gold remobilization associated with mississippi valley
TRANSCRIPT
Gold remobilization associated with Mississippi Valley–type fluids: A Pb isotope perspective
Geological Society of America Bulletin, v. 1XX, no. XX/XX 1
Gold remobilization associated with Mississippi Valley–type fluids: A Pb isotope perspective
Erik L. Haroldson†, Brian L. Beard, Aaron M. Satkoski, Philip E. Brown, and Clark M. JohnsonDepartment of Geoscience, University of Wisconsin–Madison, 1215 West Dayton Street, Madison, Wisconsin 53715, USA
ABSTRACT
Gold deposits are often observed to have complex histories with multiple mineraliza-tion and remobilization events. Lead (Pb) isotopes have been useful in identifying the source and timing of metals during the ini-tial or “primary” mineralization in these deposits; however, the source and timing of metals associated with Pb that has more radio genic isotope ratios as compared to the primary Pb are less definitive. Here, we used Pb isotope analyses, obtained by whole-rock, microdrilled, and in situ sampling tech-niques, coupled with detailed petrog raphy to identify distinct fluid histories associated with gold mineralization in the Reef Deposit, which occurs in Penokean-age (1.8 Ga) rocks in north-central Wisconsin. Early Pb-rich minerals, encased in pyrite (as is some gold), have nonradiogenic Pb isotope ratios that match the Pb isotope composition of galena from volcanogenic massive sulfide deposits that occur locally north of the Reef Deposit. In contrast, late Pb-rich minerals, inter-grown with gold, have highly radiogenic Pb isotope ratios that closely match those of ga-lena from Paleozoic carbonate rocks in the region. A low-temperature brine fluid was found in fluid inclusion assemblages in late crosscutting carbonate veinlets. Based on these data, we infer that initial gold miner-alization in the Reef Deposit was produced during the Penokean orogeny (ca. 1.8 Ga) and was subsequently overprinted by fluids similar to Mississippi Valley–type miner-alization during the Permian (ca. 270 Ma). This temporal disparity between primary mineralization and remobilization, as iden-tified by Pb isotopes, has important impli-cations for our understanding of gold and other ore deposits. This study should prove useful to guide future Pb isotope investiga-tions in various disciplines.
INTRODUCTION
Hydrothermal gold deposits may form by a variety of processes and often involve multiple overprinting systems or “events” that either emplace or remobilize gold (Large et al., 2009; Zachariáš et al., 2013). Remobilization can be driven by circulating groundwater in supergene environments (Lawrance, 1990; Butt, 1998; Larizzatti et al., 2008), hydrothermal flow during late magmatic activity (Sheets et al., 1995), or deformation that redistributes the gold (Blenkin sop and Doyle, 2014). Many studies have identified a late, lowtemperature, highsalinity brine in veingold mineralization, which may overprint primary mineralization, making it difficult to establish genetic relations with gold mineralization (e.g., Wilkinson et al., 1999; Large et al., 2009).
Lead (Pb) isotope geochemistry has proven useful in understanding the complexity of fluid sources and pathways during gold mineralization. Pb isotope compositions that match those expected for average crust (Stacey and Kramers, 1975) at the time of initial mineralization are often referred to as “primary Pb” (McNaughton and Groves, 1996). In some cases, however, Pb isotope compositions are much more radiogenic, indicating addition of Pb at a later time (Perring and McNaughton, 1990). Although geochronology of host rocks, gangue, and ore minerals has aided in providing an absolute time scale for some events, relating such results explicitly to gold mineralization/remobilization can be challenging because of uncertainties in the genetic relations between dated materials and gold (Kerrich and Cassidy, 1994; Li et al., 2008; Zhang et al., 2014; Taylor et al., 2015; Vielreicher et al., 2015).
In this study, we used a combination of Pb isotopes and UThPb geochronology on bulkrock and microdrilled samples, as well as in situ measurements by laser ablation, to evaluate the age of mineralization in the Reef Deposit in northcentral Wisconsin. To better understand the timing of gold mineralization and remobili
zation, in situ analyses focused on the textural and temporal relations among galena (PbS), rucklidgeite (PbBi2Te4), and altaite (PbTe) grains that were closely associated with gold. The nature of remobilizing fluids was additionally informed using microthermometry of fluid inclusions. We found that although initial gold mineralization may have been related to CuZn volcanogenic massive sulfide mineralization, as is observed elsewhere in the region, the final gold mineralization/remobilization in the Reef Deposit occurred in the Paleozoic during extensive regional fluid flow. This conclusion not only has important implications for ore deposit models, but it also informs our understanding of the extent and nature of largescale fluid flow in the shallow crust, an area of broad interest.
GEOLOGY
The PembineWausau subterrane, located in northern Wisconsin and the Upper Peninsula of Michigan (Fig. 1), formed during the Penokean orogeny at ca. 1890–1830 Ma (Schulz and Cannon, 2007). The PembineWausau subterrane was formed initially as a series of volcanic arcs that were thrust onto the southern Superior craton during the Penokean orogeny, when a separate Archean craton known as the Marshfield terrane collided from the south with the Superior craton to the north (Schulz and Cannon, 2007). The PembineWausau subterrane is separated into two main volcanic complexes. The LadysmithRhinelander volcanic complex to the north hosts several CuZn volcanogenic massive sulfide (DeMatties, 1996) deposits thought to have formed at ca. 1874 Ma, suggesting a period of intraarc rifting during formation (Schulz and Cannon, 2007; Quigley et al., 2016). However, a more recently measured, slightly younger age of ca. 1833 Ma for the Back Forty deposit, the most eastern deposit located within the LadysmithRhinelander vol canic complex, adds some complication to a single common timing for mineralization in the region (Quigley et al., 2016). The southern, younger
GSA Bulletin; Month/Month 2016; v. 128; no. X/X; p. 1–13; https:// doi .org /10 .1130 /B31901 .1; 9 figures; 3 tables; Data Repository item 2018115 .; published online XX Month 2016 .
†erik .haroldson@ yahoo .com
For permission to copy, contact [email protected] © 2018 Geological Society of America
Haroldson et al.
2 Geological Society of America Bulletin, v. 1XX, no. XX/XX
portion of the PembineWausau subterrane is known as the Wausau volcanic complex, and it is composed of dominantly calcalkaline felsic volcanic rocks (LaBerge and Myers, 1984). The Wausau volcanic complex was estimated to have formed between 1845 and 1835 Ma (LaBerge and Myers, 1984), and this has been supported by UPb zircon geochronology of a rhyolite microporphyry near Wausau that has an age of 1836 ± 11 Ma (Sims et al., 1989). The Wausau vol canic rocks were then folded, with steeply plunging, northeasttrending axes, and metamorphosed from middle to uppergreenschist facies and locally to amphibolite facies (LaBerge and Myers, 1984; Schulz and Cannon, 2007). At ca. 1760 Ma, a series of postorogenic granitoids intruded into the area (Sims et al., 1989). Still later, directly adjacent to and east of the deposit, the Wolf River Batholith was emplaced, which was intruded from 1484 to 1468 Ma during an extensive anorogenic plutonism event in North America (Anderson, 1983; Dewane and Van Schmus, 2007). Early to middle Paleozoic sedimentary units blanket much of the crust south of the exposed Precambrian terranes in the northern part of Wisconsin (Fig. 1). A large Bouguer gravity anomaly minimum in the central and southern central portions of the state is taken as evidence that the Wolf River Batholith is a much more extensive unit underlying much of the southern portion of the state (Allen and Hinze, 1992).
The Reef Deposit is an anomalous AuCu deposit in the sense that it is the only known primarily goldrich deposit in the region, and it is hosted in mafic intrusive and volcanic rocks of the Wausau volcanic complex. A basal gab
bro unit, with similar mineralogy and textures to that observed in the hosting mafic units, crosscuts the main quartzsulfide veins at depth. This indicates that the veins formed shortly after the hosting mafic intrusive units, possibly as the root zone of a volcanogenic massive sulfide deposit. Felsic dikes and sills, of unknown age, are found as a swarm of granophyric to porphyritic, locally aplitic units within a northeasttrending zone of deformation (Kennedy and Harding, 1990). The younger Wolf River Batholith bounds the deposit area to the east (Fig. 1).
Mineralization in the Reef Deposit consists of at least eight subparallel AuCu–bearing quartzsulfide vein zones that trend southwest and dip moderately to the northwest (Kennedy and Harding, 1990). Gold in drillcore sections is not typically visible at the hand sample scale. Pyrrhotite is the most abundant sulfide found in veins and disseminated in country rocks adjacent to vein zones, followed by chalcopyrite and pyrite, which also occur in veins. Magnetite is observed, although less commonly than FeCu sulfide minerals. Molybdenite, sphalerite, and galena are rare. Chalcopyrite is the primary copper ore mineral, but Kennedy and Harding (1990) also reported cubanite in an overlying saprolite and regolith zone, likely formed during Precambrian weathering (Driese and Medaris, 2008).
METHODOLOGY
Hand samples were selected from drillcore sections and cut using a diamond blade. Fluid inclusions were studied on separate drillcore samples dedicated for that purpose. For iso
topic analyses, individual sulfide phases were separated from hand samples using a handheld micro drill with a tungsten carbide tip. Wholerock samples were selected from material remaining from drillcore assays by lifting individual aliquots from the homogenized material using a scoop. A second sample aliquot was refined by hand magnet covered with weighing paper to isolate a magnetic concentrate, and these samples are identified with an “HM” in the name. Sulfide and wholerock samples dissolved for Pb isotope analysis ranged in mass from 12 to 108 mg.
