40Ar/39Ar Geochronology of the Eocene Green River Formation, Wyoming

40Ar/39Ar geochronology of the Eocene Green River Formation, Wyoming

M. Elliot Smith Brad Singer² Alan Carroll Department of Geology and Geophysics, University of Wisconsin, 1215 West Dayton Street, Madison, Wisconsin 53706, USA

ABSTRACT evaporative Wilkins Peak phase than the es. Much of this work has focused on past freshwater to saline Tipton (88 ؎ 34 ␮m/ climates because of the sensitivity of lakes to The deposits of Eocene Lake Gosiute that yr) and Laney (104 ؎ 18 ␮m/yr) phases. climate change and because of the potential constitute the Green River Formation of The much lower accumulation rates for the for their deposits to preserve terrestrial proxy Wyoming contain numerous tuff beds that Tipton and Laney Members are permissive records of paleoclimate. Climatic forcing of represent isochronous, correlatable strati- of an annual origin for Ͻ1-mm-thick lam- sedimentation has been inferred to operate on graphic markers. Tuff beds selected for inae and precessional forcing of 1±3-m- a variety of time scales, ranging from annual 40Ar/39Ar analysis occur within laminated thick depositional cycles in these units. to orbital (e.g., Bradley, 1929; Van Houten, mudstone, are matrix supported, and lack However, previously described cycles in the 1962; Olsen, 1986; Fischer and Roberts, evidence of reworking. These tuffs contain Wilkins Peak Member have average dura- 1991). Many of these studies link sedimentary 2%±15% euhedral phenocrysts of quartz, tions that are signi®cantly shorter than the facies successions to interpreted variations in plagioclase, sanidine, biotite, and minor 19±23 k.y. precessional modes. climatic humidity, on the basis of the expec- amphibole, pyroxene, and zircon, encased The Green River Formation encompasses tation that wetter conditions should produce in a matrix of altered glassy ash. Air abra- an ϳ5 m.y. period between ca. 53.5 and larger lakes. The translation of sediment- sion and handpicking under refractive- 48.5 Ma, spanning magnetic chrons 24n thickness patterns into time therefore repre- index oils were required to obtain clean, through 21r. The Green River Formation sents a crucial step in elucidating past rates of unaltered phenocrysts of sanidine. 40Ar/39Ar was therefore deposited during the warmest climate change. age determinations from single-crystal and period of the Cenozoic corresponding to the Owing to their common occurrence within Ͻ1 mg multigrain aliquots of sanidine and early Eocene climatic optimum as de®ned orogenic plateaus, tectonically formed lake biotite allowed the identi®cation and exclu- by the global marine O isotope record. De- basins are also well suited to deciphering re- sion of xenocrystic contamination. Laser- position of bedded evaporites (trona) of the gional interactions between climate and rising fusion experiments on phenocrysts from the Wilkins Peak Member began at ca. 51 Ma, orogenic topography. Numerous studies have Rife, Firehole, C Bed, Grey, Main, Sixth, or ϳ2 m.y. after the onset of the highest shown that lacustrine sedimentation represents inferred temperatures of the climatic opti- and Analcite tuff beds from the Tipton, a complex geomorphic response to regional mum. The Bridgerian±Wasatchian faunal Wilkins Peak, and Laney Members yielded tectonic, paleoclimatic, and, in some cases, weighted-mean ages (؎2␴ analytical uncer- turnover occurred subsequently during magmatic in¯uences (e.g., Surdam and Stan- ,tainties) of 51.25 ؎ 0.31 Ma, 50.70 ؎ 0.14 Wilkins Peak time, at ca. 50.6 Ma. Thus ley, 1980; Sullivan, 1985; Carroll and Bohacs, Ma, 50.56 ؎ 0.26 Ma, 50.39 ؎ 0.13 Ma, our 40Ar/39Ar ages strongly suggest that 1999). Understanding the rates and timing of Ma, 49.70 ؎ 0.10 Ma, and Wilkins Peak evaporite deposition and the 0.08 ؎ 49.96 sediment accumulation in such basins is es- Ma, respectively. Ages for san- turnover from Wasatchian to Bridgerian 0.12 ؎ 48.94 sential to deciphering the tectonic vs. climatic idine and biotite from the Main tuff are in- fauna were not directly caused by the ini- in¯uences on lake evolution. In addition, la- distinguishable when presumably xenocrys- tiation of maximum greenhouse conditions. tic contaminants are excluded from the age custrine and associated alluvial strata also calculation. Moreover, the 40Ar/39Ar ages Keywords: Green River Formation, sedi- contain some the most complete archives of are consistent with the stratigraphic order mentation rates, lacustrine, 40Ar/39Ar, the rates of biological evolution on the con- of the tuff beds and with provenance in the Wasatchian-Bridgerian, Early Eocene Cli- tinents (e.g., Krishtalka et al., 1987; Lucas, Absaroka and Challis volcanic ®elds. mate Optimum. 1998; Alroy et al., 2000). Our 40Ar/39Ar-based age model indicates The Green River Formation of Wyoming, that sediment accumulated three times INTRODUCTION Utah, and Colorado (Fig. 1) is one of the best- more rapidly (327 ؎ 86 ␮m/yr) during the known and most diverse lacustrine systems in Lake deposits are commonly exploited for the world (e.g., Bradley, 1929; Eugster and ²Corresponding author e-mail: bsinger@geology the long-lived and relatively continuous re- Hardie, 1975; Roehler, 1993). As such, it rep- .wisc.edu. cords they provide of Earth's surface process- resents an ideal body of lacustrine strata in

GSA Bulletin; May 2003; v. 115; no. 5; p. 549±565; 9 ®gures; 3 tables; Data Repository item 2003064.

For permission to copy, contact [email protected] ᭧ 2003 Geological Society of America 549 SMITH et al.

(Rife Bed±Tipton), to hypersaline playa-lake (Wilkins Peak), then back to saline (LaClede Bed±Laney) and eventually freshwater lake (Hart Cabin±Laney) (Fig. 2; Roehler, 1993; Carroll and Bohacs, 1999; Rhodes et al., 2002). The Green River Formation and time- equivalent alluvial strata in the Greater Green River Basin contain one of the most complete faunal records of the late Wasatchian through early Bridgerian NALMA (North American land mammal age) period (e.g., Krishtalka et al., 1987; Zonneveld et al., 2000). This transition is marked by the ®rst appearances of several primate and perissodactyl (e.g., horses and rhinoceroses) taxa, high North American mammalian diversity, and greenhouse faunas and contains 1 of 18 major turnovers in the North American Cenozoic mammalian record (Stucky, 1990; Alroy, 1999; Zonneveld et al., 2000). NALMA biostratigraphy has been for many years the principal tool for inter- and intrabasinal correlation in western North America (e.g., Wood et al., 1941; Krishtalka et al., 1987; Lucas, 1998) and can be roughly correlated with Eocene faunas in Europe and Asia (Lucas, 1998). Consequently, improve- ments in geochronology have the potential to set limits on the rates and paleogeographic patterns of Cenozoic faunal radiation and ex- Figure 1. Location of Eocene lacustrine basins, Precambrian±cored uplifts, and possible tinction in both a regional and global context. volcanic source regions in the northern Rocky Mountains. Modi®ed from Love and Chris- Until now, the timing of the Wasatchian± tiansen (1985), Roehler (1992b), and Snider and Moye (1989). Modeled 500 mbar winds Bridgerian period has been poorly determined are from Sewall et al. (2000). owing to limited and imprecise radioisotopic age control (Fig. 3; Evernden et al., 1964; which to develop a high-resolution geochro- water through hypersaline lake environments Mauger, 1977; O'Neill, 1980). This study pro- nology. We have obtained laser-fusion 40Ar/ (Bradley, 1964; Eugster and Hardie, 1975; vides the ®rst laser-fusion 40Ar/39Ar radioiso- 39Ar ages from seven tuff beds in the Green Smoot, 1983; Surdam and Stanley, 1979; topic ages for this period and allows for ac- River Formation. These ages set limits on Roehler, 1993). Carroll and Bohacs (1999) sub- curate temporal comparison of lake type and rates of sediment accumulation, periodicities divided these deposits into ¯uvio-lacustrine, faunal changes with the global marine record of micro- to mesoscale depositional cyclicity, ¯uctuating profundal, and evaporative facies of the early Eocene. the timing of faunal changes, and the mag- associations on the basis of interpretation of netostratigraphy of the Green River Forma- sedimentary facies, stratigraphic packaging, PREVIOUS GEOCHRONOLOGIC tion. Moreover, by tying the Green River For- fossil occurrence, and organic geochemistry STUDIES mation to the geomagnetic polarity time scale (see also Bohacs et al., 2000; Carroll and Bo- (GPTS; Cande and Kent, 1992, 1995; Berg- hacs, 2001). The Green River Formation in- It has been recognized at least since 1970 gren et al., 1995), the new 40Ar/39Ar ages alter®ngers with time-equivalent alluvial facies that the many silicic tephra layers within the low for correlation to other Laramide basins, of the Wasatch Formation. In the Greater Green River Formation might facilitate a pre- to Eocene volcanism in western North Amer- Green River Basin, the Green River Formation cise radioisotopic chronology for the accu- ica, and to the global marine O isotope record. is bounded at its base by the main body of the mulation of the sedimentary deposits. The pi- Wasatch Formation and is overlain by the oneering work of Mauger (1977) undertook THE GREEN RIVER FORMATION Bridger and Washakie Formations (Fig. 2; K-Ar dating of biotite separated from tuff beds Roehler, 1992a, 1992b; McCarroll et al., 1996; in the Greater Green River and Uinta Basins. The Green River Formation consists of Eo- Evanoff et al., 1998). Lacustrine facies are di- The K-Ar ages discussed here have been con- cene lacustrine strata deposited in the Greater vided into four major members: the Luman verted by using the decay constants of Steiger Green River Basin (ϭ Green River, Great Di- Tongue and Tipton, Wilkins Peak, and Laney and JaÈger (1977) following Dalrymple (1979). vide, Washakie, and Sand Wash Basins), Fos- Members (Fig. 2; Roehler, 1992b). These units Mauger's (1977) K-Ar ages between 50.9 Ϯ sil, Uinta, and Piceance Creek Basins (Fig. 1). document the geographic and chemical evo- 2.3 Ma and 46.4 Ϯ 1.9 Ma provided the ®rst The formation comprises a broad range of sed- lution of Lake Gosiute from freshwater (Lu- radioisotopic constraints on the timing of the imentary facies that were deposited in fresh man, Scheggs Bed±Tipton), to saline lake upper Green River Formation in the Greater

550 Geological Society of America Bulletin, May 2003 40Ar/39Ar GEOCHRONOLOGY OF THE EOCENE GREEN RIVER FORMATION, WYOMING

Figure 2. Generalized cross section of Eocene strata in the Greater Green River Basin along line A±A؅ in Figure 1, showing the stratigraphic position of the Green River Formation and dated tuff beds. Modi®ed from Culbertson (1961) and Roehler (1992b).

TABLE 1. CHARACTERISTICS OF TUFF BEDS

Tuff Appearance Setting² Thickness Texture % Matrix Phenocrysts Alteration Analcite White to orange with 1 lcm 15 cm Matrix supported; phenocrysts Ͼ95 Sanidine, biotite, plagioclase, Analcite cm-diameter blue concentrated at base quartz; minor amphibole, analcime crystals at zircon base Sixth White with hematite lcm 17 cm Matrix supported; phenocrysts 90±95 Plagioclase, biotite, quartz; Analcite staining at base concentrated at base minor sanidine, pyroxene, zircon Main WhiteÐlowest of three lcm 25 cm Matrix supported; possesses 90±95 Sanidine, biotite, plagioclase, Analcite correlatable 5±10 graded 1±20 mm ``fallout'' quartz; minor amphibole, cm subbeds beds pyroxene, zircon Grey Mottled gray to white sdm 4±6 cm Matrix supported 85±90 Plagioclase, biotite, sanidine, K-feldspar quartz and analcite C Bed Gray to white lcm 8 cm Matrix supported 90±95 Sanidine, biotite, plagioclase, K-feldspar quartz; minor amphibole, and analcite pyroxene, zircon Firehole Orange to yellow lcm 15 cm Matrix supported Ͼ95 Plagioclase, quartz; minor K-feldspar sanidine and analcite Rife White to gray lcm 2 cm Matrix supported; phenocrysts 80±85 Biotite, plagioclase, quartz Analcite concentrated at base ²Abbreviations: lcmÐlaminated calcareous mudstone, sdmÐsilty dolomitic mudstone.

Green River and Uinta Basins (Fig. 3). How- The only previous 40Ar/39Ar investigation of ments also utilized large bulk samples of ever, the need for large bulk samples (ϳ400 Green River Formation tuff beds was by ϳ400 mg. Individual plateau ages for the mg) severely limited the resolution of this O'Neill (1980), who performed incremental- Sand Butte, Sixth, Main, and Wavy tuffs study because of the inability to resolve po- heating experiments on biotite from tuff beds ranged from 50.1 Ϯ 2.2 Ma to 45.5 Ϯ 2.2 Ma, tential alteration effects or the incorporation of in the Greater Green River and Uinta Basins. with mean ages from replicate experiments of detrital or xenocrystic components. O'Neill's (1980) ®ve- to eight-step experi- 49.8 Ϯ 0.8 Ma and 46.6 Ϯ 1.0 Ma for the

Geological Society of America Bulletin, May 2003 551 SMITH et al. pic ages, malized to the thero and Swisher eds from Wood et es recalibrated from analytical and intercalibration uncertainties. ␴ Figure 3. Compositebiostratigraphy, age and model magnetostratigraphy for toal. (1941), the the Morris Eocene Green (1954), geomagnetic McGrew(1992), River polarity and Clyde Formation Roehler time et (1960), scale. and al. GazinCande Position (1994, associated (1965), and 1997, of West strata Kent 2001), and NALMA (1992, Dawson Prothero inintercalibration ages, 1995) (1973), (1996a), values the zones, Flynn to McCarroll of subages, (1986), Greater the et Renne Krishtalka and intercalibration al. Green et et ``Bridger'' (1996), values al. al. b River Zonneveld of (1987), (1998) Basin, Honey et Renne and (1988), al. et showing shown Pro (2000), al. with correlation and (1998) 2 Anemone of (see (2001). Fig. radioisoto Magnetic 8 reversal and ag Table DR3 [see text footnote 1 in the text]). All ages nor

552 Geological Society of America Bulletin, May 2003 40Ar/39Ar GEOCHRONOLOGY OF THE EOCENE GREEN RIVER FORMATION, WYOMING

Main and Wavy tuffs, respectively (Fig. 3). The neutron ¯uence monitor mineral used in these experiments was an 811 Ma biotite (OSU DY-8c±71) provisionally calibrated to a 519.5 Ma age for the MMhb standard (Snee and Sutter, 1982), making it dif®cult to com- pare O'Neill's (1980) results with data obtained by using modern intercalibrated standards (e.g., Renne et al., 1998). Moreover, the large 2␴ uncertainties of Ͼ1 m.y. for the plateau ages represented little improvement upon Mauger's (1977) results. Recent single-crystal laser-fusion 40Ar/39Ar dating has de®ned the timing of deposition of Bridger Formation strata (Evanoff et al., 1998) that overlie the Green River Formation in the western Greater Green River Basin (Figs. 2 and 3). For the purposes of intercomparison, all 40Ar/39Ar ages in this paper (unless other- wise noted) are reported (or recalculated) relative to the standard ages of Renne et al. (1998), including intercalibration uncertainties, at the 2␴ precision level. Murphey et al. (1999) reported 40Ar/39Ar ages for sanidine, biotite, and plagioclase from the Church Butte (base of middle Bridger B), Henry's Fork (base of upper Bridger C), and unnamed (base of Bridger E) tuffs of 48.65 Ϯ 0.30 Ma, 47.60 Ϯ 0.37 Ma, and 46.83 Ϯ 0.90 Ma, respectively. Prothero (1996b) reported an 40Ar/39Ar age of 47.56 Ϯ 0.14 Ma for Henry's Fork tuff, which, though indistinguishable from the age reported by Murphey et al. (1999), is much more precise. These age determinations con- Figure 4. Field photographs and photomicrographs (crossed nicols) of tuff beds. Abbre- ®rm existing bio- and magnetostratigraphy viations: sanÐsanidine; plagÐplagioclase; and qtzÐquartz. (Flynn, 1986; Prothero and Swisher, 1992; McCarroll et al., 1996; Prothero, 1996a) and help to de®ne the upper age limit of the Green plagioclase and minor hornblende, pyroxene, lake water. Studies of alteration in modern and River Formation (Figs. 2 and 3). They also and zircon (Table 1). The percentage of crys- ancient saline and alkaline lake deposits show temporally quantify the distal record of Eo- talline to matrix material within any speci®c that volcanic glass is initially converted to ze- cene alkali-calcic silicic volcanism in western tuff horizon is independent of its thickness. olites such as clinoptilolite and mordenite be- North America (Fig. 1) and provide a detailed Many of the beds possess angular, matrix- fore being altered to analcime (Hay, 1966; picture of post±Green River Formation basin supported phenocrysts that exhibit a gradation Goodwin and Surdam, 1967; Sheppard and development that is complementary to our re- to smaller, less abundant crystals from the Gude, 1969). Analcime is then converted to sults (Fig. 3). base upward (Fig. 4). Because these tuff beds K-feldspar under ultrasaline conditions when are generally found interbedded with profun- the Naϩ/Kϩ activity ratio is suf®ciently low GEOLOGIC SETTING OF TUFF BEDS dal mudstone (Fig. 4, Table 1), we interpret (Hay, 1966; Goodwin, 1973; Desborough, them to be primary ash-fall deposits. Re- 1975). Our observations con®rm Surdam and Lacustrine facies of the Green River For- worked tuff beds have similar compositions Parker's (1972) conclusion that tuff beds de- mation are interbedded with numerous indi- but also contain lithic fragments and carbo- posited during the most saline and alkaline vidual water-laid tuff horizons that range from naceous material and are characterized by phase of Lake Gosiute generally exhibit K- 1 mm to several meters thick; thinner beds sorting and cross-strati®cation indicative of feldspar alteration, whereas those in less saline greatly outnumber thicker beds. The tuff beds subaqueous, tractive reworking (cf. KoÈniger facies contain only zeolite and analcime alter- represent widespread isochronous markers and and Stollhofen, 2001). Reworked beds are typ- ation. Additionally, in Green River Formation have been extensively used for stratigraphic ically found in association with coarser- tuff beds we recognized the signi®cant lateral correlation (e.g., Culbertson, 1961; Bradley, grained facies indicative of higher-energy sed- alteration variations described by previous 1964; Smoot, 1983; Roehler, 1992b). Gener- imentary environments and were avoided in workers (Hay, 1966; Iijima and Hay, 1968; ally light gray to orange in outcrop, tuff beds this study. Ratterman and Surdam, 1981; Buchheim and are ®ne grained and typically contain Ͻ20% Most of the original glassy matrix of the Eugster, 1998). In particular, basin-center de- phenocrysts of quartz, biotite, sanidine, and tuff beds was altered by saline and alkaline posits typically exhibit more alteration, sug-

