Abstract

In other parts of the world, previous workers have shown that sparry dolomite in carbonate rocks may be produced by the generation and movement of hot basinal brines in response to arid paleoclimates and tectonism, and that some of these brines served as the transport medium for metals fixed in Mississippi Valley-type (MVT) and sedimentary exhalative (Sedex) deposits of Zn, Pb, Ag, Au, or barite.

Numerous occurrences of hydrothermal zebra dolomite (HZD), comprised of alternating layers of dark replacement and light void-filling sparry or saddle dolomite, are present in Paleozoic platform and slope carbonate rocks on the eastern side of the Great Basin physiographic province. Locally, it is associated with mineral deposits of barite, Ag-Pb-Zn, and Au. In this paper the spatial distribution of HZD occurrences, their stratigraphic position, morphological characteristics, textures and zoning, and chemical and stable isotopic compositions were determined to improve understanding of their age, origin, and relation to dolostone, ore deposits, and the tectonic evolution of the Great Basin.

In northern and central Nevada, HZD is coeval and cogenetic with Late Devonian and Early Mississippian Sedex Au, Zn, and barite deposits and may be related to Late Ordovician Sedex barite deposits. In southern Nevada and southwest California, it is cogenetic with small MVT Ag-Pb-Zn deposits in rocks as young as Early Mississippian. Over Paleozoic time, the Great Basin was at equatorial paleolatitudes with episodes of arid paleoclimates. Several occurrences of HZD are crosscut by Mesozoic or Cenozoic intrusions, and some host younger pluton-related polymetallic replacement and Carlin-type gold deposits.

The distribution of HZD in space (carbonate platform, margin, and slope) and stratigraphy (Late Neoproterozoic Ediacaran–Mississippian) roughly parallels that of dolostone and both are prevalent in Devonian strata. Stratabound HZD is best developed in Ediacaran and Cambrian units, whereas discordant HZD is proximal to high-angle structures at the carbonate platform margin, such as strike-slip and growth faults and dilational jogs. Fabric-selective replacement and dissolution features (e.g., collapse breccias, voids with geopetal textures) are common, with remaining void space lined with light-colored dolomite crystals that exhibit zoning under cathodoluminescence. Zoned crystals usually contain tiny (<1–3 μm) fluid inclusions with vapor bubbles, requiring Th > ∼70 °C. The oxygen isotopic compositions of HZD are consistent with formation temperatures of 50–150 °C requiring brine circulation to depths of 2–5 km, or more. The few HZD occurrences with the highest concentrations of metals (especially Fe, Mn, and Zn) and the largest isotopic shifts are closely associated with Sedex or MVT deposits known to have formed from hotter brines (e.g., Th > 150–250 °C).

These relationships permit that HZD formed at about the same time as dolostone, from brines produced by the evaporation of seawater during arid paleoclimates at equatorial paleolatitudes. Both dolostone and HZD may have formed as basinal brines, which migrated seaward from evaporative pans on the platform, with dolostone forming at low temperatures along shallow migration pathways through permeable limestones, and HZD forming at high temperatures along deeper migration pathways through basal aquifers and dilatant high-angle faults. The small MVT deposits were chemical traps where hot brines encountered rocks or fluids containing reduced sulfur. The abundant Sedex deposits mark sites where hot brine discharged at the seafloor in adjacent basins. Thus the distribution of HZD may map deep migration pathways and upflow zones between eastern shallow marine facies, where evaporative brine could have been generated, and western Sedex deposits, where heated brines discharged along faults into platform margin, slope, and basin facies. The small size and scarcity of Pb-Zn deposits and the abundance of barite deposits in the Great Basin suggests the brines were generally reduced, possibly due to reactions with carbonaceous rocks along deep migration pathways. While this scenario may have occurred at several times, the age and abundance of Sedex deposits suggest that such a hydrology was best developed in the Late Ordovician, Late Devonian, and Early Mississippian, possibly in response to episodes of extension and forebulge faults associated with the Antler orogeny. The improved understanding of HZD may aid future exploration for ore deposits in the Great Basin.

INTRODUCTION

In many mining districts in the Great Basin region of the western United States, hydrothermal zebra dolomite (HZD) is present in ore deposits of different types and ages (Fig. 1, Table 1). HZD is composed of alternating layers, ranging from a millimeter to a decimeter in thickness, of dark, fine-grained replacement dolomite and white, coarse-grained, sparry, void-filling dolomite. Such HZD has been interpreted to be approximately coeval with Paleozoic Sedex and Mississippi Valley-type (MVT) deposits (Emsbo et al., 1999; Vikre, 2001) and to host younger Cretaceous pluton-related polymetallic replacement (Vikre, 1998; Vikre and Browne, 1999) and Eocene Carlin-type gold deposits (Hofstra and Cline, 2000; Emsbo et al., 2003). Because of the proximity of HZD to economic mineral deposits, an investigation was initiated in 2002 to learn more about the distribution and attributes of HZD within and outside of mining districts in the Great Basin.

Globally, HZD has been the subject of numerous investigations because of its association with base-metal deposits and oil and gas reservoirs. It is more or less synonymous with “hydrothermal dolomite (HYD or HYTD)” in petroleum geology. Based on fluid inclusion studies, the origins of HZD from hot brines are fairly well understood (Davies and Smith, 2006). HZD is associated with dolostone, oil fields (Cooper et al., 2001; Davies and Smith, 2006), and Sedex and MVT Pb-Zn and barite deposits (Leach et al., 2005; Emsbo, 2009). It also has been shown to be a preferred host for younger pluton-related polymetallic replacement deposits in Colorado mining districts, such as Leadville and Gilman (Thompson, 1990; Taylor, 1999).

Diagenetic dolomite (or dolostone) often forms in arid climates at equatorial latitudes from evaporated seawater with high Mg/Ca ratios produced by the loss of Ca associated with precipitation of gypsum and anhydrite (e.g., Morrow, 1998; Morse and Mackenzie, 1990). Likewise, HZD associated with Sedex and MVT deposits forms from evaporative brines (10–30 wt% NaCl equiv.) that circulated to great depths where they attained elevated temperatures of 90–200 °C (Leach et al., 2005; Emsbo, 2009). The dark-gray replacement zebra dolomite layers and geopetal fabric in voids may reflect selective replacement and dissolution of limestone by hot brine. The white, sparry, cathodoluminescent (CL) zoned, dolomite-filled layers are thought to precipitate by the exsolution and loss of CO2 due to pressure decreases associated with brine movement from deeper to shallower levels (Leach et al., 1991) and dilation along faults (Davies and Smith, 2006). The stable isotopic composition of HZD reflects the original isotopic composition of the evaporative brine, as well as succeeding modifications along the flow path due to heating, reactions with limestone and organic carbon compounds, or mixing with other fluids such as meteoric water (e.g., Lohmann, 1988; Allan and Wiggins, 1993). Spatial variations in the isotopic composition of HZD have been used to elucidate the direction of brine movement (Qing and Mountjoy, 2004).

