sign in sign up
<<
Home , Cascades (ecoregion), Cascade Range, Mount Adams (Washington), Phenocryst

Earth and Planetary Science Letters 277 (2009) 38–49

Contents lists available at ScienceDirect

Earth and Planetary Science Letters

journal homepage: www.elsevier.com/locate/epsl

Discriminating assimilants and decoupling deep- vs. shallow-level records at using 238U–230Th disequilibria and Os isotopes

Brian R. Jicha a,⁎, Clark M. Johnson a, Wes Hildreth b, Brian L. Beard a, Garret L. Hart c, Steven B. Shirey d, Brad S. Singer a a Department of and Geophysics, University of Wisconsin-Madison, 1215 West Dayton Street, Madison WI 53706, USA b U.S. Geological Survey, 345 Middlefield Road MS910, Menlo Park, CA 94025, USA c School of Earth and Environmental Sciences, State University, 1228 Webster Pullman, WA 99163-2812, USA d Department of Terrestrial Magnetism, Carnegie Institution of Washington, 5241 Broad Branch Road, N.W., Washington DC 20015, USA article info abstract

Article history: A suite of 23 basaltic to dacitic erupted over the last 350 kyr from the Mount Adams volcanic field has Received 25 March 2008 been analyzed for U–Th isotope compositions to evaluate the roles of mantle versus crustal components Received in revised form 17 September 2008 during genesis. All of the lavas have (230Th/238U) N1 and span a large range in (230Th/232Th) ratios, Accepted 26 September 2008 and most have higher (230Th/232Th) ratios than and . Several of the lavas contain Available online 22 November 2008 antecrysts ( of pre-existing material), yet internal U–Th isochrons from six of seven lavas are Editor: C.P. Jaupart indistinguishable from their eruption ages. This indicates a relatively brief period of time between crystal growth and eruption for most of the phenocrysts (, clinopyroxene, , magnetite) prior to Keywords: eruption. One isochron gave a crystallization age that is ~20–25 ka older than its corresponding eruptive age, Mount Adams and is interpreted to reflect mixing of older and juvenile crystals or a protracted period of magma storage in Cascade arc the crust. Much of the eruptive volume since 350 ka consists of lavas that have small to moderate 230Th excesses – U Th isotopes (2–16%), which are likely inherited from melting of a garnet-bearing intraplate (“OIB-like”) mantle source. Os isotopes Following melt generation and subsequent migration through the upper mantle, most Mt. Adams assimilation interacted with young, mafic lower crust, as indicated by 187Os/188Os ratios that are substantially more radiogenic than the mantle or those expected via mixing of subducted material and the mantle wedge. Moreover, Os–Th isotope variations suggest that unusually large 230Th excesses (25–48%) and high 187Os/188Os ratios in some peripheral lavas reflect assimilation of small degree partial melts of pre-Quaternary basement that had residual garnet or Al-rich clinopyroxene. Despite the isotopic evidence for lower crustal assimilation, these processes are not generally recorded in the erupted phenocrysts, indicating that the crystal record of the deep-level ‘cryptic’ processes has been decoupled from shallow-level crystallization. © 2008 Elsevier B.V. All rights reserved.

1. Introduction been attributed to a variety of processes, including modification of the mantle by melts and fluids (Reagan et al.,1994), flux-ingrowth melting Identifying the physical processes and timescales involved in the (Thomas et al., 2002), slab melting (Dossetto et al., 2003; Sigmarsson origin of mafic arc magmas and their subsequent evolution to more et al.,1998), or lower crustal melting of garnet-bearing and garnet-free silicic compositions is critical for understanding eruptive histories of protoliths (Bourdon et al., 2000; Garrison et al., 2006; Jicha et al., 2007). volcanoes located above zones and ultimately the growth The majority of published 230Th excesses in lavas come of continental crust. U-series nuclides have been used for decades to from the Cascade arc. provide a temporal link between volcanic output and magma The Cascade arc is unusual among convergent margins because of dynamics in the mantle and crust (see Turner et al., 2003 for a review). the warm thermal structure of the subduction zone, the large number In most active oceanic and continental arcs, U-series data have shown of volcanoes (N2000), and compositional diversity of its eruptive that the mantle wedge is metasomatized by U-rich fluids and/or melts products (Leeman et al., 2005; Hildreth, 2007). Numerous geochem- liberated from the downgoing slab and overlying sediments, producing ical and isotopic investigations have been undertaken to identify the lavas that have 238U excesses (e.g., Elliott et al.,1997). Some continental origin of the distinctive types of primitive magmas erupted through- arc lavas, however, often have significant 230Th excesses, which have out the arc (e.g., Leeman et al., 1990; Bacon et al., 1997; Borg et al., 1997). Re–Os isotope studies from the Lassen and Adams regions were aimed at understanding the behavior of Re and Os in the mantle and ⁎ Corresponding author. crust (Borg et al., 2000; Hart et al., 2002, 2003). Hart et al. (2003) E-mail address: [email protected] (B.R. Jicha). interpreted the radiogenic Os isotope compositions of Mount Adams

0012-821X/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.epsl.2008.09.035 B.R. Jicha et al. / Earth and Planetary Science Letters 277 (2009) 38–49 39

Fig. 1. Geologic map of the Mount Adams volcanic field showing the map units analyzed in this study. All other Quaternary units are shaded gray. Solid black triangles denote vent locations. Map modified from Hildreth and Fierstein (1995). Map unit designation is that of Hildreth and Fierstein (1995). 40 B.R. Jicha et al. / Earth and Planetary Science Letters 277 (2009) 38–49 lavas to reflect assimilation of young, mafic lower crust that is not Most of the more than 120 exposed and inferred vents of the Mount detectable using Sr, Nd, or, Pb isotopes. Adams volcanic field (Hildreth and Lanphere, 1994; Hildreth and In this contribution, we integrate the large geologic, chemical, Fierstein, 1995, 1997) lie within a 6 km wide belt of mafic magmatism isotopic, and geochronologic database of the Mount Adams volcanic that extends north–south through the volcanic field for over 50 km, field with new U–Th isotope measurements of lavas and their consti- and this zone may reflect a region where mantle-derived basaltic tuent , as well as new 40Ar/39Ar ages, to explore the timescales magmas are focused through the crust (Hildreth and Fierstein, 1995). of magmatic processes in the mantle wedge and the overlying litho- sphere over the last 350 kyr. Previous U–Th isotope studies of Cascade 3. Petrology of sample suite arc lavas primarily focused on constraining the timescales of crustal processes (e.g., fractional crystallization, mixing, wallrock assimilation) Twenty three samples from the Mount Adams volcanic field were (Bennett et al., 1982; Trimble et al., 1984; Newman et al., 1986; Volpe, analyzed, all of which erupted b342 ka, and include eight basalts, four 1992; Volpe and Hammond, 1991; Reagan et al., 2003) or re-examining basaltic–andesites, eight andesites, and three dacites (Fig. 2) that were 226Ra–230Th crystallization ages (Cooper and Reid, 2003). Our focus at also analyzed by Hart et al. (2003) and Jicha et al. (2009). All lavas have Mount Adams is understanding the ascent history of the magmas. been previously analyzed for major- and trace-element contents by These results are compared to other long-lived arc volcanoes for XRF and INAA methods, as described by Bacon and Druitt (1988) and which 40Ar/39Ar and U–Th mineral isochrons are available, such as those Baedecker (1987). Mount Adams basalts are olivine phyric, and most in the Aleutians (Jicha et al., 2005) and Southern Volcanic Zone of Chile contain plagioclase and clinopyroxene (Hildreth and Fierstein, 1995). (Jicha et al., 2007). Quartz, sanidine, and biotite are absent, and amphibole is present in only two lavas, both of which predate the inception of the 2. Mount Adams (Hildreth and Fierstein, 1995). Andesites and dacites have varying percentages of olivine, clinopyroxene, orthopyroxene, The Mount Adams volcanic field is located at 46.2° N in southern plagioclase, and Fe–Ti oxides. Modal percentages of phenocrysts can Washington, ~50 km east of Mount St. Helens and ~75 km SSE of be found in Hildreth and Fierstein (1995). pyroxene an-

