PACROFI VI - Electronic Program

Fluid Inclusion Constraints on the Mineralizing Fluid from the Composite VHMS and Lode-Gold Deposits at Mt Gibson, Western Australia

Straub, K.T.1, Hagemann, S.G.1, Brown, P.E.1 and Yeats, C.J.2

  1. 1Department of Geology and Geophysics, University of Wisconsin-Madison, Madison, WI 53706
  2. Key Centre for Strategic Mineral Deposits, University of Western Australia, Perth, WA 6907

The base-metal-rich Mt Gibson gold-silver deposits are located in the southern portion of the Archean Yalgoo-Singleton greenstone belt in the Murchison Province of the Yilgarn Craton, Western Australia. The deposits are anomalous compared to other lode-gold deposits of Western Australia as they contain a highly varied sulfide assemblage, few discrete quartz veins and partially stratiform mineralization. Detailed petrographic and structural studies indicate that base-metal rich gold-silver mineralization is likely composite in nature, with early synvolcanic polymetallic volcanic hosted massive sulfide (VHMS) mineralization processes and subsequent overprinting by lode-gold style mineralization (Yeats and Groves, in press). A fluid inclusion study was undertaken in an effort to understand the physico-chemical nature of the fluids present during the two separate mineralizing events.

Geologic Setting

The "Gibson Anomaly" was defined initially as a lateritic gold deposit in 1983 by Reynolds Australian Gold and has subsequently produced 17 tonnes of gold and 10 tonnes of silver (to September, 1994) from a combination of lateritic and bedrock-hosted ore. Primary mineralization is hosted in mid-amphibolite-facies rocks composed predominately of tholeiitic metabasalts, with lesser magnesian metabasalts and felsic units. Garnet-biotite thermometry has suggested temperatures of 550 +/- 50o C for peak metamorphism (Yeats, 1996). The sequence strikes north to north-northeast, dips steeply and youngs to the east, and is intruded by a post-tectonic granitoid immediately to the east of the minesite. Mineralization is contained within the broad, approximately stratiform, north to north-northeast striking, generally steeply east dipping Mt Gibson shear zone (Yeats and Groves, in press).

Two main hydrothermal alteration events are recorded within the host sequence at Mount Gibson. The first is represented by garnet (spessartine-almandine)-gahnite bearing and corderite-muscovite schists intimately associated with gold mineralization. These assemblages are interpreted to represent the seafloor mineralization and footwall alteration associated with a VHMS-style event (Yeats and Groves, in press). The second alteration event is represented by a syn-deformational, syn-peak metamorphic, shear-hosted quartz-biotite-sulfide assemblage similar to those observed in other Western Australian amphibolite facies lode-gold deposits (Bloem et al., 1994).

Geochronological studies of the deposits also provide evidence for two discrete mineralizing events, although considerable remobilization of metals has been recognized. Pb-isotopic ratios of galena fall into two populations, with Pb/Pb model ages of 3046 +/- 30 Ma and 2772 +/- 30 Ma (Yeats and Groves, in press). In addition, these data are consistent with SHRIMP U/Pb zircon geochronology (2930 Ma and 2627 +/- 13 Ma, Yeats et al., in press). These are within error of data from other Murchison VHMS and lode-gold deposits, respectively (McNaughton et al., 1990).

The Mt Gibson deposits exhibit two main styles of alteration and sulfide mineralization, consistent with both synvolcanic and syntectonic hydrothermal activity. These unique geologic constraints provide both an unusual host rock for the second lode-gold mineralizing event as well as an opportunity to examine the extent of VHMS fluid inclusion preservation. Results of this fluid inclusion study suggest the lode-gold mineralizing event was pervasive throughout the Mt Gibson shear zone. Remnants of the VHMS alteration event have been either mixed with lode-gold fluids, are indistinguishable from lode-gold fluid inclusions in this geologic setting, or, have been destroyed by amphibolite-facies metamorphism. The most likely interpretation based on this study allows for obliteration of VHMS fluid inclusions with ensuing magmatic and metamorphic dominated fluids supplying the second lode-gold mineralizing event.

Fluid inclusion study

Microthermometry of fluid inclusions at Mt Gibson reveals five predominate types of inclusions: (I)CO2, (II)CH4, (III) H2O-salt +/- gas, (IV) CO2-CH4 and, (V) CO2-H2O-salt +/- gas (mixed). All fluid inclusions studied have been found in quartz; petrography of sphalerite, garnet and other transparent minerals (including zircon and feldspar) have revealed no preserved inclusions. Quartz textures are often granoblastic or exhibit oscillating extinction. Samples are categorized on the basis of base-metal content (primarily Zn) for purposes of comparing mineralizing events. Table 1 summarizes selected fluid inclusion data and corresponding mineralizing fluid events.

Sample TypeHost rocksMineralizationAu (ppm)Zn content (ppm)Inclusion Types (Compositional)Qt TexturesFluid inclusion textures
VHMSCMS, Sericite and Gt-bearing schistsPolymetallic sulfides (PY, PO, CPY, SPH, GA, ASP, BI)0.52-46.2>100I, III, IV, VGranoblastic Qt and qt with ocsillating extinctionClustered, internal trails, growth zone. Most measured inclusions are in pressure shadow of sulfides.
Lode-AuQT-BI schistsPY, CPY1.2-?<100I, III, IV, VAnhedral QT, sutured boundariesClustered, internal trails, growth zone. Most measured inclusions are in pressure shadow of sulfides.
Non-ore bearingFelsic volcanics porphyriesNone<0.05<100I, II, III, IV, VSub-hedral "quartz eyes" and granoblastic Qt.Clusters, internal trails. Most measured inclusions are in pressure shadows of sulfides.
Regional Granitoid NoneNA<100IIIGraniticClusters, internal trails
Late QT veining NoneNA<100late aqueousLate Qt vein crosscutting mineralization (subhedral grains).Internal and crosscutting trails.

