Protracted fluid-rock interaction in the mid-Archaean and implication for gold mineralization: Example from the Warrawoona Syncline (WA)

Protracted fluid–rock interaction in the Mesoarchaean and implication for goldmineralization: Example from the Warrawoona syncline (Pilbara, Western Australia)

Nicolas Thébaud a,b,c,⁎, Pascal Philippot b, Patrice Rey c, Joël Brugger d,Martin Van Kranendonk e, Nathalie Grassineau f

a Center for Exploration and Targeting, 35 Stirling Highway,M006 Crawley WA 6009, Australiab Equipe Géobiosphère Actuelle et Primitive, Institut de Physique du Globe de Paris et Université Paris-Diderot, CNRS, Tour 14, 2 place Jussieu, 75005 Paris, Francec EarthByte Group, School of Geosciences, The University of Sydney, Sydney, NSW 2006, Australiad School of Earth and Environmental Sciences, The University of Adelaide, SA 5005, Adelaide, Australia and The South Australian Museum, North Terrace, SA 5000, Adelaide, Australiae Geological Survey Western Australia, 100 plain street 6004 East Perth, WA, Australiaf Department of Earth Sciences Royal Holloway University of London Egham, Surrey TW20 0EX, UK

A B S T R A C TA R T I C L E I N F O

Article history:

Received 24 December 2005

Received in revised form 19 April 2007

Accepted 23 May 2008

Available online 5 June 2008

Editor: C.P. Jaupart

Keywords:

Pilbara

Archaean

hydrothermalism

gold

fluid–rock interaction

Oxygen isotopic and geochemical analyses on whole rock and quartz veins are combined with structural

observations in order to constrain the fluid circulation history within the Mesoarchaean Warrawoona

syncline of the North Pilbara Craton, Western Australia.

The plumbing system which is the focus of this study is localized in the so called Fielding's Find shear zone

(FFSZ), a km-scale shear zone formed during the burial of greenstones and coeval exhumation of granitic

complexes. This shear zone runs parallel or close to the axial plane of the syncline. It involves a prominent

quartz vein network and is lined with strongly hydrothermally-altered mafic, felsic and sedimentary rocks.

Towards the FFSZ, felsic andmafic volcanic rocks become intensely silicified with an increase in bulk rock δ18O

values from +10.8‰ to +25.1‰ for altered felsic volcanics and from +7.1‰ to 18.3‰ for alteredmafic volcanics.

Geochemical modelling ascribes the silicification to a dissolution/precipitation process. REE and most other

trace elements are strongly depleted in the silicified units, with the exceptions of elements such as V, Cr, Ni and

Co, which are enriched. Throughout the Warrawoona syncline, vein quartz δ18O data are within a small range

of +13.2±2‰, significantly lower than their silicified host rocks. These data are interpreted as the result of two

main paleo-fluid circulation stages. Intense silicification and 18O enrichment represent alteration driven by

low-temperature hydrothermal convectionprobably involvingArchaean seawater. In contrast, the quartz veins

network is related to the infiltration of metamorphic and/or magmatic fluids during a later deformation

episode. These quartz veins represent the event responsible for the bulk of economic lode-gold formation in

the area.

The protracted fluid–rock interaction history in the Warrawoona syncline may have played a major role in

setting the stage for the late mineralizing event. The early hydrothermal circulation could have formed an

efficient plumbing system characterized by high permeability, low reactivity and possibly Au-enrichment,

upgrading the Au-endowment to the late hydrothermal fluids.

Crown Copyright © 2008 Published by Elsevier B.V. All rights reserved.

1. Introduction

Mesoarchaean terranes (3.4–3.0 Ga) are minor contributors to the

World gold production. In the Pilbara Craton and Barberton Green-

stone Belt (South Africa), the pre-2000 production was 2.2 Moz and

10 Moz Au respectively (Goldfarb et al., 2001). In comparison,

Neoarchaean terranes (3.0–2.5 Ga) host giant deposits such as the

Kalgoorlie lode-gold systemwhich, with 50 Moz Au, accounts for half

of the pre-2000 production in the Yilgarn craton (Australia) (Goldfarb

et al., 2001). Numerous studies have shown that Neoarchaean lode-

gold deposits developed within a relatively narrow time window (ca

2.67–2.62 Ga) during a major episode of juvenile continental crust

formation and crustal anatexis (Goldfarb et al., 2001; Groves et al.,

2005; Rey et al., 2003). These deposits are hosted within quartz-filled

fault and fracture meshes that formed in association with brittle-

ductile shear zones. According to many economic geologists,

Neoarchaean lode gold deposits are epigenetic and associated with

deep-sourced fluids (i.e. metamorphic, magmatic or mantle derived)

channelled along crustal-scale shear zones (Goldfarb et al., 2001;

Groves, 1993; Kerrich and Wyman, 1990; Groves and Phillips, 1987;

McCuaig and Kerrich, 1998). Despite fewer investigations, comparable

geochemical and structural features have been described for lode gold

Earth and Planetary Science Letters 272 (2008) 639–655

⁎ Corresponding author. Center for Exploration and Targeting, 35 Stirling Highway,

M006 Crawley WA 6009, Australia.

E-mail address: [email protected] (N. Thébaud).

0012-821X/$ – see front matter. Crown Copyright © 2008 Published by Elsevier B.V. All rights reserved.

doi:10.1016/j.epsl.2008.05.030

Contents lists available at ScienceDirect

Earth and Planetary Science Letters

j ourna l homepage: www.e lsev ie r.com/ locate /eps l

mailto:[email protected]

http://dx.doi.org/10.1016/j.epsl.2008.05.030

http://www.sciencedirect.com/science/journal/0012821X

deposits within the Mesoarchaean terranes, suggesting a similar

genetic scenario (Huston et al., 2001; Scherrenberg et al., 2004; Zegers

et al., 2002; de Ronde et al., 1992). However, the difference in gold

productivity between Mesoarchaean and Neoarchaean terranes

suggests a difference in mineralization processes, a difference in

fluid circulation history and/or a different tectonic framework.

The aim of this paper is to document the geometry and history of

fluid circulations associated with the Mesoarchaean mining district of

the Warrawoona syncline, Pilbara Craton (Fig. 1). Several stages of

fluid circulation have been documented. An initial stage corresponds

to a phase of seafloor metasomatism, during which large volumes of

low-temperature hydrothermal fluids infiltrated the greenstone-belt

prior to or during the main stage of deformation. It is suggested that

this hydrothermal alteration stage created the necessary conditions

for the formation of economic gold deposits during a later hydro-

thermal event, highlighted by gold-bearing quartz veins within the

district. This fluid–rock interaction history is consistent with that of

the Barberton Greenstone Belt, where gold pre-concentration has

been related to an early and protracted history of fluid circulation

involving both surface and deeper fluids (Hutchinson, 1993).

2. General settings

2.1. Geological and structural setting

The Pilbara Craton Craton, Western Australia, consists of granite-

greenstone terranes (3.72–2.83 Ga) unconformably overlain by

volcanic and sedimentary rocks of the Hamersley Basin (2.77–

2.40 Ga; Fig. 1); (Van Kranendonk et al., 2002). It is one of the best

documented examples of Archaean dome-and-basin pattern (Collins

et al., 1998; Hickman, 1983).

The Warrawoona syncline, located between the Mount Edgar and

Corunna Downs Granitic Complexes (Fig. 2), consists of ultramafic,

mafic and felsic volcanic rocks and minor sedimentary and chert units

that are ascribed to the ca. 3.51–3.43 GaWarrawoona Group and 3.35–

3.31 Ga Kelly Group (Hickman, 1983, 2001; Van Kranendonk et al.,

2004a). Facing directions on both limbs point away from the granitic

complexes towards the core of the syncline (Hickman, 1983, 2001).

The northern limb displays strongly-deformed amphibolite facies of

mafic and felsic schist of the lower part of the Warrawoona Group.

These schists are in contact with the southern margin of the Mount

Edgar Granitic Complex across a kilometer-scale shear zone. The

southern limb consists of the upper part of the Warrawoona Group

and the overlying Kelly Group and is intruded by ca. 3.30 Ga granitic

rocks of the Corunna Downs Granitic Complex (Collins et al., 1998).

At a regional scale, the Warrawoona syncline shows a main, steeply

dipping foliation (Sm) parallel to the granitic/greenstone boundaries. A

mineral and stretching lineation (Lm) rotates and converges toward a

small area where the lineation is vertical (Zegers et al., 2002; Collins

et al., 1998; Teyssier et al., 1990); (Fig. 2). The finite strain ellipsoid

changes from S tectonites near the contact with the granitic complexes

to L tectonites where Lm is vertical (Zegers et al., 2002; Collins et al.,

1998; Teyssier et al.,1990). Open to isoclinal folds occur at all scales. Fold

axial planes are parallel to Sm and fold axes are parallel to Lm (Fig. 2).

The S and SWrimof theMount Edgar is affectedbya kilometric-scale

shear zonewith down-dip radial stretching lineations (Fig. 2). Kinematic

analysis reveals that the shear zones accommodate the exhumation of

the granitic dome (Collins et al.,1998; Teyssier et al.,1990; Collins,1989).