All isotopic analyses were done using a Nu Instruments Nu Plasma II multicollector–inductively coupled plasma–mass spectrometer (MCICPMS), equipped with an enhanced sensitivity interface, located in the Department of Geoscience at the University of Wisconsin–Madison. Bulkrock and microdrilled sulfide samples were dissolved in concentrated HNO3 in Savillex beakers and placed on a hotplate at 130 °C overnight. Samples were then centrifuged, and the liquid was removed and aliquoted. One aliquot was used to measure Pb isotope ratios, and the other aliquot was spiked with a mixed 208Pb235U229Th tracer to measure U, Th, and Pb concentrations by isotope dilution mass spectrometry (IDMS). The residue after dissolution consisted of silicates that were resistant to dissolution using HNO3. The chemical purification of the samples and the IDMS and Pb isotope ratio analyses followed the methods of Satkoski et al. (2015). Based on replicate analyses of mixed UThPb normals, U/Pb ratios are precise to ±0.07%, Th/Pb ratios are precise to ±0.13%, and U/Th ratios are precise to ±0.10%. Pb isotope ratio analyses are precise to ±0.014% per atomic mass unit based on repeat analyses of National Institute of Standards and Technology (NIST) standards SRM981 and SRM982 and U.S. Geological Survey (USGS) rock standards BCR2 and AGV2. All Pb isotope ratios and UThPb concentrations of samples are reported in Table 1, and standards are reported in Table DR11 to show that the Pb isotope ratios are accurate relative to consensus averages for NIST SRM981 and SRM982 and USGS rock standards BCR2 and AGV2 (Galer and Abouchami, 1998; Wilson, 1997, 1998; Woodhead and Hergt, 2000; Platzner et al., 2001; Collerson et al., 2002; Thirlwall, 2002; Ridley, 2005; Weis et al., 2005, 2006; Chauvel et al., 2011; Todd et al., 2015). Total procedural blank for Pb was 48 pg, which is 1000 times less than the Pb con
VMS Deposit
Wolf River Batholith
Wausau
Volcanic
ComplexMarshfield Terrane
Paleozoic SedimentaryRocks
UMV district
Chicago
Madison
Green Bay
Wisconsin
Minnesota
Iowa
Illinois
ReefDeposit
Michigan
Lake
Mic
higa
n
60 km
88°W89°90°91°92°
45°
44°
43°
42°N
Pembine Wausau Subterrane
N
Figure 1. Location of the Reef Deposit relative to volcano-genic massive sulfide (VMS) deposits within the Pembine-Wausau subterrane (from Can-non et al., 1997) and the Upper Mississippi Valley (UMV) Mis-sissippi Valley–type district (from Heyl et al., 1974), hosted within Paleozoic sedimentary rocks (white area). Star—the Reef Deposit, squares—vol-canogenic massive sulfide de-posits, solid lines—faults, solid line with triangles—Eau Pleine shear zone paleosuture between the Marshfield terrane and the Pembine-Wausau subterrane.
1GSA Data Repository item 2018115, further methodology of LAICPMS Pb isotope analysis, laser ablation locations and standards analysis data, is available at http:// www .geosociety .org /datarepository /2018 or by request to editing@ geosociety .org.
Gold remobilization associated with Mississippi Valley–type fluids: A Pb isotope perspective
Geological Society of America Bulletin, v. 1XX, no. XX/XX 3
tent analyzed in the samples, so no blank correction was performed.
Pb isotope ratios measured by laser ablation (LA) were done using a Teledyne Photon Machines Analytefs laser equipped with a HelEx LA cell (fsLA). The laser had a pulse width of ~150 fs, and two nonlinear crystals were used to produce a third harmonic such that the output wavelength of the laser was ~263 nm. The laser was used in focusbeam mode, where the laser spot size was controlled using an iris placed just before the objective lens and by adjusting the distance between the objective and the sample (the smaller the spacing between the objective and the sample, the larger the spot size). The laser conditions used during Pb isotope analyses (pulse energy at the sample of ~1 µJ and 125 laser shots delivered at a repetition rate of 4.03 Hz) created LA craters in pyrite that were 30 µm in diameter. Further methodology regarding fsLA and MCICPMS are described in the data repository (see footnote 1).
Fluid inclusion studies began with petrographic analysis of doubly polished thick sections and selection of inclusions for microthermometric measurements. Cathodoluminescence imaging was performed on a Hitachi S3400N scanning electron microscope (SEM) to identify carbonate textures and discern variable growth histories. Inclusions were classified using standard terminology, and fluid inclusion
assemblages were identified using criteria from Goldstein and Reynolds (1994) and Goldstein (2003). Microthermometry experiments were carried out at the University of Wisconsin–Madison using a Linkam LMS600 heatingfreezing stage mounted on an Olympus BX50 microscope. Observations were made under 100× magnification from –185 °C (limit of liquid N2 freezing) to 250 °C. Calibrations were done using synthetic H2OCO2 and separate pure H2O fluid inclusions. Calibration measurements were made at the melting temperature of CO2 (–56.6 °C), the dissociation point of clathrate (10 °C), and the critical point of pure H2O (374.1 °C). Bulk densities of inclusions were calculated within MacFlinCor (Brown and Hagemann, 1994) using equations from Bodnar and Vityk (1994).
RESULTS
Petrography
Gold occurs as 1–30µmsized grains in both an early and late textural setting (Fig. 2). The early texture consists of Au encased in pyrite (Fig. 2A); in the later, more common texture, Au occurs along gangue and sulfide grain boundaries and in microfractures (Figs. 2B–2H), and native gold and gold tellurides are often intergrown with other telluride phases (Figs. 2D, 2F,
and 2H) and calcite (Fig. 2C). Tellurides in the earlier textural setting included: altaite (PbTe), tellurobismuthite (Bi2Te3), and rucklidgeite (PbBi2Te4); tellurides in the later textural setting included hessite (Ag2Te), melonite (NiTe2), frohbergite (FeTe2), petzite (Ag3AuTe2), and calaver ite (AuTe2), as well as altaite and tellurobismuthite (at times with unmixed volynskite [AgBiTe2]). The LA Pb isotope analyses were performed on Pbrich phases that occurred in both early texture (inclusions in sulfides) and late texture (along mineral grains or in microfractures) settings (Fig. 2C; Fig. DR1 [see footnote 1]). Most of these analyses were done on Pbrich minerals intergrown with Au or in a similar textural setting in which gold was found (e.g., galena and Au both occurred as inclusions in pyrite in a sample, although Au and galena were not intergrown). Our survey for these Pbrich phases included detailed SEM petrography on 17 samples, of which we found 12 occurrences in three samples. After these were analyzed by LA, the surface was ground down by ~200 μm on 13 samples (including LA analyzed and nonanalyzed samples, as well as two newly created samples). The new surfaces were repolished and reexamined in the SEM, revealing an additional four occurrences of Pbrich minerals from two of these resurfaced samples. Several of the occurrences discovered were too small to be analyzed.
TABLE 1. SOLUTION NEBULIZATION–MEASURED U-Th-Pb CONCENTRATIONS, Pb ISOTOPE RATIOS, AND 238U/204Pb, 232Th/204Pb, AND 235U/204Pb RATIOS
SamplePb
(ppm)U
(ppm)Th
(ppm) 206Pb/204Pb 207Pb/204Pb 208Pb/204Pb 238U/204Pb 232Th/204Pb 235U/204Pb 207Pb/206Pb 208Pb/206PbMicrodrilled sulfidesEH1501py 11.66 0.07 0.03 17.009 15.401 36.485 0.383 0.157 0.003 0.9054 2.1450EH1504cpy 2.71 0.01 0.00 15.731 15.222 35.307 0.317 0.056 0.002 0.9676 2.2443EH1504po 0.98 0.02 0.01 16.553 15.333 36.006 1.042 0.318 0.008 0.9263 2.1753EH1505cpy 2.96 0.05 0.01 16.674 15.360 36.005 1.002 0.107 0.007 0.9212 2.1594EH1505po 3.97 0.08 0.01 16.383 15.309 35.717 1.197 0.148 0.009 0.9344 2.1801EH1507po 2.20 0.02 0.00 17.008 15.432 36.835 0.493 0.096 0.004 0.9073 2.1657EH1526po 6.90 0.04 0.00 16.453 15.372 35.625 0.330 0.044 0.002 0.9343 2.1653EH1528po 1.78 0.13 0.09 17.052 15.378 35.931 4.360 3.003 0.032 0.9018 2.1071EH1530py 10.53 0.05 0.01 20.597 15.830 39.977 0.319 0.048 0.002 0.7686 1.9409EH1532po 29.56 0.09 0.02 16.778 15.357 36.103 0.181 0.042 0.001 0.9153 2.1518EH1533py 8.69 0.04 0.01 18.736 15.662 38.150 0.277 0.081 0.002 0.8359 2.0362EH1534po 5.20 0.03 0.02 15.843 15.233 35.412 0.312 0.227 0.002 0.9615 2.2353
Whole rock14116LT125HM 1.43 0.05 0.02 16.475 15.322 35.528 2.042 0.706 0.015 0.9300 2.156614266LT125HM 1.77 0.07 0.03 16.285 15.284 35.379 2.339 1.036 0.017 0.9385 2.172614267HM 2.84 0.05 0.13 15.784 15.226 35.256 0.996 2.770 0.007 0.9647 2.233712470HM 1.94 0.21 0.37 19.475 15.674 36.400 6.933 12.583 0.050 0.8048 1.869114116LT125 1.45 0.06 0.03 16.583 15.334 35.596 2.578 1.241 0.019 0.9246 2.146514266LT125 2.12 0.11 0.07 16.452 15.304 35.563 2.960 1.936 0.021 0.9302 2.161714267 2.06 0.04 0.02 15.823 15.231 35.139 1.040 0.607 0.008 0.9626 2.220812470 1.24 0.26 0.34 20.974 15.814 36.896 13.913 18.544 0.101 0.7540 1.7591
Standard analysisNIST SRM-981 (n = 15) 16.939 15.495 36.717 0.9147 2.1676NIST SRM-982 (n = 12) 36.754 17.162 36.754 0.4669 1.0000BCR-2 (n = 8) 18.760 15.627 38.742 0.8330 2.0652AGV-2 (n = 8) 18.873 15.617 38.545 0.8275 2.0423Notes: Pb isotope data are accurate based on the Pb isotope analysis of U.S. Geological Survey rock standards and the National Institute of Standards and Technology
(NIST) SRM-981 and SRM-982 standards. For solution nebulization: Pb isotope ratios, based on standard analyses, are precise to ±0.014% per amu, U/Pb ratios are precise to ±0.07%, Th/Pb ratios are precise to +0.13%, and U/Th ratios are precise to +0.10%. Microdrilled sulfides are hand sample name; whole rock labels are Aquila Resources assay sample number. Abbreviations: py—pyrite, po—pyrrhotite, cpy—chalcopyrite, HM—hand magnetic fraction taken from separate aliquot, LT125—less than 125 μm size fraction.
Haroldson et al.
4 Geological Society of America Bulletin, v. 1XX, no. XX/XX
Late veinlets crosscutting primary quartz veins consisted of early dolomite mineralization with openspace euhedral growth texture, later infilled with coarsegrained (0.5–3 mm) calcite (Figs. 3B and 3C). Small, ~100µmsized pods of needleshaped chalcopyrite were found at the contact between dolomite and calcite, growing radially outward from dolomite. A 1–3 cm halo of fluid inundation was recognized in the form of ~50–100µmwide Kfeldspar microveinlets and quartz along wallrock–quartz grain boundaries, which, when viewed in cathodoluminescence imaging, showed distinctive banded crustiform texture growth zonation.