Geological Society of America Bulletin, May 2003 553 SMITH et al.

that the age of the GA-1550 primary standard remains controversial (e.g., Lanphere and Dal- rymple, 2000; Lanphere and Baadsgaard, 2001), to some extent owing to uncertainties in the 40K decay constant (Min et al., 2000; Schmitz and Bowring, 2001). The decision to use the values just given is based, in part, on improved agreement between 40Ar/39Ar and astronomical ages for the last several reversals of the geomagnetic ®eld (Renne et al., 1994) and on a U-Pb monazite intercalibration of FCs at 27.98 Ϯ 0.15 Ma (Villeneuve et al., 2000). As an internal check on accuracy, we made several laser-fusion measurements of GA-1550 biotite, FCs, MMhb, and U.S. Geo- logical Survey standard biotite SB-3, with results listed in Table DR1.1 Methods used in the extraction and analysis of argon follow Singer and Brown (2002). Mass discrimination was monitored by using an on-line air pipette and varied between 1.0025 Ϯ 0.0010 and 1.0035 Ϯ 0.0010 per amu during the 15-month analytical period. Inverse-variance weighted-mean ages and standard deviations were calculated according to Taylor (1982) by using the ArArCALC software of Koppers (2002). Where mean squared weighted deviation (MSWD) values were Ͼ1, the uncertainties were multiplied by the square root of the MSWD. Precision estimates for monitor positions, based on 14±54 isotopic measurements of standard grains sur- Figure 5. Microprobe transects and backscattered-electron images of volcanic grains. A: rounding sample positions, reveal a gradient Main tuff, sanidine. B: Grey tuff, magmatically zoned sanidine with authigenic K-feldspar of up to 0.5% per cm in neutron ¯uence across overgrowth. C: Grey tuff, sanidine with authigenic K-feldspar rim, showing the effect of the discs. By using an inverse-distance- progressive air abrasion. weighted spatial-interpolation algorithm, the uncertainty in J was estimated at between 0.12% and 0.24%, which was propagated into gesting that they were exposed to increasingly itors and irradiated for either 75 (UW-06) or 50 the ®nal weighted-mean ages for each sample saline and alkaline ¯uids as the lake shrank. (UW-11) h at the Oregon State University Triga (Fig. DR1 [see text footnote 1]). To test for reactor where they received fast-neutron doses excess argon, isochrons were regressed by us- 40Ar/39Ar METHODS of ϳ7.5±5 ϫ 1018 cm±2, respectively. Correc- ing the method of York (1969). Ages were cal- tions for undesirable nucleogenic reactions on culated with the decay constants of Steiger 40 40 40 39 Tuff beds selected for this study are from K and Ca are as follows: [ Ar/ Ar]K ϭ and JaÈger (1977) and are reported with Ϯ2␴ 36 37 39 locations exhibiting the least degree of alter- 0.00086; [ Ar/ Ar]Ca ϭ 0.000264; and [ Ar/ uncertainties. 37 ation we observed. Sanidine and biotite crys- Ar]Ca ϭ 0.000673. Uncertainties are reported in three ways: tals 75±500 (␮m in diameter were separated The 40Ar/39Ar method requires the age of a ®rst, as internal analytical uncertainties re¯ect- from seven tuffs by using a combination of sample to be calculated relative to mineral ing only precision on peak signals, system crushing, acid leaching, heavy-liquid separa- standards that have been previously dated. blanks, mass discrimination, and neutron ¯u- tion, and air abrasion (Fig. 5; Krogh, 1982). Sanidine from the Taylor Creek rhyolite (TCs; ence; second, as analytical and intercalibration Sanidine crystals (ϳ0.001 mg each) were Duf®eld and Dalrymple, 1990) was used to uncertainties that relate secondary standards handpicked under refractive-index oils, then monitor the experiments. Its age is 28.34 Ϯ 1 cleaned ultrasonically in cold 10% HF for 10 0.28 Ma relative to 98.79 Ϯ 0.96 Ma for the GSA Data Repository item 2003064, a table showing intercalibration of 40Ar/39Ar standards per- min, rinsed in deionized water, and packaged primary standard GA-1550 biotite (McDou- formed at the UW-Madison, a diagram illustrating for irradiation. Euhedral, optically fresh, 250± gall and Roksandic, 1974; Renne et al., 1998), the distribution of measured neutron ¯uence (J) 500 ␮m diameter biotite crystals were sepa- which is intercalibrated to 28.02 Ϯ 0.28 Ma across the irradiation discs, a table containing the rated prior to acid leaching and handpicked for for the sanidine from the Fish Canyon Tuff full 40Ar/39Ar data set, and a table giving the recalibrated magnetic chron boundaries, is available on irradiation. (FCs) and 523.1 Ϯ 4.6 Ma for hornblende the Web at http://www.geosociety.org/pubs/ft2003. These crystals were loaded into 1.6 cm di- from the McClure Mountain syenite (MMhb; htm. Requests may be sent also to editing@ ameter aluminum discs together with ¯ux mon- Sampson and Alexander, 1987). We recognize geosociety.org.

554 Geological Society of America Bulletin, May 2003 40Ar/39Ar GEOCHRONOLOGY OF THE EOCENE GREEN RIVER FORMATION, WYOMING

TABLE 2. SUMMARY OF 40Ar/39Ar LASER-FUSION EXPERIMENTAL RESULTS FOR SEVEN TUFF BEDS

Sample Position n Isochron analysis² Average Apparent ages² number (latitude K/Ca Ϯ 2␴ 40 36 ³ ³ § # longitude) SUMS Ar/ Ari Isochron age MSWD Weighted-mean Ϯ2␴ Ϯ2␴ (N-2) (Ma) Ϯ 2␴ age (Ma) Ϯ 2␴ Analcite tuff 41Њ21Ј01.38ЉN 19 of 19 0.85 281.2 Ϯ 48.9 49.00 Ϯ 0.21 Sanidine 0.78 48.94 Ϯ 0.12 0.18 0.85 SB-1 108Њ40Ј04.68ЉW 779 Ϯ 340 Sixth tuff 41Њ32Ј31.88ЉN 22 of 26 0.76 294.1 Ϯ 9.7 49.72 Ϯ 0.14 Biotite 0.73 49.70 Ϯ 0.10 0.17 0.86 TR-5 109Њ28Ј55.95ЉW 884 Ϯ145 8 of 25 0.58 290.8 Ϯ 16.1 49.98 Ϯ 1.20 Sanidine 0.55 49.79 Ϯ 1.04 1.05 1.34 2778 Ϯ 529 Main tuff 41Њ32Ј31.88ЉN 30 of 31 0.77 293.4 Ϯ 6.4 49.98 Ϯ 0.09 Sanidine 0.65 49.96 Ϯ 0.08 0.16 0.86 TR-1 109Њ28Ј55.95ЉW 97.5 Ϯ 11.1 16 of 30 1.05 296.1 Ϯ 8.1 50.00 Ϯ 0.20 Biotite 0.99 50.01 Ϯ 0.15 0.21 0.87 18962 Ϯ 3066 Grey tuff 41Њ39Ј20.62ЉN 18 of 18 0.73 285.7 Ϯ 11.3 50.55 Ϯ 0.21 Sanidine 0.67 50.39 Ϯ 0.13 0.19 0.87 WN-1 109Њ17Ј15.16ЉW 190 Ϯ 130 C Bed tuff 41Њ20Ј42.36ЉN 15 of 20 0.49 299.8 Ϯ 7.5 50.35 Ϯ 0.45 Biotite 0.55 50.56 Ϯ 0.26 0.30 0.90 FC-3 109Њ25Ј37.85ЉW 531 Ϯ 86 Firehole tuff 41Њ21Ј00.96ЉN 14 of 14 1.05 295.5 Ϯ 1.7 50.70 Ϯ 0.15 Sanidine 0.97 50.70 Ϯ 0.14 0.20 0.88 FC-2 109Њ22Ј60.00ЉW 190 Ϯ 130 Rife tuff 41Њ57Ј47.23ЉN 13 of 24 0.67 292.6 Ϯ 11.4 51.41 Ϯ 0.68 Biotite 0.63 51.25 Ϯ 0.31 0.34 0.93 BT-18 109Њ15Ј08.66ЉW 701 Ϯ 115 ²Ages relative to 28.34 Ma for Taylor Creek rhyolite (TCs) (Renne et al., 1998). ³Analytical uncertainties. §Analytical and intercalibration uncertainties. #Fully propagated uncertainties. with the GA-1550 primary standard; and third, ability diagram toward older ages. Because of the Tipton Member, 2 m below the contact as fully propagated uncertainties that incorpo- their euhedral shape and position at the base with the Wilkins Peak Member. Because the rate analytical and intercalibration uncertain- of apparently unreworked ash-fall beds, a de- Rife Bed is much thinner (only 4 m) at the ty in addition to uncertainty associated with trital origin for these biotite crystals is unlike- Boar's Tusk section as compared with other isotope-dilution K-Ar measurements of the ly. Instead, we interpret these outliers to result more basinward measured sections, it is dif- primary standard and rather large, but under- from the inclusion of xenocrystic biotite aris- ®cult to be precise about the stratigraphic po- appreciated uncertainty in the 40K decay con- ing from either the incorporation of older sition of the Rife tuff. However, the Rife tuff stants (Table 2; Karner and Renne, 1998; grains from the preeruptive volcanic edi®ce or was chosen for dating because it contains Schmitz and Bowring, 2001). Internal analyt- from the incomplete degassing of crystals de- abundant coarse-grained biotite, whereas other ical uncertainties are appropriate when assess- rived from basement rocks fragmented during tuffs collected from the Tipton Member have ing age differences among samples that ref- eruption. As a result, biotite analyses have as yet failed to produce datable phenocrysts. erence the same standard, whereas been excluded from isochron and weighted- Twenty-four experiments on three-grain al- intercalibration uncertainties permit direct mean calculations when their inclusion results iquots of biotite allowed the identi®cation and comparisons with ages or time scales calcu- in MSWD values Ͼ1 (Deino and Potts, 1990; exclusion of outliers from the population. The lated relative to other 40Ar/39Ar standard ages. Ton-That et al., 2001). A lack of excess argon cumulative probability plot (Fig. 6A) for the Direct comparison of our 40Ar/39Ar ages to U- is suggested by isochron analyses; however, Rife tuff shows a bimodal age distribution. Al- Pb ages (e.g., Schmitz and Bowring, 2001) the spread of data points along many of the though both populations (ca. 51.25 and ca. and other independent isotopic chronometers isochrons is limited owing to high contents of 53.0 Ma) ®t the stratigraphic order of ages of necessitates the use of fully propagated radiogenic argon, generally Ͼ80%±90% (Fig. other tuffs (Fig. 6), we prefer the minimum uncertainties. 6). Thus, we take the weighted-mean apparent age as the best estimate of time elapsed since ages as our most precise estimates of time deposition of the Rife bed. Because the tuff is 40Ar/39Ar RESULTS elapsed since eruption and deposition of these interbedded with laminated calcareous mud- tuff beds. Because our emphasis is initially on stones and shows no indications of reworking, Ages for the Green River Formation were determining the duration of lacustrine depo- a xenocrystic and not detrital interpretation for determined on the basis of 207 single and sition in the Greater Green River Basin, the older population is favored. Because each multicrystal laser-fusion 40Ar/39Ar experiments weighted-mean apparent ages and isochrons fusion incorporated three grains, we suggest on phenocrysts from seven tuff beds (Fig. 6; are discussed in the following sections with that the older apparent ages represent mixing Table 2; complete analytical data in Table 2␴ analytical uncertainties that are appropriate of one or more xenocrystic grains with juve- DR2 [see text footnote 1]). Samples sites lat- for intercomparison of each dated tuff. nile biotite grains. Thus, the actual age of the erally span a total of 90 km and 410 m of xenocrystic material likely exceeds the older vertical stratigraphy in the Greater Green Riv- Rife Tuff peak of probability density shown in Figure er Basin (Figs. 1 and 2; Table 2). Fusions of 6A. When these outliers are omitted from the 1±10 crystals each were performed in order to The Rife tuff (sample BT-18) was collected isochron and weighted-mean calculations, 13 obtain adequate signal sizes from grains com- from the Rife Bed of the Tipton Member at of a total of 24 fusions de®ne an isochron of monly smaller than 1 ␮g. In some instances, the Boar's Tusk section of Pietras et al. (2003), 51.41 Ϯ 0.68 Ma with an atmospheric 40Ar/ analyses of one- to three-grain aliquots of bi- ϳ45 km north of Rock Springs, Wyoming 36Ar intercept of 292.6 Ϯ 11.4. These 13 biotite gave ages that skew the cumulative prob- (Table 1). It is located within the Rife Bed of otite analyses yielded a weighted-mean age of

Geological Society of America Bulletin, May 2003 555 SMITH et al.

Figure 6. Cumulative probability and inverse isochron diagrams illustrating 40Ar/39Ar experimental results. Analyses excluded from age calculations shown as open sym- bols. In all cases the weighted-mean age is the preferred depositional age. Gray shading indicates 2␴ en- velope of uncertainty for each sample. Note that the tuff samples are arranged in stratigraphic order and that the determined ages are consistent with this order. MSWDÐ mean square of weighted deviates.

556 Geological Society of America Bulletin, May 2003 40Ar/39Ar GEOCHRONOLOGY OF THE EOCENE GREEN RIVER FORMATION, WYOMING

TABLE 3. SUMMARY OF AVERAGE MAJOR ELEMENT COMPOSITIONS OF PHENOCRYSTS

Oxide Sixth Main Oxide Analcite Sixth Main Grey C bed Firehole Biotite Biotite Sanidine Sanidine Plag Sanadine Plag Auth. Sanidine Plag Sanadine Plag Sanadine Plag n ϭ 26 n ϭ 10 n ϭ 137 n ϭ 19 n ϭ 4 n ϭ 18 n ϭ 142 K-spar Core n ϭ 13 n ϭ 15 n ϭ 8 n ϭ 11 n ϭ 4 n ϭ 14 n ϭ 44

SiO2 34.4 33.1 SiO2 65.2 65.2 69.0 64.3 66.6 63.6 62.6 61.1 63.8 60.3 63.0 57.9

Al2O3 13.6 13.7 Al2O3 19.0 19.1 19.9 18.6 19.4 17.6 18.7 25.0 18.8 24.7 19.1 26.0 FeO 20.7 25.6 FeO 0.12 BDL BDL BDL BDL BDL 0.08 0.16 BDL BDL BDL 0.14 MgO 9.97 6.38 CaO 0.28 0.09 BDL 0.10 BDL BDL 0.13 5.75 BDL 6.03 0.17 8.07

MnO 0.13 0.47 Na2O 6.14 1.74 12.2 2.62 11.9 0.23 2.44 8.05 1.02 7.83 3.28 6.70

CaO 0.01 0.02 K2O 7.69 14.5 0.03 12.7 0.04 15.8 12.5 0.86 15.6 0.55 11.4 0.36

Na2O 0.49 0.31 BaO BDL 0.23 BDL 0.18 BDL BDL 1.22 BDL 0.70 BDL 2.03 BDL

K2O 7.12 7.41 Total 98.5 100.8 101.3 98.6 97.9 97.3 97.7 100.9 100.0 99.5 99.1 99.2

TiO2 4.96 4.05

BaO 1.13 0.53 Anx 1 1 27 29 1 39

Total 93.0 91.7 Aby 54 15 100 24 100 1 23 68 9 68 30 59

Orz 45 84 76 99 77 5 91 3 69 2 Note: Wavelength-dispersive measurements were made with a Cameca SX51 electron microprobe using a 10 nA, 15 keV, 5 ␮m defocused beam, with 10 s peak and 10 s background counting times, and natural and synthetic mineral standards. Compilation includes 465 individual electron-microprobe analyses. Abbreviations: nÐnumber of analyses, BDLÐbelow detection limit.