Within the greenstones, hectometer to kilometer-scale shear zones

involve obliquehorizontal dextral and sinistral sense of shear generating

flower or pop-up structures (Kloppenburg et al., 2001). As the stretching

lineation rotates to becomevertical and thefinite strain ellipsoid evolves

to constriction, thehorizontal shear componentdisappears. Radiometric

data (U-Pb zircon on syn- and post granites and Ar–Ar on metamorphic

fabrics) constrain the syncline formation at ~3.32–3.30 Ga (Collins et al.,

1998; Kloppenburg et al., 2001).

The above geometry has been interpreted as the result of a gravity-

driven tectonic process, which involved the sagduction of greenstones

and coeval diapiric rise of the granitic complexes between 3.32 and

3.30 Ga (Fig. 2); (Zegers et al., 2002; Collins et al., 1998; Teyssier et

al.,1990; Collins, 1989; Delor et al., 1991; Hickman, 1984; Van

Kranendonk et al., 2004b). Alternative views are discussed by

Kloppenburg et al., (2001) and Blewett, (2002).

2.2. Fluid–rock interaction and gold mineralisation

In the Warrawoona syncline, hydrothermal chert and related

feeder dikes, sulphur vents, pillow lavas, epidote-chlorite–Ca–Na-

plagioclase–Ca-amphibole mineral assemblages and pervasive silici-

fication attest for syn-deposition fluid–rock circulations. This early

Fig. 1. Simplified geological map of the North Pilbara Terrain modified after Van Kranendonk et al. (2002). EPGGT=East Pilbara Granite-Greenstone Terrane, WPGGT=West Pilbara

Granite-Greenstone Terrane, MB=Mallina Basin; MCB=Mosquito Creek Basin, KT=Kuranna Terrane and MGB=Marble bar greenstone belt.

640 N. Thébaud et al. / Earth and Planetary Science Letters 272 (2008) 639–655

fluid–rock interaction is well documented within the weakly deformed

Marble Bar Greenstone Belt and North Pole area. It has been linked to a

primary metasomatism which involved silica-saturated hydrothermal

fluids (Van Kranendonk et al., 2004a; Barley, 1984; Buick and Barnes,

1984; Cullers et al., 1993; DiMarco and Lowe,1989; Kitajima et al., 2001;

Van Kranendonk and Pirajno, 2005). Process proposed for this early

metasomatism include: (1) shallow convecting hydrothermal cells

powered by volcanic activity (Van Kranendonk et al., 2004a; Barley,

1984; Buick and Barnes, 1984; Van Kranendonk and Pirajno, 2005) and

(2) sea-floor diagenetic alteration processes (Knauth and Lowe, 2003).

The occurrence of gold deposits hosted within a kilometer-scale

shear zones (Fig. 3) also provides evidence for significant later fluid–

rock interactions in the Warrawoona syncline. The Warrawonna

syncline is one of the largest mafic–ultramafic-hosted goldfields in the

EPPGT. It produced 745 kg of Au from 25 kt of ore at an average grade

of 29.6 g/t and recent exploration identified a resource of 9.95Mt at 1.0

g/t for the Klondyke deposit (Huston et al., 2001; Hickman, 1983).

These gold deposits are composed of quartz lodes hosted within three

main shear zones: the Klondyke shear zone, the Copenhagen shear

zone and the Fielding's Find shear zone (Fig. 3) (Huston et al., 2001).

These fluid conduits present spectacular networks of quartz±calcite±

sulphide±ankerite veins and are locally lined with heavily brecciated,

fuchsite (Cr-micas) + sericite + pyrite bearing rocks (Fig. 4a and b). Free

gold in quartz has been locally found in quartz veins collected in the

Klondyke Boulder Mine (Fig. 3). Quartz veins are either parallel to Sm(Fig. 4d and e) or oriented at a high angle (N80°) to Lm (Fig. 4g and f).

While the latter family suggests a syn-deformation emplacement,

synfolial quartz veins are often boudinaged (Fig. 4d) and folded

(Fig. 4d and e) suggesting either an early development prior to

deformation or a syn-deformation emplacement. However, the

absence of quartz veins within the less deformed Marble Bar

Greenstone Belt and North Pole area argues for a syn-deformation

origin (Fig. 1) (Thébaud et al., 2006). This late fluid-rock interaction

stage remains poorly understood. In this paper we will thus focus on

Fig. 2. Geological map of the Warrawoona syncline, (after Collins et al., 1998; Kloppenburg et al., 2001).

641N. Thébaud et al. / Earth and Planetary Science Letters 272 (2008) 639–655

fluid processes accompanying the formation of the Warrawoona

syncline and its unusual structural framework.

3. Study area and sampling strategy

In order to detail fluid rock interaction processes attending the

formation of the Warrawoona syncline, this study focuses on the

Fielding's Find shear zone (FFSZ) (Fig. 2). It displays numerous Au

prospect targets along strike and represents a main pathway for fluids

(Fig. 3). This shear zone is oriented parallel to the northern border of

the felsic volcanic Wyman Formation (Fig. 4c). It can be followed over

tens of kilometers and the continuity of the lithological units across

the shear makes it a reasonably good location for fluid rock interaction

studies.

In order to document paleo-fluid circulations within the FFSZ,

samples were collected along three, 800 m long traverses perpendi-

cular to the strike of the shear zone (Fig. 5a and b). Traverses A and B

cut across the whole lithological succession. As the FFSZ is truncated

to the East by the Klondyke shear zone, traverse C was only sampled

within the Wyman Formation. We sampled all the lithologies and

quartz veins in each cross section and analysed the samples for their

oxygen stable isotope composition. Representative fresh and altered

samples for each lithology were analysed for major and trace content.

Numerous quartz veins were sampled within the Warrawoona

syncline mining district, in order to evaluate the significance of the

results obtained from the FFSZ on a regional scale (Fig. 5c).

4. Approach and analytical methods

Stable oxygen isotope geochemistry is routinely used to infer fluid

sources and fluid–rock interaction processes. Added to major and

trace elements geochemistry, it becomes possible to constrain the

composition of the fluids as well as the fluid–rock interaction pro-

cesses. The analytical protocols followed are described hereafter.

4.1. Oxygen isotopes

4.1.1. Bulk rock

Samples were trimmed to remove weathered surfaces, and

crushed to powder (b150 µm). δ18O values were determined by

using the conventional fluorine (BrF5) extraction method (Clayton

and Mayeda, 1963) on 10 mg samples at the CSIRO Exploration and

Mining Department (Sydney, Australia). Oxygen was converted to

CO2, which was then isotopically analysed with a Finnigan 252 mass

spectrometer. Delta values were determined relative to CO2 derived

from a carbonate working standard and then referred to the SMOW

(standard mean ocean water) standard by using αCO2–H2O=1.0412

(O'Neil et al., 1972). The precision on standards and sample replicates

is b±0.2‰.

4.1.2. Quartz veins

δ18O data for the quartz minerals were obtained on a LaserPrep

system on line to a VG Isotech (now GV instruments) Optima dual

inlet (Mattey, 1997) at the Royal Holloway University of London

(Egham, UK). Samples (1.7 mg) were combusted using a CO2 laser in

the presence of excess BrF5. The laser beam is approximately 250 µm

in diameter. Liberated O2 passed through cryogenic traps for clean

up before being directly analysed in the IR-MS. Samples below 90%

of the expected yield were rejected. Three mineral standards have

been analysed during the runs to calibrate the data. Two are inter-

nal, GMG II (a garnet), and QBLC (a quartz) and the international

NBS-30 (a biotite). All δ18O values are reported relative to V-SMOW.

The overall precision on standards and sample replicates is better

than ±0.1‰.

Fig. 3. Gold occurrences in the Warrawoona syncline modified after Ferguson and Ruddock, (2001).Colour legend of Fig. 2. Black dots represent gold occurrences.


Fig. 4. Photographs of hydrothermal veins encountered in the Warrawoona syncline. a) The hydraulic breccia ridge in the core of the Fielding's Find shear zone, geologist for scale

(back arrow). b) Detail of the breccia. c) Panorama looking toward the SE of the Fielding's Find shear zone and its brecciated core. Black lines labelled A, B and C correspond to the three

sections where samples were collected. d) Boudinaged synfolial quartz-carbonates veins. e) Folded quartz vein within Klondyke shear zone. f) En-échelon quartz vein set. g)

Extensional quartz veins oriented at high angle to the stretching lineation.


Fig. 5. Sample locationmap. a)Whole rock sample locationmap. Refer toTable 1 for sample Id. b) Geologic map of the Fielding's Find shear zone, centred on the breccia ridge. A, B and

C correspond to the three sections where samples were collected. c) Quartz vein sample location map.


4.2. Major and trace elements analyses

Major and trace elements bulk rock analyses on mafic, ultramafic

and felsic volcanic rocks were performed at the Service d'Analyses

des Roches et des Minéraux, CRPG, (Nancy, France). Major elements

were obtained by ICP-AES (Jobin-Yvon JY 70), trace elements were

obtained by ICP-MS (Perkin Elmer 5000). All elements were analysed

using international geostandards. The estimated precision is 1 ppm

in the concentration range of 10 to 50 ppm, and b10% for lower

values.