U-Th-Pb Isotope Data
UThPb isochron diagrams show contrasting behavior in 238U/204Pb206Pb/204Pb, 235U/204Pb207Pb/204Pb, and 232Th/204Pb208Pb/204Pb variations relative to “PbPb” (206Pb/204Pb207Pb/204Pb) variations (Fig. 4). For UPb (Figs. 4A and 4B) and ThPb (Fig. 4C) diagrams, wholerock samples plot to the left of a 1835 Ma reference isochron, indicating postformation Pb addition. Regression of the wholerock samples produced “error chrons” for a 232Th208Pb age of 1.67 ± 0.46 Ga, a 235U207Pb age of 2.07 ± 0.26 Ga,
and a 238U206Pb age of 2.28 ± 0.47 Ga. These contrast with a PbPb age of 1.88 ± 0.09 Ga (Fig. 4D), which lies within uncertainty of the 1835 Ma reference age that is taken to record the initial mineralization age of the Reef Deposit (see earlier herein). In contrast, the micro
drilled sulfide samples all have low U/Pb and Th/Pb ratios and therefore do not define any linear relations on UThPb isochron diagrams (Fig. 4). Because the microdrilled samples have low U/Pb and Th/Pb ratios, their radiogenic Pb isotope compositions cannot be explained by in situ decay, but instead require a source of Pb that was derived from a distinctly more radiogenic reservoir. The presence of radio genic ingrowth in the wholerock samples is attributed to the incorporation of rarely observed monazite grains observed in mineralized zone samples and titanite grains observed in host gabbro.
The Pb isotope compositions of Pb sulfides and tellurides associated with Au measured by fsLA define a bimodal distribution (Table 2; Fig. 5). Pb minerals associated with early texture Au, or that occur as inclusions in Fe sulfides without Au, have nonradiogenic Pb isotope compositions (206Pb/204Pb = 15.67–15.41). In contrast, Pb associated with the late texture Au, or that which occurs in the second textural setting without Au, has very radiogenic Pb isotope compositions (206Pb/204Pb = 25.67–24.81). We note that all the LA analyses were performed on Pbrich minerals such as galena and altaite, and hence Pb isotope ratios were not affected by in situ decay. The radio genic Pb isotope compositions are far higher than those of modern average crust and plot on an extension of the arrays defined by Upper Mississippi Valley district Mississippi Valley–type (MVT) mineralization (Fig. 5).
Fluid Inclusion Data
Assemblages of relatively lowtemperature H2ONaCl inclusions (homogenization temperatures ≤103 °C) were observed as secondary
Bi2Te3
Au
CuFeS2PbTe FeS2
CuFeS2
SiO2
CaCO3
Au
PbTe
Image D
Au
A B
DAg5Te3
FeS2NiTe2
C AuAg3Te2
Image F
PbTeSiO2 FeS2
Image H
Image G
PbTeNiTe2
PbTe
NiTe2
10 μm
100 μm 500 μm
40 μm
R30PbTe5spot2(25.39)
R30PbTe5spot1(25.41)
E F
HG
40 μm
500 μm 10 μm
20 μm
Au
Au
Ag5Te3 AuAg3Te2
Figure 2. Backscattered electron images showing textural rela-tions of gold, sulfide, and tellu-ride phases. (A) Gold in earlier textural setting. (B) Gold in later textural setting. (C) Gold, telluride, and calcite in later textural setting. Circles repre-sent location of laser-ablation analysis pits. (D) Inset of im-age C with lowered brightness setting, showing intergrowth of gold with altaite, hessite, and petzite. (E) Area with multiple gold and telluride mineral clus-ters hosted in quartz fractures outside sulfide mineralization. (F) Inset of image E with low-ered brightness setting, showing intergrowth of gold with altaite. (G) Inset of image E with low-ered brightness setting, show-ing intergrowth of altaite with melonite. (H) Inset of image E with lowered brightness setting, showing intergrowth of gold with melonite, altaite, hessite, and petzite. Dark areas in A, B, and E consist of quartz.
A B
C
ED
20 μm20 μm
1 cm
1 cm
2 cm
Figure 3. Vein texture images. (A) Common, semimassive Fe-sulfide zone of primarily pyrite and pyrrhotite in vein quartz. (B) Dolomite and calcite veinlet crosscutting primary quartz. (C) Dolomite and calcite veinlet crosscutting deformed host rock. (D) Photomicrograph of assem-blage of fluid-only aqueous fluid inclusions in calcite. Image was created using stacked zoomed images. (E) Photomicrograph of assemblage of all liquid + mod-erately consistent two-phase in-clusion assemblage.
Gold remobilization associated with Mississippi Valley–type fluids: A Pb isotope perspective
Geological Society of America Bulletin, v. 1XX, no. XX/XX 5
inclusions in late calcite veins that crosscut mineralized quartz veins and host rocks of the Reef Deposit (Fig. 3). Fluid inclusion assemblage data are presented in Table 3. The lowtemperature assemblages were identified petrographically as assemblages of all liquid inclusions assumed to have formed at temperatures <50 °C (Goldstein and Reynolds, 1994), and assemblages of all liquid plus twophase inclusions with moderately consistent vapor bubbles (≤5% vapor). All liquid plus twophase inclusion assemblages homogenize in a range of temperatures (45–103 °C). An estimate of paleo depths may be made by using the fact that the deposit is overlain by a section of saprolite and regolith
associated with Precambrian weathering, indicating that the unconformity between Precambrian and Paleozoic sedimentary rocks would have been at or near the current topographic surface. Because the deposit is located upon the Wisconsin dome, well north of the Illinois Basin, the deposit was not overlain by significant sediment thickness during formation of the fluid inclusions. Coupled with the fact that these are relatively highdensity fluid inclusions with steep isochores, a pressure correction would be insignificant, indicating that homogenization temperatures closely reflect formation temperatures. The higherdensity H2ONaCl fluids would cause vapor bubble crushing (disappear
ance) upon freezing, and because, upon reheating, the vapor bubble would not nucleate prior to Tmice (ice melting), we could not use the Tmice to determine salinity. We used the technique described in Wilkinson (2017) to estimate the bulk salinity. Metastable freezing temperatures ranging from –79 °C to –82 °C were used to calculate salinities of 23.6–24.7 wt% NaCl equivalent.
DISCUSSION
In the following discussion, we examine geologic scenarios that could produce the early and late textured Pbrich minerals, the age(s) of Au mineralization, and the important constraints
A B
C
1835 Ma ref. isochron
D
1.88 +/- 0.09 Ga
MSWD = 155
sulfide micro-drillwhole rock
1835 Ma ref. is
ochron
1835 Ma re
f. isochron
sulfide micro-drillwhole rock
sulfide micro-drillwhole rock
sulfide micro-drillwhole rock
Pb addi�on
Pb addi�on
Pb addi�on
15.1
15.3
15.5
15.7
15.9
0.000 0.050 0.100 0.150
207 P
b/20
4 Pb
235U/204Pb
15
16
17
18
19
20
21
22
0 5 10 15
206 P
b/20
4 Pb
238U/204Pb
34.5
35.0
35.5
36.0
36.5
37.0
0 5 10 15 20
208 P
b/20
4 Pb
232Th/204Pb
15.0
15.2
15.4
15.6
15.8
16.0
15 17 19 21
207 P
b/20
4 Pb
206Pb/204Pb
Figure 4. U-Th-Pb isochron diagrams for (A) 238U/204Pb-206Pb/204Pb, (B) 235U/204Pb-207Pb/204Pb, (C) 232Th/204Pb-208Pb/204Pb, and (D) 206Pb/204Pb-207Pb/204Pb. A 1835 Ma reference isochron is shown for A–C, based on the estimated age for initial formation of the Reef Deposit (see text), using initial Pb isotope ratios based on the least radiogenic galena measured (Table 2). Data that fall to the left of the 1835 Ma reference isochron may be explained by Pb addition at a later time, as shown. (D) Regression of the 206Pb/204Pb-207Pb/204Pb whole-rock data produces an age of 1880 ± 90 Ma, which falls within error of the inferred initial mineralization age of 1835 Ma (age calculated using Isoplot from the whole-rock sample data [Ludwig, 2003], and a correlation coefficient of 0.9). Pb-Pb growth evolu-tion curve is from Stacey and Kramers (1975). MSWD—mean square of weighted deviates.
Haroldson et al.
6 Geological Society of America Bulletin, v. 1XX, no. XX/XX
imposed by the highly radiogenic Pb isotope compositions measured by LA. We argue that the early system was produced during the Paleoproterozoic, similar in time and from a similar source as the volcanogenic massive sulfide deposits located to the north of the Reef Deposit. The late textured Pbrich minerals, however, require a significantly younger derivation from a source that must have had very high U/Pb and Th/Pb ratios to produce the highly radiogenic Pb isotope compositions, and we postulate that this involved MVT fluid movement during the late Paleozoic. Fluid inclusion compositions are shown to be consistent with our interpretation for the late textured Pbrich minerals. We then discuss possible sources of the fluids that produced the late Pb mineralization, coincident with Au mineralization/remobilization.
Early Texture Pb-Rich Minerals
The Pbrich minerals associated with the early textured Au mineralization have Pb isotope compositions that overlap with those of galena from Penokeanage volcanogenic massive sulfide deposits to the north of the Reef Deposit (Figs. 1 and 6). We therefore suggest that the most likely source of Pb during this early gold mineralization was from sources similar to that of the Wisconsin volcanogenic massive sulfide deposits. The age of the volcanogenic massive sulfide mineralization is consistent with the intersection found in the data array with the average crustal growth curve of Stacey and Kramers (1975), between 1.9 and 1.8 Ga (Afifi et al., 1984). Although there is a spread in 206Pb/204Pb and 207Pb/204Pb ratios for the Penokeanage volcanogenic massive sulfide deposits, the range in Pb isotope compositions is much smaller within each deposit (Fig. 6), and these differences likely reflect some variation in the U/Pb source ratios. For example, the lower 206Pb/204Pb and 207Pb/204Pb ratios of the Flambeau and Reef Deposit suggest that the source of the Pb may have been more juvenile (low U/Pb), as compared to some of the other deposits such as Tomahawk, which may have been derived from a more evolved source with higher U/Pb (Fig. 6). The variations in measured 206Pb/204Pb and 207Pb/204Pb for most of the Pbrich minerals in the Reef Deposit are analytically indistinguishable, except for one analysis of altaite that has a
UMV DistrictDeposits
MVTMinor Occurrences
UMV DistrictDeposits
MVTMinor Occurrences
15.0
15.2
15.4
15.6
15.8
16.0
16.2
16.4
207 P
b/20
4 Pb
34.0
36.0
38.0
40.0
42.0
44.0
15 17 19 21 23 25
208 P
b/20
4 Pb
206Pb/204Pb
15.1
15.3
15.5
15.2 15.7
1.91.8
1.7
1.8
0 Ga
1.0 Ga
0 Ga
1.0 Ga
34.8
35.2
35.6
15.2 15.7
Bulk Analysissulfide microdrillwhole rock
early texturelate texture
VMS depositsReference
Laser Abla�on
A
B
Figure 5. (A) 207Pb/204Pb vs. 206Pb/204Pb plot. (B) 208Pb/204Pb vs. 206Pb/204Pb plot. Reference Wisconsin volcanogenic mas-sive sulfide (VMS) deposit data are from Afifi et al. (1984) and Thorpe (2008). Upper Missis-sippi Valley (UMV) district deposit field is defined by data from Millen et al. (1995). Field for minor occurrences of Missis-sippi Valley–type (MVT) min-eralization in areas of northern Illinois, southern and eastern Wisconsin, southeastern Min-nesota, and northeastern Iowa surrounding the UMV district is from Millen et al. (1995) and Luczaj et al. (2007) and extends to 206Pb/204Pb ratios of 34.7 along a similar trend. Pb evolu-tion curve is from Stacey and Kramers (1975), with marks in 100 m.y. increments. Insets show detailed plots of the nonradio-genic Pb isotope compositions.