Chemical modeling of hot brine over a range of redox conditions suggests that Fe, Mn, Pb, and Zn are more soluble under relatively oxidizing conditions, whereas Ba and Au are more soluble under reducing conditions (Cooke et al., 2000; Emsbo, 2000). Consequently, the redox state and metals present in HZD are a reflection of the mineralogy and metals present in the rocks through which hot brine moved and the distribution coefficients for each element (e.g., Veizer, 1983). Hot brines that move through rocks containing organic matter and pyrite will be more reduced and sulfidic, whereas those that move through rocks containing Fe-oxide, Fe-silicate, and Fe-carbonate minerals will be more oxidized and sulfate rich. Variations in the redox state of the fluids have been called upon to explain CL zoning in carbonate minerals—which is mainly controlled by the incorporation of Mn+2 and Fe+2, the most important activator and quencher species, respectively (e.g., Frank et al., 1982; Barnaby and Rimstidt, 1989; Fraser et al., 1989; Goette and Richter, 2004).

Sedex Pb-Zn deposits form in intracontinental or failed rifts and rifted continental margins during sedimentation or early diagenesis (Leach et al., 2005; Emsbo, 2009). In this structural setting, dense brine produced by evaporation of seawater on the carbonate platform infiltrates underlying aquifers to depths of a few km or more, moves laterally for distances as much as 100 km, and ascends buoyantly along active extensional faults in adjacent basins to form ore deposits (Leach et al., 2005; Paradis et al., 2007; Emsbo, 2009). MVT deposits form in platform carbonate sequences located within extensional zones (e.g., forebulge) inboard of orogenic belts and are generally tens of millions of years younger than their host rocks (Leach et al., 2005). The regional scale of some MVT flow systems is reflected by hydrothermal dolomite cement, with or without trace amounts of sulfides, which can be present in carbonate rocks hundreds of kilometers from the deposits (Leach et al., 2005). Petroleum reservoirs in structurally controlled HZD have been recognized in both settings (Davies and Smith, 2006). In complexly deformed margins, brine may move from platform to basin during extension or from deformed basin to platform during contraction. In successor foreland basins, there may be renewed brine generation and movement. Thus, in complex tectonic settings, multiple generations of hydrothermal dolomite may form in carbonate rocks.

These previous studies suggest that HZD may provide an indication of: (1) arid conditions at equatorial paleolatitudes, where conditions were favorable for sufficient evaporation of seawater to produce brine (e.g., Fig. 2A; www.scotese.com); (2) deep lithologic and structural brine migration pathways used during episodes of extension or contraction; (3) the anatomy of hydrothermal systems that form hydrocarbon reservoirs, Sedex, and MVT deposits associated with HZD; and (4) the isotopic evolution, redox state, and metals present in the brines.

The purpose of this paper is to summarize our current understanding of the age, distribution, and attributes (structural, textural, chemical, and isotopic) of HZD in the Great Basin. We compiled published reports and sought out anecdotal information on HZD occurrences; conducted field work to verify their locations, host strata, and structural fabrics; and collected representative samples for laboratory investigations. Diehl et al. (2005) summarize initial field observations, textural and structural fabric studies, petrographic observations, and chemical characterization of HZD by laser ablation- inductively coupled plasma-mass spectrometry (LA-ICP-MS). Herein we integrate the most important observations and data on HZD occurrences from previous work with: (1) an updated statistical treatment and interpretation of the LA-ICP-MS data, (2) new stable isotopic results on each locality, (3) information on the larger stratigraphic and geotectonic framework of the HZD occurrences, and (4) information on their relation to structures and hydrothermal ore deposits. These relationships are used to learn what HZD can tell us about the role of basinal brines in the tectonic evolution and formation of mineral deposits, i.e., metallogeny, of the Great Basin.

Geologic Setting

The Cordilleran miogeocline along western North America (Laurentia) was produced by intracontinental rifting of Rodinia and episodic Cryogenian to Devonian extension (Dickinson, 2006; Lund, 2008). Along the length of the miogeocline from Alaska to southern California, Lund (2008) has shown that northwest-striking asymmetric extensional segments are bounded by northeast-striking transform and transfer segments producing a zigzag pattern that is paralleled by younger fold and thrust belts, extensional domains, and a variety of Sedex and MVT Pb-Zn and barite deposits in slope and platform facies rocks. The widest extensional segment to the north in Canada, the Selwyn Basin, is renowned for its Sedex and MVT Pb-Zn deposits and associated occurrences of HZD (Goodfellow, 2007; Nelson et al., 2002, 2006). The widest extensional segment in the south, the Great Basin, is renowned for its Sedex barite deposits (Papke, 1984; Koski and Hein, 2003), varied gold deposits (Hofstra and Cline, 2000; Emsbo et al., 2006), and HZD (Diehl et al., 2005). Dickinson (2004, 2006) has shown that both the Selwyn Basin and the Great Basin have similar Paleozoic and early Mesozoic tectonic histories that were punctuated by a series of contractional orogenies that thrust terranes comprised of basin facies onto slope and shelf facies (e.g., Golconda [GT] and Roberts Mountains [RMT] thrusts on Figs. 1 and 2B). The Great Basin has been affected by Mesozoic contraction (e.g., Luning-Fencemaker thrust [LMT]) that culminated in the Sevier orogeny and subduction-related magmatism behind the Sierran arc terrane (Fig. 2B) that produced a wide variety of pluton-related ore deposits (Barton, 1996) that in places are hosted in HZD. In Cretaceous through Oligocene time, the locus of subduction-related magmatism migrated across the Great Basin (Fig. 2B) and in Eocene time was synchronous with the onset of extension and formation of Carlin-type gold deposits (Hofstra et al., 1999) that are locally hosted in HZD. Subsequent Basin and Range tectonism (Fig. 2B) produced the exposures of HZD that are the focus of this investigation.

SAMPLING AND ANALYTICAL METHODS

Table 1 provides information on the age and name of the formations that host HZD as well as the interpreted age(s), deposit type(s), and mine/district names that have HZD. Samples of HZD collected from mines, quarries, and outcrop were slabbed and stained. Polished thin sections of interesting features were prepared for standard petrographic, CL, and SEM study. A Citl (Cambridge Image Technology Ltd.) CCL 8200 mk3a CL scope, operating at 300 current and 15 kV, was used to document the characteristic luminescence color(s) of dolomite and calcite phases and their textural and zoning features that reflect compositional changes (Rowan, 1986). A JEOL 5800 scanning electron microscope with energy dispersive X-ray spectrometer (SEM/EDX) (USGS Denver Microbeam Laboratory) was used to document zoning and major-element compositional variations in HZD. Petrographic, CL, and SEM observations were used to select textural and zoning features that were free of mineral inclusions for LA-ICP-MS analysis. Forty-five elements were analyzed by LA-ICP-MS using a 266 nm laser and a spot size of 200 μm as described in Diehl et al. (2005). The 200 μm spot size used represents a compromise between the sensitivity required to detect many trace elements and the spatial resolution needed to characterize the textural and zoning features observed in HZD. When these data were collected, the lack of a dolomite standard prevented generation of quantitative data. Hence, the raw intensity data for each element were gas blank subtracted and normalized to the intensities of Ca plus Mg to account for variations in ablation efficiency, sample transport, and plasma conditions. The resulting qualitative data (Supplemental File 11) provide a reliable indication of trace element differences within and between samples. To facilitative multivariate statistical analyses, spots for certain trace elements that were below detection limits were assigned an arbitrary low-intensity ratio (element/Ca+Mg) of 1E−08. Statview (SAS Institute, 1995) software was used for R-mode factor analysis and to generate bar and whisker plots that display the percentile values for each trace element. The major elements that comprise dolomite (C, Mg, and Ca) were excluded from the factor analysis, as was Cr, because it contributed to a poor solution.

