A preliminary study of older hot spring alteration in Sevenmile Hole, Grand Canyon of the Yellowstone River, Yellowstone Caldera, Wyoming

Journal of Volcanology and Geothermal Research 188 (2009) 225–236

Contents lists available at ScienceDirect

Journal of Volcanology and Geothermal Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j vo l g e o r e s

A preliminary study of older hot spring alteration in Sevenmile Hole, Grand Canyon of the Yellowstone River, Yellowstone Caldera, Wyoming Peter B. Larson a,⁎, Allison Phillips a, David John b, Michael Cosca c, Chad Pritchard a, Allen Andersen a, Jennifer Manion a a b c

School of Earth and Environmental Sciences, Washington State University, Pullman, WA 99164-2812, USA United States Geological Survey, 345 Middlefield Road, Menlo Park, CA 94025, USA United States Geological Survey, Denver Federal Center, Denver, C0 80225, USA

a r t i c l e

i n f o

Article history: Received 10 April 2008 Accepted 20 July 2009 Available online 7 August 2009 Keywords: hydrothermal caldera rhyolite Yellowstone epithermal

a b s t r a c t Erosion in the Grand Canyon of the Yellowstone River, Yellowstone Caldera (640 ka), Wyoming, has exposed a cross section of older hydrothermal alteration in the canyon walls. The altered outcrops of the post-collapse tuff of Sulphur Creek (480 ka) extend from the canyon rim to more than 300 m beneath it. The hydrothermal minerals are zoned, with an advanced argillic alteration consisting of an association of quartz (opal) + kaolinite ± alunite ± dickite, and an argillic or potassic alteration association with quartz + illite ± adularia. Disseminated fine-grained pyrite or marcasite is ubiquitous in both alteration types. These alteration associations are characteristic products of shallow volcanic epithermal environments. The contact between the two alteration types is about 100 m beneath the rim. By analogy to other active geothermal systems including active hydrothermal springs in the Yellowstone Caldera, the transition from kaolinite to illite occurred at temperatures in the range 150 to 170 °C. An 40Ar/39Ar age on alunite of 154,000 ± 16,000 years suggests that hydrothermal activity has been ongoing since at least that time. A northwest-trending linear array of extinct and active hot spring centers in the Sevenmile Hole area implies a deeper structural control for the upflowing hydrothermal fluids. We interpret this deeper structure to be the Yellowstone Caldera ring fault that is covered by the younger tuff of Sulphur Creek. The Sevenmile Hole altered area lies at the eastern end of a band of hydrothermal centers that may mark the buried extension of the Yellowstone Caldera ring fault across the northern part of the Caldera. © 2009 Elsevier B.V. All rights reserved.

1. Introduction More than three hundred vertical meters of pervasively hydrothermally altered post-Yellowstone Caldera collapse rhyolites are exposed in the walls of the Grand Canyon of the Yellowstone River, Yellowstone Caldera, Wyoming. The Grand Canyon has exposed about a thousand vertical feet of older altered rocks and provides an excellent opportunity for extensive cross-sectional studies of hydrothermal alteration. Our preliminary work in the Sevenmile Hole area of the Canyon, reported here, has found various hydrothermal minerals including quartz, opal, kaolinite, dickite, alunite, illite, adularia, and pyrite in the canyon walls. These hydrothermal minerals form two major alteration types, advanced argillic and intermediate argillic, that are characteristic of shallow hydrothermal systems developed in volcanic environments (Cooke and Simmons, 2000; Simmons et al., 2005). Elsewhere, geothermal systems are used as energy resources (e.g., Broadlands–Ohaaki, New Zealand, Henley and

⁎ Corresponding author. E-mail address: [email protected] (P.B. Larson). 0377-0273/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jvolgeores.2009.07.017

Ellis, 1983), and extinct shallow volcanic hydrothermal systems are important hosts for precious and base metal mineral deposits (e.g., Summitville, Colorado, Stoffregen, 1987). By analogy to the active Yellowstone hydrothermal system, the older hydrothermal alteration in the Grand Canyon of the Yellowstone River was most likely formed in an environment of steep thermal gradients by rising and boiling fluids as observed elsewhere in the Park's thermal basins (White et al., 1975). Many Yellowstone hot springs, including the hot spring basins along the upper Firehole River (Old Faithful area), are discharging boiling, near-neutral to alkaline chloride-rich fluids as first reported by Allen and Day (1935). Allen and Day (1935) and White et al. (1971, 1988) note that these alkalinechloride fluids are the most common type of fluid discharged from the Yellowstone hot springs. These boiled fluids have pH in the range from 6 to 8 and high Cl and SiO2 concentrations. Other fluid compositions are found in many areas of the Yellowstone geothermal system (White et al., 1971, 1988; Fournier, 1989). As noted at Norris Geyser Basin by White et al. (1988), these include Cl− and SO− 4 rich acid waters, and acid-sulfate waters with low Cl concentrations. Boiling and fluid mixing are common near-surface processes at Yellowstone and influence the compositions of all four fluid types, and

226

P.B. Larson et al. / Journal of Volcanology and Geothermal Research 188 (2009) 225–236

condensation of deep steam may be important in some areas (White et al., 1971, 1988). Studies of a number of epithermal systems have shown that the alkaline-chloride and acid-sulfate-chloride fluids alter intermediate to silicic igneous rocks to distinctively different hydrothermal mineral assemblages (Henley and Ellis, 1983; Cooke and Simmons, 2000). These hydrothermal associations would, therefore, be expected to occur at Yellowstone at shallow levels in the geothermal system. Epithermal alteration forms in near-surface hydrothermal environments, and numerous descriptive classifications for epithermal systems have been proposed in the past several decades (see Simmons et al., 2005, for a thorough recent summary). Nearly all of them divide epithermal mineral deposits into two broad classes that draw important distinctions based on fluid pH and redox state that are reflected in their associated hydrothermal mineral assemblages. The most important difference in fluid chemistry is pH. Low pH fluids produce an alteration assemblage that usually includes alunite, pyrophyllite, and dickite as characteristic phases, and may contain residual vuggy quartz and kaolinite. This type of alteration has been referred to as acid sulfate (Hayba et al., 1985; Heald et al., 1987) and high sulfidation (White and Hedenquist, 1990). Neutral to slightly alkaline fluids produce an assemblage that may include quartz, adularia, illite, and carbonate minerals as characteristic minerals. This assemblage has been called adularia-sericite (Hayba et al., 1985; Heald et al., 1987) and low sulfidation (White and Hedenquist, 1990). As noted by Simmons et al. (2005), most epithermal alteration is associated with subduction-related magmatism, although occurrences are known in back-arc, continental rift, and postsubduction magmatic

environments. Yellowstone volcanism, however, is related to continental hot-spot magmatism (for example, Pierce and Morgan, 1992; Smith and Braile, 1994; Camp, 1995; Shervais and Hanan, 2008), although not all would agree (Christiansen et al., 2002). Most of the active hot springs at Yellowstone (Fig. 1) are confined within the area of the Yellowstone Caldera that formed at 0.640 Ma (Christiansen, 2001), and most lie near the Caldera margins (Fournier, 1989). A north-trending linear array of hot springs extends from Norris Junction to Mammoth Hot Springs (Fig. 1), and their distribution is most likely controlled by normal faults that serve as fluid conduits (Pierce et al., 1991; Kharaka et al., 2000). The architecture of the shallow portion of the Yellowstone hydrothermal system (White et al., 1988) is similar to the Caldera hydrothermal flow model for the 23-Ma Lake City Caldera, Colorado (Larson and Taylor, 1986a,b, 1987), and to the active system associated with the Valles Caldera, New Mexico (Goff and Gardner, 1994). At both Lake City and Valles Calderas, faults along caldera ring zones are important channelways for upflowing hydrothermal fluids. By analogy to these other caldera hydrothermal systems, a linear array of active and extinct hot spring centers in the Sevenmile Hole area could be the surficial expression of a deeper structural control over fluid flow (White et al., 1971, 1988; Christiansen, 2001). This inferred structure lies semi-parallel but inward from the Yellowstone Caldera topographic wall. This positioning suggests that this structure would most likely be a fault related to the Yellowstone Caldera margin that is concealed by the post-collapse tuff of Sulphur Creek (a ring fault?). However, direct evidence for a buried deeper structural control is lacking. The initial results from our investigations of the older hydrothermal alteration in the Sevenmile Hole altered area of the Grand Canyon of the

Fig. 1. A geologic map of the Yellowstone Caldera modified from Christiansen (2001) and Morgan et al. (2007), that shows the distribution of post-collapse rhyolites and hydrothermal areas. The heavy black line is the topographic rim of the Caldera, and it is dashed where covered by younger rhyolite flows. The two resurgent domes are outlined by light lines and are dissected by grabens. The Sour Creek dome lies to the northeast and the Mallard Lake dome to the southwest. The Elephant Back fault system extends southwest from the Sour Creek dome. The sense of fault movement is not shown on the domes' graben faults (heavy lines). The Norris–Mammoth corridor of post collapse rhyolites and active hot springs lies outside the Caldera and is shown extending northward from its central area. The corridor lies along a broad structural basin that is most likely related to basin-range tectonics. Several of the basin bounding faults are shown with balls on their downthrown sides. The area of Fig. 2 is outlined.

