South Dakota, USA
New & Recent International |
Click on image for details.
The Homestake gold deposit is located adjacent to the town of Lead, in the Whitewood mining district of Lawrence County, South Dakota, USA, ~20 km from the western border of the state, and close to the northern margin of the Black Hills. The Homestake mine closed in 2002.
The Homestake mine lies within the Black Hills uplift, which rises some 900 to 1200 m above the surrounding Great Plains, to the east of the main Cordillera. The uplift comprises an up-domed, 85 x 35 km oval shaped block of Proterozoic metamorphics and sediments, forming the core of a larger NNW elongated dome of Palaeozoic rocks covering an area of 150 x 70 km. The uplift resulted from doming of the crust above Laramide-aged intrusions. Archaean and Palaeoproterozoic rocks in the uplift record opening of a siliciclastic-dominated basin and its subsequent closure during the 'collision' between the Archaean Wyoming and Superior Provinces (Dahl et al., 1999). The rocks of the dome are in turn overlain on all sides by the Mesozoic sediments of the Interior Platform, some of which form an encircling ridge to the Proterozoic core of the uplift. The physiography and topography of the immediate area around Homestake have been modified by the emplacement and subsequent partial erosion of Laramide Tertiary granitoids, part of the batholith responsible for the Black Hills uplift (Caddy, et al. 1991; Slaughter, 1968).
Placer gold was first discovered by the Custer expedition in 1874, in the southern Black Hills, near the present town of Custer, some 65 km to the south of Homestake. This led to a gold rush which subsequently resulted in the discovery of further placer gold at Deadwood, approximately 5 km east of Lead and of Homestake. The Homestake lode, the source of this alluvial gold, was discovered by Moses Manuel and Hank Harney on April 9th, 1876 at what are now referred to as the Main Ledge, the site of the present open cut, and also in the Caledonia ore zone (Caddy, et al., 1991; Slaughter, 1968). Mining was continuous from 1976 until the mine closure in early 2002, with the exception of the World War II years in the US, ie. 1942 to 1945, when all gold mines were closed by Government directive. The orebodies are a group of 'pencil like' ore shoots parallel to the plunging fold axes to which they are confined. The orebodies outcrop sporadically within an area of 4 km2, and by 1991 had been followed some 7 km down plunge to a depth of 2440 m below the surface (Caddy, et al., 1991).
Nine main ore zones were worked. These are the elongate rod like "ledges" which follow plunging fold structures. Mining commenced on both the Caledonia ore zone and the Main Ledge in 1876. The Caledonia ceased being exploited in about 1940, while the Main Ledge was still being mined in 1991. The other seven ledges were all being stoped in 1991, commencing in the following years as follows: 9 Ledge - 1936; 7 Ledge - 1939; 11 Ledge - 1955; 17 Ledge - 1961; 19 Ledge - 1965; 13 Ledge - 1968; and 21 Ledge - 1970. Ledges 7, 9 and 11 were all discovered prior to 1958, while 13, 17, 19 and 21 were found during the 1960's, the last, number 21 being detected in 1969 (Caddy, et al., 1991).
A selection of production and reserve figures are as follows:
~1250 t Au (40 Moz) - total production to closure in 2002 (Homestake Mining, 2002),
124.9 Mt @ 8.8 g/t Au = 1101 t Au (Production, 1876 to 1988, Caddy, et al., 1991).
22.8 Mt @ 5.97 g/t Au = 136 t Au (Underground reserve, 1988, Caddy, et al., 1991).
18.5 Mt @ 6.96 g/t Au = 129 t Au (Reserve, 1993, AME, 1995).
100 Mt @ 7 g/t Au = 700 t Au (Potential Resource, Mine visit, 1993).
Historic production from the individual ledges, each of which is/was composed of a large number of orebodies, comprised approximately:
5 Mt @ 7.43 g/t Au ............................. Caledonia Zone, 75 Mt @ 8.37 g/t Au .................................. Main Ledge,
4.5 Mt @ 7.31 g/t Au ....................................... 7 Ledge, 37 Mt @ 9.05 g/t Au ........................................ 9 Ledge,
7.5 Mt @ 7.87 g/t Au ..................................... 11 Ledge, 2 Mt @ 8.71 g/t Au ........................................ 13 Ledge,
7.5 Mt @ 6.94 g/t Au ..................................... 17 Ledge, 2.5 Mt @ 7.03 g/t Au ..................................... 19 Ledge,
9 Mt @ 8.34 g/t Au ......................................... 21 Ledge.
The Homestake mine was owned and operated by the Homestake Mining Company. The orebody was mined underground by 'mechanised cut and fill' and by 'vertical crater retreat' methods, as well as by open cut. In 1993 some 2.45 Mt of ore were treated with a head grade of 5.97 g/t Au. The mill recovery was 96%, with 13.9 t Au being produced for the year. Historically, annual production peaked at a bit over 19.5 t Au in 1965 (Caddy, et al., 1991; AME, 1995). Access in 1987 was by two surface shafts and one internal shaft, with ore transported on more than 400 km of underground track (Brunker, et al., 1987).
Summary (after Morelli et al., 2010) The Palaeoproterozoic sequence of the Black Hills indicate a progression from a lower assemblage of 2560 to 2480 Ma conglomerate, quartzite and iron formation (Dahl et al., 2006) that is unconformably overlain by 2100 to 1974 Ma shallow-water deposits (Redden et al., 1990), including a range of coarse- to fine-grained clastic sedimentary rocks, iron formation, minor carbonate and mafic volcanic flows. Subsequently, this sequence was intruded by metagabbro dykes and/or sills. Deeper water turbiditic and tuffaceous rocks, temporally equivalent to shelf quartzites in the north (~1974 Ma), are found in the central and southern Black Hills, while an extensive unconformable sequence of ~1870 Ma conglomerate, turbidites, and basalts is preserved in the east-central portion of the uplift. These Palaeoproterozoic sequences suggest deposition in a long-lived intracontinental rift basin (Redden et al., 1990; Houston 1992) or in a backarc basin succeeding earlier rifting (pre-2170 Ma) of Archaean crust (Caddey et al., 1991). Palaeoproterozoic and older rocks in the Black Hills uplift were intruded by multiple felsic dykes and sills comprising the âˆ¼1715 Ma "Harney Peak granite" (Redden et al. 1990) exposed in a circular dome in the south-central portion of the uplift.
The succession within the Black Hills uplift may be summarised as follows, from the base:
Basement of granitic and pegmatitic rocks, dated at around 2500 Ma. These form a minor component of the core to the uplift and are only found locally in a few limited windows (Caddy, et al., 1991).
Poorman Formation, subdivided into the,
Yates Unit, 600 to 1200 m thick - fine grained, dark green, massive to faintly banded, and moderately to well lineated hornblende-plagioclase schist, described in the mine as an amphibolite. Ubiquitous 2 mm to 2 cm thick calcite bands and veins are characteristic. The main rock forming minerals in decreasing order of abundance are - hornblende, oligoclase-andesine, calcite, dolomite, ankerite and trace amounts of ilmenite, magnetite, titanite, leucoxene, pyrrhotite and pyrite. It is interpreted to be a metamorphosed sequence of tholeiitic basalt, tuff and epiclastics. Within it there are intercalated bands of graphitic quartz-sericite phyllite, crudely banded grunerite-bearing iron formation, coarse grained amphibole-bearing bands and cherts with minor stilpnomelane, graphite and pyrrhotite. This unit has been traced by drilling as an elongate, north-south trending, south plunging mass which has been followed for 8 km to the north of the mine, and for 6 km down plunge within the mine. Shear zones define the margins of the Yates Unit blocks, largely as a response to the competence contrast with its neighbouring lithologies. It is massive and competent and has influenced the deformational pattern of the adjacent rocks (Caddy, et al., 1991; Bachman, Pers. comm., 1993). The Yates Member has beeen dated at 2012±3 Ma (Hark et al. 2008).
Upper Poorman Formation, 600 to 1500 m thick - the upper unit of the Poorman Formation commences with a;
- Carbonate rich schist - a thin (1 to 30 m thick), moderately foliated to locally massive band which is slightly graphitic, fine grained and light grey to grey in colour. Locally it is absent or may be intercalated with the Yates Unit. It contains alternating carbonate, sericite and biotite rich layers from 3 to 20 cm thick, with moderately abundant quartz (up to 30%) and sericite. The carbonate is predominantly calcite with minor dolomite and ankerite (Caddy, et al., 1991).
- Sericite-carbonate-quartz phyllite - which overlies the basal band and is the dominant rock type of the upper Poorman Formation. It ranges in thickness from a few tens of metres to 170 m, and varies considerably in character, but is generally very fine to fine grained, with thin to medium, moderately to well developed alternating light grey to dark grey banding. The banding reflects the relative graphite content, which varies from 1 to 7%, but averages 4%. Thin, discontinuous, streaks and lenses of pyrrhotite, total around 2% of the rock with local pyrite. Lenses of dark to black phyllite with up to 30% pyrrhotite are developed locally within the Poorman Formation, although these contain no gold. Sericite is the dominant mineral, followed by variable amounts of carbonate (calcite, dolomite or ankerite), quartz and lesser biotite. Garnet is sparsely distributed in the eastern ledges (Caddy, et al., 1991; Slaughter, 1968; Brunker, et al., 1987). Where sighted on the visit it was fine to medium banded, on a 1 to 10 mm spacing, of light and dark bands with folded and lensoid to boudinaged quartz bands. These quartz bands were 2 to 3 cm thick and persisted over lengths of 20 or 30 cm, up to 1 m. The protoliths are believed to have been pelites with chert bands.
