Ga Greenstone Belts Research Articles

Late Neoarchean magmatic – metamorphic event and crustal stabilization in the North China Craton

The ca. 2.5 Ga as the time boundary between the Archean and the Proterozoic eons is a landmark, indicating the most important continental crust evolving stage of the Earth, that is, the global cratonization or the formation of supercraton(s) that was unseen before and is unrepeated in the following history of the Earth9s formation and evolution. The North China Craton (NCC) is one of the best recorders of the ca. 2.5 Ga event, and therefore studies in the thorough understanding of early Precambrian continental evolution are continuous. The period from 2.8 to 2.6 Ga is the major crustal growth period of the NCC and formed seven micro-blocks. All the micro-blocks in the NCC were surrounded by 2.6 to 2.54 Ga greenstone belts. The clear geological presentations are as follows: (1) Archaic basement rocks in North China (various micro-blocks) experienced strong partial melting and migmatization. The granitoid rocks derived from crustal partial melting include potassium, TTG and monzonitic granitoids, which come, respectively, from continental crust (sedimentary rocks with TTG gneisses), juvenile crust (mafic rocks with TTG gneisses) or mixed crust; (2) the BIF-bearing supracrustal rocks are mainly distribute in greenstone belts. The lithologic associations in the greenstone belts within the NCC are broadly similar, belonging to volcano-sedimentary sequences, with common bimodal volcanic rocks (basalt and dacite) interlayered with minor amounts of komatiites in the lower part, and calc-alkalic volcanic rocks (basalt, andesite and felsic rocks) in the upper part; (3) nearly all old rocks of >2.5 Ga underwent ∼2.52 to 2.5 Ga metamorphism of amphibolite–granulite facies. Most metamorphosed rocks show high-temperature-ultra-high-temperature (HT–UHT) characteristics and record anticlockwise P–T paths, albeit a small number of granulites seemingly underwent high-pressure granulite facies metamorphism and record clockwise P–T paths; (4) ∼2.5 Ga mafic dikes (amphibolites), granitic dikes (veins) and syenitic–ultramafic dikes developed across these archaic basements and were strongly deformed or un-deformed; (5) the extensive 2.52 to 2.48 Ga low-grade metamorphic supracrustal covers has been recognized in eastern, northern and central parts of the NCC, which are commonly composed of bi-modal volcanic rocks and sedimentary rocks. The above mentioned ∼2.5 Ga geological rocks and their characters imply that the seven micro-blocks have been united through amalgamation to form the NCC. The metamorphosed late Neoarchean greenstone belts, as syn-formed mobile belts, welded the micro-blocks at the end of the Neoarchean. However, the metamorphic thermal grades of the greenstone belts are lower than those of the high-grade terranes within the micro-blocks, suggesting that the latter might have developed under a higher geothermal gradient than the former. Besides, the greenstone belts surround the various micro-blocks in the late Neoarchean when both the old continental crust and the oceanic crust were hotter. The subduction during the amalgamation, if it happened, must have been much smaller in scale as compared to those in the Phanerozoic plate tectonic regime, and all stages occurred at crust-scale instead of lithosphere-scale or mantle-scale. This is why most rocks record HT-UHT and anti-clockwise metamorphism, while only a few samples record high-pressure granulite facies metamorphism with clockwise P–T paths. The micro-block amalgamation was accompanied by extensive crust partial melting and granitization, which finally gave rise to the stabilization of the NCC. Except for the vast granitoid intrusions, mafic-syenitic dike swarms and sedimentary covers are also landmarks of cratonization. The ca. 2.5 Ga cratonization is a global epoch-making geological event, although the accomplishment of cratonization in various cratons is somewhat different in time. Cratonization declared the formation and stabilization of global-scale supercratons or cratonic groups coupling with lithosphere, which was followed by a “silent period” with rare tectonic-thermal action lasting 150 to 200 Ma (from 2.5 Ga – 2.3 or 2.35 Ga), and then followed by the Great Oxidation Event (GOE).

Fluid inclusion and stable isotope (O, H, C) constraints on the genesis of the Pedra Branca gold deposit, Troia Massif, Borborema Province, NE Brazil: An example of hypozonal orogenic gold mineralization