P.B. Larson et al. / Journal of Volcanology and Geothermal Research 188 (2009) 225–236

Yellowstone River are reported here. Major research objectives are to: (1) determine the nature and distribution of hydrothermal minerals, (2) measure the age of the alteration, (3) define the physical conditions of the hydrothermal environment, and (4) assess the nature of the hydrothermal fluids and processes that influence their composition, such as boiling and mixing. This communication presents our first results about the nature and distribution of the hydrothermal minerals and age of alteration in the Sevenmile Hole area, and compares them to aspects of other shallow volcanic hydrothermal systems. 2. Geologic setting of the Grand Canyon of the Yellowstone River The Yellowstone Caldera collapsed during eruption of the Lava Creek Tuff at 640 ka (Christiansen, 2001). The Lava Creek Tuff is the youngest of three major caldera-forming ignimbrite eruptions from the Yellowstone area, and was emplaced after the eruption of the Huckleberry Ridge Tuff at 2.05 Ma and the Mesa Falls Tuff at 1.3 Ma. Numerous postcollapse rhyolites, the Pleistocene Plateau Rhyolites of Christiansen (2001), fill the Yellowstone Caldera and cover projected ring faults. The Caldera's topographic wall, however, is obvious in many areas along the Caldera margin. The Plateau rhyolites are mostly vitrophyric high-silica

227

flows (three are pyroclastic), and typically contain phenocrysts of quartz and sanidine, with minor clinopyroxene, magnetite, and locally fayalite (Hildreth et al., 1991; Christiansen, 2001). A few of the rhyolites also contain sodic plagioclase phenocrysts. A group of Plateau Rhyolite domes extends in a linear trend northward from the Caldera, from the Norris Geyser Basin to Mammoth Hot Springs. Together with active thermal springs along this trend, these domes define the Norris– Mammoth corridor (Fig. 1). Two resurgent domes formed in the Caldera, the Sour Creek Dome to the northeast and the Mallard Lake Dome to the southwest. Both domes have apical grabens, and the Elephant Back fault zone extends southwestward from the Sour Creek dome (Fig. 1). Active hot spring activity is widespread in the Caldera, and tends to occur in broad arcuate zones around the resurgent domes or along the Caldera margin (Fig. 1), and along the Norris–Mammoth corridor (White et al., 1988; Christiansen, 2001). The pervasively altered rocks exposed in the Grand Canyon of the Yellowstone River lie between the Sour Creek Dome to the south and the northern rim of the Yellowstone Caldera (Fig. 1) (Christiansen and Blank, 1975; Prostka et al., 1975; Christiansen, 2001). The rocks along the walls in the upper Grand Canyon of the Yellowstone River (Fig. 2) are pyroclastic flows and lava flows of the Upper

Fig. 2. Generalized geologic map of the Grand Canyon of the Yellowstone River in the Canyon Junction area modified from Christiansen and Blank (1975), Prostka et al. (1975), and Christiansen (2001). Eocene volcanic rocks of the Absaroka Volcanics lie north and outside the Yellowstone Caldera. Rhyolites of the Upper Basin Member of the Plateau Rhyolites line the canyon walls and are the main host for the hydrothermal alteration. The two important areas of older alteration are below the lower falls, in the Inspiration and Artists Point area, and at Sevenmile Hole near the Caldera margin. The rectangle encloses the area of Fig. 3 and includes the location of Sevenmile Hole.

228

P.B. Larson et al. / Journal of Volcanology and Geothermal Research 188 (2009) 225–236

Basin Member of the Yellowstone Plateau Rhyolites (Christiansen, 1975; Christiansen and Blank, 1975; Prostka et al., 1975; Christiansen, 2001). Unlike the younger postcollapse rhyolites of the Central Plateau Member, units in the Upper Basin Member contain sodic plagioclase as a major phenocryst phase and may lack sanidine. Unaltered samples of all these units are vitrophyric. From youngest to oldest, the units of the Upper Basin Member are: (1) the Dunraven Road Flow, (2) the Canyon Flow, which is exposed in the upper part of the canyon walls for several kilometers below the Lower Falls where it is pervasively hydrothermally altered, (3) the tuff of Sulphur Creek, which lies below the Canyon Flow in the canyon walls where it is also pervasively altered, and (4) the tuff of Uncle Tom's Trail, which has only limited exposure. 40 Ar/39Ar ages of the tuff of Sulphur Creek, the Canyon Flow, and the Dunraven Road Flow are closely spaced at 0.479 ± 0.010, 0.484 ± 0.015, and 0.486 ± 0.042 Ma, respectively (Gansecki et al., 1996), and here we refer to them collectively as having erupted at 480,000 years ago. These units postdate the Lava Creek Tuff by about 160,000 years. It is noteworthy that the Upper Basin Member rhyolites have very low magmatic δ18O values that are most likely inherited from source rocks that were hydrothermally altered by meteoric fluids (Friedman et al., 1974; Hildreth et al., 1984, 1991; Bindeman and Valley, 2000, 2001; Bindeman et al., 2001). Units in the Upper Basin Member also exhibit strong vertical geochemical zonation (Hildreth et al., 1984, 1991; Christiansen, 2001). Alteration in both the Canyon Flow and the tuff of Sulphur Creek extends 3 to 4 km to the northeast along the canyon walls below the Lower Falls (Fig. 2). Downstream from this point to the area of Sevenmile Hole, about 9 km below the falls, the walls consist of weakly altered to unaltered tuff of Sulphur Creek (Christiansen and Blank, 1975). At Sevenmile Hole, the canyon walls widen and the pervasive

alteration affects several square kilometers of outcrop of the tuff (Prostka et al., 1975). This altered area straddles the projection of the Yellowstone Caldera wall (Christiansen, 2001), and outside the Caldera wall, the alteration affects the older Eocene Washburn volcano, part of the Absaroka Volcanics (Feeley et al., 2002; Feeley, 2003). The Washburn volcano was dissected by the Yellowstone Caldera collapse and its remnants form a prominent part of the Caldera's northern topographic wall. Active thermal springs are scattered along the lower canyon, particularly near the Yellowstone River. The history of sedimentation on the rim and in the canyon is complex and provides constraints on the timing of glaciation, canyon incision, and hydrothermal activity (Pierce, 1974). The rims of the Yellowstone River canyon walls are capped in places with clastic sedimentary deposits that are in part glacial (Richmond, 1976, 1977). Some thin-bedded sediments are also present on lower benches in the canyon and are most likely fluvial deposits that were stranded during canyon incision. Some of the sediments on the rim are hydrothermally altered and are cemented by hydrothermal minerals. Christiansen and Blank (1975) note that the Pleistocene sediments of Inspiration Point are intensely altered and difficult to distinguish from the underlying altered volcanic rocks. Abandoned hot spring sinter deposits are also scattered on the areas around the canyon rim and in places are interbedded with older sediments (Hinman and Walter, 2005). The hot spring features, sediments, and hydrothermal cement on the areas around the canyon rim suggest that the rim area has not been eroded extensively since the last glaciation. However, erosion during and prior to the last glaciation most likely removed material so that the paleosurface at that time was stratigraphically higher than the current surface. The contact between the Canyon flow and the Dunraven Road flow, if horizontal, would project to up to 100 m higher than the

Fig. 3. Map showing areas of hydrothermal alteration and hydrothermal minerals in the Sevenmile Hole area. The heavy line in the lower right is the Yellowstone River, which flows to the northeast. Circles and squares are the locations of samples that have PIMA and XRD mineral determinations. The slightly larger open circle is the location of sample Y-05-25 that yielded the Ar age of 154 ka. Location 7741 is shown by the heavy X.

