David I. Groves

Orogenic gold and geologic time: A global synthesis

AbstractOrogenic gold deposits have formed over more than 3 billion years of Earth’s history, episodically during the MiddleArchean to younger Precambrian, and continuously throughout the Phanerozoic. This class of gold deposit is characteristi-cally associated with deformed and metamorphosed mid-crustal blocks, particularly in spatial association with major crustalstructures. A consistent spatial and temporal association with granitoids of a variety of compositions indicates that melts andfluids were both inherent products of thermal events during orogenesis. Including placer accumulations, which arecommonly intimately associated with this mineral deposit type, recognized production and resources from economicPhanerozoic orogenic-gold deposits are estimated at just over one billion ounces gold. Exclusive of the still-controversialWitwatersrand ores, known Precambrian gold concentrations are about half this amount.The recent increased applicability of global paleo-reconstructions, coupled with improved geochronology from most ofthe world’s major gold camps, allows for an improved understanding of the distribution pattern of orogenic gold in spaceand time. There are few well-preserved blocks of Middle Archean mid-crustal rocks with gold-favorable, high-strain shearzones in generally low-strain belts. The exception is the Kaapvaal craton where a number of orogenic gold deposits arescattered through the Barberton greenstone belt. A few )3.0 Ga crustal fragments also contain smaller gold systems in theUkrainian shield and the Pilbara craton. If the placer model is correct for the Witwatersrand goldfields, then it is possiblethat an exceptional Middle Archean orogenic-gold lode-system existed in the Kaapvaal craton at one time. The latter half ofthe Late Archean ca. 2.8–2.55 Ga was an extremely favorable period for orogenic gold-vein formation, and resulting oresŽ.preserved in mid-crustal rocks contain a high percentage of the world’s gold resource. Preserved major goldfields occur ingreenstone belts of the Yilgarn craton e.g., Kalgoorlie , Superior province e.g., Timmins , Dharwar craton e.g., Kolar ,Ž. Ž . Ž.Zimbabwe craton e.g., Kwekwe , Slave craton e.g., Yellowknife , Sao Francisco craton e.g., Quadrilatero Ferrifero , andŽ. Ž . Ž .Tanzania craton e.g., Bulyanhulu , with smaller deposits exposed in the Wyoming craton and Fennoscandian shield. SomeŽ.workers also suggest that the Witwatersrand ores were formed from hydrothermal fluids in this period.The third global episode of orogenic gold-vein formation occurred at ca. 2.1–1.8 Ga, as supracrustal sedimentary rocksequences became as significant hosts as greenstones for the gold ores. Greenstone–sedimentary rock sequences nowexposed in interior Australia, northwestern Africarnorthern South America, Svecofennia, and the Canadian shield were the

First evidence of >3.2 Ga continental crust in the Yangtze craton of south China and its implications for Archean crustal evolution and Phanerozoic tectonics

Ion microprobe (SHRIMP II) U-Pb zircon analyses reveal trondhjemitic magmatism at 2.90–2.95 Ga in the Kongling area of the Yangtze craton, south China, about 150 km south of the Permian-Triassic Qinling-Dabie-Sulu orogenic belt. Detrital zircons from nearby Archean metapelites are 2.87–3.28 Ga, and the rocks have Sm-Nd depleted mantle model ages of 3.07–3.21 Ga. The new data reveal, for the first time, >3.2 Ga sialic crust in the Yangtze craton, part of which predates that of the adjacent southern North China craton. Both trondhjemites and metapelites contain ca. 2.75 Ga high-grade metamorphic zircons, and ca. 1.9 Ga zircons, related to intrusion of the Quanqitang K-feldspar granite into the Archean basement. Many zircons also underwent Pb loss ca. 1.0 Ga during the Jinning orogeny, when the Cathysian block accreted to the Yangtze craton. The new data support correlation of part of the Korean Peninsula with the Yangtze craton along the eastern extension of the Qinling-Dabie-Sulu orogenic belt.

