Placer gold deposits can serve as a wet/dry paleoclimate indicator because they preserve a record of physical and chemical modification. We hypothesized that such a record would be present across the four main placer units at Rich Hill, Arizona, USA. The oldest placer unit, a paleo-erosional surface on a modern topographic high, records moderate precipitation and erosion with modest transport distance following 25–22 Ma unroofing of the lode gold source. By 17–5 Ma, high-angle Basin and Range faulting produced a shallow basin that preserved three additional placer units. The oldest is a thin gold-rich gravel within bedrock gravity traps, suggestive of a steep gradient and abundant nonseasonal precipitation. The middle unit has well-rounded gold nuggets with deep chemical weathering that record abundant nonseasonal precipitation, but with a less steep gradient and significant fluvial sediment deposition. The uppermost unit is a pulse placer unit deposited by a series of landslides and debris flows during a period of lower, seasonal precipitation. During this dry period, and continuing to the present, microbial communities may have been established within the seasonally wet bedrock traps of the lowermost placer unit. This resulted in biological modification of placer gold chemistry and production of Mn-Ba–oxide biomats, which have coated and cemented both placer gold and sediments. This record of physical and chemical change within a sequence of placer gold units is consistent with known Arizona paleoclimate conditions, and it demonstrates the potential utility of this technique for paleoenvironmental reconstructions.
The element gold occurs in natural deposits in native form, commonly alloyed with a minor amount of silver and trace amounts of copper and other elements. The occurrence of these native gold alloys (hereafter referred to as native gold) within veins of epithermal deposits is commonly interpreted as having a hypogenic hydrothermal origin (e.g., Boyle, 1979; Saunders et al., 2008, and references therein). However, the origin of placer native gold is not as well understood. Placer gold has been attributed to primary, secondary, or a combination of primary and secondary formation processes (e.g., Hough et al., 2009). In the classic interpretation of primary placer formation, it is assumed that a gold mass forms during a hypogenic hydrothermal event and survives the physical and chemical weathering stage as large fragments due to the low reactivity and malleability of gold (e.g., Lindgren, 1911; Chapman et al., 2010, 2011). However, high-purity gold overgrowths, bacterioform gold, and nanoparticle biofilms suggest a secondary origin for some placer gold (Watterson, 1994; Mossman et al., 1999; Reith et al., 2007, 2010; Southam et al., 2009). Other authors reject the notion of low-temperature or biological nugget growth for some deposits in favor of an abiological, high-temperature origin within lode deposits (e.g., Hough et al., 2007). The recent consensus of many workers is that the processes of both primary and secondary models likely play a role in placer gold formation, with the specific contribution of each end member varying from location to location. These workers also hypothesize that biology likely plays a role in aggregating large nuggets from fine-grained lode gold sources (e.g., Reith et al., 2010).
As native gold is usually alloyed with trace levels of lead (Pettke and Frei, 1996; Kamenov et al., 2007), it is possible to measure the isotopic composition of lead alloyed in lode and placer native gold to provide key indications regarding the potential source(s) and process(s) that contributed to the formation of the placer and lode gold deposits (e.g., Kamenov et al., 2013; Standish et al., 2014). The placer gold from the black and Potato Patch placer units of the Rich Hill Mining District of Arizona have Pb isotopic signatures that are far more radiogenic than gold from the inferred epithermal lode source (Figs. 9 and 10 in Kamenov et al., 2013). Significantly, placer gold from these same units was observed by the same authors to have potential biofilm coatings. The authors concluded that the most plausible scenario to explain the lead isotope signature of placer gold from Rich Hill is the addition of appreciable amounts of secondary gold within the placer environment (Kamenov et al., 2013).
We hypothesized that the characteristics of placer gold at Rich Hill would provide a record of past climatic conditions, and so we examined the geochemical and morphological characteristics of placer gold from Rich Hill, Arizona. This study combined geochemistry, sedimentology, economic geology, and paleoclimatology to address a common issue. The physical and chemical characteristics assisted with understanding the lode origins, weathering history, possible biological modification, and transport history of placer gold within all four of the distinct placer units at this locality.