- Biotite-quartz-carbonate phyllite (or schist) is commonly found towards the top, close to the overlying Homestake Formation. It is characterised by a high biotite content, with variable carbonates, quartz and lesser sericite, and ranges from a few, up to 30 m in thickness. It also is well foliated, and has thin [1 mm] to medium [1 cm] bands which have well developed brown, grey and dark grey graphitic interbands. Pyrrhotite is present from trace amounts to 2% (Caddy, et al., 1991; Slaughter, 1968).
- Graphitic quartz-sericite phyllite generally lies in the upper portions of the unit immediately below the Homestake Formation, although it is also found interbedded with other Poorman Formation lithologies. It varies in thickness from <1 m to more than 30 m. Overall it is dark grey to black in colour and very fine grained, with moderately well developed banding. In highly sheared areas it occurs as transgressive zones. Sulphides are present in trace amounts, up to 60%, and include pyrrhotite and pyrite. Pyrrhotite may occur as streaks and blebs up to several centimetres long (Caddy, et al., 1991).
Homestake Formation, 0 to 50 m thick, locally up to 125 m in thickened fold hinges - Overall the Homestake Formation is a sequence of iron rich carbonate and/or silicate dominated rocks, depending on the location within the mine, regarded to represent a metamorphosed, carbonate-silicate-sulphide facies iron formation which hosts almost all gold mineralisation at the Homestake deposit. It may be locally massive or thinly banded, marked by abrupt changes. Mixtures of Fe-carbonate and Fe-silicate minerals characterise the formation in the central sections of the mine, while end members of exclusively Fe-carbonate and Fe-silicate are found in the extreme west and east respectively. Thin chert bands are interlayered with these rocks.
Banding is characteristic of the western ledges, but is less common in the eastern ledges where the Homestake Formation is largely massive. In the east the banding has been mostly obliterated by grunerite growth accompanying the more intense metamorphism. The well developed layering in the western ledges is formed by alternating layers of chert, siderite and biotite, locally with thin chlorite, sericite and pyrrhotite bands. This rock is a "siderite schist".
Several different lithologies are mapped within the Homestake Formation, as follows:
Siderite-Dominant Phyllite - in this rock type siderite1 predominates over grunerite, with associated quartz, biotite and locally chlorite, and subordinate ankerite and graphite. Siderite may be present as a very fine grained mixture with quartz or as coarse, post-kinematic porphyroblasts. Siderite dominant phyllites predominate in the upper eastern and middle to upper western parts of the mine. Minor amounts of grunerite, stilpnomelane, muscovite and almandine are found in the phyllite. Layering comprises 1 to 5 mm thick graphite-rich bands, alternating with mm to cm thick graphite poor to graphite free layers which have accompanying biotite and/or chlorite. Locally the banding is disrupted, with diagonal fractures filled with siderite or ankerite and quartz. Chert beds are ubiquitous and variable in thickness (Caddy, et al., 1991).
Grunerite-Dominant Schist - in which grunerite dominates over siderite. It has a wide range of textures and modes, the most typical of which comprises a variably layered, olive-green, fine to medium grained grunerite-biotite-quartz±siderite±chlorite schist. As the metamorphic grade increases, siderite and quartz decline. An end member is a fine to coarse grained, massive to thick banded grunerite-biotite-quartz schist which has associated common, chert. Almandine and chlorite are important associates locally, as are appreciable sulphides (Caddy, et al., 1991).
Chloritic Biotite-Sericite-Quartz-Carbonate Phyllite - which makes up the Transitional Homestake Formation. This unit, which commonly averages 1 m in thickness, is found at the upper and lower contacts of the Homestake formation, forming a transition with the overlying Ellison Formation and the underlying Poorman Formation. It is very fine to fine grained, well foliated and moderately to well layered, with alternating grey, greenish-grey and dark brown layers. The grey layers are due to sericite and carbonate, while the greenish-grey bands are due to chlorite, and sometimes biotite, with traces of siderite and ankerite. In other grey facies the greenish-grey layers are graphitic (Caddy, et al., 1991).
Other Iron Formation rocks, including biotite-quartz-siderite phyllite or schist and cherty chlorite-quartz-siderite phyllite. The former occurs locally as a distinct unit within the Homestake formation in the western limb of the Poorman Anticline, while the other is also found in the Poorman Formation, in the core of the Lead Anticline. The latter may represent a distinct iron formation within the Poorman Formation. They range in thickness from absent to more than 10 m thick (Caddy, et al., 1991).
The iron rich Homestake Formation marks the change in stratigraphy from the underlying pelites of the Poorman Formation and the overlying clastic semi-pelites of the Ellison. The iron formation which characterises the unit is lensoid and not continuous, either locally or regionally. Within the mine there are numerous intervals where the Poorman Formation passes directly into the Ellison with no intervening iron formation. These are interpreted to be structural/boudinage breaks. The contact between the formations is sheared in these breaks and ore, sulphides, chlorite alteration and quartz veining may still be present. According to Bachman (Pers. comm., 1993) the presence of the iron formation on a regional scale can be correlated with the distribution of the underlying Yates Unit of the Poorman Formation. He also suggests, but cannot prove, that the iron formation gradually crosses the stratigraphy, diverging from the Yates Unit up plunge.
Where sighted underground during the visit, the iron formation was composed of foliated ankerite/siderite-silica-grunerite with pegmatoidal segregations and veins of silica and carbonate. The margin of the iron formation is variable, grading from a knife sharp contact to a gradational zone which is up to 3 m wide. The gradational contact results from the variation in content of phyllitic and grunerite/carbonate rich bands.
In outcrop the Homestake Formation appeared, where sighted, as an orange to brown, limonitic, weathered, finely micaceous rock with abundant quartz veining. The vein selvages are represented by a darker colouration. The veins are folded and anastomosing. Folding is evident in outcrop, reflecting the mine scale fold plunges. The adjacent Poorman Formation is a darker brown. In closer proximity to the Yates Unit of the Poorman Formation, the outcrop of the Homestake Formation is much less deformed, apparently buttressed by the amphibolites of the former unit (Pers. observ., 1993).
Ellison Formation, up to 400 m thick in the mine area, but structurally thickened locally to 1500 m. This unit is interpreted to mark regional uplift and a transition from deep-water deposition of turbidites (Poorman Formation) and iron formation (Homestake Formation) into a shallow water environment. It is a well banded to massive phyllite to quartz-mica schist, with interbedded impure quartzite and minor amphibolite. Zircons from a meta-tuff unit at the base of the formation, directly above the Homestake Formation, yielded a U-Pb age of 1974±8 Ma (Redden et al., 1990). The principal lithologies are as follows:
Quartzite is fine grained, grey to dark grey, massive to finely banded, and moderately well sorted with well rounded grains. Most are impure with subordinate sericite, biotite and plagioclase, accompanied by accessory microcline, ankerite, titanite, tourmaline, ilmenite, rutile, xenotime and/or monazite, pyrite and pyrrhotite. In cross section the quartzites are lenticular bodies that may be 0.3 to 4 m thick, while composite beds are several tens of metres thick. Due to the lack of layered silicates no foliation is generally observed (Caddy, et al., 1991). In some parts of the district quartzites are absent from the central 1500 m sections of the formation, with a lower quartzite and phyllite unit which may be up to 1500 m thick, and an upper 1200 to 1500 m thick member of similar composition (Slaughter, 1986).
Quartz-mica schist is mineralogically similar to the quartzite, but has more mica and garnet, particularly within the garnet metamorphic zone. It is moderately well foliated, although layering is only poorly laminated to absent. The colour varies from light grey, to tan to pale brown. Individual schist bands or lenses are more persistent than the quartzite layers and are locally interbedded with quartzite and mica-dominated phyllites (Caddy, et al., 1991).
Sericite-quartz phyllite is light grey to grey, well foliated and very fine to fine grained with 1 to 5 cm thick layers. It is generally discontinuous along strike, but may be up to 30 m thick, commonly being intercalated with quartzite. Sericite accounts for 35 to 70% of the rock with subordinate quartz, ankerite and biotite and trace amounts of garnet, graphite, tourmaline and chlorite (Caddy, et al., 1991).
Biotite-quartz phyllite is also very fine to fine grained, is thick bedded to massive and contains 1 to 3 mm thick local graphite layers. It also is discontinuous, although it may attain thicknesses of as much as 100 m. Sericite and tourmaline are locally present, while 1 to 4 mm pink to red almandine to almandine-spessartine garnet ranges up to 15% of the rock. Carbonates are generally subordinate, but may be locally significant. Pyrite and pyrrhotite are generally absent or only present in trace amounts (Caddy, et al., 1991).