Mount Rainier. It has produced the largest volume of Quaternary desite (56–62 wt.% SiO2; Hildreth and Fierstein, 1995) is the dominant eruptive material among Cascade arc stratovolcanoes in Washington lithology at Mount Adams, and samples of this type were taken and (315±84 km3)(Hildreth and Fierstein, 1997), and is only from geographically dispersed locations. surpassed in volume by in northern (Sherrod and Smith, 1990). Of the total eruptive volume, constitutes 4. Analytical methods 9–15%, and is primarily restricted to peripheral vents, basaltic and andesite account for 84–89%, and only 1–2% (Hildreth and 4.1. 40Ar/39Ar geochronology Lanphere, 1994). Mount Adams basalts have remarkable composi- tional diversity, including coeval alkalic, low-potassium tholeiitic, 40Ar/39Ar incremental heating experiments were undertaken on and calc-alkaline basalts. Their geographic distribution shows no eleven Mount Adams lavas to confirm the K–Ar ages of Hildreth and relation between composition and distance from the trench. The Lanphere (1994) and improve age uncertainties of selected samples compositional diversity of Mount Adams lavas has been interpreted to for which U–Th mineral isochrons were obtained, which is important reflect variable contributions from MORB- and intraplate-like mantle for discriminating between eruptive and crystallization ages. Holo- sources, modified slightly by a subducted sediment component crystalline groundmass separates were prepared from lava samples (Jicha et al., 2009). using standard magnetic and density sorting techniques. At the Detailed field mapping and K–Ar ages identified scattered pre- University of Wisconsin Rare Gas Geochronology Laboratory, ~100 mg Mount Adams vents as old as ~940 ka, and indicate that inception of groundmass packets were incrementally heated in a double-vacuum stratocone growth occurred at about 520 ka (Hildreth and Lanphere, resistance furnace attached to a 350 cm3 gas clean-up line. Fully 1994; Hildreth and Fierstein, 1995). From 450–120 ka, recurrent automated experiments consisted of 8–12 steps from 650–1300 °C; stratocone activity constructed an apron of andesite and dacite that each step included a two-minute increase to the desired temperature extends ~15 km in most directions from the present (Hildreth that was maintained for 15 min, and an additional 15 min for gas and Lanphere, 1994). Numerous peripheral basaltic eruptions occurred contemporaneously with the andesitic–dacitic central cone effusions (Hildreth and Lanphere, 1994). At ca. 120–100 ka, a surge in eruptive rate occurred as voluminous basaltic shields (unit: bqb), scoria cones (bph), and andesite–dacite fans (akc, aes, dcc) were emplaced (Hildreth and Lanphere, 1994)(Fig. 1). An apparent lull followed this period of elevated magmatic output and lasted for ~30 kyr. Post-lull activity commenced at ~70 ka with peripheral and focal eruptions of basalt and basaltic andesite (ala), followed by reemergence of andesitic lavas at ~56 ka. Between 40 and 10 ka, the last main pulse of central-vent activity produced 40–50 km3 of andesite (Hildreth and Fierstein, 1997). Postglacial volcanic activity (b15 ka) remained chemically diverse, and included eruptions that ranged from to high silica andesite up to 63 wt.% SiO2 (Hildreth and Fierstein,1997). Because of the 75.7 kyr half-life of 230Th, we focus here on the last 350 kyr of eruptive activity at Mount Adams (Fig. 1). Mount Adams lies upon Cenozoic accreted terranes, where crustal thickness is estimated to be ~35–45 km (Mooney and Weaver, 1989; Parsons et al., 1998). Oblique convergence of the in southern Washington has produced north-south trending structural Fig. 2. Total alkali-silica (TAS) diagram for Mount Adams lavas in this study. Root names features within the crust that have likely influenced vent locations. and fields on the diagram are from Le Bas et al., 1986. Data from Jicha et al., 2009. B.R. Jicha et al. / Earth and Planetary Science Letters 277 (2009) 38–49 41 cleanup. Isotopic measurements and data reduction followed the pro- and Faraday detectors (229Th and 232Th). The typical ion intensity of cedures of Singer et al. (2004). the 229Th signal for spiked samples was 7e− 14 amp. The tail of the Although single incremental heating experiments on three 232Th peak on 230Th was minimized by using a Wide Aperture samples yielded precise ages, most samples required replicate experi- Retarding Potential (WARP) filter. The average abundance sensitivity ments on subsamples to improve precision. Obtaining precise ages for was ~20 ppb. Whole-rock standards and samples were analyzed using

Mount Adams lavas proved to be difficult due to the relatively low K2O a standard-sample-standard technique. For Th isotope analyses, a concentrations, which produced low radiogenic 40Ar yields (Table 1). U500 (238U/235U=1.0003) solution was mixed with the Th analyte to Because the age plateaus of several samples are weakly saddle-shaped simultaneously determine instrumental mass fractionation during Th and the isochron regressions yield 40Ar/36Ar intercepts slightly higher isotope analyses. Daly–Faraday gain calibration was accomplished by than the atmospheric value of 295.5, we consider the isochron ages to measuring a standard of known Th isotope composition (IRMM-035), give the best estimate of the time elapsed since eruption (Table 1). which was also mixed with U500. The ratio of the mass bias cor- Reliance on isochrons, rather than apparent ages using plateau rected 232Th/230Th ratio to the true ratio is the Daly–Faraday gain, criteria, is important because isochrons take into account potential which is interpolated for each sample by bracketing standards. The non-atmospheric trapped components of excess argon that would isotopic composition of IRMM-035 that was used for Daly–Faraday otherwise bias the ages of young samples, particularly those that have gain calibration was independently confirmed relative to NBL 114 low percentages of radiogenic argon (Sharp and Renne, 2005; Singer (238U/235U=137.88) at the beginning of the analytical session and et al., 2004). All ages were calculated using the decay constants of yielded a 232Th/230Th=87,859±0.47% (all uncertainties reported as % Steiger and Jäger (1977). 2SD), which is analytically indistinguishable from the consensus value of Sims et al. (2008). 4.2. U–Th isotopes External precision, reproducibility, and accuracy of Th and U isotope measurements were evaluated through repeated analyses of spiked and Thorium concentrations in mineral phases and whole rock samples unspiked rock standards (ATHO, AGV-1, BCR-1) and thorium reference analyzed in this study were highly variable, ranging from 0.07 to solutions (IRMM-035 and IRMM-036), whose isotopic compositions 12.7 ppm, which required dissolution of different sample masses: span the range of the Mt. Adams samples. IRMM-035 yielded 232Th/ whole-rock (80–230 mg), glass/groundmass (100–210 mg), plagio- 230Th=87,867±0.49% (n=96), IRMM-036=325,771±0.70% (n=124), clase (200–900 mg), olivine (500–1000 mg), pyroxene (100–350 mg), AGV-1=199,587±0.74% (n=11), A-THO=182,008±0.52% (n=26), BCR- and magnetite (25–300 mg). All of the high-purity mineral separates 1=210,789±0.54% (n=16). The measured values for each of these stan- were pulverized in an agate mortar and pestle prior to spiking and dards are in agreement with consensus values (Sims et al., 2008). U and dissolution. Whole-rock powders and magnetite separates were Th concentrations and (238U/232Th) and (230Th/232Th) activity ratios for spiked with a mixed 235U–229Th tracer, and then dissolved using the rock standards are given in Table 2.