History and Geological Settings
Official placer gold production of the Rich Hill district (Fig. 1) is 100,000 troy ounces (e.g., Blake, 1899; Tenney, 1933; Hall, 1934; Wilson, 1952), with much of this gold occurring as very large multi-ounce nuggets (Supplemental Materials S-11; Crombie et al., 2002). Modern mining operations by JW Mining and others have yielded hundreds of troy ounces of gold, including over 100 nuggets >1 troy ounce (Supplemental Materials S-1 [see footnote 1]), with one confirmed nugget recovered in 2010 weighing 19 troy ounces. Lode gold at Rich Hill has been dominated by production of the Octave (including the Joker Shaft) and Beehive Mines from the Octave-Beehive vein (dark line, Fig. 1).
Rich Hill district bedrock geology consists of metavolcanic and metasedimentary rocks of the 1.8–1.74 Ga Yavapai Supergroup (e.g., Anderson, 1989; Karlstrom et al., 2001) and the “Granite of Rich Hill” (DeWitt, 1989). These metamorphic and granitic units are cut by diabase dikes that are mineralogically and chemically similar to the 1.1 Ga diabase veins found in other locations throughout central Arizona (Wrucke, 1989). The diabase dikes are closely associated with elevated gold where they border lode veins. Although tentatively classified as Tertiary in age by Keith et al. (1983), vein deposits in the Rich Hill and nearby Martinez metallic mineral districts are possibly middle Proterozoic in age and related to emplacement of the diabase dikes. The lode gold mineralization in the Rich Hill district must be younger than 1.4 Ga and older than 103.7 Ma, based on available data (DeWitt, 1989).
Quartz veins containing pyrite, galena, minor chalcopyrite, minor sphalerite, native gold, and electrum are typical of deposits in the Rich Hill district (e.g., Metzger, 1938). The main lode is the Octave-Beehive vein, which extends over a distance of ∼1.6 km (Fig. 1). Native gold occurs mainly as microscopic particles, and it is rarely seen in the lode deposit. The apparent disconnect between the paucity of visible native gold in the lode and the large size of local placer nuggets was not lost on early workers (Nevius, 1921). More recent work (Kamenov et al., 2013) has suggested that biological activity may have had a role in aggregation of these large nuggets from a fine-grained lode source. A detailed description of the lode mineralogy and chemistry was provided by Kamenov et al. (2013).
Placer gold occurs within an ∼16 km2 area loosely centered upon these vein deposits. While placer gold does occur within the sediments of the modern, active channel of Weaver Creek, most of the placer gold occurs within older placer units of the Devils Nest sequence. The thickest of these units, ranging from a few meters to over 15 m thick, is the “red placers” (Crombie et al., 2002). Red placers are colored by a red smectite clay matrix enclosing locally derived well-rounded to subangular lithic clasts that range in size from sand to boulders >1 m in diameter. The red placer unit is likely a series of ancient debris flows and landslides, based upon field observations, including the jumbled and unstratified nature of the deposit. Gold particles, ranging from 0.1 mm with a mass of a few milligrams to fist-sized nuggets with a mass measured in kilograms, occur throughout the red placers (Supplemental Materials S-1, S-2, and S-3 [see footnote 1]; Kamenov et al., 2013). The average grade of the red placers is 1–2 g gold/m3, with gold being up to 10× more concentrated in the bottom 10 cm of the deposit.
The underlying middle “white placer” unit (Fig. 4 in Kamenov et al., 2013) is typically 1–3 m thick, averages 5–10 g gold/m3, and consists of larger and much more rounded gold particles than those found in the overlying red placers. Though this unit is often called the “Caliche” by local miners, it typically contains little calcium carbonate. Study of this unit by scanning electron microscope (SEM) and energy dispersive spectral analysis (EDS) has revealed that the white appearance is due to bleached cobbles and sediments that are cemented by sericite and highly silicified material (Kamenov et al., 2013).
Beneath the white placers, there are thin gold-bearing gravels cemented by black manganese-iron oxide (janggunite) and barium manganese oxide (hollandite) precipitate (Kamenov et al., 2013). These “black placer” deposits average 1.5–4 g gold/m3 and contain elongate but flattened gold particles that are nearly completely encapsulated by manganese-iron oxide up to 80 μm thick (Fig. 4 in Kamenov et al., 2013). These black gravels typically occur as thin seams <1 m thick near bedrock, and as fill material for the locally deepest potholes in bedrock.