Amphibolite is present as both discordant and concordant, fine to coarse grained bodies, as well as fine grained concordant masses intercalated with the other lithologies of the Ellison Formation. These may attain thicknesses of up to 65 m. The coarse grained, discordant bodies are interpreted to be of gabbroic origin, while the concordant, fine grained varieties resemble the mafic rocks of the Poorman Formation described above, and may be of volcanic origin (Caddy, et al., 1991).
Where sighted in outcrop, the Ellison Formation is a grey sericitic quartzite, which is partly sheared, but is overall massive. It contained scattered veins like those seen in the Homestake Formation, and included well foliated pelitic/phyllitic interbands. Steeply plunging folds were evident (Pers. observ., 1993).
North-western Formation, 0 to 1300 m thick - composed wholly of biotite-quartz-sericite-garnet schist and phyllite, with a little slate. Most of the formation is structure-less with no bedding or laminations. Tourmaline and sphene are usually present (Slaughter, 1968; Caddy, et al., 1991).
Flag Rock Formation, 1600 m thick - interbedded biotite-sericite schist, graphitic phyllite, iron formation and hornblende-plagioclase schist. It is generally heterogeneous in composition and is difficult to describe. The mostly abundant lithology is a light grey sericitic phyllite or schist, which may also be greenish-grey. Some of the phyllites have fine laminations and banding due in part to graphite. There is a considerable amount of soft sooty-black schist or phyllite with abundant pyrite. These are composed of graphite and quartz alone. Quartzite is also conspicuous, with a streaky texture due to the presence of unevenly distributed graphite. Two or three bands of sideroplesite schist (MgCO3.FeCO3), which are present as cummingtonite schists in the garnet zone, are found in the lower half of the formation. Dense siliceous jasper material known as the Iron Dyke is present as an alteration product on hill tops near Lead (Slaughter, 1968; Caddy, et al., 1991).
Grizzly Formation, 1000 m thick - mainly meta-greywackes with interbedded sericite-biotite phyllites to schist. It is mainly fine grained and grey to dark grey in colour with no distinctive character and almost no bedding. Quartz and muscovite are the dominant minerals, with biotite and some graphite (Slaughter, 1968; Caddy, et al., 1991).
Cambrian to Ordovician
Deadwood Formation, 0 to 150 m thick - shale, dolomite, limestone, flat pebble conglomerate, sandstone and conglomerate. There is a 0 to 15 m, but seldom more than 3 m thick, quartz lag-pebble conglomerate at the base of the unit, on the unconformable contact with the underlying Proterozoic. The conglomerate contains sub-angular 0.5 to 2 cm clasts of white quartz in a red ferruginous sandy matrix. The conglomerate grades upwards into a 0 to 9 m thick, light brown quartzite. This in turn is overlain by a 0.65 to 5 m thick brown sandy dolomite with abundant thin partings of green shale in its upper sections. The remainder of the unit is composed of a thin-bedded green and grey, glauconitic shale with some narrow beds of limestone and dolomite, and a few bands of intraformational limestone conglomerate (Slaughter, 1968; Caddy, et al., 1991; Pers. Observ., 1993).
Eocene to Palaeocene
Intrusives - rhyolite, trachyte and phonolite sills and dykes ranging in thickness from near zero to 150 m. These have been dated at 60 to 50 Ma. A number of Tertiary stocks and laccoliths are found in the Lead district, mainly of granitic and syenitic composition. They are most abundant to the west and north-west of Homestake. The most significant of these is the 1.2 x 2.5 km Cutting Stock which is exposed near the centre of the Homestake Window, but is in turn intruded by numerous dykes and sills. Individual dykes and sills are from 0.25 to 6 m thick, and define a zone that is 90 to 120 m wide, extending over a distance of around 3 km (Slaughter, 1968; Caddy, et al., 1991).
Holocene to Oligocene
Gravel and clay deposits, 0 to 70 m thick - containing large and small, partly rounded boulders, mainly of Tertiary porphyry, with some Cambrian and Proterozoic quartzite and vein quartz. In places the gravel contains clay beds up to a metre thick, with associated Oligocene fossils (Slaughter, 1968; Caddy, et al., 1991).
While the 2500 Ma Archaean granitic rocks make up a minor part of the Black Hills, the majority is composed of metamorphosed Palaeoproterozoic sediments and interpreted volcanics which range in age from 2200 to 1870 Ma. The Archaean and Proterozoic rocks were metamorphosed and deformed prior to the emplacement of the 1720 Ma Harneys Peak granite, the youngest Precambrian rock of the Black Hills (Caddy, et al., 1991). The emplacement of this body is correlated with deformational event D3 (Bachman, 1993, Pers. comm.).
The Precambrian of the region was uplifted and eroded by 530 Ma, and during periods of relative tectonic quiescence, Palaeozoic and Mesozoic sediments were deposited. The Black Hills were uplifted again during the Laramide event at 65 to 60 Ma to their present domal form, followed by the emplacement of a WNW trending belt of 60 to 50 Ma alkali-calcic and alkalic stocks, laccoliths, dykes and sills (Caddy, et al., 1991).
The orebody area was fortuitously exposed through the overlying Cambrian sediments, with little modern weathering (Bachman & Terry, 1993, Pers. comm.).
The stratigraphy of the Black Hills window, in particular the Poorman, Homestake and Ellison Formations, is said to be of the same age as that of the Mesabi Range iron province much further to the east (Bachman, 1993, Pers. comm.)
Metamorphism & Structure
Structural and metamorphic studies at the Homestake mine indicate that regional metamorphism was syn-deformational and characterised by a dominant north to north-west trending regional foliation in the Lead area. This fabric has been tentatively dated at 1840 Ma. Contact metamorphism related to the 1720 Ma Harneys Peak Granite over prints this foliation in the southern Black Hills (Caddy, et al., 1991).
Metamorphism - The rocks within the Homestake Mine area have been subjected to an initial prograde metamorphism and are interpreted to be of upper greenschist to lower amphibolite facies. In the mine area there is a gradational change over an interval that averages 800 m in width, from upper greenschist to the west, to amphibolite in the north-east. This transition zone trends at approximately 340° and dips near vertically. The transition zone is defined mineralogically by the gradation from,
and coexistence of fine grained siderite and grunerite1 in the Homestake Formation. Typical upper greenschist assemblages in the iron formations include siderite-quartz-biotite-ankerite-ferroan clinochlore. The end member prograde assemblage found in the deepest eastern ledges comprises grunerite-quartz-biotite-almandine garnet (Caddy, et al., 1991). Grunerite begins to form at the expense of siderite at about the garnet isograd, as described below. Garnet forms at the expense of biotite and quartz at a higher grade. More grunerite is formed at the expense of quartz, garnet and biotite at a still higher grade. Towards the eastern margin of the transition zone siderite and quartz both decrease (Slaughter, 1968).
Within the mafic Yates Unit of the Poorman Formation the transition zone is characterised by the coexistence of lower amphibolite facies magnesian hornblende and plagioclase (andesine), and the upper greenschist assemblage of actinolitic hornblende2, albite-oligoclase, carbonates and epidote group minerals. The phyllites of the pelitic and semi-pelitic facies are also metamorphosed, although the presence of subsequent graphite and carbonates in the phyllites make it difficult to determine their grade. In general the phyllites were metamorphosed to create almandine and biotite during the main prograde period. Remobilised graphite which is concentrated in structural traps, is a feature commonly recognised in all rocks, particularly in the "western ledges". This remobilisation is interpreted to have taken place late during prograde metamorphism and continued well into the subsequent retrograde event described below (Caddy, et al., 1991).
A retrograde metamorphic event is also recorded in the mine area, largely manifested as a biotite halo developed within and peripheral to the iron formation of the Homestake Formation. The biotite is accompanied by siderite, chlorite and ankerite. The biotite is taken to be the product of magnesian±iron and/or potassium metasomatism, and replaces sericite in phyllites. The carbonates and chlorite are interpreted to be the result of carbonatisation, hydration ± potassium metasomatism. This retrograde alteration which is localised along the margins of the iron formation has been focused primarily along rheologically contrasting rocks, such as the iron formation and enclosing pelitic rocks where ductile shears have been preferentially developed. This phase overprints the prograde upper greenschist to lower amphibolite assemblages of the mine area and is interpreted, on the basis of cross-cutting relationships to be earlier and distinct from the emplacement of gold which was associated with a retrogressive hydrothermal alteration (Caddy, et al., 1991).
A number of isograds have been mapped at the surface and extrapolated underground within the Homestake Formation. The garnet isograd is based on the first appearance of coarse (>3 mm) garnet, usually almandine, within both the Homestake Formation and phyllites of the other units. This isograd generally coincides with the western margin of the transition zone (Caddy, et al., 1991), and closely with the grunerite isograd (Slaughter, 1968).