For stable isotope analysis, several milligrams of powdered dolomite were digested in phosphoric acid following McCrea (1950). The resulting CO2 gas was purified in a vacuum line using silver phosphate to remove any H2S that may have evolved from trace sulfide minerals. The gas samples were analyzed using a Finnigan MAT 252 mass spectrometer. The results are reported in delta notation in units of per mil relative to Vienna PeeDee Belemnite (VPDB) for carbon and Vienna Standard Mean Ocean Water (VSMOW) for oxygen. The reported values include a correction for oxygen fractionation that was made using the dolomite fractionation factor recommended by Friedman and O'Neil [1977; VSMOW = (0.9992*VSMOW) − (0.82)]. For comparison to zebra dolomite localities outside of the Great Basin, VPDB values from other researchers were converted to VSMOW, using VSMOW = 30.92 + (VPDB).

RESULTS

Distribution of HZD in the Great Basin

Given that both dolostone and HZD may form from evaporative brines, it is important to consider their respective distributions. Ludington et al. (2005) compiled a digital geologic map encompassing the Great Basin with information on the primary and secondary lithology of each map unit. Figure 3 shows the distribution of HZD localities relative to map polygons composed primarily or secondarily of dolostone. Most HZD localities occur within or near such polygons. Some of the exceptions are where HZD has only been identified in the subsurface or where HZD is localized along faults that cut across units composed of limestone (e.g., Meikle, Pequop Summit). There also are dolostone-bearing polygons without HZD occurrences. Notable among these are the Triassic foreland basin polygons in western Nevada.

If evaporative brines were generated in restricted lagoons and sabkhas on the platform, and Sedex and MVT deposits formed elsewhere after the brines circulated to substantial depths, then the distribution of HZD localities relative to depositional facies, unconformities, and ore deposits may indicate brine migration pathways. Cook and Corboy (2004) characterized the sequence stratigraphy and depositional facies across the Paleozoic passive margin in the central Great Basin. The positions of HZD localities and Sedex and MVT deposits are projected onto these profiles (Figs. 4A and 4B). Concordant and discordant HZD localities are marked, respectively, by horizontal and vertical bars. The largest group of concordant HZD localities is near the base of the section in Ediacaran and Cambrian (?) formations. Most HZD localities are in Devonian limestones (Fig. 1) and are discordant to bedding. These observations permit that dense brines descended and migrated laterally through permeable carbonate strata in the lower part of the miogeocline with upflow along high-angle faults in younger rocks at shallower levels. Most HZD localities are inboard of the platform margin, whereas most Sedex deposits are outboard of the margin. Meikle is one of the few HZD occurrences located right at the platform margin in close association with Sedex gold and barite deposits. In contrast, known MVT deposits such as Goodsprings occur in the eastern part of the platform (Fig. 2B). The distribution of ore deposits and HZD localities permits that brine migrated from eastern platform to western slope and basin facies during Sedex mineralization, and from west to east during MVT mineralization, consistent with recently described genetic models for hydrothermal dolomite and these deposit types (Gasparrini et al., 2003; Leach et al., 2005; Paradis et al., 2007; Emsbo, 2009).

Field Observations

HZD occurrences are commonly localized, either bracketed by and confined by high-angle faults, or bounded by narrow, vertical breccia zones (Diehl et al., 2005). Such fault-controlled HZD may selectively replace individual beds within an otherwise unaltered sequence of carbonate rocks, or it may be several meters thick and replace several beds (Figs. 5A–5C). Zebra texture in the Devonian Simonson Dolostone at Cedar Peak is restricted to a block bounded by bedding-parallel and high-angle faults (Thorman and Brooks, 1993). Hydrothermal dolomite at the Mendha mine, Pioche district, abruptly ends to the east by a high-angle fault and to the west by a vertically oriented zone of fault breccia with zebra-textured clasts (Diehl et al., 2005).

HZD shares many similar physical characteristics at all localities. Unaltered rocks adjacent to HZD commonly exhibit shallow-water carbonate sedimentary features such as thinly laminated tidal and ripple structures, micritic peloids, geopetal structures that record original porosity, and fragments of crinoid, brachiopod, and bryozoans (Diehl et al., 2005). Peloid and fossil shapes are commonly preserved within the replacive gray zebra layers (Fig. 6). Shallow reefal and peloidal sediments may have been favorable porous and permeable environments for precipitation of hydrothermal dolomite. At the Mendha mine in the Pioche district, HZD in Cambrian strata is concordant to bedding, and a beige-colored sparry dolomite fills moldic porosity after probable algal structure. Sparry dolomite also fills moldic porosity after shell material in a Silver Island Range, Utah, sample; there is no preservation of fossil features other than the outline of the shells. Hand samples show that shell material and bryozoans are replaced by dark dolomite at Mineral Hill, Nevada. Several sample sites also exhibit evidence of tidal zone evaporite mineralization; anhydrite replaces dolomite in the Proterozoic Noonday Dolomite at sites such as the Paddy's Pride and Blackwater Mine, California, whereas in Devonian samples from the Silver Island Range, Utah, dolomite replaces earlier gypsum.

HZD can be concordant and/or discordant to bedding and may be barren or mineralized by coeval or younger metal-bearing fluid flow events. In the Panamint Range-Death Valley area, HZD in Ediacaran and Cambrian strata is concordant to bedding and occurs at several localities over a north-south distance of 30 km (Hall and Taylor, 1983; Hall, 1984). Such stratabound occurrences in the deeper part of the section may mark aquifers utilized by migrating brines. At the Apex limestone and dolomite quarry, Las Vegas, Nevada, HZD is discordant to bedding in the Bullion Member of the Mississippian Monte Cristo Limestone. At the Windfall pit in the Eureka district, HZD is discordant, filling horizontal and vertical veins.