Samples of placer gold were recovered from placer deposits with a Keene Engineering vibrostatic dry washer for smaller pieces (0.1–0.5 g) and a MINELAB GP-Extreme metal detector for larger gold nuggets (>0.75 g). These placer samples represent individual gold grains and nuggets (nuggets defined as masses weighing >1 g or measuring >4 mm) recovered by, or in the presence of, one or more of the authors and are not bulk bullion samples or amalgamated placer gold samples. Samples for geochemical and morphological characterization were not cleaned prior to analysis. Specific sample localities are not provided due to security issues.
SEM imaging of representative gold grains mounted on carbon tape was performed with a Cameca SX-100 at the University of Arizona. The same instrument was used for probe measurements of Au and Ag major- and minor-element abundances, using standard off-peak interference and matrix corrections (Armstrong, 1988; Donovan et al., 1993). The standard deviation associated with the in situ trace-element microprobe analyses was less than ±2 wt%. The detection limit was 0.25 wt% for gold and 0.10 wt% for silver.
Samples for trace-element microanalysis were embedded within a low-volatile epoxy and polished to expose grain interiors using sequential grits down to a 0.3 μm diamond abrasive. Great care was taken to minimize sample preparation artifacts such as deposition of polishing debris into fractures or pores, localized smearing, which can obscure internal chemistry or physical features, and cross contamination (e.g., Knight and McTaggart, 1989; Douma and Knight, 1994). Analysis was performed using a ThermoFinnigan ELEMENT2 high-resolution inductively coupled plasma–mass spectrometer (ICP-MS), coupled to a CETAC LSX-213 laser-ablation (LA) system housed within a class 100 clean laboratory at the University of Arizona. The LA system delivers high-intensity 213 nm, 5 ns laser pulses at rates of 1–20 Hz, with analysis spot size of 25 μm. The standard deviation associated with the in situ LA-ICP-MS analysis of Cu is less than ±10% of the total measured abundance, with a detection limit of 3 mg/kg.
Elemental analysis of larger gold nuggets was performed using a NITON XL3t GoldD+ handheld X-ray fluorescence (XRF) instrument. The Cu/Zn Mining Mode was used for each analysis, with a dwell time of 240 s. The instrument Metals Mode was not used because our tests suggested that it had lower reproducibility, higher error, and did not test for all of the trace elements anticipated in the samples. The instrument was calibrated to certified gold-silver-copper alloy reference standards, and instrument standard deviation was determined to be less than ±10%. Detection limits certified by the manufacturer calibration are 12 mg/kg for Ag, 25 mg/kg for Au, 20 mg/kg for Cu, 80 mg/kg for Fe, and 140 mg/kg for Mn. The handheld XRF units utilized in this study generate 50 keV X-rays. Calculation of the theoretical depth of penetration for these X-rays may be made using a modified version of the Lambert-Beer law, which suggests a maximum depth of 71 μm. However, the escape depth for the fluorescent X-ray signal (Lα and Lβ) would only be ∼10 μm.
Measurement of large placer gold dimensions was performed with Fisher-brand digital calipers, with an accuracy of ±0.03 mm (Supplemental Materials S-2 [see footnote 1]). The longest axis was measured as “length,” the longest axis perpendicular to this was measured as the “width,” and the short axis perpendicular to the previous two was measured as “thickness.” The surface roughness of the large placer gold nuggets was estimated visually using the general methods outlined in Krumbein and Sloss (1956), modified to reflect roughness of the surface rather than shape.
This study examined an unprecedented number of very large placer gold nuggets (39 samples) due to fortuitous circumstances created by modern placer mining operations and the willingness of several mine owners to permit scientific study of their gold. Only four large nuggets were recovered from the Potato Patch deposit, due to the extensive historical mining activity at that specific site. It was not practical to obtain additional samples for this study due to the elevated price of gold, difficulty in maintaining access to claims, and time required for obtaining new samples (up to 6 h per sample).