The transition zone from is structurally bounded to the east and west by early ductile shear zones and forms a band of attenuation which encloses a series of porphyry dykes obliquely cutting the host sequence. It is contained within a structural domain characterised by steep upright folding. It also represents a boundary between two structurally distinct domains, as described below (Caddy, et al., 1991).
The Homestake Formation exhibits a volume reduction during the transition from upper greenschist to lower amphibolite metamorphism. This is believed to be due to both devolatolisation of the iron formation during metamorphism, and to the varied deformational strain experienced by the different composition rocks. In the 9 ledge a 69% volume decrease is observed between the upper greenschist and upper amphibolite rocks over a vertical interval of 1500 m (Caddy, et al., 1991).
Structure - The Palaeo- to Mesoproterozoic structural geology of the Homestake mine area was apparently characterised by two separate stages, termed the "older" and "younger", or the "D1a" and "D1b" events. These two stages have been further subdivided as shown below. All of the stages of D1a are interpreted to have taken place during a single regional thermal prograde metamorphic event that was initiated during Early D1a, reached its peak in Late D1a, and possibly continued into Latest D1a. This regional prograde metamorphism is generally dated at 1840 Ma. The subsequent retrograde metamorphic phase was apparently short lived and has been placed in Latest D1a. The gold mineralisation is believed to have accompanied a retrogressive hydrothermal event during the Middle D1b phase (Caddy, et al., 1991).
The phases of the D1a and D1b stages may be summarised as follows (Caddy, et al., 1991; Bachman & Terry, Pers. comm., 1993):
Early D1a - is associated with regional transpressional shearing and folding, and is believed to be related to major transcurrent movement along the Trans-Hudson orogen which separates the Wyoming Archaean Craton in the west from the Superior Province to the east. The regional deformational pattern in the orogen is represented within the Black Hills by a series of anastomosing ductile shear zones that truncate and post date nappe style folding of previously flat lying sediments. The regional ductile shearing coincided with the initial stages of prograde metamorphism and transposed the Palaeoproterozoic rocks to their current steep dipping attitude.
During this deformation, regional competence variations, such as the lensoid block of competent Yates Unit mafic rocks, played an important role, acting as strain buttresses to the shearing field. In the Homestake mine area the "iron formations" of the Homestake Formation are believed to have been in a strain shadow of the Yates Unit, developing structures which were effectively shear induced isoclinal drag folds. These folds indicate the influence of sinistral strike-slip shearing. The Homestake Formation and Yates Unit converge down plunge within the mine area.
Middle D1a - a flattening event with no accompanying lineation and no associated folding. Flattening is parallel to the D1a fabric and is expressed by flattened crystals.
Late D1a - resulted in sheath folding with an axial fabric and a stretching lineation. The plunge of the sheath folding varies, but averages around 45°SE with a trend of 145°. A penetrative foliation and distinctive stretching lineation developed in conjunction with the sheath folding. Lineations are expressed as elongation of hornblende prisms in the mafics of the Yates Unit, and as rodded quartz and pyrrhotite in mica schists. This folding particularly affect, with the lineation indicating the direction of tectonic transport. The fabric associated with this sheath folding generally over prints and destroys flattening textures and transposes Early D1a folding. The sheath folds are in turn over-printed by Latest D1a ductile shearing and probably correspond to the time of peak metamorphism.
Late D1a deformation was accompanied by the main prograde metamorphism phase, resulting in the production of grunerite and garnet in the Homestake Formation, garnet (almandine) in the Ellison and Poorman Formations and hornblende and andesine in the Yates Unit of the Poorman Formation.
Latest D1a - resulted in deformation dominated by ductile shearing with a planar mylonitic fabric. Ductile shear zones produced by this phase within the mine area are up to 1000 m wide and are laterally extensive. Where most intense this shearing produced a planar mylonitic fabric with an average foliation trending at 162° (342°) and dipping at 52°NE.
Latest D1a shears formed on bedding planes and earlier foliations to create a complex series of anastomosing shears at various scales from the very small up to several metres across. Variations in rock competency also influenced the shear set geometry. These shears are characterised by disrupted sheath fold hinges, well developed flow fabric, drag folds and high temperature mylonites. Movement is believed to have been predominantly strike-slip, with minor dip-slip which produced reclined and semi-reclined folds that plunge to the north-east and south-east respectively. These folds which are effectively drag folds formed by the ductile shearing, disrupt and transpose Early D1a folds and Late D1a sheath folds. This shearing coincides with the transition from prograde to retrograde metamorphism and the oldest recognised set of quartz veining, type I quartz veining. The last phase of the prograde metamorphism resulted in the remobilisation and concentration of graphite in structural traps. The ensuing retrograde metamorphism produced biotite which replaced sericite in phyllites in shears bounding the iron formation, and within the iron formation. In addition, calcite and/or dolomite, accompanied by chlorite, were produced, while biotite was formed in the sheared Yates Unit of the Poorman Formation.
Early D1b - produced upright folding with a vertical axial plane foliation trending on average at 160°, but ranging from 120° to 180°, and plunging at around 35°SE. This folding and its associated foliation is substantially weaker than previous phases. Domains of this folding are restricted to selected locales in the mine, where they refold most of the earlier structures described above, but do not substantially modify the Late D1a sheath folds or Latest D1a ductile shears.
Middle D1b - a ductile-brittle shearing stage which was apparently a continuation of the Early D1b deformation, but with strain conditions changing from upright folding to reverse shearing. This shearing tends to dip at 45°NE and follows bedding planes and earlier foliation surfaces, and consists in part of reactivated older shear zones. Asymmetric drag folds with near horizontal axes are developed within and adjacent to the shears indicating north-east side up reverse shearing. The Middle D1b shearing trends from 120 to 180°.
The strike and dip of Middle D1b shears were controlled by the attitude of existing foliations, which were in turn influenced by the attitude of the more competent grunerite dominant Homestake formation and the quartzites of the Ellison Formation. As with the Latest D1a ductile shearing, the Middle D1b shears are anastomosing through the mine.
This stage apparently coincides with the main ore stage mineralisation which was accompanied by a retrogressive hydrothermal event. Mineralisation and alteration may however have been initiated during Early D1b. This stage of mineralisation and hydrothermal alteration was accompanied by a second set of quartz veining, type II quartz veining.
Selected segments of the Middle D1b shears were dilated. Many of these dilated shears developed synchronously with the ore, and many are spatially coincident with the orebodies. As shearing continued, parts of some orebodies were drag folded, sheared and boudinaged
Late D1b - produced post ore brittle shearing resulting in a crenulation/kink fabric. This fabric is apparently both distinctly post-metamorphic and post ore formation, and developed in an environment of semi-brittle strain and lower temperature. The kink folds and kink shear fabrics take the form of an earlier, steeply dipping north-east to north-west set and a shallow dipping, later north-south development. The kink-shears have only produced minor movement of the order of a few metres on flat shear surfaces. These structures are interpreted to have developed in association with the formation of the Crook Mountain Granite which is only found in drill core to the north-east of the mine. They are most prevalent as narrow restricted domains in the eastern mine area. A third set of quartz veins, type III quartz veining, was emplaced during this phase of deformation.
These deformation stages are from Caddy, et al. (1991). Subsequent authors have defined different stages. Most recently Morelli et al. (2010), have equated their:
D1 - produced a gently dipping, layer-parallel foliation (S1); locally preserved relicts of S1 observed in thin section are consistently parallel to bedding. This fabric is probably equivalent to that recognized in the southern Black Hills that was dated at ca. 1785 to 1775 Ma, related to north-directed accretion of the Yavapai island arc (Dahl et al., 2005), and has been equated with D1a early.
D2 - produced upright, tight to isoclinal F2 folds that plunge to the SSE (F2 trends 146°) with an intense, steeply dipping axial plane foliation (S2) that forms the dominant cleavage in the area. This event has been attributed to continental collision between the Wyoming and Superior Cratons starting at around 1760 to 1750 Ma (Dahl et al., 1999) and resulted in the development of the north to NW-trending, upright, isoclinal folds present throughout the Black Hills and peak metamorphic mineral assemblages in the Homestake area. In particular, this collision is responsible for the large-scale geometry of two large anticlinoria ("Lead" and "Poorman" anticlinoria); equated with D1a early.
D3 - produced a heterogeneously developed sub-horizontal foliation (S3) which rotates S2 into a sub-parallel orientation where most prominent. The effect of D3 has been dramatically obscured by intense folding caused by a subsequent (D4) event; equated with D1a middle.
D4 - produced tight folds in the Homestake Formation that refold F2 folds about moderate to gently south to SSE plunging axes and that dominate map and cross-sectional patterns. The main stage of gold mineralization at Homestake is interpreted to have been synchronous with this deformational event; equated with D1a late, D1b early and D1b middle.
D5 - produced flat lying, mm-scale zones of shearing with horizontal axial planes that consistently display a top to the west shear sense. The S4 fabric initially formed vertically but now dips steeply to the east due to the top to the west rotational effects of D5; equated with D1b late.