51.25 Ϯ 0.31 Ma (MSWD ϭ 0.63) (Fig. 6A, crystals separated from the C Bed tuff were well-de®ned Gaussian peak in a cumulative Table 2). indicative of xenocrystic or detrital contami- probability plot (Fig. 6D). The 50.55 Ϯ 0.21 nation, 40Ar/39Ar experiments were undertaken Ma isochron de®ned by these experiments has Firehole Tuff on three-grain aliquots of biotite crystals. an 40Ar/36Ar intercept of 285.7 Ϯ 11.3. The Twenty experiments allowed the identi®cation weighted mean of the 18 analyses is 50.39 Ϯ The Firehole tuff (sample FC-2), or First and exclusion of outliers from the population. 0.13 Ma (MSWD ϭ 0.67) and gives the pre- tuff (Culbertson, 1961) was collected from the The cumulative probability plot (Fig. 6C) for ferred age for the Grey tuff (Fig. 6D, Table lower part of the Wilkins Peak Member of the the C Bed tuff is skewed toward older ages 2). Green River Formation at Firehole Canyon that are interpreted to be xenocrystic in origin (Table 1). It is stratigraphically ϳ60 m above and were omitted from weighted-mean and Main Tuff the base of the Wilkins Peak Member (bed isochron calculations as already described. 264, Plate 1 of Roehler, 1992a) and contains Fifteen of a total of 20 three-crystal fusions Sample TR-1 was collected from the upper chemically homogeneous, unaltered Or69 san- de®ne an isochron of 50.35 Ϯ 0.45 Ma with part of the Wilkins Peak Member at Toll Gate 40 36 idine and An39 andesine phenocrysts (Table 3). an atmospheric Ar/ Ar intercept of 299.8 Ϯ Rock (Figs. 1 and 2; Tables 1 and 2). The Six single-crystal laser fusions taken together 7.5 and a preferred weighted-mean age of Main or Third tuff of Culbertson (1961) is the with eight multicrystal measurements of san- 50.56 Ϯ 0.26 Ma (MSWD ϭ 0.55) (Fig. 6C, lowermost of three Ͼ10-cm-thick tuff beds idine form a well-de®ned Gaussian peak in a Table 2). and numerous smaller tuff beds in an 8-m- cumulative probability diagram without evi- thick span 6±14 m below the ``I'' sandstone- dence of xenocrystic contamination (Fig. 6B). Grey Tuff mudstone unit (bed 392, Plate 1 of Roehler, Collectively, these analyses de®ne an isochron 1992a). Phenocrysts separated from the low- of 50.70 Ϯ 0.15 Ma with an 40Ar/36Ar inter- The Grey tuff sample WN-1 was collected ermost unit are chemically homogeneous, un- cept of 295.5 Ϯ 1.7, consistent with an at- from the middle of the Wilkins Peak Member altered Or76 sanidine and presumably altered mospheric trapped component. The weighted at a section 5 km north of Rock Springs, Wy- Ab100 albite crystals (Fig. 5A, Table 3). As the mean of these analyses yields a preferred age oming (Figs. 1, 2, and 4A; Tables 1 and 2), glassy ash matrix of the Main tuff has been of 50.70 Ϯ 0.14 Ma (MSWD ϭ 0.97) for the and is stratigraphically ϳ9 m above the ``D'' altered to ®ne-grained analcime, we speculate Firehole tuff (Fig. 6B). sandstone-mudstone unit (ϳbed 321, Plate 1 that the albitization of plagioclase phenocrysts of Roehler, 1992a; base of DE2 oil shale of re¯ects authigenic processes.

C Bed Tuff Culbertson, 1998). Unaltered, euhedral, Or75 Sanidine from the lowermost unit of the sanidine phenocryst cores with magmatic Ba Main tuff yielded 30 (of 31) multigrain fu-

The C Bed tuff (sample FC-3) was collected zonation and andesine An27 phenocrysts are sions that de®ne an isochron of 49.98 Ϯ 0.09 40 36 ϳ5 m below the base of the ``C'' sandstone- encased in Or99 K-feldspar resulting from the Ma with an Ar/ Ar intercept of 293.4 Ϯ 6.4. mudstone unit (ϳbed 302, Plate 1 of Roehler, alteration of the original glassy ash matrix One analysis that was excluded from the iso- 1992a; base of BC6 oil shale of Culbertson, (Fig. 5B, Table 3). Plagioclase and sanidine chron and weighted-mean calculations gave an 1998) from the lower Wilkins Peak Member compositions form a near-equilibrium tie line apparent age of 54.87 Ϯ 1.18 Ma, indicating at Firehole Canyon, Wyoming (Table 1). on a ternary-feldspar diagram, suggesting the probable xenocrystic contamination. The 30

Chemically homogeneous, unaltered Or91 san- coexistence of these phases in the preeruptive included analyses form a well-de®ned Gauss-

idine and An29 andesine phenocrysts (Table 3) magma. Backscattered-electron images illus- ian distribution in a cumulative probability are encased within a matrix of authigenic K- trate the effectiveness of the air-abrasion tech- plot (Fig. 6F). The weighted mean of these feldspar and analcime, necessitating the use of nique at removing the K-feldspar overgrowths measurements, 49.96 Ϯ 0.08 Ma (MSWD ϭ air abrasion to eliminate overgrowths. from phenocryst cores (Fig. 5C). 0.78), is the preferred age for this tuff bed Because multigrain fusions of sanidine Eighteen multigrain aliquots form a single, (Fig. 6F, Table 2).

Geological Society of America Bulletin, May 2003 557 SMITH et al.

40Ar/39Ar experiments were also undertaken ferred weighted-mean age of 49.70 Ϯ 0.10 Ma aged by the fact that biotite analyses for the on biotite crystals from the lowermost unit of (MSWD ϭ 0.73) (Fig. 6G, Table 2). Main tuff yielded an age that is indistinguish- the Main tuff. Thirty-one experiments allowed able from its sanidine age when potential xen- the identi®cation and exclusion of outliers Analcite Tuff ocrystic or detrital contamination was re- from the population as already described. The moved from the calculation (Fig. 6, C and D). cumulative probability plot (Fig. 6E) for the Sample SB-1 is from the upper part of the Moreover, biotite ages for the Sixth, C Bed, Main tuff is skewed toward older ages, which LaClede Bed of the Laney Member on the and Rife tuffs are in agreement with their are interpreted to be xenocrystic in origin, and northwest face of Sand Butte, eastern Greater stratigraphic position relative to the other were omitted from weighted-mean and iso- Green River Basin (Fig. 1, Table 2). The Anal- ages. chron calculations as already described. Fif- cite tuff occurs 7 m above the Buff Marker teen of a total of 31 three-crystal fusions de- Bed (bed 433, Plate 2 of Roehler, 1992a) and DISCUSSION ®ne an isochron of 50.00 Ϯ 0.20 Ma with an is correlative across the eastern Greater Green atmospheric 40Ar/36Ar intercept of 296.1 Ϯ River Basin (Ratterman and Surdam, 1981). Accumulation Rates of the Green River 8.1. Biotite analyses yielded a weighted-mean Microprobe transects of sanidine grains reveal Formation age of 50.01 Ϯ 0.15 Ma (MSWD ϭ 0.99), homogeneous Or45 compositions and show no which is virtually identical to the more precise evidence for chemical alteration (Table 3). Previous estimates of Green River Forma- sanidine age of 49.96 Ϯ 0.08 Ma (Fig. 6F, Sanidine crystals from the Analcite tuff are tion accumulation rates were based on mea- Table 2) for the Main tuff. highly radiogenic; their mean 40Ar* values of suring the thicknesses of laminae Ͻ1mm 97.8% limit the spread along the isochron, thick in organic-rich calcimicrites (oil shales), Sixth Tuff which is 49.00 Ϯ 0.31 Ma. The 31 multigrain under the assumption that these laminae rep- analyses de®ne a Gaussian peak on a cumu- resent varves (Bradley, 1929; Crowley et al., Sample TR-5 was collected from the tran- lative probability plot (Fig. 6H) and show no 1986; Ripepe et al., 1991). Ripepe et al. sitional interval between the Wilkins Peak and evidence for xenocrystic contamination. Taken (1991) reported the average lamination thick- Laney Members at Toll Gate Rock ϳ1km together, these 31 analyses yield a weighted- ness in the Tipton and Laney Members to be west of Green River, Wyoming (Figs. 1 and mean age of 48.94 Ϯ 0.12 Ma (MSWD ϭ 96 ␮m and 107 ␮m, respectively. However, 2; Tables 1 and 2). The Sixth tuff of Culbert- 0.78) (Fig. 6H, Table 2). using laminae thickness to estimate accumu- son (1961) is located ϳ32 m above the ``I'' lation rates is problematic (cf. Dott, 1983), es- sandstone-mudstone unit (bed 423, Plate 1 of Summary of Analytical Data pecially because evidence for depositional hi- Roehler, 1992a). Microprobe analyses of phe- atuses and periods of erosion occurs nocrysts reveal chemically homogeneous, un- The weighted-mean ages determined from throughout the Green River Formation altered Or84 sanidine phenocrysts and presum- the seven tuff beds are between 51.25 Ϯ 0.31 (Smoot, 1983; Surdam and Stanley, 1979; ably altered Ab100 albite crystals (Table 3). and 48.94 Ϯ 0.12 Ma and are consistent with Roehler, 1993; Bohacs et al., 2000). In addi- Similar authigenic albitization of plagioclase their stratigraphic positions (Figs. 2 and 6). tion, much of the formation, particularly the was documented by Surdam and Parker We therefore interpret these ages to represent evaporative Wilkins Peak Member, is com- (1972) to be associated with analcime and K- the depositional ages of the sediments in posed of nonlaminated facies for which varve feldspar alteration in Green River Formation which they occur. The K-Ar and 40Ar/39Ar counting cannot be applied to estimate accu- tuff beds. ages of Mauger (1977) and O'Neill (1980) for mulation rate. Sanidine from the Sixth tuff proved to be the end of Green River Formation sedimen- Our 40Ar/39Ar results permit the ®rst direct heterogeneous in age, which limited the num- tation are 3±4 m.y. younger than those pre- measurement of accumulation rates in the ber of grains that could be grouped with con- sented here (Fig. 3), suggesting that some bi- Green River Formation that do not rely on ®dence to de®ne the juvenile eruptive age. otite grains that had undergone 40Ar loss due lamina counts. For a reference section near the Only 8 of 25 multigrain analyses, each com- to diagenetic alteration may have been incor- southern Green River Basin depocenter prising three crystals, gave stratigraphically porated into their analyses. We avoided this (Roehler, 1992a, Plate 1), we calculate accu- reasonable ages and yielded an imprecise problem by analyzing aliquots of handpicked mulation rates of 88 Ϯ 34 and 104 Ϯ 18 ␮m/ weighted-mean age of 49.79 Ϯ 1.04 Ma (Ta- euhedral biotite crystals that were ϳ100,000 yr for the Tipton and Laney Members, respec- ble 2). times smaller than those analyzed by Mauger tively. These rates are consistent with an 40Ar/39Ar experiments were also performed (1977) or O'Neill (1980). The ages for the annual origin for the laminae, as previously on euhedral biotite crystals separated from the Analcite, Sixth, Main, Grey, Firehole, and proposed (Fig. 7; Bradley, 1929; Crowley et basal 1 cm of the Sixth tuff. Twenty-six ex- Rife tuffs were the most precise, with internal al., 1986; Ripepe et al., 1991). However, be- periments allowed the identi®cation and ex- 2␴ uncertainties of Ϯ80±140 k.y. Xenocrystic cause of the evidence for depositional hiatuses clusion of outliers from the population. The or detrital contamination was clearly identi®ed throughout the Tipton and Laney Members, cumulative probability plot (Fig. 6G) for the in experiments on biotite from the Sixth, our data cannot completely exclude faster true Sixth tuff is skewed toward older ages, which Main, C Bed, and Rife tuffs, and these anal- rates for sediment accumulation. Depending are interpreted to be xenocrystic in origin, and yses were excluded from the age calculations on how much time is ``missing'' from the were omitted from the weighted-mean and for these beds. None of the mineral separates stratigraphic record, the laminae might in ac- isochron calculations as already described. analyzed showed de®nitive evidence of excess tuality be seasonal rather than annual, and the Twenty-one of a total of 26 three-crystal fu- argon (Fig. 6). individual lake cycles might be shorter than sions de®ne an isochron of 49.72 Ϯ 0.14 Ma Despite evidence from other Tertiary ash- calculated. with an atmospheric 40Ar/36Ar intercept of ¯ow tuffs that biotite gives older ages than A very different accumulation rate story 294.1 Ϯ 9.7. Biotite analyses yielded a pre- sanidine (e.g., Lipman, 2000), we are encour- emerges from the Wilkins Peak Member. For

558 Geological Society of America Bulletin, May 2003 40Ar/39Ar GEOCHRONOLOGY OF THE EOCENE GREEN RIVER FORMATION, WYOMING

Roehler, 1993), and western (DeCelles, 1994) margins of the Greater Green River Basin indicate active Eocene uplift of basin-bounding structures. Furthermore, isopach trends (Sul- livan, 1985; Beck et al., 1988; Roehler, 1993) indicate increased tectonically driven differen- tial subsidence in the eastern Greater Green River Basin during middle Wilkins Peak deposition. The increase in sediment- accumulation rate shown in Figure 7 may therefore coincide with a pulse of Laramide tectonism at ca. 50.5 Ma. Our new ages support the notion that 1±3 m lake expansion-contraction cycles in the La- ney Member may record precessional forcing (Surdam and Stanley, 1979; Rhodes et al., 2002). However, age data indicate that individual lake expansion-contraction episodes re- corded in the evaporative Wilkins Peak Mem- Figure 7. Chart showing average sediment-accumulation rates for the Green River For- ber did not occur in response to 21 k.y. mation based on reference section in the southern Greater Green River Basin (see Roehler, precessional forcing as proposed by Roehler 1992a, Plate 1). Lake-type patterns as in Figure 3. Note the abrupt increase in accumu- (1993). Roehler (1991, 1993) documented 77 lation rate corresponding to the Wilkins Peak Member. correlative lithologic cycles, each de®ned by a basal organic-rich calcimicrite (oil shale) bed, that represent episodes of lake expansion the same southern Green River Basin refer- rite facies than for lake mud, and Bobst et al. and contraction. On the basis of an estimated ence section (Roehler, 1992a, Plate 1), we cal- (2001) noted that accumulation rates during 1.6 m.y. duration for the Wilkins Peak, Roeh- culate a net average rate of 327 Ϯ 85 ␮m/yr, dry phases of the Salar de Atacama were up ler (1993) thus calculated an average period or approximately three times as fast as for the to three times as high as during wet periods. of 21,779 k.y. In contrast, our 40Ar/39Ar ages Tipton or Laney Members. A part of this dif- Net accumulation rates in these cases range give an average duration of the 69 oil shale ference might be attributable to differing between ϳ0.5 and 1.8 mm/yr, which is up to cycles that occur between the Firehole and amounts of compaction, as the Wilkins Peak ®ve to six times faster than deposition of the Sixth tuffs of 14.5 Ϯ 1.7 k.y. In addition, re- Member contains a lower proportion of pro- Wilkins Peak Member. However, higher rates cent detailed outcrop and drill-core observa- fundal lake facies compared to the Tipton and of Wilkins Peak Member evaporite accumula- tions (Smith et al., 2001) indicate that Wilkins Laney Members (Bradley, 1964; Eugster and tion may have occurred at its subsurface de- Peak expansion-contraction cycles are more Hardie, 1975; Surdam and Stanley, 1979; pocenter (to the west of the reference section). numerous than indicated by Roehler (1993). Smoot, 1983; Roehler, 1993). Reed and Oertel Our observation of faster rates in lake- Our age data therefore exclude 19±23 k.y. pre- (1978) used the degree of preferred orientation marginal facies supports that of Buchheim cessional periods as the cause of these cycles. of clay particles to estimate compaction in the (1994), who noted that rates of carbonate Tipton and Wilkins Peak Members and con- precipitation in the adjacent Fossil Basin Calibration and Correlation of the Eocene cluded that thinly laminated, organic-rich mi- were faster near the lake shore than at its cen- Paleontologic Record crite had undergone up to twice as much com- ter. In contrast, studies of clastic-sediment± paction as massive micrite. However, the dominated lake systems have documented ba- Before we can discuss the broader impli- actual difference in gross interval compaction sin-center rates that are much faster than rates cations of these new ages, they must be con- of these units should be less, because each on the basin margin (e.g., Ensley and Verosub, sidered at intercalibration uncertainties (Table unit contains both lithologies. Furthermore, la- 1982; Elston et al., 1994), presumably re¯ect- 2; Renne et al., 1998) and relative to a com- cunae appear to occur most frequently in the ing sediment bypass. parable global time scale. Use of the most re- Wilkins Peak Member (Smoot, 1983; Roehler, A tectonically derived increase in potential cent geomagnetic polarity time scale (GPTS) 1993; Bohacs et al., 2000), suggesting that the accommodation during deposition of the mid- (Cande and Kent, 1992, 1995; Berggren et al., calculated accumulation rate more seriously dle part of the Wilkins Peak Member may ex- 1995) is inappropriate because it was con- underestimates the true accumulation rates of plain the higher rates of sediment accumula- structed with radioisotopic calibration points the Wilkins Peak than is the case for the Tip- tion during this period (cf. Carroll and that are not compatible with intercalibrated ton or Laney Members. Bohacs, 1999). Uplift of surrounding ranges 40Ar/39Ar standard values (Renne et al., 1998), Our observation of faster accumulation and concurrent basin subsidence may have as noted by Obradovich and Hicks (1999). We rates in evaporative facies may seem counter- augmented erosion rates adjacent to the basin have addressed this problem by reevaluating intuitive, but it is in accord with U-series dis- while simultaneously increasing accommoda- the Eocene GPTS using calibration points cal- equilibrium dating results from Quaternary tion space. Syndepositional relationships be- culated relative to the standard values of Ren- playa-lake evaporites. Ku et al. (1998) found tween lacustrine strata and faults along the ne et al. (1998). This approach shifts the ages that accumulation rates in a core from Death northern (Steidtmann and Middleton, 1991; of the boundaries de®ning chrons 19±24 from Valley were much higher for salt-pan evapo- Pietras et al., 2003), southern (Bradley, 1964; ϩ0.36% to ϩ1.08% (Fig. 8, Table DR3 [see