HF+HNO3, followed by HCl, in Savillex Teflon beakers at ~110 °C. Because of the large sample size required for analysis and their resis- 5. Results tance to high molarity acid attack, plagioclase and pyroxene separates 40 39 were dissolved in HF–HNO3–HCl in Parr® bombs at 180 °C. Upon The new Ar/ Ar ages are on average three times more precise dissolution and spike equilibration, U and Th were co-precipitated than the previously determined K–Ar ages on the same units. Such 40 39 with Fe(OH)3 using NH4OH, and separated intoTh and U fractions using precision is important for comparing eruptive ( Ar/ Ar) ages with BioRad AG1x8 200–400 mesh anion exchange resin on 1.2 ml Teflon phenocryst residence (U–Th) ages. Below we discuss in detail the columns (Asmerom and Edwards, 1995; Jicha et al., 2007). The Th cut results of the 40Ar/39Ar and U–Th geochronology. was further purified by a second pass through 700 μl columns. Isotopic measurements were done at UW-Madison using a Micro- 5.1. 40Ar/39Ar geochronology mass IsoProbe MC-ICP-MS. Solutions were aspirated using a 50 µl/min self-aspirating, concentric-flow nebulizer tip and an Aridus® deso- Twenty three incremental heating experiments on eleven samples lvating nebulizer system. U and Th fractions were analyzed separately. yielded well-defined age plateaus and isochrons indicating ages U isotope measurements were done using Faraday detectors, and Th between 7.8 and 341.8 ka (Table 1). 40Ar/39Ar isochron ages from nine isotope measurements were made using a combination of Daly (230Th) samples are indistinguishable from their published K–Ar ages, but are

Table 1 Summary of 40Ar/39Ar incremental heating experiments of Mt. Adams lavas and comparison to published K–Ar data

40Ar/39Ar data K–Ar datac 39 40 a4036 b c 40 Map unit No. of Plateau age % Ar % Ar⁎ Ar/ Ari MSWD Isochron age K–Ar age % Ar⁎ Sample nameExpts. (ka)±2σ ±2σ (ka)±2σ (ka)±2σ Andesite of Takh Takh atm 2 7.2±1.0 96.7 1.3 295.1±0.6 0.16 7.8±1.5 ~6.5d – Basalt of Smith bsb 2 38.5±8.7 95.1 1.2 298.0±2.3 1.31 13.1±7.5 14±26 0.6 Andesite of Big Spring Creek abs 1 41.3±3.5 100.0 5.6 298.4±2.4 0.38 33.5±7.0 28±12 3.4 Andesite of Pikers Peak app 2 47.8±5.8 100.0 1.3 296.8±0.9 0.46 31.6±10.8 33±28 1.7 Andesite of Morrison Creek amc 1 51.1±2.7 100.0 9.3 296.7±2.7 0.27 49.2±5.0 56±12 8.3 Basalt of Riley Creek brc 2 56.4±2.9 100.0 6.3 295.0±0.7 0.42 58.6±3.9 63±28 2.2 Basalt of Meadow Butte bmb 4 245.0±21.0 90.8 1.3 296.9±1.5 0.76 161.0±56.0 138±116 1.4 Basalt of King bkm 3 152.4±9.4 100.0 2.6 294.9±1.2 0.59 166.2±26.3 106±50 1.4 Basalt of Herions Bridge bhb 2 223.0±11.1 100.0 4.3 294.6±0.6 0.32 221.7±13.9 209±42 8.2 Basalt west of Draper Springs bdw 3 223.0±17.0 99.2 2.2 295.4±0.8 0.24 230.0±28.0 184±58 4.0 “andesite of Sled Camp” asc-W 1 341.9±5.3 100.0 29.5 295.5±1.7 0.75 341.8±7.2 –– a% 40Ar ⁎ was calculated from all steps comprising the plateau and was weighted by the fraction of 39Ar released. bPreferred age. cData from Hildreth and Lanphere (1994). dAge constrained by tephrostratigraphy of Hildreth and Fierstein (1997). 42 B.R. Jicha et al. / Earth and Planetary Science Letters 277 (2009) 38–49

Table 2 U–Th and Os isotope data from Mt. Adams lavas

238 232 230 232 230 232 a b Sample/ SiO2 Age Min. Max. Material ( U/ Th) ( Th/ Th) ( Th/ Th)0 Th U γOs Map unit (wt.%)(ka) Volume Volume (ppm) (ppm) (km3) (km3) Whole rock analyses aaa 59.2 ~4 0.08 0.14 wr 1.036±0.002 1.108±0.004 1.111±0.004 8.099 2.765 45 amf 59.3 3.8–7.7 0.28 0.38 wr 1.043±0.004 1.112±0.006 1.116±0.006 7.411 2.548 46 bsc 48.0 21.5–41.4 1.1 2.2 wr 1.019±0.002 1.114±0.004 1.145±0.013 0.750 0.252 53 brc 51.5 59±4 0.25 0.5 wr 0.985±0.006 1.077±0.009 1.143±0.014 3.736 1.212 32 ala 55.7 63±14 0.15 0.25 wr 0.852±0.002 0.974±0.005 1.069±0.028 7.907 2.222 30 aes 59.8 111±12 0.05 0.12 wr 1.021±0.003 1.056±0.004 1.117±0.014 8.857 2.980 34 bph 54.0 111±20 0.5 1 wr 0.970±0.001 1.079±0.005 1.269±0.055 4.632 1.481 75 bqb 50.8 115±56 0.7 1 wr 1.064±0.001 1.086±0.004 1.127±0.033 3.253 1.140 dcc 64.5 117±12 0.4 0.8 wr 1.042±0.004 1.050±0.006 1.067±0.014 11.249 3.862 akc 59.6 120±14 1.75 4 wr 1.025±0.003 1.061±0.004 1.132±0.017 7.997 2.702 58 bmb 50.6 161±56 1 1.5 wr 0.915±0.001 1.015±0.005 1.351±0.224 1.869 0.564 282 bkm 51.2 166±26 4.3 9 wr 0.889±0.002 0.906 ±0.003 0.965±0.021 2.902 0.850 44 bhb 50.5 222±14 0.4 1.5 wr 0.951±0.001 1.010±0.002 1.401±0.059 2.090 0.655 bdw 52.2 230±28 0.02 1 wr 1.051±0.003 1.068±0.006 1.186±0.053 1.521 0.527 60 dsc 63.2 246±8 0.04 0.08 wr 1.014±0.002 1.025±0.002 1.119±0.021 9.774 3.267 47 asc-W 58.8 342±7 0.01 0.03 wr 0.995±0.002 1.013±0.003 1.418±0.067 5.338 1.750 264

Data for mineral isochrones atm 59.6 7.8±1.5 0.37 0.5 wr 1.040±0.002 1.093±0.004 1.097±0.004 9.740 3.338 36 gm 1.034±0.004 1.095±0.006 12.740 4.340 mt 1.220±0.006 1.114±0.008 1.374 0.552 cpx 1.034±0.002 1.099±0.006 0.517 0.176 plag 1.078±0.008 1.102±0.014 0.148 0.052 bsb 49.1 13.1±7.5 0.2 0.3 wr 1.021±0.003 1.166±0.005 1.184±0.013 3.081 1.037 54 gm 0.985±0.004 1.161±0.006 3.306 1.073 mt 1 1.749±0.005 1.378±0.005 1.699 0.980 mt 2 1.640±0.011 1.344±0.013 1.795 0.970 bic 48.0 b22.1 2 4 wr 1.076±0.003 1.191±0.005 1.216±0.006 0.703 0.249 gmass 1.229±0.004 1.216±0.006 1.148 0.465 olivine 1.531±0.016 1.247±0.039 0.036 0.018 mt 1.171±0.011 1.212±0.013 1.594 0.615 abs 58.5 33.5±7.0 0.2 0.35 wr 1.046±0.002 1.101±0.003 1.120±0.006 7.783 2.682 43 mt 1.393±0.005 1.178±0.006 0.514 0.236 gmass 1.032±0.005 1.099±0.007 10.637 3.619 plag 0.993±0.004 1.090±0.007 0.125 0.045 app 55.0 31.6±10.8 35 45 wr 0.935±0.001 1.035±0.002 1.069±0.013 5.392 1.662 46 gmass 0.940±0.002 1.038±0.004 5.961 1.848 mt 0.846±0.002 1.012±0.004 1.488 0.415 cpx 0.830±0.010 1.004±0.021 0.157 0.057 plag 1.091±0.004 1.020±0.011 0.316 0.086 olivine 1.186±0.030 1.104±0.045 0.067 0.026 amc 58.3 49.2±5.0 2 3.5 wr 1.022±0.004 1.091±0.007 1.130±0.008 6.352 2.139 45 mt 1.143±0.003 1.148±0.005 0.568 0.214 cpx 1.038±0.006 1.098±0.009 0.301 0.103 gmass 1.030±0.003 1.090±0.005 9.108 3.093 plag 1.679±0.007 1.303±0.008 0.124 0.069 dbc 68.5 115±10 0.25 0.75 wr 0.995±0.002 1.035±0.005 1.123±0.015 7.201 2.361 31 mt 1.117±0.004 1.109±0.006 0.822 0.302 cpx 1.072±0.003 1.079±0.006 0.238 0.084 gmass 0.976±0.002 1.022±0.005 9.049 2.910 plag 1.138±0.001 1.040±0.004 0.071 0.026