Additionally, major and trace element analyses of fresh end-

member samples of basalt and rhyolite were taken from published

data of the Warrawoona syncline and adjacent greenstone belts

(Cullers et al., 1993; Van Kranendonk and Pirajno, 2005; Weis and

Wasserburg, 1987). These analyses have been carefully selected on the

basis of their affiliationwith the lithologies sampled for the purpose of

this study.

5. Results

5.1. Sample description

A total of 41 rock samples and 31 veins were collected across the

FFSZ. Sampling localities are synthesized in Fig. 5 and the description

of the lithologies sampled is presented below (Fig. 5b).

Fig. 6. Photomicrograph of altered rhyolite showing quartz-bearing microfractures that

cut through an igneous quartz crystal (Sample Pb 02-239).

Table 2

Vein quartz δ18O values

Quartz Traverse A B C Warrawoona syncline

Host rock Sample δ18O

(‰)±0.1‰

Sample δ18O

(‰)±0.1‰

Basalt Pb03–77 12.2 Pb03–256 11.2

Pb03–75 12.9 Pb03–200 12.1

Pb03–201 12.2

Pb03–296 11.9

Pb03–284 11.5

Chlorito-schist Pb03–71 13.0 Pb03–273 11.9

Pb03–69 12.5 Pb03–111 12.5

Pb03–70 13.4 Pb03–70 13.4

Pb03–251 13.7

Volcano-sedimentary layer Pb03–63 12.4 Pb03–63 12.4

Breccia host rock Pb03–66 12.2

Pb03–67 13.1

Pb02–226 14.0

Ultramafic Pb03–47 13.7 Pb03–252 12.3

Felsic volcanic Pb03–59Q 16.1 Pb03–270 13.3

Pb03–205b 14.0 Pb03–212 15.1

Pb03–217 15.1

Pb03–50Q 15.1

Pb03–55Q 16.6

Pb03–214 14.6

Silicified basalt Pb02–P14 14.6

Pb03–257 12.0

See Fig. 5 for sample locations.

Table 1

Bulk rock δ18O values

Traverse A Traverse B Traverse C

Rock type Sample δ18O

(‰)±0.2‰

Sample δ18O

(‰)±0.2‰

Sample δ18O

(‰)±0.2‰

Basalt Pb02–300 7.9 Pb02–210 7.1

Chlorito-schist Pb02–301 15.3 Pb02–212 8.9

Pb02–216 8.5

Silicified Basalt Pb02–303 18.3

Chert Pb02–302 24.1 Pb02–218 20.7

Volcano-sedimentary layer Pb02–304 18.5 Pb02–221 24.7

Pb02–223 22.2

Breccia host Pb02–305 21.4 Pb02–226B 23.2

Ultramafic Pb02–307 7.8 Pb02–229 5.7

PB03–48 7.9 Pb02–230 7.5

Pb02–235 9.9

Chert Pb02–308 25.1 Pb02–237 18.5

Felsic volcanics Pb02–310 21.1 Pb02–239 25.0 Pb03–190 19.5

Pb03–50 20.0 Pb03–239bis 16.9 Pb03–202bis 11.0

Pb03–49 21.3 Pb02–39 18.5 Pb03–209bis 11.9

Pb03–51 23.3 Pb01–44 10.8 Pb03–219 13.1

Pb03–52 18.9 Pb03–159 12.8

Pb03–53 22.4

Pb03–54 22.3

Pb03–55 18.3

Pb03–56 16.1

Pb03–58 12.6

Pb03–59 13.1

Pb03–59b 11.1

Note those samples are presented following the sampling sequence across the FFSZ from north to south. See Fig. 5 for sample location.


To the north−east and away from the shear zone, the lithology

consists of microcrystalline meta basalt. Rare relicts of pillow are

preserved in places and the groundmass contains actinolite, epidote,

chlorite, albite and minor quartz. Toward the shear zone, the rock

develops a strong schistosity and is referred to as a Chloriteschist. The

mineral composition is dominated by chlorite and quartz with chlorite

alignment defining the main planar fabric. The same lithotype directly

adjacent to the FFSZ exhibits an intense silification and consists of

quartz-chlorite and fuchsite with anhedral quartz forming up to 80%

of the groundmass.

On the northern contact with the breccia ridge, a discontinuous

narrow sedimentary package consists of fine grained volcanoclastic

sediments and bedded cherts. The volcanoclastic sediment unit has a

talc-chlorite groundmass and contains up to 1mm phenocryst of quartz.

Sharp centimetre-scale banding is defined by changes in grain size.

The breccia ridge forming the core of the FFSZ is irregular in

shape with a width that ranges from 1 to 10s m. This lithotype

consists of silicified fragments up to 30 cm, within a complex

network of quartz+/−carbonate veins. The mineralogy of fragments

forming 70% of this lithotype is dominated by quartz, fuchsite and

minor sulphides. Overall, silicic alteration is intense along the shear

zone, oblitering primary textures and giving a pronounced light

green colour to the rock.

Ultramafic rocks located on the southern contact of the breccia

ridge show heterogeneous deformation. In low strained domains the

ultramafic unit preserves relics of pyroxene and olivine pseudomorphs

and contains serpentine + chlorite + Mg-carbonate±chromite±Fe

hydroxydes±brucite. In the highly schistose domains the ultramafic

rocks are referred to as talc schist and are mainly composed of talc+

chlorite + Mg-carbonate.

To the south−east of the ultramafic unit and away from the FFSZ

the lithology is dominated by felsic volcanics of theWyman Formation

(Hickman, 1983). Felsic volcanics consist of fine grained rhyolite and

present local columnar jointing. Microscopic observation shows a

porphyritic texture with the groundmass containing 2 mm pheno-

crysts of quartz, K-feldspar and minor white micas. Toward the shear

zone the rhyolites show an increase in strain and silica alteration. On

its shear contact with the FFSZ the rock is intensely silicified and at

microscopic-scale silicification is expressed by numerousmicro quartz

veins (Fig. 6).

Fig. 7. Synthetic δ18O and wt.% SiO2 evolution plotted against a synthetic lithological section of the Fielding's Find shear zone, for each of the traverses (a) and all of the traverses

confounded (b). Legend: Whole rock (filled circles) quartz veins (open circles) and arrows labelled A, B, and C refers to the distribution of δ18O values measured along traverses A+B,

and C.


5.2. Oxygen isotopic results

Bulk rock and quartz vein oxygen isotopic analyses are summarised in

Tables 1 and 2. Fig. 7a shows δ18Omean values of the various lithologies

and associated quartz veins as a function of the distance across the shear

zone for each of the traverses completed. For traverses A and B, δ18O

values of the mafic and felsic volcanic rocks range between +7.9 and

+18.3‰, and+10.8 and+25.1‰, respectively. In both cases, the lower δ18O

values are close tomagmatic values of 5.7‰ (mafic rocks) and7.9‰ (felsic

rocks) (Eiler, 2001), and the higher δ18O values correspond to

hydrothermally-altered silicified rocks. As indicated in Fig. 7, both

lithologies show a continuous increase in δ18O towards the core of the

shear zone. The chert layers, breccia host, and volcano-sedimentary unit

reveal the heaviest values between +18.5 and +25.1‰. The ultramafic

rocks located south of, and in contact with, the breccia ridge have δ18O

nearmagmatic values (Eiler, 2001); (Fig. 7a). Although all traverses show

the same range in 18O-enrichment, the distribution patterns vary

considerably from one traverse to another, the most abrupt variation

being recorded along traverse C (Fig. 7a).

δ18O values of vein quartz sampled in the FFSZ and at other

localities of the Warrawoona syncline plot within a narrow range

between +11.5 and +16.6‰ with a mean value at +13.2±2‰ (Fig. 8).

However, note that vein quartz in felsic volcanics show slightly

heavier δ18O values than those hosted by mafic and ultramafic

rocks.

5.3. Major and trace elements

Major and trace elements compositions are given in Tables 3 and

4. Felsic and mafic volcanic rocks enriched in 18O have a SiO2

content of ca. 91 wt.% and ca. 68 wt.% respectively (Fig. 7b). Silica

contents of unaltered equivalents range from 73 to 78 wt.% for the

felsic rocks and from 45 to 51 wt.% for mafic rocks (Fig. 7b). Altered

rocks are depleted in most elements with Al2O3, K2O, Rb, Th and

REE showing the strongest depletion negatively correlated with the

SiO2 content (Fig. 9). They are enriched in As, Cr, Co, Ni, and V with

a positive correlation with SiO2 (Fig. 9). The anomalous W

enrichment may be due to a grinding artefact as it is related to

the highest SiO2 contents. Ba displays no enrichment in mafic rocks

but ranges between 350 to 1739 ppm in altered felsic rocks (Fig. 9).