Geological Society of America Bulletin, May 2003 559 SMITH et al.

alent to the upper ®fth of the Wilkins Peak Member (Roehler, 1989; Zonneveld et al., 2000), can be interpolated to be 49.83 Ϯ 0.39 Ma if a position between the Main and Sixth tuffs is assumed. Note that the current limiting factor on the precision of faunal boundary ages in the Greater Green River Basin is biostratigraphic, not radioisotopic in origin. Future biostratigraphy and correlation of alluvial strata to the Green River Formation therefore have the potential to further improve the age resolution of these boundaries. Biostratigraphic correlation of radiosotopi- cally and magnetostratigraphically de®ned Bighorn Basin sedimentary rocks to the Great- er Green River Basin rocks extends the age model presented here by ϳ2 m.y. (Fig. 3). The Lysitean/Lostcabinian (Wa6/Wa7) NALMA substage boundary in the Bighorn Basin, located near the top of the Willwood Formation, was 40Ar/39Ar dated by Wing et al. (1991). In- Figure 8. Chart showing recalibration of the early and middle Eocene GPTS (Cande and cremental heating of 100 mg of sanidine from Kent, 1992, 1995; Berggren et al., 1995). Cande and Kent's (1992) scaled sea¯oor-anomaly an ash-fall bentonite bed yielded a plateau age model was ®tted by using a cubic spline curve to nine calibration points. We have recal- of 53.09 Ϯ 0.34 Ma (Figs. 1 and 3). Wing et ibrated the Eocene GPTS by replacing two of the original tie points: the ages of 65.0 Ma al.'s (1991) age was also used in this study to for the Cretaceous/Paleocene (K/T) boundary and 55.0 Ma for Paleocene/Eocene (P/E) recalibrate the GPTS based on its position at boundary. Adopting the Renne et al. (1998) intercalibration values, the age of K/T bound- the base of chron 24n.1 (Fig. 8) within the ary shifts 0.77% from 65.0 Ma to 65.5 Ma (Obradovich and Hicks, 1999). Recalculating magnetostratigraphy of Tauxe et al. (1994) the 55.0 Ma age used for the P/E boundary is more problematic, however, because of the and Clyde et al. (1994). Further, the magne- uncertain stratigraphic relationship of the dated no. 70/-17 ash unit to the P/E (NP9/NP10) tostratigraphy of Tauxe et al. (1994) and nannofossil boundary (Berggren and Aubrey, 1996, 1998), unconformities (Aubrey, 1995, Clyde et al. (1994) limits the Graybullian/Lys- 1998), and the transition from C24r to C25n (Townsend, 1985; Ali and Hailwood, 1998) itean (Wa5/Wa6) NALMA substage boundary within Deep Sea Drilling Project core 550 (cf. Wing et al., 2000). Moreover, as a further to the upper third of chron C24n.3, dated at complication to any estimates, disagreement between the 54.51 ؎ 0.10 Ma (Swisher and ca. 53.5 Ma according to the revised GPTS .(Knox, 1991; Berggren and Aubry, 1998) and 55.07 ؎ 0.32 Ma (Wing et al., 1991; Berggren (Figs. 3 and 8; Table 3 et al., 1992) 40Ar/39Ar ages for the no. 70/-17 ash used by Cande and Kent (1992, 1995) to The age constraints for the Graybullian/ estimate the age of the P/E boundary. Because of these problems with the P/E calibration Lysitean (Wa5-Wa6) and Lysitean/Lostcabi- point, we have replaced it with a stratigraphically higher 53.09 ؎ 0.34 Ma 40Ar/39Ar age nian (Wa6-Wa7) substage boundaries, which (Wing et al., 1991) for a tuff bed that is at the base of chron 24n.1 in lower Eocene Bighorn coincide with the main body of the Wasatch Basin sedimentary rocks of the Willwood Formation (Fig. 1; Clyde et al., 1994; Tauxe et Formation±Luman Tongue and the Luman al., 1994; Flynn and Tauxe, 1998). After integrating these two revised calibration points, Tongue±Niland Tongue lithostratigraphic tran- the early and middle Eocene GPTS is fully compatible with 40Ar/39Ar ages relative to the sitions (Gazin, 1965; McGrew and Roehler, intercalibration standards of Renne et al. (1998). 1960; Holroyd and Smith, 2000; Anemone, 2001), can also be applied to the Greater Green River Basin. Taken together with the text footnote 1]). We anticipate that similar cene chron boundaries are approximately Ϯ1 40Ar/39Ar ages presented here, deposition of modi®cation of the Cande and Kent (1992, m.y. the Green River Formation is shown to have 1995) time scale for the early Eocene and oth- The Wasatchian/Bridgerian NALMA bound- been initiated with the Luman Tongue at ca. er time periods will further improve the global ary is biostratigraphically assigned to the allu- 53.5 Ma and to have continued for ϳ5 m.y. GPTS. It must be noted that the precision in vial facies equivalent to the lower middle Wil- (Fig. 3). the age of any individual chron boundary is kins Peak Member (Morris, 1954; McGrew and limited by the intercalibration uncertainty, Roehler, 1960; Gazin, 1965; Krishtalka et al., Magnetostratigraphy typically Ϯ0.2±0.5 m.y., of the calibration 1987; Honey, 1988; Zonneveld et al., 2000). If points used to interpolate it. In addition, the a position between the Firehole and Grey tuffs Clyde et al. (1997) presented magnetostra- stratigraphic position of the calibration points is assumed for the Wasatchian/Bridgerian tigraphy of a section on the western edge of and the spline-curve calculation used to inter- boundary, its age can be interpolated on the the Greater Green River Basin encompassing polate between calibration points introduce basis of new 40Ar/39Ar age determinations to be the main body of the Wasatch Formation, the additional uncertainty. For comparison with 50.55 Ϯ 0.43 Ma. Likewise, the age of the Tipton and Wilkins Peak Members, the Cathe- isotopic chronometers other than 40Ar/39Ar, the Gardnerbuttean/Blacksforkian NALMA sub- dral Bluffs Tongue, and the Laney Member fully propagated uncertainties for early Eo- stage boundary, located in alluvial strata equiv- (Roehler, 1989). Clyde et al. (2001) added

560 Geological Society of America Bulletin, May 2003 40Ar/39Ar GEOCHRONOLOGY OF THE EOCENE GREEN RIVER FORMATION, WYOMING magnetostratigraphy from another section near sources for Green River Formation tuffs in et al., 2001). The EECO represents the most the northeastern edge of the Greater Green (Fig. 1). recent period of protracted greenhouse cli- River Basin encompassing the upper Cathe- The nearest and therefore most likely sourc- mate, characterized by elevated marine and dral Bluffs Tongue, the Laney Member, and es of Green River Formation tuffs lie within terrestrial temperatures between ca. 54 and 49 the lower Bridger Formation. These sections the Absaroka volcanic province, which covers Ma. In addition, the O isotope composition of document two normal-polarity zones within a 25,000 km2 area of northwest Wyoming and early Eocene planktonic foraminifera indicate the main body of the Wasatch Formation and southwest Montana 100±200 km northwest of the lowest latitudinal oceanic temperature (Za- the Cathedral Bluffs Tongue (Fig. 3), from the Greater Green River Basin (Smedes and chos et al., 1994). This interpretation is con- which two alternative correlations to the Prostka, 1972; Sundell, 1993, Fig. 1). Recent sistent with terrestrial ¯ora and fauna in the GPTS of Cande and Kent (1995) were pro- mapping and 40Ar/39Ar dating of several intru- Green River region that suggest a frost-free posed: Correlations ``1'' and ``2'' of Clyde et sive and extrusive rocks document that the high-latitude environment (Markwick, 1994; al. (1997) assign the interval of reversed po- Absaroka volcanic province consists of a Greenwood and Wing, 1995). Mean annual larity located between the two normal-polarity number of individual volcanic centers that temperatures (MAT) in the Rocky Mountain zones to chron C23r and C22r, respectively. In were active between 54 and 43 Ma; younger region inferred from paleo¯oral data varied either case, Green River Formation sedimen- eruptions occurred from volcanic centers in between 15 and 25 ЊC, and mean annual pre- tation is posited to have occurred for a max- the southwestern part of the province (Harlan cipitation (MAP) ranged from 75 to 150 cm/ imum of ϳ3 m.y. et al., 1996; Feeley et al., 1999; Hiza, 1999, yr (Bradley, 1929; Axelrod, 1968; Wilf et al., The new age model presented here, how- Figs. 1 and 3). Average modal compositions 1998). These values are supported by recent ever, suggests that neither correlation pro- of plagioclase and sanidine phenocrysts from estimates from the Green River Formation and posed by Clyde et al. (1997) is correct. Our Absaroka volcanic province ash-¯ow tuffs associated alluvial strata (Wilf, 2000). age model shows that the Green River For- range from An28 to An39 and from Or64 to Or77 Counter to suggestions by Roehler (1993) mation spans ϳ5 m.y. from ca. 53.5 Ma to (Hiza, 1999), overlapping those observed in and Matthews and Perlmutter (1994) that the 48.5 Ma, encompassing both C23r and C22r Green River Formation phenocrysts (Table 3). Wilkins Peak phase of the Green River For- (Fig. 3). We suggest that the section studied However, correlation based on phenocryst mation re¯ects a maximum in continental tem- by Clyde et al. (1997) did not record C23n chemistry to speci®c distal Green River For- perature and aridity, our new 40Ar/39Ar ages within the Tipton Member, owing to weak nat- mation tuff beds is uncertain because of lim- indicate that this most evaporative phase of ural remanent magnetic (NRM) intensities ited distal and proximal phenocryst data. lacustrine deposition in the Greater Green (Ͻ1 mA/M) and erratic demagnetization be- The Challis Volcanic Group of central Ida- River Basin occurred during the latest half of havior. Similarly, Sheriff and Shive (1982) re- ho and southwest Montana, 400±500 km the EECO, when deep-ocean temperatures ported weak NRM intensities and inconsistent west-northwest the Greater Green River Basin were relatively constant or possibly cooling demagnetization in Wilkins Peak Member (Fig. 1), may have also contributed distal ash- (Fig. 9; Shackleton, 1986; Zachos et al., rocks. Thus, weak or overprinted magnetiza- fall deposits to the Greater Green River Basin 2001). Leaf-margin MAT and MAP estima- tion may be responsible for the absence of a during the early and middle Eocene. The de- tions from the Greater Green River Basin mir- chron C23n recording in the Tipton Member. positional area of the Challis Volcanic Group ror the long-term global marine trend, but is of comparable size to the Absaroka volcanic these data are currently limited to the fresh- Correlation of the Green River Formation province; the Challis volcanoes exhibited ex- water to saline Luman, Tipton, and Laney Tuffs with Eocene Volcanism in Western plosive cauldron-forming eruptions between phases of Lake Gosiute and do not sample the North America 49.5 Ma and 45.7 Ma (Fig. 3; Snider and hypersaline Wilkins Peak phase (Fig. 9; Wilf, Moye, 1989; McGonigle and Dalrymple, 2000). Although the terrestrial and marine Crystal-bearing ash-fall deposits such as the 1996) and deposited extensive pyroclastic proxy records are based on limited data sets, one produced from the eruption of Mount St. falls and ¯ows into the Eocene basins of there does not appear to be any striking shift Helens on May 18, 1980, can occur in mea- southwest Montana and east-central Idaho in climate regionally or globally that might surable quantities more than 400 km down- during the same period (Fig. 3; Janecke and correspond to the abrupt increase in sediment- wind of an eruption (Sparks et al., 1997). Pos- Snee, 1993; Palmer and Shawkey, 1997). Cur- accumulation rate and the shift to strongly sible sources of Green River Formation tuff rent 40Ar/39Ar ages, however, limit possible evaporative conditions in Lake Gosiute during beds lie within the Absaroka volcanic prov- correlations between dated Challis Volcanic Wilkins Peak time. A novel alternative expla- ince, Challis Volcanic Group, and numerous Group eruptions and distal Greater Green Riv- nation for these changes was proposed by Pie- smaller volcanic ®elds in Montana that range er Basin tuff beds to the uppermost Green tras et al. (2003), who suggested that renewed in age from 54 Ma to 47 Ma (Marvin et al., River Formation and overlying Bridger and uplift of the Wind River Range may have 1980; Isoplatov et al., 1996; Figs. 1 and 3). Washakie Formations (Fig. 3). helped to divert rivers away from the basin, The distribution of fallout from Plinian vol- thereby reducing runoff into the lake. Our pre- canic plumes is predominantly controlled by Correlation with the Global Oxygen liminary conclusion is that regional tectonic eruptive magnitude and the strength and di- Isotope Record processes may have been at least as important rection of stratospheric winds (Schmincke and as global climate trends in controlling the de- van den Bogaard, 1991; Sparks et al., 1997). The Green River Formation was deposited position of these evaporative facies. Paleotopographically calibrated early Eocene between ca. 53.5 Ma and 48.5 Ma and there- climate models (Sewall et al., 2000) show a fore spans the entire early Eocene climatic op- CONCLUSIONS generally southeast-directed 500 mbar wind timum (EECO) de®ned by global marine O ®eld and are therefore permissive of any of isotope records from benthic foraminifera More than 200 laser-fusion 40Ar/39Ar ex- the three just-mentioned volcanic regions as (Figs. 3, 6, and 9; Shackleton, 1986; Zachos periments on sanidine and biotite crystals

Geological Society of America Bulletin, May 2003 561 SMITH et al. O record from 18 ␦ ations for formations benthic foraminifera. All agesand and members magnetic as chrons in normalized Figure to 3. the Note intercalibration that values of the Renne Green et River al. Formation (1998). coincides Lake-type with patterns the and early abbrevi Eocene climatic optima interval of Zachos et al. (2001). Figure 9. Chart showing correlation of the Green River Formation, the Greater Green River Basin (GGRB) climate estimations, and the Eocene global