P.B. Larson et al. / Journal of Volcanology and Geothermal Research 188 (2009) 225–236

current rim between Inspiration Point and Sevenmile Hole, although the preglacial extent of the Dunraven Road flow is unknown. Pierce et al. (1976) used obsidian hydration rinds together with stratigraphic relations of glacial deposits relative to dated obsidian flows to estimate the ages of the last two Yellowstone glaciations, the Bull Lake and the Pinedale glaciations, near West Yellowstone, Montana. More recent cosmogenic exposure dates (Licciardi and Pierce, 2008) have refined these glaciation ages. The Bull Lake glaciation ranged from 130,000 to about 155,000 years ago, and the Pinedale ranged from less than 20,000 to 35,000 years ago. 10Be ages of the end of Pinedale glaciation at Yellowstone range from 18.8 to 16.5 ka, and may be as young as 14.6 ka (Licciardi and Pierce, 2008). The age of the alteration in the canyon is bracketed by the age of the protolith volcanic units at 480,000 years and the incision of the canyon. Incision appears to have been older than the most recent glaciation because Bull Lake–Pinedale interglacial and other older sediments have been mapped within the canyon (Pierce, 1974; Richmond, 1977). 3. Sampling and analyses The distribution of hydrothermally altered rocks was mapped over about 1 km2 in the Sevenmile Hole area. Two to four kilogram hand samples located by a handheld GPS were collected from many outcrops for laboratory analyses. Care was taken to avoid weathered samples. A portable infrared mineral analyzer (PIMA) was employed in the field to determine the hydrothermal mineralogy for many of the samples. The PIMA spectrometer measures shortwave infrared (SWIR; 1300–2500 nm wavelength) reflectance from hand samples and is particularly useful for identifying clay (e.g., kaolinite, dickite, illite, smectite), hydrous sulfate (alunite, jarosite, gypsum), carbonate, and ammonium minerals (Thompson et al., 1999). Characteristic SWIR absorptions in spectra correspond to water and –OH bond energies, which are characteristic of each hydrous mineral. Standard X-ray diffractometer (XRD) analyses were used in the laboratory to confirm the PIMA mineral identifications and to look for minerals that have poor SWIR response (e.g., quartz and feldspars) or were not present in great enough concentrations to be detected by the PIMA. Petrographic and electron microprobe analyses of selected samples were conducted in the laboratories of the GeoAnalytical Laboratory at Washington State University. The distributions of hydrothermal mineral types or associations in the Sevenmile Hole altered area are shown in Fig. 3. The 40Ar/39Ar data were collected from a single fragment of alunite from sample Y-05-25, approximately 0.5 cm3 in size. The alunite fragment (microprobe analyses are in Table 1) together with monitors Table 1 Electron microprobe analyses of hydrothermal alunite and jarosite for four samples from the Sevenmile Hole area. Sample n Weight percent Na2O CaO K2O BaO Al2O3 Fe2O3 SO3 Oxide total

YS-05-25

YS-05-27

YS-07-31

YS-07-53

18

6

8

9

0.66 ± 0.52 0.01 ± 0.01 8.08 ± 1.82 0.59 ± 0.41 34.73 ± 1.39 0.01 ± 0.01 37.39 ± 0.48 80.86 ± 1.12

0.23 ± 0.05 0.00 ± 0.01 8.89 ± 2.64 0.14 ± 0.17 36.31 ± 3.09 0.88 ± 0.53 38.35 ± 0.48 85.26 ± 0.95

0.06 ± 0.06 0.01 ± 0.02 8.54 ± 0.30 0.23 ± 0.16 2.84 ± 1.49 40.78 ± 2.23 30.95 ± 0.51 84.48 ± 1.45

0.79 2.97 0.05 2.00 16.87

0.93 0.29 2.63 1.99 16.94

oxides 0.62 ± 0.68a 0.01 ± 0.01 6.12 ± 2.60 0.13 ± 0.12 37.44 ± 1.55 0.02 ± 0.02 36.54 ± 0.76 81.75 ± 1.77

Structural formula units based on 11 oxygen K 0.56 0.76 Al 3.13 3.00 Fe 0.00 0.00 S 2.01 1.99 Unit total 16.77 16.89 a

Plus/minus are 1σ.

229

(KCl and CaF2) for interfering neutron reactions and monitors (sanidines from both Fish Canyon Tuff and Taylor Creek Rhyolite) of neutron fluence were loaded into aluminum trays, evacuated in a quartz tube, and irradiated in the CLICIT position at Oregon State University TRIGA reactor for 10 MWH. The alunite and monitor minerals were loaded into a high-vacuum extraction line and incrementally heated with a CO2 laser. The evolved gas was exposed to a cold finger cooled to about − 130 °C and an activated getter (SAES APGP50). The purified sample gas was expanded into a Noblesse multicollector mass spectrometer fitted with a faraday and two ion counting multipliers. The present sample was analyzed using the faraday detector for mass 40 and ion counting multipliers for all other masses. Aliqouts of air were used to determine mass discrimination on all detectors as well as the intercalibration factors between detectors. Results of the alunite analysis yield a weighted mean plateau age of 0.154 +/− 0.016 (2σ) Ma, and the analytical data are given in Table 2 and Fig. 4. Although this sample is extremely poor in radiogenic argon (less than 5%), the reproducibility of the blank signals on all detectors, and especially on the ion multipliers, is excellent, resulting in precise measurements at mass 36 and a reliable correction for non-radiogenic argon. Isochron data plots result in tightly clustered arrays due to the low radiogenic yields. Trapped 40Ar/36Ar ratios are indistinguishable from present-day atmosphere. 4. Hydrothermal mineralogy and alteration associations The alteration at Sevenmile Hole is pervasive and is exposed in outcrops on the northwest side of the Yellowstone River from the rim of the canyon to its base, a vertical interval of more than 300 m (Fig. 3). The associations of hydrothermal minerals vary vertically. Lateral variations appear to exist but these may be related to topography, and additional fieldwork is required to answer this question. The protolith for the alteration described here is the 480-ka tuff of Sulphur Creek (Prostka et al., 1975; Christiansen and Blank, 1975; Gansecki et al., 1996; Christiansen, 2001). The alteration extends north of the Caldera wall along Sulphur Creek (Fig. 2) where it locally affects the Eocene Absaroka Volcanics, but that area is not covered in this discussion. The tuff of Sulphur Creek is one of the units of the Upper Basin Member of the Plateau Rhyolites (Christiansen, 2001). It is a highsilica rhyolite, with SiO2 concentrations in the reported range of 76.3 to 76.59 wt.% (Hildreth et al., 1984; Christiansen, 2001). (Our preliminary unpublished data show a range of 75.84 to 77.90 wt.%.) In thin sections, anhedral-rounded quartz phenocrysts make up 5 to 15 vol.% of the rock (0.5 to 1 mm), subhedral to anhedral-rounded sanidine phenocrysts are 20 to 35% (1 to 7 mm), and subhedral sodic plagioclase phenocrysts are approximately 5%. Primary phenocrysts are often embayed and fractured in thin section. Oxides consist of titanomagnetite, magnetite, and illmenite in trace abundance. Zircon and apatite are also present in trace amounts. Phenocryst abundances and the degree of welding are variable throughout the tuff. The matrix, where unaltered, is a dense black to purple to medium grey glass. Spherules and lithophysae are widespread and are particularly prominent in the uppermost parts of the tuff. Where exposed north of Sulphur Creek, the basal poorly welded tuff exhibits well-preserved pumice and fiamme. Stratigraphically higher exposures show discontinuous banding/fiamme in the more densely welded sections. Hydrothermal alteration is pervasive and usually completely alters the tuff to a white or pale bluish-grey color. The quartz phenocrysts typically resist hydrothermal alteration and persist in the altered rock. The feldspars are usually completely altered to pseudomorphic replacements by clay and lesser quartz or silica where the original feldspar outlines are preserved and obvious in both hand samples and thin sections. However, some original feldspar phenocrysts are only partially replaced along cleavage planes, possibly due to the armoring effects of silicification around the grain that prevents continued reaction and exchange. In some samples, hydrothermal minerals flood