RESULTS AND DISCUSSION
To explore the possibility of placer gold recording paleoclimate information, three main procedures were used to examine the placer gold from Rich Hill. These included shape analysis, LA-ICP-MS and Cameca probe chemical characterization of smaller gold grains, and XRF chemical characterization of large nuggets. Results of shape analysis for large placer gold nuggets are given in Table 1, while XRF chemical characterization data for the same gold are given in Table 2. Table 3 presents the results of detailed LA-ICP-MS and probe element analysis of small placer gold grains. Full data sets for Tables 1–3 are presented as tables in the supplemental materials (see footnote 1). Table 4 is a summary of the generalized characteristics of placer gold from each major unit, highlighting the distinct characteristics that provide clues to past environments.
Placer Gold Shape and Surface Analysis
The shape of sedimentary clasts is often characterized using the methods outlined by Zingg (1935) to understand their possible transport history. The Zingg plot of width/length versus thickness/width ratios for 39 large Rich Hill placer gold nuggets quantifies the distinctive shapes of the gold from each of the main placer units (Fig. 2A). The surface roughness of the same 39 large nuggets was characterized using the general principles outlined in Krumbein and Sloss (1956), and the resulting classification of surface roughness is presented in Figure 2B.
Red and Black Placer Gold
Gold samples from the red placers form a distinctive population in Zingg plot space, ranging linearly from discoid to blade form. Gold samples from the black placers appear to form an end-member continuation of this trend, within the blade form field. The red and black placer nuggets all have a distinctly flattened form, and angular to subangular surface texture. A significant amount of the surface roughness in the red placer samples is produced by cast imprints of quartz and pyrite, remnant skeletal octahedron gold crystal faces that occur as chevron points, and patches of striation marks indicative of glancing boulder impacts or transportation in traction (Supplemental Materials S-3 [see footnote 1]) typical of debris-flow pulse placers (McCulloch et al., 2003). Striation marks are dull and weathered, not shiny and fresh as observed when damaged during mining. Black placer samples have a surface roughness that appears etched and porous. Placer gold from the red placers is interpreted as having formed as tabular bodies within the lode source, which then experienced minimal physical/chemical/biological modification during rapid postexposure transport during debris-flow events. However, invoking such an explanation for nugget shape is problematic given the virtual absence of visible native gold within the local lode rocks as observed over the years, including the early years of mining activity (e.g., Nevius, 1921; Metzger, 1938), when native gold masses were typically encountered. It is possible that this placer gold originated within the upper portions of the local vein systems, which subsequently experienced deep and rapid erosion due to Basin and Range uplift of the Weaver Mountains. It is also possible that known local lodes are not the source of this gold, and that its source lies elsewhere, perhaps buried under the deep alluvial fill beyond the nearby range front. Black placer gold may have a similar source and history as the gold within the red placers, but it was subjected to more pronounced physical/chemical/biological modification. The black placer gold is found within bedrock potholes and deep channels, which typically serve as traps for placer gold transported in traction. Traction is known to produce more elongate forms in ductile materials (e.g., Nutting and Nuttall, 1977; Macdonald, 2007), similar to the elongate forms observed for the nuggets of the black placers. Following deposition within these traps, the observed physical/chemical/biological modification that produced the surface etching and porosity would have obscured any surface textures such as striations, which would have hinted at past traction transport.
White and Potato Patch Placer Gold
White placer gold nuggets are significantly less flattened and dominantly plot within the cylindrical to spheroid (three-dimensionality) fields, with a low surface to volume ratio (Fig. 2A). These nuggets have a very smooth and rounded surface texture (Fig. 2B), with an appearance similar to water-worn nuggets from the rivers of some California and Alaskan placer districts. Three of the four nuggets from the Potato Patch have similar form to the white placer nuggets, with the fourth nugget being distinctly discoid, similar to some red placer nuggets. This last nugget is interpreted to have been remobilized by local erosion and modified during the period of red placer development. The Potato Patch nuggets have more surface roughness than the white placer gold, mostly due to the differential weathering of quartz inclusions. The white placer and Potato Patch nuggets are interpreted as having experienced significant postexposure modification, likely due to physical and/or chemical weathering.