Empirically there are two dominant shear directions within the mine area which influence the strike of both foliation and compositional boundaries. These are related to the structural history described above, and comprise those trending at 145°, and those trending at 155° to 175°. The latter shears are the favoured set for ore. Ore is preferentially found where the iron formation occurs on the margins of these shears (Bachman & Terry, Pers. comm., 1993).
The mine area is located in a major SSE plunging Early D1a composite antiformal structure, the Lead-Poorman Anticlinorium, as traced by the Homestake Formation. The subsequent phases of deformation considerably modified this structure forming, from east to west, the Caledonia Syncline, Pierce Anticline, DeSmet Syncline, Independence Anticline, Lead Syncline and Poorman Anticline, all within the nose of the earlier structure. These later, very irregular folds superimposed on the Lead-Poorman Antiform are largely the product of shearing (Caddy, et al., 1991).
The nose zone of the Early D1a antiform, the Lead-Poorman Anticlinorium, can be divided into a number of domains. The three zones are distributed as follows (Caddy, et al., 1991; Bachman & Terry, Pers. comm., 1993):
i). To the north-east on this level, on the eastern limb of the Lead Anticline, deformation is largely limited to Early D1a folding, Middle D1a flattening fabric and a moderate Late D1a sheath folding overprint. The Lead Anticline plunges gently to the south-east at 20 to 30° at surface, steepening to an average of 45°SE at depth.
ii). The first domain is bounded to the west by a zone characterised by Late D1a sheath folding with a strong Early D1b over-print.
iii). A third zone, or pair of zones, over-prints this pattern to the west and along the contact between the first two. This latter domain is characterised by Latest D1a ductile shearing with a weak D1b over-print. To the west, structures trend north-south, dip steeply to the east, and plunge steeply at 60° to 70° to the north-east.
The second domain of more intense Middle D1a sheath folding is closely coincident with the zone of attenuation which marks the transition zone from the lower amphibolite metamorphics in the north-east, to the upper greenschist rocks to the west, as described previously. This zone of attenuation also defines a major regional domain boundary across which the structural character of the Black Hills changes from shallow south-east plunging structures to steeply east plunging, from north-east to south-west respectively. This zone of attenuation has also been the focus of Middle D1b shearing which was apparently contemporaneous with the mineralisation. It may be mapped on a regional scale, as can other shear zones, but is differentiated on the basis of scale (Bachman 1993, Pers. comm.).
As described below, the orebodies are concentrated in the nose zones, or ledges, of the Late and Latest D1a folds, but only attain ore grade where these fall within the zone of attenuation within Middle D1b shear zones.
It has also been suggested that while these shear zones are present on a semi-regional scale, the combination of shearing and the buttressing of the more competent lensoid mass of the Yates Unit has produced an unusual setting which has been exploited by the mineralisation.
Mineralisation & Alteration
Summary - (after Morelli et al., 2010) Gold mineralisation in the Homestake deposit is associated with shear zone-hosted quartz veining and is concentrated within relatively undeformed ore bodies that exhibit an overall tabular morphology. Mineralisation was emplaced into dilated segments of shear zones during a period of "retrogressive hydrothermal alteration" that overprints prograde metamorphic fabrics (Caddey et al., 1991). While ore bodies are strongly concentrated within smaller synclinal structures (the odd-numbered ore "ledges") on anticlinorium limbs, the adjacent anticlinal structures (even-numbered ore ledges) are nearly devoid of gold mineralisation. Gold is cogenetic with both pyrrhotite and arsenopyrite (Caddey et al., 1991) and is typically concentrated within alteration zones containing any or all of the minerals chlorite, biotite, sericite, carbonate and quartz.
Mineralisation at Homestake is contained within "ledges". A ledge is a plunging isoclinal fold closure, usually a synform, of Homestake Formation. Virtually all of the ore at Homestake is hosted within, and on the margins of, the Homestake Formation "iron formation". However, only 5% of the total Homestake Formation iron formation is mineralised. Where economically mineralised with gold, a ledge is termed an "ore ledge". Nine ore ledges have produced gold at Homestake, eight were still in production in 1993. These are the Caledonia, Main, 7, 9, 11, 13, 17, 19 and 21 Ledges. Odd numbered ledges are found in synformal folds made up of a complex series of associated parasitic antiforms and synforms, while even numbers denote antiformal ledges which are generally less complex. To date most even ledges have been barren, although number 6, developed on the Independence anticline has produced minor gold. Drilling in the lower levels of the mine in the number 15 Ledge position intersected 7 m @ 24.4 g/t Au, suggesting a further ore ledge (Caddy, et al., 1991; Brunker, et al., 1987).
The mine has been divided into two sections, namely the Eastern and the Western Ledges, represented respectively by the Caledonian, main, 7, 9 and 13 ledges, and by the 17, 19 and 21 ledges. The orientation of each ledge is determined by the attitude of the folded and sheared iron formation of the Homestake Formation at any point. The attitude varies considerably throughout the mine, depending on the deformational domain in which the ledge lies (Caddy, et al., 1991).
The Eastern ore ledges trend, on average at 145° and plunge at 40°SE. They generally parallel the Late D1a sheath folds and lie predominantly within the Late D1a sheath fold domain. Notable exceptions are sections, but not all of the Main and Caledonian ledges (Caddy, et al., 1991).
The Western ore ledges trend on average 80° and plunge at an angle that varies between 60°NE and 40°SE. They are parallel to the Latest D1a reclined folds and lie predominantly within the Latest D1a shear domain (Caddy, et al., 1991).
Ore ledges are composed of three geometric segments along their long axis, namely the "core" in the central section, "upper tail" up plunge and "lower tail" down plunge. The core contains the largest and most continuous orebodies, generally comprising 50% of the total ore in the ledge. The vertical height
occupied by the nine core zones ranges from 200 m (in the 13 Ledge), to 800 m (in the Main Ledge). The length of the ore ledges varies, with the 9 Ledge having been traced down, along its plunge, for a distance of 5 km (Caddy, et al., 1991). Typically individual orebodies within the core of ore ledges measure 500 x 150 x 15 m (Brunker, et al., 1987).
The mine level on which the cumulative plan area covered by the orebodies of a ledge is the maximum is known as the "centroid". The number and size of orebodies in a ledge decreases away from the centroid, both up and down plunge. The ore distribution in the tails is generally asymmetric, with the lower tail [down-plunge] usually being twice the length of the upper tail [up-plunge] (Caddy, et al., 1991).
The ore centroids, the theoretical zones of optimal gold deposition, all lie on a line trending at around 340° when plotted on plan projection, and plunge at an average of 26°SE. This parallels the Latest D1a shears, lies within the metamorphic transition zone and as such the zone of attenuation, and falls within the zone of focused Middle D1b shearing (Caddy, et al., 1991).
Average geological grades remain fairly constant throughout the orebodies of a ledge, decreasing slightly at the transition from the core to the tails. For example in the Main Ledge, the geologic grade is 1.5 g/t Au less in the lower tail compared to the core over a vertical interval of 2100 m. The economic mineralisation of the tails grades into barren Homestake Formation with <20 ppb Au. Observable physical changes from ore to barren Homestake Formation include a marked decrease in sulphides, chlorite and quartz veins, all of which are intimately associated with ore. Beyond a few hundred metres from the poorly defined limits of the upper and lower tails the trace geochemical signature of Au, Ag and As are near detection limits (Caddy, et al., 1991).
Drilling has not indicated any obvious roots to the ore along plunge, laterally or vertically, although quartz veining can be traced down plunge into the Poorman Formation (Bachman & Terry, Pers. comm., 1993). The Homestake Formation was inspected 200 to 600 m from known ore during the mine visit. At this locality there were no sulphides, although abundant chlorite and quartz veining was evident. The amount of veining was several times stronger in the iron formation compared to the adjacent phyllites.
The mineralogy and physical character of the ore within the ledges varies along the direction of plunge. The host Homestake Formation in the upper tails is siderite dominant reflecting the upper greenschist metamorphism, grading to mixed siderite-grunerite in the core and then to predominantly grunerite in the lower amphibolite metamorphic grade crossed by the lower tail (Caddy, et al., 1991).
Economic gold within ore ledges is contained within tabular to pipe-like, relatively un-deformed orebodies which occur as segregations of pyrrhotite, arsenopyrite, minor pyrite and native gold. Pyrite diminishes below the 800 level (ie. 245 m below the shaft collar). The gangue matrix mineralogy includes quartz+chlorite+siderite+biotite±garnet. The sulphide content of the total ore varies from 5 to 30%, averaging 8% overall (Caddy, et al., 1991).
Orebodies generally occur at or near the contact between the Homestake Formation and the Ellison or Poorman Formations. However, where ore is entirely within the Homestake Formation, the host is generally a siderite- or grunerite-dominant biotite-chlorite-quartz schist. Orebodies do not conform strictly to the Homestake Formation "stratigraphy", but rake across lithological banding and the pre-Middle D1b folds and foliations. Localisation of ore near formational boundaries is observed throughout the mine, and is interpreted to reflect the development of the ore-stage, Middle D1b ductile-brittle shears along rheological boundaries (Caddy, et al., 1991).