At the Archean-Paleoproterozoic Troia Massif, NE Brazil, two major Paleoproterozoic greenstone belts are documented (Algodões and Serra das Pipocas). They share similar lithostratigraphic characteristics with other 2.2–2.1 Ga greenstone belts of the surrounding cratonic domains (e.g. Guiana shield and West Africa craton). Gold mineralization in the Serra das Pipocas greenstone belt is associated with a regional NE-trending shear zone. In the mineralized area (the Pedra Branca gold deposit) the main gold mineralization is found in association with quartz veins, calc–silicate alteration (e.g. diopside, K-feldspar, amphibole, titanite, biotite, pyrite, albite, magnetite ± carbonates) and albitization. Pyrrhotite is commonly found as inclusions in pyrite, and late magnetite replaces pyrite, suggesting progressive oxidation of the ore–forming fluid. Free-milling gold commonly precipitates at the late stages of alteration, in close association with magnetite and tellurides. The main tellurides found are hessite, altaite, sylvanite, calaverite and tellurobismuthite. Two fluid inclusion assemblages were identified from mineralized quartz veins. Assemblage 1 is characterized by pseudo-secondary trails that show the coexistence of CO2–rich and low salinity (0–8 wt% NaCl equiv.) CO2–H2O–NaCl and H2O–NaCl inclusions, suggesting formation during phase separation (fluid immiscibility). Assemblage 2 is represented by late secondary low–temperature (Th < 200 °C) H2O–NaCl inclusions, probably unrelated to gold mineralization. The δ18O, δD and δ13C values of hydrothermal minerals (quartz, calcite, biotite, hornblende and magnetite) and fluid inclusions are compatible with a magmatic–hydrothermal ore-forming fluid that underwent fluid–rock interaction with the greenstone sequence. The calculated fluid δ18O values range from +8.3 to +11.0‰ (n = 59), fluid δD from −98 to –32‰ (n = 24) and δ13C values of calcite from −6.35 to −9.40‰ (n = 3). Oxygen isotope thermometry for quartz–magnetite pairs yields temperatures from 467 to 526 °C (n = 7, average 503 °C). This probably represents the temperature of gold deposition, because gold is generally associated with magnetite in the ore mineral assemblage. The estimated PT conditions of the gold mineralization range from 467 °C and 1.93 kbar (7 km) to 526 °C and 3.48 kbar (13 km), repectively, and they were obtained from the intersection of isochores with the temperature range derived from the quartz–magnetite isotope geothermometer. This PT range is akin to hypozonal orogenic gold deposits. The association of gold with magnetite and tellurides implies an ore–forming fluid derived from oxidized magmas, similar to those interpreted as ‘orogenic oxidized intrusion-related gold deposits’ in other Precambrian greenstone belts (e.g. Abitibi and Eastern Goldfields).

Spinifex-textured komatiites in the south border of the Carajas ridge, Selva Greenstone belt, Carajás Province, Brazil

Spinifex-textured komatiites in the Selva greenstone belt are the first unequivocal examples of komatiites in the Transition Subdomain of the Carajás Mineral Province. Outcrops of spinifex-textured komatiites, located ∼1.5 km to the south of the Carajás ridge, were discovered during regional exploration for Ni–Cu–(PGE) sulfide deposits by VALE. They are associated with a 3.8 km long unit consisting of variable types of ultramafic rocks (talc schist, serpentinite and spinifex-textured komatiite). This ultramafic unit follows the steep dipping NW–SE trending Selva greenstone belt composed mainly by quartz-chlorite schists (interpreted as metasediments) and chlorite-actinolite schists (interpreted as metabasalts). Greenschist facies metamorphic parageneses characterize all rock types in the Selva greenstone belt.The komatiitic rocks in the Selva belt comprise a sequence of flows consisting of an upper spinifex-textured layer and a lower olivine cumulate layer. Although the spinifex and cumulus textures are well preserved in the field, the primary mineralogy of the komatiites has been completely replaced by greenschist facies metamorphic minerals. Platy olivine spinifex texture, consisting of an array of roughly parallel olivine plates, and random spinifex texture, consisting of randomly oriented olivine plates, are the most common primary volcanic textures in komatiites in the Selva greenstone belt. Platy and random spinifex texture is defined by former plates of olivine replaced by serpentine with minor actinolite, chlorite and magnetite, alternating with former matrix replaced by abundant actinolite and minor chlorite, talc, serpentine, and magnetite. The domains between olivine plates in both platy and random spinifex-textured rocks contain irregular arrays of fine-grained parallel crystals, representing primary fine-grained “quench” clinopyroxene crystals replaced by actinolite.Spinifex-textured komatiites have MgO contents bracket between 22.8 and 26.9 wt.%, and cumulate textured komatiites have MgO contents up to 40.6 wt.%. When plotted vs MgO contents, most major and minor elements fall on well-defined linear trends indicating control by olivine fractionation or accumulation. Komatiites from the Selva and Seringa (located in the Rio Maria Domain) belts are Al-undepleted with Al2O3/TiO2 ratios close to 20. Results for CaO, Na2O, and REE suggest that these elements were mobile and their abundances have been modified during metasomatic alteration. REE contents in some samples are very high (up to 40 times primitive mantle values) and REE patterns vary from flat (La/YbMN ∼ 1) to highly enriched in LREE (La/YbMN up to ∼ 10). The REE mobility may be related to hydrothermal alteration associated to Cu–Au mineralization in the region.The identification of spinifex-textured komatiites close to the Carajás Basin suggests the continuation of 3.0–2.9 Ga greenstone belts of the Rio Maria Domain within the Transition Subdomain, and enlarges the area with potential to host komatiite-associated Ni–Cu–PGE deposits.

Aeromagnetic interpretation in the south-central Zimbabwe Craton: (reappraisal of) crustal structure and tectonic implications

