New insights into some of Australia's giant deposits through seismic reflection surveys: Exploration implications

Australia is a major mining nation – it is in the top 6 producers in world for some 20 minerals that are produced from about 400 operating mines.

Australia has a very large mineral endowment — it is the top three in the world for about 12 commodities, has the world's largest Economic Demonstrated Resources (EDR) for about 8 minerals and has the world’s largest EDR of about 8 commodities (bauxite, nickel, lead, mineral sands, silver, tantalum, uranium, and zinc).

However, most of the known ore bodies either have surface expression or were found by surface prospecting over the past 150 years.

Unlocking Australia’s potential under the variable thickness regolith and sedimentary cover requires a better knowledge of regional geology in three dimensions. But how well do we really know the 3D geology?

In this paper we will present the findings of several case histories where Geoscience Australia has used deep crustal seismic reflection profiling to image the geology around several world-class deposits.

These surveys were acquired through partnerships with the State/NT Geological Surveys, Australian Seismic Imaging Resource (ANSIR), and, in some cases, Co-operative Research Centres and industry.

The approach we adopted in the seismic imaging program can be summarised as:

• Don't try to image ore bodies
• Do what the petroleum industry does:
• Find the environment that is conducive to mineralisation
• Image the structures and pathways
• Imaging an ore body is a bonus (flat spots)
• Need to predict settings and what they will look like
• Mineralising fluids leave a trail!

This approach has proved very successful in transects ranging from hundreds down to tens of kilometres.

The available case studies around major mineral deposits undertaken in the past 10 years are shown on the slide. In this presentation we can only examine a few of these and have chosen:

• Kalgoorlie and the Eastern Goldfields of the Yilgarn Craton
• Olympic Dam in the Gawler Craton
• McArthur River deposit in the McArthur Basin, and
• Mt Isa copper-lead-zinc deposits of the Mt Isa Inlier

We will focus firstly on the eastern and north-eastern goldfields of the Yilgarn Craton – this region produces about 65% of Australian gold and most of its nickel.

Two regional seismic transects and a number of shorter lines were acquired in the Eastern Goldfields of the Yilgarn Craton over the last 15 years.

The dominant geology shown in this simplified map consists of ~2.6-2.7 Ga greenstones shown in green (mafic volcanics in lime green, dolerites and gabbros in mid green), granites in pink, mostly gneiss (purple), felsic volcanics (bright yellow), (purple), and Archaean sediments (lavender).

Also shown are the Proterozoic Albany-Fraser sediments and metamorphics (maroon), Permian sediments (pale blue), and Quaternary sediments (pale yellows). Major faults are shown as black lines.

Focussing firstly on EGF1 north of the Kalgoorlie gold deposit.

The Eastern Goldfields contains 17 gold deposits with 3 Moz or more including the super giant Kalgoorlie deposit with some 72 Moz.

This slide shows the strong structural control of the gold deposits closely associated with the major NW-trending structures, although the gold deposits are commonly contained in second and third order structures.

This slide shows the present day geometry of the Eastern Goldfields.

The basic architecture consists of three basic crustal layers cut by major low-angle structures.

1. The upper layer (pink) is seismically transparent and consists of granite and gneisses that lie to the west of and below the greenstones.
2. This is underlain by a shortened mid crustal layer shown in maroon.
3. Below this lies the ductile lower crust, shown in purple.

A seismically anomalous bland zone – shown in blue – extends from the upper to lower layers.

Key messages emerging from the transect near Kalgoorlie are:

• A few faults penetrate deep into the crust across the Yilgarn
• These are important as they focus of fluid flow from the deep crust into upper crust
• The dipping zone shown in blue is a seismically anomalous zone:
• It is inferred to be caused by fluids moving through and up to surface
• Faults such as the Bardoc Fault are important – they bring ore fluids to the surface

A similar architecture is evident in the northern transect.

Key features are:

• The Moho deepens to east
• The three broad crustal layers are evident
• The structure is dominated by prominent low-angle easterly dips – a thrust belt
• Three deep-penetrating E-dipping shear zones are evident
• Two of these are associated with substantial gold districts.
• The third – the Yamarna shear zone in the east where there is little or no outcrop - may be the site of a another gold district.

