and potential-field data. Austra-lian Journal of Earth Sciences 49, 917–933.

Goleby, B.R., Blewett, R.S., Groenewald, P.B., Cassidy, K.F.,Champion, D.C., Korsch, R.J., Whitaker, A., Jones, L.E.A., Bell,B., Carlson, G., 2003. Seismic interpretation of the northeasternYilgarn Craton seismic data. In: Goleby, B.R., Blewett, R.S.,Groenewald, P.B., Cassidy, K.F., Champion, D.C., Jones, L.E.A.,Korsch, R.J., Shevchenko, S., Apak, S.N. (Eds.), The 2001Northeastern Yilgarn Deep Seismic Reflection Survey. GeoscienceAustralia, Record 2003/28. 143 pp.

Goleby, B.R., Blewett, R.S., Korsch, R.J., Champion, D.C., Cassidy,K.F., Jones, L.E.A., Groenewald, P.B., Henson, P., 2004. Deepseismic reflection profiling in the Archaean northeastern YilgarnCraton, Western Australia: implications for crustal architecture andmineral potential. Tectonophysics 388, 119–133.

Henson, P.A., Blewett, R.S., Champion, D.C., Goleby, B.R.,Drummond, B.J., Korsch, R.J., Brennan, T., Nicoll, M., 2005.Domes: the characteristic 3D architecture of the world-class lode–Au deposits of the Eastern Yilgarn. In: Hancock, et al. (Ed.),Structure, Tectonics and Ore Mineralisation Processes. EconomicGeology Research Unit Contribution, vol. 64, p. 60.

Hill, R.I., Chappell, B.W., Campbell, I.H., 1992. Late Archaeangranites of the southeastern Yilgarn block, Western Australia: age,geochemistry and origin. Transactions of the Royal Society ofEdinburgh. Earth Sciences 83, 211–226.

Hobbs, B.E., Upton, P., Ord, A., Zhang, Y., Zhao, C., Drummond, B.,Archibald, N., 1997. Thermal and deformation modelling of theYilgarn Deep Seismic Transect. 4th Biennial SGAMeeting, Turku,Findland, pp. 859–862.

Kennett, B.L.N., 2003. Seismic structure in the mantle beneathAustralia. In: Hillis, R.R., Mueller, R.D. (Eds.), Evolution andDynamics of the Australian Plate. Geological Society of Australia,

Special Publication 22, and Geological Society of America,Special Paper, vol. 372, pp. 7–23.

Morris, P.A., Witt, W.K., 1997. Geochemical and tectonic setting oftwo contrasting Archaean felsic volcanic associations in theEastern Goldfields, Western Australia. Precambrian Research 83,83–107.

Myers, J.S., 1995. The generation and assembly of an Archaeansupercontinent—evidence from the Yilgarn Craton, WesternAustralia. In: Coward, M.P., Ries, A.C. (Eds.), Early PrecambrianProcesses. Geological Society London, Special Publication, vol.95, pp. 143–154.

Nelson, D.R., 1998. Granite–greenstone crust formation on theArchaean Earth: a consequence of two superimposed processes.Earth and Planetary Science Letters 158, 109–119.

Reading, A.M., Kennett, B.L.N., 2003. Lithospheric structure of thePilbara Craton, Capricorn Orogen and northern Yilgarn Craton,Western Australia, from teleseismic receiver functions. AustralianJournal of Earth Sciences 50, 439–445.

Reading, A.M, Kennett, B.L.N., Dentith, M.C., 2003. Seismicstructure of the Yilgarn Craton, Western Australia. AustralianJournal of Earth Sciences 50, 427–438.

Reading, A.M., Kennett, B.L.N., Goleby, B.R., in press. The deepseismic structure of Precambrian terranes within the WestAustralian Craton and implications for crustal formation andevolution. Geology.

Smithies, R.H., Champion, D.C., 1999. Geochemistry of felsic igneousalkaline rocks in the Eastern Goldfields, Yilgarn Craton, WesternAustralia: a result of lower crustal delamination? — Implicationsfor Late Archaean tectonic evolution. Journal of the GeologicalSociety (London) 156, 561–576.

Stolz, E., 2003. Direct detection of gold bearing structures at St Ives,WA, — DHEM vs DHMMR. Australian Society of ExplorationGeophysicists Conference. ASEG Preview, vol. 102, p. 77.

Swager, C.P., 1997. Tectonostratigraphy of late Archaean greenstoneterrains in the southern Eastern Goldfields, Western Australia.Precambrian Research 83, 11–41.

Swager, C.P., Griffin, T.J., 1990. An early thrust duplex in theKalgoorlie–Kambalda greenstone belt, Eastern Goldfields Prov-ince, Western Australia. Precambrian Research 48, 63–73.

Swager, C.P., Witt, W.K., Griffin, T.J., Ahmat, A.L., Hunter, W.M.,McGoldrick, P.J., Wyche, S., 1992. Late Archaean granite–greenstones of the Kalgoorlie Terrane, Yilgarn Craton, WesternAustralia. In: Glover, J.E., Ho, S.E. (Eds.), The Archaean—Terrains, Processes and Metallogeny, vol. 22. Geology Departmentand Extension Service, University of Western Australia, Publica-tion, pp. 107–122.

Swager, C.P., Goleby, B.R., Drummond, B.J., Rattenbury, M.S.,Williams, P.R., 1997. Crustal structure of granite–greenstoneterranes in the Eastern Goldfields, Yilgarn Craton, as revealed byseismic reflection profiling. Precambrian Research 83, 43–56.

van der Velden, A.J., Cook, F.A., Drummond, B.J., Goleby, B.R., 2006.Reflections of the Neoarchean: a global perspective. In: Benn, K.,Mareschal, J.-C., Condie, K. (Eds.), Archean Geodynamics andEnvironments. Geophysical Monograph Series, vol. 164.

an integrated multi-scale 3d seismic model of the archaean...

Documents