The crustal continuum model for late-Archaean lode-gold deposits of the Yilgarn Block, Western Australia

Most Archaean gold ores belong to a coherent genetic group of structurally controlled lode-deposits that are characteristically enriched in Au with variable enrichments in Ag, As, W, Sb, Bi, Te, B and Pb, but rarely Cu or Zn, and are surrounded by wallrock alteration haloes enriched in K, LILE and CO2, with variable Na and/or Ca addition. Evidence from the Yilgarn Block of Western Australia, combined with similar evidence from Canada and elsewhere, indicates that such deposits represent a crustal continuum that formed under a variety of crustal régimes over at least a 15 km crustal profile at PT conditions ranging from 180°C at <1 kb to 700°C at 5 kb. Individual deposits, separated by tens to hundreds of kilometres, collectively show transitional variations in structural style of mineralisation, vein textures, and mineralogy of wallrock alteration that relate to the PT conditions of their formation at varying crustal depths. Specific transitions within the total spectrum may be shown also by deposits within gold camps, although nowhere is the entire continuum of deposits recorded from a single gold camp or even greenstone belt.Recognition of the crustal continuum of deposits implicates the existence of giant late-Archaean hydrothermal systems with a deep source for the primary ore fluid. A number of deep fluid and solute reservoirs are possible, including the basal segments of greenstone belts, deep-level intrusive granitoids, mid-to lower-crustal granitoidgneisses, mantle lithosphere, or even subducted oceanic lithosphere, given the probable convergent-margin setting of the host greenstone terranes. Individual stable and radiogenic isotope ratios of fluid and solute components implicate fluid evolution from, or equilibrium with, a number of these reservoirs, stressing the potential complexity of pathways for fluid advection to depositional sites. Lead and strontium isotope ratios of ore-associated minerals provide the most persuasive evidence for fluid advection through deep-level intrusive granitoids or granitoid-gneiss crust, whereas preliminary oxygen isotope data show that mixing of deeply sourced fluid and surface waters only occurred at the highest crustal levels recorded by the lode gold deposits.

Constraints on crustal evolution and gold metallogeny in the Northwestern Jiaodong Peninsula, China, from SHRIMP U–Pb zircon studies of granitoids

Abstract The northwestern part of the Jiaodong Peninsula, or the Zhao-Ye Gold Belt, contains the largest lode-gold deposits in China, which are spatially related to several suites of intrusive granitoids. Previous attempts to provide constraints on the timing of gold mineralization and its tectonic setting, through studies of the granitoids, have led to conflicting data, dependant on the isotopic methodology used, and resultant genetic and tectonic models for their setting are equivocal. SHRIMP U–Pb studies of complexly zoned zircons of the Linglong, Luanjiahe and Guojialing granitoid suites suggest that the Jiaodong Peninsula is underlain by Precambrian basement with components up to 3.4 Ga old. Inherited zircons of early Mesozoic age indicate that this basement was reworked at 250–200 Ma, probably during a collisional orogeny involving the North China Craton and South China Craton. The Linglong, Luanjiahe and Guojialing suites were derived from this early Mesozoic basement between 165 and 125 Ma, and were probably emplaced as post-collisional granitoids, the latest intrusions (at 125 Ma) coinciding broadly with superplume activity or a major plume breakout event in the Palaeo-Pacific Plate. Importantly, if lode-gold mineralization is essentially a single event, as indicated by similar hosting structures and deposit characteristics, it can be dated between about 126 Ma, the age of the youngest granitoid cut by gold-bearing quartz veins, and 120 Ma, the age of one of a swarm of post-mineralization feldspar-porphyry dykes. Thus, as in most other metallogenic provinces which host so-called mesothermal lode-gold deposits, gold mineralization was late in the orogenic cycle, probably late- or post-accretion, and closely followed the emplacement of the latest major plutonic phase in the development of a series of anomalously voluminous granitoid batholiths.