In the simplest interpretation, their smooth surfaces and rounded forms likely reflect abrasion and reworking in a fluvial environment. However, it is possible that the rounded forms reflect chemical weathering, specifically in the white placers. The mineralogy of the white placer sediments may be interpreted as part of a fossil low-temperature (<200 °C) hydrothermal system. Weak to moderate silicification and sericitation (boulder bleaching) of these sediments suggest the emergence of a hot-spring system within the placer unit. Similar low-temperature hydrothermal alteration of a placer unit has been noted for the White Channel sediments of the Klondike district in Canada (Tempelman-Kluit, 1982; Dufresne et al., 1986; Dufresne, 1986). At Rich Hill, it is possible that in addition to causing the observed quartz-sericite alteration, the chemically and thermally aggressive fluids of such a system could have modified preexisting placer gold nuggets to produce smoother and rounder forms.
There is no evidence of such alteration of sediments and placer gold from the Potato Patch. Rather, the Potato Patch placer and its sediments are interpreted as having substantially longer surface residence times, because they occur along a paleo-erosional surface. The sediments of the Potato Patch occur as a planar deposit on the top of Rich Hill itself, and they contain many “exotic clasts” of rock types not found in the local bedrock. Placer gold from the Potato Patch frequently contains inclusions of distinctive purple quartz that is not found in local lode deposits or the gold of the other placer units. The purple quartz may be fragments of the original lode deposit source, which was distinct from the known local lodes, or it may represent an artifact of prolonged chemical weathering or biological modification of the quartz. We hypothesize that a prolonged period of chemical and physical weathering was responsible for the distinctive shape and surface texture of the Potato Patch gold nuggets due to the physical rounding and leaching of more mobile elements.
Handheld XRF Geochemistry of Placer Gold
Measurements of minor- and trace-element abundances were performed by handheld XRF on 39 large gold nuggets (Table 2). Despite having lower elemental detection limits than other methods, this analytical instrument was selected based upon its nondestructive analysis. This permitted elemental fingerprinting of irreplaceable museum-quality placer gold samples that otherwise could not have been analyzed by destructive techniques. Of the wide range of elements the XRF unit is capable of measuring, only Ag, Cu, Mn, and Fe were present above detection limits in the gold alloy of these samples. A plot of silver versus copper reveals distinct populations for each placer unit (Fig. 3A), demonstrating that geochemical fingerprinting with nondestructive portable XRF instrumentation can be an effective tool for classification of placer gold. Red placer gold has lower copper abundances and a distinctive range of elevated silver abundances within the nugget alloy. The gold from the white placer unit has the lowest combined abundances of silver and copper, suggesting deep leaching of these mobile elements by chemical weathering or in situ gold aggregation driven by biochemical processes. The black placer gold has low silver, but moderate copper abundances. Potato Patch gold has low silver abundances, but the highest copper abundances. The elevated abundances of mobile elements such as silver (red placers) and copper (Potato Patch) reflect either two chemically distinct lode sources, or biochemical processes occurring within the placer deposit. It would be reasonable to expect that the lower abundances of these same mobile elements in the placer gold from the white and black placer deposits represent prolonged chemical and/or biological modification of the nuggets from these same sources. The latter is confirmed by the distinct Pb isotope signal in the placer gold, which is very distinct from local lode gold (Fig. 3B), indicating growth/modification of the nuggets within the placer environment.
Manganese and iron are commonly associated with redox chemosynthetic bacteria in the shallow subsurface, particularly within deposits of metals (e.g., Ghiorse, 1984; Polgari et al., 1991; Thamdrup, 2000; Lovley, 2013) and aquifers (e.g., Weber et al., 2012). Both elements occur at elevated levels in the coating of the black placer gold (Fig. 3C). Potato Patch gold has slightly elevated manganese and iron levels. The red placers contain variable concentrations of iron, probably bound within the red clay present in deep pits and grooves in the gold. However, manganese levels are below detection limits (factoring error) for the red placer gold. White placer gold has the lowest levels of manganese and iron. Manganese minerals are not observed in any of the potential lode sources for the placer gold at Rich Hill. The combination of elevated Fe and Mn associated with the gold nuggets from the black placers hints at a potential biological origin because both elements are known to be intimately involved with chemosynthetic life within the shallow subsurface (e.g., Lovley, 2013), and these coatings have been intimately linked with biomat-like structures on the black gold (Kamenov et al., 2013).