High-angle centimeter-scale displacements within zebra dolomite layers have been suggested to indicate a tectonic control (Davies and Smith, 2006), but many high-angle small-scale faults abut low-amplitude pressure-solution seams (i.e., stylolites), which suggests that the high-angle faults are formed through accommodation by loss of rock volume (Fig. 6A). Millimeter to decimeter small-scale displacements along high-angle small-scale faults and veins in HZD are synchronous with dolomite precipitation because horst and graben “chevron” structures (Davies, 2004) are filled with coarse-grained white dolomite (Fig. 6A). These white, dolomite-filled microfaults that form the chevrons crosscut the dark-gray replacive layers. Many HZD localities exhibit such features that record multiple periods of deformation and episodes of dolomite dissolution and precipitation.

Pseudo–cross-bed structures commonly dip and merge into low-amplitude pressure-solution seams, which indicates that the “cross bedding” is a texture that forms from dissolution and tilting of zebra-textured rock (Fig. 6A). Low-amplitude horizontal stylolites are common structures parallel to the long dimension of saddle-dolomite filled vugs and veins.

Hydrothermal dolomite also occurs as cement in jigsaw-to-rubble breccias. In the Confusion Range, Utah, hydrothermal dolomite fills vugs in the overlying Sevy Dolostone but acts as a cement in breccia in the underlying Silurian Laketown Dolostone. Hydrothermal dolomite also acts as a cement in float breccias at the Rose mine in the Eureka district, at Mineral Hill, and at Cedar Peak in the Snake Range (Table 1; Fig. 5C). Breccias typically contain zebra-textured clasts, which require multiple generations of influx of Mg-bearing fluids and dolomite precipitation.

The textural features of HZD observed in the field suggest that it formed by selective dissolution and replacement of older depositional, diagenetic, and structural features in carbonate rocks. Certain laminae, trace fossils, fossil shells, breccia matrices, and other features observed in unaltered limestone are preferentially dissolved, whereas adjacent laminae or enclosing rock are preferentially dolomitized. Such porosity is subsequently filled by dolomite crystals that grow inward from adjacent replacement dolomite. This dissolution and dolomitization process is accompanied by volume loss that generates open space, horst and graben features, and pressure-solution seams (Fig. 6A).

Late-stage silicification is often associated with brecciated zebra dolomite. Silica-filled fractures crosscut zebra texture at the Mendha mine and Mineral Hill sites. Late-stage siliceous-cemented breccias are associated with the Cedar Peak and Rose mine HZD sites. Sulfide mineralization, pyrite-galena-sphalerite, is commonly associated with HZD. Hand samples show that sulfide minerals occur as: (1) a cement in solution-collapse breccias along with sparry dolomite (e.g., Mineral Hill), (2) a coating on rims of late-stage dolomite rhombs (e.g., Death Valley), (3) in cross-cutting veinlets with barite or calcite (e.g., Windfall pit), or (4) late-stage mineralization in dissolution voids in HZD (e.g., Blackwater and Queen of Sheba mines).

Petrographic Observations

Many authors use the textural terminology of Sibley and Gregg (1987) to describe the dolomite in the gray layers as nonplanar fabrics and the dolomite in the white layers as planar-S fabrics (Gasparrini, 2003; Gasparrini et al., 2003). Dolomite crystals in the gray layers are generally anhedral, with fitted irregular intercrystalline boundaries. Dolomite crystals in the white layers are subhedral to euhedral with straight intercrystalline boundaries. The planar-S fabrics of the white layers typically have grown toward one another in open cavities. At the thin-section scale, contacts between anhedral fine-grained replacive dolomite in the gray layers and coarse-grained white vug- or vein-filling dolomite cement are gradational. A saddle dolomite crystal may initiate within an inclusion-rich gray layer and increase in size and translucence toward the center of the vug (Fig. 6B).

Geopetal textures, which indicate horizontal deposition, are common in the zebra fabric. Cavities, and layered mineral fillings within them, form because of fluctuations in chemistry of the dolomitizing solutions, which alternate between corroding and dissolving, and precipitation and deposition (Bray, 1983). Common geopetal fabrics include: (1) a dark gray to black fine-grained dolomite that is an internal sediment in solution cavities and commonly along the base of the white layers (Figs. 7A–7C) (this carbonate sediment fill is different texturally from the surrounding carbonate rock), and (2) microscopic carbonate debris that settles as a microbreccia, filling microtopographic depressions (Fig. 7D). Features of these geopetal fabrics include:

  • (1) Solution cavities, which record geopetal infillings of dark fine-grained dolomite and white void-filling dolomite, are similar in size in different zebra dolomite localities (Figs. 7A and 7B). Examples shown here are solution cavities in rock slabs from the Ada Edith claims, in the southern part of Nevada, and from the Rose mine, in the northern part of Nevada. The solution cavities are rounded to irregular in outline. The dark dolomite is fluid-inclusion rich and occurs with sulfide mineralization. Davies and Wendte (2005) refer to these features as “dark-colored geopetal solution residues or internal sediments.” Dark dolomite commonly lines the base of the white coarsely crystalline dolomite (Fig. 7C).

  • (2) Microbreccia of different generations is common in zebra dolomite, emphasizing the importance of dissolution processes in the formation of HZD. Microbreccia, which infills topography on the surfaces of the gray layers and is cemented by the coarsely crystalline white dolomite, is evidence that certain layers in the host rock underwent selective dissolution; these layers were the sites of precipitation of the white vein-filling saddle dolomite. Figure 7D is an example of such an infilling of carbonate debris in a depression in a gray layer. The carbonate debris is the same material optically as the overlying gray layer.

Several zebra dolomite localities show affinities with evaporative facies. Samples from the Silver Island Range, Utah, reveal that the millimeter-scale thick gray layers were once host to acicular anhydrite and (or) gypsum crystals (Fig. 8). Interestingly, “ghosts” of the evaporite crystals project into the coarse-grained white layers, which suggests the gray replacive layers and the white, vein-filling layers formed in the same time frame. The zebra textures in the Silver Island Range samples are concordant with the thin beds of evaporite minerals and layers of “shell hash.” In the Tecopa mining district (e.g., Queen of Sheba, Paddy's Pride mines), southern California, anhydrite is common in zebra dolomite in the Noonday Dolomite.

The Mineral Hill site and mines in the Noonday Dolomite exhibit zincian and plumbian dolomite in late-stage coarse-grained terminal rhombs, commonly around voids (Fig. 9). SEM and LA-ICP-MS analyses indicate that there is element substitution of Zn for Mg in the dolomite lattice structure. The Zn and Pb zones commonly show preferential dissolution in comparison to the Mn-rich dolomite (Fig. 9D). Zincian dolomite is not uncommon in MVT deposits, such as the Navan (Tara) Zn-Pb deposit, Ireland (Kucha and Wieczorek, 1984).

Sulfide mineralization commonly occurs in veins that crosscut zebra texture, in dissolution voids, in geopetal dark dolomite, or terminal growth zones of saddle dolomite. Therefore, sulfide mineralization is generally late stage and occurs in multiple episodes of deformation. Vugs in the Sevy Dolostone are lined with zoned dolomite rhombs that show dissolution etching with very fine-grained galena occurring in the voids. Late-stage sulfide mineralization is common in vugs in samples from the Mineral Hill site, and the Queen of Sheba and Blackwater mines (Fig. 9B). In samples from the Rose Mine, sulfide mineralization occurs along calcite-filled veins that crosscut zebra texture (Fig. 6A).