562 Geological Society of America Bulletin, May 2003 40Ar/39Ar GEOCHRONOLOGY OF THE EOCENE GREEN RIVER FORMATION, WYOMING

from seven tuff beds in the Green River For- ACKNOWLEDGMENTS Berggren, W.A., Kent, D.V., Obradovich, J.D., and Swisher, C.C., III, 1992, Toward a revised Paleogene geo- mation set limits on interpretations of the We thank Jeff Pietras, Meredith Rhodes, Nick chronology, in Prothero, D.R., and Berggren, W.A., Eocene±Oligocene climatic and biotic evolution: timing and duration of deposition of ϳ410 m Delebo, and Rachel Smith for assistance in the ®eld, of lacustrine mudstone, sandstone, and car- Princeton, New Jersey, Princeton University Press, and Monica Relle and Brian Jicha for their help in p. 29±45. bonate rocks. The Green River Formation the Rare Gas Geochronology Laboratory. We also Berggren, W.A., Kent, D.V., Swisher, C.C., III, and Aubry, thus represents an unparalleled archive of thank Anthony Koppers for guidance with his pro- M.-P., 1995, A revised Cenozoic geochronology and quanti®able information about how large gram ArArCALC, used to regress, calculate, and an- chronostratigraphy, in Berggren, W.A., Kent, D.V., alyze the Ar data, and John Fournelle for sharing Aubry, M.-P., and Hardenbol, J., eds., Geochronology lakes evolve. Speci®cally, we offer the fol- his expertise with the electron probe. We appreciate time scales and global stratigraphic correlation: SEPM lowing conclusions: thoughtful reviews by D.R. Prothero and M. Ville- (Society for Sedimentary Geology) Special Publica- tion 54, p. 129±212. 40 39 neuve, and are gratful to associate editor P.R. Renne 1. Laser-fusion Ar/ Ar experiments on Bobst, A.L., Kowenstein, T.K., Jordan, J.E., Godfrey, L.V., single-crystal or Ͻ0.01 mg multicrystal sam- for helpful comments and suggestions. We are in- Ku, T.-L., and Luo, S., 2001, A 106 ka paleoclimate debted to Steve Binney and Jim Darrough at Oregon ples yielded weighted-mean ages of 51.25 Ϯ record from drill core of the Salmar de Atacama, State University Radiation Center for facilitating the northern Chile: Palaeogeography, Palaeoclimatology, 0.31 Ma (Rife tuff), 50.70 Ϯ 0.14 Ma (Fire- irradiations. Supported by National Science Foun- Palaeoecology, v. 173, p. 21±42. hole tuff), 50.56 Ϯ 0.26 Ma (C Bed tuff), dation grants EAR-9972851 (to Singer) and ATM- Bohacs, K.M., Carroll, A.R., Neal, J.E., and Mankiewicz, 50.39 Ϯ 0.13 Ma (Grey tuff), 49.96 Ϯ 0.08 0081852 (to Singer and Carroll), Geological Society P.J., 2000, Lake-basin type, source potential, and hydrocarbon character: An integrated sequence- Ma (Main tuff), 49.70 Ϯ 0.10 Ma (Sixth tuff), of America grant 6766±00, a grant-in-aid from Sig- ma Xi, and by Conoco Inc. and Texaco Inc. stratigraphic±geochemical framework, in Gierlowski- and 48.94 Ϯ 0.12 Ma (Analcite tuff) in the Kordesch, E.H., and Kelts, K.R., eds., Lake basins Green River Formation. through space and time: American Association of Pe- troleum Geologists Studies in Geology 46, p. 3±34. 2. Sediment-accumulation rates were high- REFERENCES CITED Bradley, W.H., 1929, The varves and climate of the Green est (327 Ϯ 85 ␮m/yr) during the most evap- River epoch: U.S. Geological Survey Professional Pa- orative phase (Wilkins Peak Member) of Lake Ali, J.R., and Hailwood, E.A., 1998, Magnetostratigraphic per 158-E, 110 p. (re)calibration of the Paleocene/Eocene boundary in- Bradley, W.H., 1964, The geology of the Green River For- Gosiute and lowest (88 Ϯ 34 ␮m/yr and 104 terval in Holes 550 and 549, Goban Spur, eastern mation and associated Eocene rocks in southwestern Wy- Ϯ 18 ␮m/yr) during the deposition of less North Atlantic: Earth and Planetary Science Letters, oming and adjacent parts of Colorado and Utah: U.S. evaporative, organic-rich facies (Tipton and v. 161, p. 201±213. Geological Survey Professional Paper 496-A, 86 p. Alroy, J., 1999, The fossil record of North American mam- Bryan, W.B., and Duncan, R.A., 1983, Age and provenance Laney Members). Accumulation rates are con- mals: Evidence for a Paleocene evolutionary radiation: of clastic horizons from hole 516F, in Barker, P.F., and sistent with an annual origin for the Ͻ1-mm- Systematic Biology, v. 48, p. 107±118. Johnson, D.A., Initial reports of the Deep Sea Drilling Alroy, J., Koch, P.L., and Zachos, J.C., 2000, Global cli- Project, Volume 72: Washington, D.C., Government thick laminae in the Tipton and Laney Mem- mate change and North American mammalian evolu- Printing Of®ce, p. 475±477. bers. Increased accumulation rates during the tion: Paleobiology, v. 26, supplement, p. 259±288. Buchheim, H.P., 1994, Paleoenvironments, lithofacies, and Wilkins Peak Member may re¯ect increased Anemone, R.L., 2001, New Paleogene faunas from the varves of the Fossil Butte Member of the Eocene Great Divide Basin of southwestern Wyoming, in Cli- Green River Formation, southwest Wyoming: Univer- uplift and corresponding basin subsidence mate and biota of the early Paleogene: International sity of Wyoming Contributions to Geology, v. 30, from ca. 51 to 50 Ma. Meeting and Field Conference, Powell, Wyoming, 3± no. 1, p. 3±14. 3. In contrast to Clyde et al. (2001), who 8 July: Washington, National Museum of Natural His- Buchheim, H.P., and Eugster, H.P., 1998, Eocene Fossil tory, 2001, p. 2. Lake: The Green River Formation of Fossil Basin, assigned the Green River Formation to an ϳ3 Aubry, M.-P., 1995, From chronology to stratigraphy: In- southwestern Wyoming, in Pitman, J.K., and Carroll, m.y. interval between C23r and C22r on the terpreting the lower and middle Eocene stratigraphic A.R., eds., Modern and ancient lake systems: New basis of correlating magnetostratigraphy to the record in the Atlantic Ocean, in Berggren, W.A., Kent, problems and perspectives: Utah Geological Associ- D.V., and Aubry, M.-P., and Hardenbol, J., eds., Geo- ation Publication 26, p. 191±208. 40 39 geomagnetic polarity time scale, Ar/ Ar chronology, time scales, and global stratigraphic cor- Cande, S.C., and Kent, D.V., 1992, A new geomagnetic ages indicate that it was deposited over a pe- relation: SEPM (Society for Sedimentary Geology) polarity timescale for the Late Cretaceous and Ceno- Special Publication 54, p. 213±274. riod of ϳ5 m.y. spanning magnetic chrons zoic: Journal of Geophysical Research, v. 100, Aubry, M.-P., 1998, Stratigraphic (dis)continuity and tem- p. 13,917±13,951. 24n through 21r. We suggest that Clyde et al.'s poral resolution of geologic events in the upper Cande, S.C., and Kent, D.V., 1995, Revised calibration of (1997, 2001) magnetostratigraphic sections do Paleocene±lower Eocene deep sea record, in Aubry, the geomagnetic polarity timescale for the Late Cre- M.-P., Lucas, S., and Berggren, W.A., eds., Late not record chron 23n within the Tipton Mem- taceous and Cenozoic: Journal of Geophysical Re- Paleocene±early Eocene climatic and biotic events in search, v. 100, p. 6093±6095. ber. Accordingly, the major faunal turnover the marine and terrestrial records: New York, Colum- Carroll, A.R., and Bohacs, K.M., 1999, Stratigraphic clas- bia University Press, p. 37±66. si®cation of ancient lakes: Balancing tectonic and cli- represented by the Bridgerian/Wasatchian Axelrod, D.I., 1968, Tertiary ¯oras and topographic history matic controls: Geology, v. 27, p. 99±102. of the Snake River Basin, Idaho: Geological Society boundary occurred after C23n at 50.55 Ϯ 0.43 Carroll, A.R., and Bohacs, K.M., 2001, Lake-type controls of America Bulletin, v. 79, p. 713±734. Ma. on petroleum source rock potential in nonmarine ba- Beck, R.A., Vondra, C.F., Filkins, J.E., and Olander, J.D., sins: American Association of Petroleum Geologists 4. Tuff beds of the Green River Formation 1988, Syntectonic sedimentation and Laramide base- Bulletin, v. 85, p. 1033±1053. ment thrusting, Cordilleran foreland: Timing of defor- were most likely derived from the nearby Ab- Clyde, W.C., Sheldon, N.D., Koch, P.L., Gunnell, G.F., and saroka volcanic province, although prove- mation, in Schmidt, C.J., and Perry, W.J., Jr., eds., Interaction of the Rocky Mountain foreland and the Bartels, W.S., 2001, Linking the Wasatchian/Bridger- nance of some of the Green River Formation Cordilleran thrust belt: Geological Society of America ian boundary to the Cenozoic global climate optimum: tuff beds in the Challis Volcanic Group or oth- Memoir 171, p. 465±487. New magnetostratigraphic and isotopic results from Berggren, W.A., and Aubrey, M.-P., 1996, A late South Pass, Wyoming: Palaeogeography, Palaeocli- er Eocene volcanic ®elds remains a Paleocene±early Eocene northwest Europe and matology, Palaeoecology, v. 167, p. 175±199. possibility. North Sea magnetobiochronological correlation net- Clyde, W.C., Stamatakos, J., and Gingerich, P.D., 1994, 5. The Green River Formation was depos- work, in Knox, R.W.O'B., Cor®eld, R.M., and Dunay, Chronology of the Wasatchian land-mammal age (ear- R.E., eds., Correlation of the early Paleogene in north- ly Eocene): Magnetostratigraphic results from the ited between ca. 53.5 Ma and 48.5 Ma and west Europe: Geological Society [London] Special McCullough Peaks section, northern Bighorn Basin, therefore spans the entire early Eocene cli- Publication 101, p. 309±352. Wyoming: Journal of Geology, v. 102, p. 367±377. Clyde, W.C., Zonneveld, J-P., Stamatakos, J., Gunnell, G.F., matic optimum (EECO) de®ned by global ma- Berggren, W.A., and Aubrey, M.-P., 1998, The Paleocene/ Eocene Epoch/Series boundary: Chronostratigraphic and Bartels, W.S., 1997, Magnetostratigraphy across rine O isotope records. The most evaporative framework and estimated geochronology, in Aubry, the Wasatchian/Bridgerian NAMLA boundary (early phase of the Green River Formation, i.e., the M.-P., Lucas, S., and Berggren, W.A., eds., Late to middle Eocene) in the western Green River Basin, Paleocene±early Eocene climatic and biotic events in Wyoming: Journal of Geology, v. 105, p. 657±669. Wilkins Peak Member, coincides with the last the marine and terrestrial records: New York, Colum- Crowley, K.D., Duchon, C.E., and Rhi, J., 1986, Climate half of this period of high global temperatures. bia University Press, p. 18±36. record in varved sediments of the Green River For-

Geological Society of America Bulletin, May 2003 563 SMITH et al.