The Sholl Shear Zone, West Pilbara: evidence for a domain boundary structure from integrated tectonostratigraphic analyses, SHRIMP UPb dating and isotopic and geochemical data of granitoids

Abstract The Pilbara Block provides a record of Archaean continental growth involving the tectonic accretion of outboard island-arcs and collisions with other continental-scale fragments. This record of continental growth is balanced by breakup and strike-slip dismemberment of the continent. New SHRIMP UPb in zircon ages and SmNd data provide evidence in the West Pilbara which demonstrates that subduction-related and tectonic-accretion processes at the western margin of that ancestral continent between 3.15-2.78 Ga were coeval with, and genetically related to, crustal-scale tectonics and basin formation inboard of that margin. The tectonic division of the West Pilbara is defined by integrated tectonic analyses, geochronology, geochemistry and isotopic analyses. Geochronological studies clearly indicate that the western Pilbara comparises two domains with different recorded geohistories, whereas geochemistry and isotopic systematics reflect the changing tectonic regimes through time. In combination, these studies allow the development of a reconstruction of the relative positions of the domains through time on the western margin of the Pilbara Block. The supracrustal rocks of the northern Roebourne Lithotectonic Complex (Domain 6 in a Pilbarawide scheme) were formed in an island arc setting, facing an ocean to the north-west, prior to 3260 Ma, the time of emplacement of voluminous granitoids into the complex. In contrast, the supracrustal rocks of the southern Sholl Belt (Pilbara Domain 5) were formed in a back-arc setting behind a north-west-facing arc between 3125 and 3112 Ma, with more-or-less synchronous granite emplacement at about 3115 Ma. The two domains were tectonically juxtaposed, between 2991 and 2925 Ma, by the Sholl Shear Zone, a largely sinistral shear zone, with subsequent volcanism in both domains to about 2925 Ma. The Roebourne Lithotectonic Complex (Domain 6) is interpreted to be an allochthonous terrane, which formed north-east relative to its present position, but indigenous to the Pilbara Block rather than an exotic terrane. The East Pilbara is interpreted to have acted as a cratonic hinterland during the convergent margin tectonics that affected the two West Pilbara domains.

Late-kinematic timing of orogenic gold deposits and significance for computer-based exploration techniques with emphasis on the Yilgarn Block, Western Australia