HZD at all localities contains primary submicron fluid inclusions (Fig. 10). Although we did not obtain homogenization temperatures, we observed that fluid inclusions commonly contained trapped vapor bubbles, which signify that the fluid inclusions were trapped during mineral growth at temperatures >70 °C (Roedder, 1984). Therefore we are confident that the HZD samples are hydrothermal in origin.

Cement Stratigraphy and Cathodoluminescence

Slabs and thin sections of zebra dolomite were stained with a potassium ferricyanide solution. Most evident in samples from Death Valley, the terminations of late-stage dolomite rhombs in the cavities acquired a blue coloration, indicating a greater Fe content from that of the replacive dolomite. Under CL, this relative higher Fe content in the white coarse-grained saddle dolomite is indicated by quenched, dark terminations. The coarse-grained white cement commonly shows two or more zones of bright to dark banding indicative of changing chemistry. In contrast, the fine-grained gray layers in HZD commonly show medium-dark mottled luminescence.

An attempt was made to correlate luminescence zoning between HZD occurrences in the same district and occurrences within the same stratigraphic units. This effort produced mixed results. In Figure 11, CL zones were correlated between the Cedar Peak and Pequop Summit localities (Fig. 1), which are in the Devonian Simonson Dolostone and Guilmette Formation and are spaced ∼22 miles apart. Although the zoning of the hydrothermal dolomites appears to correlate, their isotope signatures are not identical (see below) which suggests they may have formed at slightly different temperatures (∼20 °C to account for the 2‰ difference in δ18O values) from brines that evolved isotopically between each locality, or during different events. In addition, in some Cedar Peak samples, calcite pendant cements in dissolution zones suggest that HZD underwent subaerial exposure some time in its history.

CL is an invaluable tool in unraveling the deformation and carbonate cementation history in HZD deposits. For example, CL revealed a complex history of multiple generations of dolomite precipitation and brecciation at the Mineral Hill site (Fig. 12). The earliest phase is a Fe-bearing dark luminescent saddle dolomite, which in plane light (PL) can be either fluid-inclusion rich (Fig. 12A) or fluid-inclusion poor (Fig. 12C). Enclosing the early dark CL dolomite is a medium-dull to medium-bright inclusion-poor dolomite. The early Fe-rich dark luminescing dolomite and its cement were subsequently brecciated and cemented by a clear, zoned rhombic dolomite phase; CL zoning is from a bright luminescent core to a dull rim (Fig. 12D). Quartz is also a cement between breccia fragments, and the quartz has partially replaced the dolomite rhombs. Sulfide mineralization is concurrent with cementation of the breccia (Fig. 12F).

Although HZD occurs in the Bullion Member of the Monte Cristo Limestone at the Apex Quarry northeast of Las Vegas, the Green Monster mine in the Goodsprings district, and the dolomite exposures near Jean, Nevada, there is no correlation between CL zones, even though in plane light, the dolomite from all sites is simply zoned, with cloudy, fluid inclusion cores and a clear rim (Fig. 13). The clear rim is thin (<0.5 mm) in the Green Monster and Jean dolomite rhombs and wide in the Apex sample, illustrating the difficulty in correlating microstratigraphy because of varying thickness of growth zones. The Green Monster mine is mineralized and hosts Pb, Zn, and Ag. The Apex quarry is unmineralized except by Mn oxide associated with cross cutting calcite-filled veins. The sample from Jean has been extensively altered by meteoric waters, and dissolution voids in the dolomite have calcite pendant cements that are dark luminescent in comparison to the bright red luminescence of dolomite (Figs. 13E–13F). Therefore, these three localities have undergone a different history of alteration and mineralization, which may have affected CL properties.

HZD from the Meikle mine has the darkest luminescence and is the most Fe-rich locality sampled (up to 9 wt % Fe; Emsbo, 1999). CL and geochemical zoning in this dolomite, from Fe-bearing cores to Fe-poor rims (Fig. 14), suggests that brines evolved from more oxidized to reduced conditions over time culminating in formation of Sedex Au and barite deposits at shallower levels (Emsbo et al., 1999). The high Fe content of dolomite at this locality made it an ideal chemical trap for gold during younger Eocene Carlin-type hydrothermal activity (Emsbo et al., 2003).

Laser Ablation

LA-ICP-MS data was generated along traverses across gray replacement and white void-filling layers of zebra dolomite and white coarse-grained dolomite in faults (Diehl et al., 2005). The initial results of this study showed: (1) that HZD from mining districts with different deposit types commonly have different signatures, (2) that elements, such as Cu, Zn, As, Ag, Sb, Au, and Pb, are somewhat higher in the gray replacement layers than in the white void-filling layers and faults, and (3) that some interelement correlations reflect the presence of submicron inclusions of other minerals.

Because Mn and Fe are the main activators and quenchers of CL in carbonate minerals, the Mn/Ca+Mg and Fe/Ca+Mg values from each locality were displayed on an X-Y plot (Fig. 15A). The data define three fields. Most HZD's have low Mn and Fe. Meikle and three spots from Mineral Hill and one spot from Windfall have high Fe and Mn. Three spots from Mendha have high Mn and low Fe. Meikle is clearly anomalous in Mn and Fe suggesting it formed from fluids substantially different from most other HZD localities in the Great Basin. Because Zn was detected by EDS in some dolomite rims, an analogous plot was prepared that displays Zn/Ca+Mg values, using proportional symbols, as a function of Mn/Ca+Mg and Fe/Ca+Mg values (Fig. 15B). It shows that Zn generally increases as Mn and Fe increase. Other factors being equal, such a gradual increase in Mn, Fe, and Zn may reflect increasing temperature because these elements have prograde solubility (e.g., Cooke et al., 2000). The spots with high Mn and Fe but low Zn may reflect episodes of sphalerite precipitation. This interpretation is supported by the low δ18O values of HZD at Meikle (see Stable Isotope Data section) and the high temperatures (200–250 °C) recorded by fluid inclusions in associated sphalerite and quartz (Lamb, 1995).

To provide an indication of the distribution of the LA-ICP-MS intensity ratios obtained from HZD, a bar and whisker plot was constructed that displays the full range and the 10th, 25th, 50th, 75th, and 90th percentile values for each element (Fig. 16). This plot shows that most trace elements have intensity ratios that span more than three orders of magnitude and a few that span more than six orders of magnitude (Mo, Ag, Sb, and Au).

R-mode factor analysis was employed to identify element associations, or factors, in the multi-element LA-ICP-MS data generated on HZD from 12 HZD localities. A seven-factor model (Table 2) was selected because it accounts for the total variance and produced geologically meaningful element associations. The relative contribution of an element to each factor is quantified by the loading value. The percent of the total variance accounted for by each factor and the major contributing elements are listed in decreasing order at the bottom of Table 2. The intensity of a particular factor in a sample is quantified by the factor score and is available in Supplemental File 1 (see footnote 1). Below we consider each factor and try to identify those that are due to element substitutions in dolomite and those that reflect the presence of mineral inclusions in dolomite. We go on to identify the HZD localities with the highest scores for each factor and attempt to infer their geologic basis.