mation: Journal of Geophysical Research, v. 91, Goodwin, J.H., 1973, Analcime and K-feldspar in tuffs of Lipman, P.W., 2000, Central San Juan caldera cluster: Re- p. 8637±8647. the Green River Formation, Wyoming: American gional volcanic fromework, in Bethke, P.M., and Hay, Culbertson, W.C., 1961, Stratigraphy of the Wilkins Peak Mineralogist, v. 58, 93±105. R.L., eds., Ancient Lake Creede: Its volcano-tectonic Member of the Green River Formation, Firehole Basin Goodwin, J.H., and Surdam, R.C., 1967, Zeolitization of setting, history of sedimentation, and relation to min- Quadrangle, Wyoming: U.S. Geological Survey Pro- tuffaceous rocks of the Green River Formation, Wy- eralization in the Creede mining district: Geological fessional Paper 424D, p. 170±173. oming: Science, v. 157, p. 307±308. Society of America Special Paper 346, p. 9±69. Culbertson, W.C., 1998, Road log for the geology ®eld trip Greenwood, D.R., and Wing, S.L., 1995, Eocene continen- Love, J.D., and Christiansen, A.C., 1985, Geologic map of of June 13, 1997, Geology and outcrops of the trona- tal climates and latitudinal temperature gradients: Ge- Wyoming: U.S. Geological Survey, scale 1:500,000. bearing rocks of the Green River Formation: Wyo- ology, v. 23, p. 1044±1048. Lucas, S.G., 1998, Fossil mammals and the Paleocene/Eo- ming State Geological Survey Public Information Cir- Harlan, S.S., Snee, L.W., and Geissman, J.W., 1996, 40Ar/ cene series boundaries in Europe, North America, and cular 40, p. 205±211. 39Ar geochronology and paleomagnetism of Indepen- Asia, in Aubry, M.-P., Lucas, S., and Berggren, W.A., Dalrymple, G.B., 1979, Critical tables for conversion of K- dence volcano, Absaroka Volcanic Supergroup, Bear- eds., Late Paleocene±early Eocene climatic and biotic Ar ages from old to new constants: Geology, v. 7, tooth Mountains, Montana: Canadian Journal of Earth events in the marine and terrestrial records: New York, p. 558±560. Sciences, v. 33, p. 1648±1654. Columbia University Press, p. 451±500. DeCelles, P.G., 1994, Late Cretaceous±Paleocene synoro- Hay, R.L., 1966, Zeolites and zeolitic reactions in sedi- Marvin, R.F., Hearn, B.C., Jr., Menhert, H.H., Naeser, genic sedimentation and kinematic history of the Se- mentary rocks: Geological Society of America Special C.W., Zartman, R.E., and Lindsey, D.A., 1980, vier thrust belt, northeast Utah and southwest Wyo- Paper 85, 130 p. Late Cretaceous±Paleocene±Eocene igneous activi- ming: Geological Society of America Bulletin, v. 106, Hiza, M.M., 1999, The geochemistry and geochronology of ty in north-central Montana: Isochron/West, v. 29, p. 32±56. the Eocene Absaroka volcanic province, northern Wy- p. 5±25. Deino, A., and Potts, R., 1990, Single-crystal 40Ar/39Ar dat- oming and southern Montana, USA [Ph.D. thesis]: Markwick, P.J., 1994, ``Equability,'' continentality, and Ter- ing of the Olorgesailie Formation, southern Kenya Corvallis, Oregon State University, 243 p. tiary ``climate'': The crocodolian perspective: Geolo- Rift: Journal of Geophysical Research, v. 95, Holroyd, P.A., and Smith, K.T., 2000, Preliminary biostrati- gy, v. 22, p. 613±616. 8453±8470. graphic evidence for age of the Wasatch and Green Matthews, M.D., and Perlmutter, M.A., 1994, Global cy- Desborough, G.A., 1975, Authigenic albite and potassium River Formations, Washakie Basin, southwestern Wy- clostratigraphy: An application to the Eocene Green feldspar in the Green River Formation, Colorado and oming: Geological Society of America Abstracts with River Basin, in de-Boer, P.L., Smith, D.G., eds., Or- Wyoming: American Mineralogist, v. 60, p. 235±239. Programs, v. 32, no. 7, p. 498. bital forcing and cyclic sequences: International As- Dott, R.H., Jr., 1983, Episodic sedimentation: How normal Honey, J.G., 1988, A mammalian fauna from the base of sociation of Sedimentologists Special Publication 19, is average? How rare is rare? Does it matter?: Journal the Eocene Cathedral Bluffs Tongue of the Wasatch p. 459±481. of Sedimentary Petrology, v. 53, p. 5±23. Formation, Cottonwood Creek area, southeast Wa- Mauger, R.L., 1977, K-Ar ages of biotites from tuffs in Duf®eld, W.A., and Dalrymple, G.B., 1990, The Taylor satch Basin, Wyoming: U.S. Geological Survey Bul- Eocene rocks of the Green River, Washakie, and Uinta Creek Rhyolite of New Mexico: A rapidly emplaced letin 1669C, p. 1±14. basins, Utah, Wyoming, and Colorado: University of ®eld of lava domes and ¯ows: Bulletin of Volcanol- Iijima, A., and Hay, R.L., 1968, Analcime composition in Wyoming Contributions to Geology, v. 15, no. 1, ogy, v. 52, p. 475±487. tuffs of the Green River Formation: American Min- p. 17±41. Elston, D.P., Lantos, M., and Hamor, T., 1994, High reso- eralogist, v. 53, p. 184±200. McCarroll, S.M., Flynn, J.J., and Turnbull, W.D., 1996, lution polarity records and the stratigraphic and mag- Isoplatov, V.O., Dudas, F.O., Snee, L.W., and Harlan, S.S., Biostratigraphy and magnetostratigraphy of the netostratigraphic correlation of late Miocene and Pli- 1996, Precise dating of the Lowland Creek volcanics, Bridgerian±Uintan Washakie Formation, Washakie ocene (Pannonian, s.l.) deposits of Hungary, in Teleki, west-central Montana: Geological Society of America Basin, Wyoming, in Prothero, D.R., and Emry, R.J., P.G., Mattick, R.E., and KoÂkai, J., eds., Basin analysis Abstracts with Programs, v. 28, no. 7, p. 484. eds., The terrestrial Eocene±Oligocene transition in and petroleum exploration: Dordrecht, Netherlands, Janecke, S.U., and Snee, L.W., 1993, Timing and episod- North America: Cambridge, Cambridge University Kluwer Academic Publishers, p. 111±142. icity of middle Eocene volcanism and onset of con- Press, p. 25±39. Ensley, R.A., and Verosub, K.L., 1982, A magnetostrati- glomerate deposition, Idaho: Journal of Geology, McDougall, I., and Roksandic, Z., 1974, Total fusion 40Ar/ graphic study of the sediments of the Ridge Basin, v. 101, p. 603±621. 39Ar ages using HIFAR reactor: Geological Society of southern California and its tectonic and sedimentolog- Karner, D.B., and Renne, P.R., 1998, 40Ar/39Ar geochronol- Australia Journal, v. 21, p. 81±89. ic implications: Earth and Planetary Science Letters, ogy of Roman province tephra in the Tiber River val- McGonigle, J.W., and Dalrymple, G.B., 1996, 40Ar/39Ar v. 59, p. 192±207. ley: Age calibration of middle Pleistocene sea level ages of some Challis volcanic group rocks and the Eugster, H.P., and Hardie, L.A., 1975, Sedimentation in an changes: Geological Society of America Bulletin, initiation of Tertiary sedimentary basins in southwest- ancient playa lake complex, the Wilkins Peak Member v. 110, p. 740±747. ern Montana: U.S. Geological Survey Bulletin 2132, of the Green River Formation of Wyoming: Geolog- KoÈniger, S., and Stollhofen, H., 2001, Environmental and p. 1±17. ical Society of America Bulletin, v. 86, p. 319±334. tectonic controls on preservation potential of distal McGrew, P.O., and Roehler, H.W., 1960, Correlation of Ter- Evanoff, E., Brand, L.R., and Murphey, P.C., 1998, Bridger fallout ashes on ¯uvio-lacustrine settings: The tiary units in southwestern Wyoming, in Overthrust Formation (middle Eocene) of southwest Wyoming: Carboniferous±Permian Saar-Nahe Basin, south-west belt of southwestern Wyoming and adjacent areas: Widespread marker units and subdivisions of Bridger Germany, in White, J.D.L., and Riggs, N.R., eds., Vol- Casper, Wyoming Geological Association Guidebook, B through D: Dakoterra, v. 5, p. 115±122. caniclastic sedimentation in lacustrine settings: Inter- 15th Annual Field Conference, p. 157±158. Evernden, J.F., Savage, D.E., Curtis, G.H., and James, G.T., national Association of Sedimentologists Special Pub- Min, K., Mundil, R., Renne, P.R., and Ludwig, K.R., 2000, 1964, Potassium-argon dates and the Cenozoic mam- lication 30, p. 263±284. A test for systematic errors in 40Ar/39Ar geochronology malian chronology of North America: American Jour- Koppers, A.A.P., 2002, ArArCALC-software for 40Ar/39Ar through comparison with U/Pb analysis of a 1.1-Ga nal of Science, v. 262, p. 145±198. age calculations: Computers and Geoscience, v. 28, rhyolite: Geochimica et Cosmochimica Acta, v. 64, Feeley, T.C., Lindsay, C.R., and Cosca, M.A., 1999, Timing p. 605±619. p. 73±98. and compositional evolution of magmatism at the Sun- Krishtalka, L., West, R.M., Black, C.C., Dawson, M.R., Morris, W.J., 1954, An Eocene fauna from the Cathedral light volcano, Absaroka volcanic province, Wyoming: Flynn, J.J., Turnbull, W.D., Stucky, R.K., McKenna, Bluffs Tongue of the Washakie Basin: Journal of Pa- Geological Society of America Abstracts with Pro- M.C., Bown, T.M., Golz, D.J., and Lillegraven, J.A., leontology, v. 28, p. 195±203. grams, v. 31, no. 7, p. 165. 1987, Eocene (Wasatchian through Duchesnean) bio- Murphey, P.C., Lester, A., Bohor, B., Robinson, P., Evanoff, Fischer, A.G., and Roberts, L.T., 1991, Cyclicity in the chronology of North America, in Woodburne, M.O., E., and Larson, E., 1999, 40Ar/39Ar dating of volcanic Green River Formation (lacustrine Eocene) of Wyo- ed., Cenozoic mammals of North America, Geochro- ash deposits in the Bridger Formation (middle Eo- ming: Journal of Sedimentary Petrology, v. 61, nology and biostratigraphy: Berkeley, University of cene), southwestern Wyoming: Geological Society of p. 1146±1154. California Press, p. 77±117. America Abstracts with Programs, v. 31, no. 7, p. 233. Flynn, J.J., 1986, Correlation and geochronology of middle Krogh, T.E., 1982, Improved accuracy of U-Pb zircon ages Obradovich, J.D., and Hicks, J.F., 1999, A review of the Eocene strata from the western United States: Palaeo- by the creation of more concordant systems using an isotopic calibration points for the geomagnetic polarity geography, Palaeoclimatology, Palaeoecology, v. 55, air abrasion technique: Geochimica et Cosmochimica time scale, in the interval 83±33 Ma (C34n to C13n): p. 335±406. Acta, v. 46, p. 637±649. Geological Society of America Abstracts with Pro- Flynn, J.J., and Tauxe, L., 1998, Magnetostratigraphy of Ku, T.-L., Luo, S., Lowenstein, T.K., Li, J., and Spencer, grams, v. 31, no. 7, p. 71. upper Paleocene±lower Eocene marine and terrestrial R.J., 1998, U-series chronology of lacustrine deposits Olsen, P.E., 1986, A 40-million-year lake record of early sequences, in Aubry, M.-P., Lucas, S., and Berggren, in Death Valley, California: Quaternary Research, Mesozoic orbital climatic forcing: Science, v. 234, W.A., eds., Late Paleocene±early Eocene climatic and v. 50, p. 261±275. p. 842±848. biotic events in the marine and terrestrial records: New Lanphere, M.A., and Baadsgaard, H., 2001, Precise K-Ar, O'Neill, W.A., 1980, 40Ar/39Ar ages of selected tuff of the York, Columbia University Press, p. 67±90. 40Ar/39Ar, Rb-Sr and U/Pb mineral ages from the 27.5 Green River Formation: Wyoming, Colorado, and Gazin, C.L., 1965, Early Eocene mammalian faunas and Ma Fish Canyon Tuff reference standard: Chemical Utah [M.Sc. thesis]: Columbus, Ohio State University, their environment in the vicinity of the Rock Springs Geology, v. 175, p. 653±691. 142 p. uplift, Wyoming, in Sedimentation of Late Cretaceous Lanphere, M.A., and Dalrymple, G.B., 2000, First- Palmer, B.A., and Shawkey, E.P., 1997, Lacustrine sedi- and Tertiary outcrops, Rock Springs uplift: Casper, principles calibration of 38Ar tracers: Implications for mentation processes and patterns during effusive and Wyoming Geological Association Guidebook, 19th ages of 40Ar/39Ar ¯uence monitors: U.S. Geological explosive volcanism, Challis volcanic ®eld, Idaho: Annual Field Conference, p. 171±180. Survey Professional Paper 1621, 10 p. Journal of Sedimentary Research, v. 67, p. 750±774.

564 Geological Society of America Bulletin, May 2003 40Ar/39Ar GEOCHRONOLOGY OF THE EOCENE GREEN RIVER FORMATION, WYOMING

Pietras, J.T., Carroll, A.R., and Rhodes, M.K., 2003, Tec- its application to young volcanic rocks: Geochimica cene Lakes Gosiute and Uinta of Wyoming, Utah, and tonic control on lacustrine evaporite deposition in the et Cosmochimica Acta, v. 65, p. 2571±2587. Colorado: Geology, v. 8, p. 135±139. Eocene Green River Formation, Wyoming: Journal of Sewall, J.O., Sloan, L.C., Huber, M., and Wing, S., 2000, Swisher, C.C., III, and Knox, R.W.O., 1991, The age of the Paleolimnology (in press). Climate sensitivity to changes in land surface char- Paleocene/Eocene boundary: 40Ar/39Ar daring of the Prothero, D.R., 1996a, Magnetic stratigraphy and biostra- acteristics: Global and Planetary Change, v. 26, lower part of NP10, North Sea Basin and Denmark, tigraphy of the middle Eocene Uinta Formation, Uinta p. 445±465. in IGCP Project 308 (Paleocene/Eocene boundary basin, Utah, in Prothero, D.R., and Emry, R.J., eds., Shackleton, N.J., 1986, Paleogene stable isotope events: Pa- events): International Geological Correlation Pro- The terrestrial Eocene±Oligocene transition in North laeogeography, Palaeoclimatology, Palaeoecology, gramme, International Annual Meeting and Field Con- America: Cambridge, Cambridge University Press, v. 57, p. 91±102. ference, Brussels, 2±6 December, 1991: Paris, Ab- p. 3±22. Sheppard, R.A., and Gude, A.J., 1969, Diagenesis of tuffs stracts with Program, p. 16. Prothero, D.R., 1996b, Summary, in Prothero, D.R., and in the Barstow Formation, Mud Hills, San Bernardino Tauxe, L., Gee, J., Gallet, Y., Pick, T., and Bown, T., 1994, Magnetostratigraphy of the Willwood Formation, Big- Emry, R.J., eds., The terrestrial Eocene±Oligocene County, California: U.S. Geological Survey Profes- horn Basin, Wyoming: New constraints on the loca- transition in North America: Cambridge, Cambridge sional Paper 634, 35 p. tion of Paleocene/Eocene boundary: Earth and Plan- University Press, p. 664±683. Sheriff, S.D., and Shive, P.N., 1982, Unreliable paleomagnetic results from the Wilkins Peak Member of the etary Science Letters, v. 125, p. 159±172. Prothero, D.R., and Swisher, C.C., III, 1992, Magnetostra- Taylor, J.R., 1982, An introduction to error analysis: Mill Eocene Green River Formation: Geophysical Re- tigraphy and geochronology of the terrestrial Eocene± Valley, California, University Science Books, 270 p. search Letters, v. 9, p. 723±726. Oligocene transition in North America, in Prothero, Ton-That, T., Singer, B., and Paterne, M., 2001, 40Ar/39Ar D.R., and Berggren, W.A., eds., Eocene±Oligocene Singer, B., and Brown, L.L., 2002, The Santa Rosa Event: 40 39 dating of latest Pleistocene (41 ka) marine tephra in climatic and biotic evolution: Princeton, New Jersey, Ar/ Ar and paleomagnetic results from the Valles the Mediterranean Sea: Implications for global climate Princeton University Press, p. 46±73. rhyolite near Jaramillo Creek, Jemez Mountains, New records: Earth and Planetary Science Letters, v. 184, Ratterman, N.G., and Surdam, R.C., 1981, Zeolite mineral Mexico: Earth and Planetary Science Letters, v. 197, p. 645±658. reactions in a tuff in the Laney Member of the Green p. 51±64. Townsend, H.A., 1985, The paleomagnetism of sediments River Formation, Wyoming: Clays and Clay Minerals, Smedes, H.W., and Prostka, H.J., 1972, Stratigraphic frame- acquired from the Goban Spur on the Deep Sea Dril- v. 29, p. 365±377. work of the Absaroka Volcanic Supergroup in the Yel- ling Project Leg 80, in de Graciansky, P.C., Poag, Reed, W.E., and Oertel, G., 1978, Is compaction a factor lowstone National Park region: U.S. Geological Sur- C.W., et al., eds., Initial reports of the Deep Sea Dril- in organic diagenesis?: Geological Society of America vey Professional Paper 729-C, 33 p. ling Project, Volume 80: Washington, D.C., Govern- Bulletin, v. 89, p. 658±662. Smith, M.E., Pietras, J.T., Carroll, A.R., and Singer, B., ment Printing Of®ce, p. 389±410. 40 39 Renne, P.R., Deino, A.L., Walter, R.C., Turrin, B.D., Swish- 2001, Laser fusion Ar/ Ar geochronology and cy- Van Houten, F.B., 1962, Cyclic lacustrine sedimentation er, C.C., Becker, T.A., Curtiss, G.H., and Jaouni, A.- clostratigraphy of the Green River Formation: Testing and the origin of analcime-rich Upper Triassic Lock- R., 1994, Intercalibration of astronomical and radio- micro- and meso-scale climate forcing: Eos (Trans- atong Formation, west-central New Jersey and adja- isotopic time: Geology, v. 22, p. 783±786. actions, American Geophysical Union), v. 82, cent Pennsylvania: American Journal of Science, v. 260, p. 561±576. Renne, P.R., Swisher, C.C., Deino, A.L., Karner, D.B., p. F1141. Villeneuve, M., Sandeman, H., and Davis, W.J., 2000, A Owens, T.L., and DePaolo, D.J., 1998, Intercalibration Smoot, J.P., 1983, Depositional subenvironments in an arid closed basin: The Wilkins Peak Member of the Green method for intercalibration of U-Th-Pb and 40Ar/39Ar of standards, absolute ages and uncertainties in 40Ar/ River Formation (Eocene), Wyoming, U.S.A.: Sedi- ages in the Phanerozoic: Geochimica et Cosmochim- 39Ar dating: Chemical Geology, v. 145, p. 117±152. mentology, v. 30, p. 801±827. ica Acta, v. 64, p. 4017±4030. Rhodes, M.R., Carroll, A.R., Pietras, J.T., Beard, B.L., and Snee, L.W., and Sutter, J.S., 1982, Geochronological data West, R.M., and Dawson, M.R., 1973, Fossil mammals Johnson, C.M., 2002, Strontium isotopic record of pa- for Pioneer Mountains, southwestern Montana: Part I. from the upper part of the Cathedral Bluffs Tongue of leohydrology and continental weathering, Eocene 40Ar/39Ar age-spectrum and conventional K/Ar dates the Wasatch Formation (early Bridgerian), northern Green River Formation, Wyoming: Geology, v. 30, for unaltered plutons: U.S. Geological Survey Open- Green River Basin, Wyoming: University of Wyoming p. 167±170. File Report OF82±961, 20 p. Contributions to Geology, v. 12, no. 1, p. 33±41. Ripepe, M., Roberts, L.T., and Fischer, A.G., 1991, Enso Wilf, P., 2000, Late Paleocene±early Eocene climate chang- Snider, L.G., and Moye, F.J., 1989, Regional stratigraphy, and sunspot cycles in varved Eocene oil shales from es in southwestern Wyoming: Paleobotanical analysis: physical volcanology, and geochemistry of the south- image analysis: Journal of Sedimentary Petrology, Geological Society of America Bulletin, v. 112, eastern Challis volcanic ®eld: U.S. Geological Survey v. 61, p. 1155±1163. p. 292±307. Open File Report OF89±639, p. 122±127. Wilf, P., Wing, S.L., Greenwood, D.R., and Greenwood, Roehler, H.W., 1989, Correlation of surface sections of the Sparks, R.S.J., Bursik, M.I., Carey, S.N., Gilbert, J.S., intertongued Eocene Wasatch and Green River For- C.L., 1998, Using fossil leaves as paleoprecipitation Glaze, L.S., Sigurdsson, H., and Woods, A.W., 1997, indicators: An Eocene example: Geology, v. 26, mations along the western margin of the Greater Volcanic plumes: New York, John Wiley and Sons, Green River Basin in southwest Wyoming: U.S. Geo- p. 203±206. 574 p. Wing, S.L., Bao, H., and Koch, P.L., 2000, An early Eocene logical Survey Miscellaneous Field Studies Map MF- Steidtmann, J.R., and Middleton, L.T., 1991, Fault chro- 2103. cool period? Evidence for continental cooling during nology and uplift history of the southern Wind River the warmest part of the Cenezoic, in Huber, B.T., Mac- Roehler, H.W., 1991, Identi®cation of oil shale and trona Range, Wyoming: Implications for Laramide and leod, K.G., and Wing, S.L., eds., Warm climates in beds and their geophysical-log responses in Energy post-Laramide deformation in the Rocky Mountain Earth history: Cambridge, Cambridge University Research and Development Administration Black foreland: Geological Society of America Bulletin, Press, p. 197±237. Fork No. 1 corehole, Eocene Green River Formation, v. 103, p. 472±485. Wing, S.L., Bown, T.M., and Obradovich, J.D., 1991, Early southwest Wyoming: U.S. Geological Survey Miscel- Steiger, R.H., and JaÈger, E., 1977, Subcommission on geo- Eocene biotic and climatic change in interior western laneous Field Studies Map MF-2158. chronology: Convention on the use of decay constants North America: Geology, v. 19, p. 1189±1192. Roehler, H.W., 1992a, Description and correlation of Eo- in geo- and cosmochronology: Earth and Planetary Wood, H.E., Chaney, R.W., Clark, J., Colbert, E.H., Jepson, cene rocks in stratigraphic reference sections for the Science Letters, v. 36, p. 359±362. G.L., Reedside, J.B., and Stock, C., 1941, Nomencla- Green River and Washakie basins, southwest Wyo- Stucky, R., 1990, Evolution of land mammal diversity in ture and correlation of the North American continental ming: U.S. Geological Survey Professional Paper North America during the Cenozoic: Current Mam- Tertiary: Geological Society of America Bulletin, 1506-D, 83 p. malogy, v. 2, p. 375±432. v. 52, p. 1±48. Roehler, H.W., 1992b, Correlation, composition, areal dis- Sullivan, R., 1985, Origin of lacustrine rocks of the Wilkins York, D., 1969, Least squares ®tting of a straight line with tribution, and thickness of Eocene stratigraphic units, Peak Member, Wyoming: American Association of correlated errors: Earth and Planetary Science Letters, greater Green River basin, Wyoming, Utah, and Col- Petroleum Geologists Bulletin, v. 69, p. 913±922. v. 5, p. 320±324. Zachos, J., Pagani, M., Sloan, L., Thomas, E., and Billups, orado: U.S. Geological Survey Professional Paper Sundell, K.A., 1993, A geologic overview of the Absaroka K., 2001, Trends, rhythms, and aberrations in global 1506-E, 49 p. volcanic province, in Snoke, A.W., Steidtmann, J.R., Roehler, H.W., 1993, Eocene climates, depositional envi- climate 65 Ma to present: Science, v. 292, and Roberts, S.M., eds., Geology of Wyoming: Lar- p. 685±693. ronments, and geography, greater Green River basin, amie, Wyoming, Geological Survey of Wyoming Zachos, J.C., Stott, L.D., and Lohmann, K.C., 1994, Evo- Wyoming, Utah, and Colorado: U.S. Geological Sur- Memoir 5, p. 480±506. lution of early Cenozoic marine temperatures: Paleo- vey Professional Paper 1506-F, 74 p. Surdam, K.A., and Parker, R.D., 1972, Authigenic alumi- ceanography, v. 9, p. 353±387. Sampson, S.D., and Alexander, E.C., Jr., 1987, Calibration nosilicate minerals in the tuffaceous rocks of the Zonneveld, J.-P., Gunnell, G.F., and Bartels, W.S., 2000, of the interlaboratory 40Ar/39Ar dating standard, Green River Formation, Wyoming: Geological Society Early Eocene fossil vertebrates from the southwestern MMhb-1: Chemical Geology, v. 66, p. 27±34. of America Bulletin, v. 83, p. 689±700. Green River Basin, Lincoln and Uinta Counties, Wy- Schmincke, H.-U., and van den Bogaard, P., 1991, Tephra Surdam, R.C., and Stanley, K.O., 1979, Lacustrine sedi- oming: Journal of Vertebrate Paleontology, v. 20, layers and tephra events, in Einsele, G., Ricken, W., mentation during the culminating phase of Eocene p. 369±386. and Seilacher, A., eds., Cycles and events in stratig- Lake Gosiute, Wyoming (Green River Formation): raphy: Berlin, Springer-Verlag, p. 392±429. Geological Society of America Bulletin, v. 90, MANUSCRIPT RECEIVED BY THE SOCIETY 5JUNE 2002 REVISED MANUSCRIPT RECEIVED 18 OCTOBER 2002 Schmitz, M.D., and Bowring, S.A., 2001, U-Pb zircon and p. 93±110. MANUSCRIPT ACCEPTED 25 OCTOBER 2002 titanite systematics of the Fish Canyon Tuff: An as- Surdam, R.C., and Stanley, K.O., 1980, Effects of changes sessment of high-precision U-Pb geochronology and in drainage-basin boundaries on sedimentation in Eo- Printed in the USA