Detailed LA-ICP-MS Geochemistry of Placer Gold
Quantification of copper and silver abundance within smaller grains of placer gold (<500 mg) was performed with LA-ICP-MS (copper) and probe (silver). Analyses were performed on both rims and cores, in cross sections of the grains exposed by polishing to approximately the volumetric middle of each grain (Table 3).
Two significant relationships are observed on the plot of copper versus silver for the finer-grained placer gold (Fig. 3D). First, there is a significant silver and minor copper depletion in the rims relative to the cores for the red, white, and black placer gold samples. Second, there is a diagnostic geochemical signature in both rims and cores for placer gold from each major placer unit. Core chemistry is assumed to correspond with chemistry of the gold from the lode source, while rim chemistry and thickness represent a subsequent chemical or biochemical modification that may hint at exposure time and environment of transport and deposition.
The occurrence of silver- and/or copper-poor (gold-rich) rims for placer gold from other districts has been well documented for over 100 yr, with an origin attributed to differential chemical weathering similar to depletion gilding (e.g., Lindgren, 1911; Desborough, 1970; Knight et al., 1999). Previous work at Rich Hill (Kamenov et al., 2013) documented the existence of these rims, but for fine-grained placer gold from the active channels of drainages below the present study area. This rim depletion in silver and copper for placer gold from the primary placer deposits shows well under backscatter electron (BSE) imaging (Fig. 4). This chemical zonation is not separated by a gradational boundary, but rather it is a distinct and sharp front.
Cores of the gold grains from each of the major placer units were found to have their own unique chemical signature (Fig. 3D). The cores of gold grains from the red placers generally had the most elevated silver values, and moderate copper values, with a distinct correlation between decreasing silver and decreasing copper. The cores of gold grains from the white placers appear to form a continuation of this trend, but at lower silver and copper values, suggesting a similar if not identical lode source for the gold from these two placer units. The differences in chemistry along the observed trend for the white and red placer gold may reflect a change in chemistry of the lode source as it was eroded to progressively deeper levels. The trend for the gold rims from these two units may result from their different weathering and transport histories, which are known from morphology studies described earlier in this paper.
The cores of gold grains from the black placers have silver and copper values similar to the cores of red and white placer gold, but with a pronounced shift of higher values off of the observed red-white core trend. This shift may represent chemical modification distinct from the processes that formed the red-white core trend. This would not be unexpected, as the black placers occur in bedrock potholes and low spots that would produce chemical weathering conditions different from the weathering environment experienced by the red and white placers. The high surface-to-volume ratio for this flat gold would facilitate modification of grain cores. The most distinctive gold grain core population is that from the Potato Patch, which has very elevated copper and very low silver in comparison to the gold grain cores from other placer units. While gold grains from the red, white, and black placers have a core chemistry that could plausibly come from the same source, the chemistry of the Potato Patch gold grain cores is so unique as to require a separate lode source or complete bio/geochemical modification within the placer. Furthermore, as mentioned already, the Potato Patch historically contained the largest multi-ounce nuggets in the area, also suggesting growth within the placer environment. Reported Pb isotopes in a Potato Patch nugget show very distinct values as compared to the local lode gold (Fig. 3B; Kamenov et al., 2013). This is consistent with nugget formation in situ, in the placer environment.
As expected, the rims of placer gold grains from the red, white, and black placers have pronounced silver depletion and slight copper depletion relative to their cores. However, the relationship between the rims and cores of gold grains from the Potato Patch is quite different, with rims depleted significantly in copper and only slightly in silver, relative to the cores. This may result from a very low original silver content of the lode source of the Potato Patch gold grains, as implied from data on the cores. It may also represent a biochemical history that is different from that observed in other placer districts and during the time of red, white, and black placer development.
It is readily apparent that measurements of minor and trace elements in gold from specific placer units made by portable XRF for the large nuggets and by LA-ICP-MS for the fine gold grains are quite different (Fig. 3). This discrepancy cannot be explained by size differences, as no such correlation was noted among the wide range of sizes in the gold nuggets. Rather, the discrepancy seems to be a result of the copper/silver-depleted rim thickness, and the depth of penetration by XRF analysis. Thus, caution is warranted for interpretation of XRF chemistry without some knowledge of internal nugget zonation.