Factor 1, with high loadings for Yb, Er, Dy, Y, Nd, La, Gd, Ga, Sm, Ba, In, Ce, Sr, Ni, and Sn, is dominated by rare earth elements (REEs) as well as Sr and Ba that commonly substitute for Ca or Mg in carbonate minerals (e.g., Veizer, 1983). The absence of S in this factor precludes residence of Ba and Sr in sulfate mineral inclusions. Likewise, the absence of P in this factor precludes residence of REEs in phosphate mineral inclusions. Given the common occurrence of this suite of elements in HZD and the lack of petrographic or SEM evidence for mineral inclusions composed of these elements, we infer that they reside in the dolomite crystal lattice. This interpretation is supported by evidence that REEs commonly activate CL in calcite and dolomite (Habermann et al., 1996). HZD from Pequop Summit has the highest Factor 1 scores and it has dolomite crystals with luminescent cores (Fig. 11).

Factor 2, with high loadings for Pb, Sb, Ge, Zn, Cu, and Ag, contains elements that are commonly enriched in Sedex and MVT Pb-Zn deposits and pluton-related polymetallic replacement deposits. The absence of S in this factor suggests it is not due to the presence of sulfide mineral inclusions. SEM data show that Zn- and Pb-rich growth zones or rims are present in HZD from the Blackwater and Queen of Sheba MVT deposits in Death Valley and the Mineral Hill polymetallic vein and replacement deposit in Nevada (Fig. 9). Similar Zn- and Pb-bearing HZD is associated with episodes of mineralization in mid-continent MVT deposits (Rowan, 1986).

Factor 3, with high loadings for As, Fe, V, and Au, contains elements that are common in pyrite (Large et al., 2009). Diagenetic pyrite may contain V (Large et al., 2009), whereas As and Au are characteristic of ore-stage pyrite at Meikle (Emsbo et al., 2003) and other Carlin-type gold deposits (Hofstra and Cline, 2000; Cline et al., 2005). Although we selected what appeared to be fresh samples, HZD from the Meikle and Windfall Carlin-type gold deposits has the highest Factor 3 scores and exhibits a linear correlation between As and Fe+S that Diehl et al. (2005) attributed to inclusions of arsenian pyrite. Figure 17 shows that there are two linear arrays that represent a high-As and high-Au group of mineralized samples from Meikle and Windfall and a low-As, Au-poor group comprised of samples from the other localities.

Factor 4, with high loadings for Co, Bi, Mn, Sn, and Ni, contains an odd assortment of elements. Manganese is characteristic of the distal part of Sedex Pb-Zn and pluton-related polymetallic replacement deposits. Supergene Mn-oxides commonly sequester Co, Ni, and other transition elements. As cited above, Mn is a common activator of CL in dolomite. Based on evidence from natural and synthetic dolomite, Rosenberg and Foit (1979) suggest the following decreasing order of stability for transition metal substitutions in dolomite: Mg>Mn>Zn>Fe>Co>Ni>Cu. Bismuth and Sn also are transition metals, but typically occur in the proximal part of pluton-related hydrothermal systems. HZD from the Mendha polymetallic replacement deposit in the Pioche district has two populations of Factor 4 scores; one similar to most other localities and one that is very high (also see Fig. 15A). Thus we infer that Factor 4 may be due to the superposition of Mn-rich dolomite overgrowths (with Co, Bi, Sn, and Ni) produced by the Pioche pluton-related hydrothermal system on older HZD, as described in the Leadville district in Colorado (Taylor, 1999), or, weathering of HZD and concentration of transition elements in Mn-oxides that occupy microfractures.

Factor 5, with high loadings for P, U, Sc, and Th, contains elements that commonly occur in apatite, which is a common detrital, diagenetic, and hydrothermal phosphate mineral. Thus we infer that this factor reflects the presence of apatite inclusions. HZD with the highest Factor 5 scores was from Meikle, Windfall, and Goodsprings.

Factor 6, with high loadings for K, Rb, Ti, Zr, and Nb, contains a group of chemically incompatible elements that are common in detrital minerals such as K-mica, rutile or anatase [TiO2], and zircon [ZrSiO4]. In comparison to the white layers, the gray replacement layers are high in K and generally contain more clay minerals inherited from limestone protoliths. Elsewhere, Roure et al. (2005) report authigenic as well as detrital clay minerals between dolomite crystals. Ti and Nb generally exhibit positive linear correlations with the same slope (Diehl et al., 2005). The linear correlation may reflect the proportion of Ti and Nb in rutile and the amount of rutile in HZD as it is among the most stable detrital minerals in sedimentary systems and is known to contain Nb and other elements such as Al, Si, V, Cr, Fe, Zr, and W (Zack et al., 2004).

Factor 7, with a high loading for Mo, was the only single-element factor. The absence of S in this factor suggests it is not due to inclusions of molybdenite and may therefore be present in the dolomite lattice. Molybdenum is commonly enriched in black shales and granite-related hydrothermal systems. HZD from Meikle and Laketown have the highest Mo. Meikle is in a platform margin setting with HZD alteration below a Sedex horizon in black shale (Emsbo et al., 1999) and it is adjacent to Jurassic and Eocene felsic dikes. Laketown is in an inner shelf setting far from black shales or granitic intrusions. While the high Mo at Meikle is reasonable, that at Laketown is difficult to explain.

We surmise that the trace elements in Factor 1 (Yb, Er, Dy, Y, Nd, La, Gd, Ga, Sm, Ba, In, Ce, Sr, Ni, and Sn), Factor 2 (Pb, Sb, Ge, Zn, Cu, and Ag), Factor 4 (Co, Bi, Mn, Sn, and Ni), and Factor 7 (Mo) substitute into the dolomite lattice whereas the elements in Factor 3 (As, Fe, V, and Au), Factor 5 (P, U, Sc, and Th), and Factor 6 (K, Rb, Ti, Zr, and Nb) likely reside in micron to submicron mineral inclusions in dolomite. Substitution of Factor 2 elements (Zn and Pb) in HZD is supported by SEM studies that found zincian dolomite with detectable Pb. At most localities, Factors 1 and 2 are reasonably attributed to HZD that formed from basinal brine. Factors 5 and 6 may be inherited detrital or diagenetic minerals from the host rocks, and possibly, minerals that precipitated with HZD. The clear exceptions are Factors 3 and 4 that reflect overprints by Eocene Carlin-type gold systems at Meikle and Windfall and the Cretaceous pluton-related polymetallic system at Mendha. The dolomite rims with strong Factor 2 scores at Mineral Hill also are likely due to overprinting pluton-related mineralization.