P.B. Larson et al. / Journal of Volcanology and Geothermal Research 188 (2009) 225–236 2.36 1.41 0.38 0.26 0.19 0.07 0.07 0.06 0.06 0.05 0.06 0.05 0.07 0.06 0.11 0.05 0.05 0.10 (Ma)

1.17 1.01 0.45 0.40 0.34 0.27 0.19 0.19 0.18 0.21 0.18 0.12 0.11 0.11 0.07 0.15 0.09 0.06 2.9 5.3 15.1 24 32.9 44 52.7 61 68.5 75.1 79.4 83.5 86.2 89.4 90.8 95.3 98.6 100

Ar released (mol)

5.9755E− 15 5.0925E− 15 2.038E− 14 1.8735E− 14 1.8423E− 14 2.3185E− 14 1.831E− 14 1.7351E− 14 1.5603E− 14 1.3803E− 14 8.9343E− 15 8.5492E− 15 5.7282E− 15 6.6245E− 15 2.9054E− 15 9.2803E− 15 7.0752E− 15 2.8425E− 15 1.6 1.7 1.8 1.8 1.9 2 2 2 2 2 2 2 2.1 2.1 2.1 2.2 2.2 2.2

2.1318E− 17 1.8237E− 17 2.1318E− 17 1.7155E− 17 1.7155E− 17 1.6655E− 17 1.6655E− 17 1.6655E− 17 1.6155E− 17 1.7155E− 17 1.6655E− 17 1.7155E− 17 1.6655E− 17 1.5739E− 17 1.6155E− 17 1.6655E− 17 1.6655E− 17 1.6655E− 17 (moles)

2.4737E− 13 1.2594E− 13 1.3628E− 13 8.4557E− 14 6.2366E− 14 2.7279E− 14 2.0884E− 14 1.6891E− 14 1.4589E− 14 1.1629E− 14 7.7747E− 15 5.6307E− 15 4.8604E− 15 4.2119E− 15 2.6787E− 15 5.548E− 15 3.7506E− 15 2.0483E− 15

(Watts)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Data are corrected for blanks, radioactive decay subsequent to irradiation, interfering isotopic reactions, detector intercalibrations, and mass discrimination.

(mol) (mol) (mol)

3.1323E− 18 2.2234E− 18 6.6676E− 18 6.2635E− 18 6.0615E− 18 8.0818E− 18 5.3544E− 18 5.4554E− 18 5.4554E− 18 4.7484E− 18 3.2333E− 18 3.2333E− 18 2.6273E− 18 3.1323E− 18 1.7186E−18 3.1323E− 18 2.8293E− 18 1.8196E− 18

2.3182E− 16 1.4354E− 16 3.3805E− 16 2.8562E− 16 2.6721E− 16 3.0522E− 16 2.4003E− 16 2.2419E− 16 2.0154E− 16 1.7687E− 16 1.1411E− 16 1.0761E− 16 7.3172E− 17 8.3009E− 17 3.7373E− 17 1.1737E− 16 8.9021E− 17 3.6152E− 17

4.7895E− 19 3.7924E− 19 5.2906E− 19 5.49E− 19 5.49E− 19 5.5909E− 19 5.2908E− 19 4.5904E− 19 4.291E− 19 4.2905E−19 3.5952E− 19 3.1006E− 19 2.8041E− 19 2.9027E− 19 2.317E− 19 3.3975E− 19 2.9033E− 19 2.1262E− 19

4.7856E− 17 4.1806E− 17 6.3106E− 17 6.3859E− 17 6.5648E− 17 6.5701E− 17 5.4758E− 17 5.018E− 17 4.9747E− 17 4.9314E− 17 3.9395E− 17 4.4592E− 17 3.6744E− 17 4.0608E− 17 3.4536E− 17 4.6413E− 17 3.9299E− 17 3.8269E− 17

2.9038E− 18 2.8114E− 18 3.1078E− 18 2.9068E− 18 2.9092E−18 3.0094E− 18 3.0082E− 18 3.0088E− 18 2.9142E− 18 2.9139E− 18 2.9141E− 18 2.9169E− 18 2.9165E− 18 3.0127E− 18 2.9143E− 18 2.9187E− 18 2.8219E− 18 3.1179E− 18

8.3215E− 16 4.2252E− 16 4.5448E− 16 2.8071E− 16 2.0641E− 16 8.7718E− 17 6.7984E− 17 5.4709E− 17 4.7265E− 17 3.7138E− 17 2.5096E− 17 1.8281E− 17 1.5954E− 17 1.3674E− 17 8.893E− 18 1.7687E− 17 1.2159E− 17 6.7946E− 18

8.318E− 19 5.9886E− 19 6.1664E− 19 5.3679E− 19 4.8391E− 19 3.2934E− 19 2.9636E− 19 2.7205E− 19 2.4837E− 19 2.2592E− 19 2.262E− 19 1.9058E− 19 1.8417E− 19 1.9788E− 19 1.8421E− 19 1.9801E− 19 1.8458E− 19 1.7245E− 19

39

+/− Ar 36

+/− Ar

37

+/− Ar

Sample Y-05-25, J = 0.0026501 +/− 1.32E− 06

Step

Table 2 Ar isotope measurements for alunite from sample Y-05-25.

39

Ar 40

Laser power

+/−

38

Ar

+/−

Weighted mean plateau age (all steps) = 0.156 +/− 0.016 Ma

Cumulative%

40

0.59 0.85 1.41 1.84 2.12 4.75 3.57 4.02 3.98 5.33 4.33 3.69 2.72 3.69 1.69 5.38 3.74 1.73

Ar⁎

Age

1 SD

230

the groundmass adjacent to feldspar phenocrysts and the original crystal outlines are not preserved. Where present, alunite can occur where phenocryst textures are destroyed, but more commonly it occurs as feldpsar phenocryst replacement and vug or cavity filling. The glassy matrix is usually altered to quartz or opal with clots and bands of clay minerals scattered throughout. Minor disseminated fine-grained pyrite is widespread and marcasite is common locally. Carbonate and zeolite minerals have not been found. Flattened pumice fragments often are replaced by coarser hydrothermal crystals than hydrothermal minerals in the groundmass; where the alteration is not pervasive the pumice fragments tend to preferentially alter before the rest of the groundmass, probably reflecting their higher permeability. Hairline quartz or opal veinlets are common, and multiple stages of silicification or quartz growth are apparent in many of the veinlets. Two broad hydrothermal alteration associations, quartz (opal) + kaolinite and quartz + illite, are vertically distributed in the upper canyon wall (Fig. 3). Both alteration associations also are exposed along the river near the canyon base (Fig. 3), but these occurrences appear to have formed after incision of the canyon began, and their distribution does not appear related to depth beneath the paleosurface represented by the canyon rim. Opal is also widespread in deeper exposures, but is found in crosscutting late veins and veinlets and is most likely the product of hydrothermal overprinting during incision. Therefore, although we note the presence of this alteration with interest, these deeper exposures are not considered further in this discussion. The quartz (opal) + kaolinite alteration affects rocks immediately below the paleosurface (canyon rim) and is found only in the upper 75 to 100 m of the canyon wall. Altered outcrops locally contain dickite and/or alunite with the kaolinite. Minor disseminated barite is locally associated with the alunite. Massive grey to very light yellow porcelainous opal replaces the altered tuff in places close to the surface. These masses are up to 15 cm in diameter and may be openspace filling. The porcelainous opal also occurs in veins, and 1 to 2 cm wide quartz (opal) veinlets are erratically distributed in the outcrops. Trace amounts to several percent disseminated fine-grained pyrite and/or marcasite are ubiquitous. Quartz + illite alteration is found deeper in the canyon, below about 100 m beneath the canyon rim, and is predominant on the ridge that peaks at 7741 foot elevation (this ridge is referred to as location 7741 in this text). Here, several prominent 1 to 2 meter wide ledges of pervasively silicified tuff extend for up to several hundred meters along the ridge. Small vugs lined with millimeter-scale acicular quartz crystals are common in the ledges. Larger open vugs are found throughout the quartz + illite altered area. These vugs are up to 3 cm in diameter, and contain inward projecting stubby euhedral quartz crystals up to 1.5 cm long. Alunite has been found in one area of the ridge where it replaces feldspar phenocrysts. SEM imaging also noted coarse tabular illite crystals and minor quartz + illite ± adularia in veinlets or primary tuff banding at location 7741. In addition, finegrained euhedral adularia partially replaces primary feldspars along the eastern ridge at location 7741. The complex of quartz ledges at location 7741 represent important paleochannels for upwelling hydrothermal fluids. We, therefore, suggest that this area served as a center of upward flow in the older hydrothermal system. At present, we do not know if the quartz + illite alteration is a broad lateral selvage on the quartz ledges, or if it represents a vertical change in the alteration mineral assemblage reflecting higher temperatures at depth. We prefer the latter interpretation at this time. 5. Age of the alteration The age of hydrothermal activity in the Sevenmile Hole area is bracketed by the age of the protolith, the tuff of Sulphur Creek, and the age of the incision of the Grand Canyon of the Yellowstone River.