Origin and Evolution of the Placer Gold Deposits
The physical, chemical, and biological factors that left their mark on the placer gold deposits at Rich Hill suggest preservation of a past transition from wetter to drier climate in Arizona. The exact timing of this transition is not quantified, but two independent lines of qualitative evidence provide some chronologic constraints. First, the geological evidence indicates that placer formation occurred after their lode sources were unroofed, possibly by low-angle detachment faulting at 25–22 Ma. The red, white, and black Devils Nest placers formed in a basin produced by high-angle Basin and Range faulting, which occurred between 17 and 5 Ma (e.g., Menges and McFadden, 1981; Eaton, 1982; Nations and Stump, 1981; Henry and Aranda-Gomez, 1992; Colgan et al., 2006). Second, there is general agreement between the pattern of climate indicated by the placer deposit record and the climate record established by other workers (Fig. 5). Many workers have documented that Arizona and much of the southwestern United States was significantly wetter and more humid during the Pliocene (5.3–2.6 Ma) than modern times, based upon sedimentation styles, pollen studies, and fossil evidence (e.g., Melton, 1965; Morrison, 1985; Thompson, 1991; Fleming, 1994; Smith, 1994; Remeika and Fleming, 1995). In particular, the work of Remeika and Fleming (1995) suggested that it was not only much wetter, but that the transition zone between the Basin and Range and Colorado Plateau experienced a period of extensive erosion and fluvial modification between 4.5 and 1.2 Ma. Some research suggests that climate, and not tectonics, was responsible for this period of sedimentation and basin fill (Smith, 1994). This suggests that tectonic uplift exhumed the supposed local epithermal source veins for the Rich Hill placer gold, and later climatic conditions controlled weathering and transport.
The Potato Patch placer deposit is stratigraphically older than the others, and it occurs within thin sediments of a paleo-erosional surface on a topographic high (Fig. 1). In addition to locally derived sediments, these deposits contain clasts of rock types not found in the local bedrock. An additional distinctive feature of this placer deposit is that most gold contains fragments of a purple quartz that has not been linked to any local lode source. The Potato Patch placer must have formed prior to the Basin and Range faulting that uplifted the Weaver Mountains and Rich Hill between 17 Ma and 5 Ma (e.g., Menges and McFadden, 1981; Eaton, 1982; Nations and Stump, 1981; Henry and Aranda-Gomez, 1992; Colgan et al., 2006). The Potato Patch may have formed after low-angle detachment faulting in central Arizona at ca. 22–25 Ma (e.g., Reynolds et al., 1986) unroofed its lode gold source. Structural, geochemical, and mineralogical evidence suggests that this source was not the same as the source of gold for the later red, white, and black placers of the Devils Nest units. The dissimilarity of rim and core geochemistry, thin and localized silver- and copper-depleted rims, and an abundance of large quartz inclusions suggest minimal transport distances and short transport time before deposition and burial within thin sediments of the paleo-erosional surface. However, Pb isotope data presented in Kamenov et al. (2013) for one Potato Patch nugget (Fig. 3B) suggest that these large nuggets have undergone extensive chemical modification in the placer environment. Aggregation of a major part of the nugget through addition of Au under supergene conditions is in agreement with the observed low Ag content (Fig. 3D). Potato Patch gold, which is the oldest deposit and exhibits less physical weathering, most likely formed before the 3.4 Ma transition to a significantly wetter climate (Fig. 5).
The red, white, and black placers of the Devils Nest formed next, on the shoulder of a basin produced after 17–5 Ma Basin and Range faulting, providing a maximum age for their deposition. These deposits probably formed during the wetter period of Arizona’s history. Physical characteristics of the black placer gold (stratigraphically oldest Devils Nest unit), and the mechanism of gravity trapping with minimal sediment accumulation are consistent with steep slopes and elevated, nonseasonal precipitation. This is consistent with the climate period of 3.4–2.8 Ma described in Smith (1994). This unit was deposited directly on bedrock in cracks, potholes, and other bedrock lows, and in modern times is subjected to seasonally wet conditions. Oddly, this unit also shows the most radiogenic Pb isotopes (Fig. 3B), suggesting a more modern age (even active growth at present) than local stratigraphy and the principle of superposition would suggest. The placer gold from this unit has mineralized biomats (Kamenov et al., 2013) and significant rim alteration, which may result from biomineralization that has modified this stratigraphically older gold with a younger Pb isotope overprint.