Abstract Orogenic gold deposits are a widespread coherent group of epigenetic ore deposits that are sited in accretionary or collisional orogens. They formed over a large crustal-depth range from deep-seated low-salinity H2O–CO2±CH4±N2 ore fluids and with Au transported as thio-complexes. Regional structures provide the main control on deposit distribution. In many terranes, first-order faults or shear zones appear to have controlled regional fluid flow, with greatest ore-fluid fluxes in, and adjacent to, lower-order faults, shear zones and/or large folds. Highly competent and/or chemically reactive rocks are the most common hosts to the larger deposits. Focusing of supralithostatic ore fluids into dilatant zones appears to occur late during the evolutionary history of the host terranes, normally within D3 or D4 in a D1–D4 deformation sequence. Reactivation of suitably oriented pre-existing structures during a change in far-field stress orientation is a factor common to many deposits, and repeated reactivation may account for multiple mineralization episodes in some larger deposits. Absolute robust ages of mineralization support their late-kinematic timing, and, in general, suggest that deposits formed diachronously towards the end of the 100 to 200 m.y. long evolutionary history of hosting orogens. For example, in the Yilgarn Block, a region specifically emphasised in this study, orogenic gold deposits formed in the time interval between 40 and 90 m.y., with most about 60 to 70 m.y., after the youngest widespread basic-ultrabasic volcanism and towards the end of felsic magmatism. The late timing of orogenic gold deposits is pivotal to geologically-based exploration methodologies. This is because the present structural geometries of: (i) the deposits, (ii) the hosting goldfields, and (iii) the enclosing terranes are all essentially similar to those during gold mineralization, at least in their relative position to each other. Thus, interpretation of geological maps and cross-sections and three-dimensional models can be used to accurately simulate the physical conditions that existed at the time of ore deposition. It is particularly significant that the deposits are commonly related to repetitive and predictable geometries, such as structural heterogeneities within or adjacent to first-order structures, around rigid granitoid bodies, or in specific “locked-up” fold-thrust structures. Importantly, the two giant greenstone-hosted goldfields, Kalgoorlie and Timmins, show a remarkably similar geometry at the regional scale. Computer-based stress mapping and GIS-based prospectivity mapping are two computer-based quantitative methodologies that can utilize and take advantage of the late timing aspect of this deposit type to provide important geological aids in exploration, both in broad regions and more localized goldfields. Both require an accurate and consistent solid geology map, stress mapping requires knowledge of the far-field stresses during mineralization, and the empirical prospectivity mapping requires data from a significant number of known deposits in the terrane. The Kalgoorlie Terrane, in the Yilgarn Block, meets these criteria, and illustrates the potential of these methodologies in the exploration for orogenic gold deposits. Low minimum stress anomalies, interpreted to represent dilational zones during gold-related deformation, coincide well with the positions of known goldfields rather than individual gold deposits in the terrane, and there are additional as-yet unexplained anomalies. The prospectivity analysis confirms that predictable and repetitive factors controlling the siting of deposits are: (i) proximity to, and orientation and curvature of, granitoid-greenstone contacts, (ii) proximity to segments of crustal faults which strike in a preferred direction, (iii) proximity to specific lithological contacts which have similar preferred strike, (iv) proximity to anticlinal structures, and (v) the presence of preferred reactive host rocks (e.g., dolerite). The prospectivity map defines a series of anomalous areas, which broadly conform to those of the stress map (>78% correspondence). The most prospective category on this map covers less than 0.3% of the greenstone belts and yet hosts 16% of the known deposits, which have produced>80% of known gold. Thus, it discriminates in favour of the larger economically more-attractive deposits in the terrane. The successful application of stress mapping and prospectivity mapping to geology-based exploration for orogenic gold deposits indicates that more quantitative analysis of geological map data is a profitable line of research. The computer-based nature of these methodologies is ideal for the production of an ultimate, integrated, deposit target map, which can be compared to other, more conventional, targeting parameters such as geophysical and geochemical anomalies. Such an integrated strategy appears the way forward in the increasingly difficult task of cost-effective global exploration for orogenic gold deposits in poorly exposed terranes.

The nature of Archaean gold‐bearing fluids as deduced from gold deposits of Western Australia

Abstract Archaean gold deposits of Australia are most abundant in 2.8 ± 0.1 Ga granitoid‐greenstone terrains, where they form both vein‐type and disseminated occurrences. Two‐thirds of total gold production has come from metamorphic (early, peak or late) vein deposits within host rocks of metabasaltic composition and of greenschist to lower amphibolite facies grade. Wallrock alteration, isotopic readjustment and fluid inclusion data help to characterise the nature and importance of hydrothermal fluids in the generation of the vein‐type deposits. Although fluid access is controlled by major structures and more locally by permeable zones (e.g. hydraulically‐fractured country‐rocks), gold deposition is a result of fluid‐wallrock interaction within a suitable temperature range. Mineralising temperatures at least as high as 400°C (e.g. Mt Charlotte, Kalgoorlie) and slightly alkaline, reducing, H2O‐CO2‐rich fluids of low salinity, derived from lower in the sequence, are inferred. In this type of solution, gold ...