The number, size and relative internal continuity of orebodies is very variable, depending on the position of the orebody within a ledge, and on the intensity of Middle D1b shearing. Individual orebodies are discontinuous along plunge. The 9 Ledge for example contains 196 individual orebodies, which range in size from 2000 t, to more than 1 mt of ore (Caddy, et al., 1991).
In summary, the ore is restricted to "ledges" (ie. tight synformal fold closures), and is hosted predominantly by the Homestake Formation siderite-grunerite "iron formation". The pencil shaped ledges are mineralised where they rake through the broad tabular zone defined by the siderite-grunerite transition. This transition represents the overlap across a major domain boundary from greenschist to amphibole facies metamorphism. It also corresponds to the corridor of superimposed Late D1a and Middle D1b shearing, and is a zone of attenuation. The "core" and "centroid" of each ledge is localised in the centre of this zone, while the upper and lower "tails" are where the ledge approaches the margins of the zone of attenuation. Outside of this corridor the ledges contain little or no economic gold mineralisation. Each ledge contains tens to hundreds of individual elongate orebodies which are larger and more densely distributed in the "core" of the ledge.
Ore types - Several styles of ore are recognised, as follows:
Shear Ore - this style of mineralisation occurs within and adjacent to Middle D1b shears which tended to dilate preferentially in the Homestake Formation. These shears show a reverse, north-east side up sense of movement, with vertical to moderate north-east dips. The type of mineralisation that occurs within these Middle D1b shears contains type II quartz veins with an assemblage of chlorite, siderite, arsenopyrite and minor pyrrhotite. These minerals also form the selvages of the quartz veins. Extensive replacement of biotite by chlorite is a feature of the Middle Dlb shears. Quartz veining is not necessarily present in these shear zones, and good ore may occur locally in association with abundant chlorite, arsenopyrite and minor pyrrhotite alone. Gold grades in this style of ore are described as being "excellent" with values in the tens of g/t (Caddy, et al., 1991).
Replacement Ore - is regarded as representing leakage off master Middle D1b shears via Early Dlb axial plane foliation and earlier foliations. Disseminated pyrrhotite replaced certain siderite rich beds within the Homestake Formation, and forms continuous layers along foliation surfaces. Pyrrhotite within replacement style ore zones occurs in minor dilatant zones along quartz, siderite, ankerite and chlorite mineral grain boundaries. This replacement is interpreted to have resulted from an increase in permeability developed on existing foliation surfaces during Middle D1b shearing. In addition to the pyrrhotite replacement of siderite bands, moderate chlorite and siderite alteration occurred throughout the replacement ore, producing scattered coarse grains and aggregates. This siderite alteration is much coarser and younger than the fine matrix siderite found in the barren Homestake Formation. Replacement ore is everywhere associated with gold bearing and chlorite altered D1b shears (Caddy, et al., 1991).
The mineralisation, both shear and replacement styles, has not generally been subsequently deformed, except by local Late D1b kinking and brittle shearing (Caddy, et al., 1991).
The distribution of gold within an ore ledge is centred on high grade zones containing 15 g/t Au that form the nucleus of individual orebodies. Lower grade material, generally of <5 g/t Au, lies lateral to, and along strike from, the higher grade zones. The size of the high grade zones is dependent upon the intensity of Middle D1b shearing, which is in turn, apparently, controlled by it's position within the Homestake Formation. The largest high grade zones, by definition, are found in the "core" areas of the ledge as described above. Ore does also extend into the adjacent Poorman and Ellison Formations although this only represents a minor percentage of the gold produced. Within the two "tail" zones of each ledge the gold distribution is more sporadic and discontinuous. Small zones of sub-economic gold-bearing rock are found within the upper and lower tail zones throughout the mine. These may contain grades of >15 g/t Au over widths of <1 m, or lower grade material containing 1.5 to 5 g/t Au that is several metres wide. The frequency of occurrence of such sub-economic zones decreases with increasing distance, both up- and down-plunge from the core of a ledge (Caddy, et al., 1991).
As stated in the "Geology" section, the Homestake Formation iron formation may be locally absent due to boudinage structures. In such intervals the contact between the Poorman and Ellison Formations is sheared and sulphides, gold, chlorite and quartz veins may still persist (Observation, 1993, Bachman & Terry, Pers. comm., 1993).
In 1991 the cut-off was a 5 m width assaying <4.7 g/t Au (Caddy, et al., 1991).
Mineralogy - The orebodies at Homestake are sulphide rich, dominated overall by pyrrhotite. The other sulphides include arsenopyrite, with subordinate pyrite and chalcopyrite. The ratio of the two major sulphides, namely pyrrhotite:arsenopyrite, varies from 1:2 to 10:1. In general this ratio increases from the eastern to the western ore ledges, and from the upper to the lower tails. The gangue of the orebodies contain variable amounts of carbonate and silicate gangue, including, in decreasing order of abundance, chlorite, siderite, grunerite, quartz, biotite, ±garnet, with minor ankerite, muscovite and albite, in addition to graphite. Magnetite, hematite, ilmenite, galena, microscopic sphalerite and visible gold are locally present (Caddy, et al., 1991).
In 1950 the total sulphides of the ore milled averaged between 7 and 8%, in the proportions of about 50% pyrrhotite, 40% pyrite and 10% arsenopyrite. The pyrite content has decreased with time, while the percentage pyrrhotite has increased (Slaughter, 1968).
The occurrence of the principle minerals is as follows:
Pyrrhotite - is present as both the magnetic and non-magnetic varieties, the former predominates. It is found in three textural forms, namely (Caddy, et al., 1991):
i). Streaks - are common in the iron formation of the Homestake Formation and in phyllites of the adjacent Ellison and Poorman formations where it is not accompanied by anomalous gold. Pyrrhotite streaks in the iron formation are generally oriented parallel to transposed lithological banding, are crosscut or replaced by pyrrhotite blebs and layers, and are locally replaced by arsenopyrite. They are <1 mm to 3 mm thick, 1 to 10 mm long, elongated parallel to foliation, and flattened in the plane of the foliation. It is of possible early, pre-deformation, origin
ii). Blebs - occur in irregular and elliptical varieties. They measure <1 mm to several cm's in length, parallel to foliation and may locally contain fragments of quartz veins and country rocks. Blebs are common in disseminated lenses and masses within adjacent stage I and stage II quartz veins. Locally irregular blebs in stage II quartz veins are intergrown with arsenopyrite and are elsewhere associated with low grade (>1 g/t Au), to ore grade (>5 g/t Au) mineralisation. Elliptical blebs are not associated with gold. Pyrrhotite blebs are found in stage III quartz veins and are locally associated with anomalous gold (up to 300 ppb Au).
iii). Layers - of pyrrhotite occur as 1 to 2 mm thick seams, or 1 to 4 cm thick disseminations in layers alternating with parallel transposed bands of chert±biotite±chlorite in Homestake Formation. Pyrrhotite layers are found on foliation surfaces as micro-dilational fillings between lithological bands, and as replacement disseminations in certain bands, mainly siderite layers, near or adjacent to stage II quartz veins. These layers are confined to orebodies, contain sparse isolated grains as arsenopyrite porphyroblasts, and locally contain coarse grains of gold. They commonly assay >1 g/t, generally above 5 g/t Au. Pyrrhotite layers and blebs are frequently intergrown and both may locally replace pyrrhotite streaks. Zones of pyrrhotite layers invariably become discontinuous within a few metres from the margins of stage II quartz veins. Within some orebodies this textural form is exclusively associated with stage II veins. Layers are most common in the western ore ledges at lower metamorphic grades, and where D1b shear foliation is best developed.
In orebodies the combined pyrrhotite content by volume varies from 5% to more than 40% (Caddy, et al., 1991).
Arsenopyrite - is an important indicator of gold and is present in at least minor amounts in all orebodies. Its content varies from trace to more than 15%. It is almost invariably present as euhedral crystals, only rarely being found as stretched crystals. The crystals vary from 2 to 25 mm in length and are commonly striated parallel to the  axis. Arsenopyrite is disseminated in the Homestake Formation, and locally in the Ellison and Poorman Formations, occurring adjacent to stage II quartz veins. It may exist as crowded groups of crystals. The most abundant arsenopyrite generally occurs in chlorite rich, and locally pyrrhotite rich rock adjacent to stage II quartz veins, and is usually found near or within orebodies. Arsenopyrite is generally, but not exclusively, associated with ore grade (>4.7 g/t) gold, and is found as both coarse and fine grained varieties. Coarse grains are found as either single porphyroblasts or aggregates within or in the selvages of stage II veins, or as random grain aggregates within or near orebodies. Conversely stage II quartz veins occur, without exception, in the vicinity of abundant arsenopyrite concentrations. Coarse grained arsenopyrite is found intergrown with blebs and layers of pyrrhotite (Caddy, et al., 1991).