TABLE 2. LASER ABLATION–MEASURED Pb ISOTOPE COMPOSITIONS OF REEF DEPOSIT SAMPLES
SampleTextural setting
Gold association 206Pb/204Pb 207Pb/204Pb 208Pb/204Pb 207Pb/206Pb 208Pb/206Pb
R06PbTe1 Early No 15.67 15.24 35.07 0.972 2.239R09aBiPbTe1spot1 Early Yes 15.50 15.24 35.10 0.984 2.265R09aBiPbTe1spot2 Early Yes 15.50 15.25 35.05 0.984 2.263R09aPbS1spot1 Early No 15.41 15.18 34.88 0.985 2.263R09aPbS1spot2 Early No 15.43 15.18 34.96 0.984 2.266R09aPbTe1 Early No 15.50 15.24 35.08 0.983 2.261R30PbTe1 Late Yes 25.27 16.34 43.91 0.647 1.740R30PbTe2spot1 Late Yes 25.05 16.30 43.99 0.651 1.756R30PbTe2spot2 Late Yes 25.20 16.33 44.09 0.648 1.750R30PbTe4spot1 Late Yes 25.67 16.40 44.46 0.639 1.732R30PbTe4spot2 Late Yes 24.91 16.30 43.45 0.654 1.748R30PbTe4spot3 Late Yes 24.95 16.35 43.91 0.656 1.761R30PbTe4spot4 Late Yes 24.81 16.29 43.80 0.656 1.762R30PbTe4spot6 Late Yes 25.50 16.40 44.30 0.643 1.737R30PbTe4spot7 Late Yes 25.16 16.35 44.05 0.650 1.751R30PbTe5spot1 Late Yes 25.41 16.37 44.04 0.644 1.733R30PbTe5spot2 Late Yes 25.39 16.34 44.01 0.644 1.734R30PbTe6 Late Yes 25.50 16.39 44.35 0.643 1.739R30PbTe7 Late Yes 25.56 16.43 44.24 0.642 1.731Notes: For laser ablation: Pb isotope ratios are precise to 0.12% per amu. Laser ablation labels are laser
mount name followed by phase and occurrence number.
TABLE 3. FLUID INCLUSION ASSEMBLAGES MEASURED IN LATE CARBONATE VEINS OF THE REEF DEPOSIT
Sample number Chip number Host mineral OccurSize(μm)
V/T(%) n
Tmice
(°C)Tmf(°C)
ThLV-L
(°C)Salinity
(wt% NaCl equiv.)Density(g/cm3)
EH1512 5 liquid-only fluid inclusion assemblages were recognized, each with n > 10EH1512 4 cal HF 2–7 <5 8 na –82 to –81 45–81 24.3–24.7 1.15–1.18EH1512 4 cal HF 2–9 <5 7 na –80 to –79 65–103 23.6–24.0 1.11–1.17Note: Abbreviations: cal—calcite; Occur—mode of occurrence; HF—healed fracture; V/T—vapor/total ratio;Tmf—metastable freezing temperature; Th—homogenization
temperature; LV-L—homogenization to liquid; na—not applicable.
Gold remobilization associated with Mississippi Valley–type fluids: A Pb isotope perspective
Geological Society of America Bulletin, v. 1XX, no. XX/XX 7
slightly elevated 206Pb/204Pb ratio; the altaite with elevated 206Pb/204Pb is the only Pbrich mineral that was hosted in chalcopyrite, whereas all the others occurred in pyrite. This difference in Pb isotope composition may imply that copper mineralization was derived from a different source, or possibly at a younger time, as compared to the other sulfides.
The Reef Deposit is hosted in the Wausau volcanic complex, which is thought to have formed between 1845 and 1835 Ma (LaBerge and Myers, 1984). More precise ages of several of the volcanogenic massive sulfide deposits, namely Bend, Horseshoe, Lynne, and Pelican River, have been measured giving a formation age of ca. 1874 Ma, with a younger ca. 1833 Ma age for the Back Forty deposit located at the far eastern extent of the PembineWausau subterrane in Michigan (Quigley et al., 2016). The ca. 1833 Ma age is more consistent with the estimate for the Wausau volcanic complex that hosts the Reef Deposit. There are no Pb isotope data for the Back Forty deposit, but the timing could link these deposits, suggesting a distinctive secondary mineralization pulse late in the Penokean orogeny. Similar low 206Pb/204Pb and 207Pb/204Pb ratios for the Flambeau and Reef Deposit may also link the Flambeau deposit into this separate pulse of mineralization, although there is currently no accurate age for the Flambeau deposit or its host rocks.
Late Texture Pb-Rich Minerals
The Upper Mississippi Valley district is a historic PbZn MVT mining district in southwestern Wisconsin. The district boundary outlines deposits that were economic in the past, although it has long been recognized that outlying occurrences extend the mineralization footprint across the region (Heyl and West, 1982; Millen et al., 1995; Luczaj et al., 2007). Galena in the Upper Mississippi Valley district sensu stricto and outlying occurrences defines a significant range in 206Pb/204Pb207Pb/204Pb208Pb/204Pb that extends from a 206Pb/204Pb of 18.29 to 34.7 (Fig. 5). The Pb isotope composition of the later texture Pbrich minerals in the Reef Deposit is highly radiogenic and matches the Pb isotope composition of Upper Mississippi Valley district and outlying occurrences. Based on the similarity of Pb isotope compositions between the late texture Pbrich phases of the Reef Deposit and galena from Upper Mississippi Valley district deposits and vicinity, we suggest that latestage Pb minerals in the Reef Deposit may be related to the same mineralizing fluids that produced Upper Mississippi Valley galena.
Upper Mississippi Valley district mineralization is generally thought to have occurred at 270 Ma (Brannon et al., 1992). To evaluate if such a scenario is plausible, we calculated the Pb isotope evolution, assuming that the Pb
was removed from its source at various times in the past and deposited in the Reef Deposit at 270 Ma (Fig. 7A). For this model, we calculated the µ value (238U/204Pb, proportional to the U/Pb concentration ratio) that is required to produce the measured 206Pb/204Pb ratio of the later texture Reef Deposit Pbrich minerals from differentaged sources that started with Pb isotope compositions equal to that of Stacey and Kramers (1975). Figure 7A shows three curves: Curve 1 assumes a 2.6 Ga Pb source that would correspond to Marshfield terrane rocks that occur to the SW of the Reef Deposit; curve 2 assumes a Penokean age source of 1.89 Ga; and curve 3 assumes a source of 1.47 Ga that would correspond to the Wolf River Batholith, which is directly adjacent to the Reef Deposit. The best fit to this model is for Pb derived from a Wolf River Batholith age source with a µ of 42.7, where Pb isotope evolution began at 1.47 Ga until 270 Ma, when Pb was removed and deposited in the Reef Deposit.
The second Pb evolution model that was constructed (Fig. 7B) is similar to that in Figure 7A, except that final Pb mineralization is assumed to have occurred during metamorphism in the late Penokean, ca. 1.84 Ga. In other gold deposits, late textural gold that occurs along grain boundaries or in fractures is often inferred to have been remobilized during metamorphism (Large et al., 2009; Blenkinsop and Doyle, 2014). In this model, the µ values must be substantially higher than the previous model because less time is allowed for evolution. As shown in Figure 7B, µ values approaching 1000 are required to produce the measured 206Pb/204Pb ratios in the late textural gold, which seems impossibly high. Importantly, however, this model, in all cases, produces very high 207Pb/204Pb ratios due to the restriction of Pb isotope evolution to earlier times than in the first model. Such a conclusion holds for all three curves in Figure 7B: Curve 1 assumes Pb from a source with an age that would correspond to Minnesota River Valley Archean rocks of 3.5 Ga; curve 2 represents a source with a Marshfield terrane age of 2.6 Ga, and curve 3 is a Penokeanage source of 1.89 Ma that would correspond to rocks of the PembineWausau subterrane. We conclude that it is virtually impossible to match both the 206Pb/204Pb and 207Pb/204Pb ratios of the later textured Au using reasonable µ values if Pb mineralization in the latestage minerals occurred ca. 1.84 Ga, reinforcing our conclusion that the first model, where late Pb mineralization occurred at ca. 270 Ma (Fig. 7A) after circulating through 1.47 Ga anorogenic granites, is the most plausible.
Measured U and Pb contents for the Wolf River Batholith, which outcrops east of the Reef
14.9
15
15.1
15.2
15.3
15.4
15.5
15.6
15 15.2 15.4 15.6 15.8 16
207 P
b/20
4 Pb
206Pb/204Pb
Flambeau
LynneCrandon
Pelican RiverSpirit
Horse ShoeRitchie Creek
Kivela ZoneTomahawk
1.5 Ga1.6 Ga
1.4 Ga
1.9 Ga2.0 Ga
1.8 Ga
1.6 Ga
galenarucklidgeitealtaite
Average Crust
Superior Province Mantle
Figure 6. Comparison of 207Pb/204Pb vs. 206Pb/204Pb measured in situ on early textured Pb-rich phases in the Reef Deposit with volcano-genic massive sulfide deposit data from the Pembine-Wausau sub-terrane (Afifi et al., 1984; Thorpe, 2008). These results tie down the range of initial Pb isotope compositions of the Reef Deposit at ca. 1835 Ma. Average crust growth curve is from Stacey and Kramers (1975), and Superior Province mantle growth curve is from Afifi et al. (1984). Gray ellipses are error ellipses of individual data points from this study.
Haroldson et al.
8 Geological Society of America Bulletin, v. 1XX, no. XX/XX
Deposit, have unusually high U/Pb ratios and high Pb contents (Fig. 8), supporting the interpretation that the radiogenic Pb isotope compositions of the later texture Pb minerals record Permianage circulation of fluids through anorogenic granites. Assuming Pb isotope compositions equal to that of average crust (Stacey and Kramers, 1975), the mostevolved Wolf River Batholith samples have µ values (238U/204Pb) approaching 30 (Fig. 8), and, importantly, such high µ values are associated with very high Pb contents, suggesting a significant source of Pb for later fluids. Although the µ values are not as high as those required for our model (µ ~43; Fig. 7A), it seems likely that the Pb isotope compositions of the Wolf River Batholith are far more radiogenic than average crust, which, if true, would imply much higher µ values. We conclude that although there are no Pb isotope data available from the Wolf River Batholith, its unusually high U/Pb ratios support the interpretation that this is the most likely source for the radiogenic Pb isotope compositions of the later texture Pb minerals in the Reef Deposit.