Stable Isotope Data

Table 3 and Figures 18A and 18B display the stable isotopic data obtained from Great Basin HZD localities. HZD in Ediacaran rocks in southwest Nevada and the Death Valley region of California have negative δ13C PDB values and δ18O SMOW values between 19–24‰. HZD in Cambrian rocks have δ13C PDB values near 0‰ and δ18O SMOW values between 20–25‰. Isotopic data from Devonian rocks extend from positive to negative δ13C PDB values and have the widest range of δ18O SMOW values, from 16 to 28‰ (Fig. 18A). HZD in Mississippian rocks have positive δ13C PDB values with δ18O SMOW values between 20 and 23‰ (Fig. 18A). For all localities and different stratigraphic units and ages, there is no significant difference in the isotopic composition of the gray replacement and white void-filling dolomite layers (Fig. 18B). Nor are there any evolving trends in δ13C PDB or δ18O SMOW values by latitude or longitude (Figs. 19A and 19B).

The wide range of isotopic compositions is similar to that for global HZD occurrences and is lower than typical marine limestones (Fig. 20A). Figure 20B shows that the δ18O and δ13C values of HZD are lower than marine biogenic calcite deposited in the Upper Devonian and Upper Ordovician, when many Sedex barite deposits formed, and significantly lower than corresponding values inferred for sabkha or reflux dolomites deposited from evaporated seawater. If mixing between evaporative brine and meteoric water was insignificant, then the isotopic data are consistent with formation temperatures of 50 to 150 °C, which are commonly obtained on fluid inclusions in HZD elsewhere (Leach et al., 2005; Davies and Smith, 2006). Assuming a geothermal gradient of 30 °C per km, such temperatures would require circulation to depths of 2–5 km. The isotopic composition of HZD at Meikle suggests it formed at higher temperatures than most other localities, which is supported by fluid inclusion data on saline (9–20 wt% NaCl) inclusions in sphalerite and quartz that yielded temperatures of 200–250 °C (Lamb, 1995). The one HZD sample in Ediacaran rock that plots well to the left of this range is suggestive of a higher temperature in excess of 250 °C, or, mixing with meteoric water. Most of the HZD samples have δ13C values that fall within the range of precursor marine limestones for each period (Veizer et al., 1999), suggesting the brines were buffered by carbonate carbon during dolomitization. An exception might appear to be the low δ13C values from the Neoproterozoic rocks (Fig. 18A), but the primary negative δ13C excursions in the Noonday Dolomite may be due to glacial or other oceanographic events in that time period (Corsetti and Kaufman, 2003). Other samples have δ13C values that are lower than precursor marine limestones consistent with oxidation of organic matter in the rocks (Allan and Wiggins, 1993). Some of the Neoproterozoic samples lie along an inclined array that is parallel to that calculated for dolomite in equilibrium with unbuffered brine (Fig. 20B). The similarity between the early fine-grained gray replacement layers and the later white coarse-grained vein-filling dolomite cements indicates both were deposited at the same temperature with little or no change in the isotopic composition of the brines. The absence of any oxygen isotope trends as a function of latitude or longitude that (Fig. 19) make it impossible to infer any unidirectional changes in temperature or direction of brine migration.

The isotopic data mainly reflect primary variations in the isotopic composition of seawater and evaporative brine, deep circulation to attain temperatures of ∼50 to 150 °C, buffering by variable proportions of rock carbonate and organic carbon, and perhaps local precipitation from relatively unbuffered brine in major conduits. The broad spread of Devonian isotopic data may indicate a more complex history with multiple sources of dolomitizing fluids. Perhaps with more data points in the deeper aquifers it would be possible to document spatial trends, such as those documented by Qing and Mountjoy (2004) across the Presqu'ile barrier reef in British Columbia, and elucidate brine migration vectors.

DISCUSSION

Age of HZD

HZD is epigenetic, i.e., younger than the rocks in which it formed. Carbonate rocks that host zebra dolomite are Ediacaran to Early Mississippian in age (Diehl et al., 2005; this study). Interrogation of the USGS Lexicon of Geologic Names (http://ngmdb.usgs.gov/Geolex/geolex_home.html) reveals that formations in the Great Basin with dolomite or dolostone in the name also are Late Proterozoic (Ediacaran) to Mississippian in age. This broad correlation in age of carbonate deposition is consistent with paleomagnetic data and plate motions that suggest the Great Basin was at equatorial latitudes from Ediacaran to Early Jurassic time (Fig. 2A, www.scotese.com). Thus, brines could have been generated by the evaporation of seawater over this time period, facilitating the widespread formation of dolostone and scattered occurrences of HZD (Fig. 3).

The best age constraints on HZD come from localities where it is spatially and genetically associated with Sedex deposits. In the northern Carlin trend at the Meikle and Rodeo gold mines, discordant HZD in the Silurian-Devonian Bootstrap Limestone formed along coeval high-angle faults below seafloor vents in the Upper Mudstone unit of the Late Devonian Popovich Formation where Sedex Au and barite deposits formed (Emsbo et al., 1999; Emsbo and Hofstra, 2003). The HZD at the Meikle Mine is therefore Late Devonian in age. In the southern Carlin trend at the Rain gold mine, discordant HZD in the Late Devonian Devils Gate Formation is overlain by Sedex barite in the Early Mississippian Webb Formation (Thoreson, 1991; Longo et al., 2002; F.G. Poole, USGS, 2007, personal commun.). Likewise, we interpret the HZD in the Rain mine to be Early Mississippian in age. If these examples are meaningful, then HZD at other localities may have formed at about the same time as many other Sedex deposits that occur in slope and basin facies. Papke (1984) and Poole (1988; F.G. Poole, USGS, 2005, personal commun.) have shown that Sedex barite deposits in the Great Basin occur in Cambrian through Mississippian rocks with a greater number of large deposits in the Late Ordovician and Late Devonian. If all of these Sedex deposits were produced by the discharge of hot evaporative brine into marine basins, then a great deal of the HZD in the adjacent carbonate platform may have formed along brine flow paths during these episodes of mineralization. This interpretation is supported by the large number of HZD localities in Devonian rocks. Brine migration and formation of HZD and Sedex deposits appears to have been contemporaneous with episodes of extensional (or transtensional) faulting along the passive margin as well as contraction during the Early Mississippian Antler orogeny. Some of the faults that localized HZD (e.g., Rain) may have moved in response to development of the Antler forebulge (Morrow and Sandberg, 2008).

The age of HZD associated with MVT Zn, Pb, and Ag deposits in southwest Nevada and southeast California is more difficult to determine. The small MVT deposits in the Early Mississippian Monte Cristo Formation in the Goodsprings district may be related to the Antler orogeny or younger deformational events. The small MVT deposits and HZD in the Ediacaran Noonday Dolomite may be about the same age as MVT deposits in the neighboring Goodsprings district, or much older. The Sedex barite and underlying HZD at Rain and the MVTs in the Goodsprings district suggest there may have been a significant pulse of Early Mississippian brine migration during the Antler orogeny, as surmised for some of the Sedex and MVT deposits in the Selwyn Basin further north (Nelson et al., 2002, 2006).