Table 3

Major element concentrations of analysed rock samples

Sample SiO2

(wt.%)

Al2O3

(wt.%)

Fe2O3

(wt.%)

MnO

(wt.%)

MgO

(wt.%)

CaO

(wt.%)

Na2O

(wt.%)

K2O

(wt.%)

TiO2

(wt.%)

P2O5

(wt.%)

Felsic volcanics

Fresh

Pb02–44 76.81 12.73 0.35 0.22 0.07 8.21

Wyman 1a 78.27 12.36 1.09 0.02 0.43 0.22 6.11 0.08

Wyman 2a 79.34 11.19 0.43 0.01 0.36 0.03 0.11 6.62 0.11 0.02

Wyman 3a 73.36 13.11 0.08 0.02 0.12 0.21 0.14 11.05 0.50 0.07

Silicified

Pb02–239 90.85 3.37 0.54 0.29 1.34

Pb02–39 92.41 4.76 0.2 0.13 1.41 0.09

Pb02–310 92.80 4.35 1.42

Mafic

Unaltered

Pb02–210 47.60 15.77 10.59 0.15 6.72 13.62 1.09 0.97 0.1

Pb02–300 48.33 15.43 10.58 0.16 6.94 12.41 1.09 0.96 0.11

1596522159652b 50.39 13.57 10.48 0.23 8.28 7.69 4.25 0.11 0.59 0.05

1596532159653b 48.75 13.08 10.09 0.22 7.33 12.83 1.8 0.03 0.56 0.05

1778902177890b 45.27 12.19 9.45 0.16 5.19 11.44 0.18 1.32 0.6 0.08

1596482159648b 51.36 13.77 13.77 0.19 5.32 9.77 3.52 0.22 1.05 0.11

1596492159649b 51.93 13.35 13.35 0.2 6.92 8.54 3.67 0.28 0.94 0.1

Silicified

Pb02–301 68.32 6.31 7.83 0.07 6.79 2.79 0.25

Silicified ~95.00

Chert

North Pole 2b 97.96 0.16 1.1 0.04 0.04 0.12 0.01 0.01 0.03

Pilbara 3c 97.80 0.49 0.57 0.04 0.01 0.01 0.08 0.02

Breccia host and Warrawoona Chert ~97.00

~ Estimated value.aCullers et al., 1993.bVan Kranendonk and Pirajno, 2005.cWeis and Wasserburg, 1987.

Fig. 8. Vein quartz δ18O values of the different vein types sampled in the Warrawoona

syncline.


Table 4

Trace element concentrations of felsic volcanics (in ppm)

Unaltered Silicified

PB02–44 Wyman 1a Wyman 2a Wyman 3a PB02–239 PB02–39 PB02–310

V bd.l 3.0 bd.l 37.0 15.4 8.4

Cr bd.l 3.0 bd.l 18.0 8.2 8.6 8.9

Co 79.7 3.0 bd.l bd.l 277.2 119.4 204.2

Ni bd.l 5.0 2.0 7.0 10.6 6.5

Ga 17.3 17.0 bd.l bd.l 5.1 5.6 5.7

Ge 1.1 Nd Nd Nd 1.1 0.8 1.3

As bd.l bd.l bd.l bd.l 5.6 4.3 0.8

Rb 231.5 219.1 129.2 127.4 28.8 20.9 28.1

Nb 10.1 10.0 5.0 4.0 2.8 1.5 3.7

Mo 0.8 Nd Nd Nd 1.6 0.9 1.2

Sn 4.3 Nd Nd Nd 1.1 0.5 1.3

Sb bd.l Nd Nd Nd 1.0 4.5 bd.l

Cs 2.4 Nd Nd Nd 1.7 0.7 0.6

Ba 127.3 135.0 480.0 442.0 350.1 1739 1180

La 18.8 34.6 26.1 23.2 4.4 8.9 16.8

Ce 62.6 68.3 48.3 45.5 10.1 15.9 39.1

Pr 5.1 Nd Nd Nd 1.2 1.5 4.8

Nd 17.8 29.2 18.9 19.7 4.3 5.4 17.8

Sm 4.0 5.34 3.5 3.6 0.9 0.9 3.3

Eu bd.l 0.2 0.4 0.9 bd.l 0.3 0.2

Gd 3.3 3.9 3.3 3.2 0.5 0.5 2.0

Tb 0.6 Nd Nd Nd 0.1 bd.l 0.3

Dy 3.8 3.2 3.1 3.3 0.5 0.5 1.4

Ho 0.8 Nd Nd Nd bd.l 0.1 0.2

Er 2.3 2.3 2.0 2.1 0.4 0.3 0.6

Tm 0.3 Nd Nd Nd bd.l bd.l bd.l

Yb 2.9 2.7 2.1 2.2 0.4 0.2 0.7

Lu 0.4 0.4 0.3 0.3 N0.1 bd.l 0.1

Hf 4.1 Nd Nd Nd 0.4 1.1 0.9

Ta 2.6 Nd Nd Nd 3.6 1.4 2.6

W 725.3 Nd Nd Nd 2438 1154 1761

Bi bd.l Nd Nd Nd bd.l 0.6 bd.l

Pb 48.8 Nd Nd Nd 1.6 bd.l 12.3

Zn 12.7 19.0 11.0 8.0 18.8 bd.l 9.5

Sr 4.4 11.7 12.4 26.0 2.3 3.2 2.5

Y 24.8 29.0 22.0 18.0 4.1 2.9 7.6

Zr 94.7 101.0 139.0 177.0 22.9 57.7 41.2

Th 28.96 25.0 11.0 6.0 6.7 2.7 10.5

U 4.5 4.0 2.0 2.0 1.3 0.4 2.0

Trace element concentrations of mafic volcanics (in ppm)


PB02–210 PB02–300 159652b 159653b 177890b 159649b 159648b PB02–301

V 264.9 251.2 225.0 224.0 285.0 218 235

Cr 322.1 296.3 333.0 346.0 408.0 366 186 3515.0

Co 76.8 68.5 Nd Nd Nd Nd Nd 84.5

Ni 143.5 145.4 106.0 103.0 104.0 78 66 659.3

Ga 89.8 61.6 Nd Nd Nd Nd Nd 45.0

Ge 19.1 17.1 12.1 11.5 11.2 14.9 16.2 6.7

As 1.4 1.4 1.7 1.6 1.1 1.7 1.5 1.5

Rb 2.2 1.6 2.0 3.1 11.6 5.5 9.7

Nb 0.7 1.0 3.0 1.0 32.3 8 7

Mo 3.6 3.4 1.4 1.3 2.0 5.5 5.9 0.5

Sn N0.1 N0.1 Nd Nd Nd Nd Nd

Sb 0.8 0.8 Nd Nd Nd Nd Nd

Cs Nd Nd 0.3 0.1 0.4 0.6 0.2

Ba 29.4 29.4 33.0 7.0 2077.0 154 117 3.8

La 6.5 6.4 2.1 2.1 3.3 8.9 9.7 0.8

Ce 14.9 14.6 5.5 5.4 8.5 20.3 22.7 1.8

Pr 2.1 1.9 0.8 0.8 1.1 2.8 3.2 0.3

Nd 9.5 9.2 4.0 3.8 5.7 11.88 13.49 1.2

Sm 2.6 2.7 1.3 1.3 1.8 3.2 3.7 0.3

Eu 0.9 1.0 0.4 0.5 0.5 1 1.2 N0.1

Gd 3.0 3.1 1.8 1.8 2.1 3.9 4.3 0.7

Tb 0.5 0.4 0.4 0.3 0.3 0.64 0.74 N0.1

Dy 3.5 3.5 2.1 2.1 2.4 4.1 4.5 1.1

Ho 0.6 0.6 0.5 0.5 0.6 0.9 0.9 0.1

Er 2.0 2.1 1.4 1.4 1.6 2.6 2.8 0.6

Tm 0.2 0.2 Nd Nd Nd Nd Nd N0.1

Yb 2.2 2.1 1.4 1.4 1.5 2.5 2.6 0.5

Lu 0.2 0.2 0.2 0.2 0.3 0.4 0.4 N0.1

Hf 1.8 1.8 Nd Nd Nd Nd Nd 0.4

Ta 0.6 0.6 Nd Nd Nd Nd Nd 0.3


Chondrite-normalized REE patterns (Fig. 10) are relatively depleted

for the altered units.

6. Interpretations: a case for polyphase fluid circulation in the

Warrawoona Syncline

In the EPPGT we have documented that the Warrawoona Group

preserved testimonies of an early fluid–rock interaction history which

started with a syn-deposition near-seafloor low-temperature hydro-

thermal metasomatism. Recently, Van Kanendonk and Pirajno (Van

Kranendonk and Pirajno, 2005) proposed a scenario involving cyclic

volcanic and hydrothermal stages leading to interlayered volcanic

rocks and hydrothermal cherts. In this section, we focus on the syn-

deformation fluid circulation history and fluid–rock interaction

processes. On the basis of the dataset presented in this study two

successive fluid–rock interaction stages are documented.