The fluid inclusions that occur in carbonate veins in the Reef Deposit are consistent with the temperature and composition of MVT
fluids precipitating Pbrich minerals associated with the later texture gold. Although it is not possible to establish a direct paragenetic relation between later texture Au in the Reef Deposit and carbonate veins because Au or later texture Pbrich minerals were not present in any of the carbonate veins that contain fluid inclusion assemblages, carbonate often occurs intergrown with later texture Au and Pbrich minerals, suggesting that carbonate precipitation and Au remobili zation were linked. Fluid inclusions in Upper Mississippi Valley district deposits have homogenization temperatures that range from 90 °C to 150 °C, with a few measurements as high as 220 °C; salinities for MVT deposits range from 10 to 30 wt% NaCl equivalent (Rowan and Goldhaber, 1996; Leach et al., 2010). Fluid inclusions in carbonate veins from the Reef Deposit have metastable freezing temperatures that correspond to 23.6–24.7 wt% NaCl equivalent. The homogenization temperatures correspond to <103 °C for the twophase inclusion assemblages, and formation temperatures <50 °C are inferred for the singlephase inclusion assemblages. Interpretation of the maximum temperature depends on if the twophase inclusions are primary or if they yield tempera
tures that are unrealistically high due to density loss of the fluid caused by the inclusion not remaining a closed system. Because a population of twophase inclusions tends to record a narrow range of temperatures (within 10–15 °C), it is likely that these inclusions have remained closed systems. Based on the homogenization temperatures, we can conclude that the fluids in these carbonate veins were up to 103 °C but could have been as cool as 50 °C and lower.
Evidence for MVT-Age Au Mobilization
MVT age mineralization in the Reef Deposit is an unusual occurrence in that it is hosted in Precambrian bedrock, whereas MVT deposits
μ = 25.2
3.5 Ga
2.6 Ga
μ = 35
1.89 Ga
1.47 Ga
μ = 69
μ = 958
0 Ga
0.27 Ga
1.89 Ga2.6 Ga
μ = 33.3
μ = 42.7
0.27 Ga
0.27 Ga
B
A
0 Ga
curve 1
curve 2
curve 3
curve 1
curve 2
curve 3
1.84 Ga
1.84 Ga
1.84 Ga
15
15.5
16
16.5
17
13 16 19 22 25 28
207 P
b/20
4 Pb
206Pb/204Pb
13
14
15
16
17
18
19
20
21
10 15 20 25
207 P
b/20
4 Pb
206Pb/204Pb
Figure 7. Pb isotope evolution curves for production of present-day (measured) Pb isotope com-positions assuming Pb isotope evolution ceased (A) at 270 Ma or (B) 1840 Ma, as derived from Archean Minnesota River Valley crust (3.5 Ga), Archean Marsh-field terrane crust (2.6 Ga), early Penokean crust (1.89 Ga), or Wolf River Batholith crust (1.47 Ga), using the µ values (238U/204Pb) indicated. Gray tri-angles show the data for the late-stage radiogenic Pb isotope compositions obtained from the Reef Deposit by laser ablation (Fig. 5). The best model solution that matches both 206Pb/204Pb and 207Pb/204Pb is Pb isotope evolution from a 1.47 Ga Wolf River Batholith and a µ value of ~43, followed by Pb extrac-tion and cessation of Pb isotope evolution at 270 Ma (part A). All of the models that involve Pb ex-traction at 1.84 Ga (part B) pro-duce 207Pb/204Pb values that are too high. Curves ending at 0 Ga are after Stacey and Kramers (1975).
A
B0
5
10
15
20
25
30
0
0.1
0.2
0.3
0.4
0.5
238 U
/204 P
b
U/P
b
0
10
20
30
40
50
60
45 55 65 75
Pb (p
pm)
SiO2 (wt%)
Figure 8. U and Pb concentration data from the Wolf River Batholith (Anderson and Morrison, 1992; J.L. Anderson, 2017, per-sonal commun.). (A) U/Pb ratio vs. silica, with calculated µ (238U/204Pb) values on sec-ondary vertical axis. Calculated µ values assume present-day Pb isotope composi-tions equal to average crust, which likely underestimates the true µ values, given the high measured U/Pb ratios (see text for dis-cussion). (B) SiO2 vs. Pb concentration. The very high Pb concentrations for evolved samples, relative to average crust, suggest a large reservoir of Pb would be available to influence the Pb isotope compositions of fluids that migrated through the Wolf River Batholith.
Gold remobilization associated with Mississippi Valley–type fluids: A Pb isotope perspective
Geological Society of America Bulletin, v. 1XX, no. XX/XX 9
are most often observed hosted in carbonate sedimentary formations. Here, we examined the current MVT deposit models, particularly for deposits in the Upper Mississippi Valley district and minor occurrences in the surrounding region, and how they may relate to late mineralization in the Reef Deposit. In particular, we considered the relations among basin fluid flow, the Wolf River Batholith, and the host rocks for the Reef Deposit.
MVT deposits are commonly thought to have formed in passivemargin tectonic settings from basinal brine fluids enriched in metals, which, during uplift, were forced outward into the neighboring terranes as giant hydrothermal systems (Figs. 9A and 9B; Bethke and Marshak, 1990). Classically, these fluids are seen to have interacted with carbonates and deposited metal sulfides and gangue minerals, commonly galena, sphalerite, carbonate, and fluorite (Leach et al., 2005). Problems with this model remain, in part, because the model is used to describe a diverse deposit class (Wilkinson, 2014). For example, burial temperatures for pore fluids often do not match the temperatures observed in mineralization. To account for elevated tem
Appalachian
Fron
t
0
0
200 miles
200 kilometers
UMV
ArkomaBasin
ForestCity
Basin
IllinoisBasin
MichiganBasin
Wisconsin Arch
Aquifer
Gravity-driven �owPrecambrian Basement
BUMV district
1719
2123
25
Reef Deposit
A
Illinois Basin
= 206Pb/204Pb of �uid
25
N
C
WolfRiverBatholith
Meteoric H2OAquifer
?
N
?
N
Reef Deposit
21
23
32
Reef Deposit
21
= 206Pb/204Pb and modelled �ow direction of �uid
17
WRB
Precambrian Basement
Precambrian Basement
Appalachian Front
Figure 9. Map and schematic cross sec-tions showing relationships of Permian fluid movement concepts discussed in the text. (A) Map showing location of Reef Deposit in relation to multiple features in the region affected by the Appalachian foreland. Black areas are prominent Mississippi Valley–type (MVT) districts (UMV—Upper Mississippi Valley district). WRB—Wolf River Batho-lith. Modern exposure of Precambrian basement is shown. (B) Schematic (not to scale) gravity-driven fluid-flow model modi-fied from Ingebritsen and Appold (2012). 206Pb/204Pb values show a systematic in-crease moving northward with radiogenic Pb component sourced from Precambrian basement material (a large portion of which is made up of anorogenic granite material) along flow path. Buried upper surface of Precambrian basement is shown. (C) Sche-matic (not to scale) depicting high heat-producing granite model modified from Spirakis and Heyl (1996). Faulting formed during the Appalachian-Ouachita orogeny created fluid circulation pathways to trans-mit lower-temperature fluids (dark-gray arrows) to deeper portions of high heat-producing granite and return with elevated-temperature fluids (light-gray arrows). Buried upper surface of Precambrian base-ment is shown.
Haroldson et al.
10 Geological Society of America Bulletin, v. 1XX, no. XX/XX
peratures during mineralization, fluid circulation through deep underlying basement terranes is commonly invoked (Spirakis, 1995). This raises the possibility that although sedimentary platform rocks are the most common host for MVT mineralization, involvement of deep basin crystalline rocks may be important in MVT fluid flow.
The Reef Deposit is located high within the Wisconsin dome, a Precambrian high at the northern extent of the Wisconsin Arch, located between the Michigan and Illinois sedimentary basins (Fig. 9A). MVT mineralization formed by fluids flowing out of the Illinois Basin is manifested in the Upper Mississippi Valley PbZn district located in southwestern Wisconsin (Fig. 1). Minor occurrences of MVT mineralization have been identified in areas of northeastern Wisconsin and are attributed to fluids flowing outward radially from the Michigan Basin ( Luczaj, 2006; Luczaj et al., 2016). The location of the Reef Deposit allows for fluids derived from either the Illinois Basin or the Michigan Basin.
Two reasonable, although competing, models exist for the Upper Mississippi Valley district of southwestern Wisconsin. The first model involves gravitydriven fluids flowing out of the Illinois Basin (Figs. 9A and 9B; Millen et al., 1995). This is the model most commonly discussed in the literature, although a maximum burial depth of ~1 km at the time of mineralization is unable to account for the high homogeni zation temperatures observed in sphaleritehosted fluid inclusions in Upper Mississippi Valley deposits (Deming and Nunn, 1991; Spirakis, 1995; Rowan and Goldhaber, 1996). The second model for the Upper Mississippi Valley district has explained the temperature discrepancy by calling upon circulation of fluids through high heatproducing (K, Th, Urich) granites. This model was originally created to explain anomalous uranium mineralization found near granites with high uranium contents (Fehn et al., 1978). The model implies that highly radioactive granites can release heat that has built up from radioactive decay during episodes of fracturing, allowing convective flow to bring heat and metals upward from depth. Support comes from Urich granites identified in deep drilling in northern Illinois and inferred to be present underneath much of the Upper Mississippi Valley district (Spirakis and Heyl, 1996). The circulation of fluids in the highertemperature, lower portions of this radioactive granite body would have been accessed by faulting associated with the AppalachianOuachita orogeny (Fig. 9C; Spirakis and Heyl, 1996).
Outlier occurrences of trace PbZn mineralization have been recognized in areas surrounding the Upper Mississippi Valley district in all
directions (Heyl and West, 1982). These minor occurrences are similar to Upper Mississippi Valley deposits in that they contain similar mineralogy, crystal habits, Pb isotope patterns, and mineral formation sequences (Heyl and West, 1982). The areas hosting minor occurrences are often missing significant Ordovician carbonate sedimentary units due to erosion. Minor occurrences have been recognized more recently in areas of northeastern Wisconsin, and although these occurrences share similar mineralogy, they are attributed to fluid flow from the nearby Michigan Basin located to the east (Luczaj et al., 2016). The driving force for fluid flow out of the Michigan Basin is not well understood, but it has been suggested that fluid flow out of the Michigan Basin was related to reactivation of the Midcontinent Rift during the Late Devonian to Carboniferous (Ma et al., 2009).