Rock standards ATHO (n=26) wr 0.923±0.011 1.017±0.006 7.376 2.241 BCR-1 (n=16) wr 0.874±0.011 0.878±0.005 5.855 1.687 AGV-1 (n=11) wr 0.931±0.005 0.928±0.007 6.246 1.921

Abbreviations: wr, whole-rock; gm, groundmass; cpx, clinopyroxene; mt, magnetite; plag, plagioclase. (230Th/232Th) and (238U/232Th) uncertainties for samples are reported as internal 2 SE. Uncertainites for rock standards reported as 2SD. a 230 232 40 39 14 ( Th/ Th)0 ratios were calculated using the whole-rock values and the Ar/ Ar, K-Ar, or C age determinations. bOs isotope data from Hart et al. (2003).

more precise. Two experiments from the andesite of Takh Takh to sets S and J from Mount St. Helens, which have calibrated 14C ages of Meadow (unit atm), an undated postglacial lava, gave a weighted 13 and 15 ka respectively (Hildreth and Fierstein, 1997); therefore the mean isochron age of 7.8±1.5 ka (Fig. 3), which is in agreement with inferred eruptive age of ~14 ka is very similar to the 40Ar/39Ar isochron the 6.4–6.7 ka calendar age obtained for a set of phreatomagmatic age. Two experiments on the andesite of Pikers Peak (app) produced ashes believed to be correlative with this unit (Hildreth and Fierstein, a combined isochron that has a non-atmospheric intercept (40Ar/ 1997). One of the numerous thin flows of mildly alkalic olivine basalt 36Ar=296.8±0.9) and an age of 31.6±10.8 ka (Fig. 3). A small outcrop from unit bsb gave a 40Ar/39Ar isochron age of 13.1±7.5 ka (Table 1). of andesitic lava (asc-W) located between two lobes of the andesite of Unit bsb lies midway between thin layers of dacitic belonging Twin Falls Creek on the western flank of the was originally B.R. Jicha et al. / Earth and Planetary Science Letters 277 (2009) 38–49 43

Fig. 3. 40Ar/39Ar age plateau and inverse isochron diagrams for two Mount Adams lavas (atm and app) showing the various ages (±2σ errors) and spectra obtained from multiple incremental heating experiments. Isochron age is the preferred age of eruption.

mapped as the part of the b63 ka andesite of Sled Camp (asc)(Hildreth lavas yielded ages that are identical to their eruption ages within and Fierstein, 1995). However, this lava gave a 40Ar/39Ar isochron age analytical uncertainties (Table 2; Fig. 5). The U–Th mineral isochron of 341.8±7.2 ka and therefore is not part of map unit asc and is likely a age for bsb, however, is 20–25 kyr older than its eruptive age. The 36.8± remnant from the middle stratovolcano. 1.3 ka U–Th isochron age is defined by whole rock+groundmass+2 40Ar/39Ar dating of substantial shields (each N1km3) adjacent to separate analyses of magnetite (Fig. 6). This alkalic basalt also contains Mount Adams has resolved two discrete pulses of elevated rate at large (0.5–1.5 mm) olivine phenocrysts, but the 500 mg olivine sepa- ~160 and ~120–100 ka. Olivine-basalt flows on the surfaces of King rate prepared from this sample did not yield sufficient Th for isotopic Mountain (bkm) and Meadow Butte (bmb) shields yielded 40Ar/39Ar analysis. ages of 166.2±26.3 and 161.0±56.0 ka (Table 1), which is ~40–60 ka For a few samples, the whole-rock or groundmass datum lies at the older than the previous estimate of Hildreth and Lanphere (1994). left end of the isochron, and mass balance requires the presence of a These ages, coupled with the 159±31 ka K–Ar age for the Goat Butte phase that has a low U/Th ratio that was not analyzed but must have basaltic shield, suggest that all three shields may have erupted within been present in the whole rock. The plagioclase separates for app and a relatively short period of time at about 160 ka.

5.2. U–Th isotopes

Whole-rock, eruptive age-corrected (230Th/232Th) activity ratios for Mount Adams are quite variable (0.97–1.42), but fall within the range of those reported for other Cascade arc volcanoes (Bennett et al., 1982; Trimble et al., 1984; Newman et al., 1986; Volpe and Hammond, 1991; Volpe,1992; Reagan et al., 2003)(Fig. 4). All of the Mount Adams lavas have (230Th/238U) N1 (i.e., 230Th excess), a common characteristic among Cascade arc lavas, but found in only a few lavas from other continental arc volcanoes in Kamchatka (Dossetto et al., 2003; Turner et al., 1998), Nicaragua (Reagan et al., 1994; Thomas et al., 2002), Ecuador (Garrison et al., 2006), and Chile (Sigmarsson et al., 1998; Bourdon et al., 2000; Jicha et al., 2007). With the exception of sample bkm, Mount Adams basalts have higher (230Th/232Th) ratios than the andesites and dacites (Fig. 4, Table 2). Mineral and groundmass separates were analyzed for two basaltic shields adjacent to Mount Adams (bsb, bic), one basaltic andesite (app), three andesites (atm, abs, amc), and one dacite (dbc)(Table 2, Fig. 5). The measured U and Th concentrations of the mineral separates are consistent with published partition coefficient data 230 232 238 232 with the exception of olivine (Blundy and Wood, 2003), although the Fig. 4. ( Th/ Th)0 vs. ( U/ Th) equiline diagram showing whole-rock data from Mount Adams and published data from the . (230Th/232Th) ratios have been (238U/232Th) ratios were quite variable for each phase among the age-corrected to the time of eruption. Published U–Th isotope data from Cascade arc different samples (Fig. 5). The groundmass separates have the highest lavas are from Bennett et al., 1982; Trimble et al., 1984; Newman et al., 1986; Volpe and U and Th contents. Internal U–Th mineral isochrons from six of seven Hammond, 1991; Volpe, 1992; Reagan et al., 2003. 44 B.R. Jicha et al. / Earth and Planetary Science Letters 277 (2009) 38–49

Fig. 5. U–Th mineral isochrons for six Mount Adams lavas. All six internal isochron ages are statistically indistinguishable from their eruptive ages, which are constrained by 40Ar/39Ar, K–Ar, or 14C dating. U–Th mineral isochron ages and uncertainties were calculated using Isoplot version 3.50. The plagioclase separates for sample app and dbc and the magnetite separate for unit amc are denoted with an open ellipse because they are not considered when calculating the isochron ages. These separates likely contain xenocrysts of older material or inclusions of accessory phases. Abbreviations: plag, plagioclase; wr, whole rock; cpx, clinopyroxene; mt, magnetite; gm, groundmass. dbc fall below lines regressed through the other phases and whole at the same time as the phenocrysts. A similar conclusion is inferred rock compositions, suggesting that these separates may contain for the magnetite from amc, which lies on the equiline above the 43 ka xenocrysts or inclusions of accessory phases that did not crystallize isochron. The U–Th mineral isochron ages calculated for each of these

Fig. 6. U–Th mineral isochron and 40Ar/39Ar age inverse isochron diagrams for unit bsb (basalt of Smith Butte). U–Th mineral isochron age is 36.8±1.3 ka, which is ~20–25 kyr older than the 40Ar/39Ar age of 13.1±7.5 ka. B.R. Jicha et al. / Earth and Planetary Science Letters 277 (2009) 38–49 45 samples using the other minerals are consistent with eruptive ages determined by 40Ar/39Ar or K–Ar geochronology (Table 1).

6. Discussion

First we discuss the constraints imposed on phenocryst residence times by the combined 40Ar/39Ar and U–Th geochronology, including the uncertainties of such estimates. Second, we consider the unique contributions that Th and Os isotopes may provide on crustal interactions, which bear on crustal residence times of the magmas, relative to other isotopic tracers such as O, Sr, Nd, Hf, and Pb.

6.1. Mineral isochrons

The difference between crystallization ages defined by U–Th min- eral isochrons and the eruption ages derived from other approaches (e.g., 14C, K–Ar, 40Ar/39Ar methods) has been interpreted as the mini- mum residence time of the phenocrysts in the magma (Condomines et al., 2003). Recent analytical advances have facilitated the generation of 40Ar/39Ar and U–Th mineral isochron ages for latest Pleistocene to Holocene lavas and that have uncertainties of only a few thousand years (Jicha et al., 2005, 2007). Decreasing the uncertainties in 40Ar/39Ar and U–Th ages to several thousand years or less is critical to inferring meaningful phenocryst residence times in volcanic rocks that are older than historic-Holocene age. Combining high-precision eruption age geochronology with phenocryst age determinations (e.g., Jicha et al., 2005, 2007; Bourdon et al., 2000; Heumann et al., 2002) represents a significant advance in linking volcanic evolution to magmatic processes as compared to restricting U–Th geochronology to recent or historical eruptions. In Fig. 7A we summarize the U–Th isochron and eruption ages for Mt. Adams units in stratigraphic order, and in Fig. 7B we compare these ages directly with the range of phenocryst residence times that are permitted by the age uncertainties.With one exception, sample bsb, minerals (cpx, mt, plag, olivine), glass, and whole rocks measured in six Mount Adams lava flows erupted over the last ~115 ka define 238U–230Th isochrons whose age cannot be distinguished from that of the eruption (Fig. 7A,B). This finding is similar to observations at Seguam and Puyehue-Cordón Caulle volcanic complexes in the Aleutians (Jicha et al., 2005) and Andean Southern Volcanic Zone (Jicha et al., 2007), and implies that the observed phenocrysts resided Fig. 7. A) Comparison of U–Th mineral isochron ages with eruptive ages, in stratigraphic – in the magma for a short interval, less than a few kyrs, prior to order. B) U Th mineral isochron ages (ka) plotted against independently determined eruptive age (ka) for Mount Adams lavas. Dashed lines represent phenocryst residence eruption. In the Mount Adams suite, three samples (abs, amc, dbc) times in 10 kyr increments, for both negative and positive residence ages. have U–Th mineral isochron ages that are slightly younger than the eruptive ages (Fig. 7A,B), although they agree within analytical error, indicating that most of the phenocrysts likely formed immediately age which is well constrained to 14±1 ka by bracketing tephra layers prior to eruption. The major cone-building unit app has K–Ar and that are regionally well known and dated (Hildreth and Fierstein, 40Ar/39Ar ages that are identical to that of the U–Th mineral isochron 1997)(Fig. 7A,B). Possible causes of the long phenocryst residence age, although the relatively large measured errors in the K–Ar and time will be explored below. 40Ar/39Ar ages could permit phenocryst residence times up to ~10 kyr Our results show that, in general, crystal fractionation, which is (Fig. 7A,B). A ~10 kyr phenocryst residence time for unit app would likely the dominant mechanism for producing the spectrum of evolved require an eruptive age of ~20 ka for the overlying abs unit, which magma compositions at Mount Adams (Hildreth and Fierstein, 1995; conflicts with its K–Ar (28±6 ka) and 40Ar/39Ar (33.5±7 ka) ages, Jicha et al., 2009), must have occurred prior to growth of the erupted indicating that the most reasonable interpretation of the geochrono- crystals. The short crystal residence times are consistent with the logical data is that the phenocrysts in unit app had a very short relatively simple zoning patterns in the majority of the phenocrysts residence time prior to eruption. (Fig. 8A) and with inferences from trace-element zoning and diffusion Unit atm has a U–Th mineral isochron age of 11.9±5.8 ka, which profiles in phenocrysts from lavas erupted from numerous other arc lies within uncertainty of its eruptive age constrained by 40Ar/39Ar volcanoes (Zellmer et al., 1999; Costa et al., 2003; Morgan et al., 2004). geochronology (7.8±1.5 ka) and bounding ashes (~6.5 ka) from other Although models of major- and trace-element zoning in phenocrysts (Hildreth and Fierstein, 1997). It is, however, suggest very short timescales of crystal residence (tens to hundreds of possible that phenocryst residence times of ~4–5 kyr could be inferred years), these studies have tended to be restricted to only one lava or for unit atm within the uncertainties of the U–Th mineral isochron tephra, and therefore may not be applicable to all volcanic eruptive age. units. Our results complement these interpretations because we have The only unit at Mount Adams analyzed to date that contains clear illustrated that the short periods of crystal residence in magmas may evidence for significant phenocryst residence is unit bsb, where be characteristic over long lifetimes of a volcano, up to 120 kyr at crystal residence times of ~20 to ~25 kyr are permitted by the eruptive Mount Adams. 46 B.R. Jicha et al. / Earth and Planetary Science Letters 277 (2009) 38–49

analyzed that contains clear evidence for long phenocryst residence times is bsb, which is the most voluminous basalt that was erupted in the postglacial period at Mount Adams. Overlap between the U–Th mineral isochron age of 36.8±1.3 ka and the main eruptive pulse associated with the current Mount Adams cone may suggest the unit bsb incorporated phenocrysts from this earlier eruptive pulse. Hildreth and Fierstein (1997) documented unusually large ranges in chemical compositions of olivine and spinel phenocrysts in unit bsb, consistent with a protracted history. The presence of “older” magnetite in bsb,as well as amc (Fig. 5), suggests that recycling of this early liquidus phase in calc-alkaline magmas is perhaps underappreciated. Interpretation of U–Th mineral isochrons where certain phenocrysts lie above or below a line regressed through the other phases, may reflect a variety of processes. Plagioclase separates for app and dbc have (230Th/232Th) ratios that fall below isochrons regressed through the other phases and whole rock compositions (Fig. 5), and they have high U/

Th ratios, which is atypical of plagioclase (DTh NDU)(Blundy and Wood, 2003). One possible explanation for the deviation from the isochron is that some of the plagioclase phenocrysts contain micro-inclusions of a high U/Th phase such as zircon or magnetite. Photomicrographs of numerous plagioclase phenocrysts from this sample indicate that a complex magmatic record is preserved in these grains, some of which contain resorbed cores and overgrowths (Fig. 8B,C). These observations imply that careful petrography and characterization of mineral zoning patterns, when combined with precise 230Th/238U data, can provide strong constraints on the degree of open-system evolution.

6.2. Origin of 230Th excess in Mount Adams lavas

230Th excesses in continental arc lavas have been explained by various mantle and crustal processes including: 1) crystallization of

accessory minerals that have DU NDTh, 2) dynamic mantle melting in which DU NDTh, 3) addition of partial melts of the subducted slab to the mantle wedge, and 4) assimilation of 230Th-enriched lower crustal melts by mantle derived magma (Reagan et al., 2003; George et al., 2003; Dossetto et al., 2003; Thomas et al., 2002; Jicha et al., 2007; Garrison et al., 2006). Garnet, Al-rich clinopyroxene, and magnetite/ spinel preferentially retain U over Th (Wood et al., 1999; Blundy and Wood, 2003), and the presence of these minerals during any of the processes noted above could produce 230Th excesses. In all cases, preservation of 230Th excesses requires a time period of b105 years between U–Th fractionation and eruption. Identifying which process or combination of processes is responsible for generating the 230Th excesses in the Mount Adams magmas is discussed below.

Fig. 8. Photomicrographs of select plagioclase grains from Mount Adams lavas. A) Euhedral plagioclase grains from sample atm. The plagioclase separate from this sample lies on the U–Th isochron with the other phenocrysts, groundmass and whole-rock analyses. A complex magmatic record is preserved in the plagioclase grains of the voluminous eruption of unit app as evidenced by B) resorbed cores and euhedral rims, and C) resorbed rim with overgrowth.

We cast the 40Ar/39Ar and U–Th geochronological data in terms of eruptive volumes and rates in Fig. 9. In the last 140 kyr, a modest increase in eruptive rate took place between ~120 ka and 100 ka, and a major increase occurred at ~33 ka, with formation of the current cone of Mount Adams and the eruption of unit app and related lavas (Fig. 9). As discussed above, units amc and dbc have no evidence for significant Fig. 9. Cumulative eruptive volume versus time for Mount Adams lavas erupted over the phenocryst residence times, and this seems to reflect low rates of last 140 kyr. Open symbols and dashed line represent minimum volume estimates, fl eruptions from ~110 to ~40 ka, which may be interpreted to re ect whereas filled symbols and solid black line represents maximum volume estimates. A low rates of magma injection into the upper crust. The sole unit major cone-building event occurred at ~35 ka, which includes the eruption of unit app. B.R. Jicha et al. / Earth and Planetary Science Letters 277 (2009) 38–49 47

The restite mineralogy and degree of partial melting involved in made by Berlo et al. (2004) based on U–Th isotope compositions in generation of Cascade arc lavas varies greatly along the arc (Hildreth, young dacites at nearby Mount St. Helens, where they noted that 230Th 2007). Geochemical modeling by Jicha et al. (2009) indicates both excesses could not be produced by partial melting of mafic lower crust. garnet-bearing and garnet-free sources were involved in generation of A small degree partial melt (~1%) or a dynamic melt of a garnet-bearing the Mt. Adams magmas. A garnet-bearing, intraplate, asthenospheric gneiss, , or schist that contains a significant amount of mantle source is likely the major contributor to Adams magmas garnet would produce a melt that has N50% 230Th excess. The garnet- (Jicha et al., 2009). Melting calculations indicate that a low degree bearing protolith must have had a (230Th/232Th) ratio of ~1.4, which is partial batch melt (1–5%) of garnet peridotite (Ol:Cpx:Opx:Grt: higher than the inferred composition of lower crustal rocks. Given that Spl=55:15:20:6:4) will produce a melt that has 4–17% 230Th excess, Cascade arc lavas span a large range in (230Th/232Th) ratios (1.0–1.4), which is nearly identical to the range observed for the Adams lavas which indicate an origin from mantle domains that had vastly different that have low γOs values (Fig. 10). Hart et al. (2003) suggested that the isotope compositions, it is conceivable that the crustal column region- Re–Os budgets of the Mt. Adams lavas were largely controlled by ally contains rocks that have a similar spread in (230Th/232Th) ratios. interaction with mafic, lower crust and not inherited from the mantle. Hart et al. (2003) suggested that the high γOs values in the Mount This model stands in contrast to that advocated by Borg et al. (2000) Adams lavas reflect interaction with lower crustal material that is for the Lassen region where 187Os/188Os ratios that are significantly Oligocene age or older. The oldest basement exposed in the Columbia higher than mantle compositions (187Os/188OsN0.127) were attributed Embayment of the central Cascade arc may be the Rimrock Inlier to input from radiogenic slab-derived fluids into the source region. If (Miller, 1989), which is composed of Late Jurassic to early Cretaceous radiogenic Os was transferred from the slab to the mantle wedge via tonalities, gabbros, , and gneisses. The mineralogy of the

fluids beneath Mount Adams, the lavas that have the highest γOs complex is highly variable, but several of the gneiss bodies contain values would be expected to have 238U excesses or, at a minimum, the abundant garnet (Miller, 1989). Mattinson (1972) reported a measured lowest 230Th excesses, which is not observed (Fig. 10). Furthermore, 87Sr/86Sr ratio of 0.7041 for a tonalitic gneiss exposed near Mount

γOs values in the Mount Adams region do not correlate with chemical Rainier, which overlaps the range measured in Mt. Adams lavas. or isotopic indices commonly associated with subduction components Therefore, assimilation of a small degree partial melt of ancient gneiss 87 86 230 (e.g., Sr/P, Ba/Nb, Sr/ Sr ratios) (Hart et al., 2003; Jicha et al., 2009), could generate the elevated γOs values and the large Th excesses, in contrast to observations at Lassen. Therefore, the moderate 230Th and such a process would not affect the Sr, Nd, Hf, or Pb isotope excesses of 2 to 16% in Mount Adams lavas are probably inherited from composition of the lavas. the mantle, whereas the γOs values of 30–60 are interpreted to reflect All of the Adams lavas that have high γOs values, including all of the interaction with young mafic precursor magmas. N350 ka lavas that were not a part of this study (γOs =77–342) (Hart Not all of the Os–Th isotope variations in Mount Adams lavas can be et al., 2003), erupted from vents located at least 10 km away from the explained by melting of peridotite followed by assimilation of young, central stratovolcano, mainly to the west and south (Fig. 1). The 230 mafic lower crust. Several lavas have extremely large Th excesses peripheral location of the high γOs lavas suggests that 1) the basement (25–48%) and γOs values of 75 to 282 (Fig. 10). Assimilation of young in the surrounding area away from the main cone may be older and mafic underplated crust beneath Mount Adams by mantle-derived heterogeneous, or 2) the repeated flux of mafic magma into the crust magmas will not produce significant U–Th disequilibria because this beneath the stratovolcano over the last 940 kyr has diluted the older gabbroic material is likely to be composed of low-Al clinopyroxene, crustal Os isotope compositions away from those characteristic of orthopyroxene, and plagioclase, all of which have very small D's for the pre-existing lower crustal material towards those that are more both U and Th, and therefore will generate minimal U–Th disequilbria typical of mantle compositions. during partial melting (Berlo et al., 2004). A similar conclusion was 6.3. Mount Adams magma genesis

Based on our interpretations of the U–Th isotope disequilibria and Os–Th isotope variations, we offer a petrogenetic model for Mount Adams magma genesis that expands the findings of Jicha et al. (2009) to include time constraints for the various subvolcanic processes. Mafic melts of varying origin (MORB- or intraplate-like mantle sources± subduction components) migrated through the upper mantle and

interacted with mafic lower crust, as reflected in elevated γOs values relative to the mantle (Hart et al., 2003). Magmas injected into the crust beneath the central stratovolcano likely assimilated earlier Quaternary magma batches that crystallized as gabbros or cumulates, whereas those on the periphery may have incorporated small degree partial melts of 230 significantly older basement, producing high γOs and large Th excess in the peripheral lavas relative to those near the central stratovolcano. The general concordance between the U–Th mineral isochron ages and eruptive ages constrained by 40Ar/39Ar and K–Ar geochronology suggests that the time between crystal growth and eruption was relatively brief (i.e., few thousand years or less). The observed phenocrysts in erupted lavas, for the most part, did not participate in the entire magma evolutionary process. The apparent elimination of

230 238 crystals that participated in lower crustal processes suggests that Mount Fig. 10. ( Th/ U) vs. γOs for Mt. Adams lavas. Samples with the largest Th excess (i.e., 230 238 187 188 187 188 Th/ U=1.47) have high γOs values. γOs=[( Os/ Os)meas /( Os/ Os)mantle−1]× Adams magmas may have been produced in a manner similar to that 187 188 100 where / ( Os/ Os)mantle =0.127(Shirey and Walker,1998). Os isotope data from Hart proposed in the hot zone model of Annen et al. (2006). In this model, et al. (2003). Dashed curve represents mixing between an average Mount Adams basalt and evolution to intermediate and silicic compositions occurs primarily a partial melt of a garnet amphibolite (Qtz:Gt:Plag:Amph:Mt=30:30:25:10:5). Partition through assimilation-fractional-crystallization (AFC) processes in the coefficients used for melting calculations: quartz DTh=0, DU =0; garnet DTh=0.003, mid- to lower-crust, followed by adiabatic ascent of magma in a DU =0.018; plagioclase DTh =0.0003, DU =0.00006; amphibole DTh =0.007, DU =0.007; magnetite DTh=0.004, DU =0.04; see text for discussion. superheated state, which leads to the resorption of xenoliths or crystals 48 B.R. Jicha et al. / Earth and Planetary Science Letters 277 (2009) 38–49 that resided in the magma in the lower crust. Crystallization of the arc from an integrated U-Th-Ra-Be isotope study. J. Geophys. Res. 108. doi:10.1029/ 2002JB001916. erupted phenocrysts from mostly aphyric magmas in the upper crust Hart, G.L., Johnson, C.M., Shirey, S.B., Clynne, M.A., 2002. Osmium isotope constraints on likely occurred rapidly on timescales of decades or less due to lower crustal recycling and pluton preservation at Lassen Volcanic Center, CA. Earth decompression and degassing prior to eruption (e.g., Blundy and Planet. Sci. Lett. 199, 269–285. – Hart, G.L., Johnson, C.M., Hildreth, W., Shirey, S.B., 2003. New osmium isotope evidence Cashman, 2005). Short crystal residence times inferred from U Th for intra-crustal recycling of crustal domains with discrete ages. Geology 31, mineral isochron data have now been observed at several oceanic and 427–430. continental arc volcanoes that have erupted magmas interpreted to have Heumann, A., Davies, G.R., Elliott, T., 2002. Crystallization history of at Long – assimilated middle to lower crust (this study; Jicha et al., 2005, 2007; , California, inferred from combined U-series and Rb Sr isotope systematics. Geochim. Cosmochim. Acta 66, 1821–1837. Zellmer et al., 2000), which suggests that decoupling of lower and upper Hildreth, W., 2007. Quaternary magmatism in the Cascades-geologic perspectives. U.S. crustal crystal records may be quite common in arcs. Geol. Surv. Prof. Paper 1744, 125. Hildreth, W., Fierstein, J., 1995. Geologic map of the Mount Adams volcanic field, of southern Washington. U.S. Geol. Surv. Map I-2460, scale 1:50,000. Acknowledgements Hildreth, W., Fierstein, J.,1997. Recent eruptions of Mount Adams, Washington Cascades, USA. Bull. Volcanol. 58, 472–490. We thank Mike Clynne for providing age information for samples Hildreth, W., Lanphere, M.A., 1994. Potassium–argon geochronology of a basalt–andesite– fi 230 232 dacite arc system: the Mount Adams volcanic eld, Cascade Range of southern from the Lassen volcanic center so that initial ( Th/ Th) ratios Washington. Geol. Soc. Amer. Bull. 106, 1413–1429. could be determined for the Trimble et al. data. John Hora assisted Jicha, B.R., Singer, B.S., Beard, B.L., Johnson, C.M., 2005. Contrasting timescales of with Th isotope analyses. Calvin Miller, Simon Turner, Charlie Bacon, crystallization and magma storage beneath the Aleutian Island arc. Earth Planet. Sci. Lett. 236, 195–210. Jake Lowenstern, and two anonymous reviewers provided helpful Jicha, B.R., Singer, B.S., Beard, B.L., Johnson, C.M., Moreno-Roa, H., Naranjo, J.A., 2007. Rapid comments that greatly improved this manuscript. This work was magma ascent and the generation of 230Th excesses in the lower crust at Puyehue– supported by grants from NSF (EAR-9980512, EAR-0309853, -0337667, Cordón Caulle, Southern volcanic zone, Chile. Earth Planet. Sci. Lett. 255, 229–242. Jicha, B.R., Hart, G.L., Johnson, C.M., Hildreth, W., Beard, B.L., Shirey, S.B., Hildreth, W., 2009. -0738007) and the Department of Geology and Geophysics at the Isotopic and trace element constraints on the petrogenesis of lavas from the Mount University of Wisconsin-Madison. Adams volcanic field, Washington. Contrib, Mineral. Petrol. 157. doi:10.1007/s00410- 008-0329-6. fi References Le Bas, M.J., Le Maitre, R.W., Streckeisen, A., Zanettin, B., 1986. A chemical classi cation of volcanic rocks based on the total alkali-silica diagram. J. Petrol. 27, 745–750. Leeman, W.P., Smith, D.R., Hildreth, W., Palacz, Z., Rogers, N.,1990. Compositional diversity Annen, C., Blundy, J.D., Sparks, R.S.J., 2006. The genesis of intermediate and silicic of Late Cenozoic basalts in a transect across the southern Washington Cascades: magmas in deep crustal hot zones. J. Petrol. 47, 505–539. implications for subduction zone magmatism. J. Geophys. Res. 95, 19561–19582. Asmerom, Y., Edwards, R.L., 1995. U-series isotope evidence for the origin of continental Leeman, W.P., Lewis, J.F., Evarts, R.C., Conrey, R.M., Streck, M.J., 2005. Petrologic basalts. Earth Planet. Sci. Lett. 134, 1–7. constraints on the thermal structure of the Cascades arc. J. Volcanol. Geoterm. Res. Bacon, C.R., Druitt, T.H., 1988. Compositional evolution of the zoned calc-alkaline 140, 67–105. magma chamber of , , Oregon. Contrib. Mineral. Petrol. Mattinson, J.M., 1972. Jurassic metamorphism of basement gneisses near , 98, 224–256. Washington. Carnegie Institute of Washington Year Book 71, 576-578. Bacon, C.R., Bruggman, P.E., Christiansen, R.L., Clynne, M.A., Donnelly-Nolan, J.M., Miller, R.B., 1989. The Mesozoic Rimrock Lake Inlier, southern Washington Cascades: Hildreth, W., 1997. Primitive magmas at five Cascade volcanic fields: melts from hot, implications for the basement to the Columbia Embayment. Geol. Soc. Amer. Bull. heterogeneous sub-arc mantle. Can. Mineral. 35, 397–423. 101, 1289–1305. Baedecker, P.A., 1987. Methods for geochemical analysis. U.S. Geol. Surv. Bull. 1770. Mooney, W.D., Weaver, C.S., 1989. Regional crustal structure and tectonics of the Pacific Bennett, J.T., Krishnaswami, S., Turekian, K.K., Melson, W.G., Hopson, C.A., 1982. The coastal states; California, Oregon, and Washington. Geol. Soc. Am. Memoir 172, uranium and thorium decay series nulclides in Mt. St. Helens effusives. Earth Planet. 129–161. Sci. Lett. 60, 61–69. Morgan, D.L., Blake, S., Rogers, N.W., DeVivo, B., Rolandi, G., MacDonald, R., Hawkes- Berlo, K., Turner, S., Blundy, J., Hawkesworth, C., 2004. The extent of U-series disequilibria worth, C.J., 2004. Time scales of crystal residence and magma chamber volume from produced during partial melting of the lower crust with implications for the formation modeling of diffusion profiles in phenocrysts: Vesuvius 1944. Earth Planet. Sci. Lett. of the Mount St. Helens dacites. Contrib. Mineral. Petrol. 148, 122–130. 222, 933–946. Blundy, J., Cashman, K., 2005. Rapid decompression-driven crystallization recorded by Newman, S., Macdougall, J.D., Finkel, R.C., 1986. Petrogenesis and 230Th–238U disequi- melt inclusions from Mount St. Helens volcano. Geology 33, 793–796. librium at Mt. Shasta, California, and in the Cascades. Contrib. Mineral. Petrol. 93, Blundy, J., Wood, B., 2003. Mineral-melt partitioning of uranium, thorium and their 195–206. daughters. In: Bourdon, B., Henderson, G.M., Lundstrom, C.C., Turner, S.P. (Eds.), Parsons, T., Trehu, A.M., Luetgart, J.H., Miller, K., Kilbride, F., Wells, R.E., , M.A., Uranium Series Geochemistry, Reviews in Mineralogy and Geochemistry. The Fleuh, E., ten Brink, U.S., Christensen, N.I., 1998. A new view into the Cascadia Mineralogical Society of America, Washington, DC, pp. 59–123. subduction zone and : implications for hazards along the Borg, L.E., Clynne, M.A., Bullen, T.D., 1997. The variable role of slab-derived fluids in the Washington margin. Geology 26, 199–202. generation of a suite of primitive calc-alkaline lavas from the southernmost Reagan, M.K., Morris, J.D., Herrstrom, E.A., Murrell, M.T., 1994. Uranium series and Cascades, California. Can. Mineral. 35, 425–452. beryllium isotope evidence for an extended history of subduction modification of Borg, L.E., Brandon, A.D., Clynne, M.A., Walker, R.J., 2000. Re–Os isotopic systematics of the mantle below Nicaragua. Geochim. Cosmochim. Acta 58, 4199–4212. primitive lavas from the Lassen region of the Cascade Arc, California. Earth Planet. Reagan, M.K., Sims, K.W.W., Erich, J., Thomas, R.B., Cheng, H., Edwards, R.L., Layne, G., Sci. Lett. 177, 301–317. Ball, L., 2003. Time-scales of differentiation from mafic parents to in North Bourdon, B., Wörner, G., Zindler, A., 2000. U-series evidence for crustal involvement and American continental arcs. J. Petrol. 44, 1703–1726. magma residence times in the petrogenesis of Parinacota volcano, Chile. Contrib. Sherrod, D.R., Smith, J.G., 1990. Quaternary extrusion rates of the Cascade Range, Mineral. Petrol. 139, 458–469. northwestern and southern . J. Geophys. Res. 95, Condomines, M., Gauthier, P.-J., Sigmarsson, O., 2003. Timescales of magma chamber 19465–19474. processes and dating of young volcanic rocks. In: Bourdon, B., Henderson, G.M., Sharp, W., Renne, P., 2005. The 40Ar/39Ar dating of core recovered by the Hawaii Lundstrom, C.C., Turner, S.P. (Eds.), Uranium Series Geochemistry, Reviews in Scientific Driling project (phase 2), Hilo, Hawaii. Geochem. Geophys. Geosys. 6. Mineralogy and Geochemistry. The Mineralogical Society of America, Washington, doi:10.1029/2004GC000846. DC, pp. 125–174. Shirey, S.B., Walker, R.J., 1998. The Re–Os isotope system in cosmochemistry and high- Cooper, K.M., Reid, M.R., 2003. Re-examination of crystal ages in recent Mount St. temperature geochemistry. Ann. Rev. Earth Planet. Sci. 26, 423–500. Helens lavas: implications for magma reservoir processes. Earth Planet. Sci. Lett. Sigmarsson, O., Martin, H., Knowles, J., 1998. Melting of a subducting oceanic crust from 213, 149–167. U-Th disequilibria in austral Andean lavas. Nature 394, 566–569. Costa, F., Chakraborty, S., Dohmen, R., 2003. Diffusion coupling between trace and major Sims, K.W.W., Gill, J., Dossoto, A., Hoffmann, D.L., Lundstrom, C., Williams, R., Ball, L., elements and a model for calculation of magma residence times using plagioclase. Tollstrup, D., Turner, S., Prytulak, J., Glessner, J., Standish, J., Elliott, T., 2008. An inter- Geochim. Cosmochim. Acta 67, 2189–2200. laboratory assessment of the Th isotopic composition of synthetic and rock Dossetto, A., Bourdon, B., Joron, J.-L., Dupré, B., 2003. U–Th–Pa–Ra study of the standards. Geostand. Geoanal. Res. 32, 65–91. Kamchatka arc: new constraints on the genesis of arc lavas. Geochim. Cosmochim. Singer, B.S., Ackert, R.P., Guillou, H., 2004. 40Ar/39Ar and K–Ar chronology of Pleistocene Acta 67, 2857–2877. glaciations in Patagonia. Geol. Soc. Amer. Bull. 116, 434–450. Elliott, T., Plank, T., Zindler, A., White, W., Bourdon, B., 1997. Element transport from the Steiger, R.H., Jäger, E., 1977. Subcommission on geochronology; convention on the use slab to volcanic front at the Mariana arc. J. Geophys. Res. 102, 14991–15019. of decay constants in geo- and cosmochronology.EarthPlanet.Sci.Lett.36, Garrison, J., Davidson, J., Reid, M., Turner, S., 2006. Source versus differentiation controls 359–362. on U-series disequilibria: insights from Cotopaxi Volcano, Ecuador. Earth Planet. Thomas, R.B., Hirschmann, M.M., Cheng, H., Reagan, M.K., Edwards, R.L., 2002. (231Pa/235U)– Sci. Lett. 244, 548–565. (230Th/238U) of young mafic volcanic rocks from Nicaragua and Costa Rica and the George, R., Turner, S., Hawkesworth, C., Morris, J., Nye, C., Ryan, J., Zheng, S., 2003. influence of flux melting on U-series systematics of arc lavas. Geochim. Cosmochim. Melting processes and fluid and sediment transport rates along the Alaska-Aleutian Acta 66, 4287–4309. B.R. Jicha et al. / Earth and Planetary Science Letters 277 (2009) 38–49 49

Trimble, D.A., Clynne, M.A., Robsinson, S.W., 1984. The application of uranium-thorium Volpe, A.M., Hammond, P.E., 1991. 238U–230Th–226Ra disequilibria in young Mount St. systematics to rocks from the Lassen dome field, California. U.S. Geol. Surv. Open Helens rocks: time constraint for magma formation and crystallization. Earth Planet. File Report 84–371, pp. 108. Sci. Lett. 107, 475–486. Turner, S.P., McDermott, F., Hawkesworth, C., Kepezhinkas, P., 1998. A U-series study of Wood, B.J., Blundy, J.D., Robinson, J.A.C., 1999. The role of clinopyroxene in generating U- lavas from Kamchatka and the Aleutians: constraints on source composition and series disequilibrium during mantle melting. Geochim. Cosmochim. Acta 63, melting processes. Contrib. Mineral. Petrol. 133, 217–234. 1613–1620. Turner, S.P., Bourdon, B., Gill, J., 2003. Insights into magma genesis at convergent Zellmer, G.F., Blake, S., Vance, D., Hawkesworth, C., Turner, S., 1999. Plagioclase residence margins from U-series isotopes. In: Bourdon, B., Henderson, G.M., Lundstrom, C.C., times at two island arc volcanoes (Kameni Islands, Santorini, and Soufriere, St. Vincent) Turner, S.P. (Eds.), Uranium Series Geochemistry, Reviews in Mineralogy and determined by Sr diffusion systematics. Contrib. Mineral. Petrol. 136, 345–357. Geochemistry. The Mineralogical Society of America, Washington, DC, pp. 255–315. Zellmer, G., Turner, S.P., Hawkesworth, C., 2000. Timescales of destructive plate margin Volpe, A.M., 1992. 238U–230Th–226Ra disequilibrium in young Mt. Shasta andesites and magmatism: new insights from Santorini, Aegean volcanic arc. Earth Planet. Sci. Lett. dacites. J. Volcanol. Geotherm. Res. 53, 227–238. 174, 265–281.