The key findings of this survey are:

1. The signature of most crustal deformation events can be seen in the seismic image.
2. Analogy with modern orogens suggests that the detachment is controlled by fluid "ponding" at the brittle/ductile transition zone. The few faults that penetrate the detachment into the deeper crust are important for focussing fluid flow between the lower and upper crust.
• Fault zone reflectivity is due to alteration
• Alteration is caused by fluids at time of deformation

In 2004, Geoscience Australia, in collaboration with Primaries Industries and Resources South Australia (PIRSA), undertook a deep crustal seismic reflection survey across the eastern margin of the Gawler Craton in the vicinity of the Olympic Dam deposit.

In 2004, WMC upgraded the resources of the super giant Olympic Dam deposit copper-gold-uranium deposit to 3810 Mt @ 1.1% Cu, 0.4 kg/t U₃0₈, 0.5 g/t Au. – this places the Olympic Dam deposit in the top 10 copper deposits in the world by remaining resources. It is the world’s largest uranium resource.

The aim of the Gawler Craton seismic survey was to determine the architecture of the NE Gawler Craton near the Olympic Dam Cu-Au-U deposit, and the nature of the margin of the Gawler Craton near the Torrens Hinge zone.

The survey comprised 250 km of survey in two lines – a 193 km roughly N-S transect and a shorter cross-line roughly E-W to provide information on the three dimensional geometry of structures – were recorded to about 55 km depth (18s 2-way travel time).

The Gawler Craton has an Archaean core surrounded by 1.85-1.45 Ga age mobile belts comprised mostly of meta-sediments and granitoids.

In the east and central Gawler is the major 1.575-1.595 Ga age Hiltaba Suite granitoids and co-magmatic Gawler Range Volcanics. The Hiltaba Suite granitoids have a close spatial and temporal association with the iron oxide Cu-Au mineralisation of Olympic Dam and Prominent Hill and smaller deposits.

The Olympic Dam deposit lies at the eastern margin of the Gawler Craton on the Stuart Shelf. The tectonic setting at time of mineralisation is unknown:

• Extension is implied in some mineralisation models
• Similar deposits occur elsewhere around the world in a range of settings, including orogens

This slide shows the simplified geology of the deposit:

• Paleozoic basement rocks are intruded by the Hiltaba Suite granitoids, gabbroids and ultramafics (some alkaline) at ~1590 Ma
• The Olympic Dam deposit is hosted by breccias of the Burgoyne batholith
• Popular models for the formation of the Olympic Dam deposit involve mixing of deep-sourced fluids (involving mantle-derived igneous rocks), and surficially-derived waters

The Olympic Dam region preserves the near-surface ~1590 Ma environment; in Moonta-Wallaroo district and parts of Mt Woods Inlier deeper crustal levels are now exposed.

This slide shows the seismic results at the top and the interpretation below.

The Olympic Dam deposit lies between two distinctly different pieces of crust. In the north, the upper crust beneath the thin young cover sequences (0.5-2s TWT) shows S-dipping reflectors. These are interpreted as shear zones that cut an upper crust of moderate sub-horizontal reflections and a lower crust of higher amplitude reflections.

To the south, the upper and lower crust have structures indicating thrusting towards the craton, in a pattern indicative of a crustal-scale fold and thrust belt.

However, the upper crustal deformation is decoupled from the lower crustal deformation at a 3-4 km thick band of sub-horizontal reflectors at about 12 km.

The Moho is at about 40 km (13s TWT). The Burgoyne Batholith that hosts the Olympic Dam deposit is about 5 km thick. It is underlain by an anomalous but homogenised zone with few reflectors – this may be the source of region for the Burgoyne Batholith.

The tectonic elements are summarised in this slide.