6.1. Hydrothermal alteration zoning pattern of the Fielding's Find shear

zone

Towards the shear zone, basalts and rhyolites show an increase in

strain and hydrothermal alteration, including strong silicification. This

evolution is coupled with a pronounced zoning of major and trace

element data and oxygen isotope systematics across the FFSZ. This

alteration zonation is consistent with the FFSZ shear zone having

acted as preferential channel way for fluids. The various lithologies

(mainly rhyolite and basalt) are affected differently by this hydro-

thermal alteration. In what follows, the nature of the zoning is

reviewed and its origin is investigated.

6.1.1. Oxygen isotopic zonation

Oxygen isotopic results show that the bulk rock δ18O values from

both sides of the shear zone increase progressively towards its centre.

This observation is indicative of a fluid-buffered systemwithin which

the isotopic shift exhibited by the rocks was isotopically controlled by

the circulating fluids. This isotopic zonation is particularly clear in the

felsic volcanics of theWyman Formation for traverses A and B. (Fig. 7).

Traverse C shows an incomplete trend in comparisonwith traverses A

and B (Fig. 7). This incomplete trend may be explained by tectonic

removal of part of the lithological succession as the FFSZ is truncated

by the Klondyke Shear zone to the east.

Similar enrichment in heavy oxygen isotope is commonly

associated with the formation of cherts in the Archaean (Knauth

and Lowe, 2003; de Wit et al., 1982; Paris et al., 1985). However in

the Red Lake District (Superior Province, Canada) Kerrich et al.

(1981) documented similar extreme (up to 18‰) 18O-enrichment in

W 332.7 236.8 Nd Nd Nd Nd Nd 323.6

Pb 2.4 1.4 0.5 0.0 2 2

Zn 74.3 74.6 72.0 71.0 72.0 76 81 50.9

Sr 201.1 151.8 30.0 82.0 52.0 130 174 12.1

Y 21.1 20.5 13.0 12.9 13.6 24.6 26.6 7.0

Zr 73.1 71.6 33.0 29.0 46.0 110 118 13.7

Th 1.1 1.0 0.3 0.3 0.7 1.6 1.7 0.1

U 0.3 0.3 0.1 0.4 0.4 0.1

a Cullers et al., 1993.b Van Kranendonk and Pirajno, 2005.

Table 4 (continued)


PB02–210 PB02–300 159652b 159653b 177890b 159649b 159648b PB02–301

Fig. 9. Spider diagram showing elemental concentrations for mafic and felsic volcanics from the Fielding's Find shear zone. Concentrations are normalized to least-altered rhyolites

from the Wyman Formation (PB02–44) and mafic material from the Euro Basalt (PB02–210). a) Silicified felsic volcanics; b) Silicified basaltic volcanics. Dilution and dissolution/

precipitation lines represent the reference lines from which elements have been enriched or depleted according to the silicification scenario.


metabasalts (Kerrich et al., 1981). Such enrichment in heavy oxygen

isotopes suggests that the alteration process was driven by interaction

with large volumes of low temperature (~90–160°C) hydrothermal

fluids as shown elsewhere in the Pilbara (Van Kranendonk and Pirajno,

2004; Oliver and Cawood, 2001) and other Mesoarchaean terranes

(Knauth and Lowe, 2003; Paris et al., 1985; DeWit et al., 1982).

In ophiolites and modern-day seafloors, 18O-enrichment up to ca

14‰ has been described and interpreted as seawater hydrothermal

alteration (Alt et al., 1986; Alt and Teagle, 2003; Heaton and Sheppard,

1977; Ito and Clayton, 1983; Putlitz et al., 2001; Stakes and Taylor,

1992; Stakes and O'Neil, 1982). Seafloor hydrothermal alteration is

commonly described as a two-stage process (Putlitz et al., 2001).

During the first stage, the infiltration of seawater within a thick crustal

section as well as the rising temperature of the downward moving

fluid enables prolonged oxygen exchange, hence raising the δ18O of

the water. The second stage corresponds to the reverse process. As

the temperature rises, the 18O/16O enriched water flows upward in

discharge zones where the host rocks are progressively enriched with

decreasing temperatures. Accordingly, we suggest that the isotopic

enrichment of the volcanic rocks lining the FFSZ may be associated

with near sea-floor, hydrothermal convection cells at low temperature

(90–160°C).

In ultramafic rocks, the δ18O range (+5.7 to +7.9‰) is characteristic

of serpentinite (Agrinier et al., 1995). The lack of silicification and their

lower 18O-enrichment compared to adjacent units remain unclear.

One could speculate that, before its exhumation, this unit was altered

in a deeper crustal level, hence higher temperature environment,

which would have limited the 18O-enrichment.

6.1.2. Mineralogical and whole-rock major and trace elements zonation

In altered basalts and rhyolites, XRF analyses reveal silica contents

of up to 95 wt.% (Fig. 7b). Clearly, the foremost feature of the alteration

around the FFSZ is silicification. Figs. 9 and 10 show that with the

exception of V, Co, Ni, Cr and As, silica enrichment was accompanied

by a marked loss of most major and trace elements. Since mafic and

ultramafic rocks represent the dominant lithology of the greenstones

and present high contents of V, Co, Ni, Cr and As, we suggest that the

fluid responsible for the alteration may have circulated through large

volumes of these rocks.

In the case of the rhyolites, the silicification process was

accompanied by a loss of K and Al. This may be explained through

two different processes (Fig. 9). The first one is a dilution process due

to an external silica input, leading to a volume increase. Alternatively,

the second one involves an isovolumic process associated with the

dissolution of K-feldspar and precipitation of silica into micro-

fractures according to the reaction:

K � feldspar þ Hþ þ H2O ¼ AlðOHÞ3ðaqÞ þ Kþ þ 3Quartz ð1Þ

where Al(OH)3(aq) and K+ were removed by the hydrothermal fluid.

On the basis of the modal composition of a fresh rhyolite (quartz:

46.4%, K-feldspar: 39%, muscovite: 14.6%,; Pb02-44), the chemical

evolution resulting from reaction (1) was compared with that of a

simple dilution process (Fig. 11). Bulk rock density change has

negligible impact on the mass balance calculations (Grant, 1986;

Gresens, 1967). Calculations show that 100% isovolumic replacement

of K-feldspar by quartz can explain the composition of the altered

rock. In contrast, the dilution scenario requires a volume increase by a

factor of 4 to 5 which is incompatible with field observations.

The interpretation of alteration in the basalts can be made in a

similar fashion. Away from the FFSZ, the modal composition of the

basalt (e.g., albite: 12.85%, actinolite: 29.37%, chlorite: 10.38%, epidote:

41.47%, quartz: 5.96%, sample Pb 02-300) is transformed into quartz-

chlorite bearing assemblages next to the FFSZ. Hence, the silicification

process seems to be driven by mineralogical reactions where albite,

clinozoicite/epidote and amphibole were replaced by quartz and

chlorite. In the latter, the destabilisation of the calcic minerals

provides Ca, Na, and Al for the crystallization of chlorite. Calculations

show that the depletion in Al2O3, Fe2O3 and CaO can be explained by

100% replacement of albite and epidote by quartz (Fig. 11).

Alternatively, the chemical change could be the result of the

introduction of silica from an external source. This dilution model

requires a volume increase by a factor of 2 to 3.

In both cases, the chemical change observed for major elements

associated with the silification of the felsic and mafic units can be

explained by both models. However, dilution and mineral replace-

ments are not mutually exclusive.

Chondrite normalized REE plots (Fig. 10) show depletion of REE

elements in the silicified mafic and felsic rocks relative to their

equivalent away from shear zones. The spider diagrams show that in

the altered felsic rocks HREE are systematically more depleted (by a

factor ~5) relative to LREE (Fig. 10). In contrast to felsic rocks, LREE

appear to bemore depleted than the HREE in silicifiedmafic rocks. The

REE composition of a hydrothermally altered rock is a complex

function of the concentration and speciation of REE in the fresh rock.

In addition, it is also a function of the REE composition of the

hydrothermal fluid, and the speciation of REE in the fluid, which is

controlled by pressure, temperature, pH, Eh and concentrations of

ligands such as Cl− and F− (e.g., Bach et al., 2003; Bau, 1991). It is

interesting to note that a simple difference in speciation driven by

fluid–rock interaction could explain the contrasting REE behaviour in

the basalts and in the rhyolites.Assuming that fluid pH was buffered by the quartz-muscovite-K-

feldpar assemblage (reaction 1), the pHmay have been between 3 and

4 at 160 °C (aK+=0.01 m to 1 m). In such a case, REE3+ may have been

transported as REE chloride complexes or REE fluoride complexes. The

stability of the REE complexes with hard ligands such as F− increases

for the light to the heavy REE, whereas softer ligands such as Cl− are

Fig. 10. REE chondrite normalized plots for the a) felsic and b) mafic volcanics from the

Fielding's Find shear zone.


not as specific (Haas et al., 1995). Hence, it is possible that the

preferential leaching of HREE in the rhyolites reflects fluoride

complexing. In this context, the much smaller differentiation between

HREE and LREE observed in the basalts may reflect the fact that

rhyolitic rocks are good sources of fluorine.