Regional aeromagnetic data from the south-central Zimbabwe Craton have been digitally processed and enhanced for geological and structural mapping and tectonic interpretation integrated with gravity data, to constrain previous interpretations based on tentative geologic maps and provide new information to link these structural features to known tectonic events. The derived maps show excellent correlation between magnetic anomalies and the known geology, and extend lithological and structural mapping to the shallow/near subsurface. In particular, they reveal the presence of discrete crustal domains and several previously unrecognised dykes, faults, and ultramafic intrusions, as well as extensions to others. Five regional structural directions (ENE, NNE, NNW, NW, and WNW) are identified and associated with trends of geological units and cross-cutting structures. The magnetic lineament patterns cut across the >2.7 Ga greenstone belts, which are shown by gravity data to be restricted to the uppermost 10 km of the crust. Therefore, the greenstone belts were an integral part of the lithosphere before much of the upper crustal (brittle) deformation occurred. Significantly, the observed magnetic trends have representatives craton-wide, implying that our interpretation and inferences can be applied to the rest of the craton with confidence. Geological–tectonic correlation suggests that the interpreted regional trends are mainly 2.5 Ga (Great Dyke age) and younger, and relate to tectonic events including the reactivation of the Limpopo Belt at 2.0 Ga and the major regional igneous/dyking events at 1.8–2.0 Ga (Mashonaland), 1.1 Ga (Umkondo), and 180 Ma (Karoo). Thus, their origin is here inferred to be inter- and intra-cratonic collisions and block movements involving the Zimbabwe and Kaapvaal Cratons and the Limpopo Belt, and later lithospheric heating and extension associated with the break-up of Gondwana. The movements produced structures, or reactivated older fractures, that were exploited by Late Archaean and Proterozoic mafic intrusions. There was interplay between vertical and horizontal tectonics as seen in similar terrains worldwide.

Archean magmatism and crustal evolution in the northern Tarim Craton: Insights from zircon U–Pb–Hf–O isotopes and geochemistry of ∼2.7 Ga orthogneiss and amphibolite in the Korla Complex

Extensive granitoid and mafic–ultramafic magmatism and crustal growth occurred at ∼2.7Ga in many cratons of the world, but the geodynamic setting during this period is complicated because both arc- and plume-related metavolcanic rocks are closely associated in several greenstone belts. Here, we present in situ zircon U–Pb–Hf–O isotopic and whole-rock geochemical data for meta-igneous rocks from the Korla Complex, northern Tarim Craton, NW China. SHRIMP and/or LA-ICP-MS zircon U–Pb dating indicates that two orthogneisses and an amphibolite crystallized at ∼2.71–2.74Ga and were affected by at least two metamorphic events at ∼2.0–1.8Ga and 0.8–0.6Ga, respectively. These are the oldest rocks with reliable crystallization ages so far identified in the northern Tarim Craton. These rocks exhibit heterogeneous zircon Hf isotopic compositions, with the most radiogenic analyses (average ɛHf(t)=+7.4, n=8) plotting on the depleted mantle evolution array and the most unradiogenic ɛHf(t) extending down to −5.6. This implies that both ∼2.7Ga depleted mantle and ancient continental crust at least as old as 3.4–3.5Ga contributed to the magma source. This conclusion is supported by zircon O isotopic data and Hf–O isotopic modeling of the orthogneisses. Available zircon U–Pb ages and Hf isotopic data show that two important Neoarchean magmatic events occurred at ∼2.71 and ∼2.55Ga in northern Tarim, and that both events involved synchronous crustal growth and reworking. This observation suggests that the apparent peaks of zircon Hf crustal model ages of these rocks do not represent the time of crustal growth but are artifacts of magma mixing. Geochemical data show that the parent mafic magmas of the amphibolites follow two distinct evolution trends: a Fenner trend of extreme Fe–Ti enrichment (Group I) and a typical tholeiitic trend with Fe–Ti enrichment followed by Fe–Ti depletion (Group II). This was probably controlled by the fractionation of Fe–Ti oxides, which was in turn controlled by magma oxygen fugacity. The orthogneisses follow the second trend and can be interpreted as the products of assimilation and fractional crystallization of the Group II magmas. Trace element systematics indicate that the parent magma of Group I amphibolites resembled Nb-enriched or high-Nb arc basalts, whereas the Group II amphibolites were probably derived from tholeiitic arc basalts. Such a rock assemblage is similar to the island arc volcanic association found in ∼2.7Ga greenstone belts in other cratons and indicates subduction of young and hot oceanic crust under an ancient continental block, which was probably an important process in the growth and differentiation of continental crust in the Archean.

Preserved and modified mid-Archean crustal blocks in Dharwar craton: Seismological evidence

We report significant lateral variability in shear wave velocity and Moho depth in the Archean crust, beneath the Dharwar craton, using earthquake waveform data recorded over 50 broadband seismographs. The craton is a continuously exposed Archean continental fragment divided into the west Dharwar craton (WDC) of age 2.7–3.36Ga, and the east Dharwar craton (EDC) of age, dominantly, 2.5Ga. The craton progressively transitions into the Southern Granulite Terrain (SGT) with age of metamorphism around 2.6Ga.The inversion and modeling of receiver function data reveal significant variation of Moho depth, viz., 38–54km in the WDC, and 40–46km in the SGT and 32–38km in the EDC. The average shear wave velocity (Vs) of crust beneath the WDC is ∼3.85km/s as compared to ∼3.6km/s in the EDC. We infer highly variable thickness (16–30km) of mafic cumulate (Vs≥4.0km/s and Vp≥7.0km/s) beneath the WDC, in contrast with a thin one (<5km) beneath the late Archean EDC. The 3.36Ga greenstone belt in the WDC has maximum basal layer thickness of ∼30km. These results suggest the intermediate-mafic composition and exceptional thickness of crust beneath the WDC (>50km) as compared to felsic to intermediate composition for the EDC crust with almost flat Moho (down to ∼38km). These results suggest preserved mafic crustal root beneath the middle Archean terrain in the WDC that has remained inert since then. On the other hand, felsic to intermediate composition of crust with a nearly flat Moho beneath the late Archean EDC could be a consequence of regional delamination of lower crust. The preserved distinct Moho topography across the Archean terrains suggests it as compositional boundary. Considering the surface exposure of 15–20km crust, based on P–T condition, in the granulite segment of the WDC, we speculate a Himalaya-like crustal thickness (50–70km) beneath the middle Archean crust pointing toward a plate tectonic-like scenario at ∼3.0Ga.