It shows:

1. Post-ore body extension to create this basin.
2. Pre-ore body extension, Syn-ore body shortening.
3. Crust in the south has been shortened, possibly at the same time as the upper crust.
4. Crust in the north has been shortened, again possibly at the same time as the upper crust.
5. Seismically anomalous crust in the middle, this may be overprinted, possibly depleted. Magnetotelluric data indicate that this zone is one of high conductivity (less than 500 ohm/m).

Lastly, granite intrusion and then ore body formation.

The Olympic Dam deposit lies at a major crustal boundary between an older stable Archaean block to the south and the reworked crust of the Proterozoic mobile belt to the north.

The McArthur Basin hosts the Mesoproterozic ~1640 Ma McArthur River sediment-hosted, stratiform Ag-Pb-Zn deposit. The deposit has a total resource of about 14.6 Mt Zn and 6.6 Mt Pb, making it one of the world's largest Ag-Pb-Zn deposits.

Recent models for the formation of "McArthur style" Ag-Pb-Zn deposits in northern Australia fall into sedimentary exhalative ("SEDEX") or early epigenetic replacement.

Later epigenetic structurally-controlled emplacement for McArthur River is unlikely based on lack of deformation and metamorphism, base metal distribution, ore textures and organic geochemistry.

The Southern McArthur Basin is composed of sequences of mostly unmetamorphosed, relatively undeformed Palaeoproterozoic to Mesoproterozoic carbonate, siliciclastic and volcanic rocks.

Seismic reflection data were collected in late 2002 as part of a study to examine the fundamental basin architecture of the Southern McArthur Basin and the nature of underlying basement.

In particular, the seismic used to test geometric models for the Southern McArthur Basin, including the tectonostratigraphy of the basin, its fault systems and basement structure.

The results have wider applicability because the basin is considered an undeformed analogue of the Western Succession of Mt Isa.

The main seismic line - 110 km long - was oriented E-W across the Southern McArthur Basin. A short N-S cross line, 20 km long, was acquired within the trough itself, in collaboration with Anglo American.

Early models of the McArthur Basin suggested that it evolved as a framework of troughs, shelves and fault zones with basin evolution strongly influenced by intermittent strike-slip faulting, block rotation and syn-sedimentary faulting focussed in the Batten Fault Zone and essentially related to E-W extension over a period of about 200 Ma.

The fault segments – the Batten and Walker Fault Zones – were believed to have primary depositional sites ("rifts") for the McArthur and Nathan Group sediments as the Batten and Walker Troughs.

The seismic data show that the Batten Fault Zone is not a separate depocentre. There is no evidence for E-W extension.

Sedimentary successions mainly thicken to east, including east of Emu Fault (i.e. Batten "non-trough").

The Emu Fault appears to be a near vertical strike-slip fault system, containing an inverted flower structure.

The Tawallah Fault is a W-dipping thrust and part of major E-directed thrust belt.

The timing of both thrust belt and strike-slip system are post deposition of Roper Group.

The seismic data provide new insights into the origin of the McArthur River deposit.

1. The McArthur River deposit is not at the boundary of a depositional basin.
2. The Emu Fault has long been considered a conduit for ore-forming fluids moving from source to trap in the thick deep-water carbonaceous deep-water shales and siltstones of the Barney Creek formation of the middle McArthur Group. The seismic data confirms the importance of the Emu Fault as a conduit for ore forming fluids.
3. The Tawallah Fault dips gently to the west and the aquifer lies mostly below the fault. This implies that current models that invoke fluid circulation down this fault and through relatively shallow aquifers need revision.
4. There is potential in the younger thrust belt for foreland basin type deposits e.g. MVT-type deposits formed from topographically driven fluids.
5. There may be potential in the core of thrust belt (orogen) to west for orogenic gold deposits under cover.

• Steep geometry of Emu Fault permits access to deep basinal brines
• Potential for other sub-basins along strike-slip Emu Fault
• Thick McArthur Group east of Emu Fault – prospective?
• The convective fluid flow system not essential, i.e.:
• Fluids could be derived from a deeper, older part of Tawallah Formation
• Fluids possibly hotter (up to 300 deg C. This is supported by biomarkers in the Mineralisation-related organic material)
• High T means higher solubility of lead and zinc

The final case history – although done 10 years ago – was selected because it involves mine scale use of seismic in the Mt Isa Cu, and Ag-Pb-Zn mines.