Primary inclusions, as determined following criteria in Roedder (1984), of types I, III, IV and V occur in low-, high- and non-base-metal-bearing samples. Type II inclusions have only been observed in non-ore mineral bearing samples. Type I inclusions range in CO2 density from 1.08 to 0.42 g/cm3 (ave.= 0.67, n= 36). In comparison, type IV inclusions have densities from 1.13 to 0.53 g/cm3 (ave.= 0.74, n= 55) while type II inclusions average 0.29 g/cm3 (stdev= 0.03 g/cm3, n=25). Type IV inclusions appear to cluster in two sub-populations, type IVa: CO2-CH4 inclusions with <10% CH4 and type IVb: CH4-rich inclusions containing <10% CO2 (as determined by homogenization between -70.0 and -82.6oC). Figure 1 shows a histogram of TmCO2 within types I, IVa and IVb inclusions. Fluid inclusions lacking a measurable final melting of a carbonic solid are indicated at temperatures < -74.0oC. Although phase transitions were not quantifiable for these inclusions upon heating, observation under 80X objectives during cooling reveals freezing of a carbonic phase between -95 to -105oC. Type III inclusions most often contain at least one daughter mineral (some inclusions show five distinct solid phases at room temperature). The salinity of these inclusions show ranges with three distinct groups at <10 eq. wt % NaCl (ave.= 7.2, n= 14), 10-20 eq. wt % NaCl (ave.= 17.0, n= 11), and >30 eq. wt. % NaCl (ave.= 47.1, n= 14). Eutectic temperatures are also variable, ranging from -81 to -15oC. Three populations of H2O-salt inclusions are discernible from the freezing data: inclusions whose Tmice < -30.0oC, inclusions with Tmice ranging from -25.0 to 0.0oC, and inclusions with final melting temperatures at +10oC to +30.0oC. The mixed, Type V inclusions, are fairly rare containing relatively pure CO2 (ave. XCH4 <0.1 within the carbonic phase and bulk XCO2 from 0.08 to 0.14). The mixed type V inclusions also contain a very saline aqueous component with eq. wt % NaCl from 10.9 to 42.3 (ave. eq. wt % NaCl= 22.3, n= 12). The eq. CO2 density of the carbonic phase is calculated from 0.92 to 0.64 (ave.= 0.55, n= 12). Final homogenization and decrepitation temperatures for types III and V range from 213 to 600oC (stdev= 143 oC). Decrepitation of types III, IV and V also occurred within this range, often prior to final homogenization (for types III and V only). Inclusions containing daughter minerals most often experience final homogenization with daughter mineral dissolution and many inclusions containing larger daughter minerals failed to homogenize at temperatures lower than 600oC. Although the standard deviation is quite large for heating data from type III inclusions (Thl ave.= 360oC, stdev= 129oC), it is possible there are several sub-populations of H2O-salt inclusions (including texturally secondary fluid inclusions) that are at this time quantitatively indistinguishable.

Discussion and Interpretation

A comparison of fluid inclusion data from low and high base-metal samples shows little to no distinction amongst physico-chemical properties (Table 1). Figure 2 displays isochore ranges for all five inclusion types. Densities are moderate for the base-metal-rich fluid inclusion samples (0.6 to 0.9 g/cm3 for type I and IV inclusions, for example). Isochores for these densities suggest pressures from >500 to 2500 bars at temperatures of 150 to 350o C, typical temperatures of present day black smokers. Higher temperatures, established by geothermometry, of 500 to 600oC for peak metamorphism and the syn-deformational lode gold mineralizing event, give pressures of 2000 to 3000 bars for the carbonic inclusions. The calculated pressure estimates are unusually high for a VHMS environment, but some of the isochores are compatible with pressures suggestive of amphibolite-facies metamorphism. In addition, the inclusions analyzed do not conform to expected fluid inclusion compositions in VHMS systems (de Ronde, 1995; Roedder, 1984; Pisutha-Arnond & Ohmoto, 1983).The ranges of eutectic temperatures indicate cations in addition to Na and K are present in the aqueous inclusions, these most likely include complexes of Mg, Fe and Ca. This evidence and the presence of high salinity and diverse salt complexes in fluid inclusions from base-metal-rich samples seems to argue for a pervasive second mineralizing event. The primary fluid inclusions in the most base-metal-rich samples appear to be more representative of fluid compositions and P-T-X conditions found in deep (amphibolite and granulite) lode-gold deposits. It is considered likely that fluid inclusions from an early VHMS hydrothermal system were destroyed during a second high temperature and pressure mineralizing event. Further support is evident by recrystallized quartz textures present in all samples. The extremely high salinities are suggestive of a magmatic fluid component. Boiling pairs (as defined by Ramboz et al., 1982) containing high saline aqueous inclusions juxtaposed to vapor-rich fluid inclusions have been observed and measured; this evidence is possibly indicative of phase immiscibility as a mineralizing process during the lode-gold event.