Geochemistry of komatiites from the Tipasjärvi, Kuhmo, Suomussalmi, Ilomantsi and Tulppio greenstone belts, Finland: Implications for tectonic setting and Ni sulphide prospectivity

Komatiites and komatiitic basalts within the Neo-Mesoarchaean, ca. 2.8Ga greenstone belts in the western part of the Karelian craton in Finland formed in diverse palaeoenvironments. The lavas in the Tipasjärvi and Kuhmo greenstone belts formed in an oceanic plateau environment, whereas the komatiites of the Suomussalmi belt formed in a continental rift setting, consistent with Nd and Pb isotope data of Suomussalmi felsic volcanics indicating the presence of older (>3Ga) crust in the belt. The komatiites and komatiitic basalts of the Ilomantsi greenstone belt show trace element systematics similar to those of modern continental arc volcanics, implying interaction of the lavas with felsic arc volcanics. PGE contents are low (2–4ppb Pt and Pd each) relative to most other Al-undepleted komatiites from elsewhere, interpreted to result from equilibration with sulphides during emplacement. Xenocrystic zircon grains, up to 3.2Ga in age, in some felsic volcanics and granitoids that are coeval with the komatiites indicate the presence of older continental basement in the Ilomantsi area. The Ilomantsi komatiites are interpreted to have formed at an active continental margin where basement rocks were extensively intruded and assimilated by felsic plutonic massifs. The komatiites of the Tulppio belt in northern Lapland resemble those of the Tipasjärvi, Kuhmo and Suomussalmi belts, but due to limited data, their emplacement environment remains uncertain.The Ni sulphide prospectivity of the Finnish komatiite belts appears to be relatively low, despite the identification of geochemical signatures of PGE depletion and crustal contamination in the Ilomantsi belt and, to a lesser degree, Suomussalmi and Tulppio belts, the presence of locally sulphide-bearing felsic volcanics and sedimentary rocks in the footwall of some of the komatiites, and the discovery of several small magmatic sulphide deposits in the Kuhmo, Suomussalmi, Ilomantsi and Tulppio belts. The main reasons are that the Finnish komatiites are relatively differentiated compared to those in mineralised belts elsewhere, and the Finnish belts contain a relatively large proportion of compound sheet flows relative to adcumulates, suggesting that dynamic lava channel environments where significant amounts of sulphide-bearing substrate could be assimilated and deposited are uncommon. Furthermore, the samples analysed to date lack the signature of scattered, but widespread PGE enrichment that is characteristic of mineralized komatiite belts and that has been interpreted to reflect wide-scale saturation of the magmas in Ni-sulphide melt.

Fluid sources and the role of abiogenic-CH 4 in Archean gold mineralization: Constraints from noble gases and halogens

The St Ives Goldfield of Western Australia hosts high grade gold mineralization in a variety of lithologies within ∼2.7 Ga greenstone belts. The low salinity CO 2-bearing fluids responsible for gold mineralization could have been generated either by metamorphic processes within the greenstones, or from external magmatic sources, and independently sourced abiogenic-CH 4 has been suggested as an important control on gold mineralization. In order to better constrain possible fluid sources, we applied a combination of He–Ne–Ar–Kr–Xe isotope and Cl–Br–I analyses to quartz and carbonate veins containing H 2O–CO 2 fluid inclusions, pyrite ± magnetite/hematite; and quartz veins containing CH 4-rich fluid inclusions, pyrrhotite ± pyrite. Samples containing H 2O–CO 2 fluid inclusions have higher Br/Cl and I/Cl values and less radiogenic noble gas isotope signatures than suggested for CH 4-rich fluid inclusions. The samples containing CH 4 fluid inclusions have maximum 40Ar/ 36Ar of ∼50,000; 20Ne/ 22Ne of ≤9.8 and maximum 21Ne/ 22Ne of ∼0.56. In contrast, H 2O–CO 2 fluid inclusions have maximum 40Ar/ 36Ar of only ∼20,000 and maximum 21Ne/ 22Ne of ∼0.14. The use of 40Ar– 39Ar methodology to measure K simultaneously with 40Ar/ 36Ar, enables us to show the highest 40Ar/ 36Ar ratios are representative of the fluids trapped at ∼2.65 Ga, and have not been significantly modified by post-entrapment 40Ar R production ( 40Ar R = radiogenic 40Ar). Furthermore, the 40Ar/ 36Ar ratio is correlated with the 21Ne/ 22Ne and 136Xe/ 130Xe ratios, demonstrating that other heavy noble gas isotope ratios are also close to their original composition. The highly radiogenic noble gas isotope signatures are unlikely to have been generated within the greenstone host-rocks, because seawater altered-volcanic rocks have high abundances of non-radiogenic atmospheric noble gas isotopes. In contrast, the strongly radiogenic noble gas isotope signatures are likely to have been generated in ancient basement rocks underlying the greenstones at ∼2.65 Ga. The combined noble gas and halogen data are consistent with models in which redox reactions triggered by interaction of independently and distally sourced H 2O–CO 2 and CH 4 localized high-grade gold mineralization. We suggest that the fluids were intimately associated with crustal magmatism in the St Ives Goldfield and that abiogenic CH 4 could be an important moderator of redox processes in other crustal environments.

Global Variability in the Platinum-group Element Contents of Komatiites

Platinum-group element (PGE) concentrations in komatiites of similar age but different petrological affinity, lithofacies type and emplacement style are remarkably similar when the effects of olivine fractionation and accumulation are accounted for. All komatiites display a slight enrichment of PPGE (Rh, Pt and Pd) over IPGE (Ir, Ru and Os) attributable to limited retention of IPGE in their mantle sources, coupled with saturation in Ir–(Os) alloys. Significant differences and trends exist, however, including subtle provinciality in Pt/Pd ratios, and in particular a systematic linear increase in komatiite PGE contents with age from the oldest at >3·5 Ga to the late Archaean group at c. 2·7 Ga, regardless of petrogenetic type. There is a linear secular trend of increasing Pt at a given MgO content and in Pt/Ti from 3500 Ma to 2700 Ma across all komatiite types. Barberton-type komatiites younger than 3000 Ma are generally slightly depleted in Pt and Pd relative to Munro-type. Munro-type komatiites from 3200–3500 Ga greenstone belts of the east Pilbara Craton of Western Australia are strongly depleted in PGE relative to younger belts, but similar to Barberton-type komatiites of the same age. Munro-type and Barberton-type komatiites from the 2900 Ma greenstone belts of the Youanmi Terrane in the Yilgarn Craton have similar PGE contents. Age, rather than komatiite type, appears to be the dominant control. These observations lead us to the hypothesis that the age control is related to a secular variation in the PGE content of komatiite mantle source regions. We envisage that Early Archaean komatiites could have been derived from plumes sourced in the deep mantle, which had been largely stripped of its highly siderophile element budget during core segregation and as yet incompletely replenished in PGE by the Late Veneer. Late Archaean komatiites have higher PGE contents as a result of progressive mixing-in of Late Veneer material over the period between the Late Heavy Bombardment (4·3–3·9 Ga) and 2·7 Ga, at which time the mantle had become effectively homogeneous with respect to PGE. The relative constancy of gamma-osmium values in komatiites throughout the Archaean is consistent with this hypothesis.

Textural and fluid inclusion constraints on the origin of the banded-iron-formation-hosted gold deposits at Maevatanana, central Madagascar

The Maevatanana deposits consist of gold-bearing quartz–sulphide veins crosscutting banded iron formation (BIF) within a metamorphosed 2.5 Ga greenstone belt. The host rocks are dominated by a sequence of migmatites, gneisses, amphibolites, magnetite-rich quartzites and soapstones, intruded by large granitoid batholiths (e.g. the 0.8 Ga Beanana granodiorite). In the mineralised rocks, pyrite is the dominant sulphide, in addition to accessory chalcopyrite and galena. Outside the immediate ore zone, the BIF is dominated by quartz + magnetite ± hematite, accompanied by cummingtonite, albite and biotite. Gold occurs as globular grains (usually <500 μm) within quartz crystals close to the sulphides and as invisible inclusions within pyrite and chalcopyrite (up to 2,500 ppm Au content). Fluid inclusion textural and microthermometric studies indicate heterogeneous trapping of a low-salinity (∼3.6 wt.% eq. NaCl) aqueous fluid coexisting with a carbonic fluid. Evidence for fluid-phase immiscibility during ore formation includes variable L/V ratios in the inclusions and the fact that inclusions containing different phase proportions occur in the same area, growth zone, or plane. Laser Raman spectroscopy confirms that the vapour phase in these inclusions is dominated by CO2 but shows that it may contain small amounts of CH4 (<1 mol%), H2S (<0.05 mol%) and traces of N2. Fluid inclusion trapping conditions ranged from 220 to 380°C and averaged 250°C. Pressure was on the order of 1–2 kbar. The abundant CO2 and low salinity of the inclusions suggest a metamorphic origin for the fluid. Likewise, the presence of H2S in the fluid and pyritisation of the wall-rock indicate that gold was likely transported by sulphide complexing. Fluid immiscibility was probably triggered by the pressure released by fracturing of the quartzites during fault movements due to competence differences with the softer greenstones. Fracturing greatly enhanced fluid circulation through the BIF, allowing reaction of the sulphide-bearing fluids with the iron oxides. This caused pyrite deposition and concomitant Au precipitation, enhanced by fluid phase separation as H2S partitioned preferentially into the carbonic phase.

Light rare earth element depleted to enriched basaltic flows from 2.8 to 2.7 Ga greenstone belts of the Uchi Subprovince, Ontario, Canada

A compositionally diverse group of basaltic flows are present in the 2.7 Ga St. Joseph assemblage of the St. Joseph greenstone belt, Uchi Subprovince. A spectrum of light rare earth element (LREE) depleted to enriched patterns corresponds to low-, medium- and high-K series (LKS, MKS, HKS) basalts, shoshonites and leucitites of some modern arcs. More primitive, tholeiitic LKS and MKS basalts plot with counterparts from modern intraoceanic arcs, such as the South Sandwich Islands, and show no evidence of contamination by continental crust. Consequently they are interpreted to have erupted in an oceanic setting. There is an array from low Nb with high Zr/Nb (29) to high Nb but lower Zr/Nb (15) indicative of a mantle wedge similar to average mid ocean ridge basalts (MORB; average Zr/Nb = 32). These basalts also show positive to negative Nb, Hf and Zr anomalies. Negative anomalies are interpreted as the product of first stage melting, where LREE and MREE are added to the mantle wedge from dehydration of the slab, leaving a residue with high Nb/La, Hf/Nd and Zr/Nd. Second stage melts inherit the high HFSE/REE ratios. Within the Birch–Uchi greenstone belt, basalts of the ∼2.8 Ga Woman assemblage includes both LKS and HKS. These are characterized by similar Nb–Zr/Nb and HFSE/REE systematics, signifying that these are general features of sub-arc mantle wedges that developed over 100 Ma in Uchi subprovince arcs. Enriched St. Joseph assemblage basaltic rocks cannot stem from sediment melting on the slab, as Th/Ce ratios are generally < 0.1. Neither can they derive from an ocean island basalt component because Nb anomalies do not correlate with La/Yb n. Rather, the trend from primitive LKS through MKS to HKS, leucitite and shoshonite is interpreted to reflect progressively lower degrees of partial melting at increasing depths and/or less depleted mantle wedge.

New PGE and Re/Os-isotope data from lower crustal sections of the Vredefort Dome and a reinterpretation of its "crust on edge" profile

Granulites exposed in the core of the Vredefort Dome are generally thought to represent components of the Archean middle crust of the central Kaapvaal Craton. These rocks were exhumed from depths of at least 20 km following an ~2.02 Ga meteorite impact event, exposing a “crust on edge” section believed to represent a continuous magmatically-differentiated sequence from upper crust possibly down to Moho. Platinum group element concentrations suggest that mafic and ultramafic enclaves in the lower most sections are fragments of greenstone belt material metamorphosed at granulite grade. Re/Os isotopic data are consistent with previously published U-Pb zircon data, which suggest that the mafic and ultramafic rocks have a primary age of ~3.5 Ga and were metamorphosed to granulite grade at ~3.1 Ga. However, near the core of the Vredefort Dome across a major shear zone, other ~3.5 Ga greenstone belt remnants are at distinctly lower metamorphic grade. The juxtaposition of mafic-ultramafic rocks with a different metamorphic history in the core of the Vredefort structure requires a modified “crust on edge” model that incorporates tectonic stacking of different crustal segments at ~3.1 Ga.

Fluid Inclusion Characteristics of Auriferous Quartz Veins in Archean and Paleoproterozoic Greenstone Belts of Eastern and Southern Finland

Fluid inclusions in auriferous quartz veins in five deposits in the Archean (2.9–2.7 Ga) and Paleoproterozoic (2.0–1.75 Ga) greenstone belts of eastern and southwestern Finland have been characterized on the basis of their composition, reconstructed isochores, and decrepitation behavior. The dominant fluids associated with gold mineralization are H2O-CO2 ± CH4 ± halite (Archean) and H2O-CO2-CH4 or H2O-CH4 (Paleoproterozoic) in composition; phase ratios were generally uniform H2O/CO2 in the Archean but were variable in the Paleoproterozoic. In the Archean Au ± sulfide quartz veins, the mineralizing fluids contain less CH4 (<30 mol %) and have higher salinities (up to 39 wt % NaCl equiv) than the corresponding fluids in the Paleoproterozoic Au-sulfide-quartz veins (up to 38 mol % CH4 and below 13 wt % NaCl equiv). Additionally, the estimated pressure-temperature conditions for gold mineralization differ markedly between the two domains. Mineralization occurred in a pressure-temperature range of 300° to 500°C at 2 to 3.5 kbars in the Archean domain and 200° to 300°C at 1 to 2 kbars in the Paleoproterozoic domain. Decrepigrams for the Archean and Paleoproterozoic auriferous quartz veins show different characteristics and permit classification of the different fluid inclusion populations on a regional scale.

Upper mantle processes beneath the 2.7 Ga Abitibi belt, Canada: a trace element perspective

Abstract Mantle processes associated with volcanism in the Abitibi belt remain poorly understood despite numerous detailed studies of its volcano-sedimentary successions. This paper employs log plots of Nb/Y and Zr/Y to distinguish the sources of well-defined Abitibi volcanic suites and to clarify processes in the upper mantle during greenstone belt development. Mafic rocks of the Southern Volcanic Zone in the Abitibi belt define two main trends on log plots of Nb/Y versus Zr/Y. Trends for the Kidd–Munro Assemblage and Malartic Block correspond to mixtures of strongly deleted plume asthenosphere and adakitic melts. The shared trace element systematics of these volcanic successions confirm they are parts of a single authochthonous supracrustal assemblage that extended across the southern Abitibi belt. The younger Kamiskotia Assemblage and Blake River Group correspond to crustal extension episodes within an Abitibi island arc. Trace element plots establish that they were derived from mantle plume-related asthenosphere and not the late Archean MORB-source upper mantle associated with island arc assemblages elsewhere in the southern Superior Province. The data suggests widespread lateral flow of mantle plume asthenosphere occurred in the vicinity of the Abitibi arc. Mantle mixing with adakitic melts continued between ∼2716 and ∼2698 Ma. The data are not consistent with wet komatiite models that generate ultramafic volcanic rocks from typical upper mantle following hydration of a mantle wedge. Chemical mixing trends for mafic magma sources indicate melt hybridization of the Abitibi mantle wedge. The trends require a slab melt origin for adakitic magmas rather than internal differentiation of underplated mafic crust. Comparisons with other 2.7 Ga greenstone belts suggest that mantle plume interaction with Archean subduction zones, as proposed for the Abitibi belt, can be distinguished in the NbZrY systematics of associated mafic volcanic rocks.

太古代海洋における硫酸還元菌の活動と生息環境

Sulfate-reducing bacteria (SRB) are considered to be one of the oldest micro-organisms in the Earth history. Sulfur isotope data of sedimentary pyrite indicate that SRB have been active at least since 3.5 Ga. However, living environments of SRB, such as concentration level of sulfate in the Archean oceans and sulfur flux into Archean oceans, have been debated vigorously.Detailed sulfur isotope studies were performed on individual pyrite crystals in black shale and chert samples collected from several Archean greenstone belts. Sulfur isotope compositions of pyrite in all samples are fractionated, suggesting their biogenic origin. It is found that some representative sample sets from 2.7 Ga greenstone belts show a good correlation between organic carbon and pyrite sulfur concentrations. These results suggest that Archean oceans have been sulfate-rich at least since 2.7 Ga. This sulfate-rich ocean model conflicts with the sulfate-poor ocean model based on the mass independent fractionation of sulfur isotopes.

Origin and evolution of mid- to late-Archean crust in the Hanikahimajuk Lake area, Slave Province, Canada; evidence from U–Pb geochronological, geochemical and Nd–Pb isotopic data

A geochronological, major/trace element geochemical and Nd–Pb isotopic study of early- to mid-Archean basement granitoids/gneisses, ∼2.68 Ga greenstone belt and 2.58–2.62 Ga late-Archean granitoids from the Hanikahimajuk Lake area of the northern Slave province was undertaken to constrain the tectonomagmatic evolution of the north-central area of Slave province, Northwest Territories, Canada. The basement tonalites are precisely dated at 3378±1 Ma by U–Pb zircon methods and are characterized by steeply fractionated REE patterns with no negative Eu anomalies. They were likely produced through partial melting of mafic crust at >0.8 GPa. The εNd T of these rocks range from −0.3 to +2.1, implying involvement of even older crustal material in their genesis.Two different types of ∼2.68 Ga mafic to intermediate volcanic rock are recognized in the Hanikahimajuk Lake area: tholeiitic rocks with no Ta–Nb depletions and calc-alkaline rocks with slight negative Ta–Nb depletions. All of these rocks show clear evidence for contamination by the ∼3.4 Ga basement tonalite, thus precluding the possibility that some of these tholeiitic mafic rocks were produced in a mid-oceanic ridge setting. From the data presented here, we conclude that the tholeiitic rocks were produced in a back-arc setting, likely similar to that of the Japanese arc. The calc-alkaline rocks may either be a remnant of the volcanic arc or part of the same back-arc system. However, the close proximity with the tholeiitic rocks in the field favours the latter possibility. Mafic to intermediate rocks with similar geochemical and isotopic characteristics are also found from the Yellowknife area of the southwestern Slave province, and from the Indin Lake area of the westernmost Slave province (south of Grenville Lake area). This observation may indicate that the mafic-dominated volcanic belts of the western Slave province may be a remnant of a single, now collapsed and shortened back-arc basin. The existence of syn-volcanic calc-alkaline plutonic rocks with similar geochemical and isotopic features as the calc-alkaline volcanic rocks supports this possibility.The geochemistry of late-Archean granitoids suggests that they are produced mainly through partial melting of mafic igneous protoliths and clastic sedimentary rocks (and possibly tonalitic rocks). Similarity in the εNd(2.6) values between these granitoids and the supracrustal rocks in the area is in accord with this model. Such protoliths are also consistent with the tectonic setting at which the ∼2.68 Ga volcanic rocks were generated, because mafic igneous rocks and clastic sedimentary rocks (with lesser felsic igneous rocks) are the major constituents of a back-arc basin.

Evidence for extensive Proterozoic remobilization of the Aldan shield and implications for Proterozoic plate tectonic reconstructions of Siberia and Laurentia

A geological traverse across the Aldan shield along the Aldan River shows that the area is underlain by two distinct rock associations. The Middle Aldan association consists of metasedimentary rocks, mainly quartzite, that have been intruded by a potassic biotite granite. Downstream, north of the Middle Aldan association, lies the Lower Aldan association, which consists mostly of charnockite with rafts of older granulite gneiss that contains abundant metasedimentary layers. UPb dating of zircons from the Middle Aldan association granite (1900 Ma) and Lower Aldan association charnockite (maximum age 1918 Ma) and granitic gneiss (maximum age 2230 Ma) shows that the majority of the rocks exposed along the Aldan River are Proterozoic in age. Although they have Proterozic crystallization ages, the granite, granite gneiss and charnockite yield Archean Nd model ages, suggesting that they formed by remobilization—that is, partial or complete remelting—of earlier Archean crust. In contrast, pelitic gneisses included within the charnockites of the Lower Aldan association give Proterozoic Nd model ages, indicating that a substantial amount of Proterozoic rock is incorporated within the Lower Aldan association. The present results show that the rocks along the Aldan River, which are part of the Aldan block, display a significantly different history from those in the Olekma block to the west. The Olekma block contains ca 3.0 Ga greenstone belts Late Archean amphibolite-grade granitic gneisses; the Proterozoic remobilization that typifies the Aldan terrane is absent. The Olekma block was thrust under the Aldan block during the ca 1.9 Ga orogeny contemporaneous with, or slightly after the 1.9 Ga charnockitic event in the Aldan. The 1.9 Ga magmatic and granulite event seen in the Aldan is similar in age and character to the Thelon magmatic zone of northern Canada. This correlation allows the development of a preferred construction for the Precambrian Laurentia-Siberia connection in which the present day southern portion of the Siberia platform was connected to the northern margin of Laurentia.

Tectonic setting and terrane accretion of the Archean Zimbabwe craton

The Archean Zimbabwe craton is made of a number of distinct tectonostratigraphic terranes assembled by plate tectonic processes. The central Tokwe terrane consists of 3.5–2.95 Ga gneissic rocks and structurally complex inliers of possibly older greenstone belts. These are overlain unconformably by a 2.9–2.8 Ga assemblage of mafic and felsic volcanic rocks and conglomerates, and a separate 3.0–2.7 Ga southeastward thickening platform sequence of sandstone, shale, and limestone. 2.7 Ga greenstone belts form two distinctly different domains flanking the central terrane. Northwest of the ancient gneissic terrane, ca. 2.7 Ga greenstone belts comprise a series of calc-alkaline lavas and intercalated sedimentary rocks intruded by syn-volcanic plutons. Southeast of the ancient gneissic complex, 2.7 Ga greenstone belts consist of thick piles of tholeiitic basalts overlying ultramafic lavas, resting allochthonously over the shallow-water platform sequence and older gneissic terrane. This division of the Zimbabwe craton is interpreted to show that the central Tokwe terrane had a continental magmatic arc built on its northwestern edge, as its southeastern margin rifted from another fragment, forming the Sea of Umtali. A passive-margin sedimentary wedge formed on the rifted southeastern edge of this ancient continent, and prograded onto the craton during sedimentary and tectonic loading of the craton margin. The southeastern greenstone belts formed as thick oceanic crust (oceanic plateau) in this back-arc basin, and were later obducted on to the rift and passive margin sequence as the Sea of Umtali closed ca. 2.7 Ga. This was followed by intrusion of granitic plutons of the Chilimanzi suite ca. 2.6 Ga in a tectonic regime of intracontinental strike-slip faulting, representing a response to the Zimbabwe-Kaapvaal continent-continent collision. Crustal and lithospheric thickening during intrusion of these late granites may have played a role in stabilizing the Zimbabwe craton and forming the lithospheric root.

3.07-2.97 Ga greenstone belt formation in the northeastern Kaapvaal Craton; implications for the origin of the Witwatersrand Basin

3.07-2.97 Ga greenstone belt formation in the northeastern Kaapvaal Craton; implications for the origin of the Witwatersrand Basin

An overview of the relationship between granitoid intrusions and gold mineralisation in the Archaean Murchison Province, Western Australia

In the Archaean Murchison Province of Western Australia, granitoid batholiths and plutons that intruded into the ca. 2.7–2.8 Ga and ca. 3.0 Ga greenstone belts can be divided into three major suites. Suite I is a ca. 2.69 Ga monzogranite-granodiorite suite, which was derived from anatexis of old continental crust and occurs as syn-tectonic composite batholiths over the entire province. Suite II is a trondhjemite-tonalite suite (termed I-type) derived from partial melting of subducted basaltic crust, which intruded as syn- to late-tectonic plutons into the greenstone belts in the northeastern part of the province where most of the major gold deposits are situated. One of the Suite II trondhjemite plutons has a Pb−Pb isochron age of 2641±36 Ma, and one of the structurally youngest tonalite plutons has a minimum Pb−Pb isochron age of 2630.1±4.3 Ma. Suite III is a ca. 2.65–2.62 Ga A-type monzogranite-syenogranite suite which is most abundant in the largely unmineralised southwestern part of the province. Gold deposits in the province are mostly hosted in brittle-ductile shear zones, and were formed at a late stage in the history of metamorphism, deformation and granitoid emplacement. At one locality, mineralisation has been dated at 2636.8±4.2 Ma through a pyritetitanite Pb−Pb isochron. Lead and Sr isotope studies of granitoids and gold deposits indicate that, although most gold deposits have initial Pb isotope compositions most closely similar to those of Suite II intrusions, both Suite I and Suite II intrusions or their source regions could have contributed solutes to the ore fluids. These preliminary data suggest that gold mineralisation in the Murchison Province was temporally and spatially associated with Suite II I-type granitoids in the northeastern part of the province. This association is consistent with the concept that Archaean gold mineralisation was related to convergent-style tectonic settings, as generation of both Suite II I-type granitoids and hydrothermal ore fluids could have been linked to the dehydration and partial fusion of subducted oceanic crust, and old sialic crust or its anatectic products may also contribute solutes to the ore fluids. Integration of data from this study with other geological and radiogenic isotope constraints in the Yilgarn Block argue against direct derivation of gold ore fluids from specific I-type granitoid plutons, but favour a broad association with convergent tectonics and granitoid magmatism in the late Archaean.