Structural, metamorphic, and geochronologic constraints on the origin of the Clearwater core complex, northern Idaho
Published:January 01, 2007
- PDF LinkChapter PDF
P. Ted Doughty, Kevin R. Chamberlain, David A. Foster, Grant S. Sha, 2007. "Structural, metamorphic, and geochronologic constraints on the origin of the Clearwater core complex, northern Idaho", Whence the Mountains? Inquiries into the Evolution of Orogenic Systems: A Volume in Honor of Raymond A. Price, James W. Sears, Tekla A. Harms, Carol A. Evenchick
Download citation file:
New structural, metamorphic, and geochronologic data from the Clearwater complex, north-central Idaho, define the origin and exhumation history of the complex. The complex is divisible into an external zone bound by normal faults and strike-slip faults of the Lewis and Clark Line, and an internal zone of Paleoproterozoic basement exposed in two shear zone–bounded culminations. U-Pb sensitive high-resolution ion microprobe (SHRIMP) dating of metamorphic zircon overgrowths from the external zone yield zircon growth at ca. 70–72 Ma and 80–82 Ma, during peak metamorphism and before tectonic exhumation of the external zone. U-Pb SHRIMP dating of metamorphic zircon rims from the internal zone record growth at ca. 64 and between 59 and 55 Ma. The older ages record pre-extension metamorphism. The younger rim ages were derived from fractured zircons in the Jug Rock shear zone, and they document the beginning of exhumation of the internal zone along deep-seated shear zones that transported the basement rocks to the west. The 40Ar/39Ar ages record quenching of the external zone starting ca. 54 Ma and the internal zone between 53 and 47 Ma by movement along the bounding faults and internal shear zones. After ca. 47 Ma, extension was accommodated via a west-dipping detachment that was active until after ca. 41 Ma. The Clearwater complex is interpreted as an Eocene metamorphic core complex that formed in an extensional relay zone between faults of the Lewis and Clark Line.
The formation of metamorphic core complexes along low-angle normal faults in the Cordillera has been known since their recognition some 30 yr ago (e.g., summaries of Coney  and Armstrong ). While initially viewed as isolated domal structures that formed solely in response to over-thickened crust, continued studies have shown that many core complexes are kinematically linked with other fault systems. In some instances, core complexes are associated with relays or jogs in strike-slip faults (i.e., Death Valley, California; Burchfiel and Stewart, 1966). In the U.S. Cordillera, there is abundant evidence of core complexes forming along active strike-slip faults, but there are few examples older than 20 Ma (e.g., Faulds and Stewart, 1998).
The Northern Rockies is an ideal locale to examine the dynamic and kinematic relationships between metamorphic core complexes and coeval large-magnitude strike-slip faults in older rocks and across all levels of the crust. Eocene extension in northern Washington and Idaho has resulted in north-south–trending core complexes like the Kettle Dome and Priest River complex. The locus of crustal extension ceases abruptly at the Columbia Plateau and jumps 100 km to the southeast in western Montana (Fig. 1), where the Bitterroot and Anaconda complexes record large-magnitude east-west extension.
The eastward jump in the position of core complexes coincides with the Lewis and Clark Line, a >300-km-long, strike-slip fault system traceable from western Idaho to east-central Montana (Harrison et al., 1974; Reynolds, 1979; Hyndman et al., 1988). It moved in a right-lateral sense during core complex formation and behaved like a continental-scale transform fault that linked two offset domains of crustal extension during Eocene time (Sheriff et al., 1984; Hyndman et al., 1988; Doughty and Sheriff, 1992; Foster and Fanning, 1997; Sears et al., 2000; Lewis et al., 2002; Foster et al., 2007).
The Clearwater complex, also known as Boehls Butte–Goat Mountain, is a unique area of anomalously high-grade metamorphic rocks and Precambrian basement that lies within the southern part of the Lewis and Clark Line as it passes through the northern border zone of the Idaho Batholith in north-central Idaho (Figs. 1 and 2) (Hietanen, 1984; Grover et al., 1992; Burmester and Lewis, 1999; Doughty and Buddington, 2002). These rocks have been interpreted as having been uplifted in the hanging wall of Mesozoic thrust faults (Harrison et al., 1986; Skipp, 1987), but the proximity to the Lewis and Clark Line led Seyfert (1984), Doughty and Sheriff (1992), and Burmester and Lewis (1999) to propose that the basement gneisses could have been exhumed in a releasing step in the Lewis and Clark fault zone. Petrologic and isotopic data from the high-grade gneisses reveal that the rocks have been subjected to a late episode of nearly isothermal decompression (Grover et al., 1992; Larson and Sharp, 1998; Mora et al., 1999), which is consistent with their exhumation in an extensional rather than contractional setting.
This paper addresses the structure of the Clearwater complex and the style, kinematics, and chronology of shearing within the complex. We also address the timing of deformation, depth of faulting, and cooling history of the complex.
Rock Types and Metamorphism
The Clearwater complex lies in the St. Joe–Clearwater region of north-central Idaho, along the northern margin of the Idaho Batholith (Fig. 1). This part of the Cordilleran orogen is composed principally of high-grade metamorphic rocks derived from the lower part of the Middle Proterozoic Belt Supergroup and satellites of the Late Cretaceous to early Tertiary Idaho Batholith and epizonal Eocene plutons (Fig. 2) (Hietanen, 1963a, 1963c; Hyndman et al., 1988; Marvin et al., 1984; Burmester et al., 2004). The dominant pattern of metamorphism, classified as M2 by Lang and Rice (1985a), is concentric to the Idaho Batholith and is Barrovian. The peak phase of M2 metamorphism in the Clearwater complex and Bitterroot complexes reached the kyanite-sillimanite-muscovite zone (650–750 °C, 800–1100 MPa) (Grover et al., 1992; House et al., 1997; Foster et al., 2001). East of the Clearwater complex, near Snow Peak, M2 metamorphism reached 600 MPa and 565 °C (Lang and Rice, 1985b). M2 spans between ca. 117 Ma to 82 Ma in the Clearwater region (Grover et al., 1993). Foster et al. (2001) and House et al. (1997) found that peak metamorphism in the northern Bitterroot complex occurred at ≥80–56 Ma. Both estimates largely coincide with intrusion of the main phase of the Idaho Batholith and accretion of exotic terranes to the west.
A third metamorphic event, classified as M3, has been found only within the core of the Clearwater complex. This event, exhibited by complex disequilibrium mineral textures between low-pressure and high-pressure minerals, resulted from isothermal decompression of the Clearwater complex (Lang and Rice, 1985a; Grover et al., 1992) at around 50 Ma (Larson and Sharp, 1998). Similar mineral parageneses are found in the Priest River complex (Rhodes, 1986; Doughty and Price, 1999), and northern Bitterroot complex (Cheney, 1975; Foster et al., 2001).
The high-grade metamorphic rocks exposed in the northern border zone of the Idaho Batholith contain east-directed thrust faults, northwest-trending strike-slip faults of the Lewis and Clark Line, and north-trending normal faults (Fig. 2). Many of the thrust faults are ductile synmetamorphic structures that coincide with peak metamorphism and are identified on the basis of juxtaposed rock types. Other thrust faults postdate peak metamorphism and are identified by offset metamorphic isograds.
Most fault zones of the Lewis and Clark Line in the vicinity of the Clearwater complex have steeply dipping slip surfaces bearing subhorizontal mineral lineations and slip-surface striations. Childs (1982) found that brecciation and low-grade alteration locally overprint an older mylonitic fabric, suggesting that the faults were active across a range of crustal depths. The faults have also been intruded by syntectonic hypabyssal dikes (Childs, 1982) that are related to epizonal middle Eocene plutons like the Beaver Creek, Bungalow, and Roundtop plutons. These intrusions yield 52–46 Ma crystallization ages and date the right-lateral motion as middle Eocene in age (Burmester et al., 2004; Lewis et al., 2002; Marvin et al., 1984). Older left-lateral or transpressional movement is reported on parts of the Lewis and Clark Line but is not well documented in the Clearwater region (i.e., Kell and Childs, 1999; Yin and Oertel, 1995; Sears et al., 2000).