Geological Society of America Bulletin, May 2003 565 Appendix DR-1. UW-Madison standard intercomparison UW-Madison Accepted Standard n MSWD Age (Ma) ± 2σ§ Age (Ma) ± 2σ†† GA1550 7* 98.69 ± 0.57# 0.48 [98.79 ± 0.54]§§ MMhb 8* 522.39 ± 2.98** 17.7 523.1 ± 2.6## SB-3 7† 165.72 ± 0.85** 5.21 [162.9 ± 1.8]*** FCs 17* 28.02 ± 0.14# 0.81 28.02 ± 0.16††† *single crystal laser fusions †multigrain fusions §intercalibration uncertainties relative to 28.34 ± 0.16 Ma for Taylor Creek Rhyolite sanidine (Duffield and Dalrymple, 1990; Renne et al., 1998) #weighted mean age **total fusion age, reported when MSWD >2 ††intercalibration uncertainties relative to 98.79 ± 0.54 Ma for GA1550 biotite (Renne et al., 1998) §§primary standard biotite (isotope dilution, McDougall and Roksandic, 1974; Renne et al., 1998) ##McClure Mountain syenite hornblende (Sampson and Alexander, 1987; Renne et al., 1998) ***USGS standard biotite (Lanphere and Dalrymple, 2000); note that the value reported here for SB-3 (relative to 28.34 ± 0.16 for TCs) is 1.73% older than the age reported by Lanphere and Dalrymple (2000); it is, however, consistent with the GA1550 age adopted by Renne et al. (1998), which is 1.85% older than the age determined by Lanphere and Dalrymple (2000) †††Fish Canyon tuff sanidine (Renne et al., 1998) Note: Values shown in brackets represent K-Ar ages for primary standards

Appendix DR-2: Positions of samples and standards in irradiations UW-06 and UW-11 with J values Appendix DR-3: Complete 40Ar/39Ar results Sample xtal 40Ar/39Ar 37Ar/39Ar 36Ar/39Ar40Ar* %40Ar* K/Ca Apparent Age ± Experiment # (10-16 mol) 2σ Ma Rife tuff BT-18 biotite J = 0.014710 ± 0.16% µ = 1.0025 UW11F2A 3 2.982 0.00038 0.003426 14.53 65.90 1131.1 51.43 ± 1.13 *UW11F2B 3 2.621 0.00058 0.002043 9.64 76.79 744.27 52.64 ± 0.53 UW11F2C 3 2.556 0.00026 0.001947 6.13 77.31 1627.9 51.69 ± 1.50 UW11F2D 3 2.516 0.00036 0.001874 9.34 77.80 1198.1 51.21 ± 0.70 *UW11F2E 3 2.553 0.00044 0.001789 8.43 79.12 968.54 52.83 ± 1.45 UW11F2F 3 2.488 0.00047 0.001880 16.00 77.48 919.93 50.45 ± 1.10 *UW11F2G 3 2.183 0.00036 0.000422 9.63 94.08 1207.1 53.69 ± 0.89 *UW11F2H 3 2.647 0.00045 0.002054 10.74 76.90 953.18 53.23 ± 1.01 *UW11F2I 3 2.607 0.00047 0.001889 11.25 78.41 914.30 53.45 ± 1.14 UW11F2J 3 2.922 0.00046 0.003290 12.72 66.57 935.29 50.89 ± 1.06 UW11F2K 3 2.974 0.00080 0.003516 11.49 64.91 538.46 50.52 ± 1.60 *UW11F2L 3 2.329 0.00038 0.001047 9.30 86.51 1129.2 52.69 ± 0.94 *UW11F2M 3 2.903 0.00071 0.002688 8.31 72.48 605.97 54.99 ± 1.97 UW11F2N 3 2.376 0.00062 0.001457 9.38 81.69 693.69 50.79 ± 1.42 UW11F2O 3 3.112 0.00072 0.003959 7.72 62.27 598.68 50.71 ± 2.02 UW11F2P 3 2.546 0.00071 0.001866 6.41 78.16 602.50 52.05 ± 2.19 UW11F2Q 3 2.359 0.00033 0.001301 17.33 83.51 1311.8 51.54 ± 0.50 UW11F2R 3 2.398 0.00074 0.001542 8.28 80.80 581.33 50.69 ± 1.57 *UW11F2S 3 2.577 0.00077 0.001866 9.54 78.43 562.00 52.86 ± 0.87 *UW11F2T 3 2.987 0.00038 0.003257 11.30 67.62 1142.0 52.81 ± 1.11 *UW11F2U 3 2.234 0.00034 0.000683 15.72 90.76 1282.3 53.01 ± 0.53 UW11F2V 3 3.605 0.00060 0.005500 6.41 54.79 711.96 51.68 ± 1.73 *UW11F2W 3 2.364 0.00030 0.001040 7.28 86.81 1425.3 53.66 ± 1.13 UW11F2X 3 2.409 0.00035 0.001440 5.72 82.15 1245.0 51.77 ± 1.51

Isochron age ± 2σ 51.41 ± 0.68 Total fusion age ± 2σ 52.10 ± 0.25 Weighted mean age ± 2σ 51.25 ± 0.31

Firehole tuff FC-2 sanidine J = 0.014629 ± 0.14% µ = 1.0025 UW11G12A 3 7.716 0.00014 0.019616 5.53 24.82 3049.7 49.85 ± 8.40 UW11G12B 12 6.734 0.00010 0.016055 22.72 29.48 4192.6 51.65 ± 2.21 UW11G12C 6 9.625 0.00010 0.025633 15.76 21.26 4304.8 53.20 ± 4.34 UW11G12D 6 5.042 0.00008 0.010362 10.20 39.17 5376.7 51.38 ± 2.24 UW11G12E 6 18.077 0.00024 0.054615 34.85 10.70 1827.9 50.32 ± 2.61 UW11G12F 6 5.270 0.00011 0.011430 8.62 35.83 3892.5 49.15 ± 1.81 UW11G12G 5 2.043 0.00010 0.000372 4.25 94.40 4444.6 50.20 ± 0.59 UW11G12H 15 2.020 0.00011 0.000230 15.77 96.40 3748.4 50.66 ± 0.49 UW11G12I 5 2.091 0.00015 0.000461 7.33 93.27 2869.7 50.74 ± 0.49 UW11G12J 5 2.120 0.00016 0.000459 3.82 93.39 2639.3 51.50 ± 1.00 UW11G12K 15 2.054 0.00010 0.000352 15.32 94.72 4495.2 50.63 ± 0.43 UW11G12L 15 2.041 0.00012 0.000309 19.77 95.30 3682.2 50.61 ± 0.37 UW11G12M 15 2.028 0.00011 0.000265 18.60 95.90 4032.3 50.60 ± 0.29 UW11G12N 20 2.076 0.00013 0.000405 21.47 94.01 3414.5 50.79 ± 0.19

Isochron age ± 2σ 50.70 ± 0.15 Total fusion age ± 2σ 50.76 ± 0.26 Weighted mean age ± 2σ 50.70 ± 0.14

C Bed tuff BT-18 biotite J = 0.014660 ± 0.16% µ = 1.0025 UW11F4A 3 2.311 0.00015 0.001252 5.47 83.79 2903.1 50.50 ± 0.52 UW11F4B 3 2.538 0.00067 0.002004 8.15 76.48 643.57 50.61 ± 1.27 *UW11F4C 3 2.621 0.00028 0.001946 5.83 77.89 1539.7 53.21 ± 1.38 UW11F4D 3 2.619 0.00104 0.002239 7.36 74.56 411.72 50.91 ± 0.72 UW11F4E 3 2.379 0.00060 0.001559 7.71 80.44 719.17 49.93 ± 0.72 Appendix DR-3: Complete 40Ar/39Ar results 40 39 37 39 36 39 40 Sample xtal Ar/ Ar Ar/ Ar Ar/ Ar40Ar* % Ar* K/Ca Apparent Age ± Experiment # (10-16 mol) 2σ Ma C Bed tuff cont... UW11F4F 3 2.443 0.00042 0.001644 5.51 79.92 1018.9 50.90 ± 1.32 UW11F4G 3 2.701 0.00030 0.002538 7.75 72.06 1429.2 50.76 ± 1.74 UW11F4H 3 4.944 0.00083 0.010075 12.74 39.69 518.63 51.17 ± 1.67 *UW11F4I 3 2.398 0.00031 0.001016 4.28 87.29 1372.1 54.53 ± 1.04 UW11F4J 3 2.489 0.00022 0.001822 5.53 78.19 1973.0 50.76 ± 0.95 UW11F4K 3 2.768 0.00065 0.002768 6.47 70.28 661.35 50.72 ± 0.88 UW11F4L 3 2.435 0.00053 0.001617 6.96 80.19 805.97 50.91 ± 1.53 *UW11F4M 3 2.609 0.00024 0.001767 10.00 79.80 1816.8 54.24 ± 1.30 UW11F4N 3 2.451 0.00015 0.001752 8.17 78.69 2828.6 50.31 ± 0.65 UW11F4O 3 2.364 0.00043 0.001441 5.97 81.80 992.72 50.44 ± 0.78 UW11F4P 3 2.636 0.00019 0.002198 6.64 75.18 2250.7 51.66 ± 2.19 UW11F4Q 3 2.530 0.00043 0.001956 4.67 76.98 1011.7 50.79 ± 1.77 *UW11F4R 3 2.542 0.00037 0.001671 4.83 80.40 1165.5 53.26 ± 1.82 *UW11F4S 3 2.627 0.00006 0.001234 4.07 85.95 7556.1 58.76 ± 0.53 UW11F4T 3 2.348 0.00065 0.001279 3.61 83.71 664.35 51.26 ± 2.11

Isochron age ± 2σ 50.35 ± 0.45 Total fusion age ± 2σ 51.60 ± 0.31 Weighted mean age ± 2σ 51.56 ± 0.26 Grey tuff WN-1 sanidine J = 0.02222 ± 0.24% µ = 1.0035 UW0612A 10 2.022 0.00257 0.000223 16.40 96.40 167.62 50.64 ± 0.59 UW0612B 20 2.074 0.00293 0.000415 34.45 93.77 146.80 50.53 ± 0.63 UW0612C 10 2.048 0.01507 0.000322 13.21 95.08 28.53 50.60 ± 0.44 UW0612D 10 2.036 0.00233 0.000310 15.34 95.17 184.91 50.36 ± 0.53 UW0612E 10 2.046 0.00264 0.000292 11.22 95.46 163.13 50.74 ± 0.65 UW0612F 10 2.013 0.00239 0.000195 12.04 96.80 180.13 50.64 ± 0.30 UW0612G 10 2.020 0.00259 0.000236 14.59 96.21 165.89 50.50 ± 0.80 UW0612H 10 2.147 0.00329 0.000654 10.31 90.69 130.59 50.58 ± 1.13 UW0612I 10 2.053 0.00259 0.000322 9.61 95.03 166.10 50.70 ± 0.90 UW0612J 10 1.995 0.00259 0.000200 13.08 96.69 165.88 50.13 ± 0.56

sanidine J = 0.01459 ± 0.20% µ = 1.0025 UW11G11A 10 2.058 0.00010 0.000388 9.79 94.20 4218.5 50.31 ± 0.67 UW11G11B 10 2.343 0.00020 0.001351 18.73 82.76 2131.2 50.34 ± 0.35 UW11G11C 10 2.109 0.00024 0.000559 15.40 91.95 1798.4 50.34 ± 0.36 UW11G11D 10 2.063 0.00011 0.000417 18.89 93.80 3942.9 50.23 ± 0.27 UW11G11E 10 2.084 0.00028 0.000470 13.38 93.12 1562.1 50.38 ± 0.22 UW11G11F 10 2.173 0.00021 0.000785 19.73 89.11 2055.5 50.27 ± 0.19 UW11G11G 10 2.115 0.00018 0.000555 15.58 92.02 2455.4 50.51 ± 0.31 UW11G11H 10 2.280 0.00029 0.001140 20.70 85.02 1457.9 50.32 ± 0.49

