Philip E. Brown and Steffen G. Hagemann

The University of Wisconsin-Madison, Department of Geology and Geophysics, Madison, WI 53706, USA

The depth of mineralization for Archean lode-gold deposits from the Yilgarn Craton in Australia and the Superior Province in Canada has been estimated from published data and new fluid inclusion measurements (Fig. 1). Lithostatic pressure conditions were assumed for meso- and katazonal deposits, whereas a mixed litho-hydrostatic pressure gradient was assigned to epizonal deposits. Geological evidence (brittle structural style) and isotopic and fluid inclusion evidence for incursion of surface waters (Hagemann et al., 1994) supports the possibility of a transient hydrostatic pressure gradient during mineralization in the epizonal deposits. For deeper deposits (> 5 km) lithostatic fluid pressures are assumed based on the lack of evidence for surface water influx and hydrodynamic considerations such as the total load of rocks overlying the deposits.

The calculated P-depth relations shown in Figure 1A display a continuum of pressure and depths for the formation of lode-gold deposits in Western Australia. These range from shallow, epizonal levels (<1 to 1.5 kbar at <6 km depth e.g. Wiluna), to mesozonal (1.5 to 4 kbar at 6 to 12 km depth e.g. Mt Charlotte), to deep, katazonal deposits (>4 kbar at >12 km depth e.g. Griffins Find). Available geobarometric data for lode-gold deposits from the Superior Province greenstone belts in Canada, in contrast, display a distinct clustering of pressure and depths of formation at the mesozonal level (1.5 to 3.0 kbar at >6 km to <12 km depth).

Pressure-Depth (21k)

Fig.1 Pressure-depth conditions of Archean epi- meso- and katazonal lode-gold deposits from the Western Australian Yilgarn Craton (A) and Superior Province in Canada (B). Depths for deposits were estimated using ranges of pressures from isochores. Lithostatic and hydrostatic lines assume pressure gradients of 33 and 100 m/MPa. Note that for the epizonal gold deposits, pressures may vary between litho- and hydrostatic values; for meso- and katazonal deposits constant lithostatic pressures are assumed. The ranges given indicate the uncertainties of estimation. The gray bands indicate the assumed transition zone between the different crustal levels.

This study confirms earlier fluid inclusion investigations (Hagemann and Ridley, 1993) which presented P-d data that constrained the hypothesis that Archean lode-gold deposits in Western Australia formed over a remarkable range of crustal depths (Groves et al., 1991). In fact it appears that, at least in Western Australia, these deposits are hosted from epi- to katazonal depth levels and therefore contrast to other ore deposit classes, such as VMS and MVT deposits, that form only at shallow crustal levels near the seafloor, or within the top 5 km of the crust, respectively.

When comparing the depth ranges of Western Australian lode-gold deposits with deposits from the Superior Province it is obvious that the latter deposits cluster at mid-crustal levels between 4 and 12 km, except Hemlo which formed at katazonal levels of about 10 to 14 km depth. Interestingly neither the top (i.e. epizonal depths) nor the bottom (katazonal) of the crustal section appears to be represented. One reason for the apparent disparity between the two Archean terrains might be the lack of detailed fluid inclusion studies, hence lack of geobarometric data, conducted in deep and shallow level deposits. Indeed, several well known deposits such as East Main River and Red Lake lack reliable geobarometric data. These deposits are described as being situated in amphibolite facies host rocks with gold mineralization broadly syn- to peak-metamorphic. Consequently they could be classified as katazonal deposits. Conversely, geological constraints, such as structural setting and low temperature ore minerals, argue for a shallow crustal level setting (epizonal) for the Ross Mine in Ontario. However, even though there are possible examples of epi- and katazonal gold deposits in the Superior province, there are still significantly more deposits situated at mesozonal levels.

The apparent lack of giant lode-gold deposits at epi- and katazonal depth levels could be related to a variety of factors, for instance the: (1) lack of exploration in higher and lower grade metamorphic terrains due to specific exploration (genetic) models, (2) lack of exposure of lower and higher crustal rocks due to possible erosion and burial of these crustal sections, respectively, and (3) less favorable transport and precipitation mechanisms for gold at deep crustal levels.


Groves D.I., Barley M.E., Cassidy K.F., Hagemann S.G., Ho S.E., Hronsky J.M.A., Mikucki E.J., Mueller A.G., Mcnaughton N.J., Perring C.S., and Ridley J.R. 1991 Archaean lode-gold deposits: The products of crustal-scale hydrothermal systems. In Proceedings of Brazil Gold'91, An International Symposium on the Geology of Gold, (ed. E. A. LADEIRA) Belo Horizonte (1991) Balkeema, Rotterdam, 299-306.

Hagemann, S.G. and Ridley, J.R. 1993. Hydrothermal fluids in epi- and katazonal crustal levels in the Archaean; implications for P-T-X-t evolution of lode gold mineralisation. Australian Geological Survey Organization Record 1993/54, 123-130.

Hagemann, S.G., Brown, P.E., Groves, D.I., Ridley, J.R. and Valley, J.V. 1994 The Wiluna lode-gold deposits, Western Australia: Surface water influx in a shallow level Archean lode-gold system (abstr.). 12 Austr. Geol. Congress, Perth, 160-161.