In contrast to the altered rhyolites and basalts, the ultramafics do

not present evidence of intense silicification, which argues against

dilution as a controlling mechanism during fluid circulation. Petro-

graphical observations indicate that talc developed after serpentinites

in the poorly-deformed portions of rocks according to the reaction:

antigoriteðMg3Si2O5ðOHÞ4Þ þ 2 SiO2ðaqÞ¼ talcðMg3Si4O10ðOHÞ2Þ þ H2O ð2Þ

This reaction could account for the buffering of ambient dissolved

aqueous silica and production of water without invoking dissolution–

precipitation or dilution processes. Furthermore, in the absence of

significant fluid exchange, dehydration reactions of this type will have

little effect on bulk-rock oxygen isotope compositions as recorded by

the relatively small 18O-enrichment of ultramafic rocks compared to

mafic and felsic rocks.

6.1.3. Genetic models for the alteration zonation

Two models can explain the alteration pattern along the FFSZ. In

the first model, the alteration pattern results from syntectonic fluids

channelled into the shear zone (Fig. 12b). In the second model, the

alteration pattern results from the fortuitous structural juxtaposition

of geochemical trends inherited from a previous stage of seafloor

hydrothermal alteration (Fig. 12a). In the latter, the shear zone

corresponds to the strongly deformed axial planar surface where both

limbs of the Warrawoona syncline have come into contact. Along this

contact, the juxtaposition of the syn-depositional isotopic trends

explains the apparent bell-shaped oxygen isotope alteration pattern

(Fig. 12a). Several observations argue against this interpretation.

Fig. 11. Diagrams showing the evolution of major elements between unaltered and silicified samples as a function of the silica content for the felsic rocks (left) and the mafic rocks

(right). The plain line represents the evolution of the concentrations as a function of the dilution volume ratio. The dashed line represents the evolution of the concentrations

according to successive replacement of K-feldspar in the ryolite, and of the total replacement of albite and clinozoicite/epidote in mafic rocks during the silicification process.


Fig. 12. Possible scenarios regarding the alteration halo across the Fielding's Find shear zone (see text). a) Pre-deformation model: This scenario commences with development of an

enriched isotopic profile during an early syn-depositional sea-floor alteration stage (a1) followed by the juxtaposition of both limbs along the syncline axial planar surface during

greenstone burial and coeval granitic exhumation (a2). b) Syn-deformation model: The alteration zonation was formed during the early stage of deformation and shear zone

development.


Firstly, it seems implausible that shearing, deformation and coeval

fluidflowparallel to the axial surface did not affect the original alteration

trend. For instance the core of the FFSZ, hosting the breccia, is silicified

and fuchsite-rich, which highlights a syn-deformation fluid–rock

interaction process. Secondly, the monotonous isotopic trends recorded

by the rhyolites and basalts are not compatible with the metric to

hectometric wavelength of parasitic folds, as this folding should have led

to a more random distribution of oxygen isotopic compositions.

6.2. Veining episode

The quartz veins are in isotopic disequilibrium with their altered

host, which suggest that a fluid circulation event followed the

silicification around the FFSZ. Vein quartz display a relatively

restricted range of δ18O values across the Warrawoona syncline,

with a mean value of +13.2 ±2‰. This suggests that quartz

precipitated from a homogeneous fluid under near-isothermal

conditions. The slightly heavier δ18O value of the quartz veins hosted

in felsic rocks, compared to that hosted in mafic rocks, may indicate

partial and local buffering by the host rock (Fig. 7).

Commonly, an estimated or measured temperature of the forming

fluid is used in conjunction with the mineral oxygen isotopic

composition to constrain the δ18O value of the fluid from which it

precipitated. This δ18Ovalue canbe used in turn to document the origin

of the fluid reservoir. Detailed fluid inclusion studies conducted on

similar quartz veins in the Warrawoona syncline, showed CO2–NaCl–

H2O–CH4 pseudo-secondary inclusions that yielded homogenisation

temperatures between 234 and 372°C (Thébaud et al., 2006). Using

Zheng (1993) quartz/water fractionation equation, the calculated δ18O

of the water in equilibriumwith these quartz veins at 234–372 °C is in

the range +2.4 to +12.1‰. This range is indicative of a magmatic or

metamorphic origin. Thébaud et al. (2006) detailed fluid inclusion

study showed that the quartz-forming fluids contained high Cl/Br

ratios, weak salinities (0.5–7wt.%NaCl equivalent) and significant base

metal and potassium concentrations. Such a composition is regarded

as the signature of a mixed composition involving magmatic and

metamorphic fluid sources (Thébaud et al., 2006). Accordingly, it is

proposed that quartz veins were formed during a late stage of the

deformation history during infiltration of a hotter fluid of possibly

metamorphic ormagmatic origin (Thébaud et al., 2006). It is suggested

that the large production of potassic melt and progressive burial of the

greenstones led to the production of magmatic and/or metamorphic

fluids. These deeply released fluids could account for the formation of

the syn-tectonic quartz veins throughout the Warrawoona syncline.

7. Remarks on Au-prospectivity of Mesoarchaean and Neoarchaean

terranes

Most models accounting for the exceptional Au prospectivity of

Neoarchaean terranes advocate that goldmineralization resulted from

deeply sourced fluids channelled along crustal-scale shear zones

coeval with the regional metamorphic peak and thus suggesting a

single mineralisation event (‘crustal continuum’, Groves, 1993; Groves

et al., 1998). In contrast to this widely acceptedmodel, fluid circulation

Fig. 13. Synthetic diagram showing a conceptual model of fluid–rock interaction attending greenstone sagduction in the Warrawoona Syncline and granite emplacement (a) Oceanic

hydrothermal alteration stage responsible for possible gold pre-concentration in shear zones. (b) Veining/lode gold deposit stage associated with infiltration of deep magmatic/

metamorphic fluids that caused gold concentration in quartz veins.


studies in the much less prospective MesoArchaean suggest that lode

Au deposits developed through a protracted structural and hydro-

thermal history. This raises the possibility that the apparent homo-

geneity of Neoarchaean orogenic gold deposits reflects simply the

obliteration of early fluid–rock interaction histories during the 2.65 Ga

global event (Rey et al., 2003), where earlier plumbing systems were

reworked and gold re-mobilised into giant gold deposits.

7.1. Protracted history of hydrothermal circulation in Mesoarchaean

Our study contributes to a growing body of evidence arguing for a

complex history of hydrothermal circulation in Mesoarchaean

terranes. The best preserved Mesoarchaean terranes are located in

the Pilbara Craton and the Barberton Greenstone Belt (3.57–3.08 Ga,

South Africa). Results of this study suggest that that the Warrawoona

syncline was the locus of a polyphase fluid–rock interaction involving

fluids of contrasted temperatures and sources. These various stages of

fluid infiltration are summarized in Fig. 13. Based on oxygen isotopic

data, two fluid circulation stages have been constrained. The first stage

was responsible for significant 18O and silica enrichment along the

shear zone. This alteration pattern is interpreted as the product of

low temperature hydrothermal alteration developed either prior to

or during initial deformation of the greenstones. The later fluid

circulation stage is documented by gold-bearing quartz veins with

homogeneous δ18O values. These veins are interpreted as the product

of metamorphic and/or magmatic fluids released during a later stage

of deformation. A similar protracted fluid circulation history has been

documented in the Barberton greenstone belt. As in the Pilbara the

history started with syn-depositional pervasive metasomatism asso-

ciated with silicification, ferruginisation, and 18O enrichment of the

greenstones (Knauth and Lowe, 2003; de Wit et al., 1982; Lowe and

Byerly, 1986). This metasomatism has been interpreted as the result of

near sea-floor fluid–rock interaction related to hydrothermal activity

(e.g., DeWit et al., 1982; de Ronde et al., 1994). This early process has

been associated with stratabound gold concentration within early

sedimentary units (Maiden, 1984; Viljoen, 1984). Following this

episode, syn-tectonic quartz veins and associated lode gold deposits

have been documented (e.g. de Ronde et al., 1992). As in the Pilbara

they have been interpreted to develop during granite emplacement

and associated deformation (de Ronde et al., 1992).

7.2. Why are Mesoarchaean and Neoarchaean gold mineralisations

different?

While Neoarchaean terranes preserve a single dominant fluid

infiltration and mineralisation event, Mesoarchaean terranes have

preserved the geochemical fingerprints of numerous hydrothermal

episodes. Neoarchaean lode gold mineralisations are associated with a

major tectono-thermal event at around 2.67±0.05 Ga (Rey et al., 2003;

Qiu and Groves, 1999). It is therefore possible that all fluid processes

developed prior to the main mineralization event were strongly

overprinted, thus preserving the signature of the latest fluid

circulation event interpreted in turn as reflecting a single ‘crustal

continuum’ event. It is worth noting that recent investigations in the

Yilgarn Craton point toward a polyphase mineralisation history

involving contrasting fluid sources (Bateman and Hagemann, 2004;

Brown and Johnson, 2003). It seems therefore likely that Mesoarch-

aean and Neoarchaean terranes have undergone similar polyphase

and protracted fluid circulation histories. We suggest, as others before

us, that these successive hydrothermal circulations are the first steps

required to form economic gold deposits (Hutchinson, 1993; Bateman

and Hagemann, 2004; Brown and Johnson, 2003).