P.B. Larson et al. / Journal of Volcanology and Geothermal Research 188 (2009) 225–236

Fig. 4.

231

40

Ar/39Ar age spectrum for sample Y-05-25 alunite. Analytical data are provided in Table 2.

Gansecki et al. (1996) report an 40Ar/39Ar age of about 480,000 years for the three units of the Upper Basin Rhyolites that include the tuff of Sulphur Creek. The timing of the canyon's incision appears to be older than the Pinedale glaciation. Richmond (1977) found lake sediments from the Bull Lake–Pinedale interglaciation and Pinedale lake sediments and till near the Yellowstone River level in Seven Mile Hole. Pierce (1974) mapped the Sediments of Seven Mile Hole and suggested that they were of several ages and origins, including deposits of Late Pleistocene glacial lake, ice contact, and landslide origin. Licciardi and Pierce (2008) found that the Pinedale glaciation, the youngest known in the Park, ended less than 20,000 years ago. Alunite, ideally KAl3(SO4)2(OH)6, is amenable to 40Ar/39Ar age dating. Electron microprobe analyses of alunite from several of the samples are shown in Table 1. Alunite from one sample, Y-05-25, was separated from the crushed rock and cleaned ultrasonically. XRD examination of the cleaned mineral separate showed no interfering peaks. The 40Ar/39Ar age of this alunite was measured in the laboratory at the Universite de Lausanne. Analytical results are shown in Table 2, and the Ar isotope age plateau is shown in Fig. 4. The weighted mean plateau age is 0.154 ± 0.016 (2σ) Ma. This age falls within the range bracketed by the ages of the tuff of Sulphur Creek and the upper limit of the presumed age of the canyon incision. It is interesting to compare this hydrothermal age with the age of the Bull Lake glaciation maximum that occurred in the range 135,000 to 155,000 years ago (Pierce et al., 1976; Licciardi and Pierce, 2008). The effect of deglaciation and the resultant loss of hydrostatic head have been implicated in the development of other hydrothermal features in the Yellowstone Caldera (Muffler et al., 1971; Bargar and Fournier, 1988; White et al., 1988).

them in the epithermal environment (Cooke and Simmons, 2000; Simmons et al., 2005). The hot springs are most likely the surface manifestations of a large convectively driven hydrothermal system that extends several kilometers deep (Fournier, 1989), although the active system has only been drilled to just over 300 m (White et al., 1975). Seismic studies show that magma currently resides at about 8 km depth beneath the Caldera (Husen et al., 2004). Epithermal systems in general have been divided into two subclasses that are produced by distinct fluid types, an alkali-chloride fluid and an acid-sulfate fluid (e.g., Simmons et al., 2005). The alkalichloride fluids have higher chloride concentrations and have near neutral pHs. The acid-sulfate fluids have lower total solute concentrations, higher sulfate contents, are more oxidized, and have low pHs. Each produces distinctive hydrothermal alteration assemblages. Both alkali-chloride and acid-sulfate-chloride fluids have been found in the active Yellowstone hot springs (White et al., 1971, 1988). For examples, the fluids discharging in the Upper Geyser Basin, where Old Faithful is located, are alkali-chloride fluids, and acid-sulfate fluids are predominant in the Mud Volcano area and in other thermal springs throughout the eastern part of the Caldera (White et al., 1971; Fournier, 1989). Alkali-chloride fluids typically produce quartz, adularia, illite, and calcite as alteration minerals and commonly precipitate silica sinter when they discharge at the surface. In contrast, alunite, kaolinite, dickite, and residual vuggy quartz are characteristic of the acid-sulfate-chloride alteration (Simmons et al., 2005). The alunite, dickite, and kaolinite in the older altered rocks along the upper edge of the canyon at Sevenmile Hole (Fig. 3) show that the hydrothermal fluids that produced them were acid-sulfate. The quartz and illite with minor secondary potassium feldspar at location 7741 suggest that this center was altered by alkali-chloride fluids.

6. Discussion 6.2. Temperature gradient 6.1. Hydrothermal alteration characteristics Formation in the near-surface environment and mineral assemblages in both the active and extinct Yellowstone hot springs place

The vertical variation of hydrothermal clays at Sevenmile Hole imply that increasing hydrothermal temperatures with depth (Fig. 3) may have been important in their formation. Kaolinite prevails at

232

P.B. Larson et al. / Journal of Volcanology and Geothermal Research 188 (2009) 225–236

shallower levels and illite becomes dominant below about 100 m beneath the current canyon rim. Research drilling by the Carnegie Institution in the 1930's and the U.S. Geological Survey in the late 1960's provides insight into the thermal profiles in the shallow portions of active Yellowstone hydrothermal systems. These data provide some useful constraints on the paleohydrothermal profile at Sevenmile Hole that can be compared to the vertical variation in phyllosilicate mineralogy. Fenner (1936) reports the in-hole temperatures measured during Carnegie's research drilling at Upper Geyser Basin and at Norris Geyser Basin, where total hole depths were 124 and 80 m, respectively. In-hole temperature measurements were also made during drilling of the thirteen U.S.G.S. research holes that were sited in several of the Yellowstone thermal areas (White et al., 1975). The deepest U.S.G.S drill hole, Y-12, penetrated to 332 m in Norris Geyser Basin. Measured temperatures for most of these holes increased with depth in a gradient that generally follows the hydrostatic boiling curve (Haas, 1971). Many of them lie on the curve or at lower temperatures but parallel to it. The hydrostatic boiling curve probably represents a minimum temperature for the hydrothermal fluids in the active upwelling shallow parts of the systems, because sealing of flow paths could lead to overpressure and higher temperatures. Drilling into these environments could vent the fluids to the surface and could establish true hydrostatic conditions and, therefore, the in-hole measured temperatures would be expected to fall close to the boiling curve. The temperatures that were measured in most of the research drill holes at a depth of 100 m lie within the range of about 150 to 170 °C, although the possibility of overpressuring implies that these are minimum temperatures. This analysis suggests that the transition from kaolinite to illite in the Sevenmile Hole area corresponds to a temperature in the range 150 to 170 °C. Logs of alteration minerals for several of the U.S.G.S. drill holes show that the kaolinite–illite transition also occurs at these conditions in the active Yellowstone systems. This is best demonstrated in drill hole Y-2 (total depth of 157 m) that was sited in Upper Geyser Basin (Bargar and Beeson, 1981). In Y-2, only weak sporadic kaolinite is reported above 90 m, but abundant illite appears only below about 95 m. The in-hole temperature at 95 m in Y-2 was measured at about 170 °C. A similar relationship is suggested in drill hole Y-6 (total depth of 152 m, drilled in the Upper Firehole River near Upper Geyser Basin) where kaolinite is abundant above 45 m depth and illite is abundant mostly below 125 m, but there is a large vertical gap of about 80 m between these zones where neither mineral is reported (Bargar and Beeson, 1984). Alunite was not found in either of these holes, and the thermal fluids encountered in the drilling were not acid-sulfate fluids. Regardless, it is interesting to note the correlation in the depth of the transition between these drill holes and the clay minerals in Sevenmile Hole. Most of the other holes for which logs have been published show that the alteration minerals tend to be dominated by zeolites, and kaolinite and illite are sparse or are absent altogether (e.g., drill hole Y-1, Upper Geyser Basin, Honda and Muffler, 1970). Studies of other active hydrothermal systems have shown that temperature varies with the vertical distribution of clays in the epithermal environment (see summary in Simmons et al., 2005). Henley and Ellis (1983) note that illite is present in active systems at temperatures above about 150 °C. Reyes (1990) reports that kaolinite is found at temperatures up to about 200 °C in active Philippine acid hydrothermal systems, and illite is present at temperatures above 230 °C in neutral pH Philippine systems. Reyes (1990) also notes that dickite occurs with kaolinite in the temperature range 120 to 200 °C in the acid systems. This corroborates the occurrence of dickite with kaolinite in Sevenmile Hole (Fig. 3). Thus, the distribution of kaolinite, dickite, and illite at Sevenmile Hole corresponds to their distribution in active geothermal systems in other areas of the world, although the temperature of the transition from shallower kaolinite