Fine grained arsenopyrite occurs as selective replacements of biotite rich layers, or is introduced along micro-dilational positions in Middle D1b shear foliations. Arsenopyrite layers, like those of pyrrhotite, are discontinuous, commonly intergrown with pyrrhotite layers, and intimately associated with the margins of stage II quartz veins. Layers of arsenopyrite are discontinuous, commonly intergrown with pyrrhotite. It comprises randomly oriented aggregates of intergrown euhedral grains, elongated in two dimensions in the plane of the Middle D1b foliations, or sub-parallel to compositional layering (Caddy, et al., 1991).
Pyrite - of Palaeoproterozoic age is found in most orebodies. It is most abundant on the upper eastern ledges in concentrations of from trace to 10%, but is sparse in the lower eastern and western ledges where it is generally <1%. In the upper mine levels it is generally medium grained (1 to 5 mm) cubes, and very fine grained (<1 mm) and cubic in the lower mine levels. It occurs as a fine grained replacement of pyrrhotite streaks, blebs and layers. Pyrite of Tertiary age may also be found in the district. This latter variety is related to Tertiary igneous rocks and late vuggy quartz veins, and is interpreted to be a replacement of Proterozoic pyrrhotite (Caddy, et al., 1991). Regionally the pyrite content increases, at the expense of pyrrhotite, within 200 to 250 m of the Palaeozoic unconformity. This does not apparently enhance the gold content (Bachman & Terry, Pers. comm., 1993).
Gold - is invariably associated with irregular blebs and layers of pyrrhotite, and commonly with arsenopyrite. Most of the gold is anhedral, very fine grained and disseminated throughout the orebodies. The highest concentrations are found near the margins of some of the stage II quartz veins. In general gold is spatially associated with the assemblage quartz-chlorite-pyrrhotite-arsenopyrite which largely constitutes the alteration selvage in the walls of stage II quartz veins.
Gold grains occur as:
i). microscopic to sub-microscopic blebs or flakes within micro-cracks and along grain boundaries of silicate, sulphide and carbonate gangue minerals, particularly pyrrhotite, arsenopyrite and garnet;
ii). as coarse intergrown networks within quartz veins and in the selvages of quartz veins;
iii). as micrometre thin smears on chlorite parting surfaces; while
iv). 4) visible gold may be observed along grain boundaries of arsenopyrite and pyrrhotite, as well as along fractures in arsenopyrite, pyrrhotite and garnet (Caddy, et al., 1991). Visible gold accounts for less than half of the total, the remainder being micron sized (Slaughter, 1968; Brunker, et al., 1987).
The association with cross-cutting features and fractures suggest that the gold is a late stage additive (Caddy, et al., 1991).
The most abundant coarse gold was found on the upper levels of the Main and 9 Ledge orebodies, although visible gold is common throughout the mine (Caddy, et al., 1991).
On the basis of mill recoveries the gold is reported as being 899 fine, the remaining 101 being silver. Microprobe analyses of gold grains averaged 86% Au and 14% Ag. The Au:Ag ratio varies from 3 to 10, averaging between 4 and 6 (Caddy, et al., 1991; Bachman & Terry, Pers. comm., 1993).
Other Sulphides - Chalcopyrite is a common trace mineral, occurring as microscopic to <2 mm anhedral grains, commonly as streaks or blebs in pyrrhotite. Galena and sphalerite are rare, only being found in certain quartz veins. Minor lollingite has been found in trace amounts (Caddy, et al., 1991).
There is little to no magnetite in either the ore or in the original iron formation (Bachman & Terry, Pers. comm., 1993).
Microscopic work indicates three separate groupings of opaque minerals. The oldest is of arsenopyrite, pyrrhotite and pyrite. Of these, arsenopyrite is the oldest, having been fractured before the introduction of pyrrhotite. Pyrite is later than pyrrhotite. The second grouping is of specularite, magnetite and marcasite, all of which are only present in small amounts. The third assemblage is also a trace mineral association, comprising tetrahedrite, galena, sphalerite and chalcopyrite. The most positive associations with gold are arsenopyrite and galena. Gold is older than arsenopyrite, pyrrhotite, pyrite, magnetite, specularite and galena. More gold appears to be found in association with arsenopyrite than any other mineral. However, instead of being enclosed by arsenopyrite alone, it generally replaces small inclusions of pyrrhotite, pyrite or gangue within the arsenopyrite. Both pyrite and pyrrhotite contain small inclusions of gold, while in a few specimens magnetite, chalcopyrite and galena may be replaced by gold (Slaughter, 1968).
Palaeo-placer gold mineralisation is found at the base of the Cambrian, which marks the Palaeozoic unconformity, and extends for over 4 km from its intersection with the orebody. These placers are apparently un-economic but may be used as a guide to ore. It is suggested that they were derived from a deep well developed lateritic surface that is found at the unconformity. Recent alluvials have apparently in turn developed from these accumulations (Bachman & Terry, Pers. comm., 1993).
Tertiary mineralisation - In addition to the Proterozoic ore, some mineralisation is also spatially and temporally associated with Tertiary igneous dykes and sills, although it is not economically significant. It is characterised by minor calcite, quartz, pyrite, dolomite, siderite, sphalerite, galena, molybdenite, chalcopyrite, anhydrite, fluorite, gypsum, chlorite, barite, rhodochrosite, realgar, native arsenic and rarely gold-silver tellurides. These minerals are found in late Tertiary quartz veins, disseminations and open space filling in and around fractures cutting all Tertiary intrusives and adjacent Proterozoic rocks (Caddy, et al., 1991).
While Tertiary mineralisation is not of economic significance at Homestake, some 70 t of gold have historically been mined from the Tertiary ores at the nearby Bald Mountain gold field. The deposits exploited generally consisted of replacements in dolomitic units that were cut by vertical fractures and were found in close spatial relation to Tertiary intrusives. The host rocks included the Cambrian Deadwood Formation and the lower Carboniferous (Mississippian) Pahasapa Limestone, as well as locally within Tertiary calc-alkaline sills, dykes, laccoliths and stocks. These deposits had a markedly different Ag:Au ratio to those at Homestake, generally around 1:1, and contained significant Zn and Pb. A few had wolfram group minerals and tellurides, while pyrite was the dominant gangue sulphide (Caddy, et al., 1991).
Geochemistry - Comparison of a large number of samples from barren and mineralised Homestake Formation rocks indicate little change in the levels of CaO, MgO, Na2O, MnO, TiO2 and P2O5; while SiO2, FeO and K2O are depleted in the mineralised formation relative to the barren equivalents. There is an enrichment in the ore zone of S, CO2, Au, As and Ag, while Fe2O3 and some elements such as V, Cr, Co, Ni and Cu are slightly elevated. W is barely detectable, averaging 2.6 ppm in both barren and mineralised Homestake Formation. Other notable trace constituents in all Homestake Formation samples include BaO >> SrO and Rb2O. Low order Te and Se (<10 ppm) are detected in the ore-bearing, but not in the barren Homestake Formation (Caddy, et al., 1991; Bachman & Terry, Pers. comm., 1993).
Mineralisation may not be accompanied by any bulk change in the Fe and SiO2 content of the Iron Formation and adjacent rocks. It has been noted that with an increase in the sulphide content of the ore, mainly as pyrrhotite and arsenopyrite, there is a corresponding increase in the Mg:Fe ratio of chlorite. It appears that the sulphides preferentially take up Fe at the expense of chlorite, sourced from the siderite and grunerite of the iron formation. Silica is also reduced in the wall rocks of the veins and deposited as quartz veins (Bachman & Terry, Pers. comm., 1993).
The iron formation generally contains 15 to 35% Fe, with a mean of 25%, occurring as grunerite, siderite, pyrrhotite and arsenopyrite. The grunerite zone represents the highest grades. This is taken to be a reflection of the volume reduction in the conversion from siderite to grunerite, producing a higher grade without a decrease in the total amount of Fe present. For this reason the Fe content increases downwards, into the grunerite rich section of the transition zone (Terry, Pers. comm., 1993).
Geochemically the element that correlates best with gold is sulphur, a reflection of the mineralogical association with arsenopyrite and pyrrhotite. Arsenic exhibits a weak to moderate correlation with Au, although within some ledges As is a general indicator of Au grades. However, the presence of arsenopyrite is not always an indicator of associated ore grade Au, especially in the extreme upper tail of a ledge. The Au:As ratio decreases upwards from the lower tails to the centroid to the upper tails. This pattern is apparently more convincing above the centroid. In addition there appear to be no systematic trends in the Au:Ag ratio within ledges. Elevated CO2 levels in the ore zone is related to introduced carbonate accompanying hydrothermal alteration and mineralisation (Caddy, et al., 1991; Bachman & Terry, Pers. comm., 1993).
At the surface, approximately 1 to 1.5 km up plunge from ore, the Homestake Formation exhibits abundant quartz veining, with chlorite and sulphides on the selvage. Parasitic folds are obvious, plunging at 30°SSE. Limonite in outcrop was formed after both sulphides and carbonates, but is less abundant in grunerite rich zones. The outcrop contains 50 to 100 ppb Au and several hundred ppm As. It weathers to an orange to brown limonitic, finely micaceous rock which is lighter than the adjacent Poorman Formation phyllites (Pers. observ., 1993).