Lead Isotopic Evidence for MVT Fluid Sources and Pathways
Radiogenic Pb isotope compositions are a defining characteristic for Mississippi Valley Type deposits in the midcontinent United States, which are often referred to as “Jtype” (Joplintype) Pb deposits, named after a type locality in Joplin, Missouri (Heyl et al., 1974; Wilkinson, 2014). The high Pb isotope ratios reported in MVT deposits of the midcontinent United States are thought to be due to the ore fluids interacting with the basal Cambrian sandstone and underlying Precambrian basement prior to mineralization (Fig. 9B; Sicree and Barnes, 1999). Of the MVT districts in the midcontinent United States, the Upper Mississippi Valley district in southwestern Wisconsin has the most radiogenic Pb isotope compositions (Millen et al., 1995).
Galena samples measured in Paleozoic carbonates both within and peripheral to the Upper Mississippi Valley district show a progressive increase in 206Pb/204Pb and 208Pb/204Pb ratios moving northward away from the Illinois Basin, into the Upper Mississippi Valley district, and continuing in a northeast direction out of the district (Figs. 9A and 9B; Millen et al., 1995). Our Pb isotope data are consistent with a continued increase in 206Pb/204Pb and 208Pb/204Pb ratios observed in measurements from locations measured south to north. Significantly higher 206Pb/204Pb and 208Pb/204Pb ratios, however, are measured in areas to the SE of the Reef Deposit, in northeastern Wisconsin (Fig. 9A; Luczaj et al., 2007). The source of the highly radiogenic Pb isotope compositions in northeastern Wisconsin has been related to fluid flow out of the Michigan Basin (Luczaj et al., 2007, 2016). It is important to note, however, that the timing of the Illinois Basin and Michigan Basin
fluid events is variable, with the Upper Mississippi Valley district mineralization occurring at ca. 270 Ma, while radial flow outward from the Michigan Basin has authigenic illite ages of 367–322 Ma (Brannon et al., 1992; Girard and Barnes, 1995). Our Pb isotope modeling in Figure 7A cannot distinguish between Late Devonian to Carboniferous and Permian ages.
Based on the proximity of the Wolf River Batholith to the Reef Deposit, and the likelihood that it contained very radiogenic Pb isotope compositions, based on high U/Pb ratios (Fig. 8) at the time of MVT mineralization, we propose fluid flow through the batholith as an alternative to models that call upon fluids from the Michigan Basin as a source for radiogenic Pb. Given the location of the Reef Deposit on the Wisconsin Arch, we propose that fluid flow through Phanerozoic sedimentary rocks penetrated the Precambrian crystalline basement near the Reef Deposit (Fig. 9C), where fluids inter acted with highU/Pb and Pbrich rocks of the Wolf River Batholith. Such rocks would have had highly radiogenic Pb isotope compositions during the Permian, as well as the Devonian to Carboniferous.
Fluid Inclusion Evidence for MVT Fluid Sources and Pathways
Late, calcitehosted fluid inclusions in the Reef Deposit are saline, and they are consistent with basinal brine fluids associated with the gravitydriven model. Two possibilities of fluid sources are most likely for the high heatproducing granite fluid model, either local meteoric water or a shield brine present in the basement. The local meteoric water would not be saline. Archean shield brines measured in deep mine waters of northern Canada have been shown to be saline with total dissolved solids as high as 300 g/L (Bottomley et al., 1994). There are no known measurements for shield brine fluids in the Proterozoic basement of northern Wisconsin. However, we see no reason to suspect that genesis of this fluid would be significantly different than the fluids found in the Superior craton. Therefore, the source of the salinity in the late calcite–hosted fluids of the Reef Deposit could be from the gravitydriven model or the high heatproducing granite model.
Homogenization temperatures up to 103 °C observed in the Reef Deposit cannot be explained using the gravitydriven flow model alone. As discussed earlier, the same problem exists for deposits in the Upper Mississippi Valley district, and past workers usually have called on circulation of the mineralizing fluid deep into the basement in order to gain the temperatures observed (Spirakis, 1995). Another possibility for the Upper Mississippi Valley district is
Gold remobilization associated with Mississippi Valley–type fluids: A Pb isotope perspective
Geological Society of America Bulletin, v. 1XX, no. XX/XX 11
found in paleomagnetic and 40Ar/39Ar results on an intrusive breccia at Hicks dome south of the Illinois Basin, which records Permianaged igneous activity coincident with Upper Mississippi Valley fluid migration (Reynolds et al., 1997). Numerical modeling of fluid flow in the Upper Mississippi Valley district by Arnold et al. (1996) provided calculated temperatures of fluids in the district that may have approached those observed in the Upper Mississippi Valley district fluid inclusions. Considering that fluid temperature likely decreases during further migration northward, the fluid inclusion temperatures are broadly consistent with derivation from a high heatproducing granite (Fig. 9C).
Mineralization and Remobilization of AuGold (and silver) are only rarely reported in
MVT deposits (Heyl et al., 1959), so we explored if the MVT fluids that we propose to have been associated with the later texture Pbrich (and radiogenic) minerals are capable of Au mineralization. At low temperatures (<300 °C), gold is most efficiently transported in sulfide complexes, at either low or highpH conditions, whereas chloride complexes in chlorinerich fluids at lowpH conditions are only stable at higher temperature (300 °C) conditions (Stefáns son and Seward, 2004). MVT fluids, while in contact with carbonate sequences, will be pH buffered in a range of greater than ~4.5–5 (Leach et al., 2010). It has been proposed, however, that in an MVT fluid in equilibrium with crystalline rocks (Precambrian basement rocks), pH should decrease, enhancing the dissolved metal concentrations (Leach et al., 2010). Thus, it is possible that gold and other metals may have been sourced from the local Proterozoic country rocks, transported by MVT fluids, and then deposited in the Reef Deposit coincident with the late Pb and other base metal enrichment. This may explain why the Reef Deposit has a highergrade concentration of gold compared to other volcanogenic massive sulfide deposits in the region. We cannot rule out, however, that all of the Au was emplaced in the deposit during the initial Penokean mineralization and then was simply remobilized by the MVT fluids during Pbrich mineral precipitation on a local scale.
SUMMARY AND CONCLUSIONS
In this study, we used detailed petrographic examination and in situ analyses of Pbrich mineral phases along with evidence from fluid inclusions to better understand the protracted mineralization history of a Paleoproterozoic AuCu deposit. The Reef Deposit formed initially via hydrothermal activity associated with crustal construction during the Penokean orogeny (1890–1830 Ma).
Later hydrothermal overprinting remobilized and potentially enriched the gold mineralization. Pb isotope compositions from the later gold event overlap with Pbrich minerals found in MVT occurrences in the region that formed around 270 Ma. Further evidence for the presence of MVT fluids is found in lowtemperature saline fluid inclusion assemblages formed in late crosscutting carbonate veinlets.
As is apparent for the Upper Mississippi Valley district mineralization and minor occurrences in the district vicinity, a genetic model remains somewhat ambiguous for the radiogenic Pb mineralization present in the Reef Deposit. A factor unique to the Reef Deposit is that, unlike MVT deposits of the Upper Mississippi Valley district and vicinity, the mineralization is hosted in the Precambrian basement. A model that accounts for the radiogenic Pb and fluid temperatures involves a combination of gravitydriven flow of basinal brine fluids out of the Illinois Basin and circulation of hydrothermal fluid by high heatproducing granites, where both systems were driven by the same farfield stress related to the AppalachianOuachita orogeny. A separate model involving fluids driven from the Michigan Basin is also possible. Regardless, Pb isotope compositions are consistent with mineralization occurring during the late Paleozoic sourced from Mesoproterozoicaged material. Clearly, the presence of MVT fluids associated with Au mineralization/remobilization in a Proterozoic basementhosted Au deposit has important implications to our understanding of gold deposits and their potential protracted development globally. In addition, evidence for basement fluid flow during the AppalachianOuachita orogeny aids in our understanding of farfield tectonic effects across the region.
The UThPb isotope system has long provided a sensitive tracer of both the age and source of metals in ore deposits. A critical application of this system in this study, however, has been use of in situ isotopic analysis on the micron scale. We have shown that radiogenic Pb isotope compositions found by traditional sulfide analyses, even at the scale of microdrilling, are the product of incorporation of various micronscale Pbrich phases. In situ measurement of Pbrich phases, as done here, is required to see through the mixtures of Pbrich nuggets in bulkscale or even microdrilledscale samples. We posit that the mixing phenomena may be common, including prominent Archean gold and base metal terranes such as Western Australia or the southern Superior Province. Such in situ analytical methods, as applied here, make it possible to better decipher initial and subsequent mineralization events in gold and other ore deposits.
ACKNOWLEDGMENTS
This research was supported by student research grants from the Society of Economic Geology–Canada Foundation and the Geological Society of America. Haroldson thanks Aquila Resources, Inc., for access to samples. Scanning electron microscope petrography was assisted by Phil Gopon. Brian Hess created the laserablation sample mounts. The Pb isotope laserablation development was supported by National Science Foundation grant EAR1347056. Comments from Neal McNaughton, Julian Menuge, and the associate editor and editor helped to improve the manuscript. We would like to thank Gordon Medaris, Lawford Anderson, and Randy van Schmus for helpful discussions on Wisconsin geology.
REFERENCES CITED
Afifi, A., Doe, B.R., Sims, P.K., and Delevaux, M.H., 1984, UThPb isotope chronology of sulfide ores and rocks in the early Proterozoic metavolcanic belt of northern Wisconsin: Economic Geology and the Bulletin of the Society of Economic Geologists, v. 79, p. 338–353, https:// doi .org /10 .2113 /gsecongeo .79 .2 .338 .
Allen, D.J., and Hinze, W.J., 1992, Wisconsin gravity minimum: Solution of a geologic and geophysical puzzle and implications for cratonic evolution: Geology, v. 20, p. 515–518, https:// doi .org /10 .1130 /0091 7613 (1992)020 <0515: WGMSOA>2 .3 .CO;2 .
Anderson, J.L., 1983, Proterozoic anorogenic granite plutonism of North America, in Medaris, L.G., Jr., Byers, C.W., Mickelson, D.M., and Shanks, W.C., eds., Protero zoic Geology: Selected Papers from an International Proterozoic Symposium: Geological Society of America Memoir 161, p. 133–154, https:// doi .org /10 .1130 /MEM161 p133 .
Anderson, J.L., and Morrison, J., 1992, The role of anorogenic granites in the Proterozoic crustal development of North America, in Condie, K.C., ed., Proterozoic Crustal Evolution: Elsevier, p. 263–299, https:// doi .org /10 .1016 /S0166 2635 (08)70121 X .