In the Warrawoona syncline, a plumbing system in which Au-

bearing fluids were focussed in a pre-existing structural channel is

documented. As the mineralogy along the channels was already

equilibrated with hydrothermal fluids, the later, deeper mineralised

fluids could have reached upper crustal levels with minimal

interaction with the host rocks. On their way to the deposition site,

these deeper fluids may have reworked pre-enriched gold-bearing

alteration zones, further upgrading their Au-content (Hutchinson,

1993). Consequently, the early development of an active plumbing

system coupled with ore pre-concentration may be of crucial

importance for the formation of metal deposits. Nevertheless, since

this protracted history is recognised in both Mesoarchaean and

Neoarchaean terranes, it does not alone explain their contrasting

prospectivity. This suggests that the 2.67±0.05 Ga tectono-thermal

event was a key feature for the formation of World class lode gold

deposits. As it impacted only on Mesoarchaean cratons, possibly

because these were already differentiated and stabilized, this event

did not lead to the remobilization of older plumbing systems as the

one described here.

8. Conclusions

This paper argues that the Warrawoona syncline was the locus of a

protracted and polyphase fluid circulation history involving fluids of

contrasted temperatures and sources. Based on oxygen isotopic, major

and trace elements data, and fluidmodelling, we have documented two

stages of fluid infiltrations. One is highlighted by an intense silica

alteration along a main shearzone. One stage is interpreted as the

product of low temperature fluid–rock interactions prior to or during

shear zone development. This infiltration was followed by the

circulation of syn-tectonic metamorphic and/or magmatic fluids that

led to the precipitation of numerous gold-bearing quartz veins. The

protracted and polyphase fluid circulation history documented here is

similar to that documented in otherMesoarchaean greenstonebelts.We

suggest that such a polyphase fluid–rock interaction history may be of

crucial importance for the formation of economic gold deposits.

Acknowledgments

This paper has benefited from countless discussions with many

colleagues from the ‘Institut de Physique du Globe’ in Paris and the

University of Sydney in Australia. The first author would like to

acknowledge I. Gonzalez-Alvarez for last minute comments and

discussions. We are grateful for thorough and helpful comments from

the reviewers. This research was in part supported under the Australian

ResearchCouncil’sDiscovery funding scheme (ARCDP0342933) and the

Institut de Physique du Globe de Paris. M. Van Kranendonk publishes

with permission of the Director of the Geological Survey of Western

Australia.

References

Agrinier, P., Hekinian, R., Bideau, D., Javoy, M.,1995. O and H stable isotope compositionsof oceanic crust and upper mantle rocks exposed in the Hess Deep near theGalapagos Triple junction. Earth and Planetary Science Letters 136, 183–196.

Alt, J.C., Teagle, D.A.H., 2003. Hydrothermal alteration of upper oceanic crust formed at afast-spreading ridge: mineral, chemical, and isotopic evidence from ODP Site 801.Chemical Geology 201, 191–211.

Alt, J.C., Muehlenbachs, K., Honnorez, J., 1986. An oxygen isotopic profile through theupper kilometer of the oceanic crust, DSDP Hole 504B. Earth and Planetary ScienceLetters 80, 217–229.

Bach, W., Roberts, S., Vanko, D.A., Binns, R.A., Yeats, C.J., Craddock, P.R., Humphris, S.E.,2003. Controls of fluid chemistry and complexation on rare-earth element contentsof anhydrite from the Pacmanus subseafloor hydrothermal system, Manus Basin,Papua New Guinea. Mineralium Deposita 38, 916–935.

Barley, M.E., 1984. Volcanism and hydrothermal alteration, Warrawoona Group, East Pilbara.In: Muhling, J.R., Groves, D.I., Blake, T.S. (Eds.), Archaean and Proterozoic basins of thePilbara, Western Australia: Evolution and Mineralization Potential 9. GeologicalDepartment Extension University of Western Australia Special Publications, pp. 23–36.

Bau, M., 1991. Rare-earth element mobility during hydrothermal and metamorphicfluid–rock interaction and the significance of the oxidation state of europium.Chemical Geology 93, 219–230.

Blewett, R.S., 2002. Archaean tectonic processes: a case for horizontal shortening in theNorth Pilbara Granite-Greenstone Terrane, Western Australia. Precambrian Research113, 87–120.


Bateman, R., Hagemann, S., 2004. Gold mineralisation throughout about 45 Ma ofAchaean orogenesis: protracted flux of gold in the Golden Mile, Yilgarn craton,Western Australia. Mineralium Deposita 39, 536–559.

Brown, S.M., Johnson, C.A., 2003. Constraints on the composition of ore fluids andimplications for mineralising events at the Cleo gold deposit, Eastern GoldfieldsProvince, Western Australia. Australian Journal of Earth Sciences 50, 19–38.

Buick, R., Barnes, K.R., 1984. Cherts in the Warrawoona Group: early Archean silicifiedsediments deposited in shallow water environments. University of Western AustraliaGeology Department University Extension Special Publication, 9, pp. 37–53.

Clayton, R.N., Mayeda, T.K., 1963. The use of bromine pentafluoride in the extraction ofoxygen from oxides and silicates for isotopic analysis. Geochimica and Cosmochi-mica acta 27, 43–52.

Collins, W.J., 1989. Polydiapirism of the Archean Mount Edgar Batholith, Pilbara Block,Western Australia. Precambrian Research 43, 41–62.

Collins, W.J., Van Kranendonk, M.J., Teyssier, C., 1998. Partial convective overturn ofArchaean crust in the east Pilbara Craton, Western Australia: driving mechanismsand tectonic implications. Journal of Structural Geology 20, 1405–1424.

Cullers, R.L., DiMarco, M.J., Lowe, D.R., Stone, J., 1993. Geochemistry of a silicified, felsicvolcaniclastic suite from the early Archaean Panorama Formation, Pilbara Block,Western Australia: an evaluation of depositional andpost-depositional processeswithspecial emphasis on the rare-earth elements. Precambrian Research 60, 99–116.

Delor, C.A., Burg, J.P., Clarke, G., 1991. Relations diapirisme-metamorphisme dans laProvince du Pilbara (Australie Occidentale): implications pour les régimesthermiques et tectoniques à l'Archéen. Comptes Rendus de l'Académie des Sciencesde Paris 312, 257–263.

de Ronde, C.E.J., Spooner, E.T.C., de Wit, M.J., Bray, C.J., 1992. Shear zone-related, Auquartz vein deposits in the Barberton greenstone belt, South Africa: field andpetrographic characteristics, fluid properties and light stable isotope geochemistry.Economic Geology 87, 366–402.

de Ronde, C.E.J., de Wit, M.J., Spooner, E.C., 1994. Eraly Archaean (N 3.2 Ga) Fe-oxyde-rich, hydrothermal discharge vents in the Barberton greenstone belt, South Africa.Geological Society of America Bulletin 106, 84–104.

DiMarco, M.J., Lowe, D.R., 1989. Petrography and provenance of silicified early Archaeanvolcaniclastic sandstones, eastern Pilbara Block, Western Australia. Sedimentology36, 821–836.

de Wit, M.J., Hart, R., Martin, A., Abbot, P., 1982. Archaean abiogenetic and probablebiogenetic structures associated with mineralized hydrothermal vent systems andregional metassomatism, with implications for the greenstone belt studies.Economic Geology 77, 1783–1802.

Eiler, J.M., 2001. Oxygen Isotope Variations of Basaltic Lavas and Upper Mantle Rocks. In:Valley, D.R., Cole (Eds.), Reviews in mineralogy: Stable Isotopes Geochemistry 43.Mineralogical Society of America, pp. 319–364.

Ferguson, K.M., Ruddock, I., 2001. Mineral occurrences and Exploration Potential of theEast Pilbara Geological Survey of Western Australia. Report 81, 114p.

Goldfarb, R.J., Groves, D.I., Gardoll, S., 2001. Orogenic gold and geologic time: a globalsynthesis. Ore Geology Reviews 18, 1–75.

Grant, J., 1986. The isocon diagram; a simple solution to Gresens' equation formetasomatic alteration. Economic Geology 81, 1976–1982.

Gresens, R., 1967. Composition–volume relationships of metasomatism. ChemicalGeology 2, 47–65.

Groves, D., 1993. The crustal continuum model for late Archaean lode-gold deposits ofthe Yilgarn Block, Western Australia. Mineralium deposita 28, 366–374.

Groves, D.I., Phillips, G.N., 1987. The genesis and tectonic control on Archaean golddeposits of the western austalian shield— a metamorphic replacement model. OreGeology Reviews 2, 287–322.

Groves, D.I., Goldfarb, R.J., Gebre-Mariam, M., Hagemann, S.G., Robert, F., 1998. Orogenicgold deposits: a proposed classification in the context of their crustal distributionand relationship to other gold deposit types. Ore Geology Reviews 13, 7–27.