Alteration - Retrogressive hydrothermal alteration took place in the Proterozoic rocks of the Homestake Formation in response to interpreted fluid movement through dilated segments of reactivated Latest D1a and Middle D1b shear zones. This occurred during a period that also coincided with the emplacement of the early phases of the nearby 1720 Ma Crook Mountain Granite. Zones of more intense alteration are confined to dilatant parts of shear zones that formed preferentially within, and adjacent to Homestake Formation, and the Yates Unit of the Poorman Formation, owing to significant rheological contrasts with the adjacent phyllites (Caddy, et al., 1991).
The principal alteration involves:
i). Sulphidation - as described above in the mineralogy section, to produce arsenopyrite, pyrrhotite and gold which replace wall rocks within and adjacent to Middle D1b shears; and
ii). Extensive chlorite, siderite, sericite and minor garnet replacement of wall rocks within and adjacent to Middle D1b shears (Caddy, et al., 1991).
Strong alteration is also observed on the margins of the Yates Unit of the Poorman Formation. This takes the form of carbonate and chlorite alteration in zones of more intense shearing. The carbonates are predominantly ankerite. The carbonate and chlorite alteration is both pervasive and as veining, to the point of developing a well banded shear fabric which obliterates the original texture and produces a gradational change from the amphibolite to the Poorman Formation phyllites (Bachman & Terry, Pers. comm., 1993).
The main gangue alteration minerals are as follows (Caddy, et al., 1991; Bachman & Terry, Pers. comm., 1993):
Chlorite - At least three generations of chlorite were formed, the first two during the prograde metamorphism, while the third was the result of alteration associated with gold mineralisation. The Type III chlorite is not easily differentiated from the first two types in all cases. It generally exists as matted aggregates with individual chlorite grains developed in a random to nearly parallel orientation. Where it is oriented near parallel to the metamorphic foliation it is hardest to distinguish from the earlier generations.
Type III chlorite is otherwise reasonably distinctive, accompanying all orebodies. It occurs within Middle D1b shears associated with both ore grade and anomalous gold. Where associated with gold it is locally more abundant than Types I and II chlorite, and cross-cuts earlier metamorphic mineral assemblages as well as earlier deformation fabrics. With an increase in the sulphide content of the ore, and amount of Fe taken up by them, the Mg:Fe ratio of chlorite increases.
Carbonate alteration - occurred throughout the Homestake mine, especially within the iron formation of the Homestake Formation. Again, as with chlorite there are interpreted to have been a number of generations of carbonate alteration. These are said to include: 1) carbonate crystallisation during prograde metamorphism and associated devolatolisation reactions; 2) carbonate formation during retrograde metamorphism in Latest D1a; and 3) carbonate introduction during hydrothermal alteration associated with gold mineralisation.
It should be noted however that an elevated carbonate content characterises the majority of rocks in the Lead Window of the Black Hills.
The early metamorphic carbonate indigenous to the iron formation is easily distinguishable. It is intergrown with quartz to form a fine grained, heterogranular, polygonal texture. The subsequent recrystallisation or carbonate introduction related to alteration is represented by subhedral to euhedral porphyroblasts of carbonate. Most of the carbonate in the semi-pelitic rocks and iron formations appears to be early stage to original. The introduced phase is represented by abundant coarse siderite, and to a lesser extent, ankerite within the orebodies. It occurs as undeformed porphyroblasts from 2 to 10 mm in maximum dimensions that are generally scattered throughout the orebodies, especially in the zones of Middle D1b shearing.
Coarse siderite is generally associated with abundant Type III chlorite in the same shears. The presence of coarse siderite (and/or ankerite), with or without Type III chlorite, does not necessarily indicate proximity to ore or sub-economic gold. Such assemblages are also found in barren iron formation cut by Middle D1b shears.
Sericitic, bleached alteration - Sericitic alteration zones are manifested as areas of bleaching that transect lithological bands, earlier metamorphic mineral assemblages, and potassium rich metasomatic haloes (eg. the biotite halo associated with the iron formations). They are most common and/or most obvious in the graphitic phyllites of the Poorman Formation, generally being wider than 5 m and extending for more than 100 m along strike and dip. Relict rock textures are generally visible, while the margins are transitional over widths of around 1 m. The fabric has usually been transposed, parallel to the shearing direction, to form a phyllonite with a mineral assemblage dominantly comprising sericite, carbonate, quartz and plagioclase with minor biotite. Depletion of C, Fe and Mg is implied, while CO2 and possibly Ca have been added.
These zones are commonly close to orebodies or sub-economic gold mineralisation. They frequently contain quartz veins of Stage II, and in parts are associated with arsenopyrite and pyrrhotite with low anomalous gold values of <.1 ppm Au. Some such zones also carry abundant Type III chlorite below orebodies.
Quartz veining - occurs in all formations in the Homestake area, but is most common in the iron formation of the Homestake Formation. It is less common in the quartzites of the Ellison and the phyllite of the Poorman, or in the semi-pelites of the Ellison Formation. Quartz veins are rare in the amphibolite of the Yates Unit of the Poorman Formation. In general, both at surface and underground, quartz veining is diagnostic of the Homestake Formation (Bachman & Terry, Pers. comm., 1993).
In general also, the quartz in veins are milky white, discontinuous, un-zoned, coarse grained and crystalloblastic. They are readily distinguishable from the sedimentary cherts. Discordant and concordant quartz veins and irregular quartz masses are commonly found together in all ore ledges and parts of orebodies.
Three different stages of Proterozoic quartz veins development can be differentiated, as follows:
i). Stage I - veins are largely concordant to foliation and transposed bedding; are sheared and folded locally; are less than 1 m thick and extend along strike for a few tens of metres. They are generally boudinaged, drag folded and sheared in high strain zones. The long axis is parallel to the Latest D1a axial fabric. Associated pyrrhotite is generally present as elliptical blebs near the vein margins. Quartz is commonly greyish white and translucent, while in orebodies they usually occur as detached boudins, isolated floating hinges, discontinuous lenses and rounded fragments.
ii). Stage II - veins range from <1 cm to more than several metres thick and extend along strike for a few metres to more than 350 m. They are roughly tabular, concordant to discordant and contain internal streaks, pods and aggregates of chlorite and/or biotite, pyrrhotite, arsenopyrite and commonly coarse gold. Asymmetric selvages of chlorite, biotite and minor ankerite are commonly 1 m or more thick and are best developed in the iron formation of the Homestake Formation, but are also present in the Ellison and Poorman Formations. Random, isolated, coarse grained aggregates of pyrrhotite are commonly found on one quartz margin, while coarse arsenopyrite porphyroblasts or aggregates characterise the other. Ore grade (but not necessarily volume) gold is generally present in both of these vein selvages, but is most common in veins near orebodies.
The highest individual gold grades in an orebody, combined with the most abundant observed visible gold, are associated with abundant arsenopyrite in thick selvages adjacent to the widest Stage II quartz veins. The largest orebodies with the highest gold, as in the ore ledge centroids, contain numerous Stage II quartz veins with well developed vein selvages.
Stage II quartz veins are rarely deformed, although post ore shearing may produce local drag folding and boudinage. Stage II veins consistently truncate Stage I quartz veins. In many orebodies the Stage II veins develop into large, irregular masses or "blow-outs" which may crosscut earlier tabular developments and contain blocks of country rock. Arsenopyrite, pyrrhotite and fine to coarse gold are associated and intergrown with chlorite, locally along the margins of irregular quartz bodies.
Stage II veins and irregular bodies are much larger then Stage I veins. They have been interpreted to have been formed in episodic pulses by injection into active shears during Middle D1b deformation, and were then locally deformed by Late D1b shearing. It has also been suggested that the Stage II veins were introduced into the ductile to ductile-brittle transition and were directly related to gold mineralisation.
iii). Stage III - Late, barren Stage III quartz veins cut both Stage I and II quartz veins. They are markedly discordant and have poorly developed alteration selvages, if any, and are un-deformed. They average 5 cm in thickness, but may be up to 1 m wide, and extend for several tens of metres along strike. Stage III veins are planar and transect foliation and orebodies at angles of more than 30°. They were apparently emplaced during Late D1b shearing and are composed of milky quartz, commonly containing coarse ankerite and pyrrhotite with subordinate albite. Locally they may contain narrow biotite and/or chlorite selvages. They are more common in greenschist facies hosts, have open space filling textures and as such are interpreted to have developed in a brittle regime.
The most recent source geological information used to prepare this summary was dated: 2010.
This description is a summary from published sources, the chief of which are listed below.
© Copyright Porter GeoConsultancy Pty Ltd. Unauthorised copying, reproduction, storage or dissemination prohibited.
References to this deposit in the PGC Literature Collection:
Morelli R M, Bell C C, Creaser R A and Simonetti A, 2010 - Constraints on the genesis of gold mineralization at the Homestake Gold Deposit, Black Hills, South Dakota from rhenium–osmium sulfide geochronology: in Mineralium Deposita v.45 pp. 461-480|
Top | Search Again | PGC Home | Terms & Conditions