and dickite to deeper illite occurs at temperatures ranging from 150 to 230 °C. The transition from a shallower quartz–kaolinite to a deeper and hotter quartz–illite assemblage as observed at Sevenmile Hole is common to shallow systems and Yellowstone appears typical in this respect. A preliminary evaluation of mineral stability across this transition provides some insight about variations in fluid chemistry and other factors that may influence the change in mineralogy. The stability of Al-silicates at 225 °C in the system Al2O3 + K2O + SiO2 + H2O are shown in Fig. 5 (modified from Hofstra and Cline, 2000) as a function of the activity ratio of K+/H+ and the activity of SiO2 (aqueous). The diagram is constructed based on Al+++ conservation. Muscovite is typically used as a proxy for illite in activity space, and this convention is followed here. A shaded oval along the kaolinite–muscovite border above quartz saturation and below pyrophyllite and opal (amorphous silica) saturation most likely defines the environment along or near the contact between the two assemblages at Sevenmile Hole. Temperature variation between the two assemblages is important and evidence to support this is provided in the preceding discussion, but temperature alone most likely does not entirely account for the mineralogic change because muscovite/illite are K-bearing but kaolinite is not, and changes in fluid chemistry can also produce the mineral change. This is also illustrated in an equilibrium reaction between kaolinite (ka) and muscovite (mu) that is balanced on aluminum, where þ

þ

3Al2 Si2 O5 ðOHÞ4 ðkaÞ þ 2K ¼ 2KAl3 Si3 O10 ðOHÞ2 ðmuÞ þ 2H þ 3H2 O: ð1Þ Eq. (1) and Fig. 5 show that the quartz–illite assemblage may form from a higher K+ activity or a higher pH hydrothermal fluid than that which produced the quartz–kaolinite assemblage. Dilute cooler surface fluids mixing with a deeper and lower pH fluid is one possible process for changing the fluid chemistry where temperature and K+ activity would be lowered and kaolinite would be favored at shallower levels. Boiling in a rising hot fluid could also produce shallower low temperatures and changes in fluid chemistry such as increased pH that could account for the change. Both processes could be active in this environment. This analysis is preliminary and our research is proceeding with these testable multiple working hypotheses.

Fig. 5. The stability of Al-silicates in the system Al2O3 + K2O + SiO2 + H2O at 225 °C as a function of the activity ratio of K+/H+ and the activity of SiO2 (aqueous) (modified from Hofstra and Cline, 2000). The shaded oval encompasses the area of the contact between the quartz–kaolinite and quartz–illite assemblages at Sevenmile Hole, where muscovite is used as a proxy for illite.

P.B. Larson et al. / Journal of Volcanology and Geothermal Research 188 (2009) 225–236

233

Fig. 6. Ages for hydrothermal activity and for igneous rocks associated with the Yellowstone Caldera compiled from Sturchio et al. (1994), Gansecki et al. (1996), and Christiansen (2001). The star lies over the Sevenmile Hole alunite age reported here. Dashed lines for extracaldera magmatism are for basalts, solid lines are for rhyolites.

6.3. Duration of the Yellowstone hydrothermal system The hydrothermal system was active at Sevenmile Hole at least 154,000 years ago (Fig. 4) and is locally active today. These are the only two measures of the timing of the hydrothermal activity in this area, and we hesitate to place too much emphasis on only one age measurement. The hydrothermal system could have been active here in the interval between Caldera eruption and collapse at 640,000 years ago and the eruption the tuff of Sulphur Creek at 480,000 years ago. The tuff of Sulphur Creek and the other Upper Basin Rhyolites in the Sevenmile Hole area are low-18O magmas, and models for their formation include hydrothermally altered low-18O sources (Friedman et al., 1974; Hildreth et al., 1984, 1991; Bindeman and Valley, 2000, 2001; Bindeman et al., 2001). Hydrothermal systems must have been active and affecting the rhyolite sources prior to 480 ka (Bindeman et al., 2001), and could have been active and altering the rocks in this area before eruption of the Yellowstone Caldera. However, we do not know if hydrothermal activity at Sevenmile Hole has been intermittent or continuous, nor can we constrain when hydrothermal activity initiated in this area. Christiansen (2001) compiled isotopic ages for the Yellowstone pre- and post-collapse basalts and rhyolites, and Sturchio et al. (1994) measured the uranium-series ages of hydrothermal travertine deposits in the Mammoth area on the north end of the Mammoth–Norris corridor. All these ages are compiled on Fig. 6, together with our age determination for the Sevenmile Hole alunite (Fig. 4). The oldest ages

Fig. 7. The linear arrays of hot spring centers in the Sevenmile Hole area are enclosed in the box with the cross hatched pattern. Solid circles are active or extinct sinters (see text for discussion of numbers 1, 2, and 3). Open circles are eroded centers. Note the alignment of hydrothermal centers subparallel to the Caldera rim, which here is the topographic wall of the Caldera.

Fig. 8. Schematic model showing the deflection of centers of hot spring activity due to incision of the Grand Canyon of Yellowstone River. Initial hot spring centers (A) are deflected canyonward (B) as incision progresses and the water table is lowered, thus promoting shallow ground water flow toward the canyon. The ultimate effect is to focus hot spring activity near the canyon bottom (C). A deep magma (not placed to depth scale) is shown as a heat source that drives shallower convection of ambient groundwater and also contributes heat and some fluids to the hydrothermal system.

234

P.B. Larson et al. / Journal of Volcanology and Geothermal Research 188 (2009) 225–236

from the travertine are about 400,000 years (from the Terrace Mountain deposits, Sturchio et al., 1994), but most are less than 100,000 years. The periods of hydrothermal discharge as measured at Mammoth and Sevenmile Hole do not correlate precisely with periods of either the intracaldera or extracaldera magmatism. Although these data show that hydrothermal activity was active early in the history of postcollapse magmatism, neither continuity nor cyclicity of the hydrothermal system can be defined using the limited number of measured hydrothermal ages that are shown on Fig. 6. It is also possible that some older intracaldera postcollapse lavas are concealed by younger flows, and this may account for the age gap between 480,000 and 200,000 years shown on Fig. 6. An ongoing objective of our research is to more precisely define the ages of the older alteration. 6.4. Deeper structural control of hydrothermal vents The centers of hydrothermal activity in the Sevenmile Hole area lie along a linear trend (Fig. 7) that probably is the surface manifestation of buried or concealed structures that channel upflowing fluids. The centers include extinct post-glacial deposits of siliceous sinter that appear to have been abandoned only recently because they are not obviously eroded. One occurs west of the canyon edge on the plateau (location 1 in Fig. 7). This sinter mound is brecciated and large angular fragments up to several meters in diameter are scattered across the landscape. This location was once the site of an active alkali-chloride hot spring cluster, but steam vents with sulfur encrustations are now the only active thermal features here. Another abandoned coneshaped sinter deposit (location 2 in Fig. 7) lies adjacent to the trail that descends to the bottom of Sevenmile Hole. Hot springs are active and depositing sinter encrustations further down the trail (location 3). The centers at 1 and 2 on Fig. 7 were possibly abandoned during canyon incision. A model whereby canyon downcutting lowers the water table and deflects shallower hydrothermal flow canyonward can account for this change (Fig. 8). The age and rate of canyon incision would control the time of abandonment of the sinters.