Arnold, B.W., Bahr, J.M., and Fantucci, R., 1996, Paleohydro geology of the Upper Mississippi Valley zinclead district, in Sangster, D.F., CarbonateHosted LeadZinc Deposits: 75 Anniversary Volume: Society of Economic Geologists Special Publication 4, p. 378–389, https:// doi .org /10 .5382 /SP .04 .
Bethke, C.M., and Marshak, S., 1990, Brine migrations across North America—The plate tectonics of groundwater: Annual Review of Earth and Planetary Sciences, v. 18, p. 287–315, https:// doi .org /10 .1146 /annurev .ea .18 .050190 .001443 .
Blenkinsop, T.G., and Doyle, M.G., 2014, Structural controls on gold mineralization on the margin of the Yilgarn craton, AlbanyFraser orogen: The Tropicana deposit, Western Australia: Journal of Structural Geology, v. 67, p. 189–204, https:// doi .org /10 .1016 /j .jsg .2014 .01 .013 .
Bodnar, R., and Vityk, M., 1994, Interpretation of microthermometric data for H2ONaCl fluid inclusions, in DeVivo, B., and Frezzotti, M.L., eds., Fluid Inclusions in Minerals; Method and Applications: Blacksburg, Virginia, Virginia Polytechnic Institute and State University Press, p. 117–130.
Bottomley, D.J., Gregoire, D.C., and Raven, K.G., 1994, Saline groundwaters and brines in the Canadian Shield: Geochemical and isotopic evidence for a residual evaporate brine environment: Geochimica et Cosmo chimica Acta, v. 58, no. 5, p. 1483–1498, https:// doi .org /10 .1016 /0016 7037 (94)90551 7 .
Brannon, J.C., Podosek, F.A., and McLimans, R.K., 1992, Alleghenian age of the Upper Mississippi Valley zinclead deposit determined by RbSr dating of sphalerite: Nature, v. 356, p. 509–511, https:// doi .org /10 .1038 /356509a0 .
Brown, P., and Hagemann, S., 1994, MacFlinCor: A computer program for fluid inclusion data reduction and manipulation, in DeVivo, B., and Frezzotti, M.L., eds., Fluid Inclusions in Minerals; Method and Applica
Haroldson et al.
12 Geological Society of America Bulletin, v. 1XX, no. XX/XX
tions: Blacksburg, Virginia, Virginia Polytechnic Institute and State University Press, p. 231–250.
Butt, C.R.M., 1998, Supergene gold deposits: AGSO Journal of Australian Geology & Geophysics, v. 17, no. 4, p. 89–96.
Cannon, W.F., Kress, T.H., Sutphin, D.M., Morey, G.B., Meints, J., and BarberDelach, R., 1997, Digital Geologic Map and Mineral Deposits of the Lake Superior Region: U.S. Geological Survey OpenFile Report 97455, https:// pubs .usgs .gov /of /1997 /of97 455.
Chauvel, C., Bureau, S., and Poggi, C., 2011, Comprehensive chemical and isotope analyses of basalt and sediment reference materials: Geostandards and Geoanalytical Research, v. 35, no. 1, p. 125–143, https:// doi .org /10 .1111 /j .1751 908X .2010 .00086 .x .
Collerson, K.D., Kamber, B.S., and Schoenberg, R., 2002, Applications of accurate, highprecision Pb isotope ratio measurement by multicollector ICPMS: Chemical Geology, v. 188, p. 65–83, https:// doi .org /10 .1016 /S0009 2541 (02)00059 1 .
DeMatties, T.A., 1996, A geologic framework for early Proterozoic volcanogenic massive sulfide deposits in Wisconsin: An exploration model, in LaBerge, G.L., ed., Volcanogenic Massive Sulfide Deposits of Northern Wisconsin: A Commemorative Volume: Cable, Wisconsin, Institute on Lake Superior Geology, p. 31–65.
Deming, D., and Nunn, J.A., 1991, Numerical simulations of brine migration by topographically driven recharge: Journal of Geophysical Research, v. 96, no. B2, p. 2485–2499, https:// doi .org /10 .1029 /90JB02392 .
Dewane, T.J., and Van Schmus, W.R., 2007, UPb geochronology of the Wolf River Batholith, northcentral Wisconsin: Evidence for successive magmatism between 1484 Ma and 1468 Ma: Precambrian Research, v. 157, p. 215–234, https:// doi .org /10 .1016 /j .precamres .2007 .02 .018 .
Driese, S.G., and Medaris, L.G., Jr., 2008, Evidence for biological and hydrological controls on the development of a Paleoproterozoic paleoweathering profile in the Baraboo Range, Wisconsin, U.S.A.: Journal of Sedimentary Research, v. 78, no. 7, p. 443–457, https:// doi .org /10 .2110 /jsr .2008 .051 .
Fehn, U., Cathles, L.M., and Holland, H.D., 1978, Hydrothermal convection and uranium deposits in abnormally radioactive plutons: Economic Geology and the Bulletin of the Society of Economic Geologists, v. 73, p. 1556–1566, https:// doi .org /10 .2113 /gsecongeo .73 .8 .1556 .
Galer, S.J.G., and Abouchami, W., 1998, Practical application of lead triple spiking for correction of instrumental mass discrimination: Mineralogical Magazine, v. 62A, p. 491–492, https:// doi .org /10 .1180 /minmag .1998 .62A .1 .260 .
Girard, J.P., and Barnes, D.A., 1995, Illitization and paleothermal regimes in the middle Ordovician St. Peter sandstone, central Michigan Basin: KAr, oxygen isotope, and fluid inclusion data: American Association of Petroleum Geologists Bulletin, v. 79, p. 49–69.
Goldstein, R., 2003, Petrographic analysis of fluid inclusions, in Samson, I., Anderson, A., and Marshall, D., eds., Fluid Inclusions: Analysis and Interpretation: Mineralogical Association of Canada Short Course 32, p. 9–54.
Goldstein, R., and Reynolds, T.J., 1994, Systematics of Fluid Inclusions in Diagenetic Minerals: Society for Sedimentary Geology (SEPM) Short Course 31, 199 p.
Heyl, A.V., and West, W.S., 1982, Outlying mineral occurrences related to the Upper Mississippi Valley mineral district, Wisconsin, Iowa, Illinois, and Minnesota: Economic Geology and the Bulletin of the Society of Economic Geologists, v. 77, p. 1803–1817, https:// doi .org /10 .2113 /gsecongeo .77 .8 .1803 .
Heyl, A.V., Agnew, A.F., Lyons, E.J., Behre, C.H., and Flint, A.E., 1959, The Geology of the Upper Mississippi Valley ZincLead District: U.S. Geological Survey Professional Paper 309, 310 p.
Heyl, A.V., Landis, G.P., and Zartman, R.E., 1974, Isotopic evidence for the origin of Mississippi Valley–type mineral deposits: A review: Economic Geology and the Bulletin of the Society of Economic Geologists, v. 69, p. 992–1006, https:// doi .org /10 .2113 /gsecongeo .69 .6 .992 .
Ingebritsen, S.E., and Appold, M.S., 2012, The physical hydro geology of ore deposits: Economic Geology and the Bulletin of the Society of Economic Geologists, v. 107, no. 4, p. 559–584, https:// doi .org /10 .2113 /econgeo .107 .4 .559 .
Kennedy, L.P., and Harding, T.A., 1990, Summary Report of the Reef Joint Venture, Marathon County, Wisconsin: Rhinelander, Wisconsin, Noranda Exploration Inc., 63 p.
Kerrich, R., and Cassidy, K.F., 1994, Temporal relationships of lode gold mineralization to accretion, magmatism, metamorphism and deformation—Archean to present: A review: Ore Geology Reviews, v. 9, p. 263–310, https:// doi .org /10 .1016 /0169 1368 (94)90001 9 .
LaBerge, G.L., and Myers, P.E., 1984, Two early Proterozoic successions in central Wisconsin and their tectonic significance: Geological Society of America Bulletin, v. 95, no. 2, p. 246–253, https:// doi .org /10 .1130 /0016 7606 (1984)95 <246: TEPSIC>2 .0 .CO;2 .
Large, R.L., Danyushevsky, L., Hollit, C., Maslennikov, V., Meffre, S., Gilbert, S., Bull, S., Scott, R., Emsbo, P., Thomas, H., Singh, B., and Foster, J., 2009, Gold and trace element zonation in pyrite using a laser imaging technique: Implications for the timing of gold in orogenic and Carlinstyle sedimenthosted deposits: Economic Geology and the Bulletin of the Society of Economic Geologists, v. 104, p. 635–668, https:// doi .org /10 .2113 /gsecongeo .104 .5 .635 .
Larizzatti, J.H., Oliveira, S.M.B., and Butt, C.R.M., 2008, Morphology and composition of gold in a lateritic profile, Fazenda Pison “Garimpo,” Amazon, Brazil: Journal of South American Earth Sciences, v. 25, p. 359–376, https:// doi .org /10 .1016 /j .jsames .2007 .06 .002 .
Lawrance, L.M., 1990, Supergene gold deposits, in Ho, W.E., Groves, D.I., and Bennet J.M., eds., Gold Deposits of the Archaean Yilgarn Block, Western Australia: Nature, Genesis and Exploration Guides: Crawley, Western Australia, University of Western Australia Publication 20, p. 299–314.
Leach, D.L., Sangster, D.F., Kelley, K.D., Large, R.R., Garven, G., Allen, C.R., Gutzner, J., and Walters, S., 2005, Sedimenthosted leadzinc deposits: A global perspective: Economic Geology, 100th Anniversary Volume, p. 561–608.
Leach, D.L., Taylor, R.D., Fey, D.L., Diehl, S.F., and Saltus, R.W., 2010, A deposit model for Mississippi Valley–type leadzinc ores, in Mineral Deposit Models for Resource Assessment: U.S. Geological Survey Scientific Investigations Report 20105070A, 52 p.
Li, Q., Chen, F., Yang, J., and Fan, H., 2008, Single grain pyrite RbSr dating of the Linglong gold deposit, eastern China: Ore: Dizhi Lunping, v. 34, p. 263–270, doi: 10 .1016 /j .oregeorev .2007 .10 .003 .
Luczaj, J.A., 2006, Evidence against the Dorag (mixingzone) model for dolomitization along the Wisconsin arch—A case for hydrothermal diagenesis: American Association of Petroleum Geologists Bulletin, v. 90, p. 1719–1738, https:// doi .org /10 .1306 /01130605077 .
Luczaj, J.A., Martin, J.B., and Millen, T.M., 2007, A leadisotopic study of galena from eastern Wisconsin: Evidence for lead sources in Precambrian basement rocks: Geological Society of America Abstracts with Programs, v. 39, no. 3, p. 67.