The white placer deposits formed next, during a period that was wetter and dominated by fluvial systems. Significant proportions of the white placer sediments are fine-grained fluvial deposits indicative of a climate with fairly elevated and seasonal to nonseasonal precipitation. This would be consistent with the 2.8–1.6 Ma period of wet climate dominated by winter precipitation events (Smith, 1994). The smooth and highly rounded gold particle shape was most likely produced by physical weathering in these fluvial systems. However, this gold has also experienced a high degree of chemical weathering, possibly from low-temperature hydrothermal activity. Abundant bleached cobbles and sediments cemented by sericite and highly silicified material (Kamenov et al., 2013) suggest low-temperature hydrothermal alteration. It is likely that this alteration was responsible for the distinct Pb isotopes (Fig. 3B) and thick silver/copper-depleted rims noted on gold from the white placers, though the contribution of this chemical weathering to grain shape is unknown.
The red placers are the stratigraphically youngest of the Devils Nest placers. The development of the thick debris flows and fan gravels of the red placers is likely to correspond with the post–1.6 Ma period, which marks the transition to the present arid climate with seasonal precipitation (Fig. 5). As a result of this rapid transport, the gold of the red placers is relatively rough, preserving features from its original lode source, and displaying minimal rim effects produced by chemical weathering. The presence of biomats within deeper crevices and pits on the red placer gold, and uniform coatings on the black placer gold (Kamenov et al., 2013) suggest contemporaneous timing of biological colonization and modification very late in the development of placers at Rich Hill. At some fairly recent point, the modern braided drainage network began to erode the preexisting Potato Patch and Devils Nest placers, creating a modern alluvial placer deposit with its own distinctive characteristics.
The records of physical, chemical, and biological modification of the four distinct placer gold deposits at Rich Hill, Arizona, serve as a wet/dry transition paleoclimate indicator. The oldest Potato Patch placer unit occurs on a paleo-erosional surface stranded on a modern topographic high. Its placer gold shows minor geochemical weathering and is rough and three dimensional with abundant quartz inclusions. These characteristics are believed to record moderate precipitation and erosion with modest transport distance following the 25–22 Ma unroofing of the inferred lode gold source. The low silver content and the Pb isotope data, however, suggest that a major part of the large Potato Patch nuggets was aggregated within the placer environment. Following 17–5 Ma high-angle Basin and Range faulting, a shallow basin was produced on the shoulder of Rich Hill that preserves the three “Devils Nest” placer units. The Pb isotope data for nuggets from these placers indicate formation or modification of the gold nuggets during supergene processes, following the gold separation from the lode deposit (Kamenov et al., 2013). The oldest of these, the black placers, is suggestive of a steep gradient and abundant nonseasonal precipitation. Elongate forms suggest gold particle transport was under traction conditions. The middle white placer unit has well-rounded gold nuggets with deep chemical weathering that also record high levels of nonseasonal precipitation, but with a less steep gradient and abundant fluvial sediment deposition. The uppermost red placer unit is a pulse placer unit deposited by a series of landslides and debris flows during a period of much drier climate with seasonal precipitation. Microbial communities, preserved as Mn-Ba–oxide biomats, were established within the seasonally wet bedrock traps of the lowermost placer unit (Kamenov et al., 2013). This resulted in biological modification of placer gold chemistry of the black placers, and occasionally deep crevices on gold of the red placers.
Instrumentation was provided and supported by the National Science Foundation grants EAR-0115884 and EAR-0941106. X-ray fluorescence (XRF) instrumentation was provided by a grant from the W.M. Keck Foundation. Support was also provided by a National Aeronautics and Space Administration (NASA) Astrobiology Institute Minority Institution Research Sabbatical (NAI-MIRS) program award to Melchiorre, and an institutional grant by California State University–San Bernardino. We thank Ed DeWitt for his many hours of guidance over the past decade, prior to his passing in 2013. Special thanks go to Rob Allison and Scott Schuff for access to critical mining claims.
- Received 24 February 2016.
- Revision received 28 July 2016.
- Accepted 10 August 2016.
- © 2016 Geological Society of America