Isochron age ± 2σ 50.55 ± 0.21 Total fusion age ± 2σ 51.44 ± 0.17 Weighted mean age ± 2σ 51.39 ± 0.13 Main tuff TR-1 sanidine J = 0.02228 ± 0.24% µ = 1.0035 UW069A 10 1.946 0.00373 0.000049 32.78 99.14 115.27 49.86 ± 0.24 UW069B 10 1.951 0.00390 0.000065 24.84 98.90 110.24 49.85 ± 0.44 UW069C 10 1.950 0.00357 0.000049 64.41 99.14 120.53 49.95 ± 0.22 UW069D 10 1.954 0.00358 0.000054 63.75 99.07 120.07 50.01 ± 0.23 UW069E 10 1.962 0.00366 0.000076 89.97 98.74 117.40 50.05 ± 0.30 UW069F 10 1.945 0.00354 0.000043 33.28 99.23 121.53 49.87 ± 0.37 UW069G 10 1.966 0.00373 0.000085 38.65 98.61 115.38 50.10 ± 0.42 UW069H 10 1.987 0.00372 0.000175 99.57 97.29 115.65 49.94 ± 0.28 UW069I 10 1.974 0.00524 0.000135 23.31 97.88 82.00 49.92 ± 0.32 UW069J 10 1.977 0.00524 0.000135 23.35 97.88 82.00 50.01 ± 0.32 UW069K 10 1.953 0.00480 0.000032 25.02 99.41 89.50 50.17 ± 0.33 UW069L 10 1.993 0.00582 0.000191 21.30 97.07 73.93 49.98 ± 0.67 Appendix DR-3: Complete 40Ar/39Ar results Sample xtal 40Ar/39Ar 37Ar/39Ar 36Ar/39Ar40Ar* %40Ar* K/Ca Apparent Age ± Experiment # (10-16 mol) 2σ Ma Main tuff (sanidine) cont... UW069M 10 1.948 0.00466 0.000044 23.90 99.22 92.32 49.94 ± 0.45 UW069N 10 1.961 0.00490 0.000069 22.46 98.85 87.73 50.08 ± 0.45 UW069O 10 2.161 0.00516 0.000756 25.27 89.57 83.33 50.02 ± 0.43 UW069P 10 1.964 0.00486 0.000049 29.26 99.16 88.46 50.32 ± 0.55 UW069Q 10 1.958 0.00491 0.000083 29.82 98.63 87.62 49.90 ± 0.30 UW069R 10 1.958 0.00489 0.000051 9.49 99.13 87.99 50.15 ± 0.57 UW069S 10 1.955 0.00517 0.000054 27.94 99.08 83.14 50.05 ± 0.46 UW069T 10 1.971 0.00488 0.000114 34.63 98.19 88.11 50.01 ± 0.37 UW069U 10 1.957 0.00496 0.000040 25.77 99.28 86.67 50.20 ± 0.25

sanidine J = 0.014517 ± 0.12% µ = 1.0025 *UW11G6A 10 2.935 0.00008 0.002718 10.97 72.48 5353.9 54.87 ± 1.18 UW11G6B 10 4.115 0.00008 0.007426 12.87 46.56 5328.8 49.50 ± 1.93 UW11G6C 10 2.359 0.00012 0.001281 7.74 83.76 3597.6 51.01 ± 1.12 UW11G6D 10 2.118 0.00006 0.000581 6.27 91.68 6697.5 50.16 ± 0.64 UW11G6E 10 2.174 0.00007 0.000803 28.66 88.88 6080.0 49.92 ± 0.19 UW11G6F 10 1.975 0.00009 0.000124 37.52 97.90 4859.0 49.93 ± 0.10 UW11G6G 10 1.987 0.00007 0.000180 25.71 97.09 6114.6 49.83 ± 0.18 UW11G6H 10 2.065 0.00008 0.000420 25.60 93.76 5116.9 50.02 ± 0.18 UW11G6I 10 2.078 0.00012 0.000487 40.67 92.86 3494.4 49.85 ± 0.16 UW11G6J 10 2.071 0.00008 0.000439 53.47 93.50 5641.2 50.01 ± 0.17

Isochron age ± 2σ 49.98 ± 0.09 Total fusion age ± 2σ 50.03 ± 0.09 Weighted mean age ± 2σ 49.96 ± 0.08 Main tuff TR-1b biotite J = 0.014517 ± 0.12% µ = 1.0025 *UW11G6bA 3 2.019 0.00001 0.000166 6.74 97.34 78213 50.74 ± 0.97 *UW11G6bB 3 2.106 0.00002 0.000387 5.29 94.35 20892 51.31 ± 1.15 UW11G6bC 3 2.005 0.00000 0.000477 2.43 92.74 327094 48.05 ± 2.33 UW11G6bD 3 2.138 0.00003 0.000669 4.89 90.54 15936 49.99 ± 1.03 UW11G6bE 3 5.675 0.00003 0.012525 17.48 34.69 14956 50.84 ± 2.49 UW11G6bF 3 2.513 0.00005 0.002041 3.34 75.81 8503.6 49.21 ± 1.15 *UW11G6bG 3 2.306 0.00002 0.001132 29.65 85.30 26336 50.79 ± 0.35 *UW11G6bH 3 2.059 0.00001 0.000337 32.99 94.93 37218 50.48 ± 0.53 UW11G6bI 3 2.050 0.00001 0.000405 14.38 93.94 29751 49.74 ± 0.66 UW11G6bJ 3 2.083 0.00002 0.000489 18.43 92.84 19527 49.95 ± 0.49 *UW11G6bK 3 2.042 0.00001 0.000279 27.37 95.74 54740 50.48 ± 0.23 *UW11G6bL 3 2.066 0.00002 0.000326 12.44 95.12 25457 50.74 ± 0.48 *UW11G6bM 3 2.075 0.00001 0.000343 6.72 94.89 33503 50.84 ± 0.54 *UW11G6bN 3 2.417 0.00001 0.001533 22.90 81.07 59608 50.60 ± 0.32 UW11G6bO 3 2.121 0.00001 0.000590 22.43 91.56 58057 50.15 ± 0.27 *UW11G6bP 3 2.054 0.00003 0.000244 14.09 96.26 16013 51.06 ± 0.36 UW11G6bQ 3 2.019 0.00001 0.000324 11.06 95.02 51079 49.57 ± 0.54 UW11G6bR 3 2.104 0.00001 0.000575 11.36 91.71 37217 49.83 ± 0.46 *UW11G6bS 3 2.044 0.00001 0.000209 16.73 96.75 55867 51.06 ± 0.46 UW11G6bT 3 2.143 0.00002 0.000696 12.76 90.19 21085 49.92 ± 0.60 *UW11G6bU 3 2.044 0.00002 0.000222 10.68 96.57 23618 50.96 ± 0.63 UW11G6bV 3 2.091 0.00002 0.000506 19.20 92.63 23824 50.02 ± 0.50 UW11G6bW 3 2.063 0.00001 0.000389 11.00 94.21 39885 50.20 ± 0.48 UW11G6bX 3 2.330 0.00002 0.001342 14.62 82.78 21105 49.83 ± 0.56 *UW11G6bY 3 2.099 0.00001 0.000429 12.65 93.74 34418 50.81 ± 0.41 *UW11G6bZ 3 2.044 0.00003 0.000219 17.12 96.60 16895 51.00 ± 0.42 Appendix DR-3: Complete 40Ar/39Ar results Sample xtal 40Ar/39Ar 37Ar/39Ar 36Ar/39Ar40Ar* %40Ar* K/Ca Apparent Age ± Experiment # (10-16 mol) 2σ Ma Main tuff (biotite) cont... *UW11G6bAA 3 2.196 0.00003 0.000413 22.17 94.23 17148 53.39 ± 0.43 UW11G6bBB 3 2.148 0.00002 0.000710 10.00 90.01 21194 49.94 ± 0.93 UW11G6bCC 3 2.102 0.00001 0.000519 19.00 92.49 36768 50.20 ± 0.45 *UW11G6bDD 3 2.260 0.00001 0.000948 15.64 87.40 40088 51.00 ± 0.60

Isochron age ± 2σ 49.96 ± 0.20 Total fusion age ± 2σ 50.56 ± 0.12 Weighted mean age ± 2σ 49.98 ± 0.15 6th tuff TR-5 sanidine J = 0.014660 ± 0.18% µ = 1.0025 *UW11F9A 3 2.612 0.00003 0.000470 3.15 94.50 13112 64.14 ± 1.71 UW11F9B 3 2.348 0.00021 0.001559 1.18 80.19 2014.1 49.12 ± 5.35 *UW11F9C 3 8.198 0.00067 0.005996 3.65 78.33 641.45 162.31 ± 4.61 *UW11F9D 3 5.055 0.00007 0.006971 0.89 59.16 6603.7 77.41 ± 12.48 *UW11F9E 3 3.065 0.00011 0.001312 1.91 87.20 3883.6 69.33 ± 3.28 *UW11F9F 3 4.580 0.00010 0.008006 6.11 48.25 4277.3 57.52 ± 5.00 UW11F9G 3 3.626 0.00003 0.007329 0.55 40.14 16657 38.09 ± 23.86 *UW11F9H 3 226.013 0.00209 0.680165 4.30 11.07 205.67 563.72 ± 199.51 *UW11F9I 3 6.057 0.00008 0.011029 4.72 46.12 5268.7 72.41 ± 7.26 UW11F9J 3 3.674 0.00009 0.006424 1.95 48.20 4937.2 46.24 ± 8.06 *UW11F9K 3 4.713 0.00008 0.006181 5.39 61.15 5661.0 74.66 ± 2.87 *UW11F9L 3 2.641 0.00022 0.000634 2.53 92.73 1998.8 63.63 ± 0.79 *UW11F9M 3 3.028 0.00008 0.001865 3.08 81.64 5062.2 64.22 ± 1.76 *UW11F9N 3 3.079 0.00004 0.001089 3.36 89.40 10413 71.37 ± 1.84 *UW11F9O 3 3.039 0.00005 0.000516 2.10 94.82 8653.7 74.64 ± 4.38 UW11F9P 3 7.254 0.00018 0.018332 3.45 25.26 2434.4 47.82 ± 7.59 *UW11F9Q 9 6.224 0.00013 0.013609 10.95 35.31 3420.1 57.20 ± 0.86 UW11F9R 3 2.005 0.00012 0.000487 1.42 92.59 3641.2 48.43 ± 3.06 UW11F9S 3 2.338 0.00012 0.001361 2.85 82.60 3470.7 50.36 ± 1.28 *UW11F9T 3 2.601 0.00004 0.000347 5.11 95.88 11262 64.79 ± 1.05 *UW11F9U 3 2.465 0.00014 0.000914 2.63 88.86 3152.1 57.01 ± 1.08 UW11F9V 3 2.018 0.00016 0.000380 1.85 94.20 2630.7 49.59 ± 3.30 UW11F9W 3 1.953 0.00011 0.000237 1.30 96.18 3974.8 49.00 ± 4.11 *UW11F9X 3 2.168 0.00019 0.000175 3.08 97.40 2300.4 55.01 ± 1.02 *UW11F9Y 12 10.275 0.00009 0.025539 21.51 26.51 5038.2 70.63 ± 3.19

Isochron age ± 2σ 49.98 ± 1.20 Total fusion age ± 2σ 63.63 ± 0.73 Weighted mean age ± 2σ 49.79 ± 1.04 6th tuff TR-5b biotite J = 0.014660 ± 0.16% µ = 1.0025 UW11F10A 3 1.976 0.00001 0.000232 20.41 96.29 33378 49.64 ± 0.28 UW11F10B 3 2.067 0.00000 0.000568 14.45 91.65 266802 49.43 ± 0.42 UW11F10C 3 2.037 0.00002 0.000431 15.49 93.51 17878 49.69 ± 0.53 UW11F10D 3 1.996 0.00001 0.000263 8.13 95.87 38949 49.90 ± 0.56 UW11F10E 3 2.048 0.00001 0.000500 16.29 92.56 30993 49.45 ± 0.43 *UW11F10F 3 2.097 0.00002 0.000587 20.86 91.50 18318 50.06 ± 0.38 UW11F10G 3 2.003 0.00001 0.000304 17.71 95.28 32463 49.78 ± 0.51 UW11F10H 3 1.993 0.00003 0.000261 12.38 95.89 16160 49.86 ± 0.55 UW11F10I 3 2.270 0.00003 0.001252 22.33 83.50 14653 49.44 ± 0.91 *UW11F10J 3 2.143 0.00001 0.000685 10.91 90.34 48214 50.49 ± 0.58 UW11F10K 3 1.993 0.00004 0.000269 16.29 95.78 10496 49.80 ± 0.40 *UW11F10L 3 1.981 0.00000 0.000153 10.08 97.48 566802 50.36 ± 0.74 UW11F10M 3 2.204 0.00001 0.001035 12.03 85.91 52325 49.39 ± 0.70 UW11F10N 3 1.952 0.00002 0.000134 18.42 97.74 19976 49.76 ± 0.27 UW11F10O 3 1.971 0.00049 0.000207 27.52 96.66 878.18 49.70 ± 0.26 UW11F10P 3 1.994 0.00002 0.000305 20.58 95.25 26486 49.55 ± 0.27 *UW11F10Q 3 2.038 0.00002 0.000411 22.51 93.82 22699 49.89 ± 0.23 UW11F10R 3 2.210 0.00002 0.001000 42.55 86.43 17420 49.83 ± 0.30 Appendix DR-3: Complete 40Ar/39Ar results Sample xtal 40Ar/39Ar 37Ar/39Ar 36Ar/39Ar40Ar* %40Ar* K/Ca Apparent Age ± Experiment # (10-16 mol) 2σ Ma 6th tuff (biotite) cont. UW11F10S 3 2.000 0.00001 0.000279 22.68 95.64 32272 49.90 ± 0.38 *UW11F10T 3 1.966 0.00001 0.000137 25.77 97.70 48494 50.11 ± 0.23 UW11F10U 3 2.034 0.00001 0.000424 21.41 93.60 82611 49.66 ± 0.25 UW11F10V 3 2.001 0.00001 0.000317 41.84 95.08 39290 49.63 ± 0.13 UW11F10W 3 2.053 0.00001 0.000492 36.74 92.69 38171 49.64 ± 0.27 *UW11F10X 3 2.034 0.00001 0.000350 23.97 94.69 46534 50.25 ± 0.53 UW11F10Y 3 2.490 0.00001 0.001994 25.94 76.15 38894 49.47 ± 0.59 UW11F10Z 3 2.139 0.00001 0.000770 19.60 89.15 35256 49.74 ± 0.34

Isochron age ± 2σ 49.69 ± 0.14 Total fusion age ± 2σ 49.77 ± 0.11 Weighted mean age ± 2σ 49.67 ± 0.10 Analcite tuff SB-1 sanidine J = 0.022235 ± 0.24% µ = 1.0035 UW0610A 10 1.959 0.02026 0.000223 44.21 96.59 21.22 49.02 ± 0.44 UW0610B 10 1.926 0.01585 0.000117 30.68 98.14 27.13 48.95 ± 0.36 UW0610C 10 1.918 0.01789 0.000080 24.47 98.71 24.04 49.04 ± 0.62 UW0610D 10 1.905 0.01568 0.000060 30.70 99.00 27.43 48.85 ± 0.39 UW0610E 10 1.964 0.01873 0.000209 44.21 96.80 22.96 49.25 ± 0.49 UW0610F 10 1.954 0.02542 0.000233 55.10 96.45 16.92 48.82 ± 0.43 UW0610G 10 1.928 0.01696 0.000128 48.98 97.97 25.35 48.94 ± 0.26 UW0610H 10 1.922 0.01826 0.000073 51.84 98.82 23.55 49.19 ± 0.38 UW0610I 10 1.943 0.02063 0.000128 37.46 98.01 20.85 49.34 ± 0.47 UW0610J 10 1.948 0.02321 0.000186 22.14 97.15 18.53 49.01 ± 0.49

sanidine J = 0.014549 ± 0.21% µ = 1.0025 UW11G10A 5 1.918 0.00026 0.000058 10.20 98.87 1636.1 49.09 ± 0.41 UW11G10B 5 1.913 0.00035 0.000149 5.64 97.45 1222.5 48.28 ± 1.30 UW11G10C 10 1.929 0.00029 0.000118 26.15 97.95 1501.2 48.92 ± 0.16 UW11G10D 10 1.915 0.00030 0.000095 24.15 98.29 1418.4 48.73 ± 0.30 UW11G10E 10 1.939 0.00032 0.000148 24.35 97.50 1331.0 48.94 ± 0.19 UW11G10F 10 1.917 0.00036 0.000070 17.81 98.68 1210.9 48.99 ± 0.38 UW11G10G 10 1.934 0.00083 0.000145 27.32 97.55 518.43 48.85 ± 0.26 UW11G10H 10 1.927 0.00029 0.000088 22.51 98.42 1458.0 49.11 ± 0.29 UW11G10I 15 1.961 0.00034 0.000239 32.68 96.17 1270.2 48.83 ± 0.19

Isochron age ± 2σ 49.00 ± 0.21 Total fusion age ± 2σ 48.99 ± 0.14 Weighted mean age ± 2σ 48.94 ± 0.12 * indicates analyses that have been excluded from age calculations Note: All ages calculated relative to 28.34 Ma for the Taylor Creek rhyolite sanidine (Renne at al., 1998); using the decay constants of Steiger and Jäger (1977); uncertainties reported at 2σ analytical precision. Corrected for 37Ar and 39Ar decay, half lives of 35.2 days and 269 years, respectively.

References cited in Data Repository: Duffield, W. A., and G. B. Dalrymple, 1990, The Taylor Creek Rhyolite of New Mexico: A rapidly emplaced field of lava domes and flows: Bulletin of Volcanology, v. 52, p. 475-487. Lanphere, M.A., and Dalrymple, G.B., 2000, First-principles calibration of 38Ar tracers: Implications for ages of 40Ar/39Ar fluence monitors: US Geological Survey Professional Paper 1621. McDougall, I., and Roksandic, Z., 1974, Total fusion 40Ar/39Ar ages using HIFAR reactor: Geological Society of Australia Journal, v. 21, p. 81-89. Renne, P.R., Swisher, C. C., Deino, A. L., Karner, D. B., Owens, T. L., and DePaolo, D.J., 1998, Intercalibration of standards, absolute ages and uncertainties in 40Ar/39Ar dating. Chemical Geology, v. 145, p. 117-152. Sampson, S.D., and Alexander, E.C.Jr., 1987, Calibration of the interlaboratory 40Ar/39Ar dating standard, MMhb- 1: Chemical Geology, v. 66, p. 27-34. Steiger, R.H., and Jäger, E., 1977, Subcommission on geochronology: convention on the use of decay constants in geo- and cosmochronology, Earth and Planetary Science Letters, v. 36, p. 359-362.