Mt Isa hosts two world-class deposits:

• a copper ore body with 5.5 Mt contained Cu (255 Mt @ 3.3 Cu)
• a Ag-Pb-Zn lode with 8.6 Mt Zn and 7.5 Mt Pb

Around the Leichhardt River Fault Trough, the Adelheid Fault is reflective whereas the Mt Isa Fault is not (both faults are mapped at the surface).

This suggests that the western bounding fault was more important for focussing fluid flow during the formation of the Pb-Zn deposits at least.

The high-resolution work around the mine was to test whether seismic could pick up the Paroo Fault. The Paroo Fault cuts off mineralisation in the mine, and is therefore the effective depth of mineralisation for mining purposes. In the mine, the Pb-Zn-Ag is stratabound in the Urquhart Shale, whereas the Cu lies in a zone above the fault that crosses stratigraphy.

The cross section is a prediction published before the seismic work, based on drilling along the line of the seismic profile and projections from the mine northwards for several hundred metres.

The seismic data are displayed at the same scale and adjusted to the same datum. Dots show where there is control from drilling. The Paroo Fault can be interpreted as the cut off of the steeply dipping reflections attributed to the Urquhart Shale. The Paroo Fault is faulted rather than folded as predicted. There is a reflector above that which correlates spatially with the front of dolomitic alteration above the Cu ore bodies.

The Urquhart Shale can be seen penetrating through this reflector to the Paroo Fault.

There are a number of lessons to be learnt from the seismic studies.

Seismic reflection method can image structures that are ore controlling/influence mineralisation, e.g.:

• Crustal penetrating long-lived faults that were active during the change from compression to inferred strike-slip in the case of Archaean gold
• Basin margin faults in stratabound deposits

Zones that have had fluid movement are reflective:

• Alteration changes rock density and seismic wave speed
• Anisotropy caused by the alignment of phyllosilicate minerals

At deposit scales, high resolution data show the environment around the ore deposit:

• Elements for local exploration strategy
• Maps fluid signatures (e.g. alteration front etc) rather than the ore body

We are using the new information gained from the surveys together with information from structural mapping and potential field data to produce on-line 3D geological models of key provinces. An example is the Leonora-Laverton region of the Eastern Goldfield shown here in VRML.

However, we don’t always have access to seismic reflection data.

Hindsight is a wonderful tool. Clues to the structures we see in the seismic data were evident in the existing geological maps and/or potential field data, but they were not fully appreciated.

This tells us we can and need to extract more from such data and this is an area of active research at Geoscience Australia.

For example, this slide shows the results of 3D inversion modelling of magnetic and gravity data in the Olympic Dam region using the University of British Columbia inversion software. The modelling has differentiated volumes as isosurfaces of denser (blue) and more magnetic (green) material. Many of the deposits (defined by the octahedra) tend to be located near the boundaries of the two is surfaces. This type of modelling integrated with the seismic data is providing new insights into 3D geology.

The Broken Hill GeoModeller Project study - undertaken by BRGM, Intrepid Geophysics and Geoscience Australia under the pmd*CRC (Predictive Minerals Discovery Cooperative Research Centre) – was done to evaluate the ability to model and invert a geologically complex terrane.

GeoModeller software was used to develop a coherent 3D geological model using existing government geological mapping, mining company geological sections, and the Geoscience Australia regional seismic line. The model was tested and refined by inverting ground and airborne gravity data. The results suggest that this approach offers considerable potential and is a step forward in developing robust 3D geological models constrained by geophysics.

To conclude:

Seismic reflection surveys are revealing the 3D structure of the crust in mineral provinces.

The geology is not as well known in 3D as we think:

• Seismic has challenged "known geology" is each province it has been used
• Some mineral deposit models based on old geology need re-examination/revision

Crustal architecture key to determining regional scale controls on mineralisation.

Most major deposits lie at major crustal boundaries/breaks.

New mineral potential is being opened up by seismic surveys (or 3D models).