**1. Introduction**

Transport of magma within the Earth's crust has far-reaching e ffects: pluton emplacement at a variety of structural levels, development of associated contact and hydrothermal metamorphism in and around the magma conduits (plumbing systems), and commonly, but not always, volcanic eruptions. The relative timing of magma and/or fluid transport in the system is important for understanding mass and heat transfer, but also for volcanic–plutonic connections and volcanic hazard mitigation. The rock record at any one location depends on the interplay between the di fferent processes and parts of the system. The magma emplacement rate and duration [1], tectonic forces [2], pre-existing and evolving permeability structure [3], and metamorphic mineral reactions [4] all serve to modify

and cause feedbacks within the system. Simplifying assumptions, such as an instantaneous magma emplacement (e.g., Annen [5] and Reverdatto et al. [6]) or a lack of magmato-metamorphic processes (e.g., Paterson et al. [7]) exclude critical aspects of mass and heat transfer and obscure our understanding of the integrated system. The term "magmato-metamorphic" is used here to encompass a variety of near- and sub-solidus processes, such as melt and/or magmatic water flow, major and accessory phase (re)crystallization, and mineral reactions, that continue to modify the rock and obscure the record of super-solidus processes. More recently, authors have been applying sophisticated modeling that accounts for incremental pluton emplacement (e.g., Annen [8]) and pulsed conduit flow (e.g., Floess and Baumgartner [9]) to better understand the complex thermochemical feedback between magmas and wall rocks. This paper presents new preliminary petrochronology data from the Alta stock-Little Cottonwood stock system, which demonstrates the complex spatial and temporal patterns of magmatism, hydrothermal infiltration, and contact metamorphism over a crustal section from the surface to an ~11.5 km paleodepth [10].

The Wasatch Intrusive Belt comprises a series of Eocene–Oligocene plutons that intruded the thickened crust in northern Utah following the Sevier orogeny [11–13] (Figure 1A). The Wasatch Intrusive Belt magmas likely resulted from lower-crustal metasomatism and melting caused by the last stages of subduction of the Farallon plate [12,14]. The emplacement and space-making involved extension driven by a combination of far-field tectonic and gravitational forces acting on the thickened crust prior to the Basin and Range extension [15–17]. The plutons intrude a sequence of siliciclastic and carbonate rocks ranging in age from Proterozoic to Triassic. Following emplacement, Wasatch Intrusive Belt plutons were exhumed by range-bounding normal faults of the Oquirrh and Wasatch ranges (Figure 1B) beginning at ~18 Ma (17.6 ± 06 Ma K-Ar sericite date [18] and are still being actively exhumed (e.g., References [19–21]). Offset across the Wasatch fault has tilted the range by ~20◦ to the east along a roughly horizontal N–S rotation axis [22]. This rotation and exhumation exposed a crustal section from ~11.5 km at the southeast corner of the Salt Lake Valley [10] to the paleosurface (Figure 1C) east of Park City, Utah. From structurally deepest to shallowest, this section contains the Little Cottonwood & Ferguson stocks (included in the Little Cottonwood stock from here forward), Alta stock, Clayton Peak and other eastern stocks, and the Keetley volcanic deposits. Previous geochronology from the Wasatch Intrusive Belt includes a combination of multigrain thermal ionization mass spectrometry U-Pb zircon; K-Ar mica and amphibole; fission-track titanite and zircon; and (U-Th)/He and fission-track in titanite, zircon, and apatite dates [19,20,23–25]. These data suggested that pluton emplacement began at ~36 Ma with the eastern stocks, continued with the emplacement of the Alta stock at ~35 Ma, and ended with the Little Cottonwood stock at ~30 Ma. New petrochronology presented here establishes magmatism in the Little Cottonwood and Alta stocks from ~36–26 Ma, beginning and ending with different portions of the Little Cottonwood stock, thus indicating a different sequence and a much longer duration of pluton emplacement from that previously inferred.

The Alta stock is a structurally shallow (~5–5.5-km depth current exposure [22,26,27] dike-like intrusion with subvertical walls (Figure 1)). The two mappable intrusive phases that comprise the Alta stock, namely an equigranular border phase and a later porphyritic central phase, are distinguished by crystal size, microscopic rock texture, and locally recognizable cross-cutting relationships [28]. The major minerals of the Alta stock are plagioclase + K-feldspar + quartz + biotite + amphibole, with titanite + apatite + zircon + ilmenite ± magnetite accessory phases. The contact metamorphic aureole surrounding the Alta stock is especially well-studied [28–34]. Past and continuing work has largely been motivated by the apparent mismatch between the size of the intrusion (~2 km at its widest) and the surrounding contact aureole (≤1.25 km in width). Cook and Bowman [30,31] demonstrated that metamorphism of the carbonate wall rocks was largely driven by the infiltration of high-temperature (~625 ◦C), H2O-rich fluids laterally outward from the border phase of the Alta stock, which exploited and modified the existing permeability structure surrounding the pluton (Figures 1 and 2). Numerical models assuming instantaneous emplacement of the pluton showed that the locations of prograde metamorphic isograds, isotherms based on calcite-dolomite thermometry, and the pattern of 18O/16O depletion across the aureole could be duplicated with ~5000 years of advective heat and fluid flow, followed by ~20,000−30,000 years of conductive heating to produce the outer aureole [33]. This numerical modeling indicates that the observed extent of the contact aureole could be matched with heat supplied solely from cooling of the Alta stock only if the duration of pluton emplacement was <5000 years.

**Figure 1.** (**A**) A simplified geologic map (modified from Wohlers and Baumgartner [35]) shows the location of the Wasatch Intrusive Belt in northern Utah; the correlation of the Wasatch Intrusive Belt with the Uinta Anticline and the aeromagnetic anomaly (nano-Tesla scale bar) that likely represents a continuous body of intrusive rocks at depth [36]. (**B**) The more detailed geologic map overlaid on a shaded relief map and accompanying schematic cross-section (**C**) that illustrates the sample locations and the estimated paleodepth (isobaths) of the analyzed samples [10,22,37]. The cross-section is highly vertically exaggerated (see noted elevations on the map and cross-section) and was calculated using a ~15◦ rotation due to the differing azimuth from John's [22] ~20◦ rotation of the Wasatch footwall block. The western Alta and eastern Little Cottonwood contacts are ~75◦ to WSW and ~50◦ to ENE, respectively, based on three-point solutions.

**Figure 2.** Photographs and interpretation schematics showing the highly composite nature of the magmatic and hydrothermal system. (**A**) A granodiorite (grd in schematics **A**, **B**, **E**, and **F**) dike with slightly more mafic margins cross-cutting nearly identical granodiorite in the Little Cottonwood stock. (**B**) A granodiorite dike that intruded the wall rock and was later crosscut by an aplite dike near the southeastern margin of the Little Cottonwood stock. (**C**) A photograph and (**D**) interpretation of a highly composite outcrop that reflects the injection of both magma and aqueous fluid into carbonate wall rock near the southeastern margin of the Little Cottonwood stock. (**E**,**F**) Calcite marbles intruded by granodiorite sills with calcsilicate skarns developed at the contacts. Calcsilicate skarns are also found adjacent to and included in the Alta stock. Ldw = ludwigite, resulting from boron metasomatism in the Alta aureole [69].

In contrast, if space for the Alta stock was made by horizontal dilation perpendicular to the long axis of the pluton [12] at a reasonable tectonic rate for a continental lithosphere (~1–2 mm·yr<sup>−</sup>1), emplacement of the Alta stock must have taken ~1–2 Myr [38]. Thus, the preferred emplacement model based on field observations implies that the Alta stock grew too slowly by nearly three orders of magnitude to account for the thermal aureole that surrounds it. This extreme mismatch poses two important questions: What is the thermal history of the system, including the Alta conduit, aureole, and associated Little Cottonwood stock? What are the sources of heat and fluids, beyond the Alta stock, that drove metamorphism in the Alta aureole?

Because the contact aureole is centered on the Alta stock, it is unsurprising that previous studies have assumed that the Alta stock was the source of the heat and fluids that produced the aureole. However, a protracted emplacement of the Alta stock, coupled with the recognition that fluid infiltration is required to drive metamorphic reactions and 18O/16O depletion in the aureole, opens up the possibility that at least some of the hot fluid that emanated from the Alta stock originated from another source. The structurally deeper (current exposures correspond to a paleodepth range of ~6.5–11.5 km) and much larger Little Cottonwood stock is bounded by the Wasatch and Deer Creek faults to the west and south, respectively (Figure 1). Data from this study confirmed that the Ferguson stock, interpreted as a satellite to the main Little Cottonwood stock body, is cogenetic. In general, the Little Cottonwood stock is more felsic than the Alta stock, is structurally composite, and its modal mineralogy, crystal size distribution, and chemistry are highly variable [39,40] (Figure 2). The major minerals that make up the Little Cottonwood stock are plagioclase + K-feldspar + quartz + biotite ± amphibole, with titanite + apatite + zircon + ilmenite ± magnetite accessory phases (Figure 3). The Little Cottonwood stock intruded the Neoproterozoic Big Cottonwood Formation on the north and contains a ~1 km × ~200 m screen of these rocks near the western interior of the intrusion (Figure 1). The Little Cottonwood stock magmas intruded Cambrian siliciclastic rocks and Mississippian carbonates on its southeastern margin [41]. Emplacement of the Little Cottonwood stock contact metamorphosed and locally melted the surrounding wall rocks. The Big Cottonwood Formation is composed of interbedded quartzite and shale, and pelitic layers contain the peak contact metamorphic assemblage of biotite + cordierite + sillimanite + K-feldspar + melt [35].

The mineral textures (sensu lato) of intrusive igneous rocks are a topic of active conversation, and a growing body of both field (e.g., Johnson and Glazner [42]) and experimental (e.g., Lundstrom [43]) data sugges<sup>t</sup> that both major and accessory phases in granitic rocks are susceptible to recrystallization at temperatures lower than traditional solidi of ~700 ± 50 ◦C [44]. For example, the coarsening of orthoclase to so-called megacrysts that are common in calc-alkaline intrusions, including the Little Cottonwood stock, may involve melt (e.g., Higgins [45]) or may occur completely in the solid state [46]. Cathodoluminescence imaging, titanium-in-quartz thermometry in the Alta stock [47], and numerical diffusion modeling of similar data from the Tuolumne Intrusive Suite [48] sugges<sup>t</sup> that quartz in granitic plutons has commonly recrystallized at temperatures in the hydrothermal regime well below traditional granitic solidi. Accessory phases, such as titanite, have been shown to be reactive and recrystallized at similar thermal and fluid conditions [49–51]. Proper interpretation of such overprinting and continued modification using a range of processes spanning magmatic (e.g., Bartley et al. [52]) to hydrothermal (e.g., Smirnov [53]) requires a robust, multifaceted petrochronology approach to relate the chemistry and the time for the process to occur.

**Figure 3.** Plane polarized (top) and cross-polarized (bottom) photomicrographs that illustrate the range of titanite morphologies and phase relationships observed in both the Alta (**A**,**B**) and Little Cottonwood stocks (**C**). Titanites range from euhedral with few inclusions to anhedral rims on oxide phases (typically ilmenite). Zircons (not seen here) are typically inclusion free and included in a variety of early crystallizing phases, such as plagioclase and biotite. We interpreted these different groups to represent assemblages that grew via different processes. (**A**) Alta central phase sample C4, (**B**) Alta border phase sample D310b, and (**C**) Little Cottonwood stock sample MS15-03. All images were taken at 4× objective magnification.

Petrochronology exploits the accuracy and spatial context of in situ laser ablation split stream (LASS) inductively coupled plasma mass spectrometry (ICP-MS) analysis to simultaneously collect U-Pb isotopes and trace element contents in a variety of chronometer phases [54–56]. Measuring U-Pb and tetravalent cations in multiple phases, such as zircon and titanite from the same rock, gives more complete information about the thermal history of the rock due to the different reactions responsible for the paragenesis of each phase. Both zircon and titanite are common in calc-alkaline rocks like the Wasatch Intrusive Belt (WIB) (Figure 3) [57,58]. Zircon saturation depends on the Zr content and the network modifier/former ratio M of the magma [59,60]. In the Alta–Little Cottonwood system, whole-rock chemistry predicts the zircon saturation at ~725 ± 25 ◦C. Hanson [39] documented the occurrence of inherited zircon based on petrographic observations, which prolonged the Zr saturation in both the Little Cottonwood and Alta stocks. Prolonged zircon saturation suggests that the zircon saturation temperature (TZr) would be a maximum emplacement temperature for the pluton [39,61]. Titanite paragenesis in calc-alkaline rocks is more complex, and as mentioned above, titanite (re)crystallizes via multiple processes, such as de novo crystallization from melt [62], redox reactions involving Fe-Ti oxides [58], and/or (re)crystallization in the presence of fluids [63]. Titanite is stable in calc-silicate skarns [64–66] and often grows due to

the breakdown of Ti-bearing clinopyroxene, and may (re)crystallize during the infiltration of fluorine and/or H2O-rich fluids [50,67,68].

This paper presents titanite and zircon U-Pb dates, trace element concentrations, and tetravalent cation thermometry from 17 samples in the Alta–Little Cottonwood intrusive–hydrothermal system, which spans the upper ~11.5 km of the Eocene–Oligocene crust in what is now Utah. The new data indicate significant di fferences from previous conclusions regarding the timing, duration, and spatial extent of magma emplacement and hydrothermal activity within the Alta–Little Cottonwood system. The data together define a temperature–time path that spans >10 Myr from ~36–23 Ma. They further indicate that magmatic and hydrothermal processes were episodic at any given location but they were active in some portion of the system continuously throughout the >10 Myr duration.

#### **2. Materials and Methods**

Samples were taken from the freshest or least weathered outcrops and trimmed to remove any stained or visually altered rock before cutting thin section billets. As much as possible, titanite and zircon crystals were imaged and inspected for internal chemical zoning prior to analysis using backscattered electron and cathodoluminescence detectors, respectively. These images guided the laser spot placements to minimize the mechanical mixing of chemically and/or isotopically distinct domains and to ensure an analysis of the full range of textural populations in each sample. Many of the zircons analyzed, especially in the Alta stock samples, were chemically zoned at a scale smaller than the analytical beam (~20–25 μm diameter for zircon analyses), and thus the dates were certainly mechanically mixed during some analyses and represent minimum durations of zircon growth.

The majority of samples (except LCS-01,02 and MS15-01,02) were analyzed in thin sections to maintain the petrologic context of the U-Pb dates and trace-element analyses via the laser ablation split stream (LASS) technique; see the method and instrument details of References [49,54,70]. The crystals were ablated using a Photon Machines/Teledyne 193 nm Excimer laser equipped with a two-volume Helex ® stage [71] and the ablated material was introduced via He carrier gas to the multi-collector inductively-coupled plasma (ICP-MS) and quadrupole (Q-MS) mass spectrometers to measure uranium and lead isotopes and trace elements, respectively. Isotopic ratios were measured on a Nu Plasma 3 and Thermo NeptunePlus multicollector ICP-MS and trace elements were measured on Agilent 7500ce and 8900 quadrupole mass spectrometers. A laser spot diameter of 20–40 μm for titanite and 20–25 μm for zircon, and a laser repetition rate of 4–6 Hz and fluence of ~2–3 J/cm<sup>2</sup> were used during the in situ analyses.

Isotopic analyses were bracketed using and standardized to a matrix-matched primary isotopic reference material (91500 zircon, Bear Lake Road and MKED-1 titanites [72–76]). Matrix-matched secondary reference materials were analyzed during all the zircon analyses (Plesovice zircon [77]) and titanite analyses from twelve samples (BLR titanite) to determine the accuracy and propagate external uncertainties. During the titanite analyses from six samples, both "Yates Mine" titanite [70] and Plesovice zircon (standardized to 91500 primary reference material) were used as a secondary reference material due to a lack of titanite standards at the time (see the Data Repository File).

Mass spectrometry data were reduced and interpreted using the VizualAge Data Reduction Scheme [78] for the Iolite plugin within IgorPro [79]. Homogeneous portions of the time-resolved data were selected to minimize the common Pb by monitoring the 204Pb and 208Pb channels and maximize the concordance. The isotopic dates were not used as a criteria for selecting portions of the data. The Iolite output data were further reduced using the Isoplot plugin for Excel [80] and the IsoplotR package for the R statistical program [81]. Analytical uncertainty, counting statistics, and extra error for the homogeneity and accuracy of the secondary reference materials were propagated in quadrature for the final reported uncertainty (see the Data Repository File), which was typically ~2–3% or ~0.5–1.0 Ma for these samples. The typically high variance and discordance of U-Pb in titanite data, and some in situ zircon data, require an interpretation of the raw dates using a partial Pb isochron method [82]. Initial 207Pb/206Pb values were chosen using a combination of linear regression, visual data fitting, calculation using Stacey–Kramer Pb growth curves, and comparison to Pb isotope data from K-feldspar. All of the initial Pb values used to calculate the 207Pb-corrected dates fell within the expected range of 0.84–0.87, and the uncertainty in the initial Pb isotope ratio was propagated into the reported individual dates from each analysis point.

Ti-in-zircon and Zr-in-titanite apparent temperatures were calculated using the Ferry and Watson [83] and Hayden et al. [84] calibrations, respectively. Both calibrations depend on the activities of TiO2 and SiO2. The activity of TiO2 was chosen to be αTiO2 = 0.5 ± 0.2 for both thermometers based on three observations: (1) titanite was present and rutile was absent in all samples except SC-02, which also lacked titanite (<sup>α</sup>TiO2 ≥ 0.5 except SC-02 [84]); (2) non-exsolved Fe-Ti oxide pairs from the cogenetic WIB extrusive rocks produced αTiO2 ~ 0.5–0.7 [39,85]; and (3) the rhyolite-MELTS model matching the modal abundance of major phases (for the Alta stock–Little Cottonwood stock (AS–LCS) range of whole-rock chemistry) also produced αTiO2 ~ 0.5–0.7 [86]. For the Ti-in-zircon thermometer, the reported uncertainty of ~±35 ◦C includes the analytical uncertainty (~10% 2SE), uncertainties in the activity terms (~0%–40%), and P dependence (~10%). The plotted apparent temperatures include the estimated pressure dependence of 35 ◦C/GPa. The Data Repository File includes temperatures with and without the pressure corrections calculated using both the lower [83] and the higher pressure dependence of Ferriss et al. [87] of 100 ◦C/GPa. The Zr-in-titanite thermometer is much more dependent on pressure and the typical uncertainty is ±40–50 ◦C.