Luczaj, J.A., McIntire, M.J., and Hunt, M.J.O., 2016, Geochemical characterization of trace MVT mineralization in Paleozoic sedimentary rocks of northeastern Wisconsin, USA: Geosciences, v. 6, no. 2, p. 29.
Ludwig, K.R., 2003, Isoplot/EX Version 3.0: A Geochronological Toolkit for Microsoft Excel: Berkeley Geochronology Center Special Publication 4, 74 p.
Ma, L., Castro, M.C., and Hall, C.H., 2009, Crustal noble gasses in deep brines as natural traces of vertical transport processes in the Michigan Basin: Geochemistry, Geophysics, Geosystems, v. 10, Q06001, no. 6, 24 p., https:// doi .org /10 .1029 /2009GC002475 .
McNaughton, N.J., and Groves, D.I., 1996, A review of Pbisotope constraints on the genesis of lodegold deposits in the Yilgarn craton, Western Australia: Journal of the Royal Society of Western Australia, v. 79, p. 123–129.
Millen, T.M., Zartman, R.E., and Heyl, A.V., 1995, Lead Isotopes from the Upper Mississippi Valley District—A
Regional Perspective: U.S. Geological Survey Bulletin 2094B, 13 p.
Perring, C.S., and McNaughton, J., 1990, Proterozoic remobilization of ore metals within Archean gold deposits: Lead isotope evidence from Norseman, Western Australia: Australian Journal of Earth Sciences, v. 37, p. 369–372, https:// doi .org /10 .1080 /08120099008727934 .
Platzner, I., Ehrlich, S., and Halicz, L., 2001, Isotoperatio measurements of lead in NIST standard reference materials by multiplecollector inductively coupled mass spectrometry: Fresenius’ Journal of Analytical Chemistry, v. 370, no. 5, p. 624–628, https:// doi .org /10 .1007 /s002160100875 .
Quigley, A., Monecke, T., Anderson, E., Kelly, N., and Quigley, P., 2016, Setting of volcanogenic massive sulfide deposits of the Paleoproterozoic Penokean volcanic belt, in Schardt, C., and Miller, J., eds., Institute on Lake Superior Geology Annual Meeting, Volume 62: Duluth, Minnesota, Institute on Lake Superior Geology, p. 121–122.
Reynolds, R.L., Goldhaber, M.B., and Snee, L.W., 1997, Paleomagnetic and 40Ar/39Ar Results from the Grant Intrusive Breccia and Comparison to the Permian Downeys Bluff Sill—Evidence for Permian Igneous Activity at Hicks Dome, Southern Illinois Basin: U.S. Geological Survey Bulletin 2094G, 16 p.
Ridley, W.I., 2005, PlumboIsotopy: The Measurement of Lead Isotopes by MultiCollector Inductively Coupled Mass Spectrometry: U.S. Geological Survey OpenFile Report 20051393, 16 p.
Rowan, E.L., and Goldhaber, M.B., 1996, Fluid Inclusions and Biomarkers in the Upper Mississippi Valley ZincLead District—Implications for the FluidFlow and Thermal History of the Illinois Basin: U.S. Geological Survey Bulletin 2094F, 34 p.
Satkoski, A.M., Beukes, N.J., Li, W., Beard, B.L., and Johnson, C.M., 2015, A redoxstratified ocean 3.2 billion years ago: Earth and Planetary Science Letters, v. 430, p. 43–53, https:// doi .org /10 .1016 /j .epsl .2015 .08 .007 .
Schulz, K.J., and Cannon, W.F., 2007, The Penokean orogeny in the Lake Superior region: Precambrian Research, v. 157, p. 4–25, https:// doi .org /10 .1016 /j .precamres .2007 .02 .022 .
Sheets, R.W., Craig, J.R., and Bodnar, R.J., 1995, Composition and occurrence of electrum at the Morning Star deposit, San Bernardino County, California: Evidence for remobilization of gold and silver: The Canadian Mineralogist, v. 33, p. 137–151.
Sicree, A.A., and Barnes, H.L., 1999, Anomalous lead isotopes and sources of leads in the Upper Mississippi Valley zinclead district, USA: V.M. Goldschmidt Conference Abstracts, v. 9, p. 276.
Sims, P.K., Van Schmus, W.R., Schulz, K.J., and Peterman, Z.E., 1989, Tectonostratigraphic evolution of the early Proterozoic Wisconsin magmatic terranes of the Penokean orogen: Canadian Journal of Earth Sciences, v. 26, p. 2145–2158, https:// doi .org /10 .1139 /e89 180 .
Spirakis, C.S., 1995, Problems with Applying Topographically Driven Flow to Genesis of the Upper Mississippi Valley ZincLead District and to Fluid Flow in the Illinois Basin: U.S. Geological Survey Bulletin 2094C, 12 p.
Spirakis, C.S., and Heyl, A.V., 1996, Fracturing radioactive basement rocks as the energy source to drive mineralization in the Upper Mississippi Valley zinclead district, in Sangster, D.F., ed., CarbonateHosted LeadZinc Deposits: 75th Anniversary Volume: Society of Economic Geologists Special Publication 4, p. 390–399.
Stacey, J.S., and Kramers, J.D., 1975, Approximation of terrestrial lead isotope evolution by a twostage model: Earth and Planetary Science Letters, v. 26, p. 207–221, https:// doi .org /10 .1016 /0012 821X (75)90088 6 .
Stefánsson, A., and Seward, T.M., 2004, Gold(I) complexing in aqueous sulphide solutions to 500°C at 500 bar: Geochimica et Cosmochimica Acta, v. 68, no. 20, p. 4121–4143, https:// doi .org /10 .1016 /j .gca .2004 .04 .006 .
Taylor, R.D., Goldfarb, R.J., Monecke, T., Fletcher, I.R., Cosca, M.A., and Kelly, N.M., 2015, Application of UThPb phosphate geochronology to young orogenic gold deposits: New age constraints on the formation of the Grass Valley gold district, Sierra Nevada foothills province,
Gold remobilization associated with Mississippi Valley–type fluids: A Pb isotope perspective
Geological Society of America Bulletin, v. 1XX, no. XX/XX 13
California: Economic Geology and the Bulletin of the Society of Economic Geologists, v. 110, p. 1313–1337, https:// doi .org /10 .2113 /econgeo .110 .5 .1313 .
Thirlwall, M.F., 2002, Multicollector ICPMS analysis of Pb isotopes using a 207Pb204Pb double spike demonstrates up to 400 ppm/amu systematic errors in Tlnormalization: Chemical Geology, v. 184, p. 255–279, https:// doi .org /10 .1016 /S0009 2541 (01)00365 5 .
Thorpe, R.I., 2008, Release of Lead Isotope Data in 4 Databases: Canadian, Western Superior, Foreign, and whole rock and Feldspar: Geological Survey of Canada OpenFile Report 5664.
Todd, E., Stracke, A., and Scherer, E.E., 2015, Effects of simple acid leaching of crushed and powdered geological materials on highprecision Pb isotope analyses: Geochemistry Geophysics Geosystems, v. 16, p. 2276–2302, https:// doi .org /10 .1002 /2015GC005804 .
Vielreicher, N., Groves, D., McNaughton, N., and Fletcher, I., 2015, The timing of gold mineralization across the eastern Yilgarn craton using UPb geochronology of hydrothermal phosphate minerals: Mineralium Deposita, v. 50, p. 391–428, https:// doi .org /10 .1007 /s00126 015 0589 9 .
Weis, D., Kieffer, B., Maerschalk, C., Pretorius, W., and Barling, J., 2005, Highprecision PbSrNdHf isotopic characterization of USGS BHVO1 and BHVO2 reference materials: Geochemistry Geophysics Geosystems, v. 6, no. 2, Q02002, https:// doi .org /10 .1029 /2004GC000852 .
Weis, D., Kieffer, B., Maerschalk, C., Barling, J., de Jong, J., Williams, G.A., Hanano, D., Pretorius, W., Mattieli, N., Scoates, J.S., Goolaerts, A., Friedman, R.M., and Mahoney, J.B., 2006, Highprecision isotopic characterization of USGS reference materials by TIMS and MCICPMS: Geochemistry Geophysics Geosystems, v. 7, no. 8, Q08006, https:// doi .org /10 .1029 /2006GC001283 .
Wilkinson, J.J., 2014, Sedimenthosted zinclead mineralization: Processes and perspectives, in Holland, H., and Turekian, K., eds., Treatise on Geochemistry (2nd ed.): Amsterdam, Netherlands, Elsevier, v. 13, p. 219–249, https:// doi .org /10 .1016 /B978 0 08 095975 7 .01109 8
Wilkinson, J.J., 2017, Metastable freezing: A new method for the estimation of salinity in aqueous fluid inclusions: Economic Geology and the Bulletin of the Society of Economic Geologists, v. 112, p. 185–193, https:// doi .org /10 .2113 /econgeo .112 .1 .185 .
Wilkinson, J.J., Boyce, A.J., Earls, G., and Fallick, A.E., 1999, Gold remobilization by lowtemperature brines: Evidence from the Curraghinalt gold deposit, northern Ireland: Economic Geology and the Bulletin of the Society of Economic Geologists, v. 94, p. 289–296, https:// doi .org /10 .2113 /gsecongeo .94 .2 .289 .
Wilson, S.A., 1997, The Collection, Preparation and Testing of USGS Reference Material BCR2, Columbia River Basalt: U.S. Geological Survey OpenFile Report 9800x, 3 p.
Wilson, S.A., 1998, Data Compilation and Statistical Analysis of Intralaboratory Results for AGV2: U.S. Geological Survey OpenFile Report, 3 p.
Woodhead, J.D., and Hergt, J.M., 2000, Pbisotope analyses of USGS reference materials: Geostandards Newsletter, v. 24, no. 1, p. 33–38, https:// doi .org /10 .1111 /j .1751 908X .2000 .tb00584 .x .
Zachariáš, J., Žák, K., Pudilová, M., and Snee, L.W., 2013, Multiple fluid sources/pathways and severe thermal gradients during formation of the Jílové orogenic gold deposit, Bohemian Massif, Czech Republic: Ore: Dizhi Lunping, v. 54, p. 81–109, https:// doi .org /10 .1016 /j .oregeorev .2013 .02 .012 .
Zhang, J., Linnen, R., Lin, S., Davis, D., and Martin, R., 2014, Paleoproterozoic hydrothermal reactivation in a Neoarchean orogenic lodegold deposit of the southern Abitibi subprovince: UPb monazite geochronological evidence from the YoungDavidson mine, Ontario: Precambrian Research, v. 249, p. 263–272, https:// doi .org /10 .1016 /j .precamres .2014 .05 .015 .
Science editor: david i. Schofield aSSociate editor: Brian Mcconnell
ManuScript received 2 auguSt 2017 reviSed ManuScript received 13 deceMBer 2017 ManuScript accepted 23 feBruary 2018
Printed in the USA