Supercontinent cycles and the distribution of metal deposits through time

Systematic temporal variations in the distribution of several important groups of metal deposits reflect the cyclic aggregation and breakup of large continents. In particular, metal deposits that form in continental basins or are associated with anorogenic magmatism were extraordinarily abundant in the Middle Proterozoic (2.0 to 1.4 Ga), corresponding to the assembly of the first large continents. It is important to note that peaks in the abundance of continental metal deposits also coincide with a postulated Late Proterozoic supercontinent (1.0 to 0.8 Ga) and the near maximum extent of Pangea. In contrast, metal deposits that form, or are preserved, in convergent-margin orogens were most abundant in the late Archean (2.9 to 2.6 Ga), corresponding to a period of high global heat flow and rapid stabilization of continental crust, and the past 200 m.y., which corresponds to the present tectonic cycle. Similar mineralization was also present, albeit less abundant, in Early Proterozoic orogens, as well as in Late Proterozoic and Phanerozoic orogens. Future metals exploration may benefit from the application of sequence stratigraphy, as used by the oil industry, to recognize such cycles, particularly in the Precambrian rock record.

Artificial neural networks: a new method for mineral-prospectivity mapping

A multilayer feed‐forward neural network, trained with a gradient descent, back‐propagation algorithm, is used to estimate the favourability for gold deposits using a raster GIS database for the Tenterfield 1:100 000 sheet area, New South Wales. The database consists of solid geology, regional faults, airborne magnetic and gamma‐ray survey data (U, Th, K and total count channels), and 63 deposit and occurrence locations. Input to the neural network consists of feature vectors formed by combining the values from co‐registered grid cells in each GIS thematic layer. The network was trained using binary target values to indicate the presence or absence of deposits. Although the neural network was trained as a binary classifier, output values for the trained network are in the range [0.1, 0.9] and are interpreted to indicate the degree of similarity of each input vector to a composite of all the deposit vectors used in training. These values are rescaled to produce a multiclass prospectivity map. To validate and assess the effectiveness of the neural‐network method, mineral‐prospectivity maps are also prepared using the empirical weights of evidence and the conceptual fuzzy‐logic methods. The neural‐network method produces a geologically plausible mineral‐prospectivity map similar, but superior, to the fuzzy logic and weights of evidence maps. The results of this study indicate that the use of neural networks for the integration of large multisource datasets used in regional mineral exploration, and for prediction of mineral prospectivity, offers several advantages over existing methods. These include the ability of neural networks to: (i) respond to critical combinations of parameters rather than increase the estimated prospectivity in response to each individual favourable parameter; (ii) combine datasets without the loss of information inherent in existing methods; and (iii) produce results that are relatively unaffected by redundant data, spurious data and data containing multiple populations. Statistical measures of map quality indicate that the neural‐network method performs as well as, or better than, existing methods while using approximately one‐third less data than the weights of evidence method.

Geodynamic settings of mineral deposit systems

Mineral deposits represent extraordinary metal concentrations that form by magmatic, magmatic–hydrothermal or hydrothermal processes in geodynamic environments typified by anomalously high thermal and/or mechanical energy near plate boundaries. As they require the conjunction of specific environmental conditions to form, particular mineral deposit types tend to occupy specific geodynamic niches. The temporal distributions of mineral deposit types reflect both formational and preservational processes. In the Archaean and Palaeoproterozoic, these were linked because of preservation in continental crust connected to thick buoyant subcontinental lithospheric mantle (SCLM), but were decoupled by the Neoproterozoic and Phanerozoic as a result of evolution to thinner, increasingly dense SCLM. The transition marks a change from mantle plume-influenced plate tectonics to modern-style plate tectonics, with broadly coincident environmental changes and a major impact on the nature and abundance of preserved mineral deposit types. As mineral deposits represent an integral part of tectonic process, they are essential indicators of that process and geodynamic settings, and should be incorporated into any holistic tectonic terrane analysis. Their distribution also provides a particularly critical test on ancient continental reconstructions derived from palaeomagnetic data. Conversely, such reconstructions provide a first-order targeting tool for the conceptual exploration required to discover new mineral provinces and deposits under cover.