Groves, D.I., Condie, K.C., Goldfarb, R.J., Hronsky, J.M.A., VielreicherSecular, R.M., 2005.Changes in global tectonic processes and their influence on the temporaldistribution of gold-bearing mineral deposits. Economic Geology 100, 203–224.

Haas, J.R., Shock, E.L., Sassani, D.C., 1995. Rare earth elements in hydrothermal systems:estimates of standard partial molal thermodynamic properties of aqueouscomplexes of the rare earth elements at high pressures and temperatures.Geochimica et Cosmochimica Acta 59, 4329–4350.

Heaton, T.H.E., Sheppard, S.M.F., 1977. Hydrogen and oxygen isotope evidence for sea-water hydrothermal alteration and ore deposition, Troodos Complex, Cyprus.Volcanic Processes in Ore Genesis 7, Geolical Society of London Special Publication,pp. 42–57.

Hickman, A.H., 1984. Archaean diapirism in the Pilbara block, Western Australia. In:Kröner, A., Greiling, R. (Eds.), Precambrian tectonics illustrated. E. Schweizer-barts'che Verlagsbuchlandlung, pp. 113–127.

Hickman, A.H., 1983. Geology of the Pilbara Block and its environs. 1981/36. GeologicalSurvey of Western Australia, Perth. 268 pp.

Hickman, A.H., 2001. East Pilbara diapirism: new evidence from mapping. WesternAustralia Geological Survey Record 5, 23–25.

Huston, D., Blewett, R., Mernaugh, T., Sun, S.S., Kamprad, J., 2001. Gold Deposits of thePilbara Craton: Results of AGSO Research, 1998–2000. 32 pp.

Hutchinson, R.W.,1993. Amulti-stage, multi-process genetic hypothesis for greenstone-hosted gold lodes. Ore Geology Reviews 8, 349–382.

Ito, E., Clayton, R.N., 1983. Submarine metamorphism of gabbros from the Mid-CaymanRise: an oxygen isotopic study. Geochimica Cosmochimica Acta 47, 535–546.

Kerrich, R., Wyman, D., 1990. Geodynamic setting of mesothermal gold deposits: Anassociation with accretionary tectonic regimes. Geology 18, 882–885.

Kerrich, R., Fryer, B., Milner, K.J., Peirce, M.G., 1981. The geochemistry of gold-bearingchemical sediments, Dickenson mine, Red Lake, Ontario: a reconnaissance study.Canadian Journal of Earth Sciences 18, 624–637.

Kitajima, K., Maruyama, S., Utsonomita, S., Liou, J.G., 2001. Seafloor hydrothermalalteration at an Archaean mid-ocean ridge. Journal of Metamorphic Geology 19,581–597.

Kloppenburg, A., White, S.H., Zegers, T.E., 2001. Structural evolution of the WarrawoonaGreenstone Belt and adjoining granitoid complexes, Pilbara Craton, Australia:implications for Archaean tectonic processes. Precambrian Research 112, 107–147.

Knauth, L.P., Lowe, D.R., 2003. High Archaean climatic temperature inferred fromoxygen isotope geochemistry of cherts in the 3.5 Ga Swaziland Supergroup, SouthAfrica. Geological Society of America Bulletin 115, 566–580.

Lowe, D.R., Byerly, G.R., 1986. Archaean flow-top alteration zones formed initially in alow-temperature sulphate-rich environment. Nature 324, 245–248.

Maiden, K.J., 1984. Archaean gold mineralization and Komatiites in Southern Africa. In:Foster, R.P. (Ed.), Gold'82: The geology, Geochemistry and Genesis of Gold Deposits.A A Balkema, Rotterdam, pp. 325–338.

Mattey, D.P., 1997. LaserPrep: An Automatic Laser-Fluorination System for Micromass'Optima' or 'Prism' Mass Spectrometers. Micromass Application Note 107, 8pp.

McCuaig, T.C., Kerrich, R., 1998. P-T-t-deformation-fluid characteristics of lode golddeposits:evidence from alteration systematics. Ore Geology Reviews 12, 381–453.

Oliver, N.H.S., Cawood, P.A., 2001. Early tectonic dewatering and brecciation on theoverturned sequence at Marble Bar, Pilbara Craton, Western Australia: dome-related or not? Precambrian Research 105, 1–15.

O'Neil, J.R., Adami, L.H., Epstein, S., 1972. Revisited value for the 18O fractionationbetween CO2 and H2O at 25 °C. J. Res. U.S. Geol. Surv. 3, 623–624.

Paris, I., Stanistreet, I.G., Hughes, M.J., 1985. Chert of the Barberton greenstone beltinterpreted asproducts of submarine exhalative activity. Journal ofGeology93,111–129.

Putlitz, B., Katzir, Y., Matthews, A., Valley, J.W., 2001. Oceanic and orogenic fluid–rockinteraction in 18O/16O-enrichedmetagabbros of an ophiolite (Tinos, Cyclades). Earthand Planetary Science Letters 193, 99–113.

Qiu, Y., Groves, D.I., 1999. Late Archean collision and delamination in the southwestYilgarn craton: the driving force for Archean orogenic lode gold mineralization.Economic Geology 94, 115–122.

Rey, P.F., Philippot, P., Thébaud, N., 2003. Contribution of mantle plumes, crustalthickening and greenstone blanketing to the 2.75–2.65 Ga global crisis. Precam-brian Research 127, 43–60.

Scherrenberg, A.F., Hein, K.A.A., White, S., 2004. Middle Archaean gold deposits in theNorth Shaw-Coongan Greenstone Belt, Australia: the relative timing of eventsresponsible for gold mineralisation. Ore Geology Reviews 25, 175–197.

Stakes, D.S., O'Neil, J.R., 1982. Mineralogy and stable isotope geochemistry of hydro-thermally altered oceanic rocks. Earth and Planetary Science Letters 57, 285–304.

Stakes, D., Taylor, H.P., 1992. The northern Semail ophiolite: an oxygen isotope,microprobe, and field study. Journal of Geophysical Research 97, 7043–7080.

Teyssier, C., Collins, W.J., Van Kranendonk, M.J., 1990. Strain and kinematics during theemplacement of the Mount Edgar Batholith and Warrawoona Syncline, PilbaraBlock, Western Australia. Third International Archaean Symposium, ExtendedAbstracts, Geoconferences (W.A.), pp. 481–483. Perth, Western Australia.

Thébaud, N., Philippot, P., Rey, P., Cauzid, J., 2006. Composition and origin of fluidsassociated with lode gold deposits in a Mesoarchean greenstone belt (WarrawoonaSyncline, Pilbara Craton, Western Australia) using synchrotron radiation X-rayfluorescence. Contributions to Mineralogy and Petrology 19, 485–503.

Van Kranendonk, M.J., Pirajno, F., 2004. Geochemistry of metabasalts and hydrothermalalteration zones associated with c. 3.45 Ga chert±barite deposits: implications forthe geological setting of the Warrawoona Group, Pilbara Craton, Australia.Geochemistry: Exploration, Environment, Analysis 4, 1–26.

Van Kranendonk, M.J., Pirajno, F., 2005. A review of hydrothermal processes andsystems on earth and implications for martian analogues. AJES 52, 329–351.

Van Kranendonk, M.J., Hickman, A.H., Smithies, R.H., Nelson, D.R., Pike, G., 2002.Geology and tectonic evolution of the Archaean North Pilbara Terrain, PilbaraCraton, Western Australia. Economic Geology 97, 695–732.

Van Kranendonk, M.J., Smithies, R.H., Hickman, A.H., Bagas, L., Williams, I.R., Farrell, T.R.,2004a. Event stratigraphy applied to 700 million years of Archaean crustalevolution, Pilbara Craton, Western Australia. Technical papers 2003–04 AnnualReview. Geological Survey of Western Australia, Perth, pp. 49–61.

Van Kranendonk,M.J., Collins,W., Hickman, A., Pawley, M.J., 2004b. Critical tests of verticalvs. horizontal tectonic models for the Archaean East Pilbara Granite-GreenstoneTerrane, Pilbara Craton, Western Australia. Precambrian Research 131, 173–211.

Weis, D., Wasserburg, G.J., 1987. Rb–Sr and Sm–Nd isotope geochemistry andchronology of cherts from the Onverwacht Group (3.5 Ga), South Africa.Geochimica et Cosmochimica Acta 51, 973–984.

Viljoen, M., 1984. Archaean gold mineralization and Komatiites in Southern Africa. In:Foster, R.P. (Ed.), Gold'82: The geology, Geochemistry and Genesis of Gold Deposits.A A Balkema, Rotterdam.

Zegers, T.E., Barley, M.E., Groves, D.I., McNaughton, N.J., White, S.H., 2002. Oldest gold:deformation and hydrothermal alteration in the Early Archean Shear Zone—HostedBamboo Creek Deposit, Pilbara, Western Australia. Economic Geology 97, 757–773.

Zheng, Y.F., 1993. Calculation of oxygen isotope fractionation in anhydrous silicateminerals. Geochemica et Cosmochimica Acta 57, 1079–1091.


Protracted fluid-rock interaction in the mid-Archaean and implication for gold mineralization: Example from the Warrawoona Syncline (WA)

Documents