Structural controls on HZD

Several HZD localities are in well-known mineral belts or mining districts (Fig. 1; Table 1), which are interpreted to be underlain by crustal structures that were repeatedly reactivated over Phanerozoic time. For example, the HZD localities in the Carlin trend and Battle Mountain-Eureka trend (BMET) are thought to be underlain by northwest-striking faults inherited from Ediacaran rifting of Rodinia that influenced subsequent patterns of sedimentation, deformation, magmatism, and hydrothermal activity (Hofstra and Cline, 2000; Emsbo et al., 2006). The HZD at Meikle, in the northern Carlin trend, is localized by north-northwest–striking growth faults (Fig. 21A) that also served to localize younger Jurassic and Eocene dikes and Eocene high-grade Carlin-type mineralization that overprinted the HZD (Emsbo et al., 1999, their fig. 2; 2003). The HZD at Rain, in the southern Carlin trend, is in a flower structure (Fig. 21B) along a northwest-striking dextral fault system that localized younger Tertiary igneous dikes and Carlin-type gold mineralization (Williams et al., 2000). The HZD at Rose, Diamond, Windfall, and Achilles in the Eureka district is along a north-striking fault system that may have dilated during dextral translation along the northwest BMET in the Paleozoic and subsequently localized Cretaceous and Eocene dikes and corresponding polymetallic and Carlin-type gold deposits (Dilles et al., 1995; Vikre, 1998). The HZD in the Alligator Ridge district occurs near the intersection of the southern end of the Carlin trend (Bida trend of local usage) and a Jurassic or older north-striking fault system that localized Carlin-type deposits during Eocene extension (Nutt and Hofstra, 2003). The HZD at Mendha in the Pioche district is at the west end of the east-northeast Pioche-Marysvale mineral belt where it is cut by Cretaceous and Oligocene intrusions and polymetallic replacement deposits (Vikre and Browne, 1999). Hence, there is ample evidence that HZD was frequently localized by dilatant faults along crustal structures early in the tectonic evolution of the Great Basin. Such structures also have been shown to localize HZD in the other basins associated with Sedex and MVT deposits and petroleum reservoirs (Davies and Smith, 2006). This underlying structural control in combination with reactivation during subsequent events explains the superposition of Mesozoic and Cenozoic mineralization on Paleozoic HZD.

CONCLUSIONS

The Great Basin is underlain by an extensional segment of the Cordilleran miogeocline produced by the Ediacaran breakup of the Rodinian supercontinent. The structural underpinnings and deformational history of the Great Basin are very similar to the Selwyn Basin farther north in Canada and both contain an array of Sedex and MVT deposits and related HZD occurrences. However, their ore deposits differ. Sedex and MVT Pb-Zn deposits are more numerous and larger in the Selwyn basin whereas Sedex barite deposits are more numerous and larger in the Great Basin. Sedex Au is also present, and Pb-Zn deposits of either MVT or Sedex type are relatively small. The evidence collected from HZD occurrences in the Great Basin, summarized below, explains these differences.

Dolostone and occurrences of HZD are widely distributed in Ediacaran to Mississippian Formations across the carbonate platform of the miogeocline. Over this period of the development of the carbonate platform, the Great Basin was at equatorial latitudes (<30°) which would have permitted the generation of brines in restricted lagoons and playas by the evaporation of seawater. Evaporative brines are dense and tend to descend and accumulate in permeable rocks in sedimentary basins unless they are put into motion by tectonic, buoyant, or magmatic events (Garven and Raffensperger, 1997). A significant proportion of the dolostone in the Great Basin may have been produced by reactions between evaporative brine and lime sediments or underlying limestones as brines descended and migrated seaward. The widespread occurrences of HZD in the Great Basin also requires brine movement on a large scale but with circulation to greater depths and more localized discharge along high-angle faults. The oxygen isotopic compositions of HZD are consistent with formation temperatures of 50–150 °C requiring circulation to depths of 2–5 km, or more. The stratabound occurrences of HZD in Ediacaran and Cambrian strata are consistent with such depths and may mark basal aquifers of lateral flow. The discordant occurrences of HZD in Devonian strata may mark upflow zones along high-angle faults. Most HZD localities are located between eastern shallow marine facies, where evaporative brine could have been generated, and western Sedex deposits, where heated brines discharged along faults into platform margin, slope, and basin facies. The MVT deposits in the southern part of the Great Basin are on the eastern side of the platform consistent with flow from Antler orogenic highlands toward the inner shelf. If HZD was coeval with the major episodes of mineralization in Sedex and MVT deposits, then it is mostly Late Ordovician, Late Devonian, and Early Mississippian in age, based on the ages of those mineral deposits. The former two episodes coincide with pulses of extension along the platform, and the latter with the Antler orogeny (e.g., Poole et al., 1992). Deep-seated structures controlled the formation of both HZD and younger mineral deposits, thus explaining the spatial coincidence of the two features in many areas. The HZD localities with high Fe are ideal chemical traps for gold in younger Carlin-type deposits (e.g., Meikle, Emsbo et al., 2003).

The general trend of increasing Zn with increasing Mn and Fe in HZD and a corresponding decrease in δ18O values could be used in exploration to identify areas that were invaded by hot brine and thus have increased potential for Sedex or MVT Pb-Zn deposits. There appears to be a broad chemical difference in HZD occurrences between the southern and northern parts of the Great Basin. In the southern part, MVT Zn-Pb-Ag mineralization dominates, and zincian dolomite rims are readily apparent. In the northern part, Sedex barite and Au are dominant and zincian rims were not detected (although Zn is elevated in Fe-rich cores at Meikle). The exception is the Pb-Zn-Ag deposit at Mineral Hill, where zincian dolomite rims may be due to pluton-related hydrothermal activity. The low Mn, Fe, and base metals in many northern HZD localities may be a reflection of elevated H2S concentrations in the brines which would have suppressed the solubility of Fe and base metals and enhanced the solubility of Ba and Au that are present in associated Sedex deposits. Such a difference may be related to the reduced nature of the sedimentary sequence in the Great Basin (Hofstra and Emsbo, 2007). The HZD produced by the array of Paleozoic climatic, tectonic, and hydrothermal processes was permeable, reactive, and located along structures utilized by subsequent hydrothermal events making it an ideal host for younger ore deposits.

This project was funded by the U.S. Geological Survey Mineral Resources Program. Stable isotope data were collected in the Central Region Stable Isotope Geochemistry laboratory, Denver Federal Center, Colorado. Figure 3 was created in GIS by Paul Denning, U.S. Geological Survey.

1Supplemental File 1. Excel data file of laser ablation spot analyses used for factor analysis. If you are viewing the PDF of this paper or reading it offline, please visit http://dx.doi.org/10.1130/GES00530.S1 or the full-text article on www.gsapubs.org to view Supplemental File 1.

Supplementary data