Three extinct and eroded hydrothermal centers also lie along the linear trend (Figs. 3, 7). These include the center at location 7741, the 153-ka center defined by the alunite and dickite occurrences near the canyon rim, and a center located along the river. The six centers lie along a linear trend that has been a locus of hydrothermal activity for at least about 150,000 years. The linearity and longevity of the hydrothermal activity at Sevenmile Hole suggest that there is a deeper structure that controls and focuses the fluid upflow. This deeper structure is the 640,000 year-old Caldera ring fault that is buried beneath the 480,000 year-old tuff of Sulphur Creek. The Sevenmile Hole altered area lies at the eastern end of a band of hot springs in the Yellowstone Caldera (Christiansen, 2001). As noted by White et al. (1988) and Christiansen (2001), this band may trace the path of the Caldera ring fault beneath the post collapse rhyolites (Fig. 9). Larson and Taylor (1986a,b) have shown that hydrothermal fluid flow in the 23-Ma Lake City Caldera, San Juan volcanic field, Colorado, was focused along the Caldera ring faults, and this model also appears to be valid for Yellowstone Caldera. 7. Summary and conclusions Older hydrothermal alteration is well exposed in the Sevenmile Hole area of the Grand Canyon of the Yellowstone River, Yellowstone Park, Wyoming. Major findings and conclusions of our preliminary research on this altered area are summarized here. (1) The hydrothermal minerals at Sevenmile Hole include a shallower quartz (opal) +kaolinite ± alunite ±dickite assemblage and a deeper quartz (pervasive to vuggy) + illite assemblage. (2) The vertical change in alteration assemblages at Sevenmile Hole is similar to that found in active geothermal systems elsewhere. The change in clay mineralogy from shallower kaolinite to deeper illite is also found in active alkali-chloride hydrothermal centers elsewhere in Yellowstone, where it corresponds to a temperature in the range 150 to 170 °C. The

Fig. 9. The geologic map of the Yellowstone Caldera modified from Fig. 1 that highlights the broad band of hydrothermal centers that may be the surface manifestation of a buried Caldera ring fault (the broad grey band). See Fig. 1 for the map sources.

P.B. Larson et al. / Journal of Volcanology and Geothermal Research 188 (2009) 225–236

change in mineralogy is most likely related to increased K+ activity or higher pH deeper in the hydrothermal system. (3) The age of the hydrothermal activity at Sevenmile Hole extends back to about 150,000 years. Yellowstone hydrothermal activity at Mammoth Basin has been ongoing since at least 400,000 years ago, but it is not known if the activity at Mammoth or elsewhere is episodic, or if it has maintained a relatively continuous intensity. (4) The alignment of both extinct and active hydrothermal centers in the Sevenmile Hole area suggests that there is a deep structural control that focuses upwelling hydrothermal fluids. This deep structure is most likely the Yellowstone Caldera ring fault that is concealed by the younger tuff of Sulphur Creek. The Sevenmile Hole area lies at the eastern end of a zone of hydrothermal features that may trace the concealed Caldera ring fault along the northern Caldera margin. Acknowledgements We are truly appreciative of the help and suggestions provided by the science and permitting staffs of the Park Service at Yellowstone Park. We especially thank Christie Hendrix, Christine Smith, Hank Heasler, and Cheryl Jaworowski. Discussions with Todd Feeley and Dave Cole, both of whom have also provided invaluable field assistance, have been enlightening and useful. Jake Lowenstern, Alan Wallace, John Dilles, and an anonymous reviewer provided excellent and useful comments on earlier versions of the manuscript. This research has been supported by NSF grant EAR-0609475. References Allen, E.T., Day, A.L., 1935. Hot springs of the Yellowstone National Park, vol. 466. Carnegie Institute of Washington Publication. 525 p. Bargar, K.E., Beeson, M.H., 1981. Hydrothermal alteration in research drill hole Y-2, Lower Geyser Basin, Yellowstone National Park, Wyoming. American Mineralogist 66, 473–490. Bargar, K.E., Beeson, M.H., 1984. Hydrothermal alteration in research drill hole Y-6, Upper Firehole River, Yellowstone National Park, Wyoming. U.S. Geological Survey Professional Paper 1054-B. 24 p. Bargar, K.E., Fournier, R.O., 1988. Effects of glacial ice on subsurface temperatures of hydrothermal systems in Yellowstone National Park, Wyoming: fluid-inclusion evidence. Geology 16, 1077–1080. Bindeman, I.N., Valley, J.W., 2000. Formation of low-δ18O rhyolites after caldera collapse at Yellowstone, Wyoming, USA. Geology 28, 719–722. Bindeman, I.N., Valley, J.W., 2001. Low-18O rhyolites from Yellowstone: Magmatic evolution based on analyses of zircons and individual phenocrysts. Journal of Petrology 42, 1491–1517. Bindeman, I.N., Valley, J.W., Wooden, J.L., Persing, H.M., 2001. Post-caldera volcanism: in situ measurement of U–Pb age and oxygen isotope ratio in Pleistocene zircons from Yellowstone caldera. Earth and Planetary Science Letters 189, 197–206. Camp, V.E., 1995. Mid-Miocene propagation of the Yellowstone mantle plume head beneath the Columbia River Basalt source region. Geology 23, 435–438. Christiansen, R.L., 1975. Geologic map of the Norris Junction Quadrangle, Yellowstone National Park, Wyoming. U.S. Geological Survey Map GQ-1193. Christiansen, R.L., 2001. The Quaternary and Pliocene Yellowstone plateau volcanic field of Wyoming, Idaho, and Montana. U.S. Geological Survey Professional Paper 729-G. 145 p. Christiansen, R.L., Blank Jr., H.R., 1975. Geologic map of the Canyon Village Quadrangle, Yellowstone National Park, Wyoming. U.S. Geological Survey Map GQ-1192. Christiansen, R.L., Foulger, G.R., Evans, J.R., 2002. Upper-mantle origin of the Yellowstone hotspot. Geological Society of America Bulletin 114, 1245–1256. Cooke, D.R., Simmons, S.F., 2000. Characteristics and genesis of epithermal gold deposits. Economic Geology Reviews in Economic Geology 13, 221–244. Feeley, T.C., 2003. Origin and tectonic implications of across-strike geochemical variations in the Eocene Absaroka volcanic province, United States. Journal of Geology 111, 329–346. Feeley, T.C., Cosca, M.A., Lindsay, C.R., 2002. Petrogenesis and implications of cryptic hybrid magmas from Washburn volcano, Yellowstone National Park, USA. Journal of Petrology 43, 663–703. Fenner, C.N., 1936. Bore-hole investigations in Yellowstone Park. Journal of Geology 44, 225–315. Fournier, R.O., 1989. Geochemistry and dynamics of the Yellowstone National Park hydrothermal system. Annual Review of Earth and Planetary Sciences 17, 13–53. Friedman, I., Lipman, P.W., Obradovich, J.D., Gleason, J.D., Crhistiansen, R.L., 1974. Meteoric water in magmas. Science 184, 1069–1072. Gansecki, C.A., Mahood, G.A., McWilliams, M.O., 1996. 40Ar/39Ar geochronology of rhyolites erupted following collapse of the Yellowstone caldera, Yellowstone Plateau volcanic field: implications for crustal contamination. Earth and Planetary Science Letters 142, 91–107.

235

Goff, F., Gardner, J.N., 1994. Evolution of a mineralized geothermal system, Valles Caldera, New Mexico. Economic Geology 89, 1803–1832. Haas, J.L., 1971. The effect of salinity on the maximum thermal gradient of a hydrothermal system at hydrostatic pressure. Economic Geology 66, 940–946. Hayba, D.O., Bethke, P.M., Heald, P., Foley, N.K., 1985. Geologic, mineralogic and geochemical characteristics of volcanic-hosted epithermal precious metal deposits. Reviews in Economic Geology 2, 129–167. Heald, P., Hayba, D.O., Foley, N.K., 1987. Comparative anatomy of volcanic-hosted epithermal deposits: acid-sulfate and adularia-sericite types. Economic Geology 82, 1–26. Henley, R.W., Ellis, A.J., 1983. Geothermal systems ancient and modern: a geochemical review. Earth Science Reviews 19, 1–50. Hildreth, W., Christiansen, R.L., O'Neil, J.R., 1984. Oxygen isotopic study of the Yellowstone Plateau volcanic field. Journal of Geophysical Research 89, 8339–8369. Hildreth, W., Halliday, A.N., Christiansen, R.L., 1991. Isotopic and chemical evidence concerning the genesis and contamination of basaltic and rhyolitic magma beneath the Yellowstone Plateau volcanic field. Journal of Petrology 32, 63–138. Hinman, N.W., Walter, W.R., 2005. Textural preservation in siliceous hot spring deposits during early diagenesis: examples from Yellowstone National Park and Nevada, U.S.A. Journal of Sedimentary Research 75, 200–215. Hofstra, A.H., Cline, J.S., 2000. Characteristics and models for Carlin-type gold deposits. Society of Economic Geologists Reviews in Economic Geology 13, 163–220. Honda, S., Muffler, L.J.P., 1970. Hydrothermal alteration in core from research drill hole Y-1, Upper Geyser Basin, Yellowstone National Park, Wyoming. American Mineralogist 55, 1714–1737. Husen, S., Smith, R.B., Waite, G.P., 2004. Evidence for gas and magmatic sources beneath the Yellowstone volcanic field from seismic tomographic imaging. Journal of Volcanology and Geothermal Research 131, 397–410. Kharaka, Y.K., Sorey, M.L., Thordsen, J.J., 2000. Large-scale hydrothermal fluid discharges in the Norris–Mammoth corridor, Yellowstone National Park, USA. Journal of Geochemical Exploration 69, 201–205. Larson, P.B., Taylor Jr., H.P., 1986a. An oxygen isotope study of hydrothermal alteration in the Lake City Caldera, San Juan Mountains, Colorado. Journal of Volcanology and Geothermal Research 30, 47–82. Larson, P.B., Taylor Jr., H.P., 1986b. An oxygen study of water/rock interaction in the granite of Cataract Gulch, western San Juan Mountains, Colorado. Geological Society of America Bulletin 97, 505–515. Larson, P.B., Taylor Jr., H.P., 1987. Solfataric alteration in the San Juan Mountains, Colorado: oxygen isotope variations in a boiling hydrothermal environment. Economic Geology 82, 1019–1036. Licciardi, J.M., Pierce, K.L., 2008. Cosmogenic exposure-age chronologies of Pinedale and Bull Lake glaciations in greater Yellowstone and the Teton Range, USA. Quaternary Science Reviews 27, 814–831. Morgan, L.A., Shanks, W.C.P., Lee, G.K., Webring, M.W., 2007. Bathymetry and geology of the floor of Yellowstone Lake, Yellowstone National Park, Wyoming, Idaho, and Montana. U.S. Geological Survey Scientific Investigations Map 2973. Muffler, L.J.P., White, D.E., Truesdell, A.H., 1971. Hydrothermal explosion craters in Yellowstone National Park. Geological Society of America Bulletin 82, 723–740. Pierce, 1974. Surficial geologic map of the Tower Junction quadrangle and part of the Mount Wallace quadrangle, Yellowstone national Park, Wyoming and Montana. U.S. Geological Survey Miscellaneous Geologic Investigations Map I-641. Pierce, K.L., Morgan, L.A., 1992. The track of the Yellowstone hot spot; volcanism, faulting, and uplift. In. Link, P.K., Kuntz, M.A., and Platt, L.B., (editors), Regional geology of eastern Idaho and western Wyoming, Geological Society of America, Memoir 179, pp. 1–53. Pierce, K.L., Obradovich, J.D., Friedman, I., 1976. Obsidian hydration dating and correlation of Bull Lake and Pinedale glaciations near West Yellowstone, Montana. Geological Society of America Bulletin 87, 703–710. Pierce, K.L., Adams, K.D., Sturchio, N.C., 1991. Geologic setting of the Corwin Springs Known Geothermal Resources Area in and adjacent to Yellowstone National Park. In: Sorey, M.S. (Ed.), U.S. Geological Survey Water Resources Investigations Report 91-4052, pp. C1–C37. Prostka, H.J., Blank Jr., H.R., Christiansen, R.L., Ruppel, E.T., 1975. Geologic map of the Tower Junction Quadrangle, Yellowstone National Park, Wyoming and Montana. U. S. Geological Survey Map GQ-1247. Reyes, A.G., 1990. Petrology of Philippine geothermal systems and the application of alteration mineralogy to their assessment. Journal of Volcanology and Geothermal Research 43, 279–309. Richmond, G.M., 1976. Surficial geologic history of the Canyon Village Quadrangle, Yellowstone National Park, Wyoming, for use with Map I-652. U.S. Geological Survey Bulletin 1427 35 p. Richmond, G.M., 1977. Surficial geologic map of the Canyon Village Quadrangle, Yellowstone National Park, Wyoming. U.S. Geological Survey Map I-652. Shervais, J.W., Hanan, B.B., 2008. Lithospheric topography, tilted plumes, and the track of the Snake River–Yellowstone hot spot. Tectonics 27, TC5004. Simmons, S.F., White, N.C., John, D.A., 2005. Geological characteristics of epithermal precious and base metal deposits. Economic Geology 100th Anniversary Volume, pp. 485–522. Smith, R.B., Braile, L.W., 1994. The Yellowstone hotspot. In: Hill, D.P., Gasparini, P., McNutt, S., Rymer, H. (Eds.), Internal Structure of Volcanoes and Geophysical Precursors of Eruptions. Elsevier, Amsterdam, pp. 121–187. Stoffregen, R.E., 1987. Genesis of acid-sulfate alteration and Au–Cu–Ag mineralization at Summitville, Colorado. Economic Geology 82, 1575–1591. Sturchio, N.C., Pierce, K.L., Murrell, M.T., Sorey, M.L., 1994. Uranium-series ages of travertines and timing of the last glaciation in the northern Yellowstone area, Wyoming–Montana. Quaternary Research 41, 265–277.

236

P.B. Larson et al. / Journal of Volcanology and Geothermal Research 188 (2009) 225–236

Thompson, A.J.B., Hauff, P.L., Robitaille, A.J., 1999. Alteration mapping in exploration: application of short-wave infrared (SWIR) spectroscopy. Society of Economic Geology Newsletter 39 13 p. White, N.C., Hedenquist, J.W., 1990. Epithermal environments and styles of mineralization: variations in their causes, and guidelines for exploration. Journal of Geochemical Exploration 36, 445–474. White, D.E., Muffler, L.J.P., Truesdell, A.H., 1971. Vapor-dominated hydrothermal systems compared with hot-water systems. Economic Geology 66, 75–97.

White, D.E., Fournier, R.O., Muffler, L.J.P., Truesdell, A.H., 1975. Physical results of research drilling in thermal areas of Yellowstone National Park, Wyoming. U.S. Geological Survey Professional Paper 892. 70 p. White, D.E., Hutchinson, R.A., Keith, T.E.C., 1988. The geology and remarkable thermal activity of Norris Geyser Basin, Yellowstone National Park, Wyoming. U.S. Geological Survey Professional Paper 1456. 84 p.