Three large west- to northwest-trending strike-slip faults of the Lewis and Clark fault zone bound the Clearwater complex (Fig. 2). The St. Joe fault is the northern fault, and it can be traced from the southern end of the Priest River complex to the southeast along the northern margin of the Clearwater complex. Near the Clearwater complex, the St. Joe fault is expressed as a poorly defined zone of strike-slip faults and shear that has mylonitized the southern border of the ca. 52 Ma Roundtop pluton (Marvin et al., 1984). The southern boundary of the Clearwater complex is formed by the Canyon fault and Benton Creek fault (Hietanen, 1984), which is part of an 8-km-wide zone of east-west–trending strike-slip faults (Lewis et al., 2002, 2007) that extend westward from the Kelly Forks fault identified along the northwestern corner of the Bitter-root complex (Childs, 1982). Other faults, including the Clugs Jumpoff fault, splay off of the Canyon fault near the Clear-water complex. To the west, the fault system bends northward and merges with the White Rock fault on the western side of the Clearwater complex.
A relay connection between the St. Joe and the Canyon fault–Benton Creek fault system is completed by two north-south–trending normal faults, the White Rock fault on the western side and the Collins Creek fault on the eastern side of the Clearwater complex (Fig. 2). Both faults juxtapose the high-grade metamorphic rocks of the Clearwater complex against lower-grade metamorphosed Middle Belt Supergroup strata (Hietanen, 1963b; Lewis et al., 1992). The White Rock fault is a ductile-brittle detachment fault with top-to-the-west mylonites and chloritic brecciation that offset metamorphic isograds, whereas the Collins Creek fault is a top-to-the-east normal fault that has not been studied in any detail (R. Lewis, 2004, personal comm.).
In this paper, we define the Clearwater complex as a large tract of amphibolite-facies metamorphic rocks that lies between the east-west–trending St. Joe and Canyon strike-slip faults and the north-trending White Rock and Collins Creek normal faults (Fig. 2). Previous studies by Hietanen (1963a, 1984), Grover et al. (1992), Seyfert (1984), and Doughty and Sheriff (1992) considered the Clearwater complex to be a small structure centered on the high-grade Paleoproterozoic basement rocks that occur in the core of the complex. A largely intact fault block, the Boehls Butte–Goat Mountain fault block, was envisioned to host the basement rocks (Hietanen, 1984; Grover et al., 1992). Slip along the margins of this block was invoked to explain the isothermal decompression of the rocks within, as determined from metamorphic mineral assemblages and oxygen isotopes. The existence and location of these faults was inferred because of poor exposure, difficult access, and rock-type similarity across the area.
New geologic mapping during this study and by Lewis et al. (2007) significantly revises the prior mapping of Hietanen (1963a, 1963b, 1984) and the interpretations based on that mapping. Our mapping, as well as differences in rock type, strain, apparent metamorphic grade, and thermochronology, shows that the Clearwater complex is best described in terms of isolated domes, or culminations, and tectonic slivers of high-grade basement rocks that are surrounded by metasedimentary rocks. Rocks along the margins of the culminations are sheared to varying degrees, with mylonites well preserved along the eastern side of each culmination. We assign the rocks within the culminations to the internal zone of the Clearwater complex, and the rocks between the culminations and the bounding strike-slip and normal faults to the external zone of the Clear-water complex.
Rocks of the internal zone are exposed in two dome-like culminations and as fault slivers along the Clugs Jumpoff fault (Fig. 3). Internal zone rocks are composed of metamorphosed 1787 Ma Paleoproterozoic anorthosite and aluminum- and magnesium-enriched gneiss (Al-Mg–rich gneiss) that are interleaved with sill-like bodies of 1587 Ma amphibolite (Hietanen, 1956, 1963a, 1969b, 1984; Nord, 1973; Juras, 1974; Doughty and Chamberlain, 2007). Intercalation of the basement rocks with pelitic schists derived from the Belt Supergroup occurs along synmetamorphic thrust faults within the internal zone (Doughty and Chamberlain, 2007) (Fig. 3).
Metastable mineral assemblages and disequilibrium textures in metamorphic rocks of the internal zone record a complex polymetamorphic history. The oldest documented metamorphism is poorly understood and Mesoproterozoic in age, ca. 1.1 Ga (Sha et al., 2004). M1, the first regional metamorphic event, is poorly preserved in the internal zone and is best seen east of the Clearwater complex (Lang and Rice, 1985a). The second event, M2, is the peak metamorphic event, and it is characterized by the growth of kyanite and sillimanite with muscovite in Al-Mg gneisses. Quantitative estimates of the conditions for M2 range from 800 to 1100 MPa at 650 °C to 750 °C (mean of 699 °C) (Grover et al., 1992). Larson and Sharp (1998) reported temperatures for M2, based on oxygen isotopes, of 700–775 °C. So far, evidence of M3 has only been found where the unusual composition of the Al-Mg gneisses facilitated the replacement of strained kyanite with andalusite and other unique mineral assemblages (Carey et al., 1992). Conditions of metamorphism during M3 range from 400 to 600 MPa at the same temperature as M2 (Grover et al., 1992). The similarity in temperatures for M2 and M3 led Grover et al. (1992) to conclude that isothermal decompression occurred between M2 and M3.
The part of the Clearwater complex that surrounds the culminations of internal-zone basement rocks and extends out to the bounding White Rock and Collins Creek faults is defined as the external zone (Figs. 2 and 3). Hietanen (1963a, 1984) placed some of the rocks that we assign to the external zone within the uplifted Boehls Butte–Goat Mountain fault block.
Rocks of the external zone consist of rusty weathering mica schists, quartzites, and amphibolites derived from the Prichard Formation of the Middle Proterozoic Belt Supergroup metamorphosed in the garnet through kyanite-sillimanite zones of the amphibolite facies (Hietanen, 1963a, 1968; Doughty and Chamberlain, 2007). There are also several masses of Cretaceous(?) orthogneiss within this zone. These rocks are generally not sufficiently aluminous to contain aluminum silicates, although Hietanen (1963a, 1968, 1984) reported a few occurrences of kyanite or sillimanite.
Rocks along the eastern and southern flanks of the northern culmination are distinct from those found elsewhere in the external zone (Fig. 3). They consist of quartzite and fine-grained semischist that are locally calcareous enough to produce calc-silicate minerals and marbles. These rocks are very fine-grained with a weak foliation that lies at a high angle to compositional layering (relict bedding) (Fig. 4A). In many locations, the rocks exhibit very low amounts of strain, as indicated by randomly oriented porphyroblasts of tremolite or zoisite, and small poikilitic spots of orthoclase (cf. Hietanen, 1963a) (Fig. 4B). Many of the rocks, especially east of Monumental Buttes, are fine-grained, rusty weathering, siltites that look identical to the weakly metamorphosed Prichard Formation seen across much of Montana and Idaho. Small plugs of pyroxene gabbro, containing a border zone of amphibolite, intrude the metasedimentary rocks. These mafic intrusions crystallized ca. 1465 Ma (Doughty and Chamberlain, 2007).
The rocks of the external zone appear to display significant differences in composition and metamorphic history compared to the gneisses in the internal zone. Most rocks in the external zone have simple textures and low-variance mineral assemblages, and only local evidence for polymetamorphism (i.e., Lang and Rice, 1985a).
SHEAR-ZONE CHARACTERISTICS AND KINEMATICS
The contact between rocks of the internal zone and rocks of the external zone is marked by zones of intense noncoaxial shear. There are excellent exposures of these sheared contacts in glaciated cirques and ridges along Monumental Buttes, Goat Mountain, and Crescendo Peak. Elsewhere, the exposure is very poor due to intense surface weathering and thick forest growth. We have examined the contact at six different localities (Figs. 3 and 5): Jug Rock, Cedar Creek Canyon, North Monumental, and Floodwood Creek localities along the borders of the northern culmination; and Little North Fork of the Clearwater River, and Aquarius localities along the borders of the southern culmination. We also examined the Clugs Jumpoff fault as it passes through the Smith Ridge syncline (Fig. 5). Detailed structural analysis of rock fabrics and the kinematics of these shear zones are discussed next and in Sha (2004).
Northern Culmination (Jug Rock Shear Zone)
Exposures of the sheared contact between the internal and external zones are superb along the eastern edge of the northern culmination in the high country along Monumental Buttes and the Little Goat Mountains (Figs. 3 and 4). Here, shear zones are exposed on the north, east, and southern margins of the internal zone. In this paper, we define these interlinked faults and shear zones as the Jug Rock shear zone for the excellent and readily accessible exposures along the northern side of Jug Rock (Fig. 3). Along this stretch, the zone is a shallowly east-dipping ductile shear zone with downdip mineral lineations. To the north, northwest of Monumental Buttes, the Jug Rock shear zone bends sharply to the west and follows the northern edge of the anorthosite. Along this contact, it is a shallow to steep northeast-dipping shear zone with subhorizontal mineral lineations. South of Jug Rock, past Crescendo Peak, the Jug Rock shear zone turns sharply to the west and follows the southern margin of the anorthosite. Here, again, it is a steeply dipping east-west–trending zone with subhorizontal mineral lineations (Fig. 5).
The north-south–trending segment of the Jug Rock shear zone is ∼500 m thick, with gradational upper and lower boundaries. The upper boundary is a narrow zone of decreasing strain at the contact between internal zone rocks and the overlying quartzites and semischists of the external zone, which maintain a mineral lineation for a short distance above the contact. The lower boundary is a more diffuse zone of decreasing strain that occurs in the upper part of the anorthosite bodies, which change from being strongly foliated within the shear zone to massive below.
Protomylonites, mylonites, and ultramylonites are recognized within the Jug Rock shear zone. Protomylonites are developed primarily in quartz-poor lithologies, like anorthosite and plagioclase-rich Al-Mg gneiss. Mylonites and ultramylonites are best developed in mica schist and quartz-bearing Al-Mg gneiss. An increase in grain-size reduction in the mylonites correlates well with a decrease in the proportion of plagioclase in the protolith, and there is considerable heterogeneity in the character of the mylonitic fabrics across the shear zone (Sha et al., 2003).
Microstructures of the Jug Rock Shear Zone
The primary foliation in rocks of the Jug Rock shear zone formed at conditions near peak metamorphic conditions. The foliation is defined by coarse-grained reddish biotite, muscovite, fibrolite, and quartz (Fig. 4F). The quartz typically forms coarse-grained equigranular mosaics with interlobate grain boundaries indicative of static recrystallization after shearing. The micas display undulose extinction and fish-like forms. The foliation wraps around large porphyroclasts of strained kyanite, aggregates of andalusite (inverted from kyanite), garnet, and plagioclase (Fig. 4F). The porphyroclasts exhibit asymmetric sigma morphologies with tails of recrystallized quartz and pale-green biotite. Strained kyanite porphyroclasts are boudinaged with muscovite and pale-green biotite growing in the pull-apart zones. Locally, aluminosilicate porphyroclasts are surrounded by coronas of cordierite that separate the aluminosilicate from adjacent biotite. Cordierite also forms equidimensional aggregates, surrounding relict grains of andalusite, parallel to the foliation.
A second foliation overprints the primary foliation at a shallow angle (30°–40°) (Figs. 4G and 4H). Carey (1985) recognized this second fabric as a crenulation cleavage with hartschiefer texture, but these features are better interpreted as a secondary shear-band foliation related to mylonitization. The second foliation is defined by thin bands of fine-grained reddish biotite intergrown with fibrolite and dynamically recrystallized quartz grains (Figs. 4G and 4H). The average grain sizes, 0.08–0.04 mm, contrast sharply with the coarse-grained size of minerals that form the primary foliation. Most of the quartz grains have recrystallized into a fine-grained granoblastic aggregate, but locally, there are preserved quartz ribbons with undulose extinction and small elongate subgrains (0.005 mm).
Bands of ultramylonite, 2–10 cm wide, occur locally in mica schists, Al-Mg gneiss, and anorthosite. In outcrop, they form glassy, black resistant bands. The ultramylonites have a matrix of very fine-grained dark-brown biotite (0.0025 mm) enclosing small (0.5–.08 mm) rounded porphyroclasts of quartz, tourmaline, garnet, kyanite/andalusite, and aggregates of dynamically recrystallized, unannealed quartz (Figs. 4I and 4J). Sigma-type porphyroclasts of andalusite often contain tails of pale-green biotite or chlorite (Fig. 4K). Dynamic recrystallization has produced an S-C fabric with subgrains that display a strong preferred orientation at a high angle to the primary foliation.
Initial formation of the primary foliation, S1, coincides with the growth of peak M2 minerals like fibrolite and the micas, which define the foliation. Kyanite, which contains epitaxial overgrowths of fibrolite and forms strained porphyroclasts, could predate the formation of S1, although Carey (1985) concluded that it grew during the early stages of S1 development. Because S1 contains asymmetric porphyroclasts and mica fish that yield the same sense of shear as that recorded by S2 fabrics, we infer that mylonitization began during uplift of the rocks from the kyanite field into the sillimanite field. Minerals deformed by this foliation are normally annealed, which reflects the high temperatures present during mylonitization. The shear-band foliation, S2, coincides with the replacement of strained kyanite by andalusite, growth of secondary biotite, muscovite, and chlorite, and unannealed mylonitic textures. It also contains mats of fibrolite that appear to have grown along S2. Thus, S2 initially formed in the sillimanite stability field but continued to develop as the rocks were uplifted into the stability field of andalusite and chlorite, at considerably less pressure and temperature than S1. It records the strain associated with the exhumation of the Goat Mountain area to the surface. These relationships indicate that the Jug Rock shear zone formed during progressive unroofing of the rocks in the Clearwater complex.
S-C fabrics, shear-band foliations, and rotated porphyro-clasts from all three segments of the Jug Rock shear zone were used to constrain the kinematics of motion. The north-trending segment of the Jug Rock shear zone moved in a consistent down-dip top-to-the-east direction along an azimuth of 100° (Fig. 5). The east-west–trending northern segment of the Jug Rock shear zone, northwest of Monumental Buttes, moved horizontally in a right-lateral sense along an azimuth of 108° (Fig. 5). The east-west–trending southern segment of the Jug Rock shear zone, northwest of Cedar Creek, moved subhorizontally in a left-lateral sense along an azimuth of 273° (Fig. 5). These kinematic indicators document exhumation of the internal zone by east-directed transport on the Jug Rock shear zone.
Floodwood Creek (Western Boundary of Northern Culmination)
The western boundary of the northern culmination was interpreted by Hietanen (1963a, 1984) as a steep north-south–trending normal fault (Orphan Point fault), largely on the basis of a sharp demarcation between anorthosite and external zone rocks to the west. This fault was mapped as extending to the north and south of the northern culmination. During the course of our mapping, we found no obvious field evidence (such as brecciation, alteration) or minor structures supporting the presence of a steep brittle fault along the western edge of the northern culmination. The exposure, however, is poor, and there are no locations where the contact is exposed. Our interpretation of the nature of the contact is based on examination of structures and fabrics in rocks adjacent to the contact.
Rocks west of the contact in the external zone vary from weakly foliated fine-grained semischists in the south (Floodwood Creek) to mylonitic granite and metasedimentary rocks along the rugged alpine ridges of Orphan Point and Widow Mountain in the north (Fig. 3). The mylonites in the north are herein named the Widow Mountain shear zone for the excellent exposures on that mountain. The mylonites dip gently to the east with downdip lineations, but they have kinematic indicators with updip, top-to-the-west sense of shear. The mylonite zone is wholly contained within the external zone and appears to be unrelated to the juxtaposition between internal and external zone rocks to the east (but see following). The fault's updip kinematics suggest that it could be Mesozoic in age, or a younger rotated structure. If the latter, it could be part of the top-to-the-west White Rock detachment fault, which bounds the western margin of the Clearwater complex ∼20 km to the west (R. Lewis, 2004, personal commun.).
The dominant fabric present in rocks east of the contact is a gently west-dipping annealed and recrystallized foliation in anorthosite. Xenoliths of foliated anorthosite within sills of foliated hornblende-biotite orthogneiss show that this fabric is older than the orthogneiss; it is also older than stocks of foliated biotite granite along the western side of the northern culmination (Fig. 3). Lewis et al. (2007) inferred that the orthogneisses are Cretaceous in age, but without actual dates, this old fabric could range from Precambrian to Eocene in age.
A younger second, and possibly third, deformation event is evident in these rocks. Spaced centimeter-scale bands of ultra-mylonite crosscut the foliated anorthosite and sills of orthogneiss, and stocks of granite are penetratively deformed into protomylonites, mylonites, and centimeter-scale zones of ultra-mylonite. Many of these mylonitic granite stocks are similar to granites dated at 48 Ma along the southern border of the complex (Burmester et al., 2004). We conclude that this fabric is Eocene in age and it is related to exhumation of the Clearwater complex. The unannealed nature of the fabrics and presence of chlorite within mineral pull-aparts and within the mylonitic foliation show that these second-event mylonites formed at significantly lower temperatures and pressures than the primary metamorphic foliation. The second-event fabrics strike north with gentle east or west dips and downdip mineral lineations (Fig. 5). Crosscutting relationships suggest that some of the top-to-the-east mylonites are younger than the top-to-the-west mylonites. Kinematic indicators yield consistent downdip shear in all cases.
Unlike the Jug Rock shear zone, the western boundary of the internal zone along Floodwood Creek lacks a unidirectional sense of shear. There is no systematic difference between the second event west-directed and east-directed fabrics, and all rock types contain mylonite fabrics with both senses of shear. These mylonites could have formed as conjugates of one another in a regime of pure shear, or they could indicate episodes of deformation with opposing shear sense within the core of the Clearwater complex. Our preferred interpretation is that the dominant west-dipping fabric is Eocene in age and part of the Jug Rock shear zone that has been arched and back-rotated during formation of the Clearwater complex. The younger second-event mylonites formed after the initial arching of the internal zone and record overlapping top-to-the-west and top-to-the-east shear in the foot-wall of the White Rock–Widow Mountain shear zone and Collins Creek fault. Similar back-dipping mylonite zones with a shear antithetic to the main detachment are well known from near the mylonitic front of other core complexes (Reynolds and Lister, 1990; Axen and Bartley, 1997), including the Bitterroot complex (see cross section in Foster et al., 2007).
Due to very poor exposure, there were only two localities where we examined the boundary of the southern culmination in any detail. These were the Little North Fork of the Clearwater River at Dworshak Reservoir and Aquarius along the northern and eastern margins of the southern culmination, respectively (Fig. 5). At both locations, the contact is characterized by decimeters of sheared Al-Mg gneiss and anorthosite. The sheared rocks exhibit gradational upper and lower boundaries, partially annealed microstructures, heterogeneous distribution related to protolith composition, and consistent shear sense among kinematic indicators.
Little North Fork of the Clearwater River
The northern boundary of the southern culmination is best exposed where the Little North Fork of the Clearwater River enters Dworshak Reservoir (Fig. 5). At this locality, mylonitized plagioclase-rich gneiss exhibits a predominately gently northeast-dipping foliation with nearly subhorizontal mineral lineations (Fig. 5). The foliation is defined by coarse-grained sillimanite, quartz, and kinked kyanite; most of the kyanite has been replaced by andalusite, except where wholly surrounded by plagioclase porphyroclasts. A second foliation, which is north-south–oriented with gentle east dips and down-dip lineations, crenulates the first foliation. This fabric is defined by unannealed quartz ribbons, and fibrolite intergrown with fine-grained muscovite and talc. The second foliation formed during mylonitization at lower pressures and temperatures than that preserved in the first foliation, consistent with its development during uplift and exhumation of the anorthosite in the southern culmination.
Kinematic indicators yield a subhorizontal, right-lateral sense of shear along an azimuth of 103° for the older foliation and a downdip, normal (top-to-the-east) sense of shear along an azimuth of 088° for the second foliation. Both of these fabrics are consistent with shearing of the anorthosite during exhumation of the southern culmination by translation westward relative to the flanking rocks.
The eastern boundary of the southern culmination is well exposed along the north side of Dworshak Reservoir, near where the North Fork of the Clearwater River enters the reservoir at Aquarius (Fig. 5). At this locality, anorthosite beneath the contact exhibits a gently east-northeast–dipping foliation with a nearly east-trending mineral lineation. It contains the typical internal zone assemblage (kyanite-andalusite-sillimanite-rutile), textures, and parageneses found at Goat Mountain and in the Jug Rock shear zone. Metasedimentary rocks above the anorthosite display a similar attitude with an east-dipping foliation and east-trending lineation defined by fibrolite, biotite, and recrystallized quartz. The foliation wraps around porphyroclasts of plagioclase, andalusite, and minor muscovite. Rare kyanite grains are preserved within the core of plagioclase porphyroclasts. The andalusite differs from that typically found within the inner core zone. They are commonly embayed remnants of euhedral andalusite porphyroclasts with marked pleochroism. Fibrolite and muscovite surround and overgrow the andalusite.
Asymmetric porphyroclasts and crenulated foliations indicate a downdip, top-to-the-east, sense of shear along an azimuth of 081° for the rocks along this contact, very similar to the fabrics in the north-south segment of the Jug Rock shear system, and exhumation of the internal zone rocks of the southern culmination by translation to the west relative to the overlying external zone rocks (Fig. 5).
Smith Ridge Syncline
Clugs Jumpoff Fault
The area between the two large culminations of anorthosite and Al-Mg gneiss is underlain by an east-west–trending band of metasedimentary and meta-igneous rocks of the external zone, referred to as the Smith Ridge syncline by Lewis et al. (2007). Recent mapping by us and Lewis et al. (2007) has identified a major fault within this band of external zone rocks. This fault, termed the Clugs Jumpoff fault by Lewis et al. (2007) (Figs. 3 and 4), has been recognized by the presence of sparse east-west–trending masses of anorthosite and Al-Mg–rich gneiss along its very poorly exposed trace. The Clugs Jumpoff fault zone runs along the southern edge of the fine-grained quartzites and marbles that flank the northern culmination and subdivides the Smith Ridge Syncline in two (Figs. 3 and 4). The eastern and western extents of the fault are not completely mapped, but it extends past Floodwood Creek to the west and as far east as Clugs Jumpoff. We postulate that it continues its east-west trend and merges with the Canyon fault–Kelly Forks fault system somewhere east of the southern mass of anorthosite, but we do not have any idea where it extends to the west (Fig. 3).
Sheared rocks within the Clugs Jumpoff fault are shallowly southwest dipping with subhorizontal lineations along an azimuth of 109° (Fig. 5). A kinematic indicator on the north side of one anorthosite body yields a left-lateral sense of shear, whereas a kinematic indicator on the south side of another anorthosite body yields a right-lateral sense of shear. In the absence of more complete or compelling data, we conclude that the anorthosite was translated to the west relative to the rocks on either side of the Clugs Jumpoff fault.
Eight metamorphic rocks from the external zone of the Clearwater complex contain mineral assemblages suitable for the application of metamorphic thermobarometers. We employed quantitative thermobarometry in an attempt to better characterize the metamorphic conditions of the external zone. Six samples were from the domain of fine-grained metasediments and pyroxene gabbros adjacent to the Jug Rock shear zone (Fig. 3). Two samples (01-300, 01-316) were fine-grained quartz-rich schists that had a weak foliation at a high angle to relict bedding (Fig. 4A). Garnet porphyroblasts (1.5–3 mm in diameter) occur in one of two forms, subhedral, slightly embayed with inclusions of quartz, or subhedral with a core rich in quartz inclusions surrounded by a wide inclusion-poor rim. The typical mineral assemblage is garnet-biotite-muscovite-plagioclase-quartz-oxide (ilmenite, or magnetite) + pyrite or pyrrhotite. Some of the more calcareous samples contain tremolite, zoisite, or diopside (Fig. 4B). Two samples (01-313, 01–234) were garnet-bearing amphibolite from the border zone of pyroxene gabbro stocks. These amphibolites are very fine grained and contain small, 1 mm garnets in a foliated matrix of hornblende-plagioclase-quartz ± biotite (Fig. 4C). A third sample of amphibolite (01-358) was collected from Cedar Creek Canyon, south of the northern mass of anorthosite. This amphibolite is composed of coarse-grained hornblende-plagioclase-quartz-biotite with large embayed garnets surrounded by a symplectite corona of plagioclase and hornblende (Fig. 4E).
Sample (01-302) was from coarse-grained garnet-biotite-muscovite schist 2 km east of the Jug Rock shear zone. This sample is typical of the coarse-grained metamorphic rocks that characterize much of the external zone of the Clearwater complex. These schists have the same composition and mineral assemblages as the previously described samples.
Sample (02-18) was from coarse-grained mica schist that occurs between the Clugs Jumpoff fault and the southern mass of anorthosite (Fig. 3). Sample 02-18 is a coarse-grained quartzofeld-spathic gneiss containing a metamorphic foliation defined by biotite, minor muscovite, fibrolite, and thin (1–2 mm) segregations of quartz and feldspar. Garnets in 02-18 are very small, anhedral, and embayed in part. The foliation wraps around the garnets, and fibrolite needles and small biotite crystals locally grow around the garnet in pressure shadows at a high angle to foliation (Fig. 4D).
Mineral compositions were analyzed at the Washington State University (WSU) Geoanalytical Laboratory, and results are reported in Table 1, A–E 102103104105106. Sample 02-18 contains fibrolite, and we employed the relatively robust garnet-aluminosilicate-plagioclase-quartz (GASP) barometer, using the calibration of Koziol and Newton (1988) for that sample. No other metasedimentary samples contained aluminosilicate, and we employed the less robust garnet-biotite-muscovite-plagioclase-quartz (GPMQ, GPMB) barometers using the calibrations of Hodges and Crowley (1985). This calibration was chosen because it allows a direct comparison with the results of Carey (1985), who applied the same barometer to rocks in the internal zone. For amphibolites, we employed the garnet-hornblende barometer (GHPQ) of Kohn and Spear (1990). Temperatures were constrained with the garnet-biotite (GARB) exchange thermometer, using the calibration of Ferry and Spear (1978) with the garnet mixing model of Berman (1990), and the garnet-hornblende thermometer (GAHB) of Graham and Powell (1984). We also compared our computed temperatures with the Ganguly and Saxena (1984) and Indares and Martignole (1985) calibrations, which were used by previous workers in the area. Calculations were performed with the program GTB, available from Frank Spear (http://ees2.geo.rpi.edu/MetaPetaRen/GTB_Prog/GTB.html). All iron was assumed to be ferrous, based on very low ferric iron content (∼0.0019) in coexisting ilmenite and the report of very little ferric iron in similar rocks near Snow Peak (Lang and Rice, 1985b). Garnets are almandine-rich (70%–78% Fe), with small amounts of spessartine component (3%–19% Mn) and (Ca + Mn)/total cation ratios between 0.55% and 0.08%. The biotites contain 0.1%–0.03% (Al6 + Ti)/(Al6 + Ti + Fe + Mg) and Al6 contents between 0.194 and 0.476. Plagioclase ranges in composition from An17 to An87. Mineral compositions are generally within the range of compositions required for the application of these calibrations.
Results of the thermobarometric calculations are reported in Table 2, A–B 202, and Figure 6. Despite complexities, seven of the eight samples equilibrated within or near the kyanite stability field. Sample 02-18, which contains fibrolite, equilibrated in the sillimanite stability field as expected. The computed pressure and temperature conditions are consistent with the distribution of regional metamorphic isograds that places these rocks above the staurolite breakdown reaction, below the muscovite out reaction, and within or near the stability field of kyanite (Hietanen, 1963a, 1968; Lang and Rice, 1985b; Grover et al., 1992). Transects across garnets from 01 to 300 and 01-302 display a flat major-element profile in the core, with increases in Ca, Fe/Mg ratio, and Mn at the rim. This flat zoning profile is characteristic of garnets from high-grade rocks that have undergone high-temperature homogenization (e.g., Tracy et al., 1976; Frost and Chacko, 1989). The garnet rim compositions are believed to best represent the conditions of peak metamorphism in these rocks, although some of the variability in our results could be due to partial reequilibration of the garnet rims during cooling.
Metamorphic temperatures calculated with rim compositions range between 595 °C and >800 °C, depending on the calibration and thermometer used (Fig. 6). The garnet-hornblende thermometer, applied to three amphibolites with the calibration of Graham and Powell (1984), gives temperatures between 650 °C and 700 °C, with an average of 677 °C. Temperatures calculated with the garnet-biotite thermometer vary widely depending on the calibration used. The calibration of Ferry and Spear (1978) with Berman's (1990) mixing properties of garnet yields temperatures between 580 °C to 850 °C (average of 683 °C), which is anomalously high for some samples. Grover et al. (1992) and Carey et al. (1992) also struggled with anomalously high calculated temperatures in rocks of the internal zone, which they believed was due to oxidation of biotite during the last metamorphic event (M3). To overcome this problem, Grover et al. (1992) employed the Indares and Martignole (1985) calibration, whereas Carey (1985) employed the Ganguly and Saxena (1984) calibration. We examined these other calibrations in order to potentially obtain a better estimate of the temperatures and to facilitate direct comparisons of peak metamorphic pressures between our study and theirs. The Indares and Martignole (1985) calibration yields widely scattered temperatures between 100 °C less to 200 °C higher than the Ferry and Spear (1978) and Berman (1990) calibration. The Ganguly and Saxena (1984) calibration yields a tight cluster of temperatures between 575 °C and 660 °C (average of 621 °C) with only two temperatures above 850 °C. This calibration appears to provide the most robust estimate of temperature in these rocks and overlaps with temperatures derived from the garnet-hornblende thermometer, and it is consistent with the presence of primary muscovite, which constrains temperatures to be below the breakdown of muscovite + quartz at ∼700 °C. Our preferred estimate of the temperature, which includes both the biotite-garnet and garnet-hornblende thermometers, is 634 °C.
Calculated pressures fall naturally into two groups that vary between ∼900–1000 MPa or 500–600 MPa (Fig. 6). Four samples comprise the high-pressure group. Schist 01-302 and amphibolite 01–234, from east of the Jug Rock shear zone, have simple metamorphic textures that grew during one metamorphic event (Fig. 4C). Sample 01-302 equilibrated at ∼870 MPa and 660 °C, based on the GPMQ and GPMB barometers. Amphibolite 01–234 gives a higher pressure and temperature of 1000 MPa at 700 °C with the GAPQ barometer. Ziegler (1991) also obtained high pressures (1050 and 1200 MPa) from two amphibolites east of the study area.
Two of the samples from the high-pressure group (02-18 and 01-358) lie between the two culminations and have more complex textures that suggest more than one metamorphic event. Sample 01-358 is a coarse-grained amphibolite with symplectite intergrowths of hornblende and plagioclase surrounding embayed garnets (Fig. 4E). Matrix phases and core garnet compositions give a pressure and temperature of ∼925 MPa and 630 °C using the GAPQ barometer and GARB/GAHB thermometers (Fig. 6). Reliable estimates of the conditions of equilibration during symplectite growth of hornblende and plagioclase were not obtained.
Paragneiss 02-18, which is the only sample to contain an aluminosilicate phase in our suite of samples, gives conflicting results. The average composition of garnet, biotite, and plagioclase from sample 02-18 yields temperatures greater than 800 °C and pressures greater than 1200 MPa with the GASP, GPMQ, and GPMB barometers (Fig. 6). If more reasonable temperatures are used (the average of 634 °C for all samples), the pressure only drops to 950 MPa. One small part of the sample contains fibrolite overgrowing a small anhedral garnet porphyroblast (Fig. 4D). Mineral compositions from this part of the thin section yield a tight intersection of calculated equilibria at ∼475 MPa and 600 °C (Fig. 6). These data suggest that final equilibration at ∼5 MPa and 600 °C occurred after an earlier high-pressure metamorphic event.
Two samples (01-300, 01-313), from right above the Jug Rock shear zone, comprise the lower-pressure group. Both samples have very simple, fine-grained textures indicative of only one episode of metamorphic growth (Figs. 4A and 4C). Geo thermo barometers applicable to semischist 01-300 and amphibo lite 01-313 are in good agreement and give pressures and temperatures of ∼600 MPa and 619 °C.
Quantitative geothermobarometry, coupled with petrographic observation, suggests that both high-pressure and intermediate-pressure metamorphic events are present within rocks of the external zone. The presence of nonequilibrium textures in two samples and the scattered distribution of sample localities make it unlikely that a geologic structure is solely responsible for all of the different metamorphic pressures observed in the external zone. High-pressure and intermediate-pressure metamorphism occurs in both amphibolites and metasediments, with apparently simple metamorphic histories. In these rocks, either some of the pressure determinations are erroneously high (the lack of aluminosilicate in most metasediments is a concern), or recrystallization of older metamorphic minerals during subsequent metamorphic events was highly variable. Samples of this type with the lowest pressures (01-300 and 01-313) lie close to the top of the Jug Rock shear zone and could have been affected by their proximity to the fault. Perhaps the strongest evidence for more than one metamorphic event comes from the coronas of intergrown hornblende and plagioclase around embayed garnet porphyroblasts in sample 01-358 (Fig. 4E). These coronas are strikingly similar to garnet coronas documented from other core complexes that have undergone nearly isothermal decompression (e.g., Ziegler, 1991; House et al., 1997), and they argue strongly that at least part of the external zone has undergone an episode of exhumation and high-temperature decompression. If true, the process by which some samples could either escape recrystallization or undergo complete recrystallization during this event requires further investigation.
The high-pressure metamorphism recorded in the external zone occurred at conditions of ∼900 MPa and 650 °C, which is similar to the conditions of M2 in the internal zone, and this suggests that the two are correlative. Our estimate for the conditions of M2 includes pressures based on the GAPQ barometer from two amphibolites (01–234, 01-358). Ziegler (1991) showed that that the GAPQ barometer overestimated the metamorphic pressures by 100–300 MPa relative to the more robust GRIPS barometer when applied to amphibolites in both the internal and external zones of the Clearwater complex. If this is the case, our estimates for M2 in the external zone could be 100–200 MPa too high. The low-pressure metamorphism in three samples records conditions of around 600 MPa and 623 °C, which could either correlate with the older M1 event (i.e., Lang and Rice, 1985b) in some samples or, more likely, the M3 exhumation event that is so well documented in the internal zone (Carey et al., 1992).
U-Pb SHRIMP GEOCHRONOLOGY
In order to place constraints on the timing of metamorphism and deformation in rocks of the Clearwater Complex, we report new U-Pb dating of metamorphic overgrowths on zircons from rocks within the Jug Rock shear zone. These data were collected during a study of detrital and igneous zircon ages for rocks within the Clearwater complex (Doughty and Chamberlain, 2007), and more detailed sample descriptions and localities are described in that publication and in Figure 3.
We collected a total of 29 analyses of metamorphic overgrowths on zircons from four localities. Three samples (GM01-02, 02-93, and 02-106B) were collected from rocks below or within the Jug Rock shear zone. One sample (GM01-05) was collected from above the Jug Rock shear zone.
Zircon grains were extracted and concentrated with conventional mineral separation techniques at the University of Wyoming. All samples were analyzed at the Stanford–U.S. Geological Survey facility using the sensitive high-resolution ion microprobe (SHRIMP-RG [Reverse Geometry]) instrument. Data reduction followed Ludwig (1988, 1991, 2003). We report 206Pb/238U dates corrected by the 207Pb method (e.g., Williams, 1998) and 2σ errors, but our interpretations are based on concordia intercepts from total Pb Tera-Wasserburg (1972) plots to eliminate the influence of common-Pb correction choices. Although many of the analyses have high common Pb (4%–30% of 206Pb; Table 3), each age population has at least one concordant to nearly concordant analysis with low common Pb (less than 2%), which strengthens our interpretation that discordance is related to common Pb. Inheritance is unlikely in these data because each spot was positioned in a discrete cathodoluminescent (CL) domain, and the SHRIMP-RG pits are only a few microns deep.
The interpretation of metamorphic origin for these rims relies on the relatively homogeneous, unzoned nature of the domains in CL images (Fig. 7) and distinctive Th/U values (Table 3). The rims are either bright or dark in CL, but they are always distinct from the zoned magmatic and detrital cores of the grains (Fig. 7). Low Th/U values (≤0.02) are often characteristic of metamorphic zircon growth (e.g., Williams, 1998) and can be diagnostic especially when combined with textural evidence. Low Th/U zircons can also grow in some magmatic systems, although values of 0.2–0.7 are more typical. In the samples from Clearwater complex, the combination of low Th/U, CL evidence for overgrowths, and the tectonic interpretation of a Mesoproterozoic origin for these rocks (Doughty and Chamberlain, 2007) leads us to interpret the rims as metamorphic zircon growth.
Results (Internal Zone)
Sample GM01-02 (Amphibolite of Moses Butte)
Sample GM01-02 is a garnet amphibolite collected ∼500 m west of the Jug Rock shear zone (Fig. 3). We analyzed metamorphic overgrowths on seven zircons from this sample (Table 3; Fig. 8). The zircons in sample GM01-02 contain small embayed and bleb-like, zoned cores that are dark in CL, Mesoproterozoic in age (Doughty and Chamberlain, 2007), and are surrounded by banded to complexly zoned U-rich, euhedral rims (Fig. 7C) and pale, unzoned rims (Fig. 7D). In many grains, the core composes only 20% of the grain (Fig. 7D), and in some grains, there are no distinct cores at all. Data from eight spots fall into two groups with concordia intercepts of 64 ± 1.0 and 59 ± 1.9 Ma (Fig. 8A). These spots sampled a variety of CL domains and included edges, centers, and homogeneous interiors of grains, although the two youngest spots sampled unzoned rims (Fig. 7D). Zircon growth in this rock appears to be dominantly metamorphic at 65–57 Ma. Analyses 2.1 and 2.2 are from the center and rim of the same grain and yield the same ages within error. Six of the eight analyses are nearly concordant (Fig. 8A), with 206Pb/238U dates that range from 58.9 ± 1.6 Ma to 64.9 ± 2.6 Ma (Table 3). On a Tera-Wasserburg (1972) plot of total Pb, the data are consistent with two periods of zircon growth and varying degrees of common Pb producing linear scatter. Both age groups contain concordant analyses with low common Pb. Coupled with the CL evidence, we interpret the data to indicate that metamorphism occurred in two pulses ca. 64 and 59 Ma.
Sample 02-106b (Schist of Blackdome Peak) and Sample 02-93 (Schist of South Butte)
Samples 02-106b and 02-93 are both quartz-rich schists from the internal zone that are caught up in the Jug Rock shear zone. We analyzed metamorphic overgrowths on one grain from sample 02-106b and four grains from sample 02-93 (Table 3; Fig. 7). Sample 02-106b contains rounded detrital zircon cores overgrown by thin euhedral rims, which are bright in CL. The zircons from sample 02-93 are substantially more complex, with both banded, dark overgrowths and bright homogeneous rims (Figs. 7A, 7B, and 7H). Fractured cores overgrown by bright overgrowths are also present, suggesting that deformation occurred prior to metamorphic zircon growth (Figs. 7A, 7B, and 7H). One grain (3) exhibits two periods of metamorphic rim growth. The inner overgrowth (3.2) surrounds a rounded detrital core and displays a complex pattern of zoning. The outer rim (3.1) is a light-colored banded rim identical to those observed on the other grains (Fig. 7B). One analysis of the outer metamorphic overgrowth from sample 02-106b yielded a 206Pb/238U date of 66.9 ± 1 Ma (Table 3; Fig. 8C). Two concordant analyses of the inner metamorphic overgrowth from two grains in 02-93 have 206Pb/238U ages of 63.5 ± 2.6 and 64.2 ± 2 Ma (Table 3; Fig. 8C). Outer, bright overgrowths from sample 02-93 have ages of 54.2 ± 4 and 56.9 ± 2.4 Ma, with an intercept age of 57.5 ± 2.9 Ma. The textural evidence from 02-93, combined with two distinct rim ages, supports the interpretation from GM01-02 that the internal zone was flushed by at least two pulses of metamorphic fluids ca. 64 and 59–55 Ma.
Results (External Zone)
Sample GM01-05 (Quartzite of Monumental Buttes)
Sample GM01-05 is a clean quartzite within the metasedimentary rocks east of the Jug Rock shear zone (Fig. 3). We analyzed seven metamorphic overgrowths on five detrital grains from sample GM01-05 (Table 3; Figs. 8D and 8E). Sample GM01-05 contains rounded detrital cores that have been overgrown by two periods of metamorphic rim growth (Figs. 7E, 7F, and 7G). The inner overgrowth is dark-colored in CL due to low U content and well-faceted to embayed. The outer rim is very thin, light-colored (due to higher U content), and euhedral. We dated the inner dark overgrowth from five grains with six analyses. The results cluster into two ages, 80–82 Ma and 72–74 Ma (Table 3; Figs. 8D and 8E). Analysis of one bright rim on one of these grains (14, Fig. 7F) overlaps the age of the younger group. Although these analyses have variable and fairly high concentrations of common Pb, the two ages are robust, and each group contains nearly concordant analyses with low common Pb. As a whole, these results establish that metamorphic zircon growth occurred in at least two pulses in the external zone and that they are distinctly older than the metamorphic growths in the internal zone.
In order to place constraints on the timing of cooling and tectonic exhumation of the Clearwater complex, we report new 40Ar/39Ar dating of micas from metamorphic and igneous rocks within both the internal and external zones. Seven samples were collected from igneous (RTP-13, BBDF-01, -02, -03) and metasedimentary rocks (01-356, 01-229, 01-224) from the external zone. One sample (01-329) was collected from mica schist in the internal zone.
Biotite and muscovite were separated from whole-rock samples using conventional crushing, magnetic, and heavy liquid methods followed by hand selecting grains. Samples were wrapped in Al foil and stacked in a fused silica tube with the neutron flux monitor GA1550 biotite (98.5 ± 0.8 Ma: Spell and McDougall, 2003). Samples were irradiated at the Oregon State reactor facility. Correction factors for interfering neutron reactions on K and Ca were determined by analysis of K-glass and optical-grade CaF2 included in the irradiation, and the following values were used: (40Ar/39Ar)K = 2.66 × 10−2, (36Ar/37Ar)Ca = 2.70 × 10−4, and (39Ar/37Ar)Ca = 6.76 × 10−4. Following irradiation, mica grains or groups of grains were heated using a CO2 laser. The laser beam was defocused to ensure roughly uniform heating, and step heating was performed by changing the power output of the laser. Reactive gases were removed by two SAES GP-50 getters prior to expansion into to a MAP215–50 mass spectrometer. Peak intensities were measured using a Balzers electron multiplier. Mass spectrometer discrimination and sensitivity were monitored by analysis of atmospheric argon aliquots from an online pipette system. The sensitivity of the mass spectrometer was ∼6 × 10−17 mol mV−1. The data were also corrected for line blanks analyzed at regular intervals between the unknowns. Ages were calculated using a 40K decay constant of 5.543 × 10−10 yr−1 and are reported with 2σ errors.
Results (External Zone)
Sample RTP-13 (Roundtop Pluton)
Sample (RTP-13) is a coarse-grained hornblende-biotite granite obtained from the Roundtop pluton, intruded into the external zone ca. 52 Ma (Marvin et al., 1984) (Fig. 3). Biotite separated from sample RTP-13 gave a well-defined plateau age of 47.2 ± 0.8 Ma for ∼90% of the 39Ar released (Fig. 9A). The inverse isochron age for these steps is 47.6 ± 1.3 Ma (Fig. 9B).
Sample 01-356 (Schist of Cedar Creek)
Sample 01-356 is a coarse-grained calcareous schist exposed along the Cedar Creek Canyon road. Biotite from sample 01-356 gave a strongly discordant age spectrum with a total fusion age of 180.6 ± 2.2 Ma. The low- and high-temperature apparent ages are >200 Ma and drop down to a minimum age of ca. 102 Ma for the intermediate-temperature step (Table 4) 402. The shape of this age spectra is indicative of significant excess argon contamination, and, therefore, no thermo-chronological information is given by this sample.
Samples 01-229 and 01-224 (Schist of Monumental Buttes)
Samples 01-229 and 01-224 are from fine-grained mica schist-semischist ∼2 km east of the Jug Rock shear zone (Fig. 3). Muscovite separated from sample 01-229 gave a relatively flat age spectrum with a total fusion age of 58.2 ± 3.8 Ma (Fig. 9C). An inverse isochron for the four steps gives an age of 53.7 ± 2.5 Ma and an initial ratio of 40Ar/36Ar ratio greater than atmosphere (Fig. 9D). The isochron age is considered to be the best estimate of the cooling age for this muscovite because the trapped component is homogeneous and nonatmospheric in composition.
Samples BBDF-01, -02, -03 (Granitoids)
Three samples of foliated granitoids were from the western external zone, west of the Widow Mountain shear zone and east of the White Rock fault. Samples BBDF-01 and BBDF-02 were closest to the White Rock fault, and they give concordant plateau ages of 42.4 ± 0.5 Ma and 41.4 ± 0.7 Ma, respectively (Figs. 10A and 10B). BBDF-03, which came from just west of the Widow Mountain shear zone, gives a plateau age of 46.5 ± 1.0 Ma (Fig. 10C). All three of these plateau ages comprise great than 98% of the gas released from the respective samples and are considered to be robust cooling ages.
Results (Internal Zone)
Sample 01-329 (Schist of Goat Mountain)
Sample 01-329 is from a coarse-grained mica schist exposed along the north flank of Goat Mountain below the Jug Rock shear zone. Analyses of three grains of muscovite from sample 01-329 gave a discordant age spectrum with a total fusion age of 92.9 ± 2.7 Ma (Fig. 9E). A second analysis of this sample gave a plateau age of 52.5 ± 2.1 Ma for nearly 90% of the gas released (Fig. 9E). An addition total fusion analysis of one grain of muscovite from this sample gave a total fusion age of 47.3 ± 1.8 Ma (Table 4) 402. The discordant age spectrum for the first analysis was due to excess argon. The time that this sample cooled below ∼400–350 °C was probably either ca. 53 Ma or as young as ca. 47 Ma.
Excess argon was a significant problem for several of the samples, and it provides an explanation for the spread in K-Ar ages obtained by Hietanen (1969a). This required analyzing small single grains, or two to three grains at a time to find some without excess argon. This resulted in most of the age spectra having a limited number of steps. The cooling age of the eastern part of the external zone below muscovite closure was probably ca. 54 Ma, based on sample 01-229. The internal zone was probably cooling at the same time but may not have cooled below ∼300 °C until ca. 47 Ma, based on the total fusion age of the small single grain from sample 01-329.
The youngest apparent ages are recorded by samples in the footwall of the White Rock fault in the western part of the external zone (BBDF-01, -02, and -03). Samples closest to the White Rock fault did not cool below ∼350–300 °C until ca. 41–42 Ma.
The timing and pattern of exhumation of the Clearwater complex are bracketed by 40Ar/39Ar mica cooling ages from the complex that are between ca. 54 and 42 Ma (Fig. 3). Rocks in the eastern half of the external zone cooled first, below the muscovite closure temperature at ca. 54 Ma, based on sample 01-229 (Fig. 11). The internal zone of the complex cooled below muscovite closure slightly later, between 53 and 47 Ma, based on sample 01-329. The western half of the external zone cooled through biotite closure shortly thereafter in a west to east direction between 47 Ma to ca. 41 Ma, based on samples BFDF-01 to BFDF-03 (Figs. 3 and 11). This pattern of westward progressive cooling is consistent with exhumation and cooling of the Clear-water complex primarily by the west-directed White Rock fault and underlying Widow Mountain shear zone. A sharp break in cooling ages across the Jug Rock shear zone suggests that it was active during the early stages of exhumation, but was abandoned when the White Rock fault became the dominant structure sometime between 54 and 47 Ma (Fig. 11). The role of the Collins Creek fault in exhuming the complex is poorly constrained by our sparse data set, but, it was likely active during formation and exhumation of the Clearwater complex as well.
Metamorphic events in the Clearwater complex are bracketed by new U-Pb dating of metamorphic zircon overgrowths. The external zone preserves two pulses of Late Cretaceous metamorphism (ca. 72–74 Ma and 80–82 Ma) in rocks that equilibrated under conditions of ∼900 MPa and ∼650 °C before exhumation and local reequilibration of some mineral phases at ∼600 MPa and 623 °C prior to 54 Ma. Our estimate for the timing of peak M2 metamorphism in the Clearwater region is consistent with that of Grover et al. (1993), and the timing of peak metamorphism in the Shuswap complex and Priest River complex (i.e., Parrish, 1995; Doughty et al., 1998), and the early phase of high-grade metamorphism in the Bitterroot complex between 75 and 80 Ma (Toth and Stacey, 1992; Foster and Fanning, 1997; Foster et al., 2001). In contrast, rocks inside the culminations, which contain M2 minerals strongly overprinted by M3 metamorphism, underwent later metamorphic zircon growth during two episodes of fluid migration around 64 and between 59 and 55 Ma, and there is no evidence of Cretaceous metamorphism. The age of metamorphism inside the culminations documented here overlaps with the age of “main-phase” igneous activity and high-grade metamorphism in the Bitterroot complex between 65 and 53 Ma (House et al., 1997, 2002; Foster et al., 2001) and was probably related to that regional event. The youngest metamorphic zircon growth, however, could be related to an influx of hot meteoric water into the Clearwater complex during the initial phase of exhumation (i.e., Walker, 1993; Larson and Sharp, 1998; Mora et al., 1999).
The internal structure of the Clearwater complex is dominated by two doubly-plunging culminations of basement rocks (internal zone) that have been exhumed relative to the surrounding metasedimentary rocks of the external zone. Each basement culmination is mantled by shear zones that show a top-to-the-east sense of shear on the gently east-dipping eastern boundary, or dextral or sinistral shear on the steeply dipping northern and southern boundaries, respectively (Fig. 5). Thus, the overlying external zone metasedimentary rocks were transported to the east-southeast relative to the underlying Paleoproterozoic basement. The shear zones, as typified by the Jug Rock shear zone, record initial movement in the amphibolite facies with continued deformation into the greenschist facies. A lack of brittle structures shows that the shear zones ceased moving prior to reaching the upper levels of the crust.
The differences in metamorphic ages and cooling between rocks of the external and internal zones argue for significant displacement along the bounding shear zones. However, the exact timing of motion is difficult to quantify. The presence of fractured zircons overgrown by euhedral metamorphic rims within the shear zone (Figs. 7A and 7H) suggests that that metamorphic zircon growth could be related to fluid migration along the Jug Rock shear zone at 64 Ma or between 59 and 55 Ma. Although motion on the Jug Rock shear zone at this time, and earlier as a compressional structure, is allowed by the geochronologic data, we note that the youngest ages for metamorphic growth and possible deformation on the Jug Rock shear zone overlap with extension in the Valhalla complex at 59–54 Ma (Carr et al., 1987) and with the beginning of exhumation in the Bitterroot complex at 55–53 Ma (i.e., Coyner et al., 2001; House et al., 2002; Foster et al., 2007). On this basis, we conclude that the Jug Rock shear zone was active as extensional structure during the earliest phase of exhumation in the Clearwater complex (Fig. 11). Further support for this interpretation includes decreasing metamorphic conditions during shear and a sharp break in cooling ages across the shear zone. Both of these observations are more compatible with movement during extension and exhumation than during compression. Movement on the Jug Rock shear zone may have continued until around 47 Ma, when the internal zone cooled through the closure temperature for argon, and greenschist-facies mylonitic fabrics and ultramylonites formed as the internal zone and external zone were finally juxtaposed. The internal shear zones of the Clearwater complex compare favorably with the Valkyr shear zone in the Valhalla complex (Carr et al., 1987) and the Spokane dome mylonite zone in the Priest River complex (Doughty and Price, 1999), both of which are ductile extensional shear zones exposed in the core of these complexes.
The amount of uplift of the internal zone culminations relative to the outlying rocks of the external zone along these shear zones is not resolvable with metamorphic minerals and geothermobarometry alone. Average high-pressure (M2) conditions for rocks in the external zone are slightly lower than that of the internal zone, but they overlap within the uncertainty typically associated with these thermobarometers (±100 MPa and 50 °C). If the amphibolites yield erroneously high estimates of M2 pressure, as suggested by Ziegler (1991), and are excluded from the calculated estimates, there is about a 200 MPa drop in pressure across the Jug Rock shear zone. If temperatures based on oxygen isotopes are used for the internal zone (i.e., Larson and Sharp, 1998), then a temperature drop of ∼100 °C exists across this boundary. The inability of these thermobarometers to accurately define the amount of offset along the Jug Rock shear zone is partly due to the composition of the external zone rocks, which forced us to use the less robust GMBP barometer for most samples, and uncertainty associated with the complex metamorphic history and age of metamorphic events between the internal and external zones.
We envision the following sequence of events that led to creation of the Clearwater complex (Fig. 11): At the end of the Cretaceous, rocks now exposed in the Clearwater complex were undergoing metamorphism and deformation within the thickened crust of the Cordilleran orogen. Exhumation of the Clear-water complex may have begun in the Paleocene (64 Ma), but was definitely happening along the deep-seated shear zones like the Jug Rock shear zone between 59 and 55 Ma and along the bounding detachment faults by 54 Ma (Fig. 11). The St. Joe and Kelly Forks faults of the Lewis and Clark fault zone were active at this time, as was the Bitterroot complex. The eastern side of the external zone was exhumed first and cooled at around 54 Ma. Unroofing of the Clearwater complex continued on both the internal ductile shear zones and outer bounding faults until between 54 and 47 Ma, when the Jug Rock shear zone was abandoned and the west-directed White Rock–Widow Mountain fault became the dominant detachment fault that unroofed the Clearwater complex. Slip continued on the White Rock–Widow Mountain fault until ca. 42–41 Ma as a relay between the St. Joe and Kelly Forks faults. The opposing shear sense of the deep-seated mylonites and overlying ductile-brittle detachment fault is not common within most core complexes, which are typically dominated by a uniform sense of shear at all levels within the complex (i.e., Wernicke, 1985). The Priest River complex is a notable exception and provides a possible explanation for the observed structure within the Clearwater complex. Arching of the metamorphic infrastructure in the Priest River complex caused a large antithetic mylonite zone (Newport fault zone) to form on the backside of the infrastructure culmination, which broke the culmination into discrete crustal blocks. A similar scenario may have occurred in the Clearwater complex, when arching of the older, deep-seated shear zones initiated formation of the west-directed White Rock–Widow Mountain fault, which then dominated exhumation of the complex.
A combination of geologic mapping, kinematic analysis, and geochronology described in this paper provides a compelling case for interpreting the Clearwater complex as an Eocene metamorphic core complex. The direction of slip along shear zones within the complex parallels the direction of slip on the Lewis and Clark Line and in the Bitterroot complex, and 40Ar/39Ar mica cooling ages show that the complex was exhumed and cooled between 54 and 41 Ma. The cooling of the Clearwater complex coincided with an episode of crustal extension and core complex formation in the Northern Rockies (i.e., Parrish et al., 1988; Foster et al., 2007) and with motion on the Lewis and Clark Line in the vicinity (Burmester et al., 2004; Lewis et al., 2002). We conclude that the Clearwater complex formed as a mid-crustal pull-apart structure associated with a step-over in the Lewis and Clark Line.
We thank Andrew Buddington, Brian Boothe, Rebecca Pitts, Darren Tollstrup, Brandt Halver, Travis Kumm, Jason Burt, and Andrew Wiser for superlative field assistance. Potlatch Corporation and the state of Idaho provided access and some of the only outcrops available. The U.S. Geological Survey–Eastern Washington University Cooperative Program loaned us field vehicles and motorbikes. We thank Joe Wooden and staff at Stanford–U.S. Geological Survey Micro-Isotopic Analytical Center for help with collecting the SHRIMP data; Mike Hartley, Jim Vogl, and Warren Grice for assistance with the 40Ar/39Ar analyses at University of Florida; and Scotty Cornelius at the WSU (Washington State University) Geoanalytical Laboratory for help with micro-probe data acquisition. A special thanks to Andrew Buddington (Spokane Community College), whose optimistic attitude and enthusiasm made the field work enjoyable, and John Watkinson for supervising Sha's thesis and providing an open and conducive academic environment. Joe Hull, the sole attendee of the NAGT field conference, was also immensely helpful. This paper benefited substantially from constructive reviews by G. Gehrels and T. Kalakay. Reed Lewis and Russ Burmester also deserve special thanks for discussions and support. They surely don't agree with all of our interpretations. This work was supported by National Science Foundation (NSF) grant EAR01532280 to Doughty and EAR0107088 to Chamberlain.
Figures & Tables
Whence the Mountains? Inquiries into the Evolution of Orogenic Systems: A Volume in Honor of Raymond A. Price
- absolute age
- Belt Supergroup
- ion probe data
- Lewis and Clark Lineament
- mass spectra
- metamorphic core complexes
- metamorphic rocks
- metasedimentary rocks
- shear zones
- SHRIMP data
- strike-slip faults
- structural analysis
- United States
- upper Precambrian
- zircon group
- Boehls Butte Anorthosite
- Clearwater Complex
- Jug Rock shear zone
- Smith Ridge Syncline
- Boehls Butte Schist