Geological controls on refractory ore in an orogenic gold deposit, Macraes mine, New Zealand

The Macraes mine is hosted in an orogenic (mesothermal) gold deposit in metasedimentary rocks of the Otago Schist belt. Much gold occurs within altered schist with minimal silica-addition, and this study focuses on altered schist ore types. The unmineralized host schists are chemically and mineralogically uniform in composition, but include two end-member rock types: feldspathic schist and micaceous schist. Both rock types have undergone hydrothermal alteration along a shallow-dipping foliation-parallel shear zone, but their different rheological properties have affected the style of mineralisation. Micaceous schist has been extensively recrystallized and hydrothermally altered during ductile deformation, to form ores characterized by abundant, disseminated millimetre-scale pyrite cubes (typically 1–2 wt% S) and minor silicification. The earliest pyrite contained Ni and/or As in solid solution and no gold was imaged in these pyrites or later arsenopyrite grains. The ore type is refractory and gold recovery by cyanide leaching is less than 50%, with lowest recovery in rocks that have been less affected by later brittle deformation. In contrast, hydrothermally altered feldspathic schist is characterized by mineralised black microshears and veinlets formed during shear-zone related brittle deformation. Microsheared ore has relatively low sulphur content (<0.7 wt%) and muscovite has been illitised during hydrothermal alteration. Pyrite and arsenopyrite in microshears are fractured and deformed, and contain 1–10 m blebs of gold. Later pyrite veinlets also contain micron- to submicron-scale inclusions of sphalerite, chalcopyrite, galena, and gold (10 microns). Gold in microsheared ore is more readily recoverable than in the refractory ore, although encapsulation of the fine gold grains inhibits cyanidation. Both microsheared ore and disseminated pyritic ore pass laterally into mineralised black shears, which contain hydrothermal graphite and late-stage cataclastic sulphides. This black, sheared ore releases gold readily, but the gold is then adsorbed on to gangue minerals (preg-robbed) and net cyanidation recovery can be less than 50%. Hence, low gold recovery during cyanidation results from (1) poor liberation of gold encapsulated in microcrystalline quartz and unfractured sulphide grains, and (2) preg-robbing of liberated gold during cyanidation. Introduction of pressure-oxidation of ore prior to cynidation has mitigated these issues.     

Geochemistry of late metamorphic hydrothermal alteration and graphitisation of host rock, Macraes gold mine, Otago Schist, New Zealand

The Macraes gold deposit in the Otago Schist, New Zealand, formed during late metamorphic fluid flow through a lower greenschist facies shear zone. Mineralisation occurred near to the brittle-ductile transition at about 300 °C. Large volumes of host rock in a shear zone up to 120 m thick have been hydrothermally altered by this fluid activity. Most alteration is not structurally controlled apart from proximity to the shear zone. Ductile and brittle microshears traverse the most mineralised rocks and some structural control of fluid flow occurred as well. Fluid flow was slow, similar to that in metamorphic rocks (mm/year) and diffusion through interconnected fluid was a significant chemical process. Localised extensional hydrofractures (m scale) are filled with mineralised quartz. Most alteration of the host rocks was isochemical with respect to the lithophile elements, and mineralised rocks have been variably enriched in As, Au, Sb, W, Mo and Bi, but not Co or Cd. Addition of sulphur has occurred to both host rocks and mineralised rocks, up to 1 wt.% above a background of 0.1 wt.%. Host rock sulphur is mainly pyritic and is not structurally controlled. Mineralised rocks have pyrite and arsenopyrite along microshears. Pyrite, chalcopyrite, sphalerite and galena have formed from sulphidation of silicates with no addition of metals. Graphite has been added to mineralised rocks along microshears, up to 3 wt.% locally, above a background of 0.1 wt.% noncarbonate carbon. Graphite deposition may have occurred as a result of mixing of two fluids, water+methane, and water+carbon dioxide. Graphitisation and sulphidation reactions released low δD water, which accumulated in the slow-moving mineralising fluid. Distinction between this low δD reaction water and meteoric water incursion is difficult.     

Raman characterization of carbonaceous material in the Macraes orogenic gold deposit and metasedimentary host rocks,New Zealand

Raman spectroscopic and petrographic analyses were performed on samples collected from zones distal and proximal to the Macraes gold deposit in the Otago Schist of New Zealand to characterize the features and possible origins of Carbonaceous Material (CM) and to assess the potential role of CM in the formation of gold deposits. CM is a common component in meta-sedimentary orogenic gold deposits, and it has been proposed that CM contributes to gold mineralization processes, but the details of the mechanisms responsible are not fully understood. Documentation of the origins of the Otago schist CM will improve our understanding of the role of CM in gold deposits.This work has identified four types of CM of varying thermal maturity and origins from prehnite–pumpellyite grade to lower greenschist grade samples. In prehnite–pumpellyite and pumpellyite–actinolite grade rocks, low-maturity CM 1 coexists with framboidal pyrite, indicating an in-situ, sedimentary origin, with a potential association with the source of gold. Low crystallinity CM 2 is also found in low grade samples and is likely to have been deposited from fluids unrelated to gold mobilization. CM 3 is the highest maturity CM recognized. CM 3 is found in samples from the highest metamorphic grades studied (lower greenschist facies), where bands of CM 3 cross cut the foliation, CM 3 is therefore thought to have been transported by fluids, though possibly only at short length scales. CM 4 is less mature than CM 3 and is found in mineralized rocks in association with sulfide minerals and gold. CM 4 is likely to have a depositional origin but its precise role with respect to gold mineralization has not been identified.     

Kilometre-scale structural setting of ore shoots in the Frasers gold deposit,Macraes mine,New Zealand

The Hyde-Macraes Shear Zone in southern New Zealand contains the circa 10 million ounce Macraes gold deposit, one of the larger Phanerozoic orogenic gold deposits discovered to date globally. Approximately 50% of this 10 million ounce resource is hosted by 5 major ore shoots up to 400 m wide and 1500 m long in the Frasers area at the southern end of the mine. Higher grade (>1.5 g/t Au) ore shoots are located along and immediately below the Hangingwall Shear, the principal strand of the Hyde-Macraes Shear Zone at the Frasers deposit. They typically trend parallel to the intersection of the shear and foliation in the underlying schist, commonly where the foliation dips more steeply that the overlying Hangingwall Shear. Especially thick zones of higher grade mineralised rock are located between the Hangingwall Shear and underlying second order splay shears whose position correlates with minor right-hand bends in the strike of the overlying Hangingwall Shear. Lower grade (<1.2 g/t Au), but economically significant, ore shoots are located within mineralised schists below the Hangingwall Shear. Outer margins of these lower grade ore shoots are generally parallel to the strike of the foliation in the host schist. They are most extensive where open disharmonic folding has resulted in the strike of the foliation diverging from that of the overlying Hangingwall Shear. No correlation exists between the position of any ore shoots and gently dipping jogs in the Hangingwall Shear, despite mineralisation occurring during reverse movement on the Hyde-Macraes Shear Zone. Instead the angular relationship between various strands of the Hyde-Macraes Shear Zone at Frasers and foliation in underlying schists is the most consistent structural feature likely to predict the location, extent, and orientation of ore shoots within the Frasers segment of the Hyde-Macraes Shear Zone.     

Numerical modelling of mechanical controls on coeval steep and shallow dipping auriferous quartz vein formation in a thrust zone,Macraes mine,New Zealand

The Hyde-Macraes Shear Zone (HMSZ) is a regionally continuous, low-angle, NE dipping (~15°) late-metamorphic thrust zone in the Mesozoic Otago Schist. The shear zone, which is host to large volumes of mineralised schist, consists of foliated fissile schist with some massive schist pods. Two sets of quartz veins are found within the HMSZ: thrust-related, shallowly dipping veins that were emplaced parallel or sub-parallel to the shears and swarms of steeply dipping extensional veins, which cut across the metamorphic foliation. The latter are restricted to the massive schist pods. Mutual cross-cutting relationships occur between steep extensional veins and shallow-dipping veins, suggesting that they formed contemporaneously. The co-existence of these two vein types locally implies local rotation of the principal stress axes to produce extensional veins within a regional thrust setting. The steep extensional veins are spatially related to lateral and oblique ramps within the HMSZ. Three-dimensional mechanical models show that these lateral or oblique ramps can produce favourable conditions for extensional vein formation when combined with a high fluid pressure and oblique convergence. Mechanical requirements include a reduced differential stress, a positive volumetric strain and an increase in the horizontal shear stress. Our models show that under certain conditions, it is possible for extension-related structures to form during shortening because of local changes in the stress state without the need for a regional scale switch in the imposed stress field. The convergence direction across the HMSZ during formation of the steep extensional veins was ~WNW.     

Arsenic distribution during formation and capping of an oxidised sulphidic minesoil, Macraes mine, New Zealand

A minesoil has developed over 5 years oxidative exposure on sulphide concentrate tailings (ca. 1 wt.% As) at the Macraes mesothermal gold mine, New Zealand. The minesoil has a dry crust which has formed due to evaporative drying. This dry crust is enriched in arsenic (ca. 5 wt.% As) as scorodite (FeAsO₄·2H₂O) because of upward mobility of dissolved arsenic during drying. Similar enrichment of arsenic has occurred along the walls of desiccation cracks which extend over 1 m into the minesoil. Capping of the tailings and minesoil with wet tailings (pH=8) results in dissolution of scorodite and remobilization of arsenic on the millimetre scale. Experimental capping of the minesoil with wet calcium carbonate remobilized some arsenic from scorodite on the centimetre scale, but much original arsenic enrichment was preserved after 400 days. A layer of gypsum (CaSO₄·2H₂O) and iron oxyhydroxide cementation developed at the interface between the minesoil and the experimental calcium carbonate cap, restricting water flow. This layer was ca. 1 mm thick after 400 days. Theoretical comparison between advection and diffusion in the minesoil suggests that diffusion is an important mechanism for chemical mobility on the 1–50-year time scale. However, advection can be important in secondary porosity of the dry crust of the minesoil and water penetrates this zone at a rate of 1.5 mm/day.     

Consolidation and incipient oxidation of alkaline arsenopyrite-bearing mine tailings,Macraes Mine,New Zealand

Fine grained (ca. 15 μm), arsenopyrite-bearing mine tailings have been exposed to drying and oxidation for 4 a pending relocation. The tailings are still partly covered by a pond of decanted pore waters. The water table in drying tailings has lowered by 1–3 m and desiccation cracks up to 2 cm wide have formed on the 1 m scale, extending through the unsaturated zone. Tailings in the unsaturated zone have similar pore water contents to saturated tailings: typically 16–32 wt% water. Saturated tailings retain alkaline pH (ca. 10) from the mine cyanidation plant, but pH lowers progressively towards ca. 7 near the surface, or near desiccation cracks, in the unsaturated zone. The redox state of the tailings changes in parallel with pH, with an empirical relationship: Eh(mV)=−55 pH+290. Water in the remnant decant pond reflects this relationship also. Unsaturated tailings have variable but low permeabilities, typically 10⁻³ to 10⁻⁴ m/day, and more permeable horizons have allowed incursion of oxygenated air and/or rain water from desiccation cracks. Sulphide grains in all tailings examined are unaltered. Sulphides and solutions in the tailings are out of thermodynamic equilibrium predicted from the redox–pH conditions, due to kinetic constraints. Incursion of rain water locally facilitates deposition from pore waters of insoluble Fe oxide and arsenate minerals, thus fixing As in the dry unsaturated tailings.     

Geochemical signature of orogenic hydrothermal activity in an active tectonic intersection zone,Alpine Fault,New Zealand

A highly faulted and fractured rock mass has developed at the intersection of the Alpine and Hope faults, two major active faults in the South Island, New Zealand. The Alpine Fault is an oblique dextral reverse fault at the late Cenozoic-Recent Pacific-Australian plate boundary. The Hope Fault is a strike-slip fault parallel to the plate convergence vector. Hydrothermal fluids driven by the active tectonic processes have passed through the fractured rock mass, causing localised rock alteration and vein formation. Mylonites in the Alpine Fault zone are crosscut by cm-scale veins of quartz and/or ankerite with minor sulphides, with cemented breccias in dilational jogs. Breccia clasts and immediate (cm-scale) host rocks have been variably impregnated with carbonates and quartz. This generation of veins, breccias and altered rocks is post-dated by cataclasite and fault gouge zones which have been cemented by calcite, illite, smectite and chamosite. Ankerite and calcite have ¹⁸O between +10 and +30, and ¹³C between 0 and –8. These minerals are inferred to have formed from water with variable components of both meteoric and crustally exchanged fluid. Rock alteration associated with ankerite–quartz veins has added arsenic (up to 200 ppm As), strontium, and some Y to the rocks. Host-rock mylonites (<2 ppm As) have been depleted in arsenic compared to their precursors (5–15 ppm As). This depletion of arsenic in the middle crust provides the source for arsenic in shallower-level vein systems.Editorial handling: N. White     

Infrared spectroscopic characterisation of arsenate (V) ion adsorption from mine waters,Macraes mine,New Zealand

Processing waters contain up to 10 mg l⁻¹ dissolved As at the Macraes mine, New Zealand, and this is all removed by adsorption as the water percolates through a large earth dam. Laboratory experiments were set up to identify which mineral is the most effective substrate for this adsorption of As. The experiments were conducted using infrared (IR) spectroscopy of thin mineral films adhering to a ZnSe prism. Silicates, including kaolinite, adsorbed only small amounts of As which was readily washed off. Hydrated Fe oxides (HFO) were extremely effective at adsorbing As, particularly the natural amorphous HFO currently being deposited from dam discharge waters at the Macraes mine. An adsorption isotherm determined for this natural material has the adsorption constant, K_ads=(1.9±0.4)×10⁴ M⁻¹, and the substrate becomes saturated with adsorbed As when solution concentrations exceed about 50 mg l⁻¹. Saturation is not being reached at the Macraes mine. Arsenic adsorbed on to natural HFO has a distinctive IR spectrum with the absorption peak varying from 800 cm⁻¹ (alkaline solutions) to 820 cm⁻¹ (neutral to acid solutions). Much of this adsorbed As is strongly bound and difficult to wash off. Arsenate ions adsorb in a bidentate structure which may be a precursor for scorodite crystallisation.     

Occurrence of gold in hydrothermal pyrite,western Taupo Volcanic Zone,New Zealand

The Taupo Volcanic Zone (TVZ) on the North Island of New Zealand is located above the subducting Pacific slab and hosts hydrothermal systems related to subduction, arc magmatism and crustal extension. In these systems, gold is transported primarily as sulphide complexes, with gold being deposited in response to boiling and mixing of the deep geothermal fluids. Conglomerate cobbles and hydrothermal fumaroles from the upper Retaruke River in the western TVZ are mineralised deposits that have been eroded from the nearby Miocene alluvial sedimentary plateau. Abundant gold-bearing pyrite was precipitated in the conglomerates and fumaroles by late hydrothermal process, primarily occurring as veinlets, disseminations and fine-grained aggregates which consists discrete euhedral microcrystals. Scanning electron microscopy combined with X-ray energy disperses spectroscopy shows that pyrites are commendably affected by late hydrothermal fluids, possibly by the carbonate fluid. Electron probe microanalysis revealed that the pyrite contains 20–120 ppm Au (averaging 60 ppm). Wavelength dispersive spectral elemental mapping suggests that gold is distributed uniformly in pyrite, indicating structurally bound gold (solid solution) in pyrite. Gold mineralisation recognised in the conglomerates and fumaroles demonstrates that the upper Retaruke River is a promising target for future gold exploration.     

Southern Alps Cu-Au hydrothermal system,Westland, New Zealand

The Southern Alps of New Zealand is an actively rising mountain belt which displays a thermal anomaly adjacent to the Alpine Fault, the Australian-Pacific plate boundary. Extensive fluid movement occurs in this uplift zone, resulting in metallic vein mineralization. Gold mineralization is confined to greenschist facies rocks, while younger veins in amphibolite facies rocks near the Alpine Fault are enriched in copper. Transport and deposition of metals in this complex hydrothermal system is governed by interaction between rising metamorphic fluids and downward-percolating meteoric fluid. Metamorphic fluids have equilibrated with graphitic schist country rock and are relatively reduced. Infiltration and mixing of meteoric water increases oxygen activity and decreases sulphur activity in the fluid. Oxidised meteoric water heats up and dissolves Cu during downward percolation. This Cu is deposited as the fluid becomes more reduced. Hence, there is a progressive increase in copper content in the middle portions of the hydrothermal system, especially in the more permeable highly fractured rocks near the Alpine Fault.     

Delayed accumulation of placers during exhumation of orogenic gold in southern New Zealand

The giant gold placer system on the Otago Schist of southern New Zealand was derived from Mesozoic orogenic gold deposits in the underlying schist basement. The core of the schist basement was exhumed in the middle Cretaceous, coeval with the accumulation of the oldest preserved nonmarine sedimentary rocks in the area (ca 112 Ma). Those sedimentary rocks contain quartz clasts, with distinctive ductile deformation textures, that were derived from structural zones in, or adjacent to, major orogenic gold deposits. Quartz textures in these structural zones are readily distinguishable from the rest of the schist belt, and hence provide a fingerprint for erosion of gold. The earliest sedimentary rocks on the margins of the gold-bearing schist belt are immature, and were derived from unoxidised outcrops in areas of high relief. Gold was not liberated from unoxidised basement rocks during erosion, and was removed from the system without placer concentration. Placer concentration did not begin until about 20 million years later, when oxidative alteration of gold deposits had facilitated gold grain size enhancement from micron scale (primary) to millimetre scale (secondary). Subsequent erosion and recycling of gold in the early Cenozoic, and again in the late Cenozoic, caused additional concentration of gold in progressively younger deposits. The Klondike giant placer goldfield of Canada had a similar geological history to the Otago placer field, and Klondike placer accumulation occurred in the late Cenozoic, at least 70 million years after Mesozoic exhumation of orogenic gold. The giant placer deposit on the western slopes of the Sierra Nevada in California occurs in Eocene and younger sedimentary rocks, at least 40 million years younger than the timing of major exhumation of the source rocks. Circum-Pacific giant gold placers formed under entirely different tectonic regimes from the emplacement of their source orogenic deposits, and these giant placer deposits do not form in foreland basins associated with convergent orogens. Formation of giant placers requires less active erosion and more subdued topography than the collisional orogenic activity that accompanied emplacement of source gold deposits in basement rocks, as well as oxidative alteration of the primary deposits to liberate gold from sulfide minerals and enhance secondary gold grain size.     

Geological analogue for circumneutral pH mine tailings: implications for long-term storage,Macraes Mine,Otago, New Zealand

Tailings from the Macraes Au mine cyanidation process are stored in an impoundment about 0.6 km² and 80 m deep whose pH is maintained near 8 by the neutralizing capacity of the gangue minerals. The tailings are sandy (>50 μm particles), have a hydraulic conductivity of about 10⁻² m/day, and contain 0.1–1.0 wt.% S and 0.1–1.5 wt.% graphitic C from the primary deposit. Concentrations of As in the pore water of the mixed tailings, which are a combination of various tailings types, range from 0.1 to 20 ppm, HCO₃^- is 100 to 200 ppm, and dissolved SO₄ is 100–1700 ppm. The mixed tailings will be stored in this impoundment in perpetuity after mining ceases. Confidence in the long-term pH stability of these tailings can be gained from examination of mineralogically and chemically similar geological analogues in the immediate vicinity. A sequence, typically about 5 m thick, of sands and gravels derived from the Macraes mineralized zone 12–28 ka ago contains rounded detrital sulfide mineral grains which are unoxidized despite their close proximity to the surface and the occasional incursion of oxygenated waters. These sediments have a hydraulic conductivity of about 10⁻⁴ m/day. Saturating water pH is currently 7–8. Sands with 0.2–0.8 wt.% organic C host SO₄-reducing bacteria (SRB), and local cementation by authigenic framboidal pyrite has occurred. SRB were found in water-saturated sediments with decreased hydraulic conductivity and alkaline and anoxic conditions. These bacteria are involved in the formation of authigenic framboidal pyrite, reducing the cycling of dissolved Fe in the sediments. Carbon is not a limiting factor in this process as organic matter is present in the sandstone and ground water contains up to 180 ppm HCO₃^-. Comparison of the 28 ka old sediments with the modern tailings suggests that the chemical behaviour of the two will be similar, possibly with the crystallization of authigenic pyrite in the tailings over the long term. As long as the present slightly anoxic and circumneutral pH environmental conditions are maintained in the mixed tailings impoundment, sulfide decomposition and acidification are unlikely.     

Hydrothermal alteration in metasedimentary rock-hosted orogenic gold deposits,Reefton goldfield,South Island,New Zealand

Orogenic or mesothermal quartz lodes in lower Palaeozoic Greenland Group metasedimentary rocks of the Reefton area have produced 67 tonnes (t) of gold prior to 1951, and recent exploration has identified new gold resources in several deposits, including the largest past producers, Blackwater and Globe-Progress. The metasedimentary rocks consist of alternating sandstone and mudstone beds that were metamorphosed to lower greenschist facies prior to being hydrothermally altered adjacent to the quartz lodes. The sandstones are feldspathic litharenites averaging Q₆₅-F₁₀-R₂₅, with detrital grains of quartz, rock fragments, muscovite, and plagioclase and biotite that were altered to albite and chlorite, respectively, during metamorphism. Accessory minerals are graphite, apatite, zircon, tourmaline and titanite. Hydrothermal alteration of the sandstones has developed a mineral assemblage of K-mica, carbonate (dolomite, ankerite, ferroan magnesite and magnesian siderite), chlorite, pyrite and arsenopyrite. The abundance of hydrothermal chlorite is greater at Blackwater than at the other prospects studied. Hydrothermal alteration associated with the quartz lodes is marked by bleaching, magnesian siderite spots, disseminated arsenopyrite and pyrite and thin carbonate, quartz and sulphide veins. These trends are accompanied by increasing concentrations of S, As and Sb and decreasing Na, and a decrease of Fe and Mg in K-mica. The alkali alteration indices 3K/Al (representing K-mica) and Na/Al (representing albite) generally show antipathetic trends, with 3K/Al increasing near the lodes and Na/Al decreasing. These trends reflect the replacement of albite by K-mica. Carbonate alteration indices CO₂/(Ca + Mg +Fe) and CO₂/[Ca + Mg + Fe -0.5(S + As)] quantify the abundance of hydrothermal carbonates, but they show variable correlation with the lodes. They increase the width of the alteration halo in the hanging wall of the lodes at the Globe-Progress and General Gordon prospects, but the peak values are as far as 150 m from the lodes. By contrast, peak values of the carbonate alteration indices are within 10 and 2 m of the lodes, respectively, at the Merrijigs and Blackwater deposits. Data show that for deposits with wide hydrothermal alteration halos, such as at the Globe-Progress and General Gordon prospects, the use of a suite of geochemical indicators can assist exploration by indicating trends in hydrothermal alteration that provide vectors to mineralisation. They also increase the size of the exploration target. By contrast, the alteration halo of the Blackwater deposit is restricted to within less than 5 m of the quartz lode and, therefore, the geochemical indicators are of more limited assistance to exploration.     

Migration of hydrothermal systems in an evolving collisional orogen, New Zealand

The Pacific–Australian tectonic plate boundary through the South Island of New Zealand consists of the transpressional Southern Alps mountain belt and the transcurrent Marlborough Fault System, both of which have active tectonically driven hydrothermal systems, with topographically driven meteoric incursion and warm springs. The Southern Alps hydrothermal system is relatively diffuse, with little or no fault control, and is channelled through scattered extensional sites beneath the mountains, where gold mineralisation is occurring locally. The hydrothermal activity along the Marlborough Fault System is controlled by the principal faults in well-defined valleys separated by narrow high ridges. Lateral evolution of Marlborough fault strands southwestwards into the Southern Alps has caused diversion of diffuse Southern Alps hydrothermal activity into the structural superimposition zone, where fluid flow is increasingly being controlled by faults. This hydrothermal diversion was accompanied by major topographic reorientation and river drainage reversal in the late Quaternary. Vein swarms now exposed in the remnants of the Southern Alps north of the superimposition zone formed at shallow levels, with some evidence for fluid boiling, from a mixture of meteoric and deep-sourced fluid. These veins, some of which contain gold, are part of an abandoned <1 million-year-old hydrothermal zone beneath the fossil topographic divide of the Southern Alps that has now been dismembered by lateral incursion of the Marlborough fault strands. Observations on this active plate boundary provide some insights into processes that controlled orogenic gold mineralisation in ancient belts, particularly with respect to relationships between hydrothermal fluid flow, structure and topography.     

Structural controls on Tertiary orogenic gold mineralization during initiation of a mountain belt, New Zealand

Two types of structurally controlled hydrothermal mineralization have occurred during folding of fissile schist in southern New Zealand: fold-related mineralization and normal fault-related mineralization. Both types have the same mineralogy and textures, and are dominated by quartz–ankerite veins and silicified breccias with ankeritic alteration. Most mineralized zones are thin (centimetre scale), although host schist is commonly impregnated with ankerite up to 20 m away. Thick (up to 5 m wide) mineralized zones are generally gold-bearing and contain pyrite and arsenopyrite with stibnite pods locally. Some of these auriferous zones have been extensively mined historically despite rugged topography and difficult access. Mineralization occurred during regional tectonic compression in the initial stages of development of the Southern Alps mountain belt at the Pacific–Australian plate boundary in the Miocene. Most of the gold-bearing deposits occur in east to south-east, striking normal faults that cut across mesoscopic folds in a belt that coincides with the southern termination of a regional-scale north trending antiform. Mineralized zones have similar structural control and relative timing to a nearby swarm of Miocene lamprophyre dykes and carbonatites. Limited stable isotopic data (C and O) and trace element geochemistry suggest that there was probably no genetic link between the igneous activity and gold mineralization. However, these two types of fluid flow have been controlled by the same tectonically created crustal plumbing system. This Miocene hydrothermal activity and gold deposition demonstrates that orogenic (mesothermal) mineralization can occur during the inception of an orogenic belt, not just in the latter stages as is commonly believed. These Miocene structures have been preserved in the orogen because the locus of uplift has moved northwards, so the early-formed gold deposits have not yet been structurally overprinted or eroded.     

青藏高原新生代造山型金成矿系统

形成在大洋俯冲过程的造山型金矿已被广泛研究,而对随后的大陆碰撞阶段形成的造山型金矿研究较少。青藏高原是最年轻的大陆碰撞事件的产物,为揭示大洋俯冲-大陆碰撞完整构造演化背景下的造山型金成矿系统的成因提供了难得的研究窗口。研究表明,青藏高原存在三个金成矿带：（1）在大洋俯冲和大陆碰撞初期（60~43Ma）,在正向碰撞带的挤压构造中,沿雅鲁藏布江缝合带形成石英脉型金矿带;（2）在大陆侧向碰撞带的大规模走滑剪切环境中（32~21Ma）,发育受剪切带控制的石英脉型和浸染型矿体为主的金矿带;（3）在中新世印度大陆岩石圈回撤背景下（19~15Ma）,喜马拉雅穹窿带普遍发育与Sb矿化有关的浸染型和细脉型金矿带。矿床矿化-蚀变和成矿流体特征综合表明三个矿带成矿深度具有逐次变浅的系统变化规律。碰撞造山环境造山型金成矿作用发生在峰期变质和退变质之后,脉动式的成矿作用多数和印度-欧亚板块汇聚速率的多期下降具有同步性,和大洋板片断离和大陆板片回撤等地幔扰动事件同期。岩石圈结构控制了流体的运移和成矿位置,深部成矿流体在较厚岩石圈的压力下沿板块边界上涌至岩石圈厚度梯度处就位。石英脉型金矿金属沉淀受到地震泵模式和流体不混溶作用控制,浅成蚀变岩型主要受到水岩反应的控制。三个金矿带黄铁矿δ³⁴S中值大多是0左右,与不同时期的围岩地层无关;成矿流体δ¹⁸O整体上与富集地幔产生流体的氧同位素一致;与成矿有关的黄铁矿的⁴⁰Ar/³⁶Ar和³He/⁴He值表现出明显的地幔来源特征;矿石硫化物PGE特征显示成矿流体具有和岩浆热液不同的地化属性。矿床地球化学特征、金矿化整体滞后于区域进变质并与地幔扰动事件具有同步性均表明青藏高原金矿成矿流体和金属主体来自于地幔。文章进一步为造山型金矿石氢氧同位素的时空变化提供了新的可能解释,始新世金矿δD值降低指示了有超临界流体的加入,始新世至中新世金矿床δ¹⁸O值增高则可以解释为晚期有更多俯冲的大陆物质交代地幔。成矿流体的深来源以及已有造山型金矿的中-浅成矿深度显示青藏高原具有较大的寻找造山型金矿的潜力。

     

Arsenic mobility in mildly alkaline drainage from an orogenic lode gold deposit,Bralorne mine,British Columbia

The historical (1932–1971) Bralorne mine produced over 87 million grams of Au from an archetypal orogenic lode gold deposit in southwest British Columbia. High concentrations of As in mine drainage, however, represent an on-going environmental concern prompting a detailed study of effluent chemistry. The discharge rate at the mine portal was monitored continuously over a fourteen-month period during which effluent samples were collected on a quasi-weekly basis. Water samples were also collected on synoptic surveys of the adit between the portal and the main source of flow in the flooded workings. Total concentrations of As in the mildly alkaline (pH = 8.7) portal drainage average 3034 μg/L whereas at the source they average 5898 μg/L. As emergent waters from the flooded workings flow toward the portal, their dissolved oxygen content and pH increase from 0 to 10 mg/L and from 7.7 to 9, respectively. Near the emergence point, dissolved Fe precipitates rapidly, sorbing both As(III) and As(V). With increasing distance from the emergence point, dissolved As(III) concentrations drop to detection limits through sorption on hydrous ferric oxide and through oxidation to As(V). Concentrations of dissolved As(V), on the other hand, increase and stabilize, reflecting lower sorption at higher pH and the lack of available sorbent. Nonetheless, based on synoptic surveys, approximately 35% of the source As load is sequestered in the adit resulting in As sediment concentrations averaging 8.5 wt%. The remaining average As load of 1.34 kg/d is discharged from the portal. Partitioning of As(V) between dissolved and particulate phases in portal effluent is characterized by a sorption density of 0.37 mol As (mol Fe)⁻¹ and by a distribution coefficient (K_d) of 130 L/g HFO. The relatively high sorption density may reflect co-precipitation of As with Fe oxyhydroxides rather than a purely adsorption-controlled process. Results of this study show that the As self-mitigating capacity of drainage from orogenic lode gold deposits may be poor in high-pH and Fe-limited settings.     

Synmetamorphic carbon mobility and graphite enrichment in metaturbidites as a precursor to orogenic gold mineralisation, Otago Schist, New Zealand

The Macraes orogenic gold deposit is hosted by a graphitic micaceous schist containing auriferous porphyroblastic sulphides. The host rock resembles zones of unmineralised micaceous graphitic pyritic schists, derived from argillaceous protoliths, that occur locally in background pelitic Otago Schist metasediments. This study was aimed at determining the relationship between these similar rock types, and whether the relationship had implications for ore formation. Argillites in the protolith turbidites of the Otago Schist metamorphic belt contain minor amounts of detrital organic matter (<0.1 wt.%) and diagenetic pyrite (<0.3 wt.% S). The detrital organic carbon was mobilised by metamorphic–hydrothermal fluids and redeposited as graphite in low-grade metaturbidites (pumpellyite–actinolite and greenschist facies). This carbon mobility occurred through >50 million years of evolution of the metamorphic belt, from development of sheared argillite in the Jurassic, to postmetamorphic ductile extension in the Cretaceous. Introduced graphite is structurally controlled and occurs with metamorphic muscovite and chlorite as veins and slicken-sided shears, with some veins having >50% noncarbonate carbon. Graphitic foliation seams in low-grade micaceous schist and metamorphic quartz veins contain equant graphite porphyroblasts up to 2 mm across that are composed of crystallographically homogeneous graphite crystals. Graphite reflectance is anisotropic and ranges from ~1% to ~8% (green light). Texturally similar porphyroblastic pyrite has grown in micaceous schist (up to 10 wt.% S), metamorphic quartz veins and associated muscovite-rich shears. These pyritic schists are weakly enriched in arsenic (up to 60 ppm). The low-grade metamorphic mobility and concentration of graphite in micaceous schists is interpreted to be a precursor process that structurally and geochemically prepared parts of the Otago Schist belt for later (more restricted) gold mineralisation. Economic amounts of gold, and associated arsenic, were subsequently introduced to carbonaceous sulphidic schists in the Macraes gold deposit by a separate metamorphic fluid derived from high-grade metaturbidites. Fluid flow at all stages in these processes occurred at metamorphic rates (mm/year), and fluids were broadly in equilibrium